a-1: signals in b cell activation

Immunobiol., vol. 167, pp. 1-292 (1984)

16th International Leucocyte Culture Conference August 19-25, 1984 . Cambridge

Abstracts

TI antigens, LPS, PWM, PMA

Workshop A-1 Signals in B Cell Activation

University of Konstanz, Faculty of Biology, Department of Immunology, FRG

1. Permissiveness of B cell-defective mice towards congeneic immune cells

RITA BOSING-SCHNEIDER, G. LEHLE, and CH. SELINKA

There are high-responder and low-responder strains of mice with respect to the T independent antibody production against the a (1-3) glucosidic linkage of dextran B1355S (DEX). The genetic trait is linked to the constant portions of immunoglobulin heavy chains. Mice of allotype Igh' respond strongly to Dex with a crossreacting idiotype (ldX) shared by the myeloma proteins J558 and 104E, whereas animals with the genetic background of Ighb do not express these idiotypes and are low responders for dextran (1). When anti-Dex immune cells of the high-responder strain (Balb/c-Igh') are transferred into a congeneic nonresponder host (Balb-Ighb

) the grafted cells are suppressed and cannot display an idiotype-positive antidextran response. This congeneic barrier of thymus-independent immune cells is lost in mice carrying an X-linked B cell defect: when anti-Dex immune cells from responder F j (CBA/N X Balb/c) mice are transferred into congeneic nonresponder strains F j (CBA/N X Balb-Ighb),

female recipients are nonpermissive towards grafted cells whereas male Iittermates allow responder cells to develop an idiotype-positive anti-dextran response. These results show that the congeneic barrier towards Dex-immune cells is dependent on the maturation stage of B cells in the recipient. Since nude mice also suppress adoptively transferred immune cells (2) it is speculated that mature B cells display suppression by antiidiotypic reactivity.

1. BLOMBERG, B. et al. Science 177: 178-180 (1972). 2. WEILER, E. et al. In: Idiotypy in Biology and Medicine, KOHLER et al. (eds.), Academic

Press, 1984.

2 . 16th International Leucocyte Culture Conference, Cambridge

Inserm U 131, 32 rue des Carnets, 92140 Clamart, France, and Inserm U 93, 2 place du Docteur-Roux, 75475 Paris Cedex, France

2. T cells and interleukins requirement for the murine B cell response to a «T cell independent antigen»

H. DUCLOS, A. HAREL, M. C. MAILLOT, and D. FRADELIZI

The in vitro anti-trinitrophenyl (TNP) antibody response of murine spleen cells to the TNP-conjugate of polyacrylamide (Biogel P30) beads (TNP-PAA), an antigen previously designated as T independent antigen, requires macrophages and T cell help. Semi-purified III (a gift of A. MAIZEL) can replace macrophages, suggesting that they do not function as antigenpresenting cells for this particulate antigen, but III cannot substitute for T cells which appear directly involved in B cell activation. T helper cells are Lyt 1 + r T cells and are blocked by a monoclonal antibody recognizing a cell surface molecule involved in class II MHC antigens recognition (HI29-19.6, a gift of M. PIERRES). T cell-depleted response can be restored, in a dose-dependent fashion, by the addition in the first 48 hours of 5 days cultures of: a) Con Adepleted supernatant of activated murine spleen cells, b) IL2-enriched fractions of AcA 54 semi-purified EL4 and PHA-P-stimulated human lymphocyte supernatants, c) highly purified human natural IL2, and d) IL2 constituvely produced by MLA 144 monkey cell line. BCFG activity contaminating these preparations cannot account for their restorative ability as purified BCGF, constituvely produced by a murine T cell hybridoma T91 (a gift of H. J. GARCHON), is not sufficient to lead to B cell differenciation. Importantly, H 129-19.6 monoclonal antibody, present for the whole culture period, does not prevent the enhancing effect of T cell factors but prevents that of III on macrophage depleted response, suggesting that it may act on T cells involved in lymphokine production. Our results clearly establish a role for T cell-derived factors in the B cell differentiation process. The helper effect of EL4 and human IL2-enriched fractions is totally removed by absorption on an IL2-dependent cytolytic T cell line CTLL-2; however, we show that it is equally totally prevented by a potent suppressor factor contained in CTLL-2 supernatants. Thus, whether IL2 is the active component cannot be concluded. Experiments with pure recombinant IL2 are in progress to clarify this point.

National Institutes of Health, Bethesda, Maryland 20205, U.S.A.

3. T -independent activation of immature human B cells with TNPBrucella abortus

B. GOLDING, A. V. MUCHMORE, and R. M. BLAESE

Recently we showed that it is possible to activate human B cells from normal individuals (age> 3 years) to differentiate into anti-TNP plaque-forming cells (PFC) following primary in vitro stimulation with TNP-Brucella abortus (TNP-Ba). In this report we demonstrate that human newborn and Wiskott-Aldrich patient B cells are likewise capable of responding to this antigen. The cord blood cell response followed the same dose response and kinetics as that previously seen in adults. Furthermore, the newborn PFC response was also found to be: (i) TNP specific, as determined by lysis of TNP-sheep red blood cells (TNP-SRBC) but not unconjugated SRBC and by inhibitability with soluble TNP-lysine; and (ii) T-independent, since T cell removal by sequential E-rosetting did not abrogate the response. When compared to older individuals cord blood TNP-Ba stimulated cells generated a lower number of anti-

