The Interaction of Nonmitogenic and Mitogenic Lectins with T Lymphocytes: Association of Cellular Receptor Sites

The relationship between the surface receptors on neuraminidase-treated human blood lymphocytes for the mitogenic lectins Phaseolus vulgaris leukoagglutnin (La), concanavalin A (Con A), and soy bean agglutinin (SBA) and the nonmitogenic lectin Helix Pomatia A hemagglutinin (HP) was investigated. Two different techniques, co-capping with different fluorochrome-labeled lectins and cell binding-inhibition experiments with 125 I-labled lectins, were used. The results demonstrated that the nonmitogenic lectin HP and the mitogenic lectins SBA, La, and Con A bind either to the same macromolecule(s) or to different but physically linked macromolecules on the surface of human T lymphocytes. In contrast, only part of β2-microglobulin (β2-m), or β2-m-bearing complexes, appear to be physically linked to the lectin receptor complex(es). On the lectin-binding substance(s) at least two saccharide structures were recognized, one of which binds both HP and SBA and another which hinds SBA and La (and probably also Con A) but not HP.     

Human serum amyloid P component binds to peripheral blood monocytes

The binding of human serum amyloid P component (SAP) to blood cells and monocyte-derived dendritic cells was investigated by flow cytometry. Monocytes bound biotinylated SAP with avidity in a dose-dependent and saturable manner. By contrast, the binding of SAP to monocyte-derived dendritic cells was weak. No binding to erythrocytes, NK cells, T lymphocytes or B lymphocytes could be detected. The SAP-monocyte interaction was calcium-independent and readily inhibited by C1q. SAP was nonmitogenic for human mononuclear cells and had no apparent influence on lymphocyte proliferation induced by mitogenic lectins. We speculate that binding of SAP by monocytes could be of physiological relevance at extravascular sites by influencing complement regulation.     

The Sheep Erythrocyte Receptor and Both α and β Chains of the Human T-Lymphocyte Antigen Receptor Bind the Mitogenic Lectin (Phytohaemagglutinin) from Phaseolus vulgaris

We have studied the interaction of mitogenic lectins such as phytohaemagglutinin (PHA) and concanavalin A (Con A) with both surface molecules which, by the use of monoclonal antibodies, are known to trigger T-cell mitogenesis. Monoclonal antibodies recognizing the T-lymphocyte receptor for antigen (Ti) and/or its associated structure, CD3, activate T cells. More recently, a second pathway of activation has been described which involves the sheep erythrocyte binding glycoprotein CD2, a surface molecule distinct from Ti-CD3. Lysates from surface-iodinated T-leukaemia cell lines were treated with lectin and affinity purified anti-lectin antibodies coupled to protein A-Sepharose. We have shown that eluates from Con A/anti-Con A or PHA/anti-PHA immunoprecipitates contained Ti, since a rabbit anti-T alpha serum, which recognizes the native and denatured forms of the constant region of the alpha chain, immunoprecipitated Ti from these eluates. Furthermore, Ti immunoprecipitated by anti-T alpha serum from lysates of surface iodinated E+ lymphocytes was binding to PHA after elution from the immunoprecipitate. When the purified Ti molecule was reduced and alkylated, allowing the permanent dissociation of its alpha and beta subunits, PHA interacted with both chains, whereas anti-T alpha serum immunoprecipitated the alpha chain only. Altogether, these results demonstrate that PHA interacts with both chains of the T cell receptor for antigen on human peripheral T lymphocytes. With the HPB-ALL tumour line, a similar approach showed that both alpha and beta chains of Ti bind to Con A and Ulex europaeus 1 but not Helix pomatia. Affinity chromatography on immobilized lectins and immunoprecipitation with lectin/anti-lectin antibodies were employed to test whether CD2 binds to PHA and Con A. The results show that CD2 from human peripheral T lymphocytes binds both lectins but with a lower affinity for PHA than Con A.     

