(1) Seminar on “Activation of Antitumor Effector Mechanisms by Lymphokines and Biological Response Modifiers”
This meeting was held in Makaha, Hawaii, February 13-15, 1989. Eight participants from Japan and eight from the United States took part in this meeting which focused upon the mechanisms underlying the host immune response, and emphasized the manner in which biologic response modifiers can influence the host antitumor response. The meeting was organized into four major topics.
I. Hematopoiesis and lymphoid differentiation
This session included four presentations which discussed recent findings relevant to hematopoietic and lymphoid cell differentiation. Dr. Irving Weissman presented recent studies which have succeeded in phenotypically characterizing the bone marrow stem cell which has self-renewing precursor activity for all the bone marrow-derived cell lineages. This bone marrow stem cell is defined by absence of lineage-specific cell surface markers and by the expression of Sca 1 antigen. Dr. Weissman also described thymic stem cell differentiation and presented data concerning the requirements for positive and negative selection of antigen specific T cells. A modification of the SCID/HU system for studying human stem cell differentiation in the SCID mouse has also been developed by Dr. Weissman. This system involves the transfer of human fetal liver segment in addition to human T cells and allows the long-term survival of human T cells in the mouse. In this model, infection with HIV can be accomplished, providing a system for studying the human AIDS model in the mouse.
Dr. Ada Kruisbeek described two aspects of T cell repertoire selection in the thymus. She first described a set of studies in which a role of IL-2 and of the IL-2 receptor in thymic differentiation was shown. In addition, she described experimental evidence supporting the role of MHC class I and II determinants as well as the CD4 and CD8 cell surface molecules in T cell repertoire selection. The signaling role of the CD4 molecule and its associated tyrosine kinase lck molecule were described. Positive selection of T cell receptor V!!
!17 expression by class I determinants was demonstrated in the SJL mouse in a separate series of experiments employing in vivo treatment with anti-class I antibody. Dr. Hiromi Fujiwara described a newly identified thymic stromal cell-derived T cell growth factor and its putative role in thymic T cell selection. A factor isolated from thymus stromal cell supernatants and from a cloned thymus stromal cell line was shown to be active as a T cell growth factor, which is distinct from previously described lymphokines. Data were then presented, which demonstrated that the factor-producing stromal cell can eliminate cloned T cells, which are specific for antigens expressed on the stromal cells. This system thus represents a model in which self-reactive T cells would be eliminated as a result of their recognition of determinants expressed on thymus stromal cells.
Characteristics of the!!
!and!!
!T cell receptor repertoires were described respectively by Dr. Richard Hodes and Dr. Jeffrey Bluestone. Dr. Hodes described the decreased expression of mRNA specific for certainV!!!'s in selected strains of inbred mice. Substantial reductions were shown in at least 8 of the 16V!!!families analyzed to date. It was shown that deletion of specificV!!!'s correlated with expression of certain self determinants. To date, these self-determinants have included the E!!! E!!!, Mlsa and Mlsc determinants. The deletion of expression of each of theseV!!!segments was dominant and occurred in F1 animals made between deleting and non-deleting strains. The interpretation of these findings is that deletion of T cell receptorV!!!expression occurs as a result of negative selection and tolerance to self-determinants. Dr. Bluestone described several characteristics of!!
!expressing T cells. There appeared to be relatively little difference in expression patterns between strains. A role for the!!!receptor in specific antigen recognition was suggested by studies in which!!!positive,!!!negative T cell lines were generated with specificity for alloantigens. These antigens have been mapped as MHC products and further characterization is now in progress. II. Anti-tumor effector mechanisms
The second session of this symposium dealt with the cellular effector mechanisms which are active in the host response to tumors. Dr. Ko Okumura described the generation of lymphokine-activated killer (IAK) cells, and compared the specificities and activation requirements of IAK cells with those of a CTL line which had acquired broad AK-like killing potential during culture in IL-2 only. The use of bispecific cross-linked antibodies to target the cytotoxic activity of IAK cells was described. Clinical studies have been carried out employing conjugates of anti-CD3 with antibody to human glioma cells. These conjugates have been administered through Onunaya reservoirs into the cerebrospinal fluid of patients with gliomas. Preliminary data indicate that there is a high proportion of tumor regressions in those patients who received both IAK cells and bispecific cross-1inked antibodies.
Dr. Paul Sondel described both specific and nonspecific cytotoxic effector cells in vitro. LAK cell lines were analyzed for their specificity of killing and were found to differ from one another in relative lytic effect upon different target cells. The potential role of bone marrow T cells in antileukemia effect was discussed, and the mechanism by which T cells might mediate such an antitumor effect was speculated upon.
