JOURNAL OF CLINICAL MICROBIOLOGY, May 1994, p. 1302-13070095-1 137/94/$04.00+0Copyright (© 1994, American Society for Microbiology

Identification of Five Peptostreptococcus Species IsolatedPredominantly from the Female Genital Tract

by Using the Rapid ID32A SystemJAMES NG,' LAI-KING NG,2 ANTHONY W. CHOW,3 AND JO-ANNE R. DILLONI 2*

Department of Biology, University of Ottawa, and National Laboratory for Sexually Transmitted Diseases,Laboratory Centre for Disease Control, Ottawa,2 and Division of Infectious Diseases,

Department of Medicine, University of British Columbia, Vancouver,3 Canada

Received 15 October 1993/Returned for modification 23 November 1993/Accepted 4 February 1994

The rapid ID32A kit (bioMerieux Vitek, Inc., Hazelwood, Mo.) was evaluated for its ability to identifyPeptostreptococcus species compared with conventional biochemical tests and gas-liquid chromatography(Virginia Polytechnic Institute), the current "gold standard" method. A total of 5 Peptostreptococcus AmericanType Culture Collection strains and 95 clinical isolates comprising Peptostreptococcus anaerobius, P. asaccha-rolyticus, P. magnus, P. micros, and P. prevotii isolates were included for analysis. Overall, the sensitivity andspecificity of the rapid ID32A kit in the identification for five Peptostreptococcus species compared with theVirginia Polytechnic Institute method were 93 and 80%, respectively. All P. anaerobius (n = 20) and P.asaccharolyticus (n = 25) isolates were identified with 100% sensitivity and 100% specificity. For theidentification of P. magnus (n = 24) and P. micros (n = 19), the rapid ID32A kit was 100% sensitive for bothspecies; the specificity for P. magnus was 95.8% and that for P. micros was 57.9%. The sensitivity and specificityof the rapid ID32A kit for identification ofP. prevotii (n = 12) were poor (41.7 and 8.3%, respectively). The rapidID32A kit is a useful method for the rapid differentiation of P. anaerobius and P. asaccharolyticus from otherPeptostreptococcus spp. Conventional methods should be used to identify to the species level isolates of P.magnus, P. micros, and P. prevotii.

Peptostreptococcus spp. are part of the endogenous humanflora. Although isolates of this genus are considered opportu-nistic or synergistic pathogens, they have been isolated frommixed anaerobic infections or in pure cultures from soft tissueinfections and have been implicated in female genital tractinfections such as pelvic inflammatory disease and bacterialvaginosis (1, 3, 9, 10, 13, 26).

Part of the problem in ascertaining the pathogenic potentialof Peptostreptococcus spp. concerns the identification andtaxonomic classification of clinical isolates (2, 19). The ac-

cepted techniques for grouping anaerobic bacteria include thegas-liquid chromatography (GLC) and biochemical tests usedin the system of the Virginia Polytechnic Institute (VPI) (12,29). However, GLC is expensive, time-consuming, and ofteninaccessible to the routine clinical microbiology laboratory(15). Furthermore, because most Peptostreptococcus spp. areasaccharolytic (7), species characterization is based on nega-tive results (20, 22).

Alternate taxonomic markers have been used in the classi-fication of Peptostreptococcus spp., including DNA-DNA andDNA-rRNA homology (14), guanine-plus-cytosine content ofDNA (7), murein type of the cell wall (32), GLC analysis ofmajor cellular fatty acids and volatile fatty acids (17), whole-cell protein electrophoresis (30), high-pressure liquid chro-matographic analysis of amino acid utilization and productionof metabolites in a defined chemical medium (11), and sero-logical analysis of cell surface components (27, 28). Although

* Corresponding author. Mailing address: Department of Microbi-ology and Immunology, Faculty of Medicine, University of Ottawa, 451Smyth Rd., Ottawa, Ontario Canada K1H 8M5. Phone: (613) 787-6488. Fax: (613) 738-5379.

these strategies have provided additional information in theclassification of Peptostreptococcus spp., they have also pro-duced conflicting results (24).

