Attenuation and persistence of and ability to induce protective immunity to a Staphylococcus aureus aroA mutant in mice

Abstract

Staphylococcus aureus is the most important etiological agent of bovine mastitis, a disease that causes significant economic losses to the dairy industry. Several vaccines to prevent the disease have been tested, with limited success. The aim of this study was to obtain a suitable attenuated aro mutant of S. aureus by transposon mutagenesis and to demonstrate its efficacy as a live vaccine to induce protective immunity in a murine model of intramammary infection. To do this, we transformed S. aureus RN6390 with plasmid pTV1ts carrying Tn917. After screening of 3,493 erythromycin-resistant colonies, one mutant incapable of growing on plates lacking phenylalanine, tryptophan, and tyrosine was isolated and characterized. Molecular characterization of the mutant showed that the affected gene was aroA and that the insertion occurred 756 bp downstream of the aroA start codon. Complementation of the aroA mutant with a plasmid carrying aroA recovered the wild-type phenotype. The mutant exhibited a 50% lethal dose (1 x 10(6) CFU/mouse) higher than that of the parental strain (4.3 x 10(4) CFU/mouse). The aroA mutant showed decreased ability to persist in the lungs, spleens, and mammary glands of mice. Intramammary immunization with the aroA mutant stimulated both Th1 and Th2 responses in the mammary gland, as ascertained by reverse transcription-PCR, and induced significant protection from challenge with either the parental wild-type or a heterologous strain isolated from a cow with mastitis.

FIG. 1.

(A) Schematic representation of the aro operon and surrounding genes according to the published sequences of S. aureus COL (GenBank). The gene downstream of aroA (similar to the B. subtilis gene that codes for hypothetical protein YpiA) is represented by the striped arrow, the aroCBA operon is represented by open arrows, and the gray arrow represents an unknown gene. The product of the aroA gene is 3-phosphoshikimate 1-carboxyvinyltransferase; the aroB product is 3-dehydroquinate synthase, and the product of the aroC gene is chorismate synthase. (B) Schematic representation of the Tn917 insertion (open box) in the 6.1-kb HindIII fragment of the FB306 strain (black lines). Relevant restriction sites in Tn917 are shown. The left and right HindIII transposon/chromosomal DNA junction fragments, the probe (black box), and the primers (arrows) are indicated. H, HindIII; B, BglII. (C) Southern blot analysis of S. aureus aroA mutant FB306. Chromosomal DNA was digested with HindIII (lane 1) or EcoRI (lane 2) and probed with a 1.8-kb BglII internal Tn917 fragment. Molecular size markers are shown in lanes M.

FIG. 2.

Survival of mice infected with S. aureus aroA mutant FB306. Groups of 10 Swiss mice were challenged by the i.v. route with 2 × 10⁷ CFU/mouse of wild-type RN6390 or aroA mutant suspended in PSS. Each point represents the percent survival of mice infected with the parental strain RN6390 or the aroA mutant. The levels of significance by Fisher's exact test were P = 0.04 at day 8 and P = 0.01 for days 22 and 31.

FIG. 3.

Histopathological analysis of kidney tissue 3 days after i.v. bacterial challenge with lethal doses of S. aureus RN6390 (top) or the aroA mutant (bottom). Parenchyma is replaced by confluent large abscesses and inflammatory infiltrate. Tubular necrosis and reduction and/or absence of glomeruli can also be seen (stained with hematoxylin-eosin; magnification, ×100). Histological changes of lesser magnitude were induced by the aroA mutant, compared with those found in mice challenged with the wild-type S. aureus.

FIG. 4.

Persistence of S. aureus aroA mutant FB306 after i.p. inoculation. In vivo growth of wild-type RN6390 and the aroA mutant in the lungs and spleens of Swiss mice. Each point represents the median value for 8 to 12 mice. For each time point (hours), the CFU of RN6390 versus CFU of the aroA mutant recovered from different tissues is depicted. The levels of significance by the Mann-Whitney test were P = 0.002 at 3 h (A) and P = 0.04 at 3 h (B).

FIG. 5.

Persistence of S. aureus aroA mutant FB306 in vivo. In vivo growth of wild-type RN6390 and the aroA mutant in the lungs and mammary glands of Swiss mice. (A) Each strain under study (6.5 × 10⁷ CFU/mouse) was injected intravenously. (B) Bacteria (1.9 × 10⁵ CFU/gland) were administered by the ima route. Each point represents the median value for 8 to 12 mice. For each time point (hours or days), RN6390 CFU versus the aroA mutant CFU recovered from different tissues are shown. The levels of significance by the Mann-Whitney test were as follows: P = 0.04 at 1 h (A) and P = 0.004 at day 1 and P = 0.03 at day 4 (B).

FIG. 6.

Protective efficacy against intramammary challenges by local immunization with the aroA mutant. Groups of female mice received two injections of the aroA mutant (5 × 10⁵ CFU/gland) or heat-killed RN6390 (5 × 10⁵ CFU/gland) by the ima route. Fourteen days after the second injection, mice were challenged by the same route with 0.05 ml of a suspension of (A) parental wild-type S. aureus RN6390 (5 × 10⁵ CFU/gland) or (B) S. aureus MB319 (heterologous strain isolated from milk of a cow with mastitis) (1 × 10⁶ CFU/gland). Each bar represents the mean ± standard error of the mean (SEM) (error bar) of CFU recovered from 10 mammary glands. The levels of significance by the Mann-Whitney test are as follows: P = 0.01 for mice immunized with heat-killed-RN6390 versus mice immunized with the aroA mutant; P = 0.02 (A) and P = 0.01 (B) for control mice versus mice immunized with the aroA mutant (*).

FIG. 7.

Histopathology of the mammary glands of mice immunized with the aroA mutant and nonimmunized mice 96 hours after local challenge with wild-type S. aureus. (Top) Moderate polymorphonuclear leukocyte and mononuclear cell infiltration and mild vascular congestion in the mammary glands of unvaccinated and RN6390-challenged mice (stained with hematoxylin-eosin; magnification, ×400). (Bottom) The mammary glands of mice immunized with the aroA mutant and mice challenged with RN6390 did not exhibit infiltration or vascular congestion (hematoxylin-eosin; magnification, ×400).

FIG. 8.

IFN-γ and IL-4 mRNA responses in the mammary glands of mice immunized with the aroA mutant and nonimmunized mice. At 96 hours after local challenge with wild-type S. aureus, the mammary glands from mice immunized with the aroA mutant (I) and control nonimmunized mice (C) were excised, and total RNA was extracted. RT-PCR was performed using cytokine-specific primers. Amplification of the housekeeping gene β-actin was performed to ensure that similar amounts of input RNA and similar efficiencies of reverse transcription are being compared. Results are presented as the means ± standard errors of the means (SEM) (error bars) of the ratio of cytokine optical density [OD]/actin OD. The values for immunized mice were significantly different from those for control mice (P = 0.02 by the Mann-Whitney U test [★]).