Inhibition of vesicular stomatitis virus infection by spike glycoprotein. Evidence for an intracellular, G protein-requiring step.

In an assay measuring virus-directed RNA synthesis, infection of BHK cells by a standard test dose of vesicular stomatitis virus (VSV) was inhibited by ultraviolet light-irradiated wt VSV and by ts 045, one of a number of thermolabile, temperature-sensitive G protein mutants of VSV. After heat treatment for 1 h at 45 degrees C, the thermolabile mutants were no longer able to inhibit the VSV infection. In contrast, the thermolabile M protein mutant ts G31 and the nonthermolabile G protein mutant ts 044 could still inhibit the test VSV dose. Thus, the presence of G protein in its native conformation was necessary for inhibition of infection. There was little difference in the binding to cells or the internalization to a trypsin-resistant state of ts 045 or wt VSV before and after heat treatment, and there was no evidence of specific saturable receptors on the cell surface. None of the irradiated virions at concentrations that gave maximal inhibition of infection could prevent binding of infectious VSV to, or internalization by, BHK cells. The G protein-specific inhibition, therefore, did not occur at the cell surface but must have occurred at some intracellular site, which has been suggested to be the lysome. The lysosomal inhibitor chloroquine, when added with the infecting virus, completely inhibited VSV infection at all multiplicities of infection tested, and it gave 50% inhibition when added to 1.5 h after infection. The possible importance of the lysosome in the intracellular pathway of infection is discussed.

Alteration of the conformation of proteins in red blood cell membranes and in solution by fixatives used in electron microscopy.

The effects of several commonly employed fixatives on the three-dimensional conformations of two soluble proteins and the protein of intact red blood cell membranes have been studied by means of circular dichroism measurements in the spectral region of the peptide absorption bands. The fixatives used produced significant and parallel conformational changes in all of the proteins, in the increasing order: glutaraldehyde; OsO(4); glutaraldehyde followed by OsO(4); and KMnO(4). The last two treatments obliterated most of the helical character of the proteins. The significance of these observations to the preparation of specimens for electron microscopy is discussed.

The role of lysosomal enzymes in killing of mammalian cells by the lysosomotropic detergent N-dodecylimidazole.

The sensitivity of cultured human and hamster fibroblast cells to killing by the lysosomotropic detergent N-dodecylimidazole (C12-Im) was investigated as a function of cellular levels of general lysosomal hydrolase activity, and specifically of cysteine cathepsin activity. Fibroblasts from patients with mucolipidosis II (I-cell disease) lack mannose-6-phosphate-containing proteins, and therefore possess only 10-15% of the normal level of most lysosomal hydrolases. I-cell fibroblasts are about one-half as sensitive to killing by C12-Im as are normal human fibroblasts. Overall lysosomal enzyme levels of CHO cells were experimentally manipulated in several ways without affecting cell viability: Growth in the presence of 10 mM ammonium chloride resulted in a gradual decrease in lysosomal enzyme content to 10-20% of control values within 3 d. Subsequent removal of ammonium chloride from the growth medium resulted in an increase in lysosomal enzymes, to approximately 125% of control values within 24 h. Treatment with 80 mM sucrose caused extensive vacuolization within 2 h; lysosomal enzyme levels remained at control levels for at least 6 h, but increased 15-fold after 24 h of treatment. Treatment with concanavalin A (50 micrograms/ml) also caused rapid (within 2 h) vacuolation with a sevenfold rise in lysosomal enzyme levels occurring only after 24 h. The sensitivity of these experimentally manipulated cells to killing by C12-Im always paralleled the measured intracellular lysosomal enzyme levels: lower levels were associated with decreased sensitivity while higher levels were associated with increased sensitivity, regardless of the degree of vacuolization of the cells. The cytotoxicity of the cysteine proteases (chiefly cathepsin L in our cells) was tested by inactivating them with the irreversible inhibitor E-64 (100 micrograms/ml). Cell viability, protein levels, and other lysosomal enzymes were unaffected, but cysteine cathepsin activity was reduced to less than 20% of control values. E-64-treated cells were almost completely resistant to C12-Im treatment, although lysosomal disruption appeared normal by fluorescent visualization of Lucifer Yellow CH-loaded cells. It is concluded that cysteine cathepsins are the major or sole cytotoxic agents released from lysosomes by C12-Im. These observations also confirm the previous conclusions that C12-Im kills cells as a consequence of lysosomal disruption.