16th International Leucocyte Culture Conference, Cambridge· 3

TNP PFC. At the B cell level this may have been due to fewer TNP-Ba sensitive precursors andlor a reduced capacity of these percursors to proliferate. In order to test these possibilities cells were cultured at limiting dilution and precursor frequency and clone size analyses performed. These revealed that cord blood cells contained similar numbers of TNP-Ba sensitive precursors as cells from older individuals, approx. 1 per 106 cells. However, newborn culture wells, containing one or fewer precursors (by Poisson distribution) gave rise to clones not exceeding 32 anti-TNP PFC, whereas peripheral blood and tonsil cell precursors generated clones of 65-128 and 129-256, respectively. Therefore, the relatively lower cord blood cell responses were due, at least in part, to a diminished capability of newborn cells to proliferate and differentiate into anti-TNP PFC. The ability of TNP-Ba to activate neonatal and CBA/N X-linked mutant mice was used to classify this antigen as a T-independent type 1 antigen (TI-1), capable of triggering both immature and mature murine B cells. It was therefore of interest to determine whether B cells from patients with the Wiskott-Aldrich syndrome could respond to TNP-Ba. These patients have an X-linked immunodeficiency and resemble CBA/N mice in that their B cells fail to respond to polysaccharide antigens. Our findings show that indeed Wiskott-Aldrich patient B cells were mostly responsive (4/5) to TNP-Ba. Therefore, we conclude that human newborn and Wiskott-Aldrich patients do possess functionally competent B cells capable of responding to TNP-Ba which are possibly equivalent to the LyB5-subset defined in the mouse. Furthermore, these data indicate that TNP-Ba behaves as a TI-l antigen in the human and should be useful in the study of human B cell functional ontogeny.

Department of Immunology, Nagoya University School of Medicine, Nagoya 466, Japan

4. D-mannose-terminated glycoconjugates are recognized by lectin-like receptors of lymphocytes for T cell-independent type 2 antibody responses to Thy-t alloantigens

K.-I. ISOBE, F. NAGASE, and I. NAKASHIMA

Thy-l antigens are the major membrane glycoproteins of brain and thymocytes of mice and rats, but yet their functions are not clear. Brain and thymocyte Thy-l antigens share the same polypeptide that is composed of sequence-defined 111 amino acids and carries the allogeneic antigenic determinant. In addition, they contain different carbohydrates. The thymocyte Thy-1 antigen but not the brain Thy-l in uniquely immunogenic for T cell-independent type 2 (TI-2) B cell responses either in vivo or in vitro (IsoBE, K. et a!. 1983. Immuno!. Lett. 7: 111; ISOBE, K. et a!. J. Immuno!., in press). This exceptional immunogenicity might be related to the physiological activity of the thymocyte Thy-l mediating cell interactions. Recent experiments show that sugar moiety of membranes have important roles in immunological events due to natural killer cells, cytotoxic T cells and suppressor T cells, and accessory cells presenting Ia antigens for T cells. In the present study, we tested if the oligosaccharides on Thy-l molecules are involved in the strong immunogenicity. Addition of 10 or 100 mM D-mannose into the culture of murine spleen cells in vitro inhibited their primary anti-Thy-l alloantigen plaqueforming cell responses to the thymocyte antigen in over 90 %. Other monosaccharides including D-xylose, L-rhamnose, L-fucose, D-glucose, D-galactose, N-acetyl glucosamine, D-glucuronic acid and D-galacturonic acid did not display such a remarkable effect, although some of these sugars partially reduced the responses. Pretreatment of responder spleen cells with mannose before the stimulation with Thy-l antigen also prevented the responses to

occur. Further experiments have shown that the carbohydrate residues on the Thy-l antigen are crucially involved in its immunogenicity. The treatment with sodium periodate or umannosidase greatly diminished the Thy-l immunogenicity. In contrast, the serological Thy-l antigenicity was not much reduced by these treatments. Our data suggest that D-mannoseterminated oligosaccharides on the thymocyte membranes are recognized by lectin-like


receptors of B lymphocytes for anti-Thy-1 TI-2 responses. It might therefore be speculated that different carbohydrates of brain and thymocyte Thy-1 determine their tissue-specific activities.

Univ. at Buffalo, SUNY, Buffalo, NY 14214, U.S.A.

5. Selective binding of bacterial lipopolysaccharide to subpopulations of murine lymphocytes

DIANE M. JACOBS, J. H. ELDRIDGE, and F. SWARTZWELDER

Direct evidence that B cells possess surface receptors for mitogens, particularly bacterial lipopolysaccharide (LPS), has been sought in investigations designed to determine if this ligand binds preferentially to subpopulations of lymphocytes. Previous studies from a number of laboratories have given contradictory results and/or have been carried out under conditions in which binding and post-binding uptake were not distinguished. We have reexamined this question using a sensitive hapten-sandwich immunofluorescence technique to detect individual cells which bind when exposed to low concentrations of LPS in the cold in the presence of azide and determined the surface phenotype and distribution of LPS binding murine lymphocytes. Ficoll-hypaque passed spleen cell suspensions prepared from C3H/St and C3H/HeJ mice contained 40 % LPS+!!+ cells (B cells), 5 % LPS+Thy1.2+ cells (T cells) and 5 % LPS+ cells bearing neither marker. Peripheral lymph nodes of C3H1HeJ mice contained 11 % LPS+ B cells and 2 % LPS+ T cells, and mesenteric lymph nodes contained 15 % LPS+ B cells and 3 % LPS+ T cells. In contrast, the 4 % LPS+ cells in Peyer's patches were all B cells, and the thymus contained 4 % LPS+ T cells. Only the spleen contained LPS+ null cells. Within each lymphoid organ examined, the fraction of total lymphocytes identified as LPS +!! +, LPS +Ia + or LPS + /) + was similar, indicating that LPS + B cells possess the surface phenotype !!+Ia+/)+, characteristic of mature B cells. This conclusion was supported by the absence of LPS+!!+la+/)+ cells in newborn spleens. These data illustrate selective binding of LPS predominantly to mature B cells but also to small numbers of null cells and T cells, all populations containing cells which can be activated by LPS under appropriate conditions. Such binding was followed by capping of the ligand-binding site complex which is dependent on temperature, energy and a functional cytoskeleton. Surface !!, /) and la did not co-cap with LPS, indicating that the LPS binding site on B lymphocytes is distinct from these surface antigens.