The role of distinct T cell subsets in recognition and elimination of tumor cells was described by Dr. Philip Greenberg. It was found that in vivo either CD4+ or CD8+ cells can be effective against tumors, even for Class II negative tumors. The repertoires of these two T cell subsets in the response to a viral induced leukemia were shown to be quite different. CD4+ T cells were specific mainly for env viral products, whereas CD8+ CTL responded entirely to gag products in association with Class I products. The design of vaccines intended to protect against viral induced tumors was described based upon these findings.
The effector mechanisms against tumor cells in patients with lung cancer were discussed by Dr. Kosei Yasumoto. Lavage was used to harvest both non-adherent lymphocytes and adherent macrophages from patients with lung cancer. In addition, autologous tumor cells were isolated from the same patients. Cyiostatic but not cytolytic activity was seen in macrophages but not in pleural effusions of patients by direct in vivo introduction of rIL-2. In comparison to historical controls, this IL-2 therapy appeared to improve survival in some patients.
Dr. David Segal described targeted cellular cytotoxicity using bispecific antibodies. Such antibodies were used to target ADCC and CIL effector function. This targeting appeared to require that antibody be directed against receptor structures on the effector cells, which included T cells, K cells, monocytes, and neutrophils. Percoll fractionations separated the bulk of T cell and K cell effector activity from NK cell function, and these diverse populations were shown to be differentially activated to increased cytotoxic activity by preincubation with appropriate antibodies and/or lymphokines.
The nature of lymphocyte homing receptors was discussed by Dr. Irving Weissman. T and B cells, as well as their tumor equivalents, were found to have specific homing preferences for target organs in vivo. Homing preference for peripheral lymph nodes was associated with expression on lymphoid cells of the MEL-14 determinant, whereas homing preference for Peyers patches was associated with a different determinant recognized by the monoclonal antibody LPAM-1. These homing-associated determinants are expressed on resting B cells and T cells but appear to be lost after activation. Such receptors are potentially relevant both for the homing of antitumor effector cells to appropriate sites and for the metastatic spread of tumor cells.
III. Effect of lymphokines on effector cell function
Dr. Junji Yodoi described the functional analysis of an adult T cell leukemia-derived factor (ADF) and the cloning of the gene encoding ADF. This factor which is produced by ATL has the activity of inducing increased IL-2 receptor expression on T cells, including ATL cells. It was shown that B cell lines also make a factor with properties similar to ADF. ADF treatment is capable of inducing NK activity as well as enhancing anti-CD3 T cell activation. This activity may thus play a role in normal physiologic immune regulation as well as being relevant to the malignant behavior of certain T cell leukemias.
Dr. Paul Sondel described the in vitro immunologic assessment of patients receiving biologic response modifier treatments. Using these parameters it was demonstrated that continuous infusion of IL-2 was much more effective than bolus treatment in giving rebound increases in LAK activity in patients. The use of multiple cycles of IL-2 infusion was associated with only moderate toxicity and with significantly increased in vitro LAK cell activity. However the same treatment resulted in decreased proliferative responses to PHA and to allogeneic stimulator cells and a very much reduced in vitro allogeneic CTL response.
Dr. Takeshi Ogura described regulation of IL-2 induced LAK activity by macrophages and the implications of this phenomenon for immunotherapy. It was found that resting peripheral blood monocytes enhanced LAK cell generation but that alveolar macrophages inhibited this activity. Experiments indicated that the activation state of the monocyte/macrophage cell was related to this difference. This phenomenon was studied using pleural fluid from patients with malignant effusions, and LAK therapy applied locally.
IV. Preclinical and clinical immunotherapy
Murine models of bone marrow transplantation were discussed by Dr. James Ferrara. A mouse model of bone marrow transplantation across MHC and non-MHC differences was described. In a parent into F1 model, it was shown that treatment with anti-asialo GM1 in vivo after transplantation improved bone marrow engraftment. It was also shown that treatment with anti-LFA1 in vivo decreased the amount of host irradiation which was required to achieve an engraftment. The phenomenon of GVHD was analyzed in a non-MHC model. Here it was shown that in vivo treatment with anti-mouse IL-2 receptor antibody prevents GVHD, and works only early after bone marrow transplantation.
Cytomegalovirus infection is a common and often lethal complication of bone marrow transplantation. Dr. Philip Greenberg reported studies analyzing T cell responses to cytomegalovirus. It was found that the immediate early (IE) product of this virus is a highly dominant target of the human immune response, and that a single MHC product is also important to this response. Together, these findings suggest that there may be a highly limited clonality of the T cell response to this virus. These considerations are now being employed in design of potential vaccines.
The possible therapeutic usefulness of IL-1 in treatment of chemoradiotherapy-induced myelosuppression was discussed by Dr. Yoshikatsu Hirai. Human IL-1 injected into mice produced a significant increase in serum CSF activity and accelerated recovery of granulocyte count following treatment with 5FU or with radiation. The IL-1 receptor on bone marrow cells was also characterized by binding of radio-labeled IL-1 and was found to be similar to the receptors reportedly expressed on B cells.
