Subcutaneous botryomycosis due to Bibersteinia trehalosi in a Texas Longhorn steer.

A 3-year-old Texas Longhorn steer had a long history of progressive swelling of the soft tissues of the jaw and neck. At necropsy, multifocal to coalescing dermal and subcutaneous pyogranulomas were surrounded by fibrous tissue. Microscopically, the pyogranulomas contained aggregates of gram-negative coccobacilli surrounded by Splendore-Hoeppli material and were separated by bands of fibrovascular tissue (botryomycosis). Phylogenetic analysis of multilocus sequence-typing data revealed that the bacteria recovered in pure culture from swabs of submandibular tissue were most closely related to Bibersteinia [Pasteurella] trehalosi. The bacterial colonies were immunohistochemically reactive with a rabbit polyclonal anti-Pasteurella class C acid phosphatase antibody. Botryomycosis is a pyogranulomatous inflammation caused by a variety of nonbranching, nonfilamentous bacteria that elicit the formation of Splendore-Hoeppli material. This case of botryomycosis is unique for its association with Bibersteinia trehalosi.

Microbiology of the external ear canal in six African elephants (Loxodonta africana).

Samples collected from both external ear canals of six adult female African elephants (Loxodonta africana) were cultured for fungi, yeasts and aerobic and anaerobic bacteria. All the samples produced heavy growths of several aerobic bacteria, but anaerobic bacteria were rare and no fungi or yeasts were isolated. The most common bacterium isolated was Staphylococcus epidermidis, which was cultured from 11 of the 12 ears. Acinetobacter calcoaceticus lwoffi, alpha-haemolytic Streptococcus and Corynebacterium species, and Aeromonas caviae were all isolated from at least six of the 12 ears.

Characterization of multidrug resistant Salmonella recovered from diseased animals.

Three hundred and eighty Salmonella isolates recovered from animal diagnostic samples obtained from four state veterinary diagnostic laboratories (AZ, NC, MO, and TN) between 2002 and 2003 were tested for antimicrobial susceptibilities and further characterized for bla(CMY) beta-lactamase genes, class 1 integrons and genetic relatedness using PFGE. Forty-seven serovars were identified, the most common being S. Typhimurium (26%), S. Heidelberg (9%), S, Dublin (8%), S. Newport (8%), S. Derby (7%), and S. Choleraesuis (7%). Three hundred and thirteen (82%) isolates were resistant to at least one antimicrobial, and 265 (70%) to three or more antimicrobials. Resistance was most often observed to tetracycline (78%), followed by streptomycin (73%), sulfamethoxazole (68%), and ampicillin (54%), and to a lesser extent chloramphenicol (37%), kanamycin (37%), amoxicillin-clavulanic acid (20%), and ceftiofur (17%). With regards to animal of origin, swine Salmonella isolates displayed the highest rate of resistance, being resistant to at least one antimicrobial (92%), followed by those recovered from turkey (91%), cattle (77%), chicken (68%), and equine (20%). Serovars commonly showing multidrug resistance (MDR) to > or =9 antimicrobials were S. Uganda (100%), S. Agona (79%), and S. Newport (62%), compared to S. Heidelberg (11%) and S. Typhimurium (7%). Class-1 integrons were detected in 43% of all isolates, and were found to contain aadA, aadB, dhfr, cmlA and sat1 gene cassettes alone or in various combinations. All ceftiofur resistant isolates (n=66) carried the bla(CMY) beta-lactamase gene. A total of 230 PFGE patterns were generated among the 380 isolates tested using XbaI, indicating extensive genetic diversity across recovered Salmonella serovars, however, several MDR clones were repeatedly recovered from different diseased animals.

Serum levels of tumor necrosis factor-alpha in calves experimentally infected with Pasteurella haemolytica A1.

