Characterization of Molecular Species Carrying Gross Cell Surface Antigen

The Gross cell surface antigen (GCSA), associated with expression of endogenous Gross-type murine leukemia virus (G-MuLV) in tissues of mice, is defined by the cytotoxic reaction of a C57BL/6 antiserum, anti-AKR spontaneous leukemia K36, with cells of the Gross virus-induced C57BL/6 leukemia, Emale symbolG2. Sequential lactoperoxidase-catalyzed radioiodination of Emale symbolG2 cells, Nonidet P-40 lysis, precipitation with anti-K36 serum, and sodium dodecyl sulfate-polyacrylamide gel electrophoresis identified molecules with properties of polyproteins encoded by the gag region of the viral genome. These cell surface species could also be labeled by in vitro culturing of Emale symbolG2 with radioactive glucosamine. The viral specificity of these molecules and their participation in the GCSA typing system were established as follows. (i) Absorption of anti-K36 serum with GCSA(+), but not GCSA(-), leukemias led to a marked decrease in precipitation of these proteins. (ii) The same Emale symbolG2 cell surface proteins were also precipitated by antisera against the MuLV virion proteins p30 and p15. (iii) Anti-K36 was shown to possess antibodies against Gross virus p30 and p15. (iv) "Clearing" the Emale symbolG2 lysate of molecules reactive with anti-p30 or anti-p15 sera removed molecules reactive with anti-K36 serum. (v) Absorption of anti-K36 serum with disrupted G-MuLV virions or with Gross p30 or p15 removed GCSA cytotoxic antibodies; partial absorption was achieved with disrupted Rauscher-MuLV (R-MuLV) or with R-MuLV p30, and no absorption was found with R-MuLV p15. These data show that Emale symbolG2 cells express, on their surfaces, MuLV core polyproteins that apparently can be glycosylated and on which the determinants of GCSA are located.     

The immune response to Moloney murine leukemia virus-induced tumors: induction of cytolytic T lymphocytes specific for both viral and tumor-associated antigens

The specificity of CTL generated against tumors induced by murine leukemia viruses (MuLV) has been reported to parallel the expression of two serologically defined tumor cell surface antigens--the cross-reactive FMR antigen expressed on the surface of tumors induced by Friend, Moloney, and Rauscher MuLV, and the Gross cell surface antigen (GCSA) expressed on tumors induced by AKV/Gross MuLV. We examined the specificity of CTL generated against MuLV-induced tumors and identified two distinct patterns of reactivity. The first follows the traditional pattern of FMR vs GCSA reactivity as assessed on a panel of established MuLV-induced lymphomas. However, CTL exhibiting this pattern of reactivity are incapable of lysing MuLV-infected fibroblasts. CTL exhibiting the second pattern of reactivity are capable of lysing MuLV-induced lymphomas as well as MuLV-infected fibroblasts. In addition, these CTL exhibit extensive cross-reactivity between lymphomas and fibroblasts infected by both groups of MuLV. Our results suggest that CTL exhibiting the traditional FMR vs GCSA pattern of reactivity are directed against a tumor-associated antigen and not against virus-encoded antigens, and that CTL directed against MuLV-encoded antigens demonstrate extensive cross-reactivity, including the ability to lyse AKV-infected cells.     

Induction of GIX antigen and gross cell surface antigen after infection by ecotropic and xenotropic murine leukemia viruses in vitro.

A number of ecotropic and xenotropic murine leukemia viruses were examined for their ability to induce the GIX antigen and Gross cell surface antigen (GCSA) in tissue culture fibroblasts. GIX appears to be a constituent of murine leukemia virus gp70; a molecular characterization of GCSA has not yet been reported. Antigen induction was measured by the ability of productively infected cells to absorb cytotoxic activity from the standard GIX- and GCSA-typing antisera. Cells infected by ecotropic viruses displayed four distinct phenotypes GIX:+/GCSA++, GIX-/GCSA++, GIX++/GCSA+, and GIX-/GSCA+; cells infected by xenotropic viruses were either GIX-/GCSA+ or GIX-/GCSA-. GIX induction appeared to be a type-specific property of some but not all Gross-AKR type ecotropic viruses. Differences in the degree of absorption of the GCSA antiserum by ecotropic virus- and xenotropic virus-infected cells indicated that GCSA may comprise multiple antigenic determinants.     

Transformation of NIH 3T3 cells by DNA of the MCF-7 human mammary carcinoma cell line induces expression of an endogenous murine leukemia provirus.

