A requirement for antigen-specific helper T cells in the generation of cytotoxic T cells from thymocyte precursors.

Thymocytes cultured with irradiated, allogeneic stimulator cells yield no cytotoxic effector cells after a period in culture. If, however, a population of irradiated spleen cells syngeneic to the responder cells are added to these cultures, cytotoxicity is generated. The helper activity present in the irradiated syngeneic spleen cells was found to be mediated by a cell bearing theta antigens. Furthermore, it was found to be antigen specific; helper cells which were tolerant of the stimulator cell antigens were unable to help the thymocyte responder cells, although these tolerant cells did contain helpers specific for a third party antigen. These experiments are consistent with a requirement for associative recognition of linked determinants in the induction of killer precursors which is thus strictly analogous to the induction of B-cell precursors via collaboration with helper T cells. In more extensive studies, it was found that histoincompatible helper cells (H-2b, H-2p, H-2q) were able to help a cytotoxic T cell (H-2k) response to a third party stimulator cell antigen (H-2d); that is, the helper T cells which interact with cytotoxic T-cell precursors are not strain specific. It seems likely that the histocompatible helper cells induce killer precursors in an antigen-specific cooperation event similar or identical to normal syngeneic cooperation.

Ontogeny of cell-mediated immunity. I. Early development of alloantigen-specific cytotoxic T-cell precursors in postnatal mice.

Cytotoxic T-cell precursors have been shown to occur in spleens of 2-3-day-old mice. By 12 days after birth, the cytotoxic T-cell response of spleen cells to alloantigens has reached 23-32% of adult levels. Addition of extra T-helper cells did not permit cytotoxic T-cell development in spleen cells from newborn to 2-day-old mice suggesting either a lack of precursors or suppression of precursors. The ontogeny of cell-mediated immune functions has thus been shown to correlate well with other work on the development of humoral immunity, accessory cells, and graft versus host reactivity.

Cytotoxic T cell response to minor histocompatibility antigens: apparent lack of H-2 restriction in killers stimulated by membrane fragments.

The secondary cytotoxic T cell response of BALB/c to B10.D2 or DBA/2 minor histocompatibility antigens in vitro requires the participation of an adherent cell. Nylon wool-passed spleen cells were only able to respond to nonadherent intact stimulator cells, or to membrane fragments derived from those cells, if a syngeneic adherent cell were present in the cultures. When the H-2 restriction properties of cytotoxic cells generated in response to various types of stimulation were analyzed, it was found that responses to B10.D2 or DBA/2 intact cells were always H-2 restricted. Responses to syngeneic adherent cells presenting B10.D2 or DBA.2 freeze-thaw antigen were either entirely or predominantly lacking in H-2 restriction as defined by efficient competition by B10 (H-2b) cold target cells. These unrestricted killers appeared to recognize minor histocompatibility as an independent determinant rather than as an H-2d/minors moiety cross-reaction with H-2b, because they were not absorbed by BALB.B (H-2b) macrophage monolayers, but were absorbed by B10 monolayers. Similarly, B10 but not BALB.B cold targets were able to compete for the anti-B10.D2 killers. These experiments eliminate the possibility that the lack of restriction was due to an H-2b restricted receptor cross-reactive with H-2b. Possible models to explain these findings are discussed.

In vitro generation of antigen-specific helper T cells that collaborate with cytotoxic T-cell precursors.

Antigen-specific helper T cells are required in the generation of cytotoxic T cells from thymocyte precursors. We have demonstrated that these alloantigen-specific helper cells can be generated in vitro and that both the quantity and quality of the helpers appear to be superior to the help obtained from unprimed spleen cells. Optimal helper cell activity is produced at day two of culture when CBA splenic helper precursors are stimulated by irradiated allogeneic spleen cells. Helper cell precursors are antigen-specific cells which cannot be instructed to express forbidden receptor specificities and bear theta antigen on their surface. The helper effectors are radioresistant, theta-bearing, and antigen-specific cells.

Tolerance induction during ontogeny. I. Presence of active suppression in mice rendered tolerant to human gamma-globulin in utero correlates with the breakdown of the tolerant state.

