Embed Size (px)
Citation preview
Vol. 46, N
o. 3, p
p. 687–1062
JOU
RN
AL
OF
WIL
DL
IFE
DIS
EA
SE
S
JULY
2010
CONTENTS
JOURNAL OF WILDLIFE DISEASES
VOL. 46, NO. 3 JULY 2010
(Continued on inside back cover)
BACTERIOLOGY AND MYCOLOGYBrucella species survey in polar bears (Ursus maritimus) of northern Alaska. TODD M. O’HARA, DARCE HOLCOMB, PHILIP ELZER, JESSICA ESTEPP,
QUINESHA PERRY, SUE HAGIUS, and CASSANDRA KIRK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 687Effects of mycoplasmal upper respiratory tract disease on morbidity and mortality of gopher tortoises in northern and central Florida. JOAN E. DIEMER BERISH,
LORI D. WENDLAND, RICHARD A. KILTIE, ELINA P. GARRISON, and CYNDI A. GATES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 695Transmission of Mannheimia haemolytica from domestic sheep (Ovis aries) to bighorn sheep (Ovis canadensis): Unequivocal demonstration with green fluorescent
protein-tagged organisms. PAULRAJ K. LAWRENCE, SUDARVILI SHANTHALINGAM, ROHANA P. DASSANAYAKE, RENUKA SUBRAMANIAM, CAROLINE N. HERNDON, DONALD P. KNOWLES, FRED R. RURANGIRWA, WILLIAM J. FOREYT, GARY WAYMAN, ANN MARIE MARCIEL, SARAH K. HIGHLANDER, and SUBRAMANIAM SRIKUMARAN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 706
Testing for Salmonella spp. in released parrots, wild parrots, and domestic fowl in lowland Peru. OSCAR BUTRON and DONALD J. BRIGHTSMITH . . . . . . . . . . 718CLINICAL PATHOLOGYBaseline normal values and phylogenetic class of the electrocardiogram of anesthetized free-ranging brown bears (Ursus arctos). A. RAE GANDOLF, ÅSA
FAHLMAN, JON M. ARNEMO, JAMES L. DOOLEY, and ROBERT HAMLIN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 724Reference intervals for plasma biochemical and hematologic measures in loggerhead sea turtles (Caretta caretta) from Moreton Bay, Australia. MARK
FLINT, JOHN M. MORTON, COLIN J. LIMPUS, JANET C. PATTERSON-KANE, and PAUL C. MILLS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 731EPIDEMIOLOGYSurvey for antibodies to infectious bursal disease virus serotype 2 in Wild Turkeys and Sandhill Cranes of Florida, USA. KRISTEN L. CANDELORA, MARILYN
G. SPALDING, AND HOLLY S. SELLERS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 742Bovine tuberculosis in Ethiopian wildlife. R. TSCHOPP, S. BERG, K. ARGAW, E. GADISA, M. HABTAMU, E. SCHELLING, D. YOUNG, A. ASEFFA, and J.
ZINSSTAG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 753Mortality during epizootics in bighorn sheep: Effects of initial population size and cause. IVONNE CASSAIGNE G., RODRIGO A. MEDELLÍN, and JOSÉ A.
GUASCO O. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 763Health assessment of American Oystercatchers (Haematopus palliatus palliatus) in Georgia and South Carolina. DAPHNE CARLSON-BREMER, TERRY M.
NORTON, KIRSTEN V. GILARDI, ELLEN S. DIERENFELD, BRAD WINN, FELICIA J. SANDERS, CAROLYN CRAY, MARCIE OLIVA, TAI C. CHEN, SAMANTHA E. GIBBS, MARIA S. SEPÚLVEDA, and CHRISTINE K. JOHNSON . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 772
Seasonal prevalence of serum antibodies to whole cell and recombinant antigens of Borrelia burgdorferi and Anaplasma phagocytophilum in white-tailed deer in Connecticut. LOUIS A. MAGNARELLI, SCOTT C. WILLIAMS, and EROL FIKRIG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 781
Real-time PCR detection of Campylobacter spp. in free-ranging mountain gorillas (Gorilla beringei beringei). CHRISTOPHER A. WHITTIER, MICHAEL R. CRANFIELD, and MICHAEL K. STOSKOPF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 791
EXPERIMENTAL DISEASEExperimental infection in lambs with a red deer (Cervus elaphus) isolate of Anaplasma phagocytophilum. SNORRE STUEN, WIEBKE SCHARF, SONJA
SCHAUER, FELIX FREYBURGER, KARIN BERGSTRÖM, and FRIEDERIKE D. VON LOEWENICH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 803The dusky-footed woodrat (Neotoma fuscipes) is susceptible to infection by Anaplasma phagocytophilum originating from woodrats, horses, and dogs. NATHAN
C. NIETO, JOHN E. MADIGAN, and JANET E. FOLEY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 810Field evaluation of an inactivated vaccine to control raccoon rabies in Ontario, Canada. K. G. SOBEY, R. ROSATTE, P. BACHMANN, T. BUCHANAN, L. BRUCE,
D. DONOVAN, L. BROWN, J. C. DAVIES, C. FEHLNER-GARDINER, and A. WANDELER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 818PARASITOLOGYIdentification of Euryhelmis costaricensis metacercariae in the skin of Tohoku hynobiid salamanders (Hynobius lichenatus), northeastern Honshu, Japan.
HIROSHI SATO, SADAO IHARA, OSAMU INABA, and YUMI UNE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 832Baylisascaris procyonis in raccoons in Texas and its relationship to habitat characteristics. AMY E. KRESTA, SCOTT E. HENKE, and DANNY B. PENCE . . . . . 843TOXICOLOGYLead toxicity in captive and wild Mallards (Anas platyrhynchos) in Spain. JUAN JOSÉ RODRÍGUEZ, PAULA A. OLIVEIRA, LUIS EUSEBIO FIDALGO, MÁRIO M.
D. GINJA, ANTÓNIO M. SILVESTRE, CESAR ORDOÑEZ, ALICIA ESTER SERANTES, JOSÉ MANUEL GONZALO-ORDEN, and MARÍA ASUNCIÓN ORDEN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 854
Survival of radio-marked Mallards in relation to management of avian botulism. DANIEL D. EVELSIZER, TRENT K. BOLLINGER, KEVIN W. DUFOUR, and ROBERT G. CLARK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 864
VIROLOGYSurveillance of avian influenza virus in wild bird fecal samples from South Korea, 2003-2008. H. M. KANG, O. M. JEONG, M. C. KIM, J. S. KWON, M. R. PAEK,
J. G. CHOI, E. K. LEE, Y. J. KIM, J. H. KWON, and Y. J. LEE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 878Occurrence of West Nile virus infection in raptors at the Salton Sea, California. ROBERT J. DUSEK, WILLIAM M. IKO, and ERIK K. HOFMEISTER . . . . . . . . . . . 889Prevalence of antibodies to type A influenza virus in wild avian species using two serologic assays. JUSTIN D. BROWN, M. PAGE LUTTRELL, ROY D.
