Search

In this issue of Blood, Sreeramkumar and colleagues report that E-selectin ligand-1 (ESL-1) is a highly selective ligand for E-selectin on hematopoietic progenitors with unexpected important contributions to their trafficking.

Beyond its well-established roles in mediating leukocyte rolling, E-selectin is emerging as a multifunctional receptor capable of inducing integrin activation in neutrophils, and of regulating various biological processes in hematopoietic precursors. Although these effects suggest important homeostatic contributions of this selectin in the immune and hematologic systems, the ligands responsible for transducing these effects in different leukocyte lineages are not well defined. We have characterized mice deficient in E-selectin ligand-1 (ESL-1), or in both P-selectin glycoprotein-1 (PSGL-1) and ESL-1, to explore and compare the contributions of these glycoproteins in immune and hematopoietic cell trafficking. In the steady state, ESL-1 deficiency resulted in a moderate myeloid expansion that became more prominent when both glycoproteins were eliminated. During inflammation, PSGL-1 dominated E-selectin binding, rolling, integrin activation, and extravasation of mature neutrophils, but only the combined deficiency in PSGL-1 and ESL-1 completely abrogated leukocyte recruitment. Surprisingly, we find that the levels of ESL-1 were strongly elevated in hematopoietic progenitor cells. These elevations correlated with a prominent function of ESL-1 for E-selectin binding and for migration of hematopoietic progenitor cells into the bone marrow. Our results uncover dominant roles for ESL-1 in the immature compartment, and a functional shift toward PSGL-1 dependence in mature neutrophils.

Reversible interactions of glycoconjugates on leukocytes with P- and E-selectin on endothelial cells mediate tethering and rolling of leukocytes in inflamed vascular beds, the first step in their recruitment to sites of injury. Although selectin ligands on hematopoietic precursors have been identified, here we review evidence that PSGL-1, CD44, and ESL-1 on mature leukocytes are physiologic glycoprotein ligands for endothelial selectins. Each ligand has specialized adhesive functions during tethering and rolling. Furthermore, PSGL-1 and CD44 induce signals that activate the β2 integrin LFA-1 and promote slow rolling, whereas ESL-1 induces signals that activate the β2 integrin Mac-1 in adherent neutrophils. We also review evidence for glycolipids, CD43, L-selectin, and other glycoconjugates as potential physiologic ligands for endothelial selectins on neutrophils or lymphocytes. Although the physiologic characterization of these ligands has been obtained in mice, we also note reported similarities and differences with human selectin ligands.

We describe the application of a novel, bispecific antibody platform termed dual affinity retargeting (DART) to eradicate B-cell lymphoma through coengagement of the B cell-specific antigen CD19 and the TCR/CD3 complex on effector T cells. Comparison with a single-chain, bispecific antibody bearing identical CD19 and CD3 antibody Fv sequences revealed DART molecules to be more potent in directing B-cell lysis. The enhanced activity with the CD19xCD3 DART molecules was observed on all CD19-expressing target B cells evaluated using resting and prestimulated human PBMCs or purified effector T-cell populations. Characterization of a CD19xTCR bispecific DART molecule revealed equivalent potency with the CD19xCD3 DART molecule, demonstrating flexibility of the DART structure to support T-cell/B-cell associations for redirected T cell-killing applications. The enhanced level of killing mediated by DART molecules was not accompanied by any increase in nonspecific T-cell activation or lysis of CD19(-) cells. Cell-association studies indicated that the DART architecture is well suited for maintaining cell-to-cell contact, apparently contributing to the high level of target cell killing. Finally, the ability of the CD19xTCR DART to inhibit B-cell lymphoma in NOD/SCID mice when coadministered with human PBMCs supports further evaluation of DART molecules for the treatment of B-cell malignancies.

