The Veterinary Journal xxx (2013) xxx–xxx
Contents lists available at ScienceDirect
The Veterinary Journal journal homepage: www.elsevier.com/locate/tvjl
Quantitative real-time PCR for detection of neurotoxin genes of Clostridium botulinum types A, B and C in equine samples Amy L. Johnson ⇑, Susan C. McAdams-Gallagher, Raymond W. Sweeney Department of Clinical Studies, New Bolton Center, University of Pennsylvania School of Veterinary Medicine, Kennett Square, PA 19348, USA
a r t i c l e
i n f o
Article history: Accepted 18 October 2013 Available online xxxx Keywords: Botulism Equine Real-time quantitative PCR Mouse bioassay
a b s t r a c t Botulism in horses in the USA is attributed to Clostridium botulinum types A, B or C. In this study, a duplex quantitative real-time PCR (qPCR) for detection of the neurotoxin genes of C. botulinum types A and B, and a singleplex qPCR for detection of the neurotoxin gene of C. botulinum type C, were optimized and validated for equine gastrointestinal, faecal and feed samples. The performance of these assays was evaluated and compared to the standard mouse bioassay (MBA) using 148 well-characterized samples, most of which were acquired from a repository of veterinary diagnostic samples from cases of botulism: 106 samples positive for C. botulinum (25 type A, 27 type B, 28 type C, 1 type D and 25 type E) and 42 negative samples. The sensitivities of the qPCR assays were 89%, 86% and 96% for C. botulinum types A, B and C, respectively. The overall sensitivity of the mouse bioassay for types A, B and C was 81%. The specificities of the qPCR assays were 99–100% and the specificity of the mouse bioassay was 95%. Ó 2013 Elsevier Ltd. All rights reserved.
Introduction Botulism is caused by Clostridium botulinum neurotoxin (BoNT) and manifested by progressive flaccid paresis and cranial nerve deficits, such as dysphagia (Meyer, 1956; Galey, 2001). There are eight serotypes of C. botulinum, distinguished by the unique properties of their toxins (Simpson, 2004). C. botulinum type B causes most (>85%) cases of equine botulism in the USA and is endemic in the mid-Atlantic region and Kentucky (Whitlock and McAdams, 2006). Conversely, C. botulinum type A is endemic in states west of the Mississippi River (Johnson et al., 2010). Intoxication by C. botulinum type C is associated with carrion contamination of feed and occurs sporadically (Kinde et al., 1991). While only types A, B and C have been reported in horses in the USA, types C and D are most common in Europe (Gerber et al., 2006). Diagnosis of botulism often is based on compatible history and clinical signs, since laboratory confirmation is difficult. The mouse bioassay (MBA) is the gold standard laboratory test for confirming a diagnosis of botulism by detecting BoNT in clinical samples (Centers for Disease Control and Prevention, 1998). However, only 20% of samples from foals and rare samples from adults are positive for BoNT in the MBA (Whitlock and McAdams, 2006). This underdiagnosis occurs because concentrations of BoNT in non-enriched equine clinical samples are usually below the limit of detection of the MBA (Allison et al., 1976; Swerczek, 1980; Whitlock and McAdams, 2006). Culture enrichment yields more positive results ⇑ Corresponding author. Tel.: +1 610 9256283. E-mail address: [email protected] (A.L. Johnson).
in the MBA because samples from cases of botulism often contain C. botulinum spores; these spores are present at a lower frequency in samples from unaffected horses (Dowell et al., 1977; Centers for Disease Control and Prevention, 1998; Whitlock and McAdams, 2006). After culture enrichment, 70% of samples from foals and 30% of samples from adult horses are positive in the MBA (Whitlock and McAdams, 2006). PCRs for detection of BoNT genes are more sensitive and provide more rapid results than the MBA (Lindström et al., 2001; Akbulut et al., 2004; Sánchez-Hernández et al., 2008; De Medici et al., 2009; Kirchner et al., 2010; Johnson et al., 2012). With the exception of a quantitative real-time PCR (qPCR) for C. botulinum type B developed previously by our group (Johnson et al., 2012), these assays have not been optimized or validated using equine samples. The aim of this study was to optimize and validate qPCR assays for detection of the neurotoxin genes of C. botulinum types A, B and C in equine diagnostic samples.
Materials and methods Samples Table 1 provides a summary of the 148 samples used in this study. Positive samples were defined as samples that were positive for pre-formed BoNT, C. botulinum spores, or both, when tested by MBA. One-hundred-and-six positive samples were available from the Botulism Laboratory at New Bolton Center, University of Pennsylvania School of Veterinary Medicine, USA; these had been collected over a period of 15 years and were stored at 20 °C in gelatin phosphate buffer (GPB; 0.2% gelatin, 0.4% Na2PO4; pH 6.4). Positive samples originated from feces, gastrointestinal contents, feed, wound exudates and carrion (2 cats, 1 horse) associated with clinical
1090-0233/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.tvjl.2013.10.023
Please cite this article in press as: Johnson, A.L., et al. Quantitative real-time PCR for detection of neurotoxin genes of Clostridium botulinum types A, B and C in equine samples. The Veterinary Journal (2013), http://dx.doi.org/10.1016/j.tvjl.2013.10.023
2
A.L. Johnson et al. / The Veterinary Journal xxx (2013) xxx–xxx
Table 1 Samples used for detection of Clostridium botulinum neurotoxin genes by PCR. Type A C. botulinuma Type B C. botulinuma Type C C. botulinuma Type D C. botulinumb Type E C. botulinumc Non-C. botulinumd Total Feces Gastrointestinal contents Feed Wound exudate Carrion Enteric bacteria Clostridial DNA
11 9 3 2
Total
25
a b c d
16 10 1
13 9 3
20 1
13 9
60 54 7 2 3 13 9
42
148
25
3
27
28
1
25
All samples were obtained from equine botulism cases. The type D sample was obtained from a bovine botulism case. The type E samples were obtained from avian botulism cases. The 20 faecal samples were obtained from clinically healthy horses.
