The suppressive effect of a supernate from concanavalin A-activated human lymphocytes: effects of concanavalin A-activated lymphocytes and their supernates on cytotoxic and mixed lymphocyte reactions.

The effects of Concanavalin A-treated human peripheral blood lymphocytes and their supernatants were evaluated on the MLC reaction and on the generation of cytotoxic lymphocytes assessed by cell-mediated lympholysis (CML). Experiments were performed with both allogeneic and xenogenic sensitization. It was found that Con A-activated cells suppressed the MLC and CML reactions in allogeneic and xenogeneic systems. On the other hand, the SIRS-like supernate was able to suppress the MLC reaction and blastogenesis, but had no suppressive effect on the generation of cytotoxic lymphocytes. We found no difference in the magnitude of suppression, whether or not Con A-activated lymphocytes were syngeneic to the responder cells. This finding suggests that there is no requirement for allogeneic restriction in the interaction between suppressor and suppressed cells, and demonstrates a soluble human suppressor substance capable of suppressing some cell-mediated reactions.     

Linear density gradient separation of human lymphocyte subsets. II. Characterization of cells responding in secondary MLC and CML.

Lymphocytes were separated on linear density gradients (LDG) after they had been sensitized in vitro against allogeneic cells and had reverted to small cells. Cells from individual density fractions were restimulated with autologous, specific, or third-party cells and assayed 48 hr later for their response in secondary mixed leukocyte culture (MLC) and cell-mediated lympholysis (CML). Memory cells capable of responding in secondary MLC were broadly distributed and found in both heavy and light fractions. The various density classes of memory cells differed with respect to the degree of their specificity for the restimulating cells. In secondary MLC the greatest specificity for the originally sensitizing cells and the least cross-reactivity for third-party cells were primarily features of light- and medium-density cells. Memory killer cells for CML were fairly homogeneously grouped. Following restimulation, killers were enriched in light to medium fractions also, as was previously seen at the peak of the response on Day 6.     

Proliferative and cytotoxic immune functions in aging mice. IV. Effects of suppressor cell populations from aged and young mice

The changes with age in three splenic suppressor cell populations were studied in C57BL/6 mice. Allospecific Ts cells and nonspecific non-T suppressor cells were both generated in vitro in allogeneic MLC. The presence of "pre-existing" suppressor cells in fresh spleen cells from normal mice was examined. Suppressor cell activities were assayed for their ability to suppress proliferation in a fresh allogeneic MLC after treatment to prevent their own proliferation. The ability to generate both specific and nonspecific suppressor cells decreased with age, whereas pre-existing suppressor cells were detected in spleens from the majority of the aged animals but not in spleens from young animals. The decrease in suppressor cell activity was not due to any requirement for age matching between donors of suppressor and target cells. The specific and nonspecific MLC-generated suppressor cells inhibited both the proliferative response in the assay MLC and the generation of cytotoxic cells. The pre-existing suppressor cells only suppressed the proliferative response and not the generation of cytotoxic cells. The changes seen with age in these suppressor cell populations suggest that the ability to generate suppression (both allospecific and nonspecific) to newly encountered Ag declines with age, whereas a resident splenic suppressor cell population accumulates over the lifetime of the animals.     

The mechanism of cell-mediated cytotoxicity. II. The apparent biochemical requirements for cytolysis are influenced by the source and frequency of murine cytotoxic T lymphocytes

Cytotoxic T lymphocytes (CTL) were generated in allogeneic mixed-lymphocyte cultures (MLC) with the responder cells from immunized mice (secondary (II°) MLC.) These CTL could be maintained as proliferating cells for 2 to 4 weeks by culturing in a medium with T-cell growth factor (TCGF) activity and were 10 to 30 times as active as effectors obtained directly from immunized mice or those generated in primary (I°) MLC. Cytolysis by CTL from II° MLC was refractory to inhibition by azide, cyanide, or cycloheximide in contrast to previous reports that these agents were inhibitory. However, cytolysis by CTL from II° MLC was sensitive to azide when sufficient normal spleen cells were added to reduce the apparent CTL frequency to approximate that of the less active populations. Cycloheximide (Cx) partially inhibited cytolysis by CTL from immune spleen cells as shown earlier by others. However, CTL from II° MLC remained refractory to Cx even when diluted with normal spleen cells. These data show that CTL populations from various sources have both qualitative and quantitative differences that may alter their apparent biochemical requirements for CML. These considerations must be taken into account when attempting to define the minimal requirements for CTL-mediated killing.     

