Neurotoxicity in vitro: Model systems and practical applications. Comparative studies with the cholinergic neurotoxin in primary brain cultures and in rabbit retina in vivo.

In vitro neural systems can be predictive for central nervous system neurotoxicity, except where xenobiotics primarily affect the blood-brain barrier directly. The wide range of systems now used in neurobiological studies is available for mechanistic neurotoxicological investigations although the choice of system is generally arbitrary and for this reason a more rational 'stepwise' approach may now be justified. There are many culture systems available, including neural cell lines, organotypic explant or reaggregation cultures, and primary monolayer cultures of individual, or mixtures of, neural cell types: neurones, astrocytes and oligodendrocytes. Of these models much success has recently been achieved using the organotypic culture. Using rat whole-brain reaggregate cultures, good in vitro/in vivo correlations for the cholinergic neurotoxicant, ethylcholine mustard aziridinium (ECMA) have been demonstrated by comparing its actions with those in the rabbit retina in vivo. Low concentrations of ECMA (12.5 mum) cause a biphasic loss of choline acetyltransferase activity (ChAT) in foetal rat brain reaggregate cultures. Initial direct inhibition is followed by an apparently selective loss of cholinergic neurones. More widespread cytotoxicity occurs at higher ECMA concentrations. Similarly, intravitreal injection of ECMA at a final concentration of 12.5 mum into the rabbit eye in vivo for 96 hr produced a 36% loss of retinal ChAT activity with no change in the electroretinogram (ERG), whereas 600 mum-ECMA produced a 55% loss of ChAT with a significant loss of the ERG b-wave function, which is suggestive of a more generalized cellular toxicity. Exogenous nerve growth factor applied to the brain reaggregates 48 hr post-ECMA lesioning reverses the neurotoxin-induced loss of ChAT activity. By supplementing neurotoxicological information gained in the in vitro brain reaggregate culture system with tests using primary monolayer cultures of neurones or astrocytes a stepwise 'screening' system is proposed for potential neurotoxicants in vitro. In its simplest form this is: (1) screening initially by using tumour-derived neural cell line and/or primary mixed neural culture such as the 'Micromass' system; (2) testing of selected compounds in whole-brain reaggregates and (3) supplementing this information if necessary with effects on primary monolayer cultures of individual neural cell types.