(1) Seminar on “Transcription Factors in the Regulation of Oncogenesis”
This meeting was held February 9-11, 1995, at the Maui Prince Hotel in Maui, Hawaii. The organizers were Dr. Tada Taniguchi, University of Tokyo, and Dr. Tom Curran, Roche Institute of Molecular Biology.
Seven speakers from Japan and eight from the United States met to discuss the latest findings on the role of transcription factors In the regulation of oncogenesis. This is an extremely fast moving subject that links several disciplines. The talks ranged from an analysis of the genes involved in tumor formation to the regulation of the cell cycle and the interface with signal transduction mechanisms. The presentations were divided into four topics.
Session 1. Transcription factors in oncogenesis and tumor suppression
Dr. Yoshiaki Itoh opened the meeting with a report on a transcription factor termed PEBP2/CBF which has been linked to acute myeloid leukemia (AML). This transcription factor consists of alpha and beta subunits that bind to DNA in the from of a protein complex. Interestingly, gene rearrangements have been reported for both subunits in AML. The alpha subunit is related to the gene responsible for the developmental mutation runt in Drosophila. Dr. Itoh has carried out a structure/function analysis of both proteins revealing the molecular domains involved in protein dimerization and DNA binding. The alpha protein binds to DNA weakly by itself through the runt-homology region. The beta protein requires the presence of alpha for DNA binding activity. The beta protein is usually expressed in the cytoplasm but translocates to the nucleus in the presence of alpha. Dr. Itoh also provided evidence suggesting that the alpha and beta proteins need to act in concert with a third transcription factor, a member of the ets oncogene family, to activate transcription strongly in T-cells. This introduced a recurrent theme in the meeting that gene regulation occurs through multiple protein-protein interactions. Furthermore, while the identification of oncogenes and tumor suppresser genes has provided access to key molecules responsible for malignant conversion, it is now essential to elucidate the nature of their molecular functions in complex biological systems. Dr. Keith Harshman discussed the isolation and characterization of BRCA1, a recently isolated tumor suppresser gene linked to breast cancer. Although several families have been characterized with germ-1ine mutations in BRCA1 so far there is no report of a somatic mutation in this gene. Dr. Harshman suggested that it may provide a role in the early development of breast tissue. BRCA1 is expressed in testis, breast and thymus cells but it is low in other cell types tested. Its protein product contains a putative “ring-finger” structure. These cystein repeat sequences are somewhat similar to zinc-fingers but they have not been linked to DNA binding activity. At present the molecular function of the BRCA1 protein remains unknown. Future studies are aimed at biochemical characterization and an analysis of the effect of BRCA1 expression on normal and tumor cells.
Dr. Robert Eckner described some of his recent studies carried out in Dr. David Livingston’s laboratory on the adenovirus E1A binding protein p300. This protein had been identified previously as one of several bound to EIA in transformed cells. The cloning of the gene encoding p300 revealed a high degree of sequence identity with a protein termed CBP. CBP, CREB-binding protein, is a protein that binds to the phosphorylated transcription factor CREB. It is believed to function in association with CREB in cAMP-mediated transcription activation. p300 appears to be a member of the CBP family. It contains several features of transcription control proteins such as cystein/histidine repeats and a region of homology with the transcriptional coactivator ADA2. Dr. Eckner presented evidence suggesting that the role of p300 and CBP in transcription may be more complicated than previously expected. Although E1A inhibits transcription activation by CREB and protein kinase A, the CBP/CREB complex can be detected in the absence of phosphorylation. In coprecipitation assays, both T-antigen and p53 could be detected in association with p300.
Dr. Kinuko Mitani presented findings on the AML/EV-1 fusion gene that results from a translocation in AML. This fusion gene encodes a transcription factor that retains the runt-homology region of AML1 and the two zinc finger domains of EV-1. It is capable of cooperating with the BCR/abl fusion gene in cotransformation assays in rat-1 cells. This is a complex transcription factor that has many possible activities. The second zinc finger domain, which has similarities with ets proteins, is capable of activating an AP-1 reporter plasmid by itself. This domain is also necessary for cotransformation activity. The runt-homology domain of the AML1 gene is capable of inducing proliferation in a GCSF-dependent manner. However, there are two variants of normal the AML1 gene (1a and 1b). The la gene has an additional 3’ region containing PST sequences. This version of the gene is capable of inducing cellular differentiation. Thus, it is possible that the ratio of 1a to 1b determines the cellular decision to differentiate or proliferate. Thus, to understand the role of the AML1/EV-1 fusion gene in oncogenesis it will be important to determine the functional contribution of each of the protein domains and their impact on other transcription factors expressed in the transformed cell.
Session 2. The transcription factor network
Dr. Robert Eisenman introduced the myc network of transcription factors. The myc oncogene is involved in a complex series of interactions with the mad and max genes. This complexity is increased now that it is clear that there are as many as 4 mad genes. Recent data indicate that Mad and Mad/Max heterodimers can bind to a gene related to the yeast sin3 transcriptional regulator. Dr. Eisenman is investigating the growth regulatory effects of Myc and Mad by expressing them in cultured cells and by analyzing their expression in vivo. In general, Myc is associated with proliferation and Mad with differentiation. This may not be a simple competition between the two proteins as in gastric mucosa Myc is expressed early in proliferating cells then there is a delay of two days before Mad is expressed in differentiating cells. Dr. Eisenman has also attempted to identify the gene targets of Myc by immunoprecipitation of DNA-bound protein complexes. Initial results have identified potential targets such as a Myc-induced ubiquitln conjugating enzyme, a gene related to cdc34 and an RNA helicase of the DEAD box family. This provides a possible approach to the most critical question in the field which is to identify the targets of transcription factors that are responsible for induction of the oncogenic phenotype.
Dr. Yoshida discussed the interactions Involving the retroviral regulator Tax and cellular transcription factors. Tax plays a major role In the pathological phenotype of HTLV-1 which is associated with adult T-cell leukemia. Although originally identified as a regulator of viral genes, it is now clear that Tax has a profound effect on the host cell genome. The cellular targets of Tax include lymphokine receptor genes and growth factors. Dr. Yoshida has identified several transcription factors that interact directly with Tax to mediate these effects. In particular, several members of the CREB and NFkB families of transcription factors associate with Tax. However, the situation is not simple Tax can bind to the rel homology domain of several proteins in coprecipitation assays. The complex with p65 is approximately 200 times more stable than that with c-rel. The interaction of Tax with IkB alpha and beta is very unstable. These complexes are degraded so fast in cells that they cannot be detected in stable form. This may be a component of a complex feed back loop. Thus, these results indicate that transcription factors can elaborate a complex cascade of events when overexpressed and it will be necessary to analyze the consequence of these events to understand the molecular basis of oncogenesis.
