Effect of the alterations in the fusion protein of measles virus isolated from brains of patients with subacute sclerosing panencephalitis on syncytium formation

Measles virus (MV) is the causative agent of subacute sclerosing panencephalitis (SSPE) and viruses isolated from brains of the patients contain numerous mutations. We have previously demonstrated that the hemagglutinin (H) protein of MV SSPE strains can interact with the signaling lymphocyte activation molecule (SLAM) and an unidentified molecule on Vero cells, but not with CD46, as a receptor. The mechanism by which MV SSPE strains can induce cell–cell fusion in SLAM-negative Vero cells is not understood. We report here on the effect of mutations in the fusion (F) proteins of three MV SSPE strains on syncytium formation. The F proteins of the three SSPE strains were functional and co-expression with H protein from the MV wild-type or SSPE strains in this study induced formation of large syncytia in Vero cells as well as in cell lines expressing SLAM or CD46. Expression of chimeric F proteins of SSPE strains showed that amino acid substitutions in the F protein extracellular as well as cytoplasmic domain contributed to enhanced cell–cell fusion in Vero cells. These findings suggest a common molecular mechanism and a key role of the F protein for syncytium formation in cells expressing an unidentified third receptor for MV.     

Resting lymphocyte transduction with measles virus glycoprotein pseudotyped lentiviral vectors relies on CD46 and SLAM

The measles virus (MV) glycoproteins hemagglutinin (H) and fusion (F) were recently shown to mediate transduction of resting lymphocytes by lentiviral vectors. MV vaccine strains use CD46 or signaling lymphocyte activation molecule (SLAM) as receptor for cell entry. A panel of H protein mutants derived from vaccine strain or wild-type MVs that lost or gained CD46 or SLAM receptor usage were investigated for their ability to mediate gene transfer into unstimulated T lymphocytes. The results demonstrate that CD46 is sufficient for efficient vector particle association with unstimulated lymphocytes. For stable gene transfer into these cells, however, both MV receptors were found to be essential.     

Wild-type measles virus induces large syncytium formation in primary human small airway epithelial cells by a SLAM(CD150)-independent mechanism

In the natural course of measles virus (MV) infection, epithelial cells are primary targets of MV. However, it has been shown that wild-type MV utilizes signaling lymphocyte activation molecule (SLAM or CD150) as a cellular receptor, which is expressed only in some T and B cells, thymocytes, and dendritic cells. To understand how wild-type MV infects non-lymphoid cells, several non-lymphoid cells were examined for their susceptibility to wild-type MV. Here, we show that wild-type MV can infect primary human small airway epithelial cells (SAEC) and induce formation of large syncytia in vitro. mRNA specific for SLAM was not detected in SAEC, indicating that wild-type MV infects SAEC and induces syncytia formation via a SLAM-independent mechanism.     

Lymphokines and immunoregulatory molecules in subacute sclerosing panencephalitis

Frozen brain specimens from six patients with subacute sclerosing panencephalitis (SSPE) were analyzed immunohistochemically for the presence of leukocyte subpopulations and specific cytokines. In brain regions demonstrating perivascular cell infiltration and gliosis, CD4 and CD8 positive cells were identified within the brain parenchyma. Cytokine analysis revealed cells staining positively for tumor necrosis factor-alpha and interferon-gamma. These results were similar to those observed in multiple sclerosis (MS) and progressive rubella panencephalitis tissue and were different from other predominantly noninflammatory neurologic diseases and normal controls. Although SSPE and MS differ significantly in their etiology and histopathology, the similarities in leukocyte and cytokine staining patterns suggest a common mechanism of disease progression.     

Molecular assembly of CD46 with CD9, alpha3-beta1 integrin and protein tyrosine phosphatase SHP-1 in human macrophages through differentiation by GM-CSF

Human CD46, formerly membrane cofactor protein (MCP), binds and inactivates complement C3b and serves as a receptor for measles virus (MV), thereby protecting cells from homologous complement and sustaining systemic viral infection. CD46 on activated macrophages (Mphi) but not intact monocytes is presumed to be the factor responsible for virus-mediated immune modulation including down-regulation of IL-12 production. As CD46 is expressed on both Mphi and monocytes, the molecular mechanisms responsible for these distinct immune responses remain largely unknown. Here, we found that peripheral blood monocytes treated for 5--8 days with GM-CSF (i.e. mature Mphi) acquired the capacity to assemble CD9, alpha3-beta1 integrin and the tyrosine phosphatase SHP-1 with their CD46. Prior to this maturation stage, Mphi expressed sufficient amounts of CD9 and CD46 but showed no such complex formation, and as in intact monocytes MV replication was markedly suppressed. By flow cytometry and confocal microscopy, the complex was found to assemble on the surface in cells treated with approximately 6 days with GM-CSF but not for approximately 2 days. Notably, an alternative MV receptor SLAM CDw150 was neither expressed nor recruited to this complex throughout GM-CSF-mediated Mphi differentiation. These responses and molecular links were not reproduced in the hamster cell line CHO expressing human CD46 although these cells acquired high susceptibility to MV. Based on these observations, MV susceptibility in human myeloid lineages appears not to be as simple as that observed in human CD46-transfected non-myeloid cells. The molecular complex involving CD46 may confer high MV permissiveness leading to immune modulation in Mphi.     

