Search

There is great interest in testing neuroprotectants which inhibit the neurodegeneration that results from excessive activation of N-methyl-D-aspartate (NMDA) receptors. As an alternative to in vivo testing in animal models, we demonstrate here the use of a complex in vitro model to compare the efficacy and toxicity of NMDA receptor inhibitors. Organotypic hippocampal slice cultures were used to compare the effectiveness of the Alzheimer's disease drug, memantine, the Parkinson's disease drug, procyclidine, and the novel neuroprotectant, gacyclidine (GK11), against NMDA-induced toxicity. All three drugs are non-competitive NMDA receptor open-channel blockers that inhibit excitotoxic injury, and their neuroprotective capacities have been extensively investigated in vivo in animal models. They have also been evaluated as potential countermeasure agents against organophosphate poisoning. Quantitative densitometric image analysis of propidium iodide uptake in hippocampal regions CA1, CA3 and DG, showed that, after exposure to 10μM NMDA for 24 hours, GK11 was the most potent of the three drugs, with an IC50 of about 50nM and complete protection at 250nM. When applied at high doses, GK11 was still the more potent neuroprotectant, and also the least cytotoxic. These findings are consistent with those from in vivo tests in rodents. We conclude that the slice culture model provides valuable pre-clinical data, and that applying the model to the screening of neuroprotectants might significantly limit the use of in vivo tests in animals.

Most in vitro models are only used to assess the short-term effects of test compounds. However, as demonstrated here, hippocampal slice cultures can be used for long-term studies. The test compound used was the metabotropic glutamate receptor antagonist, L(+)-2-amino-3-phosphonopropionic acid (L-AP3), which is known to be toxic in vivo after subchronic, but not acute, administration. Degenerative effects were monitored by measuring the cellular uptake of propidium iodide (PI; continuously present in the medium) and lactate dehydrogenase (LDH) leakage, and by using a panel of histological stains. Hippocampal slices, derived from 2–3 day old rats and grown for 3 weeks, were subsequently exposed for the next 3 weeks to 0, 10 or 100μM L-AP3, with PI (2μM) in the culture medium. Exposure to 100μM L-AP3 induced severe toxicity after 4–6 days, as shown by massive PI uptake, LDH leakage, and changes in MAP2 and GFAP immunostaining and in Nissl and Timm staining. In contrast, 10μM L-AP3 did not induce detectable neuronal degeneration. Treatment with the NMDA receptor antagonist, MK-801, or the AMPA/KA receptor antagonist, NBQX, together with 100μM L-AP3, reduced neurodegeneration to close to control values. It is concluded that the continuous incubation of hippocampal slice cultures with PI is technically feasible for use in studies on inducible neuronal degeneration over time.

This paper reviews the current state of the use of organotypic brain slice cultures for neurotoxicological and neuropharmacological screening and mechanistic studies, as exemplified by excitotoxin application. At present, no in vitro systems have been approved by the regulatory authorities for neurotoxicity testing. For the evaluation of the slice culture method, organotypic hippocampal slice cultures were exposed to toxic doses of the excitotoxins, glutamate, N-methyl-D-aspartate (NMDA), kainic acid and 2-amino-3-hydroxy-5-methyl-4-isoxazole propionate (AMPA), and the glial toxin, DL-alpha-aminoadipic acid (DLAAA). Neuronal cell death was quantified by propidium iodide (PI) uptake, and visualised by Fluoro-Jade (FJ) staining. General cell death was monitored by lactate dehydrogenase (LDH) release into the culture medium. EC50 values for the different compounds, based on PI uptake after exposure for 48 hours in entire cultures, were: glutamate, 3.5 mM; DL-AAA, 2.3 mM; kainic acid, 13 microM; NMDA, 11 microM; and AMPA, 3.7 microM. In the slice cultures, the hippocampal subfields displayed the same differences in vulnerability as those observed in vivo. When subfield analysis was performed on the cultures, the CA1 subfield was most susceptible to glutamate, NMDA and AMPA, while CA3 was most susceptible to kainic acid. The amount of LDH release for DL-AAA was about four times that of L-glutamate, in accordance with the additional toxic effect on glial cells, which was also found by confocal microscopy to stain for FJ. In conclusion, it was found that organotypic brain slice culture, combined with standardised protocols and quantifiable markers, such as PI and FJ staining, is a relevant and feasible in vitro system for neurotoxicity testing. Considering the amount and quality of the available published data, it is recommended that the brain slice culture method could be subjected to pre-validation and formal validation for inclusion in a tiered in vitro neurotoxicity testing scheme to supplement and replace conventional animal tests.