A role for elevated H-2 antigen expression in resistance to neoplasia caused by radiation-induced leukemia virus. Enhancement of effective tumor surveillance by killer lymphocytes.

Resistance to neoplasia caused by radiation-induced leukemia virus (RadLV) is mediated by gene(s) in the H-2D region of the major histocompatibility complex. The previous observation that rapid increases in cellular synthesis and cell-surface expression of H-2 antigens are detectable immediately after virus inoculation has suggested that altered expression of H-2 antigens may play a significant role in the mechanism(s) of host defense to virus infection. This concept is supported by the following observations. First, cell-mediated immunity against RadLV transformed or infected cells can be detected with ease when H-2-positive target cells are used in the cell-mediated lympholysis (CML) assay. (Although RadLV transformed cells obtained from overtly leukemic animals and maintained in tissue culture are H-2 negative, these cells can regain their H-2 phenotype by in vivo passage in normal animals. The H-2-negative cells are poor targets in a CML assay.) Second, resistant mice develop greater numbers of effectors when infected with RadLV than do susceptible mice. Third, injection of normal (uninfected) thymocytes into syngeneic recipients of resistant or susceptible H-2 type does not stimulate a CML response. However, injection of RadLV infected thymocytes from resistant mice produces a vigorous CMI response, and such thymocytes elicit the strongest response at a time when both H-2 and viral antigen expression is elevated. By contrast, injection of infected thymocytes from susceptible mice, which express viral antigens, but low levels of H-2 antigens, does not stimulate a CML reaction. These findings may explain the easier induction of leukemia found by many investigators when virus is inoculated into neonatal mice and the preferential thymus tropism of some oncogenic type-C RNA virus. Cells expressing very low levels of H-2, such as thymocytes, may serve as permissive targets for virus infection because they lack an important component (H-2 antigens) of the dual or altered recognition signal required to trigger a defensive host immune response.

Identification of a 36,000-molecular weight, gag-related phosphoprotein in lymphoma cells transformed by radiation leukemia virus.

Radiation leukemia virus (RadLV) causes thymic lymphoma in 90% of susceptible mice after a latent period of several months. The virally encoded polypeptides produced by RadLV-induced lymphoma cells were analyzed by immunoprecipitation and NaDodSO4/polyacrylamide gel electrophoresis. Along with the expected precursor and mature forms of gag and env gene products, a polypeptide of 36,000 molecular weight (p36) was precipitated by anti-gag antisera. It was not precipitable by normal sera or anti-env antibodies. Like the gag-associated fusion proteins of some acute leukemia viruses, p36 was found to be phosphorylated in vivo, although it lacked detectable ATP-specific protein kinase activity in vitro. By kinetics during pulse-chase labeling experiments and by comparison of two-dimensional tryptic peptide maps, this protein is not an intermediate in gag precursor processing. One lymphoma cell line is described that resembles a nonproducer RadLV-transformant, synthesizing relatively large amounts of p36 in the absence of Pr66gag or p30 production. Several RadLV-induced lymphoma cell lines also produce p36, while it was not detectable in the radiation-induced lines tested. In addition, p36 was not produced by mouse or mink fibroblasts or cultured thymocyte cell lines infected with virus passaged from the RadLV-induced lymphomas. We conclude that p36 may represent a previously unrecognized transformation-related protein induced directly or indirectly by infection with RadLV.

Expression of a single major histocompatibility complex locus controls the immune response to poly-L-(tyrosine, glutamic acid)-poly-DL-alanine-poly-L-lysine.