cases of botulism. All samples positive for types A, B and C came from equine cases. The single positive type D sample came from a bovine case and all positive type E samples came from avian cases. Forty-two negative samples were included to determine specificity. These negative samples included 20 faecal samples from clinically healthy horses, 13 isolates from enteric bacterial species other than C. botulinum (Klebsiella oxytoca, Enterococcus faecalis, Shigella sonnei, Morganella morganii, Proteus vulgaris, Staphylococcus epidermidis, Salmonella poona, Escherichia coli, Staphylococcus aureus, Streptococcus agalactiae, Pseudomonas aeruginosa, Kocuria rosea and Edwardsiella tarda) and nine genomic DNA samples from non-C. botulinum Clostridium spp. (C. innocuum, C. difficile, C. clostridioforme, C hathewayi, C. symbiosum, C. orbiscindens, C. aldenense, C. bolteae and C. perfringens) obtained from the Biodefense and Emerging Infections Research Resources Repository, Manassas, Virginia, USA. Mouse bioassay The protocol for the MBA was approved by the University of Pennsylvania Institutional Animal Care and Use Committee (IACUC number 804158). All positive samples were analyzed at two different time points by MBA. Initially, MBA analysis was performed at the time of sample submission using the following protocol for detection of both pre-formed BoNT and spores (Centers for Disease Control and Prevention, 1998; Johnson et al., 2010). The sample (2–5 g) was diluted 1:2 (W:V) with sterile GPB, homogenized and refrigerated at 4 °C overnight. On the second day, 6 mL suspension were transferred to a sterile tube and centrifuged. The supernatant was injected intraperitoneally (IP) into four Swiss Webster mice (0.4 mL each) and two of the mice also received 0.1 mL specific antitoxin. Mice were observed for clinical signs of botulism for 4 days. Positive results, along with successful neutralization of the toxin, indicated the presence of BoNT (pre-formed toxin) in the original sample and identified the type of C. botulinum. Detection of spores by MBA was conducted after the original sample was culture-enriched for 5–7 days. A 1 mL aliquot of the original suspension was inoculated into a degassed tube containing chopped meat–glucose–starch (CMGS) media. The samples in CMGS media were transferred to an anaerobic chamber (88–90% N2, 5–7% H2, 5% CO2) at 37 °C for 5–7 days. After the incubation period, the culture samples were vortexed to remove toxin from chopped meat particles and 4 mL of the liquid portion were transferred into tubes containing 4 mL GPB. The resulting mixture was centrifuged and the supernatant (containing any toxin elaborated by C. botulinum during the culture period) was collected and used for MBA, as described above. Positive results indicated the presence of C. botulinum spores in the original sample and identified the type of C. botulinum. To assess the viability of C. botulinum spores from the samples after prolonged storage, a second MBA analysis was performed concurrently with the qPCR analysis on culture-enriched samples. Two mice were injected IP with 0.4 mL of the culture supernatant and observed for 4 days to demonstrate toxicity of the culture. The protection portion of the assay was not repeated. PCR Samples were processed using the Power Soil DNA Isolation Kit (MO BIO Laboratories). Oligonucleotide primers and probes (Table 2) were synthesised by Integrated DNA Technologies. The Stratagene Mx3000P qPCR System (Agilent Technologies) was utilized for amplification of BoNT genes. A duplex qPCR assay was produced for BoNT types A and B, whereas a singleplex qPCR assay was produced for BoNT type C. The primer and probe concentrations were optimized using a checkerboard titration. The efficiency of the duplex assay was compared to that of the singleplex A and B assays by testing identical samples in all three assays. For any individual sample, the cycle to threshold value (the cycle at which the amplification curve crosses the threshold of background fluorescence due to a detectable level of amplicon in the sample) was the same for the duplex and singleplex assays.
The final composition of reagents for the duplex assay was 6 lL template, 30 lL master mix (2 Brilliant QPCR Multiplex Master Mix, Agilent Technologies), 0.9 lL ROX passive reference dye (supplied with master mix), 300 nM each type A primer, 100 nM type A probe, 300 nM each type B primer, 50 nM type B probe, 12 lL exogenous internal amplification control mixture (TaqMan Exogenous IPC-VIC Probe, Applied Biosystems, Life Technologies), 2.4 lL exogenous internal amplification control template (TaqMan Exogenous IPC DNA, Applied Biosystems, Life Technologies) and distilled H2O to a final volume of 60 lL. The singleplex assay contained 300 nm each type C primer and 50 nM type C probe instead of types A or B probes or primers, but was otherwise identical. For each qPCR assay, aliquots of 25 lL were run in duplicate for each sample. The amplification cycle was 95 °C for 10 min, then 50 cycles of 95 °C for 15 s and 60 °C for 1 min. Cycle cut-offs were selected on the basis of the lowest level of detection on standard curves (40 for types A and C and 38 for type B); amplification above the threshold at or above any of these cut-offs was considered to be a negative result. The dynamic range and efficiency of the qPCR assays were assessed using a dilution series of C. botulinum DNA to generate a standard curve. Genomic DNA was commercially available for C. botulinum types A and B (bei Resources), but not for C. botulinum type C; the DNA template for type C was generated by incubating confirmed positive samples under anaerobic growth conditions and harvesting DNA from the culture-enriched sample. Comparison of qPCR and MBA The 106 positive repository samples and 20 faecal samples from clinically healthy horses were evaluated concurrently with the optimized qPCR assays and MBA. Samples were thawed overnight at 4 °C and 1 mL from each mixture was inoculated into two tubes of CMGS and incubated at 37 °C in the anaerobic chamber (88–90% N2, 5–7% H2, 5% CO2). One of the tubes was removed for DNA extraction and qPCR at 72 h and the other tube was removed for mouse bioassay at 5 days. Each PCR run included positive controls of BoNT types A, B and C DNA, and a negative (no-template) control consisting of 1 Tris–ethylene diamine tetraacetic acid (EDTA) buffer, as well as positive and negative extraction controls. Statistical analysis Sensitivity was calculated as the percentage of originally MBA-positive repository samples that yielded positive qPCR or MBA results, respectively. Samples that were negative on both qPCR and MBA were excluded from sensitivity analysis with the assumption that sample degradation during storage had affected bacterial viability. The specificity of the qPCR was calculated as the percentage of negative samples that yielded negative qPCR results. For the purposes of qPCR specificity calculation, negative samples included the 40 non-botulism samples, as well as the samples that were positive for a heterologous toxin type. For the MBA, only the 20 non-botulism faecal samples were used for the specificity analysis. Accuracy was calculated as the percentage of samples tested (positive + negative) that yielded correct results.
Results The assays all showed excellent linear range of detection over 6–7 logs (10-fold dilutions) on the standard curve with good precision (assessed by the coefficient of determination R2, the measurement of how closely data points fall to the standard curve) and amplification efficiency (see Appendix A: Supplementary Fig. 1). The standard curve for type A yielded R2 of 0.985 (perfect = 1.000) and efficiency of 93.2%, the standard curve for type B yielded R2 of
Please cite this article in press as: Johnson, A.L., et al. Quantitative real-time PCR for detection of neurotoxin genes of Clostridium botulinum types A, B and C in equine samples. The Veterinary Journal (2013), http://dx.doi.org/10.1016/j.tvjl.2013.10.023
A.L. Johnson et al. / The Veterinary Journal xxx (2013) xxx–xxx Table 2 Primers and probes for detection of Clostridium botulinum neurotoxin genes by PCR. Type A forward primera Reverse primera Probea
50 -CGAAATGGTTATGGCTCTACTCAA-30 50 -TTGCCTGCACCTAAAAGAGGAT-30 50 -5Cy5-ACTTCAAGTGACTCCTCAA AACCAAATGTAAAATCTG-3BHQ-2-30
Type B forward primera
50 -GATTATAAATGGTATACCTTA TCTTGGAGATAGAC-30 50 -CGCTCCACTTCTCCTGGATTAC-30 50 -56-FAM-TGTTCCACTCGAAGAGT TTAACACAAACATTGC-3BHQ-1-30
Reverse primera Probea Type C forward primerb Reverse primerb Probeb a b
50 -TGGTGCATTTGTGATTTATAGTAAGGT-30 50 -CACGTTCCYATCATCCATTCATATG-30 50 -56-FAM-CAAGAAAGAAACGAGATTAT-3BHQ2-30
Akbulut et al. (2004). Sánchez-Hernández et al. (2008).
