Identification of BTG2, an antiproliferative p53-dependent component of the DNA damage cellular response pathway.

Cell cycle regulation is critical for maintenance of genome integrity. A prominent factor that guarantees genomic stability of cells is p53 (ref. 1). The P53 gene encodes a transcription factor that has a role as a tumour suppressor. Identification of p53-target genes should provide greater insight into the molecular mechanisms that mediate the tumour suppressor activities of p53. The rodent Pc3/Tis21 gene was initially described as an immediate early gene induced by tumour promoters and growth factors in PC12 and Swiss 3T3 cells. It is expressed in a variety of cell and tissue types and encodes a remarkably labile protein. Pc3/Tis21 has a strong sequence similarity to the human antiproliferative BTG1 gene cloned from a chromosomal translocation of a B-cell chronic lymphocytic leukaemia. This similarity led us to speculate that BTG1 and the putative human homologue of Pc3/Tis21 (named BTG2) were members of a new family of genes involved in growth control and/or differentiation. This hypothesis was recently strengthened by the identification of a new antiproliferative protein, named TOB, which shares sequence similarity with BTG1 and PC3/TIS21 (ref. 7). Here, we cloned and localized the human BTG2 gene. We show that BTG2 expression is induced through a p53-dependent mechanism and that BTG2 function may be relevant to cell cycle control and cellular response to DNA damage.

Bone marrow function. I. Peripheral T cells are responsible for the increased auto-antiidiotype response of older mice.

After immunization with trinitrophenyl (TNP)-Ficoll, mice produced both anti-TNP antibodies and auto-anti-idiotype (auto-anti-Id) antibodies specific for the anti-TNP antibody. Older animals produced more auto-anti-Id than did young animals. When mice were exposed to a normally lethal dose of irradiation while their bone marrow (BM) was partially shielded, they survived and slowly (6 wk) regained immune function, as indicated by the number of nucleated cells in their spleen and the in vitro primary plaque-forming cell (PFC) response of their spleen cells to TNP-treated aminoethylated polyacrylamide beads. Recovery is presumably the result of repopulation of the peripheral lymphoid system by cells originating in the BM. By enzyme-linked immunosorbent assay (ELISA), and by hapten-augmentable PFC assay, we show that, after recovery from irradiation with their BM shielded, old animals produce low auto-anti-Id responses, like those of young animals. The transfer of splenic T cells into mice irradiated with their BM shielded provided evidence that the magnitude of the auto-anti-Id response is controlled by the peripheral T cells. Thus, mice that received splenic T cells from aged donors produced high levels of auto-anti-Id while those that received splenic T cells from young donors produce low levels of auto-anti-Id.

Cloning of the mouse BTG3 gene and definition of a new gene family (the BTG family) involved in the negative control of the cell cycle.

It is well known that loss of tumor suppressor genes and more generally of antiproliferative genes plays a key role in the development of most tumors. We report here the cloning of the mouse BTG3 gene and show that its human counterpart maps on chromosome 21. This evolutionarily conserved gene codes for a 30 kDa protein and is expressed in most adult murine and human tissues analyzed. However, we demonstrate that its expression is cell cycle dependent and peaks at the end of the G1 phase. This gene is homologous to the human BTG1, BTG2 and TOB genes which were demonstrated to act as inhibitors of cell proliferation. Its description allowed us to define better this seven gene family (the BTG gene family) at the structural level and to speculate about its physiological role in normal and tumoral cells. This family is mainly characterized by the presence of two conserved domains (BTG boxes A and B) of as yet undetermined function which are separated by a non-conserved 20-25 amino acid sequence.

Sequence analysis reveals that the BTG1 anti-proliferative gene is conserved throughout evolution in its coding and 3' non-coding regions.

The human BTG1 gene (expressing an anti-proliferative function) is an evolutionarily conserved gene homologous to the murine PC3/TIS21 genes. Here, we report the cloning and sequencing of the murine BTG1 coding region and chicken BTG1 cDNA. The putative human and mouse BTG1 proteins are 100% identical; the chicken BTG1 cDNA contains an open reading frame of 170 amino acids with a 91% identity to its human and murine counterparts. The 3'-untranslated region of BTG1 is also highly conserved (82% homology between human and chicken), suggesting that it plays a key role in the regulation of BTG1 expression. These data confirm that BTG1 is phylogenetically highly conserved and that BTG1 and PC3/TIS21 may constitute the first members of a new family of functionally related genes.

