Extrathymic T cell differentiation in vitro from human CD34+ stem cells.

Although it is well established that T cells are derived from CD34+ stem cells in vivo, and that T cells can develop in the absence of a functioning thymus, it has not proven possible thus far to generate human T cells in vitro from CD34+ cells in the absence of any thymic influence. We now present a limiting dilution cloning culture system that supports the differentiation of highly purified human CD34+ cells to CD3+ T cells in vitro in the complete absence of any thymic components. The culture system features the use of a serum-free medium supplemented with a cocktail of cytokines including flt-3 ligand, interleukin-3 (IL-3), stem cell factor (SCF), and IL-2. CD4+ T cell clones capable of mitogen-stimulated proliferation and response to IL-2, and expressing a varied TCR-Vbeta repertoire were obtained under these conditions. This culture system therefore supports human T lymphopoiesis in the absence of any thymic influence and may prove useful for the evaluation of extrathymic T cell differentiation in vitro.

Long-term culture of monoclonal human T lymphocytes: models for immunosenescence?

Human monoclonal T lymphocyte populations maintained in long-term culture by intermittent reactivation via the antigen receptor and supplied with exogenous interleukin 2 manifest finite proliferative lifespans. T lymphocytes cloned from mature peripheral T cells of adult donors were constantly lost from the time point of their first isolation up to an estimated maximum of 80 population doublings (PD) for the longest lived. T lymphocytes cloned from T cell progenitors in bone marrow, on the other hand, survived for a maximum of ca. 100 PD. One facet of the functional capacity of cells derived from these two different sources was assessed by measuring their autocrine proliferation after mitogenic stimulation. For a majority of T cell clones (TCC), autocrine proliferative capacity decreased as a function of culture age, becoming absent by 50 PD for adult-derived-TCC and by 70 PD for bone marrow-derived TCC, thereby clearly occurring prior to the end of the proliferative life spans of the clones. Limiting dilution frequency analysis showed that the number of autocrine proliferative precursors within these monoclonal populations declined with age, paralleling loss of autocrine proliferative capacity in the 'bulk' clones. Of a variety of surface structures monitored during culture ageing of TCC, the density of expression of the coreceptor molecule CD28 was found to correlate with decreasing autocrine proliferative capacity in two-thirds of the clones. Thus, at least for a fraction of monoclonal human T lymphocytes, decreasing autocrine proliferative capacity, a measure of clonal expansion, may correlate with decreasing numbers of CD28 molecules expressed on the surface and therefore presumably with the strength of costimulatory signal delivered via this important coreceptor.

Cytokine modulation of TH1/TH2 phenotype differentiation in directly alloresponsive CD4+ human T cells.

Organ graft rejection is caused by the recognition of allogeneic MHC molecules by recipient T cells by two different pathways. The indirect pathway of alloreactivity requires the presentation of MHC peptides from the graft by autologous APC, as with conventional antigen. The direct pathway, on the other hand, requires the recognition of foreign MHC on foreign cells. The regulatory mechanisms for this component of alloreactivity have not been extensively studied. We show here that the T cell response activated by alloantigens in the direct pathway is similarly constrained and modulated by cytokines, as has been shown for classic antigen presentation. Thus, the inclusion of IL-2 or TGF-beta in MLC performed with purified responder T cells resulted in outgrowth of cells secreting IL-2 and IFN-gamma, whereas addition of IL-4, IL-10, or anti-TGF-beta encouraged outgrowth of cells secreting IL-4 and IL-10. T cells alloactivated via the direct pathway and then cloned in IL-2 alone secreted IL-4 and IL-10 as well as IFN-gamma and IL-2 (Th0 phenotype). Established clones remained susceptible to cytokine modulation, such that IL-4 and IL-10 decreased their secretion of IL-2 and IFN-gamma, whereas TGF-beta suppressed IL-4 and IL-10 secretion. The first alterations of Th0 toward Th1 or Th2 phenotypes could already be observed after only a very brief exposure to cytokines of 48 hr, followed by extended culture with IL-2 alone. These results confirm that human T cells with Th1 and Th2 phenotypes, recognizing alloantigen via the direct pathway, derive from the same IL-2-secreting precursor and can be manipulated by cytokines in an analogous fashion to conventional antigen-reactive cells. These findings may have implications for manipulating the direct pathway of alloantigen recognition in human organ transplantation.

Cytokine regulation of the balance between alloindifferent and allospecific suppressor induction in mixed lymphocyte cultures.

