SCI Exercise


Recovery from spinal cord injuries, whether from traumatic or non-traumatic causes, is difficult because they are highly variable and multifactorial.[1] Disruption of the spinal cord leads to diminished transmission of descending control from the brain to motorneurons and ascending sensory information.[2] The effects of this disruption are innumerable. In addition to causing paralysis of voluntary musculature and loss of sensation, other secondary complications include chronic pain, bladder and bowel dysfunction, sexual dysfunction, loss of muscle mass, osteoporosis, pressure sores, joint and muscle pain, fatigue, sleep problems, depression, autonomic dysreflexia, and loss of thermoregulation.[1] Much research has been conducted on understanding how to improve function after spinal cord injury through exercise.[1] The primary mechanism that has been investigated that explains recovery due to exercise is activity-dependent plasticity.[1]

Activity-dependent Plasticity

As previously stated in the cell biology page of spinal cord injury, there are two types of plasticity that occur in the spinal cord which are spontaneous and activity-dependent. Spontaneous plasticity is discussed in detail on the cell biology page of spinal cord injury. Activity-dependent plasticity occurs in response to sensory input from the periphery and has significant effects on cellular and molecular function to promote recovery.[1] Activity-dependent plasticity has not only been examined in animals at the cellular and molecular level, but also in humans at a functional level.[3,4] In order to induce these effects in the spinal cord, activity must be both repetitive and task specific.[1,5,6]

The promotion of the development of spinal circuits has been studied extensively in animals through repetitive, task specific activities. Researchers have shown in various studies that spinalized animals can be trained to step and stand suggesting that the spinal cord can learn without supraspinal input.[1,7] In addition to stepping and standing, rats can be trained to avoid shocks by changing foot position.[7] Based on a series of studies conducted by Grau et al,[8] this type of learning is modulated by NMDA with increased expression of BDNF and is inhibited by GABA. The mechanism which includes NMDA and BDNF begins with the nociceptive afferent input in the form of a shock, which causes the release of glutamate, substance P, and BDNF.[8] Glutamate binds to NMDA, AMPA, and mGlu receptors; whereas, substance P and BDNF bind to their respective receptors NK1 and TrkB.[8] The influx of calcium as a result leads to activation of cAMP-dependent PKA, PKC, and CaMKII.[8] Along with tyrosine kinase Src, these kinases phosphorylate NMDA and AMPA receptors increasing excitability.[8] The insertion of more AMPA receptors and ERK, which phosphorylates Kv4.2 potassium channel, also enhances postsynaptic function.[8] This happens with prolonged stimulation that causes ERK translocation to the cell nucleus where it initiates the transcription factor cAMP promoting gene expression.[8] The importance of NMDA has been supported by studies that administered NMDA antagonists and found that the antagonists inhibited learning depending on the dose.[8] The effects of GABA are inhibitory to learning as GABA agonists disrupt learning depending on the dose.[8] Furthermore, supraspinal input has also been shown to be involved in spinal cord learning.[7,9,10] Rats will adapt their H-reflex response for rewards, which shows that supraspinal input is also necessary for learning as well.[7,9,10] In a study conducted by Wang et al,[9] the effects of down-conditioning the H-reflex on GABAergic input were explored. The rats in the experimental group were implanted with electrodes to stimulate the H-reflex in the right posterior tibial nerve and underwent down-conditioning protocol for the H-reflex.[9] The rats would receive a food reward for successful down-conditioning of the H-reflex.[9] Soleus motorneurons of the rats were then examined for quantity of GABAergic terminals and GAD.[9] In the rats that were successful at down-conditioning their H-reflexes there were significantly greater amount, size, and GAD density of GABAeric terminals compared to the rats that were not successful at down-conditioning or control rats.[9] These findings suggest that there is supraspinal input from the corticospinal tract affecting the H-reflex by changing motorneuron firing threshold.[9] Next, the same researchers sought to find out the opposite about GABAergic terminals.[10] They used the same design as the previous study except that they rewarded up-conditioning instead of down-conditioning of the H-reflex.[10] Unlike the previous study where down-conditioning increased the amount of GABAergic terminals, this study did not change GABAergic terminal number significantly, though there was some increase in terminal diameter and soma coverage.[10] In combination the findings of the two studies imply there are different mechanisms of plasticity involved in up-conditioning compared with down-conditioning.[10]

The quantity and timing of activity needed to promote plasticity have also been investigated. In an attempt to answer the amount of activity, Cha et al.[11] used a robotic device to move the legs of spinalized rats on a treadmill for either 1000 steps per training session or 100 steps per training session 5 days per week for 4 weeks. Training began 23 days after the spinal transection.[11] During training the robotic device counted steps and forced the hind limbs to bear as much weight as possible from body weight support.[11] After 4 weeks of training, there was not a significant difference in number of steps taken between groups.[11] The quality of stepping in the group that received 1000 steps per training session improved in their ability to bear varying amounts of weight on the hind limbs at different speeds, whereas the group that receive 100 steps per training session did not improve as much in quality.[11] This study supports the idea that more activity is better for recovery.[11] Another study that answers more activity is better was conducted by Engesser-Cesar et al.[12] and focused on frequency of activity per week rather than amount of activity per session. In this study, mice with a contusion injury were granted access to a running wheel either 3 days per week, 7 days per week for 15 weeks, or not at all.[12] Wheel running was initiated 1 week after injury.[12] Based on observations of locomotion, there was an improvement across groups.[12] Based on kinematic analysis, both wheel-running groups significantly improved stepping ability compared with controls, but those that ran 7 days per week showed higher quality stepping.[12] Based on these results it can be inferred that more frequent activity is better for recovery.[12] Another aspect of this study looked directly at the spinal tissue for evidence an increased number of neurons in the injury segment and increased length of serotonergic fibers caudal to the lesion.[12] While there was no an increased number of neurons, there was increased serotonergic fiber length caudal to the lesion in the wheel-running groups, but not the control group.[12] This shows that activity not only rearranges spinal circuits, but also promotes regenerative sprouting.[12] The timing of activity after spinal cord injury is very important for recovery as well. The general consensus is that it is best to begin activity as soon as safely possible because if the activity is too early there may be exacerbation of the injury.[1,13] In a study conducted by Smith et al,[13] which utilized swimming instead of wheel or treadmill running, found that acute training is less effective than training initiated 2 weeks after spinal cord injury. Additionally, they determined that following 8 minutes of swimming at 3 days post injury there was exacerbation of inflammation at the injury site.[13] Much more research needs to be conducted to determine the most appropriate dosage and timing of training to improve activity-dependent plasticity.

Cellular Effects in Animal Models

Benefits of activity-dependent plasticity at the cellular level are related to its ability to increase or decrease local factors.[2] Examples of cellular entities that are known to be up-regulated by activity include neurotrophins such as BDNF, GDNF, IGF, NT-3, and NT-4.[1,6,14] Neurotrophins are a family of proteins that are secreted to enhance neuronal survival.[1] Additionally, factors that inhibit axonal regeneration are decreased with activity. The nogo-binding receptor NgR is thought to be down-regulated.[1,15]


After Spinal Cord Injury, the body’s own neuronal system has the capacity to endogenously produce neurotrophins which can aid in recovery after SCI, primarily through CNS plasticity.[16] There has been a large body of literature which consistently shows that many select neurotrophins can be modulated and up-regulated by an external stress or stimulus such as exercise. First, Brain Derived Neurotrophic Factor (BDNF) will be discussed on how this relates to exercise following SCI. To get a better understanding of why this is so important, one must appreciate some of the known benefits of BDNF as related to our central and peripheral nervous systems. BDNF has been shown to increase axonal growth, improve transmission through synapses that are involved in motor recovery, stimulate hind-limb stepping response in SCI injured mice, and protect existing neurons from further damage.[16] This all leads to the conclusion that if we can modulate this important neurotrophin then activity dependant neuroplasticity may result.

A common method for tracking BDNF changes following an exercise protocol is by looking at the phosphoprotein Synapsin I.[16] Synapsin I can tell us a lot about BDNF’s downstream effects on synaptic function and the way synapses adapt and react.[16] This helps researchers narrow in on the proposed plastic effects of this particular neurotrophin. When BDNF binds with its primary receptor trkB, a signaling cascade takes place in which trkB phosphorylates a MAP kinase which in turn modulates the phosphorylation sites on Synapsin I, thus creating a causal link between BDNF and Synapsin I expression.[17] Since Synapsin I is involved in modulation of synapses, axons, and neurotransmitter release, if we are able to increase BDNF with exercise, then plasticity following SCI can take place via the terminal effects of proteins like Synapsin I.[16]

synapse.jpg Source:

Above Picture: A general schematic how growth factors like BDNF affect neurons and supporting cells making their way to the cell nucleus where plasticity can occur.

In a study by Ying et al[16], it took just three days of exercise on a running wheel for spinal cord injured rats to normalize their BDNF mRNA levels to that of control rats.[16] In addition, by the 28th day of exercise, the spinal cord injured rats’ BDNF increased 133% as compared to the control group.[16] In order to link this effect to plasticity after SCI, the researchers looked at Synapsin I and found that the levels progressively increased with exercise and at the 28 day marker the spinal cord injured rats had similar levels as compared to controls.[16] The type of injury in this study was a mid thoracic hemi section, and the mode of exercise was a running wheel with progressive resistance.[16]

It is well known that exercise can aid in the recovery following spinal cord injury in both animal and human research. The Ying et al. study is important because it looks at the improvements at a molecular level.[16] They theorize that in a hemi section SCI, decreased or absent supraspinal and sensory inputs change the way BDNF is translated.[16] This is why levels drop significantly following injury. Exercise was shown to normalize the BDNF levels to control rat values in a very short time period. When talking about timing of exercise, this study chose to initiate exercise one week post injury.[16] They based this tactic on previous studies showing that a delay in exercise can lead to better neuroplastic effects.[16] In conclusion, a proposed mechanism on how exercise can improve neuroplasticity can be inferred. Since BDNF gives a phosphate group to Synasin I through the trkB receptor, this protein kinase pathway then is free to modulate neurotransmitter release.[16] Since exercise has been shown to increase BDNF levels, this may then be the reason for the increase in Synapsin I expression, leading to improvements in synaptic plasticity.[16]

Another study by Gomez-Pinilla et al. strengthen the findings of the Ying et al. article and propose additional aspects on how exercise can lead to positive molecular effects involving neurotrophins. Exercise was performed by rats for 3 or 7 days on a standard running wheel.[18]The control group consisted of sedentary rats. Not only were BDNF and Synapsin I increased but other known neural growth proteins as well, including growth associated protein 43 (GAP-43), and cyclic AMP response element-binding molecule (CREB).[18] These molecules have been shown to be active in supraspinal learning and motor learning in feline models where the spinal cord is transected and molecular analysis is performed to study what specifically is active during functional recovery from SCI.[18] After only 3 days of voluntary exercise, BDNF levels increased 156% compared to sedentary controls and remained this high at the 7 day marker.[18] What is interesting is that there was a direct relationship between BDNF mRNA and the distance that the rodents ran on the wheel.[18] This implies a dose response which could be useful when prescribing exercise for those with SCI. To find out where the BDNF was having its action, immune-staining techniques were used and this was able to determine that BDNF was found mostly in the ventral horn motor neurons, substantia gelatinosa, and in astrocyte cells of the lumbar spinal white matter.[18] This means motor tracts, myelin areas of the cord, and nerve supporting cells are showing a positive plastic effect. In spinal cord injury this is important to know because these are the areas that are often disrupted and damaged and in need or trophic support for recovery.

In the same study, they went on determine whether muscle action of peripherally contracting muscles played a role in this mechanism.[18] To do this they paralyzed the soleus muscle of the rats and found that both locally and in the spinal cord, BDNF levels were decreased compared to rats with intact soleus muscles during exercise.[18] This demonstrates that active muscle contraction during running helps modulate BDNF levels in both the contracting muscle and in the spinal cord, however the specific mechanism is unknown. In conclusion, voluntary exercise appears to up-regulate many genes associated with both CNS and PNS plasticity.[18] BDNF is a central factor in these complex cellular mechanisms and may be a catalyst of how exercise assists in the functional improvement of the neuromuscular system.[18]

In terms of when to begin therapy Krajacic and colleagues have highlighted some important timeframe specific implementations of exercise in their research with rodents.[19] In summary, they found that specific reach training that was delayed 12 days post injury resulted in better functional performance of not only reaching but also improvements in an untrained task (in this study they used how well the rats climbed a horizontal ladder without being trained in it.)[19] At 12 days, BDNF was expressed to the same level as those rats that inititated rehab immedately post injury.[19] The main benefits stated in the article for waiting include decreased damage to the white matter near the lesion, and the delay helped prevent impairments in untrained tasks as compared to earlier initiation of training.[19]

In rat models, it has been extensively shown that exercise can increase BDNF which then leads to neuroplastic effects. In spinal cord injury it is also important to see if these effects translate to humans. It has been shown that exercise can impact BDNF and other neurotrophins in spinal cord injured humans, but considering the different techniques that must be used in humans, and lack of standardized control of injury, the response varies from animal models. A study by Rojas Vega et al.[20] sheds some important information on what we know about neuroplasticity proteins and response to exercise.[20] First, it was found that at rest, BDNF concentrations in the blood were 6 times higher in paraplegic athletes than in able bodied people. Varying levels of exercise intensity also produced different responses related to BDNF, Cortisol, and IGF-1.[20] With just 10 minutes of moderate intensity handbiking with a heart rate of 54% max, BDNF increased 1.5 times from the typical at rest levels.[20] When the intensity was increased to 89% max heart rate, a slight decrease in BDNF from basal levels was found.[20] However, stress hormones including cortisol, IGF-1, and prolactin were all elevated in response to the high intensity handbiking protocol as was expected. This is a normal response to exercise which is found in able- bodied persons under similar conditions.[20] It has been suggested that the elevated levels of stress hormones with exercise may counteract the BDNF levels so that neurogenesis may not operate at optimal levels.[20] At the cellular level, BDNF is able to cross the blood brain barrier through a saturable transport system which basically means that a trans-membrane carrier is able to bind to BDNF and carry it across the plasma membrane, however this type of system helps regulate the passage in a way that if all carriers are bound, then no more BDNF can cross into or out of the brain.[21] This is important because it is thought that a large portion of circulating BDNF originates in the CNS and is transported to the spinal cord and muscles through axons in order to aid with trophic support.[20] In summary, moderate exercise may be more beneficial in increasing neuroplasticity than extremely high levels of exercise. It is also important to note that these effects occurred in chronically injured paraplegic elite athletes, and obvious limitations in collecting data were observed secondary to not being able to study the cord and brain like is done commonly in sacrificed animal models.[20]


Glial cell line-derived neurotrophic factor (GDNF) is a neurotrophic factor capable of helping damaged axons to regenerate after SCI and can also modify and strengthen neural synapses to help promote recovery of function after SCI.[2,14,22,23,24,25] Neurotrophic factors like GDNF tend to respond best when paired with selective receptors. GDNF has been shown to have enhanced expression when coupled with receptors like GDNF-family receptor alpha 1 (GFRα1) and receptor tyrosine kinase (RET).[2,23,24] Therapeutic research for SCI has focused on how to increase the expression of this beneficial neurotrophic factor in order to optimally promote neural plasticity and recovery of function.

