Correlation between cortical plasticity, motor learning and BDNF genotype in healthy subjects
There is good evidence that synaptic plasticity in human motor cortex is involved in behavioural motor learning; in addition, it is now possible to probe mechanisms of synaptic plasticity using a variety of transcranial brain-stimulation protocols. Interactions between these protocols suggest that they both utilise common mechanisms. The aim of the present experiments was to test how well responsiveness to brain-stimulation protocols and behavioural motor learning correlate with each other in a sample of 21 healthy volunteers. We also examined whether any of these measures were influenced by the presence of a Val66Met polymorphism in the BDNF gene since this is another factor that has been suggested to be able to predict response to tests of synaptic plasticity. In 3 different experimental sessions, volunteers underwent 5-Hz rTMS, intermittent theta-burst stimulation (iTBS) and a motor learning task. Blood samples were collected from each subject for BDNF genotyping. As expected, both 5-Hz rTMS and iTBS significantly facilitated MEPs. Similarly, as expected, kinematic variables of finger movement significantly improved during the motor learning task. Although there was a significant correlation between the effect of iTBS and 5-Hz rTMS, there was no relationship in each subject between the amount of TMS-induced plasticity and the increase in kinematic variables during motor learning. Val66Val and Val66Met carriers did not differ in their response to any of the protocols. The present results emphasise that although some TMS measures of cortical plasticity may correlate with each other, they may not always relate directly to measures of behavioural learning. Similarly, presence of the Val66Met BDNF polymorphism also does not reliably predict responsiveness in small groups of individuals. Individual success in behavioural learning is unlikely to be closely related to any single measure of synaptic plasticity.
KeywordsCortical plasticity 5-Hz rTMS Theta burst stimulation BDNF
Synaptic plasticity in the motor cortex is thought to be involved in several forms of motor learning (Muellbacher et al. 2002; Baraduc et al. 2004; Kim and Jahng 2004; Richardson et al. 2006; Iezzi et al. 2010). For example, the sustained improvement in performance that can be achieved by several minutes’ practice of a skilled manual task is abolished by repetitive transcranial magnetic stimulation (rTMS) over the motor cortex (Muellbacher et al. 2001; Baraduc et al. 2004). Because the effect is seen only if rTMS is applied soon after practice, but not if given several hours later, it has been suggested that the motor cortex is necessary for the initial stages of motor learning. Motor learning also increases the amplitude of motor evoked potentials (MEPs) evoked by single pulse in the trained muscle consistent with the idea that neural elements activated by TMS and those involved in motor learning overlap.
In recent years, a number of techniques involving transcranial brain stimulation have been developed that appear to probe synaptic plasticity in motor cortex. Short-term facilitatory effects at synapses, such as augmentation and post-tetanic potentiation, can be probed with short trains of suprathreshold rTMS at 5 Hz. This form of stimulation is accompanied by an increase in the amplitude of MEPs during the train and leads to facilitatory effects that persist for up to 1 s after the end of the train (Berardelli et al. 1998; Pascual-Leone et al. 1994). The 5-Hz rTMS-induced MEP facilitation is probably due to synaptic mechanisms resembling those described in animal experiments as ‘short-term synaptic facilitation’ (Bliss and Lømo 1973; Bliss and Collingridge 1993; Castro-Alamancos and Connors 1996) and is considered a useful technique to investigate short-term plasticity in healthy humans (Ziemann et al. 2008). Although these changes act mainly at pre-synaptic sites, Bliss and Collingridge (1993), Castro-Alamancos and Connors (1996) showed that short-term synaptic enhancement also involves N-methyl-D-aspartate (NMDA) receptor-mediated components of the excitatory post-synaptic potentials.
