Experimental Aerosol Inoculation of Mycobacterium bovis in North American Opossums (Didelphis...

JOURNAL OF WILDLIFE DISEASESThursday May 22 2003 03:26 PMAllen Press • DTPro System

jwdi 39_222 Mp_418File # 22em

418

Journal of Wildlife Diseases, 39(2), 2003, pp. 418–423q Wildlife Disease Association 2003

Experimental Aerosol Inoculation of Mycobacterium bovis in NorthAmerican Opossums (Didelphis virginiana)

Scott D. Fitzgerald,1,5 Laura S. Zwick,1 Kelly L. Diegel,1 Dale E. Berry,2 Steven V. Church,2 James G.Sikarskie,3 John B. Kaneene,1,4 and Willie M. Reed1 1 Diagnostic Center for Population and Animal Health andDepartment of Pathobiology and Diagnostic Investigation, College of Veterinary Medicine, Michigan State Uni-versity, P.O. Box 30076, Lansing, Michigan 48909, USA; 2 Michigan Department of Community Health Tuber-culosis Laboratory, Lansing, Michigan 48909, USA; 3 Department of Small Animal Clinical Sciences, College ofVeterinary Medicine, Michigan State University, East Lansing, Michigan 48824, USA; 4 Department of LargeAnimal Clinical Sciences, College of Veterinary Medicine, Michigan State University, East Lansing, Michigan48824, USA; 5 Corresponding author (email: [email protected])

ABSTRACT: The goal of this study was to eval-uate the susceptibility of North American opos-sums (Didelphis virginiana) to aerosol inocu-lation of Mycobacterium bovis at two dose lev-els in order to gain information on diseasepathogenesis, fecal shedding of the organism,and the potential role that opossums play in thespread of this disease in nature. Six opossumsreceived high dose (13107 colony forming units(cfu) by aerosol inoculation, six opossums re-ceived low dose (13103 cfu inoculation, and sixopossums were sham-inoculated with sterilewater and served as controls. Lungs were themost frequently infected tissues, with nine of12 inoculated opossums positive for M. bovison culture. Gross lesions consisted of multifocalpneumonia and enlarged lymph nodes. Micro-scopically, granulomatous pneumonia and gran-ulomatous lymphadenitis associated with acid-fast bacilli were present in eight of 12 inocu-lated opossums. Fecal shedding of M. bovis wasuncommon at both inoculation doses. Whileopossums were highly susceptible to aerosol in-oculation of M. bovis, they did not becomeemaciated or develop widely disseminated le-sions. From this study, opossums may transmittuberculosis by aerosol infection to other opos-sums in close contact and serve as a source ofinfection to carnivores that feed upon them,however, transmission of the disease to largeherbivores by fecal shedding or direct contactmay be less likely.

Key words: Aerosol inoculation, Didelphisvirginiana, Mycobacterium bovis, opossum, tu-berculosis.

Endemic tuberculosis caused by Myco-bacterium bovis involving both wildlifeand domestic cattle in Michigan (USA) hasbeen under active surveillance and studysince 1994 (Schmitt et al., 1997). Whilewhite-tailed deer (Odocoileus virginianus)appear to be the principal wildlife host, avariety of wild carnivores and omnivores,including opossums, are naturally infected

with M. bovis (Bruning-Fann et al., 2001;Fitzgerald et al., 2001). Because thebrushtail possum (Trichosurus vulpecula)has been identified as a critical wildlifereservoir of M. bovis in New Zealand, withthe ability to transmit the disease in theadvanced stages of infection to grazing cat-tle, additional studies in the North Amer-ican opossum (Didelphis virginiana) wereindicated (Paterson and Morris, 1995).American opossums are in the family Di-delphidae, while brushtail possums are inthe family Phalangeridae, making thesetwo species distant relatives in the sameorder (Marsupialia) (Nowak, 1991). How-ever, both opossums and possums sharesimilar biologic traits including nocturnalactivity and omnivorous dietary habits.Surveillance in Michigan has found natu-rally infected wild opossums with pulmo-nary lesions, and preliminary studies in ourlaboratory showed that opossums can beinfected with M. bovis by oral and intra-muscular inoculation (Diegel et al., 2002).This study utilized an aerosol inoculationmethod to more closely mimic naturaldroplet transmission and evaluated suscep-tibility to infection and fecal shedding attwo different inoculation doses.