0.981 and efficiency of 89.4%, and the standard curve for type C yielded R2 of 0.995 and efficiency of 93.9%. Table 3 shows results for all samples tested, while Table 4 summarizes test performance. Of 19 samples included for BoNT type A sensitivity analysis, qPCR was positive for 17 and MBA for 16. Of 21 samples included for type B sensitivity analysis, qPCR was positive for 18 and MBA for 17. Of 23 samples included for type C sensitivity analysis, qPCR was positive for 22 and MBA for 18. Specificity analysis included 123 samples for type A, 96 samples for type B and 120 samples for type C. The type C qPCR assay had 100% specificity, while the types A and B duplex assay had 99% specificity. The type A portion of the duplex assay yielded a false positive result for a single type B sample. The type B portion of the duplex assay yielded a false positive result for a single sample from a clinically healthy horse. The 20 samples from clinically healthy horses were used to calculate the specificity of the MBA, which also yielded a false positive result for a single sample from a different clinically healthy horse. Discussion The validation of these qPCR assays expands upon our previous work (Johnson et al., 2012) by allowing detection of BoNT types A, B and C in equine samples. These diagnostic tests for equine botulism have several advantages over the MBA, including higher accuracy, more rapid results and reduced costs. In our study, the specificities of the qPCR assays and the MBA were both high, while the sensitivities of the qPCR assays were higher than that of the MBA. These results compare favorably with those of Akbulut
Table 3 Results of PCR for detection of Clostridium botulinum neurotoxin genes. Number Number Number Number Number tested positive for positive for positive for positive type A on type B on type C on on MBAb qPCRa qPCRa qPCR Positive samples Type A Type B Type C Type D Type E Negative samples Feces Enteric bacteria Clostridial DNA a b c
106 25 27 28 1 25 42 20 13 9
18 17 1 0 0 0 0 0 0 0
27 9 18 0 0 0 1 1 0 0
22 0 0 22 0 0 0 0 0 0
Type A and B PCRs performed as a duplex assay. Results for the mouse bioassay performed concurrently with PCR. NA, not applicable.
51 16 17 18 NAc NAc 1 1 NAc NAc
3
et al. (2004), in which the sensitivity of a qPCR was 90% for 10 clinical and food samples. qPCR results can be available in 4 days (3 days culture enrichment and 1 day PCR analysis), whereas the mouse bioassay on culture-enriched samples typically requires 2–3 weeks before final results are available (culture enrichment for 5–7 days in the first week, mouse bioassay in the second week and repeated mouse bioassay with protection at the end of the second week or during the third week). The mouse bioassay may be confounded by non-specific mouse deaths or difficulty with the protection portion of the assay, which increase the time for results. Using the qPCR as a screening assay will reduce live animal use and overall cost of testing, since the MBA would only be required to confirm gene expression by detection of toxin in qPCR positive samples. Creating a multiplex assay to detect all three types of C. botulinum was considered initially, but not pursued, primarily due to the limited number of available detection channels on the thermal cycler and the need to include an internal amplification control. The decision to combine types A and B into one assay, while leaving type C as a singleplex assay was based on our experience that most equine botulism cases in the United States are due to types A and B, with only sporadic occurrence of type C cases. C. botulinum type B is endemic in soil in eastern USA, whereas C. botulinum type A is endemic in soil in western USA (Meyer and Dubovsky, 1922). Since most equine cases result from ingestion of pre-formed toxin in contaminated forage (adult horses) or toxicoinfectious botulism (foals), these soil-borne types represent the majority of cases (Whitlock and McAdams, 2006). C. botulinum type C is associated with carrion and is an uncommon cause of disease in horses in the USA. The repository of previously positive samples is a unique asset for the development and validation of diagnostic tests for botulism in horses, allowing careful assessment of test performance using typical samples from veterinary diagnostic cases. Although these samples were previously confirmed as positive on MBA, prolonged storage conditions raised concerns of potential sample degradation, lack of homogeneity and loss of bacterial viability. To ensure that qPCR results were comparable to MBA results on identical samples, a second MBA test was performed concurrently for each archived specimen. Additionally, samples that were negative on both tests were removed from sensitivity analysis due to presumed sample degradation during storage. This criterion led to the exclusion of 17 samples (6 type A, 6 type B and 5 type C). Whenever possible, equine samples were utilized for this study. However, naturally-occurring cases of type D and E botulism have not been reported in horses in the USA; the repository contained only one type D sample from a bovine case. Twenty-five type E samples were included in the specificity testing, all from avian cases. Type A samples were not included in our assessment of specificity for type B qPCR due to the frequent presence of a ‘silent’ (i.e. not expressed) type B neurotoxin gene in type A strains (Franciosa et al., 1994). Nine of 17 Type A samples in our study fell in this category and were positive on qPCR for the type B neurotoxin gene. When multiple toxin genes are detected by PCR, MBA is required to determine toxigenicity. Disparate results were seen in 18/63 samples used for sensitivity analysis. Twelve of these samples (9 samples of gastrointestinal contents, 2 feed samples and 1 sample of feces) were positive on qPCR, but negative on MBA run in parallel. As discussed previously (Johnson et al., 2012), the most likely explanation for the disparity is that culture enrichment allowed proliferation of the vegetative form of the bacteria in sufficient quantity to allow detection of the toxin gene, but gene expression and toxin production were below the limit of MBA detection, potentially due to prolonged storage. Six samples were negative on qPCR but positive on MBA,
Please cite this article in press as: Johnson, A.L., et al. Quantitative real-time PCR for detection of neurotoxin genes of Clostridium botulinum types A, B and C in equine samples. The Veterinary Journal (2013), http://dx.doi.org/10.1016/j.tvjl.2013.10.023
4
A.L. Johnson et al. / The Veterinary Journal xxx (2013) xxx–xxx
Table 4 Analytical sensitivity, specificity and accuracy of PCR for detection of Clostridium botulinum neurotoxin genes. Test Type A PCRe Type B PCRe Type C PCR Mouse bioassayf a b c d e f
Sensitivitya n (%)
95% CIb
Specificityc n (%)
95% CIb
Accuracyd n (%)
95% CIb
17/19 18/21 22/23 51/63
65–98 63–96 76–100 69–89
122/123 (99%) 95/96 (99%) 120/120 (100%) 19/20 (95%)
95–100 94–100 96–100 73–100
139/142 (98%) 113/117 (97%) 142/143 (99%) 70/83 (84%)
93–99 91–99 96–100 74–91
(89%) (86%) (96%) (81%)
Sensitivity = number of true positive results divided by (number of true positive + false negative results). 95% Confidence interval. Specificity = number of true negative results divided by (number of true negative + false positive results). Accuracy = (number of true positive + true negative results) divided by (number of true positive + false negative + true negative + false positive results). Type A and B PCRs performed as a duplex assay. Results for the mouse bioassay performed concurrently with PCR. Only overall results (combining results for types A, B and C) are reported.
despite lack of evidence for PCR inhibition. These archived samples might have had delayed growth in the culture enrichment phase, such that insufficient DNA was present to allow detection after 72 h of incubation, but after 5 days there was sufficient vegetative growth and toxin elaboration to yield a positive result on MBA. However, a longer culture enrichment phase did not improve the sensitivity of a singleplex BoNT type B PCR (Johnson et al., 2012). Alternatively, sequence variation of the BoNT gene could lead to false negative qPCR results (Hill et al., 2007). Finally, the presence of a large number of spores can inhibit the PCR reaction (Franciosa et al., 1996). Faecal samples from 20 clinically healthy horses yielded two unanticipated results. One sample was positive for type B by qPCR, but negative on MBA; when this ‘false positive’ faecal sample was re-evaluated by qPCR, it was negative twice (data not shown). This finding could represent a false positive or could indicate the presence of C. botulinum in the absence of clinical disease. However, C. botulinum is not considered to be part of the normal gastrointestinal flora of the horse (Whitlock and McAdams, 2006). A faecal sample from a different horse was negative on qPCR, but positive for type B on MBA after culture enrichment, indicating the presence of C. botulinum type B spores in the original sample. However, mouse death occurred on day 3 of the 4 day MBA. This result is difficult to interpret for a clinically healthy horse; most samples from clinical cases cause death in mice within 24 h (Solomon and Lilly, 2001). Taken together, these two results indicate that faecal samples from clinically healthy horses might occasionally test positive for C. botulinum spores; this warrants further investigation.