Several commercial kits, based on detection of preformedenzymes, are available for the identification of anaerobicbacteria (15). The rapid ID32A kit (bioMerieux Vitek, Inc.,Hazelwood, Mo.) tests for saccharolytic and proteolytic en-

zymes by using chromogenic substrates (22). Previously, therewere two prototypes of the rapid ID32A kit used to identifyclinical strains of gram-positive anaerobic cocci (GPAC) (22-24). The prototype API system was the original form of the kit(24) while the ATB 32A system, which differed from theprototype API kit by excluding four biochemical reactions,was the first commercially available form (15, 22, 31). Therapid ID32A kit differs from the ATB 32A system only bycontaining a code book but no color chart for reading end-points and allows the option for automatic reading by com-

puter programs. One previous study showed that there was

complete agreement in the identification of 30 type andreference strains of GPAC including 8 Peptostreptococcusspecies (n = 23) with the prototype rapid ID32A kits and theconventional VPI method, and in another study, there was

>99% agreement for 40 reference and 256 clinical isolates ofGPAC, including 9 species of Peptostreptococcus (n = 272) (22,23, 29). The ATB 32A kit was found to be useful in differen-tiating strains within species in addition to identification at thespecies level (22).

In this report, we evaluate the rapid ID32A kit comparedwith the conventional VPI method for the identification of 5reference and 95 clinical isolates from five Peptostreptococcusspecies. Of the 95 clinical isolates, 68 were from the femalegenital tract.

1302

Vol. 32, No. 5

on January 13, 2019 by guesthttp://jcm

.asm.org/

Dow

nloaded from

IDENTIFICATION OF FIVE PEPTOSTREPTOCOCCUS SPECIES

TABLE 1. Origin of isolation and source of 100 Peptostreptococcus isolates

No. of isolates in 5 speciesOriginofisolation ~~~~~~~~~~~~~~~~~~~~~~~~~~~Supplierand/ear

P. anaerobiu-s P. asaccharolyticus P. magnus P. micros P. prevotii reference'(n = 20) (n =25) (n =24) (n =19) (n =12)

Type strain 1 1 1 1 1 ATCCBacterial vaginosis 0 0 1 0 0 ATCC"Breast abscess 0 0 0 5 0 C. EdmistonCaesarian section 1 5 1 1 1 A. Chow (33)Mucopurulent cervicitis 2 3 0 0 0 A. ChowOvarian abscess 0 0 2 0 0 T. Ezaki (8)Pelvic inflammatory disease 0 5 2 1 4 A. Chow (5)Septic abortion 11 4 5 2 0 A. Chow (6)Skin abscess 0 0 2 0 0 T. Ezaki (8)Tampon 1 1 5 4 0 A. Chow (4)Vaginal hysterectomy 1 0 2 3 0 A. ChowUnknown 3 6 3 2 6 P. Ewan (25)

aATCC, American Type Culture Collection, Rockville, Md.h Protein L-containing P. magnus isolate deposited by L. Bjorck.

MATERIALS AND METHODS prior information concerning the identities of the isolates. Therapid ID32A system (bioMerieux Vitek, Inc.) comprises 3

Bacterial isolates and growth conditions. A total of 100 empty cupules as negative controls and 29 cupules containingPeptostreptococcus isolates comprising five species were tested dehydrated enzyme substrates for six tests used in Bergey's(Table 1), including the following type strains: Peptostreptococ- manual for differentiation of Peptostreptococcus spp. (20):cus anaerobius ATCC 27337, P. asaccharolyticus ATCC 14963, urease (URE), alpha-glucosidase (otGLU), beta-glucuronidaseP. magnus ATCC 15794, P. micros ATCC 33270, and P. prevotii (IGUR), reduction of nitrates, indole production, alkalineATCC 9321. Isolates were revived either from lyophilized (2% phosphatase (PAL), and 23 other enzymes (15). An organismskim milk) cultures by growth in Robertson's cooked meat was identified by an 8-digit code with the provided rapidmedium or from storage at - 70°C in brain heart infusion with ID32A analytical profile index. Included in the product data15% (vol/vol) glycerol by plating onto Columbia blood agar base of rapid ID32A were the following Peptostreptococcus(Quelab) supplemented with 5% (vol/vol) sheep blood, vitamin species: P. anaerobius, P. asaccharolyticus, P. indolicus, P.K (1 mg/liter), and 0.1% hemin. The blood agar plates were magnus, P. micros, and P. prevotii. For most Peptostreptococcusincubated for 24 h at 37°C in an anaerobic chamber (5% C02, species, strips were inoculated in duplicate according to man-5% H2, 90% N2). Assessment of culture purity was by Gram ufacturer's instructions (15). Reactions were visually scored asstain and aerobic incubation. being positive, weakly positive, variable (very weak positive), or

