Gisela Wengler

The Fusion Glycoprotein Shell of Semliki Forest Virus: An Icosahedral Assembly Primed for Fusogenic Activation at Endosomal pH

Semliki Forest virus (SFV) has been extensively studied as a model for analyzing entry of enveloped viruses into target cells. Here we describe the trace of the polypeptide chain of the SFV fusion glycoprotein, E1, derived from an electron density map at 3.5 A resolution and describe its interactions at the surface of the virus. E1 is unexpectedly similar to the flavivirus envelope protein, with three structural domains disposed in the same primary sequence arrangement. These results introduce a new class of membrane fusion proteins which display lateral interactions to induce the necessary curvature and direct budding of closed particles. The resulting surface protein lattice is primed to cause membrane fusion when exposed to the acidic environment of the endosome.

The carboxy-terminal part of the NS 3 protein of the West Nile Flavivirus can be isolated as a soluble protein after proteolytic cleavage and represents an RNA-stimulated NTPase

Recently it has been reported that a membrane fraction can be isolated from West Nile virus-infected BHK cells which contains the viral nonstructural (NS) proteins as major constituents (Wengler et al., 1990). In this report we show that treatment of these membranes with subtilisin releases the carboxy-terminal segment of the NS 3 protein as a soluble protein of about 50 kDa apparent molecular weight. This molecule, which is called the p50-S protein, can be purified by standard chromatographic procedures. The p50-S protein binds to poly(A) and apparently represents a nucleoside triphosphatase which is stimulated in the presence of ssRNA molecules. The data represent experimental support for the predicted role of this segment of the NS 3 protein as an RNA helicase. Some properties of the p50-S protein are described and a possible function of this protein segment during RNA synthesis is discussed.

Studies on virus-specific nucleic acids synthesized in vertebrate and mosquito cells infected with flaviviruses

Abstract Virus-specific RNA molecules synthesized in BHK 21 vertebrate cells and in Aedes albopictus mosquito cells infected with the flaviviruses Uganda S (US) or West Nile (WN) have been characterized. A single-stranded (ss) RNA of plus polarity sedimenting at about 42 S was present in the virus particles. 42 S plus strand RNA was also the predominant species of virus-specific ss RNA accumulating in infected cells of both vertebrate and insect origin. No similarity was detected between the large oligonucleotides generated by ribonuclease T 1 from WN virus and US virus-specific 42 S plus strand RNA, respectively. No poly(A) sequences are present in either the US virus or the WN virus-specific 42 S plus strand RNA molecules synthesized in BHK cells. The 42 S plus strand RNA molecules present in WN virus-infected BHK cells do contain, however, a “cap” structure m 7 GpppAmpN 1 , and in some of these molecules, a further methyl group is introduced, giving rise to the “cap” structure m 7 GpppAmpN 1 mpN 2 . These “caps” are also present on the 42 S RNA of WN virus particles synthesized in BHK cells. In addition to the 42 S RNA, virus-specific ss RNA of low molecular weight (LMW-RNA) was detected in all virus-cell systems analyzed. A single species of LMW-RNA of 5 × 10 4 daltons apparent molecular weight was present in US virus-infected vertebrate and insect cells. Two LMW-RNA species of about 6.5 × 10 4 daltons (WM LMW-1 RNA) and 4.2 × 10 4 daltons (WN LMW-2 RNA) molecular weight, respectively, were isolated from WN virus-infected BHK cells. Only the larger of these was detected in WN virus-infected insect cells. The LMW-1 RNA present in WN virus-infected BHK cells has been characterized in somewhat more detail. It contains virus-specific RNA sequences of plus strand polarity, is not “capped”, and does not contain a poly(A) sequence. None of the single-stranded, virus-specific RNA molecules synthesized in either vertebrate or mosquito cells bound to oligo(dT)-cellulose. Only a single species of virus-specific RNA containing minus strand sequences was detected in WN virus-infected BHK cells. This RNA was of genome size and was present as part of a double-stranded RNA complex containing 42 S RNA of both plus and minus polarity.