Dr. Nishizawa talked about the multiple interactions among members of the Fos, Jun. Maf and CNC basic region-leucine zipper (bZip) family of transcription factors. Like fos and jun, maf was first identified as a retroviral oncogene. It has a bZip structure very similar to that of Fos and Jun and it participates in an array of heterodimeric complexes that bind to AP-1 like DNA sequences. The complexity of this family is also increased by the existence of several related family members (mafB, nrl, mafK, mafL and mafG). All of the maf family proteins have similar DNA binding specificities. In addition, there are some other more distantly related proteins such as CNC, NFe2, Nrfl and Nrf2 which, together with CREB/ATF proteins, makes this an extremely extensive transcription factor network. Thus, it is extremely difficult to delineate the in vivo target genes for any one bZip dimer. Recently, the demonstration that the NFe2 transcription factor, which regulates globin expression, is a heterodimeric complex of p45 and mafK, mafL or mafG suggests that these genes may be involved in erythroid differentiation. Indeed, expression of mafK in mouse erythroleukemia cells increased differentiation.
Session 3. Cell signaling and growth control
Dr. Tom Curran took the issues of signaling and transcription regulation by oncogenic transcription factors into whole animal systems. The c-fos promoter has been analyzed extensively in cell culture and a series of regulatory elements have been defined. Although in isolation each element can be shown to be sufficient for a particular stimuli, for example the SRE for serum induction and the CRE for responses to cAMP, they function In an interdependent fashion in the context of the natural promoter in transgenic mice. Introduction of point mutations into the SIE, SRE, AP-1 and Ca-CRE elements separately, in a c-foslacZ fusion construct, severely impaired promoter responses to a range of stimuli. Thus, in a natural context, multiple factors may be required for transcriptional responses to extracellular stimuli. Dr. Curran also presented findings on the knock-out of the erbA oncogene in mice by homologous recombination. The erbA oncogene is derived from the thyroid hormone receptor and it functions as a hormone dependent transcription factor. Mice lacking the thyroid hormone receptor beta gene exhibit a phenotype similar to that seen in humans with the genetic deficiency generalized resistance to thyroid hormone syndrome. The mice have hypothyroidism, increased levels of T3, T4 and TSH and they have a hearing impairment. It is hoped that they will provide a useful model for this debilitating human disorder that will allow identification of the physiologically important target genes of thyroid hormone receptor beta in vivo.
Dr. Yukiko Gotoh provided evidence supporting a role for the Map kinase (MapK) cascade in Xenopus oocyie maturation. MapK is activated by growth factors and it plays a critical role in a great many signal transduction pathways involved in growth and development. In Xenopus oocyies MapK can be activated by two distinct pathways. One involves ras, raf and MapK kinase (MapKK) and the other involves mos activation of MapKK. MapK itself may phosphorylate MapKK in a positive feedback loop. To investigate the role of MapK in oocyies, Dr. Gotoh prepared a constitutively active form of the kinase by substituting both serine 218 and serine 222, which are phosphorylation targets of MapKK, with glutamic acid residues. These mutations mimic the effects of phosphorylation and create a kinase that is active in the absence of any stimulus. Injection of this kinase into Xenopus oocytes is sufficient to induce maturation. In addition to induction of maturation, mos is also involved in the cytostatic factor response. MapK antibodies block mos-induced metaphase arrest. Thus, MapK can mediate both of the actions of mos in Xenopus oocyies. This conclusion is also supported by the finding that the MapK phosphatase (CL100) blocks the induction of mesoderm by fibroblast growth factor (FGF). Thus, MapK is also essential for mesoderm induction by FGF. These results expand the signal transduction events associated with MapK and provide a useful experimental model for analysis of the function of this critical enzyme.
Dr. Martine Roussel demonstrated that cyclin D1 requires myc to stimulate cell cycle progression using a cell line that is dependent on CSF-1 for growth. Mutation of tyrosine 809 to phenylalanine in the CSF-1 receptor inhibits myc activation, although other immediate early genes are induced, and blocks cells early in G1. This block could be bypassed by introduction of exogenous Myc. Interestingly, the requirement for Myc was abrogated by transfection with vectors expressing cyclin D1, D2 and D3 but not by cyclin E. Maintenance of cyclin D expression was also show to increase the levels of Rb kinase activity. Despite this dramatic effect on cell cycle progression, the cyclins did not induce cell transformation and the rescue was dependent on the presence of CSF-1 . The effect of cyclin D was also reduced by introduction of Mad and by a dominant negative mutation of Myc. Taken together, these results indicate that Myc/Max dimers are required for the stimulation of cell cycle progression by cyclin D. However, the exact relationship between these molecules and the contribution of other interacting signaling pathways to G1 progression are still to be elucidated. This cell system provides a unique opportunity for dissecting the molecular events required for cell cycle progression.
Dr. Tadatsugu Taniguchi closed the session with a discussion of the relationship between IL-2 induced oncogene expression and cell cycle progression. IL-2 activates a tripartite receptor comprising alpha, beta and gamma chains. Stimulation results in the recruitment of a host of signaling molecules to the receptor complex and the initiation of multiple signal transduction pathways. The beta chain binds the Syk, Lck and Jak1 kinases through acidic and serine rich regions. Syk induces myc expression and Lck activates fos and jun through the ras-raf-MapK pathway. Activation of Jak1 results in phosphorylation and activation of the STAT5 transcription factor. The gamma chain recruits Jak3 which ultimately activates Bcl2 through a Rap-FRAP pathway. Using a combination of mutagenesis studies, biochemistry and transfection approaches, Dr. Taniguchi demonstrated that the biological responses to IL-2 stimulation are mediated by combinations of these signaling molecules and that no single molecule or pathway is sufficient for growth. Indeed, at least three target genes on two distinct pathways are required. Using this system it was also demonstrated that Tax has a similar effect to Bcl2 on growth stimulation. This Bcl2 effect does not appear to be associated with apoptosis.