Measles virus: immunomodulation and cell tropism as pathogenicity determinants

As important determinants of measles virus (MV) pathogenicity, the MV glycoproteins play a key role in conferring the cellular tropism of this virus, but also in modulating the activity of immunocompetent cells. Whereas all MV strains are able to use CD150 (SLAM) for binding and entry into target cells, only certain, mainly vaccine, strains, can use both CD46 and CD150. Both molecules are down-regulated from the cell surface and this is brought about by both infection and contact with the MV H protein of strains that are able to interact with these molecules. Whereas down-regulation of CD46 could be linked to enhanced sensitivity to complement-mediated lysis, and may thus represent an attenuation marker for vaccine strains, pathogenetic consequences of CD150 down-regulation are unknown as yet. Although the role of CD150 is not entirely clear, viruses containing a wild-type strain-derived H protein revealed a particular tropism for human dendritic cells in vitro, and replicated well in secondary lymphatic tissues of cotton rats where they were also able to cause immunosuppression, as documented by an impaired proliferative response of lymphocytes ex vivo. Most likely, inhibition of T cell expansion by these cells is brought about by another activity of the MV glycoprotein complex, namely by disrupting a pathway important for S-phase entry of T cells, by a mere surface contact.     

Adaptation of wild-type measles virus to CD46 receptor usage

Summary.  Vaccine strains of measles virus (MV) use CD46 as receptor and downregulate CD46 from the surface of infected cells. MVs isolated and passaged on B-lymphoid cells (wild-type MVs) seem to use another receptor and do not downregulate CD46. In the present study, we found that isolation of MV on human or marmoset B-lymphoid cells did not alter the MV haemagglutinin (H) protein relative to that in the patient. The wild-type isolates were adapted to the human epithelial HEp-2 cell line or the monkey fibroblast Vero cell line. All HEp-2 cell adapted viruses and 1 out of 4 Vero cell adapted viruses acquired the capacity to use CD46 as receptor, as measured by their ability to infect murine cells expressing human CD46. Adaptation to CD46 receptor usage was coupled to substitution of amino acid 481 of the MV H protein from asparagine to tyrosine but not to CD46 downregulation. The present study demonstrates that CD46 receptor usage can be induced by adaptation of wild-type MV to cells that do not express a wild-type receptor and suggests that a similar mechanism acted on the progenitor viruses of the present MV vaccine strains during their isolation and attenuation. Received June 5, 2000 Accepted October 5, 2000     

The cellular receptor for measles virus--elusive no more

The identity of the measles virus receptor has been controversial. Several years ago CD46 was identified as a cellular receptor for the Edmonston strain of measles virus, but most clinical isolates of measles virus, which are most efficiently isolated in the marmoset B cell line B95a, cannot grow in many CD46+ cell lines. Although some researchers attributed it to post-entry block in viral replication, others believed that there is a receptor other than CD46 for wild-type measles viruses. A new study showed that human signalling lymphocytic activation molecule (SLAM; also known as CDw150) is a cellular receptor for measles virus, including the Edmonston strain. SLAM is expressed on lymphocytes and dendritic cells, and plays an important role in lymphocyte activation. The identification of SLAM as a measles virus receptor nicely explains the pathogenesis of measles virus infection.     