VPI. Clinical Peptostreptococcus isolates from C. Edmiston, negative on the basis of colorimetric indicators. Profiles of P.T. Ezaki, and P. Ewan upon receipt were reconfirmed for prevotii isolates were repeated up to six times because certainidentity by the VPI method (C. Cooper, National Laboratory preformed reactions were inconsistent because of very weakof Bacteriology, Laboratory Centre for Disease Control, Ot- reactions for some tests. The final identification was recordedtawa, Canada) (12, 29). Female genital tract clinical Peptostrep- on the basis of the majority result from the profiles.tococcus isolates from the University of British Columbia(Vancouver, British Columbia, Canada) were originally iden- RESULTStified by the VPI method. Reconfirmation of the originalidentification of the Peptostreptococcus isolates was performed Sensitivity and specificity of rapid ID32A. The overallby S. Chu (Ontario Provincial Health Laboratories, Ottawa, sensitivity and specificity in the identification of the fiveCanada) by analysis of volatile fatty acids from cultures Peptostreptococcus species by the rapid ID32A kit comparedinoculated in Robertson's cooked meat broth for 48 h by GLC with the VPI method as the "gold standard" were 93 and 80%,with a fused-silica capillary column (12, 21, 29). respectively (Table 2). All P. anaerobius (n = 20) and P.

Rapid ID32A. The rapid ID32A system was used without asaccharolyticus (n = 25) isolates were identified, compared

TABLE 2. Sensitivity and specificity of rapid ID32A for identifying five Peptostreptococcus species compared withthe gold standard VPI conventional method

Species' n P. anaerobius P. asaccharolyticus P. magnus P. micros P. prevotii No lDb % Sensitivity' % Specificity"

P. anaerobius 20 20 100.0 100.0P. asaccharolyticus 25 25 100.0 100.0P. magnus 24 23 1 100.0 95.8P. micros 19 8 11 100.0 57.9P. prevotii 12 4 1 7 41.7 8.3Total Peptostreptococcus isolates 100 93.0 80.0

"Identification obtained by the VPI conventional method including GLC.h No ID, no identification.Sensitivity: (no. of Peptostreptococcus isolates with positive test/no. of Peptostreptococcus isolates tested on the basis of VPI) x 100%.

"Specificity: (no. of Peptostreptococcus isolates correctly identified/no. of Peptostreptococcus isolates tested) x 100%.

VOL. 327 1994 1303

on January 13, 2019 by guesthttp://jcm

.asm.org/

Dow

nloaded from

1304 NG ET AL.

TABLE 3. Preformed enzyme profile of 100 Peptostreptococcus strains and isolates

Peptosteptococcus species No. with preformed enzymes"':and strains" oxGLU IND PAL ARG PRO LG PHE LEU PYR TYR ALA GLY HIS GG SER

P. anaerobiusATCC 27337 (n = 1) 1 0 t 0 1 00 0 0 0 0 0Female genital tract (n = 16) 16 t) 0 1 16 0 0 1 0 0 1 1 0 0 0Unknown clinical origin (n = 3) 3 0 0 0 3 0 0 0 0 0 0 0 0 0 0

P. asaccharolyticlusATCC 14963 (n = 1) 1 0 1 0 0 0 1 0 0 0 0 1 0 0Female genital tract (n = 18) 0 18 0 18 0 0 0 12 0 16 2 7 18 6 2Unknown clinical origin (n = 6) 0 6 0 6 0 0 0 5 0 6 0 4 6 0 0

P. magnusATCC 15794 (n = 1) 0 0 1 1 0 1 0 1 1 1 1 1 1 0 1Female genital tract (n = 18) 0 0 13 18 1 14 1 18 18 9 15 17 10 1 9Skin abscess (n = 2) 0 0 2 2 0 2 0 2 2 0 2 2 0 0 0Unknown cliniical origin (n = 3) t) 0 3 3 0 3 1 3 3 1 3 3 3 0 3