Conclusions The qPCR assays described here are well-suited for diagnostic investigation of botulism in horses and represent a faster, more sensitive method than the MBA for the detection of C. botulinum type A, B and C in culture-enriched equine biological samples.
Conflict of interest statement None of the authors of this paper has a financial or personal relationship with other people or organizations that could inappropriately influence or bias the content of the paper.
Acknowledgement The authors thank the Grayson-Jockey Club Research Foundation for providing funding for this study.
Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.tvjl.2013.10.023.
References Akbulut, D., Grant, K.A., McLauchlin, J., 2004. Development and application of realtime PCR assays to detect fragments of the Clostridium botulinum types A, B, and E neurotoxin genes for investigation of human foodborne and infant botulism. Foodborne Pathogens and Disease 1, 247–257. Allison, M.J., Maloy, S.E., Matson, R.R., 1976. Inactivation of Clostridium botulinum toxin by ruminal microbes from cattle and sheep. Applied and Environmental Microbiology 32, 685–688. Centers for Disease Control and Prevention, 1998. Botulism in the United States, 1899–1996. Handbook for Epidemiologists, Clinicians, and Laboratory Workers. Centers for Disease Control and Prevention, Atlanta, Georgia, USA, pp. 15–21. De Medici, D., Anniballi, F., Wyatt, G.M., Lindström, M., Messelhäusser, U., Aldus, C.F., Delibato, E., Korkeala, H., Peck, M.W., Fenicia, L., 2009. Multiplex PCR for detection of botulinum neurotoxin-producing clostridia in clinical, food, and environmental samples. Applied and Environmental Microbiology 75, 6457– 6461. Dowell Jr., V.R., McCroskey, L.M., Hatheway, C.L., Lombard, G.L., Hughes, J.M., Merson, M.H., 1977. Coproexamination for botulinal toxin and Clostridium botulinum. A new procedure for laboratory diagnosis of botulism. Journal of the American Medical Association 238, 1829–1832. Franciosa, G., Ferreira, J.L., Hatheway, C.L., 1994. Detection of type A, B, and E botulism neurotoxin genes in Clostridium botulinum and other Clostridium species by PCR: Evidence of unexpressed type B toxin genes in type A toxigenic organisms. Journal of Clinical Microbiology 32, 1911–1917. Franciosa, G., Fenicia, L., Caldiani, C., Aureli, P., 1996. PCR for detection of Clostridium botulinum Type C in avian and environmental samples. Journal of Clinical Microbiology 34, 882–885. Galey, F.D., 2001. Botulism in the horse. Veterinary Clinics of North America. Equine Practice 17, 579–588. Gerber, V., Straub, R., Frey, J., 2006. Equine botulism and acute pasture myodystrophy: New soil-borne emerging diseases in Switzerland? Schweizer Archiv für Tierheilkunde 148, 553–559. Hill, K.K., Smith, T.J., Helma, C.H., Ticknor, L.O., Foley, B.T., Svensson, R.T., Brown, J.L., Johnson, E.A., Smith, L.A., Okinaka, R.T., Jackson, P.J., Marks, J.D., 2007. Genetic diversity among botulinum neurotoxin-producing clostridial strains. Journal of Bacteriology 189, 818–832. Johnson, A.L., McAdams, S.C., Whitlock, R.H., 2010. Type A botulism in horses in the United States: A review of the past ten years (1998–2008). Journal of Veterinary Diagnostic Investigation 22, 165–173. Johnson, A.L., Sweeney, R.W., McAdams, S.C., Whitlock, R.H., 2012. Quantitative real-time PCR for detection of the neurotoxin gene of Clostridium botulinum type B in equine and bovine samples. The Veterinary Journal 194, 118–120. Kinde, H., Bettey, R.L., Ardans, A., Galey, F.D., Daft, B.M., Walker, R.L., Eklund, M.W., Byrd, J.W., 1991. Clostridium botulinum type-C intoxication associated with consumption of processed alfalfa hay cubes in horses. Journal of the American Veterinary Medical Association 199, 742–746. Kirchner, S., Krämer, K.M., Schulze, M., Pauly, D., Jacob, D., Gessler, F., Nitsche, A., Dorner, B.G., Dorner, M.B., 2010. Pentaplexed quantitative real-time PCR assay for the simultaneous detection and quantification of botulinum neurotoxinproducing clostridia in food and clinical samples. Applied and Environmental Microbiology 76, 4387–4395. Lindström, M., Keto, R., Markkula, A., Nevas, M., Hielm, S., Korkeala, H., 2001. Multiplex PCR assay for detection and identification of Clostridium botulinum types A, B, E, and F in food and fecal material. Applied and Environmental Microbiology 67, 5694–5699. Meyer, K.F., Dubovsky, B.J., 1922. The distribution of the spores of B. botulinus in the United States. Journal of Infectious Diseases 31, 559–594.
Please cite this article in press as: Johnson, A.L., et al. Quantitative real-time PCR for detection of neurotoxin genes of Clostridium botulinum types A, B and C in equine samples. The Veterinary Journal (2013), http://dx.doi.org/10.1016/j.tvjl.2013.10.023
A.L. Johnson et al. / The Veterinary Journal xxx (2013) xxx–xxx Meyer, K.F., 1956. The status of botulism as a world health problem. Bulletin of the World Health Organization 15, 281–298. Sánchez-Hernández, L., Cifuentes, A., Jiménez, B., Mateo, R., Gonzalez, R., 2008. Detection of Clostridium botulinum neurotoxin coding genes: Analysis of PCR products by real time versus capillary gel electrophoresis methods. European Food Research and Technology 227, 495–502. Simpson, L.L., 2004. Identification of the major steps in botulinum toxin action. Annual Review of Pharmacology and Toxicology 44, 167–193.
5
Solomon, H.M., Lilly, T., 2001. Clostridium botulinum. Bacteriological Analytical Manual, Eighth Ed., Revision A, 1998 (Chapter 17). http://www.fda.gov/Food/ FoodScienceResearch/LaboratoryMethods/ucm070879.htm (accessed 16 October 2013). Swerczek, T.W., 1980. Toxicoinfectious botulism in foals and adult horses. Journal of the American Veterinary Medical Association 176, 217–220. Whitlock, R.H., McAdams, S., 2006. Equine botulism. Clinical Techniques in Equine Practice 5, 37–42.
