INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

The anterior nares are the major reservoir in humans of the opportunistic pathogen Staphylococcus aureus. Approximately 20% of humans are persistently colonized intranasally by a single strain of S. aureus. Another 60% of individuals are intermittent nasal carriers of S. aureus strains that change with varying frequency. The remaining 20% are classified as persistent noncarriers (16). Although S. aureus colonization of the nares is asymptomatic, nasal carriage is a risk factor for subsequent infection, particularly in surgical patients, intensive-care-unit patients, patients with intravascular devices, and human immunodeficiency virus-positive individuals (16). Higher colonization rates have been observed in hospitals (among both patients and health care workers) and among intravenous drug users, insulin-dependent diabetics, human immunodeficiency virus-positive or AIDS patients, patients with S. aureus skin infections, and patients undergoing continuous ambulatory peritoneal dialysis or hemodialysis (16). Since most staphylococcal infections are of endogenous origin, measures to prevent S. aureus nasal carriage may lower the prevalence of infection in these high-risk groups.

The topical application of the antibiotic mupirocin has proven effective in attenuating staphylococcal colonization of the human nares (6, 8, 12). Furthermore, mupirocin treatment of patients undergoing hemodialysis, continuous ambulatory peritoneal dialysis, or surgery has resulted in a significant decrease in the rate of staphylococcal infections (5, 17, 22). However, mupirocin-resistant strains of S. aureus have recently been reported (9). Likewise, nosocomial epidemics remain of great concern because of the emergence of strains of S. aureus resistant to multiple antibiotics, including methicillin-resistant and vancomycin-tolerant strains. Alternative techniques for the elimination of S. aureus nasal carriage are necessary to circumvent the problems created by the use and misuse of antibiotics.

Animal models of nasopharyngeal colonization by microbial pathogens, including Streptococcus pneumoniae (20, 33, 34) and Haemophilus influenzae (15, 35), have been developed for the study of bacterial colonization factors and protective mucosal immunity. Although numerous animal models of staphylococcal infection have contributed to the knowledge of virulence factors involved in disease, only a few reported studies have examined the interactions of S. aureus with the nasal mucosa in an animal model (23, 24, 27). In this paper, we describe a new mouse model for studying S. aureus nasal colonization. Future studies will employ this model to identify staphylococcal components that are critical for colonization or that can elicit protective immunity against carriage.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Animals. Male ICR, BALB/c, and C57BL/6 mice were obtained from Harlan Sprague Dawley, Inc. (Indianapolis, Ind.). The mice were 6 to 8 weeks of age upon arrival and were given food and water ad libitum. The animals were housed with three to eight animals per cage in a modified barrier facility under viral-antibody-free conditions. Animal care was in accordance with the institutional guidelines set forth by Brigham and Women's Hospital and Harvard Medical School.

S. aureus strains and growth conditions. S. aureus Reynolds and Newman are human isolates. Strain DAK was recovered from an outbreak of murine mastitis in the Animal Resources Center at Harvard Medical School in Boston. All three isolates produce a type 5 capsule (CP5). Strain JL236 is a capsule-deficient mutant of strain Reynolds created by transposon mutagenesis (1). S. aureus strains were marked with resistance to streptomycin by selection of bacteria on tryptic soy agar (TSA) plates containing streptomycin sulfate (Sigma, St. Louis, Mo.) at 500 µg/ml. Bacteria were grown at 37°C for 24 h on Columbia medium (Difco Laboratories, Detroit, Mich.) containing 2% NaCl and 1.5% Bacto-Agar (Difco). Cells were harvested from the solid medium with a sterile loop and suspended in saline (150 mM NaCl). The total number of CFU in each inoculum was quantitated by plating duplicate serial dilutions of the inoculum on TSA. To prepare the inoculum containing mouse-passaged S. aureus, an aliquot of the pooled nasal-tissue suspension from five experimentally colonized mice (5 days after inoculation) was plated on Columbia agar with 2% NaCl and streptomycin (500 µg/ml). Bacteria were processed for inoculation as described above.