P. microsATCC 33270 (n =1) 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1Breast abscess (n =5) 0 0 5 5 5 5 4 5 5 5 5 4 5 4 5Female genital tract (n = 11) 0 0 4 11 3 9 3 11 10 6 9 10 9 3 10Unknown clinical origin (n = 2) 0 0 2 2 2 1 2 2 2 2 2 1 2 2 2

P. prevotii"ATCC 9321 (n = 1) 1 0 0 1 0 0 1 1 1 1 1 1 1 0 1Female genital tract (n = 5) 0 0 3 5 0 3 0 5 3 3 2 4 3 2 3Unknown clinical origin (n = 6) 0 0 1 6 1 3 1 4 2 2 5 6 6 1 5

Identified by conventional VPI method."Preformed enzymes of rapid ID32A: aGAL, alpha-galactosidase; IGP, ,B-galactosidase-6-phosphate; cxARA, alpha-arabinosidase; PNAG, beta-N-acetyl-

glucosaminidase; MNE, mannose fermentation; RAF, raffinose fermentation; GDC, glutamic acid decarboxylase: sxFUC, alpha-fucosidase; NIT, nitrate reduction; IND,indole production; ARG, ArgA; PRO, ProA; LG, leucyl glycine arylamidase; PHE, PheA; LEU, LeuA; PYR, PyrA; TYR, TyrA; ALA, AlaA; GLY, GlyA; HIS, HisA;GG, GGA; and SER, SerA.

' All of the other 14 preformed enzyme reactions from rapid ID32A were negative except for P. prevolii isolates.d For 12 P. prevotii isolates, 4 were URE positive, 3 were ADH positive, 1 was xGAL positive, I was ,GAL positive, 5 were siGLU positive, 1 was PGLU positive,

4 were 3GUR positive, 1 was nitrate reductase positive, and 4 were PAL positive.

with the conventional method, with 100% sensitivity and 100%specificity (Table 2). With P. magnus isolates (n = 24), thespecificity was 95.8%, with only one isolate being misidentifiedas P. micros. The rapid ID32A kit was 57.9% specific for theidentification of P. micros isolates; 8 of 19 isolates weremisidentified as P. magnus. Only I of 12 P. prevotii isolates, thetype strain ATCC 9321, was identified correctly by the rapidID32A kit, thereby producing a very low specificity of 8.3%(Table 2). Four P. prevotii isolates were identified as P. magnus.The rest were unidentifiable; thus, the sensitivity (41.7%) andspecificity (8.3%) of the kit for the identification of P. prevotiiwere lower.A total of 16 of 20 P. anaerobius isolates (index profile

0400020000) had characteristic and similar preformed enzymeprofiles; that is, they were positive only for otGLU and prolinearylamidase (ProA) (Table 3). Of 4 of the 16 female genitaltract isolates, each one had separate activities for one of thefour aminopeptidases: arginine arylamidase (ArgA), leucinearylamidase (LeuA), alanine arylamidase (AlaA), or glycinearylamidase (GlyA) (Table 3).

All P. asaccharolyticus isolates including the type strainATCC 14963 produced indole and demonstrated ArgA andhistidine arylamidase (HisA) activity (Table 3). Only 1 of 18female genital tract isolates had a preformed enzyme profilesimilar to that of the type strain P. asaccharolyticus ATCC14963. The type strain P. asaccharolyticus ATCC 14963 hadactivity for LeuA as did 12 of 18 female genital tract isolatesand 5 of 6 clinical isolates of unknown origin. Except for the

type strain ATCC 14963 and two other female genital tractisolates, all other P. asaccharolyticus isolates had activity fortyrosine arylamidase (TyrA).

All P. magnus isolates were asaccharolytic and URE, argi-nine dihydrolase (ADH), nitrate reductase, and indole nega-tive. A total of 19 of 24 P. magnus isolates, including the typestrain, produced PAL (Table 3). All isolates were positive forArgA, LeuA, and pyroglutamic acid arylamidase (PyrA). Oneisolate from the female genital tract had activity for ProA,phenylalanine arylamidase (PheA), and glutamyl glutamic acidarylamidase (GGA). As a result, this isolate was identified as P.micros.

