|Home | About | Journals | Submit | Contact Us | Français|
Measles remains a major cause of child mortality, in part due to an inability to vaccinate young infants with the current live attenuated virus vaccine (LAV). To explore new approaches to infant vaccination, chimeric Venezuelan equine encephalitis/Sindbis virus (VEE/SIN) replicon particles were used to express the hemagglutinin (H) and fusion (F) proteins of measles virus (MV). Juvenile rhesus macaques vaccinated intradermally with a single dose of VEE/SIN expressing H or H and F proteins (VEE/SIN-H or VEE/SIN-H+F, respectively) developed high titers of MV-specific neutralizing antibody and gamma-interferon (IFN-γ)-producing T cells. Infant macaques vaccinated with two doses of VEE/SIN-H+F also developed neutralizing antibody and IFN-γ-producing T cells. Control animals were vaccinated with LAV or with a formalin-inactivated measles vaccine (FIMV). Neutralizing antibody remained above the protective level for more than 1 year after vaccination with VEE/SIN-H, VEE/SIN-H+F, or LAV. When challenged with wild-type MV 12 to 17 months after vaccination, all vaccinated juvenile and infant monkeys vaccinated with VEE/SIN-H, VEE/SIN-H+F, and LAV were protected from rash and viremia, while FIMV-vaccinated monkeys were not. Antibody was boosted by challenge in all groups. T-cell responses to challenge were biphasic, with peaks at 7 to 25 days and at 90 to 110 days in all groups, except for the LAV group. Recrudescent T-cell activity coincided with the presence of MV RNA in peripheral blood mononuclear cells. We conclude that VEE/SIN expressing H or H and F induces durable immune responses that protect from measles and offers a promising new approach for measles vaccination. The viral and immunological factors associated with long-term control of MV replication require further investigation.
Measles remains a major cause of child mortality despite the availability of a safe and effective live attenuated virus vaccine (LAV). Recent efforts to improve routine vaccination and implement national immunization days have moved measles control toward the World Health Organization's goal of a 90% reduction in mortality by 2010 compared to 2000 (7). One persistent impediment to measles control in many countries remains the inability to successfully immunize young infants due to the immaturity of the immune system and interference of maternal antibodies with immune responses to LAV (1, 15, 65).
Because the decrease in maternal antibody varies from one infant to another, many children in areas with high measles virus (MV) transmission rates are at risk of acquiring measles prior to vaccination (3, 5, 12). Immaturity also affects the quality and quantity of antibody produced in response to the current vaccine, with lower levels of neutralizing antibody and deficient avidity and isotype maturation in younger than in older infants (15, 16, 37, 59). As a result, the recommended age for vaccination is generally 9 months in developing countries to balance the risk of infection with the likelihood of response to the vaccine (24).
A vaccine that could be given to children under the age of 6 months would improve measles control by allowing delivery with other infant vaccines and by closing the window of susceptibility prior to delivery of the current vaccine. Increasing the dose of LAV improved the antibody responses in young infants but resulted in an unexpected increase in mortality for girls, so this is not an acceptable approach to lowering the age of vaccination (18, 26, 29). Experience with a formalin-inactivated measles vaccine (FIMV) in the 1960s also led to unexpected complications. FIMV provided only short-term protection, and vaccinated individuals were at risk for more severe disease (atypical measles) upon infection with wild-type MV (14, 36, 54). Therefore, other strategies are necessary for development of a vaccine for young infants.
One particularly promising approach for delivery of vaccine antigens is the use of alphavirus replicon particles (55). Alphaviruses are small positive-strand RNA viruses with the nonstructural replicase proteins encoded in the 5′ two-thirds of the genome and the structural proteins in the 3′ one-third. A subgenomic promoter is used to synthesize an abundant, smaller RNA from which the structural proteins are translated (61). Replicons contain the nonstructural protein genes, the 5′ and 3′ end cis-active replication sequences, and the subgenomic promoter that directs expression of a heterologous gene rather than the viral structural proteins. The replicon RNA can be packaged into virus-like particles by providing the structural proteins in trans using transient transfection (6, 33) or with stable packaging cell lines (51) and can be engineered for efficient delivery to antigen-presenting cells (17). Advantages include high-level expression of the vaccine antigen (68), stimulation of innate immunity (25, 31, 32, 64), and general lack of preexisting immunity in the human population.