Genetic control of the immune response linked to the major histocompatibility (H-2) complex in the mouse has been described for synthetic polypeptide antigens and for low doses of native proteins. The phenomenon is well documented(1,2). Extensive screening of intra-H-2 crossover-derived recombinant strains has localized H-2-linked immune response (Ir) genes to the I-immune response region of the H-2 complex (3). For most antigens, Ir genes are autosomal, dominant, and they segregate as single loci. It is not known whether these crossover-defined loci respresent single genes with multiple alleles or clusters of tightly linked genes (4). In 1972, Stimpfling and Durham (5) postulated that two interacting loci within the H-2 complex were required for the response to the alloantigen, H-2.2 (6), and, in 1975, Dorf et. al. (7) observed a responder phenotype in a recombinant derived from two strains which were nonresponders to the synthetic linear terpolymer, L-glutamic acid, L-lysine, L-phenylaline (GLPhe). Analysis of additional recombinants and complementation tests with F(1) hybrids clearly demonstrated that genes in two intra-I-region loci controlled the immune response to GLPhe. Subsequently, requirement for genes mapping in two intra-I-region loci were reported for porcine LDH(B)(8), the alloantigen Thy-1.1 (9), and for the synthetic terpolymers L-glutamic acid, L-lysine, L-tyrosine and L-glutamic acid, L-lysine, L- leucine (6,10). Demonstration that responses to both synthetic polypeptide and native protein antigens can be controlled by genes in two distinct I-region loci prompted speculation that the phenotypic expression of two I-region genes is a general phenomenon which may provide the key for understanding the mechanism of Ir gene function and cellular collaboration in the immune response. Benacerraf and Dorf (10) have shown that Ir gene complementation is often more effective in the cis than in the trans configuration. This concept is further supported by the data reported for GLPhe (10-12) which indicate that both of the complementing genes must be expressed in each of the cell types participating in the interaction. Failure to detect complementation for the majority of antigens under H-2-linked Ir-gene control might be attributed to the limited number of available intra-I- region recombinant strains.

Genetics of susceptibility for radiation-induced leukemia. Mapping of genes involved to chromosomes 1, 2, and 4, and implications for a viral etiology in the disease.

Susceptibility to radiation-induced leukemia in (A/J x B10)F2 mice is encoded for by genes in chromosomes 1, 2, and 4. The loci involved in chromosomes 1 and 4 are close to or similar to xenotropic virus inducibility locus on chromosome 1 and a locus-affecting expression of xenotropic MuLV envelope-related cell surface antigens. Radiation-induced leukemia-1 (Ril-1) on chromosome 2 plays an overriding influence in susceptibility to the disease. This locus might encode ecotropic viral-associated genetic information or might contain cellular sequences with oncogenic potential. These findings are of interest in view of the importance of recombinant viruses to leukemogenesis. Furthermore, it is intriguing that Ril-1 is located in a chromosomal site rich in thymus differentiation-specific loci. An explanation for tissue-specific activation of endogenous viruses is that activation of the virus in question is dependent on differentiation-specific steps.

Genetic control of cell-mediated responsiveness to an AKR tumor-associated antigen: mapping of the locus involved to the I region of the H-2 complex.

The role of H-2-linked genes in controlling resistance to murine leukemia viruses has been studied by measuring the cell-mediated immune response of F1 hybrid mice (between AKR and various C3H and C57BL/10 derived, H-2 congenic strains) to an AKR tumor cell line, BW5147. The studies have shown that the ability to generate a primary or secondary cell-mediated response to an AKR tumor cell antigenic determinant is under H-2 linked control. The locus determining CML responsiveness maps in the I-J subregion. Nonresponsiveness is associated with the H-2q/k and H-2b/k hybrid genotypes, whereas responsiveness is associated with the H-2k/k homozygous genotype. Nonresponsiveness may result from (a) dominant suppression; (b) recessive responsiveness; or (c) an alternate mechanism not yet understood. This type of control may be one of several H-2-associated mechanisms of defense against virus-induced neoplasms.

Genetic control of radiation leukemia virus-induced tumorigenesis. I. Role of the major murine histocompatibility complex, H-2.

Resistance to radiation leukemia virus-induced leukemogenesis is associated with the H-2D region of the H-2 complex, or with closely linked loci. The H-2Dd allele confers resistance ot the disease, while the H-2D-Q and H-2Ds alleles are associated with susceptibility. It is not clear whether Ir genes, or an alternative mechanism are responsible for the observed H-2-linked resistance to the disease.

Genetic control of radiation leukemia virus-induced tumorigenesis II. Influence of Srlv-1, a locus not linked to H-2.

A single locus, tentatively denoted Srlv-1 (susceptibility to radiation leukemia virus [RadLV]-1), confers dominant susceptibility to RadLV-induced leukemogenesis. Srlv-1 is not linked to H-2, and appears to be distinct from Fv-1 and Fv-2. Preliminary data suggest that Srlv-1 affected virus proliferation. A striking feature of this system is that Srlv-1 overrides the protection afforded by the H2D-associated dominant resistance to RadLV-induced neoplasia.

Increased synthesis and expression of H-2 antigens on thymocytes as a result of radiation leukemia virus infection: a possible mechanism for H-2 linked control of virus-induced neoplasia.