BERGHAUS, WHITNEY KISTLER, SHAMUS P. KEELER, ANDREA HOWEY, BENJAMIN WILCOX, JEFFREY HALL, LARRY NILES, AMANDA DEY, GREGORY KNUTSEN, KRISTIN FRITZ, and DAVID E. STALLKNECHT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 896
SHORT COMMUNICATIONSAn outbreak of type C botulism in waterbirds: Incheon, Korea. NA-RI SHIN, SEONG HWAN BYUN, JEONG HOON CHUN, JEONG HWA SHIN, YUN
JEONG KIM, JEONG-HEE KIM, GI-EUN RHIE, HYEN MI CHUNG, IN-PIL MO, and CHEON-KWON YOO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 912Confirmation of Bacillus anthracis from flesh eating flies collected during a west Texas anthrax season. JASON K. BLACKBURN, ANDREW CURTIS,
TED L. HADFIELD, BOB O’SHEA, MARK A. MITCHELL, and MARTIN E. HUGH-JONES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 918Haloperidol and azaperone in drive-net captured southern chamois (Rupicapra pyrenaica). G. MENTABERRE, J.R. LÓPEZ-OLVERA, E. CASAS-DÍAZ,
I. MARCO and S. LAVÍN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 923Optimization of raccoon latrine surveys for quantifying exposure to Baylisascaris procyonis. TIMOTHY J. SMYSER, L. KRISTEN PAGE, and OLIN E.
RHODES, JR. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 929Detection of Hypoderma actaeon infestation in Cervus elaphus with ELISA and western blotting. JULIA DOMÍNGUEZ, ROSARIO PANADERO, and
CONCEPCIÓN DE LA FUENTE-LÓPEZ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 934Survey for foot-and-mouth disease in the endangered marsh deer (Blastocerus dichotomus) from marshlands of the Paraná River Basin, Brazil. JOÃO
PESSOA ARAÚJO JR., MÁRCIA F. NOGUEIRA, and JOSÉ M. B. DUARTE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 939Survey of Escherichia coli O157 in captive frogs. LUDOVICO DIPINETO, ANTONIO GARGIULO, TAMARA P. RUSSO, LUIGI M. DE LUCA BOSSA,
LUCA BORRELLI, DARIO D’OVIDIO, MARIANGELA SENSALE, LUCIA F. MENNA, and ALESSANDRO FIORETTI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 944Prevalence of selected vector-borne organisms and identification of Bartonella species DNA in North American river otters (Lontra canadensis). SATHYA K.
CHINNADURAI, ADAM J. BIRKENHEUER, HUNTER L. BLANTON, RICARDO G. MAGGI, NATALIA BELFIORE, HENRY S. MARR, EDWARD B. BREITSCHWERDT, and MICHAEL K. STOSKOPF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 947
Outbreak of botulism (Clostridium botulinum type C) in wild waterfowl: Seoul, Korea. GYE-HYEONG WOO, HA-YOUNG KIM, YOU-CHAN BAE, YOUNG HWA JEAN, SOON-SEEK YOON, EUN-JUNG BAK, EUI KYUNG HWANG, and YI-SEOK JOO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 951
First report of Hepatozoon sp. in the Oregon spotted frog, Rana pretiosa. PATRICIA L. STENBERG, and WILLIAM J. BOWERMAN . . . . . . . . . . . . . . . . . . . . . . . . . . . 956Bilateral complex microphthalmia with intraocular dermoid cyst in a neonate red deer (Cervus elaphus). DANIELA GELMETTI, IRENE BERTOLETTI,
and CHIARA GIUDICE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 961
JOURNAL OF
WILDLIFE DISEASESVOLUME 46 NUMBER 3 JULY 2010
REAL-TIME PCR DETECTION OF CAMPYLOBACTER SPP. IN FREE-
RANGING MOUNTAIN GORILLAS (GORILLA BERINGEI BERINGEI)
Christopher A. Whittier,1,4 Michael R. Cranfield,2,3 and Michael K. Stoskopf1
1 Environmental Medicine Consortium, Department of Clinical Sciences, College of Veterinary Medicine, North CarolinaState University, Raleigh, North Carolina 27606, USA2 Mountain Gorilla Veterinary Project, Inc., c/o Maryland Zoo in Baltimore, Druid Hill Park, Baltimore, Maryland 21217,USA3 Division of Comparative Medicine, School of Medicine, Johns Hopkins University, Baltimore, Maryland 21205, USA4 Corresponding author (email: [email protected])
ABSTRACT: Health monitoring of wildlife populations can greatly benefit from rapid, local,noninvasive molecular assays for pathogen detection. Fecal samples collected from free-livingVirunga mountain gorillas (Gorilla beringei beringei) between August 2002 and February 2003 weretested for Campylobacter spp. DNA using a portable, real-time polymerase chain reaction (PCR)instrument. A high prevalence of Campylobacter spp. was detected in both individually identified(22/26585%) and nest-collected samples (68/114559.6%), with no statistically significant differ-ences among different gorilla sexes or age classes or between tourist-visited versus research gorillagroups. The PCR instrument was able to discriminate two distinct groups of Campylobacter spp. inpositive gorilla samples based on the PCR product fluorescent-probe melting profiles. The rare type(6/90 positives, 7%, including three mixed cases) matched DNA sequences of Campylobacter jejuniand was significantly associated with abnormally soft stools. The more common type of positivegorilla samples (87/90 positives, 97%) were normally formed and contained a Campylobacter sp.with DNA matching no published sequences. We speculate that the high prevalence ofCampylobacter spp. detected in gorilla fecal samples in this survey mostly reflects previouslyuncharacterized and nonpathogenic intestinal flora. The real-time PCR assay was more sensitivethan bacterial culture with Campylobacter-specific media and commercially available, enzymeimmunoassay tests for detecting Campylobacter spp. in human samples.
Key words: Campylobacter, epidemiologic monitoring, gorilla, noninvasive sampling,polymerase chain reaction.
INTRODUCTION
Wild apes are most threatened byhabitat loss, poaching, and infectiousdiseases (Woodford et al., 2002). Alongwith emerging pathogens, such as Ebolavirus, the infectious disease threat com-prises a number of infectious agentspotentially transmitted from humans withwhom the wild apes interact (Wallis andLee, 1999; LeRoy et al., 2004; Kondgen etal., 2007). Investigation of infectiousagents sometimes requires invasively col-lected tissues (e.g., blood), but diagnosticadvances, sample availability, and a pref-erence for limiting wild ape disturbancehas increased the use noninvasively col-lected samples (e.g., feces) for diagnosticassays (Whittier et al., 1999; Goldberg etal., 2008; Jensen et al., 2009). Moleculardiagnostics have long been used onnoninvasively collected samples from hu-mans to diagnose infection with a variety
of pathogens, including many potentiallypathogenic to great apes (Lina et al., 1996;Houng et al., 1997). Wider implementa-tion of molecular diagnostic techniquescould facilitate free-ranging great apedisease research, especially using nonin-vasive samples coupled with durable,portable technology capable of providingrapid results in the field (McAvin et al.,2003; Tomlinson et al., 2005).
Mountain gorillas (Gorilla beringei ber-ingei) are considered at risk of a numberof infectious diseases and could benefitfrom more rapid diagnostics (MGVP Inc.and WCS, 2007). Actual and potentialdiagnostic samples, noninvasively collect-ed from mountain gorillas, include feces,urine, and partially eaten food items thatmay contain respiratory secretions andinfectious agents (Watts, 1984; Sleemanet al., 1988). Gastrointestinal parasiteshave been frequently studied in mountaingorillas because of the ease in collecting
Journal of Wildlife Diseases, 46(3), 2010, pp. 791–802# Wildlife Disease Association 2010
791
fecal samples and performing microscopicanalysis, whereas less is known aboutbacterial and viral infections (Nizeyi etal., 1999, 2001; Rwego et al., 2008).Molecular detection assays could not onlyexpand the use of noninvasively collectedsamples, but properly designed real-timepolymerase chain reaction (PCR) assayscould detect multiple organisms anddiscriminate between genetic subtypes oftarget organisms, improving the ability torapidly evaluate bacterial or viral diseases(Klaschik et al., 2004; Verweij et al., 2004).