Binding and redistribution of lectins on lymphocyte membrane

The redistribution of fluorochrome-conjugated agglutinins bound to the lymphocyte plasma membrane is investigated. Two “T” lymphocyte mitogens (concanavalin A (Con A) and phytohemagglutinin (PHA)) bind to and redistribute equally well (spotting and capping) on the lymphocyte, whether they are “B” (Ig⁺) or “T” (Ig^?). One B cell mitogen (pokeweed mitogen (PWM)) does not bind detectably either to T or to B lymphocytes, although it binds to macrophages and to thymocytes in a perfectly ring-like pattern. Its redistribution requires anti-PWM antibodies. The nonmitogenic poly-L-lysine (PLL) binds to virtually all lymphocytes, and rapidly forms caps. PLL binds more strongly to T cells than to B cells. The agglutinin concentration is critical for the formation of agglutinin caps on the cell surface, as well as for the free mobility of Ig receptors. Finally, lectin-coated beads pick up B and T cells equally well, irrelevantly of their mitogenicity. Cell attachment to agglutinin-coated beads does not induce any demonstrable capping of either agglutinin-binding sites or membrane Ig. When the cells are further treated with soluble agglutinins or anti-Ig antibodies, they form caps mainly oriented towards the lectin-coated beads. These results show the high mobility of lymphocyte membrane components, the independence of agglutinin-binding sites and of most of the Ig receptors, and a polarization of cell with regard to cap formation. They exclude any straightforward correlations between agglutinin binding, membrane component redistribution and lymphocyte triggering.     

Mitogenic lectins bind to the antigen receptor on human lymphocytes

The specificity of interactions between mitogenic and non-mitogenic lectins and disulfide-linked cell surface receptors on human lymphocytes was explored. Lysates (Nonidet-P40) of surface-radioiodinated tonsil lymphocytes and T lymphoblastoid cells (HPB-ALL) were absorbed with lectin-agarose derivatives (or bovine serum albumin, BSA-agarose) or immunoprecipitated with appropriate monoclonal antibodies (mAb). Lectin eluates and solubilized immunoprecipitates were analyzed by two-dimensional (nonreduced/reduced) sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Radiolabeled polypeptides were visualized by autoradiography. Among the various lectin-binding polypeptides, two disulfide-linked heterodimers (II and III) and two apparent homodimers (I and IV) are bound by pea lectin, concanavalin A and lentil lectin on tonsil lymphocytes; II, III and IV are bound both leukoagglutinating (L)- and erythroagglutinating (E)-phytohemagglutinins from Phaseolus vulgaris (PHA). Pokeweed mitogen recognizes only II and III. These molecules are weakly bound by peanut agglutinin, soybean agglutinin, Ulex europaeus agglutinin-I, Dolichos biflorus agglutinin, Vicia villosa agglutinin and Sophora japonica agglutinin, but are not bound by Helix pomatia agglutinin or BSA-agarose. Heterodimer II (82-88 kDa), comprised of 50-55-kDa and 40-43-kDa subunits, probably represents the alpha/beta T cell antigen receptor (TcR alpha/beta). Heterodimer III (64-72 kDa), comprised of 41-kDa and 37-kDa subunits, may represent TcR gamma. The homodimers, I (120-130 kDa) and IV (55-61 kDa), comprised of 55-60-kDa and 30-kDa polypeptides, respectively, have apparently not been previously described. Evidence that H1-2D4, a mAb directed against the antigen receptor on HPB-ALL cells, and E-PHA interact with a common molecule includes: (a) immunoprecipitation of TcR with H1-2D4 from the glycopeptide fraction specifically eluted from insolubilized lectin with N-acetylgalactosamine; and (b) adsorption of TcR from a solubilized H1-2D4 immunoprecipitate by E-PHA-agarose. Recognition of CD3 by E-PHA is indicated by immunoprecipitation of CD3 protein by UCHT1 from the glycopeptide fraction specifically eluted from E-PHA. The results are consistent with the view that mitogenic lectins interact with certain disulfide-linked molecules on human lymphocytes, including the TcR alpha/beta and perhaps TcR gamma; while some nonmitogenic lectins also recognize these receptors, the interaction is of low affinity.     

