Method for measuring neurotoxicity of aggregating polypeptides with the MTT assay on differentiated neuroblastoma cells

Abstract

Reliable in vitro assays are essential for study of the effects of neurotoxic compounds such as β-amyloid peptides (Aβ). The MTT assay has been used in cultures of different cells, e.g. SH-SY5Y neuroblastoma cells, for the quantitative measurement of Aβ toxicity. In our laboratory differentiated SH-SY5Y cells were used in the MTT assay. Cell differentiation with 10 μM all-trans-retinoic acid resulted in a constant cell number. The cells possess highly developed neurites and exhibit high sensitivity against Aβ. Owing to the constant cell number in differentiated SH-SY5Y cultures the decrease of the redox activity is directly proportional to the neurotoxicity of the substances, no correction is needed. The results of the MTT assay of Aβ peptides on differentiated SH-SY5Y cells displayed a good correlation also with the in vivo results. The present experiments reveal an effective assay for the study of potentially neurotoxic compounds.

Introduction

β-Amyloid peptide (Aβ) aggregates are toxic on neurons [17]. Aβ fibrils interact also with microglial cells and monocytes and stimulate expression of proinflammatory genes [6]. There is a need for an easy and reliable method for measuring the toxic effects of Aβ peptides and similar toxic polypeptide aggregates (e.g. prion peptides, amylin, etc.). The MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay has been widely used to measure the toxicity of different substances in cell cultures [22]. MTT enters the cells by endocytosis and is reduced to formazan by NADH reductase and other enzymes, which can be measured spectrophotometrically. All subcellular fractions can reduce MTT when supplied with NADH or NADPH. The amount of formazan reflects the reductive potential of the cytoplasm and the cell viability [15]. This method has proven to be very useful in neurotoxicity studies of Aβ peptides [14]. In most cases, neuroblastoma cells, e.g. SH-SY5Y, are applied [4]. The MTT assay is rapid and generally shows a good correlation with other viability tests 16., 25. and in vivo results 1., 8..

The neurotoxicity of Aβ-peptides (Aβ 1–42 and Aβ 25–35) has been determined with the MTT assay on PC-12 [29], C 1300 [26] and SH-SY5Y [28] cells. Mostly non-differentiated neuroblastoma cells have been used in the MTT assay. Aβ and other aggregated peptides (prions and human amylin) exert a toxic effect by inducing apoptosis in neuroblastoma cells 5., 13., 19., 21..

Working with non-differentiated cells involves some disadvantages. The number of cells is continuously growing during the experiments (e.g. Aβ-treatment for 24 or 48 h), so the results of MTT assay ought to be corrected. Yankner et al. [30] reported in 1990 that Aβ-peptides were neurotrophic to undifferentiated neurons and neurotoxic to mature neurons. In cultures, the cells are unsynchronized, which means uncertainty in toxicity measurements. The MTT assay of Aβ-peptides might lead to false results when non-differentiated neuroblastoma cells are used, because Aβ-peptides act mostly on the neurites of the neuroblastoma cells 11., 18.. This was our experience, too.

Cell membrane proteins (receptors) play a key role in the events that lead to apoptotic cell death after Aβ treatment. Cell surface receptor complexes are mediating the cellular effects of Aβ peptides [3] and the membrane composition of differentiated and non-differentiated cells are different. As a consequence, we have now attempted to improve the weak points of the MTT method by using differentiated cells. SH-SY5Y cells were incubated with all-trans-retinoic acid (RA); the resulting differentiated cell culture was used for neurotoxicity studies [12]. Our aim was to find a reliable in vitro assay for neurotoxicity measurements, an MTT assay characterized with two improvements: (1) use of highly differentiated neuroblastoma cells which are (morphologically and physiologically) very close to neurons, (2) use of constant cell number during the whole experiment; in this case the redox activity of the cells is directly proportional to the neurotoxicity of the substances and no correction of the results is needed. We wanted to find a method generally suitable for measurements of neurotoxicity of very different aggregating peptides like β-amyloids, prion sequences and human amylin.