Longer lasting synaptic facilitation can be produced by applying longer lasting trains of rTMS. Intermittent theta-burst stimulation (iTBS) is a technique that entails applying 50 Hz three-pulse subthreshold bursts repeated at 5 Hz in short trains and can elicit long-lasting changes in cortical excitability persisting up to 60 min that are thought to be caused by long-term potentiation (LTP) at glutamatergic synapses (Huang and Rothwell 2004; Huang et al. 2005; Huang et al. 2007; Teo et al. 2007). Experiments recording the corticospinal volleys in conscious humans (Di Lazzaro et al. 2008 and experiments showing that iTBS leaves the H reflex unchanged (Huang et al. 2008) suggest that iTBS acts on specific neural circuits in the primary motor cortex (M1).
Behavioural motor learning and short- or long-terms effects of rTMS all vary considerably between individuals. Many factors contribute to this variation, such as prior levels of activity, time of day and sex (Ridding and Ziemann 2010). Genetic factors are also thought to affect responses to rTMS protocols. In particular, evidence linking activity-dependent LTP in synaptic connections with neurotrophin brain-derived neurotrophic factor (BDNF) secretion has provided novel and exciting opportunities for exploring how the BDNF genotype modulates LTP-related processes in human cortex. In humans, BDNF occurs in two frequently occurring polymorphisms Val66Val and Val66Met. It has been reported that healthy subjects carrying the Val66Met have a smaller training-dependent increase in MEP size and motor map reorganisation than individuals with the most common Val66Val polymorphism, even though they show no differences in motor task performance (Kleim et al. 2006). Others report that the after-effects of rTMS-inducing LTP/LTD-like plasticity protocols such as iTBS and cTBS are reduced or absent in subjects carrying the met allele of the BDNF gene (Cheeran et al. 2008). In a recent study with quadri-pulse TMS, a technique able to induce LTP-like processes in M1; however, the investigators found no differences in the amount of QPS-induced plasticity in subjects carrying the met allele of the BDNF gene (Nakamura et al. 2011). The functional consequences of BDNF polymorphism on TMS-induced plasticity are therefore still a matter of debate.
No studies have so far investigated whether there is a correlation between individual TMS measures of short-term (5 Hz induced) and long-term (iTBS-induced) synaptic plasticity and performance in motor learning tasks, and whether all 3 are affected by BDNF in the same way. Significant correlations would indicate that responsiveness to one rTMS protocol predicts responsiveness to a different protocol, as well as indicating the expected rate of behavioural improvement in a motor learning task (Di Lazzaro et al. 2010). For this purpose, each volunteer enrolled in this study underwent 5-Hz rTMS, iTBS and a motor task that involved practice-related improvement in kinematic variables as a measure of early motor learning. All participants also underwent BDNF genotyping.
Thirty-eight right-handed healthy subjects (22 men, 16 women; mean age ± SD: 27.95 years ± 5.57 years) participated in the study after giving written informed consent. The institutional review board, following the declaration of Helsinki, approved the experimental procedures.
Subjects were counterbalanced to begin the experimental protocol with one of the three experimental protocols (motor practice, iTBS and 5-Hz rTMS). The experiments took place at least 2 weeks apart. All 38 subjects underwent blood sampling for BDNF genotyping. Of the 38 subjects recruited, 21 participated in all the neurophysiological experiments.
M1 stimulation and MEP recordings
In the different experimental sessions, single-pulse TMS was delivered with a monophasic Magstim 200 stimulator and rTMS with a biphasic Super Rapid Magstim 200 stimulator (The Magstim Company Ltd., Whitland, Dyfed, UK). The stimulators were connected to a figure of eight coil (external wing 9 cm in diameter) placed tangentially over the left hemisphere, with the handle pointing back and away from the midline at about 45°, in the optimal position (hot spot) for eliciting MEPs in the contralateral, right, first dorsal interosseous (FDI) muscle. The active motor threshold (AMT) was the lowest intensity able to evoke an MEP of at least 200 μV in five out of ten consecutive trials during a slight voluntary contraction (about 20% of maximum contraction). Resting motor threshold (RMT) was the lowest stimulus intensity able to evoke MEPs of at least 50 μV in 5 out of 10 consecutive trials. Previous studies (Berardelli et al. 1998; Pascual-Leone et al. 1994; Ziemann et al. 2008) have shown that the MEP facilitation during 5-Hz rTMS is best obtained with the target muscle relaxed. To ensure that subjects completely relaxed the target muscle during rTMS, we, therefore continuously monitored the EMG activity with visual feedback.
