Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Mar 2.
Published in final edited form as: Epilepsia. 2022 Dec 11;64(2):524–535. doi: 10.1111/epi.17469

Optogenetic activation of the superior colliculus attenuates spontaneous seizures in the pilocarpine model of temporal lobe epilepsy

Safwan K Hyder 1, Anjik Ghosh 1, Patrick A Forcelli 1,2,3,^
PMCID: PMC10907897  NIHMSID: NIHMS1959391  PMID: 36448878

Abstract

Objective:

Decades of studies have indicated that activation of the deep and intermediate layers of the superior colliculus can suppress seizures in a wide range of experimental models of epilepsy. However, prior studies have not examined efficacy against spontaneous limbic seizures. The present study aimed to address this gap through chronic optogenetic activation of the superior colliculus in the pilocarpine model of temporal lobe epilepsy.

Methods:

Sprague-Dawley rats underwent pilocarpine-induced status epilepticus and were maintained until the onset of spontaneous seizures. Virus coding for channelrhodopsin-2 was injected into the deep and intermediate layers of the superior colliculus, and animals were implanted with head-mounted LEDs at the same site. Rats were stimulated with either 5 or 100 Hz light delivery. Seizure number, seizure duration, 24-hour seizure burden, and behavioral seizure severity were monitored.

Results:

Both 5 and 100 Hz optogenetic stimulation of the deep and intermediate layers of the superior colliculus reduced daily seizure number and total seizure burden in all animals in the active vector group. Stimulation did not affect either seizure duration or behavioral seizure severity. Stimulation was without effect in opsin-negative control animals.

Conclusion:

Activation of the deep and intermediate layers of the superior colliculus reduces both the number of seizures and total daily seizure burden in the pilocarpine model of temporal lobe epilepsy. These novel data demonstrating an effect against chronic experimental seizures complement a long history of studies documenting the anti-seizure efficacy of superior colliculus activation in a range of acute seizure models.

Keywords: basal ganglia, deep brain stimulation, neuromodulation, channelrhodopsin

Introduction

Deep Brain Stimulation (DBS) continues to be a promising treatment for medically-intractable epilepsy. One approach is to stimulate seizure onset zones, e.g., hippocampal DBS. However, circuit-based strategies may offer advantages, as seizure onset zones may be multifocal or diffuse. The most common target for DBS in epilepsy is the thalamus, including the anterior nucleus of the thalamus (1) and the centromedian nucleus (2). The anterior nucleus of the thalamus is compelling based on its connectivity with the hippocampus, and the centromedian nucleus, for its suggested role in seizure generalization. However, targeting structures that are remote to areas of seizure onset and their generalization is a continuing area of investigation.

The basal ganglia – and their output targets – may represent such a circuit-based strategy (3,4). Since the late 1980s, evidence has suggested that modulating activity within the deep and intermediate layers of the superior colliculus (DLSC) can exert profound effects on seizures (5). This region, which comprises the “dorsal midbrain anticonvulsant zone” (6), was initially characterized as a crucial mediator of anticonvulsant effects evoked by focal inhibition of the substantia nigra pars reticulata (7). Accordingly, activation of the DLSC exerts powerful seizure-suppressive effects across a wide range of preclinical models. Focal pharmacological activation of the DLSC desynchronizes the cortical EEG (8–10) and exerts protection against tonic (5,6,11,12), tonic-clonic, absence (8), and limbic seizures (13).

Previous work from our lab showed that activation of the DLSC likewise suppresses seizures across a range of acutely-evoked preclinical seizure models either directly, by optogenetic activation of DLSC neurons, or indirectly, through optogenetic silencing of inhibitory projections from the substantia nigra pars reticulata (14,15). Despite its profound capacity for seizure protection, only one prior study examined the impact of superior colliculus stimulation in a model of spontaneous seizures, and this study focused on a genetic absence model (16). Thus, the effects of DLSC stimulation against spontaneous limbic seizures remain unknown.

In the present study, we tested the efficacy of optogenetic activation of the DLSC against spontaneous limbic seizures in the pilocarpine model of temporal lobe epilepsy in rats. We used an optogenetic approach to specifically target neurons of the DLSC and spare afferent terminals and fibers of passage. In our testing paradigm, individual subjects were randomized into 24-hour periods of optogenetic stimulation of the DLSC or baseline (no stimulation). In all animals with the active vector, we found a suppression of seizure activity as measured by both number of seizures and cumulative daily seizure burden. These data suggest that the findings previously reported in acute seizure models hold true for spontaneous seizures and further underscore the potential of basal ganglia and midbrain targets for controlling seizures.

Methods

Experimental design

We induced status epilepticus to produce chronically epileptic animals, which we subsequently used for experiments. On a within-subject basis, we compared the effects of optogenetic activation of the DLSC (5 or 100 Hz light delivery) to a baseline session in which no light was delivered. Video-EEG data were acquired in 3-day blocks, each containing one 24-h session of baseline (no light delivery), 5 Hz stimulation, and 100 Hz stimulation. The order of these sessions was randomized in each block and balanced across animals. Animals were tested on 1 to 4 blocks of sessions (mean 2 blocks per animal). We tested both active vector animals and opsin-negative control animals to control for off-target impacts of light delivery.