Reconstituted G protein-lipid vesicles from vesicular stomatitis virus and their inhibition of VSV infection.

The single glycoprotein (G) of vesiclar stomatitis virus (VSV) was isolated in nearly quantitative yield by extraction of the purified virions with 0.05 M octyl-beta-D- glucoside (OG) in 0.01 M sodium phosphate, pH 8.0. The extract contained essentially all of the viral phospholipids and glycolipids, and was free of other essentially all of the viral phospholipids and glycolipids, and was free of other viral proteins. Dialysis to remove OG resulted in the formation of G protein-viral lipid vesicles having a lipid-G protein ratio similar to that of the intact virions. The vesicles were 250-1,000 A in diameter, with a "fuzzy" external layer also similar to that of intact virions. The vesicles were predominantly unilamellar and sealed, with both phosphatidyl ethanolamine and gangliosides symmetrically distributed in the bilayer. G protein was asymmetrically oriented, with about 80 percent accessible to exogenous protease. Addition of soybean phospholipid to the viral extract before dialysis resulted in vesicles that incorporated viral proteins and lipids quantitatively, but that were markedly decreased in buoyant density. The G neutralized protein-lipid vesicles were effective in eliciting specific anti-G antibodies that neutralized viral infectivity. Competitive radioimmunoassay showed that both reconstituted vesicles and a soluble form of G protein (Gs) were indistinguishable from purified VSV in their antibody binding properties. Addition of G protein-lipid vesicles of BHK-21 cells before, or simultaneously with, infection by VSV inhibited viral infectivity, as measured by two independent techniques (viral RNA production in the presence of actinomycin D and a neutral red assay of cell viability). The total inhibitory activity of G protein in the vesicular form was, however, less than 5 percent of that found for intact virus particles that have been inactivated by ultraviolet light irradiation. Gs was inactive as an inhibitor as determined by the RNA production assay.

Cell killing by lysosomotropic detergents.

We have studied the mechanism by which lysosomotropic detergents kill baby hamster kidney cells. Lysosomotropic detergents are lysosomotropic amines (compounds with pK between 5 and 9, such as imidazole or morpholine) containing straight-chain hydrocarbon "tails" of 9-14 carbon atoms (Firestone, R. A., J. M. Pisano, and R. J. Bonney. 1979, J. Med. Chem., 22:1130-1133). Using lucifer yellow CH as a specific fluorescent label for lysosomes, it was shown by light microscopy that N-dodecyl (C12)-imidazole acted rapidly to damage lysosomes, causing leakage of dye into the cytoplasm. This was followed at later times by vacuolization, blebbing of the plasma membrane, cell rounding, and cell death. 3H-labeled C12-imidazole rapidly diffused into cells where much of it was trapped in lysosomes as shown by its co-migration with lysosomes in Percoll gradients. Cells preincubated with C12-imidazole released it slowly into C12-imidazole-free media, permitting the cells to be killed by the preincubation dose. Cell killing by the lysosomotropic detergents exhibited strongly sigmoidal dose-response curves. The sensitivity of baby hamster kidney cells to killing by C12-imidazole was density dependent, the cells being most sensitive at lowest cell densities, and relatively resistant at confluence. The amount of 3H-C12-imidazole taken up by the cells was also density dependent, with highest specific uptake occurring at the lowest cell density. A rise in lysosomal pH, measured in fluoresceinated dextran-labeled cells, commenced immediately upon addition of C12-imidazole to cells, and continued for over an hour. This was followed after a lag of 1-2 h by inhibition of protein and RNA synthesis and by lactate dehydrogenase release. Ionophores or lysosomotropic amines, such as methylamine, that raise intralysosomal pH provided substantial protection of the cells from killing by lysosomotropic detergents. These findings provide strong support for the idea that lysosomotropic detergents kill cells by disrupting lysosomes from within.

