Basolateral sorting signals differ in their ability to redirect apical proteins to the basolateral cell surface.

Polarized sorting of membrane proteins in epithelial cells is mediated by cytoplasmic basolateral signals or by apical signals in the transmembrane or exoplasmic domains. Basolateral signals were generally found to be dominant over apical determinants. We have generated chimeric proteins with the cytoplasmic domain of either the asialoglycoprotein receptor H1 or the transferrin receptor, two basolateral proteins, fused to the transmembrane and exoplasmic segments of aminopeptidase N, an apical protein, and analyzed them in Madin-Darby canine kidney cells. Whereas both cytoplasmic sequences induced endocytosis of the chimeras, only that of the transferrin receptor mediated basolateral expression in steady state. The H1 fusion protein, although still largely sorted to the basolateral side in biosynthetic surface transport, was subsequently resorted to the apical cell surface. We tested whether the difference in sorting between trimeric wild-type H1 and the dimeric aminopeptidase chimera was caused by the number of sorting signals presented in the oligomers. Consistent with this hypothesis, the H1 signal was fully functional in a tetrameric fusion protein with the transmembrane and exoplasmic domains of influenza neuraminidase. The results suggest that basolateral signals per se need not be dominant over apical determinants for steady-state polarity and emphasize an important contribution of the valence of signals in polarized sorting.

Multiple determinants direct the orientation of signal-anchor proteins: the topogenic role of the hydrophobic signal domain.

The orientation of signal-anchor proteins in the endoplasmic reticulum membrane is largely determined by the charged residues flanking the apolar, membrane-spanning domain and is influenced by the folding properties of the NH2-terminal sequence. However, these features are not generally sufficient to ensure a unique topology. The topogenic role of the hydrophobic signal domain was studied in vivo by expressing mutants of the asialoglycoprotein receptor subunit H1 in COS-7 cells. By replacing the 19-residue transmembrane segment of wild-type and mutant H1 by stretches of 7-25 leucine residues, we found that the length and hydrophobicity of the apolar sequence significantly affected protein orientation. Translocation of the NH2 terminus was favored by long, hydrophobic sequences and translocation of the COOH terminus by short ones. The topogenic contributions of the transmembrane domain, the flanking charges, and a hydrophilic NH2-terminal portion were additive. In combination these determinants were sufficient to achieve unique membrane insertion in either orientation.

The oligomerization domain of the asialoglycoprotein receptor preferentially forms 2:2 heterotetramers in vitro.

The human hepatic asialoglycoprotein receptor is a noncovalent hetero-oligomer composed of two homologous subunits, H1 and H2, with an as yet unknown stoichiometry. Ligand specificity and binding affinity depend on the arrangement of the subunits in the complex. An 80-amino acid segment connecting the transmembrane and the carbohydrate binding domains contains heptad repeats characteristic of alpha-helical coiled coil structure. We expressed and purified corresponding peptides, H1S and H2S, and confirmed by circular dichroism spectroscopy that they can assume alpha-helical conformation. Oxidative cross-linking of amino-terminal cysteines generated specific covalent oligomers, indicating that separately H1S forms trimers and H2S tetramers. Upon mixing, covalent heterotetramers were formed with a preferred stoichiometry of 2 H1S and 2 H2S peptides. These results suggest that the stalk segments of the receptor subunits oligomerize to constitute an alpha-helical coiled coil stalk on top of which the carbohydrate binding domains are exposed for ligand binding. We propose that the functional asialoglycoprotein receptor is a 2:2 heterotetramer.

trans-Golgi retention of a plasma membrane protein: mutations in the cytoplasmic domain of the asialoglycoprotein receptor subunit H1 result in trans-Golgi retention.

Unlike the wild-type asialoglycoprotein receptor subunit H1 which is transported to the cell surface, endocytosed and recycled, a mutant lacking residues 4-33 of the 40-amino acid cytoplasmic domain was found to be retained intracellularly upon expression in different cell lines. The mutant protein accumulated in the trans-Golgi, as judged from the acquisition of trans-Golgi-specific modifications of the protein and from the immunofluorescence staining pattern. It was localized to juxtanuclear, tubular structures that were also stained by antibodies against galactosyltransferase and gamma-adaptin. The results of further mutagenesis in the cytoplasmic domain indicated that the size rather than the specific sequence of the cytoplasmic domain determines whether H1 is retained in the trans-Golgi or transported to the cell surface. Truncation to less than 17 residues resulted in retention, and extension of a truncated tail by an unrelated sequence restored surface transport. The transmembrane segment of H1 was not sufficient for retention of a reporter molecule and it could be replaced by an artificial apolar sequence without affecting Golgi localization. The cytoplasmic domain thus appears to inhibit interaction(s) of the exoplasmic portion of H1 with trans-Golgi component(s) for example by steric hindrance or by changing the positioning of the protein in the membrane. This mechanism may also be functional in other proteins.

