MOLECULAR METHODS TO DETECT MYCOPLASMA SPP. AND TESTUDINID HERPESVIRUS 2 IN DESERT TORTOISES (GOPHERUS AGASSIZII) AND IMPLICATIONS FOR DISEASE MANAGEMENT Author(s): Josephine Braun, Mark Schrenzel, Carmel Witte, Larisa Gokool, Jennifer Burchell, and Bruce A. Rideout Source: Journal of Wildlife Diseases, 50(4):757-766. Published By: Wildlife Disease Association DOI: http://dx.doi.org/10.7589/2013-09-231 URL: http://www.bioone.org/doi/full/10.7589/2013-09-231

BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, and environmental sciences. BioOne provides a sustainable online platform for over 170 journals and books published by nonprofit societies, associations, museums, institutions, and presses. Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of BioOne’s Terms of Use, available at www.bioone.org/page/ terms_of_use. Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries or rights and permissions requests should be directed to the individual publisher as copyright holder.

BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to critical research.

DOI: 10.7589/2013-09-231

Journal of Wildlife Diseases, 50(4), 2014, pp. 757–766 # Wildlife Disease Association 2014

MOLECULAR METHODS TO DETECT MYCOPLASMA SPP. AND TESTUDINID HERPESVIRUS 2 IN DESERT TORTOISES (GOPHERUS AGASSIZII ) AND IMPLICATIONS FOR DISEASE MANAGEMENT Josephine Braun,1,2 Mark Schrenzel,1 Carmel Witte,1 Larisa Gokool,1 Jennifer Burchell,1 and Bruce A. Rideout1 1 San Diego Zoo Global, Institute for Conservation Research, Wildlife Disease Laboratories, 15600 San Pasqual Valley Road, Escondido, California 92027-7000, USA 2 Corresponding author (email: [email protected])

ABSTRACT: Mycoplasmas are an important cause of upper respiratory tract disease (URTD) in desert tortoises (Gopherus agassizii) and have been a main focus in attempts to mitigate diseasebased population declines. Infection risk can vary with an animal’s population of origin, making screening tests popular tools for determining infection status in individuals and populations. To provide additional methods for investigating URTD we developed quantitative PCR (qPCR) assays specific for agents causing clinical signs of URTD: Mycoplasma agassizii, Mycoplasma testudineum, and Testudinid herpesvirus 2 (TeHV2) and tested necropsied desert tortoises housed at the Desert Tortoise Conservation Center in Las Vegas, Nevada, USA, as well as wild desert tortoises (n53), during 2010. Findings were compared with M. agassizii enzyme-linked immunosorbent assay (ELISA) data. Based on qPCR, the prevalence of M. agassizii was 75% (33/44) and the prevalence of TeHV2 was 48% (20/42) in the evaluated population. Both agents were also present in the wild tortoises. Mycoplasma testudineum was not detected. The M. agassizii ELISA and qPCR results did not always agree. More tortoises were positive for M. agassizii by nasal mucosa testing than by nasal flush. Our findings suggest that mycoplasmas are not the only agents of concern and that a single M. agassizii ELISA or nasal flush qPCR alone failed to identify all potentially infected animals in a population. Caution should be exercised in using these tests for disposition decisions. Key words: Desert tortoise, mycoplasma, qPCR, Testudinid herpesvirus 2, upper respiratory tract disease.

et al. 1994). Mitigation of mycoplasmaassociated URTD has since been a priority for the recovery of the species (USFWS 2011). Despite the important discovery of mycoplasma in desert tortoises, disease mitigation efforts have been hampered by difficulties in diagnosing URTD. Clinical signs alone are not specific for mycoplasmosis (Jacobson et al. 1995) and may be absent altogether in animals with confirmed infections (Brown et al. 1999). A monoclonal enzyme-linked immunosorbent assay (ELISA) to detect serum antibodies against Mycoplasma agassizii was developed to provide more-sensitive and specific diagnostics (Schumacher et al. 1993). This ELISA has a sensitivity of 98% and a specificity of 99% (Brown et al. 2002; Wendland et al. 2007). Serologic testing for antibodies against M. agassizii to determine past exposure became the primary tool for deciding the disposition of

INTRODUCTION

The desert tortoise (Gopherus agassizii) population in the Mojave Desert north and west of the Colorado River was declared threatened under the US Endangered Species Act in 1990 (USFWS 1994). The listing followed population declines of up to 68% over a 9-yr period between 1979 and 1989 (Sandmeier et al. 2009). A previously undescribed upper respiratory tract disease (URTD) was thought to be a major contributing factor in the population declines (Jacobson 1994; USFWS 1994), particularly in the western Mojave Desert (Jacobson et al. 1991; Homer et al. 1998). Studies including postmortem examinations, and electron microscopic and microbiologic analyses revealed that the associated lesions were caused by a novel Mycoplasma sp., Mycoplasma agassizii, infection (Jacobson et al. 1991; Schumacher et al. 1993; Brown 757