Campylobacteriosis is one bacterial dis-ease of potential importance to wild gorillapopulations. Campylobacter is a gram-negative, microaerophilic bacteria that in-habits the gastrointestinal tract of manyanimals. The genus comprises pathogenicand nonpathogenic species as well as thosethat can be opportunistic pathogens. In thedeveloped world, virtually all campylobac-teriosis is attributed to one species, Cam-pylobacter jejuni, and is considered a food-borne illness commonly associated withanimal contamination (Blaser, 1997). Inthe developing world Campylobacter infec-tion has a different epidemiology thatincludes higher prevalence and incidenceof both symptomatic and asymptomaticinfections (Coker et al., 2002). Developingworld Campylobacter epidemiology also ischaracterized by a younger age-relatedattack-rate peak, widespread adult immu-nity to infection, and a higher proportion ofCampylobacter bacterial infections fromspecies other than C. jejuni, particularly inAfrica (Oberhelman and Taylor, 2000).Although human-human, and presumablyhuman-gorilla, transmission is believed tobe rare, Campylobacter spp. could serve asa useful model organism for molecularinvestigation in wild gorillas (Wassenaarand Newell, 2006). Previous studies indi-cate that Campylobacter spp. occurs withsufficient prevalence (19%) to warrantmolecular investigation (Nizeyi et al., 2001).
The objective of this project was toexpand molecular technologies to detectpotential pathogens in wild mountain
gorilla populations using an existing real-time PCR field-detection system withcommercially available freeze-dried assayreagents. Specifically we aimed to assessthe practicality of performing real-timePCR in the field in Africa and, secondly, toperform a basic epidemiology survey ofCampylobacter spp. in wild gorillas.
MATERIALS AND METHODS
Gorilla and wildlife sample collection
Fecal samples were collected noninvasivelyfrom human-habituated mountain gorillas inthe Parc National des Volcans in Rwanda(1u359S to 1u659S, 29u359E to 29u759E) be-tween August 2002 and February 2003. Ap-proximately 100 g of feces were collected intodisposable, sealable, polyethylene bags fromidentified gorillas immediately after observingdefecation (‘‘identified’’ samples) or from goril-la night nests (‘‘nest’’ samples). The finalepidemiologic data set included 140 gorillasamples (n5114 from nests and 26 fromidentified animals), although additional sam-ples were collected and tested for assayoptimization and melting-temperature analysesof PCR products. Duplicate sampling ofindividual gorillas was avoided by collectingonly once from a single nesting site (a group ofnests) for each gorilla group. No identifiedduplicates are included in the final 26 identifiedsamples, but because there could be duplica-tion of these individual gorillas with nestssamples, data are treated separately. Additionalwildlife samples collected from forest buffalo(Syncerus caffer, n54) and an unidentifiedsmall carnivore (n51) were incorporated intothis study to compare DNA sequences that arePCR-positive for Campylobacter spp.
For identified gorilla samples, individual ageand sex data were recorded from existingrecords and assigned to accepted age/sex classby standard definitions. Nest samples thatretained their shape (80/114, 70.2%) wereassigned to age classes based on measured,maximum lobe width, according to previouslypublished methods (McNeilage et al., 2001).Nest samples were further categorized assilverbacks (adult males .12 yr old), whenthey exceeded the maximum size for adultfemales (7.0 cm) and/or had silver/gray hairspresent, or as adult females, when nests alsocontained infant-sized feces (,3.0 cm) thatnormally occur in their shared nests. Forepidemiologic analysis, we used full age classesas well as collapsed classes that combined alladults (silverbacks, adult females, unknown
792 JOURNAL OF WILDLIFE DISEASES, VOL. 46, NO. 3, JULY 2010
adults, and blackbacks) and all nonadults(subadult, juvenile, and infant). Gorilla visita-tion type is well established for all thesehabituated groups, which at the time ofcollection were separated into those groupsvisited by tourists (‘‘tourist’’ groups, n54) andthose visited only by behavioral researchers(‘‘research’’ groups, n53).
Gorilla sample processing
Most gorilla samples (n593) were processedfor total DNA extraction within 12 hr ofcollection, whereas others (n515) were refrig-erated (2–4 C) or frozen (210 to 220 C) forprocessing within 7 days. A third group ofsamples (n532 gorillas and five other wildanimals) were used for DNA extraction afterstorage in guanidine isothiocyanate (GT) buffer(4 M guanidine isothiocyanate, 1 M sodiumcitrate, 0.7% b-mercaptoethanol, 10 mM eth-ylenediaminetetraacetic acid [EDTA], pH 7.2,Gibco BRL, Gaithersburg, Maryland, USA),which we have previously shown can be usedfor long-term storage at ambient temperature(Whittier et al., 1999). Total nucleic acids(DNA and RNA) were extracted using acommercial fecal DNA extraction kit (QIAampDNA Stool Mini Kit, QIAGEN Inc., Valencia,California, USA). Extracted DNA was eitherrefrigerated before PCR, for analysis within1 wk, or frozen, if analysis was expected tooccur after more than 1 wk of storage.
Human sample Campylobacter spp. culture andenzyme immunosorbent assay (EIA) testing
Fresh, human fecal samples were collectedvoluntarily from more than 120 local conser-vation personnel as part of the 2003 MountainGorilla Veterinary Project (MGVP) employeehealth program, described elsewhere (MGVP,2004). Eighteen of these samples were select-ed for inclusion in this study based on cultureresults for Campylobacter spp. as describedbelow (nine positive, nine negative).
Human samples were cultured for bacterialinfections in Rwanda within 2 hr of collectionand, subsequently, were frozen until EIAtesting and DNA extraction. Sample aliquotswere placed in Campylobacter thioglycollateenrichment medium (0.16% agar, trimetho-prim, vancomycin, polymyxin B, cephalothin,and amphotericin B; Remel Inc., Lenexa,Kansas, USA) and refrigerated at 2–4 C for24 hr. Enriched media were plated ontoCampylobacter-selective medium (blood agarwith same five antibiotics, Remel Inc.) andincubated in 5% CO2 at 42 C for 48 hr. Themicroaerophilic environment was achievedusing Pouch-MicroAero gas generators in the
AnaeroPouchTM System (Remel/MitsubishiGas Chemical America Inc., New York, NewYork, USA). Positive identification of Campylo-bacter spp. was based on colony morphology(small, gray to yellowish or pinkish gray, andslightly mucoid on selective media), character-istic bacterial morphology and gram-negativestaining (small, curved, or seagull-winged gram-negative rods), and biochemical characteriza-tion using APIH Strips (oxidase positive andcatalase positive; bioMerieux Clinical Diagnos-tics, Marcy l’Etoile, France). Frozen aliquotsfrom nine culture-positive and nine culture-negative human samples were later tested inthe United States using a commercial EIA forCampylobacter spp. according to the manufac-turer’s instructions (ProSpecTH CampylobacterMicroplate Assay, Alexon-Trend, Ramsey, Min-nesota, USA). After thawing, total DNA wasalso extracted for PCR according to the samemethods used for the gorilla samples.
