Noninvasive brain stimulation (NIBS) techniques have become an emerging field in clinical neuroscience due to its effect to modulate cortical excitability (Nitsche and Paulus, 2000, 2001; Reis et al., 2008b) and cognitive or motor function (Iyer et al., 2005; Hummel and Cohen, 2006; Reis et al., 2008a). Among NIBS techniques, transcranial magnetic stimulation (TMS) is widely used for brain stimulation, which can be applied for rehabilitation or therapy for neuropsychiatric disorders (Lam et al., 2008; Koganemaru et al., 2010; Rossini et al., 2010; Dayan et al., 2013; Schulz et al., 2013). Recently, transcranial direct current stimulation (tDCS) is also widely applied in clinical fields because this is a safe, well-tolerated method, which has been shown to induce prolonged excitability changes in humans' cortical regions, resulting in long-term potential (LTP)/depression (LTD)-like synaptic modifications, a cellular correlate of learning and memory (Liebetanz et al., 2002; Nitsche et al., 2003). tDCS is thought to achieve its effects by polarizing neurons and indirectly influencing their firing rates and excitability. Anodal stimulation depolarizes the cell bodies and axon hillock region of corticospinal neurons and increases their excitability, whereas cathodal tDCS has the opposite effect (Nitsche and Paulus, 2000).

Moreover, recent studies reported that local transcranial static magnetic stimulation (tSMS) over the human M1 produced by a small high-powered neodymium magnet can transiently reduce the cortical excitability (Oliviero et al., 2011; Silbert et al., 2013). And we found that the reduction of M1 excitability by tSMS is partly related to the modulation of the short interval intracortical inhibitory circuit (Nojima et al., 2015). Although the physiological mechanisms that underlie this change are not yet known for certain, animal experiments indicated the alteration of the ion channel function embedded in the membrane (Rosen, 2003b). It is possible that high-powered tSMS can affect the orientation of the membrane phospholipids due to their diamagnetic anisotropy. The effect of tSMS was also showed to reduce cortical excitability both in the sensorimotor cortex of humans (Oliviero et al., 2011; Silbert et al., 2013; Kirimoto et al., 2014; Nojima et al., 2015), and in the visual cortex of cats and monkeys (Aguila et al., 2016). Furthermore, the significant behavioral changes were recently reported with the application of tSMS to the visual cortex in monkeys and humans (Gonzalez-Rosa et al., 2015; Aguila et al., 2016).

TMS-based techniques and tDCS deliver electric currents to the cortex to obtain short or long term effects on cortical excitability. The tSMS might be the only NIBS technique that is able to produce a lasting change in cortical excitability that is not associated directly with induced electric currents (Oliviero et al., 2015). And, the advantages of tSMS were its ease of use, absence of an uncomfortable sensation, lack of the need for high operational skill and expensive devices.

However, tSMS has a disadvantage that it is hard to stimulate to the focal target area within cortex. On the other hand, TMS paired with low-frequency peripheral nerve stimulation (PNS) can induce a long lasting, reversible, and somatotopically focal alteration in the human cortical excitability (paired associative stimulation: PAS), which may be related to the associative LTP/LTD (Stefan et al., 2000). Moreover, it was reported that the combination of DCS with low-frequency stimulation in mouse M1 slice also resulted in long-lasting increases in the synaptic efficacy (Fritsch et al., 2010). Here, we hypothesized that tSMS combining with PNS can enhance the plastic change in the target cortical motor area.

To delineate the physiological mechanism of the effect of the combination tSMS with PNS, detailed TMS measurements were performed.

Regarding the MEP amplitude for the right APB, two-way repeated-measures ANOVA showed no significant main effect of Time, but significant interaction of Group × Time [F_{(3, 39)} = 2.20 (p = 0.048)]. Post-hoc analysis revealed a significant suppression of MEP amplitude in post-0 (p < 0.011) of the tSMS+PNS group compared with pre condition, suggesting that tSMS combined with PNS influenced MEP amplitude immediately after intervention (Figure 2). By contrast, there was no significant effect of Time and Time × Group in the right ADM. A summary of the mean amplitude of both muscles is given in Table 1.

For rMT, although two-way repeated-measures ANOVA showed no significant effects of Time and Group interaction, there was the tendency of increase immediately after tSMS+PNS intervention (Figure 3).

In the cortical inhibitory and excitatory neural circuit, although two-way repeated-measures ANOVA showed no significant effects of Time and Time × Group interactions for both muscles, it also showed the trend toward of enhancement of SICI immediately after intervention of tSMS+PNS (Figure 4).

