Online Issues

<< All Back-issues

<< This Issue's Table of Contents

ILAR Journal V34(1/2) 1992 [FORMERLY ILAR NEWS]
Immunodeficient Rodents

Seventh International Workshop on Immune-deficient Animals

Abstracts from the Seventh International Workshop on Immune-Deficient Animals
The Jackson Laboratory
Bar Harbor, Maine
September 25-29, 1991

CONTENTS

• Introduction
  
• Correlation between P-glycoprotein expression and multidrug resistance in human tumor xenografts
  
• Regression of a murine fibrosarcoma cell transfected with the hIL-2 gene in SCID and SCID-beige mice
  
• Role of pituitary-ovarian hormones in initiation and growth of malignant ovarian granulosa cell tumors in genetically susceptible mice
  
• Effects of HIV infection in human fetal thymus in the SCID-hu mouse
  
• Status of Ig and TCR genes in non-transformed B- and T-lineage cells of SCID mice
  
• The SCID mouse as a model for the virus-host relationship: Immune regulation of viremia and maternal-fetal virus transmission
  
• Quick screening of anticancer drugs with immunosuppressed mice
  
• Promoting engraftment of the human B-cell lineage in SCID mice by human interleukin-6
  
• Otitis media in SCID mice due to infection with an atypical pseudomonas bacteria (Fox Chase SCIDI)
  
• TCR expression during early neonatal development in euthymic and athymic (nude) mice is restricted to the intestinal lamina propria
  
• Animal models of normal and leukemic human meatopoiesis by transplantation into immune-deficient mice
  
• Differential in situ expansion of tumor-infiltrating lymphocytes (TIL) following adoptive immunotherapy of tumor-bearing immunologically-deficient mice is not sufficient to ensure a successful outcome to therapy
  
• Insights into mast cell development and function derived from analyses of mice carrying mutations at W/C-kit or Sl/MGF (SCF) loci
  
• Malignant behavior of human tumors in immunodeficient mice
  
• The use of immunodeficient mice for studying human hematopoietic neoplasms
  
• The Institute of Laboratory Animal Resources: A source of information on animal models
  
• Depletion of CD4+ cells in MHC class II deficient mice
  
• The role of Kit on murine development
  
• Abnormal thymocytopoiesis in viable motheaten (mev) mutant mice
  
• HIV infection of hu-PBMC-SCID mice
  
• Peyer's patch cells reconstitute the peripheral and mucosal immune systems of SCID mice
  
• Changes induced by nonshivering thermogenesis in macrophage populations and in NK cells of nude mice
  
• Immunological defects of nu/+ heterozygotes
  
• Reduction of lymphokine-activated killer cell activity by the introduction of the beige gene into SCID mice
  
• The study of adult nasopharyngeal mucosa xenograft and dinitrosopiperazine (DNP) carcinogenesis
  
• Acute human-versus-mouse graft-versus-host disease in normal and immune-deficient mice
  
• Prediction of clinical antitumor effect based on Clinically Equivalent Dose preestimated by animal scale-up
  
• Comparative study on Pneumocystis carinii infection in SCID and nude mice
  
• Feline immunodeficiency virus infection of SCID mice engrafted with feline tissues: A murine model for HIV drug testing
  
• Helminths infection in SCID mice
  
• Increased antitumor activity of 5-fluorouracil by 1-1eucovorin and interferon on human colon carcinoma xenograft transplanted into nude mice
  
• Orthotopic xenotransplantation of human lung cancer in the lungs of nude mice and experimental observation of invasion and metastasis
  
• Point mutation and its stability of ras oncogene in human neoplasms and tumor xenografts
  
• SCID and bg-nu-xid mice implanted with human fetal thymus and liver: A comparison
  
• Nude mouse model for development of radio-labeled monoclonal antibodies to treat human lung cancer
  
• High human IgG levels in the SCID mouse reconstituted with human splenic tissues from patients with gastric cancer
  
• Antitumor activity of human lymphoma-reactive T cells against disseminated Burkitt's (Daudi) lymphoma in SCID mice
  
• SCID thymus: A model of T-cell development in an early stage
  
• Treatment with rat stem cell factor affects thymocyte progenitor cells
  
• Comparison of resistance to Pseudomonas aeruginosa infection among nude, SCID, and drug-induced immunodeficient mice using biological response modifier S(BRM)
  
• Retrovirus-induced immunodeficiency in the mouse
  
• HIV infection and immunity in the hu-PBL-SCID mouse model
  
• Differentiation of M-CSF-independent macrophages and glucan-induced granuloma formation in osteopetrotic (op//op) mice
  
• Epidermal growth factor receptor gene alteration in human neural tumor xenografts in nude mice
  
• Pleiotropic effects of the scid mutation
  
• Epstein-Barr virus-induced lymphoproliferative disorders in SCID-hu chimeric mice
  
• SCID mice as models for human parasitic diseases
  
• Selective reconstitution of T-cell subsets in SCID mice
  
• Genetics and functional studies of minor histocompatibility loci
  
• Production of pathogen-free immune-deficient animals
  
• Comparative evaluation of congenitally athymic and euthymic rat strains bred and maintained at different institutes I. Euthymic rats
  
• Comparative evaluation of congenitally athymic and euthymic rat strains bred and maintained at different institutes II. Athymic rats
  
• Xenotransplantation of human spleen in SCID mice
  
• Malignant xenografts: A molecular cross-talk at the host-tumor interface
  
• A nude mouse model for the study of human tumor xenograft invasion and metastasis
  
• Immunomodulation of soluble Qa-2 levels in mice
  
• Chemo- and endocrino-therapy of human breast carcinoma xenografts in the dormant or exponential growth phase
  
• A newly developed hexamethylmelamine derivative (SAE 9) with an activity of aromatase inhibitor
  
• Xenotransplantation of human chancers into SCID mice
  
• Role of natural antibodies in rejection of discordant bone marrow grafts
  
• T but not B cells are required for the clearance of murine malaria parasites from SCID mice
  
• Immunodeficient animals as models for opportunistic infections
  
• Human immune response to mouse erythrocyte antigens demonstrated in SCID mouse-human chimera
  
• CD5, T independent type 2 responses and X-linked immune deficiency
  
• Some biological characteristics of highly metastatic human carcinoma in nude mice
  
• Growth and metastatic behavior of human tumors xenografted in nude and SCID mice
  
• Studies on SDI chemosensitivity patterns of surgically resected tumors before and after transplantation into athymic nude mice
  
• Metastatic behavior of murine and human cancer cells in nude mice
  
• Research on human tumor-killing cells derived from peripheral blood and tumor tissues
  
• The breeding of T, B, and NK cell immunodeficient PBI/3-xic-bg-nu mice

Introduction
Dr. Leonard D. Shultz, The Jackson Laboratory

Immune-deficient animals are widely utilized in many areas of basic and applied biomedical research. The Seventh International Workshop on Immune-Deficient Animals held at The Jackson Laboratory in Bar Harbor, Maine focused on recent advances resulting from experimentation with such animals. Results of this experimentation have provided insights into mechanisms underlying the development and regulation of the mammalian immune system in normal and pathologic states. Our understanding of mechanisms underlying heritable human immunologic diseases has been advanced through the study of experimental animals bearing single gene mutations that perturb the immune system. While certain of these mutations are considered as homologues for specific human diseases, the major value of the immunologic mutants are as tools with which to dissect the mammalian immune system. An alternate approach for the investigation of immunological mechanisms has been the development of mice bearing disrupted genes whose products are critical to normal immune system development or function. While the occurrence of spontaneous mutations opens new doors into the function of previously unknown genetic loci, the targeted disruption of specific genes facilitates the precise determination of the roles of these genes in the functioning of the immune system.

Immune-deficient animals have also been valuable tools for the study of mechanisms of resistance to infectious diseases. While such animals have been used for many years in the study of bacterial and viral diseases, recent work has focused on their use as hosts for diseases caused by protozoa, fungi, and nematode parasites. Advances in cancer research have also been facilitated by the use of these animals. Although congenitally athymic nude mice have been used for many years as hosts for the growth of human malignant cells, it has not been possible to row normal human lymphoid cells in such mice. In contrast, the growth of normal human lymphoid cells in mice homozygous for the severe combined immunodeficiency (scid) mutation has facilitated investigation of the interaction of malignant cells with tumor infiltrating lymphocytes. Moreover, scid / scid mice that harbor normal human lymphoid cells can be infected with HIV, thus providing and invaluable animal model for AIDS research. The following abstracts summarize work with immune-deficient animals in a number of areas including basic immunologic research, assessment of mechanisms of resistance to infectious diseases, evaluation of factors underlying host resistance to neoplasia, and the study of acquired immunodeficiency syndrome.

We greatly acknowledge the support from the National Institute for Allergy and Infectious Diseases (AI 31415) and from the following organizations: Accurate Chemical and Scientific Co., Amgen Inc., Bristol Myers Squibb Pharmaceutical Research Institute, Burroughs Wellcome Co., Connaught Laboratories Inc., Connaught Laboratories Ltd., The Dupont-Merck Pharmaceutical Co., Genentech Inc., Glaxo Inc. Research Institute, The Jackson Laboratory, R.W. Johnson Pharmaceutical Research Institute, Merck Research Laboratories, Miles Inc., Minnesota Mining and Manufacturing Co., Warner Lambert Co., Sandoz Research Institute, G.D. Searle and Co., Serono Laboratories, The Upjohn Co., and The Wellcome Foundation.


Correlation between P-glycoprotein expression and multidrug resistance in human tumor xenografts
Yoshiyuki Abe, Masato Nakamura, Yasuyuki Ohnishi*, Atsushi Tsugu, Yuko Katoh, Yoshito Ueyama, and Norikazu Tamaoki

Multidrug resistance is a critical subject in cancer treatment. Various mechanisms confer multidrug resistance in tumor cells. A human multidrug resistance gene (MDR1) encodes a membrane associated protein, P-glycoprotein (P-Gp). P-Gp excretes drugs outside the cells, resulting in decreased drug accumulation in the resistant cells. Solid tumors often show natural resistance to drugs. The mechanisms are still unclear in vivo.

We examined chemosensitivity to three drugs (ADR, VCR, and VLB) on 48 human tumor xenografts in nude mice, using a clinically equivalent dose. We evaluated MDR1 gene expression in human tumor xenografts by Northern blot analysis and reversed transcription PCR (RT-PCR) assay. We used the indirect immunohistochemical technique with anti P-Gp monoclonal antibody (C219) to estimate cellular P-Gp production in the tumor.

A total of 23 cell lines showed drug resistance to the three chemical agents. Six of the 23 cell lines showed low level or heterogeneous expression of the MDR1 gene by RT-PCR, while Northern blot analysis detected no apparent expression. Five of the 23 cell lines demonstrated both MDR1 expression and P-Gp production. One cell line showed apparent MDR1 gene expression without P-Gp production. Nine tumor cell lines showed no apparent MDR1 gene expression. The other eight cell lines revealed neither MDR1 gene expression nor P-Gp production. The results suggest that P-Gp has a role in multidrug resistance mechanism in certain kinds of tumor cell lines. The alternative mechanism may contribute to multidrug resistance in tumor cells.

Central Institute for Experimental Animals, Kanagawa, Japan;
* Department of Pathology, Tokai University School of Medicine, Kanagawa, Japan

Regression of a murine fibrosarcoma cell transfected with the hIL-2 gene in SCID and SCID-beige mice
T. Alosco, B.A. Croy, B. Gansbacher*, and R.B. Bankert

Local secretion of various cytokines appears to protect mice from tumor engraftment (1,2). The methylcholanthrene-induced fibrosarcoma CMS-5 is a rapidly growing tumor when transplanted subcutaneously into BALB/c mice. Its hIL-2 producing counterpart, N2A/IL2/CMS5, regresses in BALB/c mice. Regression appears to be medicated by CD8+ T cells and is accompanied by immune memory. Transplantation of CMS-5 and N2A/IL2/CMS5 into B- and T-cell deficient SCID mice gives similar results, suggesting that in addition to CD8+ mature T cells another or several other cell types are involved in effecting the IL-2 induced tumor regression. Since SCID mice have normal levels of functional NK cells, an attempt was made to determine whether or not these cells could be one of the possible cell types mediating the N2A/IL2/CMS5 regression. This is a possibility since these tumors did not regress in NK cell deficient scid-bg mice treated with anti-asialo-GM1 antibody. While the identity of the effector cell in SCID mice has not been resolved, it was established that these cells do not express memory. Histological evaluation of the sites of tumor regression in N2A/IL2/CMS5 engrafted SCID mice revealed an inflammatory response that was lymphoid in nature. We conclude that cytokine gene transfected tumor cell regression is mediated by a non-T cell, non-B cell, and possibly an NK cell or other ASGM1 positive cell in the SCID mouse.

1. Gansbacher, B., K. Zier, B. Daniels, K. Cronin, R. Bannerji, and E. Gilboa. 1990. Interleukin 2 gene transfer into tumor cells abrogates tumorigenicity and induces protective immunity. J. Exp. Med. 172:1217-1224.
2. Fearon, E.R., D.M. Pardoll, T. Itaya, P. Golumbek, H.I. Levitsky, J.W. Simons, H. Karasyuama, B. Vogelstein, and P. Frost. 1990. Interleukin-2 production by tumor cells bypasses T helper function in the generation of an antitumor response. Cell 60:397-403.

Department of Molecular Immunology, Roswell Park Cancer Institute, Buffalo, New York
*Departments of Hematology/Lymphoma, Memorial Sloan-Kettering Cancer Center, New York, New York
This work was supported in part by USPHS grants CA54491 and CA09581.