Real-time PCR using the R.A.P.I.D.TM
DNA extracts were used for analysis in theR.A.P.I.D. instrument (Idaho Technology Inc.,Salt Lake City, Utah, USA) according tomanufacturer’s instructions. This unit is aportable, rapid, forced-air thermocycler withan integrated fluorometer for real-time mon-itoring of PCR reactions. It is designed forfield deployment and can be powered by anauto battery, but security and the manufac-turer’s agreement for the unit used in thisstudy restricted its use to a laboratory inRuhengeri, Rwanda. The proprietary freeze-dried Campylobacter spp., Salmonella spp.,Listeria spp., and Escherichia coli O157reagents integrate the reaction mix, primers,and hybridization probes. The standard PCRcycle for all assays was denaturation at 94 C for1 min, followed by 35 amplification cycles of95 C denaturation (held for 0 sec), andcombined annealing and extension at 60 Cfor 20 sec. Fluorescent hybridization-probemelting-curves ramping from 50 C to 80 Cwere standard after an initial denaturation at95 C (for 0 sec). Temperature transition rateswere 20 C/sec for all steps, except the meltingcurve that changed at 2 C/sec. Each PCRreaction contained 10–15% (e.g., 2–3 ml/20 ml)volume of DNA sample, and all cyclesincluded positive and negative kit controls.
The R.A.P.I.D. instrument software (Light-CyclerH Data Analysis module, version3.5.28.250, Idaho Technology) derives posi-tive/negative results based on a cutoff valuedifference between the scores of the flat phaseand the growth phase of each fluorescent PCRproduct curve. Visual inspection of the results
WHITTIER ET AL.—CAMPYLOBACTER SPP. IN MOUNTAIN GORILLAS 793
from initial assays revealed intermittent cyclicoscillations of the PCR product curves thatoften interfered with correct software inter-pretation and instrument results. In trouble-shooting this phenomenon, we ultimatelydiscovered that decreasing the reaction vol-ume eliminated oscillations in the positivecontrols and improved concordance betweenreplicate assays (data not shown). This led usto deviate from the reagent and instrumentinstructions by running later reactions at halfthe normal volume (10 ml) and to score allresults visually to detect false-positive andfalse-negative results. Additionally, questionsabout stability of the freeze-dried reagents inRwanda, where we found the laboratorytemperature ranged from 13 to 32 C and thehumidity from 29 to 94% lead us to re-evaluate and retest initial results after assayoptimization in the United States.
PCR product sequencing
To confirm their identity, 15 R.A.P.I.D.PCR reaction products for Campylobacterspp. (two controls and 13 samples) wereelectrophoresed in a 2% agarose gel (Gibco)containing 0.2 mg ethidium bromide (Gibco)/ml in TBE buffer (40 mM Tris, 20 mM aceticacid, 1 mM sodium EDTA; Gibco) at 75 or100 V for 30–60 min. Target DNA bands wereremoved and purified with a commercial kit(QIAquick Gel Extraction Kit, QIAGEN) and,because of the proprietary nature of theirproducts, submitted to Idaho Technology forsequencing. Bidirectional sequences for the 15products were generated by the University ofUtah, DNA Sequencing Core Facility (SaltLake City, Utah, USA).
To further identify Campylobacter spp. se-quences, original samples of DNA extracts wereused as PCR templates for new reactions in aconventional thermocycler (PTC-1160 MJ Re-search, Cambridge, Massachusetts, USA). Pub-lished Campylobacter genus level primers C412f(59-GGATGACACTTTTCGGAGC-39) andC1288r (59-CATTGTAGCACGTGTGTC-39;Linton et al., 1996) were used to generate an,726–base pair (bp) product. We were unableto optimize the PCR results for all samples, butsequenced products were derived from 25-mlreactions with 1 ml of DNA extract at either fullextract concentration, 1:10 dilution, or 1:100dilution, depending on the sample. In theoptimization effort, a number of different PCRpremixtures and individual components wereused, but the standard reaction varied around0.4 mM of each primer in commercial PCRreaction mixture for final concentrations of20 mM Tris-HCl (pH 8.4), 50 mM KCl,
1.5 mM MgCl2, 200 mM dNTPs, 20 U TaqDNA polymerase. The PCR amplification was36 cycles of 94 C for 1 min, 55 C for 1 min, and72 C for 1 min, followed by 72 C for 10 min andincluded positive and negative controls. ThePCR products were electrophoresed in a 2%agarose (Gibco) gel containing 0.2 mg ethidiumbromide/ml (Gibco) in TBE buffer (40 mM Tris,20 mM acetic acid, 1 mM sodium EDTA;Gibco) at 75 or 100 V for 30–60 min. Gels wererun with appropriate DNA ladders (Gibco) onbenchtop apparatuses (Bio-Rad LaboratoriesInc., Hercules, California, USA) with PolaroidH(Bedford, Massachusetts, USA) photographusing ultraviolet transilluminators (Fisher Scien-tific, Pittsburgh, Pennsylvania, USA; UVP Inc.,Upland, California, USA). Products were puri-fied as per the R.A.P.I.D. products describedabove and were sequenced at the DukeUniversity, DNA Analysis Facility (Durham,North Carolina, USA). Eleven samples (onepositive-seeded control, four human, and sixgorilla) yielded readable DNA sequences.
Statistical analysis
The unpaired t-test was used to comparemelting peak temperatures (Steel and Torrie,1980). Fisher’s exact test for association wasused for univariate analysis of different vari-ables in 232 contingency tables (Steel andTorrie, 1980). All calculations were performedeither with GraphPad (QuickCalcs, version2002–2005, GraphPad Software, Inc., La Jolla,California, USA) or MedCalc (version 11.1.0,2009, MedCalc Software, Mariakerke, Bel-gium) online software, and P,0.05 wasconsidered statistically significant.
RESULTS
EIA, culture, and PCR comparison ofhuman samples
Table 1 shows that the EIA test dis-agreed with one of the nine culture-negative samples and three of the nineculture-positive samples. Overall, therewas agreement between the culture andEIA in 14 of 18 samples. Results fromPCR were positive for most of the culture-negative (seven of nine) and culture-positive (eight of nine) samples.
Melting peak analysis
We found that the melting peaks ofpositive samples fell into one of twogroups: those that peaked at higher
794 JOURNAL OF WILDLIFE DISEASES, VOL. 46, NO. 3, JULY 2010
temperatures around 64.5 C, and thosethat peaked at lower temperatures around59 C (Fig. 1 and Table 2). All positivecontrols, the sample seeded with C. jejuni,and most of the positive human sampleproducts, melted at the higher tempera-ture, whereas most (358/384, 93.2%) ofpositive assays for gorillas melted at thelower temperature. Using the obviousseparation at 61–62 C, there were signif-icant differences between mean melting-peak temperatures for high (.62 C) andlow (,61 C) melting groups of gorillasamples, human samples, and all com-bined samples. No detectable temperaturedifferences were found between gorillaand human samples at either the high orlow melting temperatures (Table 2). Asmall number of samples (n53 gorillasand n52 humans) were found to haveboth low- and high-temperature meltingpeaks, either in single assays or inaggregates of multiple assays. The positivehuman samples showed no association
between melting peak temperatures anddetection by culture or EIA (Table 1).