758

JOURNAL OF WILDLIFE DISEASES, VOL. 50, NO. 4, OCTOBER 2014

an animal for translocation. To minimize potential spread of M. agassizii (Wendland et al. 2007), only antibody-negative desert tortoises were translocated. The historic emphasis on M. agassizii infections overlooks the fact that other agents can also cause or contribute to URTD (Lange et al. 1989; Jacobson 1994; Jacobson et al. 1995; Brown et al. 2004). The lack of other diagnostic and research tools beyond serologic tests also limits our understanding of the relationship between antibody status, the presence of the agent, and the degree of shedding, all of which are important in mitigating disease risk. We aimed to broaden the research and management focus to include Mycoplasma testudineum and tortoise herpesvirus infections, both of which are also known to cause URTD (Lange et al. 1989; Brown et al. 2004). Our specific goals were to 1) design quantitative PCR (qPCR) assays with a high analytic sensitivity and specificity as tools to detect and quantify DNA of M. agassizii, M. testudineum, and Testudinid herpesvirus 2 (TeHV2); and 2) compare ELISA and qPCR test results. Having additional diagnostic and research tools will help us understand how the various pathogens known to cause or contribute to URTD impact desert tortoise population health and will facilitate better management decision-making. MATERIALS AND METHODS Study population

The Desert Tortoise Conservation Center (DTCC; Las Vegas, Nevada, USA) is a transitional holding facility and research center used in assisting in the recovery of the Mojave Desert tortoise populations. Since 2009, it has been operated by San Diego Zoo Global (SDZG) in partnership with the US Fish and Wildlife Service, the Bureau of Land Management, and the Nevada Department of Wildlife. The Center includes animals from a variety of origins including those found in the wild, surrendered pets, and those found in residential areas of Clark County, Nevada. We conducted health assessments on each individual in the population (including residents already at the DTCC in 2009 when San Diego Zoo Global began managing the population,

and every new arrival since that time) in consultation with an SDZG veterinarian. Assessments took place during the initial inventory or new arrival processing and on a continual cycle rotating through the population thereafter. Additionally, DTCC staff conducted individual health assessments on medical patients and as-needed based on daily observations. Animals were euthanized if they failed to thrive, did not respond to treatment, had a poor body condition score, or had suspected severe metabolic bone disease as determined by the health assessments. All euthanized tortoises and those found dead with minimal gross external and internal signs of autolysis were submitted for necropsy. All resident adult (straight median carapace length [MCL] $200 mm; n550) and immature (MCL 100–199 mm; n55) desert tortoises at the DTCC that were necropsied during 2010 were included in this study. Of the 55 (21 males; 30 females; 4 unknown sex) tortoises, 33 had been euthanized and 22 were found dead. In addition to these 55 DTCC resident desert tortoises, the study included three wild adult tortoises (2 males, 1 female) that were submitted to the DTCC after being found in their native habitat (per submitter) and euthanized within 1 day of arrival. One of these wild desert tortoises is also represented in a previous report (Jacobson et al. 2012). Forty-seven of the 58 tortoises were housed with the general DTCC population for 0– 575 days (mean 175.2 days; median 98 days). Eleven of the tortoises, including the three wild tortoises, were not mixed with the general population and spent 0–229 days (mean 25.1 days; median 7 days) in isolation. Sample collection

All sample processing and testing was carried out at the SDZG Wildlife Disease Laboratories except for the M. agassizii ELISA, which was done at University of Florida (UFL; Gainesville, FL, USA). Antemortem plasma samples for M. agassizii ELISA testing were collected and frozen (280 C) during inventory of the live animals in 2009–10. All blood samples were drawn from the subcarapacial sinus. Thirteen of the plasma samples used for the ELISA were drawn prior to 2009: six in 2008, four in 2007, two in 2003, and one in 2002. Tortoises at the DTCC were housed separately according to their M. agassizii ELISA status at arrival (positive, negative, and suspect). Samples for qPCR testing were collected postmortem. Tortoises were necropsied immediately after death or shortly after being found dead before advanced tissue autolysis

BRAUN ET AL.—PCR EVALUATION IN DESERT TORTOISE DISEASE MANAGEMENT

759

TABLE 1. Quantitative PCR (qPCR) assay parameters for three targeted agents: Mycoplasma agassizii, Mycoplasma testudineum, and Testudinid herpesvirus 2. Mycoplasma agassizii

qPCR target information qPCR target gene symbol

16S ribosomal RNA gene Sequence accession number NCBI reference sequence NR_ 025954.1 Amplicon length 69 base pairs