Animals

Male (n=7) and female (n=5) Sprague-Dawley (SD) rats from Harlan/Envigo (Indianapolis, IN) were used for these experiments. The active vector group (see above) included 4 males and 1 females; the control group included 3 males and 3 females. Animals were randomly assigned to groups. Group sizes (n=5–6) were determined based on power analysis with effect sizes estimated from our prior studies of optogenetic activation of the SC (14). Animals were maintained in the Division of Comparative Medicine facility of the Georgetown Medical Center with full-time veterinary care. Rats were pair-housed until surgery, fed ad libitum (LabDiet #5001), and maintained in temperature (21°C) and humidity-controlled (30–70%) conditions with a 12–12 light: dark cycle. All procedures were approved by Georgetown Institutional Care and Use Committee (Protocol 2016–1184) and conducted by the Guide for the Care and Use of Laboratory Animals.

Induction of status epilepticus

Status epilepticus was induced per standard protocols. Details of status epilepticus induction can be found in the Supplementary Methods.

Surgery

An average of 135 days following pilocarpine, animals underwent stereotaxic surgery. Surgery was performed essentially as we have previously described (14,15,17). Surgical details can be found in the Supplementary Methods. rAAV8-hSyn-ChR2(H134R)-mCherry or mCherry control virus (UNC Vector Core) was injected bilaterally into the DLSC. We used the same coordinates as our prior work (14). Coordinates were based on the atlas of Pellegrino and Cushman, with the incisor bar elevated 5mm over the intra-aural line (18). LED guide cannulae (Pinnacle Technologies, Lawrence, KS) were placed bilaterally into DLSC, with each targeted 200μm above the virus injection coordinates of the respective hemisphere. Dummy cannulae were placed into the guides until the animals were started on chronic monitoring, at which point fiber optics were placed in the guides.

Chronic video-EEG monitoring

An average of 28 days following surgery (163 days following pilocarpine), which allowed time for expression of opsins, animals were connected to an EEG by a tethered preamplifier routed through a commutator to enable free motion of the animal (Pinnacle Technologies, Lawrence, KS). The tether contained a headstage with a preamplifier for EEG signal acquisition and LED power delivery. EEG signals were digitized and sampled at 1 kHz and band-pass filtered from 1–50 Hz. The simultaneous video was captured at ten fps and synchronized with the EEG using Pinnacle Sirenia Acquisition software. An infrared light source was illuminated to allow video capture during the dark cycle. The light was not visible through the implant, so chronic light delivery is not expected to have impacted the light-dark cycle.

Optogenetic stimulation parameters

Stimulation was regulated with Pinnacle Sirenia Seizure Pro software. Blue light (450nm) was delivered at 50–175 mW/mm2. LEDs were fabricated and tested by Pinnacle Technologies; each LED was powered at the maximal current that avoided thermal damage to the LED according to the manufacturer’s specification. LEDs were approximately balanced in terms of power and efficiency in each animal. Light was delivered with a 10% duty cycle: 5 Hz stimulation was delivered with pulse widths of 20 msec and 100 Hz stimulation with pulse widths of 1msec. While lower than the duty cycle we have previously used (i.e., 50%), we selected this to minimize tissue heating during long-term stimulation. While we cannot exclude occult thermal damage to the brain, we do note that tissue was preserved within the SC (demonstrated by histology in Fig 1 and the consistent expression of opsin across animals).

Figure 1. Histological verification of opsin expression and fiber optic placement.

Figure 1.

Open in a new tab

(A) Coronal section at the level of the superior colliculus. * indicates the location of the cannula and fiber optic. Yellow outlined region is shown at expanded magnification in Panel (B). Arrowheads indicate mCherry positive (ChR2-expressing) somata. (C) Coronal plane from Swanson rat brain atlas (49) with blue circles indicating the location of cannula tips. Red outlined region indicates the intermediate and deep layers of the superior colliculus. PAG – periaqueductal grey, MGB – medial geniculate body.

100 Hz light delivery was selected based on the report of Sahibzada and colleagues (19), which demonstrated behavioral responses to SC activation with this stimulation frequency. 5 Hz was chosen as it is the low end of the range of the frequency of oscillations reported in the SC (20). Moreover, we have previously reported anti-seizure effects with these stimulation frequencies in acute epilepsy models (14). While the H134R variant of channelrhodopsin-2 (ChR2) displays near-perfect spike fidelity with low stimulation rates, higher stimulation rates show 20–50% spike fidelity. Assuming even the low end of this range was achieved with our 100 Hz stimulation, we would still be driving activity substantially above 5 Hz (21–23).

EEG analysis

EEGs were analyzed by an observer blinded to the animal’s treatment status. EEG signals were bandpass filtered (1–50Hz) and manually scored (see Supplementary Methods).

For each animal, we calculated the (1) daily number of seizures, which was defined as the count of seizures during a 24 hour period, (2) The average seizure duration,which was defined as the average of the lengths of individual seizures during the 24 hour period, and (3) The 24-h seizure burden, which was the total duration of electrographic seizure activity, reflecting a combined metric of both seizure number and seizure duration.

A second observer reviewed the identified electrographic seizures and assessed the behavioral responses (see Supplementary Methods).

Histological validation

After completion of data collection, histological validation of opsin expression was performed for all animals. Details can be found in the Supplementary Methods.

Statistical analysis

The mean number of seizures, mean seizure duration, and cumulative daily seizure burden were calculated across sessions for each subject. All analysis was performed blind to animal treatment and stimulation type. The median behavioral seizure severity, the maximum behavioral seizure severity, the percentage of seizures with behavioral components (scores ≥1), and the percentage of generalized (score ≥3) seizures were also calculated across sessions.

Raw seizure scores, seizure duration, and seizure burden values are presented, as are normalized data (% of baseline, % change from baseline). Data were assessed for normality and in several cases, displayed a log-normal distribution. In these cases, log-normal data were log-transformed before analysis of variance.