Session 4. Cell cycle regulation
Dr. Charles Sherr continued the analysis of the role of cyclin D in cell cycle progression. Overexpression of cyclin D has several effects on the cell cycle, including a contraction of G1, a reduced serum requirement and a reduction in cell size. The preponderance of evidence suggests that a complex between cyclin D and cdk4 functions as the critical Rb kinase in stimulating the G1 to S transition. Cyclin D also contains an LXCXE motif, not present in cyclin E, that controls association with Rb. The cyclin D/cdk4 complex is inhibited by a series of proteins (p16, p15, p18 and p19) known as INK4 p16 has been shown to represent the MSH tumor suppresser gene. Although each of these proteins functions similarly in vitro as kinase inhibitors, they must have unique properties in vivo. For example, p16 is elevated in RB- cells suggesting that it may be part of a negative feedback loop. Although p19/18 oscillate during the cell cycle, overexpression of p19 with an LTR arrests cells in G1. At present it is not clear why there are so many inhibitors of cyclin D/cdk4 and why p16 is a tumor suppresser gene whereas p19 and p18 are not .
Dr. David Beach extended the array of regulatory molecules involved in cell cycle progression by discussing the role of the phosphatase cdc25. In fission yeast, cdc25 regulates cdk/cyclin complexes by removing the phosphates put on by the weel and mikl kinases. Using a two-hybrid approach, Dr. Beach demonstrated an association of mammalian cdc25 with itself(cdc25 a and b), with 14-3-3 proteins and with ratf1. This interaction was confirmed biochemically using GST fusion proteins and was demonstrated in cell extracts bycoprecipitation. In cotransformation assays, cdc25 a and b, but not c, were capable of cooperatiori with ras. Furthermore, cdc25 expression transforms Rb- but not p53- cells.
Over expression of cdc25 was associated with aneuploidy in these assays. These studies imply that cdc25 provides a key link between the ras-raf signal transduction pathway and the machinery that drives the cell cycle.
Dr. Erin O’Shea described an additional role for cyclin-related proteins in the response of yeast to phosphate. The Pho5 gene encodes a phosphatase that scavenges inorganic phosphate. The levels of this gene are regulated negatively by the genes Pho80 and Pho85. Pho80 shows greater than 50% identity to cdc28 and Pho85 shows 35% identity to the conserved domain in cyclins. Biochemical studies revealed that Pho80 and Pho85 form a complex that is capable of phosphorylating Pho4, a regulator of Pho5 transcription. In the presence of high levels of phosphate, Pho4 is phosphorylated and transcription of Pho5 is repressed. Mutation of the phosphorylation sites on Pho4 reduces repressor activity. An upstream inhibitor of Pho85, Pho81, has been identified that has ankyrin repeat sequences similar to p16, which are sufficient to function as an inhibitor. These fuming demonstrate that the response of yeast to a low phosphate environment involves an analogous pathway to that responsible for cell cycle control. It is possible that this provides a mechanism for coupling the availability of phosphate to the regulation of cell cycle progression.
Dr. Mitsuhiro Yanagida concluded the meeting with a presentation of mitotic regulators in yeast. By analyzing the regulation of a gene involved in chromatin region maintenance (crm) in fission yeast, he implicated a yeast AP-1 protein (Pap1) in the regulation of chromatin structure. This function has not yet been demonstrated in mammalian cells, but it is clear that it could have important implications for oncogenesis. Dr. Yanagida also presented studies on the regulation of Mitosis by protein phosphatase 1 (PP1). Fission yeast has two versions of this enzyme encoded by dis2 and sds21. The activity of dis2 is reduced by phosphorylation by cdc2. The double deletion of these genes is lethal causing arrest at metaphase resulting in a short spindle and condensed chromosomes. Dis2 is regulated by association with sds22 which is an essential gene required for the metaphase/anaphase transition. The product of sds22 consists of leucine-rich repeats similar to those in RNAse. It binds and stabilizes the dephosphorylated form of dis2. These findings highlight the importance of protein phosphatases in signal transduction and cell cycle regulation.
Conclusions
In general the meeting was a great success. During the sessions there was a free exchange of information, including a high percentage of unpublished data, and lively discussions. The discussions always continued outside of the formal sessions. It was notable that, although several disciplines were represented at the meeting, there was a substantial amount of common ground. In many cases, investigators were wrestling with the same conceptual difficulties in approaching complex biological processes from a molecular perspective. It was also clear that to understand the molecular basis of cancer it will be necessary to foster such cross discipline interactions in order to elucidate the complexities of the signal transduction processes that govern transcription regulation.
PARTICIPANTS
UNITED STATES
Dr. David Beach
Cold Spring Harbor Laboratories
1 Bungtown Road
Cold Spring Harbor, NY11724
Dr. Erin O’Shea
Dept. of Biochemistry and Biophysics
University of Caledonia, San Francisco
San Francisco, CA 94143-0448
Dr. Keith Harshman
Myriad Genetics
390 Wakara Way
Salt Lake City, UT 84103
Dr. Tom Curran
Associate Director
Roche Institute of Molecular Biology
340 Kingsland Street Nutley, NJ 07110
Dr. Robert Eisenman
Fred Hutchinson Cancer Research Center
1124 Columbia Street (A2-025)
Seattle, WA 98104
Dr. Charles J. Sherr
St. Jude Children’s Research Hospital
Dept. of Tumor and Cell Biology
322 North Lauderdale Street
Memphis, TN 38105
Dr. Martine Roussel
St. Jude Children’s Research Hospital
Dept. of Tumor and Cell Biology
322 North Lauderdale Street
Memphis, TN 38105
Dr. Richard Eckner
Dana Farber Cancer Institute
44 Binney Street
Boston, MA 02115
JAPAN
Dr. Yoshiaki Itoh
Institute of Virus Research
Kyoto University
53 Shogoin-Kawaharacho
Sakyo-ku, Kyoto, 606
Dr. Yukiko Gotoh
Institute of Virus Research
Kyoto University
53 Shogoin-Kawaharacho
Sakyo-ku, Kyoto, 606
Dr. Tadatsugu Taniguchi
Dept. of Immunology
Faculty of Medicine
The University of Tokyo
Hongo 7-3-1 Bunkyo-ku, Tokyo 113
Dr. Makoto Nishizawa
Faculty of Medicine
Kyoto University
Yoshida Konoe-cho
Sakyo-ku, Kyoto 606
Dr. Kinuko Mitani
Faculty of Medicine
University of Tokyo
Hongo 7-3- l
Bunkyo-ku, Tokyo 113
Dr. Mitsuhiro Yanagida
Faculty of Sciences
Kyoto University
Oiwake-cho, Kitashirakawa
Sakyo-ku, Kyoto 606
Dr. Mitsuaki Yoshida
The Institute of Medical Science
University of Tokyo
4-6-1 Shirokanedai
Minato-ku, Tokyo 108
(2) Seminar on “Receptor-ligand Interactions and Signal Transduction in Antigen-specific Immune Responses”
A meeting focused on the topic of “Receptor-ligand interactions and signal transduction in antigen-specific immune responses” was held in Maui, Hawaii, January 9-11, 1995, and was organized by Drs. Hiromi Fujiwara and Richard Hodes. There were ten participants from the United States and seven from Japan.
