REPORTS ON SEMINARS

(1) “Molecular Mechanisms of Recognition and Regulation”
The US-Japan Cancer Cooperative meeting entitled “Molecular Mechanisms of Recognition and Regulation” was organized by Drs. Dinah S. Singer and Takehiko Sasazuki, and met February 12-13, 1996, in Maui, Hawaii. The meeting focused on signal transduction mechanisms and transcriptional regulation of genes involved in the immune response. There were 8 participants from the United States and 7 participants from Japan.
Four of the participants discussed the role of specific transcription factors in regulation of genes involved in the immune response. Dr. Grosschedl discussed the role of protein-protein interactions in synergistic activation of gene transcription, using the TCR!!!gene enhancer as a prototype. The necessity of proper orientation of the transcription factors LEF, ATF2/CREB, PEPB2!!!and ETS binding sites led to the proposal that bending of DNA by LEF allows all 4 proteins to interact. Dr. Grosschedl also reported that the DNA binding domain of LEF binds to a coactivator, currently called LIP, whose role in activation is being studied. The role of LEF-1 in T cell differentiation was examined through the production of knockout mice, which displayed no gross abnormalities in T cell development, but did exhibit a marked decrease in CD5 + B cells. This led to the proposal that there may be redundancy within the system. This proposal was confirmed in LEF-1/TCF-1 double knockout mice, which generate no double positive CD4 + CD8 + Tcells. Rather, they accumulate CD8 + precursors, leading to the speculation that elimination of these transcription factors leads to a block in this development step.
Dr. Ishii presented structural and functional studies of the transcription factor, c-Myb which is a proto-oncogene that plays an important role in the control of proliferation of hematopoietic progenitor cells. . He reported on NMR studies that established that the DNA binding domain of the protein consists of 3 repeats, only two of which (R2 and R3) directly bind DNA. The third repeat, R1, is involved in regulating binding by stabilizing the interaction. C-Myb is bi-functional, acting both as a transcriptional activator and repressor. Activation of the mim gene is achieved through interaction of the C-terminal domain of c-Myb with the mediator, CBP, which occurs independently of phosphorylation; this interaction is blocked by E1A. C-Myb also acts as a repressor of the c-erbB-2 promoter by binding to a site that overlaps the TATA box, inhibiting binding of TFIID. Oncogenic activation of c-Myb can occur either by truncation of the C-terminus which eliminates a negative regulatory domain, allowing enhanced expression of target genes, or by removal of the N-terminal R1 which may reduce binding of c-Myb to the TATA box element, allowing enhanced expression of genes involved in proliferation control. A novel bacterial system for the expression of soluble protein was also described.
Dr. Smale has been studying the TdT gene as a model in transcriptional regulation of T cell differentiation. Within the TdT enhancer region, he has identified a DNA sequence element, D, that controls cell type specific transcription. Two factors bind to D: Ikaros and elf-1. Although implicated in the development of T and B cell progenitors, Ikaros is found in sca1 + stem cells and does not activate TdT gene expression. Rather, elf-1 appears to be the primary activator of TdT. Dr. Smale speculated that Ikaros plays a regulatory role in repressing TdT gene expression in mature T lymphocytes. Ikaros is multiply spliced within internal domains, giving rise to multiple isoforms which differ in their ability to bind DNA. These isoforms multimerize, giving rise to complexes which reach sizes of 600-700kD, when Ikaros is over-expressed. The function of these complexes is unknown, but they appear to associate with at least two other cellular proteins, p30 and p70. Repression of MHC class I genes by the HIV product, Tat, was discussed by Dr. Singer, who discussed studies that demonstrated that Tat represses transcription by acting on the basal class I promoter. This promoter was shown to consist of a TATA box, and Inr, and a third promoter element; none of these elements is the direct target of Tat. It was postulated that Tat may repress through interactions with the transcription initiation complex. Structural analysis of Tat revealed that its two functional activities-HIV LTR transactivation and class I repression-are mediated by discrete, but overlapping, structural domains. In support of this novel model of separable Tat functions, it was found that Tat mediates repression but not transactivation in murine fibroblasts. It was concluded that Tat’s effect is determined by target promoter architecture, resulting in selective repression of host genes and activation of the viral promoter.
The interface between signal transduction and transcriptional regulation was addressed by three of the participants. Dr. Calame reported on regulation of expression of c-myc, which is a proto-oncogene involved in cell proliferation and apoptosis; deregulated levels are often associated with malignant transformation. Positive regulation of c-myc transcription is mediated by the v-Abl protein, a tyrosine kinase which appears to activate c-myc transcription by initiating a phosphorylation cascade that involved both ras and raf. The target DNA sequence for this signal transduction pathway is the E2F binding site; multimerized E2F binding sites upstream of a heterologous TATA box respond to v-Abl activation. v-Abl induces the appearance of E2F binding complexes which contain DP1, p107, CDK2, and cyclin A. Thus the mechanism of v-Abl transformation can be understood by its effects on cell cycle dependent genes, many of which contain E2F sites. Negative regulation of c-myc in differentiated plasmacytomas is associated with the binding of a transcription factor, Blimp-1 which is expressed at high levels in terminally differentiated cells and represses c-myc transcription.
Dr. Taki discussed the molecular mechanisms of the response to interferon, which activates two transcription factors, IRF-1 and p48, both of which bind to a common DNA sequence element, the ISRE. From analyses of knockout mice of genes encoding these factors, it is concluded that the two factors have distinct effects despite the common binding site, which are postulated to be due to differential protein/protein interactions. For example, the iNOS gene is responsive to IRF-1, but not to p48. Thus, the IRF-1 knockout mouse is sensitive to listeria and malaria, while the p48-/- mouse is resistant. Dr. Taki also implicated IRF-1 in the regulation of apoptosis in the spleen, as distinct from the p53-dependent thymic apoptosis, based on the observation that, following DNA damage, spleen cell survival is better in IRF1-/- animals than in wild-type mice, whereas there is no difference in thymocyte survival. He speculated that this resistance to apoptosis may reflect a faulty induction of an ICE family member, rather than ICE itself since the ICE knockout mouse displays normal apoptosis. Dr. Taki also discussed the role of IRF2 in B cell development: both the IRF2 knockout mouse and the IRF1 transgenic mouse have defective B cell lineage development, also suggesting that they play antagonistic roles.
Dr. Akira described a series of studies aimed at elucidating intermediates in the signal transduction pathway, using the knockout technology. NF-IL6, a transcription factor in the C/EBP family, binds to a sequence found in many cytokine genes, and is stimulated by IL6, TNF, LPS and IL1. Homozygous elimination of NF-IL6 results high prenatal mortality; survivors are highly susceptible to bacterial infection. In NF-IL6 knockout mice, unlike wild type mice, intracellular survival of listeria is observed, despite the finding that induction of interferon and TNF are normal, as is iNOS, which is normally involved in NO-mediated inactivation of listeria. Dr. Akira speculated that NF-IL6 induces an unknown gene that is required for killing of listeria. A C/EBP knockout mouse was also described, which dies at 7-8 hours after birth and shows no glycogen storage in the liver. Dr. Akira then turned to studies in macrophage differentiation, citing evidence that Stat3 plays an important role in their developmental pathway: transfection of Stat3 mutants block both differentiation and apoptosis of an IL6-responsive myelogenous leukemia, whereas wild type Stat3 accelerated the IL6 response. Homozygous Stat3 knockouts were found to be embryonic lethal. Finally, he reported on a Stat6 knockout mouse, whose homozygous phenotype was normal with respect to lymphoid populations. However, CD23 and class II were not induced in response to IL4 treatment in these animals, although the LPS stimulation was normal. Th2-type cytokine production (IL4, IL5 and IL10) was reduced, although Th1 cytokines (interferon) were normal. It was concluded that the Stat6 knockout is defective in the Th2 response. In B cells, IL4-dependent class switching was affected, showing a reduced level of IgG1 and an absence of IgE. The Stat6 knockout mouse also showed an impaired proliferative response to IL4 of small B, but a normal response to LPS. Dr. Akira concluded by postulating that IL4 signals through both a 4PS/IRS2 pathway to proliferation, or a Jak/Stat6 pathway to class I and CD23 induction.
Mechanisms of signal transduction in T cells were discussed by Drs. O’Shea and Schreiber. Dr. O’Shea reviewed the association of the Jak3 kinase with the common!!!chain that comprises part of the receptors for many cytokines, including IL2,4,7,9 and 15. Following ligation of the cytokine receptor, Jak 3 differentially phosphorylates a variety of substrates, depending on the receptor with which it is associated: Jak3 on IL2 and 15, but not IL4, phosphorylates Stat5, while Stat6 is phosphorylated by Jak3 in response to IL4. Dr. O’Shea reported that the large docking proteins, IRS-1 and IRS-2, are also substrates for Jak3 in response to IL2,4,7,9, and 15, and proposed a model in which cytokine activation results in the association of Jak3 with the IRS complex and its subsequent phosphorylation. Also associated with the IRS are grb, sos and PI3 kinase which may activate the mapk pathway. However, Jak3 function appears to be most important for the phosphorylation of Stats and proliferation. Dr. O’Shea also discussed the identification of 3 patients with Jak3 mutations, who presented with autosomal immunodeficiencies, and were found to lack Jak3 protein due to deletions within the protein.
Dr. Schreiber discussed studies addressing the biological function and specificity of Stat1 and Jak1. Interferon triggers the phosphorylation and nuclear translocation of Stat1 to the nucleus, resulting in the activation of IFN-specific genes. However, other cytokines (IL6, IL10, CSF, EGF, growth hormone, angiotensin, thrombopoeitin) are also known to activate Stat1, raising the question of how specificity to IFN is achieved. Analysis of Stat1 knockout mice revealed that they develop normally, with normal lymphoid compartments, and are able to reproduce. Although the knockout mice produces a truncated Stat1 protein and expresses normal basal levels of immune cell markers, it does not respond to IFN and is deficient in IFR1, gbp-1, CIITA, and C3. Similarly, these mice do not induce iNOS or anti-VSV activity in response to IFN + LPS. Indeed, the Stat1 knockout mice are 10⁶ times more susceptible to infection than normal mice and die rapidly following infection with either listeria or VSV. In contrast, the knockout mice show a normal response to other cytokines such as growth hormone, EGF and IL10, leading to the conclusion that Stat1 is specific for the IFN response. In contrast, the Jak1 knockout mouse which Dr. Schreiber described exhibits a perinatal lethal phenotype, with a very reduced thymus. Embryonic fibroblasts derived from the Jak1 knockout do not respond to IFN and are impaired in their IL6 signaling. Thus, unlike Stat1 which is restricted to the IFN pathway, Jak-1 serves multiple cytokine pathways.