Data were analyzed by a combination of one-sample t-tests and analysis of variance. For normalized data (% of baseline, % change from baseline), we performed 1-sample t-tests. For % of baseline data, the one-sample t-test was run against a test value of 100%; a test value of 100% indicates no change from baseline. For the % change from baseline data, the 1-sample t-test was run against a test value of 0; a test value of 0 indicates no change from baseline. Our a priori analysis plan, on which sample size was estimated, was entirely within-group. I.e., we compared on a within-subject basis no stimulation, 5 Hz, and 100 Hz stimulation for the active vector animals, and separately compared these treatments in controls. For raw data, we performed a one-way ANOVA with Holm-Sidak adjusted pairwise comparisons to compare the no-stimulation baseline condition to the 5 Hz and 100 Hz conditions. Figures display means and standard errors of the mean. P-values <0.05 were considered to be statistically significant.

Results

Figure 1A shows a representative section through the DLSC at the site of fiber optic placement. Labeling was present throughout the deep and intermediate layers of the superior colliculus. In some cases, labeling extended ventrolaterally into the medial geniculate body; in other cases, small amounts of labeling were evident along the needle track in the hippocampus. Off-target expression is unlikely to contribute to our findings. The fiber optic was placed ventral to the hippocampus (and light is delivered from the tip of the fiber optic) and several millimeters dorsomedial to the medial geniculate (power decreases rapidly with distance from the fiber optic). All animals displayed mCherry signal ventral to the cannula tip in both hemispheres, indicating that the long-term optical stimulation did not result in gross cell loss of ChR2-expressing neurons (Figure 1B). All animals included in the study had cannulae placed in or just superior to the intermediate layers of the superior colliculus. One animal was placed more superficially than the others (just dorsal to the intermediate layers, so presumably, light impacted the intermediate, but not deep layers of the superior colliculus). Interestingly, this case (indicated by the diamonds in Figures 4–6) displayed a smaller-than-average stimulation effect. Figure 2 shows a representative EEG recording of a spontaneous seizure that was evident across all three leads.

Figure 4. Optogenetic activation of the DLSC reduces daily number of seizures.

Figure 4.

Open in a new tab

(A) Normalized seizure count (% of baseline). Values shifted to the left of 100% indicate a reduction from baseline. Overlaid star and bars indicate the mean and standard error of the mean. Symbols indicate individual cases. (B) Normalized % change in seizure count from baseline; data re-expressed from Panel (A). (C) Mean+standard error of the mean of seizure count. (D-F) conventions as for Panels A-C, but for opsin-negative control rats. * = P < 0.05; ns = not significant. For the active vector, females are indicated by the ▽ and the ⬡ symbols. For the control vector, females are indicated by the ⬡, □, ○ symbols.

Figure 6. Optogenetic activation of the DLSC reduces 24-hour seizure burden.

Figure 6.

Open in a new tab

(A) Normalized seizure burden (% of baseline). Overlaid star and bars indicate the mean and standard error of the mean. Symbols indicate individual cases. (B) Normalized % change in seizure burden from baseline; data re-expressed from Panel (A). (C) Mean+standard error of the mean of seizure duration. (D-F) conventions as for Panels A-C, but for opsin-negative control rats. * = P < 0.05; ns = not significant. For the active vector, females are indicated by the ▽ and the ⬡ symbols. For the control vector, females are indicated by the ⬡, □, ○ symbols.

Figure 2. Representative electrographic recording of a spontaneous seizure.

Figure 2.

Open in a new tab

A single seizure in an active-vector animal. Baseline activity is shown in black, seizure activity is shown in red. LFr-Cb = Left frontal cortex referenced to cerebellum; RFr-Cb = right frontal cortex referenced to cerebellum, HPC Depth = bipolar depth electrode in hippocampus. Bottom set of traces show an expanded timescale of the recording shown in the top traces.

Representative data from two consecutive blocks of stimulation in one male and one female each from both the active vector and control group are shown in Figure 3. These data illustrate, and the group data confirm, a reduction in seizures with optogenetic stimulation, a pattern we observed in all animals in the active vector group. Note that for both blocks in A and C, the number of seizures and 24-h seizure burden were numerically reduced in the 5 and 100 Hz stimulation sessions compared to baseline. By contrast, no consistent pattern was observed in control vector animals (B and D). The response across blocks was highly consistent for the active vector condition. For 5 of the 6 animals in the active vector group, the same response was observed across all blocks (i.e., highest seizure number with baseline, lower with both frequencies of optogenetic stimulation). For the sixth animal, this pattern of reduced seizure number with optogenetic stimulation was observed in 3 of 4 blocks; in the fourth block, both frequencies of optogenetic stimulation increased the number of seizures. By contrast, in the control group, no animal displayed a consistent pattern across blocks. In some cases optogenetic stimulation was associated with increased seizures, in some cases no change in seizure number, and in some cases a decrease in seizures. Thus, our effects were reproducibile across blocks in the active vector group, consistent with treatment effects. Inconsistency of the effect of light in the controls was consistent with random variability. The average number of blocks included per animal was 2, with a range of 1 to 4 blocks depending on the duratiaon of the EEG implant.

Figure 3. Representative data in male and female animals from the active vector and control groups.

Figure 3.

Open in a new tab

Representative seizure burden (left Y axis) and number of seizures (right Y axis) across two consecutive blocks of testing in a representative (A) ChR2-expressing male, (B) opsin-negative male, (C) ChR2-expressing female, note that many of the seizures in this animal were non-convulsive despite the high seizure burden, and (D) opsin-negative female. Circles indicate cumulative seizure burden; squares indicate number of seizures.

