Activated B lymphocytes: stimulators of an augmented autologous mixed leukocyte reaction

The characteristics of the non-T cell(s) which stimulate T-lymphocyte proliferation in the autologous mixed leukocyte reaction (AMLR) have been at issue since this in vitro reaction was first described. Dendritic cells have been shown to be the most potent stimulator cells, but B cells, null cells, and macrophages have also been demonstrated to have the capacity to stimulate autologous T-cell proliferation. A cell preparation obtained from human peripheral blood was highly enriched for surface immunoglobulin-positive B cells. These cells were activated by brief culture with various B-cell mitogens and then compared to untreated B cells with regard to stimulatory activity in the AMLR. Mitogen-activated B cells were markedly augmented in their capacity to stimulate autologous T-cell proliferation when compared with untreated B cells. Fractionation of the B-cell preparation into high- and low-density subpopulations demonstrated that the high-density cells, enriched in resting B cells, had minimal stimulatory activity but could be activated to have increased AMLR-stimulatory capacity. Proliferation of the activated B lymphocytes was not required for the generation of the augmented AMLR. Response to both untreated and mitogen-activated B cells was a property of T4-positive T lymphocytes. The increase in stimulatory capacity was associated with a decrease in cell surface immunoglobulin, but no significant alteration in the percentage or fluorescence intensity of anti-Ia staining cells was detected. Activated B cells which are generated in vivo may acquire the capacity to generate T effector cells or factors important in the regulation of B-cell function.     

Autologous mixes lymphocyte reaction in human cord blood lymphocytes: decreased generation of helper and cytotoxic T-cell functions and increased proliferative response and induction of suppressor T cells

T lymphocytes from neonates proliferated significantly more than peripheral blood T lymphocytes from adults in autologous mixed lymphocyte reactions (AMLR). AMLR-activated cord, as compared to adult T lymphocytes, exerted significantly less nonspecific cytotoxic activity on PHA-stimulated adult mononuclear cells and Epstein-Barr virus-transformed target cells. The impaired generation of cytotoxicity of cord T cells was not corrected by Interleukin-2. Blood T lymphocytes from adults activated in AMLR synthesized a helper factor that supported PWM-induced proliferation and immunoglobulin production in both adult and cord B lymphocytes. In contrast, cord blood T lymphocytes failed to produce the helper factor for B lymphocytes. T cells from AMLR cultures established with neonatal lymphocytes showed suppressor activity, as assessed in PWM-stimulated immunoglobulin synthesis of adult peripheral-blood mononuclear cells, significantly higher than that exhibited by T cells from AMLR cultures performed with lymphocytes from adults. Finally, neonatal B lymphocytes could be activated to the production of IgM but not IgG by either adult AMLR-derived helper factor plus PWM or by Epstein-Barr virus, whereas adult B cells secreted both IgM and IgG under the same type of stimulation.     

Human T cells in cord blood: Abnormalities in Ia antigens induction by phytohemagglutinin and in autologous mixed lymphocyte reactions

About 25% of human T cells isolated from cord blood acquired Ia antigens following stimulation with phytohemagglutinin (PHA) for 72 hr. This percentage is markedly lower than that found in PHA-activated T-cell populations (PHA-T cells) isolated from peripheral blood of adults. The low expression of Ia antigens by human T cells from cord blood does not reflect abnormalities in the sensitivity to PHA stimulation and/or in the kinetics of induction of Ia antigens. PHA-T cells from cord blood display a low stimulatory activity in autologous mixed lymphocyte reactions (MLR). The defect does not reflect a nonspecific abnormality in the stimulatory activity of PHA-T cells from cord blood, since the latter do not differ from PHA-T cells from adults in their ability to stimulate allogeneic T cells from adults. Furthermore the defect does not reflect a nonspecific abnormality in the proliferative response of T cells from cord blood, since the latter display a normal proliferative response to PHA-T cells from adults. The defect in the proliferative response is not restricted to the autologous MLR with PHA-T cells, since it was found also in autologous MLR with non-T cells as stimulators. Correlation of the temporal evolution of the abnormalities of human T cells with the maturation of the immune system may contribute to our understanding of the role of Ia antigens in cell-cell interactions and of the biological significance of abnormalities of autologous MLR.     