Nasal colonization model. Mice were anesthetized by intraperitoneal injection of pentobarbital (Nembutal Sodium Solution; Abbott Laboratories, North Chicago, Ill.) at 62.5 mg/kg of body weight. The inoculum, which contained a dose of S. aureus ranging from 10⁴ to 10⁸ CFU in 10 µl of saline, was pipetted slowly onto the nares of the anesthetized mice without actually touching the pipette tip to the nose. Two to 27 days after inoculation, the animals were euthanized with CO₂ and evaluated for nasal carriage of S. aureus. The nasal region was wiped externally with 70% ethanol, and the nasal tissue was excised and dissected with sterile scissors. The sample was then vortexed vigorously in 200 µl of saline for 15 s. The total nasal flora was evaluated by plating a 50-µl aliquot of the nasal suspension on TSA with 5% sheep blood, whereas the total number of S. aureus CFU per nose was assessed by plating aliquots of neat or diluted nasal suspensions on TSA with streptomycin (500 µg/ml). The carriage rate was defined as the number of animals with 1 CFU/nose divided by the total number of mice per group. The lungs from selected groups of animals (removed 5 to 27 days after inoculation) were excised, washed with saline, and homogenized in 0.5 ml of tryptic soy broth. Homogenized lung tissue was plated in triplicate on TSA with 5% sheep blood. Heparinized blood samples were obtained by tail vein puncture on day 5 or 13, and duplicate 100-µl aliquots were plated on TSA with 5% sheep blood.

Statistics. S. aureus carriage rates of animals in each experimental group were evaluated statistically by Fisher's exact test (two tailed). The Kruskal-Wallis test was used for comparison of quantitative cultures from different experimental groups. Pairwise comparisons were performed by Dunn's multiple-comparisons procedure. The dose required to achieve nasal colonization in 50% of animals (CD₅₀) at days 7 and 14 was determined by logistic regression (31).

Quantitation of CP5 produced by S. aureus nasal isolates. S. aureus Reynolds was cultivated in vitro at 37°C for 24 h on solid or in liquid Columbia medium containing 2% NaCl. The inoculum containing bacteria cultivated on solid medium was prepared as described above. The inoculum grown in liquid medium was prepared by harvesting the cells by centrifugation, washing them once with saline, and then resuspending them in saline. Two groups of mice were inoculated intranasally (i.n.) with 10⁸ CFU of S. aureus grown on solid or in liquid medium. After 5 days, nasal suspensions were prepared from individual animals, the nasal tissue was removed from each suspension, the samples were pooled, and bacteria were recovered from pooled samples by centrifugation. Cell-associated CP5 present on in vitro-cultivated and in vivo-isolated bacteria was measured by an enzyme-linked immunosorbent inhibition assay (1, 18).

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Establishment of S. aureus nasal colonization in ICR mice. One hundred eighteen mice were inoculated i.n. with approximately 10⁸ CFU of S. aureus Reynolds marked with resistance to streptomycin. At 2 to 27 days after inoculation, groups of 5 to 10 mice were sacrificed and the nasal tissue from each animal was cultured quantitatively. All of the animals survived until the day of sacrifice with no noticeable complications. At the time of sacrifice, 95 (81%) of the 118 mice were nasally colonized with 1 CFU of S. aureus; 80 (91%) of 88 mice maintained colonization between 2 and 13 days, and 15 (50%) of 30 mice remained colonized between 14 and 27 days (Fig. 1).

View larger version (15K): [in this window] [in a new window]   FIG. 1.   Nasal colonization by S. aureus Reynolds. ICR mice, given plain water or water containing streptomycin (Sm) at 1 mg/ml, were inoculated i.n. with 108 CFU of S. aureus Reynolds. The nares were cultured for S. aureus at intervals from 2 to 27 days after inoculation. Each circle represents the number of CFU recovered from the nares of an individual mouse at the time of sacrifice. The circles below the line (log10 CFU/nose = 0) represent results with animals not colonized with S. aureus (i.e., those with <1 CFU/nose).

Effect of S. aureus mouse passage on nasal colonization. ICR mice were inoculated i.n. with 3 × 10⁸ CFU of strain Reynolds that had been cultivated in vitro or cultivated from the nares of experimentally colonized mice on day 5. Nasal carriage was assessed from groups of 5 to 15 mice 5 to 14 days after inoculation. At each time point, mice inoculated with mouse-passaged S. aureus had nasal carriage rates and total numbers of CFU of S. aureus in the nares similar to those of mice inoculated with staphylococci cultivated in vitro (data not shown).

Nasal colonization of inbred BALB/c and C57BL/6 mice. Levels of experimental nasal colonization of inbred mouse strains BALB/c and C57BL/6 were compared with that established for the outbred mouse strain ICR. The mice were inoculated i.n. with approximately 10⁸ CFU of mouse-passaged S. aureus Reynolds, and nasal carriage was assessed in groups of six to eight mice 3 to 14 days after inoculation. Results were compared with those for ICR mice colonized with mouse-passaged S. aureus. As shown in Fig. 2, the carriage rates for the different groups of mice were similar at each time point. However, 5 days after inoculation, the number of CFU recovered from the nares was greater for ICR mice than for the two inbred strains (P = 0.0035). All subsequent experiments were therefore performed with ICR outbred mice.

