Parkinsonism and Related Disorders 20S1 (2014) S123–S127
Contents lists available at SciVerse ScienceDirect
Parkinsonism and Related Disorders journal homepage: www.elsevier.com/locate/parkreldis
Exercise: Is it a neuroprotective and if so, how does it work? Michael J. Zigmond a, *, Richard J. Smeyne b a University b St.
of Pittsburgh, Pittsburgh, PA, USA Jude Children’s Research Hospital, Memphis, TN, USA
article info
summary
Keywords: 6-OHDA Environmental enrichment Exercise MPTP Mice Neuroprotection Neurotrophic factors Rats
There is clinical evidence that the symptoms of Parkinson’s disease can be ameliorated by physical exercise, and we have been using animal models to explore the hypothesis that such exercise can also be neuroprotective. To do so we have focused on models of the dopamine deficiency associated with motor symptoms of parkinsonism, including mice treated systemically with MPTP and rats treated with 6-hydroxydopamine. Our focus on exercise derives in part from the extensive literature on the ability of exercise to increase mitochondrial respiration and antioxidant defenses, and to stimulate neuroplasticity. Beginning with constraint therapy and then employing wheel running and environmental enrichment, we have shown that increased limb use can reduce the behavioral effects of dopamine-directed neurotoxins and reduce the loss of dopamine neurons that would otherwise occur. While the mechanism of these effects is not yet known, we suspect a central role for neurotrophic factors whose expression can be stimulated by exercise and which can act on dopamine neurons to reduce their vulnerability to toxins. We believe these data, together with observations from several other laboratories, suggest that exercise, as well as neurotrophic factors, is likely to be an effective neuroprotective strategy in the treatment of Parkinson’s disease. © 2013 Elsevier Ltd. All rights reserved.
1. Parkinson’s disease Parkinson’s disease (PD) is an inexorable neurodegenerative disorder involving problems of movement, emotions, and cognition, affecting some 10 million people worldwide. At the time of clinical diagnosis, substantial neurodegeneration has already occurred. Although there are drugs available that can forestall the disease symptoms in most patients for up to a decade, no treatments yet significantly retard its progression or reverse damage that has occurred by the time of diagnosis. There are many obstacles to finding such a neuroprotective or neurorestorative intervention. A first step, however, is likely to be the identification of the causes of the disease, as this should provide insights regarding the development of disease-modifying treatments. Despite our ignorance regarding the ultimate causes of PD, the considerable literature on this issue indicates that certain assumptions can be made that allow for the establishment of rational models for the development of neuroprotective strategies. We have made several such assumptions, five of which are of particular importance to the approach we will describe in this brief review. (1) Although PD is associated with complex neuropathology * Corresponding author. Michael J. Zigmond, PhD, Department of Neurology, University of Pittsburgh, 7016 BST-3, 3501 Fifth Avenue, Pittsburgh, PA 15260, USA. Tel.: +1 412 648 9720 (office), +1 412 580 0564 (mobile); fax: +1 412 648 7223. E-mail address: [email protected] (M.J. Zigmond). 1353-8020/$ – see front matter © 2013 Elsevier Ltd. All rights reserved.
involving many brain regions, the loss of dopamine (DA) neurons in the substantia nigra (SN) is responsible for most of the characteristic motor dysfunctions, and there is reason to believe that the loss of DA contributes to other aspects of the disorder, as well. (2) Environmental toxins are major risk factors. (3) Mitochondrial dysfunction is central to the pathophysiology of PD. (4) Oxidative stress – which can result from mitochondrial dysfunction, reduced antioxidant capacity, and reactive oxygen species that occur due to DA metabolism – also plays a significant role in the etiology of the disease. (5) There is a significant inflammatory component to the disease. Each of these assumptions has been reviewed in some detail by others (e.g., see [1,2]). Of course, these are not the only salient characteristics of the disease; a full list would incorporate advanced age, protein aggregations, progressive degeneration that includes the loss of neurons other than those that utilize DA, and, of course, being human! However, it is a start. 2. Models of PD – the value of neurotoxins The five assumptions we have listed above provide the basis for a model that involves the administration of one of two relatively selective neurotoxins that we have used in many of our studies over the years. One such toxin is 6-hydroxydopamine (6-OHDA), an electroactive analogue of DA first utilized by Urban Ungerstedt in 1968 to produce an animal model of parkinsonism. Upon uptake from the extracellular space by the high affinity
S124
M.J. Zigmond, R.J. Smeyne / Parkinsonism and Related Disorders 20S1 (2014) S123–S127
DA transporter (DAT), 6-OHDA is concentrated in DA nerve terminals, where it produces reactive oxygen species that lead to the death of these neurons. The other agent in common use is 1-methyl4-phenyl-1,2,3,6-tetrahydropyridine (MPTP), whose toxicity was identified by Irwin Kopin, William Langston, and their colleagues in the early 1980s. MPTP also causes a relatively selective loss of DA neurons in the SN through its active metabolite, 1-methyl-4phenylpyridinium (MPP+ ) that, like 6-OHDA, is concentrated in DA neurons. MPP+ emulates several environmental toxins, and acts to inhibit mitochondrial respiration. There are a number of reviews on the use of neurotoxins to mimic key aspects of PD, including those by Gerlach and Riederer [3] and Bove and Perier [4]. “Model bashing” has recently become fashionable in the field of PD, as it is in other areas of neuroscience. This is unfortunate, even destructive. Whereas one must always be alert to the limitations of one’s models, the use of neurotoxins in studies of PD has led to many important observations. For example, before the discovery of MPTP-induced parkinsonism there were just a handful of reports on the role that environmental toxins might play in the condition; since then there have been more than 250 such reports cited in PubMed. Before the demonstration that glial cellline neurotrophic factor (GDNF) was neuroprotective in 6-OHDA and MPTP models, there were no papers on GDNF and PD; now there are almost 500. And before the development of 6-OHDA as a model for parkinsonism, there were five reports of the reduction of PD symptoms by DA agonists; since the advent of 6-OHDA models, there have been more than 4,500! Of course, one cannot draw a straight line from toxin models to each of these reports; some certainly resulted independently. Yet, there can be no question that toxin models have had, and continue to have, a very important positive influence on the field. Recently, the discovery of genetic mutations in patients that either cause PD or serve as risk factors for the disease has led to a number of genetic animal models, and these have already begun to add insight into disease pathology and to suggest possible interventions that would not have emerged with the older toxin models. However, as the authors of the present review begin to make use of models in which the genome is manipulated, we will also retain toxin models as valuable tools in our research. 3. Physical exercise and PD At a purely logical level, physical exercise is a rational approach to developing neuroprotective and neurorestorative treatments for PD: it increases mitochondrial energy production, stimulates antioxidant defenses, reduces inflammation, causes angiogenesis, and produces synaptogenesis. The use of exercise is also consistent with an enormous body of data testifying to the value of physical therapy in treating motor impairments and improving cognition and emotional status. For many years, there has been the notion that an altered environment can have an impact on neuronal structure and function. One of the earliest mentions of this idea was by the Italian neuroanatomist Michele Vicenzo Malacarne, who reported in 1793 that dogs and birds that had undergone “training” had larger and more complex cerebellar structures than unattended littermates. In the early 1800s, Johann Spurzheim suggested that the brain was capable of increasing in size due to exercise and proposed that this idea be tested by a rigorous application of the scientific method. The idea that the size of the brain could be altered in response to changes in the environment, however, stayed within the purview of phrenologists for over 100 years. In 1947, Donald Hebb described superior maze performance of rats reared as domestic pets compared with their relatively impoverished laboratory-reared counterparts. Then, in the 1960s,