A study by Keeler et. al[2] looked at changes in the expression of a variety of factors that have been implicated in the secondary damage and functional recovery following SCI, including neurotrophins, caspases, heat shock proteins, and markers of astrocyte activity.[2] The study looked at both acute (10 days post-SCI) and more chronic (31 days post-SCI) changes in female rats following no transection (control group), transection alone, or transection coupled with exercise.[2] Initially, the investigators looked for changes in the spinal cord as a whole, and then delved even further to identify if the changes took place in the motoneurons, the intermediate gray matter, or the dorsal root ganglion (DRG).[2] The protocol divided 30 female rats into 5 groups.[2] Rats in the transection groups received a complete thoracic transection at the T10 level.[2] Rats in the exercise groups participated in passive hindlimb cycling 2 times per day, 30 minutes per session, at a rate of 45 rpm.[2] Rats began exercising 5 days after their injury, for a total of 5 days of exercise for the Tx+Ex 10 group and 20 total days of exercise for the Tx+Ex 31 group.[2] Some of the results for changes found in GDNF and its receptors are summarized in Table Number. Additionally, changes found in Neurotrophin 3 and its receptor TrkC are listed but will be described in detail in the NT-3 section.

Table 1: Changes in gene expression following transection alone or transection plus exercise[2]

Animal Group GDNF GFRα1 RET NT-3 TrKc
Tx 10 Days Whole cord protein expression increased compared to controls. Transient increase in motoneuron expression. No significant changes. No significant changes. No significant changes.
Tx 31 Days Whole cord protein expression returned to control values. Expression in motorneurons had returned to control levels. No significant changes. Whole cord protein expression increased compared to control and Tx 10. No significant changes.
Tx+Ex 10 Days Increase in whole cord protein and mRNA expression; Increased expression in motoneurons and intermediate gray matter. The increase seen with motoneuron expression in Tx 10 group was decreased significantly with exercise. Whole cord expression increased in response to exercise. Whole cord mRNA expression decreased compared to Tx 10 group. Increased expression in motoneurons and intermediate gray matter. No significant changes.
Tx+Ex 31 Days Increased expression in whole cord compared to controls but decreased from Tx+Ex 10. Sustained increased expression in motoneurons; Expression in intermediate gray matter returned to control values. Increased DRG expression. Increased whole cord expression as a result of exercise. Increased whole cord protein expression compared to controls and Tx+Ex 10 groups. Intermediate gray matter and motoneuron expression returned to control values. Increased whole cord protein expression in absence of significant changes in NT-3 expression.

As seen in Table 1, GDNF protein and mRNA increased after 10 days of exercise throughout the whole cord, but specifically in the motoneuron pools and intermediate gray area.[2] This was paired with a decreased expression of GFRα1 and an increase in RET receptor mRNA levels.[2] After 31 days of injury and 20 days of exercise, the GDNF protein and mRNA levels had declined slightly in the intermediate gray area but remained elevated in the motoneurons.[2] Both receptors increased at the 31 day mark as a result of exercise, and GFRα1 was specifically increased in the dorsal root ganglion.[2]

Cote et. al[14] conducted another study to support the beneficial role of GDNF after SCI.[14] This study used 30 female rats and randomly assigned them to 3 groups: a control group (n=8), a passive cycling exercise group (n=8), and a treadmill step training exercise group (n=14).[14] The rats received a complete spinal cord transection of the T10-11 spinal cord level and began their respective exercise regimens 5 days after the SCI.[14] The rats in the two exercising groups participated in 15 minutes of training per day, 5 days per week, for 4 weeks.[14] While both exercise groups had increased GDNF expression compared to sedentary rats, step training had a more significant impact than the passive bike-training program.[14] Above the lesion, GDNF protein levels increased in both forms of exercise as compared to the control groups.[14] These results show that treadmill step training that is initiated 5 days post-SCI may help facilitate increased expression of the neurotrophic factor GDNF.

Implications & Recommendations

These are just two animal model studies showing that exercise can be used as a means to increase levels of the neurotrophic factor GDNF, as well as its receptors in the injured spinal cord.[2,14] Both studies initiated exercise 5 days post-SCI and used cycling and stepping to elicit positive results.[2,14] The benefits appeared to last longer in the study by Cote et. al because levels of GDNF had begun to decline by day 31 in the study by Keeler et. al.[2,14] The study by Keeler et. al, however, used a greater frequency of exercise (two 30 minute sessions per day versus one 15 minute session per day) so dosage could be a factor limiting the beneficial production of GDNF.[2,14] Further studies need to be done to determine more specific dosage guidelines and also what intensity of exercise is needed to elicit these results in humans. Overall, it appears beginning 5 days post-SCI and participating in either cycling or stepping 5 days per week for several weeks can elicit positive increases in GDNF after SCI.[2,14]


IGF-1 is important for recovery following spinal cord injury because this growth factor promotes both oligodendrocyte and neuron growth and repair while also is associated with neural plasticity as it is able to be differentiated to create new pathways in the brain and spinal cord.[26] It is also thought that IGF-1 may be an upstream mediator for BDNF, and linking this fact with the known actions of BDNF found above, IGF-1 may be able to regulate neurogenesis and the pathway on how exercise as a stimulus can protect the spinal cord from further damamge.[26] Researchers also believe that IGF-1 plays an important role in mediating various growth factors and their passage through the BBB back and forth from the brain to the periphery.[26] When people with spinal cord injury perform exercise, IGF-1 is secreted in peripheral muscles. This is a catalyst for neuroprotective effects, and what is interesting is that when IGF-1 blocking antibodies are administered during the same exercise bout, the neuroprotective pathways are reduced significantly.[26] It can be hypothesized that since BDNF is responsible for many CNS and PNS plastic changes, then IGF-1 might be the upstream driving force to help regulate the action of these important neurotrophin cascade events.[26] In summary circulating IGF-1 following exercise is able to transport through the BBB and act on other neurotrophic factors including BDNF which then is responsible for positive myelin and neuron protective effects.

When looking at T8 contusion injured rodents exercising in an enriched environment, EE, (consisting of running wheel and climbing frames, paper toys and housing), it has been shown that circulating IGF-1 in combination with exercise may help improve outcomes in rodents.[27] To come to this conclusion, rats that were given an IGF-1 blocker before entering the EE had decreased functional recovery and also showed lessened improvement in the CPG area of the spinal cord when looked at 42 days into the experiment.[27] In summary when circulating IGF-1 is potentiated and combined with enriched environements, better hindlimb coordination results as compared to the same environement with lower levels of IGF-1.[27] So it appears IGF-1 is involved in how exercise promotes functional recovery, however, other neurotrophic factors may also be responsible in creating an environment for exercise to potentiate beneficial effects including BDNF, GDNF, NT-3 etc. Further studies should look at the effects each specific neurotrophin has under the same conditions as detailed in the Koopmans et al. article.

Very few studies have been done looking at how much exercise is required to elevate IGF-1 levels that are associated with positive neuronal changes in humans. However a study by Rojas et al. investigates this question.[20] It was shown that 10 minutes of moderate handbiking in paraplegic athletes was enough to raise IGF-1 levels as consistent with normal exercise response.[20] However, unlike BDNF, IGF-1 remained elevated even with high intensity long duration handbiking at the level of 85 minutes with a heart rate of 89% max.[20] This is important because since IGF-1 is thought to be an upstream regulator of BDNF, elevated levels may in the long run lead to increased BDNF and in turn higher levels of neural plasticity. Looking at IGF-1 itself, it has been shown to support injured neurons, inhibit apoptosis of brain neurons involved in memory and axon growth, and influence progenitor cells in the way neural protective proteins are expressed.[20] So based on this human SCI study, it can be recommended that both moderate and high intensity exercise will be related to possible benefits as related to IGF-1 independently and it can be hypothesized that this increase will then lead to further action of BDNF and its downstream neuroplastic and neuroprotective effects.[20]


Neurotrophin 3 (NT-3) is involved with regulating synaptic transmission, and helps promote plasticity and regeneration in both the spinal cord and muscles following SCI.[28,29,30] NT-3 and its main receptor, TrkC, have also been found to have a role with helping sensory neurons to survive after SCI.[29] NT-3 has been shown to increase with exercise in both the intact and impaired spinal cord, as well as in muscle fibers.[30]

Exercise Models on Healthy Animals

A study by Ying et. al[28] used rats to examine the effect of wheel running exercise on NT-3 and TrkC expression in the lumbar spinal cord and soleus muscle after both 3 and 7 days.[28] The study used 10 healthy male rats and split them into 3 groups: a control group, a group that exercised for 3 days, and an group that exercised for 7 days.[28] The authors were interested in looking for changes in the lumbar spinal cord because this is where the motoneuron pools responsible for hindlimb movement are located; additionally the soleus was studied because it is heavily recruited during running exercise in rats.[28] The authors analyzed NT-3 mRNA levels, NT-3 protein levels, and TrkC mRNA levels and Table 2 summarizes the findings:

Table 2: NT-3 and TrkC Expression in Spinal Cord and Soleus Muscle[28]

Group NT-3 mRNA Lumbar Cord NT-3 Protein Lumbar Cord NT-3 mRNA Soleus NT-3 Protein Soleus TrkC mRNA Lumbar Cord TrkC mRNA Soleus
3 Days Exercise Significant increase 145% Non-significant increase 117% Significant increase 187% Non-significant increase Significant increase 154% Significant increase 150%
7 Days Exercise Significant increase 187% Significant increase 144% Slight decline 139% No significant change Significant increase 157% Returned to control levels

Percentages listed are percent changes from sedentary control group. Additionally, further analysis revealed that NT-3 increases appeared more in the dorsal horn of spinal cord.[28]

As compared to sedentary controls, exercise appears to significantly increase the NT-3 mRNA and protein levels in the spinal cord noticeably after 7 days of voluntary wheel running.[28] There was some initial increase in NT-3 mRNA in the soleus, but this did not last at 7 days and was not coupled with an increase in the protein expression.[28] Also, the increase in NT-3 in the lumbar spinal cord was couple with an increase in its receptor TrkC, and there was less of a change in TrkC in the soleus, which correlates with the NT-3 changes found.[28]

The study by Cote et. al[14] that was described previously in the GDNF section, also looked at female rats' spinal cord expression of NT-3 following either no exercise, 3 days of exercise, or 7 days of exercise.[14] The study showed that both treadmill step training and passive cycling significantly increased the levels of NT-3 as compared to sedentary controls and that the increase was slightly greater for rats completing the step training.[14]

A randomized controlled trial by Pinilla et. al[30] examined the effects of neuromuscular activity on the expression of NT-3 and BDNF in the spinal cord and soleus of rats.[30] The rats were randomly divided between 2 groups: a control group (n=29) and an exercise group (n=56).[30] The rats in the exercise group participated in 30 minutes of treadmill running per day at 27 m/min on a 3% incline.[30] The rats were trained for either 1 or 5 days and then were killed either immediately (0), 2, or 6 hours after the last exercise session.[30] The authors found a progressive decrease in levels of NT-3 mRNA expression in the spinal cord after 1 day of exercise, with the lowest level found in rats killed 6 hours after exercise.[30] When trained for 5 days, however, NT-3 expression was significantly increased compared to the control groups.[30] In the soleus muscle, NT-3 levels were signficantly increased after 1 and 5 days of training.[30] Treadmill training at this intensity significantly affected the expression of NT-3 in both the lumbar spinal cord and soleus muscle in healthy rats and provides yet another burden of proof of importance for studying this in SCI injury.[30]

Implications & Recommendations

In healthy male rats, wheel running produces increased expression of NT-3 and its receptor TrkC in both the lumbar spinal cord and soleus muscle.[14,28,30] The increased expression of NT-3 mRNA in the lumbar cord and associated increase in NT-3 protein implies that the spinal cord is actually producing endogenous NT-3 in response to exercise, rather than recruiting it from other areas of the body.[28] This study shows that running can be an effective way to upregulate NT-3 in healthy rats, and leads to the justification for studies examining the effects in patients with spinal cord injury. The stronger increases in NT-3 expression seen in the substantia gelatinosa of the dorsal horn supports the idea of NT-3 being crucial in sensory regulation.[28] Besides the spinal cord, NT-3 is known to be found in the afferent cell bodies located in muscle spindles of adult rodents.[14,28] The changes in NT-3 and TrkC found in the soleus muscle in these studies suggest they may play a role in mediating the sensory components associated with muscle activation.[14,28,30] The overall bottom line from these studies on healthy animals is that exercise is crucial in modulating NT-3 and TrkC both in the spinal cord and in the muscles of healthy rats, and carry-over of results into pathologies such as SCI must be examined.

Animal SCI Models

The study by Keeler et. al[2] described above in Table 1 shows the results of SCI and exercise on expression of NT-3 and its receptor TrkB.[2] At 10 days after injury, there was no significant change in NT-3 or TrkC in throughout the cord, but by Day 31, there was a slight increase in the protein levels of NT-3.[2] Two groups of rats in this study participated in passive hindlimb cycling beginning 5 days after injury and lasting for either 5 or 20 days.[2] The rats were subjected to 2, 30 minute sessions a day.[2] At 10 days there was increased NT-3 expression in motorneurons and intermediate gray matter.[2] At 31 days there was decreased expression of NT-3 in these areas (similar to control values) but an increase in TrkC levels throughout the cord.[2] This study provides some evidence of exercise-induced changes in NT-3 expression following SCI.[2]

Another research study by Ying et. al[29] looked at the impact exercise has on the decreased levels of NT-3 found after SCI and how this can help with synaptic plasticity and axonal growth distal to the injury site.[29] The study used 76 male rats and split them into a sedentary control (Con) or hemisected (Hx) group.[29] The hemisection was at the T7-T9 level and these rats were further divided into Ex-Hx and Sed-Hx groups.[29] Starting 1 week post-SCI, the Ex-Hx rats participated in wheel running for 0, 3, 7, or 28 days.[29] Results indicated that hemisection alone did not affect the levels of NT-3 expression, but there was a significant expression in NT-3 levels following 28 days of exercise compared to sedentary Hx control rats.[29] Overall, this study indicates that when initiated 1 week post-SCI, wheel running for at least 4 weeks can induce positive increases in NT-3 as compared to sedentary rats with SCI.[29]

NT-3 is currently being studied for its role in modifying forms of chronic pain such as hyperalgesia (increased response to painful stimulus) and allodynia (response to non-painful stimulus), both of which are commonly found after SCI.[31] A study by Sharma et. al[31] sought to find a link between exercise, NT-3 levels, and chronic pain.[31] The study used 40 female mice and split them into a total of 4 groups: acid injection without exercise, acid injection with exercise, sham saline injection without exercise, and sham saline injection with exercise.[31] The acidic injection has been shown to invoke mechanical hypersensitivity in the skin, organs, and muscles and thus was used in this study.[31] The exercise protocol used consisted of treadmill running for 30-45 minutes (depending on the week) per day, 5 days per week, for 3 weeks.[31] The mice ran progressively longer and faster each week but in general the intensity was a moderate one.[31] The investigators looked at the hypersensitivity to cutaneous and muscular stimuli as well as NT-3 mRNA and protein levels in the gastrocnemius and soleus muscle tissues.[31] Cutaneous and muscular hyperalgesia was induced on both the ipsi- and contra-lateral hindlimbs for the mice injected with the acid but not the saline.[31] Exercise was able to significantly decrease the heightened response for the ipsilateral hindlimb.[31] While exercise did not show a significant effect on soleus levels of NT-3 (mRNA, protein), there was a trend for increased NT-3 mRNA in the gastrocnemius of the sham treated mice without exercise as compared to the sham treated mice with exercise.[31] Taken together, these results suggest that exercise may increase expression of NT-3 in some muscle fiber types and this may lead to a decreased response to pain.[31]

Implications & Recommendations

When examing the effects of exercise on both healthy and spinal cord injured rats, it appears that an increased expression of NT-3 and its receptor TrkC can occur both in the spinal cord and in muscle tissues. Both treadmill running and passive cycling interventions were used, with varying intensities, frequencies, and durations. It seems in general the exercise can be initiated 5-10 days post-SCI, and will need to be completed for several weeks before the benefits are noticeable. Further research should be done to find the optimal dosage in human models.