Lymphocyte-induced stimulation of the contractile response of the heart

In previous reports we have shown that phytohemagglutinin (PHA)-activated human lymphocytes had positive inotropic effects on spontaneously beating isolated rat atria. In this study, we demonstrated that the stimulatory effect on heart contractility induced by lymphocytes was linked to early events of lymphocyte activation by lectins. Active soluble factors were gradually released to the fluid phase. Similar results were obtained with both mitogenic (PHA) and nonmitogenic (WGA) lectins indicating that the stimulatory action of activated lymphocytes did not require cell division. Absence of Ca2+ inhibited both the generation of the stimulatory activity and lymphocyte proliferation. In contrast, verapamil, dexamethasone and low concentrations of cycloheximide eliminated only the appearance of the stimulatory effect.     

Non-specific effects of avian retrovirus co-incubation on lymphocyte function: abrogation of antigen- and mitogen-induced proliferative responsiveness.

Peripheral blood lymphocytes from chickens bearing tumours induced by avian retroviruses can be stimulated to divide by group-specific antigens present in supernatant fluids of avian retrovirus-infected but not normal chicken embryo fibroblast (CEF) cells. Centrifugation studies revealed that the relevant antigenic activity is non-virion in nature. Indeed, the presence of avian retrovirus particles was found to be inhibitory to the capacity of sensitized lymphocytes to be stimulated in this antigen-driven blastogenesis assay. Similar results were obtained in lymphocyte mitogenesis experiments in which any of peripheral chicken lymphocytes or mouse splenic, lymph node or thymic lymphocytes were co-incubated with either concanavalin A or phytohaemagglutinin in the presence of numerous types of virus particles. This inhibitory effect was not due to infection of lymphocytes by the viruses tested, and was obtained in the case of lymphocyte-virus combinations for which the cells lacked the surface receptors required for viral entry. Virus could be added to lymphocyte cultures as late as 26 h after co-incubation with mitogen, and still inhibit the usual mitogenic response. In addition, co-addition of virus to lymphocytes in the presence of concanavalin A was found to block the capping of ligand-bound receptors which normally ensues. Pre-added virus did not, however, affect the ability of lectins to bind to cells.     

Characterization and mitogenic activity of Haemophilus influenzae type B capsular polysaccharide.

Studies were conducted on the characterization of Haemophilus influenzae type b polysaccharide (HITB-PS) and its mitogenic activity upon peripheral lymphocytes. This capsular polysaccharide was found to contain hexosamines and hexoses in addition to the main components of ribose and ribitol phosphate. The molecular weight of HITB-PS was determined as 585,000. The affinity constant of HITB-PS to unfractionated lymphocytes was 3.13 X 10(3) M-1 with 1.11 X 10(4) binding sites per cell. HITB-PS was found to be mitogenic for both human T and B lymphocytes. At optimum doses, a three to five fold increase in 3H-thymidine incorporation into T and B cells was observed. Higher than optimum doses resulted in suppression of this mitogenicity. The effect of concanavalin A (Con A) mitogenicity was detected in T and B cells treated with effective as well as suppressive doses of HITB-PS; the mitogenic activities of Con A and HITB-PS were found to be independent of each other.     

Cooperative pathway induction of T lymphocyte mitogen stimulation.

