- Email: [email protected]
Intradermal tuberculin testing of wild African lions (Panthera leo) naturally exposed to infection with Mycobacterium bovis
Veterinary Microbiology 144 (2010) 384–391
Contents lists available at ScienceDirect
Veterinary Microbiology journal homepage: www.elsevier.com/locate/vetmic
Intradermal tuberculin testing of wild African lions (Panthera leo) naturally exposed to infection with Mycobacterium bovis D.F. Keet a,b,*, A.L. Michel b,e, R.G. Bengis a, P. Becker d,f, D.S. van Dyk a, M. van Vuuren b, V.P.M.G. Rutten b,c, B.L. Penzhorn b a
Directorate of Veterinary Services, Kruger National Park, P.O. Box 12, Skukuza 1350, South Africa Department of Veterinary Tropical Diseases, Faculty of Veterinary Science, University of Pretoria, Private Bag X04, Onderstepoort 0110, South Africa Division of Immunology, Department of Infectious Diseases & Immunology, Faculty of Veterinary Medicine, Utrecht University, Yalelaan 1, 3584 CL Utrecht, The Netherlands d Biostatistics Unit, Medical Research Council, Private Bag X385, Pretoria, South Africa e Bacteriology Section, ARC-Onderstepoort Veterinary Institute, Private Bag X05, Onderstepoort 0110, South Africa f Division of Clinical Epidemiology, Faculty of Health Sciences, University of Pretoria, Private Bag X385, Pretoria, South Africa b c
A R T I C L E I N F O
A B S T R A C T
Article history: Received 13 August 2009 Received in revised form 4 January 2010 Accepted 27 January 2010
African lions in the southern half of Kruger National Park (KNP) are infected with Mycobacterium bovis. Historically, reliable detection of mycobacteriosis in lions was limited to necropsy and microbiological analysis of lesion material collected from emaciated and ailing or repeat-offender lions. We report on a method of cervical intradermal tuberculin testing of lions and its interpretation capable of identifying natural exposure to M. bovis. Infected lions (n = 52/95) were identified by detailed necropsy and mycobacterial culture. A large proportion of these confirmed infected lions (45/52) showed distinct responses to bovine tuberculin purified protein derivative (PPD) while responses to avian tuberculin PPD were variable and smaller. Confirmed uninfected lions from non-infected areas (n = 11) responded variably to avian tuberculin PPD only. Various non-tuberculous mycobacteria (NTM) were cultured from 45/95 lions examined, of which 21/45 were co-infected with M. bovis. Co-infection with M. bovis and NTM did not influence skin reactions to bovine tuberculin PPD. Avian tuberculin PPD skin reactions were larger in M. bovis-infected lions compared to uninfected ones. Since NTM coinfections are likely to influence the outcome of skin testing, stricter test interpretation criteria were applied. When test data of bovine tuberculin PPD tests were considered on their own, as for a single skin test, sensitivity increased (80.8–86.5%) but false positive rate for true negatives (18.75%) remained unchanged. Finally, the adapted skin test procedure was shown not to be impeded by persistent Feline Immunodeficiency VirusPle co-infection. ß 2010 Elsevier B.V. All rights reserved.
Keywords: Mycobacterium bovis Lion Panthera leo Intradermal tuberculin testing Co-infection Non-tuberculous mycobacteria Feline Immunodeficiency Virus Kruger National Park
1. Introduction Mycobacterium bovis has infected part of the African buffalo (Syncerus caffer) population in the Kruger National
* Corresponding author at: Directorate of Veterinary Services, Kruger National Park, P.O. Box 12, Skukuza 1350, South Africa. Tel.: +27 82 9279650; fax: +27 13 7356693. E-mail address: [email protected] (D.F. Keet). 0378-1135/$ – see front matter ß 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.vetmic.2010.01.028
Park (KNP) causing bovine tuberculosis (bovine TB) (Bengis et al., 1996; De Vos et al., 2001). Since the African buffalo is considered a preferential prey species, African lions (Panthera leo) residing in areas of KNP with a high prevalence of bovine TB in buffaloes acquired infection through consumption of tuberculous carcasses (Keet et al., 1996, 2000). We confirmed, by genetic typing, that spillover of M. bovis from buffaloes to lions occurred in the KNP (Michel et al., 2009). The lion index case in KNP was diagnosed through necropsy and culture in 1995 (Keet
D.F. Keet et al. / Veterinary Microbiology 144 (2010) 384–391
et al., 1996). Subsequently a number of emaciated and ailing lions were euthanized and it was established that multiple organ systems were affected, suggesting several possible routes of infection followed by haematogenous and/or lymphatic spread of mycobacteria (Keet et al., 2000). It is estimated that between 16,500 and 30,000 wild lions occur in Africa of which approximately 2200 occur in the Kruger Ecosystem (Bauer and Van Der Merwe, 2004). Since African lions are listed as Vulnerable on the IUCN Red Data List (2006), this precarious population status demanded a sensitive research approach regarding various infectious diseases especially bovine TB in lions (Cousins and Florisson, 2005). There are currently still no reliable immunodiagnostic tests for feline tuberculosis (Rhodes et al., 2008). Morris et al. (1996) referred to an ELISA test that was used on three zoo lions exposed to a lion with advanced pulmonary bovine TB. A similar MPB70 ELISA test suggested a seroprevalence of 4% in Serengeti lions (Cleaveland et al., 2005) but this result was not substantiated by mycobacterial culture of lion lesion material. The same ELISA test was used on 26 confirmed and highly developed M. bovispositive cases from KNP. Only 12/26 of these infected cases could be identified by this test (Unpublished data). Serologic assays for rapid detection of M. bovis infection have been described for a number of other free-ranging wildlife species (excluding lions) with promising results (Lyashchenko et al., 2008). Ante-mortal diagnosis of bovine TB in free-living wildlife is notoriously difficult, as animals need to be located and immobilized to collect blood for in vitro diagnostic assays (Grobler et al., 2002; Palmer and Waters, 2006). In the event of skin testing, free-ranging animals need to stay in captivity or be released, tracked and immobilized again after 72 h to assess skin response. Intradermal tuberculin testing is the cornerstone of successful tuberculosis-control schemes in cattle (de la Rua-Domenech et al., 2006) and it has been used on a number of wildlife species, with varying degrees of sensitivity and specificity (Cousins and Florisson, 2005). There are other limitations to the use of intradermal tuberculin tests in wildlife, since information is lacking concerning proper test sites for different species, concentrations, dosages and preparations of tuberculins to use and guidelines for interpretation (Miller, 2008). Regarding the use of tuberculin in carnivores, reports on the use of intradermal testing in domestic cats and dogs were not favourable because they do not react adequately to intradermally administered tuberculin and results were found to be unreliable (Backues, 2008; Miller, 2008). A large proportion of lions in KNP are serologically positive for FIVPle (Van Vuuren et al., 2003). Over and above the effect that infection with NTM may have on M. bovis specific immune responsiveness, hence on the results of skin testing in lions in the KNP, these test results may also be affected by co-infection with FIVPle. In domestic cats FIVFca critically impairs cell-mediated host responses and moderate to severe CD4+ depletion has been demonstrated in infected African lions (Roelke et al., 2006). A large proportion (53%) of bovine TB infected lions in areas of KNP where bovine TB is rife, are co-infected with FIVPle (Keet,
385
unpublished data). In cattle, co-infection with M. bovis and viruses that depress the function of lymphocytes and macrophages (such as bovine viral diarrhoea virus) has been shown to affect diagnostic assays for bovine TB (Charleston et al., 2001). Similarly diminished Mycobacterium tuberculosis-specific cell-mediated immune (CMI) responses have been documented in humans where HIV immune suppression was involved (Vermund and Yamamoto, 2007). In our study FIVPle status of animals was carefully considered as a potential confounding factor in bovine TB diagnosis by skin testing. The primary objective of this study was to determine whether the intradermal bovine tuberculin PPD test can identify lions naturally infected with M. bovis and which configuration of the intradermal skin test would achieve the highest sensitivity and specificity. Secondly the ability of NTM to evoke non-specific skin responses was evaluated. Concurrently, evaluation of possible attenuating effects of FIVPle on skin test response was required. 2. Materials and methods 2.1. Study population A cohort of M. bovis-infected lions (n = 52) was identified from a group of emaciated and ailing (n = 61/84), or repeatoffender lions (n = 23/84) reported on an ad hoc basis by field staff and tourists (a repeat-offender lion being a recognized habitual cattle killer and/or an emigrating vagrant). Cases were collected from anywhere within and immediately around the KNP, with the preponderance of compromised lions originating from bovine TB infected areas. Lions from the uninfected far north of the KNP (n = 11) were assessed as non-infected control animals. These two cohorts of lions (84 and 11) were euthanized and their status confirmed through a comprehensive necropsy and mycobacterial culture process. An additional group of control lions (n = 33) from the uninfected area of the park were randomly selected and skin tested. They were not destroyed to confirm their status as they were in prime condition and not repeat-offenders. These lions were considered as negative specimens because they originated from an area where bovine TB had not been diagnosed in buffaloes or lions despite concerted efforts. This lion sub-population had been separated from the infected sub-population by a buffer zone of approximately 120 km wide. The buffer zone had a very low prevalence of bovine TB in buffalo (1.5%: 0.4–4.0, 95% CI) (Rodwell et al., 2001) and no lion cases were diagnosed. 2.2. Test procedure Lions were immobilized with a combination of tiletamine and zolazepam (Zoletil1 100, Virbac Animal Health, Halfway House, South Africa) and transported to predator isolation units in Skukuza, the KNP headquarters. Venous blood samples were obtained from the medial or lateral saphenous veins immediately after anaesthesia. Age of the animal was estimated by examining dental attrition, its condition was scored and it was weighed. The animals were then tuberculin-tested according to the protocol described below and confined for 72 h. For skin testing
386
D.F. Keet et al. / Veterinary Microbiology 144 (2010) 384–391
