Intracellular accumulation of the anti-CD20 antibody 1F5 in B-lymphoma cells.

In previous experiments, 125I-labeled 1F5 (anti-CD20)was found to kill B-lymphoma cells efficiently and specifically.Unexpectedly, the number of antibody (Ab) molecules taken up per cell was much larger than the number of antigen sites on the cell surface. The present studies were designed to explain this apparent discrepancy. Incubation with fluorophore-conjugated 1F5, using the Raji cell line, demonstrated that the Ab accumulated in large amounts in a juxtanuclear spot. Double labeling showed that the same spot was labeled by transferrin, but the transferrin labeling was much faster (45 min versus 18 h). Experiments with brefeldin A demonstrated that the spot stained was distinct from the Golgi cisternae; thus, it appears to be the endocytic recycling compartment. A fluorescent Fab fragment of 1F5 produced much weaker, barely detectable staining of the juxtanuclear spot. Experiments with three other B-lymphoma cell lines demonstrated marked heterogeneity among them. With Ramos cells, 1F5 and transferrin localized to multiple smaller intracellular spots, rather than a single large spot. There were also major differences between different Abs to CD20, as tested on Raji cells. Rituximab showed some staining of the juxtanuclear spot, but not as homogeneously as the staining with 1F5. B1 and L27 were not tested as thoroughly but did not appear to stain the juxtanuclear spot. Such internalization may have a major impact on the therapy of this tumor type with conjugates of anti-CD20 Abs. However, internalization did not correlate with sensitivity to specific killing by 125I-labeled 1F5.