Reproducibility of adipogenic responses to metabolism disrupting chemicals in the 3T3-L1 pre-adipocyte model system: An interlaboratory study

•This study evaluated the interlaboratory reproducibility in 3T3-L1 cell responses.•Ten laboratories evaluated blinded chemicals and rosiglitazone using defined protocols.•Poor reproducibility was observed for several blinded chemicals across laboratories.•However, this reproducibility largely did not impact ability to differentiate active chemicals.•We find that both cell source and differentiation protocol differences mediated the variation. The 3T3-L1 murine pre-adipocyte line is an established cell culture model for screening Metabolism Disrupting Chemicals (MDCs). Despite a need to accurately identify MDCs for further evaluation, relatively little research has been performed to comprehensively evaluate reproducibility across laboratories, assess factors that might contribute to varying degrees of differentiation between laboratories (media additives, plastics, cell source, etc.), or to standardize protocols. As such, the goals of this study were to assess interlaboratory variability of efficacy and potency outcomes for triglyceride accumulation and pre-adipocyte proliferation using the mouse 3T3-L1 pre-adipocyte cell assay to test chemicals. Ten laboratories from five different countries participated. Each laboratory evaluated one reference chemical (rosiglitazone) and three blinded test chemicals (tributyltin chloride, pyraclostrobin, and bisphenol A) using: 1) their Laboratory-specific 3T3-L1 Cells (LC) and their Laboratory-specific differentiation Protocol (LP), 2) Shared 3T3-L1 Cells (SC) with LP, 3) LC with a Shared differentiation Protocol (SP), and 4) SC with SP. Blinded test chemical responses were analyzed by the coordinating laboratory. The magnitude and range of bioactivities reported varied considerably across laboratories and test conditions, though the presence or absence of activity for each tested chemical was more consistent. Triglyceride accumulation activity determinations for rosiglitazone ranged from 90 to 100% across test conditions, but 30–70 % for pre-adipocyte proliferation; this was 40–80 % for triglyceride accumulation induced by pyraclostrobin, 80–100 % for tributyltin, and 80–100 % for bisphenol A. Consistency was much lower for pre-adipocyte proliferation, with 30–70 % active determinations for pyraclostrobin, 30–50 % for tributyltin, and 20–40 % for bisphenol A. Greater consistency was observed for the SC/SP assessment. As such, working to develop a standardized adipogenic differentiation protocol represents the best str