two rectangular areas of skin of approximately 40 mm 40 mm on the lateral aspects of the proximal third of the neck, in close proximity to the cervical lymph nodes were clipped using a pro-cordless rechargeable clipper (Oster1 Professional Products, McMinnville, TN). The mane of male lions covering the prescribed area was initially cut with curved scissors and then clipped. A double thickness fold of the skin centred in the clipped area was measured with tuberculosis callipers (Hauptner1, Solingen, Germany). A dose of 0.2 ml (6000 I.U. = 0.2 mg) instead of 0.1 ml Bovine Tuberculin PPD 3000TM (Animal Sciences Group, Wageningen UR, Lelystad, Netherlands) was injected obliquely but strictly deep intradermally on the left-hand side with a 1 ml disposable syringe and a short 23 gauge needle (Terumo1 Corporation, Tokyo, Japan) resulting in a visible nodule in the skin. The same procedure was repeated with 0.2 ml (5000 I.U. = 0.2 mg) Avian Tuberculin PPD 2500TM (Animal Sciences Group, Wageningen UR, Lelystad, Netherlands) (Anon., 2009) on the right-hand side. The bovine tuberculin PPD dose exceeds the prescribed dosage (5000 I.U.) used in cattle but not the prescribed injected volume (Anon., 2009), and was decided on due to reports of failure to obtain repeatable reactions in domestic cats and lions when the standard bovine doses were used at various inoculation sites. After 72 h, the skin fold at injection sites was then remeasured. The skin at injection sites was also examined for visible reactions, palpated and described. Two approaches to skin test result interpretation were employed; in a comparative approach (Anon., 2009) where the reaction was considered positive if the increase in skin thickness at the bovine tuberculin PPD site of injection was 2 mm and larger than the reaction measured at the avian tuberculin PPD injection site (Single Intradermal Comparative Cervical Test: (SICCT)). The other interpretation (Single Intradermal Cervical Test: (SICT)) consisted of measuring both responses, but only considering the bovine tuberculin PPD response (de la Rua-Domenech et al., 2006) of 2 mm and larger for definitive diagnosis, this despite the response to avian tuberculin PPD possibly being larger (response to avian tuberculin PPD was measured for the purpose of the study). A classification of M. bovis reactors, M. bovis nonreactors, non-M. bovis reactors or non-M. bovis nonreactors was made when mycobacterial culture results and bovine tuberculin PPD results could be combined (Table 1). With the first 15 lions tested, a full thickness skin biopsy was then collected from the centre of the clipped injection sites using a 6 mm Ø biopsy punch (Kai1 Sterile
Disposable biopsy punch, Seki City, Japan) irrespective of the presence of a visible reaction. Biopsy sections were placed in marked Teflon sachets (Shandon1, Japan) and placed in 10% buffered formalin. 2.3. Bacteriology 2.3.1. Tissue collection After evaluating test reactions, the lions were euthanized by means of intravenous administration of a lethal dosage of sodium pentobarbitone (Euthapent1, Kyron, South Africa). Detailed necropsies were performed immediately after death was confirmed. All tissues collected were thinly sectioned with ultra sharp 130 mm trimming blades (Tissue-Tek1 Accu-Edge1, Sakura Finetek, Japan) for macroscopic inspection and then divided in two equal parts for preservation in 10% buffered formalin and culture, respectively. For bacterial isolation, lymph nodes were pooled in separate sterile containers according to body regions drained, namely cranial, thoracic, abdominal, peripheral and mammary lymph nodes. The cranial lymph node cluster consisted of the parotid, mandibular and retropharyngeal nodes. The thoracic cluster consisted of the tracheobronchial nodes while the abdominal cluster consisted of gastric, mesenteric, caecal and colic as well as hepatic nodes. The peripheral cluster consisted of superficial and deep cervical, axillary and accessory cervical, popliteal, superficial inguinal and iliosacral nodes. In addition, lesions suspect for tuberculosis in organs were collected separately. All tissue samples were frozen and stored at 20 8C until transferred to the laboratory for culture. 2.3.2. M. bovis culture and characterization Specimens were identified and cultured separately as previously described (Bengis et al., 1996). Differentiation between M. tuberculosis complex and non-tuberculous mycobacteria (NTM) was also aided by PCR amplification of a genome region specific for M. tuberculosis complex organisms (Cousins et al., 1991). Isolates classified as NTM were subjected to a series of cultural and biochemical tests including temperature growth test, pigment production assay, nitratase test, arylsulphatase test, and tellurite reduction test (Collins et al., 1985). 2.4. Histology Specimens for histological examination were fixed in 10% buffered formalin. Identified lymph node sections
Table 1 Intradermal test results (mean, standard deviation and variance) of 95 lions of which infection status was confirmed through mycobacterial culture and 33 additional lions that were skin tested only. Mean skin-fold thickness increase (mm)
M. bovis reactors (n = 45) Non-M. bovis reactors (n = 6) M. bovis non-reactors (n = 7) Non-M. bovis non-reactors from the bovine TB infected zone (n = 26) Non-M. bovis non-reactors from bovine TB free zone (n = 44)
Bovine site (6000 I.U. ppd.)
Avian site (5000 I.U. ppd.)
Bovine/avian difference (mm)
4.94 2.24 4.6 2.89 1.21 0.62 0.47 0.83
1.00 1.53 (2.34) 0.36 1.47 (2.16) 0.13 0.96 (0.93) 0.35 0.896 (0.79)
4.00 2.13 4.23 2.13 1.11 1.14 0.57 0.86
1.01 1.07 (1.31)
1.05 0.96 (0.92)
(5.02) (8.36) (0.38) (0.69)
0.27 0.66 (0.44)
(4.52) (4.53) (1.31) (0.75)
D.F. Keet et al. / Veterinary Microbiology 144 (2010) 384–391
387
were placed in marked Teflon sachets. They were embedded in paraffin wax, sectioned (3 mm) and stained with haematoxylin and eosine and Ziehl-Neelsen’s stain for light-microscopic examination. Macro- and histopathology results were consulted to support culture negative status of control lions and where M. bovis nonreactors and non-M. bovis reactors were encountered (Crawshaw et al., 2008).
Table 2a Sensitivity values for comparative and single skin test interpretations in an infected lion sub-population.
2.5. FIVPle serology
Table 2b Situational applied specificity (false positive rate for true negatives) calculated from culture negative cases from the infected zone.
An enzyme-linked immunosorbent assay (ELISA) using a puma lentivirus-derived synthetic peptide as coating antigen was used to test for FIVPle on serum samples of 72/ 95 lions necropsied (Van Vuuren et al., 2003). The sensitivity of the test when using the Western blot as the gold standard was 78.6% and the specificity 100% (Van Vuuren et al., 2003). 2.6. Data analysis Data analysis was directed at the observed increases in skin thickness in response to bovine and avian tuberculin PPD. Summary statistics included mean, standard deviation, and 95% confidence interval for the ratio of group means derived from the 95% confidence intervals for differences between group means. Various subgroup analyses were done and employed, i.e. Student’s two-sample t-test, Student’s paired t-test and analysis of covariance (ancova). In the presence of small sample sizes, the outcomes of the former tests were confirmed using nonparametric tests, ranksum tests, Wilcoxon’s matched pairs signed-ranks tests and ancova for ranks. Testing was directed at the 0.05 level of significance. 3. Results 3.1. Intradermal testing Skin test responses at bovine tuberculin PPD injection sites of 51/52 infected culture-positive lions were clearly visible and swellings were well demarcated but seldom indurated. In 45 of these lions, the difference between the two readings exceeded or equaled 2 mm. Pain and discomfort was impossible to detect as the lions were completely anaesthetized during examination. Redness and superficial necrosis with exudation was seen in only a small proportion of reactions. As reactions were typically soft, warm and oedematous, they were measured prior to being palpated. Swellings decreased considerably after palpation and first-time measuring. Reacting lions (M. bovis and non-M. bovis reactors) in infected areas (51/84) showed significantly larger responses to bovine tuberculin PPD than to avian tuberculin PPD (paired t-test; p < 0.0001) (Tables 1 and 2a). Forty-five infected lions were M. bovis reactors while 7 were M. bovis non-reactors when SICT criteria were applied (Tables 1 and 2b). Forty-two infected lions were M. bovis reactors while 10 were M. bovis non-reactors when SICCT criteria were applied. Lions residing in the non-infected area showed no response to bovine tuber-
SICT SICCT
Total
SE
FN
TP
52 52
0.865 0.808
7 10
45 42
SE: sensitivity; FN: false negative reactors; TP: true positive reactors.
SICT SICCT
Total
SP
TN
FP
32 32
0.8125 0.8125
26 26
6 6
SP: specificity; TN: true negative reactors; FP: false positive reactors.
Table 2c Specificity values for comparative and single skin test interpretations in a non-infected lion sub-population. Eleven of these cases were necropsied and cultured negative.
SICT SICCT
Total
SP
TN
FP
44 44
1.00 1.00
44 44
0 0
SP: specificity; TN: true negative reactors; FP: false positive reactors.
culin PPD but varying although negligible responses to avian tuberculin PPD (Tables 1 and 3). Avian responses in general in this far northern area of KNP were larger than bovine responses (Table 2c). 3.2. Bacteriology M. bovis was cultured from 52/95 (54.7%) skin tested and necropsied lions (Table 3). Of the 52 culture-positive lions, 40 were emaciated and ailing. In addition, NTM were cultured from 45/95 (47.4%) of the lions tested of which 24 animals were NTM positive only (Table 3), demonstrating the ubiquitous nature of these organisms in the KNP population. Although the group of NTM isolates was not fully characterized it was shown that it comprised both slow and fast-growing mycobacteria including Mycobacterium avium complex, Mycobacterium chelonae and Mycobacterium fortuitum. Of the 52 M. bovis-infected lions, 31 were infected with M. bovis only, while 21 were co-infected with NTM. Nineteen (19/95) cases yielded negative mycobacterial cultures. With respect to the mean bovine tuberculin PPD response, the NTM/M. bovis co-infected and M. bovis only groups did not differ significantly (two-sample t-test: p = 0.8829; 4.5 mm vs. 4.39 mm) (Table 3). As anticipated responses to avian tuberculin PPD were significantly stronger in animals co-infected with M. bovis and NTM than in those infected with NTM only (one-sided twosample t-test: p = 0.046; 1.19 mm vs. 0.5 mm) (Table 3). No significant difference was found between bovine tuberculin PPD skin responses of culture-positive emaciated lions (n = 40, mean = 4.4175 2.396) and culture-positive lions in fair condition (n = 12, mean 4.5 2.75) two-sample t-test with equal variances (t = 0.1012; p = 0.9198).
388
D.F. Keet et al. / Veterinary Microbiology 144 (2010) 384–391
Table 3 Intradermal test results (mean, standard deviation and variance) of 76 lions culturing positive for mycobacteriosis. Culture result
Mean skin-fold thickness increase (mm) Bovine site (6000 I.U. ppd.)
Avian site (5000 I.U. ppd.)
