http://vdi.sagepub.com/Investigation

Journal of Veterinary Diagnostic

http://vdi.sagepub.com/content/11/6/553The online version of this article can be found at:

 DOI: 10.1177/104063879901100616

1999 11: 553J VET Diagn InvestDavid A. Bemis, Clark S. Patton and Edward C. Ramsay

Dermatophilosis in Captive Tortoises  

Published by:

http://www.sagepublications.com

On behalf of: 

  Official Publication of the American Association of Veterinary Laboratory Diagnosticians, Inc.

can be found at:Journal of Veterinary Diagnostic InvestigationAdditional services and information for    

  http://vdi.sagepub.com/cgi/alertsEmail Alerts:

 

http://vdi.sagepub.com/subscriptionsSubscriptions:  

http://www.sagepub.com/journalsReprints.navReprints:  

http://www.sagepub.com/journalsPermissions.navPermissions:  

http://vdi.sagepub.com/content/11/6/553.refs.htmlCitations:  

What is This? 

- Nov 1, 1999Version of Record >>

by guest on March 26, 2013vdi.sagepub.comDownloaded from

553Brief communications

ogenesis of canine parvovirus enteritis: sequential virus distri-bution and passive immunization. Vet Pathol 22:617–624.

12. Olsen CW: 1993, A review of feline infectious peritonitis virus:molecular biology, immunopathogenesis, clinical aspects, andvaccination. Vet Microbiol 36:1–37.

13. Takeuchi A, Binn LN, Jervis HR, Keenan KP: 1976, Electron

microscopic study of experimental enteric infection in neonataldogs with a canine coronavirus. Lab Invest 34:539–549.

14. Yasoshima A, Fujinami F, Doi K, et al.: 1983, Case report onmixed infection of canine parvovirus and canine coronavirus.Electron microscopy and recovery of canine coronavirus. Jpn JVet Sci 45:217–225.

J Vet Diagn Invest 11:553–557 (1999)

Dermatophilosis in captive tortoises

David A. Bemis, Clark S. Patton, Edward C. Ramsay

Dermatophilus infections have been observed in severallizard species, a boa constrictor (Constrictor constrictor),and an American alligator (Alligator mississippien-sis).1,3,5,6,10,11,13 Lesions in reptiles have been described as sur-face crusts comprised of necrotic cellular debris, keratin, andinflammatory cells, with necrosis of the epidermis or caseoussubcutaneous nodules. Dermatophilosis was diagnosed byisolation of D. congolensis in culture or identification of itsunique branching, filamentous morphology in lesions.

Until recently, D. congolensis was the only species as-signed to the genus Dermatophilus, and phenotypic proper-ties of the species were said to vary considerably.4 A newspecies designation, D. chelonae, was proposed for 3 highlydivergent isolates from chelonids in Australia.10 Chelonidisolates appear to be adapted to poikilotherms, growing bet-ter at lower temperatures than D. congolensis and havinglow infectivity for mammals.10 There is little informationavailable on disease caused by D. chelonae in reptiles andits recognition in the field. The purpose of this report is todescribe the isolation of D. chelonae-like bacteria from cu-taneous and visceral lesions in 2 species of captive tortoises.

Five bowsprit tortoises (Chersina angulata) and 3 Egyp-tian tortoises (Testudo kleinmanni) were housed in adjacentsections of an open tripartite wooden box. Another bowsprittortoise was housed separately in a glass terrarium. Sphag-num moss bedding was used. Room temperature was main-tained at approximately 30 C. A daily photoperiod of 12hours was maintained with a light source.a Fecal matter wasremoved regularly, and the boxes were emptied and washedwith detergent and half of the sphagnum moss was replacedmonthly.

The 5 bowsprit tortoises were wild-caught. They had beenhoused together since their arrival in 1991. The sixth bow-sprit tortoise had been housed with the group until Septem-ber 1995. The 3 Egyptian tortoises were of unknown origin.They were received from 2 separate donors in 1994 and werehoused in the box from September to November 1995. Sev-

From the Departments of Comparative Medicine (Bemis, Ram-say) and Pathology (Patton), College of Veterinary Medicine, Uni-versity of Tennessee, Knoxville TN 37901.

