A Novel Non-Replication-Competent Cytomegalovirus Capsid Mutant Vaccine Strategy Is Effective in Reducing Congenital Infection

Abstract

Congenital cytomegalovirus (CMV) infection is a leading cause of mental retardation and deafness in newborns. The guinea pig is the only small animal model for congenital CMV infection. A novel CMV vaccine was investigated as an intervention strategy against congenital guinea pig cytomegalovirus (GPCMV) infection. In this disabled infectious single-cycle (DISC) vaccine strategy, a GPCMV mutant virus was used that lacked the ability to express an essential capsid gene (the UL85 homolog GP85) except when grown on a complementing cell line. In vaccinated animals, the GP85 mutant virus (GP85 DISC) induced an antibody response to important glycoprotein complexes considered neutralizing target antigens (gB, gH/gL/gO, and gM/gN). The vaccine also generated a T cell response to the pp65 homolog (GP83), determined via a newly established guinea pig gamma interferon enzyme-linked immunosorbent spot assay. In a congenital infection protection study, GP85 DISC-vaccinated animals and a nonvaccinated control group were challenged during pregnancy with wild-type GPCMV (10(5) PFU). The pregnant animals carried the pups to term, and viral loads in target organs of pups were analyzed. Based on live pup births in the vaccinated and control groups (94.1% versus 63.6%), the vaccine was successful in reducing mortality (P = 0.0002). Additionally, pups from the vaccinated group had reduced CMV transmission, with 23.5% infected target organs versus 75.9% in the control group. Overall, these preliminary studies indicate that a DISC CMV vaccine strategy has the ability to induce an immune response similar to that of natural virus infection but has the increased safety of a non-replication-competent virus, which makes this approach attractive as a CMV vaccine strategy.

Importance: Congenital CMV infection is a leading cause of mental retardation and deafness in newborns. An effective vaccine against CMV remains an elusive goal despite over 50 years of CMV research. The guinea pig, with a placenta structure similar to that in humans, is the only small animal model for congenital CMV infection and recapitulates disease symptoms (e.g., deafness) in newborn pups. In this report, a novel vaccine strategy against congenital guinea pig cytomegalovirus (GPCMV) infection was developed, characterized, and tested for efficacy. This disabled infectious single-cycle (DISC) vaccine strategy induced a neutralizing antibody or a T cell response to important target antigens. In a congenital infection protection study, animals were protected against CMV in comparison to the nonvaccinated group (52% reduction of transmission). This novel vaccine was more effective than previously tested gB-based vaccines and most other strategies involving live virus vaccines. Overall, the DISC vaccine is a safe and promising approach against congenital CMV infection.

FIG 1

Nucleotide sequence of the GP85/GP86 locus. The sense genome strand sequence is shown. Both GP85 and GP86 are encoded on the complementary strand. The sequence used for flanking arms in shuttle vector (Fig. 2) recombination are highlighted. The GP85 flanking sequence includes GPCMV bases 134992 to 135906; GP86 flanking sequence includes bases 136180 to 136681; the deleted intergenic sequence includes bases 135907 to 136179.

FIG 2

Cloning strategy for generation of a shuttle vector encoding GP85 under Tet-Off (TRE-tight) promoter control. (Steps 1 and 2) Parts of the GP85 (5′) and GP86 (3′) coding sequences were separately cloned via PCR into pUC19 to generate pGP85FLK and pGP86FLK. (Step 3) The GP86 coding sequence was then cloned into pGP85FLK as a BamHI/HindIII fragment to generate pGP85GP86, which lacked the intergenic sequence between GP85 and GP86. (Steps 4 to 6) A BglII Km cassette was introduced to generate pGP8586Km+ (steps 4 and 5), which was digested with SalI to enable cloning of a SV40 poly(A) cassette to generate pGP8586SV40AKm (step 6). (Step 7) The TREtight promoter from pTREtight (Clontech) was PCR cloned as a BamHI fragment immediately upstream of the GP85 coding sequence to generate pGP8586TRESV40AKm. Shuttle vectors (steps 5, 6, and 7) linearized with PmeI were used to modify the GP85/GP86 intergenic locus in the GPCMV BAC via homologous recombination in separate reactions to introduce the modified sequence, as described in Materials and Methods, to generate the DISC GPCMV BAC and other BAC mutants.

FIG 3

BLAST alignment of the predicted minor capsid protein (mCP) nucleotide sequence from GPCMV (GP85) and HCMV Towne strain (UL85). (i) Colinear location of GP85 and UL85 genes in GPCMV and HCMV, respectively. (ii) BLAST search of the GP85 protein identified it as a member of the herpes virus V23 (capsid protein) superfamily. (iii) BLAST (NCBI BLASTp) alignment of GP85 and UL85 proteins. Score, 354 bits (908); E, = 2e−120; method used, compositional matrix adjust; identities, 172/305 (56%); positives, 226/305 (74%); gaps, 4/305 (1%).

FIG 4

Characterization of GPL Tet-Off cell lines via luciferase reporter gene transactivation. GPL cells (A to D) or candidate GP Tet-Off cells (E to H) were transfected with a luciferase reporter plasmid under TREtight promoter control (pTREtightLUC). Wells A and C were additionally transfected with the Tet-Off transactivator (tTA2) expression plasmid. Well B was transfected with a luciferase reporter gene plasmid under HCMV MIE promoter control (pcDNA3LUC). At 24 h posttransfection, luciferase substrate (D-luciferin) was added, and plates were imaged for 5 min to evaluate bioluminescence (IVIS 50; Xenogen). Shown are black and white images of six-well plates, with superimposed photon emission intensities. Vertical color bar ranges indicate the highest (red) to lowest (purple) levels of bioluminescence (in photons per second per square centimeter per steradian) for imaged samples. Tet-Off cell line assays included duplicate wells for A20 (E and G) and A21 (F and H) cells.

