Inhibition of erythroid cell growth by allogeneic murine lymphocytes. Evidence for a synergism between lymph node cells and thymocytes

Murine lymphoid cells from thymus and lymph nodes were tested for synergistic response in a graft-vs-host test. The test is based on the principle that allogeneic lymphocytes inhibit erythroid cell proliferation in the spleens of irradiated mice infused with syngeneic bone marrow cells.I was observed that mixtures of thymocytes and lymph node cells from the same parental strain yielded graft-vs-host responses in irradiated F₁-hybrids higher than expected by summing the responses of the two cell populations tested separately. A similar synergistic response was obtained using mixtures of thymocytes and lymph node cells obtained from the two parental strains of the hybrid, whereas such an effect was not detected using mixtures of lymph node cells or mixtures of thymocytes from the two parental strains. Nor could synergy be demonstrated between parental strain lymph node cells and thymocytes syngeneic with the bone marrow target cells. Thymocytes obtained from one parental strain which were injected into its irradiated F₁-hybrid transformed into a population of sensitized cells in the spleens of the recipients. This transformation was suppressed by the simultaneous injection of lymph node cells from the second parental strain. Since there is a synergistic immune response by such cell mixtures it is concluded that thymocytes may enhance the graft-vs-host response of lymph node cells. Parental strain thymocytes and lymph node cells, the latter being specifically immunologically tolerant to the bone marrow target cells, failed to give a synergistic response indicating that thymocytes do not transform unresponsive lymphocytes into responsive, but rather enhance the reactivity of existing, specifically responsive cells.The results thus show that thymocytes may enhance the response of lymph node cells in this specific graft-vs-host assay.     

Effect of tumor immunity on the distribution of labeled lymphoid cells in tumor-challenged mice

The distribution of ⁵¹Cr-labeled lymphoid cells from normal mice and mice immunized against a tumor were compared after intravenous inoculation of the labeled cells into normal syngeneic recipients. Spleen cell preparations from immune donors contained increased percentages of spleen and bone marrow-seeking cells, thus suggesting expansion of these cell populations when immunity to a tumor exists. Homing of labeled normal cells in tumor cell-injected normal animals was somewhat different from that seen in tumor cell-inoculated mice that were immunized against the tumor. In the latter case, accumulations of lymph node and spleen cells in recipient lymph nodes and bone marrow were consistently lower. In contrast, lymphoid cells from animals immunized against the tumor were found to accumulate in virtually the same percentages in lymphoid organs of normal and immune recipients. The behavior of lymphoid cell populations from thymus or bone marrow that consist mainly of precursor cells was unaffected by presence of malignancy and/or tumor immunity.     

Induction of the immune response to cell surface antigens in vitro

Summary Two tissue culture incubation systems are described in which immune responses to cell surface antigens have been demonstrated In the one-way “mixed lymphocyte interaction” system, a specific stimulation of thymidine uptake was induced by a particulate membrane antigen fraction, the microsomal lipoproteins (MLP)when low levels (0.01 to 0.001 μg per ml) were incubated with spleen or lymph node cells from nonsensitized mice. No stimulation was seen when allogeneic MLP was used at high levels, 10 μg per ml, nor at any level with syngeneic MLP. Specific effectors were demonstrated after 72-hr incubation with stimulatory levels of allogeneic MLP in three separate in vitro assays, a plaque-forming cell reduction assay, a tumor target assay, and an antigen-binding cell assay. In the latter assay, [¹²⁵I]MLP was used as the source of antigen. This system has limited potential inasmuch as mouse spleen cells do not survive in it beyond the 4th day of culture. The second tissue culture system, the Marbrook system, has much greater possibilities because at least 25% of the inoculum is recovered 7days later. In this culture system a cell-free sheep erythrocyte membrane preparation can induce, plaque-forming cells in the absence of macrophages. Using a sensitive radioimmunoassay, frees specific antibody was detected in culture supernatant fluids. With the same culture system, allogeneic lymphocytotoxic cells (killer) have been induced with spleen cells from unprimed mice in strains differing at the major histocompatibility locus (H-2). Allogeneic MLP induced very significant “killer” cell activity with spleen cells from primed mice. In a syngeneic tumor systems, significant amounts of killer cell activity were induced with unprimed spleen cell inocula, and much larger amounts induced with spleen cells from immunized mice. Presented in the formal symposium on Carcinogenesis in Vitro, at the 25th Annual Meeting of the Tissue Culture Association, Miami Beach, Florida, June 3–6, 1974. This work was supported by Public Health Service Rescarch grants CA 07973 and CA 10815 from the National Cancer Institute.     

