- Volume 78, Issue 5, 1997

Two simian immunodeficiency virus strain mac (SIVmac)/human immunodeficiency virus type 1(HIV-1) chimeric viruses (SHIVs), designated NM-3 and NM-3n, with env derived from HIV-1 and defective vpr (plus defective nef for NM-3), were inoculated into seven macaques. These macaques were transiently or persistently infected and most of them produced long-lasting neutralizing antibodies and Env-specific killer T cells to HIV-1 with no AIDS-like symptoms. When they were challenged with another SHIV with intact vpr and nef (designated NM-3rN), all were protected as judged by virus recovery, DNA detection by PCR and antibody responses. Anti-HIV-1 Env-specific killer T cells were considered to have played a major role in this protection, but a non-specific defence mechanism as well as specific immunity also appeared to be involved. Thus, these two non-pathogenic SHIVs induced long-lasting protective immunities in macaques, suggesting the possibility of gene- defective SHIVs as attenuated live vaccines for human use.

Human adenovirus type 5 can be used as a vector to elicit immune responses to antigens expressed from heterologous DNA sequences incorporated into the viral genome, for example in mice immunized intraperitoneally. We have used a recombinant adenovirus which expresses the p55gag antigen of simian immunodeficiency virus to evaluate the nature and longevity of the response elicited when administered to mice by alternative routes which translate more readily to larger animals and man. In C57Bl/6 mice immunized orally with a single dose of virus, a majority of the animals which showed evidence of responding to the immunogen by producing an anti-adenovirus response also pro-duced a plasma antibody response to Gag which persisted for more than 1 year and a Gag-specific cytotoxic T cell response that could be detected for at least 6 months. In a minority of similarly immunized responding animals, only a cytotoxic response to Gag was observed although both humoral and cellular responses to adenovirus antigens were seen; intranasal immunization produced a Gag- specific response similar to this latter pattern. These findings suggest that delivery of adenovirus recombinants orally or intranasally may be a useful strategy for eliciting long-term cytotoxic T cell memory responses in splenocytes to candidate vaccine antigens.

Recently, we and others have shown that the interaction between envelope specific antibodies and primary human immunodeficiency virus type 1 (HIV-1) isolates may result in either inhibition or enhancement of virus entry. The outcome proved to be determined by the virus isolate rather than by the specificity of the antiserum used. To study the mechanism underlying this phenomenon, a series of HIV-1 envelope glycoproteins from closely related primary virus isolates of different syncytium inducing phenotypes, together with chimeras of these proteins, were tested in an envelope transcomplementation assay for their sensitivity to either antibody mediated inhibition or enhancement of HIV-1 entry. Based on the observation that, in contrast to the inhibition of HIV-1 entry, antibody mediated enhancement was not temperature dependent and could not be mediated by F(ab) fragments, we concluded that the mechanisms underlying these phenomena are different and that antibody mediated enhancement of HIV-1 entry is largely if not exclusively mediated by HIV-1 glycoprotein cross-linking. The susceptibility of the envelope glycoprotein chimeric viruses to neutralization or enhancement of infectivity proved to be primarily determined by the configuration of the V3 loop, and the affinity of the antibodies to monomeric HIV-1 gp160 molecules, proved to be of quantitative importance only. One human monoclonal antibody directed against gp41 (IAM 2F5) inhibited entry of all the viruses studied, irrespective of their phenotype, and directly proportional to its affinity to monomeric HIV-1 gp160.

The effect of 3′-azido-3′-deoxythymidine (AZT) on in vitro infection of peripheral blood mononuclear cells (PBMCs) isolated from normal adult individuals with human T cell leukaemia/lymphoma virus type I (HTLV-I) was evaluated. Different PBMC samples were exposed to HTLV-I by cocultivation with MT-2 (a chronically infected cell line) in the presence of 20 U/ml of human recombinant interleukin 2 (IL-2) and graded concentrations of AZT. Control and drug-treated cultures, of both infected and uninfected PBMCs, were then grown for several weeks and monitored for virological and immunological parameters. The results showed a concentration- dependent anti-proliferative effect of AZT in both infected and non-infected cultures. Production of both proviral DNA and viral RNA was inhibited not only at the higher concentrations of AZT (8 μM and 32 μM) but also at concentrations as low as 0·12 μM. These results were confirmed by PCR and by flow cytometry analysis for the viral core protein p19. Moreover, treatment with AZT resulted in a decreased expression of CD25 in cultures exposed to HTLV-I as well as in non-infected PBMCs. On the other hand, HLA-DR was down-regulated to a greater extent in drug-treated, virus-exposed cultures in comparison with those not infected. No evidence of the antiviral activity of AZT was observed in PBMC cultures already infected by HTLV-I or in MT-2 cells. These findings demonstrate that treatment with AZT, when given at the time of infection with HTLV-I, has a marked protective effect on PBMCs.