The purpose of this study was to document the levels of tumor necrosis factor-alpha (TNF) in serum of calves experimentally infected intratracheally with Pasteurella haemolytica A1 and to determine if elevated TNF levels correlate with development of pneumonic pasteurellosis in the bovine. Serum samples were collected at sequential time periods from 0 h to 72 h post inoculation with P. haemolytica. TNF levels in those sera were measured by a cytotoxicity assay utilizing the TNF-sensitive WEHI 164 mouse fibrosarcoma cell line. Serum TNF levels in infected cattle began to rise at 2 h post inoculation, peaked at approximately 8 h, and decreased to near control levels by 72 h. There was extreme variability in serum TNF among the inoculated animals with levels varying from 120 pg ml-1 to 5000 pg ml-1 at 8 h post inoculation. These levels did not correspond with the degree of lung involvement. All inoculated calves developed lesions of pneumonic pasteurellosis characterized by fibrinous pleuritis with necrotizing, hemorrhagic pneumonia. These results suggest that TNF is probably a significant inflammatory mediator involved in the pathogenesis of bovine pneumonic pasteurellosis.

Use of a Salmonella typhimurium-derived virulence probe in the detection of Salmonella sp. and in the characterization of S. cholerae-suis virulence plasmids.

A genetic probe encoding a virulence gene from Salmonella typhimurium was useful in the detection of Salmonella from feces during an outbreak of salmonellosis at a local dairy. A 3.2-kb BamHI restriction endonuclease fragment of the S. typhimurium virulence plasmid, pStSR100, has been useful as a DNA probe for both detection of Salmonella sp. and characterization of virulence plasmids from numerous field isolates. This virA probe hybridizes to a highly conserved gene carried on the large virulence plasmids of invasive Salmonella isolates. Colony blots prepared from feces directly plated onto MaConkey's agar failed to detect low numbers of Salmonella sp. However, hybridization of the VirA probe to vacuum blots or colony blots prepared from feces in tetrathionate enrichment broth incubated for 16 hours at 37 C was effective for detecting Salmonella sp. and resulted in an 85.9% correlation with culture results. The probe also demonstrated the highly conserved nature (96%) of the virulence gene among S. cholerae-suis isolate plasmids detected using Southern blot analysis.

Microdilution antimicrobial susceptibilities of selected gram-negative veterinary bacterial isolates.

Gram-negative bacterial isolates (635) obtained from routine submissions to the Oklahoma Animal Disease Diagnostic Laboratory during 1983-1987 were tested for antimicrobial susceptibility. Minimal inhibitory concentrations (MICs) were determined for the following antimicrobials using commercially prepared microdilution assay materials: ampicillin, cephalothin, chloramphenicol, erythromycin, gentamicin, kanamycin, oxytetracycline, penicillin G, spectinomycin, sulfachlorpyridazine, sulfadimethoxine, and tylosin. Results for isolates from cattle, dogs, horses, and pigs are presented. In only a few instances were differences in MICs apparent among bacterial isolates from different tissues. Aminocyclitol MICs for equine uterine isolates of Klebsiella pneumoniae differed from MICs for isolates from other tissues, and ampicillin, kanamycin, and spectinomycin MICs for bovine fecal isolates of Escherichia coli differed from MICs for isolates obtained from other tissues. In several instances, bimodal distribution of susceptibilities was apparent for ampicillin, kanamycin, and/or oxytetracycline. There was also a bimodal distribution pattern for erythromycin against Pasteurella haemolytica of bovine origin.

Evaluation of a commercial automated system and software for the identification of veterinary bacterial isolates.

A commercial gram-negative bacterial autoidentification plate was originally developed using bacterial isolates of human origin. Three veterinary diagnostic laboratories conducted a 2-phase trial to enhance the database for veterinary use. The first phase consisted of testing the plate with 447 bacterial isolates of veterinary origin and incorporating that data into the existing database. Emphasis was placed on the Actinobacillus, Bordetella, Pasteurella and Enterobacteriaceae groups, since the Pseudomonas taxon was quite complete. The second phase of the trial consisted of evaluating the enhanced database using 270 clinical veterinary isolates normally encountered in veterinary laboratories. For the Actinobacillus, Bordetella, Pasteurella and Enterobacteriaceae groups, 72% of the bacterial isolates were identified correctly to genus and 85% to species after 18 hours incubation. All identifications in phase 1 and phase 2 were confirmed using conventional methods.