Structure of membranes: reaction of red blood cell membranes with phospholipase C.

Treatment of human red blood cell membranes with phospholipase C releases 68 to 74 percent of the total membrane phosphorus into solution, through hydrolysis of membrane phospholipids to diglycerides and water-soluble phosphorylated amines. In spite of this drastic change, the membrane remains intact in phase microscopy, and the average protein conformation in the membranes, as determined by circular dichroism measurements in the ultraviolet, is unaffected. These results are readily explained by a model of membrane structure that is stabilized by hydrophobic interactions and in which the polar and ionic heads of lipids are on the outer surfaces of the membrane, in contact with the bulk aqueous phase and accessible to the action of phospholipase C.

Membrane asymmetry.

The components of biological membranes are asymmetrically distributed between the membrane surfaces. Proteins are absolutely asymmetrical in that every copy of a polypeptide chain has the same orientation in the membrane, and lipids are nonabsolutely asymmetrical in that almost every type of lipid is present on both sides of the bilayer, but in different and highly variable amounts. Asymmetry is maintained by lack of transmembrane diffusion. Two types of membrane proteins, called ectoproteins and endoproteins, are distinguished. Biosynthetic pathways for both types of proteins and for membrane lipids are inferred from their topography and distribution in the formed cells. Note added in proof. A cell-free system has now been developed which permits the mechanisms of membrane protein assembly to be studied (108). The membrane glycoprotein of vesicular stomatitis virus has been synthesized by wheat germ ribosomes in the presence of rough endoplasmic reticulum from pancreas. The resulting polypeptide is incorporated into the membrane, spans the lipid bilayer asymmetrically, and is glycosylated (108). The amino terminal portion of this transmembrane protein is found inside the endoplasmic reticulum vesicle, while the carboxyl terminal portion is exposed on the outer surface of the vesicle. Furthermore, addition of the glycoprotein to membranes after protein synthesis does not result in incorporation of the protein into the membrane in the manner described above (108). Consequently, protein synthesis and incorporation into the membrane must be closely coupled. Indeed, using techniques to synchronize the growth of nascent polypeptides, it has been shown (109) that no more than one-fourth of the glycoprotein chain can be made in the absence of membranes and still cross the lipid bilayer when chains are subsequently completed in the presence of membranes. These findings demonstrate directly that the extracytoplasmic portion of an ectoprotein can cross the membrane only during biosynthesis, and not after.

Mammalian hormones in microbial cells.

Hormones and hormone-binding proteins resembling those of vertebrates are widespread in fungi, yeast and bacteria. Functional responses of microbial cells to mammalian hormones have also been found. The evolutionary roots of the vertebrate endocrine system may, therefore, be far more ancient than is generally believed.

Reduced cytotoxicity of the lysosomotropic detergent N-dodecylimidazole after differentiation of HL60 promyelocytes.

