Adult Male Sprague-Dawley rats (Experimental Animal Center of Xi’an Jiaotong University, Xi’an, China) weighing 270–320 g were used in this study (n = 130 rats). Room temperature was kept constant at 22 ± 2°C and the light/dark cycle was 12 h/12 h. Rats had free access to water and food. All experimental procedures were conducted according to the National Institute of Health Guide for the Care and Use of Laboratory Animals (NIH Publications No. 8,023, revised 1978), and approved by the Ethics Committee for Animal Experimentation of Xi’an Jiaotong University. All efforts were made to minimize the number of animals used and their suffering.

Desipramine hydrochloride, 6-OHDA hydrochloride, and apomorphine hydrochloride were purchased from Sigma–Aldrich (Sigma–Aldrich, St. Louis, MO, USA). 6-OHDA was dissolved in artificial cerebrospinal fluid containing 0.02% ascorbic acid; apomorphine was prepared in saline with 0.02% ascorbic acid; desipramine was dissolved in saline. These drugs were prepared on the day of use.

Figure 1 shows the timeline of the experiments. Four weeks after 6-OHDA injection, the lesioned rats which passed the apomorphine-induced contralateral rotation test were divided into three groups according to the number of pulses received: sham-iTBS, 1 block-iTBS (one iTBS block) and 3 block-iTBS (three blocks of iTBS). All three groups were subjected to behavioral tests, histology and immunohistochemistry at 30 or 80 min post-stimulation. In the specific case of behavioral tests including the object-place recognition (OPR) and hole-board (HB) test, the first trial of the two tests started at 30 or 80 min after real (different blocks) or sham-iTBS. Besides, hippocampal local field potential (LFP) was recorded in all three groups for 3 h, including 1 h pre-stimulation baseline and 2 h post-stimulation recordings 4 weeks after 6-OHDA injection.

The unilateral 6-OHDA lesions of the MFB in rats were carried out as previously described (Carvalho et al., 2013). Rats were anesthetized with sodium pentobarbital (40 mg/kg, i.p.) followed by the injection of desipramine (25 mg/kg, i.p.) to protect noradrenergic neurons. Then the rats were placed on a stereotaxic frame (SN-2N; Narishige, Tokyo, Japan) and 6-OHDA (12 μg/4 μl) was injected into the left MFB (AP –4.4 mm, ML –1.2 mm, DV –7.8 mm; Paxinos and Watson, 1998). After the injection, the glass pipette was left in place for 5 min to allow diffusion. Two weeks later, rats were subcutaneously injected with apomorphine (0.05 mg/kg, s.c.) to verify the effectiveness of 6-OHDA lesion (Wang et al., 2009) and those showing more than 20 contralateral turns per 5 min were selected for further experiments. All rats used in the following experiments turned consistently toward right of >30 turns per 5 min.

Four weeks after 6-OHDA injection, the lesioned rats were randomized into following groups (n = 21-22 rats/group): 30 min post sham-iTBS, 80 min post sham-iTBS, 30 min post 1 block-iTBS, 80 min post 1 block-iTBS, 30 min post 3 block-iTBS, and 80 min post 3 block-iTBS, and then received different iTBS sessions in waking relaxation state after a familiarization session. Briefly, for familiarization, the rats were gently handled, placed in a custom-made restraint device with head fixed and eyes covered, and trained to adapt to artifact noise and scalp sensation induced by iTBS daily for 3 days. During the stimulation period, different blocks of iTBS were applied using a Rui Chi magnetic stimulator (CCY–IA, YIRD, Wuhan, China) and a circular coil (inner diameter, 2.5 cm; outer diameter, 6.4 cm; YIRD, Wuhan, China) specific for rats. The coil was held parallel to the cranium with the center located 5 mm above the intersection of the sagittal suture and interaural line (Labedi et al., 2014). 1 block-iTBS consisted of 10 trains of 10 bursts (3 pulses, 50 Hz) repeated at 5 Hz with a train interval of 8 s (300 pulses; Huang et al., 2005). 3 block-iTBS were applied at 15-min intervals between each block (900 pulses). Stimulus intensity was set to 30% of maximal stimulator output. In this procedure, the stimulation conditions were almost identical, and the stimulation intensity was about 80% of motor threshold. For sham stimulation, we tilted the coil perpendicular (90°rotation) to the cranium and located the coil edge 20 cm apart from the scalp.