Previous studies identified two glycoproteins of 86 (gp86) and 72 (gp72) kilodaltons and two nonglycosylated proteins of 70 (p70) and 19 (p19) kilodaltons which were specifically expressed in NIH cells transformed by DNA of the MCF-7 human mammary carcinoma cell line. Pulse-chase experiments and the use of tunicamycin to inhibit glycosylation suggested that gp86, gp72, and p19 were related as precursor products. Characteristics of the four transformation-associated proteins resembled those of murine leukemia virus (MuLV) proteins. Sera raised against disrupted MuLV immunoprecipitated the same four proteins in extracts of NIH(MCF-7) cells and MuLV-infected NIH 3T3 cells. In addition, a monoclonal antibody against MuLV gp70 immunoprecipitated proteins gp86 and gp72, whereas a monoclonal antibody against MuLV p15(E) immunoprecipitated gp86 and p19. These results indicate that proteins gp86, gp72, and p19 expressed in NIH(MCF-7) transformants correspond to MuLV envelope proteins gp80env, gp70, and p15(E), respectively. The transformation-associated protein p70 appears to be a non-envelope MuLV protein, most likely p65gag. Northern blot analysis confirmed that transformation of NIH cells by MCF-7 mammary carcinoma DNA led to the induction of an endogenous MuLV provirus.     

Incorporation of human immunodeficiency virus type 1 Gag proteins into murine leukemia virus virions.

The retroviral Gag polyprotein is necessary and sufficient for assembly and budding of viral particles. However, the exact inter- and intramolecular interactions of the Gag polyproteins during this process are not known. To locate functional domains within Gag, we generated chimeric proviruses between human immunodeficiency virus type 1 (HIV-1) and murine leukemia virus (MuLV). In these chimeric proviruses, the matrix or capsid proteins of MuLV were precisely replaced with the matrix or capsid proteins of HIV-1. Although the chimeric proviruses were unable to efficiently assemble into mature viral particles by themselves, coexpression of wild-type MuLV Gag rescued the HIV proteins into virions. The specificity of the rescue of HIV proteins into MuLV virions shows that specific interactions involving homologous matrix or capsid regions of Gag are necessary for retroviral particle formation.     

Functional exchange of an oncoretrovirus and a lentivirus matrix protein.

To map functional domains in the retroviral Gag protein we have constructed chimeric viruses where regions of the murine leukemia virus (MuLV) Gag protein have been replaced with analogous sequences from human immunodeficiency virus type 1 (HIV-1). Here we describe the chimeric virus MuLV(MAHIV) which contains the HIV-1 matrix (MA) protein in place of the MuLV MA. MuLV(MAHIV) is infectious but grows at a reduced rate compared with wild-type MuLV. We found that the partial defect in replication of the chimeric virus is at a late stage in the viral life cycle. The MuLV(MAHIV) Gag proteins are distributed aberrantly within cells and are not associated with cellular membranes. Unlike MuLV, HIV-1 is able to integrate into growth-arrested cells. Incorporation of the HIV-1 MA, which is known to play a role in infection of nondividing cells, does not enable MuLV(MAHIV) to be expressed in growth-arrested cells. While it possesses no amino acid homology, we found that the HIV-1 MA can efficiently replace the MuLV matrix protein in infection.     

Detection of polymorphism in BALB/c leukemia viruses with mouse antisera.

Antisera produced in mice recognize primarily type-specific antigenic determinants on both the major core protein, p30, and the major envelope proteins, gp70 and p15(E), of the endogenous leukemia viruses (MuLV) of BALB/c mice. Three different mouse sera were investigated in detail. (i) Antisera prepared in C57BL/6 mice against the AKR leukemia K36 reacted with the gp70, p15(E), and p30 proteins of MuLV. Certain pools of the C57BL/6 anti-AKR K36 serum contained antibodies which serologically distinguished the p30 proteins of N-ecotropic, B-ecotropic, and xenotropic BALB/c MuLV. (ii) Antisera prepared in BALB/c mice against the BALB/c sarcoma 1315 contained antibodies that reacted with a type-specific antigen of the 1315 MuLV gp70 that is not found on other BALB/c MuLV. (iii) The normal sera of multiparous BALB/c mice contained antibodies that reacted with gp70 and p15(E) proteins of ecotropic MuLV. Sera from some of these mice contained antibodies that serologically distinguished the gp70 of N-ecotropic and B-ecotropic BALB/c viruses. These results emphasize the utility of mouse antisera in the serological typing of MuLV. Furthermore, the antigenic differences observed in the p30 and gp70 proteins should be of particular use in the future analysis of recombinant BALB/c MuLV.     