Like P. magnus isolates, all 19 P. micros isolates wereasaccharolytic and indole, URE, ADH, and nitrate reductasenegative. The type strain P. micros ATCC 33270, all breastabscess isolates, all clinical isolates of unknown origin, and 4 of11 female genital tract isolates had strong activity for PAL(Table 3). P. micros ATCC 33270 had activity for all 12aminopeptidases. Similar to P. magnus, all P. micros clinicalisolates had activity for ArgA and LeuA. All P. micros isolatesfrom breast abscesses had activity for the 12 aminopeptidasesexcept that 1 isolate was GGA negative. A total of 10 of 11isolates from the female genital tract had activity for PyrA,GlyA, and serine arylamidase (SerA). A total of 9 of 11 femalegenital tract isolates had activity for leucyl glycine arylamidase(LGA), HisA, and AlaA, whereas 3 of 11 had activity for ProA,PheA, and GGA. A total of 6 of 12 female genital tract isolateshad activity for TyrA. Both isolates of unknown clinical origin

J. CLIN. MICROBIOL.

on January 13, 2019 by guesthttp://jcm

.asm.org/

Dow

nloaded from

IDENTIFICATION OF FIVE PEPTOSTREPTOCOCCUS SPECIES

had activity for all 12 aminopeptidases except that I isolate wasLGA and GlyA negative (Table 3).

Only the type strain P. prevotii ATCC 9321 was identifiedproperly by the kit. A total of 3 of the 11 P. prevotii isolates hadURE activity, 2 produced ADH, 1 produced ,B-galactosidase(,BGAL), 4 produced oGLU, 1 produced beta-glucosidase(,BGLU), 3 produced ,GUR, 1 reduced nitrates, and 4 pro-duced PAL (Tables 3 and 4). All clinical isolates had activityfor ArgA. Certain aminopeptidase reactions were present inmore strains than others: GlyA, LeuA, HisA, SerA, AlaA, andTyrA (Table 4). The four P. prevotii clinical isolates thatwere identified as P. magnus by the kit had the followingprofile indices: 00000416405, 0000016407, 0000056700, and0000056707.

:n _E

5C

0 ,

C0

DISCUSSION

Previous studies have evaluated prototype rapid ID32A kitsfor the identification of a variety of obligate anaerobes,including GPAC (Peptostreptococcus spp.) and pigmented Bac-teroides spp. (15, 22-24, 31). In this study, the rapid ID32A kitwas used to identify five Peptostreptococcus species isolatedpredominantly from the female genital tract. Our resultsshowed an 80% agreement with conventional VPI methods,which is lower than the 100% reported in other studies(22-24). This is due to the identification of 8 of 11 P. microsisolates from the female genital tract as P. magnus and to thefailure to identify at the species level 11 of 12 P. prevotiiisolates. P. prevotii has been shown to be heterogeneous on thebasis of preformed enzyme profiles, and P. prevotii taxonomyrequires further clarification (7, 22-24). Recently, three new

species, P. vaginalis, P. lacrimalis, and P. lactolyticus, have beenidentified from five independent DNA similarity groups thathad been tentatively identified as P. prevotii (18). Only P.vaginalis was isolated from the female genital tract (18). Li etal. (18) used the preformed enzyme kits API ZYM andAN-IDENT for identification purposes. The preformed en-

zyme profile has an advantage over the conventional VPImethod because it provides an 8-digit profile specific for a

certain isolate (22). In general, most individual tests were

highly reproducible upon retesting, as long as the inoculumdensity was similar, except for P. prevotii isolates for which theproteolytic enzyme results varied upon retesting because many

were very weak. For example, the type strain P. prevotii ATCC9321 was variable for LeuA compared with seven clinicalisolates which had strong reactions. This could be a reflectionon the particular enzyme reaction, the growth conditions, or

the organism itself. TyrA, AlaA, PheA, and LGA tests were

the hardest to interpret.The preformed enzyme profile for P. anaerobius is consistent

with previous reports (22-24). Production of proline amino-peptidase and that of cxGLU are good differential characteris-tics of P. anaerobius, and the rapid ID32A kit reliably detectedthese two enzymes in all the P. anaerobius isolates tested (7,22-24).

