medium according to the method described by Kuijper et al. (32). The benzidine test was performed according to the method described by Deibel and Evans (9).
JOURNAL OF CLINICAL MICROBIOLOGY, Aug. 1993, p. 2105-2109 0095-1137/93/082105-05$02.00/0 Copyright C 1993, American Society for Microbiology
Vol. 31, No. 8
Twelve Aberrant Strains of Staphylococcus aureus subsp. aureus from Clinical Specimens CARLA FONTANA,`* LUIGINA CELLINI,2 AND BENEDETTO DAINELLI2 Dipartimento di Medicina Sperimentale e Scienze Biochimiche, Universita "Tor Vergata" di Roma, Via Orazio Raimondo 8, 00173 Roma, 1 and Istituto di Medicina Sperimentale, Universita "G. D'annunzio, " 66100 Chiet4 2 Italy Received 8 April 1993/Accepted 14 May 1993
A new biovar of Staphylococcus aureus subsp. aureus was isolated from human clinical specimens and described on the basis of studies of 12 isolates that were compared with 11 standard reference strains. Both DNA hybridization experiments and numerical taxonomy analysis demonstrated that these strains were strictly related to S. aureus subsp. aureus; however, they were significantly different from the latter. The atypical strains belonging to the new biovar can be distinguished from typical S. aureus subsp. aureus strains by their cx-chymotrypsin, ca-glucosidase, P-N-acetylglucosaminidase, lipase (C-14), and leucine arylamidase enzymatic activities and novobiocin resistance. Thus, the combination of a-glucosidase and 13-N-acetyl-glucosaminidase is more useful for distinguishing these S. aureus strains from the other, typical ones. In the past, the typically coagulase- and DNase-positive Staphylococcus aureus strains were the first and major subjects of studies regarding the genus Staphylococcus, mainly because of their role in hospital- and communityacquired infections (27, 28). More recently, however, numerous studies were performed of the most frequent strains, considered as important human pathogens that show a common pattern of coagulase-negative reaction and DNase positivity or coagulase-positive reaction and DNase negativity (17, 28, 36, 41, 43, 45, 46). The growing interest in this direction has led several investigators to introduce new species and new subspecies of staphylococci, supporting the theory of heterogeneity; therefore, 29 species and 10 subspecies are currently recognized (2, 10, 12, 16, 20, 21, 23, 24, 29, 31). Moreover, several studies dealing with more specific and innovative identification systems for the discrimination of these new taxa have been previously published (4, 8, 18, 26, 30, 33). The present article aims to describe a group of microorganisms from human specimens with unusual biochemical patterns. These strains, which belong to the genus Staphylococcus, share phenotypic characteristics with S. aureus subsp. aureus but constitute a separate biovar on the basis of both phenotypic and genotypic analyses. The main distinct enzymatic properties are a-chymotrypsin, a-glucosidase, and P-N-acetylglucosaminidase, lipase (C-14), and leucine arylamidase. MATERIALS AND METHODS Bacterial strains. A total of 23 strains of staphylococci were examined in the present study (Table 1). Twelve of these strains were from 50 Staphylococcus strains that belong to various species of the genus and were isolated during 6 months. These Staphylococcus types were isolated from 12 human specimens from different patients. In particular, nine, one, and two strains were cultured from ear swabs, from a finger abscess, and from infected wounds, respectively. These strains were examined together with the following standard reference strains: S. aureus subsp. aureus (ATCC 29213 and ATCC 25923), Staphylococcus epi*
Corresponding author.