Identification of an FMR cell surface antigen associated with murine leukemia virus-infected cells.

FMR antigens are found on the surface of cells infected with Friend, Moloney, and Rauscher murine leukemia viruses (MuLV). These antigens are serologically distinct from the G cell surface antigens that are found on cells infected with endogenous MuLV (AKR and Gross virus). Cell surface antigens of both virus groups are immunogenic in mice, and immunization with appropriate virus-infected cells leads to the production of cytotoxic antisera. The cytotoxic activity of FMR antisera can be absorbed by disrupted preparations of Rauscher MuLV, but not by AKR MuLV. FMR antisera precipitate the viral envelope proteins gp70, pl5(E), and p12(E) from detergent-disrupted preparations of [3H]leucine-labeled MuLV. The reaction of these antisera with p15(E) and p12(E) proteins is directed against group-specific antigens and can be absorbed with AKR MuLV; in contrast, the reaction of these antisera with gp70 is directed against type-specific antigens and is absorbed only by viruses of the FMR group. In immune precipitation assays with detergent-disrupted 125I surface-labeled cells, FMR antisera react only with type-specific antigens of the viral envelpe protein. On the basis of these findings we conclude that the FMR cell surface antigen is a determinant on the MuLV env gene product.     

Mechanism of formation of pseudotypes between vesicular stomatitis virus and murine leukemia virus.

Pseudotypes of vesicular stomatitis virus (VSV) and Moloney murine leukemia virus (MuLV), defined by their resistance to neutralization by anti-VSV antiserum, are released preferentially at early times after infection of MuLV-producing cells with VSV. At later times, after synthesis of MuLV proteins has been inhibited by the VSV infection, neither MuLV virions nor the VSV (MuLV) pseudotypes are made. Infection of MuLV-producing cells with mutants of VSV having temperature-sensitive lesions in either G or M protein does not generate pseudotypes at nonpermissive temperature, indicating that both proteins are needed for pseudotypes to form. Although the pseudotypes resist neutralization by anti-VSV serum, they are inactivated by anti-VSV serum plus complement, and they can be precipitated by rabbit anti-VSV serum plus goat anti-rabbit IgG. These results, coupled with experiments using a temperature-sensitive mutant of VSV G protein grown at partly restrictive temperature, suggest that small numbers of VSV G protein are obligately incorporated into VSV(MuLV) pseudotypes. There appears to be a stringent requirement for recognition of the viral core by homologous envelope components as the nucleating step in the budding process. Only after such a nucleation can the envelope components of the second virus substitute into the membrane of the budding particle.     

Sequences in the cytoplasmic tail of the gibbon ape leukemia virus envelope protein that prevent its incorporation into lentivirus vectors

Pseudotyping retrovirus and lentivirus vectors with different viral fusion proteins is a useful strategy to alter the host range of the vectors. Although lentivirus vectors are efficiently pseudotyped by Env proteins from several different subtypes of murine leukemia virus (MuLV), the related protein from gibbon ape leukemia virus (GaLV) does not form functional pseudotypes. We have determined that this arises because of an inability of GaLV Env to be incorporated into lentivirus vector particles. By exploiting the homology between the GaLV and MuLV Env proteins, we have mapped the determinants of incompatibility in the GaLV Env. Three modifications that allowed GaLV Env to pseudotype human immunodeficiency virus type 1 particles were identified: removal of the R peptide (C-terminal half of the cytoplasmic domain), replacement of the whole cytoplasmic tail with the corresponding MuLV region, and mutation of two residues upstream of the R peptide cleavage site. In addition, we have previously proposed that removal of the R peptide from MuLV Env proteins enhances their fusogenicity by transmitting a conformational change to the ectodomain of the protein (Y. Zhao et al., J. Virol. 72:5392-5398, 1998). Our analysis of chimeric MuLV/GaLV Env proteins provides further evidence in support of this model and suggests that proper Env function involves both interactions within the cytoplasmic tail and more long-range interactions between the cytoplasmic tail, the membrane-spanning region, and the ectodomain of the protein.     

Sequence analysis of amphotropic and 10A1 murine leukemia viruses: close relationship to mink cell focus-inducing viruses.

