1998

ABSTRACTS 553

A Large Scale Screen Utilizing Chromosomal Rearrangement Mutants for Novel Genes Involved in Immunoregulation ]VI.W. APPLEBYa, M. SINGHALa, D. WALKERa, W. GENEROSOb, J.E. WILKINSONb,r F. RAMSDELLa aDarwm Molecular Corporation, Bothell, W.4 98021 bBlology Dwlston, Oak Eidge National Laboratory, Oak Ridge, TN 37831 CCollege of Veterinary Medlcme, Umverstty of Tennessee, Knoxvdle TN 37901

In order to identify novel molecules involved in the regulation of the immune system we are undertaking a study designed to identify mice bearing mutations which compromise immune function. This approach utilizes a collection of novel mutant stocks, each of which carries a single reciprocal translocation induced by ionizing radiation or chemical mutagen. The use of translocations affords several advantages; in particular the karyotypic identification of the break points provides a starting point for cloning. Most intrigning however is our observation that the frequency of visible mutations within the collection of animals is high, suggesting that induced translocations will be a powerful tool in future mutagenesis studies. A preliminary immunological characterization of these animals suggests that there are several carrying novel mutations which disturb the function of the immune system. The nature of 3 of these mutants in which B cell development and T cell development is compromised will be discussed, along with the strategies involved in the implementation of the project.

Diabetic Animal Models K. BUSCHARD Barthohn Instttuttet, Kommunehospttalet, Copenhagen, Denmark

Insulin-dependent (type 1) diabetes is a frequent disease with an incidence of up to about 1%. It requires daily treatment and serious late complications are observed. Good animal models exist for studying diabetes. These can be categorized as animals with spontaneously developing diabetes (BB rats, NOD mice) and as animals with induced diabetes (e.g., by virus). Immunodeftcient nude mice have also been widely used. None of the models is perfect, but each has contributed to our present knowledge of the disease. Studies on the pathogenesis of type 1 diabetes are given as an example. Recently, experience with prophylactic treatment of animals in order to prevent diabetes has been applied to humans with promising results.

Pig Fetusses and Germ-Free Piglets: A Gnotobiological Model for the Study of Preimmune B Cell Repertoire B. CLrKROWSKAa, J. SINKORAa, Z. I~d~HAKOVAa, A. SAAL~LERb, H. TLASKALOVA-HOGENOVAa aDtvlston of Immunology and Gnotobtology, Institute of Microbiology, 142 20 Prague, Czech Repubhc bFederal Research Centrefor Virus Diseases of Ammals, Tiibingen, Germany

The effect of perinataUy administered or transferred anti-idiotypic antibodies (Abs) on the development of the B-cell repertoire has been documented in humans and mice, but in pigs the transfer of maternal immunoglobulins (Igs) into the fetal circulation does not occur through the six-layered placenta. These animals thus represent a physiological Ig-deficient experimental model. We used pig fetuses and germ-free piglets (GF) to study the occurrence and the repertoire of natural Abs, which are formed before stimulation by external antigens. Minimal amounts of serum Igs of all isotypes were found in 44-d-old fetuses (the gestation period in pigs lasts 114 d) and their level, predominately IgM, was increased during fetal ontogeny. Ab activity against autoantigeus (thyreoglobulin, hormones, ssDNA), phylogenetically conserved proteins (myosin), haptens (TNP) and bacterial components (E. coli 086, tetanic anatoxin) was detected by ELISA in sera of fetuses at the end of embryonic life as well as of newborns and older GF piglets. The antigen binding activity of natural IgM Abs increased after isolation of serum Igs on a Staphylococcus protein A (SPA)-Sepharose column. IgM reactivity similar to that detected in serum was found in supernatants from polyclonaUy stimulated cultures of spleen of 8- and 12-d-old GF piglets. Enriched fetal liver IgM+ B cells, which were able to produce IgM after polyclonal stimulation by bacterial components, did not express the CD5 molecule as shown by two-color FACS analysis. Our results suggest that the pig preimmune repertoire is comparable to that described in humans and mice, although in contrast to these species pig B-1 cells do not express CD5.