dermidis (ATCC 14990 and ATCC 12228), Staphylococcus simulans (ATCC 27851), Staphylococcus sciuri (ATCC 29060), Enterococcus faecalis (ATCC 29212), Enterococcus faecium (ATCC 19434), Enterococcus durans (ATCC 11576), and five reference strains belonging to the collection of the Department of Experimental Medicine and Biochemical Science, "Tor Vergata" University of Rome (namely, S. simulans, Staphylococcus wameri, Staphylococcus cohnii, Staphylococcus xylosus, and Staphylococcus lentus). All samples were plated on Columbia agar (Unipath Limited, Bedford, England) supplemented with 5% sheep blood. The strains were stored at -20°C in Proteose Peptone broth with 15% glycerol. Biochemical characterization. Strains were identified and biochemically characterized with the API 20 STAPH, Rapid ID 32A STAPH, and API ZYM systems (Biomerieux, Balmes les Grottes, Montalieu-Vercieu, France) according to the general outlines given by the company. The identification tests with API 20 STAPH were performed immediately on freshly isolated strains and then repeated after storage, along with tests with the Rapid ID 32 A STAPH and API ZYM systems (Biomerieux). In particular, the following characteristics were determined by the API 20 STAPH systems (Biomerieux): D-glucose, D-fructose, D-mannose, maltose, lactose, D-trehalose, D-mannitol, xylitol, D-melibiose, raffinose, xylose, sucrose, ot-methyl-D-glucoside, and N-acetylglucosamine aerobic utilization; acetylmethylcarbinol, urease, arginine dihydrolase, and alkaline phosphatase production; and reduction of nitrate to nitrite. Ribose, cellobiose, turanose, and arabinose fermentation; ornithine decarboxylase, arginine arylamidase, ,B-galactosidase, P-glucoronidase, and pyrrolidonyl arylamidase activities; and novobiocin susceptibility were tested with the Rapid ID 32 STAPH system (Biomerieux). The API ZYM galleries were used to study enzymatic properties such as esterase (C-4), esterase-lipase (C-8), lipase (C-14), leucine arylamidase, valine arylamidase, cystine arylamidase, trypsin, a-chymotrypsin, phosphatase acid, naphthol-AS-BI-phosphohydrolase, a-galactosidase, a-glucosidase, ,B-glucosidase, Nacetylglucosaminidase, o-mannosidase, and a-fucosidase (44). P-Glucosidase activities were also tested with the API LRA ZYM Oxidase system (Biomerieux), which utilizes 2105
2106
J. CLIN. MICROBIOL.
FONTANA ET AL.
TABLE 1. List of Staphylococcus strains and their sources Species
Staphylococcus species
M211(6) M212(0) M213(4) M215(0)
M215(1) M216(4) M216(6) M218(8) M219(7) M219(9) S. xylosus S. cohnii S. simulans S. lentus S. warneri S. aureus subsp. aureus
M233(3) M39 M42 M189 M47 M282 29213
25923 S. epidermidis
Source
Straina
M280
14990 12228
S. simulans
27851
S. sciun
29060
Ear swab Ear swab Ear swab
pusb PuSb Ear swab Ear swab Ear swab Ear swab Ear swab Ear swab Pusc
Coll.d Coll. Coll. Coll. Coll. American Type Culture Collection American Type Culture Collection American Type Culture Collection American Type Culture Collection American Type Culture Collection American Type Culture Collection
Dept. Dept. Dept. Dept. Dept.
a Strains isolated from different patients. Pus from infected wound. c Pus from finger abscess. d Collection of the Department of Experimental Medicine and Biochemical Science, "Tor Vergata" University of Rome. b
chromogenic substrates different from those of the API ZYM system. Each strain was also analyzed for catalase activity (22). The coagulase tube test was done with rabbit plasma (Biomerieux). DNase tests were done with the Staphynuclease kit (Biomerieux). Oxidase activities were determined according to the method described by Faller and Schleifer (14). Colony morphology and pigment were determined according to the method described by Kloos and Lambe (28). Clumping factor was determined by using a commercially available system (Slides Staph-kit; Biomerieux). Esculin hydrolysis was performed either by using the Rapid ID 32 STAPH system (Biomerieux) or by using an esculin medium according to the method described by Kuijper et al. (32). The benzidine test was performed according to the method described by Deibel and Evans (9). Growth on thioglycolate broth and hemolysis on sheep blood agar were performed according to the method described by Kloos and Jorgensen (27). Numerical taxonomy. Of the original 56 features studied, 55 were taxonomically analyzed (esculin hydrolysis was not included). The simple matching coefficient and unweighted average linkage were used (37, 39, 42). DNA preparation and digestion. Genomic DNA was prepared by a procedure described by Bialkowska-Hobrzanska et al. (3). Samples were suspended in TE (10 mM Tris-HCl [pH 8], 1 mM EDTA [pH 8]; Sigma Chemical Co., St. Louis, Mo.) and stored at -20°C until use. DNA samples (10 ,ug) were cleaved with TaqI (1 U/Iug of DNA) according to the instructions of the manufacturer (Boehringer GmbH, Mannheim, Germany).