Krech, Bennett, and Rosenzweig brought this concept into the lab for more controlled testing. The team housed groups of rats in a large cage that provided many opportunities to explore the space and the objects it contained, some of which were frequently changed. They found that this led to increased social interaction, physical exercise, and opportunities to learn that were associated with changes in the structure, neurochemistry, and function of the rodent brain [5]. This research has been continued by a number of investigators, including William Greenough and Fred Gage, and we now know that exposure to this type of environment, usually termed environmental complexity or environmental enrichment, can increase neurogenesis, learning behavior, and synaptic density [6,7]. As we have learned from our own studies and will discuss below, increased physical exercise is an important component of the impact of enriched environment. And for more than half a century, the belief that physical exercise can help to forestall the onset of PD and slow its progression has prompted many clinicians to recommend exercise to their patients. Indeed, there is now a considerable body of research to show that exercise does benefit PD patients, just as it benefits those with other conditions involving CNS damage, including Alzheimer’s disease, amyotrophic lateral sclerosis, Huntington’s disease, spinal cord injury, and stroke. The literature indicates, for example, that there is a negative correlation between the incidence of PD and lifetime level of physical activity, and that physical exercise improves movement initiation [8–10]. These clinical data are encouraging; however, they exhibit several shortcomings. The few epidemiological studies that have been performed cannot distinguish between a beneficial impact of exercise on PD and the converse – that patients with PD tend to exercise less. Furthermore, prospective studies have generally been brief, underpowered, lacking proper controls, and/or unable to differentiate between symptomatic improvement and reduction in disease progression. A large clinical trial with the capacity to differentiate between symptomatic relief and disease modification must await the funding that such a trial would require. In fact, we hope to assist in the mounting of just such a trial in the near future. But in the meantime, we have turned to animal studies. As we have noted above, we are mindful of the fact that our models do not replicate either the cause or the pathophysiology of PD. Nonetheless, we believe that they provide important insights into the possibility that exercise can reduce the vulnerability of DA neurons to stressors of many kinds and, if so, by what mechanism. We began our studies on exercise the late 1990s in collaboration with Tim Schallert and his students. We used unilateral 6-OHDA injections to deplete DA on one side of the rat brain and to examine the effects of “constraint therapy” in which the normal, ipsilateral limb was restrained to force over-use of the normally affected, contralateral limb. We found that this intervention led to a reduction in the behavioral and pathological effects of the toxin [11]. Constraint therapy has been proposed as a potential approach to the treatment of stroke, where the insult is acute and typically unilateral as in our 6-OHDA model. However, this intervention can at best have only proof-of-principle value for PD. This is because, in contrast to the 6-OHDA model, PD is a progressive and typically bilateral condition. Moreover, in our hands the effects of constraint therapy on the response to 6-OHDA were not always predictable. Thus, several years ago we turned to the use of running in a wheel or on a treadmill as our intervention, and have generated a great deal of evidence that such exercise also reduces the behavioral consequences of 6-OHDA or MPTP in the rat, mouse, and monkey. This is supported by many other studies in the 6-OHDA-rat [12–15] and the MPTP mouse [16]. In one of our studies, mice (2–4 months old) were given access to a running wheel attached to their cages for 3 months and then
M.J. Zigmond, R.J. Smeyne / Parkinsonism and Related Disorders 20S1 (2014) S123–S127
S125
Fig. 2. Correlation between loss of TH+ cells in SN and TH+ in striatum in 6-OHDA treated rats housed with a running wheel for exercise (squares) or in standard cages (circles) (S. Castro, R. Smeyne, and M. Zigmond, unpublished data).
Fig. 1. Effects of prior exercise on loss of DA neurons in response to MPTP. Mice were housed in cages with attached running wheels for 1–3 months prior to an acute MPTP treatment (4 ×20 mg/kg, i.p., at 2-h intervals). One week later, brains were collected and TH+ cells in SN counted and the results compared with those observed in control mice and MPTP-treated mice without access to wheels. Top panel: Effect of 3 months of wheel running on loss of TH+ neurons in SN. Middle panel: Effects of running for 1, 2, or 3 months on TH+ cells in SN. Bottom panel: Effects of animals allowed to run for 3 months at one-third, two-thirds, or the entire amount of number of normal wheel revolutions per day. Shown are the mean±SEM for 3–6 animals per group. *p < 0.05 compared to control; # p < 0.05 compared to MPTP alone. (Modified from Gerecke et al. 2010 [17].)
administered MPTP (4 × 20 mg/kg at 2 hour intervals, i.p.). One or two weeks later, the mice were sacrificed and the number of TH immunoreactive (TH+ ) cells in the SN estimated by stereology. MPTP-treated animals that did not have access to a running wheel showed an apparent 42% reduction in TH+ cells, whereas those with access to running wheels showed only a 9% loss of these cells (Fig. 1). In another study, adult male rats were provided continuous
access to a running wheel for 12 weeks, given 6-OHDA into the nigrostriatal projection (0.6 mg in 2 ml), and returned to the same cage with continued access to a running wheel for another 8 weeks; control animals had no access to a wheel. 6-OHDA-treated rats whose cages did not have running wheels experienced an average 47% loss of TH+ cells and a 49% loss of striatal TH+ , which was accompanied by a parallel loss of striatal DA (−36%), all of which were significantly attenuated by exercise (Fig. 2). Our colleague Judy Cameron and her students have performed parallel studies of the effects of treadmill running in MPTP-treated monkeys. These studies included a variety of behavioral assays and PET analysis of in vivo binding of 11 C-tetrabenazine (TBZ), a radiolabelled ligand that binds to the vesicular monoamine transporter, VMAT2, and several postmortem indices of DA nerve terminals in caudate and putamen. In each case, neuroprotection was observed [18]. In contrast to the broad agreement among investigators that exercise protects animals against the behavioral effects of 6-OHDA and MPTP, the literature on the protection of DA neurons is mixed. As noted, our own results are unambiguous: increased exercise on a running wheel or treadmill greatly reduces the loss of DA cells and terminals in animal models of DA deficiency, and comparable results have been reported by others [19,20]. Most of this work has involved the application of an exercise regime before toxin treatment, although we have preliminary data suggesting that exercise is also effective when initiated soon after the toxin is applied. However, there also are reports that exercise produces no reduction in the loss of striatal DA and/or TH+ cells in the SN [14,16]. What might account for the discrepancy? There are at least three possibilities: • First, the failure to see protection of DA cells and/or terminals may have resulted from insufficient time for restoration of the DA phenotype. Support for this idea comes from our observation and that of others that loss of a phenotypic marker does not necessarily indicate the death of the DA cell bodies or terminals [21]. • Second, protection of DA neurons may depend on the details of the experiment, including: (a) the type of animal (e.g., species, strain, supplier, age), (b) the therapeutic intervention (treadmill vs. wheel running; brief vs. continual; application during the light or dark cycle of the day), (c) the nature of the insult (e.g., MPTP vs. 6-OHDA; dose; acute vs. chronic; and intracerebral placement of 6-OHDA near the DA cell bodies, axons, or terminals), and perhaps most importantly, (d) the temporal relation between toxin treatment and exercise. Unfortunately,
S126
M.J. Zigmond, R.J. Smeyne / Parkinsonism and Related Disorders 20S1 (2014) S123–S127
