The effect of nucleus basalis magnocellularis deep brain stimulation on memory function in a rat model of dementia

All experiments were conducted according to the guidelines of the Institutional Animal Care and Use Committee of Yonsei University and Institute of Laboratory Anima Resources commission on Life Sciences National Research Council, USA. This study received approval of Institutional Animal Care and Use Committee (IACUC number: 2014-0206). Every effort was made to minimize animal suffering and to reduce the number of rats used. Twenty-five male Sprague–Dawley rats were used that weighed 190 − 210 g (6 week old) at the beginning of the studies. They were housed in clear plastic cages in a temperature- and humidity-controlled room with a 12-h light/12-h dark cycle. Rats were randomly assigned to one of four experimental groups before surgery. The normal group (n = 6) underwent no surgical procedures. The lesion group (n = 7) had intraventricular administration of 192 IgG-saporin. The implantation group (n = 7) and stimulation group (n = 5) had intraventricular administration of 192 IgG-saporin and implantation of an electrode in the NBM. The stimulation group had unilateral electrical stimulation.

Rats in the three experimental groups, but not the normal group, were anesthetized with a mixture of ketamine (75 mg/kg), acepromazine (0.75 mg/kg), and xylazine (4 mg/kg), and secured in a stereotaxic frame. Eight microliters (0.63 μg/μl) of 192 IgG-saporin (Chemicon, Temecula, CA, USA) were bilaterally injected into the lateral ventricle following coordinates relating to Bregma (AP - 0.8 mm, ML ± 1.2 mm, DV - 3.4 mm). The solutions were delivered at a rate of 1 μl/min and diffused for 5 min after each injection.

Right after the administration of 192 IgG-saporin, an electrode (SNEX-100, contact diameter 0.1 mm, shaft diameter 0.25 mm, impedance 0.7–1.5 MΩ; David Kopf Instruments, Tujunga, CA, USA) was inserted into the right NBM following coordinates relating to Bregma (AP -1.32 mm, ML + 2.8 mm, DV -7.4 mm) of the rats in the implantation and stimulation group. Dental cement (Long Dental Manufacturing, Wheeling, IL, USA) was used to firmly fix the electrode in place to the skull surface.

The stimuli were bipolar and electrical stimuli (120 Hz, 90 μs, 1 V) delivered via a Grass A-300 pulsemaster (Grass Instrument, Quincy, MA, USA). Stimulation parameters were monitored in real time at the beginning and end of stimulation using an oscilloscope (HDS 1022 M, Owon, Korea). Rats were unilaterally stimulated daily beginning one week after surgery until the start of behavioral testing (1 h per day, 1 week in total).

The MWM test began 2 weeks after surgery. The test consisted of 5 consecutive days of training trials and 1 day of a probe trial. A circular dark water maze tank (2 m diameter, 50 cm deep) located in a dimly lit room, was filled to a depth of 40 cm with water (maintained at 25 °C). A circular escape platform (15 cm diameter) was submerged 2 cm below the water surface. In the training trials, rats were given four acquisition trials per day with four different starting positions that were equally distributed around the perimeter of the maze. The test required the rats to swim to the hidden platform guided by distal spatial cues. Throughout acquisition, the platform was located in a fixed position in the center of the southwest quadrant of the pool. For each trial, rats were released randomly from each of the four cardinal points from the edge of the pool and gently released facing the wall. Rats were given a maximum of 60 s to find the platform. After finding the platform, rats were allowed to remain on the platform for 10 s and placed in their home cages. Animals failing to find the platform in 60 s were guided by an experimenter, then placed on the platform, and allowed to rest for 10 s. After completion of training, rats were returned to their home cages for 2 days until the probe trial. The probe trial consisted of a 60 s free swim period without a platform, during which the time spent in the target quadrant and the number of platform crossings was recorded.

After behavioral testing, three out of six rats from the normal group, three out of the seven rats from the lesion group, three out of seven rats from the implantation group, two out of five rats from the simulation group were perfused with cold 4 % paraformaldehyde in phosphate buffer saline (pH 7.2). Their brains were removed, post-fixed, and transferred to 30 % sucrose. The brains were cut into 30 μm coronal sections using a freezing microtome. The cryoprotection solution consisted of 0.1 M phosphate buffer (pH 7.2), 30 % sucrose, 1 % polyvinylpyrrolidone, and 30 % ethylene glycol. Tissue was stained with cresyl violet to confirm correct placement of the needle tracks. To detect cholinergic cells, tissue was immunohistochemically processed using polyclonal antibodies against choline acetyltransferase (ChAT; 1:100; cat# AB144P, Chemicon). The sections were stained using the avidin-biotin complex method (Vector Labs, Burlingame, CA, USA) with diaminobenzidinetetrahydrochloride as the substrate. Also fluorescence immunohistochemistry was performed with ChAT primary antibody (1:50; cat# AB144P, Chemicon) and Texas Red (1:400; cat# ab6883, abcam). Anatomical landmarks from a stereotaxic atlas [24] were used to localize the medial septum (MS).

