0095-1137/92/123112-05$02.00/0
Copyright©1992, AmericanSocietyforMicrobiology
Evaluation of
the
RapID ANA
II
and
API
ZYM
Systems
for
Identification
of Actinomyces
Species
from
Clinical
Specimens
MINNA A.BRANDER* ANDHANNELE R. JOUSIMIES-SOMER2t
Department inLappeenranta, National Public HealthInstitute, Kahilanniementie2, 53130Lappeenranta,
and Anaerobe
Reference Laboratory, National
Public HealthInstitute,
Helsinki,
2 FinlandReceived18May1992/Accepted 31 August 1992
Classification and identification offermentative actinomycetesarelabor-intensive andproblematic. In this study, we evaluated the
applicability
and reliability of the RapID ANA II system (Innovative DiagnosticSystems, Inc., Atlanta,Ga.) and the discriminatory value of the API ZYM system (Societes Analytab Products Inc., LaBalmeLesGrottes, France) in theidentification ofActinomyces-like bacteria by using conventional methods as a reference. Eighty-five strains, including 71 isolates from mixed anaerobic infections and 14 reference strains, were tested. The RapID ANA IIsystem
correctly
identified all Actinomycesodontolyticus strainsand 65% of Acfinomyces israelii strains.AllArcanobacterium haemolyticum strains were misidentified asActinomyces pyogenes. The most common isolates in the study were Actinomyces meyeri-like organisms, 84% ofwhich, however,wereaerotolerant. Theidentification of these aerotolerant strains thus remains unresolved and warrantsfurther studies. New characteristics and changes to theconventional API ZYM enzyme profiles are suggested. TheAPIZYM enzyme profiles of A.odontolyticus
and A. israelii wereverysimilar, theonly
discriminating enzyme being at-fucosidase. In differentiation between A. pyogenes and Arcanobacterium
haemolyticum, theproduction of
13-glucuronidase
by theformer and theproduction of acid phosphatase by the latter aresuggested as newhelpful characteristics for useinclinical laboratories.Insummary, theRapID ANA II and APIZYM systems can be used as rapid preliminary methods in theidentification ofActinomyces species but accurateidentification requires supplementary conventionaltestsandgas-liquid chromatography. With the exception of classical actinomycosis due toActinomycesisraelii,Actinomyces species areisolated usu-ally from bacterial infections involving mixedfloras. Thus,
the clinical significance of individual species is still
con-troversial (6-8, 15, 17). Conventional testing based on
fer-mentation and biochemical reactions together with the
de-termination of metabolic end products by gas-liquid
chromatography isconsidered themost accurateand reliable
approachforthe identificationofActinomycesspecies (1, 9,
18). However, because these procedures are both
labor-intensiveandtime-consumingandrequirespecial
instrumen-tation, theyarebeyondthecapabilities of manylaboratories
and the isolation and identification of these bacteria have remainedaspecialityofafewlaboratories.
Recently, the trend inclinical anaerobic bacteriologyhas
been towards the useofcommercially available
identifica-tion kits, including those detecting constitutive preformed
enzymes. Theperformancesof thesekitsarenotdependent onactivegrowthandrequire only4 hofaerobic incubation.
Thedifferentialchartsofthesesystems alsoinclude
Actino-mycesspecies.
The purpose ofthisstudywas toevaluate the applicabil-ity, reliability, and accuracy of the RapID ANA II system and thediscriminatory value of the API ZYM system in the
identification ofActinomyces species isolated from clinical
specimens by usingconventional methods as areference.
* Correspondingauthor.
tPresentaddress: Microbial DiseaseResearchLaboratory, 691/ 151J,VAWadsworthMedicalCenter, Los Angeles, CA 90073.
