Eiji Konishi

Dengue type 2 virus subviral extracellular particles produced by a stably transfected mammalian cell line and their evaluation for a subunit vaccine.

A dengue subunit vaccine candidate was developed using a mammalian cell line continuously expressing subviral extracellular particles (EPs) of the New Guinea C (NGC) strain of dengue type 2 virus. The cell line, designated D cell line, maintained envelope (E) antigen production for at least 10 passages. The EPs contained an E protein biochemically and antigenically equivalent to authentic E produced by NGC-infected Vero cells. Two immunizations of BALB/c mice with purified EPs containing 100ng or 400ng of E induced moderate levels of neutralizing antibody and anamnestic neutralizing antibody responses were produced when these animals were challenged with dengue virus. The yield of E antigen from D cells was comparable to that from NGC-infected Vero cells. When D cells were transfected with the anti-apoptotic bcl-2 gene, the E antigen release increased approximately two-fold. These results indicate that D cell EPs are a promising non-infectious vaccine antigen for dengue.

A DNA vaccine expressing dengue type 2 virus premembrane and envelope genes induces neutralizing antibody and memory B cells in mice.

A dengue DNA vaccine candidate was developed and evaluated for immunogenicity in mice. The vaccine, designated pcD2ME, is a pcDNA3-based plasmid encoding the signal sequence of premembrane (prM), prM and envelope (E) genes of the New Guinea C strain of dengue type 2 virus. CHO-K1 cells transfected with pcD2ME expressed prM and E as determined by immunochemical staining with monoclonal antibodies. BALB/c mice inoculated intramuscularly with 100 microg of pcD2ME two or three times at an interval of 2 weeks developed a low level of neutralizing antibody (1:10 at a 90% plaque reduction). Immunization twice with 10 microg or 1 microg of pcD2ME or three times with 100 microg of pcDNA3 did not induce detectable levels of neutralizing antibody. Mice immunized two or three times with 100 microg of pcD2ME raised neutralizing antibody titers to 1:40 or greater on days 4 and 8 after challenge with 3x10(5) plaque forming units (PFU) of the New Guinea C strain of dengue type 2 virus, showing strong anamnestic responses to the challenge. In contrast, mice immunized two or three times with 100 microg of pcDNA3 developed no detectable neutralizing antibody on days 4 and 8 after challenge. These results indicate that immunization with pcD2ME induces neutralizing antibody and dengue type 2 virus-responsive memory B cells in mice.

Safety and immunogenicity of NYVAC-JEV and ALVAC-JEV attenuated recombinant Japanese encephalitis virus — poxvirus vaccines in vaccinia-nonimmune and vaccinia-immune humans

A controlled, randomized, double-blind clinical trial evaluated whether two attenuated recombinant poxviruses with identical Japanese encephalitis virus (JEV) gene insertions, NYVAC-JEV and ALVAC-JEV, were safe and immunogenic in volunteers. Groups of 10 volunteers distinguished by vaccinia immune status received two doses of each vaccine. The vaccines appeared to be equally safe and well tolerated in volunteers, but more reactogenic than licensed formalin-inactivated JE and placebo vaccines given as controls. NYVAC-JEV and ALVAC-JEV vaccine recipients had frequent occurrence of local warmth, erythema, tenderness, and/or arm pain after vaccination. There was no apparent effect of vaccinia immune status on frequency or magnitude of local and systemic reactions. NYVAC-JEV elicited antibody responses to JEV antigens in recipients but ALVAC-JEV vaccine poorly induced antibody responses. However, NYVAC-JEV vaccine induced neutralizing antibody responses only in vaccinia-nonimmune recipients while vaccinia-immune volunteers failed to develop protective antibodies (5/5 vs. 0/5 seroconversion, p<0.01). These data suggest that preexisting immunity to poxvirus vector may suppress antibody responses to recombinant gene products.

