The Homodimerisation Model

The possibility that homodimerisation of transmembrane PTPases leads to the inhibition of phosphatase activity has received recent attention. An indication that the actions of CD45 might be regulated in this manner was suggested by studies using a chimaeric receptor with an epidermal growth factor receptor (EGFR) ectodomain and a CD45 cytoplasmic tail. This chimaera successfully restored TCR signal transduction in a CD45-deficient cell line, but TCR signalling was largely abrogated upon subsequently dimerising the artificial receptor by the addition of EGF (Desai et al. 1993). Mutation of a putative wedge domain (Fig. 1) within the CD45 tail portion of the chimaera significantly blunted the ability of EGF to inhibit TCR-mediated signals, suggesting that the wedge domain might be critical in promoting homodimeri-sation (Majeti et al. 1998). Introduction of a point mutation at the same site in CD45 in a mouse model in situ resulted in a severe lymphoproliferative and lupus-like syndrome with autoantibody production, consistent also with de-inhibition of CD45 PTPase activity (Majeti et al. 2000). However, since the point mutation was introduced into the germ-line, CD45 in all haematopoietic lineages presumably carries the same mutation, so the question of which cell-type is responsible for the observed abnormalities is important.

CD45 has been shown to dimerise in several different studies. For example, CD45 was found to dimerise to a small extent after chemical cross-linking of a Tcell line (Takeda et al. 1992) and recombinant fragments of the rat extracellular domain can exist as dimers as well as monomers (Symons et al. 1999). Is it possible that CD45 isoforms might differentially homodimerise, thereby inhibiting specific pools of CD45 PTPase activity? Interestingly, in a study using FRET, it was found that CD45R0 preferentially homodimerises on the surface of a T cell line when compared with CD45RBC or CDRABC isoforms (Dornan et al. 2002). Similar findings were obtained using chemical cross-linking and a cysteine dimer-trapping method in which it was also shown that CD45R0 preferentially homodimerises in a manner hindered by sialylation and O-glycosylation (Xu and Weiss 2002).

As with the cis-cis interaction model, the homodimerisation model is also an attractive hypothesis for explaining the differential actions of CD45 iso-forms. However, significant questions remain. For example, it is important to show by direct biochemical assay that CD45 homodimerisation causes inhibition of its PTPase activity. Furthermore, in the FRET study cited above, both CD45R0 homodimers and CD4/CD8-CD45R0 heterodimers were detected on the surface of the same cell, yet the net effect on TCR signal trans-duction of CD45R0 expression appeared to be positive (Dornan et al. 2002). In addition, in a series of transgenic lines expressing increasing amounts of CD45R0 at the T cell surface, in which comparable numbers of peripheral T cells were observed in the spleen and lymph nodes, a quantitative effect was noted in which T-dependent antigenic responses correlated positively with the CD45R0 expression level (Ogilvy et al. 2003). This result is difficult to explain if the main role of CD45R0 is to exert a negative effect on T cell responses. Therefore further work will be required to determine whether the homodimerisation model adequately explains the putative differential actions of CD45 isoforms in vivo.

Overall, our understanding of the molecular actions of specific CD45 iso-forms remains significantly incomplete. Achieving a thorough understanding of the biological reasons for the existence and tight regulation of CD45 isoform expression in immune cells remains one of the outstanding challenges in the CD45 research field.

0 0

Post a comment