there is no consensus on any of these variables and thus direct comparison across laboratories is virtually impossible. • Third, under certain conditions, exercise actually may not reduce behavioral symptoms by protecting DA cells at all, but by allowing the remaining DA cells, as well as other components of the basal ganglia circuitry, to compensate for lost DA neurons. Indeed, the ability of patients and toxin-treated animals to compensate for loss of DA neurons is well known, and almost certainly involves increased DA synthesis and release from residual DA neurons [17] although there there have been counterarguments [22]. We will come back to this point at the end of our review. 4. Identifying a neuroprotective dose of exercise We also examined the relationship between exercise-induced protection and amount of running, varying the number of revolutions per day, as well as the number of months of exercise [17]. The average number of wheel revolutions performed by mice given free access to a wheel was ~18,000 revolutions per day (3.24 km). Mice permitted to run this full distance were protected against subsequent MPTP, whereas mice running only 6,000 revolutions were not protected at all. In a separate experiment mice were given free wheel access for 1, 2, or 3 months before MPTP. Protection required 2–3 months of running, with no protection after only 1 month of running (Fig. 1). Is exercise equivalent to enrichment? Although we began our review with a discussion of environmental enrichment, most of our data – and the focus of our brief review – have been about physical exercise. The two are not the same. Studies show that although animals housed in an enriched environment are more active, some aspects of such an environment cannot be replicated by physical activity alone. Nonetheless, in our own hands the neuroprotective effects of exercise and enrichment on the loss of TH+ cells in SN of mice exposed to MPTP are equivalent [23]. Studies of preconditioning and their relevance to our toxin studies. We believe that exercise may act, in part, as a mild stressor, conditioning DA neurons to tolerate more intense stress later. Support for this hypothesis comes from our observation that exercise increases corticosterone in plasma, as well as the antiapoptotic chaperone Hsp70 in the striatum, both of which are classical stress responses implicated in preconditioning. However, the bulk of our laboratories’ insights into preconditioning to date arise from in vitro studies undertaken by Rehana Leak. For example, pre-incubation of the dopaminergic cell line MN9D with a sub-toxic concentration of a toxin was found to reduce the subsequent impact of a normally toxic concentration of either 6-OHDA or MG132. This was associated with an increase in Bcl2 protein and abolished by cyclohexamide, accompanied by an increase in pERK and pAkt, and abolished by inhibitors of kinase phosphorylation [24]. Thus, it appears that both protein synthesis and kinase activation were involved. We have also observed preconditioning in PC12 cells, primary neurons, and in an in vivo model of DA depletion. And we have recently shown that preconditioning doses of MPTP can produce a long-lasting transcriptional refractoriness in a transgenic mouse model [25], and have preliminary evidence of preconditioning in a 6-OHDA rat model, as well. Does exercise act via neurotrophic factors? We believe that the capacity of exercise to increase the availability of several neurotrophic factors (NTFs) is central to its neuroprotective effects and to its likely value as an intervention for PD. This is because the resulting multimodal action of these NTFs, each presumably being released in a physiologically appropriate concentration and location and on to one of several NTF receptors, should then be able to act simultaneously on many, if not all, of the cellular processes that are adversely affected in PD. These assumptions
are based on the following observations. First, exercise increases the levels of GDNF [23], as well as several other NTFs, including BDNF, IGF1, and FGF2 [26]. Second, our colleague Barry Hoffer and others have shown that administration of GDNF can reduce the impact of neurotoxins on DA neurons (e.g., [21,27]). Third, decreased expression of GDNF or BDNF via knockout or antisense technology results in progressive loss of TH+ neurons in the SN of mice and an increased sensitivity to MPTP [28]. Fourth, GDNF and BDNF message and/or protein are decreased in PD brains [29]. These observations led to the hypothesis that NTF administration would be a useful treatment for PD. Indeed, open label trials of intra-parenchymal infusion of GDNF were positive [30]. However, a subsequent blinded trial initiated in 2004 was declared a failure by the sponsoring company after only 6 months and halted. We believe that the termination of the trial was premature and unfortunately has given rise to the belief by some that research on GDNF and on NTFs in general should be abandoned. In fact, there are many alternative explanations for the apparent failure of that trial. These include poor dosing or poor delivery, poor or inconsistent distribution of the GDNF in brain tissue, too short a trial period, the use of patients with extremely large losses of DA neurons at the trial’s outset, and an over-reaction by the company to the appearance of circulating antibodies and to abnormal sprouting seen in a small number of monkeys in response to high dose GDNF. On the other hand, despite the decision to terminate the trial, studies of GDNF and other NTFs continue, and we predict that one or more of these factors will be found to be central to an effective neurotrophic strategy, including that of exercise. 5. Conclusions and future directions A disease-modifying strategy for the treatment of PD is desperately needed. More widespread diagnosis of individuals living in underserved communities, the continued spread of environmental contamination, increased number of individuals living to advanced age, and simple population growth will combine to raise the number of patients with PD well above its currently estimated level. One estimate suggests a doubling of the patient population by 2030. And an increased burden on caregivers and society will accompany this growth in patient numbers. Yet, although treatments for PD have existed in countries such as India and China for thousands of years and in Western medicine for at least half a century, neuroprotective interventions remain elusive. Based on our own studies and those of others, as well as a myriad of clinical reports, we believe that exercise should be seen as a primary or adjunct treatment of PD, one that has the advantage that its “side effects” are themselves health-promoting. In fact, there is little controversy regarding the ability of exercise to reduce the symptoms of PD in humans or for its ability to reduce the long-term behavioral effects of DA-directed neurotoxins in animal models. It is true that not everyone finds that exercise protects DA cells and/or terminals, and we have discussed some of the many possible explanations for this seeming contradiction between the behavioral and neurobiological effects of exercise that have been reported. A diagram summarizing some of those explanations is provided in Fig. 3, which emphasizes that DA signaling can, in principle, remain normal even if the number of DA neurons does not. Of course, many aspects of the hypotheses we have outlined in this review are yet to be validated, and there is much more that we do not yet know: Is exercise neuroprotective in patients as it is in animal models? What type of exercise is effective and for how long? How long does the effect last? When is it best to introduce exercise as a treatment, and is there a limited window of opportunity? How does exercise work and can a pharmacological treatment be substituted for those individuals who cannot exercise?
M.J. Zigmond, R.J. Smeyne / Parkinsonism and Related Disorders 20S1 (2014) S123–S127
Fig. 3. A model for exercise-induced protection against toxin-induced loss of DA function. We hypothesize that exercise increases DA signaling through increased availability of NTFs, which in turn can promote mitochondrial energy production, antioxidant defense, synaptogenesis, reduced inflammation, angiogenesis, and other processes that suppress apoptosis. (A) Shown on the left is a DA neuron being protected or rescued by exercise. (B) On the right is shown exercise increasing the capacity of otherwise healthy DA neurons to deliver a signal by increasing DA synthesis via increased tyrosine hydroxylase (TH) activity, increasing DA release, and decreasing DA reuptake. (C) This can compensate for non-functional neurons (left) via the combination of increased DA diffusion and increased target sensitivity to DA due to the absence of presynaptic DA uptake sites and increased postsynaptic receptors. Solid arrows indicate an excitatory effect; dotted lines indicate inhibition.