I. The first session dealt with the subject of receptor-ligand Interactions in T lymphocyte activation.
Dr. Lany Samelson described recent studies assessing the role of tyrosine phosphorylation in T cell receptor (TcR) activation. The roles of TcR-associated zeta chain and ZAP-70 were discussed in detail. A comparison of agonist and antagonist ligands for the TcR revealed differences in the pattern of phosphorylation of zeta chain and a change in ZAP-70 association with the TcR complex. Agonist ligand induced association of ZAP-70 with TcR as well as extensive phosphorylation of ZAP-70. In contrast, antagonist induced association of ZAP-70 with the TcR, but with substantially reduced phosphorylation of this ZAP-70. These findings provide a biochemical correlate for the difference in action of agonist and antagonist ligands for the TcR. Dr. Paul Allen described his studies in the same area, employing altered peptide ligands for the TcR, with either agonist or antagonist properties. Agonists and antagonists gave distinct patterns of zeta phosphorylation and differential association with active ZAP-70. The physiologic role of endogenous peptides as partial agonists and their effect on T cell selection during thymic development was discussed.
Dr. Takashi Saitoh discussed the molecular architecture and signalling through the TcR complex. He described the kinetics of zeta chain expression in association with cell surface TcR. Through the use of zeta knockout mice and reconstitution with zeta transgenes, Dr. Saitoh showed that a truncated zeta mutant lacking an activation motif promoted some differentiation of single positive CD4+ cells without completely rescuing thymic differentiation.
Dr. Yasuharu Nishimura described his research involving the specific presentation of a human insulin peptide to autoreactive T cells. An insulin autoimmune syndrome was described in which peptides derived from insulin are presented in association with an HLA-ORB allele that Is specifically found in patients with this syndrome. These data suggest that autoimmunity in this case results from ability of a specific HLA-DR allele to present autoantigen to self-reactive T cells.
Dr. Takeshi Watanabe described signalling through T and B cell antigen receptor complexes and through the HS1 (hematopoietic-specific gene) protein. The HSI gene product is broadly expressed in human hematopoietic cells and is a major substrate for tyrosine kinase. Activation of B cells by surface IgM cross linking increases expression of HS1 protein. Recent experiments have demonstrated that mice which are deficient in HS1 by genomic deletion have grossly normal T cell and B cell development but demonstrate a reduced capacity for B cell and T cell proliferation in response to antigen receptor cross linking. These studies indicate a role for HS1 in lymphocyte activation.
Cell cycle regulation by tumor suppressors in human T cells was discussed by Dr. Toshio Nikaidoh. These studies demonstrated a role for Rb and p53 gene products in regulating p34 cdc2 and cyclin A gene expression respectively. These regulatory events appear to play a role in normal T cell proliferation in response to IL-2. The potential role of defects in these regulatory connections during tumorigenesis was discussed.
II. The second session of the meeting centered on the topic of costimulatory signals in T lymphocyte activation.
The expression and function of B7-1 and B7-2 as alternative costimulatory ligands for T cell activation was discussed by Dr. Richard Hodes. The expression of the ligands, and the potential for their differential regulation during cell activation, was described. In addition, the in vitro and in vivo roles of these costimulatory ligands were assessed by antibody blocking. Both allograft rejection and T cell dependent antibody responses were substantially inhibited in vivo by antibodies to B7-2, indicating a role for B7-2 expressing cells, and possibly the B7-2 molecule, in these responses.
The role of CD28 and B7 in T cell activation were further discussed by Dr. Carl June Dr. June compared signal transduction induced by occupancy of CD28 with either anti-CD28 antibody or with B7-1 or B7-2 ligands. Although some differences were seen in the activation events triggered by anti-CD28 antibody and by B7 ligands, no differences were seen in the responses induced by B7-1 versus B7-2. The role of these costimulatory signals in autoimmune pathology in vivo was assessed in two models. It was found that ectopic B7-1 expression in the pancreas was not sufficient to induce disease, although B7-1 expression combined with drug-induced injury or with expression of an autoreactive TcR transgene did lead to diabetes. In another model, that of experimental autoimmune encephalomyelitis (EAE), it was found that antibody to B7-1 exacerbated disease, while anti-B7-2 did not, indicating a different role for these two costirnulatory ligands in the pathogenic process.
Drs. Jeffrey Bluestone and Laurie Glimcher presented related studies characterizing the role of costimuli in autoimmune pathology. In the NOD (non-obese diabetes) model it was found that treatment with anti-B7-1 antibody made disease worse, leading to 100% disease involvement even in male mice. In contrast, anti-B7-2 inhibited the spontaneous development of disease. Anti-B7-1 F(cab)2 fragments did not augment disease, suggesting that the role of anti-B7-l required cross linking of the B7-1 molecule. Dr. Bluestone also studied the EAE model in PL/J mice using a treatment protocol that was initiated after the first spontaneous remission of disease. Using this treatment schedule, it was found that anti-B7-1 worsened disease, whereas anti-B7-2 had no effect; anti-B7-1 Fab improved disease, again suggesting a signalling role for the B7-1 molecule. Dr. Laurie Glimcher also studied the EAE model in PL/J mice. However, in contrast to the treatment schedule employed by Dr. Bluestone, Dr. Glimcher introduced treatment at the time of induction of EAE by PLP antigen peptide 139-151. Under these conditions, anti-B7-1 ameliorated disease, and anti-B7-2 worsened disease. Taken together, the result of these investigators indicate a complex role for costimulatory signals at different stages in the development of autoimmune disease, and suggest that more complicated experimental protocols will be required to further elucidate both mechanism and potential for therapeutic intervention.
Dr. Barbara Bierer discussed details of the signalling pathway involved in CD28 function. After cross linking of CD28 in either normal human T cells or Jurkat Lymphoma, a characteristic pattern of bands was observed by immune complex kinase assays. Dr. Bierer reported that PMA pretreatment of Jurkat cells blocked the association of PI3 kinase with CD28, but did not inhibit the IL2 response of these cells; this result was different from that reported by Carl June. The roles of PKC and PI3 kinase in CD28 signalling warrant additional experimental study.
