A monoclonal antibody (MCI20.6) which inhibited measles virus (MV) binding to host cells was previously used to characterize a 57to 67-kDa cell surface glycoprotein as a potential MV receptor. In the present work, this glycoprotein (gp57/67) was immunopurified, and N-terminal amino acid sequencing identified it as human membrane cofactor protein (CD46), a member of the regulators of complement activation gene cluster. Transfection of nonpermissive murine cells with a recombinant expression vector containing CD46 cDNA conferred three major properties expected of cells permissive to MV infection. First, expression of CD46 enabled MV to bind to murine cells. Second, the CD46-expressing murine cells were able to undergo cell-cell fusion when both MV hemagglutinin and MV fusion glycoproteins were expressed after infection with a vaccinia virus recombinant encoding both MV glycoproteins. Third, M12.CD46 murine B cells were able to support MV replication, as shown by production of infectious virus and by cell biosynthesis of viral hemagglutinin after metabolic labeling of infected cells with [35S5methionine. These results show that the human CD46 molecule serves as an MV receptor allowing virus-cell binding, fusion, and viral replication and open new perspectives in the study of MV pathogenesis.

We have generated transgenic mice ubiquitously expressing the human receptor for measles virus (MV), CD46 (membrane cofactor protein). Various cell types were isolated from these transgenic mice and analyzed for their ability to support MV replication in vitro. Although MV could enter into all CD46-expressing cells, differential susceptibilities to MV infection were detected depending on the cell type. Cell cultures obtained from transgenic lungs and kidneys were found to be permissive of MV infection, since RNA specific for MV genes was detected and viral particles were released, although at a low level. Similarly to human lymphocytes, activated T and B lymphocytes isolated from transgenic mice could support MV replication; virus could enter, transcribe viral RNA, and produce new infectious particles. When expressing viral proteins, lymphocytes down-regulated CD46 from the surface. Interestingly, while activated T lymphocytes from nontransgenic mice did not support MV infection, activated nontransgenic murine B lymphocytes replicated MV as well as transgenic B lymphocytes, suggesting the use of an alternative virus receptor for entry. In contrast to the previous cell types, murine peritoneal and bone marrow-derived macrophages, regardless of whether they were activated, could not support MV replication. Furthermore, although MV entered into macrophages and virus-specific RNA transcription occurred, no virus protein or infectious virus particles could be detected. These results show the importance of the particular cell-type-specific host factors for MV replication in murine cells which may be responsible for the differential permissivity of MV infection.

SummaryMeasles virus after binding to its cell surface human CD46 receptor fuses with the plasma membrane. This fusion results in envelope hemagglutinin (H) and fusion glycoprotein (F) incorporated into the plasma membrane and injection of the nucleocapsid made of nucleoprotein (NP) into the cytosol. The influence of targeting measles virus (MV) to CD46 in the processing and presentation of MV H and NP to antigen specific MHC class II bE d-and bAd-restricted T cell hybridomas was explored using murine M12-CD46 B cell transfectants. Parent M12 cells, which lack any MV receptor, were unable to present any of these two viral proteins when incubated with MV particles. Incubating M12.CD46 cells with 200 ng and 10/zg of MV could strongly stimulate H-specific and NP-specific T cells, respectively. Neosynthesis of MV proteins was not necessary since the efficiency of antigen presentation was similar when using ultraviolet-inactivated MV. Similar enhancing effects (more than 1,000-fold) on antigen presentation were also observed when using purified native H soluble or incorporated into liposomes whereas denaturating H glycoprotein resulted in a poor efficiency in T cell stimulation, M12.CD46 being no more potent than the parental M12 counterpart. MV H and NP presentation efficiency did not depend on MV fusion with plasma membrane as revealed by the lack of effect of specific fusion inhibitors. Both MV H and NP presentations were sensitive to chloroquine inhibition indicating that antigens from CD46-mediated captured MV were likely processed in the endosome/lysosome compartment. Altogether these data indicate that (a) MV targeting via CD46 has a strong effect on the efficiency of antigen presentation by MHC class II, (b) the effect is mediated by the binding of H to CD46, and (c) though MV does fuse with plasma membrane, endocytosis, and processing of virus particles are also occurring. Since, in humans, CD46 is expressed in almost every tissue including professional antigen-presenting cells, such a targeting is likely to play a crucial role in the CD4 + T cell-mediated primary immune response against the pathogen in vivo.