Active and passive re-expression of Fc gamma receptors on human lymphocytes.

Modulation of receptors for IgG (Fc gamma R) on human lymphocytes was induced by the interaction with erythrocyte-IgG antibody (EAG) complexes followed by incubation at 37 degrees C. The re-expression of Fc gamma R could be achieved by two independent processes. (a) Active synthesis, susceptible to inhibition by puromycin or cycloheximide was shown to peak 4 to 6 h after removal of EAG complexes; it required addition of at least 2% fetal calf serum. (b) Insertion of soluble Fc gamma R into the membrane of modulated lymphocytes was shown to occur within 20 min of contact between cells and Fc gamma R-containing supernatants; it was not altered by protein synthesis inhibitors. Fc gamma R-like material, spontaneously released by unstimulated peripheral blood lymphocytes or by polymorphonuclear cells, was taken up by modulated lymphocytes. This soluble material was fully absorbed on polymerized human IgG; it was non-dialyzable, thermolabile (56 degrees C, 30 min) and partially destroyed by freezing and thawing; it was recovered as two broad peaks from chromatography on polyacrylamide gel; it was shown to bind to both T and non-T Fc gamma R-bearing lymphocytes capable of forming EAG rosettes before modulation; and it could be inserted into allogeneic lymphocytes. These results remonstrate that the Fc gamma R structure bears two active sites, one binding to the Fc gamma and the other to the surface of EAG rosette-forming cells. Rapid release of soluble Fc gamma R from the cells and their possible insertion into the membrane may have important implications with respect to the biological functions associated with these receptors.

Human T lymphocyte receptors for IgM: control by IgG-binding lymphocytes.

The capacity of human peripheral blood lymphocytes to bind antigen-IgG or antigen-IgM-antibody complexes was investigated using a rosette technique with ox erythrocytes (E) coated with rabbit IgG (AG) or IgM (AM) antibodies. EAM rosette formation was achieved only in suspensions pre-incubated for 24 h at 37 degrees C. Addition of either EAM or EAG complexes to the culture medium was shown to prevent the formation of EAM rosettes. The inhibition was reversible, it was not due to trace IgM contaminants in the IgG antibody fraction. It was not observed when lymphocytes depleted of EAG-rosetting cells were incubated with EAG complexes. Inhibition of the expression of lymphocyte receptors for IgM can be regarded as a consequence of the modulation of surface receptors for IgG and involves an interaction between the two lymphocyte subsets bearing surface receptors for IgG and IgM, respectively. However, these experiments do not exclude the possibility that a few cells which bind EAG may lose their receptors by modulation and then express a receptor for IgM.

A rosette technique for identification of human mononuclear cells bearing Fc receptors.

A rosette technique has been described which showed some human mononuclear cells to bind chicken red blood cells coated with rabbit anti-chicken erythrocyte antibodies (EA). Intact 7S Ig, but not their F(ab')2 fragments, were found to sensitize chicken erythrocytes; therefore it was concluded that this rosette technique detects cells bearing receptors for the Fc portion of IgG. Among the different parameters of the reaction which were studied, centrigugation at 4 degrees C appeared necessary for obtaining the maximum number of rosettes. Kinetics of rosette dissociation indicated low and heterogeneous binding avidities. Distribution of EA-rosette forming cells among different organs appeared extremely heterogeneous: 8-28% were found in blood, spleen and bone marrow, whereas about 3% were found in lymph nodes and less than 1% in tonsils and thymus; this suggests that EA-rosette forming cells are not restricted to a single subclass of either T or B lymphocytes.

Contribution of lymphocytes bearing Fcgamma receptors to PHA-induced cytotoxicity.

Lymphocytes participating in PHA-induced lysis of chicken erythrocytes were characterized by means of cell fractionation methods. Selective depletion of, or enrichment in, E-rosetting cells indicated that the effector cell population was heterogenous, consisting of both T and non-T lymphocytes. Most effector cells, however, were shown to bear Fcgamma receptors detected by the formation of erythrocyte-antibody (EA) rosettes, but to lack C3 receptors. This distribution of effector cells among tonsils, peripheral blood and thoracic duct lymph paralleled that of EA-rosette forming cells but not that of T or B cells. Addition of aggregated IgG resulted in a moderate decrease of PHA cytotoxicity. However, almost complete inhibition was achieved within a few hours of contact between effectors cells and insoluble immune complexes. The results support the hypothesis that cytotoxic capacity is associated with the presence of Fcgamma receptors on the cell surface.