Eighteen wild-caught opossums withnegative fecal cultures for Mycobacteriumspp. were divided into three experimentalgroups. Six opossums were dosed with13107 colony forming units (cfu) of M.bovis suspended in 2 ml of sterile water,six opossums were dosed with 13103 cfuM. bovis, and six were sham inoculatedwith 2 ml of sterile water. For inoculation,

SHORT COMMUNICATIONS 419



opossums were tranquilized with intra-muscular injection of 5 mg per kg tiletam-ine and zolazepam (Telazolt, Fort DodgeLaboratories, Fort Dodge, Iowa, USA;Stoskopf et al., 1999). Once tranquilized,opossums were placed in lateral recum-bency inside a biosafety cabinet, theirmuzzles inserted into an anesthesia cone,and they were exposed to aerosol inocu-lum for 10 min (5 min lying in recumben-cy on each side). The inoculum was deliv-ered through a commercial nebulizer(AeroVety Aerosol Drug Delivery System,AeroVet Aerosol Products, Ontario, Can-ada), which was powered by a standardvacuum pump applying 15 pounds persquare inch (PSI) room air.

Opossums were housed individually inHorsfal isolators with individual HEPA fil-ters. Opossums were weighed at 2 wk in-tervals during the 90 day study. Fecal cul-tures for Mycobacterium spp. were col-lected on days 1, 30, and 60 post-inocu-lation (PI). Two opossums from each ofthe three experimental groups were tran-quilized with Telazolt, then euthanized atdays 30, 60, and 90 PI. Information re-corded for each animal at necropsy includ-ed: total body weight and spleen, liver, andlung weights. All internal organs and sub-cutaneous lymph nodes were examinedgrossly during necropsy. The following tis-sues were collected in 10% buffered for-malin for histopathologic evaluation: brain,nasal turbinate, cranial lymph nodes in-cluding tonsil, trachea, lungs, thoraciclymph nodes, heart, liver, spleen, kidney,gonad, adrenal gland, pancreas, small andlarge intestine, and mesenteric lymphnodes. All tissues were routinely pro-cessed, paraffin-embedded, sectioned at 3mm and stained by both routine hematox-ylin and eosin stain, as well as by Ziehl-Neelsen acid-fast stain for microscopicevaluation. Fresh tissues were pooled formycobacterial isolation as follows: 1) cra-nial lymph nodes and tonsil, 2) thoraciclymph nodes and lung, 3) liver, spleen, andkidney, and 4) intestines and mesentericlymph nodes. Mycobacterial isolation and

identification techniques followed estab-lished protocols (Kent and Kubica, 1985;Butler et al., 1991; Reisner et al., 1994).

All opossums in the three experimentalgroups gained body weight over the courseof the study. All animals were active andexhibited good appetite throughout thestudy, and no animal developed externallesions associated with tuberculosis. Allopossums were negative for Mycobacteri-um spp. by fecal culture prior to inocula-tion. Fecal samples from all animals wereculture negative at day 1 and day 30 PI;only one low dose inoculated opossum waspositive for M. bovis by fecal culture onday 60 PI.

Typical gross lesions in inoculated opos-sums included multifocal granulomatouspneumonia (five of 12 inoculated opos-sums) and enlarged thoracic lymph nodes(seven of 12 inoculated opossums) (Fig. 1).While one animal in each of the high andlow dose groups exhibited splenomegaly,and one animal in each inoculated groupexhibited hepatomegaly, no statistically sig-nificant difference was found in meanspleen or liver weights between the threeexperimental groups. However, the meanlung weight was statistically significantlyhigher (P,0.05) in the high dose group(45.3617.1 gm) compared to both the lowdose group (29.9610.0 gm) and the sham-inoculated control group (24.065.4 gm).Statistical comparisons were made by Wil-coxon two-sample testing using SASt soft-ware (Statistical Analysis Systems, Release6.12, SAS Institute Inc., Cary, North Car-olina, USA).