Vol. 43

554 IV~IDA

Animal Models for Human Parasitic Diseases

M. ITO Central Institutefor ExperimentalAmmals. Kawasak1216,Japan

Parasites include a wide variety of species from protozoa to helminths, have an unique relation with their hosts termed host-parasite relationship, and live in complicated life cycles in which most of them need defmitive and intermediate hosts and have strong host-species specificity. The problems of human parasitic diseases vary. Some are caused physically by worms themselves and some result from host responses. Therefore, the establishment of animal models appropriate for individual parasites associated with their disease processes is needed. Here, our approaches to establish animal models for some human parasitic diseases using spontaneously occurring and artificially produced immunodeficient mice are presented. For helminths, studies on the development of Taenia saginata and T. solium from eggs to cysticercies in subcutaneous and peritoneal regions of SCID mice will be presented. Since the development of cysticercies of T. saginata and T. solium has been considered to date in cattle and pigs, respectively, they can offer new animal models to analyze the development of taenic worms. For protozoa, studies on the mechanisms of ulcer formation in cutaneous leishmaniasis caused by Leishmania amazonensis using OVA-TCR Tg RAG2 deficient mice will be discussed. Another topic is an attempt to establish a mouse model for human cerebral malaria. The sequestration of malaria-infected erythrocytes in brain capillaries through cell adhesion molecules is generally considered as the main cause of cerebral malaria. In order to clarify this hypothesis, Tg-SCID mice specifically expressing human ICAM-1 and ELAM-1 molecules in capillary endothelial cells have been newly developed and intravenous transfer experiments of P. falciparum-infected erythrocytes are in progress. Stromal Cell Regulation of T-Cell Development

E.J. JENKINSON,R. WILKINSON,K. HARE, J.J.T. OWEN, G. ANDERSON Department of Anatomy, MedicalSchool, The Universityof Bzrmmgham, UK

Interactions between developing thymocytes and the thymic microenvironment are important at successive stages of T-cell development. In particular, positive selection is a crucial requirement for the further development of CD4+CD8 + cortical thymocytes into single positive CD4 + or CD8 + T cells. Although it is now clear that TCR mediated interactions with peptide-MHC complexes on the thymic stroma are essential for this process, the role of the thymic microenvironment in providing other signals driving or supporting positive selection are less well defined. Using reaggregate organ cultures in which defined stromal cell types are associated with CD4+CD8 + thymocytes we have been able to show that thymic cortical epithelial cells are uniquely efficient in their ability to support positive selection suggesting that they supply additional factors required for this process. Studies using reaggregate cultures have also allowed to us define two other aspects of positive selection. Firstly, we have examined the developmental potential of CD69 +, CD4+CD8 + thymocytes that have initiated but not completed positive selection. These studies have shown that positive selection requires sustained interactions with cortical epithelial cells in addition to the initial TCR mediated triggering event. Secondly, we have found that, at least in the case of newborn cells, newly generated single positive cells undergo a wave of proliferation that may be important in the expansion of the newly selected repertoire. This wave of proliferation is also specifically dependent upon cortical epithelial cells but appears to be independent of ongoing TCR signaling. Ongoing studies are aimed at defining the nature of the support provided by thymic epithelial cells during the various phases of positive selection. In particular, using thymocytes from Bcl-2 transgenic mice, which survive independently of thymic support, we are attempting to discriminate between the role of the thymic environment in providing maintenance signals permissive for positive selection and the provision of specific differentiation signals by cortical epithelial cells. Asymmetric Involvement of ZAP-70 in Positive Selection to CD4

vs.