Neurotrophin-4 (NT-4) is a neurotrophic factor involved in the survival and preservation of neuronal conduction. NT-4 and its receptor, TkrB, have both been implicated in regeneration and plasticity in the spinal cord as a natural occurrence and following SCI.[2,14] The effects exercise on both NT-4 and TkrB have been investigated in animal models, demonstrating increasing expression of each.[2,14,32]

Healthy Animal Models and Exercise

A study by Skup et. al.,[32] used 12 healthy male Wistar rats and randomly assigned them to two groups: Exercise and no exercise.[32] The rats in the exercise group participated in treadmill training 5 days a week for 4 weeks. They walked approximately 1000 m/day, at 20-25 cm/s, which was gradually increased over the duration of the study.[32] They discovered that NT-4 immunoreactivity within the white matter was enhanced by exercise in both mRNA and protein levels.[32] Although mRNA levels were increased in the gray matter of this neurotrophin, the same increase was not seen in its protein levels. This indicates an interruption in the translation process, thus reducing the ability to facilitate neuronal survival or regeneration within the gray matter.[32]

Exercise in Animals Models with SCI

Two other studies[2,14] also investigated the effects exercise has on NT-4 and TkrB expression in rats with SCI. In rats with SCI, exercise has been proven to increase NT-4 and TkrB in the spinal cord. In a randomized control trial[14], Sprague-Dawley rats with complete spinal cord transection at T12 were assigned to three groups: no exercise, passive cycling, and treadmill training. Those participating in exercise did so for 15 minutes each day, 5 days a week, for a total of 4 weeks.[14] Expression of NT-4 and TkrB were each studied.[14] Cycling and treadmill training each demonstrated significant increases in NT-4 and TrkB in the lumbar spinal cord.[14] However, those in the cycling group demonstrated greater increases of both NT-4 and TkrB at L1-L3, but little differences between the two were evident at L4-L6.[14] This difference in expression at various lumbar spinal levels may be due to the spinal enlargements at the lower lumbar levels.[14]

The study by Keeler et. al.[2] demonstrated improvements in white matter of NT-4 mRNA levels at 10 days post-SCI, in rats that had not participated in exercise.[2] Interestingly, the same improvements were noted in TrkB, however TrkB protein levels also increased.[2] However, exercise did not promote increased expression of these.[2] When looking at the motoneurons, however, damage to the spinal cord did increase generation of NT-4 and TrkB, and this was further increased with exercise, in both proteins and mRNA levels.[2] As signified by this study, the body attempts to maintain neurotrophins and receptors as a survival mechanism following SCI.[2]

Here, research shows an interruption in the survival mechanism in the spinal cord, because of the differences noted in mRNA and proteins levels in the gray and white matter.[2,14] This demonstrates the ability of NT-4 to be transcribed, but because the protein levels did not always increase, there is some sort of interruption preventing NT-4 from being transcribed, reducing its ability to carry out its responsibilities within the white matter following SCI. This is not seen in the gray matter in rats with SCI, supporting the idea that NT-4 and TrkB act within the gray matter to attenuate damage to the central nervous system after SCI. The differences in expression observed between the white and gray matter may indicate that this neurotrophin actually acts on immature oligodendrocytes within the gray matter as a means to promote regeneration.[2]


The role of NgR in plasticity after spinal cord injury is inhibitory as it binds the three myelin-associated inhibitory molecules thus preventing plasticity.[15] This process is discussed in detail on the spinal cord injury cell biology page. The specific role of NgR after spinal cord injury in response to exercise has not been directly explored; however, Josepheson and colleagues (as documented by Borrie et al)[15] found that NgR expression decreased in the hippocampus and cortex in normal rats that participated in wheel-running. After 3 weeks of running, unfortunately these effects returned to normal levels.[15] It has also been found by Endo et al.[33] that NgR and its co-receptor LINGO-1 are down-regulated, whereas, BDNF is up-regulated in cortical areas affected by spinal cord injury as determined by using fMRI. Perhaps based on these studies a hypothesis could be made that activity further decreases NgR expression in the central nervous system after spinal cord injury.[15] Much research has been conducted in improving function after spinal cord injury by genetically or pharmacologically blocking myelin-associated inhibitory molecules from binding with the receptor NgR.[34,35] While it is known that both activity and blocking myelin-associated inhibitory molecules improves recovery after spinal cord injury, researchers now are asking which is better and are there more benefits with both.[34,35]

The first study that attempts to answer those questions was conducted by Maier et al.,[34] which compared an anti-Nogo-A antibody, treadmill training, and both treatments in rats. The treadmill training consisted of 20 minutes of bipedal training followed by 20 minutes of quadrupedal training 5 days per week for 8 weeks starting 1 week post injury.[34] The intensity of training was 7 cm/sec and eventually reached a maximum of 21 cm/sec.[34] Additionally, the inclination of the treadmill was increased each week to reach 10% by the 5th week.[34] They found that rats that received only anti-Nogo-A antibody demonstrated consistent, coordinated steps that were short and high with little toe dragging.[34] Rats that only received treadmill training had variable, coordinated steps that were long and low with much toe dragging.[34] Rats that received both demonstrated inconsistent, uncoordinated steps with a large amount of paw dragging.[34] On a novel locomotor task, both anti-Nogo-A antibody and treadmill trained rats performed significantly better.[34] The researchers could not explain the differences between groups with histological evidence.[34] There was no increase in nocieptive fiber growth, and there was no interference of treadmill training with anti-Nogo-A antibody treatment-induced regeneration and sprouting of corticospinal tract and serotonergic fibers.[34] Based on the results of this study, more research needs to be conducted to determine the timing and dosage to best maximize the benefits of both treatments.[34] Individually, both anti-Nogo-A antibody and treadmill training improve stepping in rats, but when they are used in combination the effects are detrimental.

The second study performed by Harel et al.[35] compared the effects of exercise training on wild type and NgR-null genotype rats with mild cervical spinal cord injuries. The training was conducted on novel multimodal devices including nutating wire platforms and suspended wire mesh cylinders.[35] Rats in the training group were placed in these apparatus for 1 hour each 5-6 days per week for 14 weeks. Training began 11-12 days after injury.[35] They found that neither training nor genotype affected open-field observations of function or treadmill ambulation.[35] For the task specific test of climbing an inclined wire grid, training increased the magnitude and rate of improvement in performance for both genotypes, but those with the NgR-null genotype and training at 8 weeks had only 5.75 error points compared with trained wild types that had 9.29 error points and untrained wild types that had 13.73 error points.[35] For nonspecific tests of balancing on a rotarod and grip strength, only mice with the NgR-null rats showed improvement.[35] At 13 weeks, untrained NgR-null rats performed better than trained NgR-null rats on the rotarod.[35] For grip strength, all NgR null rats improved injured forepaw strength. Histologically, there were no significant differences across genotypes or training with respect to spinal tracts and lesion sizes.[35] Trained rats tended to do better on task specific tests, whereas NgR-null rats did better on nonspecific tests.

Implications & Recommendations

In contrast with the first study, the second study shows more improvement in some activities when both activity and NgR elimination are implemented. The first study had detrimental effects when both were implemented.[34] There were differences in types and amount of time per session of exercise between the studies, which could contribute to the discrepancy.[34,35] Both studies waited at least one week after injury to implement activity.[34,35] Since in both studies rats who participated in activity improved, except for the combination rats in Maier et al., this may be a beneficial time to start activity to improve recovery. It is necessary for more research to be explored to find the appropriate therapeutic dosage for each treatment to achieve greater recovery from spinal cord injury.

Functional Effects in Human Models

In humans with incomplete spinal cord injuries activity-dependent plasticity is seen at a functional level. Common methods of exercise used in human studies include locomotor training (using manual or robotic assisted body weight supported treadmill training or overground training),[5] bicycle training[4], and repetitive upper body training.[3]

The multicenter randomized controlled trial performed by Dobkin et al compared body weigh support treadmill training with an overground mobility intervention within 8 weeks of injury.[5] Subjects were an average of 4.5 weeks post injury and received an average of 45 1-hour treatment sessions no matter the treatment type.[5] After 12 weeks of treatment there was no significant difference between the two techniques.[5] Most ASIA C and D patients achieved functional walking ability; however, few ASIA B patients did.[5] This study shows how important it is to begin locomotor rehabilitation early to gain functional ambulation and that those with less motor and sensory impairments are more likely to be able to achieve functional walking.[5]

The study conducted by Astorino et al tested the effects of two 30 minute sessions of active upper body and passive lower body exercise on individuals with chronic spinal cord injuries.[3] The intensity of the exercise was 15 rev/min for the first 10 minutes and increased to 30 and 45 rev/min at minutes 10 and 20 with the legs moving actively and the arms moving passively.[3] Exercise did significantly increase HR, systolic BP, RPE, and VO2 compared to rest, but at smaller increments than normal.[3] This suggests that this machine could be beneficial in rehabilitation of individuals with chronic spinal cord injuries.[3]

Phadke et al found that after bicycle training individuals with spinal cord injury significantly increased H-reflex depression.[4] Subjects had been injured for a range of 3 months to 3 years and performed either 20 minutes of locomotor training or bicycle training.[4] Locomotor training was conducted with 40% body weight support at a velocity of 1.2 m/s.[4] Bicycle training was also conducted at 1.2 m/s.[4] Reflexes are hyperactive after spinal cord injury, so the ability of the spinal cord to depress the H-reflex shows potential in reducing hyperreflexia, which is beneficial for walking.[4] This study sheds light on using bicycle training for improving locomotion. Bicycle training may be easier and more cost effective than locomotor training using body weight supported treadmill training.

Exercise and Apoptosis

There are several cellular components involved in apoptosis following SCI. Ascending and descending neuronal tracts are interrupted when the spinal cord undergoes trauma, but the extent of damage is dependent on the injury itself and degree of cellular mechanisms during the primary and secondary injury.[36,37] Apoptosis is carried out by various mechanisms within the cell and involve several anti-apoptotic and pro-apoptotic proteins and caspases.[37,38] Apoptosis imposes neuronal conduction, mostly through oligodendrocytes, reducing axonal regeneration in the central nervous system (CNS).[2,36] Such impacts on neuronal circuitry within the spinal cord limit recovery and functional gains following SCI.[2] In order to reduce the negative impact apoptosis has during the secondary injury of SCI, recent research has focused on the effects exercise might have on such cellular components in order to improve recovery post-SCI.


Caspases are cysteine proteases involved in promoting or executing apoptosis following SCI, depending on their structural components.[36,37] Caspases-3, -7, and -9 have each been strongly associated with apoptosis in SCI.[2,36] Refer to the SCI Cell Biology page for further explanation on how caspases promote apoptosis. Although there is little evidence on the effect exercise has on caspases in subjects with SCI, research does suggest exercise attenuates the effects of apoptosis in muscle and lymphocytes by acting on the caspase cascade.[39,40,41] However, one study be Keeler et. al.[2] observed the effects exercise has on gene expression, including caspases, in rats with SCI. Here, they studied the effects exercise has on mRNA and proteins in the whole spinal cord versus micro-dissections (intermediate gray matter and motoneurons) at two time points post-SCI (10 days and 31 days).[2] Gene expression in rats with SCI following exercise was then compared to gene expression in rats with SCI that had not undergone exercise, and studied the mRNA and protein levels of each cell component.[2] In this rat study, exercise consisted of passive cycling, twice a day for 30 minutes at 45 rpm and was performed 5 days per week.[2] In the whole cord, they observed an increase in caspase mRNA levels in rats with SCI, but only caspase-3 had a persistent increase at 31 days post-SCI.[2] They found that exercise actually decreased caspase-7 and -9 levels, but had no effect on caspase-3.[2] Interestingly, caspase-7 was the only of the caspase proteins that significantly increased post-SCI, and exercise attenuated this effect.[2] No significant differences were noted at 31 days post-SCI following exercise.[2] Micro-dissections noted little to no change in the caspases over either time interval, following SCI or exercise.[2]

Implications & Recommendations

This study suggests that caspases are primarily involved in the white matter due to their lack of recognition in the micro-dissections.[2] Increased caspase levels post-SCI can be attenuated with exercise, but primarily only effects caspases-7 and -9, indicating that apoptosis may be ongoing to some degree following SCI, in light of exercise, by the continued increase of caspase-3.[2] It appears that caspase-7 is the only caspase which has the ability to become fully translated in the spinal cord, because of its expression in both mRNA and protein, but that it can respond to exercise.[2] There is also evidence suggesting alterations in caspase levels may be expressed post-SCI and following exercise in a time-dependent manner, especially when began within the first week and continued daily, opening a window for clinicians to possibly intervene.[2] Further research needs to be done to studying the effects of apoptosis in gray matter[2] and carry over in human models.


Oligodendrocytes are one of the primary targets of apoptosis following SCI.[42,43] Their involvement in the formation of the myelin sheath makes them a pertinent component in nerve conduction.[42] When apoptosis imposes upon these glial cells, nerve conduction is compromised, resulting in a multitude of impairments following SCI.[42] In response to this, oligodendrocytes are becoming a popular aim in research with the intention of discovering ways to attenuate the damage to them after SCI.

