Interaction of transforming growth factor-beta 1 with hemopoietic growth factors in the regulation of human normal and leukemic myelopoiesis.

We studied the effect of transforming growth factor-beta 1 (TGF-beta 1) on the growth of normal and chronic myeloid leukemia (CML) granulo-monopoietic progenitors (CFU-GM) and erythroid progenitors (BFU-E) of different origins and degrees of maturation. In the presence of the supernatant of the 5637 cell line, used as a source of growth factors, TGF-beta 1 stimulates the growth of day-7 CFU-GM from Ficoll-isolated normal bone marrow cells. Maximum stimulation (172% of controls) is observed with 2.5 ng/ml TGF-beta. The results with a highly enriched progenitor cell population stimulated by recombinant granulocyte colony-stimulating factor (rG-CSF) and recombinant granulocyte-macrophage CSF (rGM-CSF) were similar, suggesting a direct effect of TGF-beta 1 on hemopoietic progenitors. In contrast to this stimulatory effect of TGF-beta 1 on normal day-7 bone marrow CFU-GM, TGF-beta 1 does not affect the growth of day-14 CFU-GM. The growth of normal bone marrow BFU-E is strongly inhibited. In the majority of cases (11/15) of CML, bone marrow day-7 CFU-GM growth is inhibited by TGF-beta 1. In few cases (4/15) leukemic progenitors respond to TGF-beta 1 as normal cells. TGF-beta 1 always inhibits the growth of day-14 bone marrow CFU-GM from CML patients.

Reduced natural killer T-cells in B-cell chronic lymphocytic leukaemia identified by three monoclonal antibodies: Leu-11, A10, AB8.28.

The proportion of E-rosette positive T-lymphocytes capable of reacting with three monoclonal antibodies (MoAbs)--Leu-11, A10, AB8.28--which appear to recognize specifically natural killer (NK) cells, was assessed in a series of untreated patients with B-cell chronic lymphocytic leukaemia (B-CLL). Irrespective of the clinical stage of the disease, the capacity of B-CLL T-lymphocytes to react with all three MoAbs was significantly reduced compared with that of normal circulating T-cells (Leu-11: 2.5% +/- 1.9 SD; A10: 2.3% +/- 1.3; AB8.28: 7% +/- 6.6 v Leu-11: 13.5% +/- 4.5; A10: 8.5% +/- 4.6; AB8.28: 12% +/- 5.5). Furthermore, a marked difference was demonstrated between the reactivity with the Leu-11, A10, AB8.28 MoAbs and the proportion of Leu-7 positive T-cells, which in B-CLL is significantly higher than in normal blood (23% +/- 12.1 v 11.9% +/- 5.9). These findings are in agreement with previous evidence of a discrepancy in B-CLL between the phenotypic expression, assessed by Leu-7 positivity, and the true functional activity of NK T-cells, and suggest that the Leu-11, A10 and AB8.28 MoAbs correlate well with the depressed NK function found in this disease.

Release of platelet-activating factor in human leukemia.

Cellular release of platelet-activating factor (PAF) was assessed in a series of human acute and chronic lymphoid and myeloid leukemias at presentation or in an active phase of the disease. PAF-like material, showing physicochemical properties similar to those of synthetic PAF and of PAF released from IgE-sensitized rabbit basophils, was found in cultures of cells from 5 of 6 acute lymphoblastic leukemias (ALL) (2 of 2 T-ALL and 3 of 4 common ALL) and from 13 of 24 B-cell chronic lymphocytic leukemias after stimulation with ionophore A23187 with or without phytohemagglutinin in the presence of acetyl coenzyme A. On the other hand, PAF was released only from 2 of 10 acute myeloblastic leukemias; both of them were of the more mature monoblastic subtype or M5 according to the French-American-British classification. Cells from all three cases of chronic myeloid leukemia studied were also capable of producing PAF. In eight cases of acute lymphoid and myeloid leukemia, the in vivo release of PAF was assessed by testing the plasma levels of this mediator. Only in two cases (one ALL and one acute myeloblastic leukemia) could detectable levels of circulating PAF be demonstrated; it is of interest that both of these cases showed clinical and hematological features of disseminated intravascular coagulation. No PAF was documented in the plasma of the five chronic leukemias tested (four B-cell chronic lymphocytic leukemias and one chronic myeloid leukemia). These findings indicate that lymphoid and myeloid leukemic cells have a different capacity of releasing PAF, possibly related to the level of cell differentiation rather than to an intrinsic property of the neoplastic cells. Furthermore, in some cases, an intravascular release of PAF may occur.