The behavioral tests were conducted 4 weeks after 6-OHDA injection. All the tests were performed in an isolated room between 9:00 and 11:00 a.m., and recorded using a digital camera (HR–550E, Sony, Tokyo, Japan).

The hippocampal LFP recordings were conducted 4 weeks after 6-OHDA injection. As previously described (Piilgaard and Lauritzen, 2009; Fordsmann et al., 2013; Li et al., 2014), Glass microelectrode (2–4 MΩ) filled with 1% pontamine sky blue in 0.5 M sodium acetate was placed in the left hippocampus (AP –5 mm, ML –3 mm, DV –3.2 mm; Paxinos and Watson, 1998). A screw placed on the occipital crest served as reference. The LFP signals were bandpass-filtered using a preamplifier (1–100 Hz) and digitized by the Spike 2 analysis system with sampling rate 100 Hz. After a 1 h stable baseline recording, the rats received sham-iTBS (n = 11 rats), 1 block-iTBS (n = 15 rats) or 3 block-iTBS (n = 11 rats), and then LFP was recorded for 2 h. At the end of the experiments, the LFP recording site was marked by iontophoretic ejection of pontamine sky blue through the microelectrode (−20 μA, 20 min).

After completing different iTBS sessions described above, the rats were returned to their home cages for 30 or 80 min. Then, the rats were sacrificed and transcardially perfused with phosphate buffered saline (PBS) followed by 4% paraformaldehyde at 30 or 80 min post-stimulation. Brains were removed intact and post-fixed in the same fixative for 4 h, and immersed in a solution of 30% sucrose for 3 days. The brains were cut coronally into 35 μm thickness sections at the substantia nigra pars compacta (SNc), hippocampus and medial septum (MS). To determine the extent of dopaminergic neuron loss in the SNc and ventral tegmental area (VTA) in 6-OHDA-lesioned rats, tyrosine hydroxylase (TH) immunohistochemistry of the SNc and VTA was carried out. The sections of the SNc and VTA were washed three times in PBS for 10 min each and pre-incubated with 0.3% H₂O₂ for 30 min at room temperature. Then the sections were washed three times in PBS and incubated with 10% goat serum and 0.3% Triton X-100 in PBS for 1 h at room temperature. The sections were then incubated with rabbit anti-TH antibody (1:500; Abcam, Cambridge, UK) for 24 h at 4°C. After several washes, the sections were incubated with biotinylated goat anti-rabbit secondary antibody (1:4; ZSGB-Bio, China) for 1 h at 37°C. Sections were then washed again, treated with Horseradish Peroxidase conjugated streptavidin (1:4; ZSGB-Bio, China) for 1 h at 37°C, washed with PBS and incubated with 3, 3′-diaminobenzidine at room temperature for 10 s-1 min. Sections were mounted onto gelatin-coated slides, dehydrated, cleared in xylene and coverslipped. Only rats with a total loss of TH immunoreactivity in the left SNc were used to analyze behavioral data, electrophysiological recordings and immunofluorescence histochemistry.

To verify the location of the LFP recording site and identify basic neuronal structures of the hippocampus, the sections were stained with cresyl violet. To examine the density of c-Fos- and parvalbumin (PV)-positive neurons in the hippocampus, c-Fos and PV immunohistochemistry of the hippocampus were conducted in the six groups of rats (n = 6-7 rats/group). Further, we observed the density of PV-positive neurons in the MS (n = 4 rats/group), which is reciprocally connected with the hippocampus and thought to be the generator of theta rhythm. PV immunohistochemistry was applied in the MS and cresyl violet staining was carried out to identify basic neuronal structures of the MS. For c-Fos and PV immunohistochemistry, the serial sections of the hippocampus and MS were washed in PBS, and pre-incubated with 10% goat serum and 0.3% Triton X-100 in PBS for 1 h at room temperature. The sections were then incubated for 24 h at 4°C with mouse anti-c-Fos antibody (1:1,000; Abcam, Cambridge, UK) or mouse anti-PV antibody (1:500; Abcam, Cambridge, UK). After several washes, the sections were incubated with goat anti-mouse Alexa 555 (1:1,000; Millipore) secondary antibody for 4 h at room temperature. After staining, the sections were mounted in mounting medium containing DAPI (Beyotime Biotech, Shanghai, China). In order to verify the specificity of the staining, we did parallel control experiments where the primary antibodies for c-Fos and PV marker were omitted, respectively.