MATERIALSANDMETHODS
Bacterial strains. A total of 85 Actinomyces and
Actino-myces-like bacterial strainswereexamined. These included
14 reference strains (13 from the American Type Culture
Collection [ATCC] and 1 from the National Collection of
T7ype
Cultures, Central Public Health Laboratory,Colin-dale, London, England): A. israelii serotypes 1 (ATCC
10049; NCTC 10236) and 2 (Actinomyces gerencseriae;
ATCC 23860), Actinomyces odontolyticus (ATCC 17929),
Actinomycesnaeslundii (ATCC 12104),Actinomyces visco-susserotype 1 (ATCC 15987),Actinomycesmeyeri (ATCC
35568),Actinomyces georgiae (ATCC 49285),Actinomyces
humiferus
(ATCC 25174),Actinomycesbovis(ATCC 13683),Actinomyces hordeovulneris (ATCC 35275), Actinomyces sp. serotype WVA 963 (ATCC 49338),Actinomyces pyo-genes (ATCC 19411), andArcanobacterium
haemolyticum
(ATCC 9345).Here,Actinomyces-like bacteriaweredefined
as gram-positive, non-spore-forming rods, most of which
werebranchingand catalasenegative and whichingas-liquid
chromatographyproducedsuccinicacidandwiththeRapID
ANA II system were identified as Actinomyces species.
All organisms except the reference strains were fresh
clinical isolatesobtainedfrompatientspecimensin 1990 and cultured and isolated at the Department of Lappeenranta,
National Public Health Institute, Lappeenranta, Finland,
mainlyaspartof mixed anaerobic floras. Afterisolation,the
strainswereexaminedbyconventionalVirginia Polytechnic
Institute reference methods (9) and the RapID ANA II system. The strains were kept at -70°C in skim milk,
thawed, and subcultured before being tested by the API
ZYMsystem. The conventional methods included aerotol-erancetestingand determinationofGramreaction, cellular
morphology,and colonial characteristicsaswell as
patterns
3112
on April 12, 2020 by guest
http://jcm.asm.org/
TABLE 1. API ZYM reactions of the study strains identified by the RapID ANA II system as A. meyeri, A.odontolyticus,and A. israelii
No.of No. of strains positive by API ZYM in
microtubea:
strains
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20A. meyerib 31 2 4 1 0 28 0 0 0 0 2 0 0 2 1 19 0 2 0 0
A. odontolyticusc 18 0 12 8 0 18 4 0 0 0 6 1 5 8 0 15 10 5 3 15
A. israeliid 14 0 14 8 0 14 5 0 0 0 1 0 3 6 1 5 3 0 0 0
a Reactionmicrotubes assayed for thefollowingenzymes: 2,alkalinephosphatase; 3, esterase; 4,esterase-lipase; 5, lipase; 6, leucine arylamidase; 7, valine arylamidase; 8, cystine arylamidase; 9, trypsin; 10, chymotrypsin; 11, acid phosphatase; 12, naphthol-AS-BI-phosphohydrolase; 13, a-galactosidase; 14,
3-galactosidase; 15, 3-glucuronidase; 16, a-glucosidase; 17, 3-glucosidase; 18, N-acetyl-,B-glucosaminidase; 19, a-mannosidase; 20,a-fucosidase.
Onlytwostrainswereobligate anaerobes.
CAllstrainskeyed outidenticallyby the conventional methods.
dOnlyeight strains keyed outidenticallyby the conventionalmethods.
of fermentation ofcarbohydrates andproductionofindole,
nitrate reductase, catalase, arginine dehydrolase, and
ure-ase, supplemented by the metabolic end product analysis
with gas-liquid chromatography (20). Some of the strains were also tested for gelatin liquefaction, peptonization of litmus milk, and reverse CAMP reaction.
RapIDANA Hsystem. TheRapIDANA IIsystemconsists
of 10 individual test wells (8 of which are bifunctional)
containing dehydrated reactants designed to test 18
bio-chemical reactions. Substrates included in the 10 wells of the test panel are urea,
p-nitrophenyl-3,D-disaccharide,
p-nitrophenyl-a,L-arabinoside,
o-nitrophenyl-,B-D-galacto-side,p-nitrophenyl-a,D-glucoside,
p-nitrophenyl-3,D-gluco-side,p-nitrophenyl-a,D-galactoside,
p-nitrophenyl-a,L-fuco-side,
p-nitrophenyl-N-acetyl-3,D-glucosaminide,
p-nitro-phenylphosphate, leucine-,-naphthylamide,
glycine-,-naphthylamide,
proline-13-naphthylamide,
phenylalanine-,-naphthylamide,
arginine-p-naphthylamide,
serine-3-naph-thylamide, pyrrolidonyl-,B-naphthylamide, and tryptophane.
