Purnell W. Choppin

Identification of biological activities of paramyxovirus glycoproteins. Activation of cell fusion, hemolysis, and infectivity by proteolytic cleavage of an inactive precursor protein of Sendai virus

Abstract The glycoproteins of Sendai virus have been isolated by a procedure involving extraction with the nonionic detergent Triton X-100 and affinity chromatography on fetuin-Sepharose. The largest Sendai virus glycoprotein (MW ~69,000) possesses both hemagglutinating and neuraminidase activities, and has been designated HN. Virions grown in MDBK cells or in the allantoic sac of the chick embryo contain similar amounts of the HN glycoprotein, but differ in their content of the other glycoproteins. Virions grown in MDBK cells contain a large amount of a glycoprotein designated F 0 (MW ~65,000). This glycoprotein is a precursor of a smaller virion glycoprotein, F (MW ~53,000) which is present in only small amount in MDBK cellgrown virions. F 0 can be cleaved to yield F by treatment of virions with trypsin in vitro . Sendai virions grown in the chick embryo lack the precursor protein F 0 , but contain a large amount of F; in this case, proteolytic cleavage of F 0 to yield F occurs in ovo . The Sendai virions grown in MDBK cells, which are deficient in the small glycoprotein F, lack hemolyzing and cell-fusing activities and cannot infect MDBK cells. Virions grown in the chick embryo, which contain much F, possess both these activities and can infect MDBK cells as well as the chick embryo. MDBK cell-grown virions acquire hemolyzing and cell-fusing activities and become infective for MDBK cells when the precursor glycoprotein F 0 is cleaved in vitro to yield F. The results indicate that the small glycoprotein of paramyxoviruses is biologically active and is involved in virus-induced hemolysis, cell fusion, and the initiation of infection. The precise mechanism by which this glycoprotein participates in these reactions remains to be determined, but is now amenable to experimentation. The precursor of this glycoprotein is biologically inactive, but is incorporated into virions grown in some host cells; it may be activated by proteolytic cleavage either in vivo or in vitro . The present results provide a biochemical basis for previously observed host-dependent variation in infectivity, and in hemolysis and cell-fusion induced by paramyxoviruses.

Enhancement of the infectivity of influenza A and B viruses by proteolytic cleavage of the hemagglutinin polypeptide

The infectivity of virions of the WSN strain of influenza A (AWSN) and the 1760 strain of influenza B (B/1760) which possess their hemagglutinin in the form of the uncleaved HA polypeptide can be enhanced as much as 100-fold by proteolytic cleavage of HA to yield HA1 and HA2. Hemagglutinating activity is unaffected by cleavage. The HA polypeptide of AWSN virions is susceptible to cleavage by trypsin and plasmin, whereas that of B/1760 virions is resistant to plasmin. The increase in infectivity can be demonstrated in different cell types, i.e., MDBK, BHK21-F, CEF, or HKCC cells, as well as in the chick embryo, and in titrations in both liquid medium and plaque assays under agar. Chymotrypsin cleaves the HA polypeptide of WSN virions in the same general region of the molecule as trypsin and plasmin, but without the enhancement of infectivity produced by the latter enzymes. Incubation of WSN virions with chymotrypsin prevents the enhancement by subsequent treatment with trypsin, but the increase in infectivity caused by an initial treatment with trypsin is not abolished by subsequent treatment with chymotrypsin. These results indicate that enhancement of infectivity of influenza virions results from proteolytic cleavage at a specific site on the HA molecule; however an increase in infectivity is not associated with cleavage at a nearby but different site. Both AWSN and B/1760 virions are capable of undergoing multiple cycles of replication in cells under conditions in which no cleaved HA polypeptides are detected on the virions produced. These results are compatible with the enhancement of infectivity being the result of an increase upon cleavage in the efficiency of some stage in the initiation of infection beyond the initial adsorption step. The increase in infectivity caused by proteolytic cleavage of the HA polypeptide provides a biochemical explanation for the previously observed enhancement of plaquing efficiency of influenza viruses by the inclusion of pancreation or trypsin in the agar overlay.