The effects of exogenous cytokines on the generation of alloindifferent, MHC-unrestricted suppressive activity early on in mixed lymphocyte culture interactions have been investigated. Interleukin 4 strongly blocked the generation of suppression, whereas IL-1, IL-2, and IL-6 enhanced it to some extent. Tumor necrosis factor-alpha, interferons-alpha and -gamma, granulocyte/macrophage colony-stimulating factor, granulocyte CSF, IL-3 and IL-5, and a number of combinations of these factors were without effect in this system. Insofar as the alloindifferent suppression studied here also inhibited the development of allospecific, MHC restricted suppressive activity later in MLC, reduction by IL-4 of its development may have relevance for the use of this cytokine to facilitate the induction of specific suppressor cell-mediated transplantation tolerance in vivo.

Cyclosporin A blocks the generation of alloindifferent but not of allospecific suppressive cells.

The induction of alloantigen-indifferent, MHC unrestricted suppressive cells (SC) early on in human mixed lymphocyte culture (MLC) or after stimulation with suppressive T cell clones was blocked in a dose-dependent fashion by cyclosporin A (CsA). This was not prevented by the addition of the following defined lymphokines: interleukin (IL) -1, -2, -4, -5, interferon-tau or GM-CSF. The addition of MLC-conditioned medium as a source of multiple lymphokines including IL-3 also failed to reconstitute suppressive activity in the presence of CsA. In contrast, the development of allospecific, HLA-restricted SC in the same CsA-blocked MLC was not prevented or was even enhanced. These results confirm that CsA 'spares' specific SC induction late in MLC, but show that it prevents induction of non-specific suppression earlier in MLC by a mechanism presumably unrelated to blocking the secretion of interferon-tau or IL-1 to IL-5.

Regulation of allospecific suppressive cell generation in vitro.

Control of the generation of suppressive cells (SC) in allogeneic mixed lymphocyte cultures (MLC) has been re-investigated. Cells taken from 3-6 days old ("early", e-MLC) suppressed lymphocyte proliferative responses when transferred to a second MLC. Suppression was not allospecific, nor restricted to the autologous responder. In contrast, cells from MLC 7-10 days old ("late", l-MLC) mediated weak but specific suppressive activity, which was HLA-restricted. Addition of early MLC cells (e-MLC1) to a second MLC (MLC2), blocked the further generation of non-specific suppressive activity in that MLC, suggesting the presence of downregulatory components for SC generation. Nonetheless, the generation of allospecific suppression in late MLC2 (l-MLC2) was enhanced rather than inhibited under these conditions. In separate experiments, a T cell clone (TCC) was isolated which mimicked the action of e-MLC cells on SC generation, namely, down-regulation of non-specific but amplification of specific SC generation. This clone carried the "alternative" tau/delta T cell receptor. Thus, tau/delta+ "antisuppressor" regulatory cells may have a role in the establishment and maintenance of allograft tolerance.

Restimulation properties of cloned alloactivated lymphocytes: detection of a novel type of HLA-D/DR region determinant and allelic subtypes for HLA-Dw3 using cloned reagents.

Lymphocytes alloactivated by Dw3 homozygous typing cells were cloned by the method of limiting dilution and cultured for prolonged periods using T-cell growth factor and irradiated pooled leukocytes (as feeder cells). Restimulation specificity of two clones functioning as primed lymphocyte typing reagents was investigated in panel and family studies. Cells from one of the clones (12-2) were always specifically stimulated by HLA-Dw3 antigens shared between the original priming cells and the stimulating panel cells. In an informative family KOH, however, cells from this clone seemed to detect a split in the Dw3 cluster. Cells from the other clone (12-8) failed to respond to Dw3 antigens as expressed by the original priming cells or by panel stimulating cells; rather, specificity of restimulation seemed to be associated with the expression of Dw4 antigens. Family segregation analysis did not support this conclusion however, since stimulating products segregated with one of the three Dw3-bearing haplotypes and with none of three Dw4-bearing haplotypes. This suggested that 12-8 cells may be responsive to antigens different from those detected by homozygous typing cells.

Dissection of suppressor cell generation in vitro.

