0095-1137/89/102169-06$02.00/0
Copyright © 1989,American Society forMicrobiology
Pasteurella
caballi,
a
New
Species
from Equine
Clinical Specimens
L. K. SCHLATER,l* DON J. BRENNER,2 A. G. STEIGERWALT,2 C. WAYNE
MOSS,3
MARY ANN LAMBERT,3AND R. A. PACKER4
National Veterinary ServicesLaboratories, Science and Technology, Animal and Plant Health Inspection Service, U.S. DepartmentofAgriculture, Ames, Iowa500101;Molecular Biology2 and Analytical Chemistry3 Laboratories, Centersfor
DiseaseControl, Atlanta, Georgia 30333; and Department of VeterinaryMicrobiology, College of Veterinary Medicine, Iowa State University, Ames, Iowa500114
Received13 March 1989/Accepted19June 1989
ThenamePasteurella caballi is proposedforagroupoforganismsrepresented by 29strainsisolated from
respiratory and other infections in horses. P. caballistrains are gram-negative,oxidase-positive, nonmotile,
fermentative rods with the key characteristics of the genus Pasteurella. These strains differed from other
Pasteurella species in that all were aerogenic and catalase negative, and some strains produced acid from
myo-inositol and L-rhamnose. The levels of DNA relatednessof 28 P. caballi strains withlabeledDNAfromthe
proposedtypestrainaveraged 91 and85% (hydroxyapatite methodat55and70°C).P.caballiwas13to53%
relatedtostrainsrepresenting 22 other species ofthefamily Pasteurellaceae. Theguanine-plus-cytosinecontent ofthe DNA of fourstrains was41 to 42mol%. The type strain is 83851 (=ATCC 49197).
The genusPasteurella is currently classified in the family
Pasteurellaceae, togetherwith the related genera Actinoba-cillus andHaemophilus. Members ofthe genusPasteurella
are gram-negative, oxidase-positive, nonmotile,
fermenta-tive, rod-shaped bacteria(4, 17). All Pasteurella species are
parasitic on the mucous membranes of the respiratory and
digestivetracts of mammals andbirds. Underconditionsof stress, these organisms can become invasive and play a significant roleinthepathogenesis ofavariety ofinfections
in animals, including pneumonia, sinusitis, abortion, masti-tis, andsepticemia (2).
Since 1983, the General Bacteriology Section ofthe
Na-tional Veterinary Services Laboratories has received
cul-tures of an aerogenic, gram-negative bacterium that had many characteristics ofthe genus Pasteurella but could not
beassignedtoanyestablishedspecieswithin the genus. Our
attention wasoriginallydrawn to thesestrains becausethey
were all isolated from horses, an uncommon source for
Pasteurella. Inapreviousreport,thebiochemical
character-istics of13 strains were described (21), but since no DNA
hybridization information wasavailable, itwas notpossible to properly classify them. Subsequent biochemical and
ge-netic studies on these 13 strains, as well as 16 additional strains, showed that the aerogenic equine strains were
genetically similar to each other but distinct from other
phenotypically similar members of the family
Pasteurel-laceae, thereby warranting their designation as a new
spe-cies,Pasteurella caballi.
(Portions of this study were done by L.K.S. in partial
fulfillment of the requirements for the Master of Science
degree from the Department of Veterinary Microbiology,
College
of Veterinary Medicine, Iowa State University,Ames).
MATERIALS ANDMETHODS
Bacterialstrains.The 29 P. caballi strains studied and their
sources are given in Table 1. Strains
representative
of species within the family Pasteurellaceae used in the DNA*Corresponding author.
hybridizationstudieswerefrom thesourceslisted in Table 2. Stock cultures of P. caballi and reference strains were stored
at -70°C in 0.4-ml samples of defibrinated rabbit blood. Isolates were subcultured twice on blood agar(37°C, 18 h)
before biochemicaltesting.
Morphology of colonies and cells. Colonial morphologywas determined by observing 24-h growth on blood agar (heart
infusion agar base; Difco Laboratories, Detroit, Mich.) containing 5% defibrinated bovine blood. Growth from the blood agarplate was removed with an inoculatingloop and examined against white filter paper for the presence of
pigment. Hemolytic activity was determined on bovine,
sheep, and rabbit blood agars. Gram staining and acid-fast stainingwere done by the methods ofKoppeloff and Ziehl-Neelson, respectively, on smears prepared from 24-hgrowth onbovine blood agar(6). Motility was determined by
exam-ination of wet mount preparations, using phase-contrast
microscopy. Wet mounts were prepared by using growth
fromchocolate agar (Columbia base; Difco) slants that had been incubated for 24 hat 25 and 37°C.