The antitumor activity mediated by targeted effector cells was discussed by Dr. David Segal. Anti-CD3 x antihuman tumor antibody conjugates were employed to study antitumor effector activity both in vitro and in vivo. In a Winn tumor neutralization assay, targeted PBL effectors were very effective in inhibiting subcutaneous tumor growth. CTL as well as targeted K effectors would function in the assay. CD8-depleted PBL had no in vitro cytotoxic activity in this assay and yet had persistent in vivo function. An in vitro tumor growth inhibition assay also demonstrated the ability of CD8-depleted cells to inhibit tumor growth in the absence of cytolytic activity. The supernatant derived from anti-CD3 activated PBL was found to block tumor growth in the absence of intact PBL. The identity of this cytokine is now being analyzed. A model for treating established ovarian cancer was then described. A human tumor was grown in nude mice, where it produced a peritoneal effusion before metastasizing and killing its host. It was demonstrated that human PBL could be targeted with anti-CD3 conjugated to an antitumor antibody and that this mode of therapy was highly effective in curing mice of this human tumor. The application of this model to human ovarian carcinoma is thus feasible.
Dr. Jeffrey Bluestone described the enhanced tumor-specific immunity found in mice which had been treated with anti-CD3 in vivo. When doses of this antibody were selected which are too low to induce massive modulation of CD3, T cell activation was still produced in vivo. Under these conditions, an increased MLR response as well as enhanced IL-2 receptor expression were seen. In mice challenged with a UV-induced sarcoma, treatment with this low dose activating anti-CD3 at the time of tumor challenge resulted in protection against tumor growth in nearly all challenged animals. Analysis of the mechanism of this effect suggests that anti-CD3 activation makes naive T cells more responsive to antigens including tumor antigen resulting in an enhanced antitumor response.
Dr. Toshiyuki Hamaoka described the identification of antigens expressed on tumor cells which are important in host resistance to these tumors. Biochemical fractionation of a mouse sarcoma, coupled with biological evaluation of the ability of these fractions to induce tumor resistance has resulted in the isolation of a 66kD protein which is highly efficient in protecting animals against subsequent tumor challenge. Antigen presenting cells can be pulsed with this tumor resistance antigen, and repeated immunizations with these pulsed APC either subcutaneously, intraperitoneally or intravenously resulted in effective tumor protection in the absence of pretreatment with intact tumor cells. This approach was discussed as potentially useful for immunization to human tumor antigens.
Dr. Kozo Imai described the use of anti-idiotypic networks to induce immune responses to tumors. An anti-idiotypic antibody which appears to represent an internal image of tumor antigen was utilized in a model system in which the antigen was carcinoembryonic antigen. Panels of anti-idiotypes were used to induce anti-anti-idiotypes. These latter antibodies were then compared to antibodies specific for peptides of the now cloned CEA molecule. Efforts have now been turned to utilize monoclonal antibodies to human melanoma high molecular weight antigens. Anti-idiotypes have been made and injected into patients who are now being evaluated for clinical effects.
The entire symposium was marked by extremely active discussion involving all participants. In several areas, the progress in basic understanding of immune mechanisms has now been applied to in vivo analysis and even treatment of cancer patients. Nonspecific effector mechanisms, "targeted" effector mechanism as well as the induction of tumor antigen-specific immune responses are all being analyzed for the potential contributions which they may make to tumor therapy. The potential of modalities such as bone marrow transplantation after ablative antitumor therapy and the use of cytokines as biologic responses modifiers were prominant parts of this discussion. The ability of the meeting to include participants who are active in clinical tumor therapy as well as leading basic scientists in the area of immunology provided a uniquely productive interaction in these areas.
(2) Oncogenes, Growth Factors and Receptors
The US-Japan Cooperative Cancer Research Program meeting entitled "Oncogenes, Growth Factors and Receptors" was held from January 30 - February 1, 1989, at the Jozankei View Hotel in Japan. This meeting which was organized by Drs. Tadamitsu Kishimoto and Stanley Korsemeyer involved the participation of six U.S. and ten Japanese scientists.
The first session was entitled "DNA Binding Proteins and Regulation of Gene Expression." Dr. Tadatsugu Taniguchi from the Institute of Molecular and Cell Biology at Osaka University, discussed his fascinating data concerning two interferon regulatory factor genes IRF-I and IRF-2 involved in the differential regulation of the interferon locus. These genes were isolated by oligonucleotide screening of!!