Role of pituitary-ovarian hormones in initiation and growth of malignant ovarian granulosa cell tumors in genetically susceptible mice
Wesley G. Beamer, Barbara J. Tennent, Kathryn L. Shultz, and Leonard D. Shultz

Ovarian tumors occur in approximately two percent of Caucasian women and are the most lethal of the gynecologic cancers. Tumors of the granulosa cells within follicles comprise 10 to 15 percent of these cases and are bimodally distributed with respect to age of onset. The initial age cluster occurs in young girls between infancy and menarche (juvenile onset) while the second occurs in mature adult and post-menopausal (adult onset) women. Both subsets of granulosa cell tumors manifest a substantial incidence of malignancy.

Human epidemiologic and laboratory experimental induction studies have provided circumstantial evidence that pituitary gonadotropins (luteinizing (LH) and follicle stimulating hormones (FSH)) have a role (or roles) in granulosa cell tumorigenic processes, although mechanisms have remained elusive. We have identified a heritable model for juvenile onset granulosa cell tumors in SWR inbred mice. These tumors develop spontaneously in one or both ovaries between 24 and 42 days of age as pubertal ovarian maturation takes place. Tumorigenesis is controlled by a small number of genes that have segregated in a set of 14 recombinant inbred strains (designated SWXJ) derived from SWR and SJL mice. These genes include: 1) granulosa cell tumorigenesis (Gct) on Chromosome 4, b) an androgen metabolism gene of uncertain location, and c) a genomic imprinting locus of uncertain location. Our studies have used these genetically susceptible mice in concert with mice carrying additional unique mutant genes to obtain insight about pituitary-ovarian hormones in primary and metastatic disease.

To pursue these studies, a stock of mice doubly homozygous for the mutant genes severe combined immune-deficiency and hypogonadal (scid/scid, hpg/hpg) was developed as a host for allogeneic ovarian tissue grafts. These doubly homozygous mice are severely immunodeficient and lack serum pituitary gonadotropins and ovarian sex steroids. Experiments were conducted by grafting genetically tumor susceptible ovaries or tumor tissue beneath the left kidney capsule of hosts. Four weeks thereafter, necropsies were performed to obtain data on tumor incidence, neoplastic growth, and histological appearance as a function of genotype, age, hormone treatment, and tumor tissue status. Tumors formed only in susceptible ovarian grafts placed in scid/scid, +/? hosts, and not in scid/scid, hpg/hpg hosts.

The combination of phenotypes characteristic of the scid and hpg mutant genes represents a powerful experimental tool that answered four critical questions posed by the genetically determined ovarian juvenile-onset granulosa cell tumor biology. The concept of combining mutant genes to address specific questions is applicable to investigations of other tumor systems, including parallel studies of human and murine tumors, with limits established only by investigator originality.

The Jackson Laboratory, Bar Harbor, Maine

Effects of HIV infection in human fetal thymus in the SCID-hu mouse
Mark L. Bonyhadi, Suzan Salimi, J.M. McCune, and Hideto Kaneshima

Individuals infected by the human immunodeficiency virus type 1 (HIV-1) exhibit progressive depletion and severely reduced function of peripheral CD4⁺ helper T lymphocytes. A variety of mechanisms have been proposed to account for the observed loss of CD4⁺ T cells, including virally mediated cytopathic killing of infected cells and deleterious infection of T-cell precursors. Even though in vitro data support both these hypotheses, it has been difficult to demonstrate or analyze these processes in vivo. We have previously shown that human fetal thymus/liver implants in SCID-hu mice support infection by HIV-1 (JR-CSF). In this study, thymus implants were inoculated with primary patient isolates of HIV (SM, TY, or EW) or with molecular HIV clones (JR-CSF or 2.4). Over time, the implants were harvested and analyzed by ELISA, flow cytrometry (FACS), and immunohistochemistry for expression of viral antigens and changes in overall thymic profile. In all cases, thymic p24 levels increased in a time-dependent manner, indicating productive infection of some thymic elements. In the case of patient isolates, FACS and immunohistochemical analysis showed that, concomitant to the increase in p24, thymus cellularity was greatly reduced, reflecting the nearly complete ablation of thymocytes of the CD^-48⁺ and CD4⁺8⁺ phenotypes. CD4^-8⁺ thymocytes and CD4^-8^- cells remained as the principal lymphoid components of infected thymi. Moreover, thymic lobes were shrunken, with cortical regions appearing extremely compressed and practically devoid of thymocytes. Expression of viral antigens could be localized in several cell types, including thymocytes and thymic stromal elements. These observations support the notion that the spread of HIV infection to the thymus can lead to the elimination of a majority of developing thymocytes, and thus preclude regeneration of the peripheral T-cell pool in HIV-infected individuals.

SyStemix, Inc., Palo Alto, California

Status of Ig and TCR genes in non-transformed B- and T-lineage cells of SCID mice
M. Bosma, A. Carroll*, M. Reichman-Fried, and R. Hardy

SCID mice lack functional lymphocytes because they carry a mutation that impairs rearrangement of immunoglobulin (Ig) and T-cell receptor (TCR) genes. Rearrangement of TCRd, but not d and b genes was routinely observed in DNA of SCID thymocytes and thymocyte hybridomas. TCRd gene rearrangements appeared to involve Dd1, Dd2, and Jd1 elements only; rearrangement of elements upstream of Dd1 (e.g., Vd1) was not observed and transcripts corresponding to fully-assembled TCRd genes (VDJd or VDDJd) were not detected in RNA from SCID thymocytes. These findings suggest that Dd1, Dd2 and Jd1 may be among the first TCR gene elements to undergo recombination and that SCID T-lineage cells are developmentally arrested during or shortly after this stage of differentiation.

SCID B-lineage cells are arrested at the pro-B stage (B220+S7+IgM-). DH elements undergo rearrangement but fully-assembled VDJH transcripts are not detected and pre-B cells (B220+S7-IgM-) fail to develop. Interestingly, the introduction of a functionally rearranged m transgene into the SCID mouse genome results in normal numbers of "pre-B cells" (B220+S7-IgM-), but no B cells (B220+S7-IgM+). However, the pre-B cells in such m transgenic scid mice (sc/sc, m mice) do not show clear evidence of IgL(k) gene rearrangements as is the case for pre-B cells from heterozygous scid littermates with an without the m transgene (sc/+, m and sc/+ mice).

Institute for Cancer Research, Fox Chase Center, Philadelphia, Pennsylvania
* Albany Medical College, Department of Microbiology and Immunology, Albany, New York

The SCID mouse as a model for the virus-host relationship: Immune regulation of viremia and maternal-fetal virus transmission
W.A. Cafruny, D.S. Bradley, K.M. Powell, and J.B. Broen

We have investigated lactate dehydrogenase-elevating virus (LDV) infection in SCID mice. LDV infects macrophages in normal and SCID mice, causing a persistent infection which in normal mice is characterized by polyclonal B-cell activation as well as a strong IgG anti-LDV response. LDV-infected SCID mice fail to produce IgG anti-LDV antibodies and display relatively elevated levels of viremia, although they may undergo limited activation of B-cell responses. Reconstitution of the IgG anti-LDV response in SCID mice by transplantation with histocompatible spleen cells resulted in suppression of viremia to levels seen in normal mice, confirming the role of IgG anti-LDV in regulating viremia. The role of immunity in the regulation of maternal-fetal virus transmission was also studied by determining the rate of infection present in the newborns of LDV-infected normal and SCID mice. Female SCID mice that had been LDV-infected for several months prior to conception, and which failed to produce IgG anti-LDV antibodies, routinely transmitted infection to 100 percent of their offspring. In contrast, normal mice which were LDV-infected just prior to conception developed IgG anti-LDV antibodies, and displayed reduced rates (about 45 percent) of transplacental virus transmission. Normal mice infected at more than 1 week after conception transmitted the virus to 100 percent of their offspring. These results show that specific IgG anti-LDV antibody is important in regulating viremia and also suggest an important role for antibodies in regulating transplacental virus transmission.

Department of Microbiology, University of South Dakota School of Medicine, Vermillion, South Dakota
This work was supported by NIH grant AI27978-01A2

Quick screening of anticancer drugs with immunosuppressed mice
Quanlu Chen, Yan Ling, and Qingfan Zhu

Nude mice (immunosuppressed and immunocompetent) were used to find optimal animal subjects and a quick test to screen anti-esophageal cancer drugs. The subrenal capsule (SRC) was used as the cancer transplantation region.

The esophageal cancer cell line Eca109 was transplanted subcutaneously to nude mice and a solid tumor was formed. The tumor pieces were then implanted into SRCs of the three groups of the mice. The immunosuppressed mice had been pretreated with 200 mg/kg of cyclophosphamide before the transplantation. The growth curves showed that in the immunocompetent mice the grafts stopped growing with obvious host reaction on the 6th day after transplantation. In the immunosuppressed mice, the graft growth did not stop on the 6th day but continued growing for an additional 2 to 4 days. Host reaction was not noticeable after the first 6 days of growth, but became much more distinct after growth stopped on around the tenth day. In the nude mice, the grafts grew continually throughout the duration of the experiment (12 days), but the growth rate was much lower than the immunosuppressed and the immunocompetent mice.

On the basis of the growth of the tumor grafts, the 6-day drug-sensitive test was performed in the three groups of mice with Bleomycin, Cis-Platinum Diammonium Dichloride, and 5-Fluroouracil. The tumor growth inhibition rates were in accordance with those in the nude mice implanted subcutaneously with the tumor grafts, and with those in the clinical practice.

The results showed that immunosuppressed mice can replace nude mice, which are expensive and for which care is inconvenient. It also showed that transplantation in SRC can replace the subcutaneous method, which requires a lengthy experimental period. Therefore, immunosuppressed mice with tumor grafts in SRC may be the optimal subjects for quick screening of anti-esophageal cancer drugs.

Apoptosis, DNA contents, silver-binding nucleolar organizer regions (Ag-NORS) and other pathological changes of the tumor cells were used as the parameters in the SRC assay.

Department of Pathology, Henan Medical University, Zhengzhou, China

Promoting engraftment of the human B-cell lineage in SCID mice by human interleukin-6
B.A. Croy, S. Williams, and R.B. Bankert

Interleukin-6 (IL-6) is a lymphohematopoietic growth factor that enhances the survival of stem cells and promotes differentiation of the B-cell lineage. Murine IL-6 does not bind to the human IL-6 receptor. To determine whether hIL-6 could promote engraftment of human peripheral blood leukocytes (hPBL) in C.B17-scid mice, hPBL were pretreated ten minutes with 4ng hrIL-6 (R & D Systems, Minneapolis, MN) or PBS and then administered intraperitoneally. One group of mice receiving treated cells also received repeated intraperitoneal injections of 4ng hrIL-6 (nine injections every 8 hours followed by 11 injections every 12 hours). Engraftment was assessed by measurement of hIg (ELISA) in serum from the SCID-hu mice. All mice receiving PBL treated with hrIL-6 displayed hIg. In comparison to control animals, the mice receiving multiple doses of hrIL-6 had twice the level of hIg in their sera within 3 weeks following engraftment and this difference was sustained. These data suggest that the levels of IL-6 in SCID mice are not optimal for engraftment of hPBL and that exogenous hrIL-6 can stimulate the human component within the SCID-hu chimera. These promising results have encouraged us to cross C57BL/6-scid/scid-bg/bg mice (1) to C57BL/6-hIL-6 transgenic mice (line Ld46) (2) to obtain improved recipients for studies of hPBL and human tumor/TIL engraftment.

1. Croy, B.A., and C. Chapeau. 1990. Evaluation of the pregnancy immunotrophism hypothesis by assessment of the reproductive performance of young adult mice of genotype scid/scid.bg/bg. J. Reprod. Fertil. 88:231-239.
2. Suematsu, S., T. Matsuda, K. Aozasa, S. Akira, N. Nakano, S. Ohno, J.-I. Miyazaki, K.-I. Yamamura, T. Hirano, and T. Kishimoto. 1989. IgG1 plasmocytosis in interleukin 6 transgenic mice. Proc. Natl. Acad. Sci. USA 86:7547-7551.

Department of Molecular Immunology, Roswell Park Cancer Institute, Buffalo, New York.
This work was supported by NSERC, Canada and NIH grant CA54491.

Otitis media in SCID mice due to infection with an atypical pseudomonas bacteria (Fox Chase SCIDI)
F. Dagnaes-Hansen, Pfister*, and M. Bisgaard**

Mice homozygous for the scid mutation (scid/scid) are deficient for immune functions mediated by T and B lymphocytes. Due to the immune-deficiency of this mouse, it is susceptible to infections with many pathogens. Few reports have dealt with specific infections such as Pneumocystis carinii (4, 7, 8) and Proteus mirabilis (6).

In a barrier bred colony of SCID mice, kept under laminar air flow at Bomholtgård Ltd., animals developing oblique head posture and circling or rolling movements were observed. During a period of 21 weeks (142 days) a total number of 108 clinical cases were examined representing approximately five percent of the total number of animals maintained in this colony during that period. Ten to 20 percent of the SCID mice delivered to a customer (RP) in that period developed otitis media approximately 3 weeks after arrival. A suppurative lesion in the middle ear was found by autopsy in 90 (83 percent) of the cases. Bacteriology was performed on 50 (72 percent) of the animals. The agent was classified as Pseudonomys gladioli.

A prophylactic treatment with oxytetracycline (Terramycin® vet) in the drinking water did not seem to have any effect on the number of cases, however after cessation of the treatment the number of infections decreased spontaneously and no cases were observed later.

Only one case of the Pseudonomys gladioli infection from clinical specimens has been observed (5), however the closely related organism Pseudonomas cepacia has been seen associated with urinary tract infections and infections of patients with cystic fibrosis (1, 2, 3).

A series of cases of otitis media in a SCID mouse breeding colony and in a SCID mouse research colony was observed. An atypical Pseudonomas bacteria classified as Pseudonomas gladioli, was isolated in a great percentage of the cases, leaving this agent as the most probable cause of the lesions.