Inactivation of stem cells by allogenic lymphocytes: competition between T1- and T2-subpopulations]

Transplantation of the bone marrow cells with allogeneic T-lymphocytes to the irradiated hosts was accompanied by inactivation of the stem elements of the graft. The lymph node cells of T-mice (those deprived of B cells) were more active than the spleen cells of these mice. The stem cells inactivation was weakly expressed or absent in case of a combined acti-n of T-cells from the lymph nodes and the spleen.     

The effect of colony stimulating factor on the synthesis of ribonucleic acid by mouse bone marrow cells in vitro.

The effect of granulocyte-macrophage colony stimulating factor (GM-CSF) on the synthesis of RNA in liquid cultures of mouse bone marrow, spleen, thymus, peritoneal, peripheral blood leukocytes and lymph node cells was investigated. GM-CSF appeared to stimulate RNA-synthesis in syngeneic bone marrow cells within ten minutes of adding it to the culture. In the presence of GM-CSF bone marrow cultures maintained their initial rate of RNA synthesis for approximately ten hours. GM-CSF had no apparent effect on the uptake of 3H-uridine into bone marrow cells. This stimulation was still observed in the presence of puromycin and cycloheximide, but was abrogated by actinomycin D. The magnitude of the stimulation was not affected by the density of cells between 1 and 20 x 10(6) cells/ml but was slightly smaller at 0.1 and 40 x 10(6) cells/ml. Increasing concentration of GM-CSF (up to 2 X 105 units per ml) led to increased stimulation of RNA synthesis in bone marrow cells, but a significant stimulation could be detected at concentrations as low as 800 units/ml. GM-CSF did not significantly stimulate RNA synthesis in spleen, thymus, mesenteric or subcutaneous lymph node cells. However a small stimulation was observed in peripheral blood leukocytes and peritoneal cells. Autoradiographic studies showed that GM-CSF stimulated RNA synthesis in blast cells, myelocytes, metamyelocytes and polymorphs. Nucleated erythroid cells showed no increased labeling with GM-CFS. Labeling in lymphoid-like cells was highly variable but the level of labeling did not appear to be influenced by GM-CSF.     

Mechanisms of in vivo generation of cytotoxic activity against allograft: 1. Local differentiation of mature cytotoxic T lymphocytes in rejection of allogeneic lymphocytes

Although cytotoxic activity was not detected within the spleen and regional lymph nodes from mice immunized sc with allogeneic lymphocytes, such activity was detected consistently in glass-nonadherent and anti-θ-sensitive peritoneal exudate cells (PE cells) from Day 5 after immunization and reached a maximum by Day 7. Immunized spleen cells developed cytotoxic T lymphocytes (CTLs) earlier and more effectively than normal spleen cells when transferred ip into X-irradiated syngeneic normal mice together with immunizing antigen, while they did not become cytotoxic when transferred without antigen. These results suggest that spleen and lymph node cells which may have differentiated into some transitional state by in vivo immunization may differentiate into mature CTLs, following direct contact with antigen at the site of graft. CTLs generated there appear to be responsible for the rejection of allogeneic lymphocytes. Cytotoxicity of PE cells was also generated in X-irradiated mice and augmented cytotoxicity was generated by treatment with cyclophosphamide.     

Studies of the T-lymphocytes resident in the spleens of thymectomized, irradiated, bone marrow-reconstituted (TXB) mice.

