• 1

    Mare Productive Loss Syndrome (MRLS). Available at: www.ca.uky.edu/gluck/mrls.index.aspAccessed July 5, 2004.

  • 2

    Harrison LR. Kentucky equine abortion storm and related conditions, in Proceedings. 105th Annu Meet US Anim Health Assoc 2001;227229.

    • Search Google Scholar
    • Export Citation
  • 3

    Dwyer RM. Epidemiological correlates of the 2001 and 2002 episodes of mare reproductive loss syndrome, in Proceedings. 1st Workshop Mare Reprod Loss Syndr 2003;3436.

    • Search Google Scholar
    • Export Citation
  • 4

    Williams NM, Donahue JM & Bolin DC, et al. Equine placental pathology: Kentucky perspective, in Proceedings. Workshop Equine Placenta 2004;8892.

  • 5

    Cohen ND, Donahue JG & Carey VJ, et al. Case-control study of early-term abortions (early fetal losses) associated with mare reproductive loss syndrome in central Kentucky. J Am Vet Med Assoc 2003;222:210217.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 6

    Dwyer RM, Garber LP & Traub-Dargatz JL, et al. Case-control study of factors associated with excessive proportions of early fetal losses associated with mare reproductive loss syndrome in central Kentucky during 2001. J Am Vet Med Assoc 2003;222:613619.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 7

    Seahorn JL, Slovis NM & Reimer JM, et al. Case-control study of factors associated with fibrinous pericarditis among horses in central Kentucky during spring 2001. J Am Vet Med Assoc 2003;223:832838.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 8

    McDowell KJ, Williams NM & Donahue JM, et al. Deductive investigations of the role of eastern tent caterpillars in mare reproductive loss syndrome, in Proceedings. Workshop Equine Placenta 2004;99103.

    • Search Google Scholar
    • Export Citation
  • 9

    Sebastian MM, Bernard WV & Harrison LR, et al. Experimental induction of mare reproductive loss syndrome with irradiated eastern tent caterpillars to assess whether the primary pathogen is a toxic molecule or a microorganism, in Proceedings. Workshop Equine Placenta 2004;103104.

    • Search Google Scholar
    • Export Citation
  • 10

    Webb BA, Barney WE & Dahlman DL, et al. Eastern tent caterpillars (Malacosoma americanum) cause mare reproductive loss syndrome. J Insect Physiol 2004;501:185193.

    • Search Google Scholar
    • Export Citation
  • 11

    Cohen ND, Carey VJ & Donahue JG, et al. Descriptive epidemiology of late-term abortions with the mare reproductive loss syndrome in central Kentucky. J Vet Diagn Invest 2003;15:295297.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 12

    Williams NM, Bolin DC & Donahue JM, et al. Gross and histopathological correlates of mare reproductive loss syndrome, in Proceedings. 1st Workshop Mare Reprod Loss Syndr 2003;2425.

    • Search Google Scholar
    • Export Citation
  • 13

    Donahue J, Sells S & Giles R, et al. Bacteria associated with mare reproductive loss syndrome: late fetal losses, in Proceedings. 1st Workshop Mare Reprod Loss Syndr 2003;2729.

    • Search Google Scholar
    • Export Citation
  • 14

    Bolin DC, Donahue JM & Vickers ML, et al. Microbiologic and pathologic findings in an epidemic of equine pericarditis. J Vet Diagn Invest 2005;17:3844.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 15

    Bisgaard M, Piechulla K & Ying YT, et al. Prevalence of organisms described as Actinobacillus suis or haemolytic Actinobacillus equuli in the oral cavity of horses. Comparative investigations of strains obtained and porcine strains of A. suis sensu stricto. Acta Pathol Microbiol Immunol Scand [B] 1984;92:291298.

    • Search Google Scholar
    • Export Citation
  • 16

    Christensen H, Bisgaard M. Revised definition of Actinobacillus sensu stricto isolated from animals. A review with special emphasis on diagnosis. Vet Microbiol 2004;99:1330.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 17

    Christensen H, Bisgaard M & Angen O, et al. Final classification of Bisgaard taxon 9 as Actinobacillus arthritidis sp. nov. and recognition of a novel genomospecies for equine strains of Actinobacillus lignieresii. Int J Syst Evol Microbiol 2002;52:12391246.

    • Search Google Scholar
    • Export Citation
  • 18

    Christensen H, Bisgaard M, Olsen JE. Reclassification of equine isolates previously reported as Actinobacillus equuli, variants of A. equuli, Actinobacillus suis or Bisgaard taxon 11 and proposal of A. equuli subsp. equuli subsp. nov. and A. equuli subsp. haemolyticus subsp. nov. Int J Syst Evol Microbiol 2002;52:15691576.

    • Search Google Scholar
    • Export Citation
  • 19

    Sternberg S. Isolation of Actinobacillus equuli from the oral cavity of healthy horses and comparison of isolates by restriction enzyme digestion and pulsed-field gel electrophoresis. Vet Microbiol 1998;59:147156.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 20

    Sternberg S, Brandstrom B. Biochemical fingerprinting and ribotyping of isolates of Actinobacillus equuli from healthy and diseased horses. Vet Microbiol 1999;66:5356.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 21

    Guettler MV, Rumler D, Jain MK. Actinobacillus succinogenes sp. nov., a novel succinic-acid-producing strain from the bovine rumen. Int J Syst Bacteriol 1999;49:207216.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 22

    Jang SS, Biberstein EL, Hirsh DC. Actinobacillus suis-like organisms in horses. Am J Vet Res 1987;48:10361038.

  • 23

    Samitz EM, Biberstein EL. Actinobacillus suis-like organisms and evidence of hemolytic strains of Actinobacillus lignieresii in horses. Am J Vet Res 1991;52:12451251.

