Bovine and rabbit models for the study of a Staphylococcus aureus avirulent mutant strain, RC122

Abstract

Staphylococcus aureus is the main etiological agent of bovine mastitis. Intramammary infections are difficult to cure and vaccination appears to be an alternative to prevent the disease. Research has focused on the development of mutants affected in the synthesis of pathogenicity determinants. We constructed a mutant strain (RC122) after chemical mutagenesis. In a mouse model, the strain was shown to be 1500 times less virulent, showed similar kinetics of disappearance in the kidney as its parental strain, and a good degree of protection against a challenge from homologous and heterologous strains.

The objective of the present report was to study the avirulent RC122 S. aureus mutant strain in rabbit and bovine infection models. The results clearly show that RC122 was less virulent than its parental strain in a rabbit skin model, and was also correlated with its avirulence as an udder pathogen. These traits make the RC122 mutant strain interesting as a potential strain for an experimental vaccine trial in dairy herds.

As a main etiological agent of bovine mastitis (1), Staphylococcus aureus causes intramammary infections that are very difficult to cure. Vaccination appears to be an interesting alternative to prevent the disease (2). With this purpose, research has focused on different types of vaccines prepared with S. aureus bacterin, or composed of pseudocapsule-enriched bacterins supplemented with alpha or beta toxoids (3,4,5,6). The success of these vaccines is difficult to establish. Also, the use of avirulent S. aureus mutants, which are affected in the synthesis of pathogenicity determinants, could be a valuable tool to prevent mastitis (7,8).

Several mutants have been constructed using different methods (7,9,10,11). Mutants isolated in our laboratory after ultra-violet (UV) mutagenesis were deficient for several characteristics; such as, the production of most exoproteins, an altered phage type, and/or colonial morphology in serum soft agar medium (12,13,14,15). They also differed in virulence when they were assayed by intraperitoneal administration in mice and showed a correlation between lower virulence and slower growth rate of the mutants (14). In other assays, we constructed a mutant (RC122), derived from S. aureus bovine mastitis strain after N-methyl-N'-nitro-N-nitrosoguanidine mutagenesis (MNNG). In a mouse model, the mutant was 1500 times less virulent, had similar kinetics of disappearance in kidney than its parental strain, and provided a good degree of protection against challenge with homologous and heterologous strains (16). The reason for the diminished virulence of RC122 remains to be elucidated. The low-virulence value of this mutant makes it interesting as a potential vaccine strain.

An essential requirement for the potential use of mutants as a vaccine is to study its behavior in different animal models. The objective of the present report was to study the avirulent RC122 S. aureus mutant in rabbit and bovine infection models.

The effects of inoculation with RC122 S. aureus mutant strain and its RC108 parental strain in rabbit skin, were assayed according to Adlam et al (17). Briefly, overnight cultures of S. aureus RC108 (a subclinical bovine mastitis strain isolated from a dairy cow in Argentina) (14), and S. aureus RC122 (an avirulent mutant derived from RC108 after MNNG) (16), were centrifuged at 9000 rotations per minute (rpm). Cells (10⁸ cfu/mL) and supernatants were collected. Rabbits (n = 5), were clipped in patches along their backs removing the hair. Each patch (n = 5), was inoculated intradermally with 0.1 mL of saline solution containing 10⁸ cfu of cells, or 0.1 mL of supernatant of RC108 and RC122 strains. One patch per rabbit was inoculated with saline solution as negative control. Macroscopic lesions were recorded for 5 d and evaluated by the diameter of swelling: negative (< 2 mm), small (2 to 5 mm), moderate (6 to 10 mm), and severe (> 10 mm). Necrosis of the area was defined as; negative, slight (< 2 mm), moderate (2 to 5 mm), and intensive (> 5 mm). At day 6 the rabbits were euthanized by cervical dislocation. Representative samples of skin along with underlying muscle were removed and fixed in buffered formalin for histological examination. Histopathological lesions were classified according to size and degree of necrosis and inflammation. These were defined as; 0) without inflammation or necrosis; 1) slight inflammation and diameter of necrosis was less than 1.2 mm; 2) moderate inflammation and diameter of necrosis was between 1.2 and 5.0 mm; and 3) intense inflammation and diameter of necrosis was more than 5.0 mm. After the lesions were transformed into metric data, they were analyzed by a statistical programme (Relative to an Identified Distribution (RIDIT), version 6.0; Epi Info software, Atlanta Center of Disease Control (CDC), Stone Mountain, Georgia, USA).

Macroscopically, the intradermal inoculations of RC108 cells showed intensive necrotic areas and a swelling larger than 11 mm after 24 h. On the other hand, cells of RC122 were significantly less virulent than parental strain (P = 0.042). None of the inoculation sites showed lesions more severe than those produced by RC108. In no case did the lesions reach a diameter of 11 mm and have from slight to moderate necrosis. The lesions produced by the intradermal inoculation of the supernatants were less severe than those produced by the inoculation of cells. Macroscopic examination showed that RC108 supernatants produced moderate swelling and slight necrosis. In contrast, supernatants of RC122 produced less damage (P = 0.031), than supernatants of the parental strain. Little inflammation was observed without necrosis.