The T-lymphocytes resident in the spleens of thymectomized, lethally irradiated mice that had been reconstituted with syngeneic bone marrow (TXB) were characterized. Both recently reconstituted N-TXB, (approximately 3 weeks after bone marrow injection) and aged (>6 months after reconstitution) A-TXB animals were studied. The T-lymphocytes from spleens of recently reconstituted N-TXB donors did not respond to PHA but did react significantly to Concanavalin A (Con A). The lack of PHA sensitivity was not due to dilution of reactive cells by other cell types. Removal of adherent cells, likewise, did not restore N-TXB spleen cell PHA responsiveness. N-TXB splenic T-cells were cortisone resistant. N-TXB spleen cells by themselves did not cause a graft vs host response. However, N-TXB spleen cells amplified the graft vs host response of normal lymph node cells but not N-TXB lymph node cells. Addition of cyclic GMP enhanced [³H]thymidine uptake of N-TXB spleen cells caused by Con A. N-TXB spleen cells were exclusively spleen seeking. The Con A reactive cell within N-TXB spleens was demonstrated to be of donor origin. Fetal liver as well as syngeneic bone marrow contained cells capable of reconstituting the Con A response. Spleen cells from aged. (>6 months) A-TXB were found to be PHA sensitive. Competitive inhibition assays measuring θ expression in A-TXB spleen cells indicate a significant increase in the θ positive lymphocyte population occurred with time. The data indicate that considerable reconstitution of θ positive cells had occurred in A-TXB donors. The results also suggest that the T-lymphocyte population of the TXB spleen may be a unique subpopulation of T-lymphocytes that resides exclusively in spleen and bone marrow.     

Elevated cyclic AMP levels in mouse lymphoid tissue after stimulation by cholera enterotoxin in vitro (38468)

Addition of CT to suspensions of thymus, lymph node, spleen, or bone marrow cells in vitro resulted in a marked accumulation of cAMP with peak levels occurring 4-5 hr after incubation of cells with CT. Thymus cells showed the largest increase in cAMP, approximately 40-fold at 10 ng/ml CT. Bone marrow cells accumulated the least cAMP (1.5x), while intermediate levels were observed for spleen and lymph node cells (10-12x). Antiserum to CT prevented stimulation of increased cAMP levels. Repopulation studies using X-irradiated mice also showed that thymus-derived spleen cells accumulated more cAMP/10-7 cells than spleen cells from recipients given spleen or marrow cells. Spleen cells from athymic (nu/nu) mice also responded much less than did spleen cells from normal mice. Thymocytes appeared to bind CT to a greater degree than bone marrow cells. Spleen and lymph node cell suspensions also contained CT-binding cells and the number of CT-binding cells in these peripheral lymphoid tissues appeared approximately equal to the summation of the numbers observed in thymocyte and bone marrow cell suspensions. Stimulation of cAMP in lymphoid cells, especially thymocytes, by CT provides a pharmacological tool to investigate the mechanism and role of this nucleotide in the early events of antibody formation.     

In vitro viability of lymphoid cells from lines of mice genetically selected for high or low responsiveness to phytohemagglutinin

The kinetics of viability of lymph node and spleen cells of mice genetically selected for "high" or "low" in vitro lymphocyte responsiveness to PHA were studied in PHA or PPD-stimulated short-term cultures. Lo/PHA cells were found to be less viable than Hi/PHA cells in unstimulated control cultures. PHA improved the viability of Lo/PHA cells while inducing proliferation of Hi/PHA cells with the appearance of more and larger lymphoblasts in the latter. PPD only improved the viability of spleen cell cultures, more so for the Hi/PHA line. The interline difference in thymidine uptake was smaller after PPD than after PHA stimulation. Modifications of culture conditions designed to decrease the interline difference in cell viability lessened but did not abolish the separation between the two lines for the PHA response as measured by thymidine uptake.     

Kinetics of CTL induction by concanavalin A.

Induction of maximal CTL activity was achieved within 12 hr of exposure to Con A in vitro in various mouse lymphoid cell populations. These included spleen cells from normal unsensitized mice, spleen cells from mice previously immunized with alloantigen, and mouse spleen cells exposed to alloantigen in long-term mixed leukocyte culture (LTMLC). Although induction of maximal incorporation of tritiated thymidine was accomplished within this same period in the cells obtained from LTMLC, a much longer period of Con A exposure (greater than 24 hr) was required for freshly prepared spleen cells from normal or previously immunized mice. These findings indicate that the increased tritiated thymidine uptake induced in freshly prepared spleen cells on continued exposure to Con A beyond 12 hr is not associated with the development of cytolytic activity, and that it probably represents stimulation of subpopulations no longer present in the LTMLC population where positive selection for cells responsive to cellular alloantigens has taken place.     