The sensitivity of the human promyelocytic cell line HL60 to killing by the lysosomotropic detergent N-dodecyl imidazole (C12-Im) has been investigated in the exponential and stationary growth states and before and after differentiation induced by suitable effector molecules. Undifferentiated HL60 cells were more sensitive to killing by C12-Im in the rapid (exponential) phase of growth than in the stationary phase, in keeping with our observations on many other cell lines. Differentiation into granulocytes induced by dimethyl sulfoxide, or into macrophages induced by phorbol ester, resulted in a further dramatic decrease in sensitivity to C12-Im, as compared to undifferentiated HL60 cells in stationary phase. Viable cells remaining after treatment with C12-Im (60 micrograms/ml, 2 h) were: 0% for exponentially growing undifferentiated cells; 16% for stationary undifferentiated cells; 41% for differentiated granulocytes; and 29% for differentiated macrophages. Treatment with the cysteine cathepsin inhibitor L-trans-epoxysuccinylleucylamido(4-guanido)butane (E64) conferred resistance to C12-Im, showing that, in these cells, as previously demonstrated for Chinese hamster ovary fibroblasts, cysteine proteases were major cytotoxic agents involved in killing by C12-Im. Cell cathepsin B + L activity levels were dramatically reduced in those cells differentiated into granulocytes (11.2 units/mg of protein) and into macrophages (9.8 units/mg of protein) as compared with undifferentiated HL60 promyelocytes in stationary phase (30.4 units/mg of protein), correlating well with reduced sensitivity to C12-Im in the differentiated cells.

Effects of growth state and amines on cytoplasmic and vacuolar pH, phosphate and polyphosphate levels in Saccharomyces cerevisiae: a 31P-nuclear magnetic resonance study.

The vacuoles of logarithmic and stationary stage cells were compared by 31P-NMR with regard to pH, orthophosphate (Pi) content and average size of polyphosphate. The vacuoles of stationary cells had lower pH, higher Pi content, and polyphosphates of longer average chain length, although total polyphosphate content was about the same as in logarithmic cells. The lower vacuolar pH in stationary cells was the major cause of a larger cytoplasmic-vacuolar pH gradient. Addition of NH4Cl, (NH4)2SO4, methylamine or amantadine at pH 8 to cells in either stage caused an increase in both cytoplasmic and vacuolar pH, with little or no change in the cytoplasmic-vacuolar pH gradient. However, the administration of ammonium salts to the cells at pH 8.0 resulted in rapid hydrolysis of the intravacuolar polyphosphate to tripolyphosphate and Pi, with attendant redistribution of Pi between the vacuolar and cytoplasmic compartments.

Permeability properties of the membrane of vesicular stomatitis virions.

Observations of the light-scattering properties of several enveloped viruses indicate that virions (vesicular stomatitis, SV5 and influenza), in common with other membrane systems, are osmotically active, responding to NaCl gradients by swelling in hypo-osmolar solutions and shrinking in hyperosmolar solutions. The permeability barrier responsible for this osmotic response in vesicular stomatitis virions was modified both by protease treatment to remove the viral glycoprotein and by treatment with the polyene antibiotic filipin, an agent known to interact with cholesterol in liposomes and membranes. Filipin altered the kinetic and equilibrium permeability behavior of virions but the extent of leakage of osmotic shocking agent was less than that in lecithin/cholesterol and lecithin/ergosterol liposomes and in ergosterol-containing ciliary membranes. Negative-staining electron microscopy revealed that filipin treatment caused structural changes in the viral membrane. Intact virions exhibited appreciably larger responses to osmotic change than did protease-treated virus particles. Thus, the osmotic barrier in intact vesicular stomatitis virions may not be exclusively lipid in nature.

Intracellular sodium content of a wall-less strain of Neurospora crassa and effects of insulin: a 23Na-NMR study.

23Na-NMR has been used to investigate some factors influencing the sodium content of a wall-less strains of Neurospora crassa. The shift reagent Tm(DOTP)H2(NH4)3 proved useful for this purpose, while several other reagents, previously used by others, were found to be unsuitable for use with these cells. When the cells were grown, washed and resuspended in medium containing sodium (25.3 mM), the intracellular sodium concentration was calculated to be 11.9 +/- 1.4 mM. This value rose within two minutes of addition of glucose (100 mM), to greater than 14 mM. Preincubation of cells with insulin (100 nM) had a significant effect on the subsequent rate of sodium accumulation during the period 3-12 minutes following glucose addition. Insulin-treated cells showed a slow, continued accumulation of sodium during this period (+1.14 +/- 0.39%/min), while control cells lost sodium very slowly (-0.63 +/- 0.29%/min; P of difference = 0.005).