Isolated peritoneal mouse macrophages pretreated with the mitogenic enzyme combination neuraminidase (EC 3.2.1.18) plus galactose oxidase (EC 1.1.3.9.) (NAGO), or with NaIO4, stimulate macrophage-depleted lymphocytes mitogenically by a macrophage-derived signal, different from the originally used mitogen. Polyethylene glycol (PEG) treatment of the cultures, although itself nonmitogenic, strongly enhances the mitogenic response of the lymphocytes. Under culture conditions the macrophage-derived signal is transmitted to lymphocytes by direct cell contact, a finding which explains the need of a critical cell density for T lymphocyte stimulation. In the absence of macrophages, lysates from mitogen-preteated macrophages stimulate column-purified lymphocytes in the presence of PEG. Our results indicate that mitogenic activation of lymphocytes is mediated through two sequential triggering events, induction (by mitogen treatment) of a macrophage-derived signal and commitment (by nonmitogenic PEG treatment) of lymphocytes to react to the signal. Reconstitution of the mitogenic response can be achieved by a sequential induction of both these events.     

Mechanism of T cell activation. I. A screening of "step one" ligands

A number of ligands which bind to T cell surfaces were screened on murine spleen cells for their functional ability to induce either of the two necessary steps in the process of T cell triggering: (a) expression of growth receptors by T lymphocytes and (b) production of T cell growth factors in cultures containing both T cells and accessory cells. Among four lectins tested, concanavalin A was found to induce both responses but is primarily a “step 2” ligand, in particular at low concentrations; leucoagglutinin was the most potent “step 1” ligand with very limited “step 2” activity; wheat germ agglutinin and soybean lectin were inactive. Direct mitogenicity of the lectin to normal spleen cell cultures was limited by the poorest of these two functional properties. A number of rabbit anti-lymphocyte antisera, anti-Thy-1 alloantisera, anti-H-2 and anti-Ia monoclonal antibodies, as well as complexes of rabbit IgG, were all found to be devoid of “step 1” activity, and consequently, non- mitogenic for spleen cells. In contrast, a rabbit anti-mouse brain antiserum, although also devoid of mitogenicity, was found to be capable of “step 1” induction. These results suggest that not all rearrangements of T cell surface membrane components upon binding of a ligand result in the functional expression of growth receptors. This appears to be the result of selective interactions with specific sites which include structures binding to concanavalin A, leucoagglutinin and rabbit anti-mouse brain antibodies.     

Human monocyte-lymphocyte interaction and its enhancement by levamisole.

We investigated the role of monocyte-lymphocyte interaction in the transformation of human peripheral blood lymphocytes by the mitogen concanavalin A (Con A). Human monocytes were separated from lymphocytes and were transiently exposed to Con A. The Con A-pretreated monocytes were able to subsequently bind autologus lymphocytes by a process that was selective for T cells. This interaction required the initial presence of Con A at the monocyte surface, and became independent of surface bound ligand after 72 hr. Levamisole, an agent thought to facilitate the participation of monocytes in the cellular immune response, enhanced the binding of lymphocytes to monocytes at low concentration of Con A (5--10 micrograms/ml). Levamisole did not lead to mitogen independent lymphocyte binding. The association between lymphocytes and Con A-pretreated monocytes resulted in the mitogenic transformation of lymphocytes in the absence of soluble Con A in the medium. These results suggest that, in addition to any possible soluble mediators, direct lymphocyte-monocyte contact is required for optimal mitogenic transformation. This T-cell-monocyte interaction over time becomes independent of cell-surface mitogen. The ability of levamisole to enhance this interaction may explain levamisole's capacity to stimulate lymphocyte proliferation.     

Characterization and Mitogenic Activity of Haemophilus Influenzae Type B Capsular Polysaccharide

Studies were conducted on the characterization of Haemophilus influenzae type b polysaccharide (HITB-PS) and its mitogenic activity upon peripheral lymphocytes. This capsular polysaccharide was found to contain hexosamines and hexoses in addition to the main components of ribose and ribitol phosphate. The molecular weight of HITB-PS was determined as 585,000. The affinity constant of HITB-PS to unfraction-ated lymphocytes was 3.13 × 10³ M^-1 with 1.11 × 10⁴ binding sites per cell.