Bovine/avian difference (mm)
M. bovis only (n = 31)
4.39 2.34 5.48 4.5 2.67 7.13 0.8 1.11 1.24
0.68 1.38 1.91 1.19 1.62 2.63 0.51 0.99 0.98
3.72 1.95 3.79 3.45 2.67 7.13 1.0 1.27 1.61
M. bovis and NTM (n = 21) NTM only (n = 24)
3.3. FIV serology The proportion of FIVPle seropositive lions amongst the tested group was 47% (34/72; 95% CI, 35.4–59.0%); of these 82% (28/34) were ailing and compromised. Co-infection with FIVPle had no visible/measurable influence on skin response. Lions with concurrent FIVPle and M. bovis infections comprised 26% (19/72) of tested lions and had a mean increase in skin thickness of 4.84 2.3 mm. Eighteen M. bovis-infected lions tested negative for FIVPle and their average skin response was 3.86 2.33 mm. These groups did not differ significantly with respect to mean skin response (two-sample t-test: p = 0.2034; 4.84 mm vs. 3.86 mm). However, the 95% confidence interval for the mean response of the FIV co-infected group as a ratio of the mean response of the M. bovis only infected group is (0.855; 1.655), i.e. the response for the FIV co-infected group can be as low as 85.5% and as high as 165.5% of the M. bovis only group response. 3.4. Histology Skin biopsies excised from visible test reactions were all characterized by varying degrees of oedema with infiltration of macrophages and lymphocytes throughout. Reactions to bovine tuberculin PPD were more severe, however, due to a marked superficial and deep perivascular round cell infiltrate associated with severe oedema. This histological pattern is consistent with a delayed-type hypersensitivity reaction at both the avian and bovine tuberculin PPD injection sites. Skin biopsies excised from test sites of non-responding and most M. bovis culture negative lions (n = 4) showed no delayed-type hypersensitivity reaction. Microscopic lesion patterns observed in various organs were of a granulomatous nature consisting of macrophages, epithelioid cells, lymphoplasma cells and numerous neutrophils, suggestive of mycobacterial infection. Severe generalized lymphoid atrophy was sometimes seen in association with granulomatous lesions. Pulmonary lesions comprised of granulomatous interstitial pneumonitis or granulomatous bronchopneumonia often associated with bronchiectasis. Intestinal lesions showed mononuclear macrophage predominance suggestive of mycobacterial mural enteritis. Granulomatous osteitis, periosteitis and osteosis were found in most of the welldeveloped cases frequently associated with myositis. Granulomatous lesions containing eosinophils and parasitic remnants were often observed in lymph nodes and the intestinal tract. An important feature of all organ sections
from culture-positive cases was the apparent absence of acid-fast bacilli in granulomas when stained with Ziehl– Neelsen stain. Microscopic non-specific granulomatous lesions were also seen in the lymph nodes, lungs and intestinal tracts of three of the 11 control cases. All control cases cultured negative for M. bovis but NTM were isolated from 4 of these cases. 4. Discussion In this study we investigated the intradermal tuberculin testing of free-living African lions naturally exposed to infection with M. bovis. We demonstrated that administration of bovine tuberculin PPD identified a significant proportion (86.5%) of the lions naturally infected as confirmed by mycobacterial culture. Testing involved administration of a single bolus of double the conventional (0.1 mg) dosage of bovine and avian tuberculin PPD and interpretation of the skin tests by assessing comparative bovine versus avian tuberculin PPD elicited readings with single bovine tuberculin PPD readings. Co-infection with FIVPle and NTM were considered as potential immunecompromising events that could influence tuberculin skin test responses and consequently their interpretation. For this reason, results of mycobacterial culture as diagnostic gold standard and tuberculin skin tests were comparatively analyzed resulting in the proposal of adapted criteria for interpretation. Preferred and absolute validation and verification of a diagnostic trial test result requires detailed necropsy followed by culturing of all organ systems separately irrespective of the presence of macroscopic lesions or whether they reacted or not to the test (Goodchild and Clifton-Hadley, 2001). This is particularly valid for lions, where multiple organ systems are affected and lesions other than pulmonary ones, are not readily identifiable macroscopically (Keet et al., 2000). For a number of compelling reasons histology results were not considered to advance the gold standard as to increase sensitivity and/ or specificity; firstly the frequent culture-confirmed presence of NTM which may result in the possible sighting of acid-fast bacilli in suspicious and microscopic lesions, and secondly, the actual observed absence of acid-fast bacilli in histological lesions from lion cases confirmed positive for M. bovis and stained with Ziehl–Neelsen (Keet et al., 1996, in preparation). Reference to histology concerning the 11 control cases was merely supplementary to culture verification of absolute negative status. Results of histological examinations are addressed in a
D.F. Keet et al. / Veterinary Microbiology 144 (2010) 384–391
pathology-specific publication (Keet et al., in preparation). Suspicious lesions of a granulomatous nature but not revealing acid-fast bacilli could also have been caused by related bacteria, parasites or fungi (Crawshaw et al., 2008). Comprehensive culture of lymph node material from lesion-negative animals at necropsy is vital as it may further identify an additional proportion, however small, of those animals infected with M. bovis (Corner, 1994). The majority of cases in a highly developed outbreak of bovine TB in any maintenance host species are subclinical infections (De Vos et al., 2001). Similar to other species, the progress of bovine TB infection in lions is also typically slow, and the majority of infected lions encountered appear healthy while being subclinically infected (Keet, unpublished data). Tests that measure CMI are currently still the most sensitive to identify such cases and for determining prevalence and geographical distribution of M. bovis infection (Pollock et al., 2001). Applying principles of prescribed comparative test criteria (Anon., 2009), and using a 2-mm increase and difference between tuberculin PPD responses, presented us with a sub-optimal sensitivity. Retrospectively evaluating cut-off points for interpretation with the benefit of positive culture results, it was observed that using 2-mm increase as cut off, both sensitivity and specificity remained highest. A 2 mm increase is already considerable as it represents a 25– 63% increase from baseline value (Time0) skin thickness of lions (results not shown). As a point of reference it must be cited that an increased difference of 2 mm is expected in positive cattle when associated with clinical signs such as diffuse or extensive oedema, exudation, necrosis, pain or inflammation of the lymphatic vessels and nodes in the region or at least 4 mm in the absence of clinical signs. In the same reference it is further concluded, however, that any palpable or visible swelling should be considered positive in M. bovis-infected cattle herds (Anon., 2009). In our study, using comparative test criteria did not detect all infected lions, due to concurrent NTM infections which resulted in equal increases at both tuberculin PPD inoculation sites in some cases, resulting in a skin test negative interpretation due to possible cross-reactivity induced by the NTM. Such cross-reactive responses elicited by exposure to NTM often confounded the interpretation for M. bovis infection in cattle (Waters et al., 2006). We therefore propose that a 2-mm increase in skin-fold thickness specific to bovine tuberculin PPD is indicative of a positive M. bovis reactor, irrespective of the size of the avian tuberculin PPD response. We hypothesize that concurrent inoculation of avian tuberculin PPD in the likely presence of mixed mycobacterial infections, would reduce the effect of cross-reactivity rendering a more specific and dependable response at the bovine tuberculin PPD inoculation site. This unconventional diagnostic approach produced a 5.8% increase in sensitivity while the false positive rate for true negative cases remained unchanged (Tables 2a and 2b). The decision to use 0.2 mg of bovine tuberculin PPD instead of the standard prescribed 0.1 mg was established after previous limited efforts to diagnose bovine TB in ad hoc lion cases resulted in no visible or palpable response when 0.1 mg was administered intrapalpebrally, intrader-
389
mally or using the subcutaneous febrile test. Unfortunately, lions on which these early tests were done were not always destroyed to confirm disease status (Keet own observation). False negative skin reactions may transpire when an insufficient tuberculin PPD dose is injected (Monaghan et al., 1994). It was further argued that administering a relatively high concentration of bovine tuberculin PPD would induce a greater and more specific immune response after exposure to various other Mycobacterium spp. This hypothesis seems to be applicable in our study as well, because most lions with mixed infections responded better to bovine tuberculin PPD than to avian tuberculin PPD. However, we also observed that a significantly stronger avian tuberculin PPD response in the presence of M. bovis/NTM co-infection when compared to that of a single NTM infection, which strongly supports our diagnostic approach of only considering bovine tuberculin PPD results (Table 3). The prevalence of NTM isolated in this study is exceptionally high when compared to that observed in other species (Hughes et al., 2005; Backues, 2008; de Lisle et al., 2008). The large number of NTM isolates cultured from lion samples suggests that lions acquire a variety of these organisms from the environment and probably from prey animals as is suggested by 33/45 isolates from cranial lymph node clusters (Table 4). Free-living wildlife has been reported to be the source of various mycobacteria (Kollias et al., 1982) and prey species are a likely source for lions. In addition lions are also rather inactive and spend a large part of their existence lying down. This could increase exposure to soil mycobacteria. Also due to the generally violent nature of their existence (hunting and intraspecific competition), skin integrity is regularly breached with consequent trans-cutaneous infection. In this study 21 lions had NTM infected peripheral lymph nodes (Table 4). The above-mentioned behavioural patterns probably expose lions to infection with various NTMs more frequently and such immunological stimulation could elicit non-specific responses to bovine and avian tuberculin PPD, influencing diagnostic interpretation. The abundance of NTM in carnivores associated with M. bovis coinfection and particularly in lions should always be considered when validating any immunodiagnostic assays. The seven tuberculous non-reactors (bovine responses not reaching 2 mm but still larger than baseline measurement) from which M. bovis was cultured (Tables 1 and 2a), were all emaciated and possibly partially anergic, as M. bovis was cultured from all lymph node sample sets submitted. Two of these animals had characteristic pulmonary lesions. A state of anergy may develop in cattle with advanced or generalized bovine TB and (temporarily) in animals subjected to stress (Pollock et al., 2001) which may also be the case in these seven lions. The six non-tuberculous reactors (only bovine tuberculin PPD result considered) (Tables 1, 2a and 2b) comprised four animals in fair condition and two emaciated lions with distinctive macroscopic pulmonary lesions histologically suggestive of mycobacteriosis. Two of the four animals in fair body condition had NTM infections. All six nontuberculous reactors originated from the high and medium infected zones.
D.F. Keet et al. / Veterinary Microbiology 144 (2010) 384–391
390
Table 4 Distribution of non-tuberculosis mycobacteria (NTM) (n = 45 cases) and M. bovis (n = 52 cases) in different body regions of lions as determined through mycobacterial specific culture.