The hemagglutinin of recent measles virus isolates induces cell fusion in a marmoset cell line, but not in other CD46-positive human and monkey cell lines, when expressed together with the F protein

Summary.  Measles virus (MV) is efficiently isolated from patients with measles by using B95a cells, a marmoset B cell line. Recent wild-type MV strains isolated using B95a cells did not produce cytopathic effects in any of CD46⁺ primate cell lines examined (except B95a cells), nor did they induce downregulation of CD46. Transfection of the hemagglutinin (H) and fusion (F) genes of the Edmonston strain of MV produced syncytia in HeLa, Cos and B95a cells. By contrast, the expression of the H gene from the two wild-type strains, together with the F gene of the Edmonston strain, resulted in syncytium production in B95a cells, but not in HeLa and Cos cells. Cocultivation of Cos cells expressing the wild-type H protein and the Edmonston strain F protein with B95a cells, but not with HeLa, Jurkat or BJAB cells, generated large syncytia. The results suggest that these recent MV isolates may use a molecule other than CD46 as the cellular receptor or require another coreceptor to infect cells. Accepted October 15, 1997 Received July 18, 1997     

Measles virus interacts with human SLAM receptor on dendritic cells to cause immunosuppression

Measles virus (MV) infects dendritic cells (DCs) resulting in immunosuppression. Human DCs express two MV receptors: CD46 and human signaling lymphocyte activation molecule (hSLAM); thus, the role played by either alone is unclear. Because wild-type (wt) MV uses hSLAM receptor preferentially, we dissected the molecular basis of MV-DC interaction and resultant immunosuppression through the hSLAM receptor by creating transgenic (tg) mice expressing hSLAM on DCs. After infection with wt MV, murine splenic DCs expressing hSLAM receptor had less B7-1, B7-2, CD40, MHC class I, and MHC class II molecules on their surfaces and displayed an increased rate of apoptosis when compared to uninfected DCs. Further, MV-infected DCs failed to stimulate allogeneic T cells and inhibited mitogen-dependent T-cell proliferation. Individual expression of human SLAM, interferon alpha/beta receptor, tumor necrosis factor-alpha, and lymphotoxin-alpha or beta from T cells was not required for MV-infected DCs to inhibit the proliferation of T cells.     

Evidence that the hypermutated M protein of a subacute sclerosing panencephalitis measles virus actively contributes to the chronic progressive CNS disease

Subacute sclerosing panencephalitis (SSPE) is a progressive degenerative disease of the brain uniformly leading to death. Although caused by measles virus (MV), the virus recovered from patients with SSPE differs from wild-type MV; biologically SSPE virus is defective and its genome displays a variety of mutations among which biased replacements of many uridine by cytidine resides primarily in the matrix (M) gene. To address the question of whether the SSPE MVs with M mutations are passive in that they are not infectious, cannot spread within the CNS, and basically represent an end-stage result of a progressive infection or alternatively SSPE viruses are infectious, and their mutations enable them to persist and thereby cause a prolonged neurodegenerative disease, we utilized reverse genetics to generate an infectious virus in which the M gene of MV was replaced with the M gene of Biken strain SSPE MV and inoculated the recombinant virus into transgenic mice bearing the MV receptor. Our results indicate that despite biased hypermutations in the M gene, the virus is infectious in vivo and produces a protracted progressive infection with death occurring as long as 30 to 50 days after that caused by MV. In primary neuron cultures, the mutated M protein is not essential for MV replication, prevents colocalization of the viral N with membrane glycoproteins, and is associated with accumulation of nucleocapsids in cells' cytoplasm and nucleus.     

Structural and functional studies of the measles virus hemagglutinin: identification of a novel site required for CD46 interaction

The entry of measles virus (MV) into human cells is mediated by the initial attachment of the viral hemagglutinin (HA) to the complement regulatory protein CD46. Two subdomains, one each within CD46 short consensus repeats (SCRs) 1 and 2, are responsible for this interaction. However, little is known about the regions within MV HA needed for a high-affinity CD46 interaction. To better define the HA-CD46 interaction, we took three approaches: chimeric domain swapping, peptide scanning, and alanine scanning mutagenesis. Chimeras of MV HA and the closely related rinderpest virus (RPV) HA were generated and tested for cell surface expression and the ability to hemadsorb CD46+ red blood cells (RBC). Exchanges with the N terminus of RPV were tolerated as MV HA could be replaced with RPV HA up to amino-acid position 154. However, both larger swaps with RPV and a small RPV HA replacement at the C terminus aborted cell-surface expression. Peptide scanning with 51 overlapping peptides derived from three MV HA regions showed one peptide, corresponding to MV HA amino acids 468-487, blocked hemagglutination of African green monkey (AGM) RBCs and inhibited MV infection of Chinese hamster ovary cells (CHO) expressing human CD46. Alanine scanning mutants mapped sites on the MV HA that were not required for trafficking to the cell surface or function in hemagglutination as well as a novel site required for CD46 interaction, amino acids 473-477.     