Distinct functions of surface receptors in the induction of neutrophil-mediated cytotoxicity.

The respective roles of cell surface receptors were studied in a model of cell-mediated cytotoxicity using 51Cr-labelled chicken erythrocytes as target cells and human polymorphonuclear neutrophils (PMN) as effector cells. The attachment of the targets to PMN, demonstrated by rosette formation, was achieved by PMN surface receptors for C3 or for Fc IgG. No receptors for Fc IgM could be demonstrated. Direct contact between targets and effector cells was required and no cell-free cytotoxic mediator was demonstrable in this model. Target cells bound to PMN-C3 receptors were not lysed unless a cytotoxicity inducing signal was delivered. This was provided by anti-PMN heteroantibodies, or by their F(ab')2 fragments as well. It was therefore possible to trigger the cytotoxic reaction by bypassing PMN-surface receptors for Fc IgG. When the target cells are coated with IgG antibodies, PMN receptors for Fc IgG ensure both the attachment and the triggering signal for the cytotoxic reaction. Polymorphonuclear neutrophils (PMN) have been reported to the effective killer cells in vitro under three different experimental conditions: during phagocytosis, in the presence of antibody directed against the target cells and in the presence of lectins. PMN accumulation has also been considered as a major component of the pathogenesis of many forms of immunologic tissue injuries since PMN may react with immune complexes bound to a surface which they cannot phagocytose. Under these circumstances, they release lysosomial enzymes, by a mechanism which has been called "reverse endocytosis" or "frustrated phagocytosis". Attachment of PMN to target involves cell surface receptors (Fc gamma R) for the Fc region of the IgG molecule and/or receptors (C3R) for the activated third component of complement. The binding of aggregated or antigen-complexed IgG to PMN surface Fc gamma R generates signals triggering the internalization phase of phagocytosis, the antibody-dependent cell cytotoxicity (ADCC) and the stimulation of glucose oxidation by the hexose monophosphate pathway. However, the latter metabolic activation was also reported to be triggered by the fixation of antibodies specific for PMN surface determinants. It was therefore conceivable that modifications induced at the membrane level on any structure distinct from Fc gamma R would produce metabolic changes leading to target cell destruction, provided that a close contact could be established between effector and target cells. In the present study we have investigated the respective roles of Fc gamma R, C3R and other yet undefined surface determinants of PMN in the induction of cytotoxic activity towards heterologous target cells.

Class-specific suppression of human B cell maturation by IgA-binding factors.

IgA-binding factors (IgA-BFs) were prepared by chromatography on Sepharose 4B beads covalently linked to dimeric and polymeric monoclonal IgA1 from supernatants of peripheral blood mononuclear lymphocytes (PBMC) and human B cell lines incubated in serum-free medium. Receptors for IgA, as revealed by the binding of biotinylated monoclonal IgA1, were expressed on monocytes, T-enriched and T-depleted lymphocytes. IgA-BFs or control eluates were added to pokeweed mitogen (PWM)-stimulated PBMC cultures, and their effects on the terminal differentiation of polyclonally activated human B cells were assessed by enumeration of intracytoplasmic IgM-, IgG- or IgA-containing cells. A selective decrease of IgA-containing cells was observed in the presence of IgA-BFs whereas IgM- and IgG-containing cells remained unchanged. Differential counts of B blasts and plasma cells revealed that only the former were decreased following addition of IgA-BFs. Kinetic studies indicated that maximum inhibition of IgA-containing cell generation was achieved when IgA-BFs were added during the first 5 days of PWM-stimulated PBMC cultures, whereas no inhibition could be demonstrated when IgA-BFs were added 24 h before harvesting. IgA-BFs did not decrease [3H]thymidine incorporation in PWM-stimulated PBMC cultures. They diminished the proliferation of the surface IgA+ monoclonal human B cell line DAKIKI, but not that of the surface IgA- IM-9 cell line. Several control eluates obtained from the same cell supernatants absorbed on Sepharose 4B, Sepharose 4B-IgG or Sepharose 4B-beta 2-microglobulin had no effect. Finally, IgA-BFs prepared from supernatants of two human B cell lines bearing receptors for IgA selectively depressed the generation of intracytoplasmic IgA+ cells in PBMC cultures stimulated by PWM. Altogether the data indicate that IgA-BFs obtained by spontaneous release from heterogeneous mononuclear cell suspensions or from IgA receptor-positive human monoclonal B cell lines selectively depress the maturation of B cells into IgA plasma cells and the proliferation of a surface IgA+ B cell line.