1. Ballard, R.W., N.J. Pallerni, M. Doudoroff, R.Y. Stanier, and M. Mandel. 1970. Taxonomy of the aerobic Pseudomonads: Pseudomonas cepacia, P. mariginata, P. allicola, and P. caryophylli. J.Gen. Microbiol. 60:199-214.
2. Cowan, S.T., J.G. Holt, J. Liston, R.G.E. Murray, C.F. Niven, A.W. Ravin, and R.Y. Stanier, eds. 1974. Bergeys Manual of Determinative Bacteriology: 8th edition. Baltimore:Williams & Wilkins Company.
3. Peir, G.B. 1985. Pulmonary disease associated with Pseudomonas aeruginosa in cystic fibrosis: Current status of the host bacterium interaction. J. Infec. Dis. 151:575-580.
4. Roth, J.B., J.D. Marshall, R.D. Allen, G.A. Carlson, and C.L. Sidman. 1990. Spontaneous Pneumocystis carinii pneumonia in immunodeficient mutant scid mice. Am. J. Pathol. 136:1173-1186.
5. Stead, D. 1991. Nat. Col. of Plant Path. Bact. Personal Communication.
6. Scott, R.A.W., B.A. Croy, D.H. Percy. 1991. Hepatitis in SCID-beige mice. Lab Anim. Sci. 41:166-168.
7. Sundberg, J.P., T. Burnsteirn, L.D. Schultz, and H. Bedigan. 1989. Identification of Pneumocystis carinii in immunodeficient mice. Lab Anim. Sci. 39:213-218.
8. Walzer, P.D., C.K. Kim, M.J. Linke, C.L. Pogue, M.J. Huerkamp, C.E. Chrisp, A.V. Lerro, S.K. Wixson, E. Hall, and L.D. Shultz. 1989. Outbreaks of Pneumocystis carinii pneumonia in colonies of immunodeficient mice. Infect. Immun. 57:62-70.

Bomholtgård Ltd., Bomholtvej, Denmark
*CIBA-GEIGY Ltd., Pharma Research and Development, Basel, Switzerland.
**Royal Agricultural and Veterinary University, Frederiksburg, Denmark
The authors would like to thank Dr. David Stead, Nat. Col. of Plant Path. Bact. (NCPPB), Hatching Green, Harpeden, Hertfordshire AL5 Gothenburg, Goldhedsgatan 10, S41346 Gothenburg for their cooperation in typing the isolated Pseudomonas.

TCR expression during early neonatal development in euthymic and athymic (nude) mice is restricted to the intestinal lamina propria
Bernard de Geus, Lex Nagelkerken, and Jan Rozing

Intestinal intra-epithelial lymphocytes (iIEL) form a heterogeneous population of T cells of which more than 50 percent differentiates independent of the thymus. This thymus independent iIEL population is composed of a population of T cells expressing the TCRgd and a T-cell population expressing the TCRab. Both populations express only the CD8a chains. It has been suggested that another cell population which is CD8a- but CD3- and TCR-, and is also present in the intestinal epithelium, forms a precursor population of the thymus independent iIEL. However, our results indicate that this hypothesis is false, since these CD3- and TCR- CD8a+ cells are present in high numbers in the intestinal epithelium of CB17-scid/scid mice and are absent at the same site in 2-7 day-old neonatal mice of euthymic as well as of athymic (nu/nu) origin. Furthermore, the localization of TCR positive cells in the same 2-7 day-old neonatal animals suggests that the intestinal lamina propria may be important in the differentiation of iIEL, since TCR+ cells in these animals are only located in the intestinal lamina propria. These TCR+ cells are CD8a- and Thy-1-.

TNO Institute of Aging and Vascular Research, Leiden, The Netherlands

Animal models of normal and leukemic human meatopoiesis by transplantation into immune-deficient mice
John E. Dick, Tsvee Lapidot, and Francoise Pflumio

A complete understanding of the organization of the human meatopoietic stem cell hierarchy and of the molecular events regulating the stem cell developmental program has been hampered by the absence of suitable in vivo stem cell assays. Perturbations in the normal stem cell program leading to neoplastic growth as the result of aberrant expression of key regulatory genes are also difficult to study because leukemic cells are difficult to grow in culture. Recent advances in the transplantation of human cells into immune-deficient mice provide an unprecedented opportunity to study human metaopoiesis-- both normal and abnormal--in the context of a small animal model. Our animal model was designed to reflect as close as possible the current approaches of bone marrow transplantation using intravenous injection of adult bone marrow into condition recipient mice. We have evidence that the murine microenvironment can support human stem cells and the provision of human growth factors can stimulate high level multi-lineage engraftment. In addition to normal cells, we have engrafted mice with cells obtained from patients with pre-B acute lymphoblastic leukemia (ALL) at different stages of their disease. All of the samples from patients at relapse grew rapidly and disseminated widely in the mice, while cells obtained from patients at diagnosis grew poorly if at all. This indicates that there is a correlation between growth in the SCID mouse and clinical outcome. These results establish that the SCID mouse is a powerful model to examine the biologic characteristics of the growth of human leukemic cells. It should also serve as an important system to test various therapeutic strategies targeted against drug-resistant leukemic cells. The combination of this in vivo model with high efficiency gene transfer methods will provide an important system to test the genetic alterations involved in leukemic progression.

Department of Genetics, Research Institute, Hospital for Sick Children, Toronto, Ontario, Canada; and Department of Molecular and Medical Genetics, University of Toronto, Toronto, Ontario, Canada

Differential in situ expansion of tumor-infiltrating lymphocytes (TIL) following adoptive immunotherapy of tumor-bearing immunologically-deficient mice is not sufficient to ensure a successful outcome to therapy
Robert Evans, Sonya J. Kamdar, and Theodore M. Duffy

C57BL/6J mice bearing the MCA/76-9 immunogenic sarcoma received combination therapy consisting of a single intraperitoneal injection of cyclophosphamide (CY) and adoptive immunotherapy (AIT) in the form of an intravenous injection of tumor-sensitized T cells. These were derived from spleens of tumor-bearers or mice immunized against the tumor cells, as well as from progressing or regressing tumors (tumor-infiltrating lymphocytes (TIL)). The use of positively-selected immune Thy-1+ cells or TIL for AIT resulted in a consistent pattern of permanent tumor regression in all experiments, despite the fact that these T-cell sources contained preponderantly either CD4+ or CD8+ T cells. Using congenic B6.PL.Thy-1a immune T cells or TIL, it was shown that 8 days after AIT, more than 95 percent of the TIL were donor in origin. When TIL were expanded for 6 to 8 days in rIL-2, resulting in populations comprised of more than 95 percent CD8+ TIL, and then used for AIT, it was seen that more T cells were required to induce permanent regression compared with unexpanded TIL. It was also seen that the incidence of permanent regressions became variable. Tumor regression induced by AIT was associated in all cases with a differential expansion of CD4+ and CD8+ TIL between days 5 and 9 after AIT with a parallel modulation of Thy-1, CD4, and CD8 antigen expression. These findings were confirmed by Northern hybridization analysis of RNA extracted from whole tumor tissue, as well as from positively selected Thy-1+, CD4+, and CD8+ TIL. AIT was shown to result in a pronounced time-related increase in Ly-2, Ly-4, IL-2 and IFNg mRNA levels supporting the view that AIT results in a differential activation of the immune and cytokine network at the tumor site. However, expansion of TIL in site appeared to be only a prelude to the induction of permanent tumor regression. When tumor-bearing immunological mutant mice, nude (b6.nu) and rhino (B6.hrrh), received AIT, a similar degree of in situ expansion of CD4+ and CD8_ TIL occurred. Permanent tumor regression was not seen despite the evidence for amplification of immune responses in these mice. It was found that TIL isolated from the treated immunological mutant mice were cytotoxic in both an in vitro and in vivo assay; there was an increased responsiveness of spleen cells to stimulation with Con A, as measured by the production of IL-2; and there was an increase in class II-MHC expression by tumor-associated macrophages. These and other findings indicated the potential of the mutant mice to respond to therapy and the inability to generate the ultimate destructive mechanisms in situ. The overall findings indicate that the expansion of TIL after AIT is a highly regulated event but in itself is not sufficient for the successful outcome of AIT. It is apparent that while expansion mainly involves those adoptively transferred TIL that accumulate at the tumor site, tumor eradication may well depend on a critical host component, such as a degree of immunoresponsiveness that determines whether expansion leads to those mechanisms needed for the induction of permanent tumor regression.

The Jackson Laboratory, Bar Harbor, Maine

Insights into mast cell development and function derived from analyses of mice carrying mutations at W/C-kit or Sl/MGF (SCF) loci
Stephen J. Galli, Barry K. Wershil, Edwin N. Geissler, John R. Gordon, Thomas R. Martin, and Mindy Tsai

Mast cell development is a complex process that results in the appearance of phenotypically distinct populations of mast cells in different anatomical sites. Mice homozygous for mutations at the W or Sl locus exhibit several phenotypic abnormalities, including a virtual absence of mast cells in all organs and tissues. Recent work indicates that W encodes the c-kit tyrosine kinase receptor, whereas Sl encodes a c-kit ligand that has been designated stem cell factor (SCF), mast cell growth factor (MGF), kit ligand (KL), and Steel factor. In vitro and in vivo studies indicate that recombinant SCF can induce the proliferation of both immature and mature mast cell populations and can induce the development of both connective tissue-type and mucosal mast cells. In addition, SCF can induce mast cell maturation and heparin synthesis. In addition to their value in identifying the products encoded at the W or Sl loci and defining the role of these products in mast cell development, W or Sl mutant mice also have been useful for the analysis of mast cell function. W/Wv mice can be locally and selectively repaired of their mast cell deficiency by adoptive transfer of in vitro-derived immature mast cells of congenic +/+ origin. Analysis of biological responses in mast cell-deficient W/Wv or Sl/Sld mice, the congenic normal (+/+) mice, and W/Wv mice locally and selectively repaired of their mast cell deficiency by adoptive transfer of mast cells of congenic +/+ origin, indicate that mast cells are essential for the expression of IgE-dependent cutaneous responses and certain other cutaneous inflammatory responses. This also indicates that TNF-a of mast cell origin represents an important mediator of the leukocyte infiltration which occurs during some of these reactions. However, mast cells appear to make no detectable contribution to the expression of T-cell dependent contact sensitivity responses in the skin. In addition, even though systemic activation of mast cells can induce anaphylactic responses associated with tachycardia, changes in pulmonary mechanics, and death, some anaphylactic responses can occur by mechanisms which are apparently completely independent of the mast cell.

Department of Pathology, Harvard Medical School, Boston, Massachusetts

Malignant behavior of human tumors in immunodeficient mice
A. Garofalo, R. Chirivi, B. Abbott, and R. Giavazzi

The malignant potential of human tumors in nude, triple deficient bg-nu-xid and SCID mice was investigated. When tumor lines (one melanoma, two colon, one kidney, and two ovarian carcinomas) were injected subcutaneously, all the mice produced local tumors with no significant differences in growth rate. A few spontaneous metastases to the lung were observed in SCID mice. Following the intravenous injection of A375 melanoma cells, bg-nu-xid mice showed a higher number of lung colonies compared to SCID and nude mice. Similarly, the intrasplenic injection of HT-29 colon carcinoma resulted in a higher number of liver colonies in bg-nu-xid mice compared to nude and SCID mice. These findings indicate that the SCID and bg-nu-xid mice may offer some advantages to study the malignant behavior of human tumor xenografts, and this may depend on the route of tumor cells injection and tumor cells type.

Mario Negri Institute for Pharmacological Research, Bergamo, Italy; and NCI-Federick Cancer Research Center, Frederick, Maryland

The use of immunodeficient mice for studying human hematopoietic neoplasms
R. Giavazzi, A. Garofalo, S. Bettoni, T. Barbui, J. Mayo, A. Biondi, and A. Rambaldi

Immunodeficient SCID and bg-nu-xid mice have been described as recipient for normal and transformed hematopoietic cell engraftment. We found that the injection of human acute leukemia cells into immunodeficient mice resulted in serially transplantable leukemic cell lines (2 myeloid and 3 lymphoid) that retained their original cytological features, immune surface antigens, and molecular characteristics. Among these leukemias, of particular interest was the observation that the intravenous injection of a human T-lymphoblastic lymphoma in bg-nu-xid mice resulted in systemic massive lymphomatous involvement with infiltration of bone marrow and various organs including spleen, liver, kidney, lung and meninges.

These findings confirm that immunodeficient mice can be used to propagate and establish human hematopoietic neoplasia. The stable features of these tumor lines and their ability to disseminate in a fashion that resembles the clinical picture of the original disease makes these animal models unique for studying the biology and therapy of human leukemia.

Mario Negri Institute for Pharmacological Research, Bergamo, Italy; and NCI-Federick Cancer Research Center, Frederick, Maryland

The Institute of Laboratory Animal Resources: A source of information on animal models
Dorothy D. Greenhouse

For 40 years, the Institute of Laboratory Animal Resources (ILAR) has assisted the biomedical research community by developing guidelines for the care and use of laboratory animals and providing information on animal models and sources of animals. ILAR is a part of the Commission on Life Sciences, National Research Council (NRC), National Academy of Sciences. The Academy is a private, nonprofit organization, created by congressional charter in 1863 to serve as an official advisor to the federal government on matters of science and technology. In 1916 the Academy founded the NRC as its principal operating agency. Nearly all substantive tasks of the NRC are carried out by carefully balanced committees of recognized scientific experts in the field of study. ILAR's ability to convene such committees has led to national recognition and acceptance of its reports.

Several aspects of ILAR's work are of particular interest to the participants of the Seventh International Workshop on Immune-Deficient Animals. The first is the report Immunodeficient Rodents: A Guide to Their Immunobiology, Husbandry, and Use. Prepared by the eight-member Committee on Immunologically Compromised Rodents and 16 invited participants, this report is aimed primarily at scientists new to the field of immunology. Following an introduction to immune function and immunodeficiency, the report addresses the genetics, pathophysiology, husbandry, and reproduction of each strain or stock of rodent known to have a spontaneously arising deficiency in immune function. The husbandry and mating systems for maintaining these models are also presented. The report was published in 1989 and is available for sale from the National Academy Press.