!gt11 library and both recognized the same upstream motif of AAGTGA. Importantly, IRF-1 when transfected into target cells has been proven to be capable to activating the interferon gene. Provocatively, IRF-2 appears to play a down-regulatory role for that same locus. The amino acid sequence and structure of IRF-1 and IRF-2 are similar but have several different regions that appear to be very good candidates for the differential regulation of this gene. Dr. Shizuo Akira also from Osaka University, followed with a description of the nuclear factors responsible for IL-1 induction of IL-6. He detailed the important transcriptional interrelationship of these two interleukins identifying cis acting regulatory elements and transacting factors responsible for IL-I effects upon IL-6 transcription. The IL-1 responsive element mapped to a 14 bp area of dyad symmetry at -180 of the IL-6 promoter region. Provocatively, this area has homology with the C-fos serum response element. Three factors were purified with molecular weights of 86, 76, and 71 Kd of which the 86 Kd cross-links to the area of dyad symmetry. Candidate genes have been obtained from a!!!gt11 library that will provide further details of the intermediate regulators in this important system. The morning session concluded with a talk by Dr. Warner Greene of Duke University concerning DNA binding proteins involved in T-cell activation. He reviewed progress in three arenas related to the!!!-chain of the IL-2 receptor, the regulation of the!!!-chain gene of the IL-2 receptor and the relationship of Rev and Rex proteins in affecting the splicing and transport of HTLV-1 messages. Monoclonal antibodies were established against the!!!-chain of the IL-2 receptor revealing 70-80 Kd phosphorylatd proteins within cells bearing intermediate affinity receptors. The transacting factors that bind to the NFkB site in the IL-2 !!!-chain were further subdivided into two inducible proteins of 51 Kd and 86 Kd and two constitutive proteins of 90 and 56 Kd. Further data was presented suggesting that there are unique factors beyond NKkB responsible for regulation of IL-2 receptors. The third area of investigation revealed that the presence of the p27 Rex protein resulted in the accumulation of a large amount of unsliced large RNAs responsible for the production of gag, pol, and envelope proteins preceeding replication. Moreover, the 19-20 Kd Rev protein of HIV could substitute for Rex in this function and both of these proteins localize to the nucleolus.The following session was related to growth factors and receptors and was initiated by Dr. Toshimitsu Uede of Sapporo Medical College. He detailed their studies of a murine monoclonal antibody, 8H3, that recognizes a cell surface structure on rat thymocytes that performs an integrin molecule function. Specifically, the 8H3 antibody would block the binding of thymocytes to fibronectin and crosslinking of this surface antigen!!
!structure would lead to the activation of the cells as measured by Ca- influx. This molecule appears to be relatively unique in that it does not possess the classic!!!1 subunit. Professor Tadamitsu Kishimoto from Osaka University followed with a detailed presentation of the IL-6 receptor and its mechanisms of signal transduction. The cloned gene for the IL-6 receptor indicated a single transmembrane spanning segment and a substantial intracytoplasmic tail while binding and cross-linking experiments indicated that it bound a single IL-6 molecule. Quite uniquely, the interaction of IL-6 with its receptor prompts the association of a second protein in this complex of 130 Kd. Deletion mutants which eliminated the intracytoplasmic tail of the IL-6 receptor still resulted in cellular activation indicating that the second chain is likely to be responsible for signal transduction. Continuing this same theme, Dr. Kazuo Sugamura of Tohoku University School of Medicine described their studies on the human IL-2 receptor subunits. Their IL-2 cross-linking experiments suggested the presence of two proteins P75 and P70 corresponding to the!!!-chains of the IL-2 receptor. They prepared two monoclonal antibodies known as TU27 and TU11 capable of discriminating between the P75 and P70. This finding suggests that the high affinity IL-2 receptor might be comprised of three subunits and that perhaps the intermediate affinity receptor could be a heterodimer of!!!+!!!IL-2 receptor chains. The next session was concerned with T-cell receptors. Dr. Yasunobu Yoshikai of Kyushu University presented his studies concerning T cell receptors in athymic nude mice and aged mice. An increase in the number of!!