    • Search Google Scholar
    • Export Citation
  • 24

    Drancourt M, Bollet C & Carlioz A, et al. 16S ribosomal DNA sequence analysis of a large collection of environmental and clinical unidentifiable bacterial isolates. J Clin Microbiol 2000;38:36233630.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 25

    Bisgaard M. Ecology and significance of Pasteurellaceae in animals. Zentrabl Bakteriol 1993;279:726.

  • 26

    Giles RC, Donahue JM & Hong CB, et al. Causes of abortion, stillbirth, and perinatal death in horses: 3,527 cases (1986–1991). J Am Vet Med Assoc 1993;203:11701175.

    • Search Google Scholar
    • Export Citation
  • 27

    Hong CB, Donahue JM & Giles RC, et al. Etiology and pathology of equine placentitis. J Vet Diagn Invest 1993;5:5663.

  • 28

    Dibb WL, Digranes A, Tonjum S. Actinobacillus lignieresii infection after a horse bite. Br Med J 1981;283:583584.

  • 29

    Peel MM, Hornidge KA & Luppino M, et al. Actinobacillus spp. and related bacteria in infected wounds of humans bitten by horses and sheep. J Clin Microbiol 1991;29:25352538.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 30

    Drancourt M, Raoult D. Sequence-based identification of new bacteria: a proposition for creation of an orphan bacterium repository. J Clin Microbiol 2005;43:43114315.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 31

    Stackebrandt E, Goebel BM. Taxonomic note: a place for DNA-DNA reassociation and 16S rRNA sequence analysis in the present species definition in bacteriology. Int J Syst Bacteriol 1994;44:846849.

    • Crossref
    • Search Google Scholar
    • Export Citation

Advertisement

Classification of Actinobacillus spp isolates from horses involved in mare reproductive loss syndrome

View More View Less
  • 1 Livestock Disease Diagnostic Center, College of Agriculture, University of Kentucky, Lexington, KY 40511-4125.
  • | 2 Livestock Disease Diagnostic Center, College of Agriculture, University of Kentucky, Lexington, KY 40511-4125.
  • | 3 Livestock Disease Diagnostic Center, College of Agriculture, University of Kentucky, Lexington, KY 40511-4125.

Abstract

Objective—To identify Actinobacillus spp isolates recovered from fetuses and pericardial fluid from horses affected with mare reproductive loss syndrome (MRLS) and determine whether these bacterial species are the same as those isolated from clinically normal horses.

Sample Population—Isolates of actinobacilli recovered from 18 horses with pericarditis and 109 fetuses aborted by mares affected by MRLS.

ProceduresActinobacillus spp isolates were identified to the level of species or subspecies by use of conventional phenotypic tests and biochemical and enzyme test kits. The 16S rRNA gene from selected isolates was amplified, purified, and sequenced. Sequence data were compared with sequence data for actinobacilli in GenBank.

Results—Of the 109 isolates obtained from fetuses, 14 were Actinobacillus equuli subsp equuli, 65 were A equuli subsp haemolyticus, 28 were Bisgaard taxon 10–like bacterium, and 2 were Actinobacillus genomospecies 1. Of the 18 isolates from horses with pericarditis, 4 were A equuli subsp equuli, 13 were A equuli subsp haemolyticus, and 1 was Bisgaard taxon 10–like bacterium. Comparisons with published data and GenBank data revealed that the isolates recovered from horses with MRLS were the same as those isolated from the oral cavity or alimentary tract of healthy horses.

Conclusions and Clinical RelevanceActinobacillus spp isolates recovered from fetuses and pericardial fluid samples of horses affected by MRLS in 2001 to 2003 were identical to Actinobacillus spp found in the oral cavity and alimentary tracts of healthy horses.

Abstract

Objective—To identify Actinobacillus spp isolates recovered from fetuses and pericardial fluid from horses affected with mare reproductive loss syndrome (MRLS) and determine whether these bacterial species are the same as those isolated from clinically normal horses.

Sample Population—Isolates of actinobacilli recovered from 18 horses with pericarditis and 109 fetuses aborted by mares affected by MRLS.

ProceduresActinobacillus spp isolates were identified to the level of species or subspecies by use of conventional phenotypic tests and biochemical and enzyme test kits. The 16S rRNA gene from selected isolates was amplified, purified, and sequenced. Sequence data were compared with sequence data for actinobacilli in GenBank.

Results—Of the 109 isolates obtained from fetuses, 14 were Actinobacillus equuli subsp equuli, 65 were A equuli subsp haemolyticus, 28 were Bisgaard taxon 10–like bacterium, and 2 were Actinobacillus genomospecies 1. Of the 18 isolates from horses with pericarditis, 4 were A equuli subsp equuli, 13 were A equuli subsp haemolyticus, and 1 was Bisgaard taxon 10–like bacterium. Comparisons with published data and GenBank data revealed that the isolates recovered from horses with MRLS were the same as those isolated from the oral cavity or alimentary tract of healthy horses.

Conclusions and Clinical RelevanceActinobacillus spp isolates recovered from fetuses and pericardial fluid samples of horses affected by MRLS in 2001 to 2003 were identical to Actinobacillus spp found in the oral cavity and alimentary tracts of healthy horses.