Evaluation of laboratory tests for confirming the diagnosis of encephalitic listeriosis in ruminants.

Retrospective analysis of 93 bovine, ovine, and caprine cases diagnosed as listerial encephalitis revealed positive bacterial isolations in only 63% of 59 cases in which bacterial culture was attempted. Only 42% of 41 attempted bovine brain cultures were successful, compared with 67% from 6 sheep brains and 92% from 12 goat brains. Gram stains and Listeria-specific immunohistochemistry were evaluated as tools for verifying the presence of bacteria or listerial antigens in 38 animals. Sixteen of 17 animals in this group with positive bacterial isolations were immunochemically positive for listerial antigens (including 5/6 cattle), but Gram stains detected only 9/17 positive animals (including 1/6 cattle). Antigen was also detected in 15 of 21 animals (including 5/9 cattle) with unsuccessful or unattempted bacterial isolations. Of all 38 animals, the histologic diagnosis could be verified in 82% by immunohistochemistry, compared to 47% verified by Gram stains. Immunohistochemical testing was especially beneficial in locating antigen in lesions with few bacteria or bacterial antigens and is a rapid method of confirming the diagnosis of encephalitic listeriosis where inappropriate material is submitted for bacterial isolation or in culture-negative cases.

Pulmonary botryomycosis in a Scottish highland steer.

Pertinent necropsy findings in a 5 1/2-year-old Scottish Highland steer with chronic intractable pneumonia and cough were limited to the pulmonary system. The accessory lobe of the lung was collapsed, scarred, and multifocally adhered to parietal pleura. A polypoid mass almost completely obstructed the lobar bronchus and protruded into the trachea; mucopurulent exudate distended more distal bronchi. Botryomycosis was diagnosed when histologic examination revealed pyogranulomatous pneumonia with eosinophilic granules and "club" formation surrounding colonies of gram-positive cocci. Staphylococcus aureus was cultured from the lung. Botryomycosis is an unusual response to infection with common bacteria and is characterized by pyogranulomatous inflammation with formation of eosinophilic granules surrounding colonies of gram-positive cocci or gram-negative bacilli. Among domestic species, staphylococcal botryomycosis is most common as a wound infection in horses or as mastitis in cows and sows. Pulmonary botryomycosis is rare in horses, humans, and guinea pigs and apparently has not been reported in cattle.

Chlamydial infection in aborted and stillborn lambs.

Chlamydia psittaci is a major cause of ovine abortion in the fourth to fifth months of gestation. During the lambing seasons of 1986, 1987, and 1988, fetuses from 52 cases of ovine abortion, stillbirth, or perinatal death were submitted to the laboratory for necropsy examination. Placenta or fetal tissues from 34 cases were cultured on mouse L cells for C. psittaci. Chlamydia psittaci was identified by immunofluorescence on cultures in 20 of these cases. The major gross lesion consistently associated with chlamydial abortion was placentitis with multifocal cotyledonary necrosis and accumulation of red-brown exudate in the intercotyledonary placenta. Chlamydiae appeared as spherical organisms, less than 1 micron in diameter, in the cytoplasm of trophoblasts in impression smears of cotyledons. Histologically, placentitis was sometimes accompanied by pneumonia or encephalitis in the fetus. Chlamydia psittaci was considered the cause for fetal death when chlamydial isolation was associated with placentitis or inflammation of other fetal tissues and when other abortifacient agents were not detected.

Antimicrobial susceptibility and serotypes of Actinobacillus (Haemophilus) pleuropneumoniae recovered from Missouri swine.