Induction of Japanese encephalitis virus-specific cytotoxic T lymphocytes in humans by poxvirus-based JE vaccine candidates

Poxvirus-based recombinant Japanese encephalitis (JE) vaccine candidates, NYVAC-JEV and ALVAC-JEV, were examined for their ability to induce JE virus-specific cytotoxic T lymphocytes (CTLs) in a phase I clinical trial. These vaccine candidates encoded the JE virus premembrane (prM), envelope (E) and non-structural 1 (NS1) proteins. The volunteers received subcutaneous inoculations with each of these candidates on days 0 and 28, and blood was drawn 2 days before vaccination and on day 58. Anti-E and anti-NS1 antibodies were elicited in most vaccinees inoculated with NYVAC-JEV and in some vaccinees inoculated with ALVAC-JEV. Peripheral blood mononuclear cells (PBMCs) obtained from approximately one half of vaccines showed positive proliferation in response to stimulation with live JE virus. Cytotoxic assays demonstrated the presence of JE virus-specific CTLs in in vitro-stimulated PBMCs obtained from two NYVAC-JEV and two ALVAC-JEV vaccinees. Cell depletion tests using PBMCs from one NYVAC-JEV recipient indicated that the phenotype of CTLs was CD8+CD4-.

Comparison of protective efficacies of plasmid DNAs encoding Japanese encephalitis virus proteins that induce neutralizing antibody or cytotoxic T lymphocytes in mice.

Mice immunized with a plasmid DNA encoding the premembrane (prM) and envelope (E) proteins of Japanese encephalitis (JE) virus (designated pcJEME) produce neutralizing antibodies and are protected from JE. To determine the role of the immune response to other viral proteins in protection, we constructed plasmid DNAs encoding other JE virus proteins and made a direct comparison among these plasmids using a mouse model. Cytotoxic T lymphocytes (CTLs) were induced by plasmids encoding capsid (C) or nonstructural proteins, NS1, NS2A, NS2B, NS3 or NS5. However, these plasmids provided only a partial protection against intraperitoneal challenge with a lethal dose of JE virus, whereas mice immunized with pcJEME were fully protected. In mice inoculated with CTL-inducing plasmids, high virus titers were detected in plasma immediately (1h) following challenge and in brain on day 4 post-challenge, but no virus infectivity was detected in plasma and brain of pcJEME-immunized mice during the 5 days following challenge. These results indicate that protection provided by the prM/E-encoding DNA consists of neutralizing antibody that prevents virus dissemination from the peripheral site to the brain, and that this antibody-mediated mechanism of protection is more efficient than the immunity induced by plasmids that generate CTL responses capable of killing JE virus-infected cells.

Ratios of subclinical to clinical Japanese encephalitis (JE) virus infections in vaccinated populations: evaluation of an inactivated JE vaccine by comparing the ratios with those in unvaccinated populations.

Japanese encephalitis (JE) virus is characterized as a virus that produces a large number of subclinical infections. In this report, we estimated a ratio of subclinical to clinical infections in vaccinated human populations who acquired natural infection with JE virus, and evaluated protective capacity of the currently approved inactivated JE vaccine by comparing the ratio with those reported for unvaccinated populations. We developed a sensitive immunostaining method for detecting nonstructural 1 (NS1) antibody to demonstrate JE virus infection in vaccinated individuals. Serum samples collected from human populations in western Japan showed NS1 antibody prevalences of approximately 10% in an urban area in 1981 and 1995 and 20% in a rural area from 1982 through 1983. Analysis of annual change in NS1 antibody titer using paired samples provided a mean duration of NS1 antibody responses of approximately 2 years, indicating that 5% of the urban population or 10% of the rural population acquired natural JE virus infection in 1 year. Based on the number of JE cases from 1982 through 1991 and the number of people acquiring natural infection, and on the assumption that annual infection rates obtained in the present study areas are representative of the infection rate in entire Japan except for non-endemic northern areas, the ratio of subclinical to clinical infections in vaccinated populations was estimated to be 2000000:1, which was 2000-80000 times higher than the ratio previously reported for unvaccinated populations.

Construction and evaluation of a chimeric pseudoinfectious virus vaccine to prevent Japanese encephalitis.

Multiple vaccines exist to control Japanese encephalitis (JE), but all suffer from problems. We have developed a new type of flavivirus vaccine, a pseudoinfectious virus (RepliVAX WN) that prevents West Nile virus (WNV)-induced disease. Here, we describe production of a chimeric RepliVAX (RepliVAX JE) that expresses the JE virus (JEV) prM and E proteins. Our prototype RepliVAX JE replicated poorly in cells, but blind passage produced a better-growing derivative, and analyses of this derivative allowed us to engineer a second-generation RepliVAX (RepliVAX JE.2) that grew to high titers. RepliVAX JE.2 elicited neutralizing antibodies in both mice and hamsters and provided 100% protection from a lethal challenge with JEV or WNV, respectively. These results demonstrate the utility our RepliVAX platform for producing a JE vaccine.