Epidemiologic results
We detected no statistically significantdifferences in prevalence of Campylobac-ter spp. among different sexes, age classes,or tourist and research gorilla groups(Table 3). Specific gorilla family groupsranged in prevalence from 42% to 100%
(Table 3), but prevalence was not associat-ed with group size or geographic locationwithin the park (data not shown). Samplesfrom identified gorillas had a significantlyhigher Campylobacter spp. prevalence(85%) than those collected from nests(60%, Fisher’s exact test, P50.022).
Considering only the identified sam-ples, which were collected and physicallyevaluated more immediately after defeca-tion than nest samples, two of the threegorilla samples that were found to havethe high melting-peak Campylobacter spp.had abnormally soft fecal samples, where-as only one of the 19 with low melting-peak Campylobacter spp. had an abnormalstool. This resulted in a statisticallysignificant association between the highmelting-peak Campylobacter spp. andabnormal stools (P50.038, Fisher’s exacttest, relative risk 5 13 [95% confidence
TABLE 1. Comparison of bacterial culture, enzymeimmunosorbent assay (EIA), real-time polymerasechain reaction (PCR), and PCR-product melting-peak results for human samples.
Sample Culture EIA PCRPCR melting
peaka
Human 01 2 2 + Highb
Human 05 2 2 + MixedHuman 07 2 2 2
Human 08 2 2 + HighHuman 11 2 2 + HighHuman 15 2 2 + HighHuman 16 2 2 2
Human 18 2 + + Highb
Human 33 2 2 + HighHuman 12 + 2 + HighHuman 13 + + + MixedHuman 14 + 2 + HighHuman 17 + 2 + HighHuman 19 + + + Lowb
Human 20 + + + Highb
Human 21 + + 2
Human 22 + + + HighHuman 24 + + + High
a Low 5 melting peak ,62 C; high 5 melting peak .62 C;mixed 5 both high and low melting peaks.
b PCR products sequenced.
FIGURE 1. Spectrum of melting peak tempera-tures for R.A.P.I.D. (Idaho Technology Inc., SaltLake City, Utah, USA) replicates from different typesof specimens.
WHITTIER ET AL.—CAMPYLOBACTER SPP. IN MOUNTAIN GORILLAS 795
interval {CI}51.6–100, P50.016]; oddsratio 5 36 [95% CI51.6–826, P50.025]).
Sequencing results
The DNA sequences from 15 differentinitial R.A.P.I.D. PCR products corre-sponded with a 265-bp sequence of theCampylobacter 16s rRNA gene. In thishighly conserved region, the R.A.P.I.D. kitcontrol, the C. jejuni–seeded PCR prod-uct, the buffalo PCR products, and thehuman and gorilla high melting-peak PCRproducts all had sequences exactly identi-cal to published sequences for C. jejuni(Table 4). The low melting-peak humanand gorilla products were, however,unique sequences that most closely re-sembled (,2%-bp difference) a sequenceattributed to Campylobacter upsaliensis(LMG8853; Gorkiewicz et al., 2003). Thesequence from the low melting-pointhuman product differed from the threegorilla sequences, which were all identical.The PCR product from the unknowncarnivore had a relatively intermediatemelting peak and a DNA sequence onebase pair different from C. jejuni.
The DNA sequences from the largerstandard PCR products verified theR.A.P.I.D. product sequences for somesamples, but not others (Table 5). Theseeded control and one human sample(H18), which had been found to havehigh melting-peak (but unsequenced),R.A.P.I.D., Campylobacter spp. product,yielded standard PCR products with se-quences 100% identical to C. jejuni refer-ence sequences. Likewise the larger, stan-
dard PCR sequences from the one humansample (H19) and three gorilla samples (G1,G6, G7) with low melting-peak R.A.P.I.D.products were similar (3–5%-bp differenc-es) to C. upsaliensis. A human sample (H01)and three gorilla samples (G2, G4, G5),whose R.A.P.I.D. products were previouslysequenced, were found to have differentstandard PCR products, although two of thegorilla sequences were poor quality withmany indeterminate bases. Three of thesestandard PCR product sequences that didnot match their prior R.A.P.I.D. productsequences were either identical (H01) ormost similar (G2 and G4) to Campylobactergracilis. An additional human sample (H20)with a previously unsequenced high melt-ing-peak R.A.P.I.D. product was mostsimilar to Campylobacter hyointestinalis.
Other assays
Fifty gorilla samples (25 from thelargest tourist group and 25 from thelargest research group) were tested usingR.A.P.I.D. for Salmonella spp., Listeriaspp., and E. coli O157 using freeze-driedreagents from Idaho Technology’s Patho-gen Identification kits. The Salmonellaspp. assay was successfully pretested usinga human sample seeded with pure Salmo-nella enterica from culture. None of the 50gorilla samples tested positive for any ofthese agents.
DISCUSSION
The high prevalence of Campylobacterspp. detected in both gorilla and human
TABLE 2. Melting peak temperature analysis for R.A.P.I.D. (Idaho Technology Inc., Salt Lake City, Utah,USA) Campylobacter spp. polymerase chain reaction products from fecal and control samples showing twodistinct products. P-values based on two-tailed unpaired t-test.
Sample type
High melting peak (.62 C) Low melting peak (,62 C)
P-valueNo. results Mean (C) SD (C) No. results Mean (C) SD (C)
All samples 207 64.36 60.56 368 59.11 60.64 ,0.0001Positive controls 113 64.23 60.55 0Seeded samples 28 64.53 60.58 0Gorillas 26 64.64 60.41 358 59.12 60.65 ,0.0001Humans 36 64.41 60.58 10 58.82 ,0.0001
796 JOURNAL OF WILDLIFE DISEASES, VOL. 46, NO. 3, JULY 2010
TA
BL
E3.
Pre
vale
nce
of
Cam
pyl
obac
ter
spp
.id
en
tifi
ed
by
R.A
.P.I
.D.
(Id
aho
Tech
nolo
gy
Inc.
,S
alt
Lak
eC
ity,
Uta
h,
US
A)
poly
mera
sech
ain
reac
tion
(PC
R)
from
mou
nta
ingori
lla
feca
lsa
mp
les,
stra
tifi
ed
by
sam
ple
cate
gori
es.