CD8 SP Thymocytes

Y. KAMETANI,T. SATO,K. HOZUMI,T. NISHIMURA,S. HABU Department of lmmunology, Tokm University, School ofMedicine Isehara-shi, Kanagawa,259-11Japan

TCR transgenic mice (Tg) are useful for exploring the intrathymic developing process of a certain T cell clone. To investigate the potential involvement of the tyrosine phosphorylation in the thymic positive

1998

ABSTRACTS 6S5

selection in vivo, freshly isolated thymocytes were obtained from TCR-Tg with MHC class 1 and class 11 restriction of selecting and non-selecting MHC background, respectively. The phosphorylation level of ZAP 70 in DP cells was higher in selector than in non-selector mice. This phosphorylation difference between selector and non-selector mice was considerably less in class 11 restricted Tg than in class 1 restricted Tg. A similar bias for ZAP-70 phosphorylation was also found between KO mice for MHC class 1 and of class 11. The monoclonal antibody-induced co-crosslinking on thymocytes of non-selector mice showed that ZAP-70 phosphorylation was induced more effectively through TCR mediated signaling with CD8 than CD4 or TCR-a alone. These findings indicate that in the thymic positive selection, coreceptors engage in a distinct manner for ZAP-70 phosphorylation induced by TCR signaling, resulting in the commitment to CD4 vs. CD8 lineages.

Primary Viremia and CNS Invasion with HIV-1 in a Novel hu-PBL Immunodeficient Mouse Strain

Y. KOYANAGIa, Y. TANAKAb, J.-I. KIRAc, Y. KAWANOa'c, M. ITOd,e, Y. UEYAMAd-f N. YAMAMOTOa aTokyo Medwal and Dental University, 1-5-45 Yushtma, Bunkyo-ku, Tokyo, bKttasato Universtty, Sagamthara, Kanagawa, eKyushu Untverszty, Fukuoka, dCentral Instttutefor ExpertmentalAmmals, Kanagawa, eKanagawaAcademy of Science and Technology, fDepartment of Pathology, Tokat Umverstty, Japan Aim: To develop a new in vivo infection system using immunodeficient mice with multiple defects for the study of HIV-1 infection, vaccine, and drug testing. Method: Immunodeficient mice were developed by either backcrossing of scid mutant onto the other mutant mice such as beige, nude, Dh, and NOD or knocking out genes responsible for lymphoid function such as recombination of the activating gene 2 (RAG-2). 6 new congenic BALB/cA-bg-scid, NOD/Shi-scid, BALB/cA-Dh-scid, BALB/cA-nu-scid, BALB/cA-RAG20/0 and C57BL/6-KAG20/0 were generated. Human T lymphocytes were reconstituted in the new immunodeficient mouse strains. HIV-1 was intraperitoneally inoculated into these hu-PBL-immunodeficient mice followed by examination of various virological and immunological parameters. Results: We found 6 new immunodeficient mouse strains which efficiently receive human PBL engraftment. Among the hu-PBL-immunodeficient strains we could reproduce the high levels of HIV-1 viremia in NOD/Shi-scid mice corffparable to or more significant than that in HIV-1 primary infection. Systemic HIV-1 infection including liver, lymph nodes, and brain was demonstrated by quantitative DNA or RNA-PCR technique. The amounts of gag p24 in plasma from the HIV-1 infected mice reached 250 ng/mL and end-point-dilution method indicated that infectious virus titer exceeded more than 106 times the tissueculture-infectious dose (TCID) per mL, which is about 1 000 fold higher viral load than that in HIV-1 infected conventional hu-PBL-C.B-17 scid mice. The viremia induced systemic HIV-1 infection involving brain. PCR in situ hybridization confirmed that HIV-1 infected cells invaded mouse brain tissue of the hu-PBL-NOD-scid mice. Conclusions: Our results suggest that the genetic background is critical in the development of primary HIV-1 viremia and subsequent CNS invasion with HIV-1. The present hu-PBL-NOD-scid-mouse represents a useful model for the study of the pathogenesis of HIV-1 in vivo, especially brain involvement, and therapy of primary HIV-1 viremia.