Biochemical tests. Dehydrated media from commercial
sources wereusedwhenever possible. Inoculated media for all tests were incubated aerobically at 37°C. Catalase, in-dophenol oxidase, phosphatase, and glucosidase enzyme testresultswererecordedat24 h. Tests for nitrate andnitrite reduction and indole productionwereperformedafter48 h of incubation. The Voges-Proskauer and methyl redtestswere done after 4days of incubation. All other biochemical tests wereobserved daily for 1 week, and resultswereconsidered
negative ifno reaction was observed by that time. Unless otherwiseindicated, methods described by Lanyiwere used for the conventional biochemical tests (12). The test for indoleproduction was done by usingEhrlichreagent follow-ing extraction with xylene. Indophenol oxidase was detected
on filter paper, using the tetramethyl-p-phenylenediamine
dihydrochloride reagent (Marion Scientific, Div. Marion
Laboratories, Inc., Kansas City, Mo.). Porphyrins were detected with a commercial disk reagent (Remel, Lenexa, Kans.). Stab cultures for detecting gelatin hydrolysis were incubated at 37°C and then chilled at4°C for 10min before resultswere recorded (5). Hydrolysis of esculin was tested
on heart infusion agarslants containing 1%(wt/vol) esculin 2169
on April 11, 2020 by guest
http://jcm.asm.org/
TABLE 1. P. caballisp. nov. strains studied
Strain Location ofsender Source Otherclinical information
83851T Washington Uterus Adultfemale
84314 California Lungabscess" Adult male,respiratory and central nervoussystemsigns preceding death
84513 New Mexico Trachea 1-yr-old female
84532 South Carolina Uterus 12-yr-old female
84678 Indiana Pleuralfluid" 3-yr-old female,pneumonia
84687 Washington Uterus 5-yr-oldfemale, metritis
85256 Kentucky Gutteralpouch 10-yr-old female, sinus infection
85683 Maryland Cervix 19-yr-old female,infertile
86026 Pennsylvania Bone 2-yr-oldmale, Actinomyces sp.also isolated
86200 Montana Leg wound 4-yr-oldfemale
86514 Wisconsin Lunga 4-yr-old female, pneumonia and meningitis
87127 Illinois Lunga 14-yr-old female,inhalationpneumonia
87151 California Lunga 4-yr-oldmale,respiratory disease, centralnervous systemsigns
87658 Alabama Lung' 3-mo-oldfemale, fibrinouspneumonia, pleuritis
87659 Alabama Lung 2-mo-old male, pneumoniaRhodococcusequialsoisolated
88036 Michigan Trachea 5-mo-oldmale,pneumonia, Actinobacillus equuli also isolated 88037 Michigan Mesentericabscess" 11-yr-oldmale
88046 Alabama Infectedincision 7-yr-old female, mixed culture
88176 Kentucky Fistulouswithers 2-yr-oldmale, Streptococcuszooepidemicus also isolated 88180 Michigan Trachea 5-mo-oldmale,Bordetella bronchiseptica also isolated
88181 Michigan Lung 6-mo-oldfemale
88193 Maryland Brain 6-yr-old female
88200 Kentucky Spleen 5-yr-old male
88228 Kentucky Lung Adultfemale,pneumonia,Streptococcuszooepidemicusand
Pseudomonasaeruginosa alsoisolated
88282 Michigan Lung 15-yr-oldfemale, centralnervoussystemsigns, Actinobacillus equuli also isolated
88283 Michigan Growthonback 6-yr-old male, mixed culture
88322 Michigan Uterus Adultfemale,Streptococcuszooepidemicus also isolated 88405 Kentucky Peritonealfluid" 11-yr-old female,colicsurgery
88406 Kentucky Gutteralpouch" 15-yr-old female
"Case histories indicatedthat P.caballiwasisolated inhigh numbersorpurecultureorboth from these tissues.
and 0.5% ferric chloride (19). The test for methylene blue milk reduction was performed as described by MacFaddin (13). Glycosidase enzyme tests were done as described by Kilian (9), using 0.1% (wt/vol) solutions of
2-nitrophenyl-P-D-galactopyranoside
(Calbiochem-Behring, La Jolla, Cal-if.),4-nitrophenyl-p-D-glucopyranosiduronic
acid (Sigma Chemical Co., St. Louis, Mo.), and4-nitrophenyl-ot-L-fu-copyranoside (Sigma). For fermentation tests, carbohy-drates (1%, wt/vol) were incorporated into heart infusion broth (Difco) with 1.0% (vol/vol) Andrade's indicator. Gas production from D-glucose was detected in a Durham tube. Cultures were observed daily for 1 week for acid and gas production.