Sam
ple
cate
go
ry
Ind
ivid
ual
lyid
en
tifi
ed
sam
ple
sN
est
coll
ect
ed
sam
ple
s
No
.sa
mp
led
No
.n
egat
ive
(%)
No
.p
osi
tive
(%)
No
.sa
mp
led
No
.n
egat
ive
(%)
No
.p
osi
tive
(%)
To
tal
po
siti
veL
ow
aH
igh
bM
ixed
cT
ota
lp
osi
tive
Lo
wa
Hig
hb
Mix
ed
c
Age
clas
s
Sil
verb
ack
71
(14)
6(8
6)
5(7
1)
1(1
4)
015
5(3
3)
10
(67)
10
(67)
00
Ad
ult
fem
ale
93
(33)
6(6
7)
5(5
6)
01
(11)
13
4(3
1)
9(6
9)
7(5
4)
1(8
)1
(8)
Un
kn
ow
nad
ult
d30
13
(43)
17
(57)
17
(57)
00
Bla
ckb
ack
30
3(1
00)
3(1
00)
00
32
(67)
1(3
3)
1(3
3)
00
Su
bad
ult
/Ju
ven
ile
40
4(1
00)
3(7
5)
1(2
5)
026
9(3
5)
17
(65)
16
(62)
01
(4)
Infa
nt
30
3(1
00)
3(1
00)
00
13
6(4
6)
7(5
4)
7(5
4)
00
Un
kn
ow
nsu
bad
ult
d14
7(5
0)
7(5
0)
7(5
0)
00
Sex M
ale
12
1(8
)11
(92)
9(7
5)
2(1
7)
018
7(3
9)
11
(61)
11
(61)
00
Fem
ale
13
3(2
3)
10
(77)
9(6
9)
01
(8)
13
4(3
1)
9(6
9)
7(5
4)
1(8
)1
(8)
Un
kn
ow
n1
01
(100)
1(1
00)
00
83
35
(42)
61
(73)
59
(71)
1(1
)1
(1)
Typ
eof
gro
up
Tou
rist
51
(20)
4(8
0)
2(4
0)
2(4
0)
053
16
(30)
37
(70)
37
(70)
00
Rese
arch
21
3(1
0)
18
(90)
17
(85)
01
(5)
61
30
(50)
31
(50)
28
(46)
1(2
)2
(3)
Sp
eci
fic
gro
up
Gro
up
116
1(6
)15
(94)
14
(88)
01
(6)
19
11
(58)
8(4
2)
6(3
2)
02
(11)
Gro
up
21
1(1
00)
00
00
27
15
(56)
12
(44)
11
(41)
1(4
)0
Gro
up
34
1(2
5)
3(7
5)
3(7
5)
00
15
4(2
7)
11
(73)
11
(73)
00
Gro
up
40
62
(33)
4(6
7)
4(6
7)
00
Gro
up
55
1(2
0)
4(8
0)
2(4
0)
2(4
0)
09
1(1
1)
8(8
8)
8(8
9)
00
Gro
up
60
90
9(1
00)
9(1
00)
00
Gro
up
70
29
13
(45)
16
(55)
16
(55)
00
All
sam
ple
s26
4(1
5)
22
(85)
19
(73)
2(8
)1
(4)
114
46
(40)
68
(60)
65
(57)
1(2
)2
(2)
aL
ow
5P
CR
pro
du
ctm
elt
ing
peak
,6
2C
.b
Hig
h5
PC
Rp
rod
uct
melt
ing
peak
.6
2C
.c
Mix
ed
5m
ult
iple
PC
Rp
rod
uct
melt
ing
peak
s.d
Un
kn
ow
nag
ecl
asse
so
nly
use
dfo
rn
est
-co
llect
ed
sam
ple
s.
WHITTIER ET AL.—CAMPYLOBACTER SPP. IN MOUNTAIN GORILLAS 797
samples in this survey was unexpected.Our overall survey result of 65–80%
prevalence (depending on collectionmethod) in gorillas is in contrast toCampylobacter spp. culture prevalencespreviously reported for wild mountaingorillas in a nearby park (19% [Kalema,1995] and 8% [Nizeyi et al., 2001]). TheR.A.P.I.D. PCR assay also detected ahigher number of positive human samplesthan either culture or EIA assays wereable to detect. Higher sensitivity ofmolecular, compared with conventional,diagnostics is well established (Munster etal., 2009), but the differences we foundcould also be a reflection of the specificityof the different methods. The cultureprotocol and media are fairly selectivefor C. jejuni and Campylobacter coli (andselective against some other species;Nachamkin et al., 2000), whereas theEIA assay detects a Campylobacter-spe-
cific antigen shared by C. jejuni and C.coli, but not by most other species. ThePCR assay is less specific, detectingCampylobacter at the genus level, and isknown to detect at least one species thatthe EIA will not (Campylobacter lari, permanufacturers’ inserts).
The high sensitivity and low specificityof a PCR assay like the one used canobscure a clinically useful result. TheR.A.P.I.D. instrument, however, is ableto quantify starting template DNA, whichcan correlate to clinical infections (Al-Robaiy et al., 2001), and to distinguishbetween different PCR products, based ontheir fluorescent probe melting patterns.This study demonstrated the use ofmelting-peak analysis in detecting anddiscriminating what appears be clinicallyrelevant C. jejuni (or C. jejuni–like)infections that were associated with softstools in wild gorillas. If we consider solely
TABLE 4. Sequence differences at variable positions among Campylobacter spp. 265–base pair R.A.P.I.D.(Idaho Technology Inc., Salt Lake City, Utah, USA) polymerase chain reaction PCR products amplified from15 different samples.a
Sequence identification Meltc
Base pair positionb
466 565 568 590 591 592 600 601 602 625 626 635
C. jejunib T A A A T G C A T G T AC. jejuni seed control 64.05d . . . . . . . . . . . .Positive kit control 63.98 . . . . . . . . . . . .Gorilla 3 64.93 . . . . . . . . . . . .Gorilla 4 64.28d . . . . . . . . . . . .Gorilla 5 64.40d . . . . . . . . . . . .Human 01 64.47 . . . . . . . . . . . .Buffalo 1 64.39 . . . . . . . . . . . .Buffalo 2 64.33 . . . . . . . . . . . .Buffalo 3 64.94 . . . . . . . . . . . .Buffalo 4 64.33 . . . . . . . . . . . .Unknown carnivore 62.18 A . . . . . . . . . . .C. upsaliensise . . . G G A T T C A . GGorilla 1 59.11d A C G G A A T T C . G GGorilla 2 59.02d A C G G A A T T C . G GGorilla 6 58.54d A C G G A A T T C . G GHuman 19 58.77 A . . G G A n T C n . n
a Underlined italics 5 reference sequences; a dot (.) 5 base identical to the C. jejuni reference; n 5 an indeterminatebase.
b Positions and sequence from Campylobacter jejuni strain RM1221 (GenBank accession CP000025 region 37396-38908).c Melting peak temperature (C).d Average of two reactions.e GenBank accession AF550641.1.
798 JOURNAL OF WILDLIFE DISEASES, VOL. 46, NO. 3, JULY 2010
the high melting peak and the mixedproducts compatible with a diagnosis of C.jejuni, the gorillas sampled in this studyhad an overall prevalence of only 3% fornest samples and 11% for identifiedsamples, which are similar to the preva-lences reported in previously publishedstudies.
Sample quality, as reflected by collec-tion method, affected the results in apredictable manner. Nest samples collect-ed up to many hours after defecation had alower proportion of positives than samplesfrom identified gorillas collected immedi-ately after defecation. One gorilla group(Group 1) was overrepresented in theidentified samples, which potentially con-founded results and inflated identified-sample prevalence. However, that samegroup had the lowest Campylobacter spp.prevalence in the more evenly sampled
nest results and, therefore, was unlikely tofalsely increase the identified-sampleprevalence. The likely explanation for thelower nest-sample prevalence is thatCampylobacter spp., a group of enteric,microaerophilic, and thermophilic bacte-ria, usually do not survive very longoutside of the gastrointestinal tract, aresusceptible to desiccation and light, andfail to grow below 30 C (Hazelberger etal., 1998). This fragility and potentialovergrowth by other organisms could limitdetection of diagnostic nucleic acids insamples left longer in the environment,such as the nest-collected samples.
In addition to the possibility of environ-mental and sample-handling factors re-sulting in undetected Campylobacter spp.infections, retention of PCR inhibitorsthat prevent product formation is also acommon cause of false-negative results for
TABLE 5. Sequence differences among Campylobacter 726–base pair, polymerase chain reaction productsamplified from different samples.