Histopathological examinations of the lesions showed that cells of RC108 produced the most intense lesions. All of the samples had some evidence of extensive inflammation at the surface, with abundant hyperemia extending into the epidermis, dermis, hypodermis, and muscle layers. The necrosis was also intensive with a diameter higher than 5.0 mm (Figure 1A). In contrast, cells of RC122 mutant strain only produced lesions in the dermis layer with moderate inflammatory reaction and a less intensive necrotic area ranging from 1.2 to 5.0 mm (Figure 1B). Supernatants of RC108 produced an inflammatory reaction with moderate hyperemia (Figure 1C). The inoculated sites showed a necrotic area ranging from 1.2 to 5.0 mm and affecting the dermis. Although the supernatant of the mutant strain induced an inflammatory reaction similar to the parental strain, it did not produce a necrotic area on any of the layers examined (Figure 1D). These results show that RC108 cells caused the most severe macroscopic and histopathologic lesions.

Figure 1. Histopathological lesions produced by intradermal inoculation in rabbit skin. A,B) Lesions produced by the inoculation of cells of S. aureus RC108 and RC122 strains, respectively. Hematoxylin and eosin times 50. C,D) Lesions produced by the inoculation of supernatant of S. aureus RC108 and RC122 strains, respectively. Hematoxylin and eosin times 50 and times 100, respectively. E) Negative control. Hematoxylin and eosin times 50.

Six quarters of 2 Holando Argentina cows were inoculated with the RC122 mutant strain and its parental strain, RC108. Cows were in the 3rd month of their 1st lactation. Teat tips were aseptically cleaned and desinfected before inoculation. Prior to injection, no bacteria were found in milk samples, the somatic cells counts were less than 100 000 cells/mL and no clinical signs were seen. Two quarters were used for each treatment. Cows were inoculated by intramammary route with 1 mL of peptonated water cointaining 10³ of RC108 strain or 10⁷ of RC122 strain. The inoculum was deposited about 3 mm inside the teat. A quarter of each cow was inoculated with peptone water alone as a control. The appearance of the milk, the condition of the udder, and the general state of health of the animal were recorded daily for 12 d. Milk samples were aseptically collected from all quarters for bacterial and somatic cell counts once a day prior to morning milking. Bacterial numbers were determined by spreading samples on duplicate brain heart agar (BHA) plates. The number of colonies were counted and the results were expressed as the log10 bacteria/mL. The results of counting of somatic cells were expressed as log10 cells/mL.

The 2 quarters inoculated with 10³ cfu of RC108 developed clinical mastitis 18 h after inoculation. The analysis of bacterial and somatic cell counts showed a high increase, with reddening of the mammary gland and milk clots. Both quarters inoculated with RC108 showed fluctuations throughout the experiment. Although the inoculum of RC122 was 10 000 times higher (10⁷ cfu) than the parental strain, these quarters remained without any clinical signs and showed a similar fluctuation in bacterial count (Figure 2A). As expected, bacteria recovered from quarters inoculated with any of the strains showed the same phenotypical characteristics and virulence after intraperitoneal inoculation in mice as S. aureus RC108 or RC122. No bacteria were recovered in control quarters. Subsequent sampling of the quarters 8 mo after the end of the experiment, showed no bacteria from the quarters inoculated with RC122. On the other hand, RC108 could be recovered from these quarters despite showing no clinical signs of mastitis. Inoculated quarters did not receive any antibiotic therapy until the next lactation. Somatic cell counts of the quarter inoculated with the RC108 parental strain increased 1.03 log after day 5. Cell counts reached a peak value of 3.25 log of cells/mL, and remained high until the end of the experiment. The quarter inoculated with RC122 mutant strain showed a strong increase of somatic cell count at day 1, with a value of 3.76 log of cells/mL and decreased to 2.8 log of cells/mL during the subsequent days (Figure 2B). A rise or fall in total cell count, and the bacterial count generally responded accordingly. The milk yield had a decrease that was related to the raised levels of somatic cell and bacterial count.

Figure 2. The effect of intramammary inoculation with RC122 and RC108 S. aureus strains. A) Bacterial numbers; and B) Somatic cell count in milk.

Day 0 — before intramammary S. aureus inoculation.

The degree of lesions after inoculation with the RC122 mutant strain in rabbit skin was significantly lower than that of its parental strain. Even when both strains developed necrosis, the lesions caused by RC122 were less severe than those produced by RC108 and compromised only the inoculated layer. Supernatant inoculation of RC122 produced only slight inflammation without necrosis. Several studies indicate a strong relationship between α-hemolysin production and dermatonecrosis. Staphylococcus aureus PS80, a high α-hemolysin producer strain, caused severe skin lesions in rabbits and killed mice at very high doses by intraperitoneal inoculations (19). The diminished degree of skin lesions shown by RC122 may be a consequence of the low level of α-hemolysin produced (16). Rabbit skin lesions were similar to those produced by subcutaneous inoculation of 10⁸ cfu/mL of RC122 in cow skin in a previous study. Brock et al (20) found that the severity of the cow skin lesions was correlated with the virulence in the bovine udder as a mastitis pathogen.

When 10³ cfu of the RC108 parental strain was inoculated directly into the teat cistern, severe inflammation was observed 24 h after the inoculation. Clinical mastitis was the result of a rapid and intense inflammatory response; as indicated by the large numbers of somatic cells and bacterial counts in the milk after day 6, while milk yield decreased drastically. In contrast, teats inoculated with 10⁷ cfu of the RC122 mutant strain did not produce clinical mastitis, even when somatic cell counts reached very high values 48 h after inoculation.

The results clearly show that the RC122 mutant strain is less virulent than its RC108 parental strain. The ability to produce skin lesions in rabbit and the reduced virulence as an udder pathogen makes the RC122 mutant strain interesting as a potential strain for an experimental vaccine trial in dairy herds.