Three of the participants presented studies on signal transduction and transcriptional regulation in B cell development. IL5 is a major cytokine involved in the differentiation of plasma cells, CD5 + B cells, and in the myeloid lineage; it binds to an receptor consisting of an!!!chain and a common!!!chain. Dr. Takatsu discussed studies on the biological effects of IL5, using both IL5 transgenic mice and IL5R!!!knockout mice. The IL5 transgenic mouse displays elevated levels of IL5 transcripts in bone marrow, spleen, and liver, elevated serum levels of IgM, A and E, and increased number of CD5 + B cells in the spleen, eosinophilic infiltrates in liver and muscle. Although elevated levels of autoantibody are produced, there is no overt autoimmune disease in these animals. Treatment of the transgenic mice with either!!!-IL5 or!!!-IL5R suppresses these manifestations. The IL5R knockout mouse appears to be developmentally normal; PEC B cells express B220 but not IL5R. Serum levels of IgM and IgG3 are about one-third to one-half normal levels, although IgG1 and IgA are normal. A decrease in the number of Mac1 + IgM + B cells in these animals suggests that IL5 contributes in part to B cell development. In peripheral blood populations, the number of eosinophils is markedly reduced, indicating IL5 plays a critical role in the induction of eosinophils. In response to infection by nematode, the IL5 transgenic is relatively resistant, whereas the IL5R knockout is relatively sensitive, suggesting the IL5 is protective as a result of elevating eosinophils. Dr. Takatsu then discussed the signal transduction pathway triggered by IL5, which activates a cascade of kinases that includes Bruton’s tyrosine kinase (Btk), and results in cellular proliferation. Btk is active in cells of the B and myeloid lineage, and is mutated in X-linked agammaglobulinemia (XLA) in humans and in XID mice. Btk knockout mice show an XID phenotype. Dr. Takatsu proposed that Btk serves as an intermediary between signaling by CD38, a membrane protein involved in B cell proliferation and Ca + mobilization, and Blimp-1, a transcription factor that is induced during B cell differentiation. Since CD38 and IL5 act synergistically, it is thought that CD38 activates Btk, which induces Blimp-1 and IL5R, whose binding of IL5 activates the Jak/Stat pathway.
Dr. Staudt discussed the isolation and characterization of a B cell specific gene product, Bcl-6 which is associated with a number of B cell lymphomas. Bcl6 translocation occurs in about 45% of cases, and mutations occur in 70-80% of cases. Translocations generally juxtapose the Bcl6 coding region with a novel promoter, activating expression of the protein. The protein consists of three recognizable domains: a Zn finger DNA binding domain, a poz box (a domain shared by many Zn finger domains) and a ser/pro rich domain that is the target of phosphorylation. When overexpressed, Bcl6 multimerizes through the poz domain to yield enormous, hollow spherical structures. The poz domain is able to function as a transcriptional repressor on a variety of test promoters, although its natural target is not known. Bcl6 is expressed in a variety of tissues and B cell lines; activation of B cells by various stimuli (i.e. LPS, PMA/ionophore, CD40L,!!!IgM) decreases the levels of Bcl6 RNA. Bcl6 protein is localized to the germinal centers of tonsils, lymph node and spleen, and is found primarily in B cells. Generally, germinal center B cells contain Bcl6 protein, whereas resting small B cells do not. Of interest, there is little difference in Bcl6 RNA levels between these cell types, suggesting post-translational regulation. Since memory B cells do not have Bcl6, Dr. Staudt proposes that Bcl6 is low in terminally differentiated cells. Based on the finding that Bcl6 inhibits proliferation, he argued that a translocation of the Bcl6 gene would prevent its autogenous repression, resulting in uncontrolled growth; this consistent with the fact that diffuse large cell lymphomas derive from germinal center B cells, where Bcl6 expression is high.
Dr. Watanabe presented studies of two proteins involved in signal transduction from the B cell receptor: lyn and HS1. Lyn is a src tyrosine kinase that is expressed in hematopoietic cells associated with the BCR that is phosphorylated after antigen-induced cross-linking. Phosphorylation of lyn leads to the activation of plc!!!2, cbl, shc, vav, syk and HS1, and leading to apoptosis. In lyn knockout mice, cross-linking by!!!-IgM did not result in phosphorylation of any of these proteins. B cells from these animals, which are reduced in number, did not display a proliferative response to CD40L,!!!IgM, or LPS, although they did respond to IL4. Since the bone marrow cells appear normal, with the exception that they lack CD43-B220 + or HSA + B220 + cells, it is concluded that in lyn knockout mice, the transition to mature B cells is impaired. Grossly, lyn-/- animals display splenomegaly, lymphadenopathy, and lymphoblastoid infiltrates. There is a increased number of Mac1 + IgM-secreting cells, resulting in elevated serum IgM antibodies that results in the development of glomerular nephritis, and an SLE-like disease. Dr. Watanabe concluded that abnormal lyn expression may result in autoimmune disease. Dr. Watanabe then discussed the characterization of a protein, HS-1, that is co-phosphorylated and co-precipitated with lyn. HS-1 is implicated in mediating apoptosis, since in cell lines chosen for their resistance to apoptotic death, levels of HS1 are markedly reduced, and the residual protein is no longer phosphorylated. Introduction of normal HS1 in to these variant cells restores the levels of protein, phosphorylation and apoptosis. Consistent with these in vitro observations, in HS1 knockout mice, splenic B cells are resistant to apoptosis and do not respond to!!!IgM cross-linking and splenic T cells do not respond to!!!CD3 treatment. These findings suggest that HS1 is in pathways leading to both proliferation and apoptosis.
Molecular mechanisms regulating cell death were discussed by Drs. Suda and Osborne. Dr. Osborne presented the model that cell death can be achieved by multiple pathways, many of which induce new gene expression, but that ultimately converge on a common execution phase. She described the isolation of genes whose expression is increased during cell death. One of these, nur77, encodes a member of the nuclear hormone receptor superfamily of zinc finger transcription factors. Unlike many of the members of this family, nur77 has no known ligand. It was demonstrated that nur77 is required for apoptosis in T cells following TCR crosslinking in T cell lines. However, nur77 does not play a role in cell death induced by other stimuli, such as radiation or glucocorticoid. The nur77 promoter consists of CD28 and PMA-response elements, as well as binding sites for NF-AT and nur77, the latter suggesting that the gene is autogenously regulated. Dr. Osborne also described the isolation of a cell mutant that is resistant to Ca + ionophore/PMA induced death, but sensitive to dexamethasone induced signals. In this mutant, nur77 is neither phosphorylated nor induced. Furthermore, nur77 remains in the cytosol, and does not translocate to the nucleus in response to death stimuli, leading to the speculation that both of these events are critical for the induction of death genes.
The role of p53 in thymocyte apoptosis was examined in p53 knockout mice. where it was observed that apoptosis occurs normally in the thymi of these mice in response to PMA/Ca + ionophores and dexamethasone, but not ionizing radiation. These data demonstrate that p53 is required for only some forms of apoptosis, and suggest that there are multiple independent pathways to death. Dr. Osborne concluded by speculating that all of these pathways converge on a common “execution” pathway, that may involve members of the ICE family.
Cell death mediated through the Fas/FasL pathway was the topic of Dr. Suda’s discussion. Whereas expression of Fas is ubiquitous, FasL expression is restricted; among the tissues are thymus, spleen, and lymph node. Various T and NK cell lines express FasL, whereas non-T and non-NK lines do not. Among T cells, FasL expression predominates in Th1 and Th0 cells, relative to Th2. Fas knockout mice display a lymphoproliferative disease characterized by cells of the CD3 + CD4-CD8-B220 + Thy1 + phenotype. Disease resembles the lpr defect, but is accelerated, and includes elevated serum levels of IgG, antibodies to double-stranded and single-stranded DNA. Interestingly, there appears to be a role for Fas/FasL in hepatocyte turn-over, since hepatocytes in the Fas knockout display enlarged nuclei, consistent with a more rapid turn-over. Dr. Suda speculated that Fas might be involved in regulating liver regeneration. Dr. Suda also discussed the biological function of soluble FasL, which is generated by metalloprotease cleavage. Inhibition of processing by BB2116 reduces the activity of soluble FasL killing. Patients with NK cell and T-LGL leukemias, but not with T or B cell lymphomas, AML, APL, or ALL, have detectable soluble FasL in their serum. Dr. Suda speculated that hepatotoxicity observed in leukemia patients may be due to soluble FasL in the serum.
Dr. Sasazuki presented studies on thymic selection. He developed the argument that both MHC class II and class I contribute to the determination of antibody levels. The model predicts that class II-restricted CD4 + T cells augment antibody production, whereas class I-restricted CD8 + T cells negatively regulate it. Current studies are directed at identifying T cell clones representative of each group. HLA DR and DQ transgenic mice have been produced and found to respond to HA peptide. Since both DR-transfected L cells and IA-/- mice elicit responses to viral peptides by the transgenic T cells, it was concluded that HLA DR is capable of interacting with murine CD4. In the DR!!!transgenic mice, distinct patterns of expression were observed depending on the 5’ flanking sequences in the transgene, such that DR!!!was observed either in thymic epithelial cells alone (strain 30) or in both thymic epithelial cells and bone marrow derived cells (strain 24). In crosses with a TCR transgenic, it was found that positive selection occurred in strain 24, but not strain 30, suggesting that expression of class II in the bone marrow derived population might be necessary for positive selection. However, in other experiments in H-2-negative mice, where the Dr!!!of strain 30 was expressed only in the thymus, positive selection was observed, suggesting that expression in the thymus is sufficient for positive selection. This discrepancy was resolved by the demonstration that in TcR!!!transgenics, positive selection occurred in crosses with both DR!!!24 and 30, leading to the conclusion that selection may require fixing the association between the!!! and!!!chains of the TcR, which normally occurs in the thymus, not in the periphery.