DNA relatedness. The procedures used for preparing un-labeled DNA, the hydroxyapatite method for DNA hybrid-ization, and the calculations used to determine relatedness have been described elsewhere (3). DNA from P. caballi
83851T (type strain) was labeled with
[32P04]dCTP
by nick translation, using a commercial kit (catalog no. 8160SB;Bethesda Research Laboratories, Inc., Gaithersburg, Md.). Because of the relatively low guanine-plus-cytosine (G+C) content of P. caballi DNA, hybridization reactions were
done at 55°C (optimal reassociation) and 70°C (stringent reassociation).
Determination of DNA base composition. The G+C content of DNA prepared from four P. caballi strains was
deter-mined
by
denaturingDNAdissolved in lx SSC(lx
SSC is 0.15 MNaClplus
0.015 M sodiumcitrate)
in a model 2600spectrophotometer
(Gilford
InstrumentLaboratories,
Inc.,Oberlin, Ohio) equipped
withamodel2527thermoprogram-mer. The G+C contents were calculated from the thermal
denaturation
midpoint
temperatures(T,,)
by
using
thefor-mula % G+C = (Tm -
69.3)/0.41 (15).
Proteus mirabilisATCC 14273 was used as acontrol.
Determinationof cellularfattyacids.Thecellular
fatty
acidcompositions
of the P. caballi and reference strains ofPasteurellaceae were determined
by
gas-liquid
chromatog-raphy
by
methods describedby
Moss and co-workers(16).
Cellsfor
fatty
acidanalysis
were obtained aftergrowth
onheart infusion agar with 5% rabbit blood.
Fatty
acids(as
methyl
esters)wereanalyzed by
gas-liquid
chromatography,
using
the HP 5898A microbial identification system(Hewlett-Packard
Co.,
Avondale, Pa.).
Identifications ofindividual
fatty
acids were confirmedby
hydrogenation,
acetylation,
and mass spectrometry.Antimicrobial susceptibility tests.
Antibiograms
were de-terminedby
the disk diffusion method of Bauer et al.(1),
using
Mueller-Hinton agar(Difco) supplemented
with 5%sheep
blood and 1% IsoVitaleX(BBL
Microbiology
Sys-tems,
Cockeysville,
Md.).
The antibiotics and theconcen-trations used are listed in Table 3. Zone sizes were
inter-preted
asresistant,
intermediate,
orsusceptible
according
toon April 11, 2020 by guest
http://jcm.asm.org/
TABLE 2. Strains used in DNA relatedness studies
Organismand strain" designationOther Source Sender'
Pasteurella multocida subsp. multicida NCTC 10322T ATCC 43137T Swine NCTC
P. multocida subsp. gallicida NCTC10204T Bovine NCTC
P. multocida subsp. septicaNCTC 11619 Human NCTC
P.multocida ATCC27883T Swine ATCC
P. anatisNCTC 11413T Duck NCTC
P. aviumATCC 29546T Chicken ATCC
P. canisNCTC 11621T Dog NCTC
P. dagmatis NCTC11617T Human NCTC
P.gallinarumP-913 ATCC 13361T Chicken K. Rhoades
P.haemolytica ATCC33396T NCTC9380T Sheep ATCC
P. langaa NCTC 11411T Chicken NCTC
P.pneumotropica ATCC 35149T NCTC8141T Mouse ATCC
P. stomnatis NCTC11623T Dog NCTC
P.testudinis ATCC33688T Tortoise ATCC
Actinobacillus equuli ATCC19392T NCTC 8529 Horse ATCC
A.lignieresii ATCC 19393T NCTC 4976 Bovine ATCC
A. suis88441 Horse Va.
A.ureae NCTC 10219T ATCC25976 Human NCTC
Actinobacillus sp. Ross 63' ATCC 27073 Swine R. F. Ross
Haemophilus influenza ATCC 33391T NCTC 8143 Human ATCC
H. aphrophilus F3436 Human R. E. Weaver
'H. somnus" ISU 8025 Bovine M. L. Kaeberle
"Straindesignations arethoseof the sender. T, Type strain. NCTC,National Collection of Type Cultures, Central Public Health Laboratory, London, England;ATCC, AmericanTypeCulture Collection, Rockville, Md.
h R. E.Weaver, Special Bacteriology Reference
Laboratory.