Additionally, and perhaps most puzzling, why do so many of us fail to take advantage of the many apparent benefits of exercise, and what can be done to change this? Acknowledgements All of the studies from our laboratories that are reported in this review were performed in compliance with the guidelines of the US National Institutes of Heath and our respective institutions. Our work has been supported by NIH (NS39006, NS45906, NS19608, and NS070825), the U.S. Army (ERMS 03281022), the American Lebanese Syrian Associated Charities, the Michael J. Fox Foundation, the National Parkinson’s Foundation, the Parkinson’s Disease Foundation, and our academic institutions. We thank Amy Rupert for her editorial assistance with this manuscript; Tim Schallert, who had the insight more than 15 years ago that exercise could be neuroprotective for PD; our present colleagues in the study of exercise and PD, Judy Cameron, Barry Hoffer, and Mart Saarma; and the many other people who have contributed to our research and our ideas in recent years. Conflict of interests The authors have no conflicts of interest to declare. References [1] Moore DJ, West AB, Dawson VL, Dawson TM. Molecular pathophysiology of Parkinson’s disease. Annu Rev Neurosci 2005;28:57–87. [2] Blandini F. Neural and immune mechanisms in the pathogenesis of Parkinson’s disease. J Neuroimmune Pharmacol 2013;8:189–201. [3] Gerlach M, Riederer P. Animal models of Parkinson’s disease: an empirical comparison with the phenomenology of the disease in man. J Neural Transm 1996;103:987–1041. [4] Bove J, Perier C. Neurotoxin-based models of Parkinson’s disease. Neuroscience 2012;211:51–76.
S127
[5] Rosenzweig MR. Modification of brain circuits through experience. In: Bermudez-Rattoni F, editor, From neural plasticity and memory: from genes to brain imaging. Boca Raton, FL: Taylor & Francis Group, LLC; 2007. [6] van Praag H, Kempermann G, Gage FH. Neural consequences of environmental enrichment. Nat Rev Neurosci 2000;1:191–8. [7] Nithianantharajah J, Hannan AJ. Enriched environments, experience-dependent plasticity and disorders of the nervous system. Nat Rev Neurosci 2006;7:697– 709. [8] Goodwin VA, Richards SH, Taylor RS, Taylor AH, Campbell JL. The effectiveness of exercise interventions for people with Parkinson’s disease: a systematic review and meta-analysis. Mov Disord 2008;23:631–40. [9] Xu Q, Park Y, Huang X, Hollenbeck A, Blair A, Schatzkin A, et al. Physical activities and future risk of Parkinson disease. Neurology 2010;75:341–8. [10] Chen H, Zhang SM, Schwarzschild MA, Hernan ´ MA, Ascherio A. Physical activity and the risk of Parkinson disease. Neurology 2005;64:664–9. [11] Tillerson JL, Cohen AD, Philhower J, Miller GW, Zigmond MJ, Schallert T. Forced limb-use effects on the behavioral and neurochemical effects of 6-hydroxydopamine. J Neurosci 2001;21:4427–35. [12] Jadavji NM, Kolb B, Metz GA. Enriched environment improves motor function in intact and unilateral dopamine-depleted rats. Neuroscience 2006;140:1127–38. [13] Mabandla M, Kellaway L, St Clair Gibson A, Russell VA. Voluntary running provides neuroprotection in rats after 6-hydroxydopamine injection into the medial forebrain bundle. Metab Brain Dis 2004;19:43–50. [14] O’Dell SJ, Gross NB, Fricks AN, Casiano BD, Nguyen TB, Marshall JF. Running wheel exercise enhances recovery from nigrostriatal dopamine injury without inducing neuroprotection. Neuroscience 2007;144:1141–51. [15] Steiner B, Winter C, Hosman K, Siebert E, Kempermann G, Petrus DS, et al. Enriched environment induces cellular plasticity in the adult substantia nigra and improves motor behavior function in the 6-OHDA rat model of Parkinson’s disease. Exp Neurol 2006;199:291–300. [16] Fisher BE, Petzinger GM, Nixon K, Hogg E, Bremmer S, Meshul CK, et al. Exercise-induced behavioral recovery and neuroplasticity in the 1-methyl-4phenyl-1,2,3,6-tetrahydropyridine-lesioned mouse basal ganglia. J Neurosci Res 2004;77:378–90. [17] Gerecke KM, Jia Y, Pani A, Pagala V, Smeyne RY. Exercise protects against MPTPinduced neurotoxicity in mice. Brain Res 2010;1341:72–83. [18] Zigmond MJ, Abercrombie ED, Berger TW, Grace AA, Stricker EM. Compensations after lesions of central dopaminergic neurons: some clinical and basic implications. Trends Neurosci 1990;13:290–6. [19] Bezard E, Dovero S, Belin D, Duconger S, Jackson-Lewis V, Przedborski S, et al. Enriched environment confers resistance to 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine and cocaine: involvement of dopamine transporter and trophic factors. J Neurosci 2003;23:10999–1007. [20] Anstrom KK, Schallert T, Woodlee MT, Shattuck A, Roberts DC. Repetitive vibrissae-elicited forelimb placing before and immediately after unilateral 6-hydroxydopamine improves outcome in a model of Parkinson’s disease. Behav Brain Res 2007;179:183–91. [21] Cohen AD, Zigmond MJ, Smith AD. Effects of intrastriatal GDNF on the response of dopamine neurons to 6-hydroxydopamine: time course of protection and neurorestoration. Brain Res 2011;1370:80–8. [22] Bezard E, Gross CE, Brotchie JM. Presymptomatic compensation in Parkinson’s disease is not dopamine-mediated. Trends Neurosci 2003;26:215–21. [23] Faherty CJ, Raviie Shepherd K, Herasimtschuk A, Smeyne RJ. Environmental enrichment in adulthood eliminates neuronal death in experimental Parkinsonism. Brain Res Mol Brain Res 2005;134:170–9. [24] Leak RK, Liou AK, Zigmond MJ. Effect of sublethal 6-hydroxydopamine on the response to subsequent oxidative stress in dopaminergic cells: evidence for preconditioning. J Neurochem 2006;99:1151–63. [25] Pattarini R, Rong Y, Shepherd KR, Jiao Y, Qu C, Smeyne RJ, et al. Long-lasting transcriptional refractoriness triggered by a single exposure to 1-methyl-4phenyl-1,2,3,6-tetrahydropyrimidine. Neuroscience 2012;214:84–105. [26] Cotman CW, Berchtold NC, Christie LA. Exercise builds brain health: key roles of growth factor cascades and inflammation. Trends Neurosci 2007;30:464–72. [27] Hoffer BJ, Hoffman A, Bowenkamp K, Huettl P, Hudson J, Martin D, et al. Glial cell line-derived neurotrophic factor reverses toxin-induced injury to midbrain dopaminergic neurons in vivo. Neurosci Lett 1994;182:107–11. [28] Boger HA, Middaugh LD, Huang P, Zaman V, Smith AC, Hoffer BJ, et al. A partial GDNF depletion leads to earlier age-related deterioration of motor function and tyrosine hydroxylase expression in the substantia nigra. Exp Neurol 2006;202: 336–47. [29] Siegel GJ, Chauhan NB. Neurotrophic factors in Alzheimer’s and Parkinson’s disease brain. Brain Res Rev 2000;33:199–227. [30] Gill SS, Patel NK, Hotton GR, O’Sullivan K, McCarter R, Bunnage M, et al. Direct brain infusion of glial cell line-derived neurotrophic factor in Parkinson disease. Nat Med 2003;9:589–95.