ILAR's most frequently used ongoing service is the Animal Models and Genetic Stocks Information Program. ILAR has developed a computerized database of commercial and investigator-held colonies of both commonly and less commonly used laboratory animals. Staff annually responds to hundreds of requests for information on sources of animals for laboratory investigation, appropriate animal models for studying human diseases and normal biologic phenomena, and animal care and treatment. On behalf of the International Committee on Standardized Genetic Nomenclature for Mice, ILAR assigns and distributes laboratory registration codes used to identify specific substrains and sublines of genetically defined rodents and rabbits and provides information on nomenclature.

Also relevant to this workshop is the quarterly journal ILAR News, which contains articles of interest to both biomedical and laboratory animal scientists. The journal pays special attention to providing information useful to institutional animal care and use committees and to scientists presenting protocols to these committees. It also contains inserts comprised of ILAR committee reports, individually authored documents, or bibliographies on alternatives to the use of live vertebrates. These inserts are a rapid, cost-effective means of disseminating information not readily available elsewhere. ILAR News is distributed free to more than 4,300 individuals and libraries worldwide.

To obtain a laboratory registration code or information on animal models, or to subscribe to ILAR News, write ILAR, National Research Council, 2101 Constitution Avenue, N.W., Washington, D.C. 20418. Tel: 202/334-2590; Fax: 202/334-1687.

Institute of Laboratory Animal Resources, National Research Council, Washington, D.C.

Depletion of CD4+ cells in MHC class II deficient mice
Michael J. Grusby, Randall S. Johnson*, Virginia E. Papaioannou**, and Laurie H. Glimcher

The normal development of mature CD4+ T lymphocytes requires their interaction with major histocompatibility (MHC) encoded class II molecules in the thymus. These highly polymorphic molecules are present on thymic cortical epithelial cells and their engagement with the ab T-cell receptor on immature thymocytes is thought to result in the positive selection of CD4+ T cells and their ultimate export to the periphery. We have used the technique of gene targeting in embryonic stem cells to disrupt the MHC class II Abb gene, and from these cells, have generated mice which lack sell surface expression of MHC class II molecules. Immunohistochemistry of frozen thymic sections and flow cytometry of peripheral lymphocytes confirmed that disruption of the Abb allele has led to a null phenotype with respect to MHC class II antigens. Analysis of these animals reveals that they are depleted of CD4+ T cells in lymphoid tissues and deficient in cell mediated immune responses, thereby providing genetic evidence for the requirement of class II molecules in the maturation and function of this T-cell subset.

Department of Cancer Biology, Harvard School of Public Health and Department of Medicine, Harvard Medical School, Boston, Massachusetts
* Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, Boston, Massachusetts
** Department of Pathology, Tufts University School of Medicine and Veterinary Medicine, Boston, Massachusetts

The role of Kit on murine development
S.-I. Hayashi, T. Kunisada, M. Ogawa, T. Era, N. Ohno, H. Yoshida, H. Maeda, K. Yoshinaga, S. Nishikawa, and S.I. Nishikawa

The murine mutation dominant white spotting (W) is in the proto-oncogene, c-kit. The receptor tyrosine kinase encoded by this gene has pleiotrophic effects on murine development including hemopoietic cells, melanocytes and germ cells. More recently, characterization of a factor that stimulates the growth of mast cells (MGF/SLF/KL) and that binds the Kit receptor has led to the identification of the Sl gene product. Phenotypic analysis of W and Sl mice has unequivocally proven the role of the receptor tyrosine kinase-mediated signaling in murine development in embryonal and postnatal life. However, previous studies have not specified which stage of these three lineage cells is functionally required.

Anti-Kit monoclonal antibodies were established by immunizing cultured mast cells which express Kit on their surface. Mast cells of W/W mice, which cannot express Kit due to a mutation at a splice donor site which results in deletion of exons including the transmembrane domain, were used as a negative control. These antibodies were used not only as a marker for Kit+ cells, but also as an antagonistic blocker capable of controlling Kit function in vivo. First, we identified at least two distinct Kit-dependent processes during melanocyte precursors in the mesodermal layer, which eventually leads to the melanocyte entry into the epidermal layer and the other during melanocyte activation along with the hair cycle in postnatal life.

Second, in gonadal tissues during postnatal development, the survival and/or proliferation of the differentiating type A spermatogonia is dependent on c-kit, but the antibody administration had no significant effect on the oocyte maturation and ovulation.

Finally, in hemopoietic progenitor cells, after Kit+ cells were removed from adult bone marrow cells, hemopoietic progenitor cells reactive to IL-3, GM-CSF, or M-CSF and also those which give rise to spleen colonies in irradiated recipients disappeared almost completely. To investigate the role of Kit in the hemopoiesis in vivo, we injected the antibody against Kit. As early as 2 days after the injection, almost all hemopoietic progenitors disappeared from the bone marrow. These results provide direct evidence that Kit is an essential molecule for constitutive intramarrow hemopoiesis, especially for the self-renewal of hemopoietic progenitor cells at various stages of differentiation. The roles of Kit on hemopoiesis in fetal stage and on lymphocyte generation will also be discussed.

Department of Pathology, Institute on Medical Immunology, Kumamoto University Medical School, Kumamoto, Japan

Abnormal thymocytopoiesis in viable motheaten (mev) mutant mice
Sandra M. Hayes, Leonard D. Shultz*, and Dale L. Greiner**

Mice homozygous for the viable motheaten (mev) allele manifest immunodeficient and autoimmune disorders. Premature thymic involution occurs at four weeks of age in homozygous mev mice, and their bone marrow prothymocytes fail to repopulate the thymus of irradiated recipients following intravenous transfer. However, mev/mev bone marrow cells are capable of generating normal numbers of donor-origin thymocytes following intrathymic adoptive transfer. Based on these observations, we have suggested that mev/mev bone marrow prothymocytes may fail to migrate to and/or enter into the thymus of adoptive recipients. To investigate this, we have developed an assay system to study the homing patterns of prothymocytes following intravenous injection. In this assay system, bone marrow cells from Ly5.2 donors are injected intravenally into irradiated Ly5.1 congenic recipients. At various times after reconstitution, the spleen, thymus, or bone marrow were recovered and the cells adoptively transferred intrathecal into irradiated secondary Ly5.1 congenic recipients. The homing pattern of prothymocytes to the tissues of the primary recipients was assessed by quantitating the number of donor-origin (Ly5.2+) thymocytes generated by each of the tissues in the secondary recipients. Our results demonstrated that mev/mev prothymocytes differed in their kinetics of seeding and in their tissue distribution in adoptive recipients as compared to that of wild type prothymocytes. Furthermore, we observed that at least some mev/mev prothymocytes are able to home to the thymus of irradiated recipients, but fail to proliferate and differentiate into thymocytes in the absence of an "accessory" cell or factor provided by normal bone marrow cells. However, the homing of mev/mev prothymocytes to the thymus of irradiated recipients, even in the presence of the "accessory" cell or factor, is not as efficient as that observed by wild type bone marrow prothymocytes. These results suggest that the abnormalities in thymocytopoiesis in mev/mev mice may in part result from environmental or "accessory" cells or factors, as well as from intrinsic genetically determined defects in prothymocytes.

Department of Pathology, University of Connecticut Health Center, Farmington, Connecticut
* The Jackson Laboratory, Bar Harbor, Maine
** Department of Medicine, University of Massachusetts Medical Center, Worcester, Massachusetts
Supported in part by ACS IM505 and by NIH grant CA20408.

HIV infection of hu-PBMC-SCID mice
RuthAnn M. Hesselton, Richard A. Koup, Kevin S. Byron, and John L. Sullivan

The development of animal models for HIV infection which can be used for the testing of therapeutic agents and vaccines is of utmost importance in the battle against AIDS. One recently described model is the human peripheral blood mononuclear cell reconstituted SCID mouse (hu-PBMC-SCID) (1). In this system, SCID mice are injected intraperitoneally with 20 x 10 6 human donor PBMC which migrate to engraft in lymphoid tissues of the mouse. These engrafted cells are then susceptible to infection with HIV (2). We have sought to further characterize human reconstitution and subsequent HIV infection of SCID mice to establish a baseline on which to evaluate the effects of antiviral and immune modulating interventions.

We have looked at the significance of reconstitution variables such as age of mice at reconstitution, number of donor cells, EBV status of donor, type of cells used for reconstitution, and variation between individual donors. Reconstituted mice produce human immunoglobulin at levels about one log lower than that seen in normal human serum, but produce soluble CD8 and IL-2R (CD25) at levels considerably higher than those seen in normal human serum. We have examined cell surface phenotype of reconstituted cells, proliferative responses of reconstituted cells, and histological appearance of lymphoid tissues. In HIV infected SCID mice, we have examined the route of administration of virus, titer of virus required for infection, optimum time for infection after reconstitution, optimum time for recovery of virus, recovery of virus from various tissues (using culture, direct p24 assay, and PCR), and quantitation of recovered virus. Preliminary results of protection studies using HIV-specific cytotoxic T-lymphocyte clones show that such CTL can be used in vivo in mice reconstituted with cells from MHC-matched donors to alter the course of HIV infection. We are also assessing the feasibility of therapeutic interventions using pharmaceutical agents.

1. Mosier, D.E., R.J. Gulizia, S.M. Baird, and D.B. Wilson. 1988. Transfer of a functional human immune system to mice with severe combined immunodeficiency. Nature 355:256-262.
2. Mosier, D.E., R.J. Gulizia, S.M. Baird, D.B. Wilson, D.H. Spector, and S.A. Spector. 1991. Human immunodeficiency virus infection of human-PBL-SCID mice. Science 251(4995):791-794.

Department of Pediatrics, Program in Molecular Medicine, University of Massachusetts medical school, Worcester, Massachusetts

Peyer's patch cells reconstitute the peripheral and mucosal immune systems of SCID mice
David M. Hilbert, K. Holmes, A.O. Anderson, and S. Rudikoff

Previous reports have suggested that murine Peyer's patches contain a selected pool of lymphocytes characterized by a high frequency of IgA precursor cells and expression of Peyer's patch specific homing receptors. In an effort to further understand the lymphocytic composition and immunologic potential of murine Peyer's patches, DBA/2 Peyer's patch cells were transferred intravenally to immunodeficient SCID mice. Using this system we assessed the ability of Peyer's patch cells to reconstitute the immune system in SCID mice as measured by serum Ig levels, lymphoid organ composition, and immunohistological analyses. Results indicate that Peyer's patch cells phenotypically and functionally reconstitute both a normal peripheral and mucosal immune system in SCID mice. Serum IgA, IgG, and IgM levels appear normal within 4 weeks of reconstitution and remain so for up to one year. The observed levels and isotypic profiles of serum Ig, as well as antibody from antigen-specific TI or TD responses, is indistinguishable from that of normal DBA/2 and BALB/c mice. The cellular reconstitution of the spleen, lamina propria, and lymph nodes also appears normal within 8 weeks of reconstitution. Each lymphoid organ contains the expected ratio of T and B cells organized into characteristic tissue-specific structures. Taken together, these results suggest that removal of cells from the Peyer's patch miroenvironment alleviates the apparent isotope restriction and homing properties previously attributed to these cells and, that the lymphoid microenvironments into which these cells subsequently migrate clearly dictate both the composition and organization of lymphocytes within it. Although the reconstitution of the peripheral lymphoid tissues appears normal, analyses of thymii from these animals revealed a markedly different cellular composition and organization than that observed in normal thymus. The thymii of Peyer's patch reconstituted SCID mice contain three lymphocytic populations. By day 84 post transfer, approximately half of the cells in the thymus are Thy1+, IL2R+, Qa-2-, CD4-, and CD8- cells. The phenotype of these cells is consistent with that previously described for endogenous immature SCID thymocytes. The remaining cells are TCR ab, donor derived cells which express either the helper (CD4+, CD8-) or cytotoxic (CD4- CD8+) phenotype characteristic of mature peripheral T cells. No double positive donor derived cells were detected.

Clearly the SCID mouse is an invaluable model for the study of immune system reconstitution, lymphocyte trafficking, and acquired immune function resulting from transfer specific lymphoid populations.

Laboratory of Genetics, National Cancer Institute, National Institutes of Health, Bethesda, Maryland

Changes induced by nonshivering thermogenesis in macrophage populations and in NK cells of nude mice
M. Holub, V. Vetvicka*, J. Houstek**, and A. Fiserova***

Exposure of individually caged mice to 22ºC induces in nude mice profound physiological alterations in comparison with nude and haired littermates exposed to 28ºC (catecholamines, 5'-deiodase, thyroid hormones, development of thermogenic mitochondria in brown fat). A long-term exposure (3 to 6 weeks), associated with a marked hyperplasia of the thermogenic tissue, provoked an increase in the percentage of phagocytosing and of MAC-1+ peritoneal macrophages. Also blood professional phagocytes increased their phagocytic performance. However, the density of FcR, MAC-1, F 4/80 and Ia markers decreased on peritoneal macrophages, among which large cells predominated in cold-exposed nu/nu. During 2, 3, and 6 weeks exposure, the percentage of MAC-1+ cells gradually increased in bone marrow and spleen; MAC-3+ cells gradually increased in the spleen; and a sharp rise occurred in FcR+ cells in all cold-exposed nude mice in bone marrow, spleen, and lymph nodes.

Also spleen NK cell activity (cytotoxic test) as higher in the nudes during the long-term exposure, and so were the numbers of asialo-GM1+ cells in different tissues.

Since some of the changes described (phagocytosis, NK activity) occur also after a short-term (6-24 hrs) exposure to 22ºC, it is likely that catecholamines have an important role in the cold-adaptation in nude. The changes in another hairless mutant (BFU mice, not immunodeficient) are much less pronounced.