Centers for DiseaseControl,Atlanta, Ga;K.Rhoades, National Animal DiseaseCenter, Ames, Iowa.;R. F.Ross, Veterinary Medical Research Institute,IowaStateUniversity,Ames; M. L. Kaeberle. College of VeterinaryMedicine, Ames, Iowa.; Va.,stateof origin ofculturesubmittedto theNationalVeterinary Services Laboratories,Ames,Iowa. ' Seereference20.
the
guidelines
of the National Committee for ClinicalLabo-ratory Standards
(18).
RESULTSANDDISCUSSION
Sources and habitats. All P. caballi strains were isolated
from
equine
clinicalspecimens
and representhorsesfrom13states. The
larger
number of strains submitted fromKen-tucky
andMichigan
mayreflect
an interest on the part ofcolleagues
in thisstudy
rather than anunusually high
inci-TABLE 3. Antimicrobialsusceptibilityresults for 29
P. caballi strainsby agar diffusion
Antimicrobial agent No. of strains
(diskconcn)" Resistant Intermediate Susceptible
Amikacin(30) 0 3 26
Ampicillin (10) 0 0 29
Carbenicillin(100) 0 0 29
Cephalothin(30) 0 0 29
Chloramphenicol (30) 0 0 29
Erythromycin(15) 0 3 26
Gentamicin(10) 0 0 29
Kanamycin(30) 0 2 27
Lincomycin(2) 29 0 0
Nalidixicacid(30) 0 1 28
Neomycin (30) 4 il 14
Penicillin G (10) 0 8 21
Streptomycin(10) 12 6 il
Sulfadiazine (300) 26 3 0
Tetracycline(30) 0 0 29
Tobramycin (10) 0 0 29
Trimethoprim (5) il 0 18
'Alldisk concentrationsexcept forpenicillinGareinmicrograms. Peni-cillin G concentration isgiven inunits.
dence of this organism. Of the 29 strains, 15 were isolated from respiratory sources, suggesting that P. caballi, like other Pasteurella spp., is an inhabitant of the mucous
membranes of the upperrespiratory tract. P. caballi strains were isolated in high numbers and in pure culture from horses with upper-respiratory-tract infection, pneumonia, andperitonitis and from a mesenteric abscess. This evidence suggests that P. caballi played a significant role in the pathogenesis of those infections. The clinical significance of P. caballi is less certain in instances inwhich theorganism was presentin mixed cultures.
Phenotypic characteristics. The 29 equine strains studied
were small, nonmotile, bipolar-stained, gram-negative rods TABLE 4. Variablephenotypic characteristics
of P. caballi strains
Cumulative %of
strainspositive Reaction oftype
Test on day: strain 83851"
1 2 7
Acidfrom:
mvo-Inositol 14 14 14
-Lactose 0 14 100 (+)
Maltose 76 83 93 +
D-Mannitol 93 97 97 +
Raffinose 0 7 72 (+)
L-Rhamnose 21 21 21
D-Sorbitol 14 14 14
D-Xylose 59 59 59 +
Growthat42°C 93 93 93 +
L-Ornithinedecarboxylase 41 59 66 +
Yellowpigment 76 76 76 +
-,Negative after 7 days;+. positive within 2days;t+),positive in 3 to 7days.
on April 11, 2020 by guest
http://jcm.asm.org/
TABLE 5. Differentiation of P. caballi from selected aerogenic species of Pasteurellaceae'
Characteristic P. P.aero- P.dag- H.aphro- Actino-caballi genes matis philus bacillusSp."
Gas fromD-glucose + + + + +
Growth on - + - - +
MacConkey agar
Catalase - + + - +
C02enhances growth - - - +
Indole - - +
Betahemolysis - - - - +
L-Ornithine d' d
decarboxylase
Urease - + + - +
Acidfrom:
L-Arabinose - d - - +
myo-Inositol d d - d +
L-Rhamnose d d - - +
D-Sorbitol d d - - +
Trehalose - - + +
D-Xylose d d - - +
Usualhost
Human - - - +
Horse +
Pig - + - - +
Dog or cat - - +
aDatafor previously described speciesareadaptedfromreferences4, 10,
11,17,and 20.
bSeereference20.