PARTICIPANTS

UNITED STATES

Dr. Kathryn Calame
c/o Dr. Bev Emerson
Salk Institute
10010 N. Torrey Pines Rd.
La Jolla, CA 92037
Tel: (619) 453-4100 X1424
E-mail: kcalame@aim.salk.edu

Permanent address:
Department of Microbiology
Columbia University
College of Physicians and Surgeons
701 W 168th St.
New York, NY 10032-2704
Tel: (212) 305-3504 Fax: (212) 305-1468

Dr. Rudolf Grosschedl
Department of Microbiology and Immunology
University of California, San Francisco
Box 0414, S447
San Francisco, CA 94143-0452
Tel: (415) 476-6954 Fax: (415) 476-8201
E-mail: rgross@itsa.ucsf.edu

Dr. Stephen Smale
5748 MacDonald Research Labs
University of California, Los Angeles
10833 Le Conte Ave.
Los Angeles, CA 90024-1662
Tel: (310) 206-4777 Fax: (310) 206-3800
E-mail: steves@hhmi.ucla.edu

Dr. Barbara Osborne
Dept. of Veterinary and Animal Sciences
304 Paige Lab
University of Massachusetts
Amherst, MA 01003
Tel: (413) 545-2427 Fax: (413) 545-6326
E-mail: osborne@vasci.umass.edu

Dr. John O’Shea
NIAMS
Building 10, Room 9N262
Bethesda, MD 20892-1360
Tel: (301) 496-6026
E-mail: osheaj@arb.niams.nih.gov

Dr. Robert Schreiber
Dept. of Pathology
Washington Univ. School of Medicine
660 Euclid Ave./Mailstop 8118
St. Louis, MO 63110-1093
Tel: (314) 362-8787 Fax: (314) 362-8888
E-mail: schreiber@immunology.wustl.edu

Dr. Dinah Singer
Experimental Immunology Branch, NCI
Bldg. 10, Room 4B-17
Bethesda, MD 20892-1360
Tel: (301) 496-9097 Fax: (301) 480-8499
E-mail: Dinah_Singer@nih.gov

Dr. Lou Staudt
Metabolism Branch, NCI
Bldg. 10, Room 4N114, NIH
Bethesda, MD 20892
(301) 496-8890
E-mail: Istaudt@alw.nih.gov

JAPAN

Dr. Shunsuke Ishii
Lab. of Molecular Genetics, RIKEN Tsukuba Center
3-1-1 Koyadai, Tsukuba, Ibaraki 305
Tel: 81-298-36-9031 Fax: 81-298-36-9030
E-mail: sishii@ rtcs1.rtc.riken.go.jp

Dr. Takehiko Sasazuki
Dept of Genetics
Medical Institute of Bioregulation
Kyushu University
3-1-1 Maidashi, Higashi-ku, Fukuoka 812-82
Tel: 81-92-641-1151 (ext 3771) Fax: 81-92-632-0150
E-mail: sasazuki@bioreg.kyushu-u.ac.jp

Dr. Takashi Suda
Dept. of Molecular Biology
Osaka Bioscience Institute
6-2-4 Furuedai, Suita, Osaka 565
Tel: 81-6-872-4850 Fax: 81-6-871-6686
E-mail: sudat@obisunl.obi.or.jp

Dr. Kyoshi Takatsu
Dept. of Immunology
Institute of Medical Science, Univ. of Tokyo
4-6-1 Shirokanedai, Minato-ku
Tokyo 108
Tel: 81-3-5449-5260 Fax: 81-3-5449-5407
E-mail: not available yet

Dr. Shinsuke Taki
Dept. of Immunology
University of Tokyo
4-6-1 Shirokanedai, Minato-ku
Tokyo 108
Tel: 81-3-5800-3220 Fax: 81-3-5689-7214

Dr. Takeshi Watanabe
Dept. of Molecular Immunology
Medical Institute of Bioregulation
Kyushu University
3-1-1 Maidashi, Higashi-ku, Fukuoka 812-82
Tel: 81-92-641-1151 (ext. 3781) Fax: 81-92-632-1499
E-mail: watanabe@bioreg.kyushi-u.ac.jp

Dr. Shizuo Akira
Dept. of Biochemistry
Hyogo College of Medicine
1-1 Mukogawa-cho, Nishinomiya
Hyogo 663
Tel/ Fax: 81-798-45-6355

Dr. Shunsuke Isii
Tsukuba Life Science Center
The Institute of Physical & Chemical Research
3-1-1 Koyadai, Tsukuba, Ibaraki 305



(2) Seminar on “Regulation of Cell Proliferation and Differentiation”
ORGANIZERS:
This Seminar on “Regulation of Cell Proliferation and Differentiation” was organized by
Dr. Robert N. Eisenman (U.S.) and
Dr. Tadatsugu Taniguchi (Japan)
March 21-23, 1996, at Hapuna Beach Prince Hotel, Hawaii