Institute for Clinical and Experimental Medicine, Czechoslovak Academy of Sciences, Prague, Czechoslovakia
* Institute of Microbiology, Czechoslovak Academy of Sciences, Prague, Czechoslovakia
** Institute of Physiology, Czechoslovak Academy of Sciences, Prague, Czechoslovakia
*** Institute of Microbiology, Czechoslovak Academy of Sciences, Prague, Czechoslovakia

Immunological defects of nu/+ heterozygotes
M. Holub, J. Madar*, and E. Necas**

It was verified in an improved CFU/s assay that hematopoietic stem cells of the bone marrow are reduced both in nu/nu and nu/+ mice (compared to +/+, C57BL/10ScSn). Thymus weight is also reduced in nu/+ mice compared to +/+, the reduction being more pronounded in males (1-2 months of age).

In BALB/c mice, the antibody-forming cells (AFC) in draining lymph nodes 9 days after immunization with a thymus-dependent antigen (hen egg lysozyme in incomplete Freund's adjuvant), were reduced by two orders in nu/nu, compared to +/+. Nu/+ mice displayed an extreme variability and their AFC counts were spread over the whole range from one to 5 x 10 3 AFC per 10 6 cells. The same antigen applied in a soluble form 10 days prior to immunization produced a clear-cut tolerance (depression of AFC by two orders) and only a very moderate decrease in AFC in nu/+ mice. On the other hand, identical reactivity of +/+ and nu/+ mice to Concanavalin A and E. coli lipopolysaccharide was observed in the in vitro spleen cell assay. The results corroborate the assumption that the phenotypic expression of the nude gene is variable in heterozygous animals.

Institute for Clinical and Experimental Medicine, Czechoslovak Academy of Sciences, Prague, Czechoslovakia
* Institute of Molecular Genetics, Czechoslovak Academy of Sciences, Prague, Czechoslovakia
** Department of Pathophysiology, Charles University, Prague, Czechoslovakia

Reduction of lymphokine-activated killer cell activity by the introduction of the beige gene into SCID mice
Chen Hua, Bo Chang*, Bao-Li Ma, Yan Li*, and Ge Wang*

The adoptive immunotherapy of cancer using lymphokine-activated killer (LAK) cells has been intensively investigated, but the lineage of LAK cells is poorly understood. Hasui reported that the SCID (T- and B-cell deficiency) had a high level of LAK cell activity, which implied that the majority of LAK activity exerted by blood or spleen lymphocytes might be attributable to IL-2-activated NK cells. But there is evidence that immature T cells are present in SCID mice. It is possible that LAK effector cells may be generated from some immature T cells of congenitally immunodeficient mice.

To further clarify whether the LAK cells come from NK cells or immature T cells, we introduced the beige gene (NK cell deficiency) into the SCID mice and observed the influence of the beige gene on the LAK activity of scid mice. The introduction of the beige gene into SCID mice was carried out by successive crossings and intercrossings, starting from C57BL/6N-bg females (f) and C.B17-scid males (m). After the beige gene was introduced into SCID mice and the new stock of scid/scid-bg/bg mice showed a combined immunodeficiency of T, B, NK cells, we further applied the new stock of scid/scid-bg/bg mice and scid mice to analyze the source of the precursor of LAK cell.

The proliferation of splenocytes of scid/scid-bg/bg mice could not be induced by addition of rlL-2 (1,100 ± 612.34 cpm at IL-2 50m/ml vs. 10,523.07 ± 1107.30 cpm for C.B17-scid mice and 3,910.12 ± 816.34 for control BALB/c mice, p<0.01). Morphological studies of incubated cells with IL-2 did not show a clonal proliferation of lymphocyte-like cells, but did show a large amount of macrophages and mast cells. They also showed a low I..AK cell activity (5.80 ± 5.50, vs. 97.7 ± 13.6 for C.B17-scid mice at IL-2 500m/ml; E/T 100.1, 3 days, p<0.01). Importantly, we have shown that rlL-2-cultured splenocytes of scid/scid-bg/bg mice did not give augmented cytotoxic activity against NK-resistant P815 cells with an increase in the incubation days or in a dose-response study. The reduction of LAK activity in the scid/scid-bg/bg mice was obviously related to the introduction of the beige gene which causes an NK cell deficiency. This implied that the LAK activity in SCID mice may have a strong relationship with the NK cell.

This is the first time that we directly demonstrated the relationship between LAK cells and NK cells by introduction of the beige gene into SCID mice. The introduction of a specific, known immune-deficient gene into various strains of immunodeficient mice would help us to understand the influence of the induced gene (bg gene) on the LAK cell activity. These methods also deliver a new and powerful way to analyze the source of highly potent cytotoxic effector cells of LAK cells for adoptive immunotherapy of cancer in experimental animals and humans.

Department of Basic Immunology, Shanghai Institute of Immunology, the Shanghai Second Medical University, Shanghai, China
*Institute of Laboratory Animal Science, Chinese Academy of Medical Science, Beijing, China

The study of adult nasopharyngeal mucosa xenograft and dinitrosopiperazine (DNP) carcinogenesis
P. G. Huang, Y. T. Cai*, M. Y. Mo, Q. Z. Cai, and S. J. Shen

Adult nasopharyngeal mucosa from 44 cases of chronic nasopharyngitis diagnosed by pathology was transplanted into 59 nude mice. The mice were divided into two groups-the control group (13 nude mice) was given transplants without any treatment, and the experimental group (46 nude mice) was subcutaneously injected with 15mg/kg of DNP twice a week, for 6 to 15 weeks beginning 10 to 16 days following transplantation. In the experimental group, the xenografts in 27 nude mice were examined after 18 to 37 weeks of transplantation, and survived in 20 mice. The survival rate was 74.1 percent. Seven out of 20 of the surviving epithelia showed focal hyperplasia and dysplasia, and four carcinomas in situ and four early infiltrating carcinomas showed carcinomatous changes. The carcinoma incidence was 40 percent. In the control group, the xenografts in seven nude mice were examined after 8 to 35 weeks of transplantation, and survived in six mice. The surviving epithelia did not develop dysplasia and carcinomatous changes. The results suggested that DNA may induce the adult naso-pharyngeal epithelia into precarcinomatous lesions and carcinomatous changes in nude mice. According to the results, the authors analyze the factors related to carcinomatous changes of nasopharyngeal epithelia.

Department of Oncology, Zhanjiang Medical College, The Second Hospital, Zhanjiang City, Guangdong, China
*Department of Otorhinolaryngology, Zhanjiang Medical College, The Second Hospital, Zhanjiang City, Guangdong, China

Acute human-versus-mouse graft-versus-host disease in normal and immune-deficient mice
W. Huppes and D. W. van Bekkum

Recent reports of persistent engraftment of human lymphocytes in hereditary immune-deficient SCID and bg-nu-xid mice have raised the question why these animals did not develop graft-versus-host disease (GvHD). We postulated that this was due to lack of human specific lymphocyte cytokines in the recipient mouse tissues.

To overcome this problem we transplanted large numbers of human peripheral blood lymphocytes (PBLs) (2 x 10⁷ cells per gram body weight and more), assuming that these numbers would produce enough human lymphokines to support proliferation of human reactive T-cell clones. The SCID mice were conditioned with 400 mg/kg of cyclophosphamide, the bg-nu-xid mice with 7-9 Gy total body irradiation (TBI). Under these conditions all mice developed acute GvHD. Similarly, acute GvHD occurred in newborn normal BCBA mice and in 3-week-old xid mice following high dose TBI and grafting of large numbers of human PBL. The clinical manifestations and the histopathology of this xenogeneic acute GvHD are quite different from those of allogeneic GvHD. The former is primarily confined to the hematolymphoid tissues and locations adjacent to accumulations of proliferating lymphoblasts, such as the peritoneal cavity in case of intraperitoneal transplantation. The discordant xenogeneic GvHD is specifically induced by human T-lymphocytes and can be abrogated by treatment with anti-human T-cell serum.

ITRI-TNO, Rijswijk, The Netherlands

Prediction of clinical antitumor effect based on Clinically Equivalent Dose preestimated by animal scale-up
Makoto Inaba, Tazuko Tashiro, Yasuyuki Ohnishi*, Yoshito Ueyama*, and Tatsuji Nomura*

In our previous reports, we have emphasized that it is very important to use Clinically Equivalent Dose (CED), previously called Rational Dose (RD), to reproduce or predict the clinical antitumor effect of a drug by human tumor/nude mouse model. CED is a dose for mice pharmacokinetically equivalent to the clinical dose. In order to determine the CED, we need not only experimental but clinical data on pharmacokinetics and maximum tolerated dose (MTD). Therefore, it is impossible to determine the CED of a new drug on the preclinical stage. As a new trial to overcome this difficulty, we attempted to preestimate the CEDs of new drugs using human pharmacokinetic parameters and MTDs predicted by an animal scale-up procedure. Practically, assuming two clinical antitumor drugs to be "new drugs," we preestimated their CEDs and examined the antitumor effects of these drugs at CEDs on human tumor xenografts.

Cancer Chemotherapy Center, Japanese Foundation for Cancer Research, Tokyo, Japan
*Central Institute for Experimental Animals, Kawasaki, Japan

Comparative study on Pneumocystis carinii infection in SCID and nude mice
Mamoru Ito, Takashi Kuramochi, Kyoji Hioki, and Tatsuji Nomura

Pneumocystis carinii is an opportunistic agent causing severe pneumonia in immunocompromised patients, notably those with AIDS. Since SCID and nude mice also spontaneously develop severe pneumonia with immunosuppressants, they can be considered useful as animal models for the research of P. carinii infection in such patients. In this study we attempted to elucidate the characteristics of P. carinii infection in SCID mice by comparing experimental infection in SCID with that in nude mice.

Twenty-five C.B17-scid and BALB/cA-nu females maintained in our institute under SPF conditions were intranasally inoculated with 0.9 x 10⁵ P. cannii cysts. After inoculation, live weights were measured each week, and day of death was recorded. In another experiment, lungs were collected each month post-inoculation and the number of P. carinii in the lungs was calculated by staining with toluidine blue O. Histological examinations of formalin-fixed lung sections were also performed.

Weight loss after infection in SCID mice was delayed compared to nude mice. Moreover, the survival rate in SCID mice was much higher than that in nude mice. Only one out of 25 SCID mice died 20 weeks after infection, compared to 13 out of 25 nude mice (52 percent). These results were reproduced in another experiment using 15 of each type of mouse.

In contrast, the number of lung cysts was higher in SCID mice than in nude mice. A histological examination showed that both types of mice developed P. carinii pneumonia typical of immunodeficient animals, and showed small numbers of infiltrating cells and extensive multiplication of protozoa. However, infiltration of lymphocytes into the lungs of SCID mice was less significant than in nude mice. These results show that the SCID mouse paradoxically survived longer than the nude mouse despite a higher degree of P. carinii multiplication. It is suggested that host cellular responses are responsible for impairment of lung function in P. carinii pneumonia rather than multiplication of the organism itself.

Central Institute for Experimental Animals, Kawasaki, Japan

Feline immunodeficiency virus infection of SCID mice engrafted with feline tissues: A murine model for HIV drug testing
C. M. Johnson, D. W. Selleseth, T. A. Childers, M. N. Ellis, M. Tompkins, and W. A. F. Tompkins

Animal models of Human Immunodeficiency Virus (HIV) infection are essential for preclinical evaluation of potential antiretroviral drugs and vaccines. SCID mice engrafted with human tissues and infected with HIV are useful models because they utilize the actual human pathogen in a host amenable to large-scale drug testing. However, public health concerns regarding accidental infection of laboratory workers with HIV restrict this model to relatively few laboratories. Feline Immunodeficieney Virus (FIV) is a lentivirus that induces immunodeficiency in cats by depleting CD4-bearing peripheral blood lymphocytcs (PBL), thereby inverting the CD4:CD8 ratio. Chronically-infected cats become susceptible to a wide variety of opportunistic pathogens, suggesting that FIV-infected cats are useful models for the immuno-pathogenesis of HIV-1 infection in humans. Furthermore, FIV is not infectious to humans and responds to reverse transcriptase-targeted drugs similarly to HIV. Therefore, SCID mice engrafted with feline tissues and infected with FIV would be valuable in preclinical screenings of potential AIDS drugs.

SCID mice were anesthetized and surgically implanted in the mammary fat pad with sections of fetal feline thymus and lymph node, then given intraperitoneal injections of feline liver, bone marrow, peripheral blood lymphocytes, and spleen lymphocytes. Two weeks after implantation, mice were injected intraperitoneally with 7 x 10⁶ FIV-NCSUl-infected feline PBL. Ten mice were given Retrovir^TM (Azidothymidine, AZT) at 125 mg/kg/day in the drinking water beginning 24 hours prior to virus challenge and continuing for 2 weeks. Two weeks post infection, mice were sacrificed and implants were analyzed for FIV proviral DNA by PCR. Grafting efficiency was determined by PCR amplification of feline-specific sequences of the c-fes protooncogene from the blood and spleen.

Feline cells were detected in 23 out of 28 mice (82 percent) in either peripheral blood or spleen lymphocytes. The number of mice positive for FIV (summarized below) indicates a lower frequency of detection of FIV provirus in AZT-treated animals as compared to untreated.

Number of Mice FIV Positive by PCR

	Untreated	AZT Treated
Thymus Implant	11/17 (65%)	2/10 (20%)
Lymph Node Implant	11/17 (65%)	4/10 (40%)
Both Implants	8/17 (47%)	0/10 (0%)


Hybridization intensities of FIV-positive samples in which equal amounts of DNA amplified by PCR were compared to determine relative levels of provirus in each sample. Comparison of five untreated mice with five AZT-treated mice showed a significant reduction in provirus burden associated with AZT treatment. These data suggest that the FIV-infected SCID-fe mouse is a safe, realistic murine model for HIV antiviral therapy.