NTM isolations M. bovis isolations
Head
Thorax
Abdomen
Peripheral
Mammary
Skeletal
33 11
11 35
8 20
21 30
10 5
5 15
It was anticipated that the presence of FIVPle may reduce skin response to tuberculin similar to the situation of combined human immunodeficiency virus (HIV) and M. tuberculosis infection (Duncan et al., 1995; Klein et al., 1999). No evidence of this phenomenon could be found in our limited study, despite analysis of mostly emaciated lions. The majority of animals diagnosed as infected through mycobacterial culture, showed a positive bovine tuberculin PPD skin test (both SICCT and SICT interpretations) despite of the fact that FIVPle is supposed to cause moderate to severe depletion of CD4 subsets in lions (Roelke et al., 2006). This cell depletion could eventually influence CMI response, which provides the most accurate assessment of prevalence of bovine TB infection (Pollock et al., 2001). In non-reactor lions with advanced bovine TB, which may have been accelerated by FIVPle co-infection, serological tests might complement intradermal tuberculin tests. However, ultimately antibody levels may also diminish due to co-infection. Considering limitations of field investigations on an endangered species, we were able to demonstrate that intradermal tuberculin testing can be used to identify M. bovis infection in lions. We further demonstrated that possible effects of cross-reactivity can be overcome by administering both diagnostic antigens but only considering bovine tuberculin PPD response. Finally, no evidence could be found that FIVPle co-infection attenuated skin reactivity in the presence of M. bovis infection. Practical constraints, however, are the cost and effort involved when performing these tests in wild lions in a large conservation area. Conflict of interest The authors declare that they do not maintain any financial and personal relationships with other people or organizations that could have inappropriately influenced their work. Acknowledgements This research was predominantly funded the South African Veterinary Foundation and the Directorate of Veterinary Services, National Department of Agriculture, Forestry and Fisheries, South Africa. We are grateful to the field staff of the Directorate of Veterinary Services, Messrs. E.H. Dekker, T.K. Muchocho and J. Oosthuizen whom assisted with technical execution of the project, as well as the physical assistance of Tradesman Aides: Messrs C. Masinga, E. Ubisi, K. Nkuna, E. Munyamela, P. Baloyi in the management of the lion holding facilities, and assistance with necropsies; the ranger staff of SANParks (KNP) for reporting and assisting with the capture of some of the study lions; the Bacteriology Section of the ARC-Onder-
stepoort Veterinary Research Institute for mycobacterial culture. Drs. W. Botha and L. du Plessis from Golden Vetpath for histological examinations of referred cases. We also acknowledge Prof. N.P.J. Kriek for valuable advice and Mrs. A. Lourens from the J. F. Soga Library, Faculty Veterinary Science Library, University of Pretoria, for continuous support. D.F. Keet was the recipient of a doctoral bursary awarded by the University of Pretoria. This project was approved by the South African National Parks conservation and ethics committees as well as the Animal Use & Care Committee of the University of Pretoria. References Anon, 2009. Bovine tuberculosis. In: Manual of Diagnostic Tests and Vaccines for Terrestrial Animals, fifth edition. Office International des Epizooties, Paris (updated August 14, 2009), Chapter 2.4.7. Backues, K.A., 2008. Mycobacterial diseases in carnivores. In: Fowler, M.E., Miller, R.E. (Eds.), Zoo and Wild Animal Medicine. sixth edition. W.B. Saunders, Saint Louis, pp. 294–298. Bauer, H., Van Der Merwe, S., 2004. Inventory of free-ranging lions Panthera leo in Africa. Oryx 38, 26–31. Bengis, R.G., Kriek, N.P.J., Keet, D.F., Raath, J.P., De Vos, V., Huchzermeyer, H.F.A.K., 1996. An outbreak of bovine tuberculosis in a free-living African buffalo (Syncerus caffer-Sparrman) population in the Kruger National Park: a preliminary report. Onderstepoort J. Vet. Res. 63, 15– 18. Charleston, B., Hope, J.C., Carr, B.V., Howard, C.J., 2001. Masking of two in vitro immunological assays for Mycobacterium bovis (BCG) in calves acutely infected with non-cytopathic bovine viral diarrhoea virus. Vet. Rec. 149, 481–484. Cleaveland, S., Mlengeya, T., Kazwala, R.R., Michel, A., Kaare, M.T., Jones, S.L., Eblate, E., Shirima, G.M., Packer, C., 2005. Tuberculosis in Tanzanian wildlife. J. Wildl. Dis. 41, 446–453. Collins, C.H., Grange, J.M., Yates, M.D., 1985. Organization and Practice in Tuberculosis Bacteriology. Butterworth & Co Limited, London. Corner, L.A., 1994. Post mortem diagnosis of Mycobacterium bovis infection in cattle. Vet. Microbiol. 40, 53–63. Cousins, D.V., Florisson, N., 2005. A review of tests available for use in the diagnosis of tuberculosis in non-bovine species. Rev. Sci. Tech. 24, 1039–1059. Cousins, D.V., Wilton, S.D., Francis, B.R., 1991. Use of DNA amplification for the rapid identification of Mycobacterium bovis. Vet. Microbiol. 27, 187–195. Crawshaw, T.R., Griffiths, I.B., Clifton-Hadley, R.S., 2008. Comparison of a standard and a detailed postmortem protocol for detecting Mycobacterium bovis in badgers. Vet. Rec. 163, 473–477. de la Rua-Domenech, R., Goodchild, A.T., Vordermeier, H.M., Hewinson, R.G., Christiansen, K.H., Clifton-Hadley, R.S., 2006. Ante mortem diagnosis of tuberculosis in cattle: a review of the tuberculin tests, g-interferon assay and other ancillary diagnostic techniques. Res. Vet. Sci. 81, 190–210. de Lisle, G.W., Kawakami, R.P., Yates, G.F., Collins, D.M., 2008. Isolation of Mycobacterium bovis and other mycobacterial species from ferrets and stoats. Vet. Microbiol. 132, 402–407. De Vos, V., Bengis, R.G., Kriek, N.P.J., Michel, A., Keet, D.F., Raath, J.P., Huchzermeyer, H.F.K.A., 2001. The epidemiology of tuberculosis in free-ranging African buffalo (Syncerus caffer) in the Kruger National Park, South Africa. Onderstepoort J. Vet. Res. 68, 119–130. Duncan, L.E., Elliott, A.M., Hayes, R.J., Hira, S.K., Tembo, G., Mumba, G.T., Ebrahim, S.H., Quigley, M., Pobee, J.O.M., McAdam, K.P.W.J., 1995. Tuberculin sensitivity and HIV-1 status of patients attending a sexually transmitted diseases clinic in Lusaka, Zambia: a cross-sectional study. Trans. R. Soc. Trop. Med. Hyg. 89, 37–40.
D.F. Keet et al. / Veterinary Microbiology 144 (2010) 384–391 Goodchild, A.V., Clifton-Hadley, R.S., 2001. Cattle-to-cattle transmission of Mycobacterium bovis. Third International Conference on Mycobacterium bovis, M. bovis 2000, August 13–16, 2000, Cambridge, UK, Tuberculosis 81, 23–41. Grobler, D.G., Michel, A.L., De Klerk, L., Bengis, R.G., 2002. The gammainterferon test: its usefulness in a bovine tuberculosis survey in African buffaloes (Syncerus caffer) in the Kruger National Park. Onderstepoort J. Vet. Res. 69, 221–227. Hughes, M.S., Ball, N.W., McCarroll, J., Erskine, M., Taylor, M.J., Pollock, J.M., Skuce, R.A., Neill, S.D., 2005. Molecular analyses of mycobacteria other than the M. tuberculosis complex isolated from Northern Ireland cattle. Vet. Microbiol. 108, 101–112. Keet, D.F., Kriek, N.P.J., Penrith, M., Michel, A., Huchzermeyer, H., 1996. Tuberculosis in buffaloes (Syncerus caffer) in the Kruger National Park: spread of the disease to other species. Onderstepoort J. Vet. Res. 63, 239–244. Keet, D.F., Michel, A., Meltzer, D.G.A., 2000. Tuberculosis in free-ranging lions (Panthera leo) in the Kruger National Park. In: Proceedings of the South African Veterinary Association Biennial Congress, September 2000, Durban, pp. 232–241. Klein, R.S., Flanigan, T., Schuman, P., Smith, D., Vlahov, D., 1999. Criteria for assessing cutaneous anergy in women with or at risk for HIV infection. J. Allergy Clin. Immunol. 103, 93–98. Kollias, G.V., Thoen, C.O., Fowler, M.E., 1982. Evaluation of comparative cervical tuberculin skin testing in cervids naturally exposed to mycobacteria. J. Am. Vet. Med. Assoc. 181, 1257–1262. Lyashchenko, K.P., Greenwald, R., Esfandiari, J., Chambers, M.A., Vicente, J., Gortazar, C., Santos, N., Correia-Neves, M., Buddle, B.M., Jackson, R., O’Brien, D.J., Schmitt, S., Palmer, M.V., Delahay, R.J., Waters, W.R., 2008. Animal-side serologic assay for rapid detection of Mycobacterium bovis infection in multiple species of free-ranging wildlife. Vet. Microbiol. 132, 283–292. Michel, A.L., Coetzee, M.L., Keet, D.F., Mare´, L., Warren, R., Cooper, D., Bengis, R.G., Kremer, K., van Helden, P., 2009. Molecular epidemiology of Mycobacterium bovis isolates from free-ranging wildlife in South African game reserves. Vet. Microbiol. 133, 335–343.
391
Miller, M.A., 2008. Current diagnostic methods for tuberculosis in zoo animals. In: Fowler, M.E., Miller, R.E. (Eds.), Zoo and Wild Animal Medicine. sixth edition. W.B. Saunders, Saint Louis, pp. 10–19. Monaghan, M.L., Doherty, M.L., Collins, J.D., Kazda, J.F., Quinn, P.J., 1994. The tuberculin test. Vet. Microbiol. 40, 111–124. Morris, P.J., Thoen, C.O., Legendre, A.M., 1996. Pulmonary tuberculosis in an African lion (Panthera leo). J. Zoo. Wild. Med. 27, 392–396. Palmer, M.V., Waters, W.R., 2006. Advances in bovine tuberculosis diagnosis and pathogenesis: what policy makers need to know. Vet. Microbiol. 112, 181–190. Pollock, J.M., McNair, J., Welsh, M.D., Girvin, R.M., Kennedy, H.E., Mackie, D.P., Neill, S.D., 2001. Immune responses in bovine tuberculosis. Tuberculosis 81, 103–107. Rhodes, S.G., Gruffydd-Jones, T., Gunn-Moore, D., Jahans, K., 2008. Adaptation of IFN-gamma ELISA and ELISPOT tests for feline tuberculosis. Vet. Immunol. Immunopathol. 124, 379–384. Rodwell, T.C., Kriek, N.P., Bengis, R.G., Whyte, I.J., Viljoen, P.C., de Vos, V., Boyce, W.M., 2001. Prevalence of bovine tuberculosis in African buffalo at Kruger National Park. J. Wildl. Dis. 37, 258–264. Roelke, M.E., Pecon-Slattery, J., Taylor, S., Citino, S., Brown, E., Packer, C., VandeWoude, S., O’Brien, S.J., 2006. T-lymphocyte profiles in FIVinfected wild lions and pumas reveal Cd4 depletion. J. Wildl. Dis. 42, 234–248. Van Vuuren, M., Stylianides, E., Kania, S.A., Zuckerman, E.E., Hardy, J.W.D., 2003. Evaluation of an indirect enzyme-linked immunosorbent assay for the detection of feline lentivirus-reactive antibodies in wild felids, employing a puma lentivirus-derived synthetic peptide antigen. Onderstepoort J. Vet. Res. 70, 1–6. Vermund, S.H., Yamamoto, N., 2007. Co-infection with human immunodeficiency virus and tuberculosis in Asia. Tuberculosis 87, S18– S25. Waters, W.R., Palmer, M.V., Thacker, T.C., Payeur, J.B., Harris, N.B., Minion, F.C., Greenwald, R., Esfandiari, J., Anderson, P., McNair, L., Pollock, J.M., Lyashchenko, K.P., 2006. Immune responses to defined antigens of Mycobacterium bovis in cattle experimentally infected with Mycobacterium kansasii. Clin. Vaccine Immunol. 13, 611–619.