Session I: Control of the Cell cycle
Dr. Leland Hartwell began the session by describing genes that are known to be involved in genomic stability and the mechanisms underlying errors in replication and mitosis that generate polyploidly, aneuploidy and chromosome aberrations. He went on to discuss the concept of cell cycle “checkpoints”: mechanisms that monitor the orderly progression of the cell cycle and arrest progression when errors occur. One aspect of checkpoint function is adaptation, in which in the continual presence of the error or damage signal the arrest is eventually overcome and the cell proceeds into the cycle despite the presence of damage. He described an approach in S. cerevisiae in which mutants in the adaptation response can be isolated.
Dr. Yumiko Kubota described her work on the identification of the proteins that are likely to limit DNA replication so that it occurs only once per cell cycle. Such “licensing factors” would be predicted to be required for initiation of replication but inactivated or degraded once replication is ongoing. Using Xenopus laevis eggs. her laboratory has purified a protein complex with the properties of the licensing factor. The complex is required for initiation of sperm DNA replication and contains at least six proteins all belonging to the MCM family (mini-chromosome maintenance) a group a zinc finger proteins previously described in S. cerevisiae and S. pombe. These proteins are either sequestered or degraded during replication but appear to fall into two groups with respect to binding affinity and inactivation in the presence of aphidicolin. She postulates that they may act at distinct stages of DNA replication.
Dr. Charles Sherr discussed the regulatory proteins that act positively or negatively on progression through the G1 phase of the mammalian cell cycle. He focused particularly on the inhibitors (CDIs) of cyclin dependent kinase (CDK) activity and presented a model in which the cyclin D/CDK4 complex sets a threshold by titrating the p27/p21 CDI proteins and thereby permitting cyclin E/CDK2 to function in late G1. In situations where p27 CDI degradation is inhibited, such as during a cAMP block of mitogenesis in CSF-1 stimulated macrophages, both cyclin D/CDK4 and cyclin E/CDK2 are inhibited. He also described the newly discovered p19ARF protein encoded by an alternative open reading frame in the p16INK4a coding region. The synthesis of p16INK4a oscillates in the normal cell cycle and its inhibitory function is dependent on the Rb protein. Mutations in the p16/p19 gene occur in the region of overlap between the two coding segments.
E2F and its dimerization partner DP-1 are transcription factors that appear to drive the cell cycle in mammalian cells, and E2F activity is governed by its interaction with the Rb tumor suppressor protein. Dr. Nick Dyson described the cloning of the homologs of all three of these proteins in Drosophila melanogaster and the use of genetic analysis in Drosophila in defining their functions. He showed that the Drosophila Rb related protein (RBF) can have effects antagonistic to those of dE2F and that such effects are dependent on the level of RBF expression. Studies using homozygous mutant dE2F embryos revealed that E2F function is required for the transcriptional program that accompanies G1 to S phase progression in embryonic cell cycle 17. Examination of the normal expression patterns of dE2F revealed that dE2F is also expressed in post-mitotic cells of the larval eye disc. Elevated dE2F expression induced cell cycle entry linked to apoptosis in the cells of the eye disc. Thus E2F has roles in cell cycle progression and development in Drosophila.
Session II: Cell Signaling and Growth Control
The cytokine Interleukin-6 (IL-6) exerts distinct biological effects on diverse cell types including growth arrest, differentiation, and proliferation. Dr. Toshio Hirano described his laboratory’s studies IL-6 signal transduction in M1 myeloid leukemia cells. The IL-6R interacts with the gp130 subunit shared by a group of cytokine receptors. Gp130 in turn is associated with the Jak1,2, and Tyk2 kinases. IL-6 binding leads to phosphorylation of gp130 and activation of the Jak-Stat transduction pathway. He demonstrated a tight correlation between Stat 3, the downregulation of c-myc and c-myb and the ability of IL-6 to arrest and differentiate M1 cells. By generating a dominant negative Stat3 he was able to demonstrate inhibition of myc and myb repression and loss of growth arrest in response to cytokine. Thus Stat3 is a critical component of the IL-6 signal transduction pathway.
Dr. Tadatsugu Taniguchi discussed work in his laboratory on the regulation of host defense mechanisms by the IRF family of transcription regulatory proteins. IRF-1 is a transcription activator that binds to a response element within interferon inducible genes. Mice bearing a targeted deletion of IRF1 are unable to mount an effective defense against infection by EMCV, but can block infection of HSV demonstrating the importance of IRF-1 in some aspects of host defense. IRF-1 also functions as a tumor suppressor and primary cells lacking it are directly transformed by an activated ras oncogene. Likewise the knockout mice display increased susceptibility to tumorigenesis by a variety of carcinogens. IRF1 is also involved in apoptosis in mature T cells: there is no induction of the ICE protease or p21CIP in IRF1 homozygous knockout MEFs. These results suggest that IRF-1 has physiologically relevant targets that regulate tumorigenesis and apoptosis.
Regulation of G1 progression was discussed by Dr. Martine Roussel. Her laboratory has employed CSF-1R bearing 3T3 cells to examine cell cycle entry in response to the CSF-l cytokine. Previous work had demonstrated that both c-myc and cyclin D1 expression are required for progression. Mutants at tyrosine 809 in the CSF-1R failed to induce these genes and also failed to enter the cell cycle, whereas ectopic expression of either c-myc or cyclin D1 could rescue the mitogenic response. Recent evidence suggests that the cyclinD1 and c-myc functions are not “in series” but likely to act in parallel since a dominant negative c-myc can block rescue by ectopically expressed cyclin D1. Myc function is also required for transformation by the bcr-abl fusion oncogene. Mutations in the SH2 FLVRES sequence can be functionally complemented by ectopic myc expression again suggesting that c-myc is required for functionality of this region. Cyclin D1 can also complement this mutation. However, cyclin E is incapable of complementing either the bcr-abl or the CSF-1R mutations.
Dr. Kohei Miyazono reported his studies on TGF signaling which, in epithelial cell types, causes growth arrest and matrix deposition as a result of separate signaling pathways through the type I TGF!!!receptor (T!!!R-I). The type I and type II TGF!!!receptors form a heterocomplex upon ligand binding and T!!!R-II phosphorylates a specific region (GS) of T!!!R-I thus activating it for further signalling. Recent evidence indicates C-terminal truncation of the type II receptor in cancer cells from HNPCC patients. In order to identify proteins involved in signal transduction downstream of T!!!R-I Dr. Miyazono’s laboratory has been using a two-hybrid screen. The screen has permitted the identification of three potential substrates or interactors for T!!!R-I.
Session III: Factors Governing Differentiation and Death
Dr. Susan McConnell discussed research from her laboratory on differentiation in the developing nervous system. Her group has carried out transplantation experiments which demonstrate that commitment to differentiation among cortical embryonic progenitor cells is made at a specific period during the cell cycle, just prior to mitosis. Differentiation to neuroblasts from multipotent progenitor cells appears to be correlated with the plane of division of the progenitors in the apical layer of the ventricular zone of the cerebral cortex. The divisions which result in formation of a neuroblast are asymmetric in that the Notch 1 protein segregates almost entirely with the daughter neuroblast. The developmental potential of early and late progenitor cells also varies and appears to correlate with the expression of the homeodomain gene Otx1.
Dr. Robert Eisenman focused on the Mad protein family: a group of bHLHZ proteins which interact with Max and appear to antagonize the function of the proliferation-promoting c-Myc oncoprotein. Mad family proteins can inhibit co-transformation of primary cells by Myc and Ras, and the Mad1 protein can block c-myc-dependent cell cycle progression in response to a mitogenic stimulus. Studies in cell lines and by in situ hybridization in embryos indicate that the expression of mad family genes is correlated with cell differentiation. The mad1 and mad4 genes are expressed in diverse cells types following downregulation of c-myc or N-myc expression. Mxi1 appears to be expressed in both proliferating and differentiating cell populations. Mad3 however can be detected in the ventricular zone of the neural tube in S phase progenitor cells (E10.5). The possibility was raised that these proteins may oppose myc function by repression of subsets of myc target genes at different times during differentiation.
Dr. Michael Levine described how enhancer-promoter interactions influence key developmental decisions in Drosophila embryogenesis. He distinguished between short-range and long-range repression. Short-range repressors, such as snail, kruppel and knirps, function over regions 100bp< and block transcription by neighboring activators. Long-range repressor can negate the effects of multiple activators, presumably by directly affecting the basal transcription machinery. However by placing a short-range repressor proximal to the initiation site, a dominant effect over multiple activators can be detected. Insulators are another feature of complex transcription units such as the Antennapedia and Bithorax gene complexes. These elements attenuate interactions between enhancers. Evidence was presented showing that protein interactions determine the activity of insulators.
Dr. Craig Thompson spoke on the mechanisms involved in apoptosis, focusing particularly on the cascade of IL-1!!!converting enzyme-related proteases, which are believed to execute the cell death program, and the Bcl-2 related proteins that can block the program. Bcl-2 will inhibit the death pathway but alone does not promote cell cycle progression and is therefore distinct from survival factors such as IL-3. Bcl-2 is but one member of group of proteins which contain conserved segments related to Bcl-2. One of these proteins Bclx also promotes cell survival independent of interaction with Bax (a Bcl-2 binding partner that promotes cell death). Deletion of a region of the protein just N-terminal to the Bcl-2 homology sites increases the survival activity of Bclx and Bcl-2 by over 100 fold. Therefore, this region may serve as an autoinhibitory domain. Bcl-2 can protect against cell death induced by a large number of chemotherapeutic agents and the experiments suggest that protection can occur at nearly any point within the cell cycle. In the presence of growth factors the vincristine-treated cells rescued by Bclx display a relatively high degree of tetraploidy. Experiments indicate that damage in the presence of mutant p53 and bclx can generate highly deregulated cells displaying a high degree of genomic instability.
Session IV: Signaling Through Cell Structure
Dr. Masayuki Yamamoto focused on the molecular mechanism underlying the decision of fission yeast to undergo meiosis in response to starvation conditions. The meiotic response is governed by cAMP levels. A decrease in cAMP triggers a pathway leading to induction of the HMG-related transcription factor Ste11. Ste11 expression is regulated in the pathway by formation of a complex between the two bZip proteins Gad7:Pcr1. Gad7 is under the positive control through the MAPK pathway and under negative control by PKA. One of the targets of Ste11 is the mei2 gene. The Mei2 protein is an RNA binding protein whose function is essential for sporulation. Mutants in mei2 can be suppressed by a small ORF-less RNA that forms a complex with Mei2. The activity of Mei2 can be reversed by phosphorylation mediated by the Pat1 kinase. Thus, S. pombe provides an excellent example of how critical cell behavioral changes can be rapidly mediated by signaling in response to environmental cues.
Many signal transduction systems in both yeast and larger eukaryotes involve the action of small G proteins. Dr. Yoshimi Takai summarized current information of the six families of small G proteins: Ras, Rho, Rab, Arf, Sar1, and Ran. Although Ras family proteins are known to mediate signaling through the MAP kinase cascade the functions of the other subfamilies are unknown. Of particular interest is the Rho family which bind GDP and GTP and regulate the cytoskeleton. Rho proteins can be inhibited by ADP ribosylation. Its activity is also regulated by stimulatory GEP, inhibitory GEP and GAP interactions. There is evidence that Rho is post-translationally modified by lipid and that such modification is essential for activity Rho is thought to regulate cytoskeletal motility and ruffling. Rho has recently been demonstrated to bind BNI1 a protein homologous to profilin binding proteins.
Dr. Shoichiro Tsukita discussed the structure and regulation of “tight junctions” between intestinal epithelial cells: these are distinct from desmosomes and adherans junctions and are involved in creating and maintaining epithelial and endothelial cell polarity. Recently a novel protein, occludin, has been identified and cloned and appears to be a major component of tight junctions. It is expressed in normal intestinal epithelium and in the blood-brain barrier. Pervious work had identified the ZO-1 and ZO-2 proteins as tight junction components and it is now demonstrated that these two proteins interact directly with the cytoplasmic region of occludin. Both ZO-1 and ZO-2 display homology to the discs large (dlg) tumor suppressor in Drosophila. This raises the possibility that proteins governing cell-cell interactions may also regulate cell behavior and play roles in neoplasia.
An example of how cell-cell interactions may be important in oncogenesis is provided by the APC tumor suppressor protein. Dr. Paul Polakis summarized information on APC’s involvement in a heritable form of familial adenomatous polyposis frequently encountered in the U.S. but less frequently in Japan. APC mutations are also found in most sporadic colon cancers. Such mutations are generally chain terminating. Both the normal and truncated APC associate with!!!-catenin which in turn interacts with cadherin to form calcium dependent adherens junctions. In transformed cells lacking functional APC a large excess of unbound cytoplasmic!!!-catenin can be detected suggesting that APC may regulate the level of free catenin, possibly by normally being involved in its turnover. Recent experiments suggest that phosphorylation of wild-type APC by the associated kinase GSKIII leads to binding and turnover of the bound!!!-catenin. One model is that free b-catenin may be involved in a signaling pathway that leads to cell proliferation.
Assessment:
This meeting brought together a diverse group of scientists working with different organisms and on distinct aspects of cell and molecular biology. The purpose was to explore from different angles the nature and function of molecules that are involved in the “choice” between differentiation and proliferation. In some areas (e.g. cell cycle inhibitors), we have a highly detailed and sophisticated view, while in others (e.g. relationship between cell structure and signaling) the important questions are only just being framed. The beautiful setting for the meeting encouraged interaction among participants and the diversity of the group encouraged the speakers to explain their work succinctly and lucidly. The length and intensity of the discussion sections suggests that most participants were deeply involved and learned a great deal. Indeed a number of collaborations were established as a result of the meeting. In the view of the organizers, the meeting was a success: new and intriguing information was presented and the result can only stimulate a deeper exploration of the questions posed at the conference.

PARTICIPANTS

UNITED STATES

Dr. Leland Hartwell
Department of Genetics
University of Washington
Seattle, WA 98195-7360

Dr. Martine Roussel
Division of Human Tumor Cell Biology
St. Jude Children’s Research Hospital
332 North Lauderdale
P.O. Box 318
Memphis, Tennessee 38101

Dr. Charles Sherr
Division of Tumor Cell Biology
St. Jude Children’s Research Hospital
331 North Lauderdale
P.O. Box 318
Memphis, Tennessee 38101

Dr. Susan McConnell
Department of Biological Sciences
Stanford University
Stanford, CA 94306-5020

Dr. Nicholas J. Dyson
MGH Cancer Center
13the Street, Bldg 149
Charlestown, MA 02129

Dr. Michael Levine
Biology Department, 0347
University of California, San Diego
9500 Gilman, UCSD
La Jolla, CA 92093-0322

Dr. Paul Polakis
Onyx Pharmaceuticals
3031 Research Drive, Bldg A
Richmond, CA 94806

Dr. Robert Eisenman
Division of Basic Sciences, Room A2-025
Fred Hutchinson Cancer Research Center
1124 Columbia Street
Seattle, WA 98104

Dr. Craig Thompson
Gwen Knapp Center
The University of Chicago
924 East 57th Street, Room 410
Chicago, IL 60637-5420

JAPAN

Dr. Toshio Hirano
Professor, Faculty of Medicine
Osaka University

Dr. Yumiko Kubota
Research Fellow of Japan Society for the Promotion of Science
Research Institute for Microbial Diseases, Osaka University

Dr. Kohei Miyazono
Member and Chief, Cancer Institute
Japanese Foundation for Cancer Research

Dr. Yoshimi Takai
Professor, Faculty of Medicine
Osaka University

Dr. Syoichiro Tsukita
Professor, Graduate School of Medicine, Faculty of Medicine
Kyoto University

Dr. Masayuki Yamamoto
Professor, Department of Biophysics and Biochemistry
University of Tokyo

Dr. Tadatsugu Taniguchi
Professor, Department of Immunology
Faculty of Medicine
University of Tokyo

Observers:

Dr. Hodaka Fujii
Research Fellow of the Japan Society for the Promotion of of Science
Department of Immunology
Faculty of Medicine
University of Tokyo