Multiple myeloma (MM) expands in the bone marrow and causes devastating bone destruction by enhancing osteoclastic bone resorption in its vicinity, suggesting a close interaction between MM cells and osteoclasts (OCs). Here, we show that peripheral blood mononuclear cell-derived OCs enhanced growth and survival of primary MM cells as well as MM cell lines more potently than stromal cells, and that OCs protected MM cells from apoptosis induced by serum depletion or doxorubicin. OCs produced osteopontin (OPN) and interleukin 6 (IL-6), and adhesion of MM cells to OCs increased IL-6 production from OCs. In addition, IL-6 and OPN in combination enhanced MM cell growth and survival. However, the effects of OCs on MM cell growth and survival were only partially suppressed by a simultaneous addition of anti-IL-6 and anti-OPN antibodies and were completely abrogated by inhibition of cellular contact between MM cells and OCs. These results demonstrate that OCs enhance MM cell growth and survival through a cell-cell contact-mediated mechanism that is partially dependent on IL-6 and OPN. It is suggested that interactions of MM cells with OCs augment MM growth and survival and, thereby, form a vicious cycle, leading to extensive bone destruction and MM cell expansion.

Movement of T-lymphocyte cell surface CD43 is associated with both antigen activation of T-cell clones and chemokine induction of T-lymphocyte motility. Here, we demonstrate that CD43 movement away from the site of T-cell receptor ligation occurs in unprimed CD4(+) T cells as well as T-cell clones. The T-cell receptor (TCR)-dependent movement of CD43 in unprimed T cells is associated with a polarized morphology and CD43 accumulation at the uropods of the cells, unlike that reported for primed T cells. The polarization of CD43 has a requirement for Src kinases and occurs in conjunction with lipid raft coalescence. Thymocytes and T-cell hybridomas, cells that have altered responses to TCR activation and lack lipid raft coalescence, do not polarize CD43 as readily as unprimed T cells. The movement of CD43 depends on the cholesterol biosynthetic pathway enzyme 3-hydroxy-3-methylglutaryl coenzyme A (HMG CoA) reductase. Blockade of this enzyme can specifically prevent CD43 redistribution without affecting cell shape polarization. The likely mechanism of this alteration in CD43 redistribution is through decreased protein prenylation because the cholesterol-dependent lipid rafts still coalesce on activation. These findings suggest that the polarization of cell shape, lipid raft coalescence, and CD43 redistribution on T-cell activation have signaling pathway distinctions. Dissecting out the relationships between various stages of molecular redistribution and lymphocyte activation may facilitate fine-tuning of immunologic responses.

Two dendritic cell (DC) subsets have been identified in the murine system on the basis of their differential CD8alpha expression. CD8alpha(+) DCs and CD8alpha(-) DCs are considered as lymphoid- and myeloid-derived, respectively, because CD8alpha(+) but not CD8alpha(-) splenic DCs were generated from lymphoid CD4(low) precursors, devoid of myeloid reconstitution potential. Although CD8alpha(-) DCs were first described as negative for CD4, our results demonstrate that approximately 70% of them are CD4(+). Besides CD4(-) CD8alpha(-) and CD4(+) CD8alpha(-) DCs displayed a similar phenotype and T-cell stimulatory potential in mixed lymphocyte reaction (MLR), although among CD8alpha(-) DCs, the CD4(+) subset appears to have a higher endocytic capacity. Finally, experiments of DC reconstitution after irradiation in which, in contrast to previous studies, donor-type DCs were analyzed without depleting CD4(+) cells, revealed that both CD8alpha(+) DCs and CD8alpha(-) DCs were generated after transfer of CD4(low) precursors. These data suggest that both CD8alpha(+) and CD8alpha(-) DCs derive from a common precursor and, hence, do not support the concept of the CD8alpha(+) lymphoid-derived and CD8alpha(-) myeloid-derived DC lineages. However, because this hypothesis has to be confirmed at the clonal level, it remains possible that CD8alpha(-) DCs arise from a myeloid precursor within the CD4(low) precursor population or, alternatively, that both CD8alpha(+) and CD8alpha(-) DCs derive from an independent nonlymphoid, nonmyeloid DC precursor. In conclusion, although we favor the hypothesis that both CD8alpha(+) and CD8alpha(-) DCs derive from a lymphoid-committed precursor, a precise study of the differentiation process of CD8alpha(+) and CD8alpha(-) DCs is required to define conclusively their origin.