HITB-PS was found to be mitogenic for both numan T and B lymphocytes. At optimum doses, a three to five fold increase in ³H-thymidine incorporation into T and B cells was observed. Higher than optimum doses resulted in suppression of this mitogenicity. The effect of concanavalin A (Con A) mitogenicity was detected in T and B cells treated with effective as well as suppressive doses of HITB-PS; the mitozenic activities of Con A and HITB-PS were found to be independent of each other.     

Stimulation of human lymphocyte subpopulations by protein A from Staphylococcus aureus

Protein A from Staphylococcus aureus is known to stimulate human lymphocytes. Using 3H-thymidine incorporation, virus plaque assay and induction of cytotoxic T lymphocytes (CTL), this study showed that soluble or insoluble protein A stimulated different lymphocyte subpopulations. Soluble protein A is highly mitogenic to T lymphocytes. In both 3H-thymidine incorporation and virus plaque assay, its maximum stimulation was as high as the stimulation by the nonspecific mitogens phytohemagglutinin and concanavalin A, and a higher CTL response than that induced by phytohemagglutinin or concanavalin A was induced. Mitogenic activity to B lymphocytes was negligible. S. aureus (Cowan I strain) is itself considered to be an insoluble form of protein A and is 3--4 times more mitogenic to B lymphocytes than pokeweed mitogen without any increase in virus plaque-forming cells. No mitogenicity was noted to T lymphocytes. Sepharose CL-4B-protein A, also known as insoluble protein A, stimulated both T and B lymphocytes effectively, but its mitogenicity to T lymphocytes was considered to be due to the soluble protein A released from the Sepharose CL-4B beads.     

Characterization and Mitogenic Activity of Haemophilus Influenzae Type B Capsular Polysaccharide

Studies were conducted on the characterization of Haemophilus influenzae type b polysaccharide (HITB-PS) and its mitogenic activity upon peripheral lymphocytes. This capsular polysaccharide was found to contain hexosamines and hexoses in addition to the main components of ribose and ribitol phosphate. The molecular weight of HITB-PS was determined as 585,000. The affinity constant of HITB-PS to unfraction-ated lymphocytes was 3.13 × 10³ M^?1 with 1.11 × 10⁴ binding sites per cell.

HITB-PS was found to be mitogenic for both numan T and B lymphocytes. At optimum doses, a three to five fold increase in ³H-thymidine incorporation into T and B cells was observed. Higher than optimum doses resulted in suppression of this mitogenicity. The effect of concanavalin A (Con A) mitogenicity was detected in T and B cells treated with effective as well as suppressive doses of HITB-PS; the mitozenic activities of Con A and HITB-PS were found to be independent of each other.     

Activation of T lymphocytes by lectins and carbohydrate-oxidizing reagents viewed as an immunological recognition of cell-surface modifications seen in the context of "self" major histocompatibility complex antigens

The experiments reported here were directed at the question of whether polyclonal activation of T cells by mitogenic substances results as a consequence of (a) a truly antigen-independent stimulation of T cells through mitogen receptors, bypassing requirements for T cell-specific recognition or (b) an immunological reaction, similar in principle and requirements to T cell activation induced by antigen. Several lines of evidence have provided strong support for the latter interpretation. First, the ability of several mitogenic substances to induce polyclonal T cell activation was found to be crucially dependent upon cell surface expression of major histocompatibility complex (MHC) products. Thus, the presence of alloantisera or monoclonal antibodies directed against either class I (SD) determinants or class II (Ia) products on the responding cell population resulted in a sizeable dose-dependent inhibition of the mitogenic response. Additional studies using spleen cells from normal F₁ mice or F₁→parent bone marrow chimeras provided the most direct and convincing evidence for a specific immunological role of MHC products in mitogenic T cell activation. Here, the concanavalin A response of normal F₁ T cells was found to be inhibited by alloantisera against either parental haplotype, while the H-2-heterozygous F₁→parent chimeras were only inhibited by antisera against the H-2 of the parent in which they matured and possessed functional, MHC-restricted helper T cells. The inherent characteristics and reactivities which endow only certain lectins to exert a mitogenic effect were specifically examined in light of the observed dependence for MHC expression in these mitogenic reactions. Using a panel of 20 different lectins as affinity absorbents for cell surface glycoproteins, it was found that only mitogenic lectins were capable of binding to cell surface-expressed H-2 products (both K/D and la). The results obtained here are discussed in support of the concept that mitogeninduced T cell responses involve an immunological recognition of cell surface modifications seen in the context of self H-2 antigens.     