Distribution of cells binding erythrocyte-antibody (EA) complexes in human lymphoid populations.

Cells bearing Fc receptors were studied using a rosette technique with chicken erythrocytes coated with rabbit antibodies (EA). By comparison with other markers and by selective depletion of T and B cells it was demonstrated that EA rosettes were formed by about 10% of the monocytes, 7% of the T lymphocytes, and 45% of the B cells. In addition, 26% of EA-rosette-forming cells (EA-RFC) in the peripheral blood carried on other markers than C3 receptors. Marked differences were found in the percentages of EA-RFC in cell suspensions from various lymphoid organs: peripheral blood, 15.2%; spleen, 26.6%; bone marrow, 30.9%; lymph nodes, 2.8%; thymus, 0.3%; and tonsils, 0.5%. It was concluded that Fc receptors were present on subsets of different lymphocyte subpopulations; their presence could not be regarded as a specific marker of one of these subpopulations.

Antibody-dependent cell cytoxicity (ADCC). Characterization of 'killer' cells in human lymphoid organs.

The effector cell(s) in human antibody-dependent cell cytotoxicity (ADCC), with antibody-coated chickens erythrocytes as targets, was studied by comparison of cell suspensions from various lymphoid organs and by means of various cell fractionation methods. Effector cells (K) were found mostly in peripheral blood, spleen, and bone marrow but not in tonsils, lymph nodes, and thymus. Effector cells bear Fc receptors and can form EA rosettes with the antibody-coated target cells. About 1,5% peripheral blood lymphocytes can form 'high-avidity' EA rosettes with targets coated at low antiserum concentration. Most of the effector cells belong to this small subset, as shown by experiments of selective depletion. Removal of most monocytes, T cells, or B cells from, or addition of T-cell-specific antiserum to, the effector cell suspensions did not affect ADCC. Effector cells in this model of ADCC therefore lack the conventional B- or T-cell markers but at least some of them are likely to bear C3 receptors.

Unexpected lack of reactivity of allogeneic anti-Ia monoclonal antibodies with MHC class II molecules expressed by mouse intestinal epithelial cells.

We have examined MHC class II molecules expression by murine gut epithelial cells using a large panel of anti-Ia antibodies. In contrast to conventional APC (i.e., B cells and macrophages), only two anti-Ia antibodies reacted with enterocytes: a xenogeneic rat anti-Ia mAb (CD311) directed against a monomorphic class II determinant, and a polyclonal antiserum directed against both I-A and I-E heterodimers. In contrast, allogeneic anti-Ia mAb were either unreactive (17 of 20) or reacted weakly (3 of 20) with enterocytes, even after in vivo treatment with IFN-gamma. This pattern of Ia reactivity of epithelial cells was tissue specific (restricted to gut mucosa) and cell specific (restricted to gut epithelial cells). Biochemical and molecular studies confirmed that enterocytes expressed I-A and I-E isotypes on their cell surface and contained mRNA of both subregion loci. Interestingly, enterocytes appeared deficient in expression of the MHC class II-associated invariant chain, and are not able to stimulate allogeneic T cells. These data suggest that gut epithelial cells express a conformation of class II molecules, antigenically distinct from that expressed on conventional APC.

[Prognostic value of delayed-hypersensitivity skin tests and rosette studies in acute non-lymphoid leukemias]. / Valeur pronostique des réactions cutanées d'hypersensibilité retardée et de l'étude des rosettes dans les leucémies aiguës non lymphoïdes.

Delayed hypersensitivity skin reactions to Tuberculin and Candidin were studied in 28 patients with non lymphoid acute leukemias. The reactions were found negative in most patients during blastic crises, whereas delayed skin reactions to Candidin were positive during remissions. The possible prognostic significance of the depressed delayed hypersensitivity response in such patients deserves further studies. Alterations of circulating T lymphocytes were observed, including low percentages of E and active E rosette-forming cells during blastic crises, and persisting low E and E active rosettes in some patients in remission; such abnormalities were less frequent in patients with remission of long duration. The percentages of EA and EAC rosettes-forming cells were found normal during blastic crises and remissions. In some patients in remission, lymphocytes were found to bind sheep erythrocytes either uncoated or coated with IgM antibodies, this penomenon which is not observed with normal lymphocytes may reveal persisting abnormalities of a yet undefined nature.