Specific inhibition of paramyxovirus and myxovirus replication by oligopeptides with amino acid sequences similar to those at the N-termini of the Fl or HA2 viral polypeptides

Abstract A series of oligopeptides have been synthesized with amino acid sequences that resemble those of the N-terminal regions of the paramyxovirus F 1 , polypeptide or the myxovirus HA 2 polypeptide, N-termini generated by proteolytic cleavage which activates infectivity. Oligopeptides with the appropriate structure are highly active, specific inhibitors of the infectivity of each virus, and of cell fusion and hemolysis induced by paramyxoviruses. Structure-activity studies have revealed the following characteristics of the inhibitory activity. Inhibition is amino acid sequence specific, and the presence of the same N-terminal amino acid on the oligopeptide as the viral polypeptide is crucial for activity. Longer peptides are more active than shorter ones with the same initial sequence. The presence of a carbobenzoxy group (Z) on the N-terminal amino acid of the oligopeptide increases inhibitory activity, and the addition of a methyl group to the carboxyl moiety of the C-terminal amino acid decreases activity. The steric configuration of the first two amino acids has a significant effect; optimum activity was obtained with oligopeptides beginning with Z- d -phenylalanine- l -phenylalanine. The results suggest that the oligopeptides competitively interfere with the N-terminal region of the F 1 or HA 2 polypeptides of paramyxoviruses or myxoviruses, respectively. The availability of these oligopeptide inhibitors provides an additional means for determining the exact site and mechanism by which the F 1 and HA 2 polypeptides initiate infection, and for investigating the biochemical and physical events in virus-induced cell fusion and hemolysis, and membrane fusion in general. The finding of specific inhibition of infectivity by oligopeptides which resemble a region of a viral polypeptide also provides a possible new approach to chemical inhibition of viral replication.

Influenza virus proteins: I. Analysis of polypeptides of the virion and identification of spike glycoproteins☆

Abstract The polypeptides of influenza virions have been analyzed by polyacrylamide gel electrophoresis. Seven polypeptides were detected in purified virions grown in three different cell types, and four of these polypeptides appear to be covalently linked with carbohydrate. Virions grown in different cell types were compared by coelectrophoresis, and host-dependent differences were detected in the electrophoretic mobility of some glycoproteins, but not of nonglycoproteins, suggesting that the carbohydrate moiety of such glycoproteins is specified by the host cell. The molecular weights of the seven proteins, estimated by coelectrophoresis with marker proteins, ranged from 83,500 to 26,500, with a total molecular weight of 380,500. Treatment of virions with the protease bromelain degraded the viral spikes resulting in particles which were bounded by a smooth-surfaced membrane, and which could be purified in a potassium tartrate gradient. Such particles were lacking three of the four glycoproteins; the four remaining viral proteins were unaltered by the enzyme. The particles devoid of spikes were noninfective, and they lacked hemagglutinin and neuraminidase activities, indicating that these properties are associated with the spike glycoproteins.

Two disulfide-linked polypeptide chains constitute the active F protein of paramyxoviruses.

Abstract The paramyxovirus glycoprotein which is required for virus-induced cell fusion, hemolysis, and the initiation of infection (F protein) has been found with three different viruses to consist of two disulfide-linked glycopolypeptide chains (F1 and F2) which are derived from the precursor glycopolypeptide (F0) by proteolytic cleavage. The larger glycopolypeptide chain (F1) previously identified in SV5, Sendai, and Newcastle disease virions has an estimated molecular weight of 48,000–54,000. The smaller polypeptide chain (F2) has been found to have a molecular weight of ∼10,000–16,000, depending on the virus. The identification of the F2 polypeptide was accomplished by isolating the disulfide-linked complex (F1,2) under nonreducing conditions, followed by reduction of the disulfide bonds and separation of the two polypeptide chains. Although both polypeptides are glycosylated, the F2 polypeptide contains more carbohydrate per unit protein than F1. No free N-terminus could be detected on the F0 or F2 polypeptides of Sendai virus, whereas N-terminal phenylalanine was found on F1. This suggests that the order of the F0 polypeptide is X-NH-F2—Phe-F1—COOH. The finding that with three different paramyxoviruses the biologically active virions possess F1 and F2 polypeptides suggests that this is a general feature of paramyxoviruses, and that the activation of infectivity, cell fusion, and hemolysis is due to a conformational change in the F protein resulting from proteolytic cleavage to form an active complex of two disulfide-linked polypeptide chains.