Supported by NIH Research Grant AI 16915.

Unite d'lmmunobiologie, Institut Pasteur, Paris, France, Department of Immunology, Clinica Puerta de Hierro, Madrid, Spain, and Institut fur Mikrobiologie II, Universitat Tubingen, FRG

6. Functional recognition of B cell mitogens by reactive B cell blasts and B cell lymphomas : Membrane proteins in lipopolysaccharide reactive and nonreactive cells.

B. KLEINE, C. MARTINEZ, W. BESSLER, and A. COUTINHO

The B lymphoma line WEHr 279.1 was found to be growth inhibited by the bacterial mitogen lipopolysaccharide (LPS). When growing tumor cells in the presence of LPS variant lines and clones of mitogen resistant cells were derived. Some of these retained their sensitivity

16th International Leucocyte Culture Conference, Cambridge . 5

to two other B cell mitogens - Dextransulfate and Tripalmitoylpentapeptide, a synthetic analogue of the E.coJi lipoprotein mitogenic region. Thus, the loss of mitogen sensitivity in the nonreactive clones was selective for LPS, suggesting specific recognition of mitogenic substances on B cell membranes. The comparison of membrane proteins from LPS reactive (lps+) and nonreactive (Ips -) cells was initiating using 35S-methionine labelled membrane proteins of a) WEHI 279.1 lps+ - and lps--subclones, b) C57BLl6 (lps+) and C57BI/I0.sc.Cr (lps-) B cell blasts, and c) C3H/Tif (lps+) and C3H/HeJ (lps-) B cell blasts. These were analyzed in twodimensional gels, but the different protein patterns in these gels do not allow direct deductions of single proteins responsible for LPS reactivity. The putative differences will have to be revealed by monoclonal antibodies to membrane antigens which will only recognize LPS reactive cells.

Departement d'Immunologie, Institut Pasteur, 75724 Paris Cedex 15, France

7. Generation of B-memory cells by TNP-Ficoll: Expression and not induction is impaired in various inbred mouse strains

A. LE MOAL, J.-H. COLLE, and P. TRUFFA-BACHI

The ability of various inbred mouse strains to mount a memory-type response against a class 2 thymus-independent (TI) antigen, TNP-Ficoll, was determined. Our studies show that no secondary anti-hapten response is induced by a challenge occurring 7 days after priming (the same optimal immunogenic dose, 25 !!g per mouse, is used for priming and challenge). These results are consistent with the generally claimed absence of memory expression to this class of antigen but contrast with our previous findings using a class 1 TI antigen, TNP-LPS, for which an anamnestic response was found to be controlled by IgH-V or closely linked genes. Using adoptive transfer experiments, we were unable to demonstrate any active role of T suppressive cells in the lack of secondary response to TNP-Ficoll. Moreover, when the TNP-Ficollprimed mice are challenged with 5 !!g of TNP-LPS, 9 out of 12 strains studied exhibit a characteristic memory-type anti-TNP response (enhanced number of antibody forming cells (AFC) and/or significant amount of IgG). This already demonstrates the ability of TNP-Ficoll to induce B memory cells. The absence of a memory-type response to this antigen is related to its failure to trigger into AFC the memory cells it has generated, since the influence of humoral factors such as anti-TNP or anti-idiotypic antibodies is still controversial. Our results can be taken as an evidence that, as already reported for thymus-dependent antigens, two distinct Bmemory cell subpopulations are involved in the responses to the 2 classes of TI antigens. A choice remains to be made between the two current hypotheses: the existence of two distinct precursors of B cell subsets versus a unique cellular lineage exhibiting different carrier sensitivity depending on its stage of development.

Dept. of Immunology, Umea University, Umea, Sweden, and Dept. of Immunobiol., Inst. Pasteur, Paris, France

8. Influence of neonatal Dextran B 512 (Dex) priming on adult Dexspecific antibody responses and B cell precursor frequencies

I. LUNDKVIST, F. IVARs, and A. COUTINHO

The in vivo immune response to the thymus independent antigen Dex is characterized by late ontogenic development. Using the mitogen LPS in limiting dilution assays, we have found,


however, that the absolute frequencies of Dex-specific B-cell precursors are constant throughout postnatal development. There are several possible explanations for the discrepancy between in vivo responses to Dex and the frequencies of Dex-specific B-cell precursors. A recent report from this laboratory (PETIERSSON et al. Eur. J. Immunol. 12: 653) shows dissociation between proliferation and maturation of neonatal B cells. Assuming that this is a general characteristic of neonatal B cells, one plausible explanation to our observations is that they are functionally competent to respond to Dex, but without subsequent maturation to plaque forming cells (PFC). Increased frequencies of responding cells (both in vivo and in limiting dilution assays), as a result of priming, would provide evidence not only for the functional competence of neonatal B-cells to respond to Dex by clonal expansion, but also for the connection between Dex-specific B cells reactive to Dex in vivo and LPS in vitro.

Dept. of Immuno-haematologie, University Hospital Utrecht, The Netherlands

9. Pokeweed mitogen induced plasma cell differentiation of human tonsil B cell subpopulations expressing multiple heavy chains on their surface

G. C. MUDDE, C. J. M. VERBERNE, and G. C. DE GAST

Human tonsil non T cells were separated into four different B cell subpopulations according to the expression of surface IgM (sIgM) and or sIgG. All fractions responded to PWM with plasma cell differentiation. The highest percentages of plasma cells were found in the sIgMfraction, whereas the lowest percentage was found in the sIgM+G-fraction. (20.8 ± 2.3 %) of which 84.1 ± 3.5 % was cytoplasmic IgM+ (cIgM). This fraction never produced cIgG+ plasma cells, indicating that under the cultured conditions used no in vitro class switch from [!~y occurred. The sIgM+G+ fraction produced 33.7 ± 5.2 % plasma cells. Both cIgM+ (47.5 ± 7.S %) and cIgG+ (35.7 ± 5.8) as well as cIgA + (16.8 ± 4.0 %) plasma cells were found. The appearance of cIgM+ plasma cells indicates that at least part of the sIgM+G+ tonsil non T cells are not yet commited to cIgG production after PWM stimulation. Therefore both the detected sIgM and sIgG molecules must be actively synthesized on the B cells in vivo. We cannot exclude the possibility of an in vitro switch from sIgM+G + A -~ cIgA +, but it is very likely to presume that the cIgA + plasma cells in this fraction have been derived from sIgM+G + A + B cells.