We found that tSMS to M1 combined with PNS to the median nerve produced a focal reduction in mean MEP amplitudes in the APB but not the ADM. We also confirmed that the lack of suppressive effects of APB in tSMS alone with short duration is in accord with the previous observation (Oliviero et al., 2011). Moreover, sham tSMS with PNS failed to induce any change. These findings suggested that the synaptic activation induced by a combination of tSMS with PNS can lead to somatotopically focal modulation in the cortical function. Our results suggested that the combination of tSMS and PNS can induce the somatotopic focal excitability change in cortical motor area.

This study revealed a significant decrease in MEP amplitude of the right APB only after the tSMS with PNS. However, this change of cortical excitability did not generalize to the right ADM muscle, which was an adjacent muscle innervated by a different nerve. This result suggested that an interesting feature of the combination of tSMS and PNS is the somatotopic focal effects which may help to shape the regional pattern of reorganization. This somatotopy is consist of other combining stimulation protocols (Stefan et al., 2000; Wolters et al., 2003; Koganemaru et al., 2009; Rizzo et al., 2009).

Our prediction was that concurrent PNS will potentiate the plastic change if the stimulation is given with tSMS. Consistent with our hypothesis, paired stimuli were tended to induce reduction in resting excitability of corticospinal output neurons. In contrast, sham tSMS with PNS for 5 min failed to induce significant change in the MEP amplitude. Previous PNS studies have demonstrated that more than 10 min of repeated PNS were required to provoke consistent increases in corticospinal excitability (Ridding et al., 2000, 2001; Pyndt and Ridding, 2004; Quartarone et al., 2006). On the other hand, original study (Oliviero et al., 2011) has reported that tSMS exposure less than for 10 min failed to reduce the MEP amplitude after the end of stimulation. The results in the present study suggested that the duration of tSMS for 5 min can suppress the cortical excitability if PNS was simultaneously applied. Although Kirimoto reported the loss of suppression of SEPs when tSMS was simultaneously combined with SEPs recording (Kirimoto et al., 2014), the results cannot be directly comparable to the present results. Firstly, our study measured the corticospinal excitability which was not measured in the previous study. Moreover, the parameters of peripheral nerve stimulation were totally different (1 Hz vs. 3.3 Hz). Concurrent afferent stimulation of peripheral nerves may produce an enhanced reduction effect of the corticospinal output.

Regarding PNS, it has been reported that high-frequency PNS (90 Hz) applied over the hand muscles in healthy volunteers for 30 min was associated with a decrease of sensory threshold and parallel decrease of corticospinal excitability (Mima et al., 2004). It suggested that long-term intervention was needed to modulate corticospinal excitability by PNS alone, and the additional tSMS exposure may potentiate the suppression effect induced by PNS. By contrast, it has also reported that prolonged PNS (>120 min) could enhance excitability in the contralateral M1 (Ridding et al., 2000, 2001). The divergence of the PNS-induced cortical excitability change might be related to the different stimulus frequencies. In patients with chronic stroke, PNS has been shown to transiently improve motor performance (Sawaki et al., 2006; Celnik et al., 2007) although little is known about the mechanisms and ability of PNS to modulate the effects of motor training. Further studies would be necessary to test the underling neural mechanism of PNS-induced cortical excitability change.

One of the plausible explanations of the effect of combining NIBS with PNS may be related to Brain-derived neurotrophic factor (BDNF) secretion. A recent ex vivo animal study in mice, in which anodal DCS applied to M1 slices was coupled with low-frequency synaptic stimulation, showed to induce long-term synaptic plasticity (Fritsch et al., 2010). Notably, these effects required activity-dependent BDNF secretion, a finding that was in agreement with previous demonstrations of the role of BDNF in NIBS-induced plasticity (Cheeran et al., 2008; Antal et al., 2010). BDNF is crucial for human motor learning, thus modulation of BDNF by external stimulation may help control the neuroplastic potential. Since the BDNF mechanism associated with cortical excitability change is still speculative, further study would be needed.

In addition to the decrease of MEP amplitudes, we found the increase of rMT just after the tSMS in accordance with previous studies (Silbert et al., 2013; Nojima et al., 2015). Although the basis of rMT is not fully certain yet, pharmacological research has suggested that it is modulated by membrane excitability (Ziemann et al., 1996a,b). These suggested that intrinsic excitability could be modulated by tSMS and may underlie some of the post-stimulation effects on cortical excitability.