Please cite this article in press as: Johnson, A.L., et al. Quantitative real-time PCR for detection of neurotoxin genes of Clostridium botulinum types A, B and C in equine samples. The Veterinary Journal (2013), http://dx.doi.org/10.1016/j.tvjl.2013.10.023
Contents lists available at ScienceDirect
The Veterinary Journal journal homepage: www.elsevier.com/locate/tvjl
Quantitative real-time PCR for detection of neurotoxin genes of Clostridium botulinum types A, B and C in equine samples Amy L. Johnson ⇑, Susan C. McAdams-Gallagher, Raymond W. Sweeney Department of Clinical Studies, New Bolton Center, University of Pennsylvania School of Veterinary Medicine, Kennett Square, PA 19348, USA
a r t i c l e
i n f o
Article history: Accepted 18 October 2013 Available online xxxx Keywords: Botulism Equine Real-time quantitative PCR Mouse bioassay
a b s t r a c t Botulism in horses in the USA is attributed to Clostridium botulinum types A, B or C. In this study, a duplex quantitative real-time PCR (qPCR) for detection of the neurotoxin genes of C. botulinum types A and B, and a singleplex qPCR for detection of the neurotoxin gene of C. botulinum type C, were optimized and validated for equine gastrointestinal, faecal and feed samples. The performance of these assays was evaluated and compared to the standard mouse bioassay (MBA) using 148 well-characterized samples, most of which were acquired from a repository of veterinary diagnostic samples from cases of botulism: 106 samples positive for C. botulinum (25 type A, 27 type B, 28 type C, 1 type D and 25 type E) and 42 negative samples. The sensitivities of the qPCR assays were 89%, 86% and 96% for C. botulinum types A, B and C, respectively. The overall sensitivity of the mouse bioassay for types A, B and C was 81%. The specificities of the qPCR assays were 99–100% and the specificity of the mouse bioassay was 95%. Ó 2013 Elsevier Ltd. All rights reserved.
Introduction Botulism is caused by Clostridium botulinum neurotoxin (BoNT) and manifested by progressive flaccid paresis and cranial nerve deficits, such as dysphagia (Meyer, 1956; Galey, 2001). There are eight serotypes of C. botulinum, distinguished by the unique properties of their toxins (Simpson, 2004). C. botulinum type B causes most (>85%) cases of equine botulism in the USA and is endemic in the mid-Atlantic region and Kentucky (Whitlock and McAdams, 2006). Conversely, C. botulinum type A is endemic in states west of the Mississippi River (Johnson et al., 2010). Intoxication by C. botulinum type C is associated with carrion contamination of feed and occurs sporadically (Kinde et al., 1991). While only types A, B and C have been reported in horses in the USA, types C and D are most common in Europe (Gerber et al., 2006). Diagnosis of botulism often is based on compatible history and clinical signs, since laboratory confirmation is difficult. The mouse bioassay (MBA) is the gold standard laboratory test for confirming a diagnosis of botulism by detecting BoNT in clinical samples (Centers for Disease Control and Prevention, 1998). However, only 20% of samples from foals and rare samples from adults are positive for BoNT in the MBA (Whitlock and McAdams, 2006). This underdiagnosis occurs because concentrations of BoNT in non-enriched equine clinical samples are usually below the limit of detection of the MBA (Allison et al., 1976; Swerczek, 1980; Whitlock and McAdams, 2006). Culture enrichment yields more positive results ⇑ Corresponding author. Tel.: +1 610 9256283. E-mail address: [email protected] (A.L. Johnson).
in the MBA because samples from cases of botulism often contain C. botulinum spores; these spores are present at a lower frequency in samples from unaffected horses (Dowell et al., 1977; Centers for Disease Control and Prevention, 1998; Whitlock and McAdams, 2006). After culture enrichment, 70% of samples from foals and 30% of samples from adult horses are positive in the MBA (Whitlock and McAdams, 2006). PCRs for detection of BoNT genes are more sensitive and provide more rapid results than the MBA (Lindström et al., 2001; Akbulut et al., 2004; Sánchez-Hernández et al., 2008; De Medici et al., 2009; Kirchner et al., 2010; Johnson et al., 2012). With the exception of a quantitative real-time PCR (qPCR) for C. botulinum type B developed previously by our group (Johnson et al., 2012), these assays have not been optimized or validated using equine samples. The aim of this study was to optimize and validate qPCR assays for detection of the neurotoxin genes of C. botulinum types A, B and C in equine diagnostic samples.
Materials and methods Samples Table 1 provides a summary of the 148 samples used in this study. Positive samples were defined as samples that were positive for pre-formed BoNT, C. botulinum spores, or both, when tested by MBA. One-hundred-and-six positive samples were available from the Botulism Laboratory at New Bolton Center, University of Pennsylvania School of Veterinary Medicine, USA; these had been collected over a period of 15 years and were stored at 20 °C in gelatin phosphate buffer (GPB; 0.2% gelatin, 0.4% Na2PO4; pH 6.4). Positive samples originated from feces, gastrointestinal contents, feed, wound exudates and carrion (2 cats, 1 horse) associated with clinical
1090-0233/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.tvjl.2013.10.023
Please cite this article in press as: Johnson, A.L., et al. Quantitative real-time PCR for detection of neurotoxin genes of Clostridium botulinum types A, B and C in equine samples. The Veterinary Journal (2013), http://dx.doi.org/10.1016/j.tvjl.2013.10.023
2
A.L. Johnson et al. / The Veterinary Journal xxx (2013) xxx–xxx
Table 1 Samples used for detection of Clostridium botulinum neurotoxin genes by PCR. Type A C. botulinuma Type B C. botulinuma Type C C. botulinuma Type D C. botulinumb Type E C. botulinumc Non-C. botulinumd Total Feces Gastrointestinal contents Feed Wound exudate Carrion Enteric bacteria Clostridial DNA
11 9 3 2
Total
25
a b c d
16 10 1
13 9 3
20 1
13 9
60 54 7 2 3 13 9
42
148
25
3
27
28
1
25
All samples were obtained from equine botulism cases. The type D sample was obtained from a bovine botulism case. The type E samples were obtained from avian botulism cases. The 20 faecal samples were obtained from clinically healthy horses.
