Could the Lateral Hypothalamus Be a New Target for Therapeutic Intervention to Prevent Seizures?

Abstract

Transient Targeting of Hypothalamic Orexin Neurons Alleviates Seizures in a Mouse Model of Epilepsy

Li HT, Viskaitis P, Bracey E, Peleg-Raibstein D, Burdakov D. Nat Commun. 2024;15(1):1249. PMID: 38341419; PMCID: PMC10858876. doi: 10.1038/s41467-024-45515-5.

Lateral hypothalamic (LH) hypocretin/orexin neurons (HONs) control brainwide electrical excitation. Abnormally high excitation produces epileptic seizures, which affect millions of people and need better treatments. HON population activity spikes from minute to minute, but the role of this in seizures is unknown. Here, we describe correlative and causal links between HON activity spikes and seizures. Applying temporally targeted HON recordings and optogenetic silencing to a male mouse model of acute epilepsy, we found that preseizure HON activity predicts and controls the electrophysiology and behavioral pathology of subsequent seizures. No such links were detected for HON activity during seizures. Having thus defined the time window where HONs influence seizures, we targeted it with LH deep brain stimulation, which inhibited HON population activity and produced seizure protection. Collectively, these results uncover a feature of brain activity linked to seizures and demonstrate a proof-of-concept treatment that controls this feature and alleviates epilepsy.

Commentary

Our inability to effectively treat seizures in approximately 30% of patients diagnosed with epilepsy necessitates the identification of new treatment strategies. Hypothalamic hypocretin/orexin neurons (HONs) have the potential to be a target for therapeutic intervention. HON cell bodies are localized to the lateral hypothalamus (LH) and are excitatory with widespread connections throughout the brain including the brainstem, thalamus, hippocampus, and neocortex.^1,2 The system has been shown to regulate the sleep–wakefulness cycle, eating behavior, and arousal.

The reason these neurons have two different names is because the peptides they contain were discovered independently by two separate laboratories; one laboratory naming the peptides hypocretins due to their similar structure to secretin and the other naming the peptides orexins due to the observation that central administration of these peptides stimulated eating behavior, with orexis the Greek word for appetite. ³ HONs synthesize and synaptically release two neuropeptides, orexin-A (hypocretin-1) and orexin-B (hypocretin-2), which act through two G-protein-coupled receptors, orexin-1 (OX-1) and orexin-2 (OX-2) respectively to exert their action. ⁴ These neurons also release glutamate which likely contributes to their excitatory action. ⁵

The widespread excitatory nature of the HON system suggests that it could contribute to seizure activity and be a possible target for therapeutic intervention. Both OX-1 and OX-2 receptors have been localized to the rat hippocampus ⁶ and micropressure application of both orexin-A and orexin-B into CA1 of the rat hippocampus significantly increased action potential firing of pyramidal cells. ⁷ Injections of both orexins in pmol concentrations into the primary motor cortex of the rat resulted in electrographic and behavioral seizures. ⁸ Consistent with the hypothesis that HON activity could be proepileptic is the observation that intracerebral injection of SB-334867, a selective blocker of the OX-1 receptor, significantly decreased behavioral seizure score and seizure duration in pentylenetetrazol kindled rats. ⁹