Viral interference studies have demonstrated the existence of four distinct murine leukemia virus (MuLV) receptors on NIH 3T3 mouse cells. The four viral interference groups are ecotropic MuLV; mink cell focus inducing virus (MCF); amphotropic MuLV; and 10A1, a recombinant derivative of amphotropic MuLV that uses a unique receptor but also retains affinity for the amphotropic MuLV receptor. We report here that 10A1 infects rat and hamster cells, unlike its amphotropic parent. We isolated an infectious molecular clone of 10A1 and present here the sequences of the env genes and enhancer regions of amphotropic MuLV and 10A1. The deduced amino acid sequences of amphotropic MuLV and 10A1 gp70su are remarkably similar to those of MCF and xenotropic MuLV (for which mouse cells lack receptors), with 64% amino acids identical in the four groups. We generated a consensus from these comparisons. Further, the differences are largely localized to a few discrete regions: (i) amphotropic MuLV has two short insertions relative to MCF, at residues 87 to 92 and 163 to 169, and (ii) amphotropic MuLV and MCF are totally different in a hypervariable region, which is greater than 30% proline, at residues approximately 253 to 304. 10A1 closely resembles amphotropic MuLV in its N terminus but contains an MCF-type hypervariable region. These results suggest the possibility that receptor specificity is localized in these short variable regions and further that the unique receptor specificity of 10A1 is due to the novel combination of amphotropic MuLV and MCF sequences rather than to the presence of any novel sequences. The Env proteins of ecotropic MuLV are far more distantly related to those of the other four groups than the latter are to each other. We also found that the enhancer regions of amphotropic MuLV and 10A1 are nearly identical, although 10A1 is far more leukemogenic than amphotropic MuLV.     

Efficient pseudotyping of murine leukemia virus particles with chimeric human foamy virus envelope proteins.

Incorporation of human foamy virus (HFV) envelope proteins into murine leukemia virus (MuLV) particles was studied in a transient transfection packaging cell system. We report here that wild-type HFV envelope protein can pseudotype MuLV particles, albeit at low efficiency. Complete or partial removal of the HFV cytoplasmic tail resulted in an abolishment or reduction of HFV-mediated infectivity, implicating a role of the HFV envelope cytoplasmic tail in the pseudotyping of MuLV particles. Mutation of the endoplasmic reticulum retention signal present in the HFV envelope cytoplasmic tail did not result in a higher relative infectivity of pseudotyped retroviral vectors. However, a chimeric envelope protein, containing an unprocessed MuLV envelope cytoplasmic domain fused to a truncated HFV envelope protein, showed an enhanced HFV specific infectivity as a result of an increased incorporation of chimeric envelope proteins into MuLV particles.     

Analysis of the cell fusion activities of chimeric simian immunodeficiency virus-murine leukemia virus envelope proteins: inhibitory effects of the R peptide.

It was previously reported that truncation or proteolytic removal of the C-terminal 16 amino acids (the R peptide) from the cytoplasmic tail of the murine leukemia virus (MuLV) envelope protein greatly increases its fusion activity. In this study, to investigate the specificity of the effect of the R peptide on the fusion activity of viral envelope proteins, we expressed simian immunodeficiency virus (SIV)-MuLV chimeric proteins in which the entire cytoplasmic tail of the SIV envelope protein was replaced by either the full-length MuLV cytoplasmic tail or a truncated MuLV cytoplasmic tail with the R peptide deleted. Extensive fusion of CD4-positive cells with the chimeric protein containing a truncated MuLV cytoplasmic tail was observed. In contrast, no cell fusion activity was found for the chimeric protein with a full-length MuLV cytoplasmic tail. We constructed another SIV-MuLV chimeric protein in which the MuLV R peptide was added to an SIV envelope protein cytoplasmic tail 17 amino acids from its membrane-spanning domain. No fusion activity was observed within this construct, while the corresponding truncated SIV envelope protein lacking the R peptide showed extensive fusion activity. No significant difference in the transport or surface expression was observed among the various SIV-MuLV chimeric proteins and the truncated SIV envelope protein. Our results thus demonstrate that the MuLV R peptide has profound inhibitory effects on virus-induced cell fusion, not only with MuLV but also in a distantly related retroviral envelope protein which utilizes a different receptor and fuses different cell types.     

Foamy virus envelope glycoprotein is sufficient for particle budding and release

Foamy viruses (FVs) are classified in the family Retroviridae, but recent data have shown that they are not conventional retroviruses. Notably, several characteristics of their particle replication strategies are more similar to those of hepatitis B virus (HBV) than those of typical retroviruses. Compared to conventional retroviruses, which require only Gag proteins for budding and release of virus-like particles (VLPs), both FV and HBV require Env proteins. In the case of HBV, Env (S protein) alone is sufficient to form subviral particles (SVPs). Because FVs also depend on Env for budding, we tested whether FV Env alone could produce SVPs. The Env proteins of FV and murine leukemia virus (MuLV) were both released into cell culture supernatants and migrated into isopycnic gradients; however, unlike MuLV Env, FV Env displayed characteristics of SVPs. FV Env particles were of greater density than those of MuLV (1.11 versus 1.07 g/ml, respectively), which strongly suggested that the released proteins of FV Env were particulate. When we examined FV SVPs by immunoelectron microscopy, we found particles that were consistent in morphology, size, and staining with gold beads, similar to FV VLPs and unlike the particle-like structures of MuLV Env, which were more consistent with vesicles produced from nonspecific membrane "blebbing." Taken together, our results demonstrated that FV Env alone is sufficient for particle budding. This finding is unique among retroviruses and further demonstrated the similarities between FV and HBV.     