Mycoplasma testudineum

Testudinid herpesvirus 2

16S ribosomal RNA gene GenBank accession AY366210

Ribonucleotide reductase large subunit (UL39) GenBank accession AY338245

67 base pairs

65 base pairs

GGTGAGTAACACGTACTCAACCTACCT CGGCATTAGCCAAAGTTTCC 6FAM-ACAGACTGGAATAACCAMGBNFQ

GCCTTAGACAAGTTGGAGAAGCA

qPCR oligonucleotides Forward primer (59R39)

Reverse primer (59R39) Hydrolysis probe (59R39)

GGGTGAGTAACACGTACCTAATCTACCT CCGGTATTAGCAACGGTTTCC VIC- AAAGATCGGAACAACAATMGBNFQ

occurred (estimated,24 hr). During necropsy nasal flush, nasal mucosa, and tongue samples were taken, snap-frozen in liquid nitrogen, and stored at 280 C until further use. Nasal flushes were performed with 1-mL sterile saline split between both nasal cavities and flushed retrograde from the choana through the nares. The collected nasal flush was preserved in 1 mL of lysogeny broth (Fisher Scientific, Hanover Park, Illinois, USA) with glycerol (Fisher Scientific). The entire nasal mucosa lining one nasal cavity was removed aseptically. The side with a lesser amount of remaining nasal septum following longitudinal sectioning of the head was selected for sampling. The tongue tip was sampled. We were unable to obtain full sets of samples or results for all tortoises (n558): 44 tortoises had a nasal flush sample result, 51 tortoises had a nasal mucosa sample result, and 45 tortoises had a tongue sample result. In total, 43 tortoises had full sets of samples for M. agassizii and M. testudineum and 42 had full sets for TeHV2 evaluation. Instruments used for sample collection were sterilized prior to use and cleaned mechanically using soapy water, then soaked in bleach (sodium hypochlorite solution), and flamed with alcohol prior to sampling and between cases to avoid contamination. Enzyme-linked immunosorbent assay

Plasma samples were shipped on ice overnight to the Mycoplasma Research Laboratory, UFL, to determine M. agassizii-specific plasma

TCGGAGGGAATGTCTGGAAA 6FAM-ATGGGTAGTAAAAGAGGCMGBNFQ

antibody titers (Brown et al. 2002; Wendland et al. 2007). DNA extraction

DNA was extracted from 200 mL of nasal flush, from the entire nasal mucosa sample, and from approximately 2–5 g of tongue crosssection including epithelium. The tissue samples were cut into small pieces using a Petri dish and sterile scalpel and homogenized using silicon carbide beads (Bio Spec Products, Bartlesville, Oklahoma, USA) prior to extraction. The Qiagen DNeasy Blood and Tissue kit (Valencia, California, USA) was used for DNA extraction according to the manufacturer’s instructions. Quantitative PCR assays

Three assays were established using realtime qPCR chemistry with a hydrolysis probe (TaqManH probe, Applied Biosystems, Foster City, California, USA). These assays target M. agassizii, M. testudineum, and TeHV2, respectively. Primers and probes for qPCR were designed using Primer Express Software v2.1 (Applied Biosystems). Table 1 summarizes qPCR assay parameters and Table 2 summarizes qPCR assay validation for each of the three assays established. Reaction mixes were composed of 12.5 mL TaqMan Environmental Master Mix 2.0 (Applied Biosystems), 0.2 mL forward primer (100 mM), 0.2 mL reverse primer (100 mM), 0.025 mL hydrolysis probe (100 mM), 2.5 mL DNA, and RNase-free water to a final volume

760

JOURNAL OF WILDLIFE DISEASES, VOL. 50, NO. 4, OCTOBER 2014

TABLE 2. Quantitative PCR (qPCR) assay validation for three targeted agents: Mycoplasma agassizii, Mycoplasma testudineum, and Testudinid herpesvirus 2.

qPCR validationa

Calibration curve – slope Calibration curve – y intercept Amplification efficiency R2 of calibration curve Linear dynamic range (copy numbers) Limit of detection (copy numbers) in 40-cycle qPCR Repeatability Maximum intra-assay variation (SD between Cq of templates) Maximum intra-assay coefficient of variance (% between Cq of templates) Reproducibility Maximum interassay variation: SD between %PCR efficiency Maximum interassay variation: SD between Cq of controls Maximum interassay coefficient of variance (% between Cq of controls) a

Mycoplasma agassizii

Mycoplasma testudineum

Testudinid herpesvirus 2

23.46 44.64 95% 0.99 5–100,000

23.46 47.04 95% 0.97 5–100,000

23.4 43.22 97% 0.99 5–100,000

50

100

16

0.97

0.9

1.04

2.46

2.09

1.49

0.54

0.6

1.49

0.33

0.87

0.05

0.93

2.18

0.15

Cq 5 quantification cycle.