Like its prototype kits, rapid ID32A was 100% successful inidentifying P. asaccharolyticus isolates. All P. asaccharolyticusisolates were indole positive, mannose negative, URE nega-

tive, PAL negative, otGLU negative, and ,BGUR negative anddid not reduce nitrate-typical characteristics of this species(20).The differentiation of P. magnus and P. micros has been

classically based on relying on cell size alone, GLC, andnegative carbohydrate utilization results (12, 29). The produc-tion of PAL was used recently to distinguish P. micros (+) fromP. magnus (-) (12), and this study showed that 19 of 24 P.

-1 -, -0 - n > x0 x x x >, >

0- O=C) ,,A ,4£ -

I+ + + I+

+ +

Ii +

+ + +

+++

+ ++ II +

I I I I I

I I I I I I

I I I I I

I I I I I

II

+ + +

+ <+ + <' + + +I

I <

+ 1< + <+

+ + I+

+ + + +

I + I++++++ +

+ <

+ + + + + S+

1-r

m

.

oD

0DQ

CD

0D

C)

r

0

c)0

0

cC

C

0

c

z

z

r,

m

CU

r

c

-o

c:-V

m

1305VOL. 32, 1994

on January 13, 2019 by guesthttp://jcm

.asm.org/

Dow

nloaded from

1306 NG ET AL.

magnus isolates had PAL activity. By this criterion for confir-mation, these isolates should be classified as P. micros. Previ-ous studies of prototype kits showed that PAL activity wasvariable or minimal for P. magnus isolates and did not discrim-inate well between the two species (22-24). Therefore, eitherthe PAL test from this strip is hard to interpret or this test isnot definitive for species level identification.Our preformed enzyme profiles of P. magnus isolates agree

well with previous prototype preformed enzyme profiles in thatthe isolates have a wide variety of aminopeptidase activityexcept that ProA, PheA, and GGA are absent (7, 22-24).There were no problems in the proper identification of the

type strain P. micros ATCC 33270, the five breast abscessisolates, and the two clinical isolates of unknown origin on thebasis of the preformed enzyme profile. All produced a strongPAL reaction and had strong proteolytic activity includingProA, PheA, and GGA. A total of 8 of the 11 P. micros isolatesfrom the female genital tract were identified as P. magnusbecause they had no ProA, PheA, or GGA activity.The identification and differentiation of P. magnus from P.

micros have been controversial (16, 20, 22-24). Identificationresults of P. magnus and P. micros from female genital tractisolates with the rapid ID32A kit support this controversy.Protein electrophoresis has been used to differentiate the twospecies but is unrealistic for routine identification purposes(30).The preformed enzyme profiles for type strain P. prevotii

ATCC 9321 and the clinical isolates confirm that this is a veryheterogeneous group (7, 18, 22-25). By comparison of com-mon preformed enzyme profiles in the three biochemical kits(API ZYM, AN-IDENT, and rapid ID32A), none of theclinical P. prevotii isolates used in this study could be identifiedas belonging to any of the five DNA similarity groups includingthe three new species P. vaginalis, P. lacnimalis, and P. lacto-lyticus. This could be a reflection of the biochemical kits usedor the difference in the inoculum density or a reflection of thewide heterogeneity of isolates tentatively identified as P.prevotii. The latter reason is confirmed by our results in thatonly the type strain ATCC 9321 could be identified as P.prevotii on the basis of the 8-digit analytical profile. Theseresults are in agreement with previous biochemical studies onclinical isolates identified as P. prevotii (7, 22-24). We agreethat further taxonomic work needs to be done on strainsphenotypically similar to P. prevotii. Until this is done, pre-formed enzyme systems will provide a species profile for suchclinical isolates, but they cannot provide a species identifica-tion for most strains in this group.

Unlike the rapid ID32A kit, color charts were provided forthe original prototype API kit and the ATB 32A prototype.Color charts would greatly help in variable reactions by makingit easier to read endpoints. In future modifications and evalu-ation of the kit, P. productus, P. tetradius, and the newlydesignated P. vaginalis, P. lacnimalis, and P. lactolyticus speciesshould be included in the data base (18).