Human CD46, a member of the family of regulators of complement activation, has been shown recently to act as a measles virus (MV) receptor, interacting with the virus envelope glycoprotein haemagglutinin (HA). Owing to alternative RNA splicing, several CD46 isoforms are co-expressed in all tissues except erythrocytes. The optional exons encode extracellular serine-, threonine-and proline-rich regions of CD46 (designated STP-A, -B and -C) which are located proximal to the plasma membrane, and alternative cytoplasmic tails (CYT1 or CYT2). The ability of the BC-CYT2, B-CYT2 and BC-CYT1 CD46 isoforms, expressed in rodent Chinese hamster ovary (CHO) cells, to mediate MV infection was tested. Every isoform was recognized by a monoclonal antibody (MAb), MCI20.6, which recognizes the MV-binding site on CD46. CHO cells expressing any of these CD46 isoforms were able to bind MV, the level of binding correlating with the CD46 expression level. Likewise, MV infection induced the cell-cell fusion of all CD46-expressing CHO cells but not of the parental CHO cells. Accordingly, MV replication was observed after infection of CHO cells expressing each CD46 isoform but not after infection of parental CHO cells. Finally, cell surface expression of every isoform was decreased after infection by MV. Altogether these data showed that the specific STP regions of CD46 played no major role in HA-mediated MV binding to CD46, virus infection and virus-induced down-regulation of CD46. Moreover, the CYT1 and CYT2 cytoplasmic tails of CD46 are either functionally similar although having distinct amino acid sequences or are dispensable for interaction with HA of MV.

The CD46 molecule is a receptor for measles virus (MV), CD46, which protects autologous cells from complement-mediated damage, exists in several isoforms which are variably expressed in different human tissues. These isoforms differ in their cytoplasmic and transmembrane regions and in a small portion of their proximal extracytoplasmic regions. To examine the role of the cytoplasmic and transmembrane regions of CD46 in MV infection, mouse M12 B cells stably expressing a transmembrane or a chimeric glycosylphosphatidylinositol (GPI)-anchored form of CD46 (CD46-GPI) were used. Both the GPI-anchored and transmembrane CD46 forms were able to mediate MV binding. MV binding mediated by the GPI-anchored form but not that mediated by the transmembrane form was abolished after treatment with phosphatidylinositol phospholipase C. MV infection of both M12.CD46 and M12.CD46-GPI cells but not parental M12 cells resulted in MV replication. Expression of hemagglutinin induced cell surface down-regulation of both CD46 and CD46-GPI. Both M12.CD46 and M12.CD46-GPI cells were able to efficiently capture MV for presentation

We have shown recently that a retrovirus vector expressing a natural mutant form of the PML-RAR alpha protein characteristic of human acute promyelocytic leukaemia can transform early chicken hematopoietic progenitors (Altabef et al., 1996). Neither truncated PML nor truncated RAR alpha alone could induce transformation which suggest that the two domains should cooperate for the oncogenicity of the fusion product. To further investigate the mechanisms of this co-operation, we have tested whether a truncated RAR alpha could cooperate with the v-erbB oncogene. This oncogene has previously been shown to co-operate with the rearranged thyroid hormone receptor, v-erbA, to transform erythrocytic progenitors. We show that v-erbB and a truncated RAR alpha co-operate when expressed simultaneously as independent products to transform very early chicken haematopoietic cells close to pluripotent stage. In addition, we show that v-erbB alters transcriptional abilities of RAR alpha by both enhancing its effects on RARE and reducing those on AP-1. Therefore, RAR alpha is able to co-operate with different kinds of proteins to induce transformation of early haematopoietic cells. This strongly suggests that RAR alpha are involved in the differentiation commitment of early haematopoietic progenitors during the normal process of haematopoietic differentiation. These data bring new insights in the mechanisms of oncogenic transformation by rearranged RAR alpha.