The antimicrobial susceptibility of 73 Actinobacillus (Haemophilus) pleuropneumoniae isolates from swine in Missouri was determined with a microdilution minimal inhibitory concentration test system. Serotyping was accomplished by means of co-agglutination. Serotype 1 (39/73) and serotype 5 (30/73) were commonly found, whereas serotype 7 (4/73) was infrequently encountered. Most isolates (MIC90) were found susceptible to ampicillin (amoxicillin), cephalothin, penicillin, erythromycin, gentamicin, and kanamycin. Marked resistance was found with oxytetracycline, tylosin, and sulfadimethoxine. The data indicate that use of ampicillin (amoxicillin) or penicillin may correlate well with the favorable outcome of treatment.

Unique Salmonella choleraesuis surface protein affecting invasiveness. Possible inv related sequence.

TnphoA mutagenesis of a Salmonella choleraesuis isolate recovered from septicemic infection of feeder pigs resulted in 56 PhoA+ KnR StrR mutants. Thirty-five mutants exhibited reduced levels of invasion in the Hep-2 cell model and were examined by SDS-PAGE Western Blot analysis using an anti-alkaline phosphatase antibody to visualize the insertion gene products. A mutant which produced a gene fusion product of 95 kDa and exhibited > 90% reduction in invasion was subcloned. A 10 Kb BamHI fragment of the chromosome containing the phoA insert was detected by hybridization and cloned into a pGEM vector. The resulting 1657 base sequence contained a 1104 bp ORF with two short regions of homology with S. typhimurium invF and invG. one region of homology with lcrD of Yersinia pseudotuberculosis but contained largely unique sequences not contained in Gene Bank. The full length sequence was not obtained as there was no stop codon detected. The % G+C was 44%, considerably lower than that of the Salmonella chromosome, but compatible with the proposed Yersinia origin of the inv genes. The NH2 387 a.a. sequence includes 5 transmembrane regions, resembling the model derived from the hydrophobicity plot of S. typhimurium InvA.

The in vitro effects of EDTA-tris, EDTA-tris-lysozyme, and antimicrobial agents on equine genital isolants of Pseudomonas aeruginosa.

Five isolants of Pseudomonas aeruginosa collected from clinical cases of equine genital infection and one standard strain of P. aeruginosa were exposed to various concentrations of ethylene-diaminetetraacetic acid (EDTA) and tris (hydroxymethyl) aminomethane (tris buffer pH 8) and EDTA-tris lysozyme. Colony forming units of the isolants and minimal inhibitory concentrations for 11 antimicrobial agents were determined with each isolant before and after exposure to the EDTA solutions. Decreased cellular viability was found with all six isolants after exposure to the EDTA-tris solutions. Reversal of antimicrobial resistance was variable and unpredictable. These effects were not enhanced by the addition of lysozyme. The results suggest that EDTA-tris could be a useful adjunct in treating equine genital infections caused by Pseudomonas aeruginosa.

A prospective study of septicaemia in colostrum-deprived foals.

Fourteen mares and their foals were attended at parturition. After mare-foal bonding, 8 colostrum-deprived (CD) foals were removed from their dams, deprived of colostrum, and provided with an alternative milk source for the first 24 h of life. The mares were milked out every 2-4 h during this period to remove colostrum, after which the CD foals were returned to their mares and allowed to nurse. Six colostrum-fed (CF) foals were allowed to suck colostrum in the normal manner. Foal serum IgG concentration was determined by single radial immunodiffusion (means, CD = 0 mg/dl; CF = 1,508 mg/dl). Accepted methods were used to minimise infections in the neonatal foals. Of the 8 CD foals, 7 demonstrated clinical signs of sepsis. Septicaemia was confirmed in 5 of the 7 septicaemic CD foals by ante-mortem blood culture or by culture of tissue at necropsy. Organisms isolated included: Actinobacillus equuli, Escherichia coli, undifferentiated coliforms, Pseudomonas spp., and Actinomyces pyogenes. Clinically ill foals were treated with antimicrobial drugs, intravenous fluid therapy, flunixin meglumine, and anti-endotoxin hyperimmune serum. Three septicaemic CD foals survived. Four of 7 septicaemic CD foals died or were destroyed. Post-mortem lesions included bacterial embolic pneumonia, glomerulonephritis/nephritis, lymphoid depletion/atrophy, splenic and lymphoid necrosis, hepatitis, septic arthritis, and systemic bacterial embolism. None of the CF foals became septicaemic. One CF foal had foal heat diarrhoea and 1 CF foal had a serum IgG concentration of 160 mg/dl (i.e. failure of passive transfer), but both foals were otherwise normal.(ABSTRACT TRUNCATED AT 250 WORDS)

Occurrence of anaerobic bacteria in diseases of the dog and cat.