Isolation of paramyxovirus glycoproteins. Association of both hemagglutinating and neuraminidase activities with the larger SV5 glycoprotein

A method has been developed for the isolation of the glycoproteins of the parainfluenza virus SV5 using the nonionic detergent Triton X-100. Full recovery of hemagglutinating and neuraminidase activities was obtained. By rate zonal centrifugation in sucrose gradients containing 1% Triton X-100 and 0.5 M or 1 M potassium chloride, it was possible to separate the two glycoproteins. Under these conditions, the sedimentation coefficient of the larger glycoprotein, virus protein 2, was 8.9 S and that of the smaller glycoprotein, virus protein 4, was 6.7 S. Each of the proteins aggregated when the detergent and KCl were removed, and the appearance of the aggregates differed with the two proteins. Both hemagglutinating and neuraminidase activities were found to be associated with protein 2; protein 4 exhibited neither activity. The results suggest that in this paramyxovirus both hemagglutinating and neuraminidase activities reside on a single glycoprotein. The biological function of the smaller SV5 glycoprotein remains to be determined.

The synthesis of Sendai virus polypeptides in infected cells

Abstract The synthesis of virus-specific proteins in several cell types infected with Sendai virus have been analyzed by high resolution polyacrylamide-gel electrophoresis. Eight structural proteins of the virion have been identified in infected cells and the kinetics of their synthesis examined. In addition, two nonstructural proteins with molecular weights (MW) of ∼ 36,000 and ∼ 22,000 have been found. The results of pulse and pulse-chase experiments suggest that large, polyprotein precursors are not involved in Sendai virus replication, and that polypeptides are synthesized from monocistronic messenger RNA species. Throughout the replicative cycle, the virion polypeptides are synthesized in approximately the same unequal proportions at which they are present in the virus particle, except for a relatively greater amount of the P polypeptide and a smaller amount of the M polypeptide in the cells, indicating that polypeptide synthesis is strictly controlled. Results of a double-isotopic label difference analysis showed that host-cell protein synthesis declined by 8% only over 12 hr and that viral protein synthesis was superimposed upon cellular synthesis with total protein synthesis in infected cells reaching 205% of that in uninfected cells. The available evidence suggests that one of the viral polypeptides (No. 7, MW ∼ 42,000) is a cellular polypeptide and that infection selectively reduces the synthesis of this polypeptide.

Influenza virus structural and nonstructural proteins in infected cells and their plasma membranes

Abstract The virus-specific proteins in three types of cultured cells infected with the WSN strain of influenza virus have been analyzed. The seven structural proteins of the WSN virion were identified in infected cells, and in addition a nonstructural protein, designated NNP, was found in large amount in the nucleus of the infected cell. This nonstructural protein has an estimated molecular weight of ∼25,000 and is slightly smaller than protein 7, the smallest viral structural protein. Proteins NNP and 7 were isolated on an agarose column and shown to be distinct by peptide mapping. The seven viral structural proteins have been found in association with the plasma membrane of infected cells. Glycoproteins 5 and 6 are the last proteins to appear at the plasma membrane. The available evidence suggests that protein 2, which is found in much larger amount in infected cells than in the mature virion, is cleaved at the plasma membrane to yield proteins 5 and 6.

Reproduction of Paramyxoviruses

The paramyxovirus group is a large one which includes the parainfluenza viruses types 1–5, Newcastle disease, and mumps viruses. Measles, canine distemper, and rinderpest viruses form a distinct subgroup on the basis of antigenicity, hemagglutinating characteristics, and lack of evidence for a virion-associated neuraminidase or neuraminic acid-containing cellular receptors. However, it is now generally accepted that these viruses should also be included in the paramyxovirus group because of their similar structural properties. Other more recently isolated viruses which have been classified as paramyxoviruses on the basis of morphological and biological properties are Yucaipa (Dinter et al., 1964) and Nariva (Walder, 1971) viruses. Table 1 lists paramyxoviruses and their primary hosts.

Evidence for a ninth influenza viral polypeptide.

Abstract An ∼11,000 molecular weight (mol. wt.) polypeptide (designated 4) has been found in cells of various types infected with the WSN strain of influenza virus and, in small amount, in purified virions. Peptides mapping and immunoprecipitation studies have shown that this polypeptide is distinct from the eight defined influenza virus gene products. A polypeptide of similar peptide composition has been synthesized in vitro in systems programmed with viral mRNA isolated from infected cells and could be precipitated with antiserum against infected cells. Polypeptide 4 was not synthesized in infected cells treated with the amino acid analog p -fluorophenylalanine to restrict viral protein synthesis to that directed by primary mRNA transcripts, nor in an in vitro system programmed with mRNA from infected cells which were treated with cycloheximide to limit viral RNA synthesis to primary transcripts. Thus the available evidence suggests that the synthesis of polypeptide 4 requires “early” protein synthesis. In infected cells, small polypeptides related to the nonstructural polypeptide NS have also been found. A hypothetical scheme for the synthesis of the mRNA for polypeptide 4 is discussed.