Received for publication August 26, 1998.

eral times during the fall and winter of 1995/1996, rain waterleaked into the box.

In November 1995, cutaneous lesions appeared simulta-neously in 3 bowsprit tortoises, including the one in the glassterrarium. Each tortoise had multiple skin lesions (.10), es-pecially on the neck and in deep recesses around the neckand legs (Fig 1). The lesions were yellow-white, 0.2–1.2-cm-diameter nodules covered with dry flaking skin with atract of yellowish material extending into the dermis. Onetortoise was euthanized because of the extensive skin nod-ules and ulcerative stomatitis. Another tortoise also had bi-lateral septic gonitis, and a fourth bowsprit developed skinlesions and bilateral septic arthritis 5 months later.

Skin and joint lesions were debrided and cleansed withdilute povidone–iodine solution. In addition, parenteral andoral antibiotics (ampicillin and amikacin or enrofloxacin, andmetronidazole) were administered. The skin lesions re-gressed in 3–8 months but recurred at different sites within12 months.

Two bowsprit tortoises and 1 Egyptian tortoise were founddead with no antemortem signs between January and April1996. Tissues from these and the euthanized bowsprit werefixed in 10% neutral buffered formalin, embedded in paraf-fin, sectioned at 4 mm, and stained with hematoxylin andeosin (HE), Gram’s, Grocott’s methanamine silver (GMS),and Fite–Faraco acid-fast procedures.

Skin biopsies from 2 bowsprit tortoises were examinedhistologically. Both samples had intracorneal accumulationsof heterophils with bacteria. Heterophils at the periphery hadundergone coagulative necrosis with uniform eosinophilicstaining of the nucleus and cytoplasm and loss of cytoplas-mic granularity. The necrotic heterophils formed poorly de-fined layers, divided by bands of keratinocytes (Fig. 2A).The epidermis at the base of the lesion was mildly acanthoticwith numerous transmigrating heterophils. The subjacentdermis contained increased fibroblasts, a moderate infiltrateof heterophils, and a few lymphocytes (Fig. 2B). The sub-cutis had a mild to moderate heterophilic infiltrate, perivas-cular lymphoid infiltrates, and myxomatous metaplasia. Or-ganisms were not visible with HE, but a Gram stain revealedgram-variable branching filamentous bacteria of varyingwidth within the exudate (Fig. 3). The bacteria were not acidfast with Fite-Faraco but stained with GMS.

by guest on March 26, 2013vdi.sagepub.comDownloaded from

554 Brief communications

Figure 1. Cutaneous heterophilic abscesses on the neck of abowsprit tortoise.

Figure 2. Bowsprit tortoise. A. intraepidermal heterophilic ab-scess with laminae due to squames and peripheral viable heterophils.HE. Bar 5 60 mm. B. Base of intraepidermal heterophilic abscess.Dermis with heterophil and mononuclear cells on right. HE. Bar 5300 mm.

The second biopsy was similar to the first with 2 excep-tions: 1) coccoid forms were also present and 2) bacteriawere not demonstrable in foci of orthokeratotic and para-keratotic hyperkeratosis.

Two bowsprit tortoises had granulomatous coelomitis andhepatic granulomas of unknown etiology. The Egyptian tor-toise had a 2.5-cm ceolomic caseous mass adhered to theplastron; the cut surface was granular, friable, and tan witha green center. Histologically, necrotic heterophils were sur-rounded by a thin zone of mononuclear cells and fewer het-erophils, all surrounded by a thin fibrous capsule. Gram-positive filaments and cocci (also visible with HE stains)were present at the periphery of the necrosis; divisions in 2planes were not seen.

