Construction, genotypic and phenotypic characterization, and immunogenicity of attenuated DeltaguaBA Salmonella enterica serovar Typhi strain CVD 915

Abstract

A promising live attenuated typhoid vaccine candidate strain for mucosal immunization was developed by introducing a deletion in the guaBA locus of pathogenic Salmonella enterica serovar Typhi strain Ty2. The resultant DeltaguaBA mutant, serovar Typhi CVD 915, has a gene encoding resistance to arsenite replacing the deleted sequence within guaBA, thereby providing a marker to readily identify the vaccine strain. CVD 915 was compared in in vitro and in vivo assays with wild-type strain Ty2, licensed live oral typhoid vaccine strain Ty21a, or attenuated serovar Typhi vaccine strain CVD 908-htrA (harboring mutations in aroC, aroD, and htrA). CVD 915 was less invasive than CVD 908-htrA in tissue culture and was more crippled in its ability to proliferate after invasion. In mice inoculated intraperitoneally with serovar Typhi and hog gastric mucin (to estimate the relative degree of attenuation), the 50% lethal dose of CVD 915 (7.7 x 10(7) CFU) was significantly higher than that of wild-type Ty2 (1.4 x 10(2) CFU) and was only slightly lower than that of Ty21a (1.9 x 10(8) CFU). Strong serum O and H antibody responses were recorded in mice inoculated intranasally with CVD 915, which were higher than those elicited by Ty21a and similar to those stimulated by CVD 908-htrA. CVD 915 also elicited potent proliferative responses in splenocytes from immunized mice stimulated with serovar Typhi antigens. Used as a live vector, CVD 915(pTETlpp) elicited high titers of serum immunoglobulin G anti-fragment C. These encouraging preclinical data pave the way for phase 1 clinical trials with CVD 915.

FIG. 1

Primers were designed to amplify the 5′-proximal and 3′-distal regions of the guaBA operon. A homologous sequence engineered within the internal primers allowed PCR fusion of these two amplification products, creating a synthetic ΔguaBA locus. Prior to introduction of this inactivated locus into the chromosome of Ty2 by allelic exchange, a synthetic ars operon was inserted into the middle of ΔguaBA. This enabled selection for the desired recombinant strain in which the wild-type guaBA locus was replaced with ΔguaBA::ars within the chromosome of serovar Typhi strain Ty2.

FIG. 2

Southern blot for Ty2 and ΔguaBA::P_tac-ars derivative strain CVD 915, confirming the expected chromosomal insertion inactivating the guaBA locus. The probe used was a DIG-labeled ΔguaBA operon SstI-XbaI restriction fragment (2.5 kb). Marker refers to a DIG-labeled DNA molecular size marker with a size range of 0.12 to 23.1 kb (Boehringer Mannheim).

FIG. 3

Serum IgG responses to serovar Typhi flagella (top) and serovar Typhi LPS (bottom). Mice were immunized intranasally on day 0 and boosted on day 28 with 2 × 10⁹ to 4 × 10⁹ CFU of the indicated vaccine strains. Data represent individual titers from 8 to 10 mice in each group. Dotted lines plot the geometric mean titers. Significant increases in antibody titers that occurred after primary and secondary immunization (day 28 versus day 42) are shown.

FIG. 4

Serum IgG responses to fragment C of tetanus toxin. Mice were immunized intranasally on days 0 and 28 with 2 × 10⁹ to 4 × 10⁹ CFU of vaccine strains CVD 915 (dashed line) or CVD 915(pTETlpp) (solid line). Data represent individual titers from 10 mice in each group. Significant increases in antibody titer after primary (day 28) and secondary (day 42) immunization are indicated. Anti-fragment C titers measured on day 42 contained more than 2 lU/ml compared with a reference serum previously calibrated by the in vivo toxin neutralization test in mice.

FIG. 5

Proliferative responses to serovar Typhi flagella and LPS. Mice were immunized as described in the legend to Fig. 3 and sacrificed 60 days after primary immunization. Splenocytes were incubated in vitro with bacterial antigens, and proliferative responses were measured. Bars indicate responses at concentrations of 2 μg of serovar Typhi flagella per ml and 2 × 10⁵ bacteria of whole-cell heat-phenolized serovar Typhi per well. The results are expressed as mean cpm and standard error of the mean (SEM). Mice immunized with bacterial strains showed responses to bacterial antigens that were significantly higher than those of the control PBS group (∗, P < 0.05). The results shown are representative of two separate experiments.