Binding of chemicals to ionotropic glutamate receptors leads to impairment of learning and memory via loss of drebrin from dendritic spines of neurons

Neurotoxicity risk assessment is an important issue for regulatory agencies. Currently, chemicals with potential risks are determined by time-consuming and costly animal testing. Therefore, in vitro testing methods are needed to rapidly evaluate thousands of chemicals for which no safety data on neurotoxicity exist. In recent years, chemicals that induce learning and memory impairment are thought to increase the risk of neurodegenerative diseases such as Parkinson's disease and Alzheimer's disease. Therefore, such risk assessment is necessary for human safety. The existing AOP No. 6 in OECD Series on AOPs (AOP48 in the AOP-Wiki) defines a molecular initiating event (MIE) as “Binding of agonists to ionotropic glutamate receptors”, causing neuronal cell death linked to impairment of learning and memory through receptor hyperactivation⁽¹⁾.

Recent studies have shown that synaptic dysfunction precedes neuronal death in the early stages of dementia accompanied with the neurodegenerative diseases. Synaptic dysfunction is presumed as a decrease in the number of dendritic spines in neurons of the cerebral cortex and hippocampus, which are essential for learning and memory⁽²⁾. Therefore, the risk for impairment of learning and memory can be assessed by the synaptic dysfunction, namely decreased number of dendritic spines⁽³⁾.

Dendritic spines are small actin-rich projections protruding from the dendrites of neurons that form excitatory synapses in the cortex and hippocampus⁽⁴⁾. Drebrin is an actin-binding protein that localizes to dendritic spines and is said to play a specific role in their formation⁽⁵⁾. Drebrin is known to decrease in Alzheimer's disease with a high correlation to symptom stage^(6,7). In low-density cultures of hippocampal neurons, the number of dendritic spines can be counted as the number of drebrin clusters with immunostaining.

We have developed an experimental protocol for low-density neuronal culture in 96-well plates and an algorithm that automatically counts the number of drebrin clusters by high-content imaging analysis⁽⁸⁾. These protocols have been shown to be useful for screening chemicals that bind to the NMDA receptor. In fact, we have examined the toxicity of phencyclidine (PCP) and PCP-analogues and published results in a paper⁽⁹⁾. We have developed not only the immunocytochemical protocol for in vitro assay using neuronal culture but also enzyme-linked immunosorbent assay (ELISA) kits to evaluate drebrin protein levels. Thus, decreased number of dendritic spines induced by chemicals can be assessed quantitatively as a loss of drebrin immunocytochemically and biochemically.

Here, we propose a new AOP with the same MIE as AOP No. 6, in which loss of drebrin as KE leads to impairment of learning and memory. Studies of genetically engineered mice have shown that drebrin deficiency is directly related to synaptic dysfunction and leads to the impairment of learning and memory, even in the absence of neuronal cell death^(10,11,12). This is the most important distinction between the proposed AOP and the existing AOP. Measurement of drebrin expression levels in neurons with immunocytochemistry and/or ELISA is easy and high-reproducible. The new KE, loss of drebrin, will promote accumulation of data of chemicals for neurotoxicity. The proposed AOP is expected to contribute to the development of many in vitro test for neurotoxicity and to establish in silico prediction to evaluate safety of many substances for human and environments.