Antimicrobial Susceptibilities of Peptostreptococcus anaerobius and the Newly Described Peptostreptococcus stomatis Isolated from Various Human Sources

Abstract

Peptostreptococcus anaerobius sensu lato, currently including two closely related species, P. anaerobius and P. stomatis, is known to be more resistant than other gram-positive anaerobic cocci. We reidentified potential Peptostreptococcus isolates and tested their susceptibilities to eight antimicrobials. Notably, P. anaerobius had constantly higher values for the MIC at which 50% of the isolates are inhibited (MIC₅₀) and the MIC₉₀ than P. stomatis.

Gram-positive anaerobic cocci (GPAC) are clinically significant organisms recovered from mixed infections involving anaerobic bacteria (16, 19). This heterogeneous group has undergone considerable taxonomic changes, with new genera being formed from species previously belonging to the genus Peptostreptococcus (10, 11). Peptostreptococcus anaerobius was the only representative of the genus until Peptostreptococcus stomatis was recently described (9). P. anaerobius sensu lato has been frequently cultured from clinical specimens from the mouth, upper respiratory tract, skin and soft tissues, bone and joints, and gastrointestinal and genitourinary tracts (5, 6, 8, 16, 19, 20, 27, 28). Interestingly, Downes and Wade (9) suggested that the novel species, P. stomatis, originates from the oral cavity, whereas P. anaerobius sensu stricto has its habitat and is involved in infections under the waistline.

GPAC have been considered susceptible to most antimicrobials used against anaerobic infections (19). However, P. anaerobius sensu lato seems to exhibit more resistance than other GPAC (2, 13, 17, 18, 21, 23). No susceptibility data on P. anaerobius sensu stricto or P. stomatis exist thus far. The aim of the present study was to obtain information on their isolation sites and in vitro susceptibilities to eight antimicrobials potentially used to treat anaerobic infections.

Sixty-four isolates that had been previously identified as P. anaerobius based on their anaerobic growth, coccoid cell morphology, sensitivity to sodium polyanethol sulfonate (SPS), and major production of isocaproic acid in gas-liquid chromatography (16) were available from various clinical specimens of 64 subjects. The isolation sites were traced from referrals that were sent together with the isolate to be characterized. The isolates, collected between 1985 and 2005, were revived from frozen (−70°C) stocks and subcultured on brucella blood agar to assure their purity before further testing. Key phenotypic tests discriminative in the separation of P. anaerobius and P. stomatis were the size of the growth inhibition zone around an SPS disk and the production of proline arylamidase (9). The zone around an SPS disk (Oxoid, Basingstoke, United Kingdom) was measured in millimeters (the minimum size should be 12 mm) on a brucella agar plate after 3 days of anaerobic incubation. Enzyme activity patterns were determined by the Rapid ID 32A test kit (API system; bioMérieux, Marcy l'Étoile, France) according to the manufacturer's instructions. For isolates with discrepant phenotypic characteristics, partial sequencing of the 16S rRNA gene was performed (15) and compared to the sequences in GenBank (3).

The antimicrobial susceptibilities to amoxicillin, amoxicillin-clavulanic acid, cefoxitin, ertapenem, azithromycin, clindamycin, metronidazole, and moxifloxacin were tested by the Etest (AB Biodisk, Solna, Sweden). Bacteria were harvested from a 48-h anaerobic culture on prereduced brucella agar and fastidious anaerobe agar, both supplemented with 5% defibrinated sheep blood, hemin, and vitamin K₁. Bacterial suspensions made in 0.9% NaCl and adjusted to a turbidity of a 1 to 2 McFarland standard were applied to fresh brucella agar and fastidious anaerobe agar plates. The plates with Etest strips were incubated in an anaerobic chamber at 37°C for 48 h (for some slow-growing strains for 72 h) before the MICs were determined. Susceptibility breakpoints were determined as follows: amoxicillin, 0.5 μg/ml; amoxicillin-clavulanic acid, 4 μg/ml; cefoxitin, 16 μg/ml; ertapenem, 4 μg/ml; clindamycin, 2 μg/ml; metronidazole, 8 μg/ml; and moxifloxacin, 2 μg/ml (7). No breakpoint for azithromycin has been approved for anaerobes. β-Lactamase production was tested by using a nitrocefin disk (AB Biodisk) test.

Of the 64 isolates, the identity of 30 isolates was confirmed as P. anaerobius based on the size of SPS inhibition zone between 12 and <19 mm (here, 14 to 18 mm) and the Rapid ID 32A code 0400020000, including positive α-glucosidase and proline arylamidase reactions (9). Thirty-two isolates were reidentified as P. stomatis based on a size of the SPS inhibition zone of ≥19 mm (here, 19 to 34 mm) and the Rapid ID 32A code 0400000000, including positive α-glucosidase and negative proline arylamidase reactions (9). The identification of two isolates remained uncertain by biochemical testing; one blood isolate was proline arylamidase negative but with an SPS inhibition zone of 16 mm, and another isolate from an abscess of the scrotum showed typical key characteristics of P. anaerobius but was negative for α-glucosidase production. The sequencing identified the former isolate as P. stomatis and the latter isolate as most closely related to Peptoniphilus.

