Abstract
Duan X, Zhang C, Wu Y, Ju J, Xu Z, Li X, Ohdah S, Constantin OM, Pan Y, Lu Z, Wang C, Chen X, Gee CE, Nagel G, Hou S-T, Gao S, and Song K. 2025. Nature Commun. 16(1):559. PMID: 39780018. Optogenetics is a valuable tool for studying the mechanisms of neurological diseases and is now being developed for therapeutic applications. In rodents and macaques, improved channelrhodopsins have been applied to achieve transcranial optogenetic stimulation. While transcranial photoexcitation of neurons has been achieved, noninvasive optogenetic inhibition for treating hyperexcitability-induced neurological disorders has remained elusive. There is a critical need for effective inhibitory optogenetic tools that are highly light-sensitive and capable of suppressing neuronal activity in deep brain tissue. In this study, we developed a highly sensitive moderately K+-selective channelrhodopsin (HcKCR1-hs) by molecular engineering of the recently discovered Hyphochytrium catenoides kalium (potassium) channelrhodopsin 1. Transcranial activation of HcKCR1-hs significantly prolongs the time to the first seizure, increases survival, and decreases seizure activity in several status epilepticus mouse models. Our approach for transcranial optogenetic inhibition of neural hyperactivity may be adapted for cell type-specific neuromodulation in both basic and preclinical settings.
Commentary
Epilepsy is an incredibly common neurological disorder with upwards of 70 million people affected worldwide. All too often, epilepsy cannot be controlled with medications, greatly contributing to morbidity, mortality, and societal cost. 1 In many of these refractory cases, surgical options may be available. 2 However, surgical treatments carry risks of infection, loss of neurological function, cognitive deficits, etc. 3 Improved therapies are needed.
Powerful tools have been developed in which light-sensitive proteins are expressed in excitable cells and can be activated or inactivated with focally applied laser light. 4 Much of this has been developed in preclinical animal models and presently requires surgical implantation of the laser light source close to the light-sensitive cells. However, human clinical applications of optogenetics have been developed for certain diseases such as retinitis pigmentosa, which does not require the implantation of a light source. 5 For epilepsy, one potential hurdle to overcome is delivery of the light into the brain. Improvements have been made in transcranial delivery of light to stimulate neurons in the brain in animal models, but inhibiting neurons, which would be desirable for disorders of excessive excitation such as epilepsy, has been more challenging.
Here, Duan et al 6 make considerable strides in this direction. They first develop a channelrhodopsin that is more K+-selective, more light-sensitive, and activated by longer wavelength light than traditional channelrhodopsins. They initially expressed these in Xenopus oocytes to identify variants with optimal K+ selectivity and light sensitivity. They next demonstrate that these can be expressed in hippocampal neurons in acute brain slices in both rats and mice and can effectively suppress action potential generation when exposed to laser light in a pulsatile (10 Hz) or continuous fashion.
Next, they move to elegant in vivo studies in which adeno-associated viral (AAV) vector injections were made into the mouse hippocampus to express the opsin in excitatory neurons under the control of the CaMKIIa promoter. Combined optogenetic fiber and recording tetrodes were implanted in the same site as the AAV injections and a cannula was implanted in the contralateral hippocampus for future kainic acid injection. Following kainic acid injections mice proceeded through focal and generalized seizures as they progressed to status epilepticus. Local field potential recordings demonstrate that hippocampal activity associated with both focal and generalized seizures could be disrupted with laser light stimulation.
In parallel experiments, the genetically encoded calcium indicator, GCaMP7f, was expressed in the contralateral hippocampus, which was also instrumented with a photometry fiber. The K+-sensitive opsin, optogenetic fiber, and kainic acid cannula were placed ipsilaterally. Here, they could identify Ca2+ transients associated with seizures following kainic acid injection. These could be disrupted with green laser light stimulation. The laser light stimulation in this paradigm both increased the latency to detection of the first seizure and reduced the number of generalized seizures observed.
Finally, they demonstrate that these proteins can be activated by transcranially applied light, eliminating the need for an invasive implant. They demonstrate that there was considerably less evidence of local inflammation compared to the invasive implant, which is a distinct advantage. Using the mouse pilocarpine-epilepsy model they demonstrate that unilateral or bilaterally applied transcranial light could reduce fiber-photometrically detected Ca2+ transients. Bilateral light prolonged the inhibition. In the kainic acid model, they demonstrate that bilateral transcranial optogenetics increased latency to the first seizure and decreased the number of generalized seizures detected. In the acute pentylenetetrazol-induced seizure model, bilateral transcranial optogenetics reduced seizure-associated mortality. Unilateral transcranial stimulation reduced mortality to a lesser extent, but did not prolong the latency to the first seizure compared to controls.
There are quite a few notable advances covered here. First, the use of K+ channels to hyperpolarize and inactivate the excitable cells is intriguing and powerful. Second, the shift to use of longer wavelength light, green in this case, instead of the shorter wavelength blue light that is typically used, could allow for deeper penetration of transcranially applied light. They believe that there could be some sensitivity of these channels to even longer red-shifted light which could penetrate even deeper into brain tissue. Third, the delivery of light transcranially, obviating the need for invasive intracranial implantation of the light source which causes local inflammation, increases the risk for infection, bleeding, and other complications, and causes a small temperature change when illuminated, is a decided advantage.
Transcranial techniques have been used for some time for imaging. For example, transcranial Dopplers are becoming more and more mainstream for neuromodulation (eg, transcranial direct current stimulation, transcranial magnetic stimulation, etc) in a variety of diagnostic and therapeutic paradigms. 7
It is a strength of this study that they were able to reduce epileptiform discharges, seizures, and status epilepticus in three different experimental animal models. In one of the models, they even reduced the likelihood of mortality from status epilepticus, an important cause of mortality in patients with seizures and epilepsy. 8
While the seizure and status epilepticus aborting possibilities of this approach are exciting and powerful, another intriguing possibility is that such therapies could be applied following traumatic brain injury to reduce epileptogenesis. While all the underpinnings of epileptogenesis are not completely understood, at least some of epileptogenesis is likely due to use-dependent anatomical and circuit reorganization in the hippocampus. 9 Perhaps ways could be devised to express the channelrhodopsins in the highly metabolically active cells undergoing epileptogenesis, and then apply transcranial light to therapeutically mitigate the development of epilepsy. While probably not ready for prime-time human application, such an approach could prove useful in preclinical models to better understand mechanisms and develop mitigation strategies that could be applied to patients.
Certainly, a limitation of this study is that intracranial AAV injections were still required to express the channelrhodopsin in hippocampal cells. This would need to be overcome before this could be applied in humans. One method to deliver the channelrhodopsin to cells of interest could be through the use of molecularly tagged nanoparticles or other peripheral delivery systems, the development of which may be enhanced through the use of modern machine learning algorithms. 10 These could be packed with the gene therapy vector, coated with antibodies against hippocampal selective antigens, and injected intravenously. A challenge with nanoparticles is that they require transcytosis to cross the blood-brain barrier. The blood-brain barrier can be disrupted with focally applied ultrasound, but alternative methods are being developed to overcome this for other applications. Another open question is whether enough transcranial applied light could penetrate the thicker human skull to reach and activate the channelrhodopsin in the larger human brain. Though there are still details to work out, this is a step in a good direction toward having additional therapies for intervention in epilepsy.
