Epilepsy is a prevalent neurological disorder affecting approximately 50 million people worldwide, as reported by the World Health Organization [WHO] (2017). Although up to 30% of epilepsy cases are claimed to have a genetic cause (Weber et al., 2017), the underlying genetic factors remain unknown for a significant number of patients (Weber et al., 2017; Lasko and Luthy, 2021). The trio (two parents and the affected proband)-based study combined with whole-exome sequencing (WES) has emerged as a popular method to identify new epilepsy candidate genes (Dupépé et al., 2019). WES produces a thorough yet precise list of candidate genes at a reasonable price by focusing on the protein-coding regions (Harnish et al., 2019; Qiao et al., 2022a). Although several large-scale genetic diagnostic tests using trio-based WES have yielded significant results (Guan et al., 2021; Zhai et al., 2021), the approach has a high rate of false positive results, like another next-generation sequencing (NGS) techniques (Zhang et al., 2015), highlighting the need for direct experimentation to fully understand the effect of a variant (Lee et al., 2014; XiangWei et al., 2019).

Experimental models, including cultured cells, Drosophila, zebrafish, and rodents such as mice and rats, are widely used to study epilepsy candidate genes. Drosophila is a particularly attractive model due to its short life cycle of only 50 days and its ability to produce a large number of offspring. Due to these factors, spatiotemporal genetic modifications, such as those enabled by the UAS/Gal4 system, can be used to conduct fly research in just a few months (Kepecs et al., 2002; Takai et al., 2020; Guerrini et al., 2021). The discovery of the Drosophila model is also easy to translate to other vertebrate models (Hunt et al., 2019; Hegazi et al., 2022).

Under various seizure-induction assays, adequate seizure-susceptible mutants of Drosophila adults, which correlate with different epilepsy subtypes, exhibit a wide spectrum of seizure-like behaviors. For instance, sudden unexpected death in epilepsy (SUDEP) mutants Shaker and Hyperkinetic display leg-shaking behavior following etherization (Kaplan and Trout, 1969; D’Adamo et al., 2020), while general epilepsy with febrile seizure plus (GEFG+) and Dravet syndrome (DS) knock-in mutations display loss of posture and random wing buzzing after heat application (Sun et al., 2012; Schutte et al., 2014), Na⁺ channel mutants display spontaneous shuddering (Kaas et al., 2016), whereas bang-sensitive (BS) mutants exhibit seizure-like activity when evoked mechanically. However, only one kind of activity, characterized by seizure and paralysis after a mechanical stimulus and observed in over twenty BS mutants, has been defined as “seizure-like” (Parker et al., 2011a). Therefore, we introduced two novel seizure phenotypes induced by vertex shock in Drosophila to extend the application of mechanical induction assays and further investigate the relationship between epileptic phenotypes and genotypes in both Drosophila and humans with greater precision.

However, a broad applicability screening system is urgently required since enormous epilepsy candidate genes must be verified by functional experiments. Here, we propose a set of simple experiments in Drosophila to evaluate seizure behaviors, brain morphology, and electrophysiology, which can expedite the progress of identifying epilepsy candidate genes. Especially, our approach introduces two novel seizure phenotypes induced by vertex shock and several optimized electrophysiological methods, thereby expanding the range of research methods available for investigating epilepsy in the Drosophila model.

Trio-based WES has been successful in defining the most likely epilepsy-implicated loci and screening out candidate genes, such as UNC13B, CELSR3, and CDK19 in numerous studies (Chung et al., 2020; Wang et al., 2021; Li et al., 2022). Besides, NGS analysis of a large cohort of patients with epilepsy revealed that approximately 14% of sequence changes in 70 common and rare mutations were detected as nonsense or deletion (Lindy et al., 2018), while missense or frameshift mutations can also lead to loss-of-function (LOF) consequences (Chen et al., 2017; Lindy et al., 2018). The Gal4/UAS system allows for precise spatiotemporal manipulations of gene expression for cell labeling or gene function analysis (Del Valle Rodríguez et al., 2011). By crossing Gal4 driver lines to UAS-RNAi lines, the expression of UAS-linked genes can be effectively blocked in the chosen Gal4 pattern. However, some missense mutations lead to gain-of-function (GOF). For example, hundreds of missense mutations, including GOF mutations, in the gene SCN1A can confer a wide spectrum of epileptic phenotypes in Drosophila as they do in humans (Schutte et al., 2014), thus knock-in Drosophila lines mimicking the missense variants should be generated to realize more precise genetic editing (Xue et al., 2014). In our previous study, c.147A > T missense variation of Ywhaz was created using the CRISPR/Cas9 system to confirm its pathogenic effects (Wan et al., 2022). Thus, the knockdown model established by the Gal4/UAS system and the knock-in model established by the CRISPR/Cas9 system can cover most scenarios of gene mutations, including LOF and GOF mutations.

Animal experimentation can be a challenging and time-consuming process for validating the extensive range of candidate genes and variants proposed by WES. However, Drosophila has emerged as an attractive model organism to examine human epilepsy due to its short life span, strong genomic conservation, the abundance of technological resources for genomic manipulations (Shulman, 2015), and the exhibition of seizure behavior similar to humans. Historically, mechanical induction assays in Drosophila have only elicited one seizure behavior. This study expands the range of seizure-like phenotypes in the Drosophila model by discovering two novel seizure behaviors. Knockdown of Tango14 results in hyperactive seizure behavior (Figure 1), while knockdown of Klp3A results in late-phase seizure behavior (Figure 2). Our findings were further validated by the demonstration of hyperactivity seizure behavior in knockdown of the Ywhaz gene (Wan et al., 2022), suggesting that this novel behavior could be applied to verifying other epilepsy candidate genes. These results imply that loss of function in different candidate genes can result in distinct phenotypes. In light of our observation, the hyperactivity seizure, manifesting increased movement and excitement, is comparable to hyperesthesia and highly consistent with the hyperactive-automatism subtype in complex partial epilepsy. The late-phase seizure may correlate with focal epilepsy originating from the sensory cortex. However, further investigation into the genotype-phenotype relationship in Drosophila is required through additional studies.