ACKNOWLEDGMENTS

This work was funded by the Canadian Bacterial Diseases Network.We thank Nancy Bigelow and Maria Carbello for the maintenance

of the Peptostreptococcus isolates; C. Edmiston (Medical College ofWisconsin, Milwaukee); C. Cooper (National Laboratory of Bacteri-ology, Laboratory Centre for Disease Control, Ottawa, Canada) forreconfirming identities of some isolates by the VPI method; and S. Chufor reconfirming the identities of Peptostreptococcus isolates by GLC.

REFERENCES1. Bourgault, A.-M., J. E. Rosenblatt, and R. H. Fitzgerald. 1980.

Peptococcus magnus: a significant human pathogen. Ann. Intern.Med. 93:244-248.

2. Brook, I. 1987. Bacterial synergy in pelvic inflammatory disease.Arch. Gynecol. Obstet. 241:133-143.

3. Brook, I., and R. I. Walker. 1984. Pathogenicity of anaerobicgram-positive cocci. Infect. Immun. 45:320-324.

4. Chow, A. W., and K. H. Bartlett. 1989. Sequential assessment ofvaginal microflora in healthy women randomly assigned to tamponor napkin use. Rev. Infect. Dis. 11:S68-S74.

5. Chow, A. W., K. L. Malkasian, J. R. Marshall, and L. B. Guze.1975. The bacteriology of acute pelvic inflammatory disease. Am.J. Obstet. Gynecol. 122:876-879.

6. Chow, A. W., J. R. Marshall, and L. B. Guze. 1977. A double-blindcomparison of clindamycin with penicillin plus chloramphenicol intreatment of septic abortion. J. Infect. Dis. 135:S35-S39.

7. Ezaki, T., and E. Yabuuchi. 1985. Oligopeptidase activity ofGram-positive anaerobic cocci used for rapid identification. J.Gen. Appl. Microbiol. 31:255-265.

8. Ezaki, T., N. Yamamoto, K. Ninomiya, S. Suzuki, and E. Yabuu-chi. 1983. Transfer of Peptococcus indolicus, Peptococcus asaccha-rolyticus, Peptococcus prevotii, and Peptococcus magnus to thegenus Peptostreptococcus and proposal of Peptostreptococcus tetra-dius sp. nov. Int. J. Syst. Bacteriol. 33:683-698.

9. Finegold, S. M. 1977. Anaerobic bacteria in human disease.Academic Press, New York.

10. Garber, G. E., and A. W. Chow. 1989. Female genital tractinfections, p. 429-454. In S. M. Finegold and W. L. George (ed.),Anaerobic infections in humans. Academic Press, San Diego,Calif.

11. Harpold, D. J., and B. L. Wasilauskas. 1987. Rapid identificationof obligately anaerobic gram-positive cocci using high-perfor-mance liquid chromatography. J. Clin. Microbiol. 25:996-1001.

12. Holdeman, L. V., E. P. Cato, and W. E. C. Moore. 1991. Anaerobiclaboratory manual, 4th ed. Anaerobic Laboratory, Virginia Poly-technic Institute and State University, Blacksburg.

13. Holst, E., B. Wathne, B. Hovelius, and P.-A. Mardh. 1987.Bacterial vaginosis: microbiological and clinical findings. Eur. J.Clin. Microbiol. 6:536-541.

14. Huss, V. A. R., H. Festl, and K. H. Schleifer. 1984. Nucleic acidhybridization studies and deoxyribonucleic acid base compositionof anaerobic, gram-positive cocci. Int. J. Syst. Bacteriol. 34:95-101.

15. Kitch, T. T., and P. C. Appelbaum. 1989. Accuracy and reproduc-ibility of the 4-hour ATB 32A method for anaerobe identification.J. Clin. Microbiol. 27:2509-2513.

16. Krepel, C. J., C. M. Gohr, A. P. Walker, S. G. Farmer, and C. E.Edmiston. 1992. Enzymatically active Peptostreptococcus magnus:association with site of infection. J. Clin. Microbiol. 30:2330-2334.

17. Lambert, M. A., and A. Y. Armfield. 1979. Differentiation ofPeptococcus and Peptostreptococcus by gas-liquid chromatographyof cellular fatty acids and metabolic products. J. Clin. Microbiol.10:464-476.