Dissociation of autologous and allogeneic mixed lymphocyte reactivity by using a monoclonal antibody specific for human T helper cells

A monoclonal antibody defining a population of human T helper cells was developed and shown to specifically block the autologous mixed lymphocyte reaction (AMLR). This antibody, termed KT69-7 (IgG1), recognized 62% of peripheral blood E rosette-positive (E+) cells while demonstrating negligible reactivity with E- cells, monocytes, granulocytes, EBV-transformed B cell lines, and mouse splenocytes. Separation of E+ cells into KT69-7+ and KT69-7- populations revealed that KT69-7+ T cells provided helper function in PWM-driven B cell differentiation, whereas KT69-7- T cells provided no help and may suppress this response. Modulation of membrane moieties by using KT69-7 or OKT4 plus goat anti-mouse IgG removed reactivity to both these antibodies, suggesting an association between these molecules recognized by these antibodies. In functional studies, KT69-7 selectively blocked the AMLR while demonstrating minimal or no effect on the allogeneic MLR (allo-MLR). Blocking of the autoreactivity occurred when either autologous B lymphocytes or macrophages were used as stimulators. The failure of KT69-7 to block the allo-MLR was not attributable to excessive allogeneic stimulus; KT69-7 failed to block even under conditions of limiting numbers of stimulator cells. KT69-7 thus appears to recognize a molecule on the surface of T helper cells required for recognition of autologous class II antigens.     

Activated human T cells can present alloantigens but cannot present soluble antigens

Human T cells, when activated by antigen or mitogen, express Ia antigens. We have examined the capacity of activated T cells to stimulate autologous and allogeneic T cells and their ability to present soluble antigen. Interleukin 2-dependent T-cell lines (TCL), free of accessory cells, were used for antigen-presenting cells. These activated T cells were potent stimulators in an autologous mixed lymphocyte reaction (AMLR), more so than autologous irradiated non-T mononuclear cells. Activated T cells were also able to stimulate proliferation of allogeneic T cells in the absence of any other accessory cells, and this stimulation was blocked by anti-Ia antibodies. Resting unstimulated T cells were unable to stimulate autologous or allogeneic responses. Thus, activated T cells were able to present self antigens and alloantigens. However, activated T cells could not present soluble antigens to autologous T cells or to antigen-specific TCL even if exogenous interleukin 1 was added to cultures. The ability of activated T cells to stimulate an AMLR in vitro may reflect an important immunologic amplification mechanism in vivo. The ability of activated T cells to present alloantigens but not soluble antigens suggests an inability to process antigen, and this may provide further insights into the complexities of antigen presentation.     

Cytotoxic T cells generated in the autologous mixed lymphocyte reaction. I. Primary autologous mixed lymphocyte reaction

In the course of the culture of an autologous mixed lymphocyte reaction (AMLR), T cells proliferated in response to autologous non-T cells, and differentiated to cytotoxic T cells (AMLR killers). DNA synthesis was necessary to generate AMLR killers, as the elimination of autoreactive proliferating cells with BUdR and UV light completely abrogated AMLR killer cytolysis. Amlr killers lysed various lymphoid cell lines, including autologous B cell lines, autologous or allogeneic mitogen blasts stimulated by Con A, PHA, or pokeweed mitogen, variious nonlymphoid cell lines derived from human, mouse, or rat, and weakly normal autologous or allogeneic non-T cells. KMT-17, methylcholanthrene-induced rat fibrosarcoma, was the only resistant cell line to have been tested. AMLR killers had characteristics similar to NK cells, Major histocompatibility antigens were not the target antigens for AMLR killers. AMLR killers distinguished the blasts stimulated by alloantigens as self from the blasts stimulated by mitogens as non-self.     

Functional involvement of the LFA-1/ICAM-1 adhesion system in the autologous mixed lymphocyte reaction

The integrin surface molecule termed lymphocyte functional antigen-1 (LFA-1), and its physiological ligand intercellular adhesion molecule-1 (ICAM-1), have been proven to play a relevant role in several immune reactions where cell-to-cell contact is required: these reactions include allogeneic mixed lymphocyte reaction (MLR) and direct cytotoxicity. In the present study, we show that monoclonal antibodies (mAbs) directed to LFA-1 as well as to ICAM-1 molecules are able to inhibit T cell proliferation in autologous MLR (AMLR). Such an in vitro reaction is generally considered a functional model of Ia-mediated immunocompetent cell cooperation, and is impaired in several pathological conditions. It is noteworthy that the LFA-1 molecule is largely represented on the T cell surface, whereas ICAM-1 is poorly expressed on resting T cells: autologous stimulation slightly increases ICAM-1 expression. Pretreatment studies indicate that the inhibitory effect of anti-ICAM-1 mAb on T cell proliferation in AMLR is exerted on responder T cells.     