Basel Institute for Immunology, Basel, Switzerland

10. Evaluation of the mitogenic effect of phorbol myristate acetate with chromosomally marked lymphocytes

W. T. WEBER and O. LASSILA

Phorbol myristate acetate (PMA) is a recognized inducer of proliferation in a variety of cell types and it is used in cultures of established macrophage and EL-4 thymoma cell lines to produce several growth and maturation factors. While mitogenic effects of PMA on lymphocytes have been reported, it is not clear to which extent T and B cells proliferate upon exposure to PMA. Peripheral blood lymphocytes from chickens containing chromosomally marked T


and B cells were cultured with optimal concentrations of PMA (10--50 ng/ml) in serum-free medium. PMA induced a 10--25 fold increase in H3 thymidine incorporation at 48-72 hrs after culture initiation. Based on chromosome analysis at 48 hrs of culture, 69-81 % of the net PMA-induced proliferation was attributable to dividing T cells, with 19-31 % of the response involving B cells. At 72 hrs, 45 % of the response involved T cells, and 55 % involved B cells. As PMA is used in the production of supernatants containing several growth and maturation factors intended for lymphocytes, it is apparent that small contaminating amounts of PMA in supernatants might induce a proliferative response of both T and B cells.

Differentiation signals, anti-Ig, anti-la, Lsd, Ag

National Institute for Medical Research, Mill Hill, London

11. Anti-immunoglobulin antibodies induce phosphoinositide breakdown in B cells

K. BI]STERBOSCH and G. G. B. KLAUS

Mouse B cells can be polyclonally stimulated by a variety of agents, including lipopolysaccharide (LPS) and F(ab')2 anti-immunoglobulin antibodies (anti-Ig). The immunosuppressive peptide cyclosporine (CS) effectively inhibits activation by anti-Ig, but not by LPS (1). This suggests that these two ligands may stimulate B cells via distinct pathways. The biochemical events leading to the activation of B cells by anti-Ig or LPS are still poorly understood. We now present data on the effects of these activators on phosphoinositide turnover. Phosphoinositides are inositol-containing phospholipids, whose breakdown to diacylglycerol and inositol phosphates has been implicated in receptor-mediated signal transduction in a wide variety of cell types (reviewed in 2). We found that anti-Ig has a considerable effect on phosphoinositide turnover in B cells; formation of inositol phosphates was detectable within a few minutes after addition of the ligand and persisted for several hours. In sharp contrast, LPS failed to induce formation of inositol phosphates. Monomeric Fab fragments of anti-Ig, which do not activate B cells, also failed to elicit a response, suggesting that cross-linking of surface immunoglobulins (sIg) is crucial. Anti-Ig induced breakdown of phosphoinositides is dosedependent, reaching a plateau at about 10 !lg/ml. This correlates quite well with the doseresponse relationship of other parameters of activation, such as RNA synthesis. Analysis of the levels of individual phosphoinositides and inositol phosphates suggested that stimulation by anti-Ig leads to breakdown of polyphosphoinositides rather than phosphatidylinositol. CS had no significant impact on anti-Ig induced phosphoinositide breakdown. This indicates that if phosphoinositide breakdown is involved in anti-Ig-induced B cell stimulation, then CS interferes with a later event in the activation cascade. Taken together, our results suggest that sIg belongs to the group of receptor systems which, when stimulated, induce enhanced phosphoinositide breakdown, whereas LPS receptors do not.

1. DONGWORTH, D. W., and KLAUS, G. G. B. (1982), Eur.]. Immunol. 12: 1018-1022. 2. FISHER, S. K., VAN ROOI]EN, L. A. A., and AGRANOFF, B. W. (1948), Trends Biochem. Sci.,

9: 53-56.


Duke University Medical Center, Durham, NC 27710, and University of North Carolina School of Medicine, Chapel Hill, NC 27514, U.S.A.

12. Immunoglobulin and Ia are signalling molecules on B cell membranes

R. B. CORLEY, M. OVNIC, N. J. LOCASCIO, L. W. ARNOLD, and G. HAUGHTON

The requirements for helper T (Th) cell-dependent B cell activation differ depending on the activation state of the responding B cell. While previously activated B cells are stimulated to further proliferation and differentiation by antigen non-specific macrophage-derived and Th cell-derived cytokines, resting B cells require both specific antigen and major histocompatibility restricted Th cell interactions for activation. The question of whether the B cell surface immunoglobulin, B cell Ia molecules, or both, are receptors involved in initiating B cell entry into cell cycle following binding by the appropriate ligands (specific antigen and T cell receptors, respectively) has been controversial and is the subject of this study. Two sources of B cells have been used, one a purified population of hapten-binding B cells, the other a B cell lymphoma, CHI2, with known antigen specificity. CH12 has the activation properties of a resting B cell. The requirement for antigen in B cell activation was studied using Th cells selected such that they interacted with B cells in the absence of sIg-antigen binding. B cells were not activated when normal ratios of Th cells and B cells were used unless the B cell antigen was also present in culture, demonstrating that sIg-ligand interactions contribute to B cell activation. The requirement for sIg-ligand binding could be overcome using higher multiplicities of T cell help, but Th cell recognition of B cell Ia was still essential. Evidence that Th cell-Ia interactions resulted in the transmission of an activation signal was provided from the study of CHI2. Although CH12 expresses both I-A and I-E molecules, only Th cells interacting with the I-E molecule were capable of activating CHI2. Recognition of the a chain of the I-E molecule is sufficient to activate CH12 cells. Interestingly, I-A-restricted Th cells were activated when T cell antigen was presented by the CH12 lymphoma. Thus, the ability of an Ia molecule to serve as a restricting element for T cell activation is separable from its ability to serve as a receptor for B cell activation. Our data suggest not only that Ia molecules serve as receptors for Th cells for B cell activation, but also that B cells may limit Th cell-B cell activation. An analysis of this question using normal B cells is currently under way. We conclude that both sIg and Ia molecules are receptors that participate in the activation of resting B cells.