The genetic relationships between type 1 polioviruses circulating in sub-Saharan Africa during the past decade have been investigated by partial genomic sequencing across the VP1/2A region of the polioviral genome. Sequencing templates were generated by single-step reverse transcription PCR amplification of the viral RNA using poliovirus-specific primers. Seven poliovirus genotypes, circulating in different geographical regions during different periods, were identified. Considerable genetic variation was exhibited by strains within several of these genotypes, indicative of sustained endemic transmission within individual countries. Two genotypes appear to be circulating in Africa at present; one major genotype, which has been in circulation since at least 1980, covers a wide geographical region and includes countries in western, central and southern Africa. Within this genotype are several smaller clusters, possibly representing strains in the process of evolving into new genotypes. The second genotype presently in circulation has been found only in Tanzania and Zambia to date, associated with a relatively small number of cases. Imported genotypes, introduced from the Middle East and the Indian subcontinent, have also recently been in circulation in eastern and central Africa. In South Africa, three genotypes, one unique to the country and the others imported from west Africa and the Middle East, co-circulated endemically between 1980 and 1985. A fourth genotype, introduced from countries to the north, displaced the endemic strains and continued to circulate until 1989. This study has generated a meaningful overview of the endemic circulation and regional transmission of type 1 polioviruses throughout sub-Saharan Africa.

Natural and experimental hosts infected with foot-and-mouth disease virus (FMDV) develop a long-lasting immune response that is closely related to the presence of anti-FMDV antibodies (Ab). We show here that spleen cells from animals which had been infected 3 or more months previously induced an anti-FMDV-Ab response in untreated animals which lasted more than 210 days after cell transfer. Persistence of infectious virus was excluded since virus isolation or detection of the viral genome by PCR in donor splenocytes were consistently negative. The role of antigen presentation (AP) in this phenomenon was studied in vivo by using irradiated splenocytes from virus-sensitized donor mice. Although these irradiated cells were unable to induce anti-FMDV-Ab in normal or irradiated recipient mice, they elicited a strong secondary reaction in FMDV-pre-sensitized recipients. The presence of AP cells (APC) presenting FMDV epitopes (FMDV/APC) was also analysed in mice sensitized to FMDV in different ways. A close correlation between FMDV/APC and the presence of anti-FMDV-Ab was found in infected mice as well as in mice immunized with different doses of inactivated virus, with or without adjuvants. Experiments in vivo and in vitro showed that the APC activity can be specifically blocked with either anti-MHC class II monoclonal antibody or anti-FMDV antiserum, and is dependent on the presence of T cell function. These results strongly suggest that persistent FMDV/APC are responsible for the existence and maintenance of an anti-virus immune response regardless of the immunization method used.

125I protein labelling of oligo(dT)-selected RNAfrom feline calicivirus (FCV)-infected cells revealed that the genomic and 2·4 kb subgenomic RNAs of FCV are linked to a 15 kDa protein (VPg). Proteinase K treatment of FCV RNA, to remove VPg, led to a decrease in the translatability of the RNA, but there was no obvious change in the site of RNA initiation. Addition of the cap analogue 7-methylGTP to in vitro translations had no effect on the translation of FCV RNA, suggesting that FCV RNA is translated by a cap-independent mechanism. Further evidence that FCV RNA is translated by an unusual mechanism was obtained by translating FCV RNA in vitro at a range of K concentrations. FCV RNA was able to direct translation at K concentrations at which cellular RNA translation was inhibited.

Cross-infection studies of normal calves infected with homologous pairs of non-cytopathic and cytopathic bovine viral diarrhoea virus (BVDV) showed significant differences in both humoral and cell- mediated immune responses against either biotype over a period of 5 months. Serological assays after primary intranasal inoculation showed striking significant (P < 0·05) differences between biotypes. Neutralizing titres were detected earlier and were much higher with the non-cytopathic strain than with the homologous cytopathic strain. Significant biotype-specific differences were also observed in the lymphocyte proliferative responses of cattle following in vitro stimulation by non-cytopathic/ cytopathic BVDV and the non-structural p80 protein (NS3). The secondary immune response seems to be largely influenced by the biotype used for the primary inoculation and only to a lesser extent by the biotype inoculated for the second time after an interval of 91 days. Animals exposed twice to the cytopathic biotype, which exhibited the lowest antibody titres, showed evidence of BVDV-specific cell- mediated immunity as measured by lympho- proliferation against BVDV. In contrast, the antibody response in the subgroup of animals inoculated twice with homologous non-cytopathic virus was inversely correlated with the proliferative responses. These differences in the immune response were not readily apparent for the two other remaining subgroups which had received cytopathic or non-cytopathic biotypes alone following the second inoculation with non-cytopathic or cyto- pathic viruses, respectively. Taken together, these data suggest that the differences in immune responses against cytopathic or non-cytopathic strains may be due to a Th1/Th2-like regulatory mechanism.