Our previous study showed that tSMS can enhance the GABAergic system (Nojima et al., 2015). SICI is measured in a paired-pulse TMS protocol involving a subthreshold conditioning stimulus followed by a suprathreshold test stimulus with a short interstimulus interval of 1–5 ms (Kujirai et al., 1993; Nakamura et al., 1997; Chen et al., 1998). This inhibitory effect is thought to result primarily from activation by the conditioning stimulus of low threshold GABAergic interneurons in the cortex (Ziemann et al., 1996c; Di Lazzaro et al., 2002; Paulus et al., 2008). Our results suggested that suppression of focal M1 function may be partly related to the modulation of the GABAergic system.

The effect of tSMS on cortical excitability has also been confirmed in the other brain areas (Kirimoto et al., 2014; Aguila et al., 2016), and was reported that had created a reversible cortical scotoma in the animal experiment. Several animal studies reported that tSMS interfere with neural function (Rosen and Lubowsky, 1987; McLean et al., 2003, 2008; Coots et al., 2004; Aguila et al., 2016). Of these, it suggested that tSMS directly interfere with the functioning of membrane ion channels and consequently with the generation of action potentials (Coots et al., 2004), possibly due to the diamagnetic anisotropic properties of membrane phospholipids (Rosen, 2003b; Miyakoshi, 2005). Previous studies revealed that the activation kinetics of both sodium (Rosen, 2003a; Coots et al., 2004) and calcium (Rosen, 1996) channels were transiently affected during tSMS. The hypothesis is that tSMS would cause reorientation of membrane phospholipids, which would cause a deformation of ion channels embedded in the membrane and therefore altering their activation kinetics (Rosen, 2003b). Other possible influences have been postulated effects on cellular growth and size alterations of the cell cytoskeleton (Rosen and Chastney, 2009).

Regarding the strength of the magnetic field, recent study reported that magnet used in this study was in range between 120 and 200 mT 2–3 cm from the surface of the magnet (Rivadulla et al., 2014; Kirimoto et al., 2016). Therefore, it seems that this range is enough to obtain biological effects.

On the other hand, the magnet size we used in the present study was slightly bigger than the previous study. In the control experiment, we confirmed that tSMS for 5 min using our magnet does not change the M1 excitability, which is consistent with the previous study (Oliviero et al., 2011). However, since the effect of magnet size/strength and duration on M1 excitability has not been systematically investigated, further studies would be necessary to clarify these points.

Previous studies have already suggested that the effect of tSMS on excitability change was disappeared for couple min after removing the magnet (Roshan et al., 2003; Ortu et al., 2008). Due to the limitation of the time during the experiment, we tested only for the hot spot for the APB muscle. It is possible that the small test MEP amplitude in ADM might influence on the results of this study. It was reported that changes in the amplitude of the test MEP had markedly different effects on SICI, especially only slight inhibition in weaker test MEP amplitude (Ziemann et al., 1996a; Sanger et al., 2001). However, the amount of SICI in the ADM (0.572 ± 0.083) at the baseline was similar to that in the APB. It suggests that the SICI phenomenon in both muscles would have been occurred in a proper way.

This study provides a new combined tSMS protocol that can be used for the induction of somatotopically focal M1 excitability change. Moreover, we confirmed that 5 min of co-stimulation induced a decrease in the excitability of the corticospinal output from the stimulated M1. These results suggested that tSMS exposure could be a valuable tool in research studies of cortical function. And enhancement of SICI function in somatotopically focal brain area by combining with PNS might be a new promising therapeutic tool for neurological disorders associated with GABA dysfunction, such as epilepsy (McLean et al., 2003, 2008) and dystonia (Ikoma et al., 1996; Garibotto et al., 2011; Boecker, 2013). Because somatotopic specificity is an important characteristic of this co-stimulation, we believe this protocol is suitable for a clinical therapeutic approach.

The Ethics Committee of Kyoto University Graduate School of Medicine (Kyoto, Japan) and Nagoya University Graduate School of Medicine (Nagoya, Japan). Each participant gave written informed consent before participation.

IN and TM designed the study. IN performed the experiment and analyzed the data. IN, SK, TM interpreted results of the experiment. IN drafted the manuscript. IN and TM edited and revised the manuscript. All authors approved the final version of the manuscript.

None of the authors have potential conflicts of interest to be disclosed. We confirm that we have read the Journal's position on issues involved in ethical publication.