WHITTIER ET AL.—CAMPYLOBACTER SPP. IN MOUNTAIN GORILLAS 799
fecal DNA extracts (Wilson, 1997). Al-though removal of PCR inhibitors wasincorporated in the standard DNA extrac-tion methods, this survey found evidenceof sporadic PCR inhibition. Even with theoptimized assays, we found a high propor-tion (.20%, data not shown) of discor-dance between replicate pairs of assaysduring single-instrument runs. We alsofound one human sample (H21) that wasculture positive and EIA positive but PCRnegative. It is, therefore, possible thatmore gorillas shed Campylobacter spp.than even the 80% prevalence we detect-ed in fresh, identified samples.
We speculate that this high prevalencemostly represents normal gorilla intestinalflora and a not yet fully described Cam-pylobacter species. Most of the Campylo-bacter spp. we detected in wild gorillas wasthe low melting-peak type that generatedmultiple, consistent DNA sequences thatwere most similar, but not identical, to C.upsaliensis. One of the drawbacks of thissurvey was the collection and handling ofgorilla fecal samples for only DNA detec-tion with PCR. Although future research isplanned, the lack of preserved samples withpotentially viable bacteria prevented isola-tion and further description of the Cam-pylobacter spp. we detected.
The initial R.A.P.I.D. results revealedfew apparent coinfections (three of 90positives) with multiple Campylobacterspp. in the same gorilla sample. Thatfinding suggests that the common (lowmelting peak) gorilla Campylobacter sp.may not only be nonpathogenic but alsomight even be somewhat protective togorillas by competitively excluding theseemingly more pathogenic, high melting-peak C. jejuni or C. jejuni–like species,similar to what has been shown withCampylobacter spp. in chickens (El-Shi-biny et al., 2007). The DNA sequencing of726-bp, standard PCR product, however,revealed a more complicated distributionof Campylobacter spp. infections in bothgorillas and humans. These sequencesincluded two human isolates not originally
detected (C. gracilis and C. hyointestina-lis–like) as well as five samples that showedeither a different Campylobacter sp. thaninitially detected or evidence of a mixtureof multiple DNA sequences. Aside fromthe DNA of the C. jejuni, which we used asa PCR control, no other Campylobacterorganisms or DNA were used in ourlaboratory, making lab contamination withthese other species unlikely. We aim toclarify the nature and spectrum of theseorganisms with future studies isolating livebacteria in addition to detecting DNA.
Employing the R.A.P.I.D. instrumenthad some additional drawbacks. In addi-tion to the problematic PCR growth-curveoscillations briefly outlined in the meth-ods, we detected a seemingly related,lower-than-optimal PCR sensitivity usingthe standard R.A.P.I.D. protocol of 20-mlreactions, and also experienced an appar-ent pre-expiration degradation ofR.A.P.I.D. reagents in the field. Thereagent issue was likely due to widelyfluctuating climactic conditions in the fieldlaboratory and was remedied with quickeruse of additional reagent lots. The cycleoscillations and suboptimal detection rateswere remedied by using smaller reactionvolumes, which we can only speculate mayhave eliminated an unknown thermody-namic effect. Lastly, at the time of ourinvestigation, the military classification ofthe R.A.P.I.D. instrument inconvenientlyprevented our ability to fully test the unitin the forest in which it is designed andcapable of operating.
Overall, however, we achieved ourobjectives of showing that the R.A.P.I.D.instrument was practical to use in a remotefield laboratory and enabled us to completea survey for Campylobacter spp. in wildgorillas. Without this or similar technology,the apparently novel low melting-peakCampylobacter sp. we found may havegone undetected or been lumped togetherwith the another type or types. Thisdistinction could be particularly importantif the typical patterns of variable Campylo-bacter spp. pathogenicity exist in wild
800 JOURNAL OF WILDLIFE DISEASES, VOL. 46, NO. 3, JULY 2010
gorillas as this study has suggested. Theability shown here to rapidly detect andeasily distinguish between similar types oforganisms, in the field, using noninvasivelycollected samples, could greatly expanddiagnostic capabilities for many wildlifeprofessionals.
ACKNOWLEDGMENTS
Funding for this study was provided by TheMaryland Zoo in Baltimore, the MountainGorilla Veterinary Project Inc., the NorthCarolina State University College of Veteri-nary Medicine, and the Graduate Assistance inAreas of National Need Fellowship. We thankthe Rwandan National Parks and TourismOffice (ORTPN) and the Dian Fossey GorillaFund International for allowing us to completethis study. Recognition is given to E. Nyir-akaragire and the Karisoke Research Centerstaff for assisting with sample collection, J.-P.Lukusa for microbiology work, M. Correa forstatistical consultation, Idaho Technology fortechnical support, and F. Nutter for logisticsupport and editing.
LITERATURE CITED
AL-ROBAIY, S., S. RUPF, AND K. ESCHRICH. 2001.Rapid competitive PCR using melting curveanalysis for DNA quantification. Biotechniques31: 1382–1386.
BLASER, M. J. 1997. Epidemiologic and clinicalfeatures of Campylobacter jejuni infections.Journal of Infectious Diseases 176(Suppl 2):S103–S105.
COKER, A. O., R. D. ISOKPEHI, B. N. THOMAS, K. O.AMISU, AND C. L. OBI. 2002. Human campylo-bacteriosis in developing countries. EmergingInfectious Diseases 8: 237–243.
EL-SHIBINY, A., P. L. CONNERTON, AND I. F.CONNERTON. 2007. Campylobacter succession inbroiler chickens. Veterinary Microbiology 125:323–332.
GOLDBERG, T. L., T. R. GILLESPIE, I. B. RWEGO, E. L.ESTOFF, AND C. A. CHAPMAN. 2008. Forestfragmentation as cause of bacterial transmissionamong nonhuman primates, humans, and live-stock, Uganda. Emerging Infectious Diseases 14:1375–1382.
GORKIEWICZ, G., G. FEIERL, C. SCHOBER, F. DIEBER, J.KOFER, R. ZECHNER, AND E. L. ZECHNER. 2003.Species-specific identification of campylobactersby partial 16S rRNA gene sequencing. Journal ofClinical Microbiology 41: 2537–2546.
HAZELBERGER, W. C., J. A. WOUTERS, F. M. ROMB-
OUTS, AND T. ABEE. 1998. Physiological activity ofCampylobacter jejuni far below the minimal
growth temperature. Applied and Environmen-tal Microbiology 64: 3917–3922.
HOUNG, H. S. H., O. SETHABUTR, AND P. ECHEVERRIA.1997. A simple polymerase chain reactiontechnique to detect and differentiate Shigellaand enteroinvasive Escherichia coli in humanfeces. Diagnostic Microbiology and InfectiousDisease 28: 19–25.
JENSEN, S. A., R. MUNDRY, C. L. NUNN, C. BOESCH,AND F. H. LEENDERTZ. 2009. Non-invasive bodytemperature measurement of wild chimpanzeesusing fecal temperature decline. Journal ofWildlife Diseases 45: 542–546.
KALEMA, G. R. N. N. 1995. Epidemiology of theintestinal parasite burden of mountain gorillas inBwindi Impenetrable Forest National Park. BritishVeterinary Zoological Society Newsletter 14: 4–6.
KLASCHIK, S., L. E. LEHMANN, A. RAADTS, M. BOOK, J.GEBEL, A. HOEFT, AND F. STUBER. 2004. Detec-tion and differentiation of in vitro-spiked bacte-ria by real-time PCR and melting-curve analysis.Journal of Clinical Microbiology 42: 512–517.
