Prevalence of agr Dysfunction among Colonizing Staphylococcus aureus Strains

Bo Shopsin () 2 3 Alex Drlica-Wagner 0 2 Barun Mathema 0 2 Rajan P. Adhikari 1 2 Barry N. Kreiswirth 0 2 Richard P. Novick 1 2 0 The Public Health Research Institute , Newark, New Jersey 1 The Molecular Pathogenesis Program, Skirball Institute of Biomolecular Medicine, New York University School of Medicine , New York , New York 2 Received 25 February 2008; accepted 25 April 2008; electronically published XX August 2008. Potential conflicts of interest: B.S. has served as an advisor to and received research support from Pfizer. Presented in part: Infectious Diseases Fellow Educational Symposium , Scottsdale, AZ , January 2007. Financial support: American Heart Association (Fellow-to-Faculty Transition Award to B.S.); National Institutes of Health (grant R01-AI30138 to R.P.N.); Cystic Fibrosis Foundation (grant to B.S.). First Ave. , 2nd Fl., Lab. 1, New York, NY 10016 3 Department of Medicine, Division of Infectious Diseases Mutations in the staphylococcal virulence regulator gene agr frequently occur during Staphylococcus aureus infection. Whether agr-defective strains are fit for colonization, an important prerequisite for infection, is unknown. Screening by means of assays to detect -hemolysin activity and agr autoinducing peptide production indicated that 15 (9%) of 160 healthy human subjects were colonized with an agr-defective strain or a mixture of agr-positive and -defective S. aureus strains. The presence of identical agr-defective strains in family members suggests that these strains are transmissible. Additionally, carriage of an agr-defective strain was associated with hospitalization, raising the possibility that such strains may be selected in a nosocomial setting. - promoters; however, the P3 transcript RNAIII is the intracellular effector of target gene regulation. agr mutants are attenuated for virulence in animal models of acute infection, and the majority of clinical isolates are agr positive. However, recent reports show that agr-defective strains can be recovered from patients with a variety of nosocomial infections, suggesting that the occurrence of these strains may be favored under certain in vivo conditions [25]. In the present work, we have addressed the possibility that agr-defective Staphylococcus aureus may be recovered from colonization sites and infection sites and may therefore participate in the transmission of S. aureus. We used an assay to detect -hemolysin, the translation product of RNAIII, and a plating assay to detect agr autoinducing peptide (AIP) production as screens to estimate the prevalence of agr dysfunction among colonizing S. aureus. Because the anterior nares are the primary site of S. aureus colonization in humans, we obtained and analyzed nasal isolates. We report here that populations of staphylococci in which agr is wholly or partially defective can be obtained from the nares of healthy individuals. In 2 families, indistinguishable populations with wholly defective agr were obtained from different individuals, suggesting that agr-defective mutants are transmissible and can therefore initiate colonization. The only significant predisposing factor for colonization with an agr-defective strain was prior hospitalization. Subjects, materials, and methods. A total of 160 colonizing S. aureus isolates were obtained via culture of anterior nares swab specimens from 500 healthy children and their guardians as part of a previous study [6]. Study subjects came from a convenience sample of 212 households that varied in size from 25 persons, with 1 4 children and 1 or 2 guardians. Bacteria from the original cultures were suspended in CYGP medium, aliquoted, snap frozen, and stored at 80C, thereby preserving diversity in the primary bacterial population. To screen for agr functionality, a frozen aliquot of each specimen was thawed and inoculated on GL agar. Dilutions of bacteria from this inoculum were plated onto sheep blood agar (SBA) that had previously been spread with a -hemolysin containing culture supernatant, as described elsewhere [3, 7]. Cultures were provisionally scored as hemolytic, nonhemolytic, or, when 10% hemolytically distinct colonies were present, mixed. At least 100 colonies were scored for each isolate. Colonies were analyzed further by cross-streaking against strain RN4220 for hemolysin production and were tested for agr AIP Isolate type, subject, isolate number A2-A1-K1-E1-M1-B1-K1-B1 A2-A1-K1-E1-M1-B1-K1-B1 U1-F1-G1-J1-B1 T1-J1-M1-B1-M1-K1 Y1-H1-G1-F1-M1-B1-Q1-B1-L1-O1 Y1-H1-G1-F1-M1-B1-Q1-B1-L1-O1 X1-K1-B1 X1-K1-B1 Z1-D1-M1-D1-M1-N1-K1-B1 Z1-D1-M1-D1-M1-N1-K1-B1 Z1-D1-M1-D1-M1-N1-K1-B1 Z1-D1-M1-D1-M1-N1-K1-B1 spa cluster NOTE. AIP, agr autoinducing peptide; C, child; G, guardian; , absent; , present. a Family members from the same household. production [8]. Pure nonhemolytic and AIP-negative isolates were analyzed by Northern blotting to confirm the lack of RNAIII production, as described elsewhere [7]. Selected isolates were spa typed [9]. agr grouping had been previously determined by polymerase chain reaction (PCR) [10]. Nucleotide sequences were determined for the agrA and agrC genes by PCR and automated DNA sequencing, using published primers [7]. Sequences were compared with the known agr sequences of S. aureus strain COL (agr group I) and RN6607 (agr group II), using a sequence analysis suite (DNAStar). The Fisher exact test was used to compare frequencies of variables between S. aureus agrpositive and -negative isolates. Crude odds ratios (ORs) and 95% confidence intervals (CIs) were calculated as appropriate. Analysis was done with SAS, version 9.1 (SAS Institute). Results. Hemolysin production in S. aureus as visualized on SBA can be used to approximate agr activity because -hemolysin is a translation product of RNAIII and because -hemolysin (hla) transcription and translation are up-regulated by RNAIII [2, 3, 5]. By screening for -hemolysin activity, we found that 15 (9%) of 160 colonized subjects harbored only nonhemolytic staphylococci (9 subjects) or mixtures of hemolytic and nonhemolytic organisms (6 subjects) (table 1). There was only 1 methicillinresistant S. aureus (MRSA) isolate in the collection; this isolate was hemolytic. To test for agr activity among pure nonhemolytic cultures, we measured RNAIII by Northern blot hybridization and found that 7 of 9 nonhemolytic strains had no detectable RNAIII or contained only a trace (data not shown). Onset of RNAIII transcription in one of the isolates that produced a significant level of RNAIII (BK3574) was delayed, compared with onset in agr-positive laboratory strain RN6734. This finding was similar to that previously described for the agr-defective strain RN4220 [7]. The other strain (BK3579) showed a normal pattern of RNAIII transcription and was considered agr positive. An important phenotypic feature of the agr system is the exoprotein profile: agr-positive strains produce many exoproteins that are not produced by agr-negative (...truncated)