The present study has examined LTD-like rapid-onset central motor plasticity and functional consequences of a transient virtual M1 lesion induced by cTBS in mild-to-moderately afflicted MS patients and healthy controls. cTBS decreased cortical excitability for a duration of about 30
min in patients and controls to a similar degree. In contrast, cTBS impaired force production performance (FPP) significantly for at least 5
min only in healthy controls, but not in MS patients.
TBS-induced plasticity shares properties with synaptic LTD [11
] although this has been questioned by some authors [20
]. The present findings suggest that LTD-like plasticity remains largely intact in MS patients. Along with previous observations [9
], these findings provide evidence that rapid-onset synaptic plasticity as assessed by neurophysiological means is not compromised in mild-to-moderately afflicted MS patients. Previous functional MRI (fMRI) studies probing rapid-onset plasticity in comparably afflicted MS patients provided inconsistent results: One recent fMRI study has demonstrated a decay of regional brain activation by a tapping task over consecutive runs, with similar changes in MS patients and matched controls [10
]. In contrast, another study has shown absent task-specific reductions in fMRI activation of some contralateral cortical regions after motor training in MS patients [5
]. However, while the first study compared activation patterns of a simple hand tapping paradigm before and after training of this movement, the latter assessed changes in activation for a trained (thumb flexion) relative to an untrained task (thumb extension). Attenuation of the blood-oxygen-level dependent (BOLD) response during repeated execution of an over-learnt motor task may reflect neuronal plasticity, but may also simply result from reduction of attention with time. Therefore, physiological mechanisms cannot be directly inferred from functional imagining studies. By employing a brain stimulation technique, the present study allows a less ambiguous, more direct insight into LTD-like phenomena in MS.
MS-related demyelination, axonal damage, and loss of neuronal synapses may critically interfere with basic prerequisites for the expression of Hebbian as well as cTBS-induced plasticity (see also [9
]). In animal models, the endocannabinoid system has been shown to mediate LTD in the neocortex, with CB1 receptors being prominently involved [21
]. Importantly, the endocannabinoid system is strongly dysregulated in acutely relapsing MS [23
]. Therefore, preserved capacity to express LTD-like central motor plasticity may either point to preserved endocannabinoid signalling during stable phases of the disease or to recruitable alternative molecular pathways in MS patients, probably converging towards negative feedback mechanisms at GABAergic and dopaminergic synapses [24
Although the FPP changes following cTBS over M1 were not statistically significant in MS patients, FPP changes were strongly correlated with CML in this group: Surprisingly, and perhaps counterintuitively, the more CML was prolonged, the less evident was the impairment of FPP by cTBS applied to M1. Although functional activation of non-canonical motor regions may contribute to maintaining motor performance in MS patients [14
], it appears highly unlikely that the functional role of M1 has completely be substituted by non-canonical motor regions. Similarly, although the functional redundancy of the motor system (“degeneracy”) might decrease its susceptibility to disruption of one of its nodes (cf. refs. [25
]), it is unlikely that degeneracy increases with higher lesion load because MS-related pathology does not affect only one node, but multiple brain regions in parallel. We favour, therefore, the alternative possibility that increasing brain injury may render the neuronal networks less responsive toward lesion-induction by cTBS. According to this idea, for TMS to induce behavioural interference, it must act on susceptible brain regions. In healthy people, the temporary behavioral deterioration following off-line cTBS possibly depends on the capacity of cortical circuits to express “noisy oscillations” in response to TMS, thereby decreasing the signal-to-noise ratio of highly organized cortical circuits [27
]. In this way, the presence of significant CNS injury (as indexed by prolonged CML) has reduced the capacity of inhibitory interneurons to initiate “noisy oscillations”. This mechanism is consistent with the lack of correlation between cTBS-induced changes of MEP amplitudes and FPP, and provides additional support for conclusions derived from parallel studies of physiology and motor behaviour in healthy subjects [28
], showing that mechanisms underlying behavioural effects of cTBS are distinct from those induced at excitatory synaptic connections.
A potential limitation of this study might be the linear extrapolation of the performance trendline, which was used in order to assess the net effect of cTBS on the FPP. However, prolonged motor training was unlikely to lead to stabile performance due to fatigue and/or lapse of concentration in healthy subjects, and possible even much more in MS patients. Decreasing task difficulty was not an alternative option, as this might have resulted in reduced sensitivity of the task towards disruption by TMS. Therefore, our paradigm can be viewed as a trade-off between feasibility, reduced resilience of MS patients and best possible sensitivity towards a virtual lesion.
Another limitation of our study is the degree of disease severity of MS patients. While LTD-like plasticity may be retained in mild-to-moderately affected MS patients, it cannot be ruled out that LTD-like plasticity may be reduced in patients with more severe CNS injury.