Optogenetic activation of the superior colliculus significantly reduced the number of seizures observed during a 24-h period. This effect was evident with both 5 Hz and 100 Hz light delivery (Figure 4A-C). On a within-subject basis, we observed a mean reduction in seizure number of 24% with 5 Hz stimulation (1-sample t-test, t = 4.15, df=4, p=0.0143) and a mean reduction in seizure number of 22% with 100 Hz stimulation (1-sample t-test, t = 5.2, df=4, p=0.0065). This was also evident in the raw (not normalized to baseline) seizure counts (Figure 4C; ANOVA; F1.656, 6.625= 2.52; p=0.0066; 5 Hz vs Baseline p=0.019; 100 Hz vs Baseline p=0.0179, Holm-Sidak corrected pairwise comparisons).

By contrast, activation of the superior colliculus was without significant effects on seizure number in opsin-negative control animals (Figure 4D-F). On a within-subject basis, we observed a mean increase in seizure number of 26.7% with 5 Hz stimulation, but this did not reach the level of statistical significance (1-sample t-test, t = 0.727, df=5, p=0.50). The mean change with 100 Hz light delivery was near-zero (0.67%) and again, did not reach the level of statistical significance (1-sample t-test, t = 0.0299, df=5, p=0.977). Consistent with this, the raw seizure counts did not differ as a function of treatment (Figure 4E; ANOVA; F1.403, 7.017= 0.861; p=0.424; 5 Hz vs Baseline p=0.758; 100 Hz vs Baseline p=0.606, Holm-Sidak corrected pairwise comparisons).

Individual seizure duration did not change as a function of optogenetic stimulation of the DLSC (Figure 5). No significant changes from baseline were observed for the active vector (5Hz vs. Baseline, 1 sample t-test, t=2.033, df=4, p=0.112; 100 Hz vs Baseline, 1 sample t-test, t=0.333, df=4, p=0.756) nor for the opsin-negative control group (5Hz vs. Baseline, 1 sample t-test, t=0.870, df=5, p=0.424; 100 Hz vs Baseline, 1 sample t-test, t=0.587, df=5, p=0.583). Consistent with this, the raw mean duration also did not differ for either the active vector (F1.939, 7.757= 2.853; p=0.119) or the control vector (F1.098, 5.489= 0.432; p=0.556). As shown in Table 1, optogenetic stimulation was also without effect on various metrics of behavioral seizure activity. Median seizure score, maximum seizure score, the percentage of seizures with behavioral manifestations, and the percentage of seizures showing generalized (bilateral) behavioral signs did not differ across treatments. Thus, while the number of seizures was decreased by stimulation (Figure 4), the duration and severity of the remaining seizures (Fig 5 and Supplemental Table 1, respectively) were unaltered.

Figure 5. Optogenetic activation of the DLSC is without effect on individual seizure duration.

Figure 5.

Open in a new tab

(A) Normalized seizure duration (% of baseline). Overlaid star and bars indicate the mean and standard error of the mean. Symbols indicate individual cases. (B) Normalized % change in seizure duration from baseline; data re-expressed from Panel (A). (C) Mean+standard error of the mean of seizure duration. (D-F) conventions as for Panels A-C, but for opsin-negative control rats. ns = not significant. For the active vector, females are indicated by the ▽ and the ⬡ symbols. For the control vector, females are indicated by the ⬡, □, ○ symbols.

Finally, we assessed 24-h seizure burden (cumulative duration of seizure activity during the 24-h period; Figure 6). Optogenetic activation of the DLSC significantly reduced the 24-h seizure burden. This effect was evident with both 5 Hz and 100 Hz light delivery (Figure 6A-C). On a within-subject basis, we observed a mean reduction in seizure burden of 19% with 5 Hz stimulation (1-sample t-test, t = 5.53, df=4, p=0.0052) and a mean reduction in seizure burden of 25% with 100 Hz stimulation (1-sample t-test, t = 4.36, df=4, p=0.012). This was also evident in the raw seizure burden (Figure 6C; ANOVA; F1.42,5.69= 10.37; p=0.0157; 5 Hz vs Baseline p=0.0105; 100 Hz vs Baseline p=0.0317, Holm-Sidak corrected pairwise comparisons). Further ruling out the non-specific effects of light delivery, activation of the superior colliculus was without significant effects on 24-h seizure burden in opsin-negative control animals (Figure 6D-F), in either the normalized (5 Hz vs Baseline t=0.041, df=6, p=0.97; 100 Hz vs Baseline t=0.66, df=6, p=0.54) or raw seizure burden measures (Fig 6E; ANOVA; F1.79,8.96= 0.231; p=0.775; 5 Hz vs Baseline p=0.97; 100 Hz vs Baseline p=0.79, Holm-Sidak corrected pairwise comparisons).

We were somewhat surprised to note that across measures, the normalized seizure metrics were more variable in the control group as compared to the active vector group (see Supplementary Results, and Figs 4–6). This suggests that it is possible that the effect of optogenetic stimulation may lead to reduced variability. I.e., the decreased variability observed in the normalized metrics may reflect the treatment effect. It is noteworthy that the stimulation data were very tightly clustered for the active vector animals.

Discussion

In the present study, we found that activation of the DLSC reduces the seizure burden of rats with spontaneous limbic seizures. Despite decades of study, this is the first application of DLSC stimulation to a chronic model of temporal lobe epilepsy. We tested low-frequency (5Hz) and high-frequency (100Hz) chronic (24-hour) optogenetic activation. Stimulation was effective at both frequencies in all animals in the active vector group, reducing the daily number of seizures and cumulative daily seizure burden.