Purecultures of theorganismswereinoculated on chocolate
agar(Oxoid).Plates wereincubatedin ananaerobic
environ-ment at35 to37°C for 48 h. The strainswere
suspended
inRapID ANA II inoculation fluid, and the inoculum was
adjustedto aturbidity ofMcFarland standard no. 3. The trays
wereinoculated accordingtothe manufacturer'sinstructions
andincubatedaerobicallyat35to37°Cfor 4 h.After
incuba-tion,reactions in the test wells wereinterpretedandrecorded
before and after the addition of reagents. Asix-digit
micro-code was generated, and identification was determined by
usingtheRapIDANA II codecompendium(10).The
Actino-mycesspecies includedinthe codecompendiumareA.bovis,
A. israelii,A. meyeri,A. naeslundii, A. odontolyticus,A.
pyogenes, andA. viscosus. The referencestrainswere exam-ined twice.
API ZYMsystem. The API ZYMgalleryiscomposedof 20
microtubes,which contain 19 substrates in buffer. The first
microtubeserves as acontrol, and the enzymes determined
in the other tubes are listed in Table 1, footnote a. Pure cultures oforganismswere obtained and incubated as with
the RapIDANA IIsystem. The strainswere suspended in
sterile distilledwaterand adjustedtoaturbidity of McFar-land standard no. 5 or greater. The trays were inoculated
accordingtothe manufacturer's instructions and incubated
aerobicallyat35 to37°Cfor 4 h. Afterincubation, reactions
in the testcupuleswereinterpretedand recorded on areport sheetafteraddition of reagents.Acolorintensityvalue from
0to5wasassignedfor each reaction accordingtothe color
chart enclosed with the kit.Onlycolorsshowinganindex of 2or morewererecorded aspositive. The API ZYMsystem
lacks an index book forspecific bacterialidentification. The reference strains were examined twice.
RESULTS
Sixty-three
of the 71clinical isolates studiedwereidenti-fiedby the RapID ANA IIsystem as A. odontolyticus (18
isolates),A. israelii (14isolates), orA. meyeni(31 isolates).
The referencestrains for each werecorrectlyidentified;A.
gerencseriae(A.israelii serotype2)wasalso identified as A.
israelii.
All ofthe 19strainsthat wereidentifiedasA.
odontolyti-cusbyRapID ANA II also wereidentifiedasA.
odontolyti-cus by the conventional methods; all were mannitol
nega-tive, 16 were raffinose and trehalose negative, and 9 were
ribose negative. Twelve different microcodes were found,
noneof which wasoutstanding.Twomicrocodes, however,
gave theidentificationprobability belongingtoA.
odontolyti-cus,Arachnia propionica (Propionibacterium
propionicus),
Lactobacillusacidophilus,orA.israelii.Asthestudy strains
produced succinic acid and nopropionic acid, the
identifi-cationasActinomyces specieswas accepted.Furthermore,
as oneof these twostrainswasa-fucosidasepositiveand the other produced red pigment, the final identification as A.
odontolyticus
instead of A. israeliiwasaccepted.Only11of the14strains identifiedasA. israeliibyRapID
ANAII wereidentified asA. israelii by conventional
meth-ods.(All strainswerea-fucosidaseand ureasenegative.)Ten
different microcodeswereobtained.
If obligately anaerobic growth (3) were used as a strict
criterion, only 2 of the 31 clinical strains identified as A.
meyeri
in the RapID ANA II system and the type strain would have beenidentifiedasA.meyeri alsobytheconven-tional methods. Even these three strains (the two clinical
isolates and the type strain) grew slightly in a 5% CO2
atmosphere.Theremaining29clinical isolateswere
aerotol-erant; however, 3 ofthese grew very slowly in air. Three
different microcodes were obtained; of these, 020671 was
produced by 26 (84%) of the 31 strains. According to the
RapID ANA II system code
compendium,
the probableidentification given to microcode 020671 overlapped A.
meyeri
andPropionibacterium
granulosum. As all the strains were catalase negative and produced succinic acidbut nopropionic acid, the identification asA. meyeri was
accepted.