Contents lists available at ScienceDirect
Veterinary Microbiology journal homepage: www.elsevier.com/locate/vetmic
Intradermal tuberculin testing of wild African lions (Panthera leo) naturally exposed to infection with Mycobacterium bovis D.F. Keet a,b,*, A.L. Michel b,e, R.G. Bengis a, P. Becker d,f, D.S. van Dyk a, M. van Vuuren b, V.P.M.G. Rutten b,c, B.L. Penzhorn b a
Directorate of Veterinary Services, Kruger National Park, P.O. Box 12, Skukuza 1350, South Africa Department of Veterinary Tropical Diseases, Faculty of Veterinary Science, University of Pretoria, Private Bag X04, Onderstepoort 0110, South Africa Division of Immunology, Department of Infectious Diseases & Immunology, Faculty of Veterinary Medicine, Utrecht University, Yalelaan 1, 3584 CL Utrecht, The Netherlands d Biostatistics Unit, Medical Research Council, Private Bag X385, Pretoria, South Africa e Bacteriology Section, ARC-Onderstepoort Veterinary Institute, Private Bag X05, Onderstepoort 0110, South Africa f Division of Clinical Epidemiology, Faculty of Health Sciences, University of Pretoria, Private Bag X385, Pretoria, South Africa b c
A R T I C L E I N F O
A B S T R A C T
Article history: Received 13 August 2009 Received in revised form 4 January 2010 Accepted 27 January 2010
African lions in the southern half of Kruger National Park (KNP) are infected with Mycobacterium bovis. Historically, reliable detection of mycobacteriosis in lions was limited to necropsy and microbiological analysis of lesion material collected from emaciated and ailing or repeat-offender lions. We report on a method of cervical intradermal tuberculin testing of lions and its interpretation capable of identifying natural exposure to M. bovis. Infected lions (n = 52/95) were identified by detailed necropsy and mycobacterial culture. A large proportion of these confirmed infected lions (45/52) showed distinct responses to bovine tuberculin purified protein derivative (PPD) while responses to avian tuberculin PPD were variable and smaller. Confirmed uninfected lions from non-infected areas (n = 11) responded variably to avian tuberculin PPD only. Various non-tuberculous mycobacteria (NTM) were cultured from 45/95 lions examined, of which 21/45 were co-infected with M. bovis. Co-infection with M. bovis and NTM did not influence skin reactions to bovine tuberculin PPD. Avian tuberculin PPD skin reactions were larger in M. bovis-infected lions compared to uninfected ones. Since NTM coinfections are likely to influence the outcome of skin testing, stricter test interpretation criteria were applied. When test data of bovine tuberculin PPD tests were considered on their own, as for a single skin test, sensitivity increased (80.8–86.5%) but false positive rate for true negatives (18.75%) remained unchanged. Finally, the adapted skin test procedure was shown not to be impeded by persistent Feline Immunodeficiency VirusPle co-infection. ß 2010 Elsevier B.V. All rights reserved.
Keywords: Mycobacterium bovis Lion Panthera leo Intradermal tuberculin testing Co-infection Non-tuberculous mycobacteria Feline Immunodeficiency Virus Kruger National Park
1. Introduction Mycobacterium bovis has infected part of the African buffalo (Syncerus caffer) population in the Kruger National
* Corresponding author at: Directorate of Veterinary Services, Kruger National Park, P.O. Box 12, Skukuza 1350, South Africa. Tel.: +27 82 9279650; fax: +27 13 7356693. E-mail address: [email protected] (D.F. Keet). 0378-1135/$ – see front matter ß 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.vetmic.2010.01.028
Park (KNP) causing bovine tuberculosis (bovine TB) (Bengis et al., 1996; De Vos et al., 2001). Since the African buffalo is considered a preferential prey species, African lions (Panthera leo) residing in areas of KNP with a high prevalence of bovine TB in buffaloes acquired infection through consumption of tuberculous carcasses (Keet et al., 1996, 2000). We confirmed, by genetic typing, that spillover of M. bovis from buffaloes to lions occurred in the KNP (Michel et al., 2009). The lion index case in KNP was diagnosed through necropsy and culture in 1995 (Keet
D.F. Keet et al. / Veterinary Microbiology 144 (2010) 384–391
et al., 1996). Subsequently a number of emaciated and ailing lions were euthanized and it was established that multiple organ systems were affected, suggesting several possible routes of infection followed by haematogenous and/or lymphatic spread of mycobacteria (Keet et al., 2000). It is estimated that between 16,500 and 30,000 wild lions occur in Africa of which approximately 2200 occur in the Kruger Ecosystem (Bauer and Van Der Merwe, 2004). Since African lions are listed as Vulnerable on the IUCN Red Data List (2006), this precarious population status demanded a sensitive research approach regarding various infectious diseases especially bovine TB in lions (Cousins and Florisson, 2005). There are currently still no reliable immunodiagnostic tests for feline tuberculosis (Rhodes et al., 2008). Morris et al. (1996) referred to an ELISA test that was used on three zoo lions exposed to a lion with advanced pulmonary bovine TB. A similar MPB70 ELISA test suggested a seroprevalence of 4% in Serengeti lions (Cleaveland et al., 2005) but this result was not substantiated by mycobacterial culture of lion lesion material. The same ELISA test was used on 26 confirmed and highly developed M. bovispositive cases from KNP. Only 12/26 of these infected cases could be identified by this test (Unpublished data). Serologic assays for rapid detection of M. bovis infection have been described for a number of other free-ranging wildlife species (excluding lions) with promising results (Lyashchenko et al., 2008). Ante-mortal diagnosis of bovine TB in free-living wildlife is notoriously difficult, as animals need to be located and immobilized to collect blood for in vitro diagnostic assays (Grobler et al., 2002; Palmer and Waters, 2006). In the event of skin testing, free-ranging animals need to stay in captivity or be released, tracked and immobilized again after 72 h to assess skin response. Intradermal tuberculin testing is the cornerstone of successful tuberculosis-control schemes in cattle (de la Rua-Domenech et al., 2006) and it has been used on a number of wildlife species, with varying degrees of sensitivity and specificity (Cousins and Florisson, 2005). There are other limitations to the use of intradermal tuberculin tests in wildlife, since information is lacking concerning proper test sites for different species, concentrations, dosages and preparations of tuberculins to use and guidelines for interpretation (Miller, 2008). Regarding the use of tuberculin in carnivores, reports on the use of intradermal testing in domestic cats and dogs were not favourable because they do not react adequately to intradermally administered tuberculin and results were found to be unreliable (Backues, 2008; Miller, 2008). A large proportion of lions in KNP are serologically positive for FIVPle (Van Vuuren et al., 2003). Over and above the effect that infection with NTM may have on M. bovis specific immune responsiveness, hence on the results of skin testing in lions in the KNP, these test results may also be affected by co-infection with FIVPle. In domestic cats FIVFca critically impairs cell-mediated host responses and moderate to severe CD4+ depletion has been demonstrated in infected African lions (Roelke et al., 2006). A large proportion (53%) of bovine TB infected lions in areas of KNP where bovine TB is rife, are co-infected with FIVPle (Keet,
385
unpublished data). In cattle, co-infection with M. bovis and viruses that depress the function of lymphocytes and macrophages (such as bovine viral diarrhoea virus) has been shown to affect diagnostic assays for bovine TB (Charleston et al., 2001). Similarly diminished Mycobacterium tuberculosis-specific cell-mediated immune (CMI) responses have been documented in humans where HIV immune suppression was involved (Vermund and Yamamoto, 2007). In our study FIVPle status of animals was carefully considered as a potential confounding factor in bovine TB diagnosis by skin testing. The primary objective of this study was to determine whether the intradermal bovine tuberculin PPD test can identify lions naturally infected with M. bovis and which configuration of the intradermal skin test would achieve the highest sensitivity and specificity. Secondly the ability of NTM to evoke non-specific skin responses was evaluated. Concurrently, evaluation of possible attenuating effects of FIVPle on skin test response was required. 2. Materials and methods 2.1. Study population A cohort of M. bovis-infected lions (n = 52) was identified from a group of emaciated and ailing (n = 61/84), or repeatoffender lions (n = 23/84) reported on an ad hoc basis by field staff and tourists (a repeat-offender lion being a recognized habitual cattle killer and/or an emigrating vagrant). Cases were collected from anywhere within and immediately around the KNP, with the preponderance of compromised lions originating from bovine TB infected areas. Lions from the uninfected far north of the KNP (n = 11) were assessed as non-infected control animals. These two cohorts of lions (84 and 11) were euthanized and their status confirmed through a comprehensive necropsy and mycobacterial culture process. An additional group of control lions (n = 33) from the uninfected area of the park were randomly selected and skin tested. They were not destroyed to confirm their status as they were in prime condition and not repeat-offenders. These lions were considered as negative specimens because they originated from an area where bovine TB had not been diagnosed in buffaloes or lions despite concerted efforts. This lion sub-population had been separated from the infected sub-population by a buffer zone of approximately 120 km wide. The buffer zone had a very low prevalence of bovine TB in buffalo (1.5%: 0.4–4.0, 95% CI) (Rodwell et al., 2001) and no lion cases were diagnosed. 2.2. Test procedure Lions were immobilized with a combination of tiletamine and zolazepam (Zoletil1 100, Virbac Animal Health, Halfway House, South Africa) and transported to predator isolation units in Skukuza, the KNP headquarters. Venous blood samples were obtained from the medial or lateral saphenous veins immediately after anaesthesia. Age of the animal was estimated by examining dental attrition, its condition was scored and it was weighed. The animals were then tuberculin-tested according to the protocol described below and confined for 72 h. For skin testing
386
D.F. Keet et al. / Veterinary Microbiology 144 (2010) 384–391