Studies on the in vitro synthesis of ppGpp and pppGpp.

The effect of a number of inhibitors of protein synthesis on ppGpp and pppGpp synthesis in vitro has been examined. As expected from in vivo results, chloramphenicol is without effect on this reaction. Aurintricarboxylic acid and chlortetracycline, on the other hand rapidly and specifically inhibit ppGpp synthesis. Fusidic acid in the presence of saturating amounts of EF G also inhibits the reaction completely, suggesting that an empty ribosomal acceptor site is necessary for this reaction. On the other hand, the 50-S subunit proteins L7 and L12 are not required for stringent factor activity. Ribosomes from Pseudomonas fluorescens can replace those from Escherichia coli in the complete system, while ribosomes from Ehrlich ascites cannot. A small but reproducible synthesis of ppGpp is observed when the ribosomal wash from E. coli is complemented with ribosomes from wheat germ cytoplasm.

Inhibition of synthesis of ribosomal proteins and of ribosome assembly after infection of L cells with vesicular stomatitis virus.

The effect of infection of mouse L cells by vesicular stomatitis virus on the synthesis of ribosomal proteins was investigated using two-dimensional polyacrylamide gel electrophoresis to analyze the ribosomal proteins. It was found that the synthesis of nearly all of the cytoplasmic ribosomal proteins examined was inhibited by infection and mostly to the same extent. Analysis of the ribosomal proteins extracted from intact ribosomes indicated that infection also reduces the incorporation of all the ribosomal proteins tested into assembled ribosomes. The inhibition of ribosome assembly was greater than the inhibition of synthesis of ribosomal proteins, suggesting that some other factor was also limiting the assembly of ribosomes. As shown in this report, infection also inhibits ribosomal RNA production. Thus, the decreased assembly of ribosomes in infected cells probably results from the inhibition of synthesis of both ribosomal proteins and ribosomal RNA.

Host cell protein kinases in nonsegmented negative-strand virus (mononegavirales) infection.

Phosphorylation of one or more viral proteins is probably an essential step in the life cycle of every member of the nonsegmented negative-strand RNA virus (mononegavirales [MNV]) group. Since no virally encoded protein kinases have been discovered in this group, phosphorylation is effected entirely by host cell kinases. The virally encoded P proteins of the MNV are the only ones consistently phosphorylated with a stoichiometry > or =1. The P protein of vesicular stomatitis virus (VSV), and perhaps also of respiratory syncytial virus, are the only ones for which a function of phosphorylation has been established. Phosphorylation by casein kinase 2 at one or more identified sites in the VSV P protein activates transcriptional activity by promoting formation of a homotrimer, which is then capable of binding the RNA polymerase and attaching it to the N protein-RNA template. A second phosphorylation of VSV P protein by a different kinase also occurs, dependent upon prior modification by casein kinase 2, but its function is not definitely known. Phosphorylation of the other MNV P proteins may serve a different purpose. No evidence has been obtained yet for any function for phosphorylation of any other MNV protein.

Effects of mammalian insulin on metabolism, growth, and morphology of a wall-less strain of Neurospora crassa.

Addition of mammalian insulin to a nutritionally rich, chemically defined culture medium affects Neurospora crassa "slime" (wall-less) cells, as indicated by enhancement of growth, extension of viability at the stationary phase of growth, alteration of morphology, and stimulation of glucose oxidation. Bovine, porcine, and recombinant human insulin had similar effects on growth and morphology, while proinsulin, reduced insulin, and several other proteins were inactive. Insulin added in the presence of excess antiinsulin antibody was without activity. Intact cells possessed high affinity insulin-binding sites, represented by a curvilinear Scatchard plot, suggesting that effects are mediated through insulin receptors on the cell surface. These findings establish a role for insulin or insulin-like molecules in regulating growth and metabolism in this fungal cell and demonstrate a close similarity to insulin effects on certain mammalian cells.