Preparation of radiolabelled DNA. Total chromosomal DNA cleaved with TaqI was labelled with 5'-[a-32P]dCTP and 5'-[a-32P]dATP (10 mCi/mmol; Amersham International, Buckinghamshire, England) with the random-priming labelling kit (Boehringer) (38). Labelled DNAs were purified on a Sephadex G-50 column (Sigma Chemical Co. [40]). The specific activities ranged from 2 x 105 to 5 x 106 cpm/,ug. Slot blot preparation and hybridization. Digested DNAs (10 ,ug/10 ,ul) were denatured at 95°C and chilled on ice for 5 min. Before DNAs were spotted onto nylon membranes (Hybond-N; Amersham International) with a slot blot machine (Hoefer Scientific Instruments, San Francisco, Calif.), a volume of 20x SSC (1 x SSC is 0.15 M NaCl plus 0.015 M sodium citrate) was added to each sample (25). The membranes were then denatured, neutralized, and fixed as recommended by the manufacturer. Prehybridization was performed at 65°C for 1 h in a mixture containing 6x SSC, 5 x Denhardt's solution, and 0.5 ml of freshly denatured salmon sperm DNA (1 mg/ml). Hybridization was performed under optimal (55°C) and stringent (70°C) conditions in a hybridization buffer (6x SSC, 5x Denhardt's solution, 500 ,tg of salmon sperm DNA), with 1 ,ug of heat-denatured 12p_ labelled DNA added, for 18 h (2, 31). The membranes were washed twice in 2x SSC-0.1% sodium dodecyl sulfate (SDS) at room temperature for 15 min, once in lx SSC-0.1% SDS at 65°C for 30 min, and then once in 0.lx SSC-0.1% SDS at the same temperature for 5 min. After overnight exposure on a Kodak-X-O-Matic film at -70°C, the film was developed with a Kodak-O-Matic M2 processor. The band patterns obtained were scanned with an Ultrascan Laser XL100 (Pharmacia LKB, Uppsala, Sweden) (37). The percent homology (percentage of relative binding) was expressed as the amount of heterologous binding divided by the amount of homologous binding times 100 (25). Divergence between DNA-DNA homology was evaluated by thermal-stability studies of DNA-DNA hybrids (ATm) (21). The percentage of guanine plus cytosine (G+C), expressed in moles percent, was determined according to the method described by Marmur and Doty (35; see also reference 1, 11, and 34).
RESULTS The 12 aberrant strains showed positivity for D-glucose, D-fructose, D-mannose, D-maltose, D-lactose, D-trehalose, D-mannitol, sucrose, N-acetyl-glucosamine, D-celiobiose, and D-turanose; meanwhile, no acid production was demonstrated by utilization of D-ribose, xylitol, xylose, D-melibiose, raffinose, L-arabinose, and a-methyl-D-glucoside. The strains gave positive results in catalase, coagulase, and benzidine reactions and were capable of nitrate reduction and acetylmethylcarbinol (acetoin) production. All strains were resistant to novobiocin. There was no production of oxidase, at-galactosidase, ,-glucoronidase, ,-galactosidase, valine arylamidase, cystine arylamidase, arginine arylamidase, trypsin, ornithine decarboxylase, a-mannosidase, and a-fucosidase. Results for DNase, clumping factor, urease, arginine dihydrolase, pyrrolidonyl arylamidase, leucine arylamidase, 3-N-acetylglucosaminidase, a-chymotrypsin, a-glucosidase, P-glucosidase, alkaline phosphatase, esterase C-4 and C-8, lipase (C-14), phosphatase acid, and naphtholAS-BI-phosphohydrolase were positive. All strains were able to grow on thioglycolate medium within 24 h. Esculin hydrolysis showed anomalous results; in fact, our atypical strains were clearly positive on esculin medium but only weakly positive on Rapid ID 32A STAPH gallery; in contrast, typical S. aureus subsp. aureus strains were negative
A NEW BIOVAR OF S. AUREUS SUBSP. AUREUS
VOL. 31, 1993
2107
