Search

Abstract Background In vitro experiments using radiolabeled molecules is fundamental for Positron emission tomography (PET) or single photon emission computed tomography (SPECT) tracer development and various metabolic assays, but no consensus on appropriate incubation conditions exists. Specifically, the use of shaking versus non-shaking conditions, cell number to medium volume and the choice of cell plating material may unintentionally influence cellular oxygenation and medium composition. This is problematic when testing the oxygen-dependence of tracers including 18F-fluoro-2-deoxyglucose ([18F]FDG) and hypoxia-selective 2-nitroimidazoles (e.g., 18F-fluoroazomycin-arabinoside, [18F]FAZA) or when doing prolonged experiments. The purpose of this study was to assess the influence of various experimental conditions on tracer retention. Methods Tumor cells were seeded in a) Glass or standard Polystyrene Petri dishes or as b) discrete droplets in polystyrene Petri dishes or on 9 mm glass coverslips positioned in glass Petri dishes. When confluent, cells were pre-equilibrated for 2 h to 21%, 0.5% or 0% O2 and [18F] FDG or [18F] FAZA was added, followed by cell harvest and analysis of radioactivity 1 h ([18F]FDG) or 3 h ([18F]FAZA) after. Experiments were conducted with/without orbital shaking. Results The influence of hypoxia on tracer retention varied widely among cell lines, but shaking-induced convection did not influence uptake. In contrast, hypoxia-driven [18F] FAZA, and to some extent [18F] FDG, retention was much lower in cells grown on polyethylene than glass. Scaling-down the number of cells did not compromise accuracy. Conclusions Tracer retention was similar under stagnant and forced convection conditions suggesting that the former approach may be appropriate even when accurate control of oxygen and tracer availability is required. In contrast, conventional plasticware should be used with caution when studying tracers and drugs that are metabolized and retained or activated at low O2 levels. Downscaling of cell number, by reducing the effective growth area, was feasible, without compromising accuracy.