Supported by USPHS grants CA36302, CA36642, CA23770, and CA29964 from the National Cancer Institute.

National Institutes of Health, Bethesda, M.D., U.S.A.

13. Activation of B cells by auto reactive T cells

A. FINNEGAN and R. J. HODES

The existence of T cells that recognize self Ia encoded determinants in heterogeneous populations has been extensively studied. In this report a number of autoreactive T cell clones have been isolated from antigen-primed spleen and lymph node cell populations. These


auto reactive T cell clones were found to directly recognize I-A or 1-E encoded molecules in the apparent absence of polymorphic non-major histocompatibility (MHC) products or exogenous foreign antigen. These studies showed that the responses of autoreactive T cell clones are resistant to chloroquine mediated inhibition of antigen processing by syngeneic stimulating cells and are in this respect indistinguishable from the responses of alloreactive T cells. Studies were also performed to characterize the relationship of autoreactive T cells to antigen-specific MHC-restricted T cells in their capacity to act as helper cells to induce B cells to synthesize immunoglobulin (Ig). It was found that autoreactive T cells function by mechanisms that resemble antigen-specific T helper (TH) cell activation of B cells through distinct MHCrestricted and MHC-unrestricted pathways. Autoreactive T cells after stimulation with syngeneic spleen cells polyclonally activated primed or unprimed B cells to synthesize IgM antibodies in the absence of specific antigen, although antigen-specific IgM antibodies were elicited in the presence of antigen. This pathway of B cell activation was MHC-unrestricted in that responses were induced in syngeneic as well as allogeneic B cells. The autoreactive T cells were also able to activate IgG responses by primed B cells through a distinct pathway that requires a) direct recognition by the autoreactive T cells of self MHC determinants expressed on the B cell surface and b) occupancy of the B cell Ig receptor with specific hapten. Thus, autoreactive T cells appear to function as helper cells in B cell activation through two distinct pathways that parallel those previously characterized for antigen-specific T H cells. If autoreactive T cells have the same potential in vivo as in vitro to activate even unprimed B cells in the absence of specific antigen, it would appear critical that mechanisms exist to regulate the activation and/or function of these T helper populations. Analysis of the regulatory mechanisms that function to control B cell activation by autoreactive T H cells is therefore of central importance to further evaluation of the physiologic role of autoreactive T cells. It has been found that B cell responses mediated by auto reactive T cells are subject to T suppressor mediated regulation which is similar to the antigen-nonspecific suppression of conventional T H dependent antibody responses.

Departement d'lmmunologie Institut Pasteur, 25, rue du Docteur Roux, 75015 Paris, France

14. Antigen-specific helper T cell clone supernatant is sufficient to induce both polyclonal proliferation and differentiation of small resting B lymphocytes

L. LECLERCQ, G. BISMUTH, and J. THEZE

Gradient-purified resting B lymphocytes can be polyclonally stimulated by antigen-specific major histocompatibility complex (MHC)-restricted helper T lymphocytes as well as by antigen-activated helper T cell supernatant. In contrast to what has been described so far, we show that helper T cell supernatant (in the absence of any other added stimulus such as that provided by anti-immunoglobulin antibodies) is sufficient to induce both proliferation of small resting B cells and their differentiation into immunoglobulin secreting cells. The stimulation induced by the helper T cell supernatant takes place in serum-free medium and is not MHC-restricted. Our findings strongly support the existence of a B cell activating factor acting on the resting B cell and causing it to enter the G, phase of the cell cycle in an MHCunrestricted manner. In addition, in the presence of helper T cell supernatant, unprimed B cells switched to IgG production with a typical thymodependent pattern.


University of North Carolina at Chapel Hill, N.C. 27514, and Duke University, Durham, N.C. 27710, U.S.A.

15. Monoclonal antibody against I-E antigens can substitute for T helper cells in the differentiation of a B cell lymphoma

N. J. LOCASCIO, R. B. CORLEY, L. W. ARNOLD, and G. HAUGHTON

CH12 is a B cell lymphoma which arose in a BI0.H-2'H-4bp/Wts mouse which had been adoptively hyperimmunized with sheep erythrocytes (SRBC). The cells of this tumor bear surface IgM specifically reactive with an epitope on SRBC, class I and class II antigens appropriate to the H-2' haplotype and Lyt-l, but lack Fe and C3 receptors. Differentiation, resulting in the secretion of haemolytic antibody, can be induced in culture by exposure to LPS or to antigen-specific, H-2 restricted T helper (Th) cells plus antigen, but not by a variety of supernatants containing combinations of non-specific, T-cell derived helper factors. With respect to these requirements for induction of differentiation, CH12 more resembles normal resting B cells than it does excited or activated B cells. Successful interaction with Th cells, leading to differentiation of CHI2, requires that the Th cells recognize antigen in the context of I-E; interaction with I-A, although resulting in activation and subsequent secretion of IL2 by the T cells, does not induce differentiation of CHI2. To test whether the differential effect of I-A and I-E restricted Th cells is due to a difference in the signals transmitted by the T cells or to a difference in the reception of those signals by the B cell, we have investigated the possibility that monoclonal antibodies might substitute for Th cells in the antigen-dependent, induced differentiation of CHI2. Three different antibodies reactive with I-Ek, at concentrations down to 10 ng/ml, in the presence of SRBC but not rabbit erythrocytes, induced rapid differentiation of CHI2, whereas 2 antibodies reactive with I_Ak, at concentrations up to 5 !-Igi ml, did not. these data demonstrate that ligand binding to sIg and I-E but not I-A molecules, on the surface of CHI2, result in transmembrane signalling which leads to induced differentiation and secretion of haemolytic antibody. They further suggest that the ability of Th cells to induce differentiation of these B cells, may not require any metabolic activity on the part of the T cells.