The cell-mediated lympholysis-inducing capacity of highly purified human monocytes and T lymphocytes in primary and secondary mixed leukocyte cultures

In this study the capacity of T cells and monocytes to induce cell-mediated lympholysis (CML) in primary and secondary MLC was investigated. The T lymphocytes were enriched by rosetting with sheep red blood cells (E) and further purified by sedimentation at unit gravity, which completely removed the contaminating monocytes. In addition, a highly purified monocyte population was obtained by 1 X G sedimentation of the non-E rosette-forming cells. These purified T cells have a poor CML-inducing capacity in primary and secondary MLC. In contrast, monocytes were very effective in inducing CML in both primary and secondary MLC. Induction of CML by monocytes in primary MLC was inhibited by heterologous anti-Ia-like antisera, indicating that the induction of CML by monocytes was related to the presence of HLA-DR (Ia-like) antigens on these cells.     

Alteration of the antigenicity of human lymphocytes by plant lectins and short-term tissue culture.

Human lymphocytes studied after being placed in culture for 1–6 wk progressively lost stimulating ability, i.e., lymphocyte defined antigens, when tested in one way mixed lymphocyte culture (MLC) but retained several other identifiable membrane components as well as the capacity to respond to mitogenic stimuli. Lymphocytes placed in culture with motogenic doses of PHA and Con-A after 1 and 2 wk strongly stimulated autologous responding fresh lymphocytes, but the MLC response of allogeneic fresh lymphocytes to stimulating lectin treated cells was even lower than the response to stimulating allogeneic cultured lymphocytes. The HL-A antigens on lectin treated cells or on lymphocytes through 6 wk in culture were clearly identifiable. Assays for T cell rosettes and B cell surface immunoglobulin showed both cell types to be present in numbers equal to fresh lymphocytes for up to 5 wk after culturing. However, the Fc receptor site on B cells was lost from cultured lymphocytes at the same time that MLC stimulation was lost. It is concluded that plant lectins can unmask new mitogenic sites on the cell surface as well as mask or delete existing sites, and that culturing lymphocytes for 1–6 wk will produce somewhat similar modulations. Modulation of surface membrane components by tissue culture or lectins may, therefore, have a profound effect in altering transplantation immunogenicity.     

Immunologic functions of isolated human lymphocyte subpopulations. III. Specific allogeneic lympholysis mediated by human T cells alone.

The studies presented herein have evaluated both the specificity and cellular basis of cell-mediated lympholysis (CML) in man. An efficient and quantitative 51Cr release assay was utilized to study the role of highly purified human T and B cells in CML. After in vitro sensitization human T cells develop the capacity to kill specifically allogeneic cells to which they were sensitized. In contrast, B cells were neither triggered to proliferate nor activated to kill allogeneic targets. B cells were not activated to kill even when sensitized in the presence of potentially "helper" T cells, nor did they block T cells from killing during the effector phase. Cell-free supernatants taken from active in vitro sensitization cultures were not lympholytic and did not modulate T cell killing. Hence, these studies show that both the afferent and efferent phases of human CML are T cell functions.     

Generation of cytotoxic T lymphocytes in vitro. VI. Effect of cell density on response in mixed leukocyte cultures.