The population of horses residing in central Kentucky1,2 during the spring of 2001 was affected by an acute-onset epidemic of early-term abortions, late-term abortions, stillbirths, fibrinous pericarditis, and uveitis. The disease outbreak was subsequently referred to as MRLS. There was a recurrent outbreak in the spring of 2002, but the incidence of disease was much lower than in 2001.3,4

Epidemiologic studies5–7 revealed that exposure of horses to ETCs (Malacosoma americanum) was the most important risk factor for abortion and fibrinous pericarditis. Researchers have subsequently reproduced the clinical signs of MRLS-associated early-and late-term abortions by feeding ETCs to pregnant mares8–10 and have also determined that ingestion of ETCs can cause abortion in pigs.8 The mechanism by which ETCs cause abortion is unknown, but only the worm exoskeleton and attached setae are required to induce abortion.8–10 Microgranulamatous lesions surrounding ETC setae embedded in the submucosal lining of the alimentary tract of pigs and horses fed the worms have been reported.8,a

Pathologic findings that are pathognomonic for MRLS have not been described.11,12 The most common characteristic observed in postmortem examinations was the finding of numerous bacteria (nearly always non–β-hemolytic streptococci or actinobacilli) in pure culture from fetal and placental tissue4,11,13 and, in nontreated horses with pericarditis, isolation of Actinobacillus spp from the myocardium or pericardial fluid.14 The same types of bacteria were also isolated from aborted fetuses of mares that were exposed to ETCs in the course of experimental reproduction of MRLS.8,10 It is postulated that ETC setae in the intestinal lumen penetrate the mucosal barrier and allow bacteria residing on the mucosa of otherwise healthy horses to enter the systemic circulation. Actinobacilli, usually considered opportunistic pathogens, are part of the normal flora in various body systems in horses and can cause the types of infections observed in MRLS.15–20 The purpose of the study reported here was to classify actinobacilli recovered from horses involved in the MRLS epidemic by use of phenotypic and genotypic techniques and to determine whether those isolates were identical to the actinobacilli that are a part of the normal resident flora of horses.

Materials and Methods

Bacteria—One hundred twenty-seven isolates of actinobacilli were obtained. One hundred twenty-three isolates were obtained from horses involved in the spontaneous outbreaks of MRLS in 2001 (isolates from 84 abortions and 12 horses with pericarditis), 2002 (isolates from 15 abortions and 5 horses with pericarditis), and 2003 (isolates from 6 abortions and 1 horse with pericarditis); and 4 isolates were obtained from fetuses of mares in which MRLS was experimentally induced in 2003. Conventional bacteriologic techniques were used to isolate actinobacilli from lung specimens or gastric contents of fetuses or from the pericardium or pericardial fluid of horses involved in the natural outbreaks. All isolates were identified as belonging to genus Actinobacillus by use of a limited number of conventional biochemical tests and were frozen in skim milk at −80°C until analysis. For purposes of comparison, 10 type species were purchased from ATCC,b including ATCC 51571 (Actinobacillus capsulatus), 19392 (Actinobacillus equuli), 13376 (Actinobacillus arthritidis [originally deposited as A equuli]), 49457 (Actinobacillus hominis), 49236 (Actinobacillus lignieresii), 49577 (Actinobacillus muris), 15768 (Actinobacillus seminis), 33415 (Actinobacillus suis), 25976 (Actinobacillus ureae), and 33384 (Actinobacillus [Haemophilus] actinomycetemcomitans).

Phenotypic characterization—To determine whether isolates were actinobacilli, all field strains with colony morphology typical of actinobacilli and all of the type strains were tested by use of the following media: triple sugar iron slants, urea agar slants, esculin agar slants, SIM medium, ortho-nitrophenyl-β-D-galactopyranoside broth, and nitrate medium. Strains were also assessed for adherence to agar medium, catalase activity by the addition of 1 mL of 3% hydrogen peroxide to colonies after 18 to 24 hours of growth on tryptose agar, oxidase activity by use of Kovàcs oxidase reagent, hemolytic activity on 5% equine blood agar and 5% ovine blood agar plates, and growth on eosin-methylene blue agar.

Biochemical analyses and determination of enzyme activities—Each of 101 field strains (26 isolates died before they could be frozen) and the 10 ATCC strains were characterized biochemically by use of a commercially available biochemical kit.c Selected strains were also tested with a kit for determination of enzyme activities.d Manufacturer's directions were followed in preparation of reagents and media. Inoculated strips were incubated at 37°C and read daily for 5 days. Pseudomonas aeruginosa (ATCC 27853) and Klebsiella pneumoniae (ATCC 35657) were used for quality control of the enzyme and biochemical kits, respectively.

Sequencing of 16S rRNA genes and data analysis—The gene encoding bacterial 16S rRNA from each of the phenotypically characterized groups was amplified, purified, and sequenced (≥ 1,450 bp). Genomic DNA was isolated from bacteria that were removed from an equine blood agar plate and placed in sterile saline (0.9% NaCl) solution.e The gene was amplified and sequenced.f Reactions were purified, and sequencing gels were run at the University of Kentucky Molecular Structure Analysis Facility in Lexington, Ky. Data were assembled and edited by use of commercial software.g