Supported by USPHS grants CA36302, CA36642, CA23770 and CA29964 from the NCr.

Dept. of Immuno-haematology, University Hospital Utrecht, The Netherlands

16. Plasma cell differentiation of normal human tonsil B lymphocytes coexpressing surface IgG (sIgG) and sIgA

G. C. MUDDE, E. C. HEESBEEN, C. J. M. VERBERNE, and G. C. DE GAST

Using the rosette technique with goat anti human heavy-chain antibodies coated ox red blood cells, we were able to identify and isolate B cells from human tonsils which coexpress sIgG and sIgA. These cells contradictory to most previously published B cell differentiation schemes, are a result of normal B lymphocyte differentiation in human tonsils and can very well be stimulated in vitro by PWM resulting in 43.2 ± 5.5 % plasma cells. Most of these

16th International Leucocyte Culture Conference, Cambridge . 11

plasma cells are cytoplasmic IgG+ (cIgG+) (49.1 + 5.3 %) but a fair amount of cIgA + (28.1 ± 7.9 %) and even cIgM+ (22.7 ± 4.4 %) cells are found. The expression of sIgG and sIgA before stimulation of the B cells was not due to passively bound sIg via Fc receptors of the B cells, therefore both surface molecules must have been actively synthesized by the B cells in vivo. The appearance of cIgM+ plasma cells in this fraction after PWM stimulation indicates the existance of sIgM+G + A + B cells in human tonsils. Since these cells do not fit in the accepted models of B cell differentiation these models have to be changed with at least one pathway comprising a switch from IgM via IgG to IgA.

Depts. of Immunology lUniversity of Miinster, FRG, and 2London Hospital, Medical College, London, England

17. Further characterization of the Lsd-alloantigen

H. PRAHL!, H. FESTENSTEIN2, E. KOLSCH 1

The cellbound Lsd (lymphocyte stimulating determinant)-antigen is capable to stimulate proliferation of allogeneic T lymphocytes and thus can be characterized through the use of alloreactive Lsd-specific T cell clones (1). We have further characterized the Lsd-antigen by isolating BI0.BR (Lsd-) anti C3H/Tif (Lsd+) T cell clones (SC3D, SC6D, 6D8), which show the Lsd-specific reactivity pattern. In addition to stimulation by C3H/Tif spleen cells they can for example be stimulated by AKR, CBA, DBAI2 (H-2d, MIs') but not by BALB/c, C3H/ HeHan, C3H/HeJ (these C3H strains are H-2k and Mls'-identical to C3H/Tif), C57BLl6 and DBA/l (H-2q, MIs') spleen cells. (BALB/c x DBA/I) Fl (H_2 d/q

, Mlsbh) and BI0.D2 x

DBA/I) Fl (H-2 d/q, Mlsb/,) spleen cells also lack stimulatory capacity. These findings exclude

the possibility that the proliferation of the presumed Lsd specific T cell clones on MIs' bearing stimulator cells is caused through a synergistic action of MIs' in conjunction with H-2d determinants (DBA/2) and that MIs' fails to stimulate those clones when associated with H-2q

(DBA/I) on stimulator cells. The findings further support the concept of two distinct alloantigenic determinants (Lsd and MIs) being expressed by chromosome 1. The Lsd-antigen is expressed on B but not on T lymphocytes. Some evidence suggests that it is a cell-interaction structure: B lymphocytes of Lsd + strains acting as cellular antigens for T cell proliferation in turn are induced to proliferate upon contact with cloned BI0.BR anti C3H/Tif T lymphocytes (clone SC6D). Third party B lymphocytes do not proliferate suggesting that B cell activation through the Lsd+ structure requires cell-cell interaction. It is possible that the T cell clones described are representatives of a certain helper cell type found among lymphocytes in primary mixed lymphocyte culture (2). For an analysis of the relationship between MIs and Lsddeterminants we have tested the proliferation of the clone 6D8 on various stimulator cells including the Mls-congenic strain BALB.D2-Mls'. This latter strain can stimulate T cell clone 6D8 but reproducibly only half as good as do DBAI2 spleen cells. This findings requires further analysis of a possible synergism between Lsd and MIs (3) determinants. Experiments are also in progress to produce Lsd specific monoclonal antibodies for functional and biochemical analysis of Lsd-structures.

1. OPALKA, B., and KOLSCH, E. (1983): Eur. J. Immuno!. 13: 24. 2. POBOR, G., PETTERSON,S., BANDEIRA, A., MARTINEZ-A, c., and COUTINHO, A. (1983):

Scand. J. Immuno!. 18: 207. 3. FESTENSTEIN, H., and DEMANT, P. (1973): Transp!' Proc. 4: 1321.

Supported by the Deutsche Forschungsgemeinschaft through SFB 104.


