Metabolism of apoB-100 in lipoproteins separated by density gradient ultracentrifugation in normal and Watanabe heritable hyperlipidemic rabbits.

We have examined the capability of a previously developed compartmental model to explain the kinetics of radioiodinated apolipoprotein (apo) B-100 in very low density lipoproteins (VLDL), intermediate density lipoproteins (IDL), and low density lipoproteins (LDL) separated by density gradient ultracentrifugation after intravenous injection of radioiodinated VLDL into New Zealand white (NZW) and Watanabe heritable hyperlipidemic (WHHL) rabbits. Our model was developed primarily from kinetics in whole blood plasma of apoB-100 in particles with and without apoE after intravenous injection of large VLDL, total VLDL, IDL, and LDL. When the initial conditions for this model were assumed to be an intravenous injection of radiolabeled VLDL, the plasma VLDL and LDL simulations for NZW rabbits and the VLDL, IDL, and LDL simulations for WHHL rabbits were found to be inconsistent with the observed density gradient data. By adding a new pathway in the VLDL portion of the model for NZW rabbits and a new compartment in VLDL for WHHL rabbits, and by assuming some cross-contamination in the density gradient ultracentrifugal separations, it was possible to bring our model, which was based upon measurements of 125I-labeled apoB-100 in whole plasma, into conformity with the data obtained by density gradient ultracentrifugation. The relatively modest changes required in the model to fit the gradient ultracentrifugation data support the suitability of our approach to the kinetic analysis of the metabolism of apoB-100 in VLDL and its conversion to IDL and LDL based upon measurements of 125I-labeled apoB-100 in whole plasma after injection of radiolabeled VLDL, IDL, and LDL. Furthermore, the differences in kinetics observed by us between data from whole plasma and data from plasma submitted to ultracentrifugal separation from the same or similar animals highlight the fact that small variations that can occur in the separation of lipoprotein classes by buoyant density can lead to confusing results.