Microscopically, five of 12 inoculatedopossums had granulomatous lymphade-nitis with associated acid-fast bacilli; fouropossums had lesions in their thoraciclymph nodes, and one opossum had a le-sion in its mesenteric lymph node (Fig. 2).Eight of 12 inoculated opossums had gran-ulomatous pneumonia associated withacid-fast bacilli typical of tuberculosis (Fig.3); two of these animals were in the lowdose group, while all six animals in thehigh dose group had this lesion. Granulo-

420 JOURNAL OF WILDLIFE DISEASES, VOL. 39, NO. 2, APRIL 2003



FIGURE 1. Lungs from a high dose M. bovis aerosol-inoculated opossum at 60 days post inoculation (PI).The lungs have dozens of scattered pale granulomas typical of pulmonary tuberculosis. Bar51 cm.

FIGURE 2. Photomicrograph of mesenteric lymph node from a high dose M. bovis aerosol-inoculatedopossum at 30 days PI. The central portion of the lymphoid follicle has been replaced by granulomatousinflammation composed of mixed infiltrates of lymphocytes and macrophages. H&E. Bar550 mm.




FIGURE 3. Photomicrograph of lung from a high dose M. bovis aerosol-inoculated opossum 90 days PI.The caseograuloma is composed of necrotic debris that is partially mineralized, and it is surrounded by mixedinfiltrates composed of lymphocytes, macrophages, and rare multi-nucleated giant cells. H&E. Bar550 mm.At high magnification, note moderate numbers of acid-fast bacilli within the necrotic debris. Ziehl-Neelsenacid-fast stain.

matous inflammatory infiltrates consistedof macrophages, lymphocytes, and neutro-phils, with rare multi-nucleated giant cellspresent in some cases. Non-tuberculosis-associated lesions present in all three ex-perimental groups included endogenouslipid pneumonia in seven opossums anddisseminated besnoitiosis in eight opos-sums.

Thoracic tissues were the most fre-quently infected with M. bovis. Nine of 12inoculated opossums (three of six low-doseanimals, six of six high-dose animals) hadM. bovis positive cultures from pooledsamples of lung and thoracic lymph nodes;all controls had negative thoracic tissuecultures. Only two inoculated opossumshad positive cranial lymph node pooledcultures, and all control opossum craniallymph node pools were culture negative.Two inoculated opossums had positive in-testinal and mesenteric lymph node poolcultures, one low dose and one high doseanimal, and all control opossum intestinalpools were culture negative. Finally, onlyone high dose opossum had a positive ab-

dominal non-intestinal (liver, spleen, andkidney) pooled culture, while all low doseand control animals were culture negative.

The aerosol route of inoculation washighly successful in producing infectedopossums, particularly in producing tho-racic lesions. Furthermore, there was a no-ticeable dose effect, in that low dose ani-mals had lower incidences of gross and mi-croscopic lesions, lighter mean lungweight, and lower incidence of M. bovispositive cultures compared to the highdose animals. The higher mean lungweight of high dose inoculated animalswas attributed to the marked granuloma-tous inflammatory infiltrate present in thisgroup of animals. In spite of highly suc-cessful infection by the aerosol route, theinoculated opossums did not exhibitweight loss. Nor did the inoculated opos-sums exhibit clinical signs such as inap-petance or depression. Finally, no mortal-ity occurred during the 90 day course ofthe study. This is in marked contrast to ex-perimental studies in brushtail possumswith high-dose (23105 cfu) inoculation

422 JOURNAL OF WILDLIFE DISEASES, VOL. 39, NO. 2, APRIL 2003



where high mortality rates occurred be-tween 25 and 100 days PI (Jackson et al.,1995a). Even in low-dose (125 cfu) inoc-ulation studies in brushtail possums, mostanimals exhibited inappetance, decreasedbody weight, and lethargy by 60 days PI(Buddle et al., 1994; Pfeiffer et al., 1994).Therefore, experimental aerosol inocula-tion of M. bovis in North American opos-sums would be a suitable model system toevaluate antibacterial agents for the treat-ment of tuberculosis, efficacy of tubercu-losis vaccines, and more extended durationexperiments to study long-term pathogen-esis, organism shedding, and immunolog-ical responses.

Only one of 12 inoculated opossums wasfound to be shedding M. bovis in its fecesat any PI culture period. Compared toprevious experimental studies in opossumsin which one of four oral-inoculated ani-mals and three of four intramuscular-in-oculated animals shed M. bovis in their fe-ces, there is a strikingly lower prevalenceof fecal shedding in aerosol inoculated an-imals (Diegel et al., 2002). It appears thataerosol infected opossums are significantlyless likely to shed M. bovis in their fecescompared to other exposure routes, or thata more advanced stage of the disease isrequired before fecal shedding develops.This aerosol inoculation model in opos-sums produced fecal culture results similarto those reported in naturally infectedbrushtail possums in New Zealand, wherepositive fecal cultures of M. bovis were un-common in cross-sectional sampling ofwild possums (Jackson et al., 1995a).