The Trimera Mouse System: A Tool for Generating Fully Human mAbs and Animal Models for Human Diseases

I. LUBIN,E. ILAN,R. EREN, O. NUSSBAUM,Y. PORAT,N. RACHAMIM,A. ZAUBERMAN,L. NEVILE, I. BEN-MOSHE,Y. ARAZI,D. TERKIELTAUB,S. MOSS, R. UHLMANN,T. TZAHOR, S. BERRE, D. GELLER, G. KITZIS,A. KISCHITZKY,S. DAGAN XTL Blopharmaceutlcals, Kirvat Wetzmann,Rehovot, Israel

Normal mice lethally irradiated and radioprotected by transplantation of BM (bone marrow) from SCID mice, are permissive for transplantation of human tissue or human immune cells. This system, the Trimera, could be used for generation of human mAbs, development of animal models for human diseases, and for creation of models for cell therapy. We have generated human mAbs to HBV by immunizing Trimera mice which were engrafted with human PBLs. High affmity fully human mAbs for HBsAg were isolated from hybridoma clones generated by fusion of Trimera splenocytes to a human-mouse heteromyeloma fusion partner. We have developed also mouse models for HBV and HCV infections by transplanting HBV or HCV

556 1WIDA

Vol. 43

infected human liver fragments under the kidney capsule of Trimera mice. Viremia was followed by the presence ofHBV DNA or HCV RNA sequences in mice sera by PCR or RT-PCR, respectively. HBV and HCV sequences started to appear 11 d after transplantation and could be detected for at least four weeks. These models are being used for screening of potential antiviral agents and human mAbs to anti-HBsAg developed by us. Using the Trimera system we were able to generate an animal model for human colon carcinoma. Growth of solid human colon carcinoma transplanted subcutaneously is observed at least 110 d post transplantation. Similar tumor fragments transplanted under the kidney capsule of Trimera mice showed invasive growth pattern. Tumors recovered from mice were CEA positive both by immunohistostaining and by PCR.

Immunological Properties of Heterozygous nu/+ Mice: Changes in Antibody Response and Inducibility of Tolerance to Protein Antigens J. MADAR,L. JANDOV.~,T. HRABA,M. BOUBEL~, M. HOLUB Instttute of Molecular Genettcs, Academy of Sclences of the Czech Repubhc, Prague and Instttutefor Clinical and Experimental Medicine, Prague, Czech Republic

Heterozygous nu/+ mice are not fully identical in their immunological properties with the mice of wild +/+ genotype. A colony of nu/nu, nu/+ and +/+ mice from the same breeding nucleus was established and their immune reactivity to human serum albumin, inducibility of adult immune tolerance to hen egg lysozyme (HEL), sensitivity of their lymphoid cells to stimulation by mitogens and ratio of CD3, CD4 and CD8 positive cell population was studied. Both the numbers of antibody-forming cells in regional lymph nodes and the antibody titers in sera of nu/+ were highly variable, between undetectable values of nu/nu and high values of +/+ homozygotes. Intravenous pretreatment with soluble HEL, leading in +/+ mice to a deep hyporeactivity to subsequent immunization with the same antigen, did not significantly decrease the response of nu/+ mice. These results indicate that the immunological alteration of nu/+ mice is not only quantitative and that T cell subpopulation might be differentially modified by the presence of nu allele. The finding of decreased CD4 : CD8 ratio in nu/+ mice also supports this idea.

Laboratory Animals: Ethics and Pseudoethics M. MATFIELD European BiomechcalResearch Association, London, UK

In the long-nmning debate about animal experirnentation, there have been two strongly held positions which have not altered very much since the debate began over a century ago. The scientific community has pointed to the fact that the use of animals in medical research and testing has played an essential role in the development of numerous medical therapies, diagnostic techniques and in the fundamental advances in science upon which these developments are based. The consequent saving of human (and animal) life and prevention of suffering more than justifies the "cost" in terms of the lives of laboratory animals and the suffeting they may have experienced in the course of the experimentation. However, the advocates of animal tights have strongly argued that the use of animals in experiments has not yielded scientifically useful information or led to significant medical advances, but has involved prolonged and gross abuse of animals. However, the animal tights perspective rejects animal experimentation on a philosophical basis and it appears likely that their arguments about the utility of animal experiments are constructed merely for campaigning purposes. In recent years, there has been a growing acceptance of the importance of animal welfare, both by moderate animal protectionists and by animal researchers themselves. The animal rights and animal welfare positions can be analyzed as examples of utilitarian and deontological ethical approaches to the question of animal experimentation. This paper will analyze the way human beings apply both utilitarian and deontological ethical analysis to human interests and animal interests. It concludes that we apply a mixed utilitarian/deontological approach to human interests and an almost entirely utilitarian approach to animal interests. On a utilitarian analysis, animal experimentation can be made more ethical by high standards of laboratory animal welfare based on the three R's of Russell and Burch.

