The endoplasmic reticulum as a protein-folding compartment

The lumen of the endoplasmic reticulum (ER) provides a dynamic and efficient environment for the folding of proteins destined for secretion and for a variety of cellular compartments and membranes. Usually, the folding process begins on the nascent chains and is completed minutes or hours later during assembly of oligomers. It is assisted by molecular chaperones and folding enzymes, some of which are unique to the ER. Quality control and selective degradation systems ensure only conformationally mature proteins are transported from the ER.     

Variant influenza virus hemagglutinin that induces fusion at elevated pH.

The hemagglutinin (HA) glycoprotein of influenza virus performs two critical roles during infection: it binds virus to cell surface sialic acids, and under mildly acidic conditions it induces fusion of the virion with intracellular membranes, liberating the genome into the cytoplasm. The pH dependence of fusion varies for different influenza virus strains. Here we report the isolation and characterization of a naturally occurring variant of the X31 strain that fuses at a pH 0.2 units higher than the parent strain does and that is less sensitive to the effects of ammonium chloride, a compound known to elevate endosomal pH. The bromelain-solubilized ectodomain of the variant HA displayed a corresponding shift in the pH at which it changed conformation and bound to liposomes. Cloning and sequencing of the variant HA gene revealed amino acid substitutions at three positions in the polypeptide. Two substitutions were in antigenic determinants in the globular region of HA1, and the third occurred in HA2 near the base of the molecule. By using chimeric HA molecules expressed in CV-1 cells from simian virus 40-based vectors, we demonstrated that the change in HA2 was solely responsible for the altered fusion phenotype. This substitution, asparagine for aspartic acid at position 132, disrupted a highly conserved interchain salt bridge between adjacent HA2 subunits. The apparent role of this residue in stabilizing the HA trimer is consistent with the idea that the trimer dissociates at low pH. Furthermore, the results demonstrate that influenza virus populations contain fusion variants, raising the possibility that such variants may play a role in the evolution of the virus.     

An efficient method for introducing macromolecules into living cells

The hemagglutinin (HA) of influenza virus was used to obtain efficient and rapid bulk delivery of antibodies and horseradish peroxidase (HRP) into the cytoplasm of living tissue culture cells. By exploiting HA's efficient cell surface expression, its high affinity for erythrocytes, and its acid-dependent membrane fusion activity, a novel delivery method was developed. The approach is unique in that the mediator of both binding and fusion (the HA) is present on the surfaces of the target cells. A recently developed 3T3 cell line which permanently expresses HA, Madin-Darby canine kidney cells infected with influenza virus, and CV-1 cells infected with a simian virus 40 vector carrying the HA gene were used as recipient cells. Protein-loaded erythrocytes were bound to the HA on the cell surface and a brief drop in pH to 5.0 was used to trigger HA's fusion activity and hence delivery. About 3 to 8 erythrocytes fused per 3T3 and CV-1 cell, respectively, and 75-95% of the cells received IgG or HRP. Quantitative analysis showed that 1.8 X 10(8) molecules of HRP and 1.4 X 10(7) IgG molecules were delivered per CV-1 cell and 6.2 X 10(7) HRP molecules per 3T3 cell. Cell viability, as judged by methionine incorporation into protein and cell growth and division, was not impaired. Electron and fluorescence microscopy showed that the fused erythrocyte membranes remained as discrete domains in the cell's plasma membrane. The method is simple, reliable, and nonlytic. The ability to simultaneously and rapidly deliver impermeable substances into large numbers of cells will permit biochemical analysis of the fate and effect of a variety of delivered molecules.     

Intermediates in influenza induced membrane fusion.

Our results show that the mechanism by which influenza virus fuses with target membranes involves sequential complex changes in the hemagglutinin (HA, the viral fusion protein) and in the contact site between virus and target membrane. To render individual steps amenable to study, we worked at 0 degree C which decreased the rate of fusion and increased the efficiency. The mechanism of fusion at 0 degree C and 37 degrees C was similar. The process began with a conformational change in HA which exposed the fusion peptides but did not lead to dissociation of the tops of the ectodomain of the trimer. The change in the protein led to immediate hydrophobic attachment of the virus to the target liposomes. Attachment was followed by a lag period (4-8 min at 0 degree C, 0.6-2 s at 37 degrees C) during which rearrangements occurred in the site of membrane contact between the virus and liposome. After a further series of changes the final bilayer merger took place. This final fusion event was not pH dependent. At 0 degree C efficient fusion occurred without dissociation of the top domains of the HA trimer, suggesting that a transient conformation of HA is responsible for fusion at physiological temperatures. The observations lead to a revised model for HA mediated fusion.     

N-linked oligosaccharides are necessary and sufficient for association of glycosylated forms of bovine RNase with calnexin and calreticulin.

Calnexin and calreticulin are lectin-like molecular chaperones that promote folding and assembly of newly synthesized glycoproteins in the endoplasmic reticulum. While it is well established that they interact with substrate monoglucosylated N-linked oligosaccharides, it has been proposed that they also interact with polypeptide moieties. To test this notion, glycosylated forms of bovine pancreatic ribonuclease (RNase) were translated in the presence of microsomes and their folding and association with calnexin and calreticulin were monitored. When expressed with two N-linked glycans in the presence of micromolar concentrations of deoxynojirimycin, this small soluble protein was found to bind firmly to both calnexin and calreticulin. The oligosaccharides were necessary for association, but it made no difference whether the RNase was folded or not. This indicated that unlike other chaperones, calnexin and calreticulin do not select their substrates on the basis of folding status. Moreover, enzymatic removal of the oligosaccharide chains using peptide N-glycosidase F or removal of the glucoses by ER glucosidase II resulted in dissociation of the complexes. This indicated that the lectin-like interaction, and not a protein-protein interaction, played the central role in stabilizing RNase-calnexin/calreticulin complexes.     