Contents lists available at SciVerse ScienceDirect
Parkinsonism and Related Disorders journal homepage: www.elsevier.com/locate/parkreldis
Exercise: Is it a neuroprotective and if so, how does it work? Michael J. Zigmond a, *, Richard J. Smeyne b a University b St.
of Pittsburgh, Pittsburgh, PA, USA Jude Children’s Research Hospital, Memphis, TN, USA
article info
summary
Keywords: 6-OHDA Environmental enrichment Exercise MPTP Mice Neuroprotection Neurotrophic factors Rats
There is clinical evidence that the symptoms of Parkinson’s disease can be ameliorated by physical exercise, and we have been using animal models to explore the hypothesis that such exercise can also be neuroprotective. To do so we have focused on models of the dopamine deficiency associated with motor symptoms of parkinsonism, including mice treated systemically with MPTP and rats treated with 6-hydroxydopamine. Our focus on exercise derives in part from the extensive literature on the ability of exercise to increase mitochondrial respiration and antioxidant defenses, and to stimulate neuroplasticity. Beginning with constraint therapy and then employing wheel running and environmental enrichment, we have shown that increased limb use can reduce the behavioral effects of dopamine-directed neurotoxins and reduce the loss of dopamine neurons that would otherwise occur. While the mechanism of these effects is not yet known, we suspect a central role for neurotrophic factors whose expression can be stimulated by exercise and which can act on dopamine neurons to reduce their vulnerability to toxins. We believe these data, together with observations from several other laboratories, suggest that exercise, as well as neurotrophic factors, is likely to be an effective neuroprotective strategy in the treatment of Parkinson’s disease. © 2013 Elsevier Ltd. All rights reserved.
1. Parkinson’s disease Parkinson’s disease (PD) is an inexorable neurodegenerative disorder involving problems of movement, emotions, and cognition, affecting some 10 million people worldwide. At the time of clinical diagnosis, substantial neurodegeneration has already occurred. Although there are drugs available that can forestall the disease symptoms in most patients for up to a decade, no treatments yet significantly retard its progression or reverse damage that has occurred by the time of diagnosis. There are many obstacles to finding such a neuroprotective or neurorestorative intervention. A first step, however, is likely to be the identification of the causes of the disease, as this should provide insights regarding the development of disease-modifying treatments. Despite our ignorance regarding the ultimate causes of PD, the considerable literature on this issue indicates that certain assumptions can be made that allow for the establishment of rational models for the development of neuroprotective strategies. We have made several such assumptions, five of which are of particular importance to the approach we will describe in this brief review. (1) Although PD is associated with complex neuropathology * Corresponding author. Michael J. Zigmond, PhD, Department of Neurology, University of Pittsburgh, 7016 BST-3, 3501 Fifth Avenue, Pittsburgh, PA 15260, USA. Tel.: +1 412 648 9720 (office), +1 412 580 0564 (mobile); fax: +1 412 648 7223. E-mail address: [email protected] (M.J. Zigmond). 1353-8020/$ – see front matter © 2013 Elsevier Ltd. All rights reserved.
involving many brain regions, the loss of dopamine (DA) neurons in the substantia nigra (SN) is responsible for most of the characteristic motor dysfunctions, and there is reason to believe that the loss of DA contributes to other aspects of the disorder, as well. (2) Environmental toxins are major risk factors. (3) Mitochondrial dysfunction is central to the pathophysiology of PD. (4) Oxidative stress – which can result from mitochondrial dysfunction, reduced antioxidant capacity, and reactive oxygen species that occur due to DA metabolism – also plays a significant role in the etiology of the disease. (5) There is a significant inflammatory component to the disease. Each of these assumptions has been reviewed in some detail by others (e.g., see [1,2]). Of course, these are not the only salient characteristics of the disease; a full list would incorporate advanced age, protein aggregations, progressive degeneration that includes the loss of neurons other than those that utilize DA, and, of course, being human! However, it is a start. 2. Models of PD – the value of neurotoxins The five assumptions we have listed above provide the basis for a model that involves the administration of one of two relatively selective neurotoxins that we have used in many of our studies over the years. One such toxin is 6-hydroxydopamine (6-OHDA), an electroactive analogue of DA first utilized by Urban Ungerstedt in 1968 to produce an animal model of parkinsonism. Upon uptake from the extracellular space by the high affinity
S124
M.J. Zigmond, R.J. Smeyne / Parkinsonism and Related Disorders 20S1 (2014) S123–S127
DA transporter (DAT), 6-OHDA is concentrated in DA nerve terminals, where it produces reactive oxygen species that lead to the death of these neurons. The other agent in common use is 1-methyl4-phenyl-1,2,3,6-tetrahydropyridine (MPTP), whose toxicity was identified by Irwin Kopin, William Langston, and their colleagues in the early 1980s. MPTP also causes a relatively selective loss of DA neurons in the SN through its active metabolite, 1-methyl-4phenylpyridinium (MPP+ ) that, like 6-OHDA, is concentrated in DA neurons. MPP+ emulates several environmental toxins, and acts to inhibit mitochondrial respiration. There are a number of reviews on the use of neurotoxins to mimic key aspects of PD, including those by Gerlach and Riederer [3] and Bove and Perier [4]. “Model bashing” has recently become fashionable in the field of PD, as it is in other areas of neuroscience. This is unfortunate, even destructive. Whereas one must always be alert to the limitations of one’s models, the use of neurotoxins in studies of PD has led to many important observations. For example, before the discovery of MPTP-induced parkinsonism there were just a handful of reports on the role that environmental toxins might play in the condition; since then there have been more than 250 such reports cited in PubMed. Before the demonstration that glial cellline neurotrophic factor (GDNF) was neuroprotective in 6-OHDA and MPTP models, there were no papers on GDNF and PD; now there are almost 500. And before the development of 6-OHDA as a model for parkinsonism, there were five reports of the reduction of PD symptoms by DA agonists; since the advent of 6-OHDA models, there have been more than 4,500! Of course, one cannot draw a straight line from toxin models to each of these reports; some certainly resulted independently. Yet, there can be no question that toxin models have had, and continue to have, a very important positive influence on the field. Recently, the discovery of genetic mutations in patients that either cause PD or serve as risk factors for the disease has led to a number of genetic animal models, and these have already begun to add insight into disease pathology and to suggest possible interventions that would not have emerged with the older toxin models. However, as the authors of the present review begin to make use of models in which the genome is manipulated, we will also retain toxin models as valuable tools in our research. 3. Physical exercise and PD At a purely logical level, physical exercise is a rational approach to developing neuroprotective and neurorestorative treatments for PD: it increases mitochondrial energy production, stimulates antioxidant defenses, reduces inflammation, causes angiogenesis, and produces synaptogenesis. The use of exercise is also consistent with an enormous body of data testifying to the value of physical therapy in treating motor impairments and improving cognition and emotional status. For many years, there has been the notion that an altered environment can have an impact on neuronal structure and function. One of the earliest mentions of this idea was by the Italian neuroanatomist Michele Vicenzo Malacarne, who reported in 1793 that dogs and birds that had undergone “training” had larger and more complex cerebellar structures than unattended littermates. In the early 1800s, Johann Spurzheim suggested that the brain was capable of increasing in size due to exercise and proposed that this idea be tested by a rigorous application of the scientific method. The idea that the size of the brain could be altered in response to changes in the environment, however, stayed within the purview of phrenologists for over 100 years. In 1947, Donald Hebb described superior maze performance of rats reared as domestic pets compared with their relatively impoverished laboratory-reared counterparts. Then, in the 1960s,