Effect of selective T cell depletion of host and/or donor bone marrow on lymphopoietic repopulation, tolerance, and graft-vs-host disease in mixed allogeneic chimeras (B10 + B10.D2----B10)

Reconstitution of lethally irradiated mice with a mixture of T cell-depleted syngeneic plus T cell-depleted allogeneic bone marrow (B10 + B10.D2----B10) leads to the induction of mixed lymphopoietic chimerism, excellent survivals, specific in vivo transplantation tolerance to subsequent donor strain skin grafts, and specific in vitro unresponsiveness to allogeneic donor lymphoid elements as assessed by mixed lymphocyte reaction (MLR) proliferative and cell-mediated lympholysis (CML) cytotoxicity assays. When B10 recipient mice received mixed marrow inocula in which the syngeneic component had not been T cell depleted, whether or not the allogeneic donor marrow was treated, they repopulated exclusively with host-type cells, promptly rejected donor-type skin allografts, and were reactive in vitro to the allogeneic donor by CML and MLR assays. In contrast, T cell depletion of the syngeneic component of the mixed marrow inocula resulted in specific acceptance of allogeneic donor strain skin grafts, whether or not the allogeneic bone marrow was T cell depleted. Such animals were specifically unreactive to allogeneic donor lymphoid elements in vitro by CML and MLR, but were reactive to third party. When both the syngeneic and allogeneic marrow were T cell depleted, variable percentages of host- and donor-type lymphoid elements were detected in the mixed reconstituted host. When only the syngeneic bone marrow was T cell depleted, animals repopulated exclusively with donor-type cells. Although these animals had detectable in vitro anti-host (B10) reactivity by CML and MLR and reconstituted as fully allogeneic chimeras, they exhibited excellent survival and had no in vivo evidence for graft-vs-host disease. In addition, experiments in which untreated donor spleen cells were added to the inocula in this last group suggest that the presence of T cell-depleted syngeneic bone marrow cells diminishes graft-vs-host disease and the mortality from it. This system may be helpful as a model for the study of alloresistance and for the identification of syngeneic cell phenotypes, which when present prevent engraftment of allogeneic marrow.     

Histamine production by alloantigen-activated mouse bone marrow cells

Histamine's contribution to the manifestations associated with graft-versus-host disease (GVHD) and/or hybrid resistance is unknown. Thus, we initiated studies to see whether or not mouse bone marrow cells could produce histamine upon alloantigen stimulation. Irradiated allogeneic spleen cells were shown to stimulate bone marrow cells to produce and secrete high levels of histamine. During 7 days of culture there was only a marginal increase in cell-associated histamine while the amount of histamine in the supernatant increased 10- to 20-fold. Optimal histamine production was dependent upon Lyt 1+2+ T cells resident in the bone marrow. Further, bone marrow cells from Nude mice failed to produce high levels of histamine following alloantigen stimulation. Soluble factors produced by alloantigen-stimulated bone marrow cells or by Con A-stimulated rat spleen cells induced high levels of histamine production in bone marrow cells in the absence of alloantigen. We suggest that histamine production by alloantigen-activated bone marrow cells may modulate immune functions following bone marrow transplantation.     

Effect of age of mouse recipients on the functional activity of transplanted hematopoietic and lymphoid cells]

When transplanting the bone marrow cells from adult C57BL mice to the lethally irradiated (CBA X C57BL) F1 hybrids of different age, the decrease of the colony forming activity of the stem haemopoietic cells was observed in the spleen of the older recipients, as compared with the 3 months old ones. The joint transplantation of the bone marrow and thymus cells resulted in both the cases in the stimulation of the growth of colonies. The number of endogenous colonies of haemopoietic cells arising in the spleen of animals following the sublethal irradiation was greater in younger hybrids. After the induction of the "transplant versus host" reaction by the lymph node or spleen cells from the CBA mice, the relative weight of spleen and regional lymph node, respectively, in the older recipients exceeded those in the younger ones.     