!type T cells was noted in mice as they aged. In addition, the expression of!!!T cell receptors may be related to antigenic exposure within the gut associated lymphoid tissues.!!!T cells were also noted in nude mice suggesting that their maturation might occur in extra thymic environments. Dr. Michael Brenner from Harvard Medical School described the distribution of!!!T cell receptors in humans.!!!T cells constitute a variable percentage from 1-16 percent of CD3 positive cells in different individuals. They constitute a phenotypically diverse T cell population distributed throughout the body. In humans, Dr. Brenner's group has not found a dramatic trophism of!!!cells for the gut epithelium. Likewise while occasional epidermal and interepithelium!!!cells are seen there is not a big involvement in the dermis of the skin. However, preferential associations of specific!!!and!!!variable regions might well occur, supporting the concept that!!!T cells substantially broaden the T cell repertoire at relevant sites. Dr. Hideo Yamagishi from Kyoto University followed with a detailed description of the excision products of the!!!T cell receptor locus. Dr. Yamagishi characterized clones from small polydisperse circular DNA isolated from mouse thymocytes containing the extra chromosomal signal joints from!!!and!!!T cell receptor rearrangements. Many of these contain signal sequences flanking variable!!!and joining!!!gene segments including newV!!!regions. They have noticed murine homologs of the T cell receptor deleting elements noted in man and have found that these can be involved in rearrangements into the!!!T cell receptor locus.The next session of the meeting concerned. signal transduction and protein kinases. Dr. Roger Perlmutter of the University of Washington described the differential regulation of the LCK gene that encodes the lymphocyte specific tyrosine kinase. Translational masking of upstream initiation codons appears to play an integral role in the regulation of this gene, and murine T cell lymphomas have been noted in which retrovirus integration eliminated these upstream sites. A transgenic mouse model had been created utilizing the LCK gene missing its upstream initiation codons and the initial data suggest that activated LCK may play a role in eliminating T cells in a fashion parallel to the deletion of self-reactive clones. Dr. Kathleen Kelly of the National Cancer Institute presented her investigations on the expression and regulation of mitogen induced genes in T cells. A large panel of mitogen induced genes were obtained by subtractive library approaches using activated T cells. These fit into categories of cyclohexamide independent primary gene transcription and those requiring protein synthesis as secondary gene transcription events. The genes can be further subdivided into various regulatory groups related to their differential response to PHA, PMA cyclohexamide, and cyclosporin A. Importantly, two new genes that are quite homologous at the protein level and appear to be homologous to the JE family of genes induced by PDGF are both secreted and appear to be new cytokines. Dr. Shigeo Ohno of the Tokyo Metropolitan Institute of Medical Science completed this session with an indepth discussion of the structure, expression and function of the protein kinase C-family. PKC is composed of four distinct molecular types!!
!, I,!!!II, and!!!encoded by three distinct genes. They all possess the same zinc finger motif and confer phorbol ester receptor activity to intact cells when transfected. Moreover, he has identified at least three additional members of a PKC-related family. This latter group is independent of calcium and provides tools to dissect their function when placed into conditional mutants. The final session was entitled Transgenic Mice and Oncogenes. Dr. Ricardo Della-Favera of New York University summarized their work on oncogenes in human B cell transformation. Utilizing a very sensitive cDNA-initiated PCR direct sequencing approach they characterized the 5' region of the myc oncogene in 40 cases of Burkitt's lymphoma. All cases possessed somatic mutation surrounding the PvU-2 site area of exon I. This region correlates well with the loss of transcription attenuation at this site. The introduction of myc into Epstein-Barr Virus lines was found to decrease the expression of LFA-1, an integrin family molecule involved in the stimulation of autologous and mixed lymphocyte responses. This appeared to be mediated at the level of!!
!but not the!!!chain of LFA in that the transcription of!!!was decreased in the presence of exogenous myc. Finally, the presence of mutated ras genes was shown at both ends of the B cell malignancy spectrum in pre-B cell ALLs and in multiple myeloma in a substantial portion but was entirely missing in the mature B cell neoplasms of CLL non-Hodgkin's lymphomas and Burkitt's lymphoma. Dr. Stanley Korsmeyer of Washington University School of Medicine followed with a discussion of their Bcl2-Ig transgenic mice. Mini-gene constructs which recapitulated the pathologic consequence of the t(14;18) were placed transgenically into the germline of mice. The expression of the transgenes resulted in a polyclonal expansion of early IgM/IgD follicular center B cells. The transgene conveyed a death sparing extended survival to this mature B cell population which was now capable of prolonged culture in vitro. In addition, data were presented using retroviral vectors indicating that Bcl-2 was capable of complementing the myc cellular oncogene in growth and neoplasia. Dr. Shinichi Aizawa of Tsukuba Life Science Center discussed tumor progression in transgenic mice. He created mice utilizing the SV40 promoter and immunoglobulin enhancer in an expression construct that contained either SV40T, myc, erB-2, Ras, Raf or myb. The erB-2 myc and Ras animals have all developed neoplasia. Moreover, mice with the erB constructs have demonstrated a growth retardation with abnormalities at the epiphyseal bone plates. The SV40T antigen mice were noticed to have an increase in B lymphocytes and in vitro CFU colony formation. Provocatively, the oncogenes separated into two groups, one in which there was expression in normal tissues and another set characteristic of erB in which the transgenes only seemed to be expressed within tumor development. Dr. Kenichi Yamamura of Kumamoto University School of Medicine concluded the meeting with a discussion of factors that affect the tumor phenotype in E-myc transgenic mice. Dr. Yamamura has further dissected the host-cell dependence of transgene oncogenesis. When the E-myc gene is placed into C57BL/6 mice they develop B cell lymphomas. In contrast, the same transgene within C3H mice leads to development of T cell tumors. Transplantation studies involving transgenic bone marrow or fetal liver from transgenic mice placed into irradiated normal hosts suggest that the micro-environment in C3H mice may predispose to the development of T cell tumors. (3) Seminar on Cytochrome P450 and Carcinogenesis
This seminar was held from January 16 to 19, 1989, at the Turtle Bay Hilton, Kahuku, Hawaii. The organizers of the seminar were Dr. Harry V. Gelboin, National Cancer Institute, Bethesda, Maryland, and Dr. Yoshiaki Fujii-Kuriyama, Tohoku University, Sendai. There were ten participants from the United States and eight from Japan.