Transferrin Receptors on Mitogen-Stimulated Human Thymus-Derived Lymphocytes

The appearance of transferrin receptors on mitogen-stimulated human thymus-derived (T) lymphocytes was studied. When indirect immunofluorescence with immunoadsorbent-purified antitransferrin antibodies was used, approximately 10% of resting T cells were stained. This proportion increased to 50-80% of the cells 3-4 days after stimulation with the mitogenic lectins concanavalin A (Con A) and leucoagglutinin (La) from Phaseolus vulgaris. Almost all blast cells (greater than or equal to 90%) were positive. Cell binding experiments with 125I-labelled transferrin indicated the presence of 1-5 x 10(5) transferrin receptor molecules/cell with high avidity for transferrin (K = 2 - 12 x 10(8) l/mol). Analysis by sodium dodecyl sulphate polyacrylamide gel electrophoresis and autoradiography of cell lysates containing 125I-labelled T-cell surface components revealed two surface peptides (90 kdaltons and 80 kdaltons, reducing conditions), which selectively bound to insolubilized antitransferrin antibodies. The 90-kdalton peptide also bound to insolubilized transferrin. The 80-kdalton peptide is most probably transferrin and the 90-kdalton peptide the transferrin receptor. Unreduced transferrin receptor had a molecular weight of 180 kdalton. It is probably a glycoprotein, since it reacted with wheat germ agglutinin, La, and probably also Con A. The properties of the lymphocyte transferrin receptor are similar to those described for transferrin receptors on various in-vitro-grown transformed cells. This speaks in favour of a common receptor present on all proliferating human cells.     

Lymphocyte cell surface glycoproteins which bind to soybean and peanut lectins.

In cellular immunology, peanut (Arachis hypogaea) lectin has been used to selectively agglutinate immature lymphoid cells and soybean (Glycine max-lectin to agglutinate B lymphocytes. We have used affinity chromatography to study the surface glycoproteins of rat and mouse lymphoid cells which bind to these lectins. Thymocyte and T and B lymphocyte glycoproteins were analysed either without modification (native) or after the removal of sialic acid with neuraminidase (asialo). The only native glycoprotein which was seen to bind to peanut lectin was the 95,000 mol. wt sialoglycoprotein from thymocytes. The equivalent molecules from T lymphocytes bound to peanut lectin only after neuraminidase digestion. Thus the selective agglutination of thymocytes by peanut lectin would seem to be due to a partial lack of sialic acid residues on the O-glycosidically-linked oligosaccharides of the thymocyte sialoglycoprotein. The B lymphocyte form of the leucocyte-common antigen was the only prominent native glycoprotein which was seen to bind to soybean lectin and this probably accounts for the specific binding of this lectin to B cells. The leucocyte-common antigens, in their asialo forms, from thymocytes and B and T lymphocytes differed in their binding to the lectins and this establishes that these glycoproteins which share antigenic determinants differ in their carbohydrate structures.     

Binding of bacterial endotoxin to murine spleen lymphocytes.