of 25 mL. The thermal profile was 2 min at 50 C, 10 min at 95 C, and 45 cycles (during establishment of the assays; 40 cycles running the samples) of 15 sec at 95 C and 1 min at 60 C. Samples were analyzed on the 7900 HT Fast Real-Time PCR System (Applied Biosystems). Protocols were established and optimized using serial dilutions of a synthetic custom gene within a pIDTSmart plasmid cloning vector (custom gene plasmid). The custom gene included the three DNA segments encompassed by the three primer pairs and was synthesized and inserted into the plasmid by IDT (Integrated DNA Technologies, San Diego, California, USA). The nine-stepped serial dilutions ranged from 5 to 105 copies of the custom gene plasmid. The nine steps included a constant logarithmic serial dilution of 101 to 105 copies and additional partial steps of 5, 16, 25, and 50 copies to narrow the limit of detection. Quantification cycle (Cq, also known as threshold cycle) values were calculated based on the baseline corrected fluorescence (DRN). To test for analytic specificity, in silico basic local alignment search tool (BLAST) analyses were carried out (NCBI, http://www.ncbi.nlm. nih.gov/blast/Blast.cgi). Additionally, both the M. agassizii and M. testudineum assays were cross-tested against the following strains of

Mycoplasma obtained through the National Institutes of Health (NIH) Biodefense and Emerging Infections Research Resources Repository, National Institute of Allergy and Infectious Diseases, NIH: Mycoplasma anatis, Strain 1340, NR-3857; Mycoplasma pulmonis, Strain Ash (PG 34), NR-3858; Mycoplasma gallisepticum, Strain PG 31, NR-3863; Mycoplasma canis, Strain PG 14, NR-3865; and Mycoplasma maculosm, Strain PG 15, NR3867. Both assays were also cross-tested against each other using M. agassizii strain PS6 (ATCC 700616) (Brown et al. 2001) and M. testudineum strain BH29 (ATCC 700618) (Brown et al. 1995) obtained from the American Type Culture Collection (ATCC, Manassas, Virginia, USA). We used 96-well optical reaction plates (Applied Biosystems) covered by MicroAmp Optical Adhesion Films (Applied Biosystems). Each plate contained triplicates of each DNA sample, triplicates of five-stepped logarithmic standard dilutions of the synthetic DNA (101 to 105 copies), duplicates of no template controls (NTC), and exogenous internal control positive controls (TaqMan Exogenous Internal Positive Control Reagents; Applied Biosystems). The standard dilutions were used as positive controls. In the NTCs no Cq value was detected within the maximum cycles run

BRAUN ET AL.—PCR EVALUATION IN DESERT TORTOISE DISEASE MANAGEMENT

(40- or 45-cycle qPCR). The qPCR results were considered positive if $2 values out of the sample triplicates had a Cq of #40. At the 40cycle cut-off the detection limit of the assays was 50 copies for M. agassizii, .100 copies for M. testudineum, and 16 copies for TeHV2.

761

poor, 0.01–0.25slight, 0.21–0.45fair, 0.41–0.60 5moderate, 0.61–0.805substantial, and 0.81– 15almost perfect. RESULTS qPCR assays

qPCR sample analysis

Using the assays described above, nasal flush (NF), nasal mucosa (NM), and tongue samples were tested for TeHV2 ribonucleotide reductase large subunit (UL39) gene DNA. The NF and NM samples were additionally tested for the M. agassizii and M. testudineum 16S ribosomal RNA gene. Prevalence estimations

Prevalence of the different agents along with the exact 95% CI was determined in the group of DTCC resident tortoises (i.e., not including the three wild individuals) by dividing the number found positive by the total number of tortoises with evaluated tissues. Tortoises were considered positive for M. agassizii, M. testudineum, and TeHV2 if any tested sample was positive by qPCR. Tortoises were considered negative if all samples (nasal flush, nasal mucosa, and tongue) were negative by qPCR. Only animals that could be classified as positive or negative based on these criteria were included in prevalence estimates (11 tortoises were excluded from M. agassizii prevalence calculation and six were excluded from M. testudineum prevalence calculation due to missing samples; one tortoise was excluded from TeHV2 prevalence estimations due to an equivocal tongue result). Prevalence of M. agassizii antibodies was determined in the same population based on the ELISA (titer,32 negative; titer 32 suspect; titer.32 positive. Evaluation of tests

The kappa statistic was calculated as a measure of agreement beyond chance between the detection of M. agassizii antibodies (by ELISA) and DNA (by qPCR); the detection of the Mycoplasma spp. (M. agassizii, M. testudineum) in different samples (nasal flush and nasal mucosa); and the detection of TeHV2 in the nasal cavity (nasal flush, nasal mucosa) and oral cavity (tongue) samples. Values for kappa range from 21 to 1, where 215perfect disagreement, 05agreement due to chance alone, and 15perfect agreement. The strength of agreement was interpreted according to Landis and Koch (1977): #05