MV encodes six structural proteins of which two, hemagglutinin (H) and fusion (F), are surface glycoproteins involved in attachment and entry. Antibodies that inhibit MV infection in neutralization assays are directed primarily against the H protein, which also contains important CD8+ T-cell epitopes (39, 41). Nonhuman primates, particularly rhesus macaques, develop a disease similar to that of humans and offer the opportunity for assessing both protection from wild-type MV challenge and priming for enhanced disease after immunization with new experimental vaccines (2, 48, 50, 66). Because protection from measles correlates best with the quality and quantity of neutralizing antibodies at the time of exposure (9, 50), most experimental vaccines have used H alone or H and F for induction of MV protective immunity (44, 50, 65, 70).
Alphaviruses that have been used for construction of replicon particle vaccines include Sindbis virus (SINV) (6, 68), Semliki Forest virus (33), and Venezuelan equine encephalitis virus (VEEV) (53). Each of the alphavirus vectors studied has its own advantages and disadvantages. For instance, VEEV replicon particles have high levels of gene expression (47), but vaccine production is disadvantaged by the requirement for biosafety level 3 manufacturing. SINV replicon particles avoid the safety concerns of VEEV, but expression levels are lower. Previous studies of a SINV-based replicon particle vaccine expressing MV H (SIN-H) in macaques showed good induction of neutralizing antibody and T-cell responses and protection from rash (44). However, vaccinated monkeys developed viremias after challenge, indicating that they were not protected from infection. In this study, we sought to improve the alphavirus replicon particle approach to vaccination for measles by using a chimeric VEE/SIN vaccine (47) expressing both the MV H and F proteins.
Ten juvenile (2 years old) and three infant (1 to 2 months old) MV-naïve rhesus macaques (Macaca mulatta) from the Johns Hopkins Primate Breeding Facility were studied. All monkeys were chemically restrained with ketamine (10 to 15 mg/kg of body weight) during procedures. All animals were maintained and studies were performed in accordance with experimental protocols approved by the Animal Care and Use Committee for Johns Hopkins University.
The H and F genes from the Edmonston strain of MV (gift from Paul Rota, Centers for Disease Control and Prevention) were inserted into the VCR-Chim2.1 alphavirus replicon backbone (47). VEE/SIN chimeric replicon particles expressing MV H and MV F (VEE/SIN-H and VEE/SIN-F, respectively) were generated as described previously, using a BHK-21 cell-based production system with defective helper constructs separately encoding the SINV capsid and envelope glycoproteins (47). Replicon particles were harvested as culture supernatant fluids, clarified by centrifugation, and purified by Fractogel EMD SO3− (M) (s-Fractogel; EM Industries) cationic exchange chromatography. The eluted VEE/SIN-H and VEE/SIN-F replicon particles were diluted into phosphate-buffered saline (PBS), pH 7.4, containing 40 mg/ml lactose, subjected to 0.2-μm-pore-size filtration, and stored at −80°C. Replicon particle titers were determined by titration in a reporter cell line that expresses β-galactosidase in response to replicon-encoded nonstructural proteins, as described previously (44). Expression of MV proteins was verified by immunoblot analysis. Preparations of VEE/SIN-F replicon particles were approximately 10-fold lower in titer than preparations of VEE/SIN-H particles. All materials used for vaccination had <0.5 endotoxin unit/ml.
Vero and Vero/SLAM cells (40) were grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS). The Chicago-1 and Moraten strains of MV were grown and assayed by plaque formation in Vero cells. The wild-type Bilthoven strain of MV (a gift from Albert Osterhaus, Erasmas University, Rotterdam, The Netherlands) was grown in phytohemagglutinin-stimulated human cord blood mononuclear cells and assayed by syncytia formation in B95-8 cells. Viremia was assessed by cocultivation of peripheral blood mononuclear cells (PBMCs) with B95-8 cells in DMEM supplemented with 10% FBS, penicillin, and streptomycin. Cocultures were scored for syncytia at 96 h, and data are reported as the number of syncytia/106 PBMCs.