The power spectrum of LFP was computed with fast Fourier transform in a range of 0–50 Hz with 0.4 Hz of resolution (Villette et al., 2010). Theta power was calculated in 3–6 Hz. For quantitative analysis, peak theta frequency was considered to be the frequency corresponding to peak power in theta band, and normalized theta power was calculated with following formula: the power in the theta frequency band divided by the total power in the entire frequency spectrum (1–50 Hz) (Villette et al., 2010). Changes in peak theta frequency and normalized theta power were analyzed per 10 min before and after real or sham-iTBS using Spike 2 analysis system. The baseline value of peak theta frequency and normalized theta power were determined as the average value recorded in six 10-min epochs pre-stimulation, and then compared to the value in each 10-min epoch post-stimulation.

After TH staining, counting of the TH-immunoreactive (TH-ir) neurons in the SNc and VTA was carried out on representative three sections per rat as previously described (Wang et al., 2009). The number of c-Fos- and PV-labeled neurons for each rat brain was calculated from the average of the numbers from five serial sections corresponding to the major part of the dorsal hippocampus and MS. The digital images were captured from the sections at 10 × and 20 × magnification. The density of c-Fos- and PV-positive neurons (the number of c-Fos- and PV-positive neurons per mm2) was calculated in the left hippocampal and MS area.

Statistical analyses were performed with the Sigmaplot 14.0 version for Windows. The data of TH-ir neuron counts and the head-dippings in the HB test were compared using a paired Student’s t-test. The discrimination ratio in the OPR task and all the immunofluorescent data were analyzed by two-way analysis of variance (ANOVA) followed by post hoc comparisons based on Bonferroni’s test. One-way repeated ANOVA was used to analyze the hippocampal LFP data. Data were expressed as the mean ± SEM in the study. A two-tailed P-value of < 0.05 (α = 0.05) was considered significant.

The loss of dopaminergic neurons is a cardinal feature of PD. Therefore, we observed the loss of dopaminergic neurons by counting the number of TH-ir neurons in the SNc and VTA 4 weeks after the injection of 6-OHDA into the MFB (Figure 2A). The SNc on the lesioned side showed a total loss of TH-ir neurons compared to the unlesioned side (Figures 2B, C; paired t-test, p < 0.001), and the number of TH-ir neurons in the VTA on the lesioned side was reduced by 80.13 ± 6.05% of those on the unlesioned side (Figures 2B, C; paired t-test, p < 0.001).

The OPR task was carried out to assess hippocampus-dependent memory after application of sham or real iTBS. A higher discrimination ratio of the OPR task reflects greater memory retention for the stationary object place. Figure 3A shows schematic illustrations of the OPR task. As shown in Figures 3B–D, the rats tended to explore displaced object more time than the stationary one at 80 min after receiving 3 block-iTBS, but not sham-iTBS or 1 block-iTBS. A two-way ANOVA analysis (stimulation × time) showed a significant effect of stimulation on discrimination ratio for the treatment [F(2,41) = 5.07, n = 7 rats/group, p < 0.05], for the time [F(1,41) = 5.57, n = 7 rats/group, p < 0.05], and for their interaction [F(2,41) = 6.50, n = 7 rats/group, p < 0.01]. For between-group comparisons, subsequent post hoc analysis showed that 3 block-iTBS significantly increased the discrimination ratio at 80 min post-stimulation compared to sham-iTBS (Figure 3E; p < 0.001). 3 block-iTBS tended to decrease the discrimination ratio at 30 min post-stimulation compared to sham-iTBS, but this did not reach statistical significance (Figure 3E). 1 block-iTBS didn’t change the discrimination ratio at 30 or 80 min post-stimulation compared to sham-iTBS (Figure 3E). Within-group comparison between different time points showed a significant increase of the discrimination ratio at 80 min compared to 30 min after 3 block-iTBS (Figure 3E; p < 0.001). There were no significant differences in the discrimination ratio between 30 and 80 min post sham or 1 block-iTBS (Figure 3E).