That aerosol inoculation tends to pro-duce lesions most frequently in the tho-racic viscera was not surprising. However,similar experimental studies using intra-tracheal inoculation in the brushtail pos-sum at lower inoculation doses resulted inmore widely disseminated lesions, signifi-cant losses in total body weight, and rapiddeterioration of clinical signs (Buddle etal., 1994; Pfeiffer et al., 1994). One pos-sible explanation for this disparity betweenstudy results is the fact that our aerosol

delivery system more closely resemblesnatural airborne infection, which allowsnormal host defense mechanisms, such asimpaction of organisms on nasal turbi-nates, and clearance by the mucociliaryapparatus in the airways. Many of the ear-lier experimental studies in brushtail pos-sum studies utilized intratracheal inocula-tion which bypassed much of the naturaldefense mechanisms, likely resulting in ahigher dose of infectious organisms in thelungs. The abdominal non-intestinal vis-cera pool (liver, spleen, and kidney) washarvested to evaluate hematogenousspread of M. bovis, because positive cul-tures would indicate generalized infection(Jackson et al., 1995b). In this study, onlyone high dose animal had evidence of gen-eralized infection, while generalized infec-tions involving these tissues are frequentlyreported in brushtail possums following ei-ther natural or experimental infection(Buddle et al., 1994; Cooke et al., 1995,Jackson et al., 1995b; Pfeiffer et al., 1994).Therefore, the North American opossumappears significantly less susceptible to M.bovis infection than is the brushtail pos-sum. The North American opossum tendsto primarily develop pulmonary tubercu-losis, while the brushtail possum developsrapidly progressive disseminated tubercu-losis.

In naturally infected brushtail possums,development of subcutaneous lymphade-nitis and subsequent draining fistuloustracts results in a much greater risk fortransmitting M. bovis to other animal spe-cies (Jackson et al., 1995a). The combina-tion of fistulous tracts, generalized debili-tation of infected possums, and the curiousnature of large ruminants has been docu-mented as a major factor in the interspe-cies transmission of M. bovis in New Zea-land (Patterson and Morris, 1995). How-ever, our study found neither debilitation,nor any gross involvement of subcutaneouslymph nodes or fistulous tracts in infectedopossums. These results suggest thatNorth American opossums are at signifi-cantly lower risk to serve as a source of




infection for cattle and cervidae than thebrushtail possums.

Opossums are still relatively susceptibleto M. bovis infection, whether they are ex-posed by aerosols, ingestion, or intramus-cular injection. This means that if tuber-culosis is introduced into these animals inthe wild, active infections can develop andaerosol spread might occur to other opos-sums, such as from a mother to her young.Other carnivores might be infected byfeeding on opossums. Our surveillance inMichigan has found several wild opossumswith natural-occurring infections. Ourstudies to date suggest that while theNorth American opossum may potentiallyserve as a reservoir host species for M.bovis it does not play a major role in trans-mission of the organism to deer and cattle.

This study was funded in part by a grantfrom the United States Department of Ag-riculture, Animal and Plant Health Inspec-tion Service, Veterinary Services, Tuber-culosis Cooperative Agreements.

LITERATURE CITED

BRUNING-FANN, C. S., S. M. SCHMITT, S. D. FITZ-

GERALD, J. S. FIERKE, P. D. FRIEDRICH, J. B. KA-

NEENE, K. A. CLARKE, K. L. BUTLER, J. B. PAY-

EUR, D. L. WHIPPLE, T. M. COOLEY, J. M. MILL-

ER, AND D. P. MUZZO. 2001. Bovine tuberculosisin free-ranging carnivores from Michigan. Jour-nal of Wildlife Diseases 37: 58–64.

BUDDLE, B. M., F. E. ALDWELL, A. PFEFFER, AND G.W. DE LISLE. 1994. Experimental Mycobacteri-um bovis infection in the brushtail possum (Tri-chosurus vulpecula): Pathology, haematology andlymphocyte stimulation responses. VeterinaryMicrobiology 38: 241–254.

