Galectins are a family of mammalian β-galactoside-binding proteins that positively and negatively regulate T cell death. Extracellular galectin-1 directly induces death of T cells and thymocytes, while intracellular galectin-3 blocks T cell death. In contrast to the antiapoptotic function of intracellular galectin-3, we demonstrate that extracellular galectin-3 directly induces death of human thymocytes and T cells. However, events in galectin-3- and galectin-1-induced cell death differ in a number of ways. Thymocyte subsets demonstrate different susceptibility to the two galectins: whereas galectin-1 kills double-negative and double-positive human thymocytes with equal efficiency, galectin-3 preferentially kills double-negative thymocytes. Galectin-3 binds to a complement of T cell surface glycoprotein receptors distinct from that recognized by galectin-1. Of these glycoprotein receptors, CD45 and CD71, but not CD29 and CD43, appear to be involved in galectin-3-induced T cell death. In addition, CD7 that is required for galectin-1-induced death is not required for death triggered by galectin-3. Following galectin-3 binding, CD45 remains uniformly distributed on the cell surface, in contrast to the CD45 clustering induced by galectin-1. Thus, extracellular galectin-3 and galectin-1 induce death of T cells through distinct cell surface events. However, as galectin-3 and galectin-1 cell death are neither additive nor synergistic, the two death pathways may converge inside the cell.

CD45 is a family of high molecular weight leukocyte cell surface glycoproteins. Recently, two related subregions of the cytoplasmic domain of CD45 have been shown to have 30-40% amino acid identity with a human placental protein phosphotyrosine phosphatase, and CD45 isolated from human spleen was found to exhibit intrinsic protein phosphotyrosine phosphatase (EC 3.1.3.48) activity. In the present studies, we demonstrate that each of the known isoforms of murine CD45 has an equivalent basal level of protein phosphotyrosine phosphatase activity and establish that this enzymatic activity is associated with the cytoplasmic domain of the glycoprotein. Studies with three independent sets of well-characterized parental CD45+, mutant CD45-, and revertant CD45+ lymphoma cell lines indicate that loss of CD45 increases the phosphorylation of the src-related leukocyte-specific tyrosine protein kinase p56lck on tyrosine-505, a putative negative regulatory site. This suggests that CD45 may play a role in leukocyte growth regulation by altering the kinase activity of p56lck.

The cell adhesion molecule CD44 is expressed on the majority of immune cells and is responsible for mediating adhesion to the extracellular matrix glycosaminoglycan, hyaluronan. The binding of CD44 to hyaluronan is induced on T lymphocytes after activation by antigen and on monocytes after stimulation by inflammatory agents. Under inflammatory conditions, CD44 on endothelial cells presents hyaluronan to CD44 on activated T lymphocytes and mediates a rolling interaction under flow conditions. This rolling interaction together with chemokine signaling upregulates integrinmediated adhesion, which induces cell arrest and leads to subsequent migration to the inflammatory site. Studies with monoclonal antibodies against CD44 in mouse models of chronic inflammatory disease showed reduced disease severity attributed to reduced leukocyte recruitment. More recent investigations, taking advantage of the availability of CD44 null mice, further established a role for CD44 in leukocyte recruitment to inflammatory sites. These studies also revealed a role for CD44 in limiting the inflammatory response and resolving inflammation in models of lung injury and hepatitis. Here we describe the contributions of CD44 and hyaluronan to an inflammatory response and discuss the role of CD44 in both promoting and resolving inflammation in various mouse models of inflammatory disease.

Lymphocyte CD44 interactions with hyaluronan localized on the endothelium have been demonstrated to mediate rolling and regulate lymphocyte entry into sites of chronic inflammation. Because neutrophils also express CD44, we investigated the role of CD44 and hyaluronan in the multistep process of neutrophil recruitment. CD44^−/− and wild-type control mice were intrascrotally injected with the neutrophil-activating chemokine, MIP-2, and leukocyte kinetics in the cremasteric microcirculation were investigated 4 h subsequently using intravital microscopy. Neither the rolling flux nor the rolling velocities were decreased in CD44^−/− mice relative to wild-type mice. In vitro, neutrophils did not roll on the CD44 ligand hyaluronan, consistent with the in vivo data that CD44/hyaluronan did not mediate rolling. However, the number of adherent leukocytes in the venule was decreased by 65% in CD44^−/− mice compared with wild-type mice. Leukocyte emigration was also greatly decreased in the CD44^−/− mice. The same decrease in adhesion and emigration was observed in the wild-type mice given hyaluronidase. Histology revealed neutrophils as being the dominant infiltrating population. We generated chimeric mice that express CD44 either on their leukocytes or on their endothelium and found that CD44 on both the endothelium and neutrophils was important for optimal leukocyte recruitment into tissues. Of those neutrophils that emigrated in wild-type and CD44^−/− mice, there was no impairment in migration through the interstitium. This study suggests that CD44 can mediate some neutrophil adhesion and emigration, but does not appear to affect subsequent migration within tissues.