In vitro alloreactivity of mouse bone marrow lymphocytes

Before characterizing alloreactive cells of the bone marrow, it was necessary to reevaluate the alloantigen response in this tissue. The results of previous studies using the parental-F₁ system in the mixed-lymphocyte reaction (MLR) are open to question because of the recently documented proliferation of F₁ stimulator cells (W. H. Adler, T.Takiguchi, B. Marsh, and R. T. Smith,J. Immunol. 105, 984, 1970; P. F. Piguet, H. K. Dewey, and P. Vassalli, J. Exp. Med. 146, 735, 1977). The culture system was optimized for measuring the MLR of bone marrow lymphocytes enriched on sucrose density gradients. The proliferative response of the enriched fraction (BML) to 2000-R irradiated allogeneic spleen cells was three times as high as the response of unfractionated bone marrow. For maximal responses, antigen concentration had to be twice as high for the BML as for the lymph node, and in a time course study the highest [³H]TdR uptake occurred on Day 3 in BML cultures and on Day 5 in the LN. In lymph node semiallogeneic cultures stimulator cell proliferation can be disregarded, while semiallogeneic BML MLR err significantly on the high side. When BML were matched with allogeneic stimulator cells at the H-2 locus, they gave good MLR responses, provided there was a minor Mls histocompatibility locus difference, while in the lymph node the response was greatly diminished in similar mixtures. The differences in the BML and lymph node alloantigen responses with respect to antigen concentration, kinetics and susceptibility to F₁ and Mls stimulation, suggest that the bone marrow alloantigen response is mediated by a cell population that is different than alloresponsive cells in the lymph node.     

Effect of stimulating antibody formation under the influence of bone marrow cells in vivo]

Injection of intact bone marrow cells to mice at the peak of the secondary immune response results in a 2.4-fold increase of the number of antibody-forming cells in the regional lymph node. Preliminary injection of bone marrow cells to donors of the immune lymph node cells decreases the stimulation effect of antibody formation when the lymph node cells are subsequently cultivated with the intact bone marrow cells. The data obtained demonstrate the cell interaction at the level of mature antibody producers in vivo.     

Studies on the maturation of immune responsiveness in the mouse. II. Role of the spleen.

Experiments were designed to investigate the role of the spleen in the development of the murine immune system. By using mice splenectomized within 24 hr of birth, as well as mice with a hereditary, congenital absence of the spleen, the primary immune response to sheep erythrocytes was examined. The immunocompetence of lymph node cells from spleenless or control mice was assessed in vitro, in organ and in cell suspension cultures, and in vivo, by transfer into lethally irradiated syngeneic recipients followed by antigenic stimulation. The immunologic capacities of thymus and bone marrow cells were similarly tested by injection separately or in combination into irradiated syngeneic mice. Lymph node cells from spleenless animals appeared fully competent both in vitro and in transfer experiments. Neither neonatal splenectomy nor congenital absence of the spleen significantly reduced the capacity of bone marrow or thymus cells to participate in the immune response to sheep erythrocytes.     

Strain difference in the radiosensitivity of immunocompetent cells and its influence on the residual host-vs-graft reaction in lethally irradiated mice grafted with semiallogeneic bone marrow

A striking difference in radiosensitivity was noted between C3H/He (H-2k) and C57BL/6J (H-2b) strain mice when assessed by primary anti-SRBC PFC response of intact animals and primary cell-mediated lympholysis (CML) response of spleen cells to allogeneic cells in vitro, the C3H strain being more radioresistant. On the other hand, when C3H and B6 mice were exposed to 6.62 to 10.40 grays (Gy) of x-rays and then were transplanted with 2 X 10(6) bone marrow cells from B6C3F1 (H-2b/k) donor mice within 3 hr or at 24 hr after radiation exposure, the early mortality caused by residual host-vs-graft (HVG) reaction was much higher when C3H mice were used as recipients. Furthermore, the proportion of surviving animals manifesting host-type lymphohemopoiesis, i.e., host-type revertants, was much higher in B6C3F1 to C3H than in B6C3F1 to B6 combination. Spleen cells from such host-type revertants manifested strong anti-donor reactivity when assessed by mixed lymphocyte reaction (MLR) and/or CML in vitro. Increase of radiation doses to the recipients to 10.40 Gy resulted in 100% survival and 100% donor-type lymphohemopoiesis in both groups of chimeras. These results indicate strongly that a genetic difference in radiosensitivity of immune system of the recipients can greatly influence the magnitude of residual HVG reactions observed in hybrid to parental strain bone marrow transplantation in mice.     

In vitro studies on the mechanism of acquired resistance to tuberculous infection. II. The effects of the culture supernatants of specifically stimulated-sensitized lymphocytes on the growth of tubercle bacilli within macrophages.