Interleukin 2 (IL 2) and interferon-gamma production by T lymphocytes from patients with B-chronic lymphocytic leukemia: evidence that normally released IL 2 is absorbed by the neoplastic B cell population.

The capacity of T lymphocytes from patients with B cell chronic lymphocytic leukemia (B-CLL) to release interleukin 2 (IL 2) and interferon (IFN)-gamma was assessed following various stimuli. The spontaneous release of IL 2 and IFN-gamma was practically absent both with B-CLL and normal T lymphocytes. By contrast, after stimulation with phytohemagglutinin (PHA) or with PHA plus 12-O-tetradecanoylphorbol-13-acetate, the production of IL 2 and IFN-gamma by B-CLL T lymphocytes was similar to that of normal T lymphocytes, irrespective of the reversed T lymphocyte subset distribution (OKT4/OKT8 ratio) observed in B-CLL. However, the titer of IL 2 was greatly reduced when autologous leukemic B cells were added to the culture system. Unlike IL 2, the presence of leukemic B cells did not affect the titer of IFN-gamma in the culture supernatants. The indication that IL 2 may be adsorbed in vivo by the neoplastic B cells was further confirmed by the demonstration of the IL 2 receptor (revealed by anti-Tac monoclonal antibody) on the leukemic B cells, particularly following mitogenic stimulation, and by the evidence that exogenous IL 2 can be directly absorbed by untreated B-CLL T lymphocytes to release IFN-gamma and IL 2 is preserved, but that IL 2 may be rapidly removed by the neoplastic B-CLL cells, thus contributing to the well-documented T lymphocyte abnormalities present in this disease.

Discrepancy between phenotypic and functional features of natural killer T-lymphocytes in B-cell chronic lymphocytic leukaemia.

The phenotypic expression and functional capacity of natural killer (NK) T-lymphocytes (E+, OKT3+) were analysed in a series of untreated patients with B-cell chronic lymphocytic leukaemia (B-CLL). The mean value of NK activity of B-CLL T-lymphocytes, tested against the K562 cell line, was significantly depressed (P less than 0.01) in the 20 cases studied, compared with that of normal T-cells. Incubation with human leucocyte interferon produced an increase (P less than 0.05) in NK activity, although the mean value was still significantly lower (P less than 0.05) than that obtained with normal T-cells. Furthermore, the formation of effector-target conjugates was significantly lower (P less than 0.01) among B-CLL T-cells compared with normal T-lymphocytes. Despite the reduced NK functions observed in the majority of B-CLL patients, the capacity of T-cells to react with the monoclonal antibody (MoAb) Leu-7 (HNK-1 clone), assessed in 60 patients, was significantly higher (P less than 0.001) in B-CLL than in normal blood (mean 24% +/- 10.6 SD v 9% +/- 4.2), irrespective of the clinical stage of the disease. These findings suggest that the reduced cytotoxic ability of B-CLL T-lymphocytes may be due either to an expanded population of immature T-cells which already express a cytotoxic-like phenotype (E+, OKT3+, HNK-1+) but which lack adequate cytotoxic functions, or, alternatively, to an intrinsic defect of the natural effectors present within the T-cell population of B-CLL. The T-cell functional abnormalities documented in this study, together with other defective functions previously described, may be implicated in some of the complications frequently associated with B-CLL, particularly the high incidence of secondary neoplasms. There is growing evidence that natural cytotoxicity may play a major role in the immune surveillance system, both against tumour cells and virus-infected cells (Herberman & Ortaldo, 1981). Natural killer (NK) cells are a morphologically homogeneous population of large granular lymphocytes with azurophilic granules in the cytoplasm (Timonen et al, 1981), which are non-adherent and express receptors for the Fc portion of IgG. About 50% of these cells form rosettes with sheep red blood cells (E-rosettes) (West et al, 1977). A monoclonal antibody (MoAb) which appears to react with practically all human NK cells (HNK-1 clone, Leu-7; Becton Dickinson) has been recently produced (Abo & Balch, 1981).(ABSTRACT TRUNCATED AT 400 WORDS)

Immature T lymphocytes in human cord blood identified by monoclonal antibodies: a model for the study of the differentiation pathway of T cells in humans.