MV RNA was detected by quantitative reverse transcription-PCR (RT-PCR) as previously described (44). Briefly, 2 × 106 PBMCs were cultured for 3 days without stimulation, RNA was isolated, and the nucleoprotein (N) gene was amplified (Applied Biosystems Prism 7700) using TaqMan primers and probe. Controls included glyceraldehyde-3-phosphate dehydrogenase (GAPDH) amplification (Applied Biosystems) and RNA isolated from cultured PBMCs from MV-naïve monkeys. Copy number was determined by construction of a standard curve from 1 to 106 copies of RNA synthesized by in vitro transcription from a plasmid encoding the Edmonston N gene. The sensitivity of the assay was 10 copies. Data were normalized to the GAPDH control, and results are expressed as follows: [(number of copies of MV N RNA)/(number of copies of GAPDH RNA)] × 100,000.
Juvenile rhesus macaques were vaccinated with a single intradermal (i.d.) injection of 108 VEE/SIN-H particles (n = 3; animals 1R, 6R, and 7R) or 108 VEE/SIN-H particles plus 3 × 106 VEE/SIN-F particles (VEE/SIN-H+F) (n = 3; animals 8R, 22R, and 25R). Two juvenile monkeys (33R and 34R) were vaccinated with one dose of 5,000 PFU of the Moraten strain of LAV intramuscularly (i.m.). Two juvenile monkeys (24R and 26R) were vaccinated with three doses of 0.5 ml of FIMV (Pfizer; a gift from Albert Kapikian, National Institutes of Allergy and Infectious Diseases) i.m. at 0, 4, and 8 weeks. Three infant rhesus macaques (9U, 25U, and 29U) were immunized i.d. with 108 VEE/SIN-H particles plus 3 × 106 VEE/SIN-F particles and boosted 2 months later. Heparinized blood was collected for evaluation of the immune response.
For MV challenge, 104 tissue culture 50% infectious doses (TCID50) of the Bilthoven strain of MV were instilled intratracheally into anesthetized animals 12 to 17 months after vaccination. Monkeys were shaved and monitored frequently for development of a rash. Heparinized blood was collected to assess viremia and immune responses.
Punch biopsy specimens were taken of normal and rash skin of FIMV-vaccinated monkeys 10 days after challenge and fixed in 10% formalin. Tissues were embedded in paraffin, sectioned, and stained with hematoxylin and eosin. MV-infected cells were identified using a mouse monoclonal antibody to the MV N protein (Chemicon), horseradish peroxidase (HRP)-conjugated rabbit anti-mouse IgG (Dako) and diaminobenzidine (DAB) as previously described (43).
Neutralizing antibody was measured by the ability of serially diluted plasma to reduce plaque formation by the Chicago-1 strain of MV on Vero cells (plaque reduction neutralization [PRN]). An internal standard serum calibrated to the international standard (66/202) was included in all assays, and data were normalized to that standard and expressed as milli-international units (mIU).
Enzyme immunoassays (EIAs) used 96-well Maxisorp plates (Nunc, Rochester, NY) coated with a lysate of MV-infected Vero cells (1.16 μg protein/well; Advanced Biotechnologies, Columbia, MD), H-expressing L cells (4), or baculovirus-expressed N. Plates were incubated with serially diluted plasma samples overnight at 4°C. Alkaline phosphatase-conjugated rabbit antibody to monkey IgG (Biomakor; Accurate Chemicals) or HRP-conjugated goat antibody to monkey IgM (Nordic, Capistrano Beach, CA) was used as the secondary antibody. For measurement of antibody avidity for the H and N proteins, plates were washed for 15 min with stepwise dilutions (0 to 3.5 M) of ammonium thiocyanate (NH4SCN) before addition of the secondary antibody, as previously described (37). The avidity index is the concentration of NH4SCN required to reduce the optical density (OD) by 50%.