Institut fiir Virologie und Immunbiologie der Universitat Wiirzburg, Versbacher Str. 7, D-8700 Wiirzburg, FRG

18. Inhibiton of LPS-induced Ig secretion by F(ab'h anti 1-1-: selective effect on I-I-s-mRNA

A. SCHIMPL, u. CHEN-BETIECKEN, and E. WECKER

F(ab'h fragments of all xeno-antibodies to mouse Ig are potent B cell mitogens. Such mitogenic antibody fragments, however, still fail to give rise to Ig production/secretion and even inhibit Ig secretion in B cells costimulated with LPS. The inhibitory effect could be explained at the basis of a) reduction in the rate of transcription of the mRNA's for heavy and/ or light chains or their rate of translation, b) a greatly increased turnover of mRNA's or c) rapid degradation of protein chains synthesized or d) secretion. To investigate the site of inhibitory action we studied mRNA's found in LPS stimulated B cells which were additionally treated with goat F(ab')2 anti mouse fl. The experiments show that, at 48 hrs, mRNA levels for fl, are very low in LPS + GF(ab'h anti fl treated cells, while those for !1m, kappa and H-2 class I are as high as in LPS only treated cells. There is no increase in mRNA for gamma. At later times (96 hrs), specific mRNA for kappa also starts to be affected, being lower than in LPS controls, while H-2 class I mRNA remains at control levels. These and additional results suggest that anti-IG-receptor complexes mediate a negative signal exerted at the level of specific corresponding mRNA transcription and/or processing and that at least initially, fl, and kappa mRNA levels are regulated separately.

Institut J. Monod, Universite Paris 7, France

19. Selective activation of murine resting B cell subsets by la-restricted antigen induced T helper cell factors.

A. DE TALANCE, V. ZILBERFARB, and M. SEMAN

T cell clones or lines from antigen primed murine peripheral lymph nodes can stimulate virgin B cells to proliferate in the presence of antigen. This polyclonal by-stander activation has been shown to be restricted by Class II histocompatibility antigens. In several but not all, lines of T helper cells, this la-restricted effect can be replaced by the supernatant of the T cell cultures. Antigen and Ia are required for factor production. Fractionation of the supernatants on ACA 54 columns indicates that the factor produced by GAT or KLH-specific cells is about 45-55 Kd M.W. This molecule produced by various BALB/c T cell lines activates syngeneic H-2d lymphocytes but not B cells from BALB.B (H-2b) or BALB.K (H_2k) allogeneic mice. The factor does not act on LPS activated blasts. Consistently it preferentially stimulates the resting B cell population from high density Percoll fractions. Macrophages can be removed from the splenic B cell population without affecting the stimulation index. Yet, proliferation requires the presence of 2-beta mercaptoethanol in the culture medium. Although antigen is required for factor production, ACA 54 purified factor seems not to be dependent on antigen to promote B cell growth. Therefore, the results suggest that this factor is an autoreactive molecule. Whether it is produced by contaminating auto reactive T cells or is normally synthetized by antigen specific helper cells is under investigation. T cell supernatants poorly stimulate B cell differentiation and Ig secretion. This low polyclonal Ig production might be due to BCGF and BCDF molecules which also seem to be present in the supernatants. Yet, both Ia restriction and MW distinguish our factor from either of these two B cell activating


lymphokines. Preliminary experiments indicate that our factor may resemble AEF on both biological and biochemical criteria.

Finally, the factor does not stimulate lymph node or bone marrow B cells suggesting that only a particular B cell subset present in the spleen of adults or neonates is susceptible to it. Experiments designed to identify this B cell subset will be presented.

IDept. of Immunology, University Children's Hospital, «Het Wilhelmina Kinderziekenhuis», Utrecht, The Netherlands, and 2Dept. of Clinical Immunology, University Hospital, Utrecht, The Netherlands

20. Cell biological aspects of the activation of human peripheral B cells with the «primary» antigen ovalbumin

B. A. T HARTt, C. J. HEI]NENl, J. ZI]LSTRAt, R. E. BALLIEUX2

Human peripheral B lymphocytes, when cultured for 6 days with ovalbumin (OA) in the presence of T cell help, differentiate into plaque-forming cells (PFC) secreting small amounts of antigen-specific IgMl. The PFC which are found in this in vitro system lack the classical characteristics of high rate antibody secreting (plasma) blasts.

In the present study it is demonstrated that the activation of resting B cells, expressing membrane-bound IgM and IgD, with OA results in the loss of IgD, in a decrease of the cell density and in the expression of a receptor for a differentiation factor produced by T cells. Similar phenomena have been observed in studies using B cells from murine origin which were stimulated with anti-immunoglobulin2 or LPS2.3• Our present findings imply that the regulation of the PFC response by T cells4 is operative in the initial phase of the B cell response to antigen in vitro. Furthermore it implies that the PFC, thus induced is not an end-stage cell. It is likely that the activated B-cell is arrested in an «excited» state, because the OA-activated T cells do not secrete any or sufficient amount(s) of the appropriate differentiation factor(s).

1. BALLIEUX, R. E. et al. Immunol. Rev. 45: 3, 1979. 2. HOWARD, M., and PAUL, W. 1., Ann. Rev. Immunol., 1: 307, 1980. 3. ANDERSSON, J., et al. P. N. A. S. USA 77: 612, 1980. 4. BALLIEUX, R. E., et al. Immunol. Rev. 74: 5, 1983.

