- Volume 67, Issue 11, 1986

Infection with u.v.-inactivated P22 bacteriophage at multiplicities higher than 30 caused a decrease in Salmonella typhimurium viability without cell lysis. Neither the action of the endoglycosidase of the P22 virion on the lipopolysaccharide of S. typhimurium nor the concomitant release of cell wall components was responsible for m.o.i.-dependent cell death. Using both free P22 tails and u.v.-inactivated P22, we have shown that p9 tail protein activity has no effect either on the integrity of host cells or on cell viability. Our results show that cell death is due to the injection of the u.v.-inactivated P22 DNA.

The entire nucleotide sequence of genomic DNA was determined for hepatitis B virus (HBV) of subtype ayr, which had been derived from the blood of a Japanese asymptomatic carrier. The genome was 3215 nucleotides long, and differed in DNA sequence by 10% from that of subtypes adw or ayw, but by only 2% from that of subtype adr. Amino acid sequences coded for by the S, C, P and X genes, as well as by the pre-S region, closely resembled those of subtype adr, indicating that the evolution of HBV/ayr from HBV/adr was more recent than the differentiation of the other three subtypes. In the product of the S gene, the mutually exclusive subtypic determinants of the surface antigen, d and y, were associated with variation of amino acid residues at only the 68th and 122nd positions from the N terminus, in contrast to the variation at as many as seven positions for the other set of subtypic determinants, w and r. Sequences representing high local hydrophilicity in the product of the S gene were involved in subtypic variation, although such sequences in the pre-S region were shared by HBV genomes of the various subtypes. In particular, a hydrophilic sequence of 19 amino acid residues, coded for by the pre-S(2) region and implicated in the presumed hepatotropism of HBV, was possessed in common by HBV/adr, HBV/ayr and HBV/ayw, and differed in HBV/adw by only one residue at the 9th position. This amino acid sequence appears to be a promising candidate for a synthetic peptide vaccine.

In this paper the kinetics of hepatitis B virus (HBV) gene expression were investigated in natural and experimentally transfected cell systems. These systems included four human hepatocellular carcinoma cell lines containing HBV DNA (TONG/PHC, HEp 3B2, PLC/PRF/5 and HA22T/VGH) as well as a mouse and a rat cell line both experimentally transfected with HBV DNA. Comparative results on the kinetics of hepatitis B surface antigen in these cell systems suggested that the S gene in the integrated state is expressed at different levels. No human cell line derived from HBV-associated hepatocellular carcinoma produced hepatitis B e antigen (HBeAg) when medium was concentrated by ultrafiltration. In distinct contrast, the two experimentally transfected cell systems produced e antigen at different levels. When all HBV-containing cell lines were grown as tumours in nude mice, no HBeAg was detected in the serum of these mice inoculated with human hepatocellular carcinoma cell lines, in the tumour homogenates, or in the tumour-derived lines, whereas e antigen was expressed both in vivo and in vitro in the experimentally transfected cell lines. These observations indicate that C gene expression is restricted to transfected cell cultures and this suggests a distinct difference in the mechanisms of HBV gene expression between the two types of in vitro model systems.

We analysed the production of interferons (IFN)-alpha and -gamma during the generation of human influenza-virus specific cytotoxic T lymphocyte (CTL) responses using monoclonal antibodies in a specific radioimmunoassay. The results showed that the peripheral blood mononuclear cells (PBM) of all donors tested produced IFN-gamma and had influenza A virus-specific CTL activity after stimulation. The amount of IFN-gamma produced and the level of CTL activity were significantly correlated. The PBM of some donors also produced IFN-alpha. The level of IFN-gamma produced was low during the first few days and increased subsequently, but IFN-alpha, when it was detected, was produced on day 1. The kinetics of the increase in IFN-gamma correlated with the increase in CTL activity. We also observed an increased percentage of cells bearing interleukin-2 receptors, which may have been a response to the production of IFN-gamma. The T cells active in lysing influenza A virus-infected target cells and in producing IFN-gamma were determined after separating effector cells with monoclonal antibodies. The CTL effector cells were mainly in the T8+ subset, but IFN-gamma-producing cells were found in both T4+ and T8+ subsets. These results suggest that influenza virus-specific T8+ CTL produce IFN-gamma in response to virus, and that T4+ cells which are not CTL effectors also produce IFN-gamma after restimulation with influenza A virus-infected cells.