Suppressor cells (SC) that nonspecifically inhibited lymphoproliferative (LP) responses were found after culturing peripheral blood mononuclear cells: with suppressor T-cell clones, in mixed lymphocyte cultures (MLC), and with recombinant interleukin 2 (IL-2), but were not found after culture in medium alone. A monoclonal anti-IL-2 receptor (R) antibody (MoAb), TU69, which blocked LP responses of IL 2-dependent T-cell lines, also blocked SC induction by T-cell clones, but completely failed to inhibit SC generation in MLC or with IL-2. This suggests that the IL-2R epitope defined by TU69 was not involved in SC induction in the latter systems. MoAb against HLA-DQ (TU22), -DR (TU34, SG157), -DP (B7/21), or DR and DP (TU43, 58), all of which were able to block stimulation of appropriately specific clones, did not block SC induction in any of the three systems studied. In contrast, the broadly reactive moAb TU39, which binds at least DR and DP but also has additional reactivity for determinants tentatively designated "DY," blocked SC induction by T-cell clones and in MLC. Finally, an anti-HLA class I MoAb, W6/32.HL, greatly decreased SC generation in MLC, but not with rIL-2 or T-cell clones. Thus, the induction of nonspecific SC was dissected into three pathways involving: class I and TU39-defined but not DR, DQ, or DP determinants (in MLC) which was independent of the IL-2R epitope bound by TU69; only TU39-defined determinants (with T-cell clones), which were IL 2R dependent; and, neither class I, class II nor TU39-defined determinants (induction by rIL-2), which was also TU69+ IL-2R independent.

Involvement of "accessory" and antigen receptor molecules in human helper T cell clone activation studied by monoclonal antibody inhibition.

The requirements for activation of autocrine proliferation in human helper T cell clones (Th-TCC) by allogeneic cells were examined in monoclonal antibody (MoAb) blocking studies. Stimulation was not blocked by CD4, CD5, CD6, CD7, or CD45 MoAbs, despite high levels of expression of these antigens on the TCC. Only CD2 and CD11a (LFA-1) MoAbs blocked activation, the latter only when peripheral blood mononuclear cells (PBMCs) and not B-lymphoblastoid cell line (B-LCL) cells were used at stimulators. Responses to interleukin 2 (IL 2) were only minimally blocked by any of the MoAbs. All TCC were CD3+ and expressed the alpha/beta chain T cell receptor (TCR) as detected by moAb WT31. Accordingly, CD3 and WT31 MoAbs consistently blocked stimulation by B-LCL, and in addition one anti-DR5 TCC and one anti-DQw3 TCC were blocked by MoAb 42/1C1, which is directed to an idiotypic determinant of the HPB-ALL leukemic line TCR. Only these two TCC reacted with moAb 42/1C1 in flow cytometry. These observations suggest that CD2- but not LFA-1-mediated interactions, as well as TCR and stimulating antigen binding, are absolutely necessary to activate Th-TCC.

Alloproliferative human T cell clones primed and cultured in vitro lose proliferative and gain suppressive activity with age.

Alloreactive human T lymphocyte clones were found to lose antigen-stimulated proliferative capacity and their ability to secrete interleukin 2 (IL 2) after a critical period in tissue culture. Instead, they gained the previously absent capacity to suppress lymphoproliferative (LP) responses in the presence or absence of exogenous IL 2. Such "ex-PLT" suppressive clones continued to grow perfectly well, retaining IL 2 and filler cell dependency, apparently normal karyotypes, and their OKT4+, OKT8- phenotypes. At least two suppressive mechanisms were demonstrated: (1) the "induction" of suppressive effectors in normal peripheral lymphocytes, and (2) a direct suppressive activity on lymphocyte proliferation shown by their ability to inhibit restimulation of cloned lymphocytes lacking suppressor cell precursors. The consistent "differentiation" from IL 2-secreting "helper" status to nonspecific suppressive status may represent a novel immunoregulatory phase in the long-term differentiation of normal human T cells.

Decreased proliferative capacity and increased susceptibility to activation-induced cell death in late-passage human CD4+ TCR2+ cultured T cell clones.