Previous studies from this laboratory have mapped resistance and/or susceptibility to radiation-induced leukemia virus (RadLV)-induced neoplasia to the H-2D region. H-2 linked effects on virus replication can be detected subsequent to the initial virus infection, and clear-cut differences in numbers of virus infected thymus cells can be detected as early as 5 wk after RadLV inoculation. Rapid increases in cellular synthesis and cell surface expression of H-2 antigens are detectable immediately after virus inoculation. These changes have been studied by immunofluorescence, absorption, cell surface iodination followed by sodium dodecyl-sulfate-polyacrylamide gel electrophoresis, and two dimensional gel electrophoretic analysis of internally labeled lymphocyte proteins. Expression of H-2K molecules is significantly increased in cells of susceptible and resistant animals. However, significant increases in expression of H-2D antigens occurs only on thymus cells from resistant strains (H-2Dd). Transformed cells of resistant and susceptible H-2 haplotypes adapted to tissue culture lack detectable H-2 antigens as determined by serological absorption studies. It is argued that altered expression of H-2 antigens plays a very significant role in the mechanism of host defense to virus infection.

Controlled propagation of replication-competent Sindbis viral vector using suicide gene strategy.

A major concern of using viral gene therapy is the potential for uncontrolled vector propagation and infection that might result in serious deleterious effects. To enhance the safety, several viral vectors, including vectors based on Sindbis virus, were engineered to lose their capability to replicate and spread after transduction of target cells. Such designs, however, could dramatically reduce the therapeutic potency of the viral vectors, resulting in the need for multiple dosages to achieve treatment goals. Earlier, we showed that a replication-defective (RD) Sindbis vector achieved specific tumor targeting without any adverse effects in vivo. Here, we present a replication-competent Sindbis viral vector that has an hsvtk suicide gene incorporated into ns3, an indispensable non-structural gene for viral survival. The capability of viral propagation significantly increases tumor-specific infection and enhances growth suppression of tumor compared with the conventional RD vectors. Furthermore, in the presence of the prodrug ganciclovir, the hsvtk suicide gene serves as a safety mechanism to prevent uncontrolled vector propagation. In addition to suppressing vector propagation, toxic metabolites, generated by prodrug activation, could spread to neighboring uninfected tumor cells to further enhance tumor killing.

Murine thymomas induced by fractionated-X-irradiation have specific T-cell receptor rearrangements and characteristics associated with day-15 to -16 fetal thymocytes.

We report here that specific T-cell receptor rearrangements were observed in fractionated-X-irradiation-induced murine leukemias. Consistent gamma-chain rearrangements, limited beta-chain rearrangements, and no detectable alpha-chain rearrangements were observed. Gene expression studies revealed that, in comparison with normal thymus tissue, expression of alpha T-cell receptor genes was lower in the thymomas, beta expression was much higher but approximately equal to that of normal thymocytes, and gamma expression was significantly increased. After coupling these data with those from analyses using reagents against other surface markers, such as Lyt-2, L3T4, H-2, IL-2R and MEL-14, we concluded that the target T cells for fractionated-X-irradiation-induced transformation resemble fetal thymocytes from days 15 and 16 of gestation.

Cell-specific targeting of Sindbis virus vectors displaying IgG-binding domains of protein A.

Sindbis virus can infect a broad range of insect and vertebrate cell types due to the widespread distribution of the cellular receptor for the virus. The development of Sindbis virus vectors that target specific cell types could have important implications for the design of gene therapy strategies. To achieve this goal we have designed and constructed Sindbis virus particles displaying the IgG-binding domain of protein A. The protein A-envelope chimeric Sindbis virus vector has minimal infectivities against baby hamster kidney and human cell lines. When used in conjunction with monoclonal antibodies that react with cell-surface antigens, however, the protein A-envelope chimeric virus was able to infect human cell lines with high efficiency. Infection rates were 90% or higher for human lymphoblastoid cells. A variety of cells could be targeted by changing the monoclonal antibody without generating a new recombinant virus.

Phosphoproteins recognized by an H-2-linked immune response gene and their association with cell proliferation.

An H-2-associated immune response gene which maps to the I-A subregion of the H-2 complex governs the ability of H-2 congenic mice to mount an antibody response to five phosphoproteins with molecular weights of 33,000, 29,000, 23,000, 17,000, and 16,000 when inoculated with BW5147, a spontaneous AKR T-cell leukemic cell line. The phosphoproteins are present in all tumor cell lines tested, including those of murine and human origins. The phosphoproteins are associated with the proliferative state of the cell as studied in many systems including growth stimulation of normal lymphoid cells with mitogens, interleukin 2 dependency for growth of a cloned T-cell line, cessation of proliferation by serum starvation of Swiss 3T3 fibroblasts, retention of the proliferative capacity of SV40-transformed 3T3 fibroblasts, and the differentiation and inhibition of proliferation of human promyelocytic leukemic cells. Phosphoproteins with molecular weights of 33,000, 29,000, 23,000, 17,000, and 16,000 are therefore not specific to a particular inducible cellular pathway but are associated with cell proliferation in general.