two rectangular areas of skin of approximately 40 mm 40 mm on the lateral aspects of the proximal third of the neck, in close proximity to the cervical lymph nodes were clipped using a pro-cordless rechargeable clipper (Oster1 Professional Products, McMinnville, TN). The mane of male lions covering the prescribed area was initially cut with curved scissors and then clipped. A double thickness fold of the skin centred in the clipped area was measured with tuberculosis callipers (Hauptner1, Solingen, Germany). A dose of 0.2 ml (6000 I.U. = 0.2 mg) instead of 0.1 ml Bovine Tuberculin PPD 3000TM (Animal Sciences Group, Wageningen UR, Lelystad, Netherlands) was injected obliquely but strictly deep intradermally on the left-hand side with a 1 ml disposable syringe and a short 23 gauge needle (Terumo1 Corporation, Tokyo, Japan) resulting in a visible nodule in the skin. The same procedure was repeated with 0.2 ml (5000 I.U. = 0.2 mg) Avian Tuberculin PPD 2500TM (Animal Sciences Group, Wageningen UR, Lelystad, Netherlands) (Anon., 2009) on the right-hand side. The bovine tuberculin PPD dose exceeds the prescribed dosage (5000 I.U.) used in cattle but not the prescribed injected volume (Anon., 2009), and was decided on due to reports of failure to obtain repeatable reactions in domestic cats and lions when the standard bovine doses were used at various inoculation sites. After 72 h, the skin fold at injection sites was then remeasured. The skin at injection sites was also examined for visible reactions, palpated and described. Two approaches to skin test result interpretation were employed; in a comparative approach (Anon., 2009) where the reaction was considered positive if the increase in skin thickness at the bovine tuberculin PPD site of injection was 2 mm and larger than the reaction measured at the avian tuberculin PPD injection site (Single Intradermal Comparative Cervical Test: (SICCT)). The other interpretation (Single Intradermal Cervical Test: (SICT)) consisted of measuring both responses, but only considering the bovine tuberculin PPD response (de la Rua-Domenech et al., 2006) of 2 mm and larger for definitive diagnosis, this despite the response to avian tuberculin PPD possibly being larger (response to avian tuberculin PPD was measured for the purpose of the study). A classification of M. bovis reactors, M. bovis nonreactors, non-M. bovis reactors or non-M. bovis nonreactors was made when mycobacterial culture results and bovine tuberculin PPD results could be combined (Table 1). With the first 15 lions tested, a full thickness skin biopsy was then collected from the centre of the clipped injection sites using a 6 mm Ø biopsy punch (Kai1 Sterile
Disposable biopsy punch, Seki City, Japan) irrespective of the presence of a visible reaction. Biopsy sections were placed in marked Teflon sachets (Shandon1, Japan) and placed in 10% buffered formalin. 2.3. Bacteriology 2.3.1. Tissue collection After evaluating test reactions, the lions were euthanized by means of intravenous administration of a lethal dosage of sodium pentobarbitone (Euthapent1, Kyron, South Africa). Detailed necropsies were performed immediately after death was confirmed. All tissues collected were thinly sectioned with ultra sharp 130 mm trimming blades (Tissue-Tek1 Accu-Edge1, Sakura Finetek, Japan) for macroscopic inspection and then divided in two equal parts for preservation in 10% buffered formalin and culture, respectively. For bacterial isolation, lymph nodes were pooled in separate sterile containers according to body regions drained, namely cranial, thoracic, abdominal, peripheral and mammary lymph nodes. The cranial lymph node cluster consisted of the parotid, mandibular and retropharyngeal nodes. The thoracic cluster consisted of the tracheobronchial nodes while the abdominal cluster consisted of gastric, mesenteric, caecal and colic as well as hepatic nodes. The peripheral cluster consisted of superficial and deep cervical, axillary and accessory cervical, popliteal, superficial inguinal and iliosacral nodes. In addition, lesions suspect for tuberculosis in organs were collected separately. All tissue samples were frozen and stored at 20 8C until transferred to the laboratory for culture. 2.3.2. M. bovis culture and characterization Specimens were identified and cultured separately as previously described (Bengis et al., 1996). Differentiation between M. tuberculosis complex and non-tuberculous mycobacteria (NTM) was also aided by PCR amplification of a genome region specific for M. tuberculosis complex organisms (Cousins et al., 1991). Isolates classified as NTM were subjected to a series of cultural and biochemical tests including temperature growth test, pigment production assay, nitratase test, arylsulphatase test, and tellurite reduction test (Collins et al., 1985). 2.4. Histology Specimens for histological examination were fixed in 10% buffered formalin. Identified lymph node sections
Table 1 Intradermal test results (mean, standard deviation and variance) of 95 lions of which infection status was confirmed through mycobacterial culture and 33 additional lions that were skin tested only. Mean skin-fold thickness increase (mm)
M. bovis reactors (n = 45) Non-M. bovis reactors (n = 6) M. bovis non-reactors (n = 7) Non-M. bovis non-reactors from the bovine TB infected zone (n = 26) Non-M. bovis non-reactors from bovine TB free zone (n = 44)
Bovine site (6000 I.U. ppd.)
Avian site (5000 I.U. ppd.)
Bovine/avian difference (mm)
4.94 2.24 4.6 2.89 1.21 0.62 0.47 0.83
1.00 1.53 (2.34) 0.36 1.47 (2.16) 0.13 0.96 (0.93) 0.35 0.896 (0.79)
4.00 2.13 4.23 2.13 1.11 1.14 0.57 0.86
1.01 1.07 (1.31)
1.05 0.96 (0.92)
(5.02) (8.36) (0.38) (0.69)
0.27 0.66 (0.44)
(4.52) (4.53) (1.31) (0.75)
D.F. Keet et al. / Veterinary Microbiology 144 (2010) 384–391
387
were placed in marked Teflon sachets. They were embedded in paraffin wax, sectioned (3 mm) and stained with haematoxylin and eosine and Ziehl-Neelsen’s stain for light-microscopic examination. Macro- and histopathology results were consulted to support culture negative status of control lions and where M. bovis nonreactors and non-M. bovis reactors were encountered (Crawshaw et al., 2008).
Table 2a Sensitivity values for comparative and single skin test interpretations in an infected lion sub-population.
2.5. FIVPle serology
Table 2b Situational applied specificity (false positive rate for true negatives) calculated from culture negative cases from the infected zone.
An enzyme-linked immunosorbent assay (ELISA) using a puma lentivirus-derived synthetic peptide as coating antigen was used to test for FIVPle on serum samples of 72/ 95 lions necropsied (Van Vuuren et al., 2003). The sensitivity of the test when using the Western blot as the gold standard was 78.6% and the specificity 100% (Van Vuuren et al., 2003). 2.6. Data analysis Data analysis was directed at the observed increases in skin thickness in response to bovine and avian tuberculin PPD. Summary statistics included mean, standard deviation, and 95% confidence interval for the ratio of group means derived from the 95% confidence intervals for differences between group means. Various subgroup analyses were done and employed, i.e. Student’s two-sample t-test, Student’s paired t-test and analysis of covariance (ancova). In the presence of small sample sizes, the outcomes of the former tests were confirmed using nonparametric tests, ranksum tests, Wilcoxon’s matched pairs signed-ranks tests and ancova for ranks. Testing was directed at the 0.05 level of significance. 3. Results 3.1. Intradermal testing Skin test responses at bovine tuberculin PPD injection sites of 51/52 infected culture-positive lions were clearly visible and swellings were well demarcated but seldom indurated. In 45 of these lions, the difference between the two readings exceeded or equaled 2 mm. Pain and discomfort was impossible to detect as the lions were completely anaesthetized during examination. Redness and superficial necrosis with exudation was seen in only a small proportion of reactions. As reactions were typically soft, warm and oedematous, they were measured prior to being palpated. Swellings decreased considerably after palpation and first-time measuring. Reacting lions (M. bovis and non-M. bovis reactors) in infected areas (51/84) showed significantly larger responses to bovine tuberculin PPD than to avian tuberculin PPD (paired t-test; p < 0.0001) (Tables 1 and 2a). Forty-five infected lions were M. bovis reactors while 7 were M. bovis non-reactors when SICT criteria were applied (Tables 1 and 2b). Forty-two infected lions were M. bovis reactors while 10 were M. bovis non-reactors when SICCT criteria were applied. Lions residing in the non-infected area showed no response to bovine tuber-
SICT SICCT
Total
SE
FN
TP
52 52
0.865 0.808
7 10
45 42
SE: sensitivity; FN: false negative reactors; TP: true positive reactors.
SICT SICCT
Total
SP
TN
FP
32 32
0.8125 0.8125
26 26
6 6
SP: specificity; TN: true negative reactors; FP: false positive reactors.
Table 2c Specificity values for comparative and single skin test interpretations in a non-infected lion sub-population. Eleven of these cases were necropsied and cultured negative.
SICT SICCT
Total
SP
TN
FP
44 44
1.00 1.00
44 44
0 0
SP: specificity; TN: true negative reactors; FP: false positive reactors.
culin PPD but varying although negligible responses to avian tuberculin PPD (Tables 1 and 3). Avian responses in general in this far northern area of KNP were larger than bovine responses (Table 2c). 3.2. Bacteriology M. bovis was cultured from 52/95 (54.7%) skin tested and necropsied lions (Table 3). Of the 52 culture-positive lions, 40 were emaciated and ailing. In addition, NTM were cultured from 45/95 (47.4%) of the lions tested of which 24 animals were NTM positive only (Table 3), demonstrating the ubiquitous nature of these organisms in the KNP population. Although the group of NTM isolates was not fully characterized it was shown that it comprised both slow and fast-growing mycobacteria including Mycobacterium avium complex, Mycobacterium chelonae and Mycobacterium fortuitum. Of the 52 M. bovis-infected lions, 31 were infected with M. bovis only, while 21 were co-infected with NTM. Nineteen (19/95) cases yielded negative mycobacterial cultures. With respect to the mean bovine tuberculin PPD response, the NTM/M. bovis co-infected and M. bovis only groups did not differ significantly (two-sample t-test: p = 0.8829; 4.5 mm vs. 4.39 mm) (Table 3). As anticipated responses to avian tuberculin PPD were significantly stronger in animals co-infected with M. bovis and NTM than in those infected with NTM only (one-sided twosample t-test: p = 0.046; 1.19 mm vs. 0.5 mm) (Table 3). No significant difference was found between bovine tuberculin PPD skin responses of culture-positive emaciated lions (n = 40, mean = 4.4175 2.396) and culture-positive lions in fair condition (n = 12, mean 4.5 2.75) two-sample t-test with equal variances (t = 0.1012; p = 0.9198).
388
D.F. Keet et al. / Veterinary Microbiology 144 (2010) 384–391
Table 3 Intradermal test results (mean, standard deviation and variance) of 76 lions culturing positive for mycobacteriosis. Culture result
Mean skin-fold thickness increase (mm) Bovine site (6000 I.U. ppd.)
Avian site (5000 I.U. ppd.)
