Mapping the Assembly of Band 3 and Rhesus Multi-Protein Complexes During Erythropoiesis

Abstract 812

Band 3 forms the core of a large multiprotein complex in the erythrocyte membrane, the Band 3 macrocomplex, which also includes proteins of the Rhesus complex (Rh and RhAG). Mutations in genes encoding proteins within this complex can result in hereditary spherocytosis with varying severity. The effect of distinct mutations and deficiencies in proteins of the Band 3 macrocomplex has been studied in detail in mature erythrocytes. This revealed important functional and structural properties of individual proteins and their relationships with other proteins within the Band 3 macrocomplex. Nevertheless, considerably less is know about the spatio-temporal mechanisms that direct the formation of the Band 3 macrocomplex, and that may explain the aberrations in the complex observed in spherocytosis. Therefore, we studied expression and mutual interactions of proteins of the band3 macrocomplex during development of proerythroblasts to reticulocytes.

Using confocal microscopy and western blotting, significant pools of intracellular Band 3 and RhAG were found in the basophilic normoblast. These intracellular pools gradually decreased in the polychromatic normoblast and were absent or low in the orthochromatic normoblast and reticulocytes, while surface expression increased. We used pronase treatment of intact cells to remove extracellular epitopes of BRIC 6 (Band 3 antibody) and LA1818 (RhAG antibody) to study the mechanism by which the intracellular pool of Band 3 and RhAG contributes to formation of the Band 3 complex on the cell surface. Pronase treatment of cells incubated with cycloheximide to block protein synthesis resulted in a reduced but still significant reappearance of BRIC6 (Band 3) and LA1818 (RhAG) epitopes on the plasma membrane confirming the presence of intracellular Band 3 and RhAG pools. It also showed that the bulk of Band 3 and RhAG is synthesized and trafficked to the membrane between the early basophilic and polychromatic stage. Immuneprecipitation of Band 3 from cell lysates of pronase treated cells pre-treated with brefeldin A to collapse the Golgi showed no increase in co-immuneprecipitated protein 4.2 albeit an increase in intracellular Band 3 expression. This suggests that protein 4.2 and Band 3 interact in the first Golgi compartment or late ER. In addition, pre-treatment of cells with cycloheximide prior to pronase treatment resulted in depletion of the intracellular Band 3 and co-immuneprecipitated protein 4.2 pool indicating that Band 3 and protein 4.2 traffic as a complex to the plasma-membrane. We were unable to co-immuneprecipitate Rh or Band 3 with intracellular pools of RhAG, whereas Rh was co-immuneprecipitated with RhAG from the plasma-membrane and from total cell lysates. Knockdown of RhAG in differentiating erythroblasts revealed a concomitant drop in membrane expression of Rh, leaving Band 3 unaffected, indicating that plasma-membrane expression of Rh but not Band 3 is dependent on RhAG.

In conclusion, despite the described association between the RhAG complex and the Band 3 complex in erythrocytes, the data suggest that the Band 3-protein 4.2 complex traffics and assembles independently from Rh and RhAG during erythroid differentiation. The experiments suggest that Rh and RhAG do not traffic as a complex to the plasma-membrane but probably assemble in the plasma-membrane. The RhAG knockdown experiments suggest that the dependency of Rh on RhAG as observed in Rhnull syndrome erythrocytes (“Rh regulator type”) originates early during erythropoiesis. Band3 surface expression was not affected upon RhAG knock down, which re-produced the unperturbed Band 3 levels seen in these patients.