A survey for anaerobic bacteria was conducted in 314 clinical specimens from dogs and cats. A total of 187 anaerobic isolates in pure and mixed culture were isolated from 111 of the specimens that contained anaerobic bacteria. Common isolated included Actinomyces (9.1%), Clostridium perfringens (19.3%), other Clostridium spp (11.2%), Peptostreptococcus anaerobius (7.5%), Bacteroides melaninogenicus (13.4%), other Bacteroides spp (17.6%), and Fusobacterium necrophorum (5.3%). Anaerobic bacteria were involved in serious lesions that often were life threatening to the animals. Antibiotic susceptibility data indicated that the lincomycin family, the penicillin family, chloramphenicol, and cephaloridine are preferred drugs for treatment of anaerobic infections. Data from the survey were used in formulation of a table to aid practitioners in clinical diagnosis of disease caused by anaerobes. Clostridium perfringens was isolated in large numbers from five of six dogs with a clinical diagnosis of canine hemorrhagic gastroenteritis and from one cat with hemorrhagic diarrhea. Experimental infections were induced in rats, using caine feces as inoculum. Induced lesions contained aerobic and anaerobic bacteria similar to those bacteria isolated in the clinical survey, indicating that feces may serve as a major source of these bacteria in clinical infections of the dog.

Effects of Fusobacterium necrophorum leukotoxin on rabbit peritoneal macrophages in vitro.

A method to demonstrate leukotoxic activity of Fusobacterium necrophorum in vitro is described. Continuous dialysis sac culture system, using reduced liquid medium was used to grow F necrophorum for 7-day periods. The continuous culture dialysis filtrate contained leukotoxic substance(s) which appeared to be less than 10,000 in molecular weight, heat resistant, and stable at 4 C for at least 10 days. Leukotoxic activity was demonstrated in vitro by determining the percentage of macrophages taking up trypan blue dye after these were exposed to a 1:2 solution of continuous culture dialysis filtrate and Eagle's minimal essential medium at 1, 2, 4, and 6 hours. Approximately 90% of these macrophages were destroyed during the 6-hour incubation period.

Comparative in vitro leukotoxin production of three bovine strains of Fusobacterium necrophorum.

The in vitro leukotoxic activity of 3 bovine isolates of Fusobacterium necrophorum which varied in pathogenicity were compared. Monolayers of mouse peritoneal macrophages were exposed to culture filtrates from each F necrophorum strain, and cell viability was determined, using the trypan blue dye exclusion test. Two methods were used for production of the leukotoxin: (1) medium M-1 continuous dialysis sac cultures and (2) brain-heart infusion agar plate cultures. Supernatant cultural fluids containing the leukotoxin were subjected to membrane-partition chromatography, using ultrafilters with approximate molecular weight (mol wt) exclusion limits of 100,000, 10,000, 2,000, and 500. All ultrafiltrates had a cytotoxic effect on the monolayers. Cytotoxic activity was not found in the ultrafilter residues or in the control media ultrafiltrates. Comparative study of leukotoxin production indicated that F necrophorum 2101, type A, produced the most leukotoxin; F necrophorum 2030, type AB, produced slightly less leukotoxin; and F necrophorum 2035, type B, produced small amounts of leukotoxin. Endotoxin activity, as demonstrated by the mouse lethality test, was found in the residues of the XM-100A ultrafilter (100,000 mol wt), but not in the filtrates. Culture supernatant fluids and the XM-100A ultrafiltrates were positive for endotoxin, using the limulus amebocyte lysate assay; however, the other ultrafiltrates with lower mol wt exclusion limits were negative.