The qPCR parameters are summarized in Tables 1 and 2. The M. agassizii and M. testudineum assays each target the 16S ribosomal RNA gene, and the TeHV2 assay targets the ribonucleotide reductase large subunit (UL39). The assays have a 100% in silico target specificity in a BLAST search. Specificity of the Mycoplasma assays was positively verified against strains of all five species of Mycoplasma cross-tested (M. anatis, M. pulmonis, M. gallisepticum, M. canis, and M. maculosm). The assays had $95% amplification efficiency as calculated from the slope and a detection limit of 50 copies for M. agassizii, .100 copies for M. testudineum, and 16 copies for TeHV2 in a 40-cycle qPCR. The average Cq values for two sets of triplicates run on separate plates (n56) were 27.3 for 100,000 copies, 30.9 for 10,000 copies, 33.8 for 1,000 copies, 37.6 for 100 copies, and 38.5 for 50 copies for M. agassizii; 29.6 for 100,000 copies, 33.1 for 10,000 copies, 36.4 for 1,000 copies, and 40.0 (of which two of six wells had a Cq.40.0) for 100 copies for M. testudineum; and 26.2 for 100,000 copies, 29.4 for 10,000 copies, 32.9 for 1,000 copies, 36.5 for 100 copies, 37.1 for 50 copies, 38.2 for 25 copies, and 38.8 for 16 copies for TeHV2. The maximum intra- and interassay variations of ,1.5 SD correspond to a variability of 0–5%, expressing repeatability and reproducibility of the data (Bustin 2000). Pathogen prevalence

The prevalence of M. agassizii (based on a combination of all NF and NM qPCR positives) was 75% (33/44; 95% CI: 60–87) among evaluated tortoises. Over half of the DTCC necropsied tortoises were

762

JOURNAL OF WILDLIFE DISEASES, VOL. 50, NO. 4, OCTOBER 2014

ELISA positive for M. agassizii antibody (30/54; 56%; 95% CI: 41–69). One tortoise had an unknown ELISA status due to loss of its identification tag. All tortoises tested for M. testudineum were negative (n549; based on a combination of all NF and NM qPCR positives). TeHV2 was detected in 20/42 (48%; 95% CI: 32–64) DTCC tortoises (based on a combination of all tongue, NM, and NF qPCR positives). Thirty-one qPCR M. agassizii-positive tortoises were also tested for TeHV2. Of these, almost equal numbers were positive (16/31; 52%; 95% CI: 33–70) and negative (15/31; 48%; 95% CI: 30–67) for TeHV2. Similarly, out of the 11 qPCR M. agassiziinegative tortoises, almost equal numbers were positive (5/11; 45%; 95% CI: 17–77) and negative (6/11; 55%; 95% CI: 23–83) for TeHV2. Thus, the presence of M. agassizii was not significantly associated with the presence of TeHV2 (Prevalence ratio51.07; 95% CI: 0.74–1.53; P50.73). The three wild desert tortoises all had an unknown M. agassizii ELISA status. One (1/3) was positive for M. agassizii (NM qPCR). Two (2/2) tortoises were negative for M. testudineum (based on a combination of NM and NF qPCR negatives) and positive for TeHV2 (based on a combination of tongue and NM qPCR positives). One of these positive TeHV2 tests was previously reported (Jacobson et al. 2012). Test evaluation Mycoplasma agassizii: Among the 43 tortoises with ELISA, NF qPCR, and NM qPCR testing results, the NM identified the largest number of M. agassizii qPCRpositive animals (n532; 74%). Twenty-six tortoises (60%) were positive by NF qPCR and 23 (53%) by ELISA. Overall, 65% (28/43) of results from antibody tests agreed with the presence or absence of DNA in both the NF and in the NM assays (all assays negative59; all positive519). The remaining 15/43 samples had various combinations of negative and positive ELISA and qPCR results (Tables 3 and 4).

TABLE 3. Comparative performance of assays and sample types for Mycoplasma agassizii (n543).a % agreement (kappa)b

Positive

Negative

19 4

7 13

74 (0.48)

21 2

11 9

70 (0.37)

26 0

6 11

86 (0.69)

ELISA NF qPCR Positive Negative NM qPCR Positive Negative NF qPCR NM qPCR Positive Negative a

Only tortoises with full samples sets were included. ELISA 5 enzyme-linked immunosorbent assay; NF 5 nasal flush; qPCR 5 quantitative PCR; NM 5 nasal mucosa.

b

The kappa statistic was used to determine agreement beyond chance. The strength of the statistic is interpreted according to Landis and Koch (1977) as follows: #0 5 poor, 0.01–0.2 5 slight, 0.21–0.4 5 fair, 0.41–0.60 5 moderate, 0.61–0.80 5 substantial, and 0.81–1 5 almost perfect.