A randomized control trial was performed by Krityakiarana et. al.[43] to study the benefits voluntary exercise might have on promoting neurogenesis in the spinal cord. This study focused on neurogenesis in rat models with an intact spinal cord, providing insight for future research.[43] Three groups were included in this study, a sedentary group, a group which performed exercise for 7 days, and a group which performed exercise for 14 days.[43] Rats included in the exercise interventions were placed in enriched environments with treadmills, providing them with the opportunity to run, running 3 km or more, daily.[43] After participating in the study, the rats were terminated and the T12 segment of the spinal cord was excised.[43] Attention was given to a marker, NG2, of oligodendrocyte progenitors, because of their ability to differentiate into oligodendrocytes.[43] They found that exercising 7 days increased the amount of NG2 protein (124%) in the spinal cord.[43] Although increases were noted in rats that participated in voluntary exercise for 14 days, these findings were not significant compared to the control group.[43] Another marker, nestin-GFP, of neural progenitor cells (NPC), was also studied; these cells have the ability to differentiate into other specific cells, including oligodendrocytes.[43] They observed a significantly increased amount (38%) of expressed nestin-GFP cells in the ependymal area after 14 days of voluntary exercise on the treadmill.[43] Rats that participated in exercise for 7 days demonstrated improvements in these cells; however this difference was not significant.[43] Finally, the researchers attempted to find differences of cell expression among the various areas of the spinal cord. Differences were noted, with greater expression of these markers within the white matter and ventral funiculus.[43] In summary, this study proved that voluntary exercise in rat models promotes neuronal generation in the intact spinal cord.[43] These findings also demonstrate that the neuronal generation is dependent on time, but further research is needed to determine more appropriate time frames.[43] Finally, greater expression of the progenitor and markers in the white matter and ventral funiculus might indicate that these locations are the primary sites for neurogenesis.[43] When considering neurogenesis after SCI, these results are limited because of the absence of subjects with SCI, being another indicator of the need for future research. Another limitation is that this study was performed on rat models, limiting clinicians' abilities to utilize this information in humans during clinical practice.

Exercise has been linked to promoting neurogenesis in rodents with SCI.[2,32,43] A receptor, TrkB, has been recognized for its various roles in transmitting messages within the central nervous system.[32] One study investigated the relationship between TrkB receptors and oligodendrocytes in healthy rats participating in a cycling exercise program.[32] They identified these receptors on oligodendroglial cells, and noted increases in the receptors following exercise.[32] Another study was looking at the effects exercise has on increasing TrkB expression within the injured spinal cord.[2] As previously described, the study by Keeler et. al. also observed the effects cycling has on TrkB in rats with SCI.[2] In this experiment, TrkB increased in response to SCI, but this increase was not maintained for long periods of time (>31 days).[2] TrkB did however respond to exercise by increasing its expression within the damaged spinal cord.[2] This response of increasing the TrkB receptor may be a mechanism to prevent neuronal damage and impairments, because increasing the receptor increases the likelihood of activating oligodendroglial cells, promoting neuronal regeneration.

Implications & Recommendations

Interestingly, TrkB receptors and oligodendrocyte progenitor cells have been identified in both gray and white matter and increase with exercise in rats with or without SCI.[2,32,43] When rats performed either locomotor training or cycling, gene expression of these receptors and progenitor cells increased.[2,32,43] They found that both mRNA and proteins of TrkB are increased with exercise, however this is only a transient effect.[2] As mentioned earlier, TrkB receptors are on oligodendroglial cells and the progenitor cells are precursors for mature oligodendrocytes.[2,43] In theory, an increase in these gene expressions should also increase the number of oligodendrocytes in the injured spinal cord. To go one step further, exercise may also induce an increase in these cells following SCI. Considering the evidence, initiating exercise early on (first week or two) following injury may reduce the degree of neuronal condution damage.[2,43] Once again, this is an area which further research is needed in order to make these distinct correlations, and to further address the fact that little is known about what type or intensity of exercise is needed to create such effects.

Heat Shock Proteins

Heat shock proteins (HSP) are proteins which protect against stress-induced apoptosis.[2,44] They are often referred to as “chaperones” because of their ability to care for damaged proteins across membranes.[2,44] In response to injury, HSPs not only protect, but also assist in the recovery process, and act against similar stresses cells may be exposed to in the future.[2,44] Because of their protective ability and role in recovery following injury, research has focused on HSPs to discover ways to further promote this process.[2,44] As described on the SCI Cell Biology page, apoptosis is induced following SCI and HSPs respond in order to inhibit apoptosis and prevent further damage during secondary injury.[44] Therefore, studies have been developed to look at the effects exercise has on HSPs following SCI.

In the study mentioned earlier, Keeler et. al.[2] also looked at the effects exercise has on HSP-27 and -70. Following SCI, they found that HSP-27 mRNA continues to increase through 31 days post-SCI, whereas HSP-70 mRNA only increases for a short period of time.[2] After rats with SCI participated in the cycling exercise, HSP-27 mRNA levels were reduced at 10 days post-SCI to match levels in rats without SCI in the whole cord; however there was no significant effect on HSP-70 levels at 31 days post-SCI.[2] When comparing protein levels of the two HSPs in the whole cord, it was evident that each slowly increased after injury, and that exercise promoted further increases in each that were sustained at 10 days post-SCI.[2] With exercise, HSP-27 remained increased at 31 days post-SCI, but HSP-70 reduced to post-SCI levels.[2] No changes were noted in HSP-27 or -70 mRNA or protein expressions in motoneurons or intermediate gray matter after SCI or with exercise.[2] The significant alterations in HSP expression when studying the whole cord and absent changes in micro-dissections indicate white matter damage as a result of SCI, attracting HSPs to a greater degree than within the gray matter.[2] Because of the ongoing insult carried out by apoptosis during the secondary injury, this indicates that stress within the white matter continues for a period of time, at least 31 days.[2] This study also indicates that passive cycling at this intensity (45 rpm) and duration (two 30 minute sessions, 5 days per week) induces HSP, promoting a protective mechanism to possibly facilitate in the recovery process of oligodendrocytes and neuronal conduction.[2] It is also apparent that there are time points at which these protective proteins may be more greatly impacted by exercise, and that further research needs to be done in order to provide a better idea of when exercise should be initiated, specify intensities and durations, and carry over into human models.

A second study[45] investigated the effects of exercise on gene expression in the lumbar spinal cord and in skeletal muscle in rats following SCI. Fifty-four female Sprague Dawley rats were involved in this longitudinal study.[45] Rats that did not have a SCI nor were involved in exercise were considered the control in this study.[45] All other rats involved in this study had a complete SCI at T10 imposed upon them and were divided into several treatment groups that underwent exercise at various time points and durations post-SCI (Table 3).[45] The exercise intervention consisted of cycling with bilateral lower extremities; one bout of exercise consisted of two, 30 minute sessions of cycling with a 10 minute rest break, at 45 rpm.[45] Rats were terminated 2-3 hours following their last determined exercise bout in order to extract a portion of the soleus muscle and of the lumbar spinal cord to assess alterations in gene expression.[45] Needle electromyography (EMG) was placed in the soleus muscle during cycling bouts in order to assess motor unit recruitment during exercise.[45] Please refer to Table 4 for results of each group.

Table 3: Descriptions of experimental groups[45]

Group Description
Control No injury, no exercise
tx2 2 days posttransection
tx5 5 days postransection
tx2ex1 2 days posttransection; 1 bout of exercise
tx5ex1 5 days posttransection; 1 bout of exercise
tx5ex4 5 days posttransection; 4 bouts of exercise (exercise started 2 days posttransection)
tx30 30 days posttransection
tx30ex20 30 days posttransection; 20 bouts of exercise (exercise started 5 days posttransection)
tx27ex20 27 days posttransection; 20 bouts of exercise (exercise started 2 days posttransection)

Table 4: Results of HSP-27 expression compared to control group[45]

Group Lumbar Spine HSP-27 Soleus Muscle HSP-27 Soleus Muscle Mass
tx2 Increased 6-fold Decreased No change
tx5 No change Decreased Decreased significantly
tx2ex1 No change compared to tx2 No change No change
tx5ex1 No change compared to tx5 Increased 6.6-fold No change
tx5ex4 Increased 2-fold Increased significantly -
tx30 - - Decreased by 46%
tx30ex20 - - Increased
tx27ex20 - - Increased

( - ) indicates data was not provided by the study.

Based on the results provided in Table 4, this study indicated the lumbar spine undergoes stress following SCI, inducing HSP-27 as a means to inhibit apoptosis.[45] HSP-27 levels within the muscle demonstrate beneficial effects of exercise in order to mediate muscle atrophy post-SCI.[45] Differences in HSP-27 expression in the spinal cord and soleus muscles early on may indicate prioritization of the protective mechanism within the spinal cord versus muscle atrophy prevention.[45] The fact that no long-term differences were evident in the soleus muscle mass between exercise groups suggests sustained increases in HSP-27 levels which may carry-over into functional recovery.[45] In short, the study provides evidence that exercise improves HSP-27 gene expression in the spinal cord after SCI, and attenuates muscles atrophy, all which may facilitate in the overall recovery process.[45]

Implications & Recommendations

As would be expected, the CNS reacts to an insult in various ways, one of which is to quickly respond to provide neuroprotection to reduce the negative impact induce by the injury. HSPs increase in the white matter as a means to protect oligodendrocytes.[45] Implementing a passive cycling treatment approach early on can increase the neural protection provided by HSPs, and continuing it for longer periods of time (~4 weeks) can reduce the degree of skeletal muscle atrophy.[45] Furthermore, although there are no exact recommendations, it appears that cycling at a low to moderate intensity early on may be most beneficial, and to be performed daily may create the greatest effects.[45]


MicroRNA (miRNA) are ribonucleic acids linked together to form a chain which degrades mRNA and other cells after translation, silencing the expression of the gene.[46,47,48] Specifically, miRNA help mediate the apoptotic pathways involving caspases, Bcl-2, phophatase and tensin homolog (PTEN), and programmed cell death 4 (PDCD4).[47,48] As described in Spinal Cord Injury – Cell Biology, several miRNAs have been discovered, and recent research has proven the involvement of specific miRNAs to act on the apoptotic cellular components mentioned previously.[46,47,48] As each specific miRNA has its own role in mediating apoptosis (Table 5), more recent research has sought to prove ways exercise can impact miRNA levels after SCI in order to reduce the effects of apoptosis.[48]

One study[48] was performed to specifically focus on the miRNAs that play a dominant role in regulating apoptosis. By equally dividing 30 Sprague-Dawley rats into five groups, including a control group (Table 6), they studied the alterations in miRNA post-SCI, and took it one step further by attempting to see how exercise might impact miRNA after SCI.[48] The exercise protocol consisted of passive cycling which was begun five days after the SCI was imposed upon the rats.[48] The exercise protocol used here was the same as that described previously in the Keeler et. al. study.[2,48] The study highlights the fact that miRNAs are altered post-SCI in ways which facilitate apoptosis. They found that following passive cycling, there were significant alterations in specific miRNAs to reduce apoptosis (Table 6).[48] In order to attenuate apoptosis after SCI, the miRNAs which degrade anti-apoptotic proteins, Bcl-2, were decreased, and those that act on caspases were increased (Table 6).[48] Based on the study, exercise acts to reduce the effects of apoptosis after SCI, and that these effects are altered in a time-dependent manner.[48]

Table 5: Roles of miRNAs in Healthy Models[48]

miRNA Role of miRNA
miR-Let-7a Pro-apoptosis: Acts on upstream cellular components
miR15b Pro-apoptosis: Inhibits Bcl-2
miR16 Pro-apoptosis: Inhibits Bcl-2
miR21 Anti-apoptosis: Inhibits PDCD4 and/or PTEN

PDCD4: Programmed Cell Death 4 - Pro-apoptosis; Activates caspases
PTEN: Phophatase and tensin homolog - Pro-apoptosis; Acts upstream of apoptosis by activating Bad

Table 6: Changes in miRNAs following transection alone or transection plus exercise[48]

Animal Group miR-Let-7a miR15b miR16 miR21
Tx 10 Days Increased significantly (11 fold) No change Increased significantly (2.5 fold) No change
Tx 31 Days No change from Tx 10 Days Decreased significantly No change from Tx 10 Days No change from Tx 10 Days
Tx+Ex 10 Days No change Decreased by >50% compared to control No change Increased significantly
Tx+Ex 31 Days No change No change No change No change

Bcl-2 Family Proteins

Pro-apoptotic and anti-apoptotic proteins play a significant role in regulating apoptotic pathways. Pro-apoptotic proteins (Bax, Bak, Bid, Bim) have the ability to alter the permeability of the mitochondrial membrane when activated, initiating a myriad of events, inducing apoptosis through the mitochondria.[49] Anti-apoptotic proteins (Bcl-2, BxL) are involved in an effort to reverse such effects by preventing pro-apoptotic proteins from binding to the mitochondrial membrane and inhibiting their downstream promotions of apoptosis.[49] There is an increased effort by the pro-apoptotic proteins to create further damage when imposed by an external threat, such as SCI, making these proteins targets in research to alleviate detrimental effects.[49,50,51] Interestingly, little is known about the benefits exercise has on these proteins following SCI. To date, most research is performed on animal models, to reduce apoptosis via these proteins in cardiac and skeletal muscle, as well as within the central nervous system in the brain.

As in SCI, traumatic brain injury (TBI) also induces apoptosis, but the response is found within the brain.[50] By creating a randomized control trial using Sprague-Dawley rats divided into four groups (control group, control exercise group, TBI group, TBI and exercise group), the short-term effects treadmill training has on apoptotic mechanisms were investigated.[50] Exercise training was performed for 10 consecutive days and consisted of treadmill training for 30 minutes a day at a relatively low intensity that was increased over the 30 minute span (2 m/min x 5 min, 5 m/min x 5 min, 8 m/min x 20 min).[50] It was apparent that Bax levels increased in the hippocampus in the TBI groups, but Bcl-2 decreased.[50] Exercise actually had the opposite effect on each of these, decreasing Bax and increasing Bcl-2 levels.[50] This indicates exercise does, in fact, have an attenuating effect on apoptosis following an insult to the CNS.