Cd, Strainsvarywithin species.
that gave positive reactions in tests for oxidase, phos-phatase, and nitrate reduction. In addition, all strains were
facultative anaerobes, and all produced acid from the
fer-mentation of carbohydrates. Additional characteristics that were common to all strains are given below in the species
description. Variable characteristics for the 29 strains are
given in Table 4. The 29 equine strains studied had the
characteristics of the family Pasteurellaceae (14). Of the three existinggenera in this family(i.e., Pasteurella,
Hae-mophilus, andActinobacillus), theseequinestrains are best
placed in the genusPasteurella (4, 17). A lack of a
require-ment for preformed growth factors excludes these strains
from the genus Haemophilus. Members of the genus
Acti-nobacillus typically produce sticky colonies, grow well on
MacConkeyagar, andhydrolyzeurea,characteristics absent in these strains. Several characteristics unusual for Pas-teurella spp. were observed: a negative catalase test,
pro-duction ofgas from D-glucose, and the production of acid
frommyo-inositol and L-rhamnose by aminority of strains.
However, we do not believe these differences warrant exclusion of these strains from thegenuswhen thenumerous morphological and biochemical characteristics in common with Pasteurella spp. are considered. The tests that are
useful fordifferentiating P. caballi from selected aerogenic members of the familyPasteurellaceaearegiven in Table 5.
Negative testsfor catalase and urease as well as a lack of
growth on MacConkey agar distinguish P. caballi from
Pasteurella aerogenes and theActinobacillus sp. described
by Ross et al. (20). P. caballi can be distinguished from Pasteurella dagmatis by its negative reactions for catalase, indole, urease, andfermentation of trehalose. P. caballi is phenotypicallymostsimilartoHaemophilus aphrophilus but
can be differentiated by its negative test for trehalose fer-mentation and its ability to grow well without additional
carbon dioxide. The source of an isolate would be of
particular importance in differentiating P. caballi from H. aphrophilus,since eachappearstohavealimited hostrange. Antibiotic susceptibility. The responses of the P. caballi strains to several antibiotics are shown in Table 3. The
antibioticsusceptibilities of these strainsaregenerally
con-sistent with those reported for other members of thegenus Pasteurella (7).
Fatty acid analyses. The fatty acid composition of 29 strains ofP.caballiwassimilartothoseof strains of 12 other species of Pasteurellaceae examined (Table 6). Relatively large amounts of tetradecanoic (C14:0), hexadecenoic (C16:1
w7c; palmitoleic), and hexadecanoic (C16:0; palmitic) acids
werefound. Alsopresent,but inlowerconcentrations,were 3-hydroxy-tetradecanoic
(3-OHC14:0),
cis-9,12-octadecadie-noic (C18:2; linoleic), cis-9-octadecenoic (C181 9c; oleic), and octadecanoic (C18:0; stearic) acid. A small amount (2%or less)ofdodecanoic (C12:0; lauric) acid and an isomer of
tetradecenoic (C14:1) were also present in all strains
exam-ined. These results are in general agreement with those of Jantzen and co-workers (8), who examined 18 species of Pasteurellaceae,exceptthatweobtained smalleramountsof
TABLE 6. Cellularfatty acidcompositionof P.caballi and otherspeciesofPasteurellaceae
%of totalfattyacidcomposition (SD)" Species No. ofstrains
C14:0 3-OHC14:0 CI6:,w7c C16:0 C18:2 C18:1w9c C18:0
Pasteurella caballi 29 30(2) 9(2) 28(3) 16(2) 5(1) 4(1) 2(1)
P.multocida il 19(3) 5(1) 29(2) 32(3) 4(1) 2(1) 3(1)
P.dagmatis 4 25(3) 8(1) 29(5) 22(2) 6(2) 3(1) 3(1)
P.pneumotropica 3 25(5) 5(1) 31 31(1) 1 2(1) 2(2)
P.aerogenes 2 22 7(1) 31(2) 28 3 2 2
P.canis 2 21(7) 6 24(5) 28(8) 9 4(1) 4(3)
P.stomatis 2 25(4) 10(3) 22(5) 21(3) 9(1) 6(1) 3
Actinobacillusequuli 4 25(5) 7(1) 27(4) 19(2) 8(3) 5(2) 3(2)
A.Iignieresii 3 25(2) 7(1) 31(2) 22(2) 5 (1) 3 2
A. ureae 3 29(5) 6 23(5) 24(5) 6(2) 5(2) 3(1)
Actinobacillus sp.c 2 32(6) 7(1) 32 18(5) 4 (1) 2 1
Haemophilusaphrophilus 7 21(40) 6(2) 26(9) 33(3) S (3) 4(2) 2(1)
H. influenzae 4 18 (6) 4(2) 35(5) 35 (3) 1 T 2(1)
aThe first number is thearithmeticmeanof the relative percentage of thefattyacid;absence ofanumber inparenthesesindicatesthat thestandarddeviation
wasless than0.6%. Allvalues were rounded to the nearest whole number.T,less than0.6%detected.