Sequence analysis of the viral core protein and the membrane-associated proteins V1 and NV2 of the flavivirus west nile virus and of the genome sequence for these proteins

Cell-associated flaviviruses contain the two membrane proteins V3 and NV2 besides the viral core protein V2 whereas extracellular viruses do contain V2 protein and the two membrane proteins V3 and V1. Since the V1 protein could not be detected in infected cells it has been suggested that V1 is generated from NV2 by proteolytic cleavage during the release of virus from cells (D. Shapiro, W. E. Brandt, and P. K. Russell (1972), Virology 50, 906-911). We have isolated the viral structural proteins V1, V2, and NV2 from the flavivirus West Nile virus and determined their amino-terminal amino acid sequences and amino acid sequences of peptides derived from these proteins. We have also transcribed parts of the viral genome into cDNA and cloned and sequenced this cDNA. The analyses of the protein structure of V1, V2, and NV2 together with the determination of the amino-terminal sequence of V3 (data not shown) have allowed us to identify the nucleotide region coding for the structural proteins V2, NV2, and V1. The primary structure of this nucleotide sequence is presented in this report. The data show that the amino terminus of the viral core protein V2 is followed by the amino termini of the proteins NV2, V1, and V3, respectively. These data for the first time identify the exact order of all structural proteins of a flavivirus identified so far. Our data strongly support the above-mentioned hypothesis that V1 is derived from NV2 by proteolytic cleavage and furthermore indicate that V1 represents the nonglycosylated carboxy-terminal part of NV2 which contains those sequences which anchor NV2 in the viral membrane. A working hypothesis is presented in which two species of cellular enzymes, signalase(s) removing signal sequences and enzymes involved in cleaving polyproteins after a pair of basic amino acids, do generate the proteins V2, NV2, and V1 from the growing peptide chain synthesized during translation of the 42 S genome RNA which functions as mRNA for these proteins.

Analyses of the terminal sequences of west nile virus structural proteins and of the in vitro translation of these proteins allow the proposal of a complete scheme of the proteolytic cleavages involved in their synthesis

The proteolytic processes involved in the synthesis of the structural proteins of the West Nile (WN) flavivirus were analyzed: The carboxy-terminal sequences of the structural proteins were determined and the proteins translated in vitro in the presence of membranes from a mRNA coding for the structural polyprotein were analyzed. The results obtained indicate that the following proteolytic activities are involved in the synthesis and assembly of WN virus structural proteins: The growing peptide chain which contains the sequences of the structural proteins in the order C-pre-M-E is cleaved at three places by cellular signalase(s). This cleavage generates the primary amino acid sequence of the mature structural proteins pre-M and E (and the amino-terminus of the ensuing nonstructural protein NS 1). The amino-terminal part of the polyprotein containing the amino acid residues 1 to 123 is released as a molecule which migrates slightly slower than the mature viral core protein and which presumably is associated to the RER membranes via its carboxy-terminal sequence. This protein is called the anchored C virus particles the anchored C protein is converted into mature C protein by removal of the carboxy-terminal hydrophobic segment containing the amino acid residues 106 to 123. Presumably a virus-coded protease which can cleave the polyprotein after two basic amino acid residues is responsible for this cleavage. The cell-associated WN virus particles are constructed from the proteins C, pre-M, and E which contain the amino residues 1-105, 124-290, and 291-787 of the polyprotein, respectively. Cleavage of the pre-M protein between amino acid residues 215 and 216, presumably by a cellular enzyme located in the Golgi vesicles, and loss of the amino-terminal fragment of this protein are associated with the release of virus from the cells.