It has been widely proposed that costimulus is most critical for induction of cytokines, especially IL-2, and for proliferation in T cells, but that other events such as the induction of IL-2 receptor are relatively independent of costimulatory requirement. Dr. Hiromi Fujiwara presented data indicating that induction of IL-2 receptor expression by naive T cells is dependent upon costimulation. In the absence of antigen presenting cells, naive T cells did not respond to alloantigen in mixed lymphocyie culture. Response was restored by anti-CD28 antibody but not by exogenous IL-2. Under these conditions, it was demonstrated that a CD28 signal allowed expression of IL-2 receptor as well as IL-2 production.
III. The third meeting session dealt with receptor-ligand interactions in B lyrnphocyte activation.
Dr. Tony DeFranco discussed the role of syk in tyrosine in phosphorylation induced by the B cell antigen receptor. He described the function of the B cell receptor (BCR) in B cell activation, B cell inactivation, and B cell development. Engagement or cross linking of the BCR results in tyrosine phosphorylation of multiple targeted substrates including components of the BCR complex, tyrosine kinases, enzymes that induce second messengers, regulators of ras and small molecular weight G proteins, as well as other targets such as HS1. Dr. DeFranco described experiments using truncated and chimeric Ig alpha or beta chains to assess the association of kinases with BCR components and their function in signal transduction.
Dr. Nobuo Sakaguchi compared the activation of mature and immature B cells. Using neonatal or adult spleen B cells or cell lines of models of mature and immature B cells, Dr. Sakaguchi demonstrated that cross linking of sIgM in mature B cells leads to proliferation, whereas in immature B cells it leads to apoptotic death. Antibodies were made to two newly identified proteins, sp52 and sp160, which are involved in activation and may play a role in B cell signal transduction.
Examples of positive and negative signal cooperativity were discussed by Dr. John Cambier Positive cooperation with the BCR and CD19/21 was described. This positive cooperativity has many parallels to that seen in signalling of T cells through the TcR and CD28. Negative cooperativity occurs between the BCR and FcgRII. Dr. Cambier described additional studies characterizing the structural requirements for positive and negative cooperativity and the signal transduction events which appear to mediate these processes.
IV. The fourth session of this meeting was dedicated to studies of signal transduction of cyiokine signalling.
Dr. Kazuo Sugamura described the involvement of a common gamma chain for multiple cytokine receptors. A mutation In the gamma chain of the IL-2 receptor has been demonstrated in patients with a severe immunodeficiency syndrome. It has subsequently been demonstrated that a common ganma chain plays a role in receptors for IL-2, IL-4, IL-7, IL-9, and IL-15, all in combination with distinct partner chains which account for the cytokine specificity of each receptor. An analysis of the common gamma chain by deletional mutation has demonstrated the existence of distinct pathways for regulation of c-mos and for c-fos for c-jun.
Dr. John O’Shea discussed the role of Janus family tyrosine kinases in cytokine signal transduction. A family of JAK kinases has now been described, with differential expression in monocytes, NK, T, and B cells. It has also been demonstrated that individual cytokines activate different members of the JAK family. It has been demonstrated, for example, that mutations in the cytokine receptor common gamma chain correlate with an inability of JAK 3 to associate with this receptor chain.
The common theme in this meeting was an analysis of receptor-mediated signal transduction events which play a role in the immune response. These events are triggered through antigen-specific receptors on T or B lymphocytes, through co-stimulatory ligand interactions, and through cytokine-receptor interactions. These pathways have in common the existence of complex cascades of events mediated by multi-chain receptors and associated molecules, which in turn trigger a series of cyioplasmic and nuclear events. The appreciation of complexity in this system is rapidly increasing, providing a more complete understanding of its biochemical basis as well as the elucidation of potential points for experimental or therapeutic intervention.
PARTICIPANTS
USA
Dr. Paul M Allen
Dept. of Pathology
Washington University School of Medicine
660 S Euclid
St. Louis MO 63110
Tel: (314) 362-8758
Fax: (314) 362-8888
Dr. Barbara E. Bierer
Harvard Medical School
Dana Farber Cancer Institute
44 Binney St., Rm. 1610B
Boston, MA 02115-6084
Tel: (617) 632-3536
Fax: (617) 632-5144
Dr. Jeffrey Bluestone
University of Chicago
Box 424
5841 S. Maryland Ave.
Chicago, IL 60637
Tel: (312) 702-0401
Fax: (312) 702-3701
Dr. John Cambier
Dept. of Pediatrics
National Jewish Center for Immunology and Respiratory Medicine
1400 Jackson St.
Denver, CO 80206
Tel: (303)398-1325
Fax: (303) 398-1396
Dr. Anthony L. DeFranco
Dept. of Microbiology and Immunology
University of California
Box 0552
San Francisco, CA 94143
Tel: (415) 476-5488
Fax: (415) 476-6185
Dr. Laurie H. Glimcher
Dept. of Cancer Biology
Harvard School of Public Health
Bldg. 1, Rm. 707
665 Huntington Ave.
Boston, MA 02115
Tel: (617) 432-0622
Fax: (617) 432-0084
Dr. Richard Hodes
Bldg. 10, Rm. 4B17
National Institutes of Health
Bethesda, MD 20892
Tel: (301)496-3129
Fax: (301) 496-0887
HODESR%/oNIHNIA31.BITNET@CU.NIH.GOV
Dr. Carl H. June
Immunobiology and Transplantation Dept.
Naval Medical Research Institute
Mail Stop 44
8901 Wisconsin Ave.
Bethesda, MD 20814-5055
Tel: (301)295-1122
Fax: (301) 295-6857
John O’Shea
Biologic Response Modifiers Program-NCI
Frederick Cancer Research Center
Frederick, MD
Tel: (301)496-0633
Fax: (301) 402-0012
Dr. Lawrence E. Samelson
CBMB, NICHD
National Institutes of Health
9000 Rockville Pike
Bethesda, MD 20892
Tel: (301)496-5216
Fax: (301) 402-0078
JAPAN
Dr. Hiromi Fujiwara
Department of Oncology
Biomedical Research Center, Osaka University Medical School
Osaka
Dr. Kazuo Sugamura
Department of Microbiology
Tohoku University Medical School
Sendai
Dr. Takashi Saitoh
Division of Molecular Genetics
Center for Biomedical Science, Chiba University Medical School
Chiba
Dr. Toshio Nikaidoh
Department of Obstetrics and Gynecology
Shinshu University Medical School
Matsumoto
Dr. Nobuo Sakaguchi
Department of Immunology
Tottori University Medical School
Yonago
Dr. Takeshi Watanabe
Department of Molecular Immunology
Medical Institute of Bioregulation, Kyushu University
Fukuoka
Dr. Yasuji Nishimura
Division of Immunogenetics
Center for Neuroscience and Immunology, Kumamoto University
Kumamoto
(3) Seminar on “Differentiation, Growth and Death of Cells”
Oakland, California, February 20-21, 1995
A meeting focused on the mechanisms of cell growth and death, in relation to the development of the immune repertoire, was held in Oakland, CA, between February 20 and 21, 1995, and was organized by Drs. Takehiko Sasazuki and Dinah Singer. There were 7 Japanese participants and 9 from the United States.