bNumbers before the colon indicate the numberofcarbon atoms, and numbers after the colonrefertothe numberof double bonds. 3-OH indicateshydroxy group atcarbon3; w7c and w9c indicatedouble-bondpositionfromhydrocarbonendof carbonchain;cis cis isomer.
CSeereference20.
on April 11, 2020 by guest
http://jcm.asm.org/
TABLE 7. DNA relatedness of P. caballi strains
Source(s) ofunlabeledDNA Pasteurella caballi 83851T 85683 88176 84679 88282 84513 88228 87127 84532 85256 87658 88283 88405 86514 88200 87151 86026 88180 84314 84687 87659 88193 88322 88046 86200 88036 88181 88037 88406
P.langaa NCTC11411T
Actinobacillus sp. ATCC 27073
PasteurellapneumotropicaATCC35149TandP.multocida subsp.gallicida NCTC 10204T
P. canisNCTC 11621T,P. multocida subsp.septicaNCTC 11619,P. multocida subsp. multocida NCTC10322T,P. avium ATCC29546', and P.dagmatis NCTC11617T
P. anatisNCTC 11413', P.stomatisNCTC 11623',P. testudinis ATCC
33688',
Actinobacillus equuli ATCC19396',
A. lignieresiiATCC 19393T,A.ureaeNCTC 10219T, Haemophilusaphrophilus F3436,H.influenza
ATCC33391T,and "H. somnus" ISU8025
Pasteurellagallinarum P-913andActinobacillussuis88441 P. aerogenesATCC27882Tand P. haemolytica ATCC33396T
Relatednessto labeled DNAfromP.caballi 83851T
RBR at55°C" oDb RBRat700C
100 99 98 97 97 94 94 94 93 93 93 93 93 92 92 91 90 90 89 89 89 89 89 88 87 87 87 85 83 53 36 30-35 0.0 1.0 1.0 1.0 1.0 1.0 2.0 1.5 1.0 1.0 1.5 1.0 1.5 1.0 1.0 1.5 1.0 1.0 1.0 1.0 1.0 1.5 1.0 1.0 1.0 1.0 1.0 1.0 1.0 11.5 14.0 100 85 77 86 88 91 89 87 86 77 82 85 86 87 87 85 88 85 85 85 84 89 84 87 82 85 88 82 82 21 il 25-29 20-24 15-19 10-14
aRBR, Relativebinding ratio (calculatedas100x percent DNA bound tohydroxyapatiteinheterologousreactions/percentDNA boundtohydroxyapatitein
homologousreactions.
b%D,Percentdivergence (calculatedontheassumptionthat1%decrease in thermalstabilityofaheterologousDNAduplexwascausedby1% unpaired bases). Itisexpressedtothenearest0.5%.
3-OH
C14:0
acid. This result is due todifferencesinmethods,because our base hydrolysis procedure does not
quantita-tively liberate all amide-linked hydroxy acids. Althoughfatty
acid composition is oflittle value in differentiating among generaand species of Pasteurellaceae, all members of the
family are easily distinguished from other small,
gram-negative, rod-shaped bacteria by their characteristic fatty
acid composition.
DNAstudies.The mean G+C contentofDNAfrom four P. caballi strains, including the type strain, was 41.2 to 42.4
mol%.These values are within the range (37.7 to 45.9) given
for the genus Pasteurella (4). The results of DNA-DNA hybridization experiments are shown in Table 7. Strains
designated P. caballi formed a single DNA relatedness group.The levelsof relatedness of labeled P. caballi DNA to DNA from 28 other P. caballi strains were 83 to 99% (average, 91%) in reactions at the optimal reassociation
temperature. DivergenceinrelatedDNAsequenceswas1 to
2%. Relatedness remained high (average, 85%) under
strin-gentreassociation conditions. Strain
83851'
was53%orlessrelated to other species ofPasteurellaceae, indicating that theequine strainsrepresenta new species. Closest related-nessoccurred with Pasteurellalangaa. Relatednesstoother species of Pasteurellaceae was 13to 34%.