Results

Phenotypic characterization of isolates—Results of conventional bacteriologic analytic methods revealed that all 127 isolates were nonmotile gram-negative bacilli that were strongly urease positive, produced an acid slant and butt without H2S formation in triple sugar iron slants, had β-D-galactosidase activity, reduced nitrate to nitrite without formation of gas, and had negative results for H2S and indole production in SIM medium. All isolates were catalase negative, except for 2 isolates of A equuli subsp equuli that had weak catalase activity. All field isolates had oxidase activity, were adherent to the surface of blood agar plates, had no growth on eosin-methylene blue agar, and had no H2S production in triple sugar iron slants, except for the 29 Bisgaard taxon 10–like isolates, which were oxidase negative, were nonadherent to the agar, grew on eosin-methylene blue agar (colonies were 1.5 to 2.0 mm in diameter after 48 hours of incubation), and produced a slight quantity of gas in triple sugar iron slants after 48 hours of incubation. Seventy-six of the 78 A equuli subsp haemolyticus isolates (ie, all but the 2 biovar-3 isolates) had positive results for hydrolysis of esculin; none of the other 49 isolates hydrolyzed esculin. None of the 18 A equuli subsp equuli or 2 Actinobacillus genomospecies-1 isolates were hemolytic on ovine blood agar or equine blood agar. Of 29 Bisgaard taxon 10–like isolates, 21 had narrow zones of hemolysis (similar to Mannheimia haemolyticus) after 48 hours of incubation on ovine blood agar and equine blood agar. All except 3 (2 biovar-1 and 1 biovar-2 isolates) of the 78 A equuli subsp haemolyticus isolates were hemolytic on ovine blood agar, but only 37 of the 78 isolates were hemolytic on equine blood agar. If an isolate was hemolytic on both types of agar, the hemolytic zone was usually much wider on the ovine blood agar plates.

All 10 type species had oxidase activity, produced an acid slant and butt in triple sugar iron media without H2S or gas production, had negative results for H2S and indol production in SIM medium, did not grow on eosin-methylene blue agar, and except for A actinomycetemcomitans, reduced nitrate to nitrite without formation of gas. Variable results were obtained with the other tests. All strains except A seminis and A actinomycetemcomitans had urease activity and, except for A muris, A seminis, and A ureae, had β-galactosidase activity in ortho-nitrophenyl-β-D-galactopyranoside broth. Only A suis was hemolytic on the blood agar plates, and the only strains that hydrolyzed esculin were A capsulatus, A muris, A seminis, and A suis. Eight of the strains were nonadherent to the agar surface; A equuli and A arthritidis were adherent. Test results for catalase production were negative for A equuli, A arthritidis, A suis, and A ureae; results were weakly positive for A hominis, A lignieresii, and A muris and were positive for A capsulatus, A seminis, and A actinomycetemcomitans.

Biochemical characterization of isolates—All 101 field strains had the same or similar reactions for 32 of the 49 tests. All strains had positive results for fermentation of glycerol, ribose, D-xylose, glucose, fructose, maltose, lactose, sucrose, and raffinose; all but 1 A equuli subsp haemolyticus–other biovar isolates had positive results for mannose fermentation; and all except the 2 Actinobacillus genomospecies–1 isolates had positive results for melibiose and trehalose fermentation. All 101 strains had negative results for fermentation of erythritol, L-xylose, adonitol, β-methyl-D-xyloside, sorbose, dulcitol, inositol, α-methyl-D-mannoside, α-methyl-D-glucoside, melezitose, xylitol, D-turanose, D-lyxose, D-tagatose, D-fucose, D-arabitol, L-arabitol, and 2-keto-gluconate. All field strains except 1 A equuli subsp haemolyticus biovar-1 strain also had negative results for fermentation of inulin, and all but 3 (2 biovar 1 and 1 biovar 2) were negative for glycogen fermentation. Variable reactions were obtained for the remaining 17 substrates (Table 1).

Table 1—

Number of isolates of Actinobacillus equuli subsp equuli (15 tested); A equuli subsp haemolyticusbiovar 1 (35 tested), biovar 2 (18 tested), biovar 3 (2 tested), and other biovars (7 tested); Actinobacillus genomospecies 1 (2 tested); and Bisgaard taxon 10–like (22 tested) that yielded positive results for selected substrate tests.

SubstrateA equuliBiovar 1Biovar 2Biovar 3OtherSpecies 1Taxon 10
D-arabinose0*000004 (1)
L-arabinose41000022
Galactose13 (2)34 (6)18 (4)2 (1)7 (3)222
Rhamnose0000016
Mannitol1501806122
Sorbitol30806030
N-acetyl-glucosamine1527 (2)18272 (1)0
Amygdalin030 (22)9 (7)0000
Arbutin001351800
Esculin035180400
Salicin0351802 (1)00
Cellobiose035180000
Starch9 (6)15 (8)9 (7)1 (1)2 (1)022 (20)
Gentiobiose035 (2)18 (1)0000
L-fucose0000024
Gluconate7 (7)34 (25)17 (9)2 (2)7 (5)1 (1)8 (3)
5-keto-glucon8 (8)24 (24)18 (18)2 (2)3 (3)3 (3)22

Number of isolates with positive results after 5 days' incubation; values in parentheses represent number of positive reactions that were delayed > 3 days.

One of the type species, A urea, did not produce discernible reactions, and the data for that strain were not included. All 9 other type species had positive results for the fermentation of glucose, fructose, and maltose and negative for the fermentation of erythritol, D-arabinose, L-xylose, adonitol, β-methyl-D-xyloside, sorbose, rhamnose, dulcitol, inositol, α-methyl-D-mannoside, α-methyl-D-glucoside, melezitose, xylitol, D-turanose, D-lyxose, D-tagatose, D-fucose, D-arabitol, L-arabitol, and 2-keto-gluconate. Variable reactions were obtained with the other 26 substrates (Table 2).

Table 2—

Fermentation reactions of 9 Actinobacillus type species for selected substrates.

Species*of Actinobacillus
Substrate123456789
Glycerol(+)++(+)+
L-arabinose++
Ribose+++++++
D-xylose+++++
Galactose++++++++
Mannose++++++++
Mannitol+++++++
Sorbitol++
N-acetyl-glucosamine++++++
Amygdalin++
Arbutin++++
Esculin++++
Salicin++++
Cellobiose+(+)++
Lactose+++++++
Melibiose++++++
Sucrose++++++++
Trehalose++++++
Inulin+
Raffinose++++++++
Starch(+)(+)+++
Glycogen(+)
Gentiobiose+++
L-fucose+
Gluconate(+)(+)+
5-keto-gluconate(+)(+)(+)(+)(+)

1 = Actinobacillus arthritidis. 2 = Actinobacillus capsulatus. 3 = A equuli subsp equuli. 4 = Actinobacillus hominis. 5 = Actinobacillus lignieresii. 6 = Actinobacillus muris. 7 = Actinobacillus seminis. 8 = Actinobacillus suis. 9 = Actinobacillus actinomycetemcomitans.