The growth characteristics in vitro of interleukin 2 (IL 2)-dependent human CD4+ alpha beta-T cell receptor-positive helper T cell clones (TCC) were studied in relation to alterations in surface phenotype, cytokine responsiveness, and susceptibility to activation-induced cell death (AICD). TCC derived from peripheral blood T cells had finite lifespans averaging 33 population doublings (PD) with a recorded maximum lifespan of 80 PD (n = 208). First analyses of the TCC were undertaken at ca. 25 PD, at which time all cells of all TCC expressed high intensity CD45RO and low intensity CD45RA, as well as high intensity CD95 (fas) and MHC class II antigens. The expression of these molecules remained elevated throughout the proliferative lifespan of the clones, but for those TCC which were initially CD28+ (the majority), the density of expression of the latter was diminished in most late-passage clones. Concomitant with this, late-passage cells showed reduced responsiveness to CD28-mediated costimulation by CHO transfectants expressing human CD80 compared to early-passage cells. Additionally, the level of expression of IL 2R gamma c and IL 7R chains was commonly reduced, as was the response to IL 2 and IL 7. Despite unchanged levels of fas expression on TCC with time, late-passage cells were more susceptible to AICD than early, passage cells. These observations further document functional and phenotypic alterations in long-term cultured human T helper cells, which may be considered as biomarkers of immunosenescence. This may contribute to an improved understanding of the mechanisms underlying depressed T cell function in old age.

Finite lifespans of T cell clones derived from CD34+ human haematopoietic stem cells in vitro.

Several studies have documented finite lifespans of at least the vast majority of cultured human T cell lines and clones. However, there is a great deal of variation among the different preparations, ranging from < 25 PD up to > 100 PD. The cultured T cells in all these studies originated from mature T cells isolated from peripheral blood of adult donors. It was, therefore, impossible to assess the contribution of differences in in vivo age to the subsequent differences between clones in in vitro aging. In an attempt to circumvent this difficulty, we have developed a culture system that supports the differentiation of highly purified human CD34+ cells into CD3+ T cells in vitro. This features the use of a serum-free medium supplemented with the cytokines flt-3 ligand, IL 3, stem cell factor (c-kit ligand) and IL 2, together with IL 7 or oncostatin M (OM). In this way it is possible to perform "longitudinal" studies on T cells derived de novo in vitro. We show here that T cell clones derived under these circumstances also manifest variable finite life expectancies, for which the only uncontrolled (nonstochastic) effects of aging must already have occurred at the stem cell level.

Relative cloning efficiencies and long-term propagation capacity for T cell clones of highly purified natural interleukin 2 compared to recombinant interleukin 2 in man.

Highly purified (HP) natural (n) and recombinant (r) interleukin 2 (IL 2) have been compared for their ability to support clonal outgrowth and long-term clonal propagation of alloactivated human T lymphocytes. The frequency of outgrowth of T cell colonies was as high as 1:2 to 1:4 when HPnIL 2 was employed for limiting dilution cloning. Separated CD4+ and CD8+ activated populations produced similar cloning efficiencies. Recloning of established CD4+ lines in HPnIL 2 suggested that a clonable cell frequency of 1:2-1:4 was the maximum possible that could be achieved in this system. In contrast, purified rIL 2 allowed outgrowth of only 1:10-1:20 cells from alloactivated populations. Again, CD4+ and CD8+ fractions generated similar cloning efficiencies. In terms of the fraction of derived clones, which could be propagated to greater than 1 X 10(6) and greater than 1 X 10(8) cells, nIL 2 again proved superior to rIL 2. In either source of IL 2, the proportion of CD4+ clones which could be extensively propagated was greater than in the CD8+ population. Surprisingly, although the addition of PHA to these lectin-free IL 2 preparations reduced the frequency of clonable cells, the proportion of clones which could be extensively propagated was increased. These results suggest that nIL 2, consisting of three differently glycosylated molecular species, may be preferable to rIL 2, which consists only of a single non-glycosylated species, for the cloning and long-term propagation of human T cells. These results may have some bearing on the choice of rIL 2 versus nIL 2 for therapeutic applications.

Relative roles of natural killer- and T cell-mediated anti-leukemia effects in chronic myelogenous leukemia patients treated with interferon-alpha.

