A novel combinational approach of microstimulation and bioluminescence imaging to study the mechanisms of action of cerebral electrical stimulation in mice

Abstract

Deep brain stimulation (DBS) is used to treat a number of neurological conditions and is currently being tested to intervene in neuropsychiatric conditions. However, a better understanding of how it works would ensure that side effects could be minimized and benefits optimized. We have thus developed a unique device to perform brain stimulation (BS) in mice and to address fundamental issues related to this methodology in the pre-clinical setting. This new microstimulator prototype was specifically designed to allow simultaneous live bioluminescence imaging of the mouse brain, allowing real time assessment of the impact of stimulation on cerebral tissue. We validated the authenticity of this tool in vivo by analysing the expression of toll-like receptor 2 (TLR2), corresponding to the microglial response, in the stimulated brain regions of TLR2-fluc-GFP transgenic mice, which we further corroborated with post-mortem analyses in these animals as well as in human brains of patients who underwent DBS to treat their Parkinson's disease. In the present study, we report on the development of the first BS device that allows for simultaneous live in vivo imaging in mice. This tool opens up a whole new range of possibilities that allow a better understanding of BS and how to optimize its effects through its use in murine models of disease.

Figure 1

Biphasic current generator: design, components and function A, circuit diagram of the biphasic current generator. B, voltage variation induced by the biphasic current generator with a 1 KΩ resistance arbitrarily selected to test the circuit.

Figure 2

In vivo BS system: from electrode design, implantation and function to impact on animal health A, components of the electrodes. The stereotactic rod serves to hold the electrode in the arm of the stereotactic frame, which allows for its implantation into the brain. Once the electrode is fixed to the skull, this rod is removed. B, schematic cross-section of the mouse brain depicting the implantation site within the primary motor cortex (M1), as well as images of the electrode implantation in a mouse. C, schematic representation of the various constituents of the in vivo BS system. D, recorded in vivo voltage variation induced by the biphasic current. E, weight fluctuation in mice did not exceed 10% and was restricted to the early post-implantation period. There was no significant weight variation detected during the stimulation protocol. Statistical analyses were performed using repeated measures ANOVA (rANOVA) followed by Tukey's post hoc tests. *P < 0.05, **P < 0.01. e, electrical field; M1, primary motor cortex; STR, striatum.

Figure 3

Cortical potential and voltage amplitude recorded at various time points during the experiment A, timeline of experimentations. B, set-up for brain voltage measurements. C, cortical potential after electrode implantation. Statistical analyses were performed using repeated measures ANOVA (rANOVA) followed by Tukey's post hoc tests. *P < 0.05. D, cortical potential within a single 5 h BS session, as recorded on day 1 and day 9 (at the beginning and end of a session). E, cortical potential after a 5 h BS session on days 3, 6 and 9 (compared to baseline). F, cortical potential on day 9 of stimulation (just prior to stimulation session), after an 18 h off stimulation period (compared to baseline). G, voltage amplitude during the postoperative recovery period (H) and during daily BS (I). Voltage amplitude was measured on the first positive current pulse. Statistical analyses were performed using a one-sample t test with a hypothetical value of zero (D–F) or rANOVA followed by Tukey's post hoc tests (H and I). *P < 0.05, **P < 0.01 and ***P < 0.001. contra St, electrode contralateral to the stimulating electrode; d, day; nSt, non-stimulating electrode; St, stimulating electrode.