− = Negative results for fermentation. + = Positive results within 72 hours. (+) = Positive results at 72 to 120 hours.

Characterization of isolates via enzyme activities—Thirty-three field isolates (A equuli subsp equuli [n = 2]; A equuli subsp haemolyticus biovar 1 [10], biovar 2 [5], biovar 3 [2], and other biovar [3]; Actinobacillus genomospecies 1 [2]; and Bisgaard taxon 10–like [9]) and all 10 type strains were tested for enzymatic activities. With a few exceptions, all yielded negative results for lipase (C14), trypsin, α-chymotrypsin, N-acetyl-β-glucosaminidase, α-mannosidase, cystine arylamidase (weak activity by A muris), β-glucosidase (weak activity by A capsulatus and A muris), α-fucosidase (moderate activity by A muris), and β-glucuronidase (strong activity by A muris and A seminis).

All strains had alkaline and acid phosphatase, leucine arylamidase, and naphthol-AS-Bl-phosphohydrolase activities. All field strains and type strains except A seminis, A ureae, and A actinomycetemcomitans had moderately to strongly positive results for β-galactosidase; those 3 strains had no β-galactosidase activity. All 33 field strains had weakly or moderately positive results for esterase (C4) activity, 31 had weakly or moderately positive results for esterase (C8) lipase activity; 2 Bisgaard taxon 10–like isolates had negative results), and 30 had weakly or moderately positive results for valine arylamidase activity (2 A equuli subsp haemolyticus and 1 Bisgaard taxon 10–like isolate had negative results). Of the type strains, only A seminis had negative results for esterase (C4) activity, A seminis and A ureae had negative results for esterase lipase (C8) activity, and A arthritidis and A seminis had negative results for valine arylamidase activity.

Activities of α-galactosidase and α-glucosidase were variable for the field isolates and type strains. Activity of α-galactosidase was weak in 14 of the A equuli subsp haemolyticus isolates and absent in 4 of the biovar-1 isolates and 2 of the biovar-2 isolates. One A equuli subsp equuli isolate had moderate activity, and the other no activity. Of the 9 Bisgaard taxon 10–like isolates, 1 had strong α-galactosidase activity, 4 had moderate activity, and 4 had weak activity. The 2 Actinobacillus genomospecies-1 isolates had no α-galactosidase activity. Activity of α-glucosidase was strong in 15 A equuli subsp haemolyticus isolates, moderate in 1 of the biovar-1 isolates, weak in 2 of the biovar-1 isolates, and weak in 1 each of biovars 2 and 3. One subspecies equuli had strong activity, and the other had moderate activity. All 9 Bisgaard taxon 10–like isolates had strong α-glucosidase activity, and the 2 Actinobacillus genomospecies-1 isolates had no activity. The type strains of A actinomycetemcomitans, A lignieresii, A seminis, and A ureae had negative results for both enzymes. The A arthritidis isolate had weak α-galactosidase activity and no α-glucosidase activity, and the A hominis isolate had no α-galactosidase but strong α-glucosidase activity. The A capsulatus isolate had weak activity for both enzymes, the A equuli isolate had weak activity for α-galactosidase and strong activity for α-glucosidase, the A muris isolate had moderate activity for α-galactosidase and strong activity for α-glucosidase, and the A suis isolate had weak activity for α-galactosidase and moderate activity for α-glucosidase.

16S rRNA sequence data and analysis—The 16S rRNA genes from 15 isolates were sequenced (≥ 1,450 bp), and the sequences were compared with sequences in the GenBank database (last search, August 2004). Laboratory case numbers, GenBank accession numbers, and results of comparisons were summarized (Table 3). Isolate 24593-01 was phenotypically identified as A lignieresii but had a 99.1% sequence homology to strain F127, which was classified as Actinobacillus genomospecies 1.17 Ten isolates were phenotypically identified as A equuli subsp haemolyticus. Six of the 10 had 99.0% to 99.9% sequence homology to strain F154T, the proposed type strain for A equuli subsp haemolyticus, and the other 4 had 99.0% to 99.2% homology to strain NCTC 8529T (ATCC 19392T), which is the type strain for A equuli subsp equuli.18 Four isolates were most phenotypically similar to Bisgaard taxon 10,21 and the closest match in GenBank for all 4 isolates was also to Bisgaard taxon 10 (97.3% to 98.1% homology). Results of genotypic tests, in conjunction with phenotypic test results, separated the 127 isolates of actinobacilli into 4 major subgroups (Table 4).

Table 3—

Results of analysis of the 16S rRNA gene in 15 isolates of actinobacilli obtained from fetuses, or cardiac tissues, or cardiac fluid of horses affected by MRLS.

Table 3—
Table 4—

Summary of Actinobacillus spp isolated from fetuses of mares affected by MRLS that had late-term abortions from 2001 to 2003, horses with pericarditis, and fetuses of mares with experimentally induced MRLS.

Species of ActinobacillusLate-term abortionsHorses with pericarditisExperimental maresTotal
200120022003
A equuli subsp equuli10314018
A equuli subsp haemolyticus
Biovar 118 (3)*827 (3)035
Biovar 210 (1)204 (1)218
Biovar 3100102
Biovar not determined21 (2)011023
Bisgaard taxon 10-like22 (3)2 (1)21229
Actinobacillus
genomospecies 12 (1)00002
Total84156184127

Data are given as number of horses(number of Actinobacillus sppstrains). Strains were identified by 16S rRNA gene analysis.