A specific state of T- and B-cell tolerance to human gamma-globulin (HGG) was induced in utero by intravenous administration of the deaggregated antigen to pregnant BALB/cCr mice. Tolerance persisted in the offspring until the 12th-wk of age and then began to gradually disappear. Suppressor cells could only be found when responsiveness to HGG ultimately appeared in the in utero-treated animals but not when they were completely unresponsives. In contrast, HGG-specific suppressors found in animals made unresponsive to HGG as adults appear to be associated with either the establishment and/or maintenance of the unresponsive state. To the extent that these experiments are consistent with natural self-tolerance to a serum protein, we conclude that active suppression is not a prerequisite from maintenance of unresponsiveness to self.

Helper T cells are required for the polyclonal stimulation of cytotoxic T cells by concanavalin A.

Concanavalin A stimulation of T-cell cytotoxicity has been shown to be absolutely dependent on helper T-cell collaboration. Thymocytes stimulated with ConA do not differentiate to yield cytotoxic effector cells. However, thymocytes cocultured with irradiated spleen cells as helpers and ConA yield high levels of cytotoxicity. The helper cell bears theta antigens on its surface, is not an adherent cell, and does not require any adherent cell functions in our culture conditions. The ConA-dependent helper cells appear to be polyclonal in specificity. Thus, polyclonal stimulation of cytotoxicity by ConA requires T helper-T precursor collaboration in analogy to antigen-specific T helper-T precursor interactions. Unlike the antigen-specific interacitons, the ConA-driven cytotoxicity does not appear to require linked associative recognition for induction of cytotoxicity.

Abnormal function of B lymphocytes from peripheral blood of multiple myeloma patients. Lack of correlation between the number of cells potentially able to secrete immunoglobulin M and serum immunoglobulin M levels.

Multiple myeloma patients are deficient in normal polyclonal serum immunoglobulins. To determine the reasons for this decrease, we quantitated and compared the number of surface IgM+ B lymphocytes, and the number of B cells susceptible to transformation by Epstein-Barr virus (EBV) with the concentration of IgM in serum. Serum IgM levels varied considerably in individual patients, temporally shifting from undetectable to normal amounts and then dropping again to undetectable levels. A transient rise to normal serum IgM concentrations was seen in 42% of patients assessed at two or more time points. Of 44 patients, 52% showed a lack of correlation between the number of surface IgM+ (sIgM+) B cells and serum IgM concentration. One subset of patients (25%) had detectable to normal numbers of sIgM+ B cells in blood but undetectable levels of serum IgM. Transformation of B cells from these patients indicated a block in IgM secretion that was extrinsic to the B cells that were fully able to transcribe, translate, and secrete IgM after EBV transformation. A second subset of patients (27%) had undetectable numbers of sIgM+ B cells but near normal levels of serum IgM, suggesting abundant secretion by few clones of B cells. Of 18 patients with monoclonal gammopathy of undetermined significance (MGUS), 26% showed a lack of correlation between the numbers of sIgM+ B cells and serum IgM concentration. We suggest that in patients with multiple myeloma, and in some with MGUS, there exists a mechanism(s) extrinsic to the B cell that mediates an arrest in terminal B lymphocyte maturation.

Severe deficiency of B lymphocytes in peripheral blood from multiple myeloma patients.

A major problem in the assessment of circulating B lymphocytes in multiple myeloma is the extent to which cells with passively absorbed Ig contribute to the assay. We have analyzed peripheral blood B cell numbers in 51 patients in various treatment categories by using an assay that is not subject to artifacts involving cytophilic Ig. We have defined a B lymphocyte by three different criteria (a) expression of a high surface density of Ig (b) expression of a high density of HLA.DR and (c) expression of a marker exclusive to surface Ig+ B cells. By these criteria, normal individuals have an average of 6% B cells. In multiple myeloma patients, B cell levels in purified mononuclear cell preparations are severely reduced. Untreated patients and the majority of patients on intermittent chemotherapy have 20-600-fold fewer B cells than do normal donors (average = 0.3%). This decrease was even greater in whole blood of patients as compared with normal donors (100-1,000-fold fewer B cells). The number of B cells did not correlate with disease status or paraprotein concentration. We found no evidence to support the idea that B lymphocytes in patients include a substantial monoclonal subset.