It is well established that a very large proportion, more than 95%, of all chemical carcinogens require metabolic activation to their procarcinogen form. The primary, but not the only system, responsible for this activation is the cytochrome P450 enzymes. A very important question posed was whether the P450-based carcinogen activation is the rate limiting step in carcinogenesis and if variation in P450 level governs individual susceptibility and sensitivity to carcinogenic compounds. The answer to this question is the primary research goal in this field. In consideration of the recent remarkable progress in the molecular biology of cytochrome P450, the seminar was organized to review the latter and to summarize the present status of studies on cytochrome P450 in relationship to carcinogenesis and to define its role in human cancer susceptibility.
Welcoming remarks were delivered by Dr. Gelboin together with a few comments on the purpose and organization of this meeting. The first speaker, Dr. Fujii-Kuriyama, Tohoku University, Sendai, summarized structural information on various forms of cytochrome P450. Since the primary structures of cytochrome P450, P450b and P450cam were first reported in 1982, more than 100 structures of cytochrome P450s from various organisms have been reported. All these structures show statistically significant similarity with one another and constitute a distinct superfamily of P450 with 11 families, according to Dayhoff’s definition. On the basis of the sequence similarity, evolutionary relations among these P450s were summarized in a phylogenetic tree, suggesting that the drug-metabolizing P450s, P450b, e, c and d, which are also involved in the metabolic activation of carcinogens, appeared 4x108 years ago. He also reported that at least two kinds of cis-acting factors are necessary for the high level of inducible expression of P450c which activates aromatic carcinogens.
Dr. Frank Gonaalez, National Cancer Institute, Bethesda, Maryland, described his recent results with P450II E1 which shows strong activities for metabolic activation of a nitrosamine as well as ketone body metabolism. Orthologous species of P450 were found constitutively in adult human, rabbit, and rat. Transcriptional activation of P450II E1 gene soon after birth was associated with specific demethylation of cytosine residues that are located upstream of the promoter region and in the second exon. When the gene was fully expressed in adult animals, the regulation of the gene expression occurred posttranscriptionally. Under certain circumstances like fasting or diabetes, the P450II E1 mRNA was markedly stabilized. When animals were given acetone or alcohol, the turnover rate of the P450II E1 protein was substantially decreased.
After reviewing briefly the papers dealing with carcinogenesis and the properties of 18 purified human P450s, Dr. Peter Guengerich, Vanderbilt University, Nashville, Tennessee, talked about his recent results on metabolic activation of various chemical carcinogens using purified preparations of human P450s. Although P450DB, P450MP- 1, or P450MP-2 did not show any noticeable activity for the metabolic activation of the carcinogens tested, the most prominent human P450 enzymes involved were P450NF (activation of aflatoxin B1 and G1, sterigmatocystin, benzo(a)pyrene-7,8-diol, 6-amino-chrysene and tris [2,3-dibromopropyl phosphate]), P450PA (2-aminofluorene, 2-acetylaminofluorene, 4aminobiphenyl, and the food pyrolysate products IQ, MeIQ, MeIQx Glu-P-1 and Glu-P-2), P450j (several N-methyl substituted N-alkylnitrosamines) and P1-450 (conversion of benao(a)pyrene to its 7,8-dihydrodiol). These results indicate several distinct differences between the activation of carcinogens by humans and the orthologs studied in experimental animals, and suggest cautious interpretation of the results obtained with experimental animals and that they can not easily be applied in the case of humans.
Dr. Yoshihiko Funae purified 25 species of cytochrome P450 from rats by HPLC using ion-exchange or hydroxyapatite columns and examined their activities in oxidative activation of various nitrosamine derivatives, bena(a)pyrene and aflatoxin B1. Dimethylnitrosamine was activated by P450j, while other nitrosamine derivatives were metabolized by other forms of P450s (phenylmethylnitrosamine, P450d and b; ethylmethylnitrosamine, P450UT-2 and d; butylmethylnitrosamine, P450UT-1 and UT-4, etc). Aflatoxin B1 was actively metabolized to genotoxic products by several constitutive forms of P450s (P450UT-2, UT-3, UT-7 and F-1) prepared from normal rats.