[Human lymphocyte receptors for the Fc ends of IgG]. / Les récepteurs des lymphocytes humains pour l'extrémité Fc des IgG.

Receptors for Fc IgG can be demonstrated by the binding of aggregated IgG or erythrocyte-IgG antibody complexes (EAG) onto subsets of B, T and "nul" lymphocytes. Among such cells are the effectors of antibody-dependent cell-mediated cytoxicity, and suppressor T cells. The binding of insoluble complexes induces a reversible modulation of the receptors associated with impaired proliferative T cell responses and transient inhibition of IgM receptors expression by adjacent T cells. Soluble receptors for Fc IgG bear a membrane binding site; they inhibit in vitro B cell differentiation induced by-T-dependent or T-independent polyclonal B cell activators.

Suppression of mitogen-induced peripheral B cell differentiation by soluble Fc gamma receptors released from lymphocytes.

Soluble receptors for FcIgG released from unstimulated human peripheral blood lymphocytes were isolated by affinity chromatography on Sepharose 4B-IgG. This material was shown to interfere with the differentiation of peripheral blood B cells into Ig-secreting cells in cultures stimulated with pokeweek or Nocardia opaca extracts. Neither cell viability nor [3H]thymidine incorporation were altered, but the number of Ig-containing cells and that of Ig-secreting cells were decreased. These effects were dose-related. They were found to be associated with Fc IgG-binding soluble material, since absorption on Sepharose 4B-IgG but not on pepsin-digested F(ab')2 fragments removed the inhibitory activity. This suppressor factor, released by unstimulated lymphocytes, may represent a human analogue of murine immunoglobulin-binding factor (IBF) produced by alloactivated T cells.

Suppression of the late stages of mitogen-induced human B cell differentiation by FC gamma receptors (FC gamma R) released from polymorphonuclear neutrophils.

During short incubation in serum-free medium, polymorphonuclear neutrophils (PMN) release soluble material that can be characterized as receptors for Fc IgG (Fc gamma R) on the following evidence: it agglutinates erythrocyte-IgG antibody (EAG) complexes, it prevents the binding of EAG to EAG-rosette-forming cells, and it binds to EAG-rosette-forming cells after modulation of their Fc gamma R, allowing the formation of 'passive' rosettes. These Fc gamma R were isolated by affinity chromatography on sepharose 4B-IgG. This material was shown to interfere with the differentiation of peripheral blood B cells into Ig-secreting cells in cultures stimulated by pokeweed (PWM) or Nocardia opaca (NOC) extracts. The number of Ig-secreting cells determined by a reverse hemolytic plaque assay using protein A-coated sheep erythrocytes was decreased by addition of Fc gamma R over a wide range of dilutions. The number of Ig-containing cells was diminished in PWM-stimulated cultures, but not in cultures stimulated with NOC. Fc gamma R did not affect cell viability, nor did they interfere with plaque-forming cell assay. Fc gamma R was not suppressive when added before the 3rd day or after the 6th day of culture. The suppressor factor was shown to remain associated with Fc gamma R after elution at acidic pH; it was removed by absorption on Sepharose 4B-IgG, but not on pepsin-digested F(ab')2 fragments. The suppressive factor as well as the capacity to restore EAG rosette formation by modulated lymphocytes were destroyed by heating (56 degrees C, 30 min) or by freezing and thawing. Properties of Fc gamma R released from PMN in this system are similar to those of Fc gamma R released from unstimulated human peripheral blood lymphocytes (PBL) and to those of mouse Ig-binding factor produced by alloactivated T cells or T cell lines.

Presence of "La-like" antigens on human T lymphocytes bearing receptors for IgG.

Human peripheral blood T cells bearing receptors for the Fc portion of immunoglobulins were characterized by the formation of double rosettes with sheep red blood cells and IgG-coated chicken erythrocytes (EA). Treatment of cells by anti-"Ia-like" antisera inhibits the binding of EA. Inihbition appeared to be specific since anti-HLA-A and HLA-B antibodies did not inhibit the formation of EA rosettes, whereas F (ab')2 fragments of anti-Ia-like IgG were found to be as potent inhibitors as the whole IgG. This result shows that in human and in mice, Ia-like determinants arepresent, close to Fc gamma receptors, on the surface of the T lymphocytes bearing such receptors.