18. Li, N., Y. Hashimoto, S. Adnan, H. Miura, H. Yamamoto, and T.Ezaki. 1992. Three new species of the genus Peptostreptococcusisolated from humans: Peptostreptococcus vaginalis sp. nov., Pep-tostreptococcus lacrimalis sp. nov., and Peptostreptococcus lactolyti-cus sp. nov. Int. J. Syst. Bacteriol. 42:602-605.

19. Mardh, P.-A., and L. Westrom. 1984. Factors that might influencethe outcome of studies on the aetiology and epidemiology of acutepelvic inflammatory disease. Scand. J. Gastroenterol. 19:73-86.

20. Moore, L. V. H., J. L. Johnson, and W. E. C. Moore. 1986. GenusPeptostreptococcus Kluyver and van Niel 1936, 402AL, p. 1083-1092. In P. H. A. Sneath, N. S. Mair, M. E. Sharpe, and J. G. Holt(ed.), Bergey's manual of systematic bacteriology, vol. 2. TheWilliams & Wilkins Co., Baltimore.

21. Moss, C. W., and 0. L. Nunez-Montiel. 1982. Analysis of short-chain acids from bacteria by gas-liquid chromatography with afused-silica capillary column. J. Clin. Microbiol. 15:308-311.

22. Murdoch, D. A., and I. J. Mitchelmore. 1991. The laboratoryidentification of gram-positive anaerobic cocci. J. Med. Microbiol.34:295-308.

23. Murdoch, D. A., I. J. Mitchelmore, R. A. Nash, J. M. Hardie, and

J. CLIN. MICROBIOL.

on January 13, 2019 by guesthttp://jcm

.asm.org/

Dow

nloaded from

IDENTIFICATION OF FIVE PEPTOSTREPTOCOCCUS SPECIES 1307

S. Tabaqchali. 1988. Preformed enzyme profiles of referencestrains of gram-positive anaerobic cocci. J. Med. Microbiol. 27:65-70.

24. Murdoch, D. A., I. J. Mitchelmore, and S. Tabaqchali. 1988.Identification of gram-positive anaerobic cocci by use of systemsfor detecting pre-formed enzymes. J. Med. Microbiol. 25:289-293.

25. Ng, L.-K., and J.-A. Dillon. 1991. Molecular fingerprinting ofisolates of the genus Peptostreptococcus using rRNA genes fromEscherichia coli and P. anaerobius. J. Gen. Microbiol. 137:1323-1331.

26. Rice, P. A., and J. Schachter. 1991. Pathogenesis of pelvic inflam-matory disease. What are the questions? JAMA 266:2587-2593.

27. Smith, G. L. F., C. G. Cumming, and P. W. Ross. 1986. Immuno-chemical characterization of the cell surface carbohydrate antigensof Peptostreptococcus anaerobius. J. Gen. Microbiol. 132:525-530.

28. Smith, G. L. F., C. G. Cumming, and P. W. Ross. 1986. Analysis ofEDTA-soluble cell surface components of Gram-positive anaero-bic cocci. J. Gen. Microbiol. 132:1591-1597.

29. Sutter, V. L., D. M. Citron, M. A. C. Edelstein, and S. M. Finegold.1985. Wadsworth anaerobic bacteriology manual, 4th ed. StarPublishing Co., Belmont, Calif.

30. Taylor, E. A., P. J. H. Jackman, and I. Phillips. 1991. Thedifferentiation of asaccharolytic anaerobic gram-positive cocci byprotein electrophoresis. J. Med. Microbiol. 34:339-348.

31. van Winkelhof, A. J., M. Clement, and J. de Graaff. 1988. Rapidcharacterization of oral and nonoral pigmented Bacteroides spe-cies with the ATB anaerobes ID system. J. Clin. Microbiol.26:1063-1065.

32. Weiss, N. 1981. Cell wall structure of anaerobic cocci. Rev. Inst.Pasteur Lyon 14:53-59.

33. Williams, C. M., D. M. Okada, J. R. Marshall, and A. W. Chow.1987. Clinical and microbiologic risk evaluation for post-cesareansection endometritis by multivariate discriminant analysis: role ofintraoperative mycoplasmas, aerobes, and anaerobes. Am. J. Ob-stet. Gynecol. 156:967-974.

VOL. 32, 1994

on January 13, 2019 by guesthttp://jcm

.asm.org/

Dow

nloaded from