The API ZYMreactions of A.
odontolyticus,
A. israelii,and A.
meyeri
aresummarized in Table 1.Of thestudystrains,seven wereidentified asA.pyogenes
bytheRapIDANA IIsystem. These strains wereidentified
on April 12, 2020 by guest
http://jcm.asm.org/
TABLE 2. Reactivities of referencestrains of Actinomycesspp.in the API ZYMsystem
Colorintensity of API ZYM reaction inmicrotubea:
Reference strain
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
A. meyeri - - - - +- - - - +
A. odontolyticus - - - - +. . . +
A. israelii
ATCC10049 - - - - + --- w + - w + -
-NCTC10236 - - - - + . . . w + w + -
-Serotype2, ATCC 23860 - - - - + w . . . w + + +
A. pyogenes - - - - + . . . w w w
Arcanobacteniumhaemolyticum - - - - w + . . . .w
A. naeslundii - - - - + . . . w + ww
-A. viscosus - w + - + w . . . w + + -
-A. hordeovulneris - - w - + + . . . w + +
-A. humiferus - - - - +
A. georgiae - - - - +. . . +--+
Actinomycessp. serotypeWVA963 - - - - + w w + ww
-A. bovis - - - - + w . . . + + + -
-aFormicrotubenumberingsystem, see Table 1, footnotea. -, w, and +, rangesof colorintensity of 0 to 2, 2 to 3, and 3 to 5, respectively.
by the conventional methods as
Arcanobacterium
hae-molyticum.
All strains were reverseCAMP reaction positive(5), none liquefied gelatin, and only one strain fermented
xyloseslightly and wasabletopeptonize litmusmilk(18).In
the API ZYM system, all these strains produced at least someleucinearylamidase,
P-galactosidase,
andN-acetyl-p-glucosaminidase, all of which are enzymesproduced byA.
pyogenes accordingtoBergey's Manualof Systematic
Bac-teriology (18). In addition, all the strains produced acid
phosphatase,which hasnotbeenreportedasbeing produced
byA.pyogenes.The API ZYM reactions of the type strain of
Arcanobacterium haemolyticum wereidentical to the
reac-tions of the seven strains; only the production of
3-galac-tosidase was very weak. The type strain of A. pyogenes
produced leucine arylamidase,
P-galactosidase,
,-glucuron-idase, and a-glucosidase and a very small amount of
N-acetyl-,B-glucosaminidase butnoacidphosphatase.
There was only one catalase-positive strain among the clinical strains tested. The RapID ANA II microcode
ob-tainedfor thisorganism,whichbythe conventionalmethods
most closely resembled A. viscosus, was not listed in the
codecompendium.Unlisted microcodeswerealso obtained
for the twocatalase-positive reference strains,A. viscosus
ATCC 15987 and A. hordeovulnenis ATCC 35275. The API ZYM reactions of these two ATCC strains as well as the
otherreference strains aresummarized in Table 2.
DISCUSSION
Thetaxonomy ofActinomyces speciesiscomplicated,and classification and identification of fermentative actino-mycetesstill present problems despite the contributions of modemtechnology. In this study, twocommercially
avail-able kits, the RapID ANA II system and the API ZYM
system, were evaluated for the identification of Actinomy-ces-like bacteria and compared with the conventional
Vir-ginia Polytechnic Institute reference methods. In earlier
evaluations of the RapID ANA II system, about60%of the
Actinomycesstrains could bereliably identified to the genus
level but not always to the species level (8). In those
evaluations,however,onlyafewspecies and strains of these
bacteriawere included and they constituted only a minor fraction of the tested strains(4,8, 12-14, 16, 19).Kilian(13) described the results for 162 strains oftheActinomycetaceae
and related taxa for 20 different enzymatic activities, of
which 10 were the same as in the API ZYM system. Our
resultsconcurwiththose of Kilian.