North Carolina State College of Veterinary Medicine, Raleigh, North Carolina; and Burroughs Wellcome Co., Research Triangle Park, North Carolina

Helminths infection in SCID mice
Masao Kamiya, Yuzaburo Oku, Hong-Kean Ooi, Matthew C. Playford, Kenji Ishiwata, Munehiro Okamoto, Hirofumi Sakai, Haruo Kamiya*, Takashi Inaba*, and Mamoru Itot

The course and characteristics of several helminths infections in SCID (C.B17-scid/scid) mice and in normal C.B17 mice were compared. More Taenia taeniaeformis cysticerci were recovered from SCID mice, in which almost no eosinophil and macrophage infiltration nor necrosis of the hepatic cells around the liver cyst were observed. Rapid asexual propagation of Echbtococcus multilocularis cysts was observed in SCID mice intraperitoneally injected with cyst suspension. Numerous but mostly sterile fluid-filled cysts of E. granulosis were recovered from the abdominal cavities of SCID mice 10 months after intraperitoneal injection with hydatid sand. Higher intestinal worm recovery and a longer period of parasitism of Trichinella pseudospiralis were observed in SCID mice than in normal mice. Non-specific acute phase serum protein and intestinal eosinophilia response were observed in both the Trichinella-infected SCID and normal mice. However, more muscle larvae were recovered from the SCID than from the normal mice. Unexpectedly, fewer worms were recovered from SCID mice than from the normal mice in primary infection with Schistosoma mansoni, but no evidence of acquired immunity to the reinfection was observed in the SCID mice. Experiments were also carried out to examine the possible role of the SCID mouse as an alternative definitive host for Taenia crassiceps, E. multiloculatis, and Angiostrongylus cantonensis. Underdeveloped tapeworms of T. crassiceps were recovered from the intestine 12 days postinfection from prednisolone-treated SCID mice inoculated orally with cysticerci but not from the un-treated SCID mice. The protoscoleces of E. multilocularis failed to establish in the intestine of the SCID mice and were expelled 1 day postinfection. A high mortality rate was observed in SCID mice infected with A. cantonensis at around 30 days postinfection. In the few mice that survived longer, A. cantonensis young adult worms failed to migrate to the lungs, which is the final predilection site in its natural rat host. On the whole, our results showed that there is not much difference between SCID and normal C.B17 mice in the ability to serve as an alternative definitive host for the tested helminths.

Department of Parasitology, Faculty of Veterinary Medicine, Hokkaido University, Sapporo, Japan
*Department of Parasitology, Hirosaki University School of Medicine, Hirosaki, Japan the Central Institute for Experimental Animals, Kawasaki, Japan

Increased antitumor activity of 5-fluorouracil by 1-1eucovorin and interferon on human colon carcinoma xenograft transplanted into nude mice
Suguru Kase, Tetsuro Kubota, Masahiko Watanabe, Hirokazu Tanino, Tooru Takeuchi, Toshiharu Furukawa, Susumu Kodaira, Kyuya Ishibiki, and Masaki Kitajima

We have investigated the modulated antitumor activity of 5-fluorouracil (5-FU) by 1-leucovorin (LV) and recombi-nant human interferon a-2a (IFN) against human colon carcinoma xenograft (Co-4) serially transplanted into nude mice. 5-FU showed dose-dependent antitumor activity on Co-4 and a significant combination antitumor effect was obtained, when 200 mg of LV per kg was administered intraperitoneally on -1 and 0 hours before 5-FU (90 mg/kg, intraperitoneally) treatment. The modulated antitumor effect of 5-FU was dependent on the dose of LV and correlated with the increment of the thymidylate synthase (TS) inhibition. When IFN was administered subcutaneously in a schedule of qdx14 at doses of 6 x 10³, 6 x 10⁴ and 6 x 10⁵ IU/mouse, IFN alone indicated the antitumor activity in a dose-dependent manner. The combination treatment of 5-FU (60 mg/kg intraperitoneal, q4dx3) and IFN (6 x 10⁴ IU/mouse subcutaneously, qdx14) showed an additive antitumor effect on Co-4 without any changes of TS inhibition. These results showed that the modes of action of 5-FU modulator are diverse and a human tumor xenograftnude mouse system would be useful in evaluating these modes of action.

Department of Surgery, School of Medicine, Keio University, Tokyo, Japan

Orthotopic xenotransplantation of human lung cancer in the lungs of nude mice and experimental observation of invasion and metastasis
Chang-shu Ke and Wei-hua Li*

With the purpose of establishing a model for orthotopic xenotransplantation of human lung cancer in nude mice, we first made the methodological experiment replacing lung cancer cell suspension by 1 percent Methylene Blue solution. The success rate was 26/32. Then we implanted human lung giant cell carcinoma cell strain into the lung of BALB/cA nude mice by the same via bronchial procedure. We observed the tumorigenicity, local invasion, and metastatic pattern of orthotopic xenografts.

When implanting 2.0 x 10⁵ tumor cells per animal, tumorigenicity was 0/4; when implanting 1.0 x 10⁶, it was 1/5; when implanting 2.0 x 10⁶, it was 7/18. The dissemination of xenografts within airway (2/8) and invasion to diaphragm (5/8) were discovered. Lymphatic, vascular, and seeding metastasis all occurred in tumor-bearing animals. The pathohistological, ultrastructural features and expression of Vimentin confirmed by immunohistochemistry were consistent with the parent human lung giant carcinoma. These results indicate that the tumor cell strains grow autonomously when implanted orthotopically and the invasive and metastatic pattern of xenografts more closely imitates the clinical manifestation of lung cancer patients than the subcutaneous counterparts. It suggests that the implant site of the host affects the biological behavior of the xenografts. The same Vimentin expression implicates the similarity between the microenvironment of human and nude mice lungs. Nude mice lungs can be used as a better model for study of invasion, metastasis, and experimental therapy of lung cancer.

Department of Pathology, Wuhan General ttospital, Wuhan, China
*Department of Pathology, 301 Hospital, Beijing, China

Point mutation and its stability of ras oncogene in human neoplasms and tumor xenografts
Hiroshi Kijima, Masato Nakamura, Yasuyuki Ohnishi*, Takashi Imanishi, Yoshito Ueyama, Keiichi Watanabe, and Norikazu Tamaoki

We have examined the active c-Ki-ras gene by point mutation at the twelfth codon in human neoplasms and tumor xenografts, using the polymerase chain reaction and oligonucleotide hybridization methods. Point mutations at the twelfth codon of the Ki-ras gene were detected in 20.3 percent (15/74) of the primary human neoplasms and 19.3 percent (11/57) of the human tumor xenografts. Xenografts of human pancreas carcinomas revealed high frequency (83.8 percent) of the point mutation as well as the primary pancreas carcinomas (80 percent). The mutation of GGT (Gly) to GAT (Asp) was most frequent in the tumor xenografts (63.3 percent) as well as the primary human neoplasms (73.3 percent). The twelfth codon of Ki-ras showed no discrepancy between the original human neoplasms and their xenografts in 16 (84.2 percent) of the 19 cases. Two of the 19 cases revealed point mutation of GAT (Asp) in the xenografts despite having had no mutation in the original tumors. The other one case showed point mutation of GAT (Asp) in the original tumor despite an absence of mutation in the xenograft. These results suggested that point mutation at the twelfth codon of Ki-ras gene was stable in human neoplasms and their tumor xenografts.

Department of Pathology, Tokai University School of Medicine, Kanagawa, Japan
*Central Institute for Experimental Animals, Kanagawa, Japan

SCID and bg-nu-xid mice implanted with human fetal thymus and liver: A comparison
T. R. Kollman, M. M. Goldstein, A. Rubenstein, M. Hachamovitch, and H. Goldstein

We implanted human fetal (17-19 gw) thymus and liver under the renal capsule of 56 SCID (C.B17-scid/scid) mice and nine bg-nu-xid mice. Histological examination revealed that implants of human fetal thymus and liver grew in SCID (examined after I and 3 months) and in bg-nu-xid mice (examined after 1 month) into a structure comparable to normal fetal thymus but with interspersed areas of myeloid hematopoiesis. No difference between the appearance of the implant in SCID and bg-nu-xid mice at 1 month post-implantation was observed. Human and mouse serum immunoglobulin levels were measured 1 month after the implantation of tissue from the same fetus. Human IgG was detected in most of the implanted SCID mice and in all implanted bg-nu-xid mice, whereas human IgM was detected in few of the implanted SCID mice but was absent in all of the implanted bg-nu-xid mice. Growth of the implant was not affected by the presence of leakiness (>50 g/mi mouse IgG) in SCID mice, and leakiness was not induced in most non-leaky SCID mice following implantation. In contrast, most implanted bg-nu-xid mice showed a marked increase of mouse IgG production, suggesting that mouse B-cell function in bg-nu-xid mice was en-hanced by the implantation of human fetal thymus and liver. We conclude that unirradiated bg-nu-xid mice do not seem to offer an apparent advantage over unirradiated SCID mice as recipients for human fetal thymus and liver. In addition, the hypogammaglobulinaemia in bg-nu-xid mice is normalized following implantation of human fetal thymus and liver.

Albert Einstein College of Medicine, New York, New York

Nude mouse model for development of radio-labeled monoclonal antibodies to treat human lung cancer
Aurelia M. C. Koros and Geoffrey Levine*

Lung cancer continues to be a serious public health problem despite an increased understanding of the biology of the tumors. Presently, lung cancer is the leading cause of cancer-related deaths among both men and women in the United States with approximately 142,000 new cases diagnosed each year and about the same number of deaths. Survival rates have not improved in 20 years, even though there have been advances in categorization of lung tumor types.

Currently, lung cancers can be categorized on a clinical basis into two broad groups, small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC). SCLC comprise about 25 percent of all lung cancers, and are frequently metastatic at diagnosis; they are, however, more responsive to chemotherapy than are the NSCLC. Therefore, treatment for SCLC has been chiefly chemotherapy, whereas treatment for NSCLC has been primarily surgery and radiation. A small percentage of NSCLC have been identified that share neuroendocrine features with SCLC and may respond to chemotherapy. Although there have been advances in diagnosis as well as in chemotherapy and radiation therapy, patient survival rates are poor. Current strategies are inadequate.

Chemotherapy is not sufficiently specific for neoplastic cells; toxicity kills patients either directly, or indirectly via suppression of their immune system which leads to patients' susceptibility to infections and death. There is heterogeneity in tumors; not only in initial lesions, but there is evolution of cell types during metastatic progression leading to multidrug resistance. More specific treatments as well as diagnostic techniques are needed.

Newer diagnostic methods based on monoclonal antibody (MOAB) panels, immunohistochemistry, flow cytometry, and even molecular probing may ultimately lead to "individual" patient management. Although there probably are no unique tumor specific antigens, there is preferential expression of certain antigens by particular tumors; for example, neuroendocrine tumors including SCLC and neuroblastomas express CD56 and CD57 antigens as defined by MOABs from the fourth Human Leucocyte Differentiation Antigen Workshop. Through use of panels of MOABs, individual tumors may be phenotyped very specifically, and in the future, appropriate MOABs may be selected for targeted therapy.

One example of in vivo testing of a candidate MOAB for future patient use is provided by anti-NKH1. Anti-NKH1 not conjugated to either an immunotoxin or radioisotope has very limited efficacy in inhibiting the growth of an unusual SCLC line, SHP-77, which has biochemical features of an SCLC "classic," but histological features of an SCLC "variant." It is especially useful for emphasizing the plasticity of human tumors which evolve as they metastasize during therapy.

To test potential therapeutic efficacy of anti-NKH1 (Coulter Immunology), athymic nude mice (Harlan Sprague Dawley) were injected with SHP-77 cells in each of two sites on the abdomen. When visible tumors appeared, mice were divided into two groups. One group was injected intraperitoneally weekly with anti-NKH1 murine MOAB in RPMI 1640 cell culture medium. The other group received RPMI 1640 alone. Tumors were measured weekly. The anti-NKHl-treated group of mice had xenografts that grew slightly more slowly and were phenotypically different as defined by an MOAB panel, but were histologically indistinguishable in H and E sections compared with controls. These results show treatment of mice with anti-NHKl must be started early after tumor cell inoculation and possibly could be given more frequently than once a week to inhibit tumor growth. More importantly, however, this model empha-sizes that for future extrapolation to patient treatment, unconjugated MOAB treatment is not effective in eliminating tumors. Results of in vitro studies with SHP-77 are similar.

University of Pittsburgh, Department of Infectious Diseases and Microbiology, Graduate School of Public Health, Pittsburgh, Pennsylvania
*Division of Nuclear Medicine, Department of Radiology, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania

High human IgG levels in the SCID mouse reconstituted with human splenic tissues from patients with gastric cancer
Tetsuro Kubota, Hiroshi Yamaguchi, Masahiko Watana-be, Tetsuya Takahara, Tooru Takeuchi, Toshiharu Furukawa, Suguru Kase, Hirokazu Tanino, Susumu Kodaira, Kyuya Ishibiki, and Masaki Kitajima

We have implanted normal peripheral blood lymphocyte (PBL) from healthy donors and splenic tissues from patients with gastric cancers into the SCID mouse, demonstrating that the SCID mouse with splenic tissue can produce a high level human immunoglobulin G (IgG). The normal PBLs at numbers of 107 and 10⁸ per mouse were implanted intraperitoneally, and three splenic tissues with a diameter of 3 x 3 x 3 mm from patients with gastric cancers were inoculated subcutane-ously into the bilateral backs of the mouse. On 2, 4, 6, and 8 weeks after the inoculation the mice were sacri-ficed, and the human IgG was assessed by ELISA method. The SCID mice containing the splenic tissue revealed a high level human IgG from 2 weeks after the inoculation and approximately 2 mg IgG per mi was observed 8 weeks after the implantation, while the levels of IgG in mice treated with PBLs were limited. Since the half time of the extrinsic human IgG was 10.2 daysl the high level human IgG in the SCID mice was sup-posed to be produced by human plasma cells in the splenic tissue from the patients with gastric cancer. This model was thought to be adequate in evaluating the human immunological functions in vivo.