Discussion

Actinobacilli were the most common bacteria recovered from horses with MRLS-associated pericarditis14 and the second most common bacteria recovered from aborted fetuses.13

In the present study, results of phenotypic and genotypic testing enabled allocation of the 127 isolates of actinobacilli into 4 major taxonomic groups: A equuli subsp haemolyticus (78 isolates), A equuli subsp equuli (18 isolates), Bisgaard taxon 10–like bacterium (29 isolates), and Actinobacillus genomospecies 1 (2 isolates).

Fifty-five of the 78 A equuli subsp haemolyticus isolates were further divided into 3 biovars on the basis of phenotypic results proposed by Christensen et al.18 The other 23 isolates were not assigned a biovar because the bacteria died before complete phenotypic analysis was performed (16 isolates) or were phenotypically different from described biovars (7 isolates). Certain strains of A equuli subsp haemolyticus have been reported15,22,23 as not fitting into these 3 biovars.

Three of the 78 A equuli subsp haemolyticus isolates were nonhemolytic on ovine blood agar and equine blood agar. However, on the basis of phenotypic tests, (ie, hydrolysis of esculin and production of acid in arbutin, cellobiose, esculin, gentiobiose, and salicin) those 3 isolates were considered to be A equuli subsp haemolyticus. The source of blood (ovine or equine) was important in assessment of hemolytic activity for subspecies haemolyticus strains. Ovine blood agar plates yielded a much higher percentage (96%) of isolates with hemolytic activity than did blood agar plates with added equine blood (47%). Moreover, the zone of hemolysis was always wider on the ovine blood agar plates. Descriptions of hemolytic patterns of Actinobacillus spp or any group of bacteria should take into account the source of blood used.

Ten of the A equuli subsp haemolyticus isolates were further characterized via 16S rRNA gene analysis. Six isolates were most closely related to the A equuli subsp haemolyticus type strain F154T (same as CCUG 19799T and NCTC 13195T), and 4 isolates were most closely related to the A equuli subsp equuli type strain ATCC 19392T (same as NCTC 8529T). Detection of strains that were more homologous to A equuli subsp equuli was not surprising because phylogenetic analysis on the basis of 16S rRNA gene analysis is of little value for resolving species within the genus.18 In the present study, the biovar of A equuli subsp haemolyticus was not directly related to the genotypic strain isolated. Four of the biovar-1 isolates were more closely related to the subspecies haemolyticus type strain, but 2 were more closely related to the subspecies equuli type strain.

A problem was encountered in the present study during comparison of the 6 isolates containing the 16S rRNA gene sequence most similar to type strain F154T (the type strain for A equuli subsp haemolyticus) with all strains of actinobacilli in GenBank. Five of the 6 isolates were more closely related (differing only in 2 to 11 bp) to GenBank accession No. AF145255, which was named A capsulatus.24 This strain was probably named on the basis of close homology to GenBank accession No. M75067, which was misnamed A capsulatus when its sequence was deposited.18 Type strains M75067 and AF247716 (same as F154T) represent identical deposits under different strain designations and represent the type strain for A equuli subsp haemolyticus.18 For identification of unknown bacteria by use of 16S rRNA gene sequence data, comparisons with GenBank strains should only be made with wellcharacterized type strains.

The phenotypic reactions for the 15 A equuli subsp equuli isolates were remarkably consistent and were nearly identical to type strain ATCC 19362T. Results for the tests commonly used for their identification were as follows: negative for hemolytic activity; esculin hydrolysis; and acid production in cellobiose, esculin, and salicin and positive for urease production and production of acid in glucose, mannitol, and trehalose. On the basis of fermentation of L-arabinose, 2 variants of subspecies equuli have been described.15,19 Both of those variants were isolated during the present study; 4 of the 15 isolates fermented L-arabinose. Fermentation of sorbitol by A equuli subsp equuli is variable, a finding in accordance with the present study, in which 3 of the 15 strains fermented sorbitol.

Both subspecies of A equuli have been associated with disease in horses.18 Subspecies haemolyticus represents a group of opportunistic pathogens that are associated with various disease conditions in foals and older horses, including stillborn fetuses, metritis, mastitis, septicemia, arthritis, endocarditis, meningitis, respiratory infections, and wound inflammation.18,25 Subspecies equuli is an opportunistic pathogen of horses and pigs but is most prominently associated with acute, highly fatal septicemia of newborn foals.18,25 Investigators in 2 extensive studies26,27 completed prior to the 2001 foaling season in central Kentucky reported that A equuli was a rare cause of abortion or stillbirths in mares. One of those studies26 was a 5-year study in which only 14 of 1,183 (1.2%) fetoplacental infections were caused by A equuli, and the other was a 2-year study27 in which it was reported that, of placentitis cases involving aborted fetuses or stillborn foals, only 2 of 236 (0.8%) cases were associated with A equuli. In contrast, during the MRLS outbreak in 2001, approximately 20% of the abortions and stillbirths involved actinobacilli.13 The 2 subspecies of A equuli are residents of mucous membranes of healthy horses and can be isolated from up to 100% of clinically normal horses as part of the normal resident bacterial flora of the oral cavity.15,19,22

Actinobacillus genomospecies 1 has only recently been described17 and revealed by means of 16S rRNA gene sequencing and DNA-binding studies to represent a different species from A lignieresii. Phenotypic characteristics that would be needed to separate this group from A lignieresii were not observed; therefore, no name was given for this taxon.16,17 In a study17 of 11 isolates, 6 isolates were obtained from the oral cavity of healthy horses, 2 were from the oral cavity of horses with stomatitis, and 3 were from human wounds caused by horse bites. In later studies28,29 in which 2 of those 3 human wound isolates were evaluated, the bacterium was identified as A lignieresii on the basis of phenotypic tests. Because all isolates pre-dating the present study originated from the oral cavity of horses,17 it is probable that Actinobacillus genomospecies 1 is a component of the normal flora of horses.