The oligomerization reaction of the Semliki Forest virus membrane protein subunits.

The Semliki Forest virus (SFV) spike is composed of three copies of a membrane protein heterodimer. The two subunits of this heterodimer (p62 and E1) are synthesized sequentially from a common mRNA together with the capsid (C) in the order C-p62-E1. In this work heterodimerization of the spike proteins has been studied in BHK 21 cells. The results indicate that: (a) the polyprotein is cotranslationally cleaved into individual chains; (b) the two membrane protein subunits are initially not associated with each other in the endoplasmic reticulum (ER); (c) heterodimerization occurs predominantly between subunits that originate from the same translation product (heterodimerization in cis); (d) the kinetics of subunit association are very fast (t1/2 = 4 min); and (e) this heterodimerization is highly efficient. To explain the cis-directed heterodimerization reaction we suggest that the p62 protein, which is made before E1 during 26S mRNA translation, is retained at its translocation site until also the E1 chain has been synthesized and translocated at this same site. The mechanism for p62 retention could either be that the p62 anchor sequence cannot diffuse out from an "active" translocation site or that the p62 protein is complexed with a protein folding facilitating machinery that is physically linked to the translocation apparatus.

Assembly and entry mechanisms of Semliki Forest virus.

The alphavirus Semliki Forest (SFV) is an enveloped virus with a positive single-stranded RNA genome. The genome is complexed with 240 copies of a capsid protein into a nucleocapsid structure. In the membrane the virus carries an equal number of copies of a membrane protein heterodimer. The latter oligomers are grouped into clusters of three. These structures form the spikes of the virus and carry its entry functions, that is receptor binding and membrane fusion activity. The membrane protein heterodimer is synthesized as a p62E1 precursor protein which upon transport to the cell surface is cleaved into the mature E2E1 form. Recent studies have given much new information on the assembly and entry mechanism of this simple RNA virus. Much of this work has been possible through the construction of a complete cDNA clone of the SFV genome which can be used for in vitro transcription of infectious RNA. One important finding has been to show that a spike deletion variant and a capsid protein deletion variant are budding-negative when expressed separately but can easily complement each other when transfected into the same cell. This shows clearly that enveloped viruses use different budding strategies: one which depends on a nucleocapsid-spike interaction as exemplified by SFV and another one which is based on a direct core-lipid bilayer interaction as shown before to be the case with retroviruses. Another important finding concerns the activation process of the presumed fusion protein of SFV, the E1 subunit. In the original p62E1 heterodimer E1 is completely inactive. Activation proceeds in several steps. First p62 cleavage activates the potential for low pH inducible fusion. Next the low pH which surrounds incoming virus in endosomes induces dissociation of the heterodimeric structure. This is followed by a rearrangement of E1 subunits into homotrimers which are fusion active.

CD13 (human aminopeptidase N) mediates human cytomegalovirus infection.

Human cytomegalovirus (HCMV) infects cells by a series of processes including attachment, penetration via fusion of the envelope with the plasma membrane, and transport of the viral DNA to the nucleus. The details of the early events of HCMV infection are poorly understood. We have recently reported that CD13, human aminopeptidase N, a metalloprotease, is present on blood cells susceptible in vitro to HCMV infection (C. Söderberg, S. Larsson, S. Bergstedt-Lindqvist, and E. Möller, J. Virol. 67:3166-3175, 1993). Here we report that human CD13 is involved in HCMV infection. Antibodies directed against human CD13 not only inhibit infection but also block binding of HCMV virions to susceptible cells. Compounds known to inhibit aminopeptidase activity block HCMV infection. HCMV-resistant murine fibroblasts have heightened susceptibility to HCMV infection after transfection with complementary DNA encoding human CD13. A significant increase in binding of HCMV was observed in the CD13-expressing transfectants compared with neomycin-resistant control mouse cells. However, murine fibroblasts transfected with mutant CD13, lacking a portion of the aminopeptidase active site, remained susceptible to HCMV infection. Thus, human CD13 appears to mediate HCMV infection by a process that increases binding, but its enzymatic domain is not necessary for infection.