Terminal sequences of the genome and replicatioe-form RNA of the flavivirus west nile virus: absence of poly(A) and possible role in RNA replication

Abstract The structures of the infectious 42 S genome RNA of the flavivirus West Nile (WN) virus and of the replicative-form (RF) RNA containing 42 S RNA of positive and negative polarity have been investigated. The RF RNA has been labeled in vitro at the 3′ and 5′ termini and the terminal sequences have been determined by the mobility shift method. The results obtained indicate that both RNA molecules are exact complements of each other and that the 3′ terminus of the 42 S plus-strand RNA component of the RF RNA does not contain a poly(A) sequence but terminates with a heteropolymeric AACACAGGAUCU OH sequence. The 3′ terminus of the 42 S minus-strand RNA has the sequence CUCACACAGGCGAACUACU OH . Comparison of these sequences shows that both molecules contain the 3′-terminal dinucleotide CUOH and the heptanucleotide ACACAGG which is separated from the 3′-terminal dinucleotide by two and seven nucleotides in 42 S plus- and minus-strand RNA, respectively. The 42 S viral genome RNA also does not contain a 3′-terminal poly(A) sequence but terminates with the 3′-terminal sequence identified in the 42 S plus-strand RNA of the RF. Analysis of the nucleotides adjacent to the cap at the 5′ terminus of the viral genome RNA together with the 3′-terminal sequence analysis indicates that the nucleotide sequence of the viral genome RNA is identical to that of the 42 S plus-strand RNA component of the virus-specific RF RNA.

Identification of a sequence element in the alphavirus core protein which mediates interaction of cores with ribosomes and the disassembly of cores

Early in infection core protein is transferred from alphavirus cores to ribosomes (Wengler and Wengler, 1984, Virology 134, 435-442) and it has been suggested that ribosome binding is a property of alphavirus core protein which is involved in core disassembly. Here we describe in vitro analyses of this transfer. Sindbis virus cores, incubated with ribosomes either in a reticulocyte lysate or in buffer, are disassembled with a concomitant transfer of core protein to the large ribosomal subunit. Preincubation of ribosomes with core protein blocks disassembly. Limited proteolysis of Sindbis virus core releases the carboxy-terminal core protein domain as a soluble fragment (Strong and Harrison, 1990, J. Virol. 64, 3992-3994). Trypsin- or proteinase Lys-C-released fragments contain the amino-terminal residue met (106) or gln (94), respectively. The fragment generated by proteinase Lys-C binds to ribosomes and interferes with core disassembly whereas the slightly shorter tryptic fragment has none of these activities. These and further analyses indicate that a conserved sequence element which surrounds amino acid met (106) of SIN CP, the so-called RBSc element, leads to binding of core protein to ribosomes and thereby to core disassembly. Implications of the experiments for regulation of assembly of alphavirus cores and for the core protein-induced resistance to viral multiplication observed in plant virus systems are discussed.

Localization of the 26-S RNA sequence on the viral genome type 42-S RNA isolated from SFV-infected cells.

Abstract 32 P-labeled virus-specific 42- and 26-S RNA has been prepared from BHK 21 cells infected with the togavirus Semliki Forest virus. Analyses of the oligonucleotides generated from these RNA species after digestion with RNAse T 1 by two-dimensional polyacrylamide gel electrophoresis (T 1 -fingerprints) show that both RNA species contain poly(A) sequences and that the oligonucleotides of the 26-S RNA represent a subset of those present in 42-S RNA. The following findings indicate that the different nucleotide sequences are present on the 42-S RNA in the order 5′-terminus-42-S RNA-specific sequences-26-S RNA sequences-poly(A)-A OH . (1) Identical oligonucleotide patterns are found in the T 1 -fingerprints of the 26-S RNA accumulating in infected cells and of poly(A)-containing fragments of the 42-S RNA which sediment around 26 S on sucrose density gradients. (2) The poly(A) fragments were isolated from T 1 -fingerprints of 42-and 26-S RNA, digested with pancreatic RNAse, and the reaction products analyzed by electrophoresis on DEAE paper. The results indicate that both fragments contain a heteropolymeric sequence consisting of U 6 C 2 (AC) 1 (AU) 1 (AAU) 1 . (3) GMP has been detected in all oligonucleotides derived from T 1 -fingerprints which have been analyzed so far, except in the poly(A)-containing oligonucleotides, indicating that the latter contain the 3′-termini of the molecules from which they were derived. (4) [ 3 H]Adenosine-labeled 42-S RNA was digested with RNAse T 1 and pancreatic RNAse, the poly(A) fragments were isolated, and an [ 3 H]adenosine to [ 3 H]AMP radioactivity ratio of 1 to 97 was determined for these fragments. Possible implications of the organization of the different sequences on the 42-S RNA on its translation and the replication of SFV are discussed.