Autologous immune responses to the major oncornavirus polypeptides in unmanipulated AKR/J mice.

The autologous immune response of AKR/J mice to the structural proteins of murine leukemia virus (MuLV) was examined. Immunoglobulins from the renal glomeruli were chemically eluted, separated from antigens, recovered, and tested for immunological reactivity against MuLV structural proteins. Analyzing immune precipitates obtained after mixing radiolabeled Tween-disrupted MuLV preparations with eluates from AKR/J mice on sodium dodecyl sulfategel electrophoresis, we found evidence of antibodies to the major classes of MuLV structural components: gp70, gp45, p30, and one or more proteins in the 10,000- to 15,000-dalton class. Using rate zonal centrifugation we confirmed that the eluates from AKR/J glomeruli contained antibody(s) that bound specifically to p30. These results indicate that AKR/J mice spontaneously mount immune responses against the major oncornavirus polypeptide antigens.     

Antibodies to murine leukemia virus gp70 and p15(E) in sera of BALB/c mice immunized with syngeneic chemically induced sarcomas

The sera of BALB/c mice immunized with syngeneic methylcholanthrene-induced sarcomas commonly contain antibodies that react with the immunizing tumor line and also with other chemically induced sarcomas. Three lines of evidence suggest that the cross-reactive antibodies are directed against antigenic determinants of gp70 and p15(E), envelope proteins of endogenous murine leukemia virus (MuLV). 1) The antibodies bind only to sarcomas expressing MuLV antigens, 2) they are removed by absorption with purified MuLV antigens, and 3) they precipitate proteins identified as gp70 and p15(E) from 125I-labeled sarcoma cell lysates.     

The carboxy-terminal fragment of nucleolin interacts with the nucleocapsid domain of retroviral gag proteins and inhibits virion assembly

A yeast two-hybrid screen for cellular proteins that interact with the murine leukemia virus (MuLV) Gag protein resulted in the identification of nucleolin, a host protein known to function in ribosome assembly. The interacting fusions contained the carboxy-terminal 212 amino acids of nucleolin [Nuc(212)]. The nucleocapsid (NC) portion of Gag was necessary and sufficient to mediate the binding to Nuc(212). The interaction of Gag with Nuc(212) could be demonstrated in vitro and was manifested in vivo by the NC-dependent incorporation of Nuc(212) inside MuLV virions. Overexpression of Nuc(212), but not full-length nucleolin, potently and specifically blocked MuLV virion assembly and/or release. A mutant of MuLV, selected to specifically disrupt the binding to Nuc(212), was found to be severely defective for virion assembly. This mutant harbors a single point mutation in capsid (CA) adjacent to the CA-NC junction, suggesting a role for this region in Moloney MuLV assembly. These experiments demonstrate that selection for proteins that bind assembly domain(s) can yield potent inhibitors of virion assembly. These experiments also raise the possibility that a nucleolin-Gag interaction may be involved in virion assembly.     

The prion protein gene is dispensable for the development of spongiform myeloencephalopathy induced by the neurovirulent Cas-Br-E murine leukemia virus.

The Cas-Br-E murine leukemia virus (MuLV) induces paralysis in susceptible mice that is accompanied by a severe spongiform myeloencephalopathy. These neurodegenerative lesions are very similar to those observed in prion diseases. To determine whether the prion protein gene (Prn-p) product was a downstream effector of this neurovirulent MuLV, we inoculated Prn-p(-/-) knockout homozygote and control heterozygote or wild-type mice with this retrovirus. All groups developed typical paralysis and spongiform encephalopathy, and no differences in clinical or histological phenotypes were observed between these groups. These results indicate that the Cas-Br-E MuLV does not require the prion protein to induce lesions. Thus, MuLV and prion proteins may induce a very similar disease through distinct pathways, or the viral Env protein, which harbors the primary determinant of pathogenicity, may act in a common pathway but downstream of the prion protein.