cases of botulism. All samples positive for types A, B and C came from equine cases. The single positive type D sample came from a bovine case and all positive type E samples came from avian cases. Forty-two negative samples were included to determine specificity. These negative samples included 20 faecal samples from clinically healthy horses, 13 isolates from enteric bacterial species other than C. botulinum (Klebsiella oxytoca, Enterococcus faecalis, Shigella sonnei, Morganella morganii, Proteus vulgaris, Staphylococcus epidermidis, Salmonella poona, Escherichia coli, Staphylococcus aureus, Streptococcus agalactiae, Pseudomonas aeruginosa, Kocuria rosea and Edwardsiella tarda) and nine genomic DNA samples from non-C. botulinum Clostridium spp. (C. innocuum, C. difficile, C. clostridioforme, C hathewayi, C. symbiosum, C. orbiscindens, C. aldenense, C. bolteae and C. perfringens) obtained from the Biodefense and Emerging Infections Research Resources Repository, Manassas, Virginia, USA. Mouse bioassay The protocol for the MBA was approved by the University of Pennsylvania Institutional Animal Care and Use Committee (IACUC number 804158). All positive samples were analyzed at two different time points by MBA. Initially, MBA analysis was performed at the time of sample submission using the following protocol for detection of both pre-formed BoNT and spores (Centers for Disease Control and Prevention, 1998; Johnson et al., 2010). The sample (2–5 g) was diluted 1:2 (W:V) with sterile GPB, homogenized and refrigerated at 4 °C overnight. On the second day, 6 mL suspension were transferred to a sterile tube and centrifuged. The supernatant was injected intraperitoneally (IP) into four Swiss Webster mice (0.4 mL each) and two of the mice also received 0.1 mL specific antitoxin. Mice were observed for clinical signs of botulism for 4 days. Positive results, along with successful neutralization of the toxin, indicated the presence of BoNT (pre-formed toxin) in the original sample and identified the type of C. botulinum. Detection of spores by MBA was conducted after the original sample was culture-enriched for 5–7 days. A 1 mL aliquot of the original suspension was inoculated into a degassed tube containing chopped meat–glucose–starch (CMGS) media. The samples in CMGS media were transferred to an anaerobic chamber (88–90% N2, 5–7% H2, 5% CO2) at 37 °C for 5–7 days. After the incubation period, the culture samples were vortexed to remove toxin from chopped meat particles and 4 mL of the liquid portion were transferred into tubes containing 4 mL GPB. The resulting mixture was centrifuged and the supernatant (containing any toxin elaborated by C. botulinum during the culture period) was collected and used for MBA, as described above. Positive results indicated the presence of C. botulinum spores in the original sample and identified the type of C. botulinum. To assess the viability of C. botulinum spores from the samples after prolonged storage, a second MBA analysis was performed concurrently with the qPCR analysis on culture-enriched samples. Two mice were injected IP with 0.4 mL of the culture supernatant and observed for 4 days to demonstrate toxicity of the culture. The protection portion of the assay was not repeated. PCR Samples were processed using the Power Soil DNA Isolation Kit (MO BIO Laboratories). Oligonucleotide primers and probes (Table 2) were synthesised by Integrated DNA Technologies. The Stratagene Mx3000P qPCR System (Agilent Technologies) was utilized for amplification of BoNT genes. A duplex qPCR assay was produced for BoNT types A and B, whereas a singleplex qPCR assay was produced for BoNT type C. The primer and probe concentrations were optimized using a checkerboard titration. The efficiency of the duplex assay was compared to that of the singleplex A and B assays by testing identical samples in all three assays. For any individual sample, the cycle to threshold value (the cycle at which the amplification curve crosses the threshold of background fluorescence due to a detectable level of amplicon in the sample) was the same for the duplex and singleplex assays.
The final composition of reagents for the duplex assay was 6 lL template, 30 lL master mix (2 Brilliant QPCR Multiplex Master Mix, Agilent Technologies), 0.9 lL ROX passive reference dye (supplied with master mix), 300 nM each type A primer, 100 nM type A probe, 300 nM each type B primer, 50 nM type B probe, 12 lL exogenous internal amplification control mixture (TaqMan Exogenous IPC-VIC Probe, Applied Biosystems, Life Technologies), 2.4 lL exogenous internal amplification control template (TaqMan Exogenous IPC DNA, Applied Biosystems, Life Technologies) and distilled H2O to a final volume of 60 lL. The singleplex assay contained 300 nm each type C primer and 50 nM type C probe instead of types A or B probes or primers, but was otherwise identical. For each qPCR assay, aliquots of 25 lL were run in duplicate for each sample. The amplification cycle was 95 °C for 10 min, then 50 cycles of 95 °C for 15 s and 60 °C for 1 min. Cycle cut-offs were selected on the basis of the lowest level of detection on standard curves (40 for types A and C and 38 for type B); amplification above the threshold at or above any of these cut-offs was considered to be a negative result. The dynamic range and efficiency of the qPCR assays were assessed using a dilution series of C. botulinum DNA to generate a standard curve. Genomic DNA was commercially available for C. botulinum types A and B (bei Resources), but not for C. botulinum type C; the DNA template for type C was generated by incubating confirmed positive samples under anaerobic growth conditions and harvesting DNA from the culture-enriched sample. Comparison of qPCR and MBA The 106 positive repository samples and 20 faecal samples from clinically healthy horses were evaluated concurrently with the optimized qPCR assays and MBA. Samples were thawed overnight at 4 °C and 1 mL from each mixture was inoculated into two tubes of CMGS and incubated at 37 °C in the anaerobic chamber (88–90% N2, 5–7% H2, 5% CO2). One of the tubes was removed for DNA extraction and qPCR at 72 h and the other tube was removed for mouse bioassay at 5 days. Each PCR run included positive controls of BoNT types A, B and C DNA, and a negative (no-template) control consisting of 1 Tris–ethylene diamine tetraacetic acid (EDTA) buffer, as well as positive and negative extraction controls. Statistical analysis Sensitivity was calculated as the percentage of originally MBA-positive repository samples that yielded positive qPCR or MBA results, respectively. Samples that were negative on both qPCR and MBA were excluded from sensitivity analysis with the assumption that sample degradation during storage had affected bacterial viability. The specificity of the qPCR was calculated as the percentage of negative samples that yielded negative qPCR results. For the purposes of qPCR specificity calculation, negative samples included the 40 non-botulism samples, as well as the samples that were positive for a heterologous toxin type. For the MBA, only the 20 non-botulism faecal samples were used for the specificity analysis. Accuracy was calculated as the percentage of samples tested (positive + negative) that yielded correct results.
Results The assays all showed excellent linear range of detection over 6–7 logs (10-fold dilutions) on the standard curve with good precision (assessed by the coefficient of determination R2, the measurement of how closely data points fall to the standard curve) and amplification efficiency (see Appendix A: Supplementary Fig. 1). The standard curve for type A yielded R2 of 0.985 (perfect = 1.000) and efficiency of 93.2%, the standard curve for type B yielded R2 of
Please cite this article in press as: Johnson, A.L., et al. Quantitative real-time PCR for detection of neurotoxin genes of Clostridium botulinum types A, B and C in equine samples. The Veterinary Journal (2013), http://dx.doi.org/10.1016/j.tvjl.2013.10.023
A.L. Johnson et al. / The Veterinary Journal xxx (2013) xxx–xxx Table 2 Primers and probes for detection of Clostridium botulinum neurotoxin genes by PCR. Type A forward primera Reverse primera Probea
50 -CGAAATGGTTATGGCTCTACTCAA-30 50 -TTGCCTGCACCTAAAAGAGGAT-30 50 -5Cy5-ACTTCAAGTGACTCCTCAA AACCAAATGTAAAATCTG-3BHQ-2-30
Type B forward primera
50 -GATTATAAATGGTATACCTTA TCTTGGAGATAGAC-30 50 -CGCTCCACTTCTCCTGGATTAC-30 50 -56-FAM-TGTTCCACTCGAAGAGT TTAACACAAACATTGC-3BHQ-1-30
Reverse primera Probea Type C forward primerb Reverse primerb Probeb a b
50 -TGGTGCATTTGTGATTTATAGTAAGGT-30 50 -CACGTTCCYATCATCCATTCATATG-30 50 -56-FAM-CAAGAAAGAAACGAGATTAT-3BHQ2-30
Akbulut et al. (2004). Sánchez-Hernández et al. (2008).