Department of Surgery, School of Medicine, Keio University, Tokyo, Japan

Antitumor activity of human lymphoma-reactive T cells against disseminated Burkitt's (Daudi) lymphoma in SCID mice
V. Malkovska, F. Cigel, N. Armstrong, K. J. Schell, W. Borcherding, and R. Hong

Human Burkitt's lymphoma (Daudi) ceils grow as progressive disseminated tumors in SCID mice after either intravenous or intraperitoneal injection (1). These cells are lysed efficiently in vitro by human Vd9/Vd2 T cells which recognize the groEL homolog on the Daudi cell surface (2). The in vitro culture of normal peripheral blood mononuclear cells (PBMC) with irradiated Daudi cells results in selective outgrowth of the Vd9/Vd2 T-cell subset. We studied the in vivo antitumor activity of these highly cytotoxic cell populations against Daudi lymphoma in a SCID mouse model. Normal human PBMC were incubated for 2 weeks with irradiated Daudi cells in complete RPMI 1640 with 10 percent fetal calf serum. The resulting cell populations (PBMC-Daudi) contained 42 percent Vd9/Vd2 T cells and showed 21 percent specific lysis of ⁵¹Cr labelled Daudi cells at 1.5:1 E:T ratio in a 4-hour ⁵¹ release assay. Groups of SCID mice were injected with 1) 10⁵ Daudi cells alone, either intraperitoneally or intravenously; 2) 10⁵ Daudi cells and 10⁷ PBMC-Daudi, intraperitoneally or intravenously through separate routes; and 3) 10⁵ Daudi cells and 10⁷ unstimulated PBMC, intraperitoneally or intravenously through separate routes. All animals injected with Daudi cells alone or Daudi ceils and unstimulated PBMC developed disseminated tumors (in the kidneys, ovaries, adrenals, spinal cord, bone marrow, and other organs) and their survival was significantly shorter than that of mice injected with Daudi cells and PBMC-Daudi (p<0.0001 for both intraperitoneal and intravenous routes). Similar results were obtained when PBMC-Daudi was administered intraperitoneally 4 days before or 4 days after Daudi cells. Some of the PBMC-Daudi treated mice are still alive at 6 months with no signs of tumors. The above data suggest that the SCID mouse model can be used for studying in vivo antitumor effects of human tumor-reactive lymphocytes.

1. Ghetie, M. A., J. Richardson, T. Tucker, D. Jones, J. W. Uhr, and E. S. Vitetta. 1990. Disseminated on localized growth of human B-cell tumor (Daudi) in SCID mice. Int. J. Cancer 45:481-485.
2. Fisch, P., M. Malkovsky, S. Kovats, E. Sturm, E. Braakneu, B. S. Klein, S. D. Voss, L. W. Morrissey, IL De Mans, W. S. Welch, IL L. H. Bolhuis, and P. M. Soudel. 1990. Recognition by human V69/Vd2 T cells of a gro-EL homolog on Daudi Burkitt's lymphoma cells. Science 250:1269-1273.

University of Wisconsin Hospital, Madison, Wisconsin

SCID thymus: A model of T-cell development in an early stage
Takashi Matsuyama, Gao Xiahua, Takashi Nishimura, and Sonoko Habu

Lymphoid stem cells migrate into the thymus where they develop to be mature T cells in the ordered manner. Since immature thymocytes, CD4^~ CD8^~ (DN) cells do not express T-cell antigen receptor (TCR), it is assumed that TCR may not be responsible for developing pathway from DN to CD4⁺CD⁸ (DP) cells. To examine this issue, we used SCID thymocytes in which TCR are not expressed on their cell surface because of abnormal or defective rearrangement of TCR genes. Expression of cell surface molecules on the SCID thymocytes was determined by staining with monoclonal antibodies (mAb). The majority (~ 90 percent) of SCID thymocytes were Thy-1⁺, Jlld^~, IL-2RA⁺ B^~, CD2^~, CD3^~, CD4^~, CD8^~, and TCR^~. IL-2RB and CD2 were co-expressed on a minor population of the thymocytes. This phenotype is similar to embryonic thymocytes on day 13 of gestation which are considered to be the most immature thymocytes. Among the SCID thymocytes, IL-2RB⁺ cell population showed a good responsiveness to IL-2 in vitro but IL-2RB-IL-2RA⁺ did not, while both cells produced IL-2 by culture with PMA and A 23187. These characteristics are not different from those of DN embryonic 2PBLthymocytes in normal mice. However, IL-2RB-A⁺ embryonic thymocytes were found to be different from SCID thymocytes with the same phenotype in the follow-ing ways. 1) During co-culture with our established thymic stromal cell clone, TNC-R3.1, embryonic IL2RA⁺ cells lost IL-2RA expression; and 2) In the same culture, the culIL-2RB^~ were able to differentiate into DP cells. Such changes were not observed in the SCID thymocytes. The importance of TCR for differentiation and proliferation of immature thymocytes will be discussed.

Department of Immunology, Tokai University School of Medicine, Kanagawa, Japan

Treatment with rat stem cell factor affects thymocyte progenitor cells
Eugene Medlock, Russell Migita, Lisa Trebasky, Dale Greiner*, Wayne Hendren, Randy DePrince, Chrysa Mineo-Daley, and Kris Zsebo

Recombinant rat stem cell factor (rrSCF¹⁶⁴), a form of the c-kit ligand, has numerous developmental effects on a multitude of stem/progenitor cells particularly important to the erythroid and myeloid lineages. However, the relevance of the kit-ligand to lymphoid cell development is less clear since c-kit-ligand deficient, SI/SI^d mice and c-kit defective, W/W^v mice have apparently normal peripheral B and T lymphocytes. We have recently shown that the cellularity of 12-week-old SI/SI^d thymi is reduced by approximately 50 percent when compared with age- and weight-matched normal F₁ +/+ litter-mates and certain IL-7 responsive B-cell precursors have enhanced growth in the presence of rrSCF and IL-7. In the present study, we have analyzed thymocyte progenitor cells (cells able to repopulate the thymus of adoptive recipients following intravenous injection) to determine if rrSCF affects their development in vivo. A quantitative in vivo assay for thymocyte progenitors has been used to assess relative numbers of thymocyte progenitor cells in rrSCF^164PEG-treated and saline-treated rat bone marrow and spleen. Donor/host determinations were made using the RT-7.1 and 7.2 allotypic pan lymphocyte antigens to distinguish M520 rat (donor) cell regeneration in irradiated BUF rats (host). Time course and cell dose studies revealed no significant differences in thymocyte regenera-tion resulting from transplantation of bone marrow cells from either saline-treated rats or rats treated for 7 days with rrSCF^164PEG. In contrast, splenocytes from rrSCF^164PEG-treated rats were significantly (p<0.05) more effective in regenerating the thymus than splenocytes from saline-treated rats. These data suggest that rrSCF¹⁶⁴ treatment in vivo induces an increase in either the number or activity of spleen cells able to migrate to and colonize the thymus of an irradiated recipient. Future studies will be directed towards comparison of these induced splenic thymocyte progenitors with normal bone marrow or spleen thymocyte progenitors.

Amgen, Inc., Thousand Oaks, California
*University of Massachusetts at Worcester

Comparison of resistance to Pseudomonas aeruginosa infection among nude, SCID, and drug-induced immunodeficient mice using biological response modifier S(BRM)
Masaharu Mita, Takaya Yamada*, Hiroyuki Inagaki*, Yumiko Suzuki*, Toshifumi Hibi**, Sonoko Habu**, Matsuo Muraoka***:, Sashiko Oh-ishi****, Tatsuo Suzuki* and Eiji Kawamura******

Infections are continuously changing, and these changes are especially marked in cases of bacterial infections. Infections by highly contagious bacteria are decreasing, while those by bacteria considered as weakly toxic or nonpathogenic are increasing. These are known as opportunistic infections. We investigated the relative importance of the various component factors of the vital defense mechanism in the elimination of foreign matter and studied the mutual relation between the host and microorganisms on the basis of an analysis of comparative importance. At present, we are performing experiments on opportunistic infections in compromised hosts and the role of phagocytes against the causative bacteria in such cases using an experimental Pseudomonas infection system. We have also performed a comparative investigation on activators which show non-specific, rapid, and strong effects on the vital defense mechanism in the early stage of infections in compromised hosts with SCID, nude, and drug-induced immunodefieient mice. SCID mice and BALB/c mice pretreated with Cyclophosphamide (60mg/kg, intravenously for 3 days) all failed to survive the challenge of Pseudomonas aeruginosa M-24 (2 x 10⁷ CFU intraperitoneal). The survival rate of BALB/c mice pretreated with Cicrosporin A (30mg/kg, subcutaneously for 7 days) and nude mice was 40 percent and 100 percent. Phagocytic activity of abdominalis cavity cells was high in nude, BALB/c mice, SCID mice.

Laboratory Animal Research Center, School of Pharmaceutical Sciences, School of Medicine, Kitasato University, Japan
* Biomedical Laboratory, Kitasato Institute Hospital, Japan
** Department of Internal Medicine, School of Medicine, Keio University, Tokyo, Japan
*** Department of Cell Biology, School of Medicine, Tokai University, Isehara, Japan
**** Department of Internal Medicine, Kitasato Institute Medical Center Hospital, Kitamoto, Japan
***** Department of Pharmacology, School of Pharmaceutical Sciences, School of Medicine, Kitasato University, Japan
****** Department of Surgery, Kitasato Institute Hospital, Japan

Retrovirus-induced immunodeficiency in the mouse
H. C. Morse III, S. K. Chattopadhyay, Y. Tang, A. W. Hagin, A. Cerny, M. Makino, R. Gazzinelli, T. N. Frederickson, and J. W. Hartley

Susceptible strains of mice infected with the mixture of murine leukemia viruses (MuLV), termed LP-BM5, develop progressive lymphoproliferation and profound immunodeficiency, a syndrome designated MAIDS. The disease-inducing agent, a replication defective virus, encodes a Pr60^gag protein that behaves like a "supantigen," stimulating marked T-cell activation. Targets for infection and expression of the defective virus include macrophages, B cells, and CD4⁺ T cells with expression in B cells being crucial for T-cell activation. The consequences of T-cell activation include production of cytokines with the spectrum of genes expressed varying with time after infection. At one week, IFN-g, IL-2, -4, -5, and -10 are produced without restimulation by mitogen or antigen. Thereafter only low levels of spontaneous production are detected, but stimulation with mitogen results in high level production of IL-4, -5, and -10 with impaired production of IL-2 and IFN-g. Levels of IL-10 mRNA expression also increase at these later times, as do levels of serum IgE. The latter findings suggest that Th2-type helper cells are preferentially activated in vivo in the later stages of disease. As IL-10 impairs cytokine expression by Thl cells and CD8⁺ T cells and IL-4 and -5 stimulate B cells, activation of Th2 cytokines may explain impaired T-cell responses to infectious agents (T. gondii, cytomegalovirus) or conventional antigens as well as chronic B-cell activation. The central role ascribed to cytokine dysregulation in the pathogenesis of MAIDS is supported by the finding that treatment of mice with cyclosporin A retards the development of B- and T-cell dysfunction and lymphoproliferation. Analyses of disease resistance exhibited by some mouse strains was found to be mediated by genes both within and outside the MHC. MHC influences are dominated by class I genes, their effect being to direct potent CD8⁺ T-cell responses capable of ridding the host of the defective virus genome.

National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland

HIV infection and immunity in the hu-PBL-SCID mouse model
D. E. Mosier

Human peripheral blood leukocytes have been injected intraperitoneally into C.B17-scid mice to create hu-PBL-SCID animals. The human cells surviving in these animals have been characterized as predominantly T cells, with CD4:CD8 ratios ranging from 4:1 to 1:1, and most of the T cells expressing the CD45RO activa-tion/memory cell marker. The hu-PBL-SCID mouse is highly susceptible to HIV-1 infection with T-cell- and MO-tropic laboratory isolates (IIIB, SF2, MN, SF13, SF33, SF167, ADA-5, JR-CSF) and fresh clinical isolates. The minimal infectious dose of the virus is 1-10 infectious particles. The hu-PBL-SCID mouse allows the stimulation of secondary antibody responses, and has been used to evaluate protective immunity to HIV-1. Two normal donors vaccinated originally with gpl60-vaccinia and subsequently boosted repeatedly with recombinant gpl60 have been evaluated repeatedly. Hu-PBL-SCID animals created from vaccinated donors were challenged with 10³ TCID₅₀ of homologous strain IIIB HIV-1, a virus inoculum that has infected 92 ± 8 percent of 357 control hu-PBL-SCID mice generated from naive donors at 1, 2, 6, and 12 months after boosting with 160/µg rgp 160. For donor 1, 60 percent (6/10) resisted infection at 1 month post-boost, 50 percent at 6 months post-boost, and all mice became infected at 12 months post-boost. For donor 2, who was producing neutralizing antibody, 42 percent (5/12) of mice resisted infection at 1 to 2 months post-boost, and all animals were infected at 6 and 12 months post-boost. These results suggest that adoptive transfer of protective immunity from vaccinees to SCID mice is possible, but that cells capable of such transfer persist for relatively short periods after boosting. Whether such immune cells disappear altogether, or simply cannot be recovered from peripheral blood, is unknown.