1998

ABSTRACTS 557

Single Dose of the nu Gene Affects the Bone M a r r o w Stem Cell Potential

E. NE~ASa, V. ZNOJILa, M. HOLUBb aDepartment of Pathophysiology, Ist School of Medicine, Charles University bDlvtslon of Immunology, Institute of Mtcrobtology, 142 20 Prague, Czech Repubhc

In the Till-McCulloch assay it was established that in mice of BALB/c background the number of spleen colony-forming units per 105 femoral bone marrow cells injected into lethally irradiated syngeneic recipients was reduced to 27 % in nu/nu donors and 36 % in nu/+ donors, compared to +/+ donors. In the B10LP strain the reduction was 22 % in nu/nu, 24 % nu/+, in CBA/J strain it was 68 % in nu/nu and 58 % in nu/+. In the C57B1/10ScSn strain the reduction was 67 % of the +/+ count in nu/nu and 79 % in nu/+. Per femur, the average values were 4788 CFU/s for +/+, 2533 for nu/nu and 3580 for nu/+. The proliferation value of CFU-S was not significantly different between the 3 genotypes. It follows that the presence of the nu gene in the population affects the haematopoetic potential and that this potential is not dependent on the presence of the fully differentiated thymic tissue (which in nu/+ heterozygotes is only hypoplastic, not dysgenetic).

An Approach to Obtain a Suitable Dose of AUC-Dependent Antitumor Agents Showing Clinically Equivalent Effects in the Human Tumor Nude Mouse Model

Y. OHNISHIa, M. INABA b, Y. SUGIYAMA c aCentral Institutefor ExpertmentalAmmals, Kawasakt, bdapanFoundatzonfor Cancer Research, Tokyo, CTokyoUniversity, Tokyo, Japan

Prediction of clinical efficacy of new drugs is very difficult, but it is an important aim of preclinical testing. In antitumor drug evaluation, candidates often respond well in preclinical models although they do not show any obvious clinical efficacy. In this regard, evaluation with pharmocokinetically equivalent conditions to humans may be a key to improving the predictability of preclinical testing. We have reported that agents with type I cell-killing action showed AUC dependent antitumor effects. If it is possible to predict clinically achievable AUC of new agents with type I cell killing action, we can assess their clinical efficacy more precisely by comparing plasma AUC of human tumor bearing mice and that of humans. In this study, assuming that AUCs on treatment with the maximum dose (AUCMTD) are more closely related between humans and monkeys than between humans and mice, we measured AUCMTo in monkeys, and compared them with the respective clinical data reported. It was found that the AUCMTD of type I agents such as MMC, DDP, VP-16 and CPM showed good agreement with their respective clinical values, although that of ACNU in monkeys was somewhat higher than that of humans. However, with type II drug CPT-11, the AUCMTD of its active metabolite SN-38 in monkeys was much lower than that of humans. These results indicate that prediction of clinically achievable plasma AUC of type I agents may be possible by analyzing pharmaco/toxicokinetics of the agents in monkeys.