Adhesion of multiple myeloma peripheral blood B cells to bone marrow fibroblasts: a requirement for CD44 and alpha4beta7.

We have earlier described the presence of phenotypically unusual monoclonal B cells within the peripheral blood of multiple myeloma (MM) patients. To determine the biological properties of these B cells as compared to B cells from normal donors, we investigated the potential of CD19+ MM blood B cells to adhere to endothelial cell and bone marrow (BM)-fibroblast monolayers. We find that 30-60% of freshly isolated CD19+ MM blood B cells adhere to endothelial cell monolayers, and 50-80% adhere to BM fibroblast monolayers. The adhesion of MM blood B cells to either monolayer was not increased by in vitro activation, suggesting that these cells were activated in vivo. In contrast, fewer than 10% of CD19+ B cells from peripheral blood of normal donors adhered. Function-blocking monoclonal antibodies (mAbs) were used to determine which adhesion receptors were involved in CD19+ MM blood B cell interaction with BM fibroblasts. mAbs against very late antigen 4, the beta7-integrin subunit, and CD44, but not mAbs against very late antigen 5 and beta1, inhibited adhesion 61, 50, and 30%, respectively. The lack of inhibition with mAbs against beta1 implicates alpha4beta7 but not alpha4beta1 in adhesion of CD19+ MM blood B cells. To determine the alpha4beta7 ligand that mediated MM blood B cell adhesion, mAbs against vascular cellular adhesion molecule 1 and fibronectin, as well as CS1 and RGD peptides, were used as inhibitors. These were unable to reduce the adhesion of CD19+ MM blood B cells to BM fibroblasts, suggesting that fibronectin and vascular cellular adhesion molecule 1 are not involved in adhesion. Also, adhesion of MM blood B cells to mucosal addressin cell adhesion molecule 1-transfected Chinese hamster ovary cells was not enhanced compared to control-transfected Chinese hamster ovary cells, suggesting that mucosal addressin cell adhesion molecule 1 was not promoting adhesion of these cells. These data implicate CD44:HA interactions, as well as alpha4beta7 and an as yet unidentified ligand in the adhesion of in vivo activated MM blood B cell adhesion to BM fibroblasts. The adhesion properties of MM CD19+ B cells distinguishes them from normal B cells. Although the malignant status of these cells is as yet undefined, their adhesion properties implicate MM blood B cells in migratory spread of the disease.

In vitro and in vivo targeting of immunoliposomal doxorubicin to human B-cell lymphoma.

The ability to selectively target liposomal anticancer drugs via specific ligands against antigens expressed on malignant cells could improve the therapeutic effectiveness of the liposomal preparations as well as reduce adverse side effects associated with chemotherapy. Long-circulating formulations of liposomes containing lipid derivatives of poly(ethyleneglycol) [sterically stabilized liposomes (SLs)] have been described previously, and new techniques have recently been developed for coupling monoclonal antibodies (Abs) at the poly(ethyleneglycol) terminus of these liposomes. Ab-targeted SLs [immunoliposomes (SILs)] containing entrapped anticancer drugs are predicted to be useful in the treatment of hematological malignancies such as B-cell lymphomas or multiple myeloma, in which the target cells are present in the vasculature. The specific binding, in vitro cytotoxicity, and in vivo antineoplastic activity of doxorubicin (DXR) encapsulated in SILs coupled to monoclonal Ab anti-CD19 (SIL[anti-CD19]) were investigated against malignant B cells expressing CD19 surface antigens. Binding experiments with SIL[anti-CD19] resulted in a 3-fold higher association of the SILs with a human CD19+ B lymphoma cell line (Namalwa) in comparison with nontargeted SLs. Using flow cytometry, fluorescently labeled SIL[anti-CD19] bound to B cells with no recognition of T cells in a mixture of B cells and T cells in culture. Nontargeted SLs demonstrated significantly lower recognition of either B cells or T cells. Targeted DXR-SIL[anti-CD19] displayed a higher cytotoxicity to B cells relative to DXR entrapped in nontargeted SLs. Therapeutic experiments in severe combined immunodeficient mice implanted with Namalwa cells by the i.v. or i.p. routes resulted in significantly increased effectiveness of DXR-SIL[anti-CD19] compared to similar amounts of free DXR, DXR-SL (no Ab), or isotype-matched nonspecific Abs attached to DXR-SL. Single doses (3 mg/kg) of DXR-SIL[anti-CD19] administered i.v. resulted in a significantly improved therapeutic benefit, including some long-term survivors. From our results, we infer that targeted anti-CD19 liposomes containing the anticancer drug DXR may be selectively cytotoxic for B cells and may be useful in the selective elimination of circulating malignant B cells in vivo.

