Online LI-rTMS during a Visual Learning Task: Differential Impacts on Visual Circuit and Behavioral Plasticity in Adult Ephrin-A2A5^–/–Mice

Abstract

Repetitive transcranial magnetic stimulation (rTMS) induces plasticity in normal and abnormal neural circuitries, an effect that may be influenced by intrinsic brain activity during treatment. Here, we study potential synergistic effects between low-intensity rTMS (LI-rTMS) and concurrent neural activity in promoting circuit reorganization and enhancing visual behavior. We used ephrin-A2A5–/– mice, which are known to possess visuotopic mapping errors that are ameliorated by LI-rTMS, and assessed the impact of stimulation when mice were engaged in a visual learning task. A detachable coil was affixed to each mouse, and animals underwent 2 wk of 10-min daily training in a two-choice visual discrimination task with concurrent LI-rTMS or sham stimulation. No-task controls (+LI-rTMS/sham) were placed in the task arena without visual task training. At the end of the experiment, visuomotor tracking behavior was assessed, and corticotectal and geniculocortical pathway organization was mapped by injections of fluorescent tracers into the primary visual cortex. Consistent with previous results, LI-rTMS alone improved geniculocortical and corticotectal topography, but combining LI-rTMS with the visual learning task prevented beneficial corticotectal reorganization and had no additional effect on geniculocortical topography or visuomotor tracking performance. Unexpectedly, there was a significant increase in the total number of trials completed by task + LI-rTMS mice in the visual learning task. Comparison with wild-type mice revealed that ephrin-A2A5–/– mice had reduced accuracy and response rates, suggesting a goal-directed behavioral deficit, which was improved by LI-rTMS. Our results suggest that concurrent brain activity during behavior interacts with LI-rTMS, altering behavior and different visual circuits in an abnormal system.

Manipulating the Level of Sensorimotor Stimulation during LI-rTMS Can Improve Visual Circuit Reorganisation in Adult Ephrin-A2A5-/- Mice

“…Importantly, there was also a nonsignificant relationship in mice treated with online LI-rTMS in a dark environment ( F (1,13) = 4.026, p = 0.066; Figure 1 C, bottom left). This finding is in contrast to our previous results showing a significant spatial relationship between the injection site and TZ location in ephrin-A2A5 -/- mice treated with online LI-rTMS under normal light conditions [ 23 ]. Finally, we observed a moderate and significant relationship between the injection site and TZ locations in LI-rTMS treated ephrin-A2A5 -/- mice that had free access to a running wheel during every stimulation session ( F (1,15) = 10.70, p = 0.005; Figure 1 C, bottom right panel).…”

“…Furthermore, LI-rTMS with concurrent locomotion also did not increase the number of head tracking responses as compared to sham (locomotion+LI-rTMS: 2.56 ± 0.39; locomotion+sham: 2.12 ± 0.39; U = 40, p = 0.481). These results are in contrast to previous findings, where we have shown that LI-rTMS alone or in combination with a visually engaging task—both in normal light conditions—improves head tracking behaviour [ 21 , 23 ].…”

“…To examine the potential reorganisation in the corticotectal circuit, we performed a linear regression analysis to determine whether there was a relationship between V1 injection location in the lateromedial axis and TZ location in the rostrocaudal axis. As expected, our results showed a very weak and nonsignificant linear relationship in sham-treated ephrin-A2A5 -/- mice in both groups (dark+sham: F (1,21) = 1.906, p = 0.182; locomotion+sham: F (1,13) = 2.354, p = 0.149; Figure 1 C, top panels), consistent with previous results showing a lack of corticotectal topography in this strain [ 17 , 22 , 23 ]. Importantly, there was also a nonsignificant relationship in mice treated with online LI-rTMS in a dark environment ( F (1,13) = 4.026, p = 0.066; Figure 1 C, bottom left).…”

“…Despite the nonsignificant main effect of interaction between environment and treatment group for all dLGN measures (convex-hull volumes: F (1, 34) = 2.042, p = 0.162, observed power = 0.284; cluster area: F (1, 34) = 2.963, p = 0.094, observed power = 0.387; number of labelled neurons: F (1, 34) = 3.307, p = 0.078, observed power = 0.424), we were interested in knowing whether chronic online LI-rTMS alters the abnormal dispersion of labelled dLGN neurons as compared to sham in their respective groups, a finding that has been observed in previous studies [ 22 , 23 ]. The pairwise comparisons with Sidak correction revealed a significant reduction in convex-hull volumes by chronic LI-rTMS in both dark ( p < 0.001) and locomotion groups ( p = 0.034).…”

“…Previously, we showed that increasing the salience of visual information by training ephrin-A2A5 -/- mice to perform a visually engaging task during LI-rTMS prevented the beneficial reorganisation of the corticotectal map as compared to no-task subjects [ 23 ]. A possible explanation is that the combination of visually evoked activity, subthreshold stimulation induced by LI-rTMS [ 24 ] and the modulation of cortical activity by other factors including attention, motivation, and locomotion during the task [ 25 , 26 ], may have non-selectively strengthened both ‘normal’ and ‘ectopic’ projections within the corticotectal pathway [ 23 ].…”