The authorities in Europe and United States request information regarding possible toxicity for substances that are produced in one tonne or more per year. Estimation of acute systemic toxicity is conducted in vivo using mice or rats. These tests can be time consuming, costly, unethical, and in some cases irrelevant due to the lack of similarity between humans and rodents. It has been proposed that determining general cytotoxicity together with more tissue-specific effects assessed by using in vitro test systems, e.g. reflecting adverse structure or function in the nervous system, can be an alternative approach to the in vivo tests. Neurotoxicity studies in vitro can be performed by using primary cell cultures from fresh tissue or established cell lines, the latter being often preferred as they are beneficial both economically and ethically.

Here, I present a murine neural progenitor cell line called C17.2 with the potential to differentiate to a mixed culture of both neurons and astrocytes. The differentiation process was examined using 3 different media compositions and 3 different exposures, totally 9 different scenarios. After 7 days in culture with DMEM/F-12 medium containing N2 supplements and 10 ng/mL nerve growth factor and 10 ng/mL brain derived neurotrophic factor, the culture contained two morphological distinguishable cell types, assumed to be neurons and astrocytes. Reverse Transcriptase Polymerase Chain Reaction (RT-PCR) and Western blot analyses were performed which confirmed the presence of neurons and astrocytes in the differentiated cultures. The mRNA and protein levels of the neuronal marker bIII-tubulin and the astrocyte marker glial fibrillary acidic protein (GFAP) were up-regulated during differentiation, while the progenitor cell marker nestin was down-regulated.

To further investigate how this differentiated neural cell model responded to neurotoxic and non-neurotoxic substances, cell membrane potential (CMP) and mRNA expression of bIII-tubulin, GFAP and heat shock protein-32 were examined after exposure to nicotine, atropine, strychnine, ethanol, digoxin, and acetylsalicylic acid. The concentrations that induced effects on the CMP and biomarker expression were compared to general cytotoxicity (Inhibitory Concentration 50%) determined by the neutral red uptake assay in a mouse fibroblast 3T3 cell line, i.e. the 3T3/NRU assay. The CMP assay showed that nicotine, atropine and strychnine exposure induced depolarisation of the cell membrane. However, no effect on the CMP was seen when the cells were treated with acetylsalicylic acid, digoxin, and ethanol at non-cytotoxic concentrations. Alternation in the mRNA expression levels for one of the three biomarkers was seen at non-cytotoxic concentrations for all the compounds that induced acute toxicity by neuronal modes of action, i.e. nicotine, atropine and strychnine. No significant alteration was seen in any of the biomarker mRNA levels when the differentiated C17.2 cells were exposed to compounds that do not induce acute toxicity by neuronal modes of action, i.e. digoxin and acetylsalicylic acid and ethanol.

In conclusion, acute toxicity, which could be induced by neuronal modes of action, may be detected in the differentiated C17.2 cell model by using CMP and gene expression biomarkers as endpoints. The simple cell culture requirements for culturing and differentiating the C17.2 cells into a mixed culture of neurons and astrocytes, the robustness in toxicity read-out, and the cost-effectiveness of the assay make the C17.2 cell line attractive as a model for acute neurotoxicity studies.

We are surrounded by chemicals, thus understanding how exposure to these chemicals affect us during our life is of great social importance. In order to predict human acute toxicity of chemicals, cosmetics or drugs, development of novel in vitro test strategies is required. The overall aim of this thesis was to evaluate whether two different cell line models could be used to predict acute neurotoxicity or developmental neurotoxicity. In paper one, we identified changes in cell membrane potential (CMP) as the most sensitive indicator of toxicity in neuroblastoma SH-SY5Y cells.

In the following studies, we evaluated the capacity of the murine neural progenitor cell line C17.2 to differentiate into mixed cell cultures. Upon differentiation of the C17.2 cells we could identify two morphologically distinguishable cell types; astrocytes and neurons (Paper II). We then investigated how differentiated C17.2 cells responded to non-cytotoxic concentrations of three known neurotoxic and three non-neurotoxic substances. The neurotoxicants induced depolarisation of CMP and alteration in the mRNA expression of at least one of the three biomarkers studied, i.e. βIII-tubulin, glial fibrillary acidic protein or heat shock protein-32. In contrast, no significant effects were observed when exposed to non-neurotoxic compounds (Paper IV).

To further characterise the C17.2 cell model during differentiation, an mRNA microarray analysis of the whole genome was performed. The 30 most significantly altered biomarkers with association to neuronal development were identified. The mRNA expression of the 30 biomarkers were used as a panel to alert for developmental neurotoxicity by exposing C17.2 cells during differentiation to toxicants known to induce impaired nervous system development. All but two of the selected genes were significantly altered by at least one of the chemicals, but none of the 30 genes were affected when treated with the negative control (Paper III).  

In conclusion, the differentiated C17.2 neural progenitor cell line seems to be an attractive model for studying and predicting acute and developmental neurotoxicity.