Reexposure of day 14 murine mixed leukocyte culture (MLC) populations to the original irradiated allogeneic stimulating spleen cells has previously been found to result in the ratpid generation of cytolytic T lymphocytes (CTL) associated with a net increase in cultured cell number. Under the experimental conditions used, day 5 MLC cells appeared unable to respond to the allogeneic stimulus. In order to characterize further the development of the potential for anamnestic reactivity during the course of MLC, C57BL/6 spleen cells were incubated with irradiated (1000 rads) DBA/2 spleen cells (primary MLC) for up to 3 weeks. At various time intervals after the onset of the primary MLC, the surviving cells were collected and reexposed, at varying cell concentrations, to irradiated DBA/2 spleen cells (secondary MLC). At daily intervals thereafter, CTL activity was assessed using a quantitative 51Cr-release assay system. A paradoxic effect of responding cell concentration on generation of CTL activity was observed; relatively greater increase in CTL activity was observed as the concentration of responding cells was decreased over a 100-fold range. This effect was more pronounced with responding cells reexposed to antigen after primary MLC for 20 days, but was observed even with normal cells. The apparent unresponsiveness of day 5 MLC cells to alloantigen restimulation could be overcome by simple dilution of responding cells. Cytotoxic activity at the time of restimulation with antigen seems to be a major factor determining the magnitude of the secondary response. Since intact cells bearing alloantigens are required for the generation of CTL in MLC, residual cytotoxic cells reduce the effective antigenic stimulus by destroying stimulating cells. This effect of concentration of responding cells on generation of CTL in MLC complicates interpretation of experiments investigating the role of "inhibitor" and "helper" cell in cell-mediated immune responses occurring in vitro. Under optimal conditions, the highest CTL activity and the largest increase in total cell number was observed 4 days after restimulation of day 10 MLC cells. On a per cell basis, the lytic activity was up to 4 times greater than that observed at the peak of a primary response, and the number of viable cells recovered was nearly 20 times higher than that at the onset. Such secondary MLC are thus a convenient source of lymphoid cells selected primarily on the basis of proliferation induced by alloantigens.     

Generation of cytotoxic lymphocytes in mixed lymphocyte reactions II. Importance of private and publicH-2 alloantigens on the expression of cytotoxicity

MLC were established to test for the generation of specific cytotoxic effector cells in CML. The target cell used to assay for CML in the five combinations tested was of a differentH-2 haplotype from the stimulating cell population. Cytotoxicity was observed against this target only when it shared private alloantigens (antigens that are specific for theH-2D andH-2K region of differentH-2 haplotypes) with the stimulating cell population. Very weak or no Cytotoxicity was found when such alloantigens were not shared, although cross-reactive publicH-2 specificities were. These findings indicate that T cells display a cytotoxic potential against privateH-2 antigens in a primary response in vitro and are not capable of responding to publicH-2 specificities to the same level.BSS balanced salt solution - CML cell-mediated lympholysis - GPC guinea pig complement - ¹²⁵IUdR ¹²⁵I-iodo-deoxyuridine - MLC mixed lymphocyte culture - SE standard error     

T cell-mediated cytotoxicity against trinitrophenyl-modified cells: effect of glutaraldehyde treatment on the immunogenicity and antigenicity of trinitrophenyl-modified cells.

Cells treated with low concentrations of glutaraldehyde for a 10-sec interval were unable to incorporate 3H-leucine into TCA precipitable protein, respond to H-2 allogeneic cells in mixed lymphocyte reactions (MLR) and cell-mediated lympholysis (CML) assays, or display capping of cell surface immunoglobulin (Ig) with a fluoresceinated anti-Ig reagent. Such cells could stimulate and specifically block H-2 allogeneic CML activity but could not stimulate an H-2 allogeneic MLR response. Trinitrobenzenesulfonic acid (TNBS) treated spleen cells were used to sensitize syngeneic splenocytes into displaying a cytotoxic effect against trinitrophenyl (TNP)-modified target cells. Treatment of the stimulator cells with glutaraldehyde immediately after modification with TNBS did not impair their immunogenic activity. Similar treatment of TNP-modified concanavalin A-stimulated lymphoblasts that were used as inhibitors in a CML cold target competition assay allowed such cells to retain their antigenicity. Cells treated with glutaraldehyde before TNP-modification, however, were not antigenic in the cold target competition assay. These data are compatible with TNBS acting on plasma membrane molecules directly to cause cells to be antigenic and immunogenic in the CML assay rather than affecting internal cellular components.     

Cellular and genetic restrictions in the immunoregulatory activity of alpha-fetoprotein. III. Role of the MLC-stimulating cell population in alpha-fetoprotein-induced suppression of T cell-mediated cytotoxicity

Alpha-fetoprotein (AFP), a normal component of fetal and newborn sera, exerts selective suppressive effects on various functions of thymus-derived (T) lymphocytes, including the T cell-mediated cytotoxic reaction. In the present study, AFP-induced suppression of the T cell-mediated allogeneic response was examined to define more precisely the mechanism by which AFP acts. Data presented indicate that AFP elicits its effect within the first 25 to 36 hr of culture. However, AFP apparently has no direct effect on T lymphocytes. T cells incubated in the presence of AFP retain full capacity to respond in MLC and CML. In contrast, AFP induces major changes in the functional status of monocyte-enriched, MLC-stimulating cell population. This monocyte-enriched population, after pretreatment with AFP, possesses strong stimulating capacity for the T suppressor cell limb of the immune response, while at the same time, suppresses the activity leading to the normal generation of cytotoxic T lymphocytes. These data correlate with the immune activity present in the newborn mouse.     

