Antibody-induced activation of the hyaluronan receptor function of CD44 requires multivalent binding by antibody.

CD44 can function as a receptor for hyaluronan (HA). However, many cell lines and normal hematopoietic cells that express CD44 do not constitutively bind HA. A monoclonal antibody (mAb) specific for CD44 (IRAWB 14) has been described previously which induces CD44-mediated binding of HA rapidly (seconds to minutes) in some cell lines and in normal murine T cells. Of 16 CD44-specific mAb tested in the present study, only 3 exhibited this activity. Monovalent Fab fragments were prepared from two IgG2a antibodies that induce HA binding (IRAWB 14 and IRAWB 26) and used to determine whether multivalent binding was required for induction of HA receptor function. Fab from both antibodies had a tendency to form multivalent aggregates. After addition of iodoacetamide to prevent further aggregation, multimeric and monovalent forms were separated by gel filtration. This made it possible to compare the inducing activity of monovalent and multivalent antibody fragments of identical composition in the absence of Fc determinants. Multimeric forms were very active at inducing binding of fluorescein-conjugated HA (Fl-HA). Monovalent Fab fragments of both antibodies had 20- to 50-fold lower binding activity than intact antibody or multimer. IRAWB 26 Fab monomers were completely inactive in the induction of HA-binding. The observed weak inducing activity of IRAWB 14 Fab monomer could be attributed to very low levels of contaminating multimer. Induction of HA binding could also be achieved by using anti-immunoglobulin to cross-link Fab monomers of IRAWB 26. Thus, multivalent binding was required for the activation of HA binding by CD44-specific antibody, suggesting that the distribution of CD44 molecules on the cell surface is important for HA receptor function. In kinetic studies, induction of HA receptor function occurred simultaneously with antibody binding at 0 degrees C (ice water bath). Furthermore, antibody could induce HA binding in paraformaldehyde-fixed cells, which were permeable to propidium iodide and trypan blue, suggesting that intracellular signaling mechanisms were not involved in induction of receptor function. We conclude, therefore, that these CD44-specific antibodies are inducing HA binding by directly influencing the distribution of CD44 on the cell surface. The possibility of a concurrent change in CD44 conformation is not ruled out. We discuss possible mechanisms by which CD44 might be activated to bind HA in vivo.