0.981 and efficiency of 89.4%, and the standard curve for type C yielded R2 of 0.995 and efficiency of 93.9%. Table 3 shows results for all samples tested, while Table 4 summarizes test performance. Of 19 samples included for BoNT type A sensitivity analysis, qPCR was positive for 17 and MBA for 16. Of 21 samples included for type B sensitivity analysis, qPCR was positive for 18 and MBA for 17. Of 23 samples included for type C sensitivity analysis, qPCR was positive for 22 and MBA for 18. Specificity analysis included 123 samples for type A, 96 samples for type B and 120 samples for type C. The type C qPCR assay had 100% specificity, while the types A and B duplex assay had 99% specificity. The type A portion of the duplex assay yielded a false positive result for a single type B sample. The type B portion of the duplex assay yielded a false positive result for a single sample from a clinically healthy horse. The 20 samples from clinically healthy horses were used to calculate the specificity of the MBA, which also yielded a false positive result for a single sample from a different clinically healthy horse. Discussion The validation of these qPCR assays expands upon our previous work (Johnson et al., 2012) by allowing detection of BoNT types A, B and C in equine samples. These diagnostic tests for equine botulism have several advantages over the MBA, including higher accuracy, more rapid results and reduced costs. In our study, the specificities of the qPCR assays and the MBA were both high, while the sensitivities of the qPCR assays were higher than that of the MBA. These results compare favorably with those of Akbulut
Table 3 Results of PCR for detection of Clostridium botulinum neurotoxin genes. Number Number Number Number Number tested positive for positive for positive for positive type A on type B on type C on on MBAb qPCRa qPCRa qPCR Positive samples Type A Type B Type C Type D Type E Negative samples Feces Enteric bacteria Clostridial DNA a b c
106 25 27 28 1 25 42 20 13 9
18 17 1 0 0 0 0 0 0 0
27 9 18 0 0 0 1 1 0 0
22 0 0 22 0 0 0 0 0 0
Type A and B PCRs performed as a duplex assay. Results for the mouse bioassay performed concurrently with PCR. NA, not applicable.
51 16 17 18 NAc NAc 1 1 NAc NAc
3
et al. (2004), in which the sensitivity of a qPCR was 90% for 10 clinical and food samples. qPCR results can be available in 4 days (3 days culture enrichment and 1 day PCR analysis), whereas the mouse bioassay on culture-enriched samples typically requires 2–3 weeks before final results are available (culture enrichment for 5–7 days in the first week, mouse bioassay in the second week and repeated mouse bioassay with protection at the end of the second week or during the third week). The mouse bioassay may be confounded by non-specific mouse deaths or difficulty with the protection portion of the assay, which increase the time for results. Using the qPCR as a screening assay will reduce live animal use and overall cost of testing, since the MBA would only be required to confirm gene expression by detection of toxin in qPCR positive samples. Creating a multiplex assay to detect all three types of C. botulinum was considered initially, but not pursued, primarily due to the limited number of available detection channels on the thermal cycler and the need to include an internal amplification control. The decision to combine types A and B into one assay, while leaving type C as a singleplex assay was based on our experience that most equine botulism cases in the United States are due to types A and B, with only sporadic occurrence of type C cases. C. botulinum type B is endemic in soil in eastern USA, whereas C. botulinum type A is endemic in soil in western USA (Meyer and Dubovsky, 1922). Since most equine cases result from ingestion of pre-formed toxin in contaminated forage (adult horses) or toxicoinfectious botulism (foals), these soil-borne types represent the majority of cases (Whitlock and McAdams, 2006). C. botulinum type C is associated with carrion and is an uncommon cause of disease in horses in the USA. The repository of previously positive samples is a unique asset for the development and validation of diagnostic tests for botulism in horses, allowing careful assessment of test performance using typical samples from veterinary diagnostic cases. Although these samples were previously confirmed as positive on MBA, prolonged storage conditions raised concerns of potential sample degradation, lack of homogeneity and loss of bacterial viability. To ensure that qPCR results were comparable to MBA results on identical samples, a second MBA test was performed concurrently for each archived specimen. Additionally, samples that were negative on both tests were removed from sensitivity analysis due to presumed sample degradation during storage. This criterion led to the exclusion of 17 samples (6 type A, 6 type B and 5 type C). Whenever possible, equine samples were utilized for this study. However, naturally-occurring cases of type D and E botulism have not been reported in horses in the USA; the repository contained only one type D sample from a bovine case. Twenty-five type E samples were included in the specificity testing, all from avian cases. Type A samples were not included in our assessment of specificity for type B qPCR due to the frequent presence of a ‘silent’ (i.e. not expressed) type B neurotoxin gene in type A strains (Franciosa et al., 1994). Nine of 17 Type A samples in our study fell in this category and were positive on qPCR for the type B neurotoxin gene. When multiple toxin genes are detected by PCR, MBA is required to determine toxigenicity. Disparate results were seen in 18/63 samples used for sensitivity analysis. Twelve of these samples (9 samples of gastrointestinal contents, 2 feed samples and 1 sample of feces) were positive on qPCR, but negative on MBA run in parallel. As discussed previously (Johnson et al., 2012), the most likely explanation for the disparity is that culture enrichment allowed proliferation of the vegetative form of the bacteria in sufficient quantity to allow detection of the toxin gene, but gene expression and toxin production were below the limit of MBA detection, potentially due to prolonged storage. Six samples were negative on qPCR but positive on MBA,
Please cite this article in press as: Johnson, A.L., et al. Quantitative real-time PCR for detection of neurotoxin genes of Clostridium botulinum types A, B and C in equine samples. The Veterinary Journal (2013), http://dx.doi.org/10.1016/j.tvjl.2013.10.023
4
A.L. Johnson et al. / The Veterinary Journal xxx (2013) xxx–xxx
Table 4 Analytical sensitivity, specificity and accuracy of PCR for detection of Clostridium botulinum neurotoxin genes. Test Type A PCRe Type B PCRe Type C PCR Mouse bioassayf a b c d e f
Sensitivitya n (%)
95% CIb
Specificityc n (%)
95% CIb
Accuracyd n (%)
95% CIb
17/19 18/21 22/23 51/63
65–98 63–96 76–100 69–89
122/123 (99%) 95/96 (99%) 120/120 (100%) 19/20 (95%)
95–100 94–100 96–100 73–100
139/142 (98%) 113/117 (97%) 142/143 (99%) 70/83 (84%)
93–99 91–99 96–100 74–91
(89%) (86%) (96%) (81%)
Sensitivity = number of true positive results divided by (number of true positive + false negative results). 95% Confidence interval. Specificity = number of true negative results divided by (number of true negative + false positive results). Accuracy = (number of true positive + true negative results) divided by (number of true positive + false negative + true negative + false positive results). Type A and B PCRs performed as a duplex assay. Results for the mouse bioassay performed concurrently with PCR. Only overall results (combining results for types A, B and C) are reported.
despite lack of evidence for PCR inhibition. These archived samples might have had delayed growth in the culture enrichment phase, such that insufficient DNA was present to allow detection after 72 h of incubation, but after 5 days there was sufficient vegetative growth and toxin elaboration to yield a positive result on MBA. However, a longer culture enrichment phase did not improve the sensitivity of a singleplex BoNT type B PCR (Johnson et al., 2012). Alternatively, sequence variation of the BoNT gene could lead to false negative qPCR results (Hill et al., 2007). Finally, the presence of a large number of spores can inhibit the PCR reaction (Franciosa et al., 1996). Faecal samples from 20 clinically healthy horses yielded two unanticipated results. One sample was positive for type B by qPCR, but negative on MBA; when this ‘false positive’ faecal sample was re-evaluated by qPCR, it was negative twice (data not shown). This finding could represent a false positive or could indicate the presence of C. botulinum in the absence of clinical disease. However, C. botulinum is not considered to be part of the normal gastrointestinal flora of the horse (Whitlock and McAdams, 2006). A faecal sample from a different horse was negative on qPCR, but positive for type B on MBA after culture enrichment, indicating the presence of C. botulinum type B spores in the original sample. However, mouse death occurred on day 3 of the 4 day MBA. This result is difficult to interpret for a clinically healthy horse; most samples from clinical cases cause death in mice within 24 h (Solomon and Lilly, 2001). Taken together, these two results indicate that faecal samples from clinically healthy horses might occasionally test positive for C. botulinum spores; this warrants further investigation.