5-Hz rTMS was delivered with a biphasic stimulator and consisted in trains of 10 stimuli at an intensity of 120% of RMT. To avoid cumulative after-effects (Gilio et al. 2007; Suppa et al. 2008), 15 trains were delivered with an intertrain interval of 1–2 min. MEPs from I to X were measured peak-to-peak (mV) and then averaged for each subject. As a measure of the 5-Hz rTMS-induced MEP facilitation, we calculated a ratio between the Xth and the Ist MEP in the train (MEP X/I%).
iTBS was delivered with a biphasic stimulator and consisted of three-pulse bursts at 80% of AMT and 50 Hz frequency, repeated every 200 ms (5 Hz) and delivered in short trains lasting 2 s separated an 8 s pause in 20 repetitions for a total number of 600 pulses (Huang et al. 2005). Twenty single TMS pulses were delivered at 0.25 Hz to evoke MEPs of about 1 mV at baseline (T0). The same intensity was used to evoke 20 MEPs at 5 (T1), 15 (T2) and 30 (T3) minutes after iTBS. MEP was measured peak–peak (mV) and then averaged at each time point. As a measure of iTBS-induced M1 plasticity, we also calculated a ratio between MEPs obtained at T2 and those obtained at T0 (MEPs T2/T0%).
As it is known that the MEP facilitation after iTBS is prevented if the target muscle is under contraction (Huang et al. 2008), to ensure that subjects completely relaxed the target muscle during iTBS, we therefore continuously monitored the EMG activity with visual feedback.
The electromyographic (EMG) activity was recorded through a pair of surface Ag/AgCl electrodes placed over the FDI muscle in a belly-tendon fashion. Raw signal, sampled at 5 kHz with a CED 1401 A/D laboratory interface (Cambridge Electronic Design, Cambridge, UK), was amplified and filtered (bandwidth 20 Hz–1 kHz) with a Digitimer D 360 (Digitimer Ltd., Welwyn Garden City, Hertfordshire, UK). Data were stored on a laboratory computer for offline analysis (Signal software, Cambridge Electronic Design, Cambridge, UK).
Motor practice and movement recordings
During the motor task, subjects were comfortably seated in an armchair beside a table. The right arm was firmly placed on the table maintaining the distance between the armchair and the table constant throughout the experiment. The arm was abducted at the shoulder by about 45–50°, the elbow joint was flexed at about 90°, and the wrist was kept in the neutral position. The forearm and the palm of the hand at the metacarpophalangeal joints were firmly secured to the table, and the thumb was abducted, the other fingers were adducted and flexed at the metacarpophalangeal joints by about 70–80° and extended at the interphalangeal joints (Agostino et al. 2007, 2008, Iezzi et al. 2010).
Index finger movements in the 3-D space were recorded with the SMART motion analyser (BTS Engineering, Milan, Italy). The system comprises three infrared cameras (sampling rate, 120 Hz) able to follow a passive optical marker taped over the distal phalanx of the right index finger. A dedicated software reconstructed offline the marker displacement in three main directions: vertical, anteroposterior and mediolateral. Early motor learning was assessed by asking subjects to perform 600 index finger movements ‘as fast as possible’ and continuously encouraging them to do so throughout the motor task. For the motor task, subjects extended the index finger until the metacarpophalangeal joints reached the neutral position after a verbal ‘ready’ signal. Then, after a verbal ‘go’ signal, they rapidly abducted the index finger. Soon after a verbal ‘stop’ signal, they returned the finger to the starting position (Agostino et al. 2007). After a few practice movements, index finger abductions were executed in blocks of 20 movements. To prevent fatigue, an interval of 3 s elapsed between the end of one movement and the beginning of the next and an interval of 5 s elapsed between blocks. These intervals were paced with a stopwatch. Motor practice lasted about 40 min.