We describe here a mouse model for rhinovirus infection using a variant of human rhinovirus type 2 (HRV2/H) which replicated 50- to 300-fold in the lungs of BALB/c mice. The variant virus differed only marginally from HRV2/H according to various biochemical parameters. Use of a photosensitive inoculum and pretreatment of the animals with actinomycin D were necessary for detection of reproducible and significant levels of virus replication. This mouse model of rhinovirus infection is the first example of human rhinovirus replication in a non-primate mammal, and provides an important link for the development of rhinovirus therapy or prophylaxis.

HeLa cells generally do not respond well to interferon (IFN). We have used is-1, an IFN-sensitive mutant of mengovirus, to select a clone of IFN-responsive HeLa cells (F-H12). At moderate levels of human α/β IFN, is-1 yields were fivefold lower in these cells than in similarly protected control cells. In contrast, wild-type mengovirus, vesicular stomatitis virus and a wild-type and thymidine kinase-negative strains of herpes simplex virus type 1 grew equally well in both cell lines. By a cell survival assay, the F-H12 line was up to 100 times more responsive to IFN than the parental line when challenged by is-1. 2′-5′-Oligo(A)-dependent endonuclease activity was the same in both lines. These observations cannot be accounted for by enhanced induction of IFN following infection.

We have determined the DNA sequence of the herpes simplex virus type 1 (HSV-1) gene encoding the immediate early protein IE110, which is involved in transcriptional activation of later virus genes. The locations of the 5′ and 3′ termini of IE110 mRNA, together with the positions of two introns, were identified. Examination of the DNA sequence suggested that translation starts at the first ATG after the 5′ terminus of the mRNA, and that both introns occur in protein-coding sequence. The predicted IE110 polypeptide contains 775 amino acids, and has a molecular weight of 78452. It contains a cysteine-rich region resembling regions found in several proteins which interact functionally with DNA. An antiserum was raised to the predicted C terminal amino acid sequence of the IE110 polypeptide and was shown to immunoprecipitate the native protein from HSV-1-infected cell extracts. The functional importance of regions of the protein was evaluated by construction of frameshift and deletion mutants of a plasmid-borne IE110 gene. The mutants were tested for IE110 function by short-term transfection assays, and the results were correlated with the DNA sequence and RNA mapping studies.

Seven virus-specific RNA transcripts have been identified in tumours induced by bovine papillomavirus type 4 (BPV-4). The RNAs measured 4.2, 3.6, 3.0, 2.8, 1.9, 1.6 and 1.0 kilobases (kb). They were mapped on the viral genome by Northern blot hybridization to subgenomic probes, by cDNA hybridization to viral DNA fragments and by S1 analysis of unlabelled and 3′ and 5′ end-labelled DNA fragments. All the RNA species are transcribed from the same DNA strand, are polyadenylated and with the exception of the 1.0 kb RNA internally spliced. The 3.0, 1.9, 1.6 and 1.0 kb RNAs share the same 3′ polyadenylation site at nucleotide 4009 whereas the 4.2 kb RNA and the 2.8 kb RNA terminate near a polyadenylation site at nucleotide 7187. The 4.2 kb and the 2.8 kb RNAs are transcribed from the late open reading frames and encode the structural polypeptides; the 3.0, 1.9, 1.6 and 1.0 kb RNAs are transcribed from the early open reading frames and are the transcripts involved in viral replication and cellular transformation.

The organization of proviruses derived from infecting transforming viruses rescued from hamster tumour cells was studied. Southern blot analysis indicated that the provirus from the F6 cell line was organized as long terminal repeat (LTR)-src-LTR, and S1 mapping experiments suggested that it was probably derived by reverse transcription of src mRNA followed by integration. In the E6 cell line, the provirus unit was arranged as LTR-Δ gag-src-LTR, indicating a recombination event between the rescued transforming virus and the helper virus. These results suggest that transforming defective viruses containing only the src gene can be rescued from non-permissive mammalian cells.

Synthetic peptides have been made corresponding to the C-terminal portion of each of the three presumptive genome-linked proteins (VPgs) of foot-and-mouth disease virus type A10. Antisera against each of these peptides efficiently precipitated only the homologous VPg, and the reactions were inhibited by prior absorption with homologous, but not heterologous synthetic peptide. The peptide antisera precipitated a number of proteins from infected cell extracts with mol. wt. of 100, 84, 56, 36, 27, 25 and 20, all × 103; all these reactions were inhibited by absorption with homologous peptide, indicating that they were probable precursors of VPg. The relationship between these proteins is at present unclear.

