Performance evaluation of cytometric bead assays for the measurement of lung cytokines in two rodent models.

There is a growing demand for a cost-effective, efficient, and high-throughput method for measuring cytokines. Currently, many studies are using flow cytometric bead-based multiplex assays in the measurement of cytokines. However, limited data are available regarding the performance of these cytometric bead assays versus enzyme-linked immunosorbent assay (ELISA) or correlation with mRNA expression using real time reverse transcriptase-polymerase chain reaction (RT-PCR). In one of our studies, cytometric bead array (CBA) was used to measure inflammatory cytokine protein levels in bronchoalveolar lavage (BAL) samples from mice exposed to welding fume, an inflammatory particulate. The results were then compared to whole lung mRNA levels of the same cytokines measured by real time RT-PCR in the same mouse model. It was found that the trends in cytokine profiles measured via CBA agreed with the whole lung mRNA results. In a separate experiment, we used a rat zymosan infectivity model to induce a pulmonary immunomodulatory response and determined cytokine concentrations in recovered BAL fluid by ELISA and two different types of cytometric bead-based assays, CBA and FlowCytomix (FC). The sample-to-sample correlation was good between ELISA and CBA with correlation coefficient R values of 0.76, 0.66, and 0.92 for rat IFN-gamma, TNF-alpha, and IL-6, respectively. ELISA only correlated significantly with the FC assay for TNF-alpha with R=0.43. Patterns of cytokine response in our rat model also differed among the assays but overall, the ELISA and CBA yielded similar results. For a method-to-method comparison, we assayed supplied cytokine standards from ELISA kits using both ELISA and CBA to determine the R values and found it to be greater than 0.90 for all the cytokines tested. It was found that the ELISA was more sensitive in the low range of the standard curve while the bead assays were capable of detecting higher protein concentrations, which would allow for direct measurement of concentrated samples. There was a lack of agreement between the absolute protein values for the ELISA and flow cytometric bead-based assays; in most cases, the latter method tended to give higher protein concentrations than ELISA. In conclusion, direct comparisons between absolute protein values did not agree among the assays tested in this study, but patterns of cytokine response generally agreed between ELISA and CBA. In the case of the mouse CBA, a companion measurement is recommended if samples with low concentrations of an analyte are reported and extrapolated below sensitivity or zero.