Kappa statistics for all assay and sample comparisons are provided in Tables 3 and 4. There was substantial agreement between the NF qPCR and NM qPCR results; all of the animals that had a positive NF result were also positive when testing the NM. Thus, the animals with discordant qPCR results represented those with a negative NF and a positive NM result (n56). There was fair agreement between the ELISA and NM qPCR results and moderate agreement between the M. agassizii ELISA and NF qPCR results (Table 3). Testudinid herpesvirus 2

All tortoises positive for TeHV2 (n520) had positive tongue samples, with the exception of one animal positive only by NM qPCR. There was substantial agreement between the NF qPCR and NM qPCR results, moderate agreement between the tongue qPCR and NM qPCR results, and fair agreement between the tongue qPCR and NF qPCR results

BRAUN ET AL.—PCR EVALUATION IN DESERT TORTOISE DISEASE MANAGEMENT

TABLE 4. Comparative performance of assays and sample types for Testudinid herpesvirus 2 (n542).a

Positive

Negative

% agreement (kappa)b

NF qPCR Tongue Positive Negative

7 0

12 23

70 (0.39)

11 1

8 22

79 (0.55)

7 0

5 30

88 (0.67)

NM qPCR Tongue Positive Negative NF qPCR NM qPCR Positive Negative a

Only tortoises with full samples sets were included. NF 5 nasal flush; qPCR 5 quantitative PCR; NM 5 nasal mucosa.

b

The kappa statistic was used to determine agreement beyond chance. The strength of the statistic is interpreted according to Landis and Koch (1977) as follows: #0 5 poor, 0.01–0.2 5 slight, 0.21–0.4 5 fair, 0.41–0.60 5 moderate, 0.61–0.80 5 substantial, and 0.81–1 5 almost perfect.

(Table 4). Results of test comparisons are further described in Tables 3 and 4. DISCUSSION

Upper respiratory tract disease is thought to have contributed to past population declines of wild desert tortoises in the Mojave Desert and remains a concern. The ELISA for detecting antibodies to M. agassizii, one of the primary agents associated with URTD, has greatly facilitated studies of the prevalence of antibody to this agent and has advanced our understanding of the geographic distribution and levels of exposure within populations. This study adds new tools for investigating URTD by establishing the first hydrolysis probe (TaqMan probe)-based real-time quantitative PCR assays for three agents associated with clinical signs of URTD: M. agassizii (Brown et al. 1994), M. testudineum (Brown et al. 2004), and Testudinid herpesvirus 2 (Lange et al.

763

1989; Mu¨ller et al. 1990). These assays are specific for these three agents, respectively, based on BLAST analyses against all known strains of organisms within the NCBI BLAST database, and verified by cross-testing the oligonucleotides against select relevant and closely related strains. Their high specificity reduces the probability of false positive results that may result from the amplification of unrelated products compared to conventional and SYBR green–based PCR assays. The cutoff value of Cq40 to determine positive samples was based on Cq values .40 to be reportedly suspect (Bustin et al. 2009). The assays have a high sensitivity with a limit of detection of 50 copies for M. agassizii, .100 copies for M. testudineum, and 16 copies for TeHV2 at 40 amplification cycles. Applying these new tests to our study population revealed a high prevalence of both M. agassizii and TeHV2 in the necropsied population of DTCC resident tortoises, reinforcing the view that investigations of URTD should not focus exclusively on M. agassizii infections. These two agents were also detected in the wild desert tortoises tested. The failure to detect M. testudineum, another agent of URTD, was surprising but may simply reflect a limited or disjunct distribution of the agent. Correlating the M. agassizii ELISA and qPCR results revealed only fair to moderate agreement between test results. Twenty-six percent (11/43) of tortoises were antibody-negative yet tested positive with NM qPCR for M. agassizii. Conversely, 5% (2/43) of tortoises were antibodypositive yet tested NM qPCR negative for M. agassizii. The antibody negativity in the former group could result from the tortoises being tested prior to seroconversion, which can take 6–8 wk to develop (Schumacher et al. 1993; Jacobson et al. 1995; Wendland et al. 2007), or from failure to develop an immune response (Jacobson et al. 1995). In the latter group, a negative qPCR result in an antibody-