Mechanisms of thymic selection and the role of cell death in this process were discussed by three of the speakers. Dr. Takehiko Sasazuki presented recent studies that suggest that the development of single positive CD4+ and CD8+ thymocytes requires a two-step process of positive selection. Transgenic mice expressing human DR only in the thymus and a clonal T cell receptor (TcR), 2B4 (which recognizes peptide presented by the DR heterodimer), fall to positively select. This suggests that thymic expression of MHC is not sufficient to positively select mature cell. Dr. Sasazuki proposed that double negative (DN) thymocytes are initially selected on bone marrow derived cells, permitting maturation to the double positive (DP) stage which then depends on thymic epithelium for further positive selection. Selection of the DN thymocyte appears to require the expression of a TcR complex containing the pT precursor. Thus, in contrast to the TcR transgenic, MHC double knock-out mice expressing DR only in the thymus are capable of generating CD4+ single positive (SP) cells. Similarly, a transgenic mouse containing only the 2B4 TcR chain also undergoes positive selection. From these studies, Dr. Sasazuki concluded that the first step in positive selection involved the interaction of an immature TcR complex on a DN thymocyte with a natural, non-MHC ligand, on the bone marrow derived cell to generate a DP thymocyte. Subsequently, the mature TcR on the DP thymocyte undergoes positive selection by interaction with MHC on a thymic epithelial cell.
Dr. Jonathan Sprent continued the discussion of thymic selection, focusing on mechanisms of negative selection. Thymocytes undergo one of three possible fates: death by neglect, positive selection, or negative selection. The process of death by neglect was shown to occur largely in the cortex, as visualized by a method for detecting DNA strand scissions in dying cells. Thymocytes engulfed by macrophages are replete in the normal cortex, but absent from the medulla.. Negative selection is not clearly restricted to a single thymic location. Thus, negative selection of TcR-transgenic thymocytes in the presence of exogenous antigen occurs in the cortex, as assayed by the staining technique. In contrast, negative selection mediated by endogenous superantigen is restricted to the medulla.
Dr. Sprent also presented evidence that multiple signalling events participate in negative selection. Peptide presentation by the MHC class I molecule, Ld, expressed on the surface of Drosophila cells elicits apoptosis in the responder T cell line. Coexpression of B7 or B7+ICAM with Ld augments apoptosis. Surprisingly, coexpression of only ICAM with Ld inhibits apoptosis. The data suggest that costimulatory molecules regulate negative selection. Dr. Sprent proposed that the presence of ICAM in the thymic cortex, where there is no B7 expression, leads to positive selection, whereas negative selection predominates in the medulla, where both ICAM and B7 are expressed.
Dr. Jonathan Ashwell continued the discussion of thymic selection, presenting studies on the role of steroids in positive selection. Early studies from Dr. Ashwell’s lab demonstrated that whereas steroid and antigen separately induce apoptosis, in combination their effects are antagonistic. Dr. Ashwell has proposed that this antagonism is a natural regulatory mechanism for selection. The model he set forth is that all TcR signalling of double positive (DP) thymocytes leads to death, but a steroid signal can antagonize the death signal, leading to positive selection. In support of this model, he showed evidence that thymic nurse cells synthesize steroids. Blocking of steroid synthesis augments thymocyte death. Using transgenic mice expressing a receptor to the male HY antigen, he demonstrated that inhibition of steroid expression enhances negative selection, and reverses positive selection, but has no effect in the absence of selection. Further, a transgenic mouse with an antisense glucocorticoid receptor gene expresses reduced level of receptor in the thymus and spleen. In the thymus of these mice, there is increased cell death, possibly resulting from enhanced negative selection. This effect occurs at 14-15d of development, concurrent with the appearance of the preTcR chain, leading to the proposal that during normal development a death signal to double negative thymocytes is mediated by a natural ligand binding the surrogate TcR and that this signal is reversed by steroids. Thus, steroids play two roles in thymic development: l) regulating death at the double negative thymocyte “check point” and 2) regulating antigen stimulated cell death of double positive thymocytes.
The mechanism of cell death in peripheral lymphocytes was discussed by four speakers. Dr. Takeshi Watanabe began a discussion of apoptosis in peripheral lymphocytes by discussing the characterization of the structure and function of a gene, HS1, that appears to be involved in triggering apoptosis in lymphocytes. The HS1 gene product is a helix-loop-helix protein, with the characteristics of a DNA binding protein. It is not expressed in non-lymphoid cells, but is expressed in B220+IgM+ cells and CD4+ single positives, but not in double positive thymocytes. Following stimulation of B cells with anti-IgM, HS1 associates with a large number of cellular kinases (including lyn, fyn, blk, lck, hck, and src), becomes phosphorylated, and appears to translocate from the cytoplasm to the nucleus. The role of HS1 in apoptosis was demonstrated by the finding that mutation of HS1 in WEHI231 renders them resistant to anti-IgM induced cell death. Transfection of wild type HS1 into the mutant cell line restores susceptibility. Furthermore, B cells from an HS1-knock-out mouse are resistant to anti-IgM induced deletion; double positive thymocytes are also resistant to antigen-induced deletion In mice carrying both the HY TcR and the HS1 knock-out, the female thymocyte profile is normal, whereas negative selection fails to occur in the males, implicating HS1 in negative selection. Dr. Watanabe proposes that HS1 plays a critical role in antigen-mediated ceil death, both in the thymus and the periphery.