Mononuclear cell production of gamma interferon (IFN-γ) and interleukin-4 (IL-4) was measured by enzyme-linked immunospot (ELISPOT) assay. Multiscreen plates (Millipore) were coated with anti-human IFN-γ antibody (2 μg/ml) or anti-human IL-4 antibody (5 μg/ml; BD PharMingen). After plates were washed and blocked with culture medium, 1 ×105 to 5 ×105 fresh PBMCs were added along with 1 μg/ml pooled MV peptides (20-mers overlapping by 11 amino acids) from the H or F proteins or 5 μg/ml concanavalin A (Sigma). After 40 h of incubation, plates were washed and incubated with biotinylated antibody to IFN-γ (1 μg/ml; Mabtech) or IL-4 (2 μg/ml; BD PharMingen) for 2 h at 37°C. After plates were washed, HRP-conjugated avidin (1:2,000 in PBS-1% FBS; Research Laboratory Inc.) was added and incubated 1 h at 37°C. The assays were developed with stable DAB solution (Invitrogen, Carlsbad, CA). The reaction was stopped with tap water, plates were allowed to dry, and wells were scanned in an ImmunoSpot reader and analyzed using ImmunoSpot, version 2.0.5, software (CTL, Cleveland, OH). Data are presented as the number of spot-forming cells (SFCs)/106 PBMCs.
To determine the immunogenicity of the chimeric VEE/SIN alphavirus replicon particle vaccine, groups of three juvenile rhesus macaques were vaccinated i.d. once with VEE/SIN-H replicon particles (108) or with VEE/SIN-H (108) + F (106.5) replicon particles. To compare these vaccine candidates with other measles vaccines, control monkeys were vaccinated i.m. with either one dose (5,000 PFU) of LAV or three 0.5-ml doses of FIMV. MV-specific neutralizing antibodies were induced in all VEE/SIN-H- and VEE/SIN-H+F-vaccinated monkeys (Fig. (Fig.1A).1A). Titers were similar to those induced by LAV and were above the predicted protective level (120 mIU/ml) within 2 to 4 weeks after vaccination and were sustained at or above that level for at least 1 year. PRN responses of FIMV-vaccinated monkeys increased briefly with each booster dose but remained below the predicted protective level. The peak geometric mean titers were 676 ± 415 mIU/ml for VEE/SIN-H, 1,380 ± 398 mIU/ml for VEE/SIN-H+F, 70 ± 48 for FIMV mIU/ml, and 1,549 ± 338 mIU/ml for LAV.
MV-specific IgG, as measured by EIA, was also induced in all vaccinated monkeys and showed a 3- to 8-fold increase by 2 weeks (Fig. (Fig.1B).1B). FIMV-vaccinated monkeys showed responses that were similar to those induced by LAV and higher than the VEE/SIN-H or VEE/SIN-H+F responses. The average MV-specific IgG OD values (1:400 dilution) were 0.418 ± 0.12 for VEE/SIN-H, 0.405 ± 0.13 for VEE/SIN-H+F, 1.26 ± 0.23 for FIMV, and 1.049 ± 0.28 for LAV. However, FIMV-vaccinated animals did not develop detectable H-specific IgG (Fig. (Fig.1C)1C) but did develop N-specific IgG (Fig. (Fig.1E).1E). Avidity maturation of antibody to H occurred over several months and was similar for animals vaccinated with VEE/SIN-H, VEE/SIN-H+F, and LAV (Fig. (Fig.1D).1D). FIMV-vaccinated animals had insufficient H-specific antibody to measure avidity but had only a modest increase in avidity of N-specific antibody (Fig. (Fig.1F1F).