764

JOURNAL OF WILDLIFE DISEASES, VOL. 50, NO. 4, OCTOBER 2014

positive tortoise could mean that the number of organisms was below the detection limit of the qPCR. Although the negative qPCR might also suggest clearance of the agent from the nasal cavity by the time of testing, clearance of mycoplasmal infections is thought to be a rare event (Simecka 1992; Jacobson et al. 1995; Brown et al. 1999). The time lag between the collection of serum and tissue samples may have influenced the overall agreement of these tests. PCR testing occurred 0 days to 7.6 yr (average 1.3 years, SD 1.7) after samples for the initial M. agassizii ELISA were collected. During this time period tortoises were housed in pens with individuals of identical ELISA test results; however, it is possible that exposure to a misclassified, positive individual could have elicited an undetected seroconversion event. Comparison of M. agassizii ELISA and qPCR results from samples taken on the same day would be the most informative for assessing test agreement; unfortunately, we did not have the biomaterials available. It is therefore possible that some antibody-negative tortoises were in the process of seroconverting and would have been ELISA positive at the time of postmortem sampling for qPCR. However, in that case it is also possible that these individuals would have tested qPCR positive at the time of plasma sampling. Fourteen percent (6/43) of tortoises were NF qPCR negative yet tested NM qPCR positive for M. agassizii. This may be explained by the nasal flush failing to flush out the organisms. The close association of mycoplasmas with the apical surface of epithelial cells, and the anatomic structure of the ventral niche of the nasal cavity, may render small numbers of organisms difficult to isolate with nasal flush procedures. Alternatively, the flushed mycoplasmas may have been present in numbers below the detection limit of the qPCR. Thus, similar to previously published data in gopher tortoises (Gopherus polyphemus), the pres-

ence of serum antibodies against M. agassizii in desert tortoises only partially correlates with the presence of Mycoplasma DNA in nasal flushes (McLaughlin et al. 2000). Furthermore, absence of M. agassizii in the nasal flush does not rule out the presence of M. agassizii in the nasal cavity of the tortoise. While NM qPCR identified the most positive tortoises, it only had moderate agreement with the NF and fair agreement with ELISA. The discordance between the presence or absence of antibodies and the presence or absence of DNA for M. agassizii indicates that a single test result may not be effective in achieving the management goal of excluding M. agassizii-infected animals from release cohorts: 1) Antibody-negative and qPCR-positive carriers of M. agassizii may be mixed in with true negative animals; and 2) antibody-positive animals may be treated as infectious and excluded from a pool of potential translocatees despite lack of detectable M. agassizii DNA in their nasal cavity. In conclusion, we developed qPCR assays to detect M. agassizii, M. testudineum, and TeHV2. The assays have a 100% in silico and experimental analytic specificity and high sensitivity of 50 copies, .100 copies, and 16 copies, respectively. Using these assays, we determined a 75% prevalence of M. agassizii and a 48% prevalence of TeHV2 in DTCC resident desert tortoises (MCL$100 mm) necropsied in 2010. Both agents were detected in DTCC residents as well as in wild desert tortoises. This supports previously published evidence that both M. agassizii and TeHV2 are agents of URTD in captive and wild desert tortoises (Jacobson et al. 1991; Jacobson et al. 2012). These newly established tests will open new areas of research by allowing sensitive detection and quantification of agent DNA. The tests can be used to quantify these microorganisms and evaluate their abundance with respect to presence and degree of clinical signs and postmortem

BRAUN ET AL.—PCR EVALUATION IN DESERT TORTOISE DISEASE MANAGEMENT

lesions. However, the relatively poor correlation between tests suggests that individual test results should not be used for animal disposition decisions in desert tortoise translocations. Additional studies would be required to evaluate the positive and negative predictive values or likelihood ratios of various test combinations before testing recommendations could be made. ACKNOWLEDGMENTS

This study was supported by the US Fish and Wildlife Service and the Ellen Browning Scripps Foundation. We thank Mary Schwartz for helping develop the real-time PCR assays. We thank the Desert Tortoise Conservation Center staff and San Diego Zoo Global veterinarian Nadine Lamberski for their help and support. LITERATURE CITED Brown DR, Crenshaw BC, McLaughlin GS, Schumacher IM, McKenna CE, Klein PA, Jacobson ER, Brown MB. 1995. Taxonomic analysis of the tortoise mycoplasmas Mycoplasma agassizii and Mycoplasma testudinis by 16S rRNA gene sequence comparison. Int J Syst Bacteriol 45:348–350. Brown DR, Merritt JL, Jacobson ER, Klein PA, Tully JG, Brown MB. 2004. Mycoplasma testudineum sp. nov., from a desert tortoise (Gopherus agassizii) with upper respiratory tract disease. Int J Syst Evol Microbiol 54:1527–1529. Brown DR, Schumacher IM, McLaughlin GS, Wendland LD, Brown MB, Klein PA, Jacobson ER. 2002. Application of diagnostic tests for mycoplasmal infections of desert and gopher tortoises, with management recommendations. Chelonian Conserv Biol 4:497–507. Brown MB, Berry KH, Schumacher IM, Nagy KA, Christopher MM, Klein PA. 1999. Seroepidemiology of upper respiratory tract disease in the desert tortoise in the western Mojave Desert of California. J Wildl Dis 35:716–727. Brown MB, Brown DR, Klein PA, McLaughlin GS, Schumacher IM, Jacobson ER, Adams HP, Tully JG. 2001. Mycoplasma agassizii sp. nov., isolated from the upper respiratory tract of the desert tortoise (Gopherus agassizii) and the gopher tortoise (Gopherus polyphemus). Int J Syst Evol Microbiol 51:413–418. Brown MB, Schumacher IM, Klein PA, Harris K, Correll T, Jacobson ER. 1994. Mycoplasma agassizii causes upper respiratory tract disease in the desert tortoise. Infect Immun 62:4580– 4586.