Development of N e w scid Mouse Models for AIDS Research

L.D. SHULTZa, R. HESSELTONb, D.L. GREINTERb aTheJackson Laboratory, Bar Harbor, ME 04609, USA

bUmverstty of Massachusetts Medical Center, Worcester,MA 01605, USA

The continued progression of the AIDS pandemic has stimulated increased emphasis on the development of effective animal models for this disea.~e. These models are essential for the evaluation of HIV vaccines and therapy and for increasing the understanding of HIV pathogenesis. A number of mouse models for AIDS research have been described including scid/scid mice engrafted with human hematopoietic tissues or lymphoid cells, HIV transgenic mice, and MAIDS. Our studies have focused on the development of new genetic stocks of scid/scid mice as hosts for engraftment with human hematolymphoid cells. Initial investigations focused on stocks of NOD/LtSz-scid/scid mice deficient in innate immune function. These mice have defects in NK cell function, macrophage activity, and hemolytic complement, facilitating heightened levels of engraftment with human lymphoid cells and with human hematopoietic progenitor cells compared with conventional C.B-17-scid/scid mice. The recent construction, by genetic engineering, of transgenes, or disrupted alleles that improve the ability of immunodeficient mice to support engraftment with human lympho-

Vol. 43

558 IW1DA

hematopoietic cells has provided a tremendous opportunity for development of effective mouse models for AIDS research. Recently, we carded out genetic crosses to produce NOD/LtSz-scid/scid mice homozygous for the 132-microglobulin null allele. NOD/LtSz-scid/scid B2M null mice lacked MHC class I expression and had no detectable NK cell activity. These mice supported markedly elevated levels of human CD4 + T cell engrathnent. The absence of 132-microglobulin in these mice also results in hemachromatosis and a shortened half life of circulating human IgG. The rapid IgG clearance is associated with the absence of the FcRp that normally protects IgG from catabolism. Additional immunodeficient mouse models under development should accelerate the analyses of HIV vaccines.

Germ-Free Animals as a Sensitive Tool for Testing the Safety of Live Vectors

J. SINKORAa, Z. PxEHAKOV.h,a,R. STI~PANKOV,~a, T. HUDCOVICa, H. TLASKALOV,h,-HOGENOV,h, a, E. HOLODAb, B. DE GEUSc, A. BIANCHIc alnsntute of MIcrobtology, Academy of Sciences of the Czech Repubhc, 142 20 Prague, Czech Republic bVetermary Umverslty, Kogice, Slovakm Clnstttutefor Animal Science and Health, Lelystad, The Netherlands

In GF mice and piglets we studied the behavior of live vectors that had proved to be non-virulent in conventional (CV) animals. Salmonella typhimurium x4072 strain expressing a non-toxic form ofDNT (Pasteurella multocida) and an attenuated pseudorabies virus (PRV)-based vector bearing a hemagglutinin gene from influenza virus were tested in young GF minipigs and GF SCID mice (Balb/c) of different ages. In contrast to their CV counterparts, all but one Salmonella x4072 monoassociated piglets were susceptible to Salmonella-caused sepsis and died a couple of days after colonization. Immunocompromised GF mice with the same vector survived for several weeks. I.m. injection of the PRV vector had no effect on GF piglets. I.n. route of administration, however, caused lung oedema and death in 2 out of 3 GF piglets and the virus could be isolated from lung tissue. We conclude that live vectors, which appear to be safe in CV animals, may be virulent under GF conditions because of the underdeveloped immune mechanisms and/or nutrition deficiency. Marked interspecies differences were observed in natural resistance to S. typhimurium live vectors. We have found a sensitive system for studying safety limits of live vectors and confirmed that attention must be paid to the choice of animal species for safety experiments.

Conditional Gene Targeting of PU.1

C. SOMOZAa, D. AUDa, T. MCNEILa, L. LUCIANa, C. HUETTNERb, D. T

NENb, R. MURRAYa

aDNAX Research Instztute of Molecular and Cellular Bmlogy, Palo Alto, Cahforma, CA 94304, USA bBeth Israel Medical Center, Boston, MA 02215, USA