Our findings build on a long history of studies documenting the anti-seizure efficacy of superior colliculus activation in a range of acute seizure models. Pharmacological activation of the SC suppresses maximal electroshock seizures (5,6), piriform cortex (Area Tempestas)-evoked limbic seizures (13), and spike-and-wave discharges in the pentylenetetrazole and GAERS models of absence seizures (8,16). Optogenetic activation of the SC, or optogenetic inhibition of GABAergic inputs to the SC, similarly reduces Area Tempestas-evoked limbic seizures, pentylenetetrazole-evoked tonic and clonic seizures, audiogenic seizures in genetically epilepsy-prone rats, and spike-and-wave discharges in the gamma-butyrolactone model of absence epilepsy (14,15). While no prior study has evaluated the impact of SC activation in a model of spontaneous limbic seizures, our previous study reported attenuation of acutely evoked limbic seizures. The magnitude of this effect was larger than in the present study (71% vs. 27%). In the acute model, we found a suppression of behavioral seizure severity when stimulation was initiated before the onset of behavioral seizure activity. In contrast, in the present study, our manipulations did not affect behavioral seizures (Supplemental Table 1). Thus, while chronic stimulation of the DLSC displayed efficacy against seizure onset (i.e., a reduction in seizure number), it was without effect – at least with the parameters we used – on seizure propagation and generalization in the pilocarpine model.

Several differences in approach might explain the smaller effect we observed in this study. First, this may reflect a difference between models – while the semiology of seizures in the pilocarpine and Area Tempestas models is very similar (24,25), the underlying substrate of seizure initiation differs. In the Area Tempestas model, the seizure focus is clearly defined in the piriform cortex. By contrast, spontaneous seizures after status epilepticus can have multifocal onset (26–29). While the broader literature on basal ganglia networks and seizure control suggests that the site of seizure onset is not critical for anti-seizure effects, it remains possible that anti-seizure effects may be more potent for some initiation sites than for others.

Further studies comparing the efficacy of stimulation as a function of seizure onset location in the pilocarpine model may thus be informative. Second, the pilocarpine model is associated with rather severe damage (25). While valuable as a model of spontaneous seizures, the damage caused by status epilepticus in this model is more significant – and more widespread – than observed in temporal lobe epilepsy (30). While the DLSC itself does appear to be spared after SE (25), robust degeneration in the substantia nigra pars reticulata – a key regulator of DLSC function, has been reported (25,31). As the networks mediating DLSC stimulation effects are not entirely understood, the degree to which SE-associated damage impacts other circuit components is unknown. Third, and perhaps most importantly, the stimulation parameters we used in the present study differed from our prior study (14). Here, because of the chronic nature of our stimulation and the limits associated with heat dissipation from the head-mounted LEDs, we opted for a 10% duty cycle. This is substantially lower than our prior studies, which employed a 50% duty cycle. It also differs significantly from the effects of focal pharmacological activation, which would provide near-constant stimulation of the region. Recent work focused on cerebellar stimulation in the intrahippocampal kainate model of temporal lobe epilepsy (32) revealed that Bayesian optimization of stimulation parameters on a subject-by-subject basis produced robust and significant improvement in responses to closed-loop modulation. Further exploration of the parameter space for stimulation within the DLSC can reasonably be expected to do the same.

The only prior study to evaluate superior colliculus manipulations in a model of chronic epilepsy employed a combination of electrical and pharmacological stimulation in the genetic absence epilepsy rats from Strasbourg (GAERS) model (16). Unlike the majority of prior studies, which reported anti-seizure effects predominantly confined to the deep and intermediate layers of the SC, electrical stimulation in GAERS was effective even in superficial layers. The most parsimonious explanation for this discrepancy is current spread. Electrical stimulation of the SC, which includes dense bands of white matter, is likely to activate afferent and efferent fibers, as well as fibers of passage. Even electrical stimulation of the superficial layers, at sufficient current amplitudes, may result in the recruitment of the deep and intermediate layers. One animal in our present study had LEDs placed more superficially than the others (just dorsal to the intermediate layers rather than within the intermediate or deep layers). Thus, light likely impacted this subject’s intermediate but not deep layers of the superior colliculus.

Interestingly, this subject displayed the smallest effect size (4 and 7% reduction in seizure count with 5 and 100 Hz stimulation, respectively). The remaining animals displayed a mean reduction of 32% and 26%, with 5 and 100 Hz stimulation, respectively. This pattern is consistent with a preferential role for the deep/intermediate layers on seizure activity.

We did not monitor estrous cycle in this study. In the pilocarpine model, ovariectomy does not appreciably change seizure burden over a 30 day period (33), and >80% of female rats that undergo pilocarpine-induced status epilepticus display either a cessation of the estrous cycle or abnormal estrous cycling (34). Additionally, routine vaginal cytology would have posed an additional stressor to animals. Stress can worsen epileptiform activity in the pilocarpine model in mice (35), handling-associated stress can induce seizures in some epilepsy models (36), and the interplay between stress and the interaction between glucorticoids and ovarian hormes is complex across the estrous cycle (37). Finally, the order of testing was randomized across days and blocks of stimulation, and animals were run at different times, reducing the likelihood of estrous-cycle synchronization associated confounds. We analyzed the variability of seizure activity during baseline (no optogenetic stimulation) sessions and found no signficiant differences between males and females (see Supplementary Results). However, future studies that are powered to address estrous cycle variability in stimulation response may be warrented given the robust impact that ovarian hormones can have on seizure activity (38).