Three out of six rats from the normal group, four out of the seven rats from the lesion group, four out of seven rats from the implantation group, three out of five rats from the simulation group were decapitated with a guillotine, and their brains were quickly removed to acquire protein for AChE assay and Western blot. The medial prefrontal cortex and hippocampus were dissected with fine forceps from 1 mm thick coronal brain slices. The samples were homogenized in lysis buffer (Intron, Seongnam, Korea) and centrifuged for 10 min at 12,000 rpm, and the protein in the supernatant was measured using the bicinchoninic acid protein assay reagent kit (Pierce, Rockford, IL, USA). The enzymatic activity of AChE was determined using the method of Ellman et al. [25] with some modifications. Twenty microliter triplicate samples identical to those used in western blot analyses were mixed with a reaction mixture [0.2 mM dithiobisnitrobenzoic acid (Sigma-Aldrich, St. Louis, MO, USA), 0.56 mM acetylthiocholine iodide (Sigma-Aldrich), 10 μM tetraisopropylpyrophosphoramide (Sigma-Aldrich), 39 mM phosphate buffer; pH 7.2] at 37 °C. After 30 min, the optical density was measured at 405 nm.

To acquire western blot data, proteins were separated using a 10 − 15 % sodium dodecyl sulfate-polyacrylamide gel and transferred onto polyvinylidene fluoride membranes. The membranes were incubated with blocking buffer [5 % nonfat dry milk in phosphate-buffered saline containing 0.05 % Tween-20 (PBST)] for 1 h at room temperature. They were then incubated with the indicated antibodies overnight at 4 °C and washed three times with PBST. The membranes were incubated with corresponding secondary antibodies for 1 h at room temperature. After washing with PBST, proteins were detected with enhanced chemiluminescence solution (Pierce) and LAS-4000 (Fujifilm, Tokyo, Japan). The intensity of each band was determined using an analysis system (Multi Gauge version 3.0, Fujifilm, Tokyo, Japan). The membranes were incubated with antibodies to glutamate acid decarboxylase 65/67 (GAD 65&67) (1:500; Millipore, Temecula, CA, USA), glutamate transporter (GT) (1:4000; Abcam, Cambridge, UK), and β-actin (1:1000; Sigma-Aldrich).

The results of all experiments were expressed as a percentage of the values for the normal group. The results of the western blots were normalized to β-actin for each sample and expressed as a percentage of the normal values. One-way ANOVA was used for overall analysis of experiments. ANOVA followed by least significant difference was used as a post hoc test at each time point for statistical analysis. The p values less than 0.05 were considered statistically significant. All statistical analyses were performed by PASW (version 18; SPSS Inc., Chicago, IL, USA).

Figure 1a and b shows the location of MS in the atlas [24] and in a brain slice stained with cresyl violet. The 192 IgG-saporin injections produced denervation of ChAT-immunopositive neurons in the MS (Fig. 1). In normal rats (C), the ChAT-immunopositive neurons were evenly distributed in the MS. By contrast, lesion (D), implantation (E), and stimulation (F) groups showed a remarkable decrease. The number of ChAT-immunopositive cells of the normal group was 125 ± 10.11. However, it was significantly decreased by administration of 192 IgG-saporin (lesion group, 18.3 ± 3.3; implantation group, 25 ± 4.5; stimulation group, 28 ± 5.6). In addition, 192 IgG-saporin damaged ChAT-immunopositive cells in the NBM (Fig. 2). The normal group (A) averagely had 47 ± 6.2 ChAT positive cells whereas 192 IgG-saporin injection groups (lesion, B; implantation, C; stimulation, D) remarkably decreased ChAT positive cells (lesion group, 26 ± 4.0; implantation group, 25.5 ± 6.4; stimulation group, 21.1 ± 6.8).

We confirmed the location of the inserted electrode in the NBM using cresyl violet staining. Figure 3a shows the location of NBM in a atlas. And also, Fig. 3b and c shows the location of NBM in the implantation and stimulation group, respectively. Although the electrodes were very close to the pallidum and internal capsule, there were no side effects such as seizure, abnormal behavior when we implant the electrodes and stimulate the nucleus basalis.

On the last day of training trials, all groups found the platform in 20 s, which was evidence of learning the platform location (Fig. 4a). During the probe trial, all groups showed similar motor-related behaviors, evidenced by equivalent swim distances and speeds. These findings suggest no effect of cholinergic lesion, electrode implantation, or electrical stimulation on motor function (Fig. 4b). Compared to the normal group, the lesion (no implantation) group showed less time in the target quadrant, less time in the platform zone, and fewer platform crossings (*p < 0.05). The stimulation group showed similar performance as compared with the normal group in terms of time spent in the target quadrant and number of platform crossings. Notably, there were statistically significant differences between the implantation and stimulation group in time on the platform zones and platform crossings (+p < 0.05).

There were no significant differences of AChE activity in the mPFC among groups (Fig. 5a). In addition, there were no differences among groups in AChE activity in the hippocampus, except for the normal group (Fig. 5b).

In the mPFC, expression of GAD 65&67 in the lesion and lesion-implantation groups was decreased compared with the normal group. However, the stimulation group showed a similar level as the normal group (Fig. 6a). In the hippocampus, the expression level of GAD 65/67 was decreased to 25.32 % in animals receiving DBS. However, there were no significant differences of GAD65&67 expressions among groups in the hippocampus (Fig. 6c). The lesion and implantation groups showed a significant increase in expression of GT both in the mPFC and hippocampus (Fig. 6b and d) (*p < 0.05). However, the stimulation group showed a similar GT expression as compared to the normal group in the both regions. And also, the stimulation group showed notably decreased expression of GT both mPFC and hippocampus as compared with the implantation group (Fig. 6b and d) (+p < 0.05).