Interleukin 2 induces proliferation of normal "resting" human T cells in the absence of other known external stimulation

Interleukin 2(IL-2) is known to stimulate the progression of activated T cells from G1 through the rest of the cell cycle. We have demonstrated that addition of purified recombinant human IL-2 (rIL-2) to fresh normal human peripheral blood mononuclear cells (PBM), which were IL-2 receptor (Tac) negative by FACS analysis, stimulated marked proliferation of the PBM. IL-2-induced proliferation was also observed with umbilical cord blood mononuclear cells. Monocyte depletion of PBM resulted in a marked reduction of rIL-2-induced proliferative response which could be restored by adding back autologous irradiated monocytes but not by interleukin 1. The T cells preincubated with rIL-2 showed a five to six times enhanced autologous mixed-lymphocyte reaction (AMLR) compared to controls. The rIL-2-induced proliferative response of PBM was inhibited in a concentration-dependent fashion by preincubation of PBM with an anti-HLA-DR framework monoclonal antibody. The proliferating cells were shown by two-color flow cytometric analysis to be primarily Leu-1+ and Leu-4+ T cells (both leu-3+ and Leu-2+ subsets); however, 6 to 19% of responding cells had surface markers for B cells or NK cells. The data demonstrate that rIL-2 can induce proliferation of "resting" human T cells. The phenomenon may be related to a monocyte-dependent AMLR which induces IL-2 receptors and IL-2 responsiveness in a subset of T cells.     

Analysis of the role of xenogeneic antigens in the proliferation of human T cells stimulated with autologous non-T cells and phytohemagglutinin-activated T cells

Since conflicting results have been reported about the role of xenoantigens in the proliferation of T cells stimulated with autologous non-T cells, the effect of the exposure of cells to xenogeneic proteins during the isolation procedure and/or the culture period on autologous mixed lymphocyte reactions (AMLR) with non-T cells and phytohemagglutinin-activated T cells as stimulators was investigated. T and non-T cells were isolated by rosetting with 2-aminoethylisothiuronium bromide-treated sheep red blood cells (AET-SRBC), by nylon-wool nitration, and by positive or negative selection with anti-class II HLA antigens and anti-T-cell monoclonal antibodies. Isolation and cultures were performed in presence of fetal calf serum (FCS) or of autologous serum. In both types of AMLR, proliferation of responding cells did not require exposure to xenoantigens. However xenoantigens enhanced the proliferation of cells from some, although not all, the donors tested. There were differences in the degree of proliferation of the cells from the donors tested, but without correlation with the two types of AMLR. These results suggest that both types of AMLR reflect a self-recognition event and not a response to xenoantigens. However the potential interference of xenoantigens, as well as the individual variability, should be taken into account when interpreting the significance of abnormalities of AMLR in immunopathologic processes.     

Helper factors for hapten-self cytotoxic T cells occur in NZB culture supernatants despite deficient AMLR

The ability of helper T cells from NZB mice to produce non-interleukin 2 (IL-2) lymphokines in the autologous mixed lymphocyte reaction (AMLR) was examined. Factors present in normal AMLR culture have been previously reported to mediate the development of a cytotoxic T-cell response to trinitrophenyl (TNP)-modified syngeneic thymocytes. Young NZB mice, like the normal strains, were able to produce the helper factors in the AMLR and to utilize these mediators in the cytotoxic induction system. Old autoimmune NZB mice demonstrated a poor proliferative response in the AMLR and were unable to activate hapten-specific cytotoxic cells in the presence of AMLR culture supernatant from either young or old mice. This was not due to a lack of cytotoxic precursors, nor was it a normal consequence of aging, but may be related to decreased IL-2 production by helper T cells. Interestingly, supernatant from AMLR proliferation deficient old NZB mice contained normal amounts of the AMLR helper factor. These data suggest that AMLR helper factor production is not directly related to the proliferative response and that two different helper-T-cell subpopulations may be responsible for these activities. The production of these mediators in mice which cannot utilize them raise questions about their role in autoimmunity.     