Measles virus-induced down-regulation of CD46 is associated with enhanced sensitivity to complement-mediated lysis of infected cells

CD46, the major component of the measles virus (MV) receptor complex and a member of the regulators of complement activity (RCA) gene cluster, is down-regulated in MV-infected cells. We investigated whether the reduction of surface CD46 correlates with enhanced sensitivity of lymphoid and monocytic cells to lysis by activated complement. On human U937 cells, acutely or persistently infected with MV-Edmonston (ED) vaccine strain, infection-dependent down-regulation of CD46 confers sensitivity to activated complement, regardless of the pathway of activation and the specificity of the activating antibodies. Interestingly, down-regulation of CD46 alone is sufficient to confer susceptibility of cells to complement lysis despite the continued surface expression of other RCA proteins such as CD35 and CD55. In primary cultures, both peripheral blood lymphocytes and macrophages are efficiently lysed in the presence of complement activated via the alternative pathway after MV infection. In contrast to the MV-ED infection, infection of cells with the lymphotropic MV wild-type strain WTF does not down-regulate CD46. Cells infected with MV-WTF do not exhibit enhanced susceptibility to complement lysis. These data suggest that MV strains similar to WTF that do not down-regulate CD46 may have an enhanced potential for replication and dissemination within the human host, whereas complement-mediated elimination of cells infected with CD46-down-regulating strains of MV, such as ED, may limit the spread of MV infection, and could thus represent an attenuating factor for MV.     

Alterations and diversity in the cytoplasmic tail of the fusion protein of subacute sclerosing panencephalitis virus strains isolated in Osaka,Japan

We determined the nucleotide sequence of the fusion (F) gene of three strains (Osaka-1, -2, and -3) of nonproductive variants of measles virus (MV). These viral strains were isolated in Osaka, Japan, from brain tissues of patients with subacute sclerosing panencephalitis (SSPE). Phylogenetic analysis revealed a close relationship among the three strains of SSPE virus. The cytoplasmic tail of the F protein, predicted from sequence analysis of the gene, is altered in all three SSPE strains when compared to the MV field strains. However, the extent and mode of alteration are different in each strain. The F protein of the Osaka-1 strain has six nonconservative amino acid substitutions and a 29-residue elongation of its cytoplasmic tail. The F protein of the Osaka-3 strain has two nonconservative substitutions and a 5-residue truncation of its C-terminus. Although the termination codon is not altered in the F protein of the Osaka-2 strain, five or six amino acids are changed in the cytoplasmic tail of the F protein of the two sibling viruses of this strain. The significance of the altered cytoplasmic domain of the SSPE viruses in the SSPE pathogenesis is discussed.     

Detection of measles virus genomic sequences in SSPE brain tissue by the polymerase chain reaction

The polymerase chain reaction (PCR) was modified to detect RNA genomic sequences by generating cDNA copies of these sequences as a preliminary step. Oligonucleotide primer pairs complementary to sequences in each of the five major structural protein genes of the measles virus (nucleocapsid protein, phosphoprotein, matrix protein, fusion protein, and hemagglutinin protein) were synthesized. PCR products were tentatively identified by visualization of bands of the appropriate size by ethidium bromide staining after gel electrophoresis, and identity was confirmed by subsequent restriction enzyme cleavage of the products at predetermined sites to yield fragments of predicted size. This method successfully amplified 400-500 base regions from each of these five genes in RNA extracts of wild measles virus cultured in Vero cells and in RNA extracted from most of the SSPE brain tissues tested, but not in RNA from any control brain tissues. Measles virus genome was detected in SSPE brain tissues stored frozen for as long as 27 years and formalin-fixed paraffin-embedded subacute sclerosing panencephalitis (SSPE) brain tissues as old as 9 years. This method provides a simple, rapid and highly sensitive means of detecting and identifying sequences of RNA genomes by PCR. The success of this method in detecting measles virus in SSPE brain tissue suggests that PCR is appropriate to investigate the possible presence of RNA viruses in other neurological disorders of unknown etiology.     

Molecular analysis of measles virus genome derived from SSPE and acute measles patients in Papua,New Guinea

A very high annual incidence of 56 per million population below the age of 20 years for subacute sclerosing panencephalitis (SSPE) has been reported from Papua New Guinea (PNG). In a more recent study, we have confirmed this unusual high incidence for Eastern Highlands Province (EHP) of PNG. In the study, it was observed that the vaccination rate among SSPE patients registered at Goroka Base General Hospital (GBGH) in EHP was higher than that of other infants in the province in recent years. To identify the measles virus (MV) responsible for SSPE in EHP, sequence analysis of hypervariable region of the N gene was performed from 13 MV genomes: 2 amplified from clinical specimens of SSPE patients and 11 from acute measles patients. In 2 cases among the 11 with acute measles, nucleotide sequence of the entire H gene derived from isolated viruses was determined. Both nucleotide sequence and phylogenetic tree analyses showed that the amplified MV cDNAs were closely related to one another and belonged to the D3 genotype though they were different from any previously reported MV sequences. No genome sequences of vaccine strains were detected. These findings suggest that the MV strains prevailing in the highlands of PNG belong to genotype D3 of the MV and this wild-type MV rather than the vaccine strains was likely to be responsible for SSPE in these patients.     