The breast mucin MUCI as a novel adhesion ligand for endothelial intercellular adhesion molecule 1 in breast cancer.

The MUC 1 mucin is expressed on normal breast epithelium and in 90% of breast cancers. We report here that tumor-associated MUC1 is a ligand for intercellular adhesion molecule 1 (ICAM-1). Antibodies to ICAM-1 and to MUC1 inhibited adhesion of human and transfected mouse MUC1-positive cell lines to human umbilical vein endothelial cell monolayers and immobilized recombinant human ICAM-1-immunoglobulin fusion protein. Purified MUC1 pretreatment of recombinant human ICAM-1 was an equally effective inhibitor of adhesion. The interaction between MUC1 and ICAM-1 may be critical to the process of bloodborne metastases in breast cancer.

Selective targeting of immunoliposomal doxorubicin against human multiple myeloma in vitro and ex vivo.

Circulating malignant CD19(+) B cells have been implicated in the pathogenesis and relapse of multiple myeloma (MM). This study investigated the therapeutic applicability of using long-circulating liposome-encapsulated doxorubicin (DXR) targeted against the internalizing CD19 antigens present on human MM cells. In vitro binding studies using the CD19(+) MM cell line ARH77 demonstrated that CD19-directed immunoliposomes (SIL[anti-CD19]) specifically attached to these cells. Formulations of immunoliposomal doxorubicin (DXR-SIL[anti-CD19]) showed a higher association with, and higher cytotoxicity against, ARH77 cells than did non-targeted liposomal doxorubicin (DXR-SL) or isotype-matched controls (DXR-NSIL[IgG2a]). By using the pH-sensitive fluorophore, 1-hydroxypyrene-3,6, 8-trisulfonic acid, binding of SIL[anti-CD19] to CD19 antigens was shown to trigger receptor-mediated internalization of the antibody-antigen complexes into endosomes. Targeting of SIL[anti-CD19] to CD19(+) B cells was also demonstrated in a heterogeneous mixture of peripheral blood mononuclear cells (PBMC) from MM patients. A decrease in cellular DNA (which is an indicator of apoptosis) caused by the cytotoxicity of DXR-SIL[anti-CD19] to myeloma PBMC was determined by using flow cytometry. While PBMC treatment with free DXR resulted in non-specific cytotoxicity to both B and T cells, DXR-SL were only minimally cytotoxic to either. In contrast, DXR-SIL[anti-CD19] were selectively cytotoxic for B cells in PBMC, indicating that this treatment may be effective in eliminating circulating malignant B cells in MM patients.

The blood B-cells and bone marrow plasma cells in patients with multiple myeloma share identical IgH rearrangements.

Previous reports have described the phenotypic and functional properties of monotypic late stage B cells in the blood of patients with multiple myeloma and have speculated that these B cells represent a malignant circulating component of myeloma. Here we show that blood B cells have IgH rearrangements identical to those expressed by the bone marrow plasma cells by using Ig Fingerprint and Allele-Specific Oligomer (ASO) polymerase chain reaction (PCR) methods. DNA from purified blood B cells and bone marrow plasma cells taken at the same time, and blood B cells taken at subsequent patient visits was amplified using consensus IgH primers, or ASO primers. In 10/16 patients, a single IgH rearrangement was amplified from the bone marrow plasma cells. In all 10 of those patients the same clonotypic rearrangement was amplified from the purified blood B cells. The relationship of these clonal blood B cells to the malignant bone marrow plasma cells remains undetermined.

Drug resistance in multiple myeloma: cyclosporin A analogues and their metabolites as potential chemosensitizers.