The early events in lipopolysaccharide (LPS)-induced B-cell activation were investigated by studying the binding of 14C-labeled LPS to murine lymphocytes in vitro. In these studies we utilized intrinsically labeled 14C-labeled LPS from Salmonella minnesota or the 14C-labeled glycolipid derived from the Re mutant of S. minnesota (R595). Bone marrow-derived (B) lymphocytes bound more LPS than did thymus-derived (T) lymphocytes. Binding of LPS to murine spleen lymphocytes from strain C3H/HeN was compared with the binding to spleen lymphocytes from strain C3H/HeJ, a strain resistant to certain biological activities of LPS including mitogenesis. Spleen cells from both strains bound LPS equally well, suggesting that unresponsiveness of C3H/HeJ mice to LPS is due to factors other than a defect in binding of LPS. LPS binding to cells appeared to be due to a nonspecific interaction between the lipid moiety of LPS and the lipid components of the cell membrane. Thus, the highly lipophilic, polysaccharide-deficient glycolipid from R595 bound at least 20 times better than did LPS. Furthermore, partial removal of cell surface proteins with trypsin or sialic acids with neuraminidase enhanced glycolipid binding, suggesting that binding is not through a protein- or sialic acid-containing receptor. The binding of glycolipid to lymphocytes was only partially specific since unlabeled glycolipid R595, lipid A, and LPS did not completely inhibit the uptake of 14C-labeled glycolipid R595. In addition, binding could be inhibited by a nonmitogenic phospholipid (phosphatidyl ethanolamine), which also is consistent with a nonspecific lipid-lipid interaction. Experiments were performed to determine the relationship of LPS binding to lymphocyte activation in the lymphocytes. The process of activation of lymphocytes by LPS was a slow one, since LPS was required to be present in culture for at least 24 h in order to obtain significant lymphocyte activation, suggesting that the amounts of LPS bound earlier are either quantitatively or qualitatively insufficient to irreversibly activate the cell.     

T cell activation by anti-T3 antibodies: comparison of IgG1 and IgG2b switch variants and direct evidence for accessory function of macrophage Fc receptors

The human T3 antigen is closely associated with the T cell receptor. Some anti-T3 antibodies cause T cell proliferation in the presence of monocytes which have Fc receptors (FcR) that bind particular antibody subclasses. Such an interaction is thought to determine whether or not an anti-T3 antibody is mitogenic. We examined the mitogenicity of an IgG1 antibody, UCHT1, and an IgG2b switch variant of identical specificity, UCHT1B. With autologous monocytes, 76% of individuals responded to UCHT1 and 9% to UCHT1B, falling into three patterns of responsiveness. Both antibodies in the absence of monocytes induced responsiveness to recombinant interleukin 2, even for UCHT1B nonresponder T cells. The proliferation induced by UCHT1B, however, was always less than that induced by UCHT1. These findings demonstrate the critical role played by the Fc region for mitogenesis, and suggest a possible role for the hinge region. We then obtained direct evidence that mitogenicity can be mediated exclusively via FcR. Mouse macrophages have distinct FcR: FcRI binds IgG2a but FcRII binds IgG1 and IgG2b and its function can be inhibited by the specific antibody 2.4G2. Because UCHT1 and UCHT1B were of the correct subclass to interact with FcRII we examined the accessory function of mouse peritoneal macrophages. Without exception, human T cells now responded to both antibodies. Proliferation was drastically inhibited by 2.4G2 but not by an irrelevant anti-macrophage antibody, F4/80, nor by an anti-human neutrophil FcR antibody, 3G8. Furthermore, 2.4G2 did not inhibit the accessory function of mouse macrophages for OKT3, an IgG2a antibody that presumably interacts with FcRI, and did not inhibit the function of human monocytes for UCHT1 and UCHT1B. Mouse B cells, in contrast to macrophages, have an FcR which binds all three subclasses, but which can be inhibited by 2.4G2. B cells, however, were not accessory cells for mitogenesis with UCHT1, UCHT1B or OKT3. These findings are discussed in relation to other requirements for T cell activation by anti-T3 antibodies.