In the present study, A. meyeri was the most common identificationgiven bytheRapID ANA II system(31 of 71 study strains). By the conventionalmethods,all these strains
wouldhave received the same identification had 26 strains
notgrownaerobically.Ifthe criterionofnoaerobicgrowth
(3) is obligatory, the identification ofthese strains remains
unresolved. Whether these aerotolerant A. meyeri-like
strains represent variants of A. meyeri or a new species
warrantsfurther studies. Recently, three of thestrains have
been investigated byLillian V. H. Moore atVirginia
Poly-technicInstitute and StateUniversity, Blacksburg, and the strains were unlike any described species that she has
studied (1Sa). It is also noteworthy that none of the 31
clinicalA. meyeri-like strainsnortheA. meyeri typestrain
produced valine arylamidase or cystine arylamidase,
al-thoughthesearelisted inBergey'smanualascharacteristic
products of A. meyeri.
All19 strains identifiedby theRapID ANA II system asA.
odontolyticuswerealso identifiedidentically bythe
conven-tionalmethods. Theproductionof a-fucosidasewas aunique
reaction for A. odontolyticus strains, being produced byno otherspeciesinthisstudy,aresult similartothoseofearlier studies (18). Onlythree of the A.
odontolyticus
strains didnotproduceanya-fucosidase.AlthoughKilian(13) reported
thatA. viscosus serotype 1produced a-fucosidaseintheAPI ZYM system,inthepresentstudy this reactionwasnegative
for the type strain. The colonies of only 37% ofA.
odon-tolyticus
strains developed redpigment. This percentage is somewhat lowerthan thosereportedearlier(2, 11,15). Thus,the demonstration of pigment formation might not be as
importantinthe identification ofthisActinomycesspeciesas
has beenthought. Thespecimensin the earlier studies have
beenmainlyof dentalorigin,butthiswas notthecasein the
presentstudy,as83%werefrom nonoralsources.However,
this difference inspecimensourcemaynotsufficetoexplain
thedivergentresults. Somebrownishred colorcouldbeseen
inmostof the ATCCstrains included in thisstudy,provided
the strainswere incubatedlong enough(5 to7 days). Also,
Johnsonetal. could demonstratetypicalred colonies inonly 58% of the serotype I strains and 43% of the serotype II strains of A.
odontolyticus (11).
on April 12, 2020 by guest
http://jcm.asm.org/
According to Bergey's manual, there are only three
en-zymes produced by A.
odontolyticus
in the API ZYM system: leucine arylamidase, 1-glucosidase, and a-fucosi-dase. The A. odontolyticus strains in the present study produced more enzymes, a result concordant with that ofKilian (13). More than halfofourA.
odontolyticus
strainsproducedesteraseand a-glucosidase, andatleast27% ofthe
strains produced esterase-lipase, 3-galactosidase, and
N-acetyl-fi-glucosaminidase.TwoA. meyeri-likestrainsalso
producedsomeN-acetyl-,-glucosaminidase, an enzyme
re-portedasbeingproducedonly byA. bovis and A.pyogenes
accordingtoSchaal (18). Of the reference strains, A. bovis
did not produce that enzyme at all and A. pyogenes
pro-duced onlya verysmallamountofit. The referencestrains which produced the enzyme were A. hordeovulneris and
Arcanobacterium haemolyticum.
Wepropose addingesterase, esterase-lipase,
a-galactosi-dase, 1-galactosidase, a-glucosidase, and
N-acetyl-f3-glu-cosaminidase to the list of enzymes produced by some
strains of A. odontolyticus.
Atotal of 11 (all 3 reference strains and 8 clinical strains) of the 17strains, including the referencestrainsidentified as
A. israelii by the RapID ANA II system, were identified
correctlytothespecies level. Only three clinical strainsgave
thesamereactionsasthereference strains in the APIZYM system. Theremaining five clinical strains didnot produce
any ,-galactosidase or a- or 3-glucosidase, enzymes
pro-duced byat least 90% of A. israelii strains (18). However, thecarbohydratefermentationresults andanegative urease
reaction were consistent with the characteristics of A.
is-raelii. Most of the A. israelii strains in the study were
smooth-colony variants and were aerotolerant. Only two
study strains and ATCC strain 23860 did not grow
aerobi-cally.