Insummary, resultsofphenotypiccharacterization, fatty acid analyses, and DNA-DNA hybridization studies show
that these equinestrainsform anew, homogenous taxon in the genus Pasteurella. We propose that these strains be
designated Pasteurella caballi.
Description of P. caballi sp. nov. Pasteurella caballi
(ca.bal'li. L.n. caballus, a horse; L.gen.n. caballi, of a horse; Pasteurella caballi, thePasteurella ofa horse) cells
arerod-shaped (0.8to 1.0by1.3 to 1.9F.m), bipolar stained,
and arranged singly or in pairs with occasional swollen,
on April 11, 2020 by guest
http://jcm.asm.org/
curved, or filamentous forms observed. The bacterium is gram negative, not acid fast, nonmotile, and facultatively anaerobic. Endospores are notformed. Colonies are 1.0 to 1.5 mm in diameter, smooth, slightly raised, and grayish yellow on blood agar after 24 h of incubation. Colonies grow atroomtemperature and at 37°C butnotat 4°C. Colonies are nothemolyticon bovine, sheep, or rabbit blood agar but do impart a brownish discoloration to these media under areas
ofconfluent growth. Growth scraped up from the surface of the blood agar is cream (less frequently) to bright yellow (usual). Growth on MacConkey agar is scant or absent. The
bacterium is X factor and V factor independent and positive
for oxidase, phosphatase, and
P-galactosidase.
Nitrate is reduced to nitrite. Methylene blue milk is reduced. Porphy-rins are produced from 8-aminolevulinic acid. Catalase,indole,lysinedecarboxylase,arginine dihydrolase,
a.-fucosi-dase, and 3-glucuronidase are not produced. Esculin,
gela-tin, and urea are not hydrolyzed. Citrate is not utilized. Voges-Proskauer and methyl red tests are negative.
D-Glucose is fermented, with the production of acid and gas.
Acid is produced from D-fructose, D-galactose, lactose, D-mannose, and sucrose. Fermentation of lactose is delayed. Acid is not produced from adonitol, L-arabinose, cellobiose,
dulcitol, inulin, salicin, starch, or trehalose. The bacterium
contains 3-OH
C14:0
and approximately 25% each ofC14:0,
C16,
7c, andC16:0
acids, which is the characteristic fattyacid profile of all members of the family Pastuerellaceae. The G+C content of DNA is 41 to 42mol%
(T,,,).
P. caballistrains have been isolated fromequine clinical specimens.
The type strain of P. caballi is 83851 (=ATCC 49197), and it
exhibits all of the characteristics described above.
Addi-tional variable characteristicsare givenin Table 4. The type
strain was isolated in 1983 from the uterus ofa mare. The
G+C contentofthe DNA was 42 mol%.
ACKNOWLEDGMENTS
WethankJ. M.Donohue, R.Walker,A.Render, andothers who submittedisolates.
LITERATURE CITED
1. Bauer, A. W., W. M. M. Kirby, J. C. Sherris, and M. Turck. 1966. Antibiotic susceptibility testing by standardized single diskmethod. Am. J. Clin. Pathol. 45:493-496.
2. Biberstein, E. L. 1981. Haemophilus-Pasteurella-Actinobacil-lus: their significance in veterinary medicine, p. 61-74. In M. Kilian,W. Fredericksen, andE. L.Biberstein(ed)., Haemoph-ilus, Pasteurella, and Actinobacillus. Academic Press, Inc. (London), Ltd., London.
3. Brenner, D. J., A. C. McWhorter, J. K. Leete Knutson, and A.G. Steigerwalt. 1982. Escherichiavulneris:a newspecies of Enterobacteriaceae associated with human wounds. J. Clin. Microbiol. 15:1133-1140.
4. Carter,G. R. 1980.Genus I.Pasteurella,p. 552-557. In N. R. Krieg and J. G. Holt (ed.), Bergey's manual of systematic bacteriology, vol. 1. TheWilliams & WilkinsCo.,Baltimore. 5. Clark, W.A., D. G. Hollis,R.E. Weaver, and P.Riley. 1984.