The HB test was also performed to evaluate hippocampus-dependent memory after sham or real iTBS. A reduction of head-dippings into the holes indicates better recognition of novel context feature. Figure 4A shows schematic illustrations of the HB test. Paired t-test showed 3 block-iTBS significantly decreased the number of head-dippings in the second trial compared to the first trail only at 80 min post-stimulation (Figure 4D; n = 8 rats/group, p < 0.01), not at 30 min (Figure 4D). Sham or 1 block-iTBS did not produce significant alterations in head-dipping behavior when the rats re-explored the hole-board at any time points (Figures 4B, C).

In the present study, the LFP recording site used was verified to be within the hippocampus by iontophoretic ejection of pontamine sky blue (Figure 5A). Similar to the results in the behavioral tests, different blocks of iTBS also produced different effects on hippocampal theta rhythm. Sham-iTBS or 1 block-iTBS didn’t affect peak theta frequency or normalized theta power during a 2 h period post-stimulation compared to the pre-stimulation baseline (Figures 5B–E, 6). Surprisingly, 3 block-iTBS elicited a time-dependent effect on hippocampal theta rhythm [Figures 5B–E; F(12,141) = 4.93, n = 11 rats/group, p < 0.001]. It significantly decreased the normalized power of hippocampal theta rhythm after 10 min, and this inhibitory effect lasted for 30 min (Figure 5E; p < 0.05). Furthermore, it caused an increase in normalized theta power, which successively reached statistical significance at 50 min post-stimulation, and the excitatory effect lasted for 50 min (Figure 5E; p < 0.05). 3 block-iTBS didn’t change peak theta frequency during a 2 h post-stimulation compared to the baseline value (Figure 6).

The density of c-Fos-positive neurons in the dorsal hippocampus was calculated after rats receiving sham or real iTBS. Representative sections of hippocampal immunofluorescent staining for c-Fos and DAPI are shown in Figure 7. A two-way ANOVA analysis (stimulation × time) showed a significant difference on the density of c-Fos-positive neurons in the dorsal hippocampus induced by iTBS for the treatment [F(2,38) = 2.80, n = 6-7 rats/group, p < 0.05], for the time [F(1,38) = 5.60, n = 6-7 rats/group, p < 0.05], but not for their interaction. Compared to sham-iTBS, 1 block-iTBS did not alter the density of c-Fos-positive neurons in the dorsal hippocampus regardless of whether 30 or 80 min post-stimulation (Figure 8A). Interestingly, changes in the density of c-Fos-positive neurons in the dorsal hippocampus induced by 3 block-iTBS showed a progressive increase over time compared to sham-iTBS. The mean density of c-Fos-positive neurons was still unchanged at 30 min post-stimulation compared to sham-iTBS (Figure 8A). At 80 min, the increase in the density of c-Fos-positive neurons induced by 3 block-iTBS was statistically significant (Figure 8A; p < 0.01).

Furthermore, the density of PV-positive neurons in the dorsal hippocampus and MS were calculated after rats receiving sham or real iTBS. Representative sections of immunofluorescent staining for PV and DAPI in the dorsal hippocampus and MS are shown in Figures 9, 10. Unexpectedly, 3 block-iTBS did not alter the mean density of PV-positive neurons in the dorsal hippocampus (Figure 8B), but changed the mean density of those neurons in the MS (Figure 8C). A two-way ANOVA analysis (stimulation × time) showed a significant difference on the mean density of PV-positive neurons in the MS for the treatment [F(2,23) = 8.90, n = 4 rats/group, p < 0.01], for the time [F(1,23) = 9.95, n = 4 rats/group, p < 0.01] and for their interaction [F(2,23) = 10.09, n = 4 rats/group, p < 0.01]. Post hoc analysis showed that the mean density of PV-positive neurons in the MS was significantly lower at 30 min following 3 block-iTBS compared to sham-iTBS (Figure 8C; p < 0.001). However, 3 block-iTBS didn’t induce any change in the mean density of PV-positive neurons in the MS at 80 min post-stimulation compared to sham-iTBS (Figure 8C). Within-group comparison between different time points showed a significant increase in the mean density of PV-positive neurons in the MS at 80 min compared to 30 min after 3 block-iTBS (Figure 8C; p < 0.001). 1 block-iTBS did not change the density of PV-positive neurons in the dorsal hippocampus or MS when compared to sham-iTBS (Figure 8C).