Joint fluid, dermal samples, liver, and coelomic granulomawere submitted for aerobic and anaerobic bacterial culture.Samples for aerobic culture were inoculated onto 1) brain-heart infusion agarb supplemented with 1% yeast extract,b

1% heat inactivated horse serum,c and 5% defibrinated sheepbloodd (EBA plate), 2) MacConkey agarb (MAC), 3) eitherColistin-Nalidixic acid agarc (CNA) or phenylethyl alcoholagarb (PEA) supplemented with 5% defibrinated sheepblood, and 4) thioglycolate brothb supplemented with0.0005% hemin and 0.0001% vitamin K. EBA, PEA, andCNA plates were incubated at 35 C in a 7% CO2 atmosphere,MAC plates and thioglycolate broth were incubated at 35 Cunder atmospheric conditions. In addition, anaerobic cultureswere inoculated onto brain-heart infusion agar supplementedwith 0.5% yeast extract, 0.0005% hemin, 0.04% L-cystine,0.001% vitamin K, and 5% defibrinated sheep blood (CDCplate) and anaerobic PEA (APEA). Plates were incubated at35 C in an atmosphere of 5% hydrogen, 5% CO2, and 90%nitrogen.

Tests for catalase, urease, gelatin hydrolysis, and nitratereduction were performed using standard laboratory proce-dures. Reference strains of D. congolensis (ATCC 14634)and D. chelonae (ATCC 51576) were obtained from theAmerican Type Culture Collection.e

Paraffin-embedded specimens were processed for trans-mission electron microscopy (TEM) by removing the par-affin with xylene and alcohol, rinsing in cacodylate buffer,

and postfixing in osmium tetroxide. Epon-embedded sectionswere stained with uranyl acetate and lead citrate. For TEMof cultured organisms, a drop of medium was placed on aformvar-coated copper grid; organisms were negativelystained with potassium phosphotungstic acid and examinedwith a transmission electron microscope.f

A D. chelonae-like bacterium was isolated as the predom-inant organism, often in pure culture, from 8 chelonid spec-imens, including 6 from cutaneous lesions in 3 bowsprit tor-toises, 1 from the stifle joint of a bowsprit tortoise, and 1from the coelomic granuloma in an Egyptian tortoise. Small(,1 mm) white to off-white, dry adherent colonies withprominent hemolysis appeared on blood agar after 3 days ofeither aerobic or anaerobic incubation at 37 C. Eight- to 21-day-old cultures on blood agar were membranous and pro-duced exfoliative-like rugose folds, which created a rose budappearance of isolated colonies (Fig. 4).

Isolates were identified as gram-positive branching fila-ments; however, transition during growth from branching fil-aments to cocci was often abrupt (Fig. 5). Zoospores wereobserved by light microscopy in aqueous suspensions ofgrowth from blood agar plates after at least 7 days incuba-tion at room temperature. Flagellated zoospores were ob-served by TEM (Fig. 6). Elongate mulberry-like clusters ofcocci were abundant in thioglycolate broth cultures incubat-

by guest on March 26, 2013vdi.sagepub.comDownloaded from

555Brief communications

Figure 3. Branching, gram-positive filamentous bacteria withinan intraepidermal heterophilic abscess in a bowsprit tortoise. Gramstain. Bar 5 20 mm.

Figure 5. Dermatophilus chelonae-like isolate in a 7-day thio-glycolate broth culture. Morphology is variable. Dark to light con-trast reflects positive to negative Gram staining of the organism. Bar5 20 mm.

Figure 6. Electron micrograph of zoospore from chelonid Der-matophilus isolate from aqueous suspension of 14-day culture. Neg-ative stain, 0.5% potassium phosphotungstic acid, pH 7. Bar 5 1.0mm.

Figure 4. Colony morphology of a chelonid Dermatophilus iso-late after 7 days incubation on blood agar at 24 C. The confluentfirst quadrant of growth is membranous, with rugose folds (arrow).After 21 days of incubation, colonies have a rose-bud appearance(inset).

ed at room temperature for at least 3 days; however, defin-itive filaments dividing in 2 planes were rarely observed.Filaments containing longitudinal and transverse septa wereoccasionally found in paraffin-embedded tissue from thecoelomic granuloma processed for TEM (Fig. 7). The stockstrain of D. congolensis consistently produced intact, easilyrecognizable branching filaments with longitudinal andtransverse septa.