To assess the development of MV-specific T-cell responses, IFN-γ and IL-4 production in response to MV H and F peptide stimulation was measured by ELISPOT assay (Fig. (Fig.2).2). In general, peak IFN-γ responses were higher than peak IL-4 responses in all VEE/SIN- and LAV-vaccinated monkeys while FIMV-vaccinated monkeys had numbers of H- and F-specific IL-4-producing cells that were similar to the numbers of IFN-γ-producing cells (Fig. 2A and B). The highest numbers of IFN-γ-producing cells were present in blood 1 to 2 weeks after vaccination (Fig. 2C and D), and the number of H-specific cells remained above background levels in all but the FIMV-vaccinated animals for many weeks. The peaks of H-specific IFN-γ-producing cells were 79 ± 6 SFCs/106 PBMCs for VEE/SIN-H, 145 ± 30 SFCs/106 PBMCs for VEE/SIN-H+F, 44 ± 44 SFCs/106 PBMCs for FIMV, and 89 ± 0 SFCs/106 PBMCs for LAV-vaccinated monkeys (P = 0.0008, one-way analysis of variance [ANOVA]). The peaks of F-specific IFN-γ-producing cells were 89 ± 54 SFCs/106 PBMCs for VEE/SIN-H+F, 11 ± 3 SFCs/106 PBMCs for FIMV, and 29 ± 1 SFCs/106 PBMCs for LAV-vaccinated monkeys (P > 0.05).
To evaluate the protective efficacy of VEE/SIN measles vaccines in comparison with LAV and FIMV, all monkeys were challenged with 104 TCID50 of wild-type MV intratracheally 17 months after vaccination. Petechial rashes developed in both FIMV-vaccinated monkeys at day 10 to 14 on the face (Fig. 3A and B) and lower abdomen and extremities (Fig. 3C and D). Histological examination of skin biopsy specimens showed severe infiltration of lymphocytes, monocytes, neutrophils, and eosinophils into the superficial epithelium and dermis (Fig. (Fig.3E).3E). Epithelial cells stained positively for MV N protein (Fig. (Fig.3F).3F). No rashes appeared in VEE/SIN-H-, VEE/SIN-H+F- or LAV-vaccinated monkeys.
Viremias, detected by cocultivation of PBMCs with B95-8 cells, occurred in both FIMV-vaccinated monkeys, peaked at 7 days, and resolved by 14 days after the challenge (Fig. (Fig.4A).4A). No viremias were detected in VEE/SIN-H-, VEE/SIN-H+F- or LAV-vaccinated monkeys. Likewise, only FIMV-vaccinated monkeys produced MV-specific IgM in response to challenge, with a peak at day 14 (Fig. (Fig.4B).4B). Titers of MV-specific PRN antibody were above the predicted protective level (120 mIU/ml) (9) in all vaccinated monkeys at the time of challenge except in those vaccinated with FIMV (Fig. (Fig.4C).4C). PRN titers increased after challenge in all monkeys and peaked at 14 to 18 days, with the highest titers in FIMV-vaccinated monkeys. The peak geometric mean PRN titers were 3,776 ± 2319 mIU/ml for VEE/SIN-H, 2,673 ± 2,456 mIU/ml for VEE/SIN-H+F, 13,397 ± 1,258 mIU/ml for FIMV, and 2,128 ± 991 mIU/ml for LAV. The change in PRN titer was most rapid and greatest for FIMV-immunized monkeys. MV-specific IgG also increased after challenge in all monkeys (Fig. (Fig.4D).4D). IgG OD values were highest in FIMV-vaccinated monkeys. The peak IgG OD readings (1:400 dilution) were 1.86 ± 0.16 for VEE/SIN-H, 1.6 ± 0.54 for VEE/SIN-H+F, 2.23 ± 0.2 for FIMV, and 1.6 ± 0.1 for LAV.
To monitor the cell-mediated immune responses to viral challenge, PBMC IFN-γ and IL-4 production was measured by ELISPOT assays. An increase in H-specific IFN-γ production was detected within 1 month after challenge in all monkeys (Fig. (Fig.5A).5A). Monkeys vaccinated with FIMV showed a later peak (day 25) than VEE/SIN- and LAV-vaccinated monkeys (days 7 to 10). The mean peak numbers of H-specific IFN-γ producing cells were 98 ± 19 SFCs/106 PBMCs for VEE/SIN-H, 58 ± 19 SFCs/106 PBMCs for VEE/SIN-H+F, 127 ± 12 SFCs/106 PBMCs for FIMV, and 100 ± 6 SFCs/106 PBMCs for LAV. F-specific IFN-γ production was detected in VEE/SIN-H+F-, FIMV-, and LAV-immunized monkeys (Fig. (Fig.5B).5B). The peak numbers of F-specific IFN-γ-producing cells were 22 ± 19 SFCs/106 PBMCs for VEE/SIN-H, 45 ± 13 SFCs/106 PBMCs for VEE/SIN-H+F, 43 ± 1.4 SFCs/106 PBMCs for FIMV, and 79 ± 17 SFCs/106 PBMCs for LAV. Only FIMV-vaccinated animals showed H- and F-specific IL-4 production after challenge with an initial peak at days 20 to 25 (Fig. 5C and D).
