Bone marrow purging using monoclonal antibodies

Intracellular and surface distribution of CD9 in human eosinophils

Expression of CD9 is a feature of both eosinophils and platelets. We have investigated the CD9 expression on resting and activated eosinophils with regard to possibly interacting platelets. Mixed leukocytes were obtained from the platelet-containing (PC) and platelet-depleted (PD) peripheral blood of healthy donors. A cell membrane permeabilization technique, the FOG method, enabled us to detect the eosinophils as a separate population and permitted flow cytometric analysis of both surface and intracellular antigens. Monoclonal antibodies against CD61 were used to identify platelets. The CD9/CD61 ratio indicated that CD9 on resting eosinophils originates mainly from eosinophils and not from adhered platelets. No difference in CD9 expression was obtained between resting eosinophils in PC and PD blood. However, the expression of CD9 was decreased (p < 0.05) on eosinophils in PMA-activated PD blood but increased (p = 0.001) in PC blood, probably due to interacting platelets since CD61 increased simultaneously. In addition, we were able to detect an intracellularly stored pool of CD9 in eosinophils which decreased after activation with PMA. Together these results indicate a translocation of intracellularly stored CD9 to the cell membrane upon activation, probably followed by a subsequent shedding.

Antibody-induced modulation and intracellular transport of CD10 and CD19 antigens in human malignant B cells

Antibody-induced antigenic modulation (AIAM) is a complex biological phenomenon closely resembling other receptor-ligand interactions. Following exposure to specific antibodies, surface antigens are usually rapidly redistributed on the cell surface and internalized. A subsequent intracellular processing results in dissociation of the antigen-antibody complexes, degradation, exocytosis and recycling. AIAM plays an important role in MoAb-targeted therapy of hematopoietic malignancies contributing to escape of tumor cells from immunodestruction. On the other hand, internalization of MoAbs used as carriers of toxins and drugs is a prerequisite of therapeutic efficacy. Even though MoAbs directed against CD10 and CD19 have been used in immunotherapy of B cell malignancies, some aspects regarding AIAM of these Ags are not yet fully understood. Both Ags are modulated by specific MoAbs and internalized through the same pathway, however, the kinetics of AIAM vary from one Ag to another and from one cell type to another. Recent studies with malignant B-cell lines show that, under certain experimental conditions, the extent and rate of surface clearing, uptake and intracellular transport are considerably higher in the case of CD19 than in CD10 and higher in less mature cells compared with more mature cells. These observations may be useful in the selection of MoAbs for immunotherapy, although they need to be confirmed with fresh malignant B cells.

Modulation and intracellular transport of CD20 and CD21 antigens induced by B1 and B2 monoclonal antibodies in RAJI and JOK-1 cells--an immunofluorescence and immunoelectron microscopy study

By fluorescence microscopy (FM), flow cytometry (FCM) and immunoelectron microscopy (IEM) we have shown that B1 and B2 monoclonal antibodies (MoAbs) were able to induce modulation of CD20 and CD21 in RAJI and JOK-1 cell lines. Redistribution and internalization of both antigens (Ags) after binding with MoAbs was readily demonstrated by FM, and by IEM CD20 and CD21 were found to be processed by the pathway of receptor-mediated endocytosis. The rate of intracellular transport varied: CD21 > CD20 and RAJI > JOK-1. Approximately 65 and 55% of CD20 and 60 and 45% of CD21 were cleared from the surface of RAJI and JOK-1 cells, respectively (FCM and IEM). These values, however, clearly exceeded those corresponding to internalization (11, 9, 24 and 16%) indicating shedding of Ag-MoAb complexes. No evidence of recycling was found. The present data support the hypothesis that the kinetics of modulation vary from one Ag to another and probably also reflect the stage of differentiation of the malignant B-cells. The results are discussed in the context of the possible usefulness of B1 and B2 MoAbs in the therapy of B-cell malignancies.

Bone marrow purging in acute lymphoblastic leukemia: biological and clinical features

A group of 56 patients with immunophenotyped acute lymphoblastic leukemia had their bone marrow purged using monoclonal antibody (MAb) and complement. Mononuclear marrow cells were treated with panels of MAb directed against T cells (n = 19), CD10-CALLA (n = 16), and B cells (n = 21), resulting in median specific cell depletions of 96.50, 97.59, and 96.97%, respectively. The resulting CFU-GM recoveries were 60.15, 110.6, and 72.9%. Long-term bone marrow cultures showed normal formation of adherent cell layers in 17 of 23 cases, and failure to form an adherent layer correlated with poor hematological reconstitution. A total of 32 patients (aged 4-40 years) received purged marrow grafts containing a median of 1.32 x 10(7) mononuclear cells/kg and 1.64 x 10(4) CFU-GM/kg; 7 patients required backup marrow. Time to engraftment of 500 granulocytes, 1000 leukocytes, 20,000 platelets, and 50,000 platelets was 37, 38, 39, and 57 days, respectively. A total of 16 patients (50%) remain alive and in complete remission 313-913 days posttransplant (median 565 days); 13 patients (41%) relapsed between 50 and 365 days posttransplant (median 100 days), and 3 patients (9%) died from transplant-associated complications.