The reactivity of human cord blood lymphocytes was assessed against a panel of monoclonal antibodies (MoAb). The mean proportion of OKT3+ cells (pan-T) was significantly lower in cord blood (52 +/- 13.8%; mean +/- SD) compared with that of adult blood (75 +/- 8.9%) and paralleled well with the E-rosette-forming capacity (50 +/- 16.3%). Both the proportions of OKT4+ cells (helper/inducer phenotype) and of OKT8+ cells (suppressor/cytotoxic phenotype) were significantly reduced in cord blood (43 +/- 11.8% vs 50.3 +/- 7.4% and 20 +/- 10.3% vs 25.6 +/- 6.0%, respectively), while the overall OKT4/OKT8 ratio was increased compared with adult blood (2.87 +/- 1.83 vs 2.04 +/- 0.61). Unlike adult blood, in 30 of the 35 samples of cord blood an overlap was observed between the total proportion of OKT4+ and OKT8+ cells (65 +/- 15.2%) and that of OKT3+ cells (52 +/- 14.3%). Although small numbers of cells coexpressing both antigens were occasionally found, double-staining analysis showed that the overlap in cord blood was mostly due to an expanded proportion of OKT3 (Leu-4)-/OKT8 (Leu-2)+ cells. Relevant proportions of OKT6+ (common thymocyte antigen) and OKT10+ (thymocytes, activated T cells, precursor cells) cells were found in cord blood as opposed to adult blood (10.8 +/- 8.6% vs 0.6 +/- 0.6% and 67 +/- 18.0% vs 8 +/- 2.1%, respectively), while terminal deoxynucleotidyl transferase-positive cells were observed only in two samples of cord blood. A small proportion of T cells (E-rosette+) reacted with the MoAb OKIa1 (HLA-DR). Finally, the proportion of cord blood cells recognized by the MoAb Leu-7 (HNK-1 clone) was almost negligible compared with adult blood (2.8 +/- 2.4% vs 15 +/- 7.5%). These data confirm the immaturity and heterogeneity of cord blood lymphocytes and demonstrate the presence at birth of circulating lymphocytes which express a surface phenotype reminiscent of that found in the late stages of intrathymic differentiation and in some human T-cell leukemias. Human cord blood may thus represent a suitable model for the study of the differentiation pathway of normal and pathological T-cells in humans.

Release of platelet-activating factor (PAF)-like material from human lymphoid cell lines.

The release of platelet-activating factor (PAF)-like material was evaluated from eight B and T human leukemic cell lines and from three lymphoblastoid cell lines obtained after Epstein-Barr virus (EBV) transformation. All cell lines were stimulated with ionophore A23187, phytohemagglutinin (PHA), or both either in the presence of or without acetyl coenzyme A (AcCoA). After appropriate stimuli, all cell lines tested released PAF-like material, irrespective of the surface phenotype and of the level of differentiation. Some cell lines released PAF-like material after stimulation with A23187 (Molt-4, Jurkat, 1301, Raji, FM007), others only if AcCoA was added to A23187 (CCRF/CEM, HUT 78, Daudi, Nalm-1). In some cases, PHA was capable of enhancing the effect of A23187 (Molt-4, HUT 78, Daudi, IBW4) and finally, in a lymphoblastoid cell line (WT20), the release of PAF-like material was achieved only in the presence of both A23187 and PHA. These findings demonstrate for the first time that human lymphoid cells both of B- and of T-cell origin can release a PAF-like material after appropriate stimuli.

Effects of 12-o-tetradecanoyl-phorbol-13-acetate (TPA) on the colony growth of human T lymphocytes.

The T colony promoting activity of 12-o-tetradecanoyl-phorbol-13-acetate (TPA) was assessed in a double layer culture assay which is dependent on the simultaneous presence of phytohaemagglutinin (PHA) and a leucocyte rich underlayer. TPA (10(-8) M) incorporated in the overlayer in place of PHA was capable of promoting T cell growth in the form of clusters in all 37 experiments performed and in the form of colonies in more than 50% of the samples tested. However, the T colony promoting activity of TPA alone was markedly less evident and consistent than that of PHA (mean 13 +/- 19.9 s.d. colonies vs 168 +/- 78.6). TPA concentrations of 10(-6) M, 10(-9) M and 10(-10) M were practically ineffective. On the other hand, the number of colonies obtained when both TPA 10(8) M and PHA were incorporated in the overlayer was significantly higher (P less than 0.01) than that observed with PHA alone (mean 250 +/- 108.2 vs 178 +/- 84.5 colonies). When TPA concentrations of 10(-9) M and 10(-10) M were used in addition to PHA, the enhancing effect was less evident, while an inhibition of T colony growth was observed with TPA 10(-6) M + PHA. TPA 10(-8) M was also capable of enhancing T colony growth when incorporated in the leucocyte rich underlayer (222 +/- 98.6 vs 172 +/- 80.9 colonies). In all cultures with TPA the peak of growth was delayed compared with that of control experiments with PHA. These findings demonstrate that TPA, particularly when co-cultured with PHA, is an effective T colony promoting agent. The observation that the number of colonies formed in the presence of TPA plus PHA is higher than the sum of those observed with the two stimulators independently, suggests that their synergistic effect may be mediated via the production of colony stimulating soluble factors.