Recruitment of blood monocytes into tissues is a central event in the inflammatory response and in atherogenesis. The mechanisms leading to monocyte adhesion and migration through endothelium are not completely defined. We recently reported that MAb L11, against the leukocyte sialomucin CD43, blocks T-lymphocyte binding to lymph node and Peyer's patch high endothelial venules (HEV) and inhibits T-cell extravasation from the blood into organized secondary lymphoid tissues. We have now assessed the ability of L11 to inhibit monocyte-endothelial (EC) interactions and trafficking. L11 blocks binding of WEHI78/24 cells, a murine monocytoid cell line, to inflamed lymph node HEV and inhibits recruitment of monocytes and neutrophils to thioglycollate-inflamed peritoneum. Because monocyte adhesion to the endothelium and diapedesis in lesion-prone regions of the vasculature is among the earliest events in atherogenesis, leading to formation of lipid-laden foam cells, the ability of L11 to block monocyte recognition of aortic endothelial cells was assessed in a novel ex vivo assay of monocyte binding to intact rabbit aortic endothelium. Cholesterol feeding of rabbits induces enhanced aortic adhesiveness for monocytes and WEHI78/24 monocytoid cells, and this adhesion is inhibited by L11. The inhibitory effect of L11 is additive with that of a cocktail of anti-L-selectin and anti-alpha4 and beta2 integrin monoclonal antibodies. Thus, CD43 represents a novel target for manipulation of monocyte recruitment in inflammation and atherogenesis.

CD43 (the major sialoglycoprotein of leukocytes) is an adhesion molecule broadly expressed on hematopoietic cells. A monoclonal antibody recognizing this molecule induces apoptosis of lineage marker-negative bone marrow hematopoietic progenitor cells (HPCs) that express CD34 at a high density (CD34hiLIN-). However, not all cells within this population undergo apoptosis on stimulation via CD43. Dividing progenitor cells are most highly affected, whereas more primitive quiescent cells survive anti-CD43 monoclonal antibody treatment. These surviving cells (1) are enriched for cobblestone area-forming cells, (2) repopulate fragments for human fetal bone implanted into C.B-17 scid/scid mice, (3) have a potential to differentiate in vivo to myeloid and lymphoid cells, and (4) have a high proliferative potential in long-term stromal cell-free liquid culture. These data indicate that cells with hematopoietic stem cell characteristics are relatively resistant to CD43-mediated apoptosis as compared with HPCs. Thus, CD43 may be specifically involved in the regulation of HPC proliferation.

Autoantibodies to lymphocytes have been detected in sera from human immunodeficiency virus type 1 (HIV-1)-infected individuals, and several autoantigens have been described. Among them, hyposialylated CD43 has been shown to be a target for autoantibodies in up to 47% of HIV+ individuals. However, the corresponding autoantigen (ie, the incompletely sialylated CD43) has not been isolated from blood cells of HIV-1-infected individuals. Recently, we have observed in vitro that HIV-1 productively or latently infected CEM cells (CEMLAI/NP) express CD43 molecules with modified glycosylation (mogly CD43). Using CEMLAI/NP cells, which do not express any structural viral antigen, we show now that all of the tested HIV+ sera from asymptomatic individuals, and up to 86% of those from subjects at the acquired immunodeficiency syndrome stage contain antibodies (mainly IgM and, to a lesser degree, IgG) that recognize the surface of CEMLAI/NP cells, and precipitate mogly CD43 molecules from the cells lysates. Taken together with our previous demonstration of altered glycosylation of CD43 from HIV-1-infected CEM cells in vitro, the constant antimogly CD43 autoimmune response observed from asymptomatic HIV-1+ subjects is likely to illustrate the occurrence of an altered glycosylation in vivo of the major lymphocyte surface CD43 glycoprotein, associated with HIV-1 infection.

