Chemically induced strong cellular hypertrophy often reduces the accuracy of cytotoxicity measurements obtained using the ATP assay

Sensitization of Airway Epithelial Cells to Toxin-Induced Death by TNF Superfamily Cytokines

The TNF superfamily of proinflammatory and proapoptotic cytokines influence tissue-wide responses to molecular insults such as small molecules, toxins, and viral infections that perturb cellular homeostasis at the level of DNA replication, transcription, and translation. In the context of acute lung injury, for example, TNF superfamily members like TNF-α and TRAIL can severely exacerbate disease pathophysiology. This chapter describes a systematic approach to optimization of mammalian cell viability assays and transcriptional profiling through nCounter^® Technology to permit a detailed examination of how TNF-α and TRAIL modulate programmed cell death pathways in concert with ricin toxin, a ribosome-inactivating protein (RIP) and a potent inducer of acute respiratory distress. We compare two widely used luciferase- and colorimetric-based cell viability assays and provide optimization protocols for adherent and non-adherent cell lines. We provide a computational workflow to facilitate downstream analysis of datasets generated from nCounter^® gene expression panels. While combined treatment with ricin toxin and TRAIL serves as the exemplar, the methodologies are applicable to any TNF superfamily member in combination with any biological agent of interest.

High-content image analysis (HCIA) assay has the highest correlation with direct counting cell suspension compared to the ATP, WST-8 and Alamar blue assays for measurement of cytotoxicity

INTRODUCTION

Various cytotoxicity assays measuring indicators such as enzyme activity, dye uptake, or cellular ATP content are often performed using 96-well microplates. However, recent reports show that cytotoxicity assays such as the ATP assay and MTS assay underestimate cytotoxicity when compounds such as anti-cancer drugs or mutagens induce cell hypertrophy whilst increasing intracellular ATP content. Therefore, we attempted to evaluate the reliability of a high-content image analysis (HCIA) assay to count cell number in a 96-well microplate automatically without using a cell-number indicator.

METHODS

We compared cytotoxicity results of 25 compounds obtained from ATP, WST-8, Alamar blue, and HCIA assays with those directly measured using an automatic cell counter, and repeating individual experiments thrice.

RESULTS

The number of compounds showing low correlation in cell viability measured using cytotoxicity assays compared to automatic cell counting (r²<0.8, at least 2 of 3 experiments) were follows: ATP assay; 7; WST-8 assay, 2; Alamar blue assay, 3; HCIA cytotoxicity assay, 0. Compounds for which correlation was poor in 3 assays, except the HCIA assay, induced an increase in nuclear and cell size. However, correlation between cell viability measured by automatic cell counter and the HCIA assay was strong regardless of nuclear and cell size. Additionally, correlation coefficients between IC₅₀ values obtained from automatic cell counter and from cytotoxicity assays were as follows: ATP assay, 0.80; WST-8 assay, 0.84; Alamar blue assay, 0.84; and HCIA assay, 0.98.

DISCUSSION

From the above, we showed that the HCIA cytotoxicity assay produces similar data to the automatic cell counter and is highly accurate in measuring cytotoxicity.