Biochemical and growth characteristics of the chelonidDermatophilus isolates were similar and most closely resem-bled D. chelonae (Table 1). Differential characteristics in-cluded better growth at 23 C than at 35 C, nitrate reduction,and failure to produce urease or carotenoid pigment.

Dermatophilosis is a widely recognized skin disease ofmammals caused by the actinomycete D. congolensis.Known by such colloquial names as lumpy wool, strawberryfoot rot, and rain scald, dermatophilosis affects a great va-riety of domestic and wild mammals and is economicallyimportant in cattle and sheep in high-rainfall areas.2,12 Mam-

malian epidermal dermatophilosis is associated with ortho-keratotic and parakeratotic hyperkeratosis with layers of neu-trophils and branching filaments with longitudinal and trans-verse divisions. No organisms were visible in HE-stainedsections of the epidermal lesions in the 2 bowsprit tortoises,but they were revealed with Gram’s and GMS stains.

by guest on March 26, 2013vdi.sagepub.comDownloaded from

556 Brief communications

Figure 7. Electron micrograph of chelonid Dermatophilus in ul-trathin section of coelomic granuloma. Cells that have divided inlongitudinal and transverse planes are seen within a fragmented bac-terial filament. Paraffin-embedded tissue processed for TEM. Leadcitrate and uranyl acetate. Bar 5 1.0 mm.

Table 1. Comparison of chelonid Dermatophilus isolates withD. congolensis and D. chelonae.

Characteristic

Chelonidisolates(n 5 8)

D.chelonae(ATCC51576)

D. con-golensis(ATCC14634)

Motile (flagellated zoospores pro-duced) 1 1 1

Branching gram-positive filamentwith longitudinal and transversesepta 1 1 1

Small, dry, adherent hemolyticcolonies 1 1 1

Anaerobic and aerobic growth 1 1 1Catalase produced 1 1 1Gelatin hydrolyzed 1 1 1Acid-fast stain 2 2 2Orange colonies 2 2 1Urease produced 2 2* 1Growth at 23 C faster than growth

at 35 C 1 1 2Nitrate reduced 1 1 2

* This result is in disagreement with that reported for the originalisolate; however, lack of urease was considered a distinctive char-acteristic of other chelonid isolates in the original report.

Skin lesions attributed to D. congolensis in several reptil-ian species differ from those presented here. Spontaneousepidermal lesions in 6 Senegal chameleons (Chamaeleo se-negalensis), a collared lizard (Crotaphytus collaris), and agreen iguana (Iguana iguana) consisted of hyperkeratosis,inflammation, and necrosis of the underlying epidermis.6 Theorganisms did not stain with HE, but the Giemsa stain re-vealed filaments with divisions in 2 planes. Nodules of epi-dermal hyperplasia with hyperkeratotic caps had heterophilsin the underlying epidermis, and organisms were visible withHE.11 Two marble lizards (Calotes mystaceous) had nodulesof hyperkeratosis without inflammation; typical filamentswere visible with HE.1 Cutaneous dermatophilosis reportedhere differs in some ways from these earlier descriptions.Necrotic heterophils were present within hyperkeratosis, andnumerous heterophils infiltrated the underlying parakeratoticepidermis. Hyperkeratosis without heterophilic infiltrationwas seen in one of 2 samples, but no bacteria were seen.Epidermal necrosis was not seen. Bacteria were not stainedwith HE; with Gram’s or GMS stains, divisions in 2 planeswere not seen. Cocci were prominent in 1 sample.