For monkeys vaccinated with FIMV and VEE/SIN, the T-cell responses were biphasic. There was a peak early after challenge that, for FIMV-vaccinated monkeys, coincided with clearance of infectious virus from the blood (Fig. (Fig.4A).4A). The second peak occurred approximately 3 months after challenge and was comprised of cells producing IFN-γ for VEE/SIN-vaccinated monkeys and of both IFN-γ and IL-4 for FIMV-vaccinated monkeys (Fig. (Fig.5).5). To determine whether this reflected recurrence of viremia, quantitative RT-PCR was performed for MV RNA (Fig. (Fig.6).6). Large amounts of RNA were detected during the previously documented viremia (Fig. (Fig.4A)4A) for FIMV-vaccinated monkeys that were not protected from rash, but RNA was not detected at these times for the other groups of monkeys that had no detectable viremia or rash. However, at the time of the late T-cell responses (day 90), viral RNA was detected in all VEE/SIN-H- and VEE/SIN-H+F-vaccinated monkeys but not in LAV- or FIMV-vaccinated monkeys. No monkeys showed evidence of disease during this period.
To determine the protective efficacy of the VEE/SIN-H+F replicon particle vaccine in infant monkeys, three 1- to 2-month-old rhesus monkeys born to measles-naïve mothers were immunized with VEE/SIN-H+F and boosted 2 months later (Fig. (Fig.7).7). Infant monkeys showed an increase in PRN antibody by 4 weeks after vaccination, with a maximum at 8 weeks (geometric mean, 1791 ± 211 mIU/ml), and levels were maintained above 120 mIU/ml for 1 year (Fig. (Fig.7A).7A). MV-specific IgG OD values were also increased after vaccination and showed a similar pattern to the PRN responses (Fig. (Fig.7B).7B). The mean peak OD reading was 0.78 ± 0.07 (at week 8). H-specific IgG showed a similar pattern (Fig. (Fig.7C),7C), and avidity of the antibody to H matured over a time course similar to that of the juvenile monkeys (Fig. (Fig.11 and and7D7D).
MV-specific IFN-γ and IL-4 responses were measured by ELISPOT assay. Similar to the antibody responses, H-specific IFN-γ-producing T cells appeared by week 4 (Fig. (Fig.7E).7E). Peak H-specific IFN-γ-producing cells were 96 ± 30 SFCs/106 PBMCs. F-specific IFN-γ-producing cells were also induced in all vaccinated infant monkeys but amounts were lower than the H response (Fig. (Fig.7F).7F). The peak of the F-specific IFN-γ SFCs was 38 ± 12 per 106 PBMCs. IL-4 responses were not detected (data not shown).
To test the protective capacity of VEE/SIN-H+F measles vaccine in infants, monkeys were challenged with wild-type MV 1 year after vaccination. No monkey developed a rash, viremia, or IgM response (data not shown). PRN antibody titers were boosted after challenge to similar levels with a geometric mean peak titer of 2,382 ± 1,030 (Fig. (Fig.8A).8A). MV-specific IgG measured by EIA also showed a 2- to 3-fold increase after challenge in all vaccinated monkeys, with a peak OD of 1.145 ± 0.493 (Fig. (Fig.8B).8B). MV-specific IFN-γ, but not IL-4, responses were induced by challenge in all monkeys. The IFN-γ-producing SFCs were 130 ± 47 per 106 PBMCs (Fig. (Fig.8C).8C). Similar to H, F-specific IFN-γ production increased significantly at day 14, with 73 ± 22 SFCs/106 PBMCs (Fig. (Fig.8D8D).