Large, as yet undefined, numbers of P450s with various degrees of sequence similarity are present in a single species of animal. It is often impossible to biochemically separate the P450 forms with minimal sequence variance. However, molecular cloning technology enables us to achieve relatively easily the isolation of closely related P450s. In order to characterize isolated P450s in a CDNA form it is necessary to have good CDNA expression systems. Dr. Aoyama reported success in applying a vector composed of a TK promoter sequence from vaccinia virus for cDNA-directed expression. He found that P450c, d and a, were expressed in Hep G2 cells transfected with the respective cDNA-inserted vectors and suggested that this expression system would be useful for the expression of other P450 cDNAS.
Dr. Kaname Kawajiri, Saitama Cancer Center Research Institute, Saitama, discussed the change in the amounts of P450, epoxide hydratase, glutathione S-transferase and UDP-glucuronosyl transferase in rat liver after continuous dietary administration of 3'-methyl-4-dimethylaminoazobenzene. The mRNA content of cytochrome P450d increased in the liver soon after treatment with the drug and reached a maximum level in 3 weeks, whereas the other phase II drug-metabolizing enzymes and P450c (detoxicating 3'-methyl-DAB) increased slowly and reached a plateau at the 5-week stage of development of hyperplastic nodules. How or whether the difference in the inducibility between 3'-methyl-DAB-activating P450d and detoxicating enzymes is related to the cancer incidence remains of interest for study.
After a brief review of aromatic amines as causative agents in occupational carcinogenesis and their presence as carcinogens in tobacco smoke, synthetic fuels, agricultural chemicals, and cooked foods, Dr. Fred Kadlubar, National Center for Toxicological Research, Jefferson, Arkansas, presented evidence that rat P450U and its human ortholog, P450 (phenacetin O-deethylase), are primarily responsible for the N-oxidation of 4-amino- biphenyl, 2-naphthylamine and several heterocyclic amines. Dr. Kadlubar found that the distribution of the monooxygenase activity among humans was bimodal with the difference in the activities being about 100-fold, thereby suggesting that the expression of the monooxygenase is under genotypic or phenotypic regulation. Since the level of hepatic P450PA may be an important determinant in susceptibility of different individuals to arylamine-induced cancer, his research group sought to identify a non toxic agent that could be used to describe the human N-oxidation phenotype. He found that caffeine 3-demethylation is a selective indicator for P450PA activity. Using an HPLC method for analysis of urinary metabolites of administered caffeine, they are currently assessing the utility of this approach for metabolic phenotyping in human populations. This approach is based on sound scientific information and is expected to provide an effective tool for assessing the susceptibility of individuals to cancers.
The relationship between the level of P450d and susceptibility to arylamine-induced cancer was studied extensively in experimental animals such as mice and rats by Dr. Yoshiyuki Hashimoto, Tohoku University, Sendai. He reported that the differences in susceptibility to arylamine-induced hepatic cancer between male and female mice could be accounted for by the levels of P450d expression in the livers of female and male animals. 3-methyl-aminoazobenzene induced higher levels of P450d expression in the livers of female mice than in that of males and thus caused a higher incidence of tumors in the female. The same was true for TrpP-1 and TrpP-2. The difference in the level of P450d expression was found to be due to the level of androgens which appeared to exert a suppressive effect in expression. He concluded that the organ selectivity of cancers induced by aromatic amines was also determined by the expression of P450d in a given organ.
Dr. Tetsuya Kamataki, Hokkaido University, Sapporo, discussed the purification and enzymatic properties of P450s from various experimental animals such as monkey, dog and rat in order to identify orthologous species of P450. Metabolic activation of various carcinogenic substances by microsomes from various animals varied among species. Inmunological quantitation showed that the variation in activity toward GluP-1, TrpP-2 and IQ was determined by the content of P450d orthologs in the microsomes. He also reported the sequence analysis of several isolated P450 cDNAS from dog, monkey, and human.
The utility of monoclonal antibodies for quantitating specific forms of P450 was described by Dr. Sang Shin Park, National Cancer Institute, Bethesda, Maryland. He raised inhibitory monoclonal antibodies to P450c, d, b, e, j and h. Using these antibodies, he quantitated the contents of each P450 species in microsomes from various manmalian sources and found that more than 90 percent of the AHH activity of placentas from women who smoke cigarettes was inhibited by antibody 1-7-1 that is cross-reactive with P450c and d. This result indicated that cigarette smoking actually induced P450c or d or both in the human body.
Not only exogenous but also endogenous substances are known to be involved in the regulatory expression of P450 genes. Dr. Ryuishi Kato, Keio University, Tokyo, presented data on the expression of P450 genes by endogenous substances. Hypophysectomy stimulated the synthesis of P450b, P450e and P450j in rats of both sexes and P450(M-1) (male-specific form) in female rats, whereas it reduced the expression of P450(M-1) in male rats and P450 F-1 (female-specific form) in female rats. Consequently, hypophysectomy obscured the sex-specific difference in the expression of P450. Neither androgen nor estrogen reversed the effects of hypophysectomy. The administration of growth hormone mimicked both the male and female-specific secretion pattern and restored the normal levels of P450s in hypophysectomized rats.