Calnexin and calreticulin promote folding, delay oligomerization and suppress degradation of influenza hemagglutinin in microsomes.

Calnexin (CNX) and calreticulin (CRT) are molecular chaperones that bind preferentially to monoglucosylated trimming intermediates of glycoproteins in the endoplasmic reticulum. To determine their role in the maturation of newly synthesized glycoproteins, we analyzed the folding and trimerization of in vitro translated influenza hemagglutinin (HA) in canine pancreas microsomes under conditions in which HA's interactions with CNX and CRT could be manipulated. While CNX bound to all folding intermediates (IT1, IT2 and NT), CRT was found to associate preferentially with the earliest oxidative form (IT1). If HA's binding to CNX and CRT was inhibited using a glucosidase inhibitor, castanospermine (CST), the rate of disulfide formation and oligomerization was doubled but the overall efficiency of maturation of HA decreased due to aggregation and degradation. If, on the other hand, HA was arrested in CNX-CRT complexes, folding and trimerization were inhibited. This suggested that the action of CNX and CRT, like that of other chaperones, depended on an 'on-and-off' cycle. Taken together, these results indicated that CNX and CRT promote correct folding by inhibiting aggregation, preventing premature oxidation and oligomerization, and by suppressing degradation of incompletely folded glycopolypeptides.     

Visual performance of the toad (Bufo bufo) at low light levels: retinal ganglion cell responses and prey-catching accuracy

The accuracy of toad snapping towards moving worm dummies under various levels of dim illumination (from absolute threshold to moonlight) was videorecorded and related to spike responses of retinal ganglion cells exposed to equivalent stimuli. Some toads (at ca. 16 °C) successfully snapped at dummies that produced only one photoisomerization per 50 rods per second in the retina, in good agreement with thresholds of sensitive retinal ganglion cells. One factor underlying such high sensitivity is extensive temporal summation by the ganglion cells. This, however, is inevitably accompanied by very long response latencies (around 3 s near threshold), whereby the information reaching the brain shows the dummy in a position where it was several seconds earlier. Indeed, as the light was dimmed, snaps were displaced successively further to the rear of the dummy, finally missing it. The results in weak but clearly supra-threshold illumination indicate that snaps were aimed at the advancing head as seen by the brain, but landed further backwards in proportion to the retinal latency. Near absolute threshold, however, accuracy was too good, suggesting that the animal had recourse to a neural representation of the regularly moving dummies to correct for the slowness of vision.     

Hyperphosphorylation of mutant influenza virus matrix protein, M1, causes its retention in the nucleus.

The matrix (M1) protein of influenza virus is a major structural component, involved in regulation of viral ribonucleoprotein transport into and out of the nucleus. Early in infection, M1 is distributed in the nucleus, whereas later, it is localized predominantly in the cytoplasm. Using immunofluorescence microscopy and the influenza virus mutant ts51, we found that at the nonpermissive temperature M1 was retained in the nucleus, even at late times after infection. In contrast, the viral nucleoprotein (NP), after a temporary retention in the nucleus, was distributed in the cytoplasm. Therefore, mutant M1 supported the release of the viral ribonucleoproteins from the nucleus, but not the formation of infectious virions. The point mutation in the ts51 M1 gene was predicted to encode an additional phosphorylation site. We observed a substantial increase in the incorporation of 32Pi into M1 at the nonpermissive temperature. The critical role of this phosphorylation site was demonstrated by using H89, a protein kinase inhibitor; it inhibited the expression of the mutant phenotype, as judged by M1 distribution in the cell. Immunofluorescence analysis of ts51-infected cells after treatment with H89 showed a wild-type phenotype. In summary, the data indicated that the ts51 M1 protein was hyperphosphorylated at the nonpermissive temperature and that this phosphorylation was responsible for its aberrant nuclear retention.     

Characterization and cloning of lgp110, a lysosomal membrane glycoprotein from mouse and rat cells

lgp110 is a heavily glycosylated intrinsic protein of lysosomal membranes. Initially defined by monoclonal antibodies against mouse liver lysosomes, it consists of a 45-kilodalton core polypeptide with O-linked and 17 asparagine-linked oligosaccharide side chains in mouse cells. Sialic acid residues make the mature protein extremely acidic, with an isoelectric point of between 2 and 4 in both normal tissues and most cultured cell lines. Partial sequencing of mouse lgp110 allowed oligonucleotide probes to be constructed for the screening of several mouse cDNA libraries. A partial cDNA clone for mouse lgp110 was found and used for additional library screening, generating a cDNA clone covering all of the coding sequence of mature rat lgp110 as well as genomic clones covering most of the mouse gene. These new clones bring to seven the number of lysosomal membrane proteins whose amino acid sequences can be deduced, and two distinct but highly similar groups (designated lgp-A and lgp-B) can now be defined. Sequence comparisons suggest that differences within each group reflect species variations of the same protein and that lgp-A and lgp-B probably diverged from a common ancestor prior to the evolup4f1ary divergence of birds and mammals. Individual cells and individual lysosomes possess both lgp-A and lgp-B, suggesting that these two proteins have different functions. Mouse lgp110 is encoded by at least seven exons; intron positions suggest that the two homologous ectodomains of each lgp arose through gene duplication.