Membrane fusion of Semliki Forest virus in a model system: correlation between fusion kinetics and structural changes in the envelope glycoprotein.

This paper presents a kinetic analysis of low-pH-induced fusion of Semliki Forest virus (SFV) with cholesterol-containing unilamellar lipid vesicles (liposomes), consisting otherwise of phosphatidylcholine, phosphatidylethanolamine and sphingomyelin. Fusion is monitored continuously with a lipid mixing assay, involving virus bio-synthetically labeled with the fluorophore pyrene. At pH 5.55, 37 degrees C, SFV-liposome fusion occurs on the time scale of seconds. Extensive fusion (up to 60% of the virus) requires an excess of liposomes, while a low-pH preincubation of the virus alone results in inactivation of its fusion capacity. The onset of fusion after acidification of virus-liposome mixtures is preceded by a pH- and temperature-dependent lag phase. Early in this lag phase, a conformational change in the E2E1 spike glycoprotein occurs, involving formation of a trypsin-resistant E1 homotrimer, exposing a conformation-specific epitope (E1"). These changes are followed by a rapid, cholesterol-dependent binding of the virus to the liposomes (as assessed by sucrose density gradient analysis), subsequent fusion starting only after an additional delay. This sequence of events strongly suggests that the E1 homotrimeric structure represents the fusion-active conformation of the SFV spike, the actual fusion complex possibly involving a higher order oligomer of E1 trimers.

Membrane fusion of Semliki Forest virus involves homotrimers of the fusion protein.

Infection of cells with enveloped viruses is accomplished through membrane fusion. The binding and fusion processes are mediated by the spike proteins in the envelope of the virus particle and usually involve a series of conformational changes in these proteins. We have studied the low-pH-mediated fusion process of the alphavirus Semliki Forest virus (SFV). The spike protein of SFV is composed of three copies of the protein heterodimer E2E1. This structure is resistant to solubilization in mild detergents such as Nonidet P-40 (NP40). We have recently shown that the spike structure is reorganized during virus entry into acidic endosomes (J. M. Wahlberg and H. Garoff, J. Cell Biol. 116:339-348, 1992). The original NP40-resistant heterodimer is dissociated, and the E1 subunits form new NP40-resistant protein oligomers. Here, we show that the new oligomer is represented by an E1 trimer. From studies that use an in vitro assay for fusion of SFV with liposomes, we show that the E1 trimer is efficiently expressed during virus-mediated membrane fusion. Time course studies show that both E1 trimer formation and fusion are fast processes, occurring in seconds. It was also possible to inhibit virus binding and fusion with a monoclonal antibody directed toward the trimeric E1. These results give support for a model in which the E1 trimeric structure is involved in the SFV-mediated fusion reaction.

Membrane fusion process of Semliki Forest virus. I: Low pH-induced rearrangement in spike protein quaternary structure precedes virus penetration into cells.

The Semliki Forest virus (SFV) directs the synthesis of a heterodimeric membrane protein complex which is used for virus membrane assembly during budding at the surface of the infected cell, as well as for low pH-induced membrane fusion in the endosomes when particles enter new host cells. Existing evidence suggests that the E1 protein subunit carries the fusion potential of the heterodimer, whereas the E2 subunit, or its intracellular precursor p62, is required for binding to the nucleocapsid. We show here that during virus uptake into acidic endosomes the original E2E1 heterodimer is destabilized and the E1 proteins form new oligomers, presumably homooligomers, with altered E1 structure. This altered structure of E1 is specifically recognized by a monoclonal antibody which can also inhibit penetration of SFV into host cells as well as SFV-mediated cell-cell fusion, thus suggesting that the altered E1 structure is important for the membrane fusion. These results give further support for a membrane protein oligomerization-mediated control mechanism for the membrane fusion potential in alphaviruses.

Membrane fusion process of Semliki Forest virus. II: Cleavage-dependent reorganization of the spike protein complex controls virus entry.