Bovine/avian difference (mm)
M. bovis only (n = 31)
4.39 2.34 5.48 4.5 2.67 7.13 0.8 1.11 1.24
0.68 1.38 1.91 1.19 1.62 2.63 0.51 0.99 0.98
3.72 1.95 3.79 3.45 2.67 7.13 1.0 1.27 1.61
M. bovis and NTM (n = 21) NTM only (n = 24)
3.3. FIV serology The proportion of FIVPle seropositive lions amongst the tested group was 47% (34/72; 95% CI, 35.4–59.0%); of these 82% (28/34) were ailing and compromised. Co-infection with FIVPle had no visible/measurable influence on skin response. Lions with concurrent FIVPle and M. bovis infections comprised 26% (19/72) of tested lions and had a mean increase in skin thickness of 4.84 2.3 mm. Eighteen M. bovis-infected lions tested negative for FIVPle and their average skin response was 3.86 2.33 mm. These groups did not differ significantly with respect to mean skin response (two-sample t-test: p = 0.2034; 4.84 mm vs. 3.86 mm). However, the 95% confidence interval for the mean response of the FIV co-infected group as a ratio of the mean response of the M. bovis only infected group is (0.855; 1.655), i.e. the response for the FIV co-infected group can be as low as 85.5% and as high as 165.5% of the M. bovis only group response. 3.4. Histology Skin biopsies excised from visible test reactions were all characterized by varying degrees of oedema with infiltration of macrophages and lymphocytes throughout. Reactions to bovine tuberculin PPD were more severe, however, due to a marked superficial and deep perivascular round cell infiltrate associated with severe oedema. This histological pattern is consistent with a delayed-type hypersensitivity reaction at both the avian and bovine tuberculin PPD injection sites. Skin biopsies excised from test sites of non-responding and most M. bovis culture negative lions (n = 4) showed no delayed-type hypersensitivity reaction. Microscopic lesion patterns observed in various organs were of a granulomatous nature consisting of macrophages, epithelioid cells, lymphoplasma cells and numerous neutrophils, suggestive of mycobacterial infection. Severe generalized lymphoid atrophy was sometimes seen in association with granulomatous lesions. Pulmonary lesions comprised of granulomatous interstitial pneumonitis or granulomatous bronchopneumonia often associated with bronchiectasis. Intestinal lesions showed mononuclear macrophage predominance suggestive of mycobacterial mural enteritis. Granulomatous osteitis, periosteitis and osteosis were found in most of the welldeveloped cases frequently associated with myositis. Granulomatous lesions containing eosinophils and parasitic remnants were often observed in lymph nodes and the intestinal tract. An important feature of all organ sections
from culture-positive cases was the apparent absence of acid-fast bacilli in granulomas when stained with Ziehl– Neelsen stain. Microscopic non-specific granulomatous lesions were also seen in the lymph nodes, lungs and intestinal tracts of three of the 11 control cases. All control cases cultured negative for M. bovis but NTM were isolated from 4 of these cases. 4. Discussion In this study we investigated the intradermal tuberculin testing of free-living African lions naturally exposed to infection with M. bovis. We demonstrated that administration of bovine tuberculin PPD identified a significant proportion (86.5%) of the lions naturally infected as confirmed by mycobacterial culture. Testing involved administration of a single bolus of double the conventional (0.1 mg) dosage of bovine and avian tuberculin PPD and interpretation of the skin tests by assessing comparative bovine versus avian tuberculin PPD elicited readings with single bovine tuberculin PPD readings. Co-infection with FIVPle and NTM were considered as potential immunecompromising events that could influence tuberculin skin test responses and consequently their interpretation. For this reason, results of mycobacterial culture as diagnostic gold standard and tuberculin skin tests were comparatively analyzed resulting in the proposal of adapted criteria for interpretation. Preferred and absolute validation and verification of a diagnostic trial test result requires detailed necropsy followed by culturing of all organ systems separately irrespective of the presence of macroscopic lesions or whether they reacted or not to the test (Goodchild and Clifton-Hadley, 2001). This is particularly valid for lions, where multiple organ systems are affected and lesions other than pulmonary ones, are not readily identifiable macroscopically (Keet et al., 2000). For a number of compelling reasons histology results were not considered to advance the gold standard as to increase sensitivity and/ or specificity; firstly the frequent culture-confirmed presence of NTM which may result in the possible sighting of acid-fast bacilli in suspicious and microscopic lesions, and secondly, the actual observed absence of acid-fast bacilli in histological lesions from lion cases confirmed positive for M. bovis and stained with Ziehl–Neelsen (Keet et al., 1996, in preparation). Reference to histology concerning the 11 control cases was merely supplementary to culture verification of absolute negative status. Results of histological examinations are addressed in a
D.F. Keet et al. / Veterinary Microbiology 144 (2010) 384–391
pathology-specific publication (Keet et al., in preparation). Suspicious lesions of a granulomatous nature but not revealing acid-fast bacilli could also have been caused by related bacteria, parasites or fungi (Crawshaw et al., 2008). Comprehensive culture of lymph node material from lesion-negative animals at necropsy is vital as it may further identify an additional proportion, however small, of those animals infected with M. bovis (Corner, 1994). The majority of cases in a highly developed outbreak of bovine TB in any maintenance host species are subclinical infections (De Vos et al., 2001). Similar to other species, the progress of bovine TB infection in lions is also typically slow, and the majority of infected lions encountered appear healthy while being subclinically infected (Keet, unpublished data). Tests that measure CMI are currently still the most sensitive to identify such cases and for determining prevalence and geographical distribution of M. bovis infection (Pollock et al., 2001). Applying principles of prescribed comparative test criteria (Anon., 2009), and using a 2-mm increase and difference between tuberculin PPD responses, presented us with a sub-optimal sensitivity. Retrospectively evaluating cut-off points for interpretation with the benefit of positive culture results, it was observed that using 2-mm increase as cut off, both sensitivity and specificity remained highest. A 2 mm increase is already considerable as it represents a 25– 63% increase from baseline value (Time0) skin thickness of lions (results not shown). As a point of reference it must be cited that an increased difference of 2 mm is expected in positive cattle when associated with clinical signs such as diffuse or extensive oedema, exudation, necrosis, pain or inflammation of the lymphatic vessels and nodes in the region or at least 4 mm in the absence of clinical signs. In the same reference it is further concluded, however, that any palpable or visible swelling should be considered positive in M. bovis-infected cattle herds (Anon., 2009). In our study, using comparative test criteria did not detect all infected lions, due to concurrent NTM infections which resulted in equal increases at both tuberculin PPD inoculation sites in some cases, resulting in a skin test negative interpretation due to possible cross-reactivity induced by the NTM. Such cross-reactive responses elicited by exposure to NTM often confounded the interpretation for M. bovis infection in cattle (Waters et al., 2006). We therefore propose that a 2-mm increase in skin-fold thickness specific to bovine tuberculin PPD is indicative of a positive M. bovis reactor, irrespective of the size of the avian tuberculin PPD response. We hypothesize that concurrent inoculation of avian tuberculin PPD in the likely presence of mixed mycobacterial infections, would reduce the effect of cross-reactivity rendering a more specific and dependable response at the bovine tuberculin PPD inoculation site. This unconventional diagnostic approach produced a 5.8% increase in sensitivity while the false positive rate for true negative cases remained unchanged (Tables 2a and 2b). The decision to use 0.2 mg of bovine tuberculin PPD instead of the standard prescribed 0.1 mg was established after previous limited efforts to diagnose bovine TB in ad hoc lion cases resulted in no visible or palpable response when 0.1 mg was administered intrapalpebrally, intrader-
389
mally or using the subcutaneous febrile test. Unfortunately, lions on which these early tests were done were not always destroyed to confirm disease status (Keet own observation). False negative skin reactions may transpire when an insufficient tuberculin PPD dose is injected (Monaghan et al., 1994). It was further argued that administering a relatively high concentration of bovine tuberculin PPD would induce a greater and more specific immune response after exposure to various other Mycobacterium spp. This hypothesis seems to be applicable in our study as well, because most lions with mixed infections responded better to bovine tuberculin PPD than to avian tuberculin PPD. However, we also observed that a significantly stronger avian tuberculin PPD response in the presence of M. bovis/NTM co-infection when compared to that of a single NTM infection, which strongly supports our diagnostic approach of only considering bovine tuberculin PPD results (Table 3). The prevalence of NTM isolated in this study is exceptionally high when compared to that observed in other species (Hughes et al., 2005; Backues, 2008; de Lisle et al., 2008). The large number of NTM isolates cultured from lion samples suggests that lions acquire a variety of these organisms from the environment and probably from prey animals as is suggested by 33/45 isolates from cranial lymph node clusters (Table 4). Free-living wildlife has been reported to be the source of various mycobacteria (Kollias et al., 1982) and prey species are a likely source for lions. In addition lions are also rather inactive and spend a large part of their existence lying down. This could increase exposure to soil mycobacteria. Also due to the generally violent nature of their existence (hunting and intraspecific competition), skin integrity is regularly breached with consequent trans-cutaneous infection. In this study 21 lions had NTM infected peripheral lymph nodes (Table 4). The above-mentioned behavioural patterns probably expose lions to infection with various NTMs more frequently and such immunological stimulation could elicit non-specific responses to bovine and avian tuberculin PPD, influencing diagnostic interpretation. The abundance of NTM in carnivores associated with M. bovis coinfection and particularly in lions should always be considered when validating any immunodiagnostic assays. The seven tuberculous non-reactors (bovine responses not reaching 2 mm but still larger than baseline measurement) from which M. bovis was cultured (Tables 1 and 2a), were all emaciated and possibly partially anergic, as M. bovis was cultured from all lymph node sample sets submitted. Two of these animals had characteristic pulmonary lesions. A state of anergy may develop in cattle with advanced or generalized bovine TB and (temporarily) in animals subjected to stress (Pollock et al., 2001) which may also be the case in these seven lions. The six non-tuberculous reactors (only bovine tuberculin PPD result considered) (Tables 1, 2a and 2b) comprised four animals in fair condition and two emaciated lions with distinctive macroscopic pulmonary lesions histologically suggestive of mycobacteriosis. Two of the four animals in fair body condition had NTM infections. All six nontuberculous reactors originated from the high and medium infected zones.
D.F. Keet et al. / Veterinary Microbiology 144 (2010) 384–391
390
Table 4 Distribution of non-tuberculosis mycobacteria (NTM) (n = 45 cases) and M. bovis (n = 52 cases) in different body regions of lions as determined through mycobacterial specific culture.