View larger version (16K): [in this window] [in a new window]   FIG. 2.   Nasal colonization of outbred (ICR) and inbred (BALB/c or C57BL/6) mouse strains with S. aureus Reynolds. Mice were inoculated i.n. with approximately 108 CFU of mouse-passaged S. aureus Reynolds, and the nares were cultured for S. aureus 3 to 14 days after inoculation. Each symbol represents the number of CFU recovered from the nares of an individual mouse at the time of sacrifice. The circles below the line (log10 CFU/nose = 0) represent results with animals not colonized with S. aureus (i.e., those with <1 CFU/nose).

Effect of bacterial dose on nasal colonization. The effect of inoculum size was evaluated with untreated mice and mice given streptomycin in their drinking water. Groups of animals were inoculated i.n. with S. aureus inocula ranging from 10⁴ to 10⁸ CFU/mouse. Nasal carriage was quantitated 7 days later. At a dose of 10⁸ CFU of strain Reynolds, similar colonization rates were achieved with streptomycin-treated and untreated mice (Fig. 3). Untreated mice inoculated with 10⁸ CFU of two other S. aureus isolates, Newman (human) and DAK (murine), also had carriage rates equivalent to those obtained with strain Reynolds (data not shown).

View larger version (22K): [in this window] [in a new window]   FIG. 3.   Nasal colonization of ICR mice with S. aureus Reynolds at various inoculum doses. Animals, given plain water or water containing streptomycin (Sm) at 1 mg/ml, were inoculated i.n. with doses ranging from 104 to 108 CFU of S. aureus. Each circle represents the number of CFU recovered from the nares of an individual mouse 7 days after inoculation. The median log10 CFU/nose value for each group of animals is represented with a horizontal line. The circles below the line (log10 CFU/nose = 0) represent results with animals not colonized with S. aureus (i.e., those with <1 CFU/nose). The P values derived from the statistical comparison of colonization rates and numbers of CFU recovered from the nares of untreated or streptomycin-treated mice are shown below the graph. A P of <0.05 was considered significant. n.s., not significant.

Influence of CP5 expression on nasal colonization. S. aureus capsule expression is highly influenced by environmental growth conditions (11, 13, 18, 28, 29). Although we have shown that CP8 is expressed in vivo in a rat endocarditis model (18), Herbert et al. reported that CP5 was not expressed in a rat granuloma pouch model (13). To determine whether CP5 was expressed during nasal colonization by S. aureus, two groups of mice were inoculated i.n. with strain Reynolds cultivated under conditions of high CP5 expression (solid Columbia salt medium) or low CP5 expression (liquid Columbia salt medium). After 5 days, nasal samples were obtained from individual animals and pooled in order to obtain a sufficient bacterial biomass for quantitation of CP5 expressed by S. aureus recovered (without subculture) from mouse nares.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Many studies have addressed the role of S. aureus virulence determinants in the pathogenesis of infections. In contrast, very little is known about microbial factors that influence staphylococcal colonization. The nose is regarded as the primary site of S. aureus carriage, the source from which bacteria spread to other parts of the body. Nasal carriage of this endogenous pathogen can lead to skin colonization, which in turn poses a risk of infection if the skin is traumatized (e.g., by wounds, surgical procedures, or catheter insertion). Moreover, colonization of health care workers may contribute to the transmission of staphylococci to patients. Colonization is a multifactorial process that requires a variety of adaptive mechanisms, including nutrient acquisition, adherence to host tissues, and evasion of, or protection against, host defenses. Identifying bacterial factors that influence nasal colonization is best accomplished with an intact living animal.

We have developed a model for S. aureus nasal colonization in mice, establishing stable nasal carriage with three S. aureus strains, over a range of bacterial inocula, in inbred or outbred mouse strains over periods of 2 to 27 days. Although colonization persisted for 2 weeks in untreated animals, mice given streptomycin in their drinking water had enhanced long-term colonization (persisting for at least 27 days). Furthermore, treatment with streptomycin lowered the inoculum dose necessary to achieve persistent colonization. These observations suggest that suppression of the indigenous nasal flora results in reduced bacterial interference and creates an environment amenable to S. aureus colonization. The natural flora may inhibit experimental S. aureus nasal colonization by competing for binding sites or available nutrients. Alternatively, secreted products elaborated by members of the indigenous flora may inhibit staphylococcal growth. In untreated mice, a large inoculum dose (10⁸ CFU) of S. aureus was capable of overcoming this apparent interference. Streptomycin treatment may be critical in comparisons of the abilities of different strains or mutants to colonize the nares, since a large inoculum may overcome the effects of a phenotypic difference or mutation in staphylococci. We considered that human S. aureus isolates (Reynolds and Newman) might not colonize the nares of mice as readily as a staphylococcal strain originally isolated from mice (DAK). However, all three strains had similar patterns of nasal colonization. Although inbred mice offer the advantage of genetic relatedness, we chose to develop the nasal colonization model in outbred animals since their makeup is more representative of the variability observed within the human population.