BUTLER, W. R., K. C. JOST, JR., AND J. O. KILBURN.1991. Identification of mycobacteria by high per-formance liquid chromatography. Journal ofClinical Microbiology 29: 2468–2472.

COOKE, M. M., R. JACKSON, J. D. COLEMAN, AND M.R. ALLEY. 1995. Naturally occurring tuberculo-sis caused by Mycobacterium bovis in brushtailpossums (Trichosurus vulpecula): II. Pathology.New Zealand Veterinary Journal 43: 315–321.

DIEGEL, K. L., S. D. FITZGERALD, D. E. BERRY, S. V.CHURCH, W. M. REED, J. G. SIKARSKIE, AND J. B.KANEENE. 2002. Experimental inoculation ofNorth American opossums (Didelphis virgini-

ana) with Mycobacterium bovis. Journal of Wild-life Diseases 38: 275–281.

FITZGERALD, S. D., K. L. BUTLER, L. S. ZWICK, K. R.CLARKE, S. M. SCHMITT, D. J. O’BRIEN, J. S.FIERKE, C. S. BRUNING-FANN, J. B. KANEENE,AND W. M. REED. 2001. Are we winning thefight against bovine tuberculosis in Michigan’swildlife? Proceedings of the Xth InternationalSymposium of Veterinary Laboratory Diagnosti-cians, Salsomaggiore—Parma, Italy, pp. 53–54.

JACKSON, R., M. M. COOKE, J. D. COLEMAN, R. S.MORRIS, G. W. DE LISLE, AND G. F. YATES.1995a. Naturally occurring tuberculosis causedby Mycobacterium bovis in brushtail possums(Trichosurus vulpecula): III. Routes of infectionand excretion. New Zealand Veterinary Journal43: 322–327.

, , , AND . 1995b. Nat-urally occurring tuberculosis caused by Myco-bacterium bovis in brushtail possums (Trichosu-rus vulpecula): I. An epidemiological analysis oflesion distribution. New Zealand Veterinary Jour-nal 43: 306–314.

KENT, P. T., AND G. P. KUBICA. 1985. Public healthmycobacteriology. A guide for the level III lab-oratory. US Department of Health and HumanServices. Atlanta, Georgia.

NOWAK, R. M. 1991. Order Marsupialia. In Walker’smammals of the world, Vol. 1, 5th Edition, R. M.Nowak (ed.). The Johns Hopkins UniversityPress, Baltimore, Maryland, pp. 10–113.

PATERSON, B. M., AND R. S. MORRIS. 1995. Interac-tions between beef cattle and simulated tuber-culous possums on pasture. New Zealand Vet-erinary Journal 43: 289–293.

PFEIFFER, A., B. M. BUDDLE, AND F. E. ALDWELL.1994. Tuberculosis in the brushtail possum (Tri-chosurus vulpecula) after intratracheal inocula-tion with a low dose of Mycobacterium bovis.Journal of Comparative Pathology 111: 353–363.

REISNER, B. S., A. M. GATSON, AND G. L. WOODS.1994. Use of gen-probe accuprobes to identifyMycobacterium avium complex, Mycobacteriumtuberculosis complex, Mycobacterium kansasiiand Mycobacterium gordonae directly from Bac-tec TB broth cultures. Journal of Clinical Micro-biology 32: 2995–2998.

SCHMITT, S. M., S. D. FITZGERALD, T. M. COOLEY, C.S. BRUNING-FANN, L. SULLIVAN, D. E. BERRY, T.CARLSON, R. B. MINNIS, J. B. PAYEUR, AND J. SI-

KARSKIE. 1997. Bovine tuberculosis in free-rang-ing white-tailed deer from Michigan. Journal ofWildlife Diseases 33: 749–758.

STOSKOPF, M. K., R. E. MEYER, M. JONES, AND D. O.BAUMBARGER. 1999. Field immobilization andeuthanasia of American opossum. Journal ofWildlife Diseases 35: 145–149.

Accepted for publication 10 April 2002.

Experimental Aerosol Inoculation of Mycobacterium bovis in North American Opossums (Didelphis...

Documents

Transcript of Experimental Aerosol Inoculation of Mycobacterium bovis in North American Opossums (Didelphis...