Aging also induces the natural occurring process of apoptosis, by interrupting the Bcl-2 family proteins as previously described.[51] Specifically within the myocardium, Kwak et. al.[51] looked at the effects exercise has on reducing the degree of apoptosis in aging rats. Two ages of Fischer rats were used (3 months and 24 months), and were then divided further into exercise and no exercise groups. Here, exercise consisted of a 12 week protocol of 60 min/day, 5 days/week of treadmill exercise at 75% VO2max.[51] Their study supported earlier reports that apoptosis does occur with aging in the heart muscle.[51] Similar to the study by Kim et al.[50], exercise in this population reduces apoptosis through the Bcl-2 family proteins.[51] Because these proteins act on the mitochondria in order to induce apoptosis, exercise reduces the mitochondrial-mediated pathway of apoptosis.[51]

Implications & Recommendations

Much research has provided evidence that exercise reduces apoptosis through mediating the effects of Bcl-2 family proteins in various conditions. This exemplifies the idea that exercise affects the mitochondria because of the ability of these proteins to alter the mitochondrial membrane[49], however little to no research has been identified directly linking the two. On the same note, there is minimal research demonstrating these effects in subjects with SCI. It might be expected that this attenuation will carry over into this population, in order to reduce apoptosis, preventing further damage to the spinal cord. Considering the research provided, implementing a form of daily aerobic exercise can help attenuate apoptosis-induced damage following SCI.[50,51]This is a an area that should be recognized for future research in order to provide evidence of such effects, recommendations for types and intensities of exercise, and carry over into human models.

Exercise and Excitotoxicity

Excitotoxicity after SCI can occur from increases in the neurotransmitter glutamate, increased activation of glutamate receptors, and associated ion imbalances, especially that of calcium (Ca2+).[36,52,53] Excitotoxicity and calcium-mediated secondary damage can be further understood by reading the SCI cell biology page. Extracellular levels of glutamate spike as quickly as 15 minutes after SCI, and can remain heightened throughout the first hour before attenuating to normal levels by 2 hours post-SCI.[54] Because of all the detrimental cascades excitotoxicity sets off, such as cell lysis, increased production of free radicals, and mitochondria-induced apoptosis, it seems that interventions seeking to decrease glutamate levels could potentially mediate some of the secondary damage after SCI. However, as suggested in the adaptive plasticity section above, glutamate may also play an important role in adaptive plasticity after the spinal cord injury by increasing the activation of receptors involved with plasticity pathways. Taken together, interventions must be able to decrease the initial harmful spikes in glutamate and then later maximize its therapeutic potential in the spinal cord. Unfortunately, research on exercise effects of glutamate in the SCI model is lacking. Table Number 7 below summarizes studies conducted on both healthy rats and rats with ischemic-reperfusion injuries of the middle cerebral artery.

Table 7: Exercise Effects on Glutamate

Study Subjects Interventions Outcome Measures Results
Jia et. al[55] 72 rats were randomly divided into 3 groups: Sham, Ischemic, and Ischemic Treadmill Pre-training Pre-training: Treadmill running at 20 m/min, 30 minutes per day, 5 days per week for 2 weeks. MCAO (Middle cerebral artery occlusion) was performed for the 2 ischemic groups after the 2 week pre-training/sedentary period. Glutamate, mGluR-1 (glutamate receptor in the brain) RNA expression, Neurological Deficit Score Glutamate levels spiked in both ischemic groups around 80 minutes post-injury as compared to sham controls; however exercise led to a significant decrease in this spike. mGluR-1 increased in both ischemic groups compared to sham control but exercise significantly lowered this increase; Rats that participated in treadmill training also had a lower neurologic deficits than sedentary controls.
Chang et. al[56] 64 rats were divided into 4 groups: Ischemic Control (IC), Ischemic Training (IT), Normal Control (NC), Normal Training (NT) Exercise: Treadmill training was initiated 24 hours after MCAO and the rats ran at a speed of 20 m/min for 30 min/day for 14 days. MCAO: Middle cerebral artery occlusion (ischemic-reperfusion injury) Extracellular glutamate, motor function Both groups that received training improved on motor function tests while the other groups did not. Extracellular glutamate levels significantly increased after exercise groups compared to sedentary controls and there was not a significant difference between exercising rats that did or did not have the ischemic injury. Overall, the glutamate levels in ischemic groups increased 5-20 times within the first few minutes to hours after the injury but returned to normal levels by 2 weeks post-injury (start of exercise or sedentary).
Zhang et. al[57] 36 rats divided into 3 groups: Sedentary Sham Group, Exercise with Operation, Sedentary Operation Exercise: Treadmill running at 25 m/min for 30 min/day, 6 days/week for 4 weeks. Sedentary: Remained in Cages for 4 weeks. Operation: After exercise or sedentary pre-training, rats in the operation groups underwent a middle cerebral artery ischemic-reperfusion injury. Assessed 80 minutes after injury: mRNA expression of mGluR5, NR2B (two glutamate receptors in the striatum) Exercise protocol prior to ischemic injury led to decreased mRNA expression of two glutamate receptors as compared to the sham control group and the non-exercised operation group.
Real et. al[58] 65 rats divided into 5 groups: sedentary (S), exercise training for 3 days (EX3), exercise training for 7 days (EX7), exercise training for 15 days (EX15), and exercise training for 30 days (EX30) Exercise: Treadmill running at 10 m/min for 40 minutes for either 3,7,15, or 30 days consecutively. Two subunits of the glutamate receptor AMPA (GluR1 and GluR2/3) in the motor cortex, cerebellum, and striatum Motor Cortex: Both receptors significantly increased after 30 days of training in the M1 region; GluR1 decreased in M2 at 15 days of exercise. GluR2/3 increased by 30 days in M2. Cerebellum: Significant increase Glu1 by day 30. Striatum: Both increased by day 30.

Implications & Recommendations

Two of the studies described used an ischemic-reperfusion injury to study how a pre-training program can alter the negative effects of glutamate excitotoxicity that occur after the injury.[56,58] Both studies found that a pre-training program incorporating treadmill running for about 30 minutes a day, for 5-6 days a week, for 2-4 weeks before the injury led to decreased expression of glutamate receptors or glutamate levels after injury.[56,58] Because most individuals do not necessarily expect to experience a SCI, these results can only be applied with caution and show that more research is needed on the effects of exercise after injury. The study by Chang et. al[56] did show that after an injury glutamate levels increase initially, but when exercise is initiated 24 hours after injury, beneficial increases were found in glutamate levels.[56] Overall, more SCI-specific research is needed to understand how to best play up the beneficial effects of glutamate while downplaying its excitotoxic effects.

Exercise and Oxidative Stress

In able-bodied individuals, high intensity aerobic exercise (between 75 and 100% of VO2 max) has been shown to increase oxidative stress, which is closely followed by an increase in endogenous antioxidants including superoxide dismutase and glutathione peroxidase.[59] A recent study on humans examined whether an 8 week exercise intervention could alter the oxidant/antioxidant balance. Functional electrical stimulation (FES) with a leg cycle ergometer was used for 30 minute bouts, 2 to 3 days per week.[59] This mode was used because it has been shown to induce physiologic responses similar to other forms of aerobic exercise.[59] The results yielded no significant changes in the blood levels of malondialdehyde (free radical end-product of lipid peroxidation), superoxide dismutase, or glutathione peroxidase (two endogenous antioxidants).[59] The lack of expected changes may have been due to an inadequate exercise intensity, the mode of exercise used (FES versus an upper body ergometer), or from inadequate measures of antioxidants (the study measured them indirectly through the blood).[59]

In the brain and cerebellum, studies have found exposure to low levels of oxidative stress through regular exercise can actually be beneficial.[60] This consistent exposure to moderate levels of stress allows the body's natural redox/antioxidant processes to adapt, ultimately lowering the levels of free radicals (FR) and reactive oxygen species (ROS).[60] Neurotrophins such as BDNF and GDNF, which help promote neuronal survival, can also be altered through the exercise-induced adaptations in the antioxidant balance. A study by Siamilis et. al[60] looked at the effects of exercise and oxidant/antioxidant injections on the levels of free radicals, GDNF, and BDNF in the rat spinal cord.[60] This study used 36 male rats split into 6 groups: sedentary control with saline injection (NEC), sedentary with H2O2 injection, sedentary with N-tert butyl-α-phenyl nitrone (PBN is an antioxidant) injection, exercised control with saline injection (EC), exercised with H2O2 injection, and exercised with PBN injection.[60] All rats in the exercise groups participated in treadmill running for 1 hour per day, 5 days a week, over a 10 week intervention period. Following the intervention, all rats were decapitated so the cervical spinal cords could be analyzed.[60] Ultimately, the results of the study indicated free radical levels in the spinal cord decreased with exercise training.[60] There was no additional decrease in free radicals with an increased oxidative challenge from H2O2 injections.[60] PBN (the antioxidant) injections alone decreased free radical levels compared to the other sedentary groups, indicating a potential therapeutic target for individuals with SCI who are unable to complete a moderate intensity exercise program.[60] This study found that exercise combined with antioxidant injections did not change the level of free radical reductions, suggesting that either exercise or antioxidant therapy are sufficient to cause the same level of positive change.[60] Additionally, the researchers found decreased BDNF levels with exercise, which may be due to the fact that one of the transcription factors for BDNF is sensitive to redox, meaning that decreased free radical levels leads to decreased BDNF expression.[60] While the study did not find an increase in GDNF levels with exercise, it did show an increased expression in rats injected with H2O2, supporting the neuroprotective role of GDNF in the spinal cord.[60] Overall, this study showed that regular exercise and/or antioxidant therapy can both serve to decrease free radical concentrations in the spinal cord region.[60] Additionally, there is a link between free radical levels and BDNF expression in the spinal cord, which suggests that regular exercise could promote better BDNF expression and thus better neuronal survival and regeneration.[60]

Studies shows specific links between exercise, oxidative stress, and antioxidant productions are lacking the SCI model. Some available research on these links in SCI models as well as healthy models are listed below in Table 8.

Table 8: Studies examining links between exercise, oxidative stress, and antioxidant production

Study Subjects Intervention Outcomes Results
Park et. al[62] 24 male rats randomly divided into 4 groups: Control, SCI, SCI + Ex, SCI + Ex + MT; a weight-drop contusion model of SCI was used Exercise: Treadmill stepping 2 times a day for 15 minutes, 6 days per week, for 4 weeks. Melatonin: MT injection twice a day for 4 weeks (dose 10 mg/kg) Locomotor recovery as measured by the BBB Scale, iNOS expression (free radical marker), markers of stress and apoptosis Rats in the SCI + Ex + MT group had significantly decreased iNOS and significantly improved motor function compared to the SCI + Ex group; the functional changes were seen within 7 days after SCI.
Revan et. al[63] 37 healthy males Treadmill running at 100% VO2 maximum until exhaustion; protocol was stopped when age predicted heart rate max was reached or subject initiated the stop. LOOH, GPx, CAT, Lactate dehydrogenase (marker of muscle damage), Time to Exhaustion No significant changes in markers of oxidative damage or antioxidant levels
Ogonovszky et. al[64] 28 rats randomly assigned to 4 groups: Control, Moderate Training (MT), Strenuous Training (ST), or Over Training (OT) MT: 1 hour swimming per day, 5 days per week for 8 weeks; ST: 1 hour swimming per day with an additional 30 minutes added each week up to 4 ½ hours by the 8th week, 5 days per week for 8 weeks; OT: 1 hour swimming per day for 6 weeks and 4 ½ hours per day for last 2 weeks, 5 days per week Markers of oxidative damage to nuclear DNA, marker of oxidative protein damage, lipid peroxidation level No changes in oxidative damage in any group; markers of oxidative protein damage decreased in all groups with a significant decrease in ST/OT, no change in lipid peroxidation

Implications & Recommendations

The SCI model studied by Park et. al[62] found positive on the BBB Scale after 4 weeks of treadmill stepping, and these were enhanced with the addition of an exogenous antioxidant supplement (melatonin).[62] The study by Revan et. al[63] found that acute, exhaustive exercise is not beneficial to the antioxidant system but also did not produce excessive oxidant levels.[63] Similarly, Ogonovszky et. al[64] did not find any increased oxidative damage from either a moderate, strenuous, or over-training program in rats.[64] These studies, combined with the ones described above, suggest that a variety of exercise intensities (acute exhaustive, moderate, strenous, over-training) are not sufficient to produce negative changes in the oxidant/antioxidant balance. However, because only the study by Park et. al used a spinal cord model, clearly more research is needed in this area. Additionally, there was conflicting results on whether or not the administration of exogenous antioxidants could enhance the exercise-induced changes in oxidant/antioxidant balance. More research should determine appropriate dosage and timing of administering antioxidants, and should also look specifically at the spinal cord injury population.[59,60,62,63,64]

Exercise and Hormones


Physical stress disrupts homeostasis, leading the body to initialize mechanisms to control and minimize this deviation. One of the control pathways the body utilizes is releasing hormones such as catecholamines and cortisol. Both of these hormones have been found to increase with the introduction stress in the form of exercise.[65,66,67] Exercise at levels that disrupt homeostasis leads to immune mediated inflammatory reactions that release the pro-inflammatory cytokines TNF-α, IL-1 and 6.[68] These cytokines have been found to stimulate corticotrophin releasing hormine (CRH), thus increasing circulating cortisol levels via ACTH.[68] Increased levels of cortisol with exercise has been found to have widespread effects on the body such as stimulating gluconeogenesis, muscle protein degradation, inhibit synthesis of the actin/myosin heavy chains, and immuno-suppression.[65,68]

Evidence in the literature demonstrates cortisol levels in the blood are dependent on both duration and intensity.[69] In order for a significant increase in cortisol levels to occur, the exercise parameters need to be higher intensity over a long duration.[69] Duclos conducted a study of 8 subjects (4 sedentary individuals, and 4 endurance athletes) to examine the impact of exercise at different intensities and durations on circulating cortisol levels.[69] No significant increase in cortisol occurred for low intensity (50% HR max) exercises for brief (20 min.) or prolonged (120 min.) time periods, or high intensity (80% HR max) and low duration for either group.[69] Significant increases in cortisol only occurred for both groups after participating in intense exercise at prolonged duration.[69] Results of this study are in accordance with other research demonstrating that ~60% max HR needs to be achieved before significant amount of cortisol are released into circulation, at which point a linear correlation exists between HR and circulating cortisol levels.[70,71] The highest circulating levels of cortisol have been observed with high volume and intensity resistance training that produces the greatest amount of lactate and creatine kinase.[65]

Interestingly, it has been found that men that participate in routine endurance training have roughly the same resting cortisol levels of sedentary men. [69,72] Also, when subjected to an intense bout of exercise with a prolonged duration, both sedentary and endurance trained individuals had similar increases in cortisol level.[69] However, repeated exposure to high intensity endurance exercise decreases the sensitivity of monocytes to cortisol.[72] This was found in the case of healthy men, yet it remains reasonable to believe that patients with SCI should still have some benefit related to this adaption of repetitive exercise and decreased sensitivity to immuno-suppression from cortisol by monocytes.