The malignant clone in myeloma is not eradicated by chemotherapy. Cyclosporins inhibit drug transport mechanisms, particularly the multidrug transporter p-glycoprotein 170, leading to their use as chemosensitizers. In myeloma, clonotypic blood B cells represent the major drug-resistant subset. This study compares the ability of cyclosporin A analogues and metabolites to inhibit cellular transporter(s) in myeloma and normal B cells in vitro, and evaluates their potential role in vivo. Cyclosporin A (CsA), CsG, PSC 833 or SDZ 280-446, and primary CsA and CsG metabolites, were tested for their ability to inhibit drug transport mechanisms of ex vivo malignant B cells from 81 patients with multiple myeloma as compared to B cells from normal donors, as measured by the export of the dye rhodamine 123 (Rh123) using multiparameter flow cytometry. The majority of myeloma B and normal B cells had efficient transporter function as measured by their CsA-sensitive export of Rh123. CsA and CsA analogues mediated efficient inhibition of this transport. Inhibition of dye transport by normal B cells required an approximately six-fold greater concentration of the synthetic peptolide SDZ 280-446 than was needed to optimally inhibit transport by myeloma B cells. PSC 833 and CsG were inhibitory at concentrations approximately five-fold lower than were required for CsA. Assessment of inhibitory potency in vivo indicated that the in vivo chemosensitizer levels of CsA and PSC 833 exceeded the transporter inhibitory concentration by four- and 20-fold respectively. In vivo, cyclosporins are rapidly and almost completely converted to metabolites. AM1 and AM4N, primary metabolites of CsA, mediated inhibition of transport, as did CsG metabolites GM1, GM4N and GM9. AM1 and GM9 are known to reach steady-state in vivo levels that exceed the inhibitory concentration identified here by 1.1- to 1.9-fold. Thus, cyclosporin metabolites, which accumulate in the blood during infusion of CsA and other cyclosporins, are shown here to be effective chemosensitizers for normally drug-resistant myeloma cells in vitro. Cyclosporin metabolites are considered to be less toxic than the parent drugs, suggesting that novel chemosensitization strategies designed to minimize concentrations of parent drug and maximize accumulation of primary metabolites in vivo may optimize cytotoxicity to the malignant clone in myeloma.

In multiple myeloma, circulating hyperdiploid B cells have clonotypic immunoglobulin heavy chain rearrangements and may mediate spread of disease.

DNA aneuploidy characterizes a proportion of malignant bone marrow (BM)-localized plasma cells in multiple myeloma (MM). This analysis shows that for most MM patients, circulating clonotypic B cells in MM are also hyperdiploid. Although all normal B cells and some malignant B cells are diploid, hyperdiploidy is likely to be exclusive to those that are malignant. Hyperdiploid MM B cells express CD34 and have clonotypic IgH transcripts, confirming them as part of the malignant clone. For MM, 92% (70/76) of patients had a DNA hyperdiploid subset [5-30% of peripheral blood mononuclear cells (PBMCs)] of CD19+ B cells. All CD19+ PBMCs in MM expressed CD19 and IgH variable diversity joining (VDJ) transcripts, confirming them as B cells. DNA aneuploid cells were undetectable in T or B lymphocytes from normal blood, spleen or thymus, or in blood from patients with B chronic lymphocytic leukemia. In MM, untreated patients had the highest DNA index (1.12). DNA hyperdiploid PBMCs were most frequent among untreated patients and were significantly reduced after chemotherapy. Diploid B cells were significantly more frequent after chemotherapy than at diagnosis. Of the hyperdiploid PBMCs, 81 +/- 3% expressed CD34 and CD19. In contrast to circulating CD34+ B cells, CD34- B cells in MM are diploid. In MM, unlike hyperdiploid PBMC B cells, hyperdiploid BM plasma cells lack both CD34 and CD19, suggesting that loss of CD34 correlates with differentiation and BM anchoring. In situ reverse transcription-PCR of the CD34+ (hyperdiploid) and CD34- (diploid) PBMC B-cell subsets was performed using patient-specific primers to amplify clonotypic IgH VDJ transcripts. Confirming previous work, CD34+ hyperdiploid MM PBMCs were clonotypic (86 +/- 5%). In contrast, CD34- diploid MM PBMCs had few monoclonal cells (4.8 +/- 2%). The lack of hyperdiploidy, together with the relative absence of cells having clonotypic transcripts, suggests these polyclonal CD34- B cells are normal. After culture in colchicine to arrest mitosis, hyperdiploid B cells were reduced and MM B cells accumulated in a diploid G2-M, suggesting that hyperdiploid in MM may represent a transient S-phase arrest rather than an aneuploid G0 phase. The DNA hyperdiploidy of CD34+ clonotypic B cells suggests these cells may be clinically important constituents of the myeloma clone and that they may play a direct role in the spread of myeloma.