Krech, Bennett, and Rosenzweig brought this concept into the lab for more controlled testing. The team housed groups of rats in a large cage that provided many opportunities to explore the space and the objects it contained, some of which were frequently changed. They found that this led to increased social interaction, physical exercise, and opportunities to learn that were associated with changes in the structure, neurochemistry, and function of the rodent brain [5]. This research has been continued by a number of investigators, including William Greenough and Fred Gage, and we now know that exposure to this type of environment, usually termed environmental complexity or environmental enrichment, can increase neurogenesis, learning behavior, and synaptic density [6,7]. As we have learned from our own studies and will discuss below, increased physical exercise is an important component of the impact of enriched environment. And for more than half a century, the belief that physical exercise can help to forestall the onset of PD and slow its progression has prompted many clinicians to recommend exercise to their patients. Indeed, there is now a considerable body of research to show that exercise does benefit PD patients, just as it benefits those with other conditions involving CNS damage, including Alzheimer’s disease, amyotrophic lateral sclerosis, Huntington’s disease, spinal cord injury, and stroke. The literature indicates, for example, that there is a negative correlation between the incidence of PD and lifetime level of physical activity, and that physical exercise improves movement initiation [8–10]. These clinical data are encouraging; however, they exhibit several shortcomings. The few epidemiological studies that have been performed cannot distinguish between a beneficial impact of exercise on PD and the converse – that patients with PD tend to exercise less. Furthermore, prospective studies have generally been brief, underpowered, lacking proper controls, and/or unable to differentiate between symptomatic improvement and reduction in disease progression. A large clinical trial with the capacity to differentiate between symptomatic relief and disease modification must await the funding that such a trial would require. In fact, we hope to assist in the mounting of just such a trial in the near future. But in the meantime, we have turned to animal studies. As we have noted above, we are mindful of the fact that our models do not replicate either the cause or the pathophysiology of PD. Nonetheless, we believe that they provide important insights into the possibility that exercise can reduce the vulnerability of DA neurons to stressors of many kinds and, if so, by what mechanism. We began our studies on exercise the late 1990s in collaboration with Tim Schallert and his students. We used unilateral 6-OHDA injections to deplete DA on one side of the rat brain and to examine the effects of “constraint therapy” in which the normal, ipsilateral limb was restrained to force over-use of the normally affected, contralateral limb. We found that this intervention led to a reduction in the behavioral and pathological effects of the toxin [11]. Constraint therapy has been proposed as a potential approach to the treatment of stroke, where the insult is acute and typically unilateral as in our 6-OHDA model. However, this intervention can at best have only proof-of-principle value for PD. This is because, in contrast to the 6-OHDA model, PD is a progressive and typically bilateral condition. Moreover, in our hands the effects of constraint therapy on the response to 6-OHDA were not always predictable. Thus, several years ago we turned to the use of running in a wheel or on a treadmill as our intervention, and have generated a great deal of evidence that such exercise also reduces the behavioral consequences of 6-OHDA or MPTP in the rat, mouse, and monkey. This is supported by many other studies in the 6-OHDA-rat [12–15] and the MPTP mouse [16]. In one of our studies, mice (2–4 months old) were given access to a running wheel attached to their cages for 3 months and then
M.J. Zigmond, R.J. Smeyne / Parkinsonism and Related Disorders 20S1 (2014) S123–S127
S125
Fig. 2. Correlation between loss of TH+ cells in SN and TH+ in striatum in 6-OHDA treated rats housed with a running wheel for exercise (squares) or in standard cages (circles) (S. Castro, R. Smeyne, and M. Zigmond, unpublished data).
Fig. 1. Effects of prior exercise on loss of DA neurons in response to MPTP. Mice were housed in cages with attached running wheels for 1–3 months prior to an acute MPTP treatment (4 ×20 mg/kg, i.p., at 2-h intervals). One week later, brains were collected and TH+ cells in SN counted and the results compared with those observed in control mice and MPTP-treated mice without access to wheels. Top panel: Effect of 3 months of wheel running on loss of TH+ neurons in SN. Middle panel: Effects of running for 1, 2, or 3 months on TH+ cells in SN. Bottom panel: Effects of animals allowed to run for 3 months at one-third, two-thirds, or the entire amount of number of normal wheel revolutions per day. Shown are the mean±SEM for 3–6 animals per group. *p < 0.05 compared to control; # p < 0.05 compared to MPTP alone. (Modified from Gerecke et al. 2010 [17].)
administered MPTP (4 × 20 mg/kg at 2 hour intervals, i.p.). One or two weeks later, the mice were sacrificed and the number of TH immunoreactive (TH+ ) cells in the SN estimated by stereology. MPTP-treated animals that did not have access to a running wheel showed an apparent 42% reduction in TH+ cells, whereas those with access to running wheels showed only a 9% loss of these cells (Fig. 1). In another study, adult male rats were provided continuous
access to a running wheel for 12 weeks, given 6-OHDA into the nigrostriatal projection (0.6 mg in 2 ml), and returned to the same cage with continued access to a running wheel for another 8 weeks; control animals had no access to a wheel. 6-OHDA-treated rats whose cages did not have running wheels experienced an average 47% loss of TH+ cells and a 49% loss of striatal TH+ , which was accompanied by a parallel loss of striatal DA (−36%), all of which were significantly attenuated by exercise (Fig. 2). Our colleague Judy Cameron and her students have performed parallel studies of the effects of treadmill running in MPTP-treated monkeys. These studies included a variety of behavioral assays and PET analysis of in vivo binding of 11 C-tetrabenazine (TBZ), a radiolabelled ligand that binds to the vesicular monoamine transporter, VMAT2, and several postmortem indices of DA nerve terminals in caudate and putamen. In each case, neuroprotection was observed [18]. In contrast to the broad agreement among investigators that exercise protects animals against the behavioral effects of 6-OHDA and MPTP, the literature on the protection of DA neurons is mixed. As noted, our own results are unambiguous: increased exercise on a running wheel or treadmill greatly reduces the loss of DA cells and terminals in animal models of DA deficiency, and comparable results have been reported by others [19,20]. Most of this work has involved the application of an exercise regime before toxin treatment, although we have preliminary data suggesting that exercise is also effective when initiated soon after the toxin is applied. However, there also are reports that exercise produces no reduction in the loss of striatal DA and/or TH+ cells in the SN [14,16]. What might account for the discrepancy? There are at least three possibilities: • First, the failure to see protection of DA cells and/or terminals may have resulted from insufficient time for restoration of the DA phenotype. Support for this idea comes from our observation and that of others that loss of a phenotypic marker does not necessarily indicate the death of the DA cell bodies or terminals [21]. • Second, protection of DA neurons may depend on the details of the experiment, including: (a) the type of animal (e.g., species, strain, supplier, age), (b) the therapeutic intervention (treadmill vs. wheel running; brief vs. continual; application during the light or dark cycle of the day), (c) the nature of the insult (e.g., MPTP vs. 6-OHDA; dose; acute vs. chronic; and intracerebral placement of 6-OHDA near the DA cell bodies, axons, or terminals), and perhaps most importantly, (d) the temporal relation between toxin treatment and exercise. Unfortunately,
S126
M.J. Zigmond, R.J. Smeyne / Parkinsonism and Related Disorders 20S1 (2014) S123–S127