Postural/Exercise Impact And Age

Catecholamine serum concentration is largely impacted by physical activity, and has been widely studied in the literature.[67,73] It has been found that even small activity such as a simple change in posture from supine to upright will increase the release of catecholamines.[74,75,66]

Kohrt studied the impact postural changes and exercise has on circulating plasma catecholamine between groups of 106 healthy older (64 +/- 1yr) and 24 younger (24+/- 1yr) subjects following a 9 month exercise program.[66] At baseline, it was found that significant increases in plasma catecholamine concentrations, specifically norepinephrine, occur with postural changes from supine to standing in both age groups.[66] Also, during a 15 minute exercise session at about 78% VO2 max, significant increases of epinephrine and nor epinephrine occurred.[66] While increases were observed in both groups, they were not equal for both activities. Older subjects had a higher increase in the release of catecholamines during supine to standing (P<0.05), and during exercise, concentrations were more pronounced in the younger subjects (P<0.01).[66] Following the 9 month exercise program, there was not a significant difference between resting and supine to standing catecholamine concentration, however they were lower during exercise and five minutes post-exercise.[66] Kohrt found a correlation between increases in heart rate and catecholamine concentration in response to exercise. It was observed that higher heart rates were accompanied by higher concentrations of circulatory catecholamines.[66]

Lesion Level

Circulating concentration of epinephrine and norepinephrine is further affected by spinal cord injury and where it is sustained.[74,75] As previously mentioned, cell bodies of pre-ganglionic sympathetic neurons are found between T3 and L3, with the primary innervations derived between T5 and T9.[74] Therefore, depending on where the spinal cord injury is, brain centers such as the hypothalamus may have reduced ability to regulate the sympathetic nervous system. Different levels of spinal cord injury have been found to significantly alter the body’s normal response to release catecholamines into circulation during periods of rest and physical activity.[74,75]

Schmid conducted a study of 75 subjects with chronic complete spinal cord injuries, separating them into groups depending on lesion level. Groups consisted of 30 tetraplegics and 15 paraplegics in each group of T1-T4, T5-T10, and below T10. Plasma concentrations of catecholamines were measured at rest, exercise at 60% VO2 max, and max O2 consumption and compared to each other and a group of 16 non injured controls.[74] Tetraplegics were found to have a slight increase in epinephrine and norepinephrine at maximal exercise, however concentrations remained lower at rest, sub max and max exercise than all other groups (P<0.01).[74] Norepinephrine increased in all paraplegic groups and the control at both sub max and max exercise, while increased epinephrine was only observed at maximum exercise.[74] The high thoracic group (T1-T4) showed similar norepinephrine levels to the control group during all three examined conditions, however epinephrine levels remained lower throughout (P<0.05).[74] The mid and lower thoracic groups catecholamine concentrations remained similar to each other throughout the study, however their epinephrine and norepinephrine concentrations were significantly higher at rest and after max exercise that all other groups including the control (P<0.01).[74] These results are supported by Steinberg in a study demonstrating subjects with chronic spinal cord injuries of the thoracic spine have significantly increased norepinephrine levels with maximal exercise; however lesions below T6 had higher resting and exercise values.[75] Resting levels of epinephrine, though not significant, appear higher in subjects with lesions below T6, and during exercise only the low thoracic group had a significant increase in epinephrine.[75]


Research has demonstrated that exercise intensity and duration have a significant influence on the body’s response to excrete catecholamines during exercise. As mentioned previously, a correlation exits between heart rate and catecholamines.[66] Significantly higher serum catecholamine concentrations are observed as exercise intensity increases.[76] Norepinephrine has significant increases with all the progressions of intensity from minimal to moderate then max, while epinephrine typically has high increases at higher intensities such as moderate to max.[66,77,78] Throughout the duration of exercise, norepinephrine has been shown to continually increase until complete exhaustion exhaustion.[67] Also, participation in multiple exercise sessions per day can lead to higher blood catecholamine concentrations.[79]


There are many physiologic differences between men and women that may lead one to expect differences in the body’s endocrine response to exercise. Men generally have increased muscle mass, and women tend to have higher body fat percentages. Additionally, there are very specific differences in circulating hormones such as increased testosterone in men and increased estrogen/estrodiol in pre-menopausal women. Though these differences exist, evidence in the literature is widely contradictory.[67] There is conflicting evidence for plasma catecholamine levels between gender at rest, during moderate and intense exercise.[67] At this point, no clear conclusion can be made regarding the differences in the body’s catecholamine response to exercise between gender.

Mental Stress

Mental stress in combination with physical stress has some evidence of increasing levels of catecholamines, specifically norepinephrine.[80] This relationship is important to consider when applying exercise interventions to patients following spinal cord injuries. Increased mental stress levels from both coping and rehabilitation may lead to increased levels of circulating catecholamines. Webb demonstrated increased circulating norepinephrine in 8 healthy male subjects who were simultaneously cycling at 60% max VO2 , and completing a mental challenge as compared to cycling alone.[80] There was no significant change in epinephrine levels between the two testing conditions.[80]

Overcoming the Glial scar

In humans, the mature Central Nervous System (CNS) is restricted in its ability to form new connections or to reorganize its current connections following injury when compared to an immature CNS. This is due to the mature CNS being outside the time frame known as the “critical period.”[81] Typically the critical period of the CNS occurs early in life most notably in newborns. This has been previously demonstrated in studies comparing functional recovery following SCI between immature and mature animals.[82]

Immediately following adult Spinal Cord Injuries (SCI), glial cells collect around the injury site to incase the lesion. During this process specific molecules that are contained within the CNS’s extracellular matrix (ECM) are unregulated by astrocytes and the glial scar is formed. The purpose of the scar is to limit damage to the CNS and to protect the Blood Brain Barrier (BBB) that was breached during the injury. However, due to the protective nature of the glial scare many of the molecules that are up-regulated to promote its formation have been shown to inhibit axonal growth or reorganization over and across the scar site. Because of the limited ability to grow across this site, functional recovery of the involved tracts is limited. One of the main inhibitory molecules released by the astrocytes during glial scar formation is Chondroitin sulfate proteoglycans (CSPGs). As mentioned in the cell biology portion, CSPGs inhibit new growth by binding to growth promoting lamanin and preventing it from interacting with its receptor in the growth cone integrin. Recent studies have shown that CSPGs have also been found to participate in the formation of perineuronal nets (PNNs). PNNs are most often formed towards the end of the critical period for neurogenic changes. Thus, with the formation of PNNs we see limited regrowth in the CNS. Therefore, it is vital that rehabilitation takes place in the time period before the formation of PNNs in order to maximize the functional gains a patient with SCI could obtain. In order to combat the presence of CSPGs and the formation of PNNs in the CNS, research is being done with the use of chondroitinase ABC (ChABC) to decrease CSPGs effect on axonal growth. There is now evidence significant evidence showing that ChABC therapy reduces the effects of CSPGs in SCI and also promotes significant axonal repair.[83] Also, more evidence is immerging showing that disturbing the action of CSPGs with ChABC therapy encourages collateral outgrowth of new axons from uninjured axons.

In order to take advantage of the capabilities of ChABC therapy and its’ effect on regeneration and new growth of axons at the lesion site, it will be important to incorporate the correct exercise for patients with SCI. This includes the type and timing of that exercise. Improved UE function including hand usage has been found to be the most desirable function to improve quality of life in tetraplegia patients.[84] Therefore, a study was conducted testing the ability of corticospinal tract to regenerate with the use of ChABC therapy and task specific exercises in rodents and its relation to functional gains. The study found that although ChABC therapy promoted new growth and regeneration by itself, when not coupled with task specific training the rodents only had only a moderate recovery of function. However, when ChABC therapy was combined with task specific training, the rodents showed significant improvements in that specific task.[85] However, the gain was not transferred into other tasks. For example, the group that only focused on paw movements following SCI did not have any carry over into the locomotor outcome and the opposite was true for the locomotor group. This shows us that in order to maximize a functional gain in patients with SCI injury, they must repeatedly practice that task which they want to recover. In the previous study the animals practice their individual task 1 hour/day every day for 6 weeks. The ChABC treatment was starting on day 0 post injury andc continued for 10 days. Tasks specific rehabilitation began on day 7. The authors of the study concluded that the timing of their treatment would be analogous to begin ChABC treatment in humans 14-28 days in order to coincide with the beginning of rehabilitation.[85]

Based on the current evidence it seems that the most appropriate exercise prescription for axonal re-growth and repair across the glial scar formation leading to functional gains would include repetitive task specific training in combination with ChABC therapy at a minimum of 1 hour/day for at least 6 weeks starting 2-4 weeks after the initial injury.

Other Considerations

Bone Density and Osteoporosis

As alluded to on the SCI Cell Bio WIKI, individuals with SCI are both acutely and chronically at risk for changes in bone structure, bone mineral density, osteoporosis, and fractures.[86,87,88,89,90,91] A vast majority of the bone changes can be found at the distal femur or proximal portion of the tibia.[89,90] Preventing BMD loss and preventing fractures is crucial because fractures can make an individual with SCI more susceptible to complications such as prolonged immobilization (which could hinder any current progress from therapeutic interventions), heterotopic ossification, pressure sores, and difficulty surgically fixing the fracture because of the increased porous composition of the bone itself.[89,90] Thus, by better understanding the pathology behind bone loss and optimizing treatments to increase bone density, physical therapists can help prevent these devastating and costly complications.

Bone Changes from Exercise in Healthy Models

Exercise has been shown to have positive effects on both the prevention and treatment of decreased bone mineral density (BMD).[92,93] Research has shown that both in humans and animal models continuous exercise can promote the bone resorption process needed to maintain bone health, and that a lesser intensity of exercise may be adequate to maintain initial gains.[93] In human models, some controversy exists on whether endurance-type exercises promotes or hinders improving bone mineral density and the answer may lie in what type and frequency of exercise is being done.[92,93]

A study using healthy male competitive cyclists found that after 9 months of training, BMD loss occurred in multiple areas of the lower extremities including the lumbar spine, hip, femoral neck, greater trochanter, and femoral shaft.[92] After a 3 month recovery period some athletes showed attenuation of these effects, but the majority still has lower BMD than when they began a year prior.[92] The change in calcium homeostasis in the bones was not related to loss of calcium through the sweat and occurred despite the athletes taking either a high or low dose calcium supplement each day.[92]

A study by Ooi et. al[93] looked at the level of exercise needed to achieve significant bone changes in healthy rats, and then determined a minimum intensity required to maintain these gains throughout the rest of the exercise intervention.[93] The study showed rats involved in the 8 week exercise program, regardless of number of jumps per week, still had improved bone mass, bone strength, and bone perimeters than sedentary controls.[93] The best improvements were found with rats completing 200 jumps per week and these gains were able to be maintained with between 10% (21 jumps per week) and 18% (36 jumps per week) of their original workout intensity.[93] One additional finding in this study was that better bone health maintenance was found when rats were jumping for 3 days per week as opposed to only 1 day.[93]

Bone Changes from Exercise in SCI Models

While these are just two studies examining the effects of different exercise programs in healthy humans and rats, the results show the need to study the effects of exercise and bone health in persons with SCI. The human study used cycling, and whether passive or combined with functional electrical stimulation, cycling is often used as an intervention in both human and animal models for SCI interventions.[86,91] The potential negative effects of long-term non-weightbearing activities such as cycling should therefore be considered when treating a patient with SCI, who is already at risk for decreased BMD. On the other hand, the weight-bearing jumping activity studied in rats was associated with positive gains in BMD, so activities such as treadmill training may be more likely to produce positive changes in BMD for individuals with SCI.[87,88,93]

Quality studies, including randomized controlled trials, on the effects of exercise and bone health in individuals with SCI are definitely lacking. Additionally, there is the need to understand how any changes can be additionally impacted by level of lesion, and whether the SCI is acute or chronic. When studying bone mineral density in human models, a dual x-ray absorptiometer (DXA) is often used.[86,87,88,89,90,91] Markers of bone health that are frequently found in the literature include serum osteocalcin (OC) which is a marker of bone formation, urine deoxypyridinoline (DPD) which is a marker of bone resorption or breakdown, and also bone-alkaline phosphatase (B-ALP) which is another marker of bone formation.[86,87,88,89,90,91] In general, the studies want to see if bone resorption rate is different from bone formation rate..[86,87,88,89,90,91] If bone is broken down faster than it can be built, bone loss will occur.[86,87,88,89,90,91] Table 9 below summarizes some of the studies examining the effects of exercise on bone health in individuals with SCI.

Table 9: Impact of Exercise on Bone Health in Persons with SCI

Citation Subjects Intervention Bone Markers Overall Findings
Bloomfield et. al[86] 9 subjects(5 men, 4 women) with chronic SCI (average was 6 years post-SCI); Injury levels ranged from C5-T7; 8 individuals (5 male, 3 female) with chronic SCI served sedentary as controls. Exercise: Started with FES-induced quadricep extensions; 45 extensions per session, 3 days per week, until 4 ½ kg could be lifted then cycle protocol was initiated (average 6 weeks). FES cycle ergometry was done at a rate of 50 rpm for 30 minutes per session, 3 times per week for a total of 80 sessions (around 9 months total). Control: Testing was done at baseline, 3, and 6 months. BMD at the lumbar spine, femoral neck,distal femur, and proximal tibia were measured using DXA; Serum OC as a marker of bone formation BMD: Both groups had significantly decreased BMD at distal and proximal femur at baseline compared to able-bodied individuals; After 9 months of training, exercise group showed significant improvements only at lumbar spine. It is interesting to note that subjects with paraplegia had a greater increase in BMD (18%) at the distal femur than tetraplegic and they did not experience a decrease in proximal tibia BMD as seen with the tetraplegics. OC: Increased in exercised group by 78% in first 6 months and was maintained at 9 months
Giangregorio et. al[87] 5 subjects with acute cervical SCI (2-6 months post-SCI) BWSTT two days per week for 48 sessions. Serum OC, Urine DPD, BMD lumbar spine, proximal femur, distal femur, proximal tibia, functional gait changes Decreased BMD in all areas with no significant trends for bone markers.
Giangregorio et. al[88] 14 subjects (11 males, 3 females) with chronic SCI (at least 12 months post-SCI) with varying levels of incomplete motor lesions (range C4-T12). Also included 4 individuals with SCI who did not participate in BWSTT to serve as controls. BWSTT, three days per week for a total of 144 sessions over a year. Serum OC, Urine DPD, BMD lumbar spine, proximal femur, distal femur, proximal tibia, functional gait changes. No significant changes in BMD for any site and no significant changes for markers of bone formation/resoprtion.
Mohr et. al[91] 10 subjects (8 men, 2 women) with chronic SCI (average 12 years post-SCI) and 5 able-bodied individuals served as controls. 6 subjects were tetraplegic (C6) and 4 were paraplegic (T4) and all were motor complete. FES with leg cycle ergometer for 30 minutes a day, 3 days per week for 12 months; Decreased to 1 day per week for an additional 6 months. Serum OC, Urine DPD, BMD lumbar spine, femoral neck, proximal tibia Overall BMD was significantly lower than able-bodied controls. BMD significantly increased in proximal tibia after 12 months but not spine or femoral neck. After 6 months with decreased frequency, BMD of proximal tibia decreased back toward baseline levels. No significant change in OC or DPD.