The remaining 29 isolates were classified as Bisgaard taxon 10–like bacterium. Phenotypically, the 29 isolates yielded reactions for more than 75 characteristics that were similar to those reported for Bisgaard taxon 10, strain CCUG 15572.21 Differences were obtained with tests for α-glucosidase activity (Bisgaard taxon 10 bacterium yielded a negative result; 8/8 Bisgaard taxon 10–like isolates had positive results) and with acid production from sorbitol metabolism (Bisgaard taxon 10 bacterium yielded a negative result; 29/29 Bisgaard taxon 10–like isolates produced acid) and D-xylose (Bisgaard taxon 10 bacterium was negative; 28/29 Bisgaard taxon 10–like isolates produced acid). Four of the Bisgaard taxon 10–like isolates were further characterized by 16S rRNA gene analysis and determined to be 97.3% to 98.1% similar to Bisgaard taxon 10, strain CCUG 15572. Similarities of 97% and higher indicate that organisms may be the same species. However, additional molecular characterization such as DNA-DNA reassociation, 23S rRNA gene sequencing, or gyrB gene sequencing should be completed before bacteria are deemed the same species.30,31 Because additional molecular characterization was not completed on these 29 isolates, the authors choose to use the designation Bisgaard taxon 10–like bacterium.

Little is known about the natural habitat for bacteria of Bisgaard taxon 10. The strain CCUG 15572 was isolated from the oral cavity of a horse, but no publications were found regarding the circumstances surrounding its isolation. The closely related bacteria Actinobacillus succinogenes and Bisgaard taxon 6 have been isolated from the rumen of healthy bovids and the pharynx, ear, and intestine of healthy guinea pigs, respectively.21,25 One of the authors (JMD) has occasionally isolated Bisgaard taxon 10–like bacteria from the oral cavity and various sites in the intestinal tract of horses. In those horses, recovery of the bacteria was not associated with a disease process and it was believed that they were a component of the horse's resident flora.

In conclusion, isolates of actinobacilli recovered from late-term aborted fetusus and stillborn foals of mares involved with MRLS were identical to strains of actinobacilli recovered from the alimentary tract of healthy horses. The role of the ETC and the mechanism by which it results in fetoplacental infection still requires elucidation.

ABBREVIATIONS

MRLS

Mare reproductive loss syndrome

ETC

Eastern tent caterpillar

ATCC

American Type Culture Collection

SIM

Sulfide-indole motility

NCTC

National Collection of Type Cultures

CCUG

Culture Collection, University of Goteborg

a.

Williams NM, Livestock Disease Diagnostic Center, University of Kentucky, Lexington, Ky: Personal communication, 2004.

b.

ATCC, Manassas, Va.

c.

API 50 CH, bioMerieux Inc, Hazelwood, Mo.

d.

API ZYM, bioMerieux Inc, Hazelwood, Mo.

e.

Ultraclean microbial DNA isolation kit, Mo Bio Laboratories, Solana Beach, Calif.

f.

MicroSeq 16S rDNA gene kit, Applied Biosystems, Foster City, Calif.

g.

Sequencher 4.0 software, Gene Codes Corp, Ann Arbor, Mich.

References

  • 1

    Mare Productive Loss Syndrome (MRLS). Available at: www.ca.uky.edu/gluck/mrls.index.aspAccessed July 5, 2004.

  • 2

    Harrison LR. Kentucky equine abortion storm and related conditions, in Proceedings. 105th Annu Meet US Anim Health Assoc 2001;227229.

    • Search Google Scholar
    • Export Citation
  • 3

    Dwyer RM. Epidemiological correlates of the 2001 and 2002 episodes of mare reproductive loss syndrome, in Proceedings. 1st Workshop Mare Reprod Loss Syndr 2003;3436.

    • Search Google Scholar
    • Export Citation
  • 4

    Williams NM, Donahue JM & Bolin DC, et al. Equine placental pathology: Kentucky perspective, in Proceedings. Workshop Equine Placenta 2004;8892.

  • 5

    Cohen ND, Donahue JG & Carey VJ, et al. Case-control study of early-term abortions (early fetal losses) associated with mare reproductive loss syndrome in central Kentucky. J Am Vet Med Assoc 2003;222:210217.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 6

    Dwyer RM, Garber LP & Traub-Dargatz JL, et al. Case-control study of factors associated with excessive proportions of early fetal losses associated with mare reproductive loss syndrome in central Kentucky during 2001. J Am Vet Med Assoc 2003;222:613619.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 7

    Seahorn JL, Slovis NM & Reimer JM, et al. Case-control study of factors associated with fibrinous pericarditis among horses in central Kentucky during spring 2001. J Am Vet Med Assoc 2003;223:832838.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 8

    McDowell KJ, Williams NM & Donahue JM, et al. Deductive investigations of the role of eastern tent caterpillars in mare reproductive loss syndrome, in Proceedings. Workshop Equine Placenta 2004;99103.

    • Search Google Scholar
    • Export Citation
  • 9

    Sebastian MM, Bernard WV & Harrison LR, et al. Experimental induction of mare reproductive loss syndrome with irradiated eastern tent caterpillars to assess whether the primary pathogen is a toxic molecule or a microorganism, in Proceedings. Workshop Equine Placenta 2004;103104.