Restricted expression of measles virus in primary rat astroglial cells

Persistent infection of the central nervous system (CNS) with measles virus (MV) is associated with characteristic restrictions of viral envelope gene expression as documented in subacute sclerosing panencephalitis (SSPE), measles inclusion body encephalitis (MIBE), or subacute measles encephalitis (SAME) in rats. To determine whether these restrictions are the result of a long lasting virus-host cell interaction or primarily based on intrinsic brain cell factors MV gene expression was analyzed in primary rat astroglial cultures. It could be shown that MV infection of these cells led to a defective replication cycle with a reduced synthesis of viral envelope proteins and a steep expression gradient of the monocistronic viral mRNAs similar to the findings in brain tissue of SSPE, MIBE, and SAME. This restriction of MV gene expression has not been observed in cells of nonneural origin. We suggest that this cell-type specific regulation of MV gene expression contributes to early events in the establishment of MV persistent infection in CNS tissue.     

Expression of CD34 and bcl-2 in phyllodes tumours, fibroadenomas and spindle cell lesions of the breast

AIMS: Strong expression of CD34 and bcl-2 has been described in solitary fibrous tumours. It has been proposed that these lesions arise from long-lived mesenchymal cells. We tested the hypothesis that spindle cell lesions of the breast arise from similar mesenchymal cells in the mammary stroma. METHODS and RESULTS: Sections of phyllodes tumours (26), fibroadenomas (15), myofibroblastomas (two), pseudoangiomatous hyperplasia (five) and myoid hamartoma (one) were stained immunohistochemically for CD34 and bcl-2. Conventional mammary carcinoma is known to be CD34-negative: we therefore stained 11 spindle cell carcinomas and one adenosquamous carcinoma. The mammary stroma, particularly around lobules, stained for CD34. All the lesions (except the carcinomas) showed spindle cell staining for CD34. There was more staining in fibroadenomas than in phyllodes tumours (especially malignant tumours). The staining in phyllodes tumours was typically patchy. In some there was increased or decreased staining adjacent to the epithelium. There were occasional spindle cells positive for bcl-2 in the normal perilobular stroma. bcl-2 was frequently expressed in spindle cells in fibroadenomas, phyllodes tumours and pseudoangiomatous hyperplasia, and rarely in the other lesions. CONCLUSIONS: The combined expression of CD34 and bcl-2 suggests that fibroadenomas, phyllodes tumours and pseudoangiomatous hyperplasia may arise from long-lived bcl-2-positive mesenchymal cells in the breast in a manner similar to that proposed for solitary fibrous tumours. The absence of CD34 staining in spindle cell carcinomas is of potential diagnostic value in the distinction from malignant phyllodes tumours in difficult cases.     

Tetraspanins in the immune response against cancer

The SLAM (CD150) family receptors are leukocyte cell-surface glycoproteins involved in leukocyte activation. These molecules and their adaptor protein SAP contribute to the effective germinal center formation, generation of high-affinity antibody-secreting plasma cells, and memory B cells, thereby facilitating long-term humoral immune response. Multi-color flow cytometric analysis was performed to determine the expression of CD48 (SLAMF2), CD84 (SLAMF5), CD150 (SLAM or SLAMF1), CD229 (Ly9 or SLAMF3), CD244 (2B4 or SLAMF4), CD319 (CRACC, CS1, or SLAMF7), and CD352 (NTB-A or SLAMF6) on human cell lines and B-cell subsets. The following subsets were assessed: pro-B, pre-B, immature-B, and mature-B cells from bone marrow; transitional and B1/B2 subsets from peripheral blood; and na?ve, pre-germinal center, germinal center, memory, plasmablasts, and plasma cells from tonsil and spleen. All receptors were expressed on B cells, with the exception of CD244. SLAM family molecules were widely distributed during B-cell development, maturation and terminal differentiation into plasmablasts and plasma cells, but their expression among various B-cell subsets differed significantly. Such heterogeneous expression patterns suggest that SLAM molecules play an essential and non-redundant role in the control of humoral immune responses.