NTM isolations M. bovis isolations
Head
Thorax
Abdomen
Peripheral
Mammary
Skeletal
33 11
11 35
8 20
21 30
10 5
5 15
It was anticipated that the presence of FIVPle may reduce skin response to tuberculin similar to the situation of combined human immunodeficiency virus (HIV) and M. tuberculosis infection (Duncan et al., 1995; Klein et al., 1999). No evidence of this phenomenon could be found in our limited study, despite analysis of mostly emaciated lions. The majority of animals diagnosed as infected through mycobacterial culture, showed a positive bovine tuberculin PPD skin test (both SICCT and SICT interpretations) despite of the fact that FIVPle is supposed to cause moderate to severe depletion of CD4 subsets in lions (Roelke et al., 2006). This cell depletion could eventually influence CMI response, which provides the most accurate assessment of prevalence of bovine TB infection (Pollock et al., 2001). In non-reactor lions with advanced bovine TB, which may have been accelerated by FIVPle co-infection, serological tests might complement intradermal tuberculin tests. However, ultimately antibody levels may also diminish due to co-infection. Considering limitations of field investigations on an endangered species, we were able to demonstrate that intradermal tuberculin testing can be used to identify M. bovis infection in lions. We further demonstrated that possible effects of cross-reactivity can be overcome by administering both diagnostic antigens but only considering bovine tuberculin PPD response. Finally, no evidence could be found that FIVPle co-infection attenuated skin reactivity in the presence of M. bovis infection. Practical constraints, however, are the cost and effort involved when performing these tests in wild lions in a large conservation area. Conflict of interest The authors declare that they do not maintain any financial and personal relationships with other people or organizations that could have inappropriately influenced their work. Acknowledgements This research was predominantly funded the South African Veterinary Foundation and the Directorate of Veterinary Services, National Department of Agriculture, Forestry and Fisheries, South Africa. We are grateful to the field staff of the Directorate of Veterinary Services, Messrs. E.H. Dekker, T.K. Muchocho and J. Oosthuizen whom assisted with technical execution of the project, as well as the physical assistance of Tradesman Aides: Messrs C. Masinga, E. Ubisi, K. Nkuna, E. Munyamela, P. Baloyi in the management of the lion holding facilities, and assistance with necropsies; the ranger staff of SANParks (KNP) for reporting and assisting with the capture of some of the study lions; the Bacteriology Section of the ARC-Onder-
stepoort Veterinary Research Institute for mycobacterial culture. Drs. W. Botha and L. du Plessis from Golden Vetpath for histological examinations of referred cases. We also acknowledge Prof. N.P.J. Kriek for valuable advice and Mrs. A. Lourens from the J. F. Soga Library, Faculty Veterinary Science Library, University of Pretoria, for continuous support. D.F. Keet was the recipient of a doctoral bursary awarded by the University of Pretoria. This project was approved by the South African National Parks conservation and ethics committees as well as the Animal Use & Care Committee of the University of Pretoria. References Anon, 2009. Bovine tuberculosis. In: Manual of Diagnostic Tests and Vaccines for Terrestrial Animals, fifth edition. Office International des Epizooties, Paris (updated August 14, 2009), Chapter 2.4.7. Backues, K.A., 2008. Mycobacterial diseases in carnivores. In: Fowler, M.E., Miller, R.E. (Eds.), Zoo and Wild Animal Medicine. sixth edition. W.B. Saunders, Saint Louis, pp. 294–298. Bauer, H., Van Der Merwe, S., 2004. Inventory of free-ranging lions Panthera leo in Africa. Oryx 38, 26–31. Bengis, R.G., Kriek, N.P.J., Keet, D.F., Raath, J.P., De Vos, V., Huchzermeyer, H.F.A.K., 1996. An outbreak of bovine tuberculosis in a free-living African buffalo (Syncerus caffer-Sparrman) population in the Kruger National Park: a preliminary report. Onderstepoort J. Vet. Res. 63, 15– 18. Charleston, B., Hope, J.C., Carr, B.V., Howard, C.J., 2001. Masking of two in vitro immunological assays for Mycobacterium bovis (BCG) in calves acutely infected with non-cytopathic bovine viral diarrhoea virus. Vet. Rec. 149, 481–484. Cleaveland, S., Mlengeya, T., Kazwala, R.R., Michel, A., Kaare, M.T., Jones, S.L., Eblate, E., Shirima, G.M., Packer, C., 2005. Tuberculosis in Tanzanian wildlife. J. Wildl. Dis. 41, 446–453. Collins, C.H., Grange, J.M., Yates, M.D., 1985. Organization and Practice in Tuberculosis Bacteriology. Butterworth & Co Limited, London. Corner, L.A., 1994. Post mortem diagnosis of Mycobacterium bovis infection in cattle. Vet. Microbiol. 40, 53–63. Cousins, D.V., Florisson, N., 2005. A review of tests available for use in the diagnosis of tuberculosis in non-bovine species. Rev. Sci. Tech. 24, 1039–1059. Cousins, D.V., Wilton, S.D., Francis, B.R., 1991. Use of DNA amplification for the rapid identification of Mycobacterium bovis. Vet. Microbiol. 27, 187–195. Crawshaw, T.R., Griffiths, I.B., Clifton-Hadley, R.S., 2008. Comparison of a standard and a detailed postmortem protocol for detecting Mycobacterium bovis in badgers. Vet. Rec. 163, 473–477. de la Rua-Domenech, R., Goodchild, A.T., Vordermeier, H.M., Hewinson, R.G., Christiansen, K.H., Clifton-Hadley, R.S., 2006. Ante mortem diagnosis of tuberculosis in cattle: a review of the tuberculin tests, g-interferon assay and other ancillary diagnostic techniques. Res. Vet. Sci. 81, 190–210. de Lisle, G.W., Kawakami, R.P., Yates, G.F., Collins, D.M., 2008. Isolation of Mycobacterium bovis and other mycobacterial species from ferrets and stoats. Vet. Microbiol. 132, 402–407. De Vos, V., Bengis, R.G., Kriek, N.P.J., Michel, A., Keet, D.F., Raath, J.P., Huchzermeyer, H.F.K.A., 2001. The epidemiology of tuberculosis in free-ranging African buffalo (Syncerus caffer) in the Kruger National Park, South Africa. Onderstepoort J. Vet. Res. 68, 119–130. Duncan, L.E., Elliott, A.M., Hayes, R.J., Hira, S.K., Tembo, G., Mumba, G.T., Ebrahim, S.H., Quigley, M., Pobee, J.O.M., McAdam, K.P.W.J., 1995. Tuberculin sensitivity and HIV-1 status of patients attending a sexually transmitted diseases clinic in Lusaka, Zambia: a cross-sectional study. Trans. R. Soc. Trop. Med. Hyg. 89, 37–40.
D.F. Keet et al. / Veterinary Microbiology 144 (2010) 384–391 Goodchild, A.V., Clifton-Hadley, R.S., 2001. Cattle-to-cattle transmission of Mycobacterium bovis. Third International Conference on Mycobacterium bovis, M. bovis 2000, August 13–16, 2000, Cambridge, UK, Tuberculosis 81, 23–41. Grobler, D.G., Michel, A.L., De Klerk, L., Bengis, R.G., 2002. The gammainterferon test: its usefulness in a bovine tuberculosis survey in African buffaloes (Syncerus caffer) in the Kruger National Park. Onderstepoort J. Vet. Res. 69, 221–227. Hughes, M.S., Ball, N.W., McCarroll, J., Erskine, M., Taylor, M.J., Pollock, J.M., Skuce, R.A., Neill, S.D., 2005. Molecular analyses of mycobacteria other than the M. tuberculosis complex isolated from Northern Ireland cattle. Vet. Microbiol. 108, 101–112. Keet, D.F., Kriek, N.P.J., Penrith, M., Michel, A., Huchzermeyer, H., 1996. Tuberculosis in buffaloes (Syncerus caffer) in the Kruger National Park: spread of the disease to other species. Onderstepoort J. Vet. Res. 63, 239–244. Keet, D.F., Michel, A., Meltzer, D.G.A., 2000. Tuberculosis in free-ranging lions (Panthera leo) in the Kruger National Park. In: Proceedings of the South African Veterinary Association Biennial Congress, September 2000, Durban, pp. 232–241. Klein, R.S., Flanigan, T., Schuman, P., Smith, D., Vlahov, D., 1999. Criteria for assessing cutaneous anergy in women with or at risk for HIV infection. J. Allergy Clin. Immunol. 103, 93–98. Kollias, G.V., Thoen, C.O., Fowler, M.E., 1982. Evaluation of comparative cervical tuberculin skin testing in cervids naturally exposed to mycobacteria. J. Am. Vet. Med. Assoc. 181, 1257–1262. Lyashchenko, K.P., Greenwald, R., Esfandiari, J., Chambers, M.A., Vicente, J., Gortazar, C., Santos, N., Correia-Neves, M., Buddle, B.M., Jackson, R., O’Brien, D.J., Schmitt, S., Palmer, M.V., Delahay, R.J., Waters, W.R., 2008. Animal-side serologic assay for rapid detection of Mycobacterium bovis infection in multiple species of free-ranging wildlife. Vet. Microbiol. 132, 283–292. Michel, A.L., Coetzee, M.L., Keet, D.F., Mare´, L., Warren, R., Cooper, D., Bengis, R.G., Kremer, K., van Helden, P., 2009. Molecular epidemiology of Mycobacterium bovis isolates from free-ranging wildlife in South African game reserves. Vet. Microbiol. 133, 335–343.
391
Miller, M.A., 2008. Current diagnostic methods for tuberculosis in zoo animals. In: Fowler, M.E., Miller, R.E. (Eds.), Zoo and Wild Animal Medicine. sixth edition. W.B. Saunders, Saint Louis, pp. 10–19. Monaghan, M.L., Doherty, M.L., Collins, J.D., Kazda, J.F., Quinn, P.J., 1994. The tuberculin test. Vet. Microbiol. 40, 111–124. Morris, P.J., Thoen, C.O., Legendre, A.M., 1996. Pulmonary tuberculosis in an African lion (Panthera leo). J. Zoo. Wild. Med. 27, 392–396. Palmer, M.V., Waters, W.R., 2006. Advances in bovine tuberculosis diagnosis and pathogenesis: what policy makers need to know. Vet. Microbiol. 112, 181–190. Pollock, J.M., McNair, J., Welsh, M.D., Girvin, R.M., Kennedy, H.E., Mackie, D.P., Neill, S.D., 2001. Immune responses in bovine tuberculosis. Tuberculosis 81, 103–107. Rhodes, S.G., Gruffydd-Jones, T., Gunn-Moore, D., Jahans, K., 2008. Adaptation of IFN-gamma ELISA and ELISPOT tests for feline tuberculosis. Vet. Immunol. Immunopathol. 124, 379–384. Rodwell, T.C., Kriek, N.P., Bengis, R.G., Whyte, I.J., Viljoen, P.C., de Vos, V., Boyce, W.M., 2001. Prevalence of bovine tuberculosis in African buffalo at Kruger National Park. J. Wildl. Dis. 37, 258–264. Roelke, M.E., Pecon-Slattery, J., Taylor, S., Citino, S., Brown, E., Packer, C., VandeWoude, S., O’Brien, S.J., 2006. T-lymphocyte profiles in FIVinfected wild lions and pumas reveal Cd4 depletion. J. Wildl. Dis. 42, 234–248. Van Vuuren, M., Stylianides, E., Kania, S.A., Zuckerman, E.E., Hardy, J.W.D., 2003. Evaluation of an indirect enzyme-linked immunosorbent assay for the detection of feline lentivirus-reactive antibodies in wild felids, employing a puma lentivirus-derived synthetic peptide antigen. Onderstepoort J. Vet. Res. 70, 1–6. Vermund, S.H., Yamamoto, N., 2007. Co-infection with human immunodeficiency virus and tuberculosis in Asia. Tuberculosis 87, S18– S25. Waters, W.R., Palmer, M.V., Thacker, T.C., Payeur, J.B., Harris, N.B., Minion, F.C., Greenwald, R., Esfandiari, J., Anderson, P., McNair, L., Pollock, J.M., Lyashchenko, K.P., 2006. Immune responses to defined antigens of Mycobacterium bovis in cattle experimentally infected with Mycobacterium kansasii. Clin. Vaccine Immunol. 13, 611–619.