Enzyme-linked immunosorbent assay using recombinant antigens for serodiagnosis of Japanese encephalitis

Recombinant Japanese encephalitis (JE) virus proteins were evaluated as antigens for serodiagnosis of JE using an enzyme‐linked immunosorbent assay (ELISA). The premembrane/membrane (prM/M) and envelope (E) proteins of JE virus were expressed in HeLa cells infected with a recombinant vaccinia virus that encodes the JE virus prM and E genes and were released from cells in a particulate form. The particulate antigens were partially purified from culture fluid from the infected cells by precipitation of particles with polyethylene glycol and then dissociated from the particles with 0.1% Triton X‐100. This antigen preparation was used to evaluate one preimmune and two postvaccination sera from 20 volunteers given three inoculations of the commercial JE vaccine (Biken vaccine) by a conventional ELISA. The results from this assay correlated with neutralization data. The results of an lgM capture ELISA carried out with the recombinant antigen also correlated with the results of an existing lgM capture ELISA performed with JE virus‐infected mouse brain, when tested with 29 serum and 13 cerebrospinal fluid samples from JE patients. These results indicated that recombinant JE virus antigens are useful for ELISA as an antigenically equivalent, highly productive, and safe alternative to authentic JE virus antigens.

Particulate vaccine candidate for Japanese encephalitis induces long-lasting virus-specific memory T lymphocytes in mice

We previously reported that extracellular particles (EPs) composed of premembrane (prM) and envelope (E) proteins were released from cells infected with recombinant vaccinia viruses encoding Japanese encephalitis (JE) virus prM and E genes. In the present study, EPs were evaluated for induction of JE virus-specific antibody and specific T lymphocytes in mice. Six- to 8-week-old male Balb/c mice were inoculated intraperitoneally once or twice (at a 3-week interval) with purified EPs containing 1 microgram of E without adjuvant. Neutralizing antibody was detected and spleen cells proliferated against JE viral antigen 3 weeks after the second immunization with EPs. Neutralizing antibody and JE virus-specific T lymphocytes were also detected 10 months after immunization with EPs containing 2 micrograms of E. Spleen cells obtained from EP-immunized mice and stimulated in vitro with live JE virus, expressed JE virus-specific cytotoxic activity. The cytotoxic activity was reduced by treatment with anti-CD3 antibody and complement. These results indicate that immunization with EPs induces long-lasting specific antibody and memory T cells in mice.

Evidence for antigen production in muscles by dengue and Japanese encephalitis DNA vaccines and a relation to their immunogenicity in mice.

This study demonstrated viral antigen production in muscle tissues following inoculation with DNA vaccines and examined its relation to antibody induction in mice using the flavivirus system. To achieve detectable levels of antigen production, we used a needle-free jet injector and examined 10% homogenate of quadriceps muscle for viral antigens in a sandwich enzyme-linked immunosorbent assay. We compared DNA vaccines against dengue type 1 (designated pcD1ME), dengue type 2 (pcD2ME) and Japanese encephalitis (pcJEME). The amounts of viral envelope (E) antigen contained in muscle homogenate 1, 2, 3 and 4 days following inoculation with 50 microg of pcJEME were 1.1, 1.0, 0.3 and <0.1 ng/ml, respectively. Muscles from pcD2ME- and pcD1ME-inoculated mice did not contain detectable levels of E antigen (<0.1 ng/ml) during 4 days following inoculation. The E amounts released from Vero cells transfected with DNAs were in the order pcJEME>pcD2ME>pcD1ME. Levels of neutralizing antibody induced by two immunizations with 100 microg of each DNA vaccine using needle-free or normal needle/syringe injection systems also were in the order pcJEME>pcD2ME>pcD1ME, 2-11 weeks after the first immunization. However, the difference in antibody levels among three DNA vaccines 14-18 weeks after immunization was smaller than that in the early phase of immunization. These results provide fundamental information useful for developing combination DNA vaccines, such as a dengue tetravalent DNA vaccine, which require adjustment of immunogenicity of each component.