Non-capsid poliovirus proteins of the P2 region in extracts of infected cells were solubilized by SDS, separated by PAGE, electroeluted from the gels and used to immunize mice. The sera obtained were rendered monospecific by extensive absorption with proteins from uninfected cell extracts, and the spleen cells of these animals were used to establish antibody-secreting hybridomas. The monoclonal and the monospecific antibodies recognized denatured antigen (proteins 2C, 2BC and P2) in ELISA, immunoblot and immunoprecipitation tests and also combined with the native proteins of the P2 region in immunoprecipitation. In addition, the antibodies could be used successfully for immunofluorescence and electron microscopic immunocytochemistry.

The interaction between the flavivirus West Nile virus (WNV) and cells of the mouse macrophage-like cell line, P388D1, was assayed by transmission electron microscopy, by following the association of [35S]methionine-labelled virus with cells, and by using a radiobinding assay with an 125I-labelled F(ab′)2 fragment of a monoclonal antibody directed against the major viral envelope surface glycoprotein. Using electron microscopy, both fusion and endocytosis were observed at pH 6.4, but at pH 8.0 only endocytosis was observed. When 35S-labelled WNV was bound to the P388D1 cell surface at 0 °C, less virus eluted on warming to 37 °C at mildly acidic than at alkaline or neutral pH values. The monoclonal antibody fragment had an increased affinity for cell surface viral E glycoprotein after prebound WNV was warmed at mildly acidic pH values. It is proposed that the warming of cell-virus mixtures at low pH results in fusion with a consequent reduction in elution of virus and an increase in the recognition of cell surface-expressed viral envelope glycoprotein by labelled antibody.

The presence of five structural proteins of measles virus in brain material obtained at autopsy from four patients with subacute sclerosing panencephalitis (SSPE) was examined by immunofluorescence employing monoclonal antibodies. In addition, the humoral immune response against measles virus antigens in serum and cerebrospinal fluid was analysed by immunoprecipitation in combination with gel electrophoresis, revealing a reduced response mainly to the matrix (M) protein. In none of the brain material were all five structural proteins simultaneously detected. Nucleocapsid protein and phosphoprotein were found in every diseased brain area, whereas haemagglutinin (H) protein was detected in two, fusion (F) protein in three and M protein only in one SSPE case. In two cases, variations in the occurrence of H and F proteins could be observed between regions displaying different degrees of neuropathological changes. No correlation was observed between the humoral immune response and the immunohistological findings. These data support the hypothesis of a restricted synthesis of measles virus proteins, in particular the envelope and M proteins, in SSPE.

With a reassortant from a cross of human rotavirus DS-1 (serotype 2) and OSU (serotype 5) it was determined that the OSU major neutralization glycoprotein antigen (VP7) was encoded by gene segment 9. A full-sized cloned cDNA copy of the OSU gene 9 was produced and sequenced. Hybridization of such labelled cDNA with the corresponding segment of a reassortant DS-1 × OSU virus confirmed the coding assignment. Comparison of the deduced amino acid sequence of the VP7 of OSU with those previously determined for five other rotavirus strains, representing four distinct serotypes, revealed some hydrophilic regions that exhibited significant homology and other hydrophilic domains with greater amino acid divergence. In one of the latter hydrophilic domains each of the five serotypes had a distinct amino acid substitution at residue 146, suggesting that it may be involved in serotype specificity.

Five neutralizing monoclonal antibodies produced against human rotavirus (HRV) serotypes 1, 2, 3 and 4 and the simian rotavirus (SA11) were used to study 59 rotavirus isolates of human, simian and feline origin previously serotyped using polyclonal antisera. In neutralization tests, 19 of 26 HRV serotype 1 isolates, both strains of HRV serotype 2, 14 of 24 HRV serotype 3 isolates and all of seven serotype 4 isolates were neutralized by the homologous serotype-specific monoclonal antibodies. Use of the panel of monoclonal antibodies revealed antigenic differences between strains within serotypes 1 and 3 and, in the case of the serotype 3 strains, each variant had a unique RNA electropherotype. An enzyme immunoassay (EIA) which utilized the monoclonal antibodies essentially confirmed the neutralization results. Preliminary results show that direct serotyping in faecal extracts by EIA using these monoclonal antibodies is specific but lacks sensitivity.