PU. 1 is a transcription factor of the ETS family expressed exclusively in hematopoietic ceils, in particular in the monocytic and B cell lineages. PU.1 has been shown to regulate the expression of multiple genes involved in monocyte/macrophage growth, differentiation and function, including MCSF and GMCSF receptors, Fc receptors, CD1 lb, CDlg, and scavenger receptors. In B cells, PU.1 is important for the expression of the genes encoding the immunoglobulin light and heavy chains, J chain and IgA molecule. Disruption of the PU. 1 gene using standard gene targeting techniques leads to late embryonic or neonatal lethality of unclear cause, accompanied by ablation of multiple hematopoietic lineages. Our objective was twofold: (1) delete PU. 1 from the monocyte/macrophage lineage at selected developmental time points and (2) delete PU. 1 in the adult animal (as opposed to embryonic deletion, as in conventional null mutation). Both approaches involve the use of the CRE/loxp recombination system. The PU. 1 gene was modified in ES cells by flanking exons encoding the PEST and the DNA binding domains of the PU. 1 with loxp sites. In vitro CRE-mediated recombination generated a functional PU. 1 gene flanked by loxP sites (flox) with the Neo cassette removed, and this modification has been introduced in the mouse germline. To selectively inactivate PU.1 from the myeloid lineage we have created transgenic mice that express CRE under the myeloid specific promoters CD1 lb and lysozyme. Several lysozyme-CRE and CD1 lb-CRE transgenic founders have been crossed to mice that carry the flox PU.1 gene to examine the monocyte/macrophage specific deletion of PU.1. In addition, we have created a fusion protein between a mutant estrogen receptor and CRE (CREmER). The CREmER construct, under the control of ubiquitous promoters, was transfected into ES cells that contain one allele of the PU. 1 flox gene. Culture of the cells in

1998

ABSTRACTS w

the presence, but not in the absence, of 4-HT induces recombination and deletion of exons 4 and 5 of the PU.1 flox allele. ES clones preselected for inducible CRE activity have been injected into blastocysts to generate mice that carry the CRE.-mER transgene and the flox PU. 1 gene.

Ribozyme Mediated Cleavage of Aberrant Epidermal Growth Factor Receptor (EGFR)-mRNA Inhibits the in Vivo Tumor Growth H. YAMAZAKIa, N. TAMAOKIa, H. KIJIMAa, Y. OHNISHIb, Y. ABEa, Y. OSHIKAa, T. TUCHIDAa, T. TOKUNAGAa, M. NAKAMURAa, Y. UEYAMAa'b aDepartment of Pathology, School of Medicine, Tokal Umverslty, Bohseidai, Isehara, Kanagawa 259-11 bCentral Institutefor Experimental Ammals, Nogawa 1430, Kawasakl, Kanagawa 213, Japan

The mutant EGFR with a specific deletion encompassing exon 2 to exon 6 might correlate with malignant transformation of glial cells. The mutation of EGFR gene creates a sequence cleavable by the ribozyme in the mRNA. We examine the effects of the ribozyme-mediated site-specific cleavage of the aberrant EGFR-mRNA on the growth of cell lines in vivo. The ribozyme (abEGFR-rib) efficiently cleaved the synthetic EGFR-RNA substrate at 37 ~ in the range of 0.5 mmol/L to 6 mmol/L of Mg 2+ in vitro. The disabled ribozyme did not cleave the EGFR-RNA substrate even at 40 mmol/L of Mg 2§ These data suggested that the ribozyme abEGFR-rib efficiently, specifically, and enzymically cleaves the aberrant EGFR-RNA substrate in vitro. The I]-actin promoter driven ribozyme sequence was introduced into the ERM5-1 cell line transformed by the aberrant EGFR cDNA. The growth ability of the cell line was estimated by transplantation in SCID mice. The ERM5-1 cell line transfected by ribozyme (pHl~-abEGFR-rib) showed a significantly decreased mitotic activity, growth and transplantability as compared with the original ERM5-1 cell line in the SCID mice in vivo. The mitotic index with BrdU incorporation also decreased in the ERM5-1 cell line transfected by pH[3-abEGFR-rib. The ham merhead ribozyme efficiently and specifically cleaves the aberrant EGFR-mRNA in vitro, and inhibits the growth of the transformed cell line in vivo. These results indicate that immune-deficient mice/tumor system is useful for testing ribozyme mediated cancer gene therapy in vivo.