Electrophysiology has shown to be a valuable tool for the study of epilepsy, with numerous studies reporting its effectiveness in assessing electrical activity in the giant fiber and neuromuscular junction (NMJ) (Kuebler et al., 2001; Marley and Baines, 2011; Parker et al., 2011a,b; Duan et al., 2012; Xue et al., 2014). However, its application in the central nervous system (CNS) studies of epilepsy using the Drosophila model is less common (Guo et al., 2019; Savitsky et al., 2020; Chi et al., 2022). Our study adopted three electrophysiological methods, namely, whole-cell recording, cell-attached recording, and evoked EPSP recording, to directly explore the electrophysiological characteristics in the CNS. Whole-cell/cell-attached recordings allow us to assess the neuronal excitability of mutants compared to wild-type neurons in specific GFP-expressed regions of the Drosophila brain, such as the mushroom body (MB), a major high-level brain structure in flies. We could determine whether the cells are firing action currents and record the firing modes, such as tonic or burst firing. Moreover, the approach of recording evoked EPSP from AL to MB, previously used to reveal input-timing-dependent plasticity in the MB circuit during olfactory learning (Qiao et al., 2022b), has the potential to investigate the synaptic and molecular causes of epilepsy. The latency of evoked EPSP can measure the time interval of monosynaptic transmission in target synapses (Gil and Amitai, 1996). Our preliminary results suggest that the latency of evoked EPSP in PN-gamma neurons is around 1.6 ms (Figure 6D), validating the efficacy of our techniques in accurately detecting the time interval of target synaptic transmission. While we have yet to observe positive results in any epilepsy candidate genes, these methods can provide a strong tool for studying genes involved in synaptic function coding in epilepsy. Combining these electrophysiological methods with the LPF recording technique in the CNS of Drosophila (Iyengar and Wu, 2021) can develop an instrumental platform for investigating the cellular and synaptic mechanisms underlying neural function or dysfunction (Roemmich et al., 2018), with excellent accessibility for studying epilepsy’s pathological and pharmacological aspects (Sheeba et al., 2008).

The prevalence of seizures and epilepsies is alarmingly high among infants and preschoolers, with 10–25% of these young patients suffering from intractable seizures and varying degrees of intellectual/developmental disabilities (ID/DD) (Chung et al., 2020). Understanding the genetic basis of these conditions is crucial for accurate diagnosis, prognosis, genetic counseling, and treatment. Recent advances in NGS have revealed that de novo missense mutations in the CACNA1A gene contribute to ID/DD and therapy-resistant epilepsy (Indelicato and Boesch, 2021). As a result, our development study utilized the knockdown of the Cac gene (the ortholog of CACNA1A) to further investigate the impact of epilepsy candidate genes on development. Although approximately 50% knockdown did not demonstrate a significant effect (Figure 7), the study provided a reliable protocol for examining the impact of epilepsy candidate genes on development. The use of Gal4 drivers in conjunction with Gal80ts, which exhibit minimal activity at 18^°C and maximal activity at 29^°C (Duffy, 2002), can precisely restrict the gene knockdown to the first, second, and third instar larval or pupal stages through temperature control. By measuring parameters such as larval or pupal weight, pupation rate, eclosion rate, and body length (Zhang et al., 2021), we can conduct a more comprehensive analysis of candidate genes that exclusively express during specific developmental stages.

The potential of the genetic screening system, which models epilepsy using genotypic and phenotypic strategies, could be further demonstrated by applying it to a larger collection of genes implicated in epilepsy, such as a gene set. In addition, it is important to consider the sex of the flies in current experiments, as gender differences in epilepsy exist in both humans and Drosophila (Sharma et al., 2009).

Notably, epilepsy in individuals may be caused by multiple pathogenic variants, both common and rare, with a range of effect sizes (Ferraro and Buono, 2006; Epi25 Collaborative, 2019; Thijs et al., 2019). Drosophila is advantageous in the study of epilepsy due to its ability to map genetic interactions, which is facilitated by powerful resources and techniques such as genome-wide RNAi transgenic stocks (Dietzl et al., 2007; Ni et al., 2011), collections of mapped transposon insertion alleles, Gal4/UAS system, and the recently developed CRISPR/Cas9 system. Notably, integrating the RNAi with Gal4/UAS system could achieve double- or even triple-knockdown in one generation without complicated genetic crosses (Okumura et al., 2015). Drosophila is a premier genetic model system for elucidating mechanisms responsible for many neurological genetic disorders (Jackson et al., 1998; Finelli et al., 2004). By making the most of resources and strategies, Drosophila holds great potential for advancing our understanding of the pathogenesis of epilepsy and other complex polygenic disorders, as well as the development of potential therapies.

The original contributions presented in this study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author/s.

C-QL, J-DQ, and K-CW designed the study and wrote the manuscript. X-CQ, M-FH, D-HL, C-QL, S-MX, X-XZ, Y-ML, and W-JZ performed the behavior experiments. J-DQ performed the electrophysiological experiments. All authors contributed to the article and approved the submitted version.