Medical Biology Institute, La Jolla, California
In collaboration with B. E. Torbett, R. J. Gulizia, P. MacIsaac, B. J. Mathieson, O. Smith, S. -L. Hu, L. Corey and P. Greenberg
Supported by NIH grants AI29182, A127703, and AI30238

Differentiation of M-CSF-independent macrophages and glucan-induced granuloma formation in osteopetrotic (op/op) mice
Makoto Naito, Shin-Ichi Hayashi, Shin-Ichi Nishikawa, Leonard D. Shultz*, and Kiyoshi Takahashi

Osteopetrotic (op/op) mice are characterized by a generalized skeletal sclerosis due to failure of bone resorption and remodeling resulting from a defect of osteoclasts. The op mutation is within the coding region of the macrophage-colony-stimulating factor (Csfm) gene. In the mutant, peripheral blood monocytes, macrophages in various visceral organs, and osteoclasts in the bone are markedly reduced. However, we found ultrastructurally immature tissue macrophages in many organs and tissues. In co-cultures of normal mouse bone marrow cells with fibroblast cell lines prepared from the lung of the op/op mice, we confirmed a defective differentiation of monocytes into macrophages. However, a small number of macrophages developed most probably under the influence of granulocyte/macrophage-colony-stimulating factor produced by the op/op fibroblast cell lines. This macrophage population seems to correspond to the M-CSF-independent immature macrophages in the tissues of the mutant mice.

The process of glucan-induced hepatic granuloma formation was compared between monocytopenic op/op mice and normal littermates. In contrast to marked monocytosis in the littermates after glucan injection, blood monocytes were deficient in the mutant and the differentiation process of monocytes into macrophages was severely defective. However, granulomas were formed 5 days after glucan injection. They were com-posed of macrophages intermingled with a few lymphocytes. Although the size and the number of granulomas were smaller in op/op mice than those in littermates, proliferative capacity of hepatic macrophages was as high as that of littermates 5 days after glucan administration. These results suggest that the M-CSF-independent macrophage population is derived from granulocyte/macrophage-colony forming cells or earlier hematopoietic progenitors but not from monocytes, and that this macrophage population participates in glucan-induced granuloma formation without contribution from blood monocytes.

Second Department of Pathology, Kumamoto University Medical School, Kumamoto, Japan; and Department of Pathology, Institute of Medical Immunology, Kumamoto University Medical School, Kumamoto, Japan
*The Jackson Laboratory, Bar Harbor, Maine

Epidermal growth factor receptor gene alteration in human neural tumor xenografts in nude mice
Yasuyuki Ohnishi, Atsushi Tsugu*, Masato Nakamura*, Yoko Yamaguchi*, Hitoshi Yamazaki*, Yoshito Ueyama*, Norikazu Tamaoki*, and Tatsuji Nomura

Tumors of the nervous system showed various histological and biological features. While the proper meaning of the features is still unclear, it is considered that various mechanisms, including activation of oncogenes, may affect carcinogenesis and progression of tumors. Primary neural tumor specimens limit fine analysis because of their small size. The human tumor/nude mouse system provides useful tools for studying human neoplasm. We have reported that some glioma xenografts showed structural and functional alterations of the epidermal growth factor receptor (EGFR) gene. We studied 20 lines of human neural tumor xenografts including glioma, ependymoma and schwannoma. We examined the structure of the EGFR gene by Southern blot analysis. We estimated the expression of the EGFR gene by Northern blot analysis and reverse transcription-polymer-ase chain reaction (RT-PCR) methods. Three of 15 glioblastoma cell lines showed amplification and rearrangement of the EGFR gene. RT-PCR analysis demon-strated aberrant transcripts of the EGFR gene with deletion (800 bp) at the extracellular domain coding region in the three cell lines. One of three malignant schwannomas also revealed EGFR gene amplification, while the cell line showed no rearrangement. Four tumors overexpressed mRNA of the EGFR gene. The other six tumors (two ependymomas and four glioblastomas) expressed the EGFR gene moderately without gene amplification. Our results suggest that various EGFR gene alterations were associated with neoplasms of the peripheral nervous system as well as the central nervous system.

Central Institute for Experimental Animals, Kanagawa, Japan
*Department of Pathology, Tokai University School of Medicine, Kanagawa, Japan

Pleiotropic effects of the scid mutation
R. A. Phillips

The immune-deficiency in SCID mice results from their inability to correctly rearrange their immunoglobulin in T-cell receptor genes. However, the basic defect in SCID mice is a DNA repair gene which affects the ability of the mice to repair double-strand DNA breaks. Although this defect affects all tissues equally, lymphoid tissues are severely affected since their normal function depends on the use of this repair gene. The effect of the scid mutation on many tissues make this mouse a unique recipient for studies requiring transfer of hematopoietic cells. Thus, SCID mice are ideal recipients for analyzing lymphoid progenitors. The increased radiation sensitivity also makes sublethally irradiated SCID mice excellent recipients for myeloid progenitors despite the fact that this lineage exhibits no detectable defect in these mice. In addition, sublethally irradiated SCID mice offer many advantages for studying autoimmune diseases, such as graft-vs-host disease (GvHD). GvHD is particularly interesting since the disease in these recipients is markedly different from that observed in other mice. The differences in pathology may be due to the absence of host lymphocytes which are potential targets for GvHD. Host lymphocytes may also be required to initiate some components of the GvH reaction. In any case, the extremely low numbers of lymphocytes in SCID mice makes it feasible to investigate cellular interactions and the role of different lymphocyte subpopulations in the pathogenesis of many autoimmune diseases.

Division of Immunology and Cancer Research, Hospital for Sick Children, Toronto, Canada

Epstein-Barr virus-induced lymphoproliferative disorders in SCID-hu chimeric mice
P. Pisa, J. L. Garnier*, M. Cirone*, R. Rochford*, R. I. Fox*, M. J. Cannon*, and N. R. Cooper*

EBV B95-8 infection of SCID mice reconstituted with PBL from human donors is followed 20-40 days later by development of invasive and rapidly lethal peritoneal tumors of human B-cell origin. Histopathologic studies indicated that tumors were composed of plasmacytoid immunoblasts, with well-defined areas of necrosis, and Southern blots showed that tumors carried EBV DNA. Analysis of Ig gene rearrangements and EBV terminal repeat (TR) sequences demonstrated that SCID/hu tumors were oligoclonal. Southern blot analysis also showed that the EBV genome in tumors resultant from B95-8 infection of SCID/hu mice was episomal, whereas spontaneous tumors arising in SCID mice reconstituted with PBL from EBV-seropositive donors possessed both episomal and linear, replicating EBV genomes.

Cellular and viral gene expression in SCID/hu tumors broadly paralleled patterns seen in LCL rather than Burkitt lymphoma cells, except that CD20 and CD23 expression was down-regulated in tumor biopsy cells, and, most notably, that EBV latent gene expression may also be down-regulated. EBV-specific T-cell recognition was found to correlate with levels of EBV EBNA-2 and LMP expression by tumor cells and autologous LCL.

The data suggest that down-regulation of EBV latent gene expression in tumor cells, and consequent reduction in vulnerability of EBV-specific T-cell recognition, may further contribute to pathogenesis in immunosuppressed patients with impaired T-cell function.

Department of Immunology, Karolinska Institute, Stockholm, Sweden
*Research Institute of .Scripps Clinic, La Jolla, California

SCID mice as models for human parasitic diseases
T.V. Rajan, F.K. Nelson, L.D. Shultz*, and Dale Greiner**

Parasitic diseases represent an enormous economic burden in many tropical and subtropical countries. Research into basic pathophysiology of these diseases has been hampered by the fact that many human parasites are not transmissible to mice. We are developing SCID mice as models for two parasitic diseases of global significance.

Human lymphatic filariasis is a major public health problem in a number of developing countries, where there are an estimated 100 million cases, caused by the organisms Wucheria bancrofti and Brugia malayi (3). We have shown that subcutaneous injection of infective L3 larvae into SCID mice recapitulates the human disease, in that: (a) the parasites home to lymphatics where they cause lymphatic dilation and retention of lymph; and where their presence causes inflammatory changes including lymphangitis and lymphadenitis; and (b) immunological reconstitution of SCID mice prior to injection of infective larvae prevents the establishment of infection, whereas a normal, functional immune system is unable to cure established infection. An observation that is puzzling is that "leaky" SCID mice appear to be resistant to infection. There is evidence in the literature that these mice generate a very limited repertoire of T-and B-cell clones, and it would appear unlikely that any one of the oligoclones would be filiarial specifc.

The nematode parasite Onchocerca volvulus is responsible for the second largest incidence of preventable blindness in the world. O. volvulus is an obligate human parasite and its study has been difficult due to an inability to maintain it outside the human host. Over the past year, we have been successful in transplanting Onchocercomata containing live adult O. volvulus worms into immunodeficient C.B17 scid/scid (scid) or athymic mu/mu (nude) rats. Live, motile worms containing viable mf are present in Onchocercomata recovered from SCID mice or nude rats for up to 20 weeks, establishing a novel animal model for future investigation of the biology of O. volvulus.

Department of Pathology, University of Connecticut Health Center, Farmington, Connecticut
*The Jackson Laboratory, Bar Harbor, Maine
** The Department of Medicine, University of Massachusetts Medical Center, Worcester, Massachusetts

Selective reconstitution of T-cell subsets in SCID mice
Jorg Reimann

Adoptive transfer of purified CD4⁺CD8^- T cells from thymus, spleen, or lymph node of congenic (BALB/c, C.B17) or scmi-allogeneic (dm2), but not completely allogeneic (C57BL/6) adult donor mice selectively repopulated the respective T-cell compartments of spleen and peritoneal cavity in young SCID recipient mice. Engraftment of T cells 3 to 5 months post-transfer was assessed by different reconstitution parameters: cytofluorographic analyses, histological examinations, in vitro establishment of donor-derived CD3⁺CD4⁺CD8^- T-cell lines from transplanted SCID mice, and repeated serial transfers of donor-type T-cell populations through young SCID recipients. Transplanted CD4⁺ T cells repopulated spleen and peritoneal cavity, but not lymph nodes or thymus of SCID recipients. Transfer of limiting numbers of purified CD4⁺CD8^- T cells (10³ to 10⁵ cells per mouse) led to partial and selective reconstitution of T-dependent splenic areas of SCID recipients. Only a small fraction of injected CD4⁺ T cells is apparently selected for engraftment by the SCID mouse environment. CD4⁺ T cells obtained from the spleen of transplanted SCID mice displayed in vitro proliferative, A^d-restricted self reactivity, but no evidence for autoimmune lesions was found in vivo in CD4⁺ T cell-transplanted SCID mice. Extensive in vivo expansion of donor-type CD4⁺ T cells was observed in the course of repeated passages of CD4⁺ T cells through young SCID recipients for more than one year. Attempts to antigen-specifically activate transplanted CD4⁺ T cells in SCID hosts were unsuccessful.

Institute of Microbiology, University of UIm, Ulm, Germany

Genetics and functional studies of minor histocompatibility loci
Derry C. Roopenian, Larry E. Mobraaten, Sheau-Chainn Wang, Greg Christianson, and David Higgins

There are many control points at which a given stimuli can have profound effects on both qualitative and quantitative aspects of an ensuing immune response. To better understand these control points, we used the minor histocompatibility (H) antigen system because it is a model system pertinent not only for transplantation immunity but also auto and tumor immunity. We found previously that two minor H loci H-3 and H-4 are genetically complex and can be subdivided into two gene types (1,2). One type encodes products that stimulate MHC class I-restricted cytotoxic T cells (Tc) and the other encodes products that stimulate MHC class II-restricted helper T cells (Th). Thus, the nature of the minor H locus gene product is a control point that dictates which T-cell subset will be stimulated. We produced new mouse strains that carry a crossover between the above two gene types in order to assess the characteristics of immunity that each type of gene product would elicit. Tc were not detected when the mice were primed with spleen cells expressing the Tc-defined antigens alone; Th stimulation was required. Thus, stimulation of the Th subset during immunization was a control point that profoundly influenced the Tc immune response. Moreover, skin allograft rejection across Tc-defined antigen barriers proceeded quite inefficiently, and in many cases not at all. The type of immunizing cell also heavily influenced the Tc response since the requirement for Th was less pronounced when the primary immune stimulus was semipurified dendritic cells or skin allografts. These results indicated the nature of the antigen presenting cell also is a control point for the Tc response. Skin allografts solely across a Th-defined antigen barrier were retained indefinitely. Thus, Th in the absence of Tc did not lead to effective skin graft rejection. Allografts bearing both Th and Tc-defined antigens were rejected rapidly, as were Tc-defined antigen-disparate grafts after priming across Th + Tc antigen barriers. Thus, Th are not necessary proximal effectors of rejection, but contribute to the rejection process by facilitating the differentiation of precursors into mature Tc. We discuss these results in the context of control points involved in transplantation, auto- and tumor-immunity.

1. Roopenian, D. C., and A. P. Davis. 1990. Responses against antigens encoded by the H-3 histocompatibility locus: Antigens stimulation class I MItC- and class II MHC-restricted T cells are encoded by separate genes. Immunogenetics 30:335-343.
2. Davis, A. P., and D. C. Roopenian. 1990. Complexity at the mouse minor histocompatibility locus H-4. Immunogenetics 31:7-12.

The Jackson Laboratory, Bar Harbor, Maine

Production of pathogen-free immune-deficient animals
Robert J. Russell

Today, animals are produced in barrier buildings and isolator housing systems to prevent infection with adventitious viruses, bacteria, Mycoplasma sp., fungi, and parasites. Pathogenic agents of special concern in immune-deficient mice include Sendai virus, mouse hepatitis virus, Staphylococcus aureus, Pseudomonas aeruginosa, and Pneumocystis sp. Contaminated animal stocks and strains are rederived by the use of hysterectomy and embryo transfer techniques. Rederived animals are then maintained in either barrier buildings or isolators. Because of potential contact with the animal care staff, barrier building produced animals may harbor commensal and other nonpathogenic bacteria that can interfere with the research utilization of immune-deficient animals, e.g., athymic nude, scid, and bg-nu-xid. There-fore, immune-deficient animals are produced and maintained in flexible film isolators free from direct human contact until they are delivered to investigators. Isolators are designed and operated to provide total animal protection and maintain animals free of unwanted organisms. Isolator techniques and procedures include complete sterilization of the flexible film isolator system prior to animal pop