Normal helper T-cell function in hairy-cell leukaemia.

The function of peripheral blood T-lymphocytes was tested in a series of untreated patients with hairy-cell leukaemia (HCL). The in vitro T-colony forming capacity fell within the normal range in 20 of the 22 cases studied. No difference in colony growth was found between splenectomized and non splenectomized patients. The ability to induce differentiation of normal B-lymphocytes into antibody producing plasma cells, in a pokeweed mitogen (PWM) stimulated system, was similar to that of normal T-cells in all 6 cases studied. Both the T-colony growth and the helper capacity were unrelated to the distribution of T-lymphocyte subsets defined by monoclonal antibodies, despite an increase in OKT8+ cells (suppressor/cytotoxic phenotype) and a decrease in OKT4+ cells (helper/inducer phenotype) in 44% of the patients. The significance of these findings is discussed in relation to T-cell abnormalities previously described in other B-cell leukaemias, chiefly chronic lymphocytic leukaemia (B-CLL). The preserved T-cell function in HCL correlates well with the normal humoral immunity found in this disease.

[Effect of human plasma on the in vitro formation of T-lymphocyte colonies]. / Effetto del plasma umano sulla formazione in vitro di colonie T-linfocitarie.

T-lymphocyte colonies can be obtained from normal human peripheral blood and bone marrow, using a double layer technique which requires the simultaneous presence of peripheral blood irradiated leucocytes in the underlayer and phytohaemagglutinin in the overlayer (Foa & Catovsky, Clin. Exp. Immunol., 36, 488, 1979). The absence of either of these two factors unables colony formation. In an attempt to improve the plating efficiency 15% autologous plasma was added to the leucocyte rich underlayer, or tested alone. The results demonstrated that both leucocytes and plasma alone show a similar colony promoting activity, giving rise respectively to 160 +/- 80.4 SD and 180 +/- 113.5 SD T-colonies. The combination of both factors produced a significant increase in growth, with an overall mean of 410 +/- 190.5 colonies. Increasing concentration of plasma (1%, 5%, 10%, 15%) gave rise to a dose-dependent increase in colony growth. Furthermore, the simultaneous presence of leucocytes and plasma in the underlayer enabled good colony formation with as little as 0.1 x 105 tes cells in the overlayer. These findings indicate that both human peripheral blood leucocytes and plasma possess T-colony stimulating activity, and that optimal growth is achieved with the combination of these two factors in the underlayer.

[OK T-positivity in stimulated T-lymphocytes]. / OK T-positività' in linfociti T stimolati.

T-lymphocytes express different antigenic determinants which can be recognized using specific anti-T monoclonal antibodies. OK T3 (Ortho Diagnostics, Raritan, N.J.) detects 95% circulating T-lymphocytes, while OK T4 reacts with helper/inducer T-lymphocytes and OK T8 with suppressor/cytotoxic T-lymphocytes. In normal peripheral blood the proportion of mononuclear cells (after "Lymphoprep' separation) positive with the various anti-T monoclonal antibodies is, according to our standards, as follows: OK T3 77 +/- 9.4%, OK T4 51 +/- 7.8%, OK T8 27+/- 6.4%. In this study we have evaluated the positivity with OK T MoAb after activation and proliferation of the T-cell population in a double layer T-lymphocyte colony assay. After 4-5 days of incubation, the proportion of OK T3 + cells had increased to 94 +/- 4.6%, while that of OK T4+ and OK T8+ had raised to 62 +/- 14.1% and 65 +/- 7.1% respectively. These data suggest that T-colony formation gives rise to an increased expression of OK T8 positivity, possibly through a mechanism of T-cell activation (shown also by the 'Ia' positivity), and/or of proliferation of T-cells with a double antigenic phenotype.