FIG 5

Modification of the GP85 5′ UTR in a GPCMV BAC and generation of a DISC GP85 strain BAC. (i) Schematic of the GPCMV genome, with location of the BAC insertion and HindIII sites indicated. The GP85/GP86 locus is encoded in the HindIII A fragment. (ii) Modifications to the 5′ UTR in the GP85/GP86 locus in the wild type (wt; 1) and mutants GP85/GP86KmR BAC (2, top), GP85/GP86KmRpoly(A) BAC (2, bottom), and TRE GP85 (DISC) BAC (3). Also shown are the approximate sizes of the wild-type and mutant loci, based on the external primer used to generate the original GP85/GP86 shuttle vector. (iii) EcoRI restriction profile analysis of wild-type and mutant GPCMV BACs described in panel ii, including wild-type GPCMV BAC (1), GP85/GP86KmR BAC (2), and TRE GP85 (DISC) BAC (3). The original HindIII A genomic fragment is indicated in blue, and the modified fragment indicated by red dots. (iv) PCR amplification of the GP85/GP86 locus in wild-type and mutant GPCMV BACs: wild type (1), GP85/GP86 KmR (2), and TRE GP85 DISC (3). The DNA ladder (measurements in kilobases) was obtained from Invitrogen.

FIG 6

Regeneration of a DISC GP85 GPCMV from GP85 BAC mutants requires a TRE promoter and a cell line expressing a Tet-Off transactivator (tTA2). (A to C) GP85 mutant GPCMV BACs were individually transfected into GPL cells: GP85/GP86 KmR (A), GP85/86 KmR poly(A) (B), and TRE GP85 DISC (C). (D) Mutants were also transfected onto GPL Tet-Off cells. The TRE GP85 DISC is shown. Individual transfected cells expressed GFP encoded by the BAC. Virus spread was detected by GFP spread across the cell monolayer. Images were taken at 16 days posttransfection.

FIG 7

Growth kinetics of DISC GP85 GPCMV versus wild-type GPCMV on GPL Tet-Off cells. GPL Tet-Off cells were infected at a multiplicity of infection of 1 PFU/cell with each respective virus in separate wells of six-well dishes. Samples were taken on different days postinfection and titrated in duplicate as previously described (33). Results are plotted as the virus titer versus day postinfection.

FIG 8

Western blots of wild-type GPCMV. GPCMV (purified virus particles or total cell lysate) were separated by SDS-PAGE and immunoblotted with anti-GPCMV sera (1:2,000 dilution) from DISC-vaccinated animals and with anti-guinea pig IgG–HRP (1:500; Sigma) as described in Materials and Methods. (A) Sucrose-purified GPCMV. (B) GPCMV-infected GPL cell lysate. MI, control lanes with uninfected GPL cells. Protein sizes are indicated in kilodaltons (Bio-Rad ladder).

FIG 9

Overview of DISC GP85 vaccination schedules and the preconception vaccine study. (a and b) In the initial characterization of the immune response to the DISC vaccine, two vaccine regimens were employed: R1 (a) and R2 (b). DISC virus inoculations are indicated with the arrow along with a virus symbol. Animals were bled for immune response measurements at the indicated weeks post-initial vaccination. Animals were euthanized at the end of the vaccination schedule (and after confirmation of seroconversion) to determine the T cell response via an ELISPOT assay as described in Materials and Methods. (c) Preconception vaccine study (group 1) following vaccine regimen R1. During the late second trimester of pregnancy, animals were challenged with wild-type GPCMV (10⁵ PFU) and then allowed to go to term. (d) Control nonvaccinated animals (group 2) from the congenital vaccine study.

FIG 10

Antibody immune responses (determined in ELISAs) to GPCMV and specific viral glycoprotein complexes (gB, gH/gL, and gM/gN) in DISC GP85-vaccinated guinea pigs. (A) Immune responses of animals vaccinated with DISC GP85 under the R1 or R2 regimen were compared to animals hyperimmunized with wild-type GPCMV. Results shown are means for each group with standard deviations represented by error bars. Statistically significant differences (P < 0.05; Mann-Whitney test) are shown by lowercase italic letters for comparisons as follows: a, hyperimmune versus R1; b, R1 versus R2; c, hyperimmune versus R2. (B) Neutralizing antibody titers on fibroblast cells from hyperimmune sera, compared to those from the R1 and R2 regimens. Statistically significant differences (P < 0.05; Mann-Whitney test) are indicated by lowercase italic letters for comparisons as follows: d, hyperimmune versus R1; e, hyperimmune versus R2.

FIG 11

Evaluation of animal T cell response to the GPCMV vaccine. Gamma interferon ELISPOT assay results are shown for splenocyte responses to GPCMV GP83 peptide pools in R1 or R2 vaccinated and wild-type (wt) GPCMV-infected animals. The assays were carried out as described in Materials and Methods, with splenocytes isolated from DISC GP85-vaccinated animals (vaccine regimens R1 and R2) or wild-type GPCMV-infected animals. Overlapping peptides (9-mers) spanning the complete GP83 protein sequence were assigned to 24 peptide pools (I to XXIV). Results are shown for selected GP83 peptide pools: II, IV, X, and XX. ConA was used as a positive control, and negative controls included unstimulated and DMSO-treated cells. Final counts were calculated based on SFC per 10⁶ cells after background spots (cells only, without any stimulation) were subtracted. GP83 peptide pools X and XX produced highly stimulated cells, whereas pools II and IV resulted in poorly stimulated cells.