Dr. Henkart discussed the molecular mechanisms of programmed cell death in peripheral T lymphocytes. Cross-linking of the TcR triggers Ca+2-dependent PCD, which is inhibited by calpain protease inhibitors. Since these inhibitors do not block steroid-induced PCD, Dr. Henkart concludes that TcR-induced PCD has unique protease requirements. He proposed that TcR cross-linking induces calpain cleavage of spectrin, which leads to apoptosis. He further demonstrated that TcR cross-linking increases the expression of both Fas and Fas ligand; protease inhibitors prevent this increase. Consistent with his finding that calpain expression is high in mature T cells and low in double positive thymocytes, Dr. Henkart reported that protease inhibitors target mature, meduallary single positive thymocytes and peripheral T cells, but do not protect immature double positive thymocytes. Dr. Henkart also addressed the question of the cellular mechanism of TcR-induced PCD, which could either be mediated in cis directly by an antigen presenting cell or be mediated in trans through both an APC and a T-T interaction. Since the extent of cell death is depedent on initial cell density, Dr. Henkart concluded that death occurs in trans, which is also consistent with the induction of Fas and Fas ligand by TcR induction. Dr. Henkart further discussed a potentially important application of these findings to the therapy of HIV-infected individuals. Lymphocytes from HIV+ donors, which have lost their recall antigen response, show markedly increased cell death upon stimulation, in both CD4 and CD8 populations. However, protease inhibitors are capable of protecting these cells from death, and significantly restoring proliferative responses.
Dr. Goldsby discussed the role of oxygen in cell death, arguing that low oxygen triggers signal transduction pathways that result in death in peripheral T cells. Thus, PMA/Ca ionophore-induced death of a T cell hybridoma is blocked by the removal of oxygen, although the induction of a “death” gene, nur77, is not. However, removal of oxygen does not protect from steroid-induced death, indicating that it functions through a distinct pathway. Dr. Goldsby proposed an oxygen-death pathway, mediated by oxidants. Consistent with this view, N-acetylcysteine, a pharmacological reducing agent, blocks PMA/Ca ionophore-mediated cell death. Finally, he proposed that the oxygen-death pathway may play a role in negative selection. Thus, in thymic organ culture, addition of the superantigen, SEB, results in deletion. However, addition of N-acetylcysteine with the SEB diminishes thymocyte death.
Finally, Dr. Tsujimoto presented studies on the role of Bcl2 in protecting from oxygen radical mediated death. Mice deficient In Bcl2 have elevated levels of reactive oxygen species. However, Dr. Tsujimoto proposed that a Bcl2-independent pathway to cell death also exists, since Bcl2 does not protect from death induced by increasing reactive oxygen species by diethyl maleate, which binds glutathione. Anoxia also leads to cell death, which Is also protected by Bcl2 or BclXL. Dr. Tsujimoto proposed that the cysteine protease ICE may be involved in triggenng hypoxic cell death, since introduction of a ICE-inhibitor protects cells. Three speakers discussed the regulation of cell growth. Dr. Okayama described a novel G1 cell cycle checkpoint. Three cell cycle checkpoints exist, in S, G2, and G1. In G2, DNA damage blocks the activity of CDC25 phosphatase that normally dephosphorylates CDC2 kinase; this blocks further progression through G2. In G1, a p21-dependent check point has been described which is disrupted by damage to p53. Dr. Okayama described a second G2 checkpoint independent of p21 which is regulated by a novel CDC25A phosphatase, homologous to CDC25.
Dr. Okayama presented evidence that the substrate of the CDC25A phosphatase is CDK4. Mutants of CDK4 in which the phosphorylation site has been removed are unable to arrest at G1 after UV or X-ray irradiation; such mutants accumulate chromosomeal aberrations and have elevated cell death. Anti-CDC25A normally blocks entry of cells into S phase; however it has no effect on the CDK4 mutant. Based on these studies, Dr. Okayama suggested that CDK4 is phosphorylated in G1 and that CDC25A dephosphorylation is necessary for entry into S. Further, Dr. Okayama argued that this checkpoint functions independently of p53/21.
Dr. Tokino then described the identification and characterization of downstream targets of p53. p53 is a transcription factor known to play a role In both apoptosis and G1 arrest, presumably through the activation of target genes. By subtractive hybridization of p53 induced and uninduced glioblastoma cells, Dr. Tokino isolated a downstream target, waf1 (p21) which he showed inhibits cell cycle by complexing with cyclin/CDK. Introduction of waf1 into tumors reduces cell growth of SW480 cells as efficiently as p53. Analysis of the waf1 gene upstream regulatory region revealed the presence of p53 binding sites; deletion analysis suggested that one intact site is sufficient for p53 binding and activity. Since p53 is presumed to have multiple downstream targets, Dr. Tokino has developed a novel screening approach to identify all p53 targets in the genome. The screening approach is based on a yeast activation system, in which fragments of human DNA are inserted upstream of a gal promoter and reporter gene; activation of the promoter requires binding of p53 to the human DNA fragment. Using this approach, Dr. Tokino has isolated and sequenced a large number of p53 binding DNA fragments, called p53 tagged sites, which are now being further analysed for their regulation by p53.
Cytokine signalling for growth and differentiation was discussed by Dr. Kishimoto, who focused on studies of NF-Il6 and the receptor subunit, gp130, common to many cytokine receptors. Binding of the Il6 receptor induces the dimerization of gp130, which in turn triggers association of stat3 and jak1 or jak2 kinases, that ultimately result in the activation of factors that enter the nucieus and either induce growth inhibition or differentiation. Dr. Kishimoto presented evidence that distinct domains of gp130 bind stat3 and jak, and that distinct domains also transduce growth regulation signals or differentiation signals. Expression of gp130 is ubiquitous and deletion of the gp130 gene through homologous recombination is lethal. The major morphological defect in homozygous animals is abnormal development of the heart; erythroid differentiation is also abnormal, with many nucleated cells. Conversely, a double transgenic of Il6 and soluble Il6 receptor induces extensive dimerization of gp130 and cardiac hypermyotrophy. He further described NFIl6 knock-out mice, which survive but are runted and highly susceptible to infections by listeria and salmonella. Since both of these infections are handled by the macrophage, Dr. Kishimoto concluded that the macrophage bacteriocidal pathway is lost in these NFIl6-deficient mice. However, no levels were found to be normal, as were TNF-!
!!and IFN-!