765

Bustin SA. 2000. Absolute quantification of mRNA using real-time reverse transcription polymerase chain reaction assays. J Mol Endocrinol 25:169– 193. Bustin SA, Benes V, Garson JA, Hellemans J, Huggett J, Kubista M, Mueller R, Nolan T, Pfaffl MW, Shipley GL, et al. 2009. The MIQE guidelines: Minimum information for publication of quantitative real-time PCR experiments. Clin Chem 55:611–622. Homer BL, Berry KH, Brown MB, Ellis G, Jacobson ER. 1998. Pathology of diseases in wild desert tortoises from California. J Wildl Dis 34:508–523. Jacobson ER. 1994. Causes of mortality and disease in tortoises: A review. J Zoo Wildlife Med 25:2– 17. Jacobson ER, Gaskin JM, Brown MB, Harris RK, Gardiner CH, Lapointe JL, Adams HP, Reggiardo C. 1991. Chronic upper respiratory tract disease of free-ranging desert tortoises (Xerobates agassizii). J Wildl Dis 27:296–316. Jacobson ER, Brown MB, Schumacher IM, Collins BR, Harris RK, Klein PA. 1995. Mycoplasmosis and the desert tortoise (Gopherus agassizii) in Las Vegas Valley, Nevada. Chelonian Conserv Biol 1:279–284. Jacobson ER, Berry KH, Wellehan JF Jr, Origgi F, Childress AL, Braun J, Schrenzel M, Yee J, Rideout B. 2012. Serologic and molecular evidence for Testudinid herpesvirus 2 infection in wild Agassiz’s desert tortoises, Gopherus agassizii. J Wildl Dis 48:747–757. Lange H, Herbst W, Wiechert JM, Schleter T. 1989. Elektronenmikroskopischer Nachweis von Herpesviren bei einem Massensterben von Griechischen Landschildkro¨ten (Testudo hermanni) und Vierzehenschildkro¨ten (Agrionemzs horsfieldii). Tierarztliche Praxis 17:319–321. Landis J, Koch G. 1977. The measurement of observer agreement for categorical data. Biometrics 33:159–174. McLaughlin GS, Jacobson ER, Brown DR, McKenna CE, Schumacher IM, Adams HP, Brown MB, Klein PA. 2000. Pathology of upper respiratory tract disease of gopher tortoises in Florida. J Wildl Dis 36:272–283. Mu¨ller M, Sachsse W, Zangger N. 1990. Herpesvirus-Epidemie bei der Griechischen (Testudo hermanni) und der Maurischen Landschildkro¨te (Testudo graeca) in der Schweiz. Schweizer Archiv fu¨r Tierheilkunde 132:199–203. Sandmeier FC, Tracy CR, Dupre S, Hunter K. 2009. Upper respiratory tract disease (URTD) as a threat to desert tortoise populations: A reevaluation. Biol Conserv 142:1255–1268. Schumacher IM, Brown MB, Jacobson ER, Collins BR, Klein PA. 1993. Detection of antibodies to a pathogenic mycoplasma in desert tortoises (Gopherus agassizii) with upper respiratory tract disease. J Clin Microbiol 31:1454–1460.

766

JOURNAL OF WILDLIFE DISEASES, VOL. 50, NO. 4, OCTOBER 2014

Simecka JW, Davis JK, Davidson MK, Ross SE, Stadtlander CTK-H, Cassell GH. 1992. Mycoplasma diseases of animals. In: Mycoplasmas: Molecular biology and pathogenesis, Maniloff J, McElhaney R, Finch L, Baseman J, editors. American Society for Microbiology, Washington, DC, pp. 391– 415. US Fish and Wildlife Service (USFWS). 1994. Desert tortoise (Mojave population) recovery plan. US Fish and Wildlife Service, Portland, Oregon, 73 pp. US Fish and Wildlife Service (USFWS). 2011. Revised recovery plan for the Mojave population

of the desert tortoise (Gopherus agassizii). US Fish and Wildlife Service, Pacific Southwest Region, Sacramento, California, 222 pp. Wendland LD, Zacher LA, Klein PA, Brown DR, Demcovitz D, Littell R, Brown MB. 2007. Improved enzyme-linked immunosorbent assay to reveal Mycoplasma agassizii exposure: A valuable tool in the management of environmentally sensitive tortoise populations. Clin Vaccine Immunol 14:1190–1195. Submitted for publication 4 September 2013. Accepted 2 June 2014.