Dr. Sachiko Tsukita
Professor
College of Medical Technology
Kyoto University



(3) Seminar on “The Role of Cytokines in Cancer”
Introduction:
The meeting organized by Dr. Joost Oppenheim and Dr. Hiromi Fujiwara was held as an open forum at the Natcher Conference Center on the Bethesda NIH Campus from January 15-17, 1996, and was attended by 19 Americans and 7 Japanese invited participants. The effect of cytokines on tumors was considered over the course of the two and one half day meeting from a variety of perspectives. Most of the first day was devoted to a discussion of the role of cytokines in modulating angiogenesis and the consequent effect of this on tumor growth and metastases. This was followed by sessions on the effect of various cytokines in enhancing or suppressing immunological responses to tumors. Several presentations focused on the direct effects of some cytokines in either inhibiting or at times promoting tumor growth. The final session consisted of a comparison of the efficacy of different approaches to tumor vaccination including gene therapy, enhanced antigen presentation, use of organic carries or of DNA vectors.
We will summarize in brief the crucial observations and concepts conveyed by the speakers. The first session dealt with the role of cytokines in angiogenesis and tumor progression. Dr. Richard O’Reilly presented the results of his studies performed with his collaborators in Dr. Judah Falkman’s laboratory. He has found that primary tumors generate an inhibitor of angiogenesis which suppresses the growth of tumor metastasis, thus accounting for observations that removal of the primary promotes the development of metastasis. A suppressive factor present in plasma and urine of tumor bearing mice was purified and identified, by microsequence analysis, as an internal fragment of plasminogen. We have named this angiogenesis inhibitor angiostatin. Another different angiostatic factor has subsequently been purified from a murine hemangio-endothelioma is a fragment of collagen XVIII and has been called endostatin.
When given systemically, angiostatin, produced by a limited proteolytic digest of human plasminogen, potently inhibits angiogenesis induced by basic fibroblast growth factor (bFGF) in a mouse model of corneal neovascularization. Systemic therapy with angiostatin potently inhibits both metastatic and primary tumor growth of murine and human tumors in mice. For the human tumors, the suppression of angiogenesis by angiostatin induces a state of dormancy defined by a balance of proliferation and apoptosis of the tumor cells. Unlike many other agents, there is no detectable toxicity in animals treated for prolonged periods will angiostatin. Discussion revealed that plasminogen itself is inactive, and that high concentrations of the angiostatin are needed to obtain in vivo effects.
Dr. Hal Dvorak and his colleagues have been studying a vascular permeability inducing factor for over 20 years. Vascular Permeability Factor (VPF), variously known as Vascular Endothelial Growth Factor (VEGF) and Vasculotropin (VAS), is a multi-functional cytokine with important roles in vasculogenesis and in both pathological and physiological angiogenesis. Originally described in the late 1970s as a tumor-secreted protein that potently increases microvascular permeability to plasma proteins, VPF/VEGF exerts a variety of effect on vascular endothelial cells which together promote the formation and growth of new blood vessels. In addition to rendering microvessels hyperpermeable with a potency some 50,000 times that of histamine, VPF/VEGF induces calcium transients, stimulates endothelial cells to migrate and divide, and profoundly alters their pattern of gene expression. In particular, VPF/VEGF induces the expression of genes associated with clotting and fibrinolysis. VPF/VEGF and both of its endothelial cell receptors are strikingly overexpressed in the angiogenesis associated with most common human tumors but also in certain pathological but non-neoplastic settings such as, for example, wound healing, rheumatoid arthritis, psoriasis, delayed hypersensitivity, diabetic retinopathy, etc. Similar events occur in the physiological angiogenesis of endometrial cycling and corpus luteum formation. Thus, common mechanisms appear to be involved in all of these expressions of angiogenesis. Tumor cells transfected to overexpress VPF/VEGF become more tumorigenic and antibodies to VPF suppress tumor growth.
VPFNEGF is a dimeric protein joined together by disulphide bonds; as the result of alternative splicing, it exists in four different isoforms, all of which seem to have the basic biological activities. VPF/VEGF exerts its effects on endothelial cells by way of two high affinity transemebrane tyrosine kinase receptors, flt-1 and KDR/flk-1. These receptors for VPF/VEGF are overexpressed on endothelial cells in wound healing and chronic inflammatory states. VPF/VEGF increases microvascular permeability to circulating proteins by activating a recently discovered endothelial cell cytoplasmic organelle, the VVO. Increased microvascular permeability apparently occurs as an early step in all forms of angiogenesis so far investigated. This increase in permeability leads to an outpouring of plasma proteins and alters the extracellular matrix, converting it form an “anti-angiogenic” to a “pro-angiogenic” state, depositing a fibrin gel that serves as a favorable matrix for endothelial cell migration. VPF/VEGF expression is upregulated by the malignant phenotype but also by hypoxia, a variety of cytokines, hormones, etc.
Discussion revealed that VPF/VEGF is a most potent and pivotal angiogenic factor that is induced by proinflammatory cytokines as well as growth factors. However, it is not induced by IL-2 and not a cause of the vascular leak syndrome.
Dr. Bob Kerbel presented data in support of the idea that well known oncogenes such as mutant ras may actually promote tumor growth by promoting angiogenesis. Their results indicate that mutant ras oncogenes upregulate VPF/VEGF. Conversely, farnesyl transferase inhibitors of mutant ras block the production of VPFNEGF in H-ras transformed cell lines. This may account for the lack of in vitro cytotoxic effect, but dramatic in vivo antitumor effects of these inhibitors.
Furthermore, transfection of an immortalized epithelial cell line with mutant ras also suppressed apoptosis due to programmed cell death (PCD). Cells that no longer undergo PCD characteristically lose their anchorage independence and grow three dimensionally. This is presumably promoted in vivo by m-ras induced VPF/VEGF. In addition, wt p53 suppresses the production of VPF/VEGF and this is lost in the many tumors with mutated p53.
Studies of the role of chemokines in modulating angiogenesis have been pioneered by Dr. Robert Strieter He pointed out that these CXC chemokines containing the ELR motif in their amino acid sequence are potent angiogenic factors, inducing both in vitro endothelial chemotaxis and in vivo corneal neovascularization. In contrast, the CXC chemokines lacking the ELR motif, PF4, IP-10, and MIG, not only failed to induce significant in vitro endothelial cell chemotaxis or in vivo corneal neovacularization, but were shown to be potent angiostatic factors in the presence of either CXC chemokines containing the ELR motif or the unrelated angiogenic factor, basic fibroblast growth factor (bFGF).
IL-8 (contains ELR motif) mediates both endothelial cell chemotactic and proliferative activity in vitro and angiogenic activity in vivo. In contrast, PF4 (lacking the ELR motif) has been shown to have angiostatic properties, and attenuates growth of tumors in vivo. The interferons (IFN-!!!, IFN-!!!, and IFN-!!!) are all known inhibitors of wound repair, especially angiogenesis. These cytokines, however, upregulate IP-10 and MIG from a number of cells, including keratinocytes, fibroblasts, endothelial cells, and mononuclear phagocytes. Finally, they and others have found that IFN-!!!, IFN-!!!, and IFN-!!!are potent inhibitors of the production of monocyte-derived IL-8, GRO-!!!, and ENA-78, supporting the notion that IFN-!!!, IFN-!!!, and IFN-!!!may shift the biological balance of ELR-and non-ELR-CXC chemokines toward a preponderance of angiostatic (non-ELR) CXC chemokines.
Tumor-derived IL-8 production directly correlated with the rate of growth of the two human NSCLC cells lines in vivo. IL-8 was not found to behave as an autocrine growth factor for the proliferation of NSCLC cells. Moreover, when IL-8 was depleted in vivo by a strategy of passive immunization with neutralizing antibodies, tumorigenesis was markedly reduced via a reduction in tumor-derived angiogenic activity. These findings support the contention that IL-8 mediated angiogenesis is critical to NSCLC tumorigenesis. On the other hand, intratumor injections of IP-10 inhibited tumor growth.
These observations were reinforced and extended by the presentation by Dr. Giovanna Tosata from the FDA who replaced Dr. Josh Fidler who was incapacitated by the flu. She reported that a murine athymic mouse model was developed in which regression of malignant Burkitt’s lymphoma can be induced reproducibly by intratumor injections of EBV-immortalized normal B cells. Regressing tumor tissues expressed significantly higher levels of IL-6, TNF-!!!, IFN-!!!, IL-12 p35 subunit, Mig, RANTES and IP-10 than progressive tumor tissues. In contrast, the IL-12 p40 subunit, MIP-1!!!, MIP-1!!!and JE were expressed at similar levels in progressive and regressing Burkitt’s tumors. Because endothelial cell damage and intravascular thrombosis were prominent features of regressing tumor tissues, she tested for the possibility that inhibition of angiogenesis is responsible for tumor regression. Using an angiogenesis assay in which a Matrigel plug impregnated with bFGF is injected subcutaneously into athymic mice, human IP-10 and murine IL-12 consistently inhibited neovascularization. Because IL-12 is an inducer of IFN-!!!, and IFN-!!! is an inducer of IP-10, she tested whether the antiangiogenic effect of IL-12 could be mediated by IP-10. A neutralizing antibody to murine IP-10 removed most of the inhibition of angiogenesis induced by IL-12. In additional experiments, inoculation of Burkitt’s tumors established subcutaneously into athymic mice with human or murine IP-10 caused extensive tumor necrosis in most tumors. Histologically, IP-10 treated Burkitt’s tumors had widespread evidence of capillary damage, including intimal thickening and vascular thrombosis. Thus, IP-10 is an inhibitor of angiogenesis and a potent antitumor agent.
Dr. Fijio Suzuki studied cartilage because it is avascular and resists invasion by tumor cells. He has purified cartilage extracts and obtained chondromodulin I (ChM-I), an angiostatic factor that inhibits the proliferation and tube formation by endothelial cells (EC). ChM-1 counters bFGF and the bone forming activity of Bone Morphogenetic Proteins (BMP). Thus, ChM-1 stimulates cartilage formation and prevents its conversion to bone. ChM-1 also promotes cartilage growth and stimulates osteoclast mediated bone formation, but does not inhibit EC tube formation. Injection of CHMI do suppress growth of B16 melanoma tumors in mice; the angiostatic properties of ChM-I may make it a nontoxic antitumor agent.
By studying endothelial cell genes activated by tumor necrosis factor (TNF), Dr. Vishva Dixit has identified B61 as another potent angiogenic factor. This factor may provide the means by which TNF can induce corneal angiogenesis in the absence of an inflammatory cell infiltrate. This effect of TNF can be blocked by antibody to B61. This antibody blocks interaction of the B61 with its Eck receptor protein kinase. He also identified the p85 subunit of PI3 kinase and an Src-like adapter protein (SLAP) as participants in B61 induced signal transduction.
Dr. Tony Passanite studied the fact that tumors in aged mice don’t vascularize well. He uses an in vitro Matrigel to grow endothelial cells as a model of angiogenesis. Injection of Matrigel with bFGF promotes angiogenesis and this can be blocked by TIMP-1. The EC from 26 wk old mice do not grow well even in vivo in response bFGF. Furthermore, extracts of tumor from aged inhibit EC differentiation in this vitro model. Extracts from young mice don’t inhibit. He believes that host factors from aged mice inhibit tumor growth. However, the factor(s) remains to be identified. There is no upregulation of IP-10 in aged mice .
The final presentation relevant to the role of angiogenesis was that of Dr. Barbara Ensoli who studied kaposi sarcoma (KS) Formation in AIDS patients. It is still not clear if this is truly a malignant tumor or a polyclonal EC dependent benign growth. Spindle cells of vascular origin from the tumor produce a vast number of cytokines and growth factors. Transient KS lesions can be grown in nude mice and are inhibited by antigens to bFGF. Growth of KS cells is promoted by TNF-!!!, IFN- !!!and HIV-TAT proteins. IFN-!!!primes EC to respond to HIV-1 TAT problems and exposure to the latter enables EC to cross basement membranes and invade tissues. HIV-1 TAT also synergizes with bFGF to promote angiogenesis and contains an RGD domain that promotes the adhesion of EC to!!!5!!!1 and!!!V!!!3 integrins. KS derived spindle cells produce angiogenic cytokines such as bFGF, VEGF and IL-8. By promoting adhesion and interacting with these cytokines CD4 + cell derived HIV-1 TAT may be responsible for the increased frequency and aggressiveness of KS in AIDS patients.
Several speakers in the second session addressed other possible consequences of cytokines in the influencing tumor growth and metastasis. Dr. Garth Nicolson has been interested in the role of endothelial cell derived factors that attract tumor cells and may be important determinants of the site of localization and organ preference of metastases. He identified C3b as a hepatic sinusoidal EC motility factor. He also established that transformed RAW 117 monomyelocytic cell line cells are chemoattractant by conditional medium from murine lungs. Purification identified the chemoattractant effect was blocked in vitro by anti-MCP-1. Brain cells produce a 70 kd factor, which has not as yet been identified. Organs also produce tumor mitogens, one of which is homologous to transferrin (TF). They have cloned 3 such TF-like tumor growth factors. Metastases in brain > lung > liver express receptors for these TF like factors.
Although tumor bearing mice that were injected with a colon 26 tumor cell variant (clone 20) developed cachexia and produced IL-6, antibody to IL-6 only partially reversed the cachexia according to Dr. Kouji Matsushima. Thus, IL-6 is necessary, but not sufficient for induction of cachexia according to Dr. Matsushima. Thus, IL-6 is necessary, but not sufficient for induction of cachexia. In order to identify additional factor(s) regulating cachectic state of clone 20 bearing mice, he has applied differential display analysis for comparison of the expression pattern of genes between clone 20 and clone 5 tumors. Three cDNA clones were identified as genes which were selectively expressed in the clone 20 tumor in vivo. Sequence analysis of cDNA clones revealed that two clones originated from the same gene and belong to a mouse retrotransposon VL30 family, indicating that the signaling pathway leading to the activation of LTR of a certain class of VL30 is activated in the cachexigenic clone 20 tumor in vivo. The third cDNA clone, highly expressed in the clone 20 tumor is a novel gene and named cachexigenic tumor associated gene-1 (CTAG-1). A full length cDNA of CTAG-1 was cloned and deduced amino acid sequence revealed that CTAG-1 gene encodes a novel metalloprotease-distegrin protein with thrombospondin type I motifs. CTAG-1 mRNA was inducible by stimulating clone 20 cells with IL-1 in vitro and selectively inducible in heart and kidney in mice by administrating LPS iv. These data suggest that CTAG-1 may be involved, not only in the inflammatory processes at the tumor site causing cachexia, but also acute inflammation in some organs.