Accordingtoourresultsand those of Kilian, the APIZYM enzymeprofilesof A. israelii and A. odontolyticus are very
similar, the onlydiscriminatingenzymebeinga-fucosidase.
Allsevenstrainsidentified asA.pyogenes bythe RapID
ANA II system were identified asArcanobacterium
hae-molyticumby theconventionalmethods. The cell
morphol-ogy and biochemical reactions of these two species very
muchresemble each other, anddifferentiationbetweenthem is difficult. Because the two species are nowadays
consid-ered to belong to two distinct taxa (18) and because A. pyogenesisolates areoften ofanimal origin, the
identifica-tion ofArcanobacterium haemolyticum asA. pyogenes is
misleading in terms of both the source and the disease association of the bacterium and ought to be corrected. Comparing the typestrains of thesetwo species, we found
that thehemolysisproducedbyArcanobactenium
haemolyti-cumwasmuch strongerand developedmorequickly(1-day
incubation) thanthat produced by A.pyogenes(3-day
incu-bation).Arcanobacterium haemolyticum also grew equally
well in allthreeatmospheres, butA.pyogenesgrewbetter in
anaerobic and CO2 atmospheres. Other important discrimi-native reactions were the reverse CAMP reaction, gelatin
liquefaction, and xylosefermentation (5). In theAPI ZYM
system, the most distinctive difference was between the
production ofacid phosphatase by Arcanobacterium
hae-molyticum and the production of ,3-glucuronidase by A.
pyogenes, characteristics which could beofhelptoclinical laboratories in differentiating thesetwospecies.
The reactions observed with the RapID ANA II system
weremostlyeasytointerpret, although borderline reactions occurred in the hydrolysisofnaphthylamine substrates. In the APIZYMsystem, minorerrorsininterpretingthe color
intensities were possible. Although some of the reactions
were included in both systems, the substrates were not
consistently the same, nor were the inoculum densities
equal.Therefore, directcomparison of the enzyme reactions
was not considered justified.
In thisstudy, theRapID ANA II system correctly
identi-fied to thespecies level all the A.
odontolyticus
strains and 65% oftheA.israelii
strains. However, allArcanobactenumhaemolyticum
strains were misidentified as A. pyogenes.Theenzyme reactions inthe API ZYM system aided in the
identification of the strains but did not always concur with
those given in Bergey's manual for equivalent species;
accordingly, additions totheenzyme profiles arepresented.
API ZYM profiles of some newer Actinomyces spp. not
tested before are reported here. In conclusion, both rapid
methodsareuseful aids in the identification of
Actinomyces-like bacteria, although accurate identification will often
requiresupplementary conventional methods.
ACKNOWLEDGMENTS
Wethank SeppoPaltemaa, RailiHaanpaa,and TuulaRasanenfor excellentmicrobiological assistance andLillianV. H.Moore, Vir-ginia Polytechnic Institute and State University, Blacksburg, for analyzing threerepresentatives of the aerotolerantA. meyeri-resem-bling strains. We thank P. H. Makela, director ofInfectious Dis easesDivision, National Public HealthInstitute, Helsinki, Finland, for criticalreview of the manuscript.
REFERENCES
1. Allen, S. D. 1985. Gram-positive, nonsporeforming anaerobic bacilli, p. 461-472. In E. H. Lennette, A. Balows, W. J. Hausler, Jr., and H. J. Shadomy (ed.), Manual of clinical microbiology, 4th ed. American Society for Microbiology, Washington, D.C.
2. Batty, I. 1958. Actinomyces odontolyticus, a new species of actinomycete regularly isolated from deep carious dentine. J. Pathol. Bacteriol. 75:455-459.
3. Cato, E. P., W. E. C.Moore, G.Nygaard,and L. V.Holdeman. 1984.Actinomycesmeyeri sp. nov., specific epithet rev. Int. J. Syst.Bacteriol. 34:487-489.