Our results suggest that the presence of S. aureus in the nares of inoculated mice is a result of stable colonization rather than temporary contamination. The persistence of nasal carriage of S. aureus and the comparable abilities of lower and higher inoculum doses to result in colonization support this conclusion. The paucity of S. aureus recovered from the lungs and blood of mice inoculated i.n. indicates that nasal carriage was not dependent upon or associated with infection at another site. These results are consistent with the fact that S. aureus causes infection in competent hosts only when mucosal or skin barriers are breached. Because sequential culturing (swabbing) of the nares of a single animal might traumatize the nasal mucosa and result in a wound infection rather than colonization, we thought it necessary to culture the nasal tissue of individual animals at a single time point. Nasal carriage of H. influenzae (15) and Streptococcus pneumoniae (20) had previously been assessed by sequential cultures of nasal lavage fluid from individual animals. However, this method may adversely affect colonization by reducing bacterial numbers or altering environmental conditions of the nasal mucosa.

Our model of S. aureus nasal colonization should be useful for identifying bacterial factors that influence carriage and for assessing targets for protective immunity. Animal models have been developed for studying nasopharyngeal colonization by encapsulated bacterial pathogens such as H. influenzae and Streptococcus pneumoniae. Although the role of capsular polysaccharides in nasopharyngeal carriage is unknown, capsular antibodies have been shown to reduce nasopharyngeal colonization by H. influenzae type b in infant rats (15) and to protect against transmission of Streptococcus pneumoniae from colonized infant rats to their littermates (20). In addition, intranasal immunization of adult mice with a capsular polysaccharide 6B-tetanus toxoid conjugate reduced nasopharyngeal carriage of Streptococcus pneumoniae (33). The S. aureus capsule is antiphagocytic and enhances virulence in certain animal models of infection (14, 21, 30). In the present study the levels of CP5 produced by S. aureus recovered from the nares of experimentally colonized mice were similar to the high levels of CP5 produced by staphylococci cultivated in vitro on solid medium. Likewise, high levels of CP8 were expressed by S. aureus in endocardial vegetations recovered from a rat model of endocarditis (18). Furthermore, a capsule-deficient mutant of S. aureus did not colonize the nares as effectively as the parent strain. Capsule production may influence nasal colonization by circumventing immunoglobulin A-mediated clearance from the nasal mucosa and by protecting the bacterium from desiccation in the nasal vestibule.

Most experiments addressing mechanisms of S. aureus adherence have used cultured mammalian cells or microtiter plates coated with purified host matrix proteins. Although this work has revealed a large array of staphylococcal adhesins, the role that these adhesins play in colonization can be determined only in humans or in an animal model of S. aureus carriage. Shuter et al. (26) examined the in vitro interactions between S. aureus and desquamated human nasal epithelial cells by light microscopy. They observed avid binding of the staphylococci to both cell-bound and cell-free mucus. When S. aureus was added to purified human nasal mucin bound to microtiter plates, they found that the adhesin-receptor interaction involved bacterial proteins and the carbohydrate moiety in mucin. The premise that S. aureus adherence to nasal mucin may be more important than its interaction with epithelial cells was corroborated by the results of Sanford et al. (23, 24). These investigators challenged ferrets i.n. with S. aureus and then sacrificed the animals after 60 to 90 min to examine the early interaction between the microbe and host. Most of the staphylococci were associated with the mucous gel overlaying the mucosa. Moreover, in vitro binding to nasal mucin was inhibited by pretreatment of the staphylococci with trypsin (24). Although these data support the premise that S. aureus adherence to mucin is protein mediated, other reports have indicated that lipoteichoic acid or cell wall teichoic acid may be involved in S. aureus adherence to nasal epithelial cells in vitro (2-4, 7, 32). Our model might be used to observe the possible inhibitory impact of mutations or antibodies directed against such putative adhesins on the establishment of nasal colonization. It might also serve as a model to identify genes expressed during or essential to the colonization process by methods such as in vivo expression technology or signature-tagged mutagenesis, both of which have proven to be successful in identifying virulence factors expressed during experimental S. aureus infection of animals (10, 19, 25).