The third session on the role of the immune response to tumors was introduced by Dr. Takashi Yokota with a discussion of the role of TH1 and TH2 cytokines in cancer. Mouse TH1 cells produce IL-2 and IFN-!!!and mediate cellular immunity, whereas TH2 cells produce IL-4, IL-5, and IL-10, and stimulate humoral immunity. Th1 and Th2 cells may differ either in the signal transduction pathways for cytokine production or in the transcription factors that regulate cytokine production. Differential sensitivity of TH1 and TH2 cells to reagents that elevate intracellular cAMP such as PGE2, cholera toxin, and forskolin is in keeping with these views. After elevation of intracellular cAMP, cytokine production from TH1 clones but not TH2 clones is inhibited, whereas IL-5 production is increased. Therefore, signaling molecules and/or transcription factors that are differentially required for TH1- and TH2- type cytokine production may be the target of cAMP actions. The mechanism of reciprocal action of cAMP on cytokine production by TH1 and TH2 cells was studied to shed light on the molecular basis for differences between TH1 and TH2 cells. Transient transfection analyses showed that the IL-2 promoter and IL-5 promoter are oppositely regulated by cAMP in EL-4 cells, and differentially regulated in TH1 clone, HDK1, and TH2 clone, D10.G4.1 cells. Using deletion and mutation analyses, he identified four cis-regulatory elements necessary for full activity of the IL-5 promoter. He designated these elements as IL-5A (-948 to -933), IL-5P (-117 to -92), IL-5C (-74 to 056) and IL-5 CLEO (-55 to -38). He found that IL-5P bears homology to the binding site for the nuclear factor of activated T cells (NF-AT), and interacted with protein factors in nuclear extracts prepared from EL-4 cells stimulated with PMA and Bt2cAMP (designated NFIL-5P). The NFIL-5P complex was inhibited in the presence of an excess NF-AT and AP1 oligonucleotides and supershifted by antisera raised against NF-ATp, c-Fos, and c-Jun. It thus seems likely that an NF-AT-related factor is involved in the regulation of IL-5 gene transcription. The IL-2 promoter contains several transcriptional factor binding sites, including NF-AT, NF-kB and AP1, and cooperation of these transcription factors is required for maximal activation of the promoter. Comparative electrophoretic mobility shift assay using nuclear extracts prepared from PMA/A23187-stimulated HDK1 and D10.G4.1 cells showed that binding of the NF-kB (p50/p65) heterodimer to the NF-kB site was observed in HDK1 cell extracts, but not in D10G4.1 cell extracts. Immunobloting using nuclear and cytoplasmic extracts of these cells demonstrated that p65 was present in both cells. However, nuclear translocation of p65 in response to stimulation occurred only in HDK1 cells. Thus, nuclear translocation of p50/p65 NF-kB complex is differentially regulated between TH1 and TH2 cells.
The effect on tumor models of IL-2 which promotes TH1 type of cytokine production, was presented by Dr. Takashi Nishimura. In vivo administration of IL-2 caused the inhibition of both transplantable and primary tumor growth concomitant with the augmentation of serum IFN-!!!levels and the generation of tumor-specific protective immunity. Activation of TH1-dominant immunity by IL-2 in vivo was also demonstrated to be effective on Trypanosoma cruzi infection model and TH-2-dependent autoimmune host versus graft disease (HVGD). Conversely, IL-2 accelerated the onset off several immune diseases such as TH-1-dependent liver injury and graft verses host disease (GVHD). In such diseases, in vivo administration of monoclonal anti IL-2 antibody had a potent therapeutic effect. Thus, it seems to be true that the balance between TH1-TH-2 play an important role in various immune diseases and is possible to regulate this balance by IL-12 or its antagonist. The discussion also solicited some additional important point such as the fact that IL-2 has a relatively long half life of six hours which may contribute to its rather prolonged effects. Transfection of tumors with a combination of IL-12 plus B7 resulted in almost complete rejection of the very tumorigenic B16 melanoma tumor in mice, while transfection with either of these agents by themselves was not effective. These are also genetically controlled non-MHC related factors that influence the expression of TH1 and TH2 reactivity. For example, Balb/c mice have much greater number of IL-4 producing memory cells than C57B1 mice. The former have more CD4 + RB low, CD62L-cells, whereas the latter have more CD4 + CD62L +T cells that produce more IFN!!!.
Dr. Stephen Ullrich and his collaborators in Dr. Margaret Kripke’s laboratory have elucidated the basis for the immunosuppressive effect of ultravioret (UV)B irradiation. UV-B is a cause of both non-melanoma and melanoma skin cancer. The UV-B induced immune suppression can be transferred to normal recipient mice by adoptive transfer of antigen-specific, CD3+,CD4+,CD8-T cells, cells that are responsible for controlling the induction of the primary tumor in the UV-Irradiated host. One mechanism responsible for the systemic wide modulation of immune function after UV exposure, is the release of immunomodulatory cytokines by UV-irradiated keratinocytes. One such cytokine is interleukin (IL)-10. IL-10 released by UV-irradiated keratinocytes, is found in the serum of UV-irradiated mice, and neutralizing its activity in vivo blocks the induction of immune suppression. One consequence of increased serum IL-10 levels following UV exposure appears to be shift towards a TH-2-like immune reaction. While spleen cells from UV-irradiated mice do not present antigen to TH-1 clones, the ability of antigen-presenting cells from UV-irradiated mice to present antigen to IL-10 reverses these effects. Moreover, Dr. Ullrich suggests that one effect of UV-induced IL-10 on antigen presentation leads to the preferential expansion of Th2 cells, that upon activation secrete more IL-10 and IL-4, and suppress inflammatory immune reactions. In preliminary experiments, he found that antibodies to IL-4 and IL-10 will block the activity of the transferred suppressor cells and restore immune function. In addition, treatment of UV-irradiated mice with IL-12, a cytokine that promotes the activation of TH1 cells, and blocks the activation/differentiation of TH2 cells, reverses the induction of immune suppression, and the induction of suppressor cells in UV-irradiated mice.
Dr. Ullrich proposed that IL-10, released by UV-damaged keratinocytes, modulates systemic antigen presenting cell function. The net effect of UV-induced IL-10 on antigen presenting cell function appears to be an inhibition of TH-1 cell stimulation with a preferential activation of TH2 cells. When these cells are stimulated with antigen, they secrete the TH2 cytokines, IL-10 and IL-4, which can down regulate inflammatory immune reactions, such as delayed-in-time hypersensitivity reactions and tumor rejection, and permit the outgrowth of highly antigenic. UV-induced skin cancers.
Another known immunosuppressive cytokine, TGF!!!, is reported to promote the growth of some tumors. Dr. Sanford Markovitz presented his studies showing that the signal transducing human type II TGF!!!!receptor is a tumor suppressor gene. TGF!!!is a potent Inhibitor of colon epithelial cell growth that when studied in vitro completely abrogates growth of two independent nontransformed colon adenoma cell lines. In addition to blocking cell growth, TGF!!!also induces apoptosis in nontransformed colon epithelial cells. In vitro progression of a colon adenoma to a carcinoma phenotype is accompanied by acquisition of resistance to TGF!!!growth inhibition, suggesting a functional role for TGF!!!in colon cancer progression. To determine if alteration in TGF!!!receptors could underline TGF!!!resistance in colon cancer, receptor expression was examined by crosslinking iodinated TGF!!!to a panel of 37 colon cancer cell lines. Absence of crosslinking to the RI and RII components of the receptor was noted in eight cases. each of which was also demonstrated to be resistant to TGF!!!induced inhibition. Unexpectedly, each of these cases also demonstrated the genetic phenotype of microsatellite instability (also known as replication error or RER) in which tumors show numerous alterations in the lengths of the generally noncoding DNA sequences which comprise the DNA microsatellite sequences scattered throughout the genome. This RER phenotype is present in many colon cancers found sporadically in the general population when the cancers arise on the right side of the abdomen. RER is also typical of colon cancers that arise in individuals with familial cancer syndrome of hereditary nonpolyposis colon cancer (HNPCC). RER is also found in many endometrial and gastric cancers both among those tumors that arise sporadically, as well as among those that occur with high frequency HNPCC kindreds. Recent studies have shown that defects in genes encoding the DNA base-base mismatch repair system account for many, though not all cases of RER cancers. Furthermore, expressed gene sequences, as well as DNA microsatellite, become hypermutable in cancers.
To determine whether loss of cell surface TGF!!!receptors in RER cancers was due to a mutation, the DNA sequence of the receptors was examined. RII mutation were demonstrated to be present in. each of the RER colon cancers. Moreover, these mutations were also demonstrated present in each of antecedent colon tumors from which three of the RER colon cancer cell lines had been established, but were absent from DNA extracted from normal tissue obtained from the same individuals. RII was inactivated by BAT-RII frameshift mutations in 100 of 111 RER colon cancers (90%). This included 75 of 84 cases (89%) of primary RER colon cancer tumors and included 25 of 27 (90%) RER colon cancers examined as tumor xeographts or immortalized cell lines. To determine whether BAT-RII mutations also occur in extracolonic RER cancers, 7 cases of RER gastric cancer were also examined. Five of 7 cases (71%) of the gastric cancers demonstrated BAT-RII frameshift mutations. Thus, RII mutation are not restricted to a single tissue only, but play a clear role in the generation of malignancies of both the upper and lower gastrointestinal tract. Unexpectedly, BAT-RII mutation were only occasionally present in RER endometrial tumors (4 of 24 cases, 17%) and were absent in 5 of 6 endometrial cancers obtained from HNPCC kindreds as well as in 2 of 2 endometrial cancer cell lines each known to bear the same type of mutations in DNA repair genes as are typically are associated with the HNPCC syndrome. Thus, BAT-RII mutation are not simply markers of the RER phenotype. Rather, RII inactivation must be actively selected for in RER gastric and colon cancers. This conclusion is supported by the finding that in three RER colon cancers in which only one RII allele was inactivated by a BAT-RII mutation, the remaining RII allele was also inactivated by novel RII mutations located elsewhere in the gene. Functional demonstration that RII tumor suppressor gene activity comes from demonstrating that transfecting a wild type RII CDNA into Hct116 an RER colon cancer cell line in which the endogenous RII alleles have been inactivated, resulted in suppression of Hctll6 anchorage independent growth in soft agar and loss of Hct116 tumorigenic growth when injected into athymic mice. Thus, RII is a bona fide tumor suppressor gene, and RII mutation are progression events during evolution of RER colon cancers. As TGF!!!resistance is common among many different human cancers, it is likely that many other mechanisms will be found in human cancers both for inactivation RII as well as for inactivating other elements of the TGF!!!signal transduction pathway.
Session concluded with a discussion of the role of FAS mediated apoptosis by Dr. Shige Nagata. Fas is a type I membrane protein belong to the TNF/NGF receptor family, the Fas ligand (FasL) is a member of the TNF family, and exists as a type II membrane protein as well as a soluble cytokine. The soluble FasL is a trimer. Binding of FasL or agonistic anti-Fas monoclonal antibody to Fas rapidly induces apoptosis in Fas-bearing cells. Fas-mediated apoptosis can be inhibited by Bcl-2 as well as inhibitors of ICE (interleukin-1!!!-converting enzymes). The cytosolic fraction from Fas-activated cells quickly induces apoptosis in vitro in nuclei from mouse liver, and causes DNA degradation. There is no evidence as yet that Fas or Fas ligand act as tumor suppressor genes. Mutant mice with the lymphoproliferation mutation (lpr) and generalized lymphoproliferative disease (gld) defects which cause lymphadenopathy and autoimmune disease, are leaky loss-of-function mutations in the Fas and /FasL genes, respectively. The Fas system is involved in the activation-induced suicide of T cells, or peripheral clonal deletion. The Fas mRNA is abundantly expressed in the liver, heart, lung, and ovary. Fas-null mice generated by gene targeting showed lymphadenopathy and splenomegaly, which are much more accelerated and pronounced than those in lpr mice. Tumor development is not a prominent complication in these mutant mice defective Fas or FasL.
The fourth session focused on the immune and nonimmune mediated effects of cytokines on tumors. In this content, Dr. Toshiyuki Hamaoka pointed out that as tumors grow and the host becomes immunosuppressed, the host becomes much less able to produce immunostimulating cytokines such as IL-2, IL-4, TNF!!!, and IFN!!!. In contrast, the serum levels of immunosuppressive cytokines such as TGF!!!and IL-6, a TH2 product inhibits IFN!!!and TNF!!!production, while TGF!!!effectively suppresses the production of all cytokines. Administration of IL-12 to such tumor bearing hosts restores IFN!!!production completely and IL-2 and TNF!!!production partially.
This theme was discussed further by his collaborator, Dr. Hiromi Fujiwara, who added that IL-12 induces IFN!!!production by the tumor infiltrating T lymphocytes. IFN!!!has direct antiproliferative effect on some tumors and enhances the antitumor activities of cytotoxic T cells (CTL) and macrophages. IFN!!!also induces macrophage to produce nitric oxide which is cytotoxic for tumor cells. The discussion indicated that antibody to IFN!!!counters the antitumor effects of IL-12, and IL-12 does not reject tumors in IFN!!!knockout mice according to Dr. Mike Brunda of Hoffmann LaRoshe. This accentuates the pivotal role of IFN!!!as a mediator of the anti-tumor of IL-12.
Dr. Robert Wiltrout obtained regression of a murine renal cell carcinoma RENCA in response to IL-12 therapy in only 50% of mice, whereas IL-2 treatment had little or no effect. However, these two cytokines together caused 75% of the RENCA tumors to regress and a 50% remission rate in PRP-3 transgenic mice with spontaneous breast tumors. Evaluation of tumor growth of implants in collagen gelatin sponges revealed that IL-12 plus Il-2, only about 2/3 of mice developed subsequent specific immune resistance to rechallenge with RENCA tumor cells and the nonresistant mice developed recurrence with metastasis at distant sites. Consequently, additional measure are needed to further potentiate this immunotherapeutic approach.

