Development of a new live rough vaccine against bovine brucellosis

Full text: Brucella abortus S-19 is the most commonly used attenuated live vaccine to prevent bovine brucellosis. The vaccine induces good levels of protection in cattle, preventing premature abortion. Although B. abortus S-19 is the most used vaccine in eradication campaigns worldwide, it has two major problems: (i) it produces abortion when administered to pregnant cattle and is fully virulent for humans and (ii), the presence of smooth lipopolysaccharide interferes with the discrimination between infected and vaccinated animals during immunescreening procedures. In our laboratory we have previously cloned, sequenced and disrupted the gene coding for the enzyme phosphoglucomutase (pgm), responsible for the interconversion of glucose-6P to glucose-1P. The mutant does not synthesize the sugar nucleotide UDP-glucose and/or ADP-glucose and thus is unable to form any polysaccharide containing glucose, galactose or any other sugars whose synthesis proceeds through a glucose-nucleotide intermediate. The mutant has a rough phenotype, is avirulent in mice but retains the ability to multiply inside HeLa cells, although it shows a delay of the exponential intracellular replication. These characteristics prompt us to evaluate the potential use of this strain as a new live rough vaccine. We generated an unmarked deletion mutant of pgm. Western blot analysis of purified lipopolysaccharide from Δpgm indicated that it is devoid of O-antigen, however Δpgm whole cell extracts contained detectable amounts of O-antigen with a MW of 45 kDa, indicating that mutant strain is able to synthesize O-antigen but incapable to assemble a complete LPS probably due to the presence of an altered core structure. When administered intraperitoneally in Balb/C mice, the number of viable Δpgm recovered from spleens were, at all tested times, significantly lower than those inoculated with the parental virulent strain S2308 and was completely cleared at 8 weeks p.i., thus indicating a severe virulence reduction. These results indicate that even at high doses, strain Δpgm is cleared from the animal in a short period of time. The antibody response against O-antigen was evaluated using Fluorescence Polarization Assay with FITC-coupled O-antigen as a tracer. Mice receiving S2308 developed antibodies to the O-antigen that reached its maximal value at 49 days p.i (152.71±27.65 mP). In contrast, Δpgm-vaccinated mice failed, as the saline control mice, to elicit antibodies against O-antigen at any tested time (92.33±3.66 mP and 92.20±7.10 mP at 49 p.i. respectively). These results indicate that the O-antigen present in Δpgm is incapable to elicit a detectable specific antibody response. It is well established that for intracellular pathogens like Brucellae, cellular immunity plays a central role in protection. To investigate the cellular immune response induced by Δpgm, we analyzed the proliferative splenocytes response of vaccinated and non-vaccinated mice upon stimulation with heat-inactivated S2308 whole cells. At 8 weeks post inoculation, splenocytes recovered from mice vaccinated with Δpgm proliferated upon stimulation in a specific manner, in contrast to the non-vaccinated control group (60,142 ± 7,443 c.p.m. vs 20,855 ± 2,541 c.p.m., P<0.001). All immunized animals responded equally to the nonspecific mitogen Concanavalin A. The spleen cells from Δpgm vaccinated animals were induced to secrete high levels of IFN-γ (112.0 ng/ml) after stimulation. A significant induction was also observed upon stimulation with ConA (53.8 ng/ml). In contrast, the splenocytes from PBS-inoculated control animals only released IFN-γ upon stimulation with ConA. IL-4 was not detected in the supernatants of splenocytes obtained from both immunized and non-immunized animals. These results indicate that vaccination with Δpgm induce a classical ThI cellular proliferative response. In order to examine the protection induced by Δpgm, a vaccine-challenge experiment was performed. Groups of five mice were vaccinated intraperitoneally with 107 CFU of Δpgm, saline or 105 CFU of B19. Eight weeks post-vaccination animals were challenged with 105 CFU of strain S2308. Protection was defined as the difference between the numbers of viable bacteria recovered from spleens of immunized mice compared to those receiving saline. Vaccine efficacy was expressed as log10 units of protection. Δpgm generated a significant protection 2 and 4 weeks post challenge, with 2.25 and 1.93 protection units, respectively. As expected B. abortus strain S19 induced also a significant protection at 4 weeks (1.78 protection units). These results, together with the ability of the mutant to generate a strong cellular Th1 response without eliciting specific O-antigen antibodies, emphasize the potential use of this mutant as a new live vaccine for cattle. (author)