KONDGEN, S., H. KUHL, P. K. N’GORAN, P. D. WALSH,S. SCHENK, N. ERNST, R. BIEK, P. FORMENTY, K.MATZ-RENSING, B. SCHWEIGER, S. JUNGLEN, H.ELLERBROK, A. NITSCHE, T. BRIESE, W. I. LIPKIN,G. PAULI, C. BOESCH, AND F. H. LEENDERTZ.2007. Pandemic human viruses cause decline ofendangered great apes. Current Biology 18:260–264.
LEROY, E. M., P. ROUQUET, P. FORMENTY, S.SOUQUIERE, A. KILBOURNE, J. M. FROMENT, M.BERMEJO, S. SMIT, W. KARESH, R. SWANEPOEL, S.R. ZAKI, AND P. E. ROLLIN. 2004. Multiple Ebolavirus transmission events and rapid decline ofcentral African wildlife. Science 303: 387–390.
LINA, B., M. VALETTE, S. FORAY, J. LUCIANE, J.STAGNARA, D. M. SEE, AND D. AYMARD. 1996.Surveillance of community-acquired viral infec-tions due to respiratory virus in Rhone-Alpes(France) during winter 1994 to 1995. Journal ofClinical Microbiology 34: 3007–3011.
MCAVIN, J. C., M. A. MCCONATHY, A. J. ROHRER, W.B. HUFF, W. J. BARNES, AND K. L. LOHMAN. 2003.A real-time fluorescence polymerase chainreaction assay for the identification of Yersiniapestis using a field-deployable thermocycler.Military Medicine 168: 852–855.
MCNEILAGE, A., A. J. PLUMPTRE, A. BROCK-DOYLE,AND A. VEDDER. 2001. Bwindi ImpenetrableNational Park, Uganda: Gorilla census 1997.Oryx 35: 39–47.
[MGVP] THE MOUNTAIN GORILLA VETERINARY PROJ-
ECT 2002 EMPLOYEE HEALTH GROUP. 2004. Riskof disease transmission between conservationpersonnel and the mountain gorillas: Resultsfrom an employee health program in Rwanda.EcoHealth 1: 351–361.
[MGVP INC. AND WCS] THE MOUNTAIN GORILLA
VETERINARY PROJECT INC., AND THE WILDLIFE
WHITTIER ET AL.—CAMPYLOBACTER SPP. IN MOUNTAIN GORILLAS 801
CONSERVATION SOCIETY. 2007. Conservation med-icine for gorilla conservation. In Conservation inthe 21st century: Gorillas as a case study, T. S.Stoinski, H. D. Steklis, and P. Mehlman (eds.).Springer-Verlag, New York, New York, pp. 57–78.
MUNSTER, V. J., C. BAAS, P. LEXMOND, T. M.BESTEBROER, J. GULDEMEESTER, W. E. P. BEYER,E. DE WIT, M. SCHUTTEN, G. F. RIMMELZWAAN, A.D. M. E. OSTERHAUS, AND R. A. M. FOUCHIER.2009. Practical considerations for high-through-put influenza A virus surveillance studies of wildbirds by use of molecular diagnostic tests.Journal of Clinical Microbiology 47: 666–673.
NACHAMKIN, I., J. ENGBERG, AND F. M. AEROSTRUP.2000. Diagnosis and microbial susceptibility ofCampylobacter species. In Campylobacter, 2ndEdition, I. Nachamkin and M. J. Blaser (eds.).American Society for Microbiology, Washington,D.C., pp. 45–66.
NIZEYI, J. B., R. MWEBE, A. NANTEZA, M. R.CRANFIELD, G. R. N. N. KALEMA, AND T. K.GRACZYK. 1999. Cryptosporidium sp. and Giar-dia sp. infections in mountain gorillas (Gorillagorilla beringei) of the Bwindi ImpenetrableNational Park, Uganda. Journal of Parasitology85: 1084–1088.
———, R. B. INNOCENT, J. ERUME, G. R. N. N.KALEMA, M. R. CRANFIELD, AND T. K. GRACZYK.2001. Campylobacteriosis, salmonellosis, andshigellosis in free-ranging human-habituatedmountain gorillas of Uganda. Journal of WildlifeDiseases 37: 239–244.
OBERHELMAN, R. A., AND D. N. TAYLOR. 2000.Campylobacter infections in developing coun-tries. In Campylobacter, 2nd Edition, I. Na-chamkin and M. J. Blaser (eds.). AmericanSociety for Microbiology, Washington, D.C.,pp. 139–153.
RWEGO, I. B., G. ISABIRYE-BASUTA, T. R. GILLESPIE,AND T. L. GOLDBERG. 2008. Gastrointestinalbacterial transmission among humans, mountaingorillas, and livestock in Bwindi ImpenetrableNational Park, Uganda. Conservation Biology22: 1600–1607.
SLEEMAN, J. M., AND A. B. MUDAKIKWA. 1998. Analysisof urine from free-ranging mountain gorillas
(Gorilla gorilla beringei) for normal physiologicvalues. Journal of Zoo and Wildlife Medicine 29:432–434.
STEEL, R. G. D., AND J. H. TORRIE. 1980. Principlesand procedures of statistics, a biometrical ap-proach, 2nd Edition. McGraw Hill, New York,New York, 633 pp.
TOMLINSON, J. A., N. BOONHAM, K. J. D. HUGHES, R.L. GRIFFIN, AND I. BARKER. 2005. On-site DNAextraction and real-time PCR for detection ofPhytophthora ramorum in the field. Applied andEnvironmental Microbiology 71: 6702–6710.
VERWEIJ, J. J., R. A. BLANGE, K. TEMPLETON, J.SCHINKEL, E. A. BRIENEN, M. A. VAN ROOYEN, L.VAN LIESHOUT, AND A. M. POLDERMAN. 2004.Simultaneous detection of Entamoeba histoly-tica, Giardia lamblia, and Cryptosporidiumparvum in fecal samples by using multiplexreal-time PCR. Journal of Clinical Microbiology42: 1220–1223.
WALLIS, J., AND D. R. LEE. 1999. Primate conservation:the prevention of disease transmission. Interna-tional Journal of Primatology 20: 803–826.
WASSENAAR, T. M., AND D. G. NEWELL. 2006. The genusCampylobacter. In The prokaryotes: A handbookon the biology of bacteria, Vol. 7, 3rd Edition, M.Dworkin, S. Falkow, E. Rosenberg, K. Schleiferand E. Stackebrandt (eds.). Springer-Verlag, NewYork, New York, pp. 119–138.
WATTS, D. 1984. Composition and variability ofmountain gorilla (Gorilla g. beringei) diet inthe central Virungas (central Africa). AmericanJournal of Primatology 7: 323–356.
WHITTIER, C. A., A. K. DHAR, C. STEM, J. GOODALL,AND A. A. ALCIVAR-WARREN. 1999. DNA extrac-tion from preserved primate fecal samples: Acomparison of five different methods. Biotech-nology Techniques 13: 771–779.
WILSON, I. G. 1997. Inhibition and facilitation ofnucleic acid amplification. Applied and Envi-ronmental Microbiology 63: 3741–3751.
WOODFORD, M. H., T. M. BUTYNSKI, AND W. B.KARESH. 2002. Habituating the apes: The diseaserisks. Oryx 36: 153–160.
Submitted for publication 5 December 2008.Accepted 25 February 2010.
802 JOURNAL OF WILDLIFE DISEASES, VOL. 46, NO. 3, JULY 2010