Dr. Theresa Whiteside studied the “dark side” of cytokines by evaluation of the potential tumor growth promoting effect of cytokines (e.g. squamous cell carcinoma of the head and neck, gastric carcinoma, renal cell carcinoma) and other tumors. These carcinomas expressed IL-2R!!!and!!!and exhibited the capacity to bind IL-2 with intermediate affinity. Human carcinomas were also shown to constitutively produce IL-1 protein and expressed mRNA for IL-2 in vivo an in vitro. Although no IL-2 was detectable in supernatant of carcinoma cells, cell surface-associated IL-2 was detectable by flow cytometry of a viable proportion of tumor cells. Both IL-2 and IL-2R expression was highly upgraded in actively proliferating tumor ceils, compared to tumor cells in the G0/G1 phase of cell cycle. They synchronized tumor cells using aphidicolin and showed that cells in the M phase of the cell cycle contained significantly more intracytoplasmic IL-2 protein (by flow cytometry of permeabilized tumor ceils) and 5-10 fold more mRNA for IL-2 protein (by quantitative competitive RT-PCR) than cells in the G0/G1 phase. Thus, IL-2 behaved like cell-cycle associated protein in carcinoma cells. Antibodies to IL-2R!!and the anti-sense!!!chain construct each inhibited growth of human carcinoma cells in vitro. Hence, both the IL-2R!!and presumably its ligand, endogenous IL-2, were necessary for tumor growth. Carcinoma cell stably transfected with the human IL-2 gene to produce IL-2 proliferated significantly better in vitro than parental cells. Although the growth of tumor cells was not inhibited by antibodies to IL-2 (at <10 µg/ml), it was significantly inhibited by antibodies to IL-2 2R!!.
These data indicate that an autocline IL-2 pathway may exist in tumor cells. In contrast to endogenously synthesized IL-2, which seems to be needed for cell division for cell division and stimulates tumor growth, exogenous IL-2 (at > 10 nM concentration) delivers a negative growth signal tumor cells. In the presence of high concentrations of exogenous IL-2, tumor cess were seen to exocytose vesicles containing intracellular IL-2. No intracytoplasmic IL-2 was detectable in tumor cells incubated in the presence o > 10 nM concentrations IL-2. It appears that exogenous IL-2 at these high concentrations interferes with the intracellular IL-2 pathway, which may be necessary for tumor cell growth. Dr. Whiteside added that although IL-15 does not influence tumor growth, anti-IL-15 is inhibitory. Thus, IL-15 may have analogous endogenous tumor growth effect.
The last session of the meeting compared various immunizing vaccine approaches to induce more effective antitumor responses. Dr. Steven Rosenberg proposed that IL-2 induced tumor regression is dependent on tumor infiltrating (TIL) cell recognition of tumor associated antigens (TAA). He and his colleagues, therefore, used the TIL cells to identify unique antigens coded for by subtractive hybridization. This led to the identification of TAA commonly present on melanoma cells such as tyrosinase, gp100, MART, and mutated!!!catenin. They then generated CTL in quantity in vitro that are direct agent the immunogenic epitopes from these TAA for eventual administration to the patients. In addition, Dr. Rosenberg is administering TAA derived peptides that have been modified to increase their affinity for MHC (to improve their presentation by antigen presenting cells) together with adjuvants to increase or restore the tumor immune response of patients.
The capacity of dendritic cells, which are the most effective antigen presenting cells. To promote greater tumor was studied by Dr. Michael Lotzc. He obtained large numbers of cells enriched in dendritic cells (DC) by culturing bone marrow cells with IL-4 and GM-CSF. He reported that stimulation of DC with a polyclonal stimulant such as antibodies that cross-link MHC class II antigens on DC induces to produce IL-12. In addition, co-stimulation by CD-40-CD40L interactions is essential to DC-T cell interactions. He was able to vaccinate mice against tumors using DC pulsed with peptides derived from TAA to induce resistance in mice to tumor challenge. Furthermore, intravenous administration of such DC pulsed with peptides can even reject an established tumor, provided it is not too far advanced. He and his colleagues have even stripped TAA peptides off tumor cell surfaces using acid elution. Administration of DC pulsed with such an eluate results in specific tumor immunity in mice against several types of tumors. This immunity could be adoptively transferred using spleen cells from immunized donor, and could be blocked by antibodies to CD4 T cells, CD8 T cells, anti-IFN!!!or anti-IL-12. The tumors become infiltrated with TIL cells and regress. This approach was potentiated by giving IL-12 and has the advantage that the responsible TAA(s) does not need to be purified and may not even be known.
In order to circumvent, the problems and short comings of gene therapy, Dr. Kam Leong in collaboration with Dr. Drew Pardol has generated biodegradable polymer microspheres to deliver immunopotentiating cytokines such as GM-CSF-containing microspheres made of gelatin and chondroitin sulfate were synthesized by complex coacervation. To gain insight into the relationship between the cytokine level and the antitumor response, the concentration of the cytokine in plasma and in tissue surrounding the microspheres was determined. The efficacy of the hybrid tumor vaccine was tested in the murine B16-F10 melanoma model and evaluated by different challenge schemes, for rate and duration of cytokine release, and the ratio of tumor antigen to cytokine.
These studies demonstrated the presence of the critical elements necessary for the success of the proposed tumor vaccine design: 1) a mild encapsulation process to produce cytokine-containing microspheres, 2) local delivery of bioactive cytokine, whose level and duration can be adjusted, and 3) convincing evidence that the source of cytokine does not have to originate from the same tumor cell for stimulation of antigen-specific immune responses. The results showed the GM-CSF containing beads plus irradiated tumor cells do induce much greater inflammatory responses to the tumor than control beads.
Dr. Hildegund Ertl presented the advantage of using vaccines consisting of DNA in plasmid vectors administered intramuscularly. She evaluated the development of antibodies to rabies following intramuscular immunization of mice with a plasmid vector expressing the rabies virus glycoprotein under the control of the Simian virus (SV) 40 promoter, termed pSG5rab.gp. This was shown to result in the induction of a protective immune response to rabies virus. The immune response consisted of irus neutralizing antibodies, T helper cells of the THO/TH1 type and low levels of cytolytic T cells, Although the immune response was low compared to that elicited by conventional vaccines such as inactivated rabies virus or a vaccinia rabies virus glycoprotein recombinant virus, it was exceptionally long-lasting, providing partial protection for up to about a year after a single inoculation of plasmid.
To boost the immunizing effect of the DNA vaccine, the vector was supplemented with cytokine DNA. GM-CSF-DNA was found to improve the antibody response and to favor TH1 cytokine responses. Although IL-4 DNA induced low antibody and T cell responses (paradoxically of the TH1 type), the antibody response was long lasting and had increased by 5 months. IFN!!!DNA resulted in a lower response, while the DNA for the p40 chain of IL-12 was more effective and p35 DNA inhibitory. IL-10 DNA also improved the response but IL-5 DNA had no effect. Overall, the technique of genetic immunization has a number of advantages over methods that have been developed to address this issue. First of all, this method is exceedingly simple only requiring some basic knowledge in construction and purification of plasmids and a system to test antigen-specific T and B cells responses. Secondly, this method mimics physiological events better than other more complex approaches. Genetic immunization has the advantage of providing cytokines locally together with the antigen. Disadvantages such as our limited knowledge of the levels or the duration of cytokine secretion can be addressed experimentally by determining the dose responses or by the inducible promoters. Additional parameters need to be explored to increase the rate and degree of immune response in these studies.
The final presentation of the meeting by Dr. Hideaki Tahara, a collaborator from Dr. Lotze’s laboratory, evaluated cytokine gene therapy in a number of tumor models in mice and man. He reported that tumors transfected with IL-12 are rejected. Dr. Tahara had similar results using murine IL-10. In contrast, viral IL-10 expressed by the BCRF1 gene of the EBV virus, which only has immunosuppressive and B cell stimulating effects of mammalian IL-0 promoted tumor immunity. They have initiated gene therapy of cancer patients using IL-12 or IL-4 and the results are pending. With that, the conference closed with the hope that a future meeting will focus on the role of cytokines in cancer when warranted by further progress in this specialized area of tumor immunology.