EBV transformation

lU.K. Transplant Service, Southmead, Bristol BSIO 5ND, and 2PHLS, Centre for Applied Microbiology and Research, Porton Down, Salisbury Wilts. SP4 OJG

21. In vitro development and characterization of human monoclonal anti-rhesus D antibodies produced by stable, cloned B-lymphoblastoid cell-lines (BLCL)

A polyclonal antibody-producing BLCL was established from an immunized female donor, by transformation in vitro with Epstein-Barr virus (EBV). The BLCL underwent repeated selection by a rosetting procedure and was subsequently cloned by the limiting dilutions


technique. Multiple, stable, antibody-producing clones were established, using a technique that involved the use of mouse peritoneal macrophage feeder-layers, arginine supplements and a cocktail of anti-mycoplasma agents. Formal evidence of monoclonality has been demonstrated by gel electrophoresis techniques, pre- and post-immunoabsorption. Most clones were of IgG1 isotype and secreted specific antibody at 15-20 f.lg/ml. Antibody-producing clones have shown long term stability in continuous culture for over 12 months and are easily converted to growth in Iscoves serum-free medium. The monoclonal cell-lines contained typical plasmacytomas with the possible anomaly that they expressed HLA-class II antigens on their surface. However, a constant proportion of cells within the presumed monoclonal lines appeared negative for cytoplasmic Immunoglobulin, suggestive of undifferentiated, or memory, B cells. The plasmacytomas stained positively for the EBV derived Nuclear Antigen (EBNA) and exhibited a normal human karyotype. Although the cell-lines originated from peripheral blood they appear to have differentiated, in vitro, into cells with surface characteristics usually associated with germinal centre B cells. This study exemplifies a successful protocol for the cloning of Human BLCL which produce high levels of human monoclonal antibodies without the necessity of hybridisation. Adaptation of these techniques to the production of human monoclonal antibodies against HLA antigens is underway.

Departments of Medicine and Laboratory Medicine, University of Alberta, Edmonton, Alberta, Canada

22. Spontaneous B lymphocyte transformation in patients with infectious mononucleosis

T. KOVITHAVONGS, M. O'BRIEN, J. SCHLAUT, M. K. DASGUPTA, and J. B. DOSSETOR

Infectious mononucleosis (LM.) is a self-limiting lymphoproliferative disease in which B lymphocytes infected with Epstein-Barr virus in vivo become transformed and can be maintained in culture indefinitely as B-lymphoblastoid cell lines (B-LCL). In order to explore the suitability of using these lines as target cells in transplant monitoring and for HLA antibody absorptions, attempts were made to grow B-LCL from blood of LM. patients, in this case, 20 university students seen at the Student Health Service on campus who presented with symptoms suggestive of LM. and who had a positive mononucleosis «spot» test. Peripheral blood mononuclear cells (PBMC) from ficoll hypaque density gradient separation, were further separated into T and non-T cells by SRBC rosetting technique. Non-T cells were kept in culture with RPM I 1640 with 10 % FCS and observed at weekly intervals for transformation. PBMC were studied for K and NK cell activities against alloantibody coated Raji (AbRaji) and K562 respectively. In a few instances, cytotoxic T cell (CTL) activity against uncoated Raji cells were also done. T helper to T cytotoxic/suppressor ratios (T.h:T.c/s) were determined by 51-Cr release with monoclonal antibodies 66.1 and 51.1 plus Low-Tox rabbit complement. At the time of the abstract, 13 of 17 I.M. patients' non-T cells in culture for longer than 3 weeks have grown into B-LCL and have been maintained for as long as 5 months. All 13 patients had T.h:T.c/s ratios of less than 0.6. In contrast, all 4 patients whose B cells failed to transform had ratios of 0.7 or greater (0.7, 1.7, 1.8, 1.9). NK activity was reduced to less than 30 % of normals whereas K cell activity against Ab-Raji was maintained or even enhanced and no CTL activity against Raji could be detected simultaneously. HLA-ABC and DR typing of B-LCL did not show preponderance of any HLA antigens in B-LCL developed. Most of the transformed cells in all B-LCL can be shown to have surface immunoglobulins and 6 of the lines secrete significant amounts of IgG in culture (1(}'-33 f.lg/mI!L, IgM and IgA not


tested). Thus, B-LCL can be cultivated from blood of I.M. patients with relative ease at the time of acute infection as reflected by a low T.h:T.c/s ratio. Susceptibility of these lines to lysis by CTL, as well as complement dependent or cell dependent antibody mediated lysis are currently under study.

Central Laboratory Netherlands Red Cross Blood Transf. Service and the Lab. of Exp. and C1in. Immunology of the Univ. of Amsterdam, Amsterdam, The Netherlands

23. Establishment of continuous human antibody-producing cell lines

R. F. TIEBOUT, E. A. M. STRICKER, and W. P. ZEIJLEMAKER

Immortalization of human B lymphocytes and the subsequent cloning of antibody-producing cells are essential steps towards production of human monoclonal antibodies. Such immortalization can be achieved by transformation of B cells with Epstein-Barr virus (EBV) as well as by fusion with continuously growing (myeloma) cell lines.

In our hands, EBV-transformation readily yields Iymphoblastoid cell lines producing immunoglobulins and specific antibody, but these cell lines are unstable; they lose antibody production within one or two months and they cannot be cloned at the single cell level. However, stable monoclonal cell lines can sometimes be set up. We have obtained two stable EBV-transformed cell lines (TT-1 and HBsAg), producing human monoclonal IgG antibodies against tetanus toxoid and HBsAg (up to 50 !!g/ml of culture supernatant).

In general, we found that cloning efficiency could be much improved by fusion of EBVtransformed cells with mouse myeloma cell lines; the resulting xenohybrids can be easily cloned at one celliweli. After expansion of cloned cells, immunoglobulin productions of > 30 !!g/ml are readily achieved and the xenohybrids are quite stable: upon bimonthly recloning, over 90 % of the subclones produce the specific antibody. Properties of 4 stable cloned cell lines are given in table 1.

Table 1.

cell line CLB-Hu-TT1 CLB-Hu-TT 1/1 CLB-Hu-TT 2 CLB-Hu-HBsAgl

antibody spec. tetanus toxoid tetanus toxoid tetanus toxoid HBsAg

Ig yl, K yl, K yH ytA

method EBV EBV + fusion EBV + fusion fusion (EBV+)

The advantage of EBV-transformation and cell fusion will be discussed as well as applications of the human monoclonal antibodies produced: two of the monoclonal anti-TT are biologically active in a mouse protection assay against tetanus toxin; the anti-HBsAg appears to be useful for screening tests for HBsAg.

a-1: signals in b cell activation

Documents