[Frontiers in Bioscience 3, d631-636, July 1, 1998]

Table of Conents
 Previous Section   Next Section


Dacai Liu, Tong Liu, Ruliang, Li, and Man-Sun Sy

Institute of Pathology, Cancer Research Center, Case Western Reserve University School of Medicine, Cleveland, OH 44120

Received 4/27/98 Accepted 5/12/98


Activation of CD44+ Jurkat cells is required for binding high levels of HA. Cellular activation results in the up-regulation of CD44 expression, reorganization of the cytoskeletal proteins, clustering of CD44, covalent dimerization of CD44 and binding of HA. Activation-induced covalent disulfide dimerization of receptors represents a novel signal transduction mechanism for regulating receptor-ligand interactions. Some of the proposed mechanisms may be applicable to other receptor molecules with cysteine residues in the transmembrane domain and multivalent ligands. While we have provided a general picture for activation induced binding of CD44 to HA, many specific questions remained unanswered. We shall briefly summarize some of the questions that need to be addressed in the future.

Up-regulation of CD44 may be essential for dimerization of CD44. The need for up-regulation of CD44 expression may reflect a simply quantitative requirement for CD44 dimerization. A certain threshold level of CD44 expression must be reached in order for CD44 to form clusters and to dimerize. A high CD44 density on the cell surface may enhance the probability of CD44-CD44 interaction. PMA may induce dimerization by simply up-regulating CD44 protein, to a level favorable for CD44-CD44 interaction. A more quantitative study of the relationship between the number of CD44 molecules expressed on the cell surface and CD44 dimerization is needed to address this issue.

Alternatively, up-regulation of CD44 expression may reflect a qualitative alteration in CD44. Newly synthesized CD44 proteins may be more prone to dimerization than pre-existing CD44. The newly synthesized CD44 protein in an activated cell may be post-translationally modified differently from the existing CD44. These differences may allow newly synthesized CD44 protein to dimerize more readily. At least three potential post-translational modifications may be different between pre-existing CD44 and newly synthesized CD44 in activated cells; glycosylation in the extracellular domain, palmitoylation in the transmembrane domain or phosphorylation in the cytoplasmic domain. Additional experiments are needed to verify whether newly synthesized CD44 in activated cells is preferentially dimerized and whether newly synthesized CD44s are biochemically different from pre-existing CD44. It is also possible that both quantitative and qualitative changes in CD44 expression are required for CD44 dimerization on the cell surface.

The molecular mechanism responsible for CD44 dimerization is not known. As with other proteins, disulfide bonding occurs subsequent to protein-protein interactions mediated by other non-covalent interactions. Mutational analysis of the zeta chain of the CD3 complex revealed that an aspartic acid in the transmembrane domain is critical for the spontaneous dimerization of CD3 (44). Since the aspartic acid can be replaced with a lysine residue, it is a charged residue within the a helix that is essential for dimerization. Whether interactions between charged residues within the transmembrane domain of CD44 is required for CD44 dimerization is not known.

The putative cytoplasmic domain of human CD44 is comprised of 72 amino acids (a.a). After activation, CD44 transfectants lacking the last 23 a.a. (CD44.C23) are able to aggregate, dimerize and bind HA. In contrast, CD44 transfectants lacking the last 57 a.a. (CD44.C57) neither aggregate nor dimerizes. The motif important in the clustering and dimerization of CD44 is located within the region between amino acid 305 to 338. (S-G-N-G-A-V-E-D-R-K-P-S-G-L-N-G E-A-S-K-S-Q-E-M-V-H-L-V-N-K-P-S-E). Within this sequence there are 5 serines. In murine CD44, Ser 323 and Ser 325 (typed in bold) are constitutively phosphorylated. PMA increased the portion of CD44 molecules that were phosphorylated (24). This region of CD44 is also important in the binding of murine CD44 to ankyrin. Phosphorylation of murine CD44 enhanced the interaction between CD44 and ankyrin (45). Whether phosphorylation of the cytoplasmic domain of CD44 is critical for the clustering or dimerization of CD44 remain to be investigated.

In addition to hyaluronic acid, CD44 also binds to other ligands including collagen (46), fibronectin (47), serglycin (48) and osteopontin (49). It has been reported that monovalent ligand occupancy of the integrin receptor induces different cellular responses than multivalent aggregation of the integrin receptor (50). Therefore, the requirements for interaction between CD44 and its ligands may also depend on the nature of the ligand. In the case of polymeric hyaluronic acid, multiple binding sites may be essential. On the other hand, binding of monovalent ligands like osteopontin may not require clustering or dimerization of CD44. Therefore, activation-induced dimerization of CD44 adds an additional flexibility to the capacity of CD44 to interact with multiple ligands. However, the mechanisms by which the interactions between CD44 and its ligands other than HA are regulated have not been studied. The system developed in this thesis work will be employed to study the interaction between CD44 and other ligands.