CD43 is a cell surface-associated mucin that is abundantly expressed by most leukocytes, and that appears to function as a negative regulator of cell surface interactions, providing a repulsive barrier around cells. We have analyzed herein the ability of anti-CD43 monoclonal antibody (MoAb) to upregulate both beta 1 and beta 2 integrin-mediated cell adhesion and to promote redistribution of the CD43 molecule into a cellular uropod. Engagement of CD43 with specific antibodies enhanced the cell adhesion to both 80- and 38-kD fibronectin fragments as well as to the endothelial cell ligands vascular cell adhesion molecule-1 and intercellular adhesion molecule-1, an effect that was mediated through the alpha 5 beta 1, alpha 4 beta 1, and alpha L beta 2 integrins, respectively. This effect on cell adhesion was achieved in Jurkat leukemic T cells by anti-CD43 MoAb alone; however, in T lymphoblasts, the activation of cell adhesion required the concomitant ligation of CD3 with suboptimal doses of anti-CD3 MoAb. Immunofluorescence analysis showed that the engagement of CD43 was accompanied by a differential redistribution of CD43 into a well-defined cytoplasmic projection or uropod, whereas the beta 1 or beta 2 integrins remained uniformly located on the contact area with substrata. This change in the localization of CD43 did not require costimulation and was induced directly by engagement of CD43 in T lymphoblasts. Other stimuli of cell adhesion in the form of cross-linked anti-CD3 MoAb or phorbol esters did not induce uropod formation or CD43 redistribution. In addition, we observed that prolonged co-culture of resting peripheral blood T lymphocytes with endothelial cells, in the absence of anti-CD43 MoAb, induced uropod formation and redistribution of CD43 in T cells. Interestingly, the myosin-disrupting drug butanedione monoxime inhibited the redistribution of CD43 induced by the specific MoAb, whereas the stimulation of cell adhesion induced by engagement of CD43 was preserved in the presence of this drug. These observations indicate that the signaling inducing integrin-mediated cell adhesion by CD43 takes place independently from the receptor redistribution. Altogether, these results indicate that CD43 has a regulatory role on both integrin-mediated T-cell adhesion and cellular morphology.

Interactions of hematopoietic progenitor cells (HPC) with bone marrow stroma, mediated by adhesion molecules, are assumed to be critically important to the regulation of hematopoiesis. However, the specific roles of individual adhesion molecules involved in these interactions are poorly understood. Here, a monoclonal antibody, MEM-59, recognizing CD43, an adhesion molecule highly expressed on HPC, is shown to induce apoptosis in this cell population. This process operates at the single-cell level, and its initiation requires crosslinking of surface CD43 and the presence of cytokines. In contrast to HPC, more differentiated cells originating from this primitive cell population, as well as peripheral lymphocytes, do not undergo apoptosis in response to the CD43-mediated stimulation. Thus, CD43 may function as a negative regulator of early hematopoietic events, delivering a signal for apoptosis of HPC.

A monoclonal IgMK (IgMGAS) cold agglutinin (CA) of the infrequent anti-Gd specificity, found in a patient with Waldenström macroglobulinemia, has been characterized. IgMGAS uses a VH gene homologous (93.7%) to the reported VH251 germ-line, one of the two functional genes of the VH5 family, with differences in both framework regions and complementary determining regions (CDR). The VL gene is homologous to the reported 15AVK1 germ-line gene, recently described in an anti-i CA, with differences mostly clustered in CDR. In the patient's serum, IgMGAS coexisted with a monoclonal IgG3K (IgGGAS) that lacked CA activity but expressed the private idiotopes found on IgMGAS. Both Ig lacked reactivity with antibodies detecting VH1 or VHIII or VKIII subgroup regions or the VH4-21 gene product that is expressed by anti-I/i CAs. The K chains from both Igs showed the same isoelectrical mobility. Moreover, the k chains from both serum Igs showed the same N-terminal amino acid sequence. This sequence was identical to that predicted by the nucleotide sequence of VK1GAS gene segment, including one discrepancy at position 15 (Ile for Val) with respect to the consensus VK1 subgroup regions. Although these data do not exclude a possible independent clonal origin, they are consistent with the notion that IgGGAS might be clonally related to IgMGAS.

A monoclonal antibody (LICR.LON.R10) specific for the major sialoglycoprotein of the erythroid cell membrane, glycophorin A (alpha), has been used to test the possibility that "cryptic" erythroleukemia may be diagnosed as acute lymphoblastic leukemia (ALL) or acute myeloblastic leukemia (AML). In addition to 27 overt erythroleukemias, 724 leukemias, including 329 ALL (103 in relapse), 205 AML, and 109 blast crises of Ph1-positive chronic myeloid leukemia, were analyzed. Twenty cases with a significant proportion of glycophorin-A-positive (gA+) cells were found; 8 of these (5 AML and 3 blast crises of chronic myeloid leukemia, CML) had an obvious erythroid component, but 12 others were diagnosed as AML (2), AMML (1), CML in myeloid blast crisis (4) or megakaryoblastic blast crisis (1), acute megakaryoblastic leukemia (2), or acute lymphoblastic leukemia (2). The latter two patients had no immunologic evidence supporting a diagnosis of ALL and were resistant to chemotherapy. We conclude that AML and ALL only very rarely express gA, and these are probably genuine "cryptic" erythroleukemias. Other gA+ leukemias (megakaryoblastic and CML blast crises) may arise from bi- or pluripotent stem cells and contain distinct and separable blast cell populations.