Immune lymph node cells were obtained from mice immunized with bovine gamma globulin (BGG) in complete Freund's adjuvant or allogeneic MH134 tumor cells. They showed the capacity of conferring bactericidal activity on macrophages infected with Mycobacterium tuberculosis, H37Rv, when they were incubated on macrophage monolayers together with the corresponding antigen, i.e., BGG or solubilized cellular antigen of the tumor cells. However, such capacity was lower than that of tubercle bacilli-immune lymph node cells. Culture supernatants were harvested after incubation of tubercle bacilli-immune, BGG-immune or allogeneic tumor-immune lymph node cells with the corresponding antigen for 24 hr. Macrophages were altered so as to suppress intracellular bacillary growth when macrophage monolayers were exposed to the supernatants for more than 2 days. When normal lymph node cells were incubated on normal macrophage monolayers together with a mitogen such as PHA or concanavalin A, growth of tubercle bacilli within the macrophages was slightly but difinitely suppressed. The mechanism of elicitation of cellular immunity to the infection with tubercle bacilli is discussed on the basis of results presented in this and the preceding paper.     

Administration of IL-7 to normal mice stimulates B-lymphopoiesis and peripheral lymphadenopathy.

Normal mice were injected with IL-7 (500 ng, twice daily) for various periods of time up to 6 days and the cellularity and phenotypic composition of the thymus, spleen, lymph node, and bone marrow was assessed. After 6 days of treatment, significant increases in the cellularity of the spleen, lymph node, and bone marrow were observed which returned to the normal range within 6 days after cessation of treatment. After 3 days of IL-7 treatment, increased numbers of B220+/surface(s) IgM- bone marrow cells were observed. After 6 days of treatment, these numbers were still further increased and a significant population of B220+/sIgM- cells were observed in the spleen. The numbers of c mu+/sIgM- cells were also increased in the IL-7-treated mice. Analysis of the expression of B220 and BP-1 on the sIgM- bone marrow cells revealed that the B220+/BP-1+ population was dramatically increased after IL-7 treatment and the size of the B220+/BP-1- population did not differ from control mice. The pre-B cell numbers declined rapidly after the cessation of IL-7 treatment. After 6 days of IL-7 treatment, a twofold increase in the number of B cells in the spleen and lymph node was observed. The B cell numbers declined to normal values within 6 days after the cessation of IL-7 administration. In the spleens of the IL-7-treated mice, there was a significant increase in the number of B cells with an immature phenotype (e.g., sIgMhi/sIgDlo, decreased levels of Ia and FcR expression). The numbers of CD8+ and CD4+ T cells were also increased in the lymph node and spleen of the IL-7-treated mice. These numbers declined to normal levels after the cessation of IL-7 treatment.     

Induction of graft versus leukemia effects by cell-mediated lymphokine-activated immunotherapy after syngeneic bone marrow transplantation in murine B cell leukemia

 The feasibility of inducing graft versus leukemia (GVL) effects with allogeneic T cells in recipients of autologous bone marrow transplantation (BMT) was studied in a murine model (BCL 1) of human B cell leukemia/lymphoma. Allogeneic cell therapy, induced by infusion with peripheral blood lymphocytes, a mixture of allogeneic spleen and lymph node cells and allogeneic activated cell therapy, induced by in vitro recombinant-interleukin-2(rIL-2)-activated allogeneic bone marrow cells in tumor-bearing mice, prevented disease development in adoptive BALB/c recipients. Concomitant in vivo activation of allogeneic lymphocytes with rIL-2 suppressed even more effectively the development of leukemia in secondary adoptive recipients of spleen cells obtained from treated mice. In contrast, in vivo administration of rIL-2 after syngeneic BMT, with or without equal numbers of syngeneic lymphocytes, led to disease development in secondary recipients. Our data suggest that effective cell therapy can be achieved after SBMT by allogeneic but not syngeneic lymphocytes and that anti-leukemic effects induced by allogeneic lymphocytes can be further enhanced by in vitro or in vivo activation of allogeneic effector cells with rIL-2. Therefore, cell therapy by allogeneic lymphocytes following autologous BMT could become an effective method for inducing GVL-like effects on minimal residual disease provided that graft versus host disease can be prevented or adequately controlled. Received: 14 May 1996 / Accepted: 6 August 1996