The higher level of variability we observed in the control group, as compared to the active vector group, may reflect random chance, but may also reflect a treatment effect. Given that seizure burden naturally waxes and wanes in this model (see (39)) in control animals we would expect that for a given session, some animals would have displayed increases, some decreases, and some no change. Consistent with this, we observed divergent seizure burdens with optogenetic stimulation in control animals; half of animals in the control group displayed increases in number of seizures with stimulation, half displayed decreases. Stimulation was consistently effective at both frequencies and across measures in all animals in the active group, but this was not the case for the control group.. By contrast, we both expected and observed that our intervention would lead to a decrease in all animals. However, it is noteworthy that relatively large percent decreases were observed in some control animals (in fact the largest percent decrses in the study were observed in one control animal). This raises the possibility that stimulation of the SC may drive the seizure-generating networks into a state of modest seizure probability – reducing the likelihood of days with very low or very high seizure burden, and instead consistently providing a reduction from the average baseline.

Our present findings, while providing proof of efficacy in a model of temporal lobe epilepsy, lead to several further questions. First, would closed-loop (responsive) neurostimulation display similar or enhanced efficacy? We focused initially on open-loop neuromodulation, as in our prior study in acute models; effects were most robust when stimulation preceded seizure onset. However, closed-loop neuromodulation is used clinically (40) and has been explored extensively in the preclinical literature (32,41–47). Future experiments evaluating closed-loop modulation may reveal a different pattern against chronic, spontaneous seizures. Second, what effect does chronic neurostimulation have on general behavior, EEG, and sleep? While we did not observe any overt behavioral responses to stimulation of the SC in this or our prior studies, the utility of DLSC-targeted stimulation would require a benign side effect profile. Given that DLSC activation can trigger desynchronization of the cortical EEG, the impact of stimulation on sleep is of particular interest for future studies. Moreover, given that electrical and pharmacological activation (although apparently not optogenetic activation) of the DLSC triggers orienting and defense responses (19,48), further characterization is needed.

Here we have demonstrated that chronic, 24-hour stimulation of the DLSC displays efficacy against seizures. These data build on a body of literature – almost exclusively in acute models of seizure activity – and extend it to spontaneous seizures in the rat pilocarpine model of temporal lobe epilepsy. While our current stimulation parameters did not alter the severity of seizures when they occurred, they significantly reduced both the number of seizures and cumulative seizure burden, which is of therapeutic value. Future studies optimizing stimulation parameters, examining side effect profiles, and identifying the circuit mechanism underlying this effect may further increase the translational rationale for this approach.

Supplementary Material

Supplementary Material

Key Points:

  • Activating the deep and intermediate layers of the superior colliculus (DLSC) in acute models of epilepsy suppresses seizures.

  • Effects of DLSC activation in chronic models of epilepsy with spontaneous limbic seizures were previously unknown.

  • We found that 5 and 100 Hz optogenetic stimulation of DLSC suppressed seizures in the pilocarpine model of temporal lobe epilepsy in rats.

  • The DLSC may represent a target for deep brain stimulation-based therapy for epilepsy.

Funding Statement:

This work was supported by R01NS097762 from the National Institutes of Health (NIH) / National Institute for Neurological Disorders and Stroke to PAF and F30NS110318 to SKH.

Footnotes

Ethics approval statement: We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

Conflict of interest disclosure: None of the authors have any conflict of interest to disclose.

Data availability statement:

Data will be made available upon request.