there is no consensus on any of these variables and thus direct comparison across laboratories is virtually impossible. • Third, under certain conditions, exercise actually may not reduce behavioral symptoms by protecting DA cells at all, but by allowing the remaining DA cells, as well as other components of the basal ganglia circuitry, to compensate for lost DA neurons. Indeed, the ability of patients and toxin-treated animals to compensate for loss of DA neurons is well known, and almost certainly involves increased DA synthesis and release from residual DA neurons [17] although there there have been counterarguments [22]. We will come back to this point at the end of our review. 4. Identifying a neuroprotective dose of exercise We also examined the relationship between exercise-induced protection and amount of running, varying the number of revolutions per day, as well as the number of months of exercise [17]. The average number of wheel revolutions performed by mice given free access to a wheel was ~18,000 revolutions per day (3.24 km). Mice permitted to run this full distance were protected against subsequent MPTP, whereas mice running only 6,000 revolutions were not protected at all. In a separate experiment mice were given free wheel access for 1, 2, or 3 months before MPTP. Protection required 2–3 months of running, with no protection after only 1 month of running (Fig. 1). Is exercise equivalent to enrichment? Although we began our review with a discussion of environmental enrichment, most of our data – and the focus of our brief review – have been about physical exercise. The two are not the same. Studies show that although animals housed in an enriched environment are more active, some aspects of such an environment cannot be replicated by physical activity alone. Nonetheless, in our own hands the neuroprotective effects of exercise and enrichment on the loss of TH+ cells in SN of mice exposed to MPTP are equivalent [23]. Studies of preconditioning and their relevance to our toxin studies. We believe that exercise may act, in part, as a mild stressor, conditioning DA neurons to tolerate more intense stress later. Support for this hypothesis comes from our observation that exercise increases corticosterone in plasma, as well as the antiapoptotic chaperone Hsp70 in the striatum, both of which are classical stress responses implicated in preconditioning. However, the bulk of our laboratories’ insights into preconditioning to date arise from in vitro studies undertaken by Rehana Leak. For example, pre-incubation of the dopaminergic cell line MN9D with a sub-toxic concentration of a toxin was found to reduce the subsequent impact of a normally toxic concentration of either 6-OHDA or MG132. This was associated with an increase in Bcl2 protein and abolished by cyclohexamide, accompanied by an increase in pERK and pAkt, and abolished by inhibitors of kinase phosphorylation [24]. Thus, it appears that both protein synthesis and kinase activation were involved. We have also observed preconditioning in PC12 cells, primary neurons, and in an in vivo model of DA depletion. And we have recently shown that preconditioning doses of MPTP can produce a long-lasting transcriptional refractoriness in a transgenic mouse model [25], and have preliminary evidence of preconditioning in a 6-OHDA rat model, as well. Does exercise act via neurotrophic factors? We believe that the capacity of exercise to increase the availability of several neurotrophic factors (NTFs) is central to its neuroprotective effects and to its likely value as an intervention for PD. This is because the resulting multimodal action of these NTFs, each presumably being released in a physiologically appropriate concentration and location and on to one of several NTF receptors, should then be able to act simultaneously on many, if not all, of the cellular processes that are adversely affected in PD. These assumptions
are based on the following observations. First, exercise increases the levels of GDNF [23], as well as several other NTFs, including BDNF, IGF1, and FGF2 [26]. Second, our colleague Barry Hoffer and others have shown that administration of GDNF can reduce the impact of neurotoxins on DA neurons (e.g., [21,27]). Third, decreased expression of GDNF or BDNF via knockout or antisense technology results in progressive loss of TH+ neurons in the SN of mice and an increased sensitivity to MPTP [28]. Fourth, GDNF and BDNF message and/or protein are decreased in PD brains [29]. These observations led to the hypothesis that NTF administration would be a useful treatment for PD. Indeed, open label trials of intra-parenchymal infusion of GDNF were positive [30]. However, a subsequent blinded trial initiated in 2004 was declared a failure by the sponsoring company after only 6 months and halted. We believe that the termination of the trial was premature and unfortunately has given rise to the belief by some that research on GDNF and on NTFs in general should be abandoned. In fact, there are many alternative explanations for the apparent failure of that trial. These include poor dosing or poor delivery, poor or inconsistent distribution of the GDNF in brain tissue, too short a trial period, the use of patients with extremely large losses of DA neurons at the trial’s outset, and an over-reaction by the company to the appearance of circulating antibodies and to abnormal sprouting seen in a small number of monkeys in response to high dose GDNF. On the other hand, despite the decision to terminate the trial, studies of GDNF and other NTFs continue, and we predict that one or more of these factors will be found to be central to an effective neurotrophic strategy, including that of exercise. 5. Conclusions and future directions A disease-modifying strategy for the treatment of PD is desperately needed. More widespread diagnosis of individuals living in underserved communities, the continued spread of environmental contamination, increased number of individuals living to advanced age, and simple population growth will combine to raise the number of patients with PD well above its currently estimated level. One estimate suggests a doubling of the patient population by 2030. And an increased burden on caregivers and society will accompany this growth in patient numbers. Yet, although treatments for PD have existed in countries such as India and China for thousands of years and in Western medicine for at least half a century, neuroprotective interventions remain elusive. Based on our own studies and those of others, as well as a myriad of clinical reports, we believe that exercise should be seen as a primary or adjunct treatment of PD, one that has the advantage that its “side effects” are themselves health-promoting. In fact, there is little controversy regarding the ability of exercise to reduce the symptoms of PD in humans or for its ability to reduce the long-term behavioral effects of DA-directed neurotoxins in animal models. It is true that not everyone finds that exercise protects DA cells and/or terminals, and we have discussed some of the many possible explanations for this seeming contradiction between the behavioral and neurobiological effects of exercise that have been reported. A diagram summarizing some of those explanations is provided in Fig. 3, which emphasizes that DA signaling can, in principle, remain normal even if the number of DA neurons does not. Of course, many aspects of the hypotheses we have outlined in this review are yet to be validated, and there is much more that we do not yet know: Is exercise neuroprotective in patients as it is in animal models? What type of exercise is effective and for how long? How long does the effect last? When is it best to introduce exercise as a treatment, and is there a limited window of opportunity? How does exercise work and can a pharmacological treatment be substituted for those individuals who cannot exercise?
M.J. Zigmond, R.J. Smeyne / Parkinsonism and Related Disorders 20S1 (2014) S123–S127
Fig. 3. A model for exercise-induced protection against toxin-induced loss of DA function. We hypothesize that exercise increases DA signaling through increased availability of NTFs, which in turn can promote mitochondrial energy production, antioxidant defense, synaptogenesis, reduced inflammation, angiogenesis, and other processes that suppress apoptosis. (A) Shown on the left is a DA neuron being protected or rescued by exercise. (B) On the right is shown exercise increasing the capacity of otherwise healthy DA neurons to deliver a signal by increasing DA synthesis via increased tyrosine hydroxylase (TH) activity, increasing DA release, and decreasing DA reuptake. (C) This can compensate for non-functional neurons (left) via the combination of increased DA diffusion and increased target sensitivity to DA due to the absence of presynaptic DA uptake sites and increased postsynaptic receptors. Solid arrows indicate an excitatory effect; dotted lines indicate inhibition.