TABLE 2. Differential characteristics of Staphylococcus spp. and S. aureus subsp. aureus No. of strains of S. aureus subsp. % of strains that No. of strains of Staphylococcus aureus with positive results spp. with positive results Characteristic were positive (n = wer2psitve(n =2) 2 50 6 Colony size of .5 mm 2 16 2 Colony pigment 2 100 12 Catalase 2 100 12 Coagulase 2 25 3b DNase 0 91 11 a-Chymotrypsin 0 83 10b ca-Glucosidase 91 11 2c 3-Glucosidase 0 83 10 P-N-Acetylglucosaminidase 0 91 11 Arginine arylamidase 0 100 12 Lipase (C-14) 0 100 12 Leucine arylamidase 1 100 12 Novobiocin resistance n, total number of strains tested. residual strains showing a very weak positive reaction. c Refers to the activity determined by the API LRA ZYM Oidase system.
a
b The
on both. All strains produced a wide zone of strong hemolysis within 24 to 36 h. The variable characteristics of these strains compared with those of S. aureus subsp. aureus are indicated in Table 2. For the standard reference strains of S. aureus subsp. aureus, tests for 0-glucosidase gave an unusual result. These strains gave negative results with the API ZYM system but positive results with the API LRA ZYM Oxidase system. This phenomenon could be explained by the different natures of the chromogenic substrates used in the two systems. Numerical taxonomy showed that our 12 strains clustered in a single homogeneous phenon with a similarity of 92%; the species with the closest linkage, showing a similarity of 82%, is S. aureus subsp. aureus. The remaining phena showing a linkage of similarity of less than 45% with S aureus subsp. aureus are formed of strains that belong to various species of staphylococci and members of the genus Enterococcus (Fig. 1). The genomic relationships among our strains and other strains belonging to different species of staphylococci and enterococci are indicated in Table 3. We also reported the G+C contents (in moles percent) and the values of divergence in thermal stability
(ATm).
Enterococcus faecalis
DISCUSSION During recent years, the genus Staphylococcus has been expanded to include more species and more subspecies (2, 10, 12, 16, 21, 23, 24, 31). Some researchers maintain that this goal should concern only the enterotoxic strains, and they consider a general expansion unnecessary (22). In contrast, various publications regarding coagulase-positive as well as coagulase-negative staphylococci clearly demonstrated the role of these as significant clinical pathogens apart from their enterotoxigenicity (5-7, 13, 15, 19). Thus, it has been suggested that an expansion in Staphylococcus classification is not only useful but should be greater than the one reported in the most recent edition of Bergey's manual (29). In accordance with this viewpoint and with the aim of correctly classifying some aberrant strains of S. aureus, we have demonstrated in the present paper a close relationship among typical S. aureus subsp. aureus strains and 12 atypical ones. Evidence for this was provided by a phenotypic similarity of 92% among the SAL (S. aureus-like) strains, which were shown to belong to a homogeneous phenon, and of 82% with members of S. aureus subsp. aureus, as
I
Staphylococcus epidermidis Staphylococcus simulans Staphylococcus sdurii
Staphylococcus aureus
I
subsp. aureus Staphylococcus aureLus (new biovar)
100 Percent Similarity FIG. 1. Dendrogram constructed from phenotypic data.
40
2108
J. CLIN. MICROBIOL.
FONTANA ET AL. TABLE 3. DNA homology of 12 aberrant Staphylococcus strains % Relative binding with labelled DNAs from:
Source of unlabelled DNA
S. aureus ATCC 25923
Staphylococcus sp. M280
ATaG+C (mol%)
ATma
55°C
70°C
55°C
70°C
94 90
90 88 84 90 89 93 90 87 92 90 91 83 100 30 16 20 25 10 9 8