A derivative operator was finally used to compute speed and acceleration and determine peak movement values. Movement amplitude, peak velocity and peak acceleration were calculated for each movement and then averaged for each block. As a measure of early motor learning, we also calculated a ratio between the thirtieth and the first movement block for each kinematic variable (movement amplitude XXX block/I block%, peak velocity XXX block/I block% and peak acceleration XXX block/I block%).
To investigate the effects of motor practice on M1 excitability, we delivered 20 single TMS pulses at 0.25 Hz able to evoke MEPs of about 1 mV at baseline (T0). The same intensity was used to evoke 20 MEPs immediately after motor practice ended (T1). MEPs were measured peak–peak (mV) and then averaged at each time point. As a measure of motor learning-induced M1 plasticity, we also calculated a ratio between MEPs obtained at T1 and those obtained at T0 (MEPs T1/T0%).
Genomic DNA was extracted from whole blood. The Val66Met single nucleotide polymorphism (SNP) in the BDNF gene was typed by polymerase chain reaction (PCR) in a total volume of 50 μl containing 150 ng of template DNA, 200 μM of dNTP, 20 pmol of oligonucleotide and 2.5 U of Taq Gold (Roche) in its 1× buffer, 1.5 mM MgCl2 using a DNA thermal cycler (Master Cycler, Eppendorf). Primer sequences were 5′-ACTCTGGAGAGCGTGAAT-3′ and 5′-ATACTGTCACACACGCTG-3′. PCR started with an initial denaturation at 95°C for 12 min, followed by 95°C for 30 s, 60°C for 45 s and 72°C for 60 s for 30 cycles, with a final extension at 72°C for 4 min. PCR products were analysed by denaturing high-performance liquid chromatography (DHPLC, Wave-system Transgenomic, Crewe, UK). Before DHPLC analysis, 10 µl of the PCR product was submitted to a denaturation at 95°C for 10 min and reannealing step for the heteroduplex formation.
The subject’s chromatograms for the amplified fragment were compared with previously sequenced DNA from controls. The abnormal fragments were sequenced by BigDye Deoxy Terminator Kit and analysed on an ABI PRISM Genetic Analyzer 3130 (Applied Biosystems, Foster City, California, USA).
We first analysed the data collected for each subject for all the variables tested independently from the BDNF genotype. We then statistically compared the values obtained in the Val66Val group (n = 14) with those obtained in the Met carriers group (n = 7). We used a repeated measures analysis of variance (ANOVA) with factor ‘Number of stimuli’ (I vs. II, III, IV, V, VI, VII, VIII, IX and X) to test changes in MEP size evoked by each stimulus during the train in all subjects in the study. Tukey honest significance difference test was used for post hoc analysis. One-way ANOVA was used to test whether the size of the first MEP evoked by the first stimulus progressively increased or decreased from the first to the tenth train. We then ran a between-group repeated-measures ANOVA with factor ‘Number of stimuli’ (I vs. II, III, IV, V, VI, VII, VIII, IX and X) to investigate possible differences in MEP facilitation during the 5 Hz train between the Val66Val group and Met carriers group. We used a repeated measures ANOVA with factor ‘Time’ to analyse MEP size at baseline (T0) and immediately after movements (T1) in the motor practice experiment and MEP size at baseline (T0) and 5 (T1), 15 (T2) and 30 (T3) minutes after iTBS. One-way ANOVA was used for post hoc analysis. We then ran a between-group repeated-measures ANOVA with factor ‘Time’ to analyse MEP size and RMT at baseline (T0) and immediately after movements (T1) in the motor practice experiment, and at baseline (T0) and 5 (T1), 15 (T2) and 30 (T3) minutes in the iTBS experiment in Val66Val group and Met carriers group.