Expression of both I-A and I-E/C subregion antigens on accessory cells required for in vitro generation of cytotoxic T lymphocytes against alloantigens or TNBS-modified syngeneic cells

We have previously reported that radioresistant, Thy 1-negative accessory cells (SAC) are required for the in vitro generation of cytotoxic T-effector cells to allogeneic or trinitrophenyl-modified syngeneic cells. These SAC were found to provide accessory functions irrespective of whether they were syngeneic, semi-syngeneic, or allogeneic to the responding cells. To further characterize the accessory cells active in CML, the expression of Ia antigens on this functional population was assessed by pretreated SAC with anti-Ia reagents and complement and by testing the accessory cell function of these treated populations. The results of these studies demonstrated that the relevant accessory cells for allogeneic and TNP-self CTL express Ia determinants encoded by genes mapping in the I-A and I-E/C subregions. For the TNP-self CTL the accessory function of both SAC syngeneic or allogeneic to the responding and stimulating cells was specifically abrogated by treatment with anti-Ia reagents and complement.     

Properties of reticulum cell sarcomas in SJL/J mice. III. Promotion of tumor growth in irradiated mice by normal lymphoid cells.

Experiments were performed to elucidate the mechanisms by which lymphocytes obtained from an M-antigen-incompatible strain reduce the specific mixed lymphocyte culture (MLC) response of lymphoid cell populations after injection into allogeneic recipients. Mice of strain CBA were injected with spleen cells from hybrids of the H-2-compatible, M-antigen-incompatible strain C3H. Normal C3H × CBA spleen cells increased the MLC reactivity of the host's lymphocytes during the first 1–3 days, and thereafter the response against C3H was drastically reduced. Mitomycin-treated or antibody-coated C3H × CBA cells rather enhanced the MLC responsiveness. Roughly similar results were obtained by injecting untreated H-2-incompatible C3H hybrid lymphocytes. Lymph node or spleen cell populations from CBA mice, injected with C3H × CBA cells up to 2 weeks earlier, were found to depress the MLC reactivity against C3H when transferred to new CBA hosts. The results indicate that injected cells had survived for 2 weeks in the host. On the other hand, H-2-incompatible C3H hybrid cells could not be detected even at day 3 after injection into CBA mice. The results also indicate that C3H hybrid lymphocytes have to be functionally intact and able to survive in the host for a relatively long period of time to be able to reduce the specific MLC response of the host's lymphocytes.     

Cellular requirements for the generation of primary cell-mediated lympholysis responses to Qa-1 antigens

Mixed leukocyte cultures (MLC) between NZB responder spleen cells and Qa-1-disparate stimulator spleen cells were employed to determine the cellular requirements for the generation of primary anti-Qa-1 cell-mediated lympholysis (CML) responses. Although primary anti-Qa-1 cytotoxic lymphocytes (CTL) were generated during H-2-homologous stimulation, anti-Qa-1 CTL were not detectable from MLC in which the stimulators were H-2 allogeneic. Anti-Qa-1 CTL also were not generated from MLC in which the stimulators were semiallogeneic. Thus, H-2 identity between responder and stimulator cells was not sufficient to permit the generation of primary anti-Qa-1 CTL when H-2 disparity was also present. The capacity for H-2 disparity to prevent anti-Qa-1 CML responses was further demonstrated in MLC containing both H-2-allogeneic and H-2-homologous stimulator cells. Therefore, in subsequent studies we employed NZB responders and H-2-homologous, Qa-1-disparate stimulators. When various subpopulations of stimulator cells were studied for their ability to induce anti-Qa-1 CTL, nylon wool-adherent cells were found to be potent stimulators, but nylon wool-nonadherent cells were not. Furthermore, depletion of macrophages from the stimulator population abrogated the generation of anti-Qa-1 CML responses, despite the presence of responder macrophages in the culture. In contrast, all fractionated subpopulations stimulated anti-H-2 CML responses. When macrophage-enriched cells were used as stimulators, anti-Qa-1 CTL could be generated with approximately 80-fold fewer stimulator cells than when unfractionated splenocytes were used as stimulators. These findings indicated that stimulator macrophages were essential for the generation of primary anti-Qa-1 CTL. Direct evidence for macrophage expression of Qa-1-antigens was obtained by using a Qa-1b-specific CTL clone. These studies provide i) the first evidence for Qa-1 expression on macrophages, ii) a basis for comparison of the cellular interactions necessary to generate CTL against H-2K/D-encoded vs Qa-1-encoded class 1 antigens, and iii) a model for investigating the mechanisms responsible for the immunodominance of H-2K/D alloantigens.     