4. Celig, D. M., and P. C.Schreckenberger. 1991. Clinical evalua-tion oftheRapID-ANA II panel for identification of anaerobic bacteria. J.Chin. Microbiol. 29:457-462.
5. Clarmdge, J. E. 1989. The recognition and significance of Arcanobacterium haemolyticum. Clin. Microbiol. Newsl. 11:
41-45.
6. Finegold, S. M. 1989. General aspects of anaerobic infection, p. 137-153. In S. M. Finegold and W. L. George (ed.),Anaerobic infections in humans. Academic Press, Inc., San Diego, Calif. 7. George, W. L. 1989. Actinomycosis, p. 529-539. In S. M.
Finegold and W. L. George (ed.), Anaerobic infections in humans.Academic Press, Inc., San Diego, Calif.
8. Hillier, S., and B. J. Moncla. 1991. Anaerobic gram-positive nonsporeforming bacilli and cocci, p. 522-537. In A. Balows, W. J.Hausler, Jr., K. L.Herrmann,H. D.Isenberg, and H. J.
Shadomy(ed.),Manual of clinicalmicrobiology, 5th ed.
Amer-icanSociety forMicrobiology, Washington, D.C.
9. Holdeman, L.V., E. P. Cato, and W. E. C.Moore (ed.). 1977. Anaerobe laboratory manual, 4th ed. Virginia Polytechnic In-stitute and StateUniversity, Blacksburg.
10. Innovative Diagnostic Systems, Inc. 1989. RapID ANA II code compendium. Innovative Diagnostic Systems, Inc., Atlanta. 11. Johnson, J.L., L. V. H. Moore, B. Kaneko, and W. E. C. Moore.
1990.Actinomyces
geoigiae
sp. nov.,Actinomycesgerencse-riaesp. nov., designation of two genospecies of Actinomyces naeslundii, andinclusion of A. naeslundii serotypes II and III andActinomyces viscosus serotype II in A. naeslundii geno-species 2. Int. J. Syst. Bacteriol.40:273-286.
12. Karachewski, N. O., E. L. Busch, and C. L. Wells. 1985. Comparison of PRAS II, RapID ANA, and API 20A systems for
on April 12, 2020 by guest
http://jcm.asm.org/
identification of anaerobic bacteria. J. Clin. Microbiol. 21:122-126.
13. Kilian, M. 1978. Rapididentification ofActinomycetaceae and related bacteria. J. Clin. Microbiol. 8:127-133.
14. Marler, L. M., J. A. Siders, L. C. Wolters, Y.Pettigrew, B. L. Skitt,and S.D.Allen.1991. Evaluation ofthenewRapID-ANA II system for theidentification of clinical anaerobic isolates. J. Clin. Microbiol. 29:874-878.
15. Mitchell, R. G., and M. R. Crow. 1984.Actinomyces odontolyti-cus isolated from the female genital tract. J. Clin. Pathol. 37:1379-1383.
15a.Moore, L. V. H. Personal communication.
16. Murray, P. R., C. J. Weber, and A. C. Niles. 1985. Comparative evaluation ofthree identification systems for anaerobes. J. Clin. Microbiol.22:52-55.
17. Pordy, R. C. 1988. Lumpyjaw due toActinomyces meyeni: report ofthefirstcaseand review of theliterature. Mt. Sinai J. Med.55:190-193.
18. Schaal,K. P. 1986. GenusActinomyces Harz 1877, 133-', p. 1383-1418. In P. H. A.Sneath,N.S.Mair, M. E. Sharpe, and J. G.Holt (ed.), Bergey's manual of systematicbacteriology, vol. 2.Williams &Wilkins,Baltimore.
19. Summanen, P., and H. Jousimies-Somer. 1988. Comparative evaluation of RapID-ANA and API 20A for identification of anaerobic bacteria. Eur. J. Clin. Microbiol. Infect. Dis. 7:771-775.
20. Sutter,V. L., D. M. Citron, M. A. C. Edelstein, and S. M. Finegold (ed.). 1985. Wadsworth anaerobic bacteriology man-ual, 4th ed. Star Publishing Co., Belmont, Calif.