Implications & Recommendations

Taken together, the studies above have examined the effects on a variety of exercises including BWSTT and FES cycle ergometer on bone mineral density and other markers of bone formation or resoprtion in individuals with both acute and chronic SCI.[86,87,88,91] It appears that acute exercise has little effect on bone density, which has been shown to decrease for the first 6-12 months following SCI.[87] In more chronic SCI, FES cycle ergometer did seem to have a small, yet positive, effect on BMD as compared to control individuals with SCI who were not completing exercise.[86,91] Overall, BMD values in the SCI population still remain lower than able-bodied controls despite the use of exercise as an intervention.[91] One interesting finding from Giangregorio and colleagues[87] was that the individual in their study who showed the smallest decrease in BMD showed the greatest improvement in functional locomotion.[87] Conversely, the subject who participated in the fewest BWSTT sessions had the greatest overall decrease in BMD.[87] It seems that perhaps prolonged exercise is needed to elicit any changes in BMD and that this needs to be initiated chronically after SCI (perhaps after the year). FES cycle ergometry proves more effective as shown above, but with the locomotion benefits from BWSTT, further studies should be done to find positive benefits on BMD.[86,87,88,90,91]

PGC-1α and Exercise in Healthy Individuals

Acute and chronic endurance exercise has been shown in both human and animal models to increase skeletal muscle levels of peroxisome proliferator-activated receptor γ co-activator 1 α(PGC-1α), which is a known regulator of mitochondrial biogenesis.[94,95,96] This positive increased expression of PGC-1α with exercise leads to improved mitochondrial biogenesis and eventually to improved oxidative function.[94] PGC-1α-mediated improvement in mitochondrial biogenesis has also been linked to prevention of aging and death in an animal study that looked at wheel running in mice.[95] Mice with dysfunctional mitochondria were used and subjected to either a sedentary control group or to an endurance exercise group.[95] The exercise mice ran on a wheel for 45 minutes per day, at a rate of 15 m/min, 3 times a week for 5 months.[95] At the end of this randomized control trial, mice completing endurance exercise training had improved endurance capacity, decreased aging expressions (gray hair, balding), and a prevention of mortality.[95]

Resistance exercise has been shown to cause muscular adaptations through a pathway involving target of rapamycin (mTOR) signaling that can lead to the formation of proteins and muscle.[96] There has been some evidence that this mTOR pathway may have an impact on certain aspects of the mitochondrial biogenesis cascade, so it is possible these two paths are linked.[96] A human study showed that a 10 week resistance training program can activate AMP-activated protein kinase (AMPK; kinase involved in exercise-mediated PGC-1α expression and also skeletal muscle adaptations to exercise) but did not have an effect on the mRNA levels of PGC-1α or citrate synthase activity (which is a measure of oxidative capacity).[96]

A study by Wang et. al[96] looked at whether or not resistance exercise in combination with endurance training could also produce beneficial effects on PGC-1α and mitochondria in the skeletal muscle fibers of the quadriceps femoris.[96] The study used ten healthy individuals (7 men, 3 women) and had them each perform 2 different exercise interventions, with a 2 week break in between.[96] Subjects were randomly assigned to the order in which they completed either the endurance only intervention (E) or the endurance plus resistance exercise program (ER). The endurance component consisted of 1 hour of cycling at a moderate intensity of 65% of VO2 max, followed by a 2 hour rest period.[96] In the ER phase, subjects were then given a 15 minute break before they began the resistance training program.[96] This study used the leg press and performed 6 sets of 15 repetitions (at varying percentages of their 1 repetition maximum load); this was followed by a 1 hour-20 minute rest period.[96] Muscle biopsies and blood samples were taken at the start of the program, after the first hour of endurance exercise, after the second hour (either rest or at the end of ER), and after the third hour.[96] The researchers were looking at mRNA levels of various genes and proteins involved with mitochondrial biogenesis, but the ones of main interest were PCG-1α, PGC-1-related coactivator (PRC; similar role as PGC-1α), pyruvate dehydrogenase kinase 4 (PDK4) which is an enzyme that can regulate carbohydrate oxidation and lipid peroxidation, and in general the various markers of the mTOR pathway.[96] Overall the study found that resistance training performed after endurance exercise can increase the expression of genes that are part of the signaling pathway that ultimately leads to mitochondrial biogenesis, improved oxidative capacity, and increased activation of proteins involved with the mTOR pathway of protein synthesis.[96] Specifically, the mRNA content of PGC-1α, PRC, and PDK4 were around 2 times greater in the resistance and endurance training program as compared to the endurance alone.[96] This gives further proof of not only the benefits of endurance exercise on mitochondria, but also brings to light the benefit of combining endurance and resistance exercise.

Implications & Recommendations

The extensive mitchondrial impact following SCI is described in detail on the SCI cell biology page. While it is true that the majority of studies look at PGC-1α-mediated mitochondrial biogenesis in skeletal muscle, the results of these studies indicate the need for further research into the effects of PGC-1α and SCI, as well as the effects of exercise and PGC-1α-induced mitochondrial biogenesis in the spinal cord itself. Many subjects with SCI will be participating in various forms of endurance and/or resistance training throughout their rehabilitation so some of the results above can be extrapolated to provide recommendations. Using exercise to optimally promote the induction of PGC-1α could lead to improved mitochondrial efficiency and oxidative function in the remaining mitochondria after the injury. With such profound effects being shown in human and animal models, studies need to determine how PGC-1α can promote mitochondrial biogenesis in individuals with pathologies, such as SCI, and what that optimal dose may be.

Overall Recommendations

Many cellular processes are involved in the secondary pathology that follows SCI. As shown throughout this page, exercise has the potential to impact many of these cellular factors. Table 10 below is an overall summary of the exercise recommendations following SCI based off of available research presented throughout this page.

Table 10: Overall Exercise Recommendations After SCI

Time Frame Type of Exercise Intensity Duration Cellular Rationale
Day 1 cell-content cell-content cell-content cell-content
Day 2-7 Treadmill step-training, passive-cycle training, or swimming Need further research 15-30 minutes, 5 days per week, for at least a month for treadmill step-training and passive-cycle training and 2 sessions of 4-minute swims on day 3 after injury (in rat models) When initiated 5 days post-injury, both treadmill step-training and passive cycle training have led to increased expression of GDNF, NT-3, NT-4, and reduces caspases-7 and -9 in the spinal cord.[2,14]; When initiated 5 days post-SCI, daily passive cycling increases HSP expression in the spinal cord, and if continued for approximately one month, increases HSP levels within skeletal muscle.[45] Despite these positive findings it may not be wise to initiate exercise at this time because exacerbation of inflammation at the injury site has been found with swimming 3 days post injury.[13]
Weeks 1-3 Voluntary treadmill training Need further research; however acute-exhaustive, moderate, strenuous, and overtraining has not been shown to significantly produce harmful levels of oxidative stress in models studied. ~3 km daily at self-selected pace (in rat models) 2-4 weeks of duration has shown positive effects on glutamate levels after ischemic-reperfusion injuries Increases precursor cells for oligodendrocytes which may provide a degree of neuroprotection following SCI.[43] Additional benefits may be seen with changes to oxidative stress, however exogenous antioxidant therapy should be studied further and may enhance effects.[59,60,62,63,64] Glutamate and exercise has not been studied in the SCI population, but work in healthy and ischemic-reperfusion injuries has showed that exercise can decrease glutamate expression after injury when the environment is already toxic.[56,58]
Chronic (1 to many years post-SCI) FES cycle ergometer, Treadmill training and/or resistance training. Also seated handcycling. Moderate intensity Severals weeks to months In patients with chronic SCI, decreases in BMD begin to plateau around 9-12 months post-SCI and studies have found small, yet positive changes in BMD in chronic SCI.[91] While studies examining the effects of PGC-1α in SCI models is lacking, chronic endurance exercise, and also resistance training coupled with endurance exercise, has been shown to promote mitochondria biogenesis, improved oxidative capacity, and even helped prevent aging in healthy animal and human models.[94,95,96]

Overall Need for Future Research

There has been a common theme in the research when studying the effects exercise has on the cellular mechanisms following SCI. Although current evidence demonstrates alterations in gene expression in animal models with SCI, there are gaps within the research which may help to attenuate the damage caused during secondary injury and facilitate recovery after SCI. The apoptotic pathways have been studied, but there is a lack of animal and human models with SCI when studying particular cellular components. There are also several cellular components which need to be studied further to have a better picture of when and where within the the spinal cord exercise might reduce apoptosis or promote neuronal protection. Even though it would be a huge task, it would be very interesting to see how various endongenously produced growth factors respond to exercise in human populations, and then do long term studies to see how this impacted real world function of the individual. This would blend the cellular biology research with the more rehabilitation/functional based studies. This synthesis would most likely lead to some very profound conclusions about how we can better treat individuals with SCI.