Potential anti-leukemia effects mediated by T cells or by natural killer (NK) cells were investigated in chronic myelogenous leukemia (CML) patients treated with interferon-alpha. Therapy-associated modulation of T cell and NK reactivity was monitored for one year from initiation in autologous mixed lymphocyte-tumor cell reactions and cytotoxicity directed against autologous CML cells, respectively. During the course of IFN-therapy, NK activity against autologous CML cells increased steadily, whereas T cell reactivity fluctuated randomly. Despite the high level of T cell reactivity to autologous tumor cells in short-term (6 days) culture, 1) they failed to respond to synthetic peptides corresponding to the bcr/abl fusion sequence of the patient, and 2) only one proliferative T cell clone (TCC) was isolated which specifically recognized HLA-DR-matched CML cells. This TCC appeared not to recognize synthetic peptides corresponding to the bcr/abl fusion sequence of the patient; the antigen to which it responds remains unknown. To assess potential immunogenicity of bcr/abl peptides, it was attempted to sensitize T cells from normal donors in vitro. Of 109 cell lines obtained from seven different donors, eleven showed peptide-dependent proliferation. Therefore, although these results show that it is possible to isolate apparently CML-specific T cells from patients, as well as to prime T cells against tumor-specific peptide in vitro, the frequency of such T cell-mediated reactivity appears low and its relevance to anti-leukemic effects questionable. On the other hand, the strong time-dependent enhancement of natural killing of autologous CML blasts during IFN-alpha treatment, a phenomenon not observed for T cell reactivity, suggests that natural immunity may be more important in controlling disease.

Immunosenescence, suppression and tumour progression.

There are good arguments for suggesting that two seminal papers published 50 years ago can be taken as the beginning of modern tumour immunology. These papers by R. Baldwin, "Immunity to transplanted tumour: the effect of tumour extracts on the growth of homologous tumours in rats" and "Immunity to methylcholanthrene-induced tumours in inbred rats following atrophy and regression of the implanted tumours" (Br J Cancer 9:646-51 and 652-657, 1955) showed that once tumours are established, they and their products can be recognised by the adaptive immune system and rejected. However, the tumour normally co-evolves with immunity, like a parasite, rather than being suddenly introduced as in these, and many other, experimental models. Dynamics of this co-evolution are illustrated by findings that inflammation enhances tumorigenicity, yet is important to enable T cells to respond properly to tumour antigen and exert anti-tumour effects. The important thing is to maintain the balance between effective anti-tumour immunity and tumour escape and/or stimulatory mechanisms. Tumours almost always co-exist with immune defence systems over extended periods and interact chronically with T cells. The effect of this is potentially similar to other situations of chronic antigenic stress, particularly lifelong persistent virus infection, most strikingly, CMV infection. The questions briefly explored in this symposium paper are what happens when T lymphocyte clones are chronically stimulated by antigen which is not or cannot be eliminated? What are the similarities and differences between chronic antigenic stimulation by tumour antigen versus CMV antigen? What can we learn in one system which may illuminate the other?

IFN-alpha regulates IL 10 production by CML cells in vitro.

High levels of spontaneous in vitro IL 10 secretion by a subset of untreated chronic phase CML patients' cells are shown to be decreased in the presence of IFN-alpha. However, the lower level of spontaneous IL 10 secretion by healthy control cells are was not depressed by IFN-alpha. In contrast to its effects on IL 10 production, IFN-alpha increased the low spontaneous secretion of IL 1alpha by patients' cells, bud did not further increase the higher levels of spontaneous IL 1beta secretion by normal cells. It had no effect on secretion of TNF-alpha by patients or normals. Spontaneous secretion of IL-1alpha (or IFN-gamma) by patients' cells was not observed whether or not IFN-alpha was present. Therefore, one mechanism of action of IFN-alpha in vivo may involve decreasing endogenous IL 10 secretion (thereby reducing suppressive effects on T cell reactivity) and increasing IL 1beta secretion (thereby enhancing antigen presentation).

Interleukin 10 protects activated human T lymphocytes against growth factor withdrawal-induced cell death but only anti-fas antibody can prevent activation-induced cell death.

Interleukin 10 (IL-10) is a pleiotropic T cell-derived cytokine best known for its negative regulatory effects on T cell immunity. It inhibits responses indirectly by downregulating expression of major histocompatibility complex (MHC) molecules and co-stimulatory molecules such as CD80 on antigen presenting cells as well as directly via its effects on responding cells. On the other hand, IL-10 has been shown to protect activated T cells against apoptosis caused by withdrawal of the major growth factor, IL-2, and allow proliferation of T cells in the absence of IL-2. However, we show here that this IL-10-dependent, IL-2-independent proliferative response is short-lived, and that IL-10-responsive T cells cannot multiply in its presence. Moreover, inclusion of exogenous IL-10 in clonal cultures propagated with IL-2 results in suppression of their growth. These findings, together with the observation that IL-10 fails to protect T cells against activation-induced cell death (a fas/fas-ligand-dependent phenomenon blocked only by certain antagonistic anti-fas reagents), suggest that the negative regulatory effects of IL-10 outweigh the upregulated proliferation observed on some T cell clones (TCC) in the absence of IL-2.