Contrary to our first hypothesis, the premanifest and early HD participants did not show enhancement of synaptic plasticity following high-intensity interval exercise, in comparison to either moderate intensity exercise or rest. Specifically, the HD group showed no increases in cortico-motor excitability, glutamatergic facilitation, or decreases in GABAAergic inhibition following either high- or moderate-intensity exercise, a finding which was further supported by the follow-up Bayesian analyses. The HD group showed lower inhibition at baseline which may have attenuated the effect of exercise on plasticity, consistent with our second hypothesis. Taken together, the current findings indicate that HD is associated with an abnormal, attenuated plasticity response to an acute bout of cardiorespiratory exercise in premanifest and early HD, which has implications for the design of exercise interventions in this population.
Several possible mechanisms may account for the absence of exercise induced plasticity in the HD group. The first is that the HD group showed low baseline levels of GABAAergic intracortical inhibition compared to controls, which has been observed in several previous studies27,28,50. In healthy adults, exercise transiently reduces GABAAergic intracortical inhibition17,18. In HD, due to homeostatic mechanisms, the already reduced SICI in HD at baseline may have precluded further reductions in response to exercise 51. However, baseline levels of corticomotor excitability and intracortical facilitation were not different than controls. Given the HD group also did not show the expected plasticity response for these variables, homeostatic mechanisms are unlikely to wholly account for the lack of responsivity of neuroplasticity to exercise.
The absence of the normal neuroplasticity response to exercise may be attributable to altered or dysregulated dopamine signaling, as well as reduced production of BDNF in HD. Specifically, the indirect (D2) dopaminergic pathway within the striatum is affected early in HD, which connects to the motor cortex via the nigrostriatal pathway52, and the dopamine D2 receptor has been shown to mediate the effects of exercise on motor learning53. Mutant huntingtin has also been found to decrease the level of BDNF and its receptor tropomyosin-related kinase B (TrkB) in human and mouse brains, and reduced release of BDNF has been observed in cortical neurons of an HD mouse-model54. This is important, because in healthy adults, acute exercise triggers the release of lactate, dopamine, and the synthesis and release of BDNF55, which are associated with decreased cortical inhibition and corresponding increase in neuroplasticity55,56. The disruption of these processes in early HD, therefore, may interfere with the usual increases in dopamine and BDNF observed following acute exercise, and result in reduced motor cortex plasticity.
These findings have important implications for studies of exercise in HD, as attenuated brain responses to exercise may contribute to the mixed outcomes reported for exercise interventions in HD to date. Although some non-randomized controlled trials of exercise interventions have reported promising results, a recent meta-analysis of motor and cognitive effects from randomized controlled trials indicated no significant effects from the interventions on either the primary outcome (UHDRS motor score) or secondary outcomes (cognitive, health status or physical)11. The current findings indicate that more research is urgently needed to understanding under what circumstances exercise may elicit an optimal neurophysiological response in HD (e.g., by investigating response to different types of exercise, or if there are effects over the longer-term following multiple exercise sessions), to inform the future direction of exercise intervention research in this population.
Our findings contrast with reports of exercise effects on the brain in HD mouse models. For example, wheel running in R6/1 transgenic HD mice increases BDNF gene expression3,4 and delays onset of motor signs7, whereas treadmill exercise in CAG140 knock-in HD mice restores dopamine D2 receptor expression6. The exercise interventions in these studies were long-term, however, rather than a single bout of exercise as implemented in the current study. In HD, longer-term exercise may be needed to have potent beneficial effects on brain chemistry, such as BDNF and dopamine levels.
Our study included people with the HD CAG expansion in very early stage, as much as 43 years before predicted onset, but also participants who had already been diagnosed. Manifest HD participants may have had too much neurodegeneration to respond optimally to exercise. However, we did not find support for this possibility, in that disease burden score in our HD sample was not associated with the size of plasticity response following exercise. Thus neuroplasticity in HD does not necessarily track with disease progression.
Unlike the control group, who showed an increased facilitatory response to iTBS following high-intensity exercise, HD participants did not show any change to iTBS response following either exercise intensity. Only one previous study has utilized a theta-burst paradigm within a HD population. Orth et al. investigated responses to continuous TBS (cTBS) in a mixed premanifest and manifest HD group, and found that cTBS had no effect on inhibition in HD, whereas cTBS resulted in significantly increased inhibition in the control group26. Our current study extends this to show that in premanifest and early HD, there was no detectable effect of iTBS on excitability, inhibition or facilitation, either alone or primed with moderate- or high-intensity exercise. This suggests that these neuroplasticity mechanisms are affected early on in the disease course of HD, which has important implications for the consideration of non-invasive brain stimulation interventions in HD.
The current study has a number of limitations. We did not include an objective measure of cardiorespiratory fitness (e.g., VO2max), and therefore we could not examine any relationships between physical fitness and synaptic plasticity. However, groups did not differ on self-reported physical activity levels, resting heart rate, or objective measures of exercise performance, suggesting our groups were reasonably well matched. We also did not assess the potential role of BDNF genotype as a mediator of the plasticity response within the HD group, due to sample size limitations. Our group and others have previously found BDNF genotype to mediate the plasticity response to iTBS in healthy adults24,57. Further, our small sample size, particularly in the HD group, means that we may have been underpowered to detect an effect of exercise on neuroplasticity in the HD sample. Our follow up Bayesian analyses, however, provide moderate evidence that the results likely reflect a true absence of effect of exercise.