Regulation of H-2 class I gene expression in virally transformed and infected cells.

Early studies of the resistance and susceptibility of mouse strains to radiation-induced leukemia virus have demonstrated the important role of altered histocompatibility (H-2) antigen expression in the effectiveness of the immune response of the host to virus-infected and transformed cells. Changes in H-2 gene expression have now been correlated with disease resistance in a variety of viral systems. The experiments discussed indicate that viruses may directly or indirectly affect H-2 antigen expression at various levels of gene expression. These investigations generate a framework for approaching a molecular understanding of viral-induced changes in H-2 gene expression.

Infection of human cells by murine ecotropic viruses: retroviral vectors carrying the hygromycin resistance-encoding gene.

The construction of a new retroviral vector, pSKV, is described. This vector carries two unique cloning sites, located between two Moloney leukemia virus-derived LTR, into which genes of interest may be introduced. The gene encoding hygromycin resistance (HyR) was subsequently introduced into one of the two sites, producing a second vector (pSKV/HyR) containing a unique SfiI site for the introduction of cDNA clones under the control of the cytomegalovirus (CMV) promoter (P-CMV). The cDNA (mH13), encoding a protein that has been shown to serve as a murine ecotropic retroviral receptor in transient assays, was cloned into the SfiI site (pSKV/HyR/mH13). Both constructs can be packaged into retroviral particles following transfection into an appropriate packaging cell line. Stable transfectants of the human glioblastoma cell line (U118MG) carrying each of these two constructs were generated by transfection and subsequent Hy selection. Clones expressing both the selectable marker and the mH13 gene, but not those expressing only the selectable marker, are shown to be susceptible to infection with murine ecotropic retroviral particles. These cells (HyR and mH13 positive) were then exposed to CRE/Xtk culture supernatant, a packaging cell line producing ecotropic retroviral particles carrying the HSV-TK (Herpes simplex virus-thymidine kinase) and neoR (neomycin-resistance) genes. Selection was in the presence of G418. In vitro growth of the U118MG/HyR/mH13/TK cells, but not that of the U118MG/HyR/mH13 cells, was inhibited by ganciclovir (GCV), indicating the successful transfer of HSV-TK by infection of human cells with murine retroviruses via the mH13 product.

Inactivation of the human immunodeficiency virus by hypericin: evidence for photochemical alterations of p24 and a block in uncoating.

Following attachment and entry of human immunodeficiency virus (HIV) into a host cell, the HIV genomic RNA is reverse transcribed to cDNA. This step may be inhibited by hypericin, a compound that induces alterations of the retroviral capsid. Incubation of HIV with hypericin rendered the virus noninfectious. The replication of HIV was blocked early; HIV cDNA could not be detected in cells challenged with hypericin-treated HIV. Hypericin did not inhibit the binding of recombinant gp120 to CD4+ cells, nor did hypericin inhibit syncytium formation. However, reverse transcriptase activity could not be released from hypericin-treated virions. Western blot analysis revealed altered mobility of the HIV major capsid protein (p24) following hypericin treatment. Hypericin-treated recombinant HIV p24 exhibited similar altered mobility. The inactivation of HIV infectivity and the alterations in p24 mobility required hypericin incubations in the presence of visible light. Collectively, these data suggest that photochemical alterations of the HIV capsid may contribute to the hypericin-mediated inactivation of HIV. Such alterations may inhibit the release of RT activity from treated HIV, and prevent uncoating and subsequent reverse transcription of the HIV genome within a target cell.

A molecular and genetic approach to understanding the mechanisms by which fractionated X-irradiation induces leukemia in mice.

Our laboratory's approach to try to shed light on the question of a viral etiology for radiation-induced leukemia has focused on defining, localizing and understanding the mode of action of genes involved in susceptibility to FXI-induced disease. These studies have indicated that multiple genes control the process of leukemogenesis. In addition not every mouse strain which shows some susceptibility to FXI-induced leukemia carries the susceptible gene at each of the multiple loci involved in the disease process. Thus, it is plausible to conclude that more than one mechanism of leukemogenesis can be triggered by FXI. Our studies have focused on the mode of action of one such locus Ril-1. Several reagents have been developed to help us clone and characterize this locus. Currently chromosomal "walking" and "hopping" techniques are being used in conjunction with an RFLP molecular probe which is adjacent to Ril-1. In addition a cDNA library has been prepared from a radiation-induced thymoma and subtraction hybridization analysis is being used in the search for Ril-1.