Using a murine monoclonal antibody (RF-HBs-1) which has been shown to be capable of neutralizing both ad and ay subtypes of hepatitis B virus (HBV), we have devised a competitive inhibition assay to measure the presence of virus-neutralizing antibodies in the sera of patients who have recovered from acute type B hepatitis. The majority of patients have this antibody in their serum. We also show that this antibody inhibits the binding of polymerized human serum albumin (pHSA) to the pHSA receptor site of the HBV particle, which has been proposed as an important site for the entry of HBV into liver cells. We have demonstrated that the epitope recognized by this antibody is dependent on the linkage of 24000 and 28000 mol. wt. polypeptides via a disulphide bond. This conformational determinant in the coat of the virus which is part of or near to the pHSA binding site is important in evoking a virus-neutralizing response.

Using [3H]glucosamine and [3H]mannose labels, two virus-specific glycosylated polypeptide species with M r values of about 200000 (200K) and in the 75K to 100K range, respectively, were recognized in Berne virus-infected embryonic mule skin cells. In purified virions only the latter glycoprotein occurred. Concanavalin A was bound to the virion as evidenced by reduction in infectivity. Analyses using SDS-PAGE, blotting and glycoprotein identification with concanavalin A and horseradish peroxidase showed coincidence of the virion glycoprotein signals with the maximum infectivity and haemagglutinating activity in an isokinetic sucrose gradient. Polyclonal rabbit immune serum and a neutralizing and haemagglutination-inhibiting monoclonal antibody raised against Berne virus recognized both the 75K to 100K and the ‘200K’ glycoproteins. Using tunicamycin, a concentration-dependent inhibition of infectivity was noted; however, non-infectious particles containing the two major polypeptides (20K and 22K) were released from the cells in small quantities. The glycoproteins were absent from cytoplasmic extracts and a novel polypeptide of about 150K was identified instead. Translation of poly(A)-selected intracellular RNA from infected cells in a rabbit reticulocyte cell-free system also resulted in the appearance of a new high M r polypeptide (about 170K). Using pulse-chase labelling and radioimmunoprecipitation, suggestive evidence for a precursor-product relationship between the intracellular ‘200K’ and the virion glycoproteins has been obtained. These experiments identify the N-glycosylated proteins in the 75K to 100K range as constituents of the peplomeric envelope projection of Berne virus; they probably arise by post-translational processing of a 150K to 170K precursor molecule involving glycosylation and subsequent cleavage.

Berne virus possesses haemagglutinating activity which is inhibited by antisera that neutralize the infectivity of the virus. In decreasing order, human, rabbit and guinea-pig erythrocytes were agglutinated whereas agglutination was not observed with rat, goose, chicken or horse red blood cells. This pattern is different from that seen with the closely related Breda virus of cattle. Haemagglutinin was found to co-sediment with viral infectivity in sucrose density gradients. Transmission electron microscopy showed that intact virus particles form bridges between adjacent erythrocytes. The viral envelope was seen at a distance from the erythrocyte surface suggesting that the peplomers possess haemagglutinating activity. Haemagglutination was decreased in the presence of fetuin and gangliosides and also by pretreatment of the erythrocytes with periodate, suggesting that the virus binds to glycoproteins and/or glycolipids on the erythrocyte surface.

Human embryonic lung cells were infected with herpes simplex virus (HSV), treated with 10 µg/ml or more of cycloheximide for 24 h, incubated at 37 °C, and then shifted to 40.5 °C for various periods of time (0 to 40 days) without cycloheximide treatment. No infectious virus was detected after freezing and thawing of the cultures; however, infectious virus was recovered after temperature shift-down to 37 °C or superinfection with human cytomegalovirus (HCMV). The time course for formation of infectious centres after temperature shift-down was examined with and without HCMV superinfection during incubation at 40.5 °C. Two patterns of latently infected cells were identified: one pattern showed spontaneous reactivation of virus after temperature shift-down, and the second showed reactivation of HSV after superinfection with HCMV. The first pattern showed a rapid decrease in the number of infectious centres with time, whereas the second maintained a steady reactivation rate up to 40 days at 40.5 °C. The same tendency was observed for infectious centre formation at 37 °C with and without HCMV superinfection in the HSV latency system established with (E)-5-(2-bromovinyl)-2′-deoxyuridine and interferon treatment.