The highlighted study used a comprehensive, multidisciplinary approach to determine the contribution of HON activity to evoked seizure activity and whether blocking HON activity could have an antiepileptic effect. ¹⁰ All experiments were performed in adult, male C57BL/6 mice and seizures were induced using an optogenetic stimulation model. To induce seizures a viral vector coding a red-shifted excitatory opsin (AAV9-CaMKIIa.C1V1(t/T).TS.EYSP) was stereotaxically injected unilaterally into the dorsal hippocampus. A fiber optic (200 µm) was implanted with the tip 500 µm above the target of the viral vector injection and seizures were triggered by exciting CaMKIIa.C1V1 expressing neurons using a 532 nm laser. To record electrographic seizure activity, a three-channel EEG/EMG headstage with surface screw electrodes was also implanted. Behavioral seizures were video recorded, and severity was scored using a modified Racine scale. ¹¹ Additional animals received stereotaxic injections into the LH of the viral vector AAV1-hORX.GCaMP6s to deliver GCaMP6s to HONs to allow for photometric measurement of HON activity. To determine the effect of optogenetic inhibition of HON activity, the viral vector AAV1-hORX.ArchT.TdTomato was bilaterally injected into the LH to deliver the inhibitory opsin, ArchT. Control animals were injected with the non-opsin hORX.TdTomato. For experiments that examined the effect of deep brain stimulation (DBS)-induced silencing of HONs on seizure activity, a bipolar electrode was implanted in the LH. Separate experiments were performed to determine (1) the contribution of HON activity to seizure activity, (2) the effect of pharmacologic blockade of the OX-1 receptor on seizure activity with SB-334867, and (3) the effect of suppression of HON activity both optogenetically through activation of inhibitory opsins and through DBS of the LH. Given the number of behaviors influenced by HON activity, behavioral testing for anhedonia, depression, appetite, reward-related behavior, and sleep were also evaluated in animals where HON activity was suppressed by DBS.

Optogenetic stimulation of the dorsal hippocampus consistently evoked electrographic and behavioral seizures with over 50% of the animals exhibiting generalized behavioral seizures. In animals systemically treated with the orexin receptor antagonist SB-334867 there was a significant decrease in the EEG power of optogenetically induced seizures. Although not significant, there appeared to be a decrease in the likelihood that the stimulation would evoke a seizure and there was a trend toward a decrease in seizure duration. However, treatment with SB-334867 did cause a significant decrease in motor activity, which highlights the potential side effects of blocking orexin receptors throughout brain. To determine the potential contribution of HON activity to an evoked seizure, HON activity was measured before, at the onset and at the end of the evoked seizure. HON activity increased during an evoked seizure but the presence or absence of HON activity during the seizure did not correlate with seizure intensity. However, there was a significant correlation between HON activity and seizure intensity when HON activity was elevated preceding seizure onset suggesting that increased HON activity preceding seizure onset could contribute to seizure intensity. Consistent with the above observation, preemptive, optogenetic silencing of ArchT expressing HONs before seizure onset resulted in a decrease in seizure intensity and behavioral seizure score while optogenetic silencing of HONs during a seizure had no effect. The positive effect of preemptive optogenetic silencing of HONs on seizure activity had a limited duration of <10 min. Similar results were obtained when HON activity was inhibited by preemptive DBS (sinusoidal, 120 Hz) delivered for 1 min to the LH. High frequency DBS decreased HON activity but was more effective than optogenetic silencing, significantly decreasing seizure probability, intensity and behavioral seizure scores for up to 20 min. DBS silencing of HONs was not accompanied by any of the expected side effects such as impaired movement, narcolepsy, stress, depressed appetite, or altered sleep which might be expected given the wide range of functions influenced by this system.

The results from the study by Li et al ¹⁰ provide clear evidence that excitatory HON activity is proepileptic and that suppressing HON activity can decrease seizure severity in an evoked seizure model. Additional research needs to be performed to determine if manipulation of the HON system in a spontaneous seizure model such as kainic acid-induced status epilepticus, yields similar antiepileptic results. The widespread excitatory connections of HONs raise the possibility that HON activity could be an unrecognized contributor to a decrease in seizure threshold. There is evidence that treatment of seizures associated with hypothalamic hamartomas is most effective when surgery is initiated early after seizure onset with the possibility of secondary epileptogenesis increasing the longer surgical intervention is delayed. ¹² Clinically, the LH could be a new target for DBS, however, the number of functions influenced by HONs suggests any manipulation has a high potential for significant side effects and would need to be targeted to the site of seizure onset. It is interesting to speculate that hamartoma-induced gelastic seizures could be mediated through HON dysfunction.