Multi-drug delivery system using streptavidin-transforming growth factor-alpha chimeric protein.

Tissue-specific delivery of a variety of molecules has been a valuable technique for biological and medical research. Therefore, we have constructed a recombinant plasmid containing the coding regions for streptavidin core and mature human transforming growth factor-alpha (TGF-alpha). The recombinant plasmid has been expressed in Escherichia coli to produce a chimeric protein with both streptavidin and TGF-alpha activity. The streptavidin-TGF-alpha chimeric protein (ST-TGF-alpha) could efficiently transfer biotinylated beta-galactosidase into A431 cells via the epidermal growth factor receptor. More than 99% of the cells contained the enzyme transferred. Furthermore, ST-TGF-alpha complexed with biotinylated-glucose oxidase had a significant cytotoxic effect when incubated with A431 cells. These findings suggest that the ST-TGF-alpha chimeric protein could be used to deliver proteins of interest into target cells without the need for chemical linkage or genetic construction. Essentially, ST-TGF-alpha serves as a high-modular "molecular bridge" for the passage of a wide variety of effector molecules into target cells.

Specific cell targeting for delivery of toxins into small-cell lung cancer using a streptavidin fusion protein complex.

New modalities of treatment for small-cell lung cancer (SCLC) are needed, because the majority of patients continue to die of disseminated disease despite an initial response to conventional chemotherapy. Abnormal surface expression of the neural-cell adhesion molecule (NCAM) has been noted to be highly associated with SCLC. We examined the ability and efficiency of a streptavidin-Protein A (ST-PA) fusion protein complexed with an anti-NCAM monoclonal antibody (Mab) to transfer biotinylated beta-galactosidase into human SCLC cell lines NCI-H69, NCI-H526, and NCI-H446. When the surface molecule NCAM was targeted with this system, more than 99% of the targeted cells internalized and exhibited beta-galactosidase activity. In addition, we evaluated cytotoxic activity against SCLC lines NCI-H69 and NCI-H526 by efficient delivery of biotinylated glucose oxidase using the same ST-PA/anti-NCAM Mab complex. Cytotoxicity of the transduced cells (SCLC) was 10-fold and 100-fold greater, respectively, than the glucose oxidase control. This system could be widely applied for specific therapy of cancer cells by targeting unique surface molecules (antigens) using the corresponding Mab/ST-PA complex to transfer a variety of effector molecules; e.g., immunotoxic compounds, into target cells with a high degree of efficiency and specificity.

Targeted gene transfer system using a streptavidin-transforming growth factor-alpha chimeric protein.

The previously reported streptavidin-TGFalpha chimeric protein-based delivery system (Ohno and Meruelo, DNA Cell Biol. 15:401-406, 1996) could efficiently transfer protein molecules into A431 cells via the epidermal growth factor (EGF) receptor. We have modified this delivery system for the transfer of DNA. For this purpose, we have linked the chimeric protein ST-TGFalpha to DNA through biotinylated polylysine molecules. We show with this system, in the presence of the endosome-destabilizing reagent chloroquine, an average of 50-fold increase in reporter gene expression in comparison with polylysine DNA complexes alone. This gene expression is specific for EGF receptor-expressing cells and is blocked by EGF-binding molecules. These results suggest that the ST-TGFalpha biotinylated polylysine system could be used to deliver DNA to targeted cells.

Genomic analysis and localization of murine Deltex, a modulator of notch activity, to mouse chromosome 5 and its human homolog to chromosome 12.

Deltex is a component of the Notch signaling network, which mediates cellular differentiation, proliferation, and apoptosis during development. Murine Deltex was initially isolated as a cDNA transcript that displayed increased expression in T-cell tumors induced by gamma irradiation. The in vivo function of Deltex is unknown; however, the emerging role of Notch signaling in T-cell development and lymphomagenesis indirectly supports a role for Deltex in these processes. To investigate the regulation of Deltex expression in both normal and transformed tissue, we have begun analyzing the Deltex genomic locus. Here, we report the exon-intron organization of Deltex and map the locus to the middistal region of mouse chromosome 5, tightly linked to the Adam1a, Lnk, Tbx5, and Nos1 loci. The human homolog of Deltex has been localized to chromosome 12.