The effect of motor practice on movement variables (amplitude, peak velocity and peak acceleration) was tested with a repeated measures ANOVA with main factor ‘Movement blocks’ (I vs. all the other blocks, from II to XXX). A further statistical analysis with a between-group repeated-measures ANOVA was used to compare changes in kinematic variables during motor learning in the Val66Val group and Met carriers group. Spearman correlation was used to assess a possible correlation between 5-Hz rTMS-induced MEP size facilitation (MEP X/I%), iTBS-induced changes in MEP size (MEPs T2/T0%), kinematic measure evaluating motor learning (peak velocity XXX block/I block%), motor practice-induced changes in MEP size and BDNF allele variations. For correlation analysis, BDNF allele was considered a binary variable. Holm’s correction for multiple comparisons was used to discover false significance.
For all statistical analyses P ≤ 0.05 were considered to indicate statistical significance.
None of the participants experienced any adverse effect during or after TMS.
M1 stimulation (5-Hz rTMS and iTBS)
Repeated measures ANOVA for MEP size during 5-Hz rTMS in all subjects showed a significant effect of factor ‘Number of stimuli’ (F9, 180 = 10.04, P < 0.0001). Post hoc test showed that the increase in MEP size became significant after the second stimulus.
Repeated measures ANOVA for MEP size after iTBS showed a significant effect of the factor ‘Time’ (F3, 60 = 11.25, P < 0.00001). Post hoc tests showed that the increase in MEP size was significant at T2 (15 min) and T3 (30 min) after iTBS.
Repeated measures ANOVA for MEP size after motor practice showed a significant effect of ‘Time’ (F1, 20 = 14.77, P < 0.001).
Between-group repeated-measures ANOVA for the MEP size after motor practice showed a significant effect of ‘Time’ (F = 16.23, P < 0.001), but no significant effect of ‘Group’ (F = 0.54, P > 0.05) and no significant interaction (F = 0.91, P = 0.35). Motor practice therefore produced a significant increase in the MEP size, which was similar in both groups of subjects.
Repeated measures ANOVA for kinematic variables (peak amplitude, peak velocity and peak acceleration) showed a significant effect of the factor ‘Blocks’ for peak velocity (F29, 580 = 5.13, P < 0.01) and peak acceleration (F29, 580 = 5.49, P < 0.01) but no significant effect of ‘Blocks’ for peak amplitude (F29, 580 = 0.12, P > 0.05). Post hoc analysis showed that peak velocity significantly increased from the third block and peak acceleration significantly increased from the eighteenth block indicating practice-related improvement in motor performance.
Correlations between kinematic and neurophysiological measures in Val/Val and Met carriers
In the present study in healthy subjects, we first confirmed previous findings that 5-Hz rTMS significantly facilitated MEP size during the train and that iTBS-induced significant long-lasting changes in MEP size (Pascual-Leone et al. 1994; Berardelli et al. 1998; Huang et al. 2005; Ziemann et al. 2008). We also found that when subjects repeated a finger movement ‘as fast as possible’, peak velocity increased significantly across the movement blocks consistent with early motor learning. As expected, when tested soon after motor learning, MEP size significantly increased in the trained muscle.
When we looked for relationships between these measures of motor cortex plasticity, we found that the 5 Hz induced increase in MEP size correlated with the iTBS-induced increase in MEP size, whereas Spearman correlation tests disclosed no significant correlations between motor learning and the amount of TMS-induced MEP facilitation. The main finding in our study is therefore that the susceptibility to develop STP- (5-Hz rTMS) and LTP-like (iTBS)- TMS-induced plasticity in the individual subject does not parallel the ability to develop motor practice. Furthermore, similar to Nakamura et al. (2011), we failed to replicate previous findings that the presence of the Val66Met polymorphism in the human BDNF gene influences the degree of iTBS-induced plasticity (Cheeran et al. 2008) and motor learning. Since the investigators who performed neurophysiological assessments were blinded to the BDNF findings, we feel confident that our findings were not operatorbiased.