!!. Therefore, Dr. Kishimoto proposes that NFIl6 is involved in a novel major bacteriocidal pathway. The next four speakers discussed the role of Fas and Fas ligand in cell death. Dr. Green presented the hypothesis that although there are multiple pathways to trigger a death signal, there is a single, common mechanism for apopotosis. Further, he argued that this common mechanism is necessary not only to mediate cell death, but also for cell viability. He then discussed the possibility that this common pathway includes vav and ras. Cross-linking of the TcR by anti-CD3 results in the induction of both Fas and FasL, and results in cell death by apoptosis. Unlike Dr. Henkart, Dr. Green reported observing no dependence of cell death on ceil density. Induction of Fas results in the activation of sphingomyelinase, which increases cellular ceramide levels, leading to cell death. Dr. Green described studies on cells derived from patients with Niemann-Pick disease. These cells have low to no sphingomyelinase and accumulate sphingomyelin, but do express Fas. Ligation of Fas in these cells does not Induce cell death. However, following transient introduction of a sphingomyelln cDNA, the Niemann-Pick cells become transiently sensitive to anti-Fas mediated apoptosis, leading Dr. Green to conclude that sphingomyelinase is required for Fas-induced death. Dr. Green proposed a model in which Fas induces increased levels of sphingomyelinase, leading to elevated ceramide levels which activate vav and ras, leading to apoptosis. Consistent with this model, it was shown that a dominant negative mutant of ras inhibits apoptosis. Finally, initial results in developing an in vitro, Inducible model of apoptosis were described
Dr.. Nagata discussed the cloning of both Fas and FasL. FasL is a membrane protein that induces apoptosis in Fas-expressing cells. It is expressed mostly in activated T cells, including CTL, TH1 and TH2 cells, but is not expressed in B cells, plasmacytomas, or thymic macrophages. Activated human PBL secrete soluble FasL as a trimer, capable of cross-1inking Fas to cause apoptosis. Since an inhibitor of Interleukin-1 converting enzyme (ICE) also inhibits Fas-mediated cell death, Dr. Nagata suggested that Fas works through ICE. The lpr mouse, which is defective in Fas, accumulates double negative T cells in the lymph node and spleen; a FasFc fusion can induce death in these cells. However, Fas is not involved in thymocyte death, since thymocytes develop normally in these mice and in Fas knock-out mice.
Dr. L ynch presented a model for the role of Fas and FasL in the homeostatic regulation of the normal immune response, and discussed how dysregulation of the Fas apoptotrc pathway might contribute to autoimmune diseases and HIV-induced depletion of CD4+ T cells. Although ligation of Fas leads to apoptosis in cultured cell lines, its cross-linking on freshly isolated T cells leads to cellular activation and proliferation. Activation of T cells induces Fas, leading to the suggestion that Th1 cells and APC interact to provide a costimulatory signal through Fas. FasL is also induced by TcR ligation, but only on Th1 cells, not on Th2. Dr. Lynch proposed that the role of FasL is to remove activated CTL, through cell-cell contact. Thus, Fas acts during T cell activation, while FasL acts to restore homeostasis. Dr. Lynch also discussed the development of peptide inhibitors of FasL as therapeutics in AIDS based on the notion that CD4+ cells undergo apoptosis as a result of chronic activation.
Dr. Mountz discussed the signal transduction pathway of Fas and the possible role of soluble Fas in autoimmune disease. Fas RNA can be alternatively spliced to yield both membrane and soluble forms of Fas. Soluble Fas is elevated in SLE patients; mice injected with soluble Fas develop elevated numbers of T and B cells in the spleen. Dr. Mountz suggested that soluble Fas blocks normal Fas-mediated apoptosis that eliminates autoreactive T cells. He also presented evidence that Fas signalling results in protein tyrosine dephosphorylation by the enzyme hematopoeitic cell protein-tyrosine phosphatase (HCP). Fas-apoptosis MOLT-4 cells do not contain HCP; introduction of HCP into MOLT-4 cells renders them sensitive to Fas-mediated cell death. Further, treatment of HCP-defective mev mutant mice with anti-Fas antibody induces cell proliferation, not cell death. Thus, like Dr. Lynch, Dr. Mountz suggested that Fas activation leads to two distinct signals: one for cell proliferation and one for cell death. He also discussed the role of the Nur77 gene product in delivering a cell death signal during negative selection in the thymus.
Dr. Osborne concluded the meeting by discussing genes that regulated cell death in lymphocytes. Whereas p53 is required for induction of death by ionizing radiation, nur77 is induced in TcR-mediated cell death. The nur77 gene encodes a Zn+ finger transcription factor that is a member of the steroid hormone receptor family, although it has no known ligand. Nur77 gene expression is induced in both activated and dying cells, but is only transient in the former and sustained in the latter. Analysis of the upstream flanking regions of the nur77 gene revealed the presence of sites for various known transcription factors, including NF-AT. In addition, nur 77 autogenously regulates its transcription through a binding site located between -288 and -248bp. Dr. Osborne also discussed the role of phosphorylation in regulating nur77 activity. A mutant hybridoma, resistant to TcR-mediated cell death transcribes the nur77 gene normally, but does not phosphorylate the nur77 protein. Since in activated cells nur77 is also not phosphorylated, Dr. Osborne speculated that the phosphorylation state of nur77 may determine the cellular outcome of death or proliferation. Dr. Osborne also raised the intriguing question of the relationship between TcR-mediated and steroid-mediated death, since nur77 and the glucocorticosteroid receptor are both members of the steroid hormone receptor family.
UNITED STATES
Dr. Dinah Singer Organizer
NIH
Dr. Jonathan Ashwell
NIH
Dr. Richard Goldsby
Amherst College
Dr. Douglas Green
La Jolla Institute of Allergy and Immunology
Dr. Pierre Henkart
NIH
Dr. David Lynch
Immunex Corp.
Dr. John Mountz
University of Alabama at Birnxingham
Dr. Barbara Osborne
University of Massachusetts
Dr. Jonathan Sprent
Scripps Research Institute
JAPAN
Dr. Takehiko Saszuki
Department of Genetics
Medical Institute of Bioregulation, Kyushu University
3-1-1, Maidashi Higashi-ku, Fukuoka 812
Dr. Takeshi Watanabe
Department of Molecular Immunology
Medical Institute of Bioregulation, Kyushu University
3-1-1 Maidashi, Higashi-ku, Fukuoka 812
Dr. Takashi Tokino
Japanese Foundation for Cancer Research
1-37-1 Kamiikebukuro, Toshima-ku, Tokyo 170
Dr. Yoshihide Tsujimoto
Biomedical Research Center
Osaka University Medical School
2-2 Yamadaoka, Suita 565
Dr. Tadamitsu Kishimoto
Professor, Osaka University Medical School
2-2 Yamadaoka, Suita 565
Dr. Hiroto Okayama
Professor, Faculty of Medincine
University of Tokyo 7-3-1 Hongo, Bunkyo-ku, Tokyo 113
Dr. Shigekazu Nagata
Osaka Bioscience Institute
6-2-4 Furuedai, Suita 565