Identification ofpathogenic gram-negative aerobicand faculta-tively anaerobic bacteria, p. 14. Centers for Disease Control, PublicHealthService, U.S. Departmentof Health and Human
Services, Atlanta.
6. Cottral, G. E. (ed). 1978. Manual of standardized methods for veterinary microbiology, p. 697-698, 701. Cornell University Press, Ithaca, N.Y.
7. Fales, W. H., L. A. Selby, J. J. Webber, L. J. Hoffman, L. D. Kinter, S. L. Nelson, R. B. Miller, J. G. Thorne, J. T. McGinity, and D. K. Smith. 1982. Antimicrobial resistance among Pas-teurella spp. recovered from Missouri and Iowa cattle with bovine respiratory disease complex. J. Am. Vet. Med. Assoc. 181:477-479.
8. Jantzen, E., B. P. Berdal, and T.Omland. 1981. Cellularfatty acidtaxonomy ofHaemophilus, Pasteurella, and Actinobacil-lus, p. 197-204. In M. Kilian, W. Fredericksen, and E. L. Biberstein (ed.), Haemophilus,Pasteurella, andActinobacillus. Academic Press, Inc. (London), Ltd., London.
9. Kilian, M.1978. Rapid identificationof Actinomycetaceae and relatedbacteria.J.Clin. Microbiol. 8:127-133.
10. Kilian, M., and E. L. Biberstein. 1980.Genus Il. Haemophilus, p.558-569. In N. R.Kriegand J. G. Holt(ed.), Bergey'smanual ofsystematic bacteriology, vol. 1.TheWilliams & Wilkins Co., Baltimore.
11. Kilian, M., and W. Frederickson. 1981. Identificationtablesfor theHaemophilus-Pasteurella-Actinobacillus group, p.281-290. In M. Kilian, W. Frederickson, and E. L. Biberstein (ed.), Haemophilus, Pasteurella, and Actinobacillus. Academic Press, Inc. (London), Ltd.,London.
12. Lanyi, B. 1987. Classical andrapid identification methods for medicallyimportantbacteria, p. 1-67. In R. R. Colwell and R. Grigorova (ed.), Methods in microbiology, vol. 19. Academic Press, Inc.(London), Ltd., London.
13. MacFaddin, J. F. 1980. Biochemical tests for identification of medicalbacteria, 2nd ed.,p.205-209. TheWilliams& Wilkins Co., Baltimore.
14. Mannheim, W. 1980. Family lII. Pasteurellaceae Pohl, p. 550-552. In N. R.Krieg andJ.G. Holt(ed.), Bergey'smanual ofsystematic bacteriology, vol.1.The Williams &WilkinsCo., Baltimore.
15. Marmur, J., and P. Doty. 1962. Determination of the base composition of deoxyribonucleic acid from its denaturation temperature. J. Mol. Biol.5:109-118.
16. Moss, C. W.,P. L. Wallace, D. G. Hollis, and R. E. Weaver. 1988. Cultural and chemical characterization of CDC groups EO-2, M-5, andM-6, Moraxella (Moraxella) species, Oligella
urethralis, Acinetobacter species, andPsychrobacter immobi-lis. J.Clin. Microbiol. 26:484-492.
17. Mutters, R., P. Ihm, S. Pohl,W. Frederiksen, and W. Mann-heim. 1985. Reclassification of thegenusPasteurellaTrevisan 1887 on the basis of deoxyribonucleic acid homology, with proposalsforthe newspecies Pasteurella dagmatis, Pasteurella canis,Pasteurellastomatis,Pasteurellaanatis, and Pasteurella langaa. Int.J. Syst. Bacteriol. 35:309-322.
18. National Committee for Clinical Laboratory Standards. 1975. Performance standards for antimicrobial disc susceptibility tests. National Committee for Clinical Laboratory Standards, Villanova, Pa.
19. Phillips,E., and P. Nash.1985.Culturemedia,p. 1051-1092. In E. H. Lennette, A. Balows, W. J. Hausler, Jr., and H. J. Shadomy (ed.), Manual of clinicalmicrobiology,4thed. Amer-icanSociety forMicrobiology,Washington, D.C.
20. Ross, R. F., J. E. Hall, A. P. Orning, and S. E. Dale. 1972. Characterization of an Actinobacillus isolated from the sow
vagina. Int. J. Syst. Bacteriol. 22:39-46.
21. Schlater, L. K. 1989. Anaerogenic Pasteurella-like organism isolated from horses. J. Vet. Diagn. Invest. 1:3-5.