During human thymic development, beta 1 integrins regulate adhesion, motility, and the outcome of RHAMM/hyaluronan engagement.

During human thymic differentiation, interactions between fibronectin (Fn)/beta1 integrins and hyaluronan (HA)/RHAMM control motility and Fn/beta1 integrins mediate spontaneous Fn-dependent adhesion. Multinegative (MN, CD3-4-8-) thymocytes exhibit strong spontaneous adherence to Fn (75%) that was efficiently inhibited by anti-alpha5beta1 and only weakly inhibited by anti-alpha4beta1. The relatively weak adherence of unfractionated thymocytes to Fn required both alpha4beta1 and alpha5beta1. Video time-lapse microscopy indicates that a subset of thymocytes also undergo spontaneous Fn-dependent motility mediated by alpha5beta1, alpha4beta1, and the HA-receptor RHAMM, but not by CD44. The loss of motility after hyaluronidase treatment of thymocytes indicated that motility is strongly dependent on HA. Of motile cells, 55% were DP, 19% were DN, and 24% were CD4+SP, but only 1% were CD8+SP. Overall, for MN thymocytes, beta1 integrin mediated Fn-adhesion, but after expression of CD4/CD8, beta1 integrins mediated Fn-dependent motility. Treatment with the activating anti-beta1 mAb QE.2E5 inhibited thymic motility and converted otherwise nonadherent thymocytes to an adherent state. High-avidity interactions via integrins appear to supercede the motogenicity of RHAMM and HA, suggesting that integrin avidity may regulate RHAMM. During thymic development, changes in adhesion or motility appear to be mediated by integrin avidity modulation.

Expression of IL-6 and IL-6 receptors by circulating clonotypic B cells in multiple myeloma: potential for autocrine and paracrine networks.

OBJECTIVE: To investigate the participation of clonotypic MM B cells in the IL-6 network in patients with multiple myeloma. METHODS: CD19(+) B cells from 45 patients with multiple myeloma and from 18 healthy donors were sorted and their expression of IL-6, IL-6 receptor (CD126) characterized by flow cytometry, in situ RT-PCR, and ELISA measurement of IL-6 and soluble IL-6R. Expression of CD31 was detected by flow cytometry. RESULTS: Interleukin-6 (IL-6) is a pleiotropic cytokine often overexpressed in multiple myeloma (MM). IL-6 induces growth and inhibits apoptosis of MM plasma cells, and upregulates the activity of osteoclasts. MM plasma cells, the most mature component of the MM clone, secrete IL-6 and induce IL-6 production from other cell types. However, the MM clone also includes circulating clonotypic B lymphocytes. Using ELISA and in situ RT-PCR we demonstrate here that, unlike the healthy control B cells, MM B cells express IL-6 mRNA and secrete IL-6 protein. In vitro, MM B cells were the major producers of IL-6 in peripheral blood mononuclear cells. On average, 50% of MM B cells express the IL-6 receptor (IL-6R, CD126), suggestive of autocrine stimulation. They also express CD31, potentially facilitating their paracrine interactions with osteoclast precursors. CONCLUSION: Secretion of IL-6 by circulating clonotypic B cells in MM may contribute to the autocrine and paracrine cytokine networks that maintain the malignant clone and are responsible for disruption of normal bone metabolism in this incurable disease.

Treatment of plasma cell dyscrasias by antibody-mediated immunotherapy.