Conclusions The qPCR assays described here are well-suited for diagnostic investigation of botulism in horses and represent a faster, more sensitive method than the MBA for the detection of C. botulinum type A, B and C in culture-enriched equine biological samples.
Conflict of interest statement None of the authors of this paper has a financial or personal relationship with other people or organizations that could inappropriately influence or bias the content of the paper.
Acknowledgement The authors thank the Grayson-Jockey Club Research Foundation for providing funding for this study.
Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.tvjl.2013.10.023.
References Akbulut, D., Grant, K.A., McLauchlin, J., 2004. Development and application of realtime PCR assays to detect fragments of the Clostridium botulinum types A, B, and E neurotoxin genes for investigation of human foodborne and infant botulism. Foodborne Pathogens and Disease 1, 247–257. Allison, M.J., Maloy, S.E., Matson, R.R., 1976. Inactivation of Clostridium botulinum toxin by ruminal microbes from cattle and sheep. Applied and Environmental Microbiology 32, 685–688. Centers for Disease Control and Prevention, 1998. Botulism in the United States, 1899–1996. Handbook for Epidemiologists, Clinicians, and Laboratory Workers. Centers for Disease Control and Prevention, Atlanta, Georgia, USA, pp. 15–21. De Medici, D., Anniballi, F., Wyatt, G.M., Lindström, M., Messelhäusser, U., Aldus, C.F., Delibato, E., Korkeala, H., Peck, M.W., Fenicia, L., 2009. Multiplex PCR for detection of botulinum neurotoxin-producing clostridia in clinical, food, and environmental samples. Applied and Environmental Microbiology 75, 6457– 6461. Dowell Jr., V.R., McCroskey, L.M., Hatheway, C.L., Lombard, G.L., Hughes, J.M., Merson, M.H., 1977. Coproexamination for botulinal toxin and Clostridium botulinum. A new procedure for laboratory diagnosis of botulism. Journal of the American Medical Association 238, 1829–1832. Franciosa, G., Ferreira, J.L., Hatheway, C.L., 1994. Detection of type A, B, and E botulism neurotoxin genes in Clostridium botulinum and other Clostridium species by PCR: Evidence of unexpressed type B toxin genes in type A toxigenic organisms. Journal of Clinical Microbiology 32, 1911–1917. Franciosa, G., Fenicia, L., Caldiani, C., Aureli, P., 1996. PCR for detection of Clostridium botulinum Type C in avian and environmental samples. Journal of Clinical Microbiology 34, 882–885. Galey, F.D., 2001. Botulism in the horse. Veterinary Clinics of North America. Equine Practice 17, 579–588. Gerber, V., Straub, R., Frey, J., 2006. Equine botulism and acute pasture myodystrophy: New soil-borne emerging diseases in Switzerland? Schweizer Archiv für Tierheilkunde 148, 553–559. Hill, K.K., Smith, T.J., Helma, C.H., Ticknor, L.O., Foley, B.T., Svensson, R.T., Brown, J.L., Johnson, E.A., Smith, L.A., Okinaka, R.T., Jackson, P.J., Marks, J.D., 2007. Genetic diversity among botulinum neurotoxin-producing clostridial strains. Journal of Bacteriology 189, 818–832. Johnson, A.L., McAdams, S.C., Whitlock, R.H., 2010. Type A botulism in horses in the United States: A review of the past ten years (1998–2008). Journal of Veterinary Diagnostic Investigation 22, 165–173. Johnson, A.L., Sweeney, R.W., McAdams, S.C., Whitlock, R.H., 2012. Quantitative real-time PCR for detection of the neurotoxin gene of Clostridium botulinum type B in equine and bovine samples. The Veterinary Journal 194, 118–120. Kinde, H., Bettey, R.L., Ardans, A., Galey, F.D., Daft, B.M., Walker, R.L., Eklund, M.W., Byrd, J.W., 1991. Clostridium botulinum type-C intoxication associated with consumption of processed alfalfa hay cubes in horses. Journal of the American Veterinary Medical Association 199, 742–746. Kirchner, S., Krämer, K.M., Schulze, M., Pauly, D., Jacob, D., Gessler, F., Nitsche, A., Dorner, B.G., Dorner, M.B., 2010. Pentaplexed quantitative real-time PCR assay for the simultaneous detection and quantification of botulinum neurotoxinproducing clostridia in food and clinical samples. Applied and Environmental Microbiology 76, 4387–4395. Lindström, M., Keto, R., Markkula, A., Nevas, M., Hielm, S., Korkeala, H., 2001. Multiplex PCR assay for detection and identification of Clostridium botulinum types A, B, E, and F in food and fecal material. Applied and Environmental Microbiology 67, 5694–5699. Meyer, K.F., Dubovsky, B.J., 1922. The distribution of the spores of B. botulinus in the United States. Journal of Infectious Diseases 31, 559–594.
Please cite this article in press as: Johnson, A.L., et al. Quantitative real-time PCR for detection of neurotoxin genes of Clostridium botulinum types A, B and C in equine samples. The Veterinary Journal (2013), http://dx.doi.org/10.1016/j.tvjl.2013.10.023
A.L. Johnson et al. / The Veterinary Journal xxx (2013) xxx–xxx Meyer, K.F., 1956. The status of botulism as a world health problem. Bulletin of the World Health Organization 15, 281–298. Sánchez-Hernández, L., Cifuentes, A., Jiménez, B., Mateo, R., Gonzalez, R., 2008. Detection of Clostridium botulinum neurotoxin coding genes: Analysis of PCR products by real time versus capillary gel electrophoresis methods. European Food Research and Technology 227, 495–502. Simpson, L.L., 2004. Identification of the major steps in botulinum toxin action. Annual Review of Pharmacology and Toxicology 44, 167–193.
5
Solomon, H.M., Lilly, T., 2001. Clostridium botulinum. Bacteriological Analytical Manual, Eighth Ed., Revision A, 1998 (Chapter 17). http://www.fda.gov/Food/ FoodScienceResearch/LaboratoryMethods/ucm070879.htm (accessed 16 October 2013). Swerczek, T.W., 1980. Toxicoinfectious botulism in foals and adult horses. Journal of the American Veterinary Medical Association 176, 217–220. Whitlock, R.H., McAdams, S., 2006. Equine botulism. Clinical Techniques in Equine Practice 5, 37–42.
Please cite this article in press as: Johnson, A.L., et al. Quantitative real-time PCR for detection of neurotoxin genes of Clostridium botulinum types A, B and C in equine samples. The Veterinary Journal (2013), http://dx.doi.org/10.1016/j.tvjl.2013.10.023