The first new finding in this study is the significant correlation between the effects on MEPs of an STP-like inducing protocol (5-Hz rTMS) and an LTP-like inducing protocol (iTBS). The mechanisms responsible for 5-Hz rTMS facilitation probably resemble NMDA-dependent STP described in animal experiments (Bliss and Lømo 1973; Zucker 1989; Cooke and Bliss 2006). Similarly, the after-effects of iTBS seem to involve NMDA neurotransmission because in humans NMDA receptor antagonists such as memantine (Huang et al. 2007) and d-cycloserine (Teo et al. 2007) can prevent them from developing. The correlation we found between STP tested with 5-Hz rTMS and LTP tested with iTBS is therefore in line with animal experiments showing that, although STP and LTP have different time course, they may interact and both involve NMDA and Ca2+ currents (Castro-Alamancos and Connors 1996). It may be that the effect of 5 Hz is more pre-synaptic than NMDA based (Ziemann et al. 2008). If so, then the correlation between iTBS and 5 Hz might depend on the fact that in most models of synaptic plasticity, long-term effects only occur as a consequence of passing through a stage of short-term effects. Thus, despite their very different protocols, it may be that iTBS and 5 Hz both recruit similar mechanisms of short-term plasticity with the longer-term effects of iTBS developing from that.
The second new finding in our study is that our statistical analyses found no correlation between the amount of STP induced by 5-Hz rTMS or the amount of LTP induced by iTBS and our healthy subjects’ ability to improve motor performance. Because the kinematic variables (peak velocity and acceleration) significantly improved during the motor task and because the MEP size in M1 significantly increased after the motor task, we do not think this is due to inadequate motor practice. A number of TMS studies have investigated practice-related changes in cortical excitability during motor learning and examined the time course of changes in functional reorganisation of the human motor cortex that are associated with motor learning (Liepert et al. 1999; Pascual-Leone et al. 1995; Ljubisavljevic 2006). However, no correlations have been reported between the extent of behavioural improvement and the magnitude of the changes in cortical excitability. The lack of correlation between TMS-induced plasticity measures and behaviourally induced plasticity implies that mechanisms of M1 cortical plasticity induced by TMS only partially overlap with those recruited in motor learning. Supporting the hypothesis that mechanisms of M1 cortical plasticity induced by rTMS protocols and behavioural motor learning do not completely overlap, a recent study by Dileone et al. (2010) showed an enhanced response to PAS technique in patients with Costello syndrome who have impaired motor learning and memory. Although this is consistent with two previous studies showing that TMS-induced plasticity over M1 failed to improve motor task performance in healthy subjects (Agostino et al. 2007, 2008), it contrasts with the results of Ziemann and co-workers (Jung and Ziemann 2009) who found both homeostatic and nonhomeostatic interactions between an rTMS protocol and behavioural learning. Indeed, these interactions were similar to those found between TMS protocols alone, suggesting a large overlap between mechanisms of motor learning and TMS probes of plasticity. One reason for these different results may be that Ziemann and colleagues use the paired associative plasticity protocol in which a median nerve stimulus is repeatedly paired with a TMS pulse in order to induce spike timing-dependent plasticity in motor cortex (Stefan et al. 2000). This may engage different synaptic mechanisms to those involved in LTP-like effects of repetitive TMS.
The final result of this study was that the polymorphism of the BDNF gene had no effect on any of the measures of synaptic plasticity that we employed. The lack of influence on rTMS-induced STP-like plasticity has not been reported previously. BDNF potentiates spontaneous and activity-dependent synaptic transmission (Lohof et al. 1993; Levine et al. 1995; Stoop and Poo 1996; Carmignoto et al. 1997; Takei et al. 1997, 1998; Li et al. 1998; Numakawa et al. 1999; Sherwood and Lo 1999; He et al. 2000; Numakawa et al. 2001; Yang et al. 2001) with the Val66Met polymorphism reducing activity-dependent secretion of BDNF (Egan et al. 2003; Chen et al. 2006). In comparison with animal experiments, which use prolonged high frequency trains, 5-Hz rTMS with only short-lasting trains (ten pulses) is probably insufficient to activate cellular processes of activity-dependent secretion of BDNF. Pharmacological studies also showed that 5-Hz induced after-effects on M1 are preferentially prevented by drugs acting on ion channels more than by drugs modulating glutamatergic transmission (Inghilleri et al. 2005). Both of these reflections would explain why we found no effect of the BDNF polymorphism on STP-like plasticity tested with 5-Hz rTMS.