Infectious cDNA clones of two strains of tick-borne encephalitis (TBE) virus, i.e. European subtype prototypic strain Neudoerfl and the closely related but more virulent strain Hypr, were constructed. The recombinant constructs consisted of cDNAs stably inserted into the bacterial plasmid pBR322 under the control of T7 promoter elements. The genome of TBE virus strain Neudoerfl was successfully cloned, both as a full-length cDNA and as two partial cDNAs. In the case of strain Hypr, the genome is represented by two cDNA clones corresponding to the 5′ - and 3′ - terminal halves of the genome. Highly infectious RNAs can be produced from the full-length cDNA clone or from the partial clones ligated in vitro to form full-length cDNA templates prior to T7 transcription. The biological properties of the recombinant progeny viruses, including virulence characteristics, were indistinguishable from the corresponding parent virus strains. Thus, the described infectious cDNA clones represent a useful and reliable experimental system for the specific mutagenesis of TBE virus.

We have previously shown that the influenza virus NS1 protein exhibits stable binding to a model mini vRNA which is stronger than its binding to dsRNA. In this study, we confirmed that the binding depended on a higher-order structure of the model RNA, probably the panhandle structure formed by pairing between the 5′ - and 3′ -terminal common sequences. The formation of the structure was enhanced by the NS1 protein itself. We propose that an A bulge in a stretch of double helix results in a binding site with greater affinity for the NS1 protein.

The cytotoxicT cell (CTL) response in C57/Bl6 (H-2b) mice to rotavirus has been analysed using a cognate set of vaccinia virus recombinants covering the 12 primary gene products of the UKtc strain of bovine rotavirus. The gene products of RNA segments 5 (VP5/NSP-1) and 8 (VP7) both elicited a classic CD8 Class I MHC restricted CTL response. Using L cells transfected with specific Class I MHC loci as targets the VP5/NSP-1 response was found to be restricted at Db and the VP7 response at Kb. Vaccinia virus recombinants expressing VP7 genes from seven G serotypes were used to show that the CTL response to this antigen is completely cross-reactive. By contrast, using the same strategy the CTL response to VP5/NSP-1 was found to be virus strain specific. A vaccinia virus recombinant carrying RNA segment 5 from the deletion mutant P9DA5 was used to localize at least one CTL epitope in VP5/ NSP-1 to the first 150 amino acids of the protein. The expression of a number of fragments of VP7 in vaccinia virus recombinants was used to show that the CTL epitope (amino acids 31–40) previously identified through the use of synthetic peptides is virus serotype specific rather than cross-reactive.

The characteristic tubules that are produced during the orbivirus infection cycle are composed of a major viral nonstructural protein, NS1. To characte`rize the NS1 gene and gene product of African horsesickness virus (AHSV), a full-length cDNA copy of the NS1 gene of AHSV-6 was cloned and the nucleotide sequence determined. NS1 was highly conserved within the AHSV serogroup with between 95–98% conservation of amino acids among NS1 of AHSV-6, AHSV-4 and AHSV-9. The structure of AHSV NS1 tubules was investigated by in vitro translation of the AHSV-6 NS1 gene followed by expression of the gene in insect cells. The NS1 protein assembled in tubular structures with a diameter of approximately 23 nmand lengths of up to 4 μm. The absence of a ladder-like structure and lower sedimentation value of AHSV NS1 tubules clearly distinguished them from those of blue- tongue virus.

Hepatitis B virus (HBV) preS1 fused to the GAL4 DNA-binding domain functioned as a transcriptional activation domain in yeast and mammalian cells. The GAL4-preS1 fusion proteins derived from the preSI of all three tested HBV subtypes (adr, adwand ayw) specifically activated the transcription of a lacZ or chloramphenicol acetyltransferase reporter gene linked to a GAL4-responsive promoter in transient transfection assays using yeast or HepG2 cells, respectively. Deletion analyses showed that the segments of preSI from residues 21 to 90 and from residues 21 to 56 are sufficient and essential for the activity, respectively. Stable expression of GAL4-preS1 in Chinese hamster ovary cells also produced transactivator activity. These results suggest that preS1 fused to any DNA- binding domain of transcription factors would have transactivation potential.

