Plasmid vectors encoding cholera toxin or the heat-labile enterotoxin from Escherichia coli are strong adjuvants for DNA vaccines

Abstract

Two plasmid vectors encoding the A and B subunits of cholera toxin (CT) and two additional vectors encoding the A and B subunits of the Escherichia coli heat-labile enterotoxin (LT) were evaluated for their ability to serve as genetic adjuvants for particle-mediated DNA vaccines administered to the epidermis of laboratory animals. Both the CT and the LT vectors strongly augmented Th1 cytokine responses (gamma interferon [IFN-gamma]) to multiple viral antigens when codelivered with DNA vaccines. In addition, Th2 cytokine responses (interleukin 4 [IL-4]) were also augmented by both sets of vectors, with the effects of the LT vectors on IL-4 responses being more antigen dependent. The activities of both sets of vectors on antibody responses were antigen dependent and ranged from no effect to sharp reductions in the immunoglobulin G1 (IgG1)-to-IgG2a ratios. Overall, the LT vectors exhibited stronger adjuvant effects in terms of T-cell responses than did the CT vectors, and this was correlated with the induction of greater levels of cyclic AMP by the LT vectors following vector transfection into cultured cells. The adjuvant effects observed in vivo were due to the biological effects of the encoded proteins and not due to CpG motifs in the bacterial genes. Interestingly, the individual LT A and B subunit vectors exhibited partial adjuvant activity that was strongly influenced by the presence or absence of signal peptide coding sequences directing the encoded subunit to either intracellular or extracellular locations. Particle-mediated delivery of either the CT or LT adjuvant vectors in rodents and domestic pigs was well tolerated, suggesting that bacterial toxin-based genetic adjuvants may be a safe and effective strategy to enhance the potency of both prophylactic and therapeutic DNA vaccines for the induction of strong cellular immunity.

FIG. 1.

Functional map of vector pPJV2002. pPJV2002 is a DNA vaccine adjuvant vector encoding the mature form of the CT A subunit fused in frame with the tissue plasminogen activator signal peptide. Expression is driven by the human CMV immediate early promoter, which is functionally linked to the intron A element. Polyadenylation is controlled by the bovine growth hormone polyadenylation sequence. hCMV Pro, human CMV immediate early promoter; TPAsig CDS, tissue plasminogen activator signal peptide coding sequence; CTA CDS, CT A subunit coding sequence; and bGHpA, bovine growth hormone polyadenylation sequence. Vectors pPJV2003, pPJV2004, and pPJV2005 (not shown) encode the CT B subunit, the LT toxin A subunit, and the LT toxin B subunit, respectively.

FIG. 2.

Induction of cAMP production in Caco-2 cells following transfection with vectors encoding CT and LT. Four hours following transfection of Caco-2 cells with pPJV2002 + pPJV2003 (CT) or with pPJV2004 + pPJV2005 (LT), growth medium was replaced with fresh medium containing 1 mM 3-isobutyl-1-methylxanthine. Individual wells were harvested at 6, 12, 18, and 24 h posttransfection for quantification of cAMP content. Positive control cells were treated with purified CT protein (1 μg) for 4 h, after which medium was replaced with fresh medium containing 1 mM 3-isobutyl-1-methylxanthine. Two types of negative controls were employed: cells transfected with empty-vector DNA (pWRG7054) or nontransfected cells.

FIG. 3.

gp120-specific humoral and cellular responses in mice following codelivery of vectors encoding HIV-1 gp120 and CT or LT. All groups (n = 4) received primary and booster immunizations spaced 4 weeks apart. Serum samples and splenocytes were collected 2 weeks following the second immunization. (A) IgG1 and IgG2a responses. (B) IgG1-to-IgG2a ratios. (C) CTL responses. P values shown in panels B and C were calculated using Student's t test. Bars represent standard error.

FIG. 4.

M2-specific humoral and cellular immune responses in mice following codelivery of pM2-FL with one of four dosage levels of the CT or LT vector. All groups (n = 8) received primary and booster immunizations spaced 4 weeks apart. Blood samples and splenocytes were collected 2 weeks following the second immunization. (A) Total M2-specific IgG responses. (B) M2-specific IgG1-to-IgG2a ratio. (C) IFN-γ ELISPOT responses. (D) IL-4 ELISPOT responses. As noted in the text, * indicates statistical significance using Student's t test when compared to the control group that did not receive adjuvant. Bars represent standard error.

FIG. 5.

HBsAg (sAg) and HBcAg (cAg)-specific IFN-γ and IL-4 ELISPOT responses following immunization with a dual HBsAg/HBcAg vector and CT A + B or LT A + B vectors containing and lacking signal peptide coding sequences. All groups (n = 5) received primary and booster immunizations spaced 4 weeks apart. Blood samples and splenocytes were collected 2 weeks following the second immunization. (A) HBsAg-specific IFN-γ responses. (B) HBsAg-specific IL-4 responses. (C) HBcAg-specific IFN-γ responses. (D) HBcAg-specific IL-4 responses. ∗ indicates statistical significance using Student's t test when compared to the control group that did not receive adjuvant. Bars represent standard error. “w/o TPA” indicates that the vectors did not contain the human TPA signal peptide coding sequence.

FIG. 6.

HBsAg (sAg) and HBcAg (cAg)-specific IFN-γ and IL-4 ELISPOT responses following immunization with a dual HBsAg/HBcAg vector and individual CT A or B or LT A or B vectors containing and lacking signal peptide coding sequences. All groups (n = 5) received primary and booster immunizations spaced 4 weeks apart. Blood samples and splenocytes were collected 2 weeks following the second immunization. (A) HBsAg-specific IFN-γ responses. (B) HBsAg-specific IL-4 responses. (C) HBcAg-specific IFN-γ responses. (D) HBcAg-specific IL-4 responses. ∗ indicates statistical significance using Student's t test when compared to the control group that did not receive adjuvant. Bars represent standard error. “w/o TPA” indicates that the vector did not contain the human TPA signal peptide coding sequence.

FIG. 7.

Local skin reactogenicity in domestic pigs following epidermal inoculation of CT and LT vectors. CT and LT vectors, as well as a control sample of empty vector, were formulated onto gold particles and delivered into the inguinal epidermis of domestic pigs using the PowderJect XR gene delivery device at a helium pressure of 500 lb/in². Delivery conditions involved 0.5 mg of gold and 1.0 μg of DNA total per inoculation. Skin sites were observed on days 0, 2, 7, and 14. Days 0 and 7 are shown in panels A and B, respectively. EV, empty vector; CT, CT A and B subunit vectors; and LT, LT toxin A and B subunit vectors.