Autologous induction of the human T-cell-dependent monocyte procoagulant activity: a possible link between immunoregulatory phenomena and blood coagulation

Activated monocytes acquire the ability to induce clot formation in platelet-poor citrated human plasma. The generation of this procoagulant activity (PCA) is dependent upon an interactive pathway between monocytes and T lymphocytes. Here we show that an ongoing autologous mixed lymphocyte reaction (AMLR) can elicit a T-cell-instructed PCA. PCA was measured by the ability of the cells to accelerate the clotting time of pooled citrated platelet-poor human plasma. AMLR was measured by tritiated thymidine incorporation. PCA and AMLR had very similar kinetics. Correlation coefficients between both reactions ranged from 0.59 to 0.99. Addition of an anti-DR monoclonal antibody blocked both reactions. T-Lymphocyte-depleted cell populations did not increase their level of PCA after 6 days in culture. Addition of autologous T cells to the T-depleted population restored its ability to produce PCA. Cyclosporin A blocked the peripheral blood mononuclear cell ability to generate PCA. A lymphokine generated during the AMLR was able to induce PCA in normal mononuclear cells. The results indicate that self recognition activates monocytes to produce PCA and suggests that this mechanism may represent a link between immunoregulatory phenomena and blood coagulation.     

T cell proliferation induced by autologous non-T cells is a response to apoptotic cells processed by dendritic cells

Self-reactive T cells are present in the mature immune repertoire as demonstrated by T cell proliferation induced by autologous non-T cells in the autologous mixed lymphocyte reaction. This reaction generates regulatory T cells in vitro and may reflect immune regulatory pathways in vivo, but the antigenic peptides recognized remain uncharacterized. We revisited this issue in light of the importance of apoptosis in immune regulation. We found that apoptosis among peripheral blood non-T stimulator cells is associated with augmented induction of autologous T cell proliferation. Our data show that caspase activity in the non-T stimulator population is essential for induction of autologous T cell proliferation, suggesting that cellular components in the non-T cell fraction are enzymatically modified, most likely by effector caspases, and have a direct or indirect effect on autoreactive T cell activation. Furthermore, exposure of macrophage-derived dendritic cells to apoptotic non-T cells augments autologous T cell proliferation, and blockade of alpha(v)beta(5) integrin, but not alpha(v)beta(3), inhibits the capacity of irradiated non-T cells or dendritic cells to stimulate autologous T cell proliferation. These experiments, using an entirely autologous system, suggest the interpretation that autoreactive T cells may recognize self-Ags modified through the actions of caspases and presented to T cells by dendritic cells. Induction of an in vivo autologous mixed lymphocyte reaction by caspase-modified self-Ags present in apoptotic cells may represent a mechanism to maintain peripheral immune tolerance.     

Lymphocyte transformation induced by autologous cells.

Human peripheral blood T lymphocytes are stimulated to proliferate when cultured with autologous B-lymphoblastoid cell lines, autologous mitogen-induced lymphoblasts, or autologous non-T blood lymphocytes. This reaction, the autologous mixed lymphocyte reaction, has attributes of an immune response possessing both memory and specificity. The capacity to stimulate autologous T lymphocyte proliferation depends on the lineage of the lymphoid cell and not on its establishment in continuous culture or carriage of the EB viral genome. The determinant on non-T lymphocytes which stimulates the autologous mixed lymphocyte reaction appears to be an Ia determinant. Thus, allogeneic graft rejection and the allogenic mixed lymphocyte reaction are very likely extensions of an immune response expressed within the host.     

Gamma irradiation of human dendritic cells influences proliferation and cytokine profile of T cells in autologous mixed lymphocyte reaction