These studies have shown that a single i.d. injection of a chimeric VEE/SIN replicon particle vaccine expressing the MV H protein or H and F proteins induced long-lasting MV-specific neutralizing antibody and IFN-γ-producing T cells in juvenile rhesus macaques at levels similar to those induced by LAV. Two doses induced similar responses in infant macaques. Vaccination protected both juvenile and infant monkeys from rash and viremia after wild-type MV challenge. A late appearance, without evident accompanying disease, of viral RNA in blood 3 months after challenge in otherwise VEE/SIN-protected monkeys was associated with a renewed increase in circulation of MV-specific T cells and subsequent clearance of viral RNA. We conclude that the chimeric VEE/SIN replicon particle vaccine expressing either H or H and F is a promising candidate for development of a new measles vaccine for infants and that further study of MV infection and clearance in the face of apparent protective immunity is warranted.
Alphavirus replicon particle vaccines are a promising platform for the development of new vaccines as they combine the advantages of a live virus vaccine with safety due to one-cycle replication and the ability of a highly expressed subgenomic RNA to drive synthesis of mRNA for the relevant antigen. Both i.m. and intranasal routes of delivery for alphavirus replicon vaccines can induce protection from parenteral or mucosal challenge (13, 19, 20, 27, 34, 35), and, in this study, we have shown that the i.d. route is also effective. This route was chosen because it offers ready access to antigen-presenting cells and secondary lymphoid tissue that results in increased immunogenicity and dose sparing for other vaccines (10, 11, 28, 38, 57). Furthermore, a number of needleless approaches to i.d. vaccination have been developed that hold promise for practical routine delivery by this route (30, 52).
Chimeric VEE/SIN-based replicon particle vaccines for human immunodeficiency virus (HIV) and parainfluenza virus type 3 tested in mice, hamsters, and rhesus macaques have shown induction of antibody and IFN-γ-producing T cells and at least partial protection from challenge (19-22, 47, 69). VEE/SIN-based vaccines offer expression and immunogenicity advantages over SIN and manufacturing advantages over VEE vaccines (8, 13, 47). Previous studies directly comparing VEE/SIN with SIN replicon particles showed higher levels of HIV envelope antibody in macaques (22) and better Gag T-cell responses in mice after immunization with VEE/SIN (21). In the current studies, levels of MV PRN antibody induced by VEE/SIN-H were similar to those previously reported for SIN-H-vaccinated monkeys that were not protected from viremia and only partially protected from rash (44). This was postulated to be associated with a need to include F in the vaccine for broader immune responses. A direct comparison of VEE/SIN-H and VEE/SIN-H+F in this study showed similar protection and similar levels of PRN antibody, consistent with the fact that H is the dominant protein against which neutralizing antibody is directed. It is possible that the quality or quantity of the T-cell immune response induced by SIN-H was different in a way that is important for prevention of viremia after challenge.
Previous studies have shown higher levels of PRN antibody after vaccination with H alone than with both H and F when given in equivalent amounts (44, 45, 50), perhaps due to the differences in T-cell cytokine induction by H and F (49). Because of difficulties producing high titers of VEE/SIN-F replicon particles, the dose of VEE/SIN-F was lower than that of VEE/SIN-H in the VEE/SIN-H+F vaccine. Interference with H responses by coexpression of F may have been eliminated with the lower dose of F, but difficulties with production could be a barrier to development of a VEE/SIN-H+F vaccine.
FIMV provided no protection from challenge. FIMV-vaccinated monkeys developed higher MV-specific IgG but lower PRN titers than LAV-, VEE/SIN-H-, or VEE/SIN-H+F-vaccinated monkeys. The N antigen is present in LAV and FIMV and in the MV-infected cell lysate used as antigen for the EIA, but antibody to this protein is not neutralizing and is thus not detected in the PRN assay. H- and N-specific EIAs confirmed that antibody induced by FIMV vaccination was primarily directed to the N protein. This lack of neutralizing antibody is likely to be a major contributor to the poor protection provided by FIMV. LAV, VEE/SIN-H, and VEE/SIN-H+F induced MV-specific T-cell responses with a Th1 cytokine bias while FIMV-induced T-cell responses were less vigorous and showed a Th2 cytokine bias. These responses may contribute to enhanced disease after challenge in FIMV-primed animals.