Additionally, and perhaps most puzzling, why do so many of us fail to take advantage of the many apparent benefits of exercise, and what can be done to change this? Acknowledgements All of the studies from our laboratories that are reported in this review were performed in compliance with the guidelines of the US National Institutes of Heath and our respective institutions. Our work has been supported by NIH (NS39006, NS45906, NS19608, and NS070825), the U.S. Army (ERMS 03281022), the American Lebanese Syrian Associated Charities, the Michael J. Fox Foundation, the National Parkinson’s Foundation, the Parkinson’s Disease Foundation, and our academic institutions. We thank Amy Rupert for her editorial assistance with this manuscript; Tim Schallert, who had the insight more than 15 years ago that exercise could be neuroprotective for PD; our present colleagues in the study of exercise and PD, Judy Cameron, Barry Hoffer, and Mart Saarma; and the many other people who have contributed to our research and our ideas in recent years. Conflict of interests The authors have no conflicts of interest to declare. References [1] Moore DJ, West AB, Dawson VL, Dawson TM. Molecular pathophysiology of Parkinson’s disease. Annu Rev Neurosci 2005;28:57–87. [2] Blandini F. Neural and immune mechanisms in the pathogenesis of Parkinson’s disease. J Neuroimmune Pharmacol 2013;8:189–201. [3] Gerlach M, Riederer P. Animal models of Parkinson’s disease: an empirical comparison with the phenomenology of the disease in man. J Neural Transm 1996;103:987–1041. [4] Bove J, Perier C. Neurotoxin-based models of Parkinson’s disease. Neuroscience 2012;211:51–76.
S127
[5] Rosenzweig MR. Modification of brain circuits through experience. In: Bermudez-Rattoni F, editor, From neural plasticity and memory: from genes to brain imaging. Boca Raton, FL: Taylor & Francis Group, LLC; 2007. [6] van Praag H, Kempermann G, Gage FH. Neural consequences of environmental enrichment. Nat Rev Neurosci 2000;1:191–8. [7] Nithianantharajah J, Hannan AJ. Enriched environments, experience-dependent plasticity and disorders of the nervous system. Nat Rev Neurosci 2006;7:697– 709. [8] Goodwin VA, Richards SH, Taylor RS, Taylor AH, Campbell JL. The effectiveness of exercise interventions for people with Parkinson’s disease: a systematic review and meta-analysis. Mov Disord 2008;23:631–40. [9] Xu Q, Park Y, Huang X, Hollenbeck A, Blair A, Schatzkin A, et al. Physical activities and future risk of Parkinson disease. Neurology 2010;75:341–8. [10] Chen H, Zhang SM, Schwarzschild MA, Hernan ´ MA, Ascherio A. Physical activity and the risk of Parkinson disease. Neurology 2005;64:664–9. [11] Tillerson JL, Cohen AD, Philhower J, Miller GW, Zigmond MJ, Schallert T. Forced limb-use effects on the behavioral and neurochemical effects of 6-hydroxydopamine. J Neurosci 2001;21:4427–35. [12] Jadavji NM, Kolb B, Metz GA. Enriched environment improves motor function in intact and unilateral dopamine-depleted rats. Neuroscience 2006;140:1127–38. [13] Mabandla M, Kellaway L, St Clair Gibson A, Russell VA. Voluntary running provides neuroprotection in rats after 6-hydroxydopamine injection into the medial forebrain bundle. Metab Brain Dis 2004;19:43–50. [14] O’Dell SJ, Gross NB, Fricks AN, Casiano BD, Nguyen TB, Marshall JF. Running wheel exercise enhances recovery from nigrostriatal dopamine injury without inducing neuroprotection. Neuroscience 2007;144:1141–51. [15] Steiner B, Winter C, Hosman K, Siebert E, Kempermann G, Petrus DS, et al. Enriched environment induces cellular plasticity in the adult substantia nigra and improves motor behavior function in the 6-OHDA rat model of Parkinson’s disease. Exp Neurol 2006;199:291–300. [16] Fisher BE, Petzinger GM, Nixon K, Hogg E, Bremmer S, Meshul CK, et al. Exercise-induced behavioral recovery and neuroplasticity in the 1-methyl-4phenyl-1,2,3,6-tetrahydropyridine-lesioned mouse basal ganglia. J Neurosci Res 2004;77:378–90. [17] Gerecke KM, Jia Y, Pani A, Pagala V, Smeyne RY. Exercise protects against MPTPinduced neurotoxicity in mice. Brain Res 2010;1341:72–83. [18] Zigmond MJ, Abercrombie ED, Berger TW, Grace AA, Stricker EM. Compensations after lesions of central dopaminergic neurons: some clinical and basic implications. Trends Neurosci 1990;13:290–6. [19] Bezard E, Dovero S, Belin D, Duconger S, Jackson-Lewis V, Przedborski S, et al. Enriched environment confers resistance to 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine and cocaine: involvement of dopamine transporter and trophic factors. J Neurosci 2003;23:10999–1007. [20] Anstrom KK, Schallert T, Woodlee MT, Shattuck A, Roberts DC. Repetitive vibrissae-elicited forelimb placing before and immediately after unilateral 6-hydroxydopamine improves outcome in a model of Parkinson’s disease. Behav Brain Res 2007;179:183–91. [21] Cohen AD, Zigmond MJ, Smith AD. Effects of intrastriatal GDNF on the response of dopamine neurons to 6-hydroxydopamine: time course of protection and neurorestoration. Brain Res 2011;1370:80–8. [22] Bezard E, Gross CE, Brotchie JM. Presymptomatic compensation in Parkinson’s disease is not dopamine-mediated. Trends Neurosci 2003;26:215–21. [23] Faherty CJ, Raviie Shepherd K, Herasimtschuk A, Smeyne RJ. Environmental enrichment in adulthood eliminates neuronal death in experimental Parkinsonism. Brain Res Mol Brain Res 2005;134:170–9. [24] Leak RK, Liou AK, Zigmond MJ. Effect of sublethal 6-hydroxydopamine on the response to subsequent oxidative stress in dopaminergic cells: evidence for preconditioning. J Neurochem 2006;99:1151–63. [25] Pattarini R, Rong Y, Shepherd KR, Jiao Y, Qu C, Smeyne RJ, et al. Long-lasting transcriptional refractoriness triggered by a single exposure to 1-methyl-4phenyl-1,2,3,6-tetrahydropyrimidine. Neuroscience 2012;214:84–105. [26] Cotman CW, Berchtold NC, Christie LA. Exercise builds brain health: key roles of growth factor cascades and inflammation. Trends Neurosci 2007;30:464–72. [27] Hoffer BJ, Hoffman A, Bowenkamp K, Huettl P, Hudson J, Martin D, et al. Glial cell line-derived neurotrophic factor reverses toxin-induced injury to midbrain dopaminergic neurons in vivo. Neurosci Lett 1994;182:107–11. [28] Boger HA, Middaugh LD, Huang P, Zaman V, Smith AC, Hoffer BJ, et al. A partial GDNF depletion leads to earlier age-related deterioration of motor function and tyrosine hydroxylase expression in the substantia nigra. Exp Neurol 2006;202: 336–47. [29] Siegel GJ, Chauhan NB. Neurotrophic factors in Alzheimer’s and Parkinson’s disease brain. Brain Res Rev 2000;33:199–227. [30] Gill SS, Patel NK, Hotton GR, O’Sullivan K, McCarter R, Bunnage M, et al. Direct brain infusion of glial cell line-derived neurotrophic factor in Parkinson disease. Nat Med 2003;9:589–95.