Recognition of human minor alloantigen(s) by cytotoxic lymphocytes in vitro

Lymphocytes from an extensively transfused patient with aplastic anemia were induced to cytotoxicity against target cells from several HLA-matched siblings by in vitro stimulation with allogeneic cells. Effective stimulating cells shared HLA-B7 with the patient, but not all B7 individuals were effective. An additional factor, which was found to segregate in both the patient's and an unrelated sibship, was also necessary. Segregation of this minor alloantigen, W, was also revealed among the patient's HLA-matched sibs by differential susceptibility to lysis by effectors from the patient. The ratio of six positive to four negative siblings suggests that the antigen difference might be coded by a single locus. Lymphocytes from a normal sib, who like the patient is lacking the minor antigen, could not be induced to cytotoxicity against positive targets. Thus in vivo sensitization of the donor of the responding cells appears to be necessary for the demonstration of the cytotoxic response to the minor antigen in vitro. No correlation was observed between the segregation pattern of W and of known blood group antigens, and no cytotoxic antibody to W was detected in the patient's serum in several trials.Abbreviations used in this paper MHC major histocompatibility complex - Tc thymus dependent cells capable of mediating cytotoxicity in the absence of Immoral antibody - GVHD graft versus host disease - CML cell mediated lympholysis - MLC unidirectional mixed lymphocyte culture - ADCC antibody-dependent cell mediated cytotoxicity     

Effect of corticosteroids on human in vitro anamnestic response.

When human peripheral blood lymphocytes were cultured in the presence of the antigens streptodornase (SKSD) and allogeneic lymphocytes (MLC), increased rates of DNA synthesis were observed 4 to 6 days laer. If these cell cultures were continued to 12 days, the rates of DNA synthesis returned to control levels. On rechallenging these cells with the same antigens, an accelerated response was observed. The corticosteroid preparations hydrocortisone and methylprednisolone suppressed both of these responses and also impaired the ability of the 12-day cultures to generate cells responsible for the accelerated secondary response. Although cells cultured for 12 days in the presence of SKSD and hydrocortisone responded poorly to SKSD rechallenge, they responded well in MLC. Hence, the suppressive effect was not completely unselective.     

Isogeneic and allogeneic lymphocyte interactions may be controlled by cell surface immunoglobulin tropism.

A study of allogeneic MLC (mixed lymphocyte culture) and two types of isogeneic MLC has been conducted in which subsets of B-cells were serologically removed from pools prior to using the remainder as stimulator cells. Cellular division in the two types of ILC (isogeneic lymphocyte culture) was found to be triggered by lymphocytes with IgG₁ on their surfaces. In contrast, the stimulator cells in ALC (allogeneic lymphocyte culture) possessed membrane-bound IgG_2A and/or possibly IgG_2B. Splenic T-cells were incapable of stimulating replication of splenic or lymph nodal T-cells in the absence of B-cells. Splenic T-cell preparations served as weak stimulators of other allogeneic T-cells but only when B-cells, either isogeneic or allogeneic to the stimulator T-cells, were present. We propose that stimulation in the MLC occurs in two distinct steps. First, immunoglobulins on cell surfaces may function to bring appropriate subsets of cells together. Next, antigenic recognition occurs that results in blastogenesis. Furthermore, the tropism or attraction that certain immunoglobulins have for some cell types may determine which subsets of cells participate in allogeneic and which take part in isogeneic MLC.     

Genetics of cell-mediated lympholysis in man.