A combination of 9,10-dimethyl-1,2-benzanthracene (DMBA) application and excisional wounding on the lingual tips of Syrian Golden hamsters (Mesocricetus auratus) induces dysplastic and malignant mucosal lesions. Papillomavirus genus- specific antigen and viral particles, measuring 55 nm in diameter, were demonstrated in the nuclei of squamous cells of dysplastic lesions showing koilocytotic change. In this study, we cloned a circular genome at a single KpnI site from one of these dysplastic lesions. The genomic sequence of this clone, consisting of 7647 bp, was shown to be that of a novel papillomavirus with a conserved genomic organization. We named the new virus hamster oral papillomavirus (HOPV). All dysplastic lesions induced by this combination of DMBA application and excisional wounding contained viral DNA. Although Southern blot hybridization analysis could not detect the HOPV genome, PCR analysis demonstrated the latent HOPV genome in the tongue and skin of an untreated hamster. These results suggest that latently present HOPV genome is reactivated by the DMBA/wounding procedures. Lingual HOPV infection may be an important model for gaining insight into the interactions between papillomavirus infection, chemical carcinogens and physical irritations in carcinogenesis or malignant transformation.

We have matched a PCR assay which detects disruptions in the E2 reading frame of human papillomavirus type 16, with RNA in situ hybridization patterns and shown that in 15 out of 16 cervical intraepithelial neoplastic (CIN) III lesions and in 19 out of 19 tumours, the E2 gene is disrupted with no detectable E2 transcripts. Varying levels of E6–E7 transcripts are detected in CIN III lesions, with stronger signals in tumours. The cytokeratin profile of most tumours: cytokeratin 10-, 14- and 19-positive and 4-, 13- and 18negative, is also detected in CIN III lesions. The changes in levels of α2, β1 and β4 integrins, CD44 and E-cadherin occur during the evolution of high-grade CIN lesions. Increases in the levels of expression of CD44 and E6-E7 transcripts, coupled with changes in the cellular localization of the Notch protein, define the transition from CIN III lesions to tumours.

The vagina and medulla of the adrenal gland of mice vaginally infected with herpes simplex virus (HSV) types 1 and 2 were examined in the latent stage of infection (5 to 51 weeks post-infection). RNA in situ hybridization with HSV-1 and -2 latency- associated transcript (LAT) RNA probes resulted in positively stained neuronal cell nuclei in the uterovaginal plexus, but not in the medulla of the adrenal gland. These organs were chosen because HSV antigens can be detected not only in the vaginal epithelium, but also in neurons of the uterovaginal plexus and in the medulla of the adrenal gland at the acute stage of genital infection. To our knowledge, this is the first report describing LATs in neurons of the uterovaginal plexus in the genital tract of latently HSV-infected mice.

Neuronal and lymphoid tissues of 15 randomly selected horses were analysed post mortem by liquid nested-PCR to study the tropism of equine herpesvirus 4 (EHV-4). In four animals the trigeminal ganglia and in one case the lung were positive. Using a direct in situ PCR the EHV-4 genome was localized in the nuclei of neurons and in the bronchiolar as well as alveolar epithelium of the lung. In none of these tissues could infectious virus or viral antigens be detected. Applying the more sensitive liquid RT-PCR, however, an acute infection was demonstrated in one of the trigeminal ganglia by amplification of viral transcripts coding for glycoprotein B. The failure to detect these transcripts in the other trigeminal ganglia and the lung indicates a latent infection. This report formally proves that, like other members of the Alpha- herpesvirinae, EHV-4 establishes latency in the trigeminal ganglia.

The distribution of equine herpesvirus 2 (EHV-2) DNA within neurological and lymphoid tissues from 12 EHV-2 seropositive Welsh mountain ponies was determined by PCR. The lymphoid sites sampled in this study were almost universally PCR positive, thus confirming the existing virus co-cultivation data which suggest that the lymph nodes draining the respiratory tract are the main reservoirs of EHV- 2 DNA. In addition, EHV-2 DNA was also detected, albeit with lower frequency, within both the peripheral and central nervous systems (PNS and CNS) of these animals. Of the CNS sites sampled 11% were PCR-positive and in the PNS the trigeminal ganglion proved PCR-positive in 50% of the animals tested. Since the nasal epithelium is innervated by the maxillary division of the trigeminal nerve, these observations suggest that the trigeminal ganglion may represent a biologically important site for EHV-2 latency.

Intranasal inoculation of BALB/c mice with a recent clinical isolate of equine herpesvirus 2 (EHV-2 strain MR) produced a productive infection characterized by clinical signs, including weight-loss and conjunctivitis, but no mortality. Infectious virus was isolated from the lung, trachea and nasal turbinates with the highest titres present in lung tissue; EHV-2 neutralizing antibody was detected in the serum on day 21 post-inoculation. No infectious virus was detected in neural tissue, blood or lymphoid tissues. An EHV-2-specific nested PCR confirmed the presence of EHV-2 DNA in the respiratory tissues. In addition, EHV-2 DNA was detected in the trigeminal ganglia and olfactory bulbs from the first day after inoculation and these tissues remained positive for EHV-2 DNA for at least 30 days, by which time EHV- 2 DNA had been cleared from all other tissues.