The meeting also developed the fact that there are several new extrahepatic P450s that seem specific for nasopharyngeal and lung tissues. Dr. Xinxin Ding, University of Michigan, Ann Arbor, Michigan, gave a brief review on nasal cancers and carcinogens and reported his recent work on the purification and properties of rabbit nasal P450s. Multiple forms of P450 were identified in olfactory and respiratory nasal microsomes by immunoblot analysis and purification procedures. The olfactory mucosa contains two new forns of P450NMa and NMb as well as isozyme 4 (originally found in livers) as major P450 components and several other forms of P450 found in livers as minor ones. On the other hand, the respiratory mucosa contains mainly two forms of P450, P450NMa and isozyme 2. P450NMa and NMb show metabolizing activities toward specific nasal carcinogens such as hexamethylphosphoramide and phenacetin, though the former P450 is much more predominant.
Dr. Minro Watanabe, Tohoku University, Sendai, talked about P450s in the lungs of rats and hamsters treated with 3-methylcholanthrene, since lungs seem to be more susceptible to carcinogenesis induced by polycyclic aromatic hydrocarbons. A major form of P450 in the lungs of 3-methylcholanthrene-treated rats appeared to differ slightly in peptide mapping from the P450 counterpart of liver. Treatment with 3-methylcholanthrene could induce benao(a)pyrene hydroxylase activity in the lung of golden hamster, but not in liver. The purified methylcholanthrene-induced P450 from hamster lung resembled methylcholanthrene-inducible P450c of rat liver in its enzymatic, spectrophotometric and immunological properties.
A number of phenolic antioxidants are known to alter the course of carcinogenesis, if administered with toxic chemicals. Dr. Russell Prough, University of Louisville, Louisville, Kentucky, discussed the inhibitory mechanism of P450 monooxygenase activity by 3-t-bytyl-4-hydroxyanisole (BHA, a food additive). Using the well-characterized P450cam hydroxylase, they established that BHA enhances the decomposition of the oxyferrous, substrate-bound complex of P450cam, resulting in the inhibition of the P450 function. The oxyferrous intermediate complex of mammalian P450s is also drastically diminished in the presence of BHA and direct inhibition of P450 activity can be demonstrated. He suggested the utility of this sort of material with less cytotoxicity as anticarcinogens. He also reported that the GRE (glucocorticoid responsive element) in the first intron of the P450c gene acts synergistically with XRE to enhance the transcription of the gene.
Several widely used anticancer drugs are metabolized in the liver by the P450-dependant pathway that leads to the formation of therapeutically significant cytotoxic metabolites. Dr. David Waxman, Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, reported that the activation of cyclophosphamide was catalysed by phenobarbital-inducible P450b and constitutive P450II C6 and II C 11 and also presented evidence that thio-TEPA underwent metabolic activation by the P450-dependant oxidative pathway to express an anticancer activity. Finally, the complex changes in hepatic P450 profiles that resulted from administration of the chemotherapeutic drug cisplatin were discussed. Cisplatin treatment of adult male rats resulted in a marked and prolonged reduction of the male-specific P450s such as P4502c (or M1), 2a and RLM2 by 70 to 90%, with parallel decreases in their respectively associated hydroxylase activities. Concomitantly, hepatic levels of the female-dominant P450s, P450 3(or a) and j, were elevated two- to fourfold, while no elevation of female-specific P450 2d (or F-1) was detected. He suggested that these changes in P450 content were caused by disturbed hormonal balances in cisplatin-treated animals and pointed out that the combined administration of cisplatin with other antitumor drugs such as cyclophosphamide and thio-TEPA which need activation by the P450 system might have profound effects on the efficacy of drug therapy. Carcinogenesis is a result of multiple biological steps and, therefore, it may be very difficult to demonstrate the primary or direct involvement of P450 in chemical carcinogenesis. A very important question is still open as to whether the P450-based carcinogen activation is the rate-limiting step in carcinogenesis, and whether variations in P450 level govern individual susceptibility and resistance to carcinogenic compounds. An answer to this question is a significant goal of P450 research related to carcinogenesis. As Dr. Ronald Estabrook, University of Texas, Dallas, Texas, stated in his summary of the meeting, remarkable progress in the concept and facts about the role of P450s was presented in this meeting and provides us with more work to approach our final goal. This is a clear indication of a successful meeting.
SEMINAR AGENDA AND PARTICIPANTS
(1) SEMINAR ON ACTIVATION OF ANTITUMOR EFFECTOR MECHANISMS BY LYMPHOKAINS AND BIOLOGICAL RESPONSE MODIFIERS
Makaha, Hawaii, February 13-15, 1989
AGENDA