Cognitive impairment is a core feature of cyclin-dependent kinase-like 5 (CDKL5) deficiency, a neurodevelopmental disorder characterized by early epileptic seizures, intellectual disability, and autistic behaviors. Although loss of CDKL5 affects a number of molecular pathways, very little has been discovered about the physiological effects of these changes on the neural circuitry. We therefore studied synaptic plasticity and local circuit activity in the dentate gyrus of bothCdkl5^−/yandCdkl5^+/−mutant mice. We found that CDKL5 haploinsufficiency in both male and female mice impairs hippocampus-dependent learning and memory in multiple tasks.In vivo, loss of CDKL5 reduced LTP of the perforant path to the dentate gyrus and augmented feedforward inhibition in this pathway;ex vivoexperiments confirmed that excitatory/inhibitory input into the dentate gyrus is skewed toward inhibition. Injecting the GABAergic antagonist gabazine into the dentate improved contextual fear memory inCdkl5^−/ymice. Finally, chronic forniceal deep brain stimulation rescued hippocampal memory deficits, restored synaptic plasticity, and relieved feedforward inhibition inCdkl5^+/−mice. These results indicate that CDKL5 is important for maintaining proper dentate excitatory/inhibitory balance, with consequences for hippocampal memory.SIGNIFICANCE STATEMENTCognitive impairment is a core feature of cyclin-dependent kinase-like 5 (CDKL5) deficiency disorder. Although CDKL5 deficiency has been found to affect a number of molecular pathways, little is known about its physiological effects on the neural circuitry. We find that CDKL5 loss reduces hippocampal synaptic plasticity and augments feedforward inhibition in the perforant path to the dentate gyrusin vivoinCdkl5mutant mice. Chronic forniceal deep brain stimulation rescued hippocampal memory deficits, restored synaptic plasticity, and relieved feedforward inhibition inCdkl5^+/−mice, as it had previously done with Rett syndrome mice, suggesting that such stimulation may be useful for other neurodevelopmental disorders.

Rett syndrome (RTT), caused by mutations in the X-linked gene encoding methyl-CpG binding protein 2 (MeCP2), severely impairs learning and memory. We previously showed that forniceal deep brain stimulation (DBS) stimulates hippocampal neurogenesis with concomitant improvements in hippocampal-dependent learning and memory in a mouse model of RTT. To determine the duration of DBS benefits; characterize DBS effects on hippocampal neurogenesis; and determine whether DBS influences MECP2 genotype and survival of newborn dentate granular cells (DGCs) in RTT mice. Chronic DBS was delivered through an electrode implanted in the fimbria-fornix. We tested separate cohorts of mice in contextual and cued fear memory at different time points after DBS. We then examined neurogenesis, DGC apoptosis, and the expression of brain-derived neurotrophic factor (BDNF) and vascular endothelial growth factor (VEGF) after DBS by immunohistochemistry. After two weeks of forniceal DBS, memory improvements lasted between 6 and 9 weeks. Repeating DBS every 6 weeks was sufficient to maintain the improvement. Forniceal DBS stimulated the birth of more MeCP2-positive than MeCP2-negative DGCs and had no effect on DGC survival. It also increased the expression of BDNF but not VEGF in the RTT mouse dentate gyrus. Improvements in learning and memory from forniceal DBS in RTT mice extends well beyond the treatment period and can be maintained by repeated DBS. Stimulation of BDNF expression correlates with improvements in hippocampal neurogenesis and memory benefits.

The physiological similarities between mice and humans make them vital animal models in biological and medical research. This paper explores the application of artificial intelligence (AI) in analyzing mice behavior, emphasizing AI’s potential to identify and classify these behaviors. Traditional methods struggle to capture subtle behavioral features, whereas AI can automatically extract quantitative features from large datasets. Consequently, this study aims to leverage AI to enhance the efficiency and accuracy of mice behavior analysis. The paper reviews various applications of mice behavior analysis, categorizes deep learning tasks based on an AI pyramid, and summarizes AI methods for addressing these tasks. The findings indicate that AI technologies are increasingly applied in mice behavior analysis, including disease detection, assessment of external stimuli effects, social behavior analysis, and neurobehavioral assessment. The selection of AI methods is crucial and must align with specific applications. Despite AI’s promising potential in mice behavior analysis, challenges such as insufficient datasets and benchmarks remain. Furthermore, there is a need for a more integrated AI platform, along with standardized datasets and benchmarks, to support these analyses and further advance AI-driven mice behavior analysis.

The automatic recognition and quantitative analysis of abnormal behavior in mice play a crucial role in behavioral observation experiments in neuroscience, pharmacology, and toxicology. Due to the challenging definition of abnormal behavior and difficulty in collecting training samples, directly applying behavior recognition methods to identify abnormal behavior is often infeasible. This paper proposes ABNet, an AI-empowered abnormal action recognition approach for mice. ABNet utilizes an enhanced Spatio-Temporal Graph Convolutional Network (ST-GCN) as an encoder; ST-GCN combines graph convolution and temporal convolution to efficiently capture and analyze spatio-temporal dynamic features in graph-structured data, making it suitable for complex tasks such as action recognition and traffic prediction. ABNet trains the encoding network with normal behavior samples, then employs unsupervised clustering to identify abnormal behavior in mice. Compared to the original ST-GCN network, the method significantly enhances the capabilities of feature extraction and encoding. We conduct comprehensive experiments on the Kinetics-Skeleton dataset and the mouse behavior dataset to evaluate and validate the performance of ABNet in behavior recognition and abnormal motion detection. In the behavior recognition experiments conducted on the Kinetics-Skeleton dataset, ABNet achieves an accuracy of 32.7% for the top one and 55.2% for the top five. Moreover, in the abnormal behavior analysis experiments conducted on the mouse behavior dataset, ABNet achieves an average accuracy of 83.1%.