The core protein of the alphavirus sindbis virus assembles into core-like nucleoproteins with the viral genome RNA and with other single-stranded nucleic acids in vitro

Abstract A system has been developed that allows the reconstruction of a core-like (CL) ribonucleoprotein (RNP) from Sindbis virus-specific core protein and genome RNA in vitro . The RNP particles were analyzed by equilibrium density gradient centrifugation and electron microscopy. The CL RNP is similar in size, shape, and texture to authentic viral core. The assembly of the homologous CL RNP in vitro depends on the relative concentrations of protein and RNA in the reaction: At low concentrations of protein incomplete particles of rather high density are made; increase of the protein concentration leads to an optimum concentration at which the protein is quantitatively incorporated into complete CL particles; further increase in protein concentration leads to the formation of a precipitate which has not been analyzed in detail. No identifiable structures were generated in vitro in the absence of nucleic acid, but all single-stranded deoxyribonucleic and ribonucleic acids analyzed were incorporated into particles similar to those formed in the presence of viral genome RNA. These complexes are called heterologous CL nucleoproteins. Since nucleic acids differing in size between about 100 and 6000 nucleotides (e.g., tRNA and fd DNA), which vary widely in secondary structure are efficiently incorporated into heterologous CL particles, probably all single-stranded nucleic acids in this size range can be efficiently incorporated into such particles in vitro . Some implications of a possible interaction between viral core protein and single-stranded nucleic acids other than the viral genome in vivo , e.g., during the synthesis of defective interfering particles or during inhibition of host cell DNA synthesis, are discussed.

Identification of a transfer of viral core protein to cellular ribosomes during the early stages of alphavirus infection

Sindbis virus containing [35S]methionine-labeled structural proteins was allowed to be taken up by primary chick embryo fibroblasts, and the fate of the core protein was studied. The experiments show that core protein of incoming viral particles is transferred to the large subunit of cellular ribosomes during the initial steps of virus infection. A similar transfer occurs in vitro if cores isolated from SIN virus particles are incubated in the postmitochondrial cytoplasmic fraction of cell lysates. In vivo transfer also occurs if the protein synthesis-inhibiting drugs puromycin or cycloheximide are present during virus uptake, whereas in the presence of chloroquine, which inhibits the release of viral cores into the cytoplasm, which is necessary for productive infection, a transfer of core protein to ribosomes cannot be observed. The latter result indicates that the transfer probably is part of the reactions leading to the release of viral genomic RNA into the cellular cytoplasm during the early stages of productive infection, and presumably does not reflect side reactions. It has been shown earlier that newly synthesized alphavirus core protein binds to the large ribosomal subunit prior to the assembly of viral core particles in infected cells [I. Ulmanen, H. Sonderlund, and L. Kääriäinen (1979) Virology 99, 265-276]. These data lead to the suggestion that the disassembly and assembly of alphavirus cores might be regulated by a process which could be called receptor-mediated core disassembly, in which acceptors exist for the protein components of viral nucleoproteins in uninfected cells which early in infection bind these proteins and thereby lead to disassembly of these complexes, and which later on have to be saturated with newly synthesized protein before efficient assembly of nucleoproteins can occur, and that the large ribosomal subunit functions as such a receptor during alphavirus replication.