Elimination of clonogenic tumor cells from bone marrow using methylprednisolone (MP) and etoposide VP16: an in vitro pharmacologic study

The ability to eliminate malignant cells from bone marrow (BM) while retaining sufficient numbers of normal progenitors to ensure engraftment, may well establish the future of autologous BM transplantation (ABMT) for hematologic malignancies. In this study, we describe the effects of methylprednisolone (MP) and etoposide (VP16) alone or in combination on 5 tumor cell lines (HL-60, a promyelocytic cell line; Molt-4, a T cell leukemia; Daudi, a Burkitt's lymphoma and R10/8226 and R40/8226, doxorubicin-resistant myeloma cell lines). The tumor cell kill efficiency of the drugs was assayed using the limiting dilution assay. We determined the toxic effect on progenitor cells by assaying granulocyte-macrophage colony-forming units (CFU). With a combination of MP at 10(-3) M and VP16 at 75 microM, we observed the following log reduction in tumor cell clones: HL-60, 4.695 +/- 0.001; Molt-4, 3.626 +/- 0.036; Daudi, 5.633 +/- 0.001; R10/8226, 3.052 +/- 0.544; R40/8226, 3.126 +/- 0.080. CFU recovery was 24% +/- 5%. Mixing tumor cell lines with a 20-fold excess of normal irradiated BM cells did not eliminate the inhibitory effect of the drug combination. We propose that MP and VP16 used in concert produce effective purging of malignant hematopoietic cells from BM while sparing normal progenitors needed for engraftment.

Treatment of high-risk solid tumors of childhood with intensive therapy and autologous bone marrow transplantation

Autologous bone marrow transplantation (ABMT) allows delivery of intensive, marrow-ablative chemotherapy or chemoradiotherapy to children with high-risk solid tumors. Results from several studies of neuroblastoma suggest that outcome is improved by ABMT; however, relapses can occur months to years after complete clinical remission. Other high-risk tumors including peripheral neuroepithelioma, Ewing's sarcoma, rhabdomyosarcoma, Wilms' tumor, and brain tumors also appear to be responsive to intensive marrow-ablative therapy, although few studies have been reported. For tumors that can metastasize to marrow, a sensitive method is necessary for detecting tumor cell contamination. Immunocytologic analysis with monoclonal antibodies can identify one neuroblastoma cell per 10(5) normal marrow cells; this method also is applicable to other tumors with appropriate antibodies. Ex vivo removal (purging) of tumor cells decreases the probability of infusing tumorigenic cells with the ABMT. There is considerable experience in tumor detection and purging for neuroblastoma, but little has been done for other childhood solid tumors. Future investigations of ABMT will aim to further increase disease-free survival by intensifying induction and marrow-ablative regimens and by developing therapies to be given after ABMT that are directed at minimal residual disease. As pilot investigations mature, the efficacy of ABMT and conventional chemotherapy will be compared in multi-institution randomized studies.

Receptor-mediated cell attachment and detachment kinetics. I. Probabilistic model and analysis

The kinetics of receptor-mediated cell adhesion to a ligand-coated surface play a key role in many physiological and biotechnology-related processes. We present a probabilistic model of receptor-ligand bond formation between a cell and surface to describe the probability of adhesion in a fluid shear field. Our model extends the deterministic model of Hammer and Lauffenburger (Hammer, D.A., and D.A. Lauffenburger. 1987. Biophys. J. 52:475-487) to a probabilistic framework, in which we calculate the probability that a certain number of bonds between a cell and surface exists at any given time. The probabilistic framework is used to account for deviations from ideal, deterministic behavior, inherent in chemical reactions involving relatively small numbers of reacting molecules. Two situations are investigated: first, cell attachment in the absence of fluid stress; and, second, cell detachment in the presence of fluid stress. In the attachment case, we examine the expected variance in bond formation as a function of attachment time; this also provides an initial condition for the detachment case. Focusing then on detachment, we predict transient behavior as a function of key system parameters, such as the distractive fluid force, the receptor-ligand bond affinity and rate constants, and the receptor and ligand densities. We compare the predictions of the probabilistic model with those of a deterministic model, and show how a deterministic approach can yield some inaccurate results; e.g., it cannot account for temporally continuous cell attach mentor detachment, it can underestimate the time needed for cell attachment, it can overestimate the time required for cell detachment for a given level of force, and it can overestimate the force necessary for cell detachment.