References

  • 1.Fisher R, Salanova V, Witt T, Worth R, Henry T, Gross R, et al. Electrical stimulation of the anterior nucleus of thalamus for treatment of refractory epilepsy. Epilepsia. 2010. May;51(5):899–908. [DOI] [PubMed] [Google Scholar]
  • 2.Velasco F, Velasco AL, Velasco M, Jiménez F, Carrillo-Ruiz JD, Castro G. Deep brain stimulation for treatment of the epilepsies: the centromedian thalamic target. Acta Neurochir Suppl. 2007;97(Pt 2):337–42. [DOI] [PubMed] [Google Scholar]
  • 3.Depaulis A, Vergnes M, Marescaux C. Endogenous control of epilepsy: the nigral inhibitory system. Prog Neurobiol. 1994;42(1):33–52. [DOI] [PubMed] [Google Scholar]
  • 4.Loddenkemper T, Pan A, Neme S, Baker KB, Rezai AR, Dinner DS, et al. Deep brain stimulation in epilepsy. J Clin Neurophysiol. 2001. Nov;18(6):514–32. [DOI] [PubMed] [Google Scholar]
  • 5.Dean P, Gale K. Anticonvulsant action of GABA receptor blockade in the nigrotectal target region. Brain Res. 1989. Jan 16;477(1–2):391–5. [DOI] [PubMed] [Google Scholar]
  • 6.Shehab S, Simkins M, Dean P, Redgrave P. The dorsal midbrain anticonvulsant zone--I. Effects of locally administered excitatory amino acids or bicuculline on maximal electroshock seizures. Neuroscience. 1995. Apr;65(3):671–9. [DOI] [PubMed] [Google Scholar]
  • 7.Garant DS, Gale K. Substantia nigra-mediated anticonvulsant actions: role of nigral output pathways. Exp Neurol. 1987. Jul;97(1):143–59. [DOI] [PubMed] [Google Scholar]
  • 8.Redgrave P, Dean P, Simkins M. Intratectal glutamate suppresses pentylenetetrazole-induced spike-and-wave discharges. Eur J Pharmacol. 1988. Dec 13;158(3):283–7. [DOI] [PubMed] [Google Scholar]
  • 9.Dean P, Simkins M, Hetherington L, Mitchell IJ, Redgrave P. Tectal induction of cortical arousal: evidence implicating multiple output pathways. Brain Res Bull. 1991. Jan;26(1):1–10. [DOI] [PubMed] [Google Scholar]
  • 10.Redgrave P, Dean P. Tonic desynchronisation of cortical electroencephalogram by electrical and chemical stimulation of superior colliculus and surrounding structures in urethane-anaesthetised rats. Neuroscience. 1985. Nov;16(3):659–71. [DOI] [PubMed] [Google Scholar]
  • 11.Redgrave P, Marrow LP, Dean P. Anticonvulsant role of nigrotectal projection in the maximal electroshock model of epilepsy--II. Pathways from substantia nigra pars lateralis and adjacent peripeduncular area to the dorsal midbrain. Neuroscience. 1992;46(2):391–406. [DOI] [PubMed] [Google Scholar]
  • 12.Redgrave P, Simkins M, overton P, Dean P. Anticonvulsant role of nigrotectal projection in the maximal electroshock model of epilepsy--I. Mapping of dorsal midbrain with bicuculline. Neuroscience. 1992;46(2):379–90. [DOI] [PubMed] [Google Scholar]
  • 13.Gale K, Pazos A, Maggio R, Japikse K, Pritchard P. Blockade of GABA receptors in superior colliculus protects against focally evoked limbic motor seizures. Brain Res. 1993. Feb 19;603(2):279–83. [DOI] [PubMed] [Google Scholar]
  • 14.Soper C, Wicker E, Kulick CV, N’Gouemo P, Forcelli PA. Optogenetic activation of superior colliculus neurons suppresses seizures originating in diverse brain networks. Neurobiol Dis. 2016. Mar;87:102–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Wicker E, Beck VC, Kulick-Soper C, Kulick-Soper CV, Hyder SK, Campos-Rodriguez C, et al. Descending projections from the substantia nigra pars reticulata differentially control seizures. Proc Natl Acad Sci U S A. 2019. Dec 16;201908176. [DOI] [PMC free article] [PubMed]
  • 16.Nail-Boucherie K, Lê-Pham BT, Marescaux C, Depaulis A. Suppression of absence seizures by electrical and pharmacological activation of the caudal superior colliculus in a genetic model of absence epilepsy in the rat. Exp Neurol. 2002. Oct;177(2):503–14. [DOI] [PubMed] [Google Scholar]
  • 17.Wicker E, Forcelli PA. Optogenetic activation of the reticular nucleus of the thalamus attenuates limbic seizures via inhibition of the midline thalamus. Epilepsia. 2021. Sep;62(9):2283–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Pellegrino LJ, Cushman AJ. A stereotaxic atlas of the rat brain [by] Pellegrino Louis J. [and] Cushman Anna J. New York: Appleton-Century-Crofts; 1967. [Google Scholar]
  • 19.Sahibzada N, Dean P, Redgrave P. Movements resembling orientation or avoidance elicited by electrical stimulation of the superior colliculus in rats. J Neurosci. 1986. Mar;6(3):723–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Brecht M, Singer W, Engel AK. Patterns of synchronization in the superior colliculus of anesthetized cats. J Neurosci. 1999. May 1;19(9):3567–79. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Cardin JA, Carlen M, Meletis K, Knoblich U, Zhang F, Deisseroth K, et al. Targeted optogenetic stimulation and recording of neurons in vivo using cell-type-specific expression of Channelrhodopsin-2. Nat Protoc. 2010;5(2):247–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Berndt A, Schoenenberger P, Mattis J, Tye KM, Deisseroth K, Hegemann P, et al. High-efficiency channelrhodopsins for fast neuronal stimulation at low light levels. Proc Natl Acad Sci USA. 2011. May 3;108(18):7595–600. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Nakamura S, Baratta MV, Pomrenze MB, Dolzani SD, Cooper DC. High fidelity optogenetic control of individual prefrontal cortical pyramidal neurons in vivo. F1000Res. 2012;1:7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Piredda S, Gale K. A crucial epileptogenic site in the deep prepiriform cortex. Nature. 1985;317(6038):623–5. [DOI] [PubMed] [Google Scholar]
  • 25.Turski WA, Cavalheiro EA, Schwarz M, Czuczwar SJ, Kleinrok Z, Turski L. Limbic seizures produced by pilocarpine in rats: behavioural, electroencephalographic and neuropathological study. Behav Brain Res. 1983. Sep;9(3):315–35. [DOI] [PubMed] [Google Scholar]