Endotoxin from Fusobacterium necrophorum of bovine hepatic abscess origin.

The endotoxic activity of Fusobacterium necrophorum bov 5 was investigated. The supernatant (S) fluid and cell wall (CW) preparation, obtained after differential centrifugation of the ruptured cell mass, were lethal for mice. The toxicity of the S fluid was stable during prolonged storage, treatment with formalin, and heating for 15 minutes at 80, 100, and 121 C, but was destroyed by alkaline hydrolysis with 0.25 N NaOH. The toxic factor was found in a high molecular weight (MW) fraction after gel filtration. The properties exhibited by the toxic S fluid resembled those of endotoxic lipopolysaccharide (LPS). Extracted and partially purified LPS (endotoxin) from F necrophorum bov 5 demonstrated a mouse median lethal dose (mouse LD50) of 16.8 mg/kg of body weight. The toxic LPS material, a high molecular weight moiety as estimated by gel filtration, was resistant to ribonuclease (RNase), deoxyribonuclease (DNase), and pronase treatment. A positive Shwartzman reaction (median skin lesion dose (SLD50) equal to 3.32 mug/kg of body weight) and biphasic fever response (minimal dose required to produce a fever index of 40 sq cm which falls on the linear portion of dose-response curve (FL40) equal to 0.41 mug/kg of body weight) further indicated the toxin was endotoxin in nature. The LPS from F necrophorum bov 5 was less toxic than Salmonella typhimurium LPS; but had considerable toxicity for experimental animals. The toxic activity of the partially purified F necrophorum bov 5 endotoxin was separated into 2 fraction regions by diethylaminoethyl (DEAE)-cellulose chromatography. The data provide evidence for the production of a potent endotoxin, possibly composed of more than one toxic component, which may be released upon cell disruption.

Clinical models for anaerobic bacterial infections in dogs and their use in testing the efficacy of clindamycin and lincomycin.

Two canine models of clinical anaerobic bacterial infections were developed for the study of the clinical parameters associated with these infections and for evaluation of antimicrobial agents that might be useful in therapy. In model I, a mixed culture of Bacteroides fragilis, B melaninogenicus, and Fusobacterium necrophorum was used as the inoculum. In model II, a mixed culture of B fragilis and Clostridium perfringens combined with an infection enhancer (sterile cinder dust) was used as the inoculum. In both models, reproducible localized pyogenic or gangrenous infections were induced. Clinical signs of fever, depression, and leukocytosis with a left shift were present. A depression in the nonspecific cell-mediated immune response was noticed. The 2 models were used to evaluate clindamycin and lincomycin for therapy of anaerobic bacterial infections, using a subjective scoring system for severity of lesions and general clinical appearance. Clindamycin at dosage levels of 5.5 mg or 11 mg/kg of body weight, twice a day was highly efficacious in the treatment of anaerobic bacterial infections in both models. Dogs given lincomycin (22 mg/kg twice a day) responded to the treatment, but the response was less than that seen with clindamycin.

An epizootic of hemorrhagic disease in white-tailed deer (Odocoileus virginianus) in Missouri: necropsy findings and population impact.

An epizootic occurred among white-tailed deer (Odocoileus virginianus) from July through October 1988 in Missouri (USA). From late July through September, nine necropsied deer had lesions of the peracute or acute forms of hemorrhagic disease (HD) or no apparent lesions, whereas two deer necropsied in October had lesions of the chronic form of HD. Epizootic hemorrhagic disease virus was isolated from two necropsied deer. Based on changes in population indices, there is evidence that deer populations declined in seven of Missouri's 57 deer management units from 1987 to 1990. Based on a deterministic model designed to simulate deer populations in management units, it appeared that summer and fall 1988 mortality ranging from 6% to 16% accounted for the population decreases in deer management units with population declines. Heavily hunted areas where high deer mortality was not reported in the summer and fall of 1988 did not have population declines. Based on these results, we believe that HD mortality was high and resulted in deer population declines in parts of Missouri when combined with hunting harvest.