Dendritic cells (DC) are the most potent antigen-presenting cells (APC); their ability to induce proliferation of T cells in a mixed lymphocyte reaction (MLR) assay is commonly used for the evaluation of their function. It is a general thought that gamma irradiation of APC does not influence their ability to activate T-cell proliferation, but the data from several studies are controversial. To further determine the mechanisms involved in DC-induced T-cell activation in MLR assay, human DC induced from peripheral blood mononuclear cells (PBMC) were gamma-irradiated and determine their effects on the proliferation and cytokine profiles of T cells in an autologous MLR. DC were induced from the PBMC of 11 multiple sclerosis (MS) patients with RMPI 640 medium containing recombinant human GM-CSF (rhGM-CSF; 800 U/ml) and recombinant human IL-4 (rhIL-4; 500 U/ml). DC harvested on day 7 were divided into two equal parts. One part was not irradiated (naive DC); the other was gamma-irradiated at a dose of 30 Gy. Cell surface molecules were analyzed by flow cytometry. T-cell proliferation was determined using a beta-scintillation counter. The levels of IL-2, IL-4, IL-6 and IL-10 in co-culture supernatants were measured by ELISA. The results indicated that gamma irradiation reduced expression of CD86, CD80 and HLA-DR molecules on DC, especially CD86 (P=0.0072). DC, irradiated or non-irradiated, effectively stimulated autologous T-cell proliferation. Compared to naive DC, irradiated DC showed a markedly lower capacity to promote T-cell proliferation (P=0.0073), and strikingly up-regulated secretion of IL-4 (P=0.0145) and IL-2 (P=0.0323) by autologous T cells. No significant differences were noted in IL-6 and IL-10 production between T cells co-cultured with naive DC and irradiated DC (P>0.05). It is concluded that gamma irradiation of DC not only influences the phenotype of DC but also alters their capacity to stimulate the proliferation and the cytokine profiles of autologous T cells in a MLR.     

The effect of low dose gamma irradiation on the differentiation and maturation of monocyte derived dendritic cells

The effects of gamma-irradiation on the differentiation of peripheral blood monocytes (PBM) into monocyte derived dendritic cells (MDC), their maturation, and subsequent ability to present antigen to T cells was studied. Undifferentiated MDC were more sensitive to gamma-irradiation induced apoptosis than mature MDC. Irradiation of immature MDC with 5 Gy of gamma-rays down regulated the expression of the costimulatory receptors CD80/CD86 and may compromise their ability to capture and present antigen. By contrast, gamma-Irradiation of mature MDC did not affect the expression of CD86/CD80, and HLA-DR. Gamma-irradiation increased the apoptosis of MDC; but did not affect the ability of mMDC to stimulate autologous MLR. T cell proliferative response in the MLR and in response to tetanus antigen was reduced when gamma-irradiated primary DC1 were used to either stimulate or present antigen to T cells.     

The human autologous mixed lymphocyte reaction. I. Suppression by macrophages and T cells

The autologous mixed lymphocyte reaction (AMLR) is a proliferative response of T cells to signals from autologous non-T cells. The AMLR has been an enigma to immunologists because spontaneous proliferation of cells removed from the body is usually substantially less than that observed with a strong AMLR. However, the AMLR is thought to represent an important in vitro function, since it has the attributes of other immune responses, and it is abnormal in a variety of disease states thought to have an immune basis. We reasoned that if the AMLR represented a fundamental immune phenomenon, it should be subject to regulation. In the present study, we present evidence for suppression of the AMLR by macrophages and by T cells. Macrophages inhibited the T cell proliferation to (B + null) cells in a dose-dependent fashion and throughout the time course of the AMLR. Elimination of suppressor T cells by a specific antiserum led to an increase in the AMLR, which was again suppressed in a dose-dependent way by addition of the suppressive T cells. It may be concluded that the AMLR itself is subject to immune regulation and that the suppressive influences observed probably strongly inhibit the AMLR in vivo. Removal of the suppressive principles allows the maximal expression of the AMLR in vitro. We believe that our demonstration of regulation of the AMLR should remove the enigma associated with it and lead to a better understanding of normal cell-cell interactions as well as the basis for abnormalities in a variety of immune-mediated diseases.     

Abnormalities of T cells isolated from mediastinal lymph nodes and peripheral blood of patients with lung carcinoma: deficit in PHA-induced expression of HLA class II antigens and in autologous mixed lymphocyte reactions

Summary A reduced percentage of T cells isolated from mediastinal lymph nodes and peripheral blood of patients with lung carcinoma acquired HLA Class II antigens following in vitro stimulation with PHA. Furthermore T cells were functionally abnormal in autologous and allogeneic mixed lymphocyte reactions (MLR). The immunoregulatory properties of HLA Class II antigens and autologous MLRs suggest that these abnormalities may affect the interaction of the host's immune system with tumor cells.     