Figure 4

Live bioluminescence imaging of the microglial/TLR2 response after electrode implantation and during BS A, timeline of experimentations. Note that the last scan was performed during the postoperative period (days 12–15) and served as a baseline for the BS protocol. B, examples of TLR2 signals in non-stimulated (postoperative period), stimulated (daily BS) and control (implanted without BS) mouse brains. Each row corresponds to images collected in a single mouse (numbers were arbitrarily assigned to compare the same animal on different rows). Mouse #3 did not undergo stimulation and thus illustrates the stability of the TLR2 signal during the postoperative period. Ca method of quantification of the TLR2 bioluminescence signal. The photon emission was measured in two distinct regions (delineated by black circles) corresponding to the areas of implantation of the electrodes. The white arrow points to the constitutive TLR2 signal detected at baseline in the olfactory bulb. Graphs represent the mean ± SEM of the total flux of bioluminescence. During the post-implantation period (Cb) electrodes were compiled separately (left vs. right; n = 12 for 6 mice). The TLR2 signal was most robust on days 1–3 and stable but significantly lower from day 6 to days 12–15. BS generated a significant increase in the TLR2 signal on day 3 (Cb; see also B). TLR2 expression was constant and similar to baseline with the non-stimulating electrode (Cc). Statistical analyses were performed using rANOVA followed by Tukey's post hoc test (Cb) or a Dunnett's post hoc (Cc). *P < 0.05, **P < 0.01. bsl, baseline; d, day.

Figure 5

3D reconstruction of the TLR2 bioluminescent signal Representative 3D reconstructions of a single mouse (chosen from six imaged mice and different from those illustrated in Fig. 4) of the TLR2 signal 1 day post-implantation and 8 days after the initiation of the BS protocol. At day 1 post-implantation, two very localized and specific TLR2 signals are visible at the site of the electrode (arrows). The TLR2 signal is also observable within the region of the olfactory bulb (arrowhead), illustrating constitutive baseline TLR2 expression in this region, as described previously (Lalancette-Hebert et al. ; Drouin-Ouellet et al. ; Lalancette-Hebert et al. 2011). Similar patterns of expression are observed during the BS protocol, although a single signal is visible given the use of a unilateral stimulation regimen. Again, baseline TLR2 expression in the olfactory bulb is detectable. All coronal, saggital and transaxial planes illustrate the depth of the signal within the cranial vault and therefore clearly indicate that the signal originates from within the brain. Colour scales indicate the source intensity (photons s^–1).

Figure 6

Post-mortem histological evaluation of the microglial response in mice and humans after BS Immunodetection of the microglial response (Iba1+ cells) using nickel intensified 3,3′-diaminobenzidine tetrahydrochloride (A, C and E), as well as the autoradiographic detection of TLR2 mRNA (B). Arrowheads in (A) and (B) highlight the Iba1+ immunoreactivity and the TLR mRNA signal detected around the stimulating electrode. Arrows indicate the cavity created by the electrodes. Electrodes, drawn schematically (i.e. following an approximate scale), were superimposed on the cavities shown in (A) to further illustrate how it corresponds to the region targeted by the electric current flow (e, curved arrows) generated between the electrode poles (see also Fig. 2B). D, E, higher magnifications of the Iba1+ immunoreactivity around the electrode cavities, disclosing different patterns of microglial responses with the stimulating (E) and non-stimulating electrode (D). Isolated regions near the non-stimulating electrodes (arrowheads) depict an increased density of microglial cells that have a resting phenotype (day 1). Outside these regions, the tissue appears slightly compressed, although the density of resting microglia is similar to that seen in normal tissue (day 2) or near the electrode implantation site. Histological changes are prominent in the regions of electric flow on the stimulated side (E). This region has diffuse Iba+ staining (between arrowheads), in which there is a concentration of activated microglia (e1). Reactive microglia (e2) are observed at the boundaries of this region, showing that there is a transition between the activated microglia around the electrode and the resting microglia in the unaltered surrounding tissue. F–I, Iba1 immunofluorescence staining (green) revealed a mild microglial response, as observed by the presence of amoeboid Iba1+ cells in close vicinity to the electrode tip (subthalamic nucleus) in PD cases who received long-term DBS (F, G), similar to that observed in PD control cases (H, I). Nuclei are stained with 4′,6-diamidino-2-phenylindole (blue) in all high magnification images. Scale bars (A–C) = 1 mm; (D, E), = 200 μm; (F) = 100 μm (low magnification image) and 25 μm (high magnification panels). The same scales were used for the remaining panels.