You are here:
An Evaluation of In Silico Predicted Neurotoxicity in Embryonic Rat Dorsal Root Ganglion (DRG) Cultures: Effects on Cytotoxicity, Neurite Length, and Neurophysiology
Citation:
Johnstone, A., C. Mack, M. Valdez, D. Herr, T. Shafer, AND P. Kodavanti. An Evaluation of In Silico Predicted Neurotoxicity in Embryonic Rat Dorsal Root Ganglion (DRG) Cultures: Effects on Cytotoxicity, Neurite Length, and Neurophysiology. Society of Toxicology, Baltimore, Maryland, March 10 - 14, 2019.
Impact/Purpose:
Many chemicals are proposed to produce toxic responses through non-receptor mediated actions of the parent compound (or its metabolites) with endogenous proteins and cellular macromolecules (Thomas et al., 2013). For reproductive and developmental toxicants, predictions based on structure activity modeling have been proposed (Blackburn et al., 2011). A method based on chemical structure, known as the Hard and Soft Acids and Bases (HSAB) hypothesis, has recently been proposed to predict peripheral neurotoxicity (LoPachin et al., 2012). Peripheral neurotoxicity (i.e. paresthesia, hypesthesia, dysesthesia, anaphia, etc) continues to be a human health problem that is not easily predicted in advance. In spite of recent advances in chemical screening (e.g., EPA’s ToxCast) such programs do not have exhaustive coverage of molecular initiating events (MIE’s) for all forms of toxicity (e.g. neurotoxicity and reproductive toxicity) and are based on “acute exposures” in vitro that may not account for the effects of metabolism on chemicals. To address these data gaps in EPA’s existing screening programs, we propose to develop an AOP for certain types of toxicities based on HSAB predictions from chemical structures . The HSAB hypothesis was developed for neurotoxicants, and will be expanded and tested using additional chemicals. Neuropathies will be evaluated by the evaluation of protein adduct formation in functionally crucial neuronal macromolecules. We will use the existing findings to “follow the data”, provide proof of principle, and expand the number of chemicals (chemical space) tested using the HSAB theory, thereby helping to establish its predictive validity. Linking the predictions from the HSAB theory with traditional adverse outcomes is important for increasing its use in the risk assessment process. Successful predictions will allow the development of Adverse Outcome Pathways (AOPs) for peripheral neurotoxicity. If successful in predicting neurotoxicants, the theory can be further expanded into other types of toxicity. This research project will increase the predictive capability of EPA’s screening process for multiple types of toxicity.
Description:
Numerous chemicals are proposed to produce non-receptor/ion-channel mediated neurotoxic responses through adverse interactions with intracellular proteins and macromolecules. These molecular initiating events may induce an adverse outcome pathway (AOP) that can hinder synaptic nerve terminal protein function and result in peripheral neuropathy. To test the Hard-Soft Acid and Base (HSAB) hypothesis which is based on the potential chemical bio-reactivity (electrophilicity) to cause a neuropathic effect, a test set of chemicals from the USEPA ToxCast database was evaluated in vitro in DRG cultures. Chemicals predicted to be neurotoxic were tested acutely, using a 3-tiered assessment of: 1. Cytotoxicity (% lactate dehydrogenase (%LDH) release); 2. Structural alteration (total neurite length per neuron via high content microscopy) and 3. DRG neurophysiology, as measured by mean firing rate (MFR), recorded on microelectrode arrays. DRG primary cultures were generated from embryonic day 16 Long-Evans rats and treated with cytosine arabinoside (500 nM) on day 3 in vitro (DIV 3) to inhibit glial overgrowth. DRGs were dosed acutely on DIV 7 with 1, 5, 10, 50 or 100 µM of a chemical and LDH release was measured at 24 and 48 hrs after exposure. To assess for neurite length, DRGs where fixed and antibody (PGP9.5) stained 48 hrs after dosing (DIV 7). Of the 10 chemicals tested, only 4-cyclohexylhexanone (4C) increased %LDH release at 50µM (22%) and 100 µM (49%) after 48 hrs. Two chemicals, 4C and Phenylacetylaldehyde (PAA) decreased mean neurite length per neuron at 48 hrs; 4C at 50 µM (28%) and 100 µM (60%) and PAA at 100 µM (46%). Separately, mature DRG cultures were dosed with a single concentration (10, 50 or 100 µM) of one of 13 chemicals on DIV 14 and changes in MFR were recorded for 1 hr. Octanal and 2,5 hexanedione had no effect on MFR. However, 11 compounds altered MFR in a dose-dependent manner, the most potent being PAA (-67%), 4C (-82%), and 4-Tertbutyl cyclohexanone (-92%). With the exception of 4C, the observed changes in MFR or neurite length occurred in the absence of cytotoxicity. These results provide support that HSAB physical properties may be used to predict chemical-induced neurotoxicity. (This abstract does not necessarily reflect USEPA policy).