The use of serotherapy to treat patients with plasma cell dyscrasias (PCDs) has been sought by us and others. Candidate antigens that have been targeted or proposed for targeting in PCDs include the immunoglobulin idiotype, CD19, CD38, CD54, CD126, HM1.24, and Muc-1 core protein. Unfortunately, many of these antigens are not ideal for use in serotherapy since they are not selectively expressed, are either shed or secreted, or have not been fully characterized. Serotherapy with an anti-CD19 monoclonal antibody (B4) conjugated to a blocked ricin toxin had no significant activity in patients with multiple myeloma (MM). Circulating CD20+ clonotypic B cells have been detected in the circulation of most MM and Waldenstrom's macroglobulinemia (WM) patients. Plasma cells from most WM patients express CD20, but most MM patient plasma cells either lack CD20 or express it weakly. In view of recent successes with anti-CD20-directed serotherapy in other B-cell malignancies, we initiated a phase II trial to study the anti-CD20 monoclonal antibody rituximab (Rituxan; IDEC Pharmaceuticals, San Diego, CA, and Genentech, Inc, San Francisco, CA) in patients with MM. We describe two PCD patients (one with WM and one with MM) who responded to therapy. By flow cytometric analysis, CD20+ plasma cells and B cells present in the bone marrow and peripheral blood of a patient with MM disappeared with response to rituximab therapy. However, residual CD20- tumor cells remained in the bone marrow following rituximab therapy, and after 6 months this patient progressed with CD20- myeloma cells. As a potential strategy to overcome this limitation, we demonstrated that interferon-gamma at pharmacologically achievable levels induced CD20 expression on these CD20- plasma cells, consistent with our recent findings that interferon-gamma is a potent inducer of CD20 expression on MM patient plasma cells and B cells. We also characterize a response to rituximab with a decrease in paraprotein and resolution of anemia in a patient with WM whose response to rituximab is ongoing after 19+ months. This preliminary experience supports the potential use of serotherapy targeting CD20 in PCDs. Our studies further suggest that interferon-gamma may enhance CD20 expression on MM plasma cells, thereby increasing their susceptibility to anti-CD20 monoclonal antibody therapies.

In vitro cytotoxic T cell responses to cell-free minor histocompatibility antigens. Requirement for antigen-presenting cells.

Although it has been documented that antigen-presenting cells (APC) are required for the response to minor histocompatibility antigens (MIHA) on membrane fragments, there is no evidence to indicate whether or not these APC actually process MIHA. In this manuscript we identify these APC as Ia+, Thy 1.2-, Ly6.1-, Ly7.2-, Ly5- adherent cells. They cannot be replaced by soluble adherent cell products, including interleukin-1 (IL-1), or by interleukin-2, and are required for at least the first 4 hr of a cytotoxic T lymphocyte (CTL) response. After this period, the response can continue in the absence of either adherent cells or the MIHA-bearing membrane fragments, indicating that the APC function is required for the initiation of a response. If APC are inactivated by heat treatment prior to exposure to the MIHA-bearing fragments, they are unable to contribute to the triggering of a CTL response. However, if these cells are inactivated by heat after exposure to MIHA, they are fully able to contribute to the induction of a CTL response. We interpret these results as an indication that metabolic activity is required for antigen uptake--but, once this has occurred, an inactivated APC is still able to present MIHA. This suggests the involvement of a requisite early antigen-processing event by APC that is distinct from the putative association of MIHA with H-2.

Regulation of the cytotoxic T cell response to minor histocompatibility antigens.

A system is described in which the regulation of the cytotoxic T cell response to minor histocompatibility antigens can be analyzed. Killer precursor cells are derived from spleen cells of mice previously primed in vivo with minor H antigens; at low cell numbers, the helper activity in primed spleen cell populations is diluted away leaving a population of killer precursor cells which is relatively deficient in helper activity. Addition to these cultures of irradiated helper cells which have been generated in vitro allows induction of a strong cytotoxic response to minor H antigens. This helper function is mediated by specifically induced radioresistant T lymphocytes; the helper cells may recognize Mls antigens. In the same system, culture of higher numbers of primed responder cells yields a cytotoxic response which can be inhibited by irradiated suppressor cells which have been generated in vitro. These suppressor cells are also specifically induced T lymphocytes capable of radioresistant function; they appear to be generated only from populations of primed spleen cells. The fact that in the above system both cytotoxic T cells and suppressor T cells are obtained only from primed cells, while helper cells may be generated from unprimed cells, suggests a difference between these cells in specificity, induction requirements, or both. The suppressor cells must be present within the first 24 to 48 hr of culture; after this point, the differentiating cytotoxic response becomes resistant to suppressive effects.