Young age is a barrier to measles immunization. The time course of the production of both antibody and IFN-γ-producing T cells by infant macaques after VEE/SIN-H+F immunization was similar to that of juvenile macaques. Previous studies of SIN-H in infant macaques showed slower production of protective antibody levels (44). Vaxfectin-adjuvanted H+F DNA induced antibody in infants that was similar to that of juvenile monkeys and gave protection from challenge (42). Protection of infant macaques was also observed using a prime-boost strategy involving a SINV replicon-based H+F DNA vaccine prime and intranasal LAV boost (45).
In addition to immunologic immaturity, maternal antibody is also a significant barrier to the immunization of young infants (1, 15). Alphavirus replicon particles offer a theoretical advantage in that preexisting antibody to MV will not inhibit interaction of the alphavirus glycoproteins with target antigen-presenting cells. Previous studies have shown that VEE replicon particles expressing dengue virus antigen are able to successfully immunize mice born to immune mothers (67), but more study is required to determine the effect of maternal antibody on infant responses to VEE/SIN-H+F. Cumulatively, these studies suggest that the barrier of young age can be overcome with new approaches.
Vaccine-induced protective immunity for virus infections is poorly understood even for diseases such as measles for which successful live virus vaccines have been in use for many years. Protection against respiratory pathogens such as MV may require both antibody and T cells localized to the respiratory mucosa. Parenteral alphavirus vaccination of mice stimulates a mucosal inductive environment in local lymph nodes (62, 64) and mucosal immunity (62, 63), further suggesting that this platform may be advantageous for measles virus immunization. However, after challenge, MV-specific antibody and T-cell responses were boosted at least transiently, indicating that vaccine-induced immunity was not “sterilizing”; i.e., sufficient infection occurs to stimulate a recall immune response. How frequently this occurs in more natural settings, where the dose of virus is likely to be less than that used in these studies, is not completely clear. Antibody responses have been observed during measles outbreaks in individuals who do not develop disease, particularly when preexisting neutralizing antibody levels are between 120 and 1,000 mIU/ml (9).
Further evidence that infection was established despite undetectable viremia is the fact that both VEE/SIN-H- and VEE/SIN-H+F-vaccinated animals showed recrudescence of MV RNA in PBMCs 3 to 4 months after challenge accompanied by reappearance of IFN-γ-producing MV-specific T cells in circulation. This recurrent viremia was not observed in LAV- or FIMV-vaccinated monkeys, but FIMV-vaccinated monkeys had a second wave of T cells in circulation, suggesting that MV RNA was present but not detected, probably due to an inadequate frequency of sampling. Reappearance of T cells was not noted in LAV-vaccinated animals, suggesting that they were protected from this recurrence of MV replication.
The site of silent replication is presumed to be the respiratory tract. Further investigation is needed to determine whether this escape from immune control represents a change in the virus or regulatory suppression of the initial T-cell response to infection prior to complete virus clearance. These studies further indicate that MV is able to persist for substantial periods of time (23, 58, 60). Children with natural measles continue to have detectable MV RNA in respiratory secretions, blood, and urine for at least 3 to 4 months after resolution of the rash despite rapid clearance of infectious virus (46, 56). Viral RNA has previously been detected in PBMCs up to 5 months after wild-type MV infection in both naïve macaques and macaques immunized with the partially protective SIN-H vaccine (44). The current studies further indicate that clearance is a complicated and prolonged process that is influenced by prior vaccine-induced priming of the immune system. Further studies are needed to determine whether prevention of this reactivation requires the inclusion of additional MV antigens in the vaccine or a different vaccine platform.
This work was funded by research grants from the National Institutes of Health (R01 AI23047 to D.E.G) and the Bill and Melinda Gates Foundation (RG3522 to D.E.G.) with additional support from the Gilbert F. Otto Young Investigator Fund (C.-H.P.) and a sabbatical fellowship from Chungbuk University (E.-Y.L.).
We thank Paul Rota for Edmonston MV cDNAs, Albert Osterhaus for the Bilthoven wild-type strain of MV, and Albert Kapikian for FIMV.
Published ahead of print on 3 February 2010.