Subcutaneous caseous lesions due to D. congolensis havebeen reported in 3 reptilian species.5,6,13 Lesions in a beardedlizard (dragon) (Amphibolurus barbatus) consisted of eitheran encapsulated caseous abscess or encapsulated calcifica-tion; by inference, organisms did not stain with HE but wereseen as gram-positive branching filaments.13

The morphology of D. chelonae in tissue has not beendescribed previously, and the longitudinal septum formationis difficult to observe in cultures.10 The pleomorphic micro-scopic morphology of Dermatophilus may be misleading inlesions and cultures from chelonids.

Dermatophilus spp. have a morphologically complex lifecycle.12 In the present filaments are gram variable, divisionin 2 planes was difficult to observe, and cocci are abundant.Under some conditions, bacteria with gram-positive cellwalls may lose their ability to retain crystal violet dye in theGram’s staining procedure.8 Zoospores of D. chelonae aregram negative more frequently than the filamentous formsof the organism, perhaps because the zoospores are older,less actively dividing cells.

Dermatophilus chelonae-like bacteria were recovered in 6of 7 culture attempts from cutaneous lesions. Definitive clas-sification and more precise identification of D. chelonae-likebacteria requires further molecular characterization of theseisolates.

The distribution and mode of transmission of D. chelonaeis not known. The relatively abrupt onset of the present Der-matophilus infections in a restricted number of chelonid spe-cies with opportunities for close contact suggested that theremight have been a common source. Bowsprit and Egyptiantortoises are primarily dry habitat animals; prolonged ex-posure to an abnormally wet environment may have been apredisposing factor. The D. chelonae-like bacteria were notisolated from bedding material or from a sample of waterthat had leaked through the roof. Given the timing of infec-tion, the Egyptian tortoises could also have been the sourceof infection.

Dermatophilus chelonae infections most frequently in-volve the epidermis. Infection of a joint cavity and a coe-lomic mass may represent extension of infection from a bodysurface in a host with lowered resistance. Such a mechanismhas been proposed for the occasional extraepidermal infec-tions that occur in mammals.7,9 The effect of D. chelonaeinfection on overall reptile health deserves further study.

Acknowledgements. We thank Mary Jean Bryant, Mary

by guest on March 26, 2013vdi.sagepub.comDownloaded from

557Brief communications

Lambert, Jason Scott, Dee Stevenson, Dr. Jim Dugan, Dr.Jonathan Sleeman, Melanie Gregory, Bern Tryon, and KreisWeigel for technical assistance in the preparation of thismanuscript.

Sources and manufacturersa. VitaLite, Duro-Test Corp., New Bergen, NJ.b. Becton Dickinson Microbiology Systems, Cockeysville, MD.c. GIBCO/BRL, Grand Island, NY.d. Hemostat Laboratories, Dixon, CA.e. American Type Culture Collection, Rockville, MD.f. Phillips 300, Phillips Electronic Instruments, Mahwah, NJ.

References1. Anver MR, Park JS, Rush HG: 1976, Dermatophilus in the mar-

ble lizard (Calotes mystaceus). Lab Anim Sci 26:817–823.2. Biberstein EL: 1990, The skin as a microbial habitat: bacterial

skin infections. In: Review of veterinary microbiology, ed. Bi-berstein EL, Zee YC, pp. 268–270. Blackwell, Cambridge, MA.

3. Chineme CN, Addo PB: 1980, Pathologic changes in lizards(Agama agama) experimentally infected with Dermatophiluscongolensis. J Wild Dis 16:407–412.

4. Gordon MA: 1964, The genus Dermatophilus. J Bacteriol 88:509–522.

5. Jacobson ER: 1989, Dermatophilosis in reptiles. Int ColloqPathol Reptiles Amphib 3:47. [Abstr.]

6. Jacobson ER: 1991, Diseases of the integumentary system ofreptiles. In: Dermatology for the small animal practitioner, ex-otics, feline, canine, ed. Nesbitt GH, Ackerman LJ, pp. 225–239. Veterinary Learning Systems, Trenton, NJ.

7. Jones RT: 1976, Subcutaneous infection with Dermatophiluscongolensis in a cat. J Comp Pathol 86:415–421.

8. LaScola B, Raoult D: 1998. Molecular identification of Gemellaspecies from three patients with endocarditis. J Clin Microbiol36:866–871.