The murine monoclonal antibody (MAb) DF3 was prepared against a human breast carcinoma. Previous studies have demonstrated that DF3 antigen levels are elevated in plasma of patients with breast cancer. Furthermore, MAb DF3 reacts with circulating glycoproteins of different molecular weights ranging from approximately 300 to 450 kd. The present study demonstrates that plasma DF3 antigen is comprised of at least four moieties with slow (S), intermediate (I), rapid (R) and very rapid (VR) electrophoretic mobilities. The electrophoretic mobility patterns for circulating DF3 antigen differ among individuals. Moreover, DF3 antigen is detectable in urine, and the electrophoretic mobility of the urinary moieties is similar, but not identical, to that in the plasma. Studies in family members suggest that the electrophoretic heterogeneity of plasma DF3 antigen is determined by codominant expression of multiple alleles at a single locus. This locus may code for the core protein of DF3 antigen. These findings thus identify a genetically determined polymorphism of a circulating tumor-associated glycoprotein.

Sialophorin, previously called gpL115, is the heavily sialylated surface protein that is defective in lymphocytes of Wiskott-Aldrich syndrome patients. Using the monoclonal antibody L10 as a probe, sialophorin expression was detected on isolated T lymphocytes and thymocytes, B cell lines, monocytes, neutrophils, and platelets, but not on erythrocytes, fibroblasts, and glioblastoma cells. This unusual distribution pattern suggests that sialophorin is expressed on all circulating cells except erythrocytes. Trace amounts of the sialophorin molecules on lymphocytes are incompletely sialylated, but significant amounts of the molecules on thymocytes are incompletely sialylated. The molecular form of sialophorin on T lymphocytes, thymocytes, and monocytes is the previously characterized species of apparent mol wt 115,000. A newly described sialophorin species of apparent mol wt 135,000 was found on neutrophils and platelets. The 115,000 lymphocyte/monocyte form and the 135,000 platelet/neutrophil form were shown to be substantially similar. The two forms have approximately the same content of sialylated O-linked carbohydrate units since both undergo the same atypical shift in electrophoretic mobility on desialylation. Both contain the epitope recognized by the monoclonal antibody L2 and the epitope recognized by L10 antibody. Moreover, evidence from another study indicates that the polypeptide portions are identical, cumulatively suggesting that 115,000 sialophorin and 135,000 sialophorin are identical except for the presence on the latter of additional neutral saccharide residues.

The effect of well-characterized monoclonal antibodies to red cell surface molecules on the invasion of human red cells by the malarial parasites Plasmodium falciparum and Plasmodium knowlesi was examined. Antibodies to glycophorin A (GP alpha) inhibit invasion for both parasite species, and this is highly correlated with the degree to which they decrease red cell membrane deformability as measured by ektacytometry. This effect on rigidity and invasion was also seen with monovalent Fab fragments. The closer the antibody binding site was to the membrane bilayer, the greater was its effect on inducing membrane rigidity and decreasing parasite invasion. Antibodies to the Wright determinant in particular were the most inhibitory. This differential effect of the various antibodies was not correlated with their binding affinities or the number of sites bound per cell. Antibodies to surface molecules other than GP alpha were without effect. A novel mechanism is described whereby monoclonal antibodies and their Fab fragments directed at determinants on the external surface of red cells might act to inhibit invasion by malarial parasites by altering membrane material properties.

The M/N blood groups are carried by the major human red cell sialoglycoprotein, glycophorin A. O-glyosidic carbohydrate is needed for the activity, but the M/N specificity is due to amino acid replacements in the NH2-terminal portion of the molecule. We have used monoclonal antibodies specific for M and N blood groups to study their expression during normal erythropoiesis. Here we report that the M/N blood group activities are very weakly or not at all expressed before the polychromatic normoblast stage. Using polyclonal anti-glycophorin A antiserum, it was shown that glycophorin A molecules are already abundantly present on the earliest morphologically recognizable erythroid precursor, the proerythroblast. These findings can be explained by our previous observation that the O-glycosylation of glycophorin A gradually increases during erythroid maturation.