Development and validation of a herring gull embryo toxicokinetic model for PCBs

A toxicokinetic model was developed to describe polychlorinated biphenyl (PCB) accumulation by herring gull (Larus argentatus) embryos during development. The model consists of two components, a bioenergetics model that predicts the lipid mass balance of embryo and yolk compartments with time and an empirical toxicokinetic model that describes PCB partitioning between lipid compartments in the egg. The model was calibrated using data on PCB and lipid partitioning between embryo and yolk + albumen at four time points during incubation in herring gull eggs injected with a PCB mixture, combined with data sets on herring gull embryo growth rates and bioenergetic demands with time. The model was validated using independent data consisting of maternally exposed, field-incubated Lake Superior herring gull eggs that varied in incubation ages over the range of 8.5 d to pipping age (26-28 days). PCB concentrations in 6-9 d embryos were nearly an order of magnitude less than predicted by equilibrium lipid partitioning between the embryo and yolk + albumen compartments of the eggs. PCB concentrations in embryos were adequately predicted by equilibrium partitioning, however, for eggs incubated for 23-28 d. An empirical relationship was developed to account for the apparent nonequilibrium behaviour of PCBs during early development. The model was sensitive to the mass of yolk lipids and the mass of PCBs deposited to fresh eggs and much of the variability in embryo PCB concentrations could by explained by accounting for variability in these input parameters. Consistent with experimental data for other avian species, the model predicts that the highest PCB concentrations in the embryo/chick occur during pipping or soon after when yolk lipids have been completely resorbed by the embryo.