9. Lloyd DH: 1984, Immunology of Dermatophilus: recent devel-opments and prospects for control. Prev Vet Med 2:93–102.

10. Masters AM, Ellis TM, Carson JM, et al.: 1995, Dermatophiluschelonae sp. nov., isolated from chelonids in Australia. Int JSyst Bacteriol 45:50–56.

11. Montali RJ, Smith EE, Davenport M, et al.: 1975, Dermatophi-losis in Australian bearded lizards. J Am Vet Med Assoc 167:553–555.

12. Roberts DS: 1961, The life cycle of Dermatophilus dermaton-omus, the causal agent of ovine mycotic dermatitis. Aust J ExpBiol Med Sci 39:463–476.

13. Simmons GC, Sullivan ND, Green PE: 1972, Dermatophilus ina lizard (Amphibolurus barbatus). Aust Vet J 48:465–466.

J Vet Diagn Invest 11:557–560 (1999)

Prevalences of some virulence genes among Escherichia coli isolates from swinepresented to a diagnostic laboratory in Iowa

Harley W. Moon, Lorraine J. Hoffman, Nancy A. Cornick, Sheridan L. Booher, Brad T. Bosworth

Escherichia coli strains that carry genes encoding for spe-cific virulence attributes cause diarrhea and edema diseasein swine.1,2,8 Enterotoxigenic E. coli (ETEC) have genes forenterotoxins that stimulate secretion of electrolytes and wa-ter by the small intestine.6 To colonize the small intestineand cause diarrhea, ETEC must also produce fimbriae(pili).16 Escherechia coli strains that cause edema diseaseproduce E. coli Shiga toxin (Verotoxin) and are designatedas STEC.7 Shiga toxin is absorbed from the intestine intoblood and causes systemic vascular damage resulting in ede-ma disease. STEC must also produce fimbriae to colonizethe small intestine and cause disease.2 Some E. coli strainsare designated as attaching/effacing E. coli (AEEC) becauseof their ability to attach intimately to the surface of intestinalepithelial cells and efface microvilli.10 The attaching/effacing

From the Departments of Veterinary Pathology (Moon) and Vet-erinary Diagnostic and Production Animal Medicine (Hoffman) andthe Veterinary Medical Research Institute (Moon, Cornick, Booker),College of Veterinary Medicine, Iowa State University, Ames, IA50011, and the National Animal Disease Center, Agricultural Re-search Service, USDA, Ames, IA 50011 (Bosworth). Current ad-dress (Bosworth): Pig Improvement Company, Franklin, KY 42134.

Received for publication September 4, 1998.

attribute is encoded by a series of chromosomal genes lo-cated in a pathogenicity island called the locus of enterocyteeffacement. ETEC, STEC, and AEEC are considered to bedifferent pathotypes of E. coli. However, some of the viru-lence genes that characterize them can be located on mobilegenetic elements (plasmids, transposons, bacteriophages),and combinations of pathotypes occur. For example, someAEEC such as the human pathogen E. coli O157:H7 alsohave genes for Shiga toxin production,11,14 and some strainsassociated with edema disease of swine have genes for bothShiga toxin and enterotoxin production.2

The objectives of the work reported here were to deter-mine 1) the prevalences of ETEC, STEC, and AEEC amongswine E. coli isolates obtained at the Iowa State UniversityVeterinary Diagnostic Laboratory, 2) the comparative prev-alences of genes for different enterotoxin and pilus typesamong such isolates, and 3) whether there are differences inthe prevalences of toxin and fimbrial gene types isolatedfrom pigs in different age groups.

Escherichia coli isolates recovered from 539 swine fecalor tissue samples submitted to the Iowa State University Vet-erinary Diagnostic Laboratory from August 1996 throughDecember 1997 were analyzed. More than 95% of the spec-imens were obtained from swine herds in the midwestern

by guest on March 26, 2013vdi.sagepub.comDownloaded from