Regulation of the growth of Epstein-Barr virus-infected B cells. I. Growth regression by E rosetting cells from VCA-positive donors is a combined effect of autologous mixed leukocyte reaction and activation of T8+ memory cells

Regression of B cell proliferation in co-cultures of EBV-infected B cells (BEBV) and autologous T cells at 1:4 ratio was studied. 3H-TdR incorporation was used to measure proliferation by the participating lymphocyte populations and a 51Cr release assay was used to document the generation of cells capable of killing autologous EBV-transformed B lymphoblastoid cell lines (LCLEBV). EBV-infected B cells cultured alone transformed to blasts by culture day 10, and continued to proliferate throughout the 22 day observation period. When EBV-infected B cells were co-cultured with E rosetted cells from VCA-positive donors, there was a characteristic proliferative response on day 10 (an augmented autologous mixed lymphocyte reaction; AMLR), followed by the development of T8+ cells capable of killing autologous LCLEBV, as well as over 90% suppression of EBV growth by day 22 as assessed by 3H-TdR incorporation, and confirmed in a visual outgrowth assay. Negative and positive selection techniques were used to define the regulatory components in the T cell population. Depletion of T8+ cells from the blood lymphocytes of VCA-positive donors did not significantly reduce the 10 day proliferative response, but the subsequent development of cytotoxic cells and the regression of BEBV outgrowth was not observed. Thus, the circulating T8+ cells are required for the subsequent appearance of autologous LCLEBV cytotoxicity and BEBV growth regulation. However, when the responder population consisted only of T8+ cells, the augmented AMLR response was absent, cytotoxic cell development was weak or absent, and there was no regression of EBV outgrowth. Therefore, the cells participating in the AMLR, as well as T8+ memory cells from VCA-positive donors, are necessary for the control of the in vitro EBV infection. Growth regression is dependent on the proliferation of the regulatory T cells. Mitomycin C treatment of fresh E rosetting cells or those exposed to BEBV for up to 10 days in culture abrogates growth regression and the subsequent appearance of LCLEBV killer cells. However, E rosetting cells exposed to BEBV for 14 days or more already have developed the ability to kill LCLEBV and no longer need to proliferate to induce growth regression when cultured with newly infected BEBV. These results lend additional support to the view that the control of EBV-induced B cell expansion requires a AMLR-dependent clonal amplification of EBV-specific, T8+ cytotoxic cells.     

Regulation of T cell proliferation by cloned interferon-alpha mediated by Leu-11b-positive cells

The autologous T lymphocyte proliferative response (AMLR) induced by a B lymphocyte-enriched non-T, nonadherent cell population (NT, NAC) and by a macrophage-enriched population were both suppressed by the addition of a cloned interferon-alpha (IFN-alpha Con1) directly to the cultures. Preincubation of the stimulating NT, NAC with IFN-alpha Con1 resulted in comparable suppression. In contrast, preincubation of the macrophages with IFN-alpha Con1 resulted in significant augmentation of T cell proliferation. Depletion of Leu-11b-positive cells from the NT, NAC exposed to IFN-alpha Con1 restored the autologous T cell response. Addition of IFN-alpha Con1 activated Leu-11b-positive cells, isolated from the NT, NAC population, was suppressive of the AMLR. Although NK cytotoxicity was irradiation sensitive, suppression of the AMLR by IFN-alpha Con1-activated NT, NAC was resistant, suggesting that different subsets of cells or mechanisms by the same cells may have been responsible. These observations may offer insights into the potential role of cells with the NK phenotype, Leu-11b, and IFN in contributing to immuno-regulatory changes observed in clinical states associated with elevated concentrations of IFN.     

Responsiveness of human T-lymphocyte subpopulations in autologous mixed-lymphocyte reaction using xenoprotein-free separated cells: autologous reactivity lies chiefly in a low density T-cell fraction

The responsiveness in the autologous mixed-lymphocyte reaction (AMLR) by human T cells separated using two different methods not involving xenoantigen contract was examined. Although T cells from nylon-wool columns were active in AMLR, T cells separated by a Percoll gradient method responded poorly. Further separation of T cells from nylon-wool columns into low density (TL) and high density (TH) fractions by Percoll revealed that TL was enriched, while TH was depleted, in AMLR responsiveness. This difference could not be accounted for by differences in the helper or suppressor cells in the fractions. Moreover, TH responded well in secondary AMLR. Therefore the T cells reactive in AMLR reside chiefly, although not exclusively, in the low density fraction.