PARTICIPANTS

UNITED STATES

Dr. Joost J. Oppenheim
Chief, Laboratory of Molecular Immunology
Bldg. 560, Room 21-89A
National Cancer Institute
Frederick Cancer Research and Development Center
Frederick, MD 21702-1201
Tel: 301-846-1551 Fax: 301-846-7042

Dr. Harold F. Dvorak
Chief, Dept. of Pathology
Beth Israel Hospital
330 Brookline Avenue
Boston, MA 02215
Tel: 617-667-4343 Fax: 617-667-2943

Dr. Vishva M. Dixit
The University of Michigan Medical School
Dept. of Pathology
1301 Catherine Road
Medical Science Research Bldg.
Room 7520
Ann Arbor, MI
Tel: 313-747-0264 Fax: 313-764-4308

Dr. Barbara Ensoli
Laboratory of Tumor Cell Biology
National Cancer Institute
Bldg. 37, Room 6D06
9000 Rockville Pike
Bethesda, MD 20892
Tel: 301-402-0442 Fax: 301-496-8394

Dr. Hildegund C. J. Etrl
The Wistar Institute
3601 Spruce Street
Philadelphia, PA 19104-4268
Tel: 215-898-3700

Dr. Isaiah J. Fidler
Dept. of Cell Biology-173
University of Texas
MD Anderson Cancer Center
1515 Holcombe Blvd.
Houston, TX 77030
Tel: 713-792-8577 Fax: 713-792-8747

Dr. Robert S. Kerbel
Sunnybrook Health Science Centre
Reichmann Research Building, S-218
2075 Bayview Avenue
Toronto, Ontario M4N 3M5
Canada
Tel: 416-480-6100, ext. 5711 Fax: 416-480-5703

Dr. Kam W. Leong
Dept. of Biomedical Engineering
Johns Hopkins University
720 Rutland Avenue
Baltimore, MD 21205-2196
Tel: 410-955-3131 Fax: 410-955-0549

Dr. Michael T. Lotze
Molecular Genetics and Biochemistry
Pittsburgh Cancer Institute
University of Pittsburgh Medical Center
300 Kaufmann Bldg.
3471 Fifth Avenue
Pittsburgh, PA 15213
Tel: 412-624-9375 Fax: 412-624-1172

Dr. Sanford Markowitz
Case Western Reserve University
U.C.R.C. #2, Room 200
11001 Cedar Road
Cleveland, OH 44106
Tel: 216-844-8236 Fax: 216-844-8230

Dr. Garth L. Nicolson
Department of Tumor Biology
The University of Texas
M.D. Anderson Cancer Center
1515 Holcombe Blvd. Box 108
Houston, TX 77030
Tel: 713-792-7477 Fax: 713-794-0209

Dr. Michael S. O’Reilly
Children’s Hospital
Enders 10
300 Longwood Avenue
Boston, MA 02115
Tel: 617-355-6084 Fax: 617-335-7043

Dr. Steven A. Rosenberg
Chief, Surgery Branch
National Cancer Institute
Bldg, 10, Room 2B44
9000 Rockville Pike
Bethesda, MD 20892
Tel: 310-496-4164

Dr. Robert M. Strieter
University of Michigan Medical Center
Dept. of Internal Medicine
Division of Pulmonary and Critical Care Medicine
3916 Taubman Center
Ann Arbor, MI 48109-0360
Tel: 313-936-9370 Fax: 313-764-4556

Dr. Hideaki Tahara
University of Pittsburgh
Dept. of Surgery
300 Kaufmann Bldg.
3471 Fifth Avenue
Pittsburgh, PA 15213
Tel: 412-692-2126


Dr. Stephen E. Ullrich
University of Texas
MD Anderson Medical Center
Dept. of Immunology-178
1515 Holcombe Blvd.
Houston, TX 77030
Tel: 713-792-8593 Fax: 713-745-1633

Dr. Theresa L. Whiteside
Director, Immunologic Monitoring Laboratory
Pittsburgh Cancer Institute
Biomedical Science Tower
DeSoto & O’hara Street
10th Floor West Wing
Pittsburgh, PA 15213
Tel: 412-624-0080 Fax: 412-624-0264

Dr. Robert H. Wiltrout
National Cancer Institute
Frederick Cancer Research and Development Center
Bldg. 560, Room 31-93
Frederick, MD 21702-1201
Tel: 301-846-1258 Fax: 301-846-1673

JAPAN

Dr. Hiromi Fujiwara
Division of Oncogenesis
Department of Oncology
Biomedical Research Center
Osaka University Medical School
2-2, Yamadaoka, Suita,
Osaka 565

Dr. Toshiyuki Hamaoka
Division of Oncogenesis
Department of Oncology
Biomedical Research Center
Osaka University Medical School
2-2, Yamadaoka, Suita,
Osaka 565

Dr. Kouji Matsushima
Department of Pharmacology
Cancer Research Institute
Kanazawa University
13- 1, Takaramachi , Kanazawa 920

Dr. Shigekazu Nagata
Department of Genetics
Osaka University Medical School
2-2, Yamadaoka, Suita
Osaka 565

Dr. Takashi Nishimura
Division of Host Defense Mechanisms
Department of Immunology
Tokai University, School of Medicine
Boheseidai, Isehara,
Kanagawa 259-11

Dr. Fujio Suzuki
Department of Biochemistry
Faculty of Dentistry, Osaka University
1-8 Yamadaoka, Suita,
Osaka 565

Dr. Takashi Yokota
Department of Stem Cell Regulation
Institute of Medical Science
University of Tokyo
4-6-1, Shirokanedai. Minato-ku