Our observation that iTBS-induced cortical plasticity does not differ between Met carriers, and Val/Val agree with recent findings of Nakamura et al. (2011). Conversely, Cheeran et al. (2008) reported that the responses to two probes of synaptic plasticity (iTBS and cTBS) were reduced in Val66Met carriers. Why was this effect not replicated in the present study? The difference might reflect methodological features. In our study, we evaluated iTBS-induced after-effects by collecting 20 MEPs at three time points (5, 15, and 30 min after iTBS) whereas Cheeran et al. (2008) collected batches of 20 MEPs at more frequent intervals. Previous investigators (Plewnia et al. 2003; Heide et al. 2006; Houdayer et al. 2008; Delvendahl et al. 2010) reported that even very low frequency rTMS such as used to evoke MEPs before and after iTBS can reduce cortical excitability under certain conditions. It could therefore be that BDNF genotype is important in influencing the interaction between these two opposing factors (facilitation from iTBS and suppression from low frequency MEP testing). The lack of differences we found in motor practice between Val/Val and Met carriers agrees with a previous study by Kleim et al. (2006) who found no effect of the polymorphism on motor learning in a very small sample of individuals practicing a hand movement task. However, they did find that there was reduced cortical map reorganisation after learning in individuals with the Val66Met polymorphism. A more recent study investigating 7 Met carries on a population of 21 subjects, similar to our population study, also found no significant interaction between the decline in error rate in a computer driving test and BDNF polymorphism (McHughen et al. 2010). The most likely explanation for our findings is that the polymorphism has a variable effect on activity-dependent BDNF secretion (30% for Met/Met and 18% for Val/Met (Egan et al. 2003; Chen et al. 2006). Since approximately 6–30 min (Hartmann et al. 2001; Poo 2001; Balkowiec and Katz 2002; Zhang and Poo 2002; Tanaka et al. 2008) are necessary for BDNF to be released and to exert its effects on cellular function, any influence of the BDNF polymorphism on the TMS measures we used here might be small and variable.
In conclusion, the present results emphasise the complex relationships between M1 cortical plasticity, motor learning and genotype. The correlation between the interindividual effects of iTBS and 5-Hz rTMS suggests that they may utilise similar mechanisms, whereas the lack of correlation with behavioural effects of learning on MEP suggests that volitional learning engages other pathways, which seem likely to involve areas of brain outside motor cortex.
Conflict of interest
There are no funding sources and potential conflicts of interest from each author that relate to the research covered in the article submitted.
- Egan MF, Kojima M, Callicott JH, Goldberg TE, Kolachana BS, Bertolino A, Zaitsev E, Gold B, Goldman D, Dean M, Lu B, Weinberger DR (2003) The BDNF val66met polymorphism affects activity-dependent secretion of BDNF and human memory and hippocampal function. Cell 112:257–269PubMedCrossRefGoogle Scholar
- Nakamura K, Enomoto H, Hanajima R, Hamada M, Shimizu E, Kawamura Y, Sasaki T, Matsuzawa D, Sutoh C, Shirota Y, Terao Y, Ugawa Y (2011) Quadri-pulse stimulation induced LTP/LTD was not affected by Val66Met polymorphism in the brain-derived neurotrophic factor gene. Neurosci Lett 487:264–267PubMedCrossRefGoogle Scholar
- Takei N, Sasaoka K, Inoue K, Takahashi M, Endo Y, Hatanaka H (1997) Brain-derived neurotrophic factor increases the stimulation-evoked release of glutamate and the levels of exocytosis-associated proteins in cultured cortical neurons from embryonic rats. J Neurochem 68:370–375PubMedCrossRefGoogle Scholar