P. anaerobius isolates originated from miscellaneous clinical specimens representing various anatomical sites, especially from ulcer and skin specimens of the lower extremities (33%) and pus specimens from the genitourinary tract (27%), but only occasionally from specimens of the head and gastrointestinal tract, whereas 36% of the P. stomatis isolates originated from oropharyngeal specimens (six oral and six pharyngeal isolates) and 45% of the isolates originated from gastrointestinal specimens (12 appendicidal and three fecal isolates). One P. anaerobius isolate and two P. stomatis isolates were from blood.

The susceptibility data presented in Table 1 come from 61 Peptostreptococcus isolates, since two P. stomatis isolates failed repeatedly to grow with the Etest. P. stomatis was susceptible to all drugs except for one isolate with a reduced susceptibility (MIC = 3 μg/ml) to clindamycin, whereas four P. anaerobius isolates demonstrated intermediate or resistant MICs to one or more drugs: one isolate to amoxicillin, amoxicillin-clavulanic acid, cefoxitin, and azithromycin (48, 96, 24, and 24 μg/ml, respectively); one isolate to amoxicillin, amoxicillin-clavulanic acid, and cefoxitin (16, 32, and 24 μg/ml, respectively); one isolate to amoxicillin and amoxicillin-clavulanic acid (3 and 6 μg/ml, respectively); and one isolate to moxifloxacin (16 μg/ml). Although no breakpoint for azithromycin has been approved for anaerobes, an MIC of 24 μg/ml indicates resistance. Among 24 isolates collected during the 1980s, no resistance was seen, whereas 1 of 24 isolates (4%) from the 1990s and 4 of 13 (31%) isolates from 2000 onward showed resistance. P. anaerobius had constantly higher MICs at which 50% of the isolates are inhibited (MIC₅₀s) and MIC₉₀s than P. stomatis (Table 1). No β-lactamase production was detected.

TABLE 1.

Antimicrobial susceptibilities of the tested strains of P. anaerobius and P. stomatis

Antimicrobial agent	MIC (μg/ml) fora:
	P. anaerobius (n = 30)			P. stomatis (n = 31)
	Range	MIC50	MIC90	Range	MIC50	MIC90
Amoxicillin	0.047-48	0.125	0.38	<0.016-0.19	<0.016	0.094
Amoxicillin-clavulanate	0.047-96	0.125	0.5	<0.016-0.25	<0.016	0.094
Cefoxitin	0.25-24	0.38	0.75	<0.016-1.5	0.064	0.38
Ertapenem	0.094-4	0.25	0.38	0.016-1	0.047	0.19
Azithromycin	0.19-24	0.5	1	<0.016-1	0.19	0.75
Clindamycin	<0.016-1	0.19	0.75	<0.016-3	<0.016	0.016
Metronidazole	0.047-0.5	0.094	0.19	<0.016-0.25	<0.016	0.047
Moxifloxacin	0.064-16	0.125	0.125	0.023-0.75	0.094	0.125

^an, Number of strains.

Instead of grouping microorganisms together, the accurate characterization of bacterial species can reveal differences in their presumed habitats or infection patterns. Downes and Wade (9) suggested that P. stomatis originates from the oral cavity, whereas P. anaerobius sensu stricto is involved in infections below the waistline. Their suggestion was consistent with a previous report of Riggio and Lennon (22), who failed to detect P. anaerobius in oral specimens by species-specific PCR. In line with these studies was our finding that most P. anaerobius isolates were from miscellaneous infectious specimens of the lower extremities and genitourinary tract. In contrast, the majority of the P. stomatis isolates were not only from oral and pharyngeal sources but also from gastrointestinal specimens.

An interesting observation was that P. anaerobius had constantly higher MIC₅₀ and MIC₉₀ values than did P. stomatis. In studies dealing with antimicrobial susceptibilities of anaerobes, GPAC usually appear as an otherwise nonspecified group. However, an accurate identification can reveal unexpected differences in susceptibility patterns of related species within a genus (12, 24). In the present study, a considerable number of isolates representing both P. anaerobius and P. stomatis were available for testing. In general, the drugs demonstrated excellent activities, especially against P. stomatis. However, 13% of the P. anaerobius isolates were resistant to one or more drugs. The combination of amoxicillin-clavulanic acid has been reported to be less effective against P. anaerobius than other GPAC (4, 17). Indeed, 10% of the present P. anaerobius isolates showed resistance to amoxicillin-clavulanic acid. Since no β-lactamase production was detected, a resistance pattern for amoxicillin-clavulanic acid was expected to be similar to that for amoxicillin alone. A novel parenteral carbapenem antibiotic, ertapenem, is reported to be active against most anaerobic species, including GPAC (14), and this was the case among the present Peptostreptococcus isolates. Moxifloxacin belongs to fluoroquinolones that have shown broad-spectrum activity also against anaerobic organisms, including “peptostreptococci,” the MICs usually being ≤2.0 μg/ml (1). However, in the present study, one P. anaerobius isolate from a surgical wound specimen demonstrated repeatedly a resistant MIC (16 μg/ml) to moxifloxacin. Metronidazole proved to have an excellent activity against both Peptostreptococcus species. Of concern may be that nim genes, which encode nitroimidazole resistance, are frequently present in GPAC, including P. anaerobius (26).

In conclusion, both Peptostreptococcus species are involved in infections, including bacteremia, and they may become increasingly resistant to antibiotics. It is important that clinicians and microbiology laboratories are aware of differences between these two species. Recently, an easy-to-use flow chart for the identification of GPAC has been created for clinical microbiology laboratories (25). If the species identification is not considered, this may result in poor therapy outcomes, especially when infections with the involvement of P. anaerobius sensu stricto are being treated.