Development of novel disease-modifying treatment strategies for neurological disorders, which at present have no cure, represents a major challenge for today's neurology. Translation of findings from animal models to humans represents an unresolved gap in most of the preclinical studies. Gene therapy is an evolving innovative approach that may prove useful for clinical applications. In animal models of temporal lobe epilepsy (TLE), gene therapy treatments based on viral vectors encoding NPY or galanin have been shown to effectively suppress seizures. However, how this translates to human TLE remains unknown. A unique possibility to validate these animal studies is provided by a surgical therapeutic approach, whereby resected epileptic tissue from temporal lobes of pharmacoresistant patients are available for neurophysiological studies in vitro. To test whether NPY and galanin have antiepileptic actions in human epileptic tissue as well, we applied these neuropeptides directly to human hippocampal slices in vitro. NPY strongly decreased stimulation-induced EPSPs in dentate gyrus and CA1 (up to 30 and 55%, respectively) via Y2 receptors, while galanin had no significant effect. Receptor autoradiographic binding revealed the presence of both NPY and galanin receptors, while functional receptor binding was only detected for NPY, suggesting that galanin receptor signaling may be impaired. These results underline the importance of validating findings from animal studies in human brain tissue, and advocate for NPY as a more appropriate candidate than galanin for future gene therapy trials in pharmacoresistant TLE patients.

No abstract

Key points The median raphe is a key subcortical modulatory centre involved in several brain functions, such as regulation of the sleep–wake cycle, emotions and memory storage.A large proportion of median raphe neurones are glutamatergic and implement a radically different mode of communication compared to serotonergic cells, although their in vivo activity is unknown.We provide the first description of the in vivo, brain state‐dependent firing properties of median raphe glutamatergic neurones identified by immunopositivity for the vesicular glutamate transporter type 3 (VGluT3) and serotonin (5‐HT). Glutamatergic populations (VGluT3+/5‐HT– and VGluT3+/5‐HT+) were compared with the purely serotonergic (VGluT3–/5‐HT+ and VGluT3–/5‐HT–) neurones.VGluT3+/5‐HT+ neurones fired similar to VGluT3–/5‐HT+ cells, whereas they significantly diverged from the VGluT3+/5‐HT– population. Activity of the latter subgroup resembled the spiking of VGluT3–/5‐HT– cells, except for their diverging response to sensory stimulation.The VGluT3+ population of the median raphe may broadcast rapidly varying signals on top of a state‐dependent, tonic modulation. AbstractSubcortical modulation is crucial for information processing in the cerebral cortex. Besides the canonical neuromodulators, glutamate has recently been identified as a key cotransmitter of numerous monoaminergic projections. In the median raphe, a pure glutamatergic neurone population projecting to limbic areas was also discovered with a possibly novel, yet undetermined function. In the present study, we report the first functional description of the vesicular glutamate transporter type 3 (VGluT3)‐expressing median raphe neurones. Because there is no appropriate genetic marker for the separation of serotonergic (5‐HT+) and non‐serotonergic (5‐HT–) VGluT3+ neurones, we utilized immunohistochemistry after recording and juxtacellular labelling in anaesthetized rats. VGluT3+/5‐HT– neurones fired faster, more variably and were permanently activated during sensory stimulation, as opposed to the transient response of the slow firing VGluT3–/5‐HT+ subgroup. VGluT3+/5‐HT– cells were also more active during hippocampal theta. In addition, the VGluT3–/5‐HT– population, comprising putative GABAergic cells, resembled the firing of VGluT3+/5‐HT– neurones but without any significant reaction to the sensory stimulus. Interestingly, the VGluT3+/5‐HT+ group, spiking slower than the VGluT3+/5‐HT– population, exhibited a mixed response (i.e. the initial transient activation was followed by a sustained elevation of firing). Phase coupling to hippocampal and prefrontal slow oscillations was found in VGluT3+/5‐HT– neurones, also differentiating them from the VGluT3+/5‐HT+ subpopulation. Taken together, glutamatergic neurones in the median raphe may implement multiple, highly divergent forms of modulation in parallel: a slow, tonic mode interrupted by sensory‐evoked rapid transients, as well as a fast one capable of conveying complex patterns influenced by sensory inputs.

Chloride ions play an important role in controlling excitability of principal neurons in the central nervous system. When neurotransmitter GABA is released from inhibitory interneurons, activated GABA type A (GABAA) receptors on principal neurons become permeable to chloride. Typically, chloride flows through activated GABAA receptors into the neurons causing hyperpolarization or shunting inhibition, and in turn inhibits action potential (AP) generation. However, in situations when intracellular chloride concentration is increased, chloride ions can flow in opposite direction, depolarize neurons, and promote AP generation. It is generally recognized that altered chloride homeostasis per se has no effect on the AP threshold. Here, we demonstrate that chloride overload of mouse principal CA3 pyramidal neurons not only makes these cells more excitable through GABAA receptor activation but also lowers the AP threshold, further aggravating excitability. This phenomenon has not been described in principal neurons and adds to our understanding of mechanisms regulating neuronal and network excitability, particularly in developing brain and during pathological situations with altered chloride homeostasis. This finding further broadens the spectrum of neuronal plasticity regulated by ionic compositions across the cellular membrane.