Cell-mediated lympholysis (CML) was studied in a family containing two siblings in who genetic recombinaiton had occurred in the HLA comples. In one sibling, recombination occurred between the HLA-A locus and the HLA-B locus. In the second sibling recombination occurred between the HLA-B locus and the HLA-D locus. Strong CML activity was generated in mixed lymphocyte cultures (MLC) when stimulator and responder cells differed in HLA-A, B, and D antigens. MLC involving HLA-D differences alone did not generate CML. Weak, but definite CML activity was generated during MLC with cells differing at HLA-A and HLA-B but sharing HLA-D. HLA-B antigens were good targets for lysis in all combinations studied. HLA-A antigens were poor targets in some but not in all combinations. However, combinations where HLA-A antigens seemed to be good targets could have involved HLA-B differences due to polymorphism of HLA-B7 antigens each inherited from a different parent. HLA-D antigens did not serve as targets for lysis. In three cell experiments, excellent CML activity was generated when responder cells were stimulated by HLA-D antigens and by HLA-A and B antigens present on separate stimulator cells.     

Requirements for the stimulation of allogeneic T lymphocytes by acute non-lymphoblastic leukaemia cells

Summary Blast cell populations from 32 patients with acute non-lymphoblastic leukaemia of various morphological types have been examined for their ability to stimulate allogeneic T lymphocytes from normal donors in one-way mixed leucocyte culture (MLC). At the same time, these leukaemic cell populations were examined for the amounts of major histocompatibility complex Class I and Class II antigens they expressed, and their ability to release interleukin 1 (IL1) in culture both with and without stimulation by lipopolysaccharide. The abilities of the leukaemic cell populations to stimulate in MLC, and to produce IL1, were found to be associated with the expression of morphological characteristics of monocytic differentiation, and correlated significantly. In contrast, no correlation was observed between the extent of Class I or Class II expression and MLC stimulatory ability. Many myeloblast populations of immature phenotype were unable to stimulate allogeneic T cells despite their strong expression of these antigens. This lack of stimulatory ability was not overcome by the addition of exogenous IL1. We therefore conclude that the correlation between the production of IL1 and MLC stimulatory ability does not necessarily imply a cause/effect relationship, and that the interaction between allo-antigen and the T cell receptor together with a supply of lymphokine co-stimulator is not sufficient to activate resting T lymphocytes. The failure of some Class I and II antigen positive leukaemic blasts to stimulate in MLC even in the presence of exogenous IL1 may be due to the lack of other differentiation-associated cell surface molecules necessary for stable cell-cell interaction.     

Effect of ribonucleases on cell-mediated lympholysis reaction and on GM-CFC colonies in bone marrow culture

Natural dimer of bovine seminal ribonuclease (AS RNase) suppressed markedly DNA synthesis in allogeneic mixed lymphocyte culture (MLC) of normal human lymphocytes and simultaneously inhibited induction of cytotoxic effector cells within the sensitization phase of indirect cell-mediated lympholysis (CML) reaction. The last purification step of the AS RNase isolation procedure did not increase the suppressive activity of AS RNase compared to a less purified preparation (ZS RNase), thus, the later preparation was mostly used. ZS RNase (10 micrograms/ml) caused 50% inhibition of MLC reaction whereas pancreatic ribonuclease (A RNase) was 10 times less effective. The suppressive effect of RNases added in the beginning of the sensitization phase of the CML reaction correlated with that observed in the MLC reaction. The concentrations of ZS RNase (10 micrograms/ml), A RNase (100 micrograms/ml), and additionally tested cyclosporin A (0.5 microgram/ml) resulted in nearly total abrogation of cytolysis in CML. ZS RNase added after the sensitization of effector cells did not influence their cytolytic action on target cells within the destruction phase of CML. Natural killer and killer cell activities in normal peripheral lymphocytes were not inhibited by ZS RNase at the concentration of 330 micrograms/ml. ZS RNase (20 micrograms/ml), cocultivated 1 h with normal human bone marrow cells and then washed off, enhanced formation of GM-CFC colonies in semisolid agar culture up to 200%. Simultaneously tested antilymphocyte globulin increased the number of GM-CFC colonies at the average of 128%. This stimulating effect on colony formation appeared also in bone marrow culture of patients suffering with various hematological disorders. The possibility of utilizing the preparations gained from seminal plasma in clinical bone marrow transplantation is discussed.