GI-19007, a Novel Saccharomyces cerevisiae-Based Therapeutic Vaccine against Tuberculosis

Abstract

As yet, very few vaccine candidates with activity in animals against Mycobacterium tuberculosis infection have been tested as therapeutic postexposure vaccines. We recently described two pools of mycobacterial proteins with this activity, and here we describe further studies in which four of these proteins (Rv1738, Rv2032, Rv3130, and Rv3841) were generated as a fusion polypeptide and then delivered in a novel yeast-based platform (Tarmogen) which itself has immunostimulatory properties, including activation of Toll-like receptors. This platform can deliver antigens into both the class I and class II antigen presentation pathways and stimulate strong Th1 and Th17 responses. In mice this fusion vaccine, designated GI-19007, was immunogenic and elicited strong gamma interferon (IFN-γ) and interleukin-17 (IL-17) responses; despite this, they displayed minimal prophylactic activity in mice that were subsequently infected with a virulent clinical strain. In contrast, in a therapeutic model in the guinea pig, GI-19007 significantly reduced the lung bacterial load and reduced lung pathology, particularly in terms of secondary lesion development, while significantly improving survival in one-third of these animals. In further studies in which guinea pigs were vaccinated with BCG before challenge, therapeutic vaccination with GI-19007 initially improved survival versus that of animals given BCG alone, although this protective effect was gradually lost at around 400 days after challenge. Given its apparent ability to substantially limit bacterial dissemination within and from the lungs, GI-19007 potentially can be used to limit lung damage as well as facilitating chemotherapeutic regimens in infected individuals.

Keywords: BCG vaccine; T cells; Th1 cells; Th17 cells; recombinant yeast vaccine; therapeutic; therapeutic vaccine; tuberculosis.

FIG 1

Structure and expression of the four target antigens in the yeast-based Tarmogen platform to construct the fusion protein vaccine GI-19007. (A) The genes encoding hypoxia antigens Rv1738, Rv2320, Rv3130, and Rv3841 were fused in frame and subcloned into a yeast 2μm expression plasmid under the control of the copper-inducible CUP1 promoter. A six-amino-acid (aa) N-terminal leader (MADEAP) was added for improved metabolic stability, and a C-terminal hexahistidine tag was included for antigen detection and quantification. (B) GI-19007 was cultured in medium lacking uridine (U2) and treated with 500 μM CuSO₄ to induce protein antigen expression. Two micrograms of yeast lysate protein (GI-19007) and a dilution series of 1 to 10 pmol of His-tagged HCV protein standard (NS3-his std; observed molecular weight, 32) were subjected to SDS-PAGE and Western blot analysis with an anti-(His)₆ tag monoclonal antibody; 1 YU = 10⁷ yeast cells. (Far left lane) Precision Plus protein standard with sizes listed in kDa. Expression levels of the ~118 kDa 4-hypoxia antigen ranged from 9,000 ng/YU to ~14,200 ng/YU (shown) for different lots of GI-19007.

FIG 2

Immunization with GI-19007 elicits an antigen-specific IFN-γ response in mice. C57BL/6 mice were immunized intradermally with 2.5 YU per flank of GI-19007 or empty yeast vector (YVEC) on days 0 and 56. Ten days later, spleens were removed and stimulated in a 24-h ELISpot assay with 6 μg/ml of pooled recombinant antigens or the same concentration of irrelevant antigen. The ratio of IFN-γ ELISpot counts between the two are expressed as a stimulation index.

FIG 3

Vaccination of mice elicits a substantial IL-17 response. Mice were vaccinated with GI-19007 as described above and then challenged by low-dose aerosol with Beijing strain 212. Results are compared to those of controls given BCG or combination vaccination in a prime-boost format. Lung cells were harvested by tissue digestion and analyzed by flow cytometry. (A) Gating strategy. (B) GI-19007 induced strong Th17 responses, including within the CD8 T cell population.

FIG 4

Dose ranging studies in guinea pigs. Animals were infected with approximately 10 Beijing strain 212 bacteria and then vaccinated with GI-19007 10, 25, and 40 days later. Empty platform (YVEC) was used as a negative control. The bacterial load in the lungs was determined on day 70. While the bacterial load in the lungs was significantly reduced (P < 0.03) by 0.9 log in animals given 1 YU of vaccine, no protection was seen at higher doses. Data shown as mean ± SEM (n = 5 guinea pigs).

FIG 5

Whole-lung scans of representative samples stained with hematoxylin and eosin. (A) Saline control; (B) 3 YU YVEC control; (C) 1 YU GI-19007; (D) 3 YU GI-19007; (E) 10 YU GI-19007. Scans shown are from animals infected 70 days earlier with 10 to 20 M. tuberculosis Beijing strain 212 bacteria.

FIG 6

Kaplan-Meier survival curves for saline controls (open circles) and YVEC controls (open squares) versus animals vaccinated therapeutically with GI-19007 (closed circles). Because of curve overlap over the first 150 days, the curves were not significantly different, but substantially increased survival was seen in about one-third of the vaccinated animals past day 260. Each group contained 10 guinea pigs.

FIG 7

Demonstration that therapeutic vaccination with GI-19007 in guinea pigs that were previously given BCG vaccination prior to challenge with Beijing strain 212 did not improve protection in the lungs compared to BCG alone but significantly prevented dissemination to the spleen (P < 0.01). Data are shown as means ± standard errors of the means (SEM) (n = 5 guinea pigs).

FIG 8

Whole-lung scans of representative samples stained with hematoxylin and eosin. (A) Saline control; (B) YVEC control; (C) 1-YU GI-19007; (D) animals vaccinated with BCG prior to challenge; (E) BCG-vaccinated animals given GI-19007 therapeutically. Scans shown are from animals infected 70 days earlier with 10 to 20 M. tuberculosis Beijing strain 212 bacteria.

FIG 9

Kaplan-Meier survival curves for guinea pigs vaccinated with BCG prior to challenge (circles) compared to animals given BCG and then GI-19007 (squares) on days 10, 25, and 40 after challenge with Beijing strain 212. Through 300 days, survival in the double-vaccinated group was significantly improved compared to that of the BCG control group, but after this time the curves converged to the point this significance was lost. Each group contained 10 guinea pigs.

FIG 10

Whole-lung scans of representative samples stained with hematoxylin and eosin after harvest on day 330 of the infection. (A and B) BCG only; (C and D) BCG and then GI-19007 therapeutically.