    • Search Google Scholar
    • Export Citation
  • 10

    Webb BA, Barney WE & Dahlman DL, et al. Eastern tent caterpillars (Malacosoma americanum) cause mare reproductive loss syndrome. J Insect Physiol 2004;501:185193.

    • Search Google Scholar
    • Export Citation
  • 11

    Cohen ND, Carey VJ & Donahue JG, et al. Descriptive epidemiology of late-term abortions with the mare reproductive loss syndrome in central Kentucky. J Vet Diagn Invest 2003;15:295297.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 12

    Williams NM, Bolin DC & Donahue JM, et al. Gross and histopathological correlates of mare reproductive loss syndrome, in Proceedings. 1st Workshop Mare Reprod Loss Syndr 2003;2425.

    • Search Google Scholar
    • Export Citation
  • 13

    Donahue J, Sells S & Giles R, et al. Bacteria associated with mare reproductive loss syndrome: late fetal losses, in Proceedings. 1st Workshop Mare Reprod Loss Syndr 2003;2729.

    • Search Google Scholar
    • Export Citation
  • 14

    Bolin DC, Donahue JM & Vickers ML, et al. Microbiologic and pathologic findings in an epidemic of equine pericarditis. J Vet Diagn Invest 2005;17:3844.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 15

    Bisgaard M, Piechulla K & Ying YT, et al. Prevalence of organisms described as Actinobacillus suis or haemolytic Actinobacillus equuli in the oral cavity of horses. Comparative investigations of strains obtained and porcine strains of A. suis sensu stricto. Acta Pathol Microbiol Immunol Scand [B] 1984;92:291298.

    • Search Google Scholar
    • Export Citation
  • 16

    Christensen H, Bisgaard M. Revised definition of Actinobacillus sensu stricto isolated from animals. A review with special emphasis on diagnosis. Vet Microbiol 2004;99:1330.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 17

    Christensen H, Bisgaard M & Angen O, et al. Final classification of Bisgaard taxon 9 as Actinobacillus arthritidis sp. nov. and recognition of a novel genomospecies for equine strains of Actinobacillus lignieresii. Int J Syst Evol Microbiol 2002;52:12391246.

    • Search Google Scholar
    • Export Citation
  • 18

    Christensen H, Bisgaard M, Olsen JE. Reclassification of equine isolates previously reported as Actinobacillus equuli, variants of A. equuli, Actinobacillus suis or Bisgaard taxon 11 and proposal of A. equuli subsp. equuli subsp. nov. and A. equuli subsp. haemolyticus subsp. nov. Int J Syst Evol Microbiol 2002;52:15691576.

    • Search Google Scholar
    • Export Citation
  • 19

    Sternberg S. Isolation of Actinobacillus equuli from the oral cavity of healthy horses and comparison of isolates by restriction enzyme digestion and pulsed-field gel electrophoresis. Vet Microbiol 1998;59:147156.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 20

    Sternberg S, Brandstrom B. Biochemical fingerprinting and ribotyping of isolates of Actinobacillus equuli from healthy and diseased horses. Vet Microbiol 1999;66:5356.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 21

    Guettler MV, Rumler D, Jain MK. Actinobacillus succinogenes sp. nov., a novel succinic-acid-producing strain from the bovine rumen. Int J Syst Bacteriol 1999;49:207216.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 22

    Jang SS, Biberstein EL, Hirsh DC. Actinobacillus suis-like organisms in horses. Am J Vet Res 1987;48:10361038.

  • 23

    Samitz EM, Biberstein EL. Actinobacillus suis-like organisms and evidence of hemolytic strains of Actinobacillus lignieresii in horses. Am J Vet Res 1991;52:12451251.

    • Search Google Scholar
    • Export Citation
  • 24

    Drancourt M, Bollet C & Carlioz A, et al. 16S ribosomal DNA sequence analysis of a large collection of environmental and clinical unidentifiable bacterial isolates. J Clin Microbiol 2000;38:36233630.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 25

    Bisgaard M. Ecology and significance of Pasteurellaceae in animals. Zentrabl Bakteriol 1993;279:726.

  • 26

    Giles RC, Donahue JM & Hong CB, et al. Causes of abortion, stillbirth, and perinatal death in horses: 3,527 cases (1986–1991). J Am Vet Med Assoc 1993;203:11701175.

    • Search Google Scholar
    • Export Citation
  • 27

    Hong CB, Donahue JM & Giles RC, et al. Etiology and pathology of equine placentitis. J Vet Diagn Invest 1993;5:5663.

  • 28

    Dibb WL, Digranes A, Tonjum S. Actinobacillus lignieresii infection after a horse bite. Br Med J 1981;283:583584.

  • 29

    Peel MM, Hornidge KA & Luppino M, et al. Actinobacillus spp. and related bacteria in infected wounds of humans bitten by horses and sheep. J Clin Microbiol 1991;29:25352538.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 30

    Drancourt M, Raoult D. Sequence-based identification of new bacteria: a proposition for creation of an orphan bacterium repository. J Clin Microbiol 2005;43:43114315.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 31

    Stackebrandt E, Goebel BM. Taxonomic note: a place for DNA-DNA reassociation and 16S rRNA sequence analysis in the present species definition in bacteriology. Int J Syst Bacteriol 1994;44:846849.

    • Crossref
    • Search Google Scholar
    • Export Citation

Contributor Notes

Supported in part by the Grayson-Jockey Club Research Foundation Incorporated.

This report (No. 05-14-111) was published by permission of the Dean and Director of the Kentucky Agricultural Experiment Station and the College of Agriculture, University of Kentucky, Lexington, Ky.

Address correspondence to Dr. Donahue.