The envelope of the Semliki Forest virus (SFV) contains two transmembrane proteins, E2 and E1, in a heterodimeric complex. The E2 subunit is initially synthesized as a precursor protein p62, which is proteolytically processed to the mature E2 form before virus budding at the plasma membrane. The p62 (E2) protein mediates binding of the heterodimer to the nucleocapsid during virus budding, whereas E1 carries the entry functions of the virus, that is, cell binding and low pH-mediated membrane fusion activity. We have investigated the significance of the cleavage event for the maturation and entry of the virus. To express SFV with an uncleaved p62 phenotype, BHK-21 cells were transfected by electroporation with infectious viral RNA transcribed from a full-length SFV cDNA clone in which the p62 cleavage site had been changed. The uncleaved p62E1 heterodimer was found to be used for the formation of virus particles with an efficiency comparable to the wild type E2E1 form. However, in contrast to the wild type virus, the mutant virus was virtually noninfectious. Noninfectivity resulted from impaired uptake into cells, as well as from the inability of the virus to promote membrane fusion in the mildly acidic conditions of the endosome. This inability could be reversed by mild trypsin treatment, which converted the viral p62E1 form into the mature E2E1 form, or by treating the virus with a pH 4.5 wash, which in contrast to the more mild pH conditions of endosomes, effectively disrupted the p62E1 subunit association. We conclude that the p62 cleavage is not needed for virus budding, but regulates entry functions of the E1 subunit by controlling the heterodimer stability in acidic conditions.

Two mutations in the envelope glycoprotein E2 of Semliki Forest virus affecting the maturation and entry patterns of the virus alter pathogenicity for mice.

The prototype strain of Semliki Forest virus (SFV) of known sequence and virus produced by the cDNA clone derived from it were lethal following intranasal (i.n.) infection of 40-day-old and intraperitoneal (i.p.) infection of pregnant BALB/c mice; this lethality was related to neuronal necrosis in the central nervous system (CNS). We conclude that the virulence of the prototype strain, and virus from the cDNA clone derived from it, is similar to that of L10 (the original SFV isolate). The effects of two mutations in the p62 envelope protein region of the clone were determined. Substitution of Glu for Lys at position 162 (mut64) extended the mean time of death following i.n. inoculation of 40-day-old mice. Pregnant mice infected with this virus survived but lethal infection of some fetuses did occur. Substitution of Leu for Arg at position 66 (mL), the cleavage site of the E2 and E3 proteins, results in the production of particles containing uncleaved p62. These particles were less virulent than the prototype strain when inoculated i.n. and induced immunity to virulent SFV challenge. The virus also induced the formation of multifocal glial nodules in the CNS of surviving mice. The differences in pathogenicity between the two mutants and the virulent parental virus are probably related to differences in the efficiency of virus multiplication in infected mice. The mut64 mutation attenuated the virus and allowed survival of pregnant mice infected i.p. so that the effects of fetal infection could be detected. The mL mutation allowed survival of i.n.-infected mice so that the later effects of virus multiplication in the CNS could be assessed. In the former case, this is probably a result of reduced virus release, whereas in the latter case it is due to inefficient entry of host cells. The results are consistent with our previous suggestion that lethality for virulent SFV infection results from a lethal threshold of damage to neurons in the CNS and that attenuating mutations may reduce neuronal damage below this threshold level.

Spike protein oligomerization control of Semliki Forest virus fusion.

We have recently shown, using cleavage-deficient mutants of the p62-E1 membrane protein complex of Semliki Forest virus that p62 cleavage to E2 is necessary for the activation of the fusion function of the complex at pH 5.8 (a pH optimal for virus fusion) (M. Lobigs and H. Garoff, J. Virol. 64:1233-1240, 1990). In this study, we show that the mutant precursor complexes can be induced to activate membrane fusion when treated with more acidic buffers (pH 5.0 and 4.5), which also appear to dissociate most of the p62-E1 complexes and change the conformation of the E1 subunit (the supposed fusion protein of Semliki Forest virus into a form which is resistant to trypsin digestion. These data suggest that p62 cleavage is not essential for membrane fusion per se but that the crucial event activating this process seems to be the apparent dissociation of the heterodimer, which in turn is facilitated by the spike precursor cleavage.

The heterodimeric association between the membrane proteins of Semliki Forest virus changes its sensitivity to low pH during virus maturation.

The budding and the fusion processes of the enveloped animal virus Semliki Forest virus serve the purpose of transporting its nucleocapsid, containing its genome, from the cytoplasm of an infected cell into that of an uninfected one. We show here that, in the infected cell, the viral membrane (spike) proteins p62 and E1 are organized as heterodimers which are very resistant to dissociation in acidic conditions. In contrast, the mature form of the heterodimer, E2E1, which is found in the virus particle and which is generated by proteolytic processing of p62, is very prone to dissociate upon treatment with mildly acidic buffers. We discuss the possibility that this difference in behavior of the intracellular precursor form and the mature form of the spike protein complex represents an important regulatory mechanism for the processes involving membrane binding around the nucleocapsid during budding and membrane release from the nucleocapsid at the stage of virus fusion.