Go To SCI Cell Biology Page

Return To Welcome Page

1. Dunlop SA. Activity-dependent plasticity: implications for recovery after spinal cord injury. TINS. 2008;31(8):410-418.
2. Keeler BE, Liu G, Siegfried RN, et. al. Acute and prolonged hindlimb exercise elicits different gene expression in motoneurons than sensory neurons after spinal cord injury. Brain Res. 2012;1438:8-21.
3. Astorino TA, Tyerman N, Wong K, Harness E. Efficacy of a new rehabilitation device for individuals with spinal cord injury. J. Spinal Cord Med. 2008;31:586-591.
4. Phadke CP, Flynn SM, Thompson FJ, Behrman AL, Trimble MH, Kukulka CG. Comparison of single bout effects of bicycle training versus locomotor training on paired reflex depression in soleus H-reflex after motor incomplete spinal cord injury. Arch Phys Med Rehabil. 2009;90:1218-1228.
5. Dobkin B, Barbeau H, Deforge D, Ditunno J, Elashoff R, Apple D, Basso M, Behrman A, Herkema S, Saulino M, Scott M. The evolution of walking-related outcomes over the first 12 weeks of rehabilitation for incomplete spinal cord injury: the multicenter randomized spinal cord locomotor trial. Neurorehabil Neural Repair. 2007;21(1):25-35.
6. Lynskey JV, Belanger A, Jung R. Activity-dependent plasticity in spinal cord injury. J Rehabil Res Dev. 2008;45(2):229-240.
7. Edgerton VR, Roy RR. Activity-dependent plasticity of spinal locomotion: implications for sensory processing. Exerc Sport Sci Rev. 2009;37(4):171-178.
8. Grau JW, Crown ED, Ferguson AR, Washburn SN, Hook MA, Miranda RC. Instrumental learning within the spinal cord: underlying mechanisms and implications for recovery after injury. Behav Cogn Neurosci Rev. 2006;5:191-239.
9. Wang Y, Pillai S, Wolpaw JR, Chen XY. Motor learning changes GABAergic terminals on spinal motoneurons in normal rats. European Journal of Neuroscience. 2006;23:141-150.
10. Pillai S, Wang Y, Wolpaw JR, Chen XY. Effects of H-reflex up-conditioning on GABAergic terminals on rat soleus motoneurons. European Journal of Neuroscience. 2008;28:668-674.
11. Cha J, Heng C, Reinkensmeyer DJ, Roy RR, Edgerton VR, de Leon RD. Locomotor ability in spinal rats is dependent on the amount of activity imposed on the hindlimbs during treadmill training. J Neurotrauma. 2007;24:1000-1012.
12. Engesser-Cesar C. Wheel running following spinal cord injury improves locomotor recovery and stimulates serotonergic fiber growth. J Neurosci. 2007;25:1931-1939.
13. Smith RR, Brown EH, Shum-Siu A, Whelan A, Burke DA, Benton RL, Magnuson DSK. Swim training initiated acutely after spinal cord injury is ineffective and induces extravasation in and around the epicenter. J Neurotrauma. 2009;26(7):1017-1027.
14. Cote MP, Azzam GA, Lemay MA, Zhukareva V, Houle JD. Activity-dependent increase in neurotrophic factors is associated with an enhanced modulation of spinal reflexes after spinal cord injury. J Neurotrauma. 2011;28:299-309.
15. Borrie SC and Baeumer BE. The nogo-66 receptor family in the intact and diseased CNS. Cell Tissue Res. 2011.
16. Ying Z, Roy RR, Edgerton VR, Gómez-Pinilla F. Exercise restores levels of neurotrophins and synaptic plasticity following spinal cord injury. Exp Neurol. 2005;193(2):411-9.
17. Jovanovic JN, Czernik AJ, Fienberg AA, Greengard P, Sihra TS. Synapsins as mediators of BDNF-enhanced neurotransmitter release. Nat Neurosci. 2000;3(4):323-9.
18. Gomez-Pinilla F, Ying Z, Roy R, Molteni R, Edgerton R. Voluntary exercise induces a BDNF medated mechanism that promotes neuroplasticity. J Neurophysiol. 2002;88:2187-2195.
19. Krajacic A, Ghosh M, Puentes R, Pearse DD, Fouad K. Advantages of delaying the onset of rehabilitative reaching training in rats with incomplete spinal cord injury. Eur J Neurosci. 2009;29(3):641-651
20. Rojas Vega S, Abel T, Lindschulten R, et al. Impact of exercise on neuroplasticity-related proteins in spinal cord injured humans. Neuroscience. 2008;153(4):1064-1070.
21. Pan W, Banks WA, Fasold MB, Bluth J. Transport of brain-derived neurotrophic factor across the blood-brain barrier. Neuropharmacology. 1998;37(12):1553.
22. Deng LX, Hu J, Liu N, Wang X, Smith GM, Wen X, Xu XM. GDNF modifies reactive astrogliosis following robust axonal regeneration through Schwann cell-seeded guidance channels after spinal cord injury. Exp Neurol. 2011;229:238-250.
23. Dolbeare D, Houle JD. Restriction of axonal retraction and promotion of axonal regeneration by chronically injured neurons after intraspinal treatment with glial cell line-derived neurotrophic factor (GDNF). J Neurotrauma. 2003;20(11):1251-1261.
24. Ledda F, Paratcha G, Guzman T, Ibanez CF. GDNF and GFRa1 promote formation of neuronal synapses by ligand-induced cell adhesion. Nat Neurosci. 2007;10(3):293-300.
25. Pedersen LM, Jacobsen LM, Mollerup s, Gjerstad J. Spinal cord long-term potentiation (LTP) is associated with increased dorsal horn gene expression of IL-1B, GDNF, and iNOS. Eur J Pain. 2010;14:255-260.
26. Cotman CW, Berchtold NC. Exercise: A behavioral intervention to enhance brain health and plasticity. Trends Neurosci. 2002;25(6).
27. Koopmans GC, Brans M, Gómez-Pinilla F, et al. Circulating insulin-like growth factor I and functional recovery from spinal cord injury under enriched housing conditions. Eur J Neurosci. 2006;23(4):1035-1046.
28. Ying Z, Roy RR, Edgerton VR, Gomez-Pinilla F. Voluntary exercise increases neurotrophin-3 and its receptor TrkC in the spinal cord. Brain Res. 2003;987:93-99.
29. Ying Z, Roy RR, Edgerton VR, Gomez-Pinilla F. Exercise restores levels of neurotrophins and synaptic plasticity following spinal cord injury. Exp Neurol. 2005;193:411-419.
30. Gomez-Pinilla F, Ying Z, Opazo P, Roy RR, Edgerton VR. Differential regulation by exercise of BDNF and NT-3 in rat spinal cord and skeletal muscle. Eur J Neurosci. 2001;13:1078-1084.
31. Sharma NK, Ryals JM, Gajewski BJ, Wright DE. Aerobic exercise alters analgesia and neurotrophin-3 synthesis in an animal model of chronic widespread pain. Phys Ther. 2010;90:714-725.
32. Skup M, Dwornik A, Macias M, et. al. Long-term locomotor training up-regulates TrkBFL receptor-like proteins, brain-derived neurotrophic factor, and neurotrophin 4 with different topographies of expression in oligodendroglia and neurons in the spinal cord. Exp Neurol. 2002;176:289-307.
33. Endo T, Spenger C, Tominaga T, Brene S, Olson L. Corticol sensory map rearrangement after spinal cord injury: fMRI responses linked to Nogo signaling. Brain. 2007;130:2951-2961.
34. Maier IC, Ichiyama RM, Courtine G, Schnell L, Lavrov I, Edgerton VR, Schwab ME. Differential effects of anti-nogo-A antibody treatment and treadmill training in rats with incomplete spinal cord injury. Brain. 2009;132:1426-1440.
35. Harel NY, Song KH, Tang X, Strittmatter. Nogo receptor deletion and multimodal exercise improve distinct aspects of recovery in cervical spinal cord injury. J Neurotrauma. 2010;27:2055-2066.
36. Profyris C, Cheema SS, Zang D, Azari MF, Boyle K, Petratos S. Degenerative and regenerative mechanisms governing spinal cord injury. Neurobiol Dis. 2004;15:415-436.
37. Sekhon LH, Fehlings MG. Epidemiology, demographics, and pathophysiology of acute spinal cord injury. Spine. 2001;26(24S):S2-S12.
38. Eldadah BA, Faden AI. Caspase pathways, neuronal apoptosis, and CNS injury. J Neurotraum. 2000;17:811-824.
39. Song W, Kwak HB, Lawler JM. Exercise training attenuates age-induced changes in apoptotic signaling in rat skeletal muscle. Antioxid Redox Sign. 2006;8:517-528.
40. Hoffman-Goetz L, Spagnuolo PA. Freewheel exercise training modifies pro- and anti-apoptotic protein expression in mouse splenic lymphocytes. J Sports Med. 2007;28:787-791.
41. Kruger K, Frost S, Most E, et. al. Exercise affects tissue lymphocyte apoptosis via redox-sensitive and Fas-dependent signaling pathways. Am J Physiol Regul Integr Comp Physiol. 2009;296:R1518-R1527.
42. Shuman SL, Bresnahan JC, Beattie MS. Apoptosis of microglia and oligodendrocytes after spinal cord contusion in rats. J Neurosci Res. 1997;50:798-808.
43. Krityakiarana W, Espinosa-Jeffrey A, Ghiani CA, et. al. Voluntary exercise increases oligodendrogenesis in spinal cord. Int J Neurosci. 2010;120:280-290.
44. Beere HM, Wolf BB, Cain K, et. al. Heat shock protein 70 inhibits apoptosis by preventing recruitment of procaspase-9 to the Apaf-1 apoptosome. Nat Cell Biol. 2000;2:469-475.
45. Dupont-Versteegden EE, Houle JD, Dennis RA, et. al. Exercise-induced gene expression in soleus muscle is dependent on time after spinal cord injury in rats. Muscle Nerve. 2004;29:73-81.
46. Nakinishi K, Tanaka N, Ishikawa M, et. al. Responses of microRNAs 124a and 223 following spinal cord injury in mice. Spinal Cord. 2010;48:192-196.
47. Liu NK, Wang XF, Lu QB, Xu XM. Altered microRNA expression following traumatic spinal cord injury. Exp Neurol.2009;219:424-429.
48. Liu G, Keeler BE, Zhukareva V, Houle JD. Cycling exercise affects the expression of apoptosis-induced microRNAs after spinal cord injury in rats. Exp Neurol. 2010;226:200-206.
49. Jeong SY, Seol DW. The role of mitochondria in apoptosis. BMB Reports. 2008;41:11-22.
50. Kim DH, Ko IG, Kim Bo, et. al. Treadmill exercise inhibits traumatic brain injury-induced hippocampal apoptosis. Physiol Behav.2010;121:660-665.
51. Kwak HB, Song W, Lawler JM. Exercise training attenuates age-induced elevation in Bax/Bcl-2 ratio, apoptosis, and remodeling in the rat heart. FASEB J. 2006; 1-21.
52. Dumont RJ, Okonkwo DO, Verma S, Hurlbert RJ, Boulos PT, Ellegala DB, Dumont AS. Acute spinal cord injury, part I: Pathophysiologic mechanisms. Clin Neuropharmacol. 2001;24(5):254-264.
53. Doble A. The role of excitotoxicity in neurodegenerative disease: implications for therapy. Pharmacol Ther. 1999;81(3):163-211.
54. Verea-Portocarrero LP, Mills CD, Ye Z, Fullwood SD, McAdoo DJ, Hulsebosch CE, Westlund KN. Rapid changes in expression of glutamate transporters after spinal cord injury. Brain Res. 2002;927:104-110.
55. Jia J, Zhang F, Hu YS, Wu Y, Yu HX, Liu G, Zhu DN, Xia CM, Cao ZJ, Zhang X, Guo QC. Treadmill pre-training suppresses the release of glutamate resulting from cerebral ischemia in rats. Exp Brain Res. 2010;204:173-179.
56. Chang HC, Yang YR, Wang SGP, Wang RY. Effects of treadmill training on motor performance and extracellular glutamate level in striatum in rats with or without transient middle cerebral artery occlusion. Behav Brain Res. 2009;205:450-455.
57. Zhang Z, Jia J, Wu Y, Hu Y, Wang Y. The effect of treadmill training pre-exercise on glutamate receptor expression in rats after cerebral ischemia. Int J Mol Sci. 2010;11:2658-2669.
58. Real CC, Ferreira AFB, Hernandes MS, Britto LRG, Pires RS. Exercise-induced plasticity of AMPA-type glutamate receptor subunits in the rat brain. Brain Res. 2010;1363:63-71.
59. van Duijnhoven N, Hesse E, Janssen T, Wodzig W, Scheffer P, Hopman M. Impact of exercise training on oxidative stress in individuals with a spinal cord injury. Eur J Appl Physiol. 2010;109:1059-1066.
60. Siamilis S, Jakus J, Nyakas C, Costa A, Mihalik B, Falus A, Radak Z. The effect of exercise and oxidant-antioxidant intervention on the levels of neurotrophins and free radicals in spinal cord of rats. Spinal Cord. 2009;47(6):453-456.
61. Hong Y, Palaksha KJ, park K, Park S, Kim HD, Reiter RJ, Chang Kt. Melatonin plus exercise-based neurorehabilitative therapy for spinal cord injury. J Pineal Res. 2010;49:201-209.
62. Park K, Lee Y, Park S, Lee S, Hong Y, Lee S, Hong Y. Synergistic effect of melatonin on exercise-induced neuronal reconstruction and functional recovery in a spinal cord injury animal model. J Pineal Res. 2010;48:270-281.
63. Revan S, Balci SS, Pepe H, Kurtoglu F, Erol AE, Akkus H. Short duration exhaustive running exercise does not modify lipid hydroperoxide, glutathione peroxidase and catalase. J Sports Med Phys Fitness. 2010;50:235-240.
64. Ogonvszky H, Berkes I, Kumagai S, et al. The effects of moderate-, strenuous-, and over-training on oxidative stress markers, DNA repair, and memory in rat brain. Neurochem. Int. 2005; 46:635-640.
65. Fragala MS, Kraemer WJ, Denegar CR., et al. Neuroendocrine-immune interactions and responses to exercise. Sports Med. 2011; 41:8 621-639.
66. Kohrt WM, Spina RJ, Ehsani AA, Cryer PE, Holloszy JO. Effects of age, adiposity, and fitness level on plasma catecholamine responses to standing and exercise. J Appl Physiol. 1993;75(4):1828-1835.
67. Zouhal H, Jacob C, Delamarche P, Gratas-Delamarche A. Catecholamines and the effects of exercise, training and gender. Sports Med. 2008;38(5):401-423.
68. Mastorkos G, Pavlatou M, Diamanti-Kandarakis E, et al. Exercise and the stress system. Hormones 2005;4(2):73-89.
69. Duclos M, Corcuff M, Rashedi V, et al. Trained versus untrained men: different immediate post-exercise responses of pituitary adrenal axis: a preliminary study. Eur J Appl Physiol. 1997;75:343-350.
70. Luger A, Deutser P, Kyle S, et al. Acute hypothalamic-pituitary-adrenal responses to the stress of treadmill exercise. New Engl J Med. 1987;616:1309-1315.
71. Kjaer M, Secher N, Bach F, et al. Role of motor center activity for hormonal changes and substrate mobilization in humans. Am J Physiol. 1987;253:687-695.
72. Duclos M, Gouarne C, and Bonnemaison D. Acute and chronic effects of exercise on tissue sensitivity to glucocorticoids. J Appl Physiol. 2003;94:869-875.
73. Coyle EF. Physical activity as a metabolic stressor. Am J Clin Nutr. 2000;72(2 Suppl):512S-20S.
74. Schmid A, Huonker M, Stahl F, et al. Free plasma catecholamines in spinal cord injured persons with different injury levels at rest and during exercise. J Auton Nerv Syst. 1998;68(1-2):96-100.
75. Steinberg LL, Lauro FA, Sposito MM, et al. Catecholamine response to exercise in individuals with different levels of paraplegia. Braz J Med Biol Res. 2000;33(8):913-918.
76. Fry AC, Kraemer WJ, Van Borselen F, et al. Catecholamine responses to short-term high-intensity resistance exercise overtraining. J Appl Physiol. 1994;77(2):941-946.
77. McMurray RG, Forsythe WA, Mar MH, Hardy CJ. Exercise intensity-related responses of beta-endorphin and catecholamines. Med Sci Sports Exerc. 1987;19(6):570-574.
78. Jezova D, Vigas M, Tatar P, et al. Plasma testosterone and catecholamine responses to physical exercise of different intensities in men. Eur J Appl Physiol Occup Physiol. 1985;54(1):62-66.
79. Ronsen O, Haug E, Pedersen BK, Bahr R. Increased neuroendocrine response to a repeated bout of endurance exercise. Med Sci Sports Exerc. 2001;33(4):568-575.
80. Webb HE, Weldy ML, Fabianke-Kadue EC, Orndorff GR, Kamimori GH, Acevedo EO. Psychological stress during exercise: Cardiorespiratory and hormonal responses. Eur J Appl Physiol. 2008;104(6):973-981.
81. Jakeman L., Hoschouer E., Basso D.M. Injured mice at the gym: Review, results and considerations for combining chondroitinase and locomotor exercise to enhance recovery after spinal cord injury. Brain Research Bulletin. 2011;84:317-3326.
82. Bregman B.S., Goldberger M.E. Infant lesion effct. III. Anatomicalcorrelates of sparing and recovery of function after spinal cord damage in newborn and adult cats. Dev Brain Res. 1983;9:137-154.
83. Bradbury E., Carter L. Manipulating the glial scar: Chondroitinase ABC as a therapy for spinal cord injury. Brain Rresearch Bulletin. 2011;84: 306-316.
84. Anderson, K.D. Targeting recovery: Priorities of the spinal cord-injured population. J Neurotrauma. 2004;21:1371–1383 (2004).
85. Garcia-Alias G., Barkhuysen S., Buckle M., Fawcett J. Chondroitinase ABC treatment opens a window of opportunity for task specific rehabilitation. Nature Neuroscience. 2009;9(12):1145-1152.
86. Bloomfield SA, Mysiw WJ, Jackson RD. Bone mass and endocrine adaptations to training in spinal cord injured individuals. Bone. 1996;19(1):61-68.
87. Giangregorio LM, Hicks AL, Webber CE, Phillips SM, Craven BC, Bugaresti JM, McCartney N. Body weight supported treadmill training in acute spinal cord injury: impact on muscle and bone. Spinal Cord. 2005;43:649-657.
88. Giangregorio LM, Webber CE, Phillips SM, Hicks AL, Craven BC, Bugaresti JM, McCartney N. Can body weight supported treadmill training increase bone mass and reverse muscle atrophy in individuals with chronic incomplete spinal cord injury? Appl Physiol Nutr Metab. 2006;31:283-291.
89. Jiang SD, Dai LY, Jiang LS. Osteoporosis after spinal cord injury. Osteoporos Int. 2006;17:180-192.
90. Maimoun L, Fattal C, Sultan C. Bone remodeling and calcium homeostasis in patients with spinal cord injury: a review. Metabolism clinical and experimental. 2011;60:1655-1663.
91. Mohr T, Podenphant J, Biering-Sorensen F, Galbo H, Thamsborg G, Kjar M. Increased bone mineral density after prolonged electrically induced cycle training of paralyzed limbs in spinal cord injured man. Calcif Tissue Int. 1997;61:22-25.
92. Barry DW, Kohrt WM. BMD decreases over the course of a year in competitive male cyclists. J Bone Miner Res. 2007;23(4):484-491.
93. Ooi FK, Singh R, Singh HR, Umemura Y. Minimum level of jumping exercise required to maintain exercise-induced bone gains in female rats. Osteoporos Int. 2009;20:963-972.
94. Safdar A, Little JP, Stokl AJ, Hettinga BP, Akhtar M, Tarnopolsky MA. Exercise increases mitochondrial PGC-1 α content and promotes nuclear-mitochondrial cross-talk to coordinate mitochondrial biogenesis. The Journal of Biological Chemistry. 2011.286(12):10605-10617.
95. Safdar A, Bourgeois JM, Ogborn DI, Little JP, Hettinga BP, Akhtar M, Thompson JE, Melov S, Mocellin NJ, Kujoth GC, Prolla TA, Tarnopolsky MA. Endurance exercise resuces progeroid aging and induces systemic mitochondrial rejuvenation in mtDNA mutator mice. PNAS. 2011;1-6.
96. Wang L, Mascher H, Psilander N, Blomstrand E, Sahlin K. Resistance exercise enhances the molecular signaling of mitochondrial biogenesis induced by endurance exercise in human skeletal muscle. J Appl Physiol. 2011;111:1335-1344.


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