Stimulation by mammalian insulin of glycogen metabolism in a wall-less strain of Neurospora crassa.

Addition of bovine insulin to cells of the wall-less variant FGSC4761 of Neurospora crassa ("slime") produced several significant effects on glycogen metabolism. 1) Intracellular levels of the glycogen precursor UDP-glucose decreased 17-18% (P less than 0.01) within 30 min of insulin addition. 2) Cells grown with insulin possessed 40% more glycogen than did control cells. 3) The incorporation of 14C-labeled glucose into glycogen increased 41% after 30-min treatment with 100 nM bovine insulin (P less than 0.01). 4) Insulin treatment of the cells caused activation of the enzyme glycogen synthase from a glucose-6-phosphate-dependent form to an independent form. Half-maximum activation occurred with 2 nM insulin. These are similar to insulin-induced effects in some mammalian cells. In contrast, no insulin-induced effect on glucose transport could be demonstrated in these cells.

Characterization of PDR4, a Saccharomyces cerevisiae gene that confers pleiotropic drug resistance in high-copy number: identity with YAP1, encoding a transcriptional activator [corrected].

PDR4 is a gene that confers pleiotropic drug resistance (pdr) to the yeast Saccharomyces cerevisiae when present in high copy number [Leppert et al., Genetics 125 (1990) 13-20]. Transposon insertion mutations had identified the active region of the gene as a 3.7-kb SalI-EcoRI restriction fragment of the 8-kb cloned fragment. We have confirmed this by showing that this fragment is sufficient to confer pdr, and have sequenced its entire 3761 bp. It contains a single complete open reading frame (ORF) extending from nucleotide (nt) position 1631-3580, coding for a protein of 650 amino acids (aa). A 2.7-kb fragment containing this ORF is also sufficient to confer pdr. The aa sequence contains no recognizable homologies or consensus sequences, so it is a novel protein of unknown function. It is apparently soluble, since no transmembrane-type sequences were predicted. A second, partial ORF was also found, on the opposite strand, extending from nt position 774 to past the SalI site, which is apparently unrelated to pdr.

Seripauperins of Saccharomyces cerevisiae: a new multigene family encoding serine-poor relatives of serine-rich proteins.

A gene, PAU1, has been cloned from Saccharomyces cerevisiae and sequenced. It is located in a telomeric region, probably on chromosome IV, and contains an open reading frame encoding a protein of 120 amino acids (aa) (approx. 13 kDa). The deduced sequence is nearly identical to two other genes found in GenBank (named PAU2 and PAU3 by us), which are located close to the ends of chromosomes V and III, respectively. Blotting of separated chromosomes with a PAU1 probe at high stringency revealed that at least six chromosomes in addition to III, IV and V possessed related sequences, suggesting a large gene family. Probing of an ordered array of phage lambda clones containing yeast genomic DNA inserts ('Olson filters') revealed ten additional hybridizing sequences, located close to the ends of the left and/or right arms of chromosomes I, II, VII, VIII, X, XII, XIV and XV. Transcription of these sequences could not be demonstrated, however, under a wide variety of growth and culture conditions. The deduced PAU1, PAU2 and PAU3 aa sequences are all highly homologous with the SRP1 aa sequences, which contains eight serine-rich tandem repeats of 12 aa each, at its C terminus. This homology is limited, however, to the N-terminal half of SRP1, and does not include the repeats. In fact, PAU1 is quite serine-poor (5.8%), leading to the suggested name of seripauperins for this family of genes.

A preproinsulin-like pseudogene from Neurospora crassa.

A segment of DNA was amplified from the Neurospora crassa genome by the polymerase chain reaction using several oligonucleotides coding for highly conserved domains in proinsulin as primers and probe. A genomic clone corresponding to this segment was isolated and the nucleotide sequence was determined. The deduced amino acid sequence of a part of this segment bears remarkable resemblance to preproinsulin, but lacks several requirements for transcription or translation and must therefore be considered to be a pseudogene.