  • 26.Buckmaster PS, Reyes B, Kahn T, Wyeth M. Ventral hippocampal formation is the primary epileptogenic zone in a rat model of temporal lobe epilepsy. J Neurosci. 2022. Aug 19;JN-RM-0429–22. [DOI] [PMC free article] [PubMed]
  • 27.Toyoda I, Bower MR, Leyva F, Buckmaster PS. Early activation of ventral hippocampus and subiculum during spontaneous seizures in a rat model of temporal lobe epilepsy. J Neurosci. 2013. Jul 3;33(27):11100–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Connell P, Bayat A, Joshi S, Koubeissi MZ. Acute and spontaneous seizure onset zones in the intraperitoneal kainic acid model. Epilepsy Behav. 2017. Mar;68:66–70. [DOI] [PubMed] [Google Scholar]
  • 29.Bertram EH. Functional anatomy of spontaneous seizures in a rat model of limbic epilepsy. Epilepsia. 1997. Jan;38(1):95–105. [DOI] [PubMed] [Google Scholar]
  • 30.Sloviter RS. The neurobiology of temporal lobe epilepsy: too much information, not enough knowledge. C R Biol. 2005. Feb;328(2):143–53. [DOI] [PubMed] [Google Scholar]
  • 31.Schmidt-Kastner R, Heim C, Sontag KH. Damage of substantia nigra pars reticulata during pilocarpine-induced status epilepticus in the rat: immunohistochemical study of neurons, astrocytes and serum-protein extravasation. Exp Brain Res. 1991;86(1):125–40. [DOI] [PubMed] [Google Scholar]
  • 32.Stieve BJ, Richner TJ, Krook-Magnuson C, Netoff TI, Krook-Magnuson E. Optimization of closed-loop electrical stimulation enables robust cerebellar-directed seizure control. Brain. 2022. Feb 8;awac051. [DOI] [PMC free article] [PubMed]
  • 33.Valente SG, Marques RH, Baracat EC, Cavalheiro EA, Naffah-Mazzacoratti MG, Amado D. Effect of hormonal replacement therapy in the hippocampus of ovariectomized epileptic female rats using the pilocarpine experimental model. Epilepsy Research. 2008. Nov;82(1):46–56. [DOI] [PubMed] [Google Scholar]
  • 34.Joshi S, Sun H, Rajasekaran K, Williamson J, Perez-Reyes E, Kapur J. A novel therapeutic approach for treatment of catamenial epilepsy. Neurobiol Dis. 2018. Mar;111:127–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Castro OW, Santos VR, Pun RYK, McKlveen JM, Batie M, Holland KD, et al. Impact of corticosterone treatment on spontaneous seizure frequency and epileptiform activity in mice with chronic epilepsy. PLoS One. 2012;7(9):e46044. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Forcelli PA, Orefice LL, Heinrichs SC. Neural, endocrine and electroencephalographic hyperreactivity to human contact: a diathesis-stress model of seizure susceptibility in El mice. Brain Res. 2007. May 4;1144:248–56. [DOI] [PubMed] [Google Scholar]
  • 37.Atkinson HC, Waddell BJ. Circadian Variation in Basal Plasma Corticosterone and Adrenocorticotropin in the Rat: Sexual Dimorphism and Changes across the Estrous Cycle*. Endocrinology. 1997. Sep 1;138(9):3842–8. [DOI] [PubMed] [Google Scholar]
  • 38.Christian CA, Reddy DS, Maguire J, Forcelli PA. Sex Differences in the Epilepsies and Associated Comorbidities: Implications for Use and Development of Pharmacotherapies. Pharmacol Rev. 2020. Oct;72(4):767–800. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Smith ZZ, Benison AM, Bercum FM, Dudek FE, Barth DS. Progression of convulsive and nonconvulsive seizures during epileptogenesis after pilocarpine-induced status epilepticus. Journal of Neurophysiology. 2018. May 1;119(5):1818–35. [DOI] [PubMed] [Google Scholar]
  • 40.Morrell MJ. Responsive cortical stimulation for the treatment of medically intractable partial epilepsy. Neurology. 2011. Sep 27;77(13):1295–304. [DOI] [PubMed] [Google Scholar]
  • 41.Krook-Magnuson E, Armstrong C, Oijala M, Soltesz I. On-demand optogenetic control of spontaneous seizures in temporal lobe epilepsy. Nat Commun. 2013;4:1376. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Krook-Magnuson E, Szabo GG, Armstrong C, Oijala M, Soltesz I. Cerebellar Directed Optogenetic Intervention Inhibits Spontaneous Hippocampal Seizures in a Mouse Model of Temporal Lobe Epilepsy. eneuro. 2014. Dec 31;1(1):ENEURO.0005–14.2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Streng ML, Tetzlaff MR, Krook-Magnuson E. Distinct Fastigial Output Channels and Their Impact on Temporal Lobe Seizures. J Neurosci. 2021. Dec 8;41(49):10091–107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Armstrong C, Krook-Magnuson E, Oijala M, Soltesz I. Closed-loop optogenetic intervention in mice. Nat Protoc. 2013. Aug;8(8):1475–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Paz JT, Davidson TJ, Frechette ES, Delord B, Parada I, Peng K, et al. Closed-loop optogenetic control of thalamus as a tool for interrupting seizures after cortical injury. Nat Neurosci. 2013. Jan;16(1):64–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Kros L, Eelkman Rooda OHJ, Spanke JK, Alva P, van Dongen MN, Karapatis A, et al. Cerebellar output controls generalized spike-and-wave discharge occurrence. Ann Neurol. 2015. Jun;77(6):1027–49. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Berényi A, Belluscio M, Mao D, Buzsáki G. Closed-loop control of epilepsy by transcranial electrical stimulation. Science. 2012. Aug 10;337(6095):735–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.DesJardin JT, Holmes AL, Forcelli PA, Cole CE, Gale JT, Wellman LL, et al. Defense-like behaviors evoked by pharmacological disinhibition of the superior colliculus in the primate. J Neurosci. 2013. Jan 2;33(1):150–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Swanson LW. Brain maps 4.0-Structure of the rat brain: An open access atlas with global nervous system nomenclature ontology and flatmaps. J Comp Neurol. 2018. Apr 15;526(6):935–43. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Material
NIHMS1959391-supplement-Supplementary_Material.docx (509.7KB, docx)

Data Availability Statement

Data will be made available upon request.

RESOURCES