Research Article

Development of a Taqman Real-Time PCR Assay for Detection of Bordetella bronchiseptica

Tomas Jinnerot1*, Karin Malm1, Erik Eriksson1 and Jonas Johansson Wensman2

1Department of Microbiology, National Veterinary Institute, SE-751 89 Uppsala, Sweden; 2Department of Clinical Sciences, Swedish University of Agricultural Sciences, P.O Box 7054, SE-750 07 Uppsala, Sweden.

Editor | Muhammad Abubakar, National Veterinary Laboratories, Park Road, Islamabad, Pakistan.

Received | December 01, 2015; Accepted | December 17, 2015; Published | December 29, 2015

*Correspondence | Tomas Jinnerot, National Veterinary Institute, SE-751 89 Uppsala, Sweden; e-mail | [email protected]

Citation | Jinnerot, T., K. Malm, E. Eriksson and J. J. Wensman. 2015. Development of a Taqman real-time PCR assay for detection of Bordetella bronchiseptica. Veterinary Sciences: Research and Reviews, 1(1): 14-20.

DOI | http://dx.doi.org/10.17582/journal.vsrr/2015.1.1.14.20

Introduction

Bordetellae are gram-negative ß-proteobacteria that colonize the respiratory tract of mammals. B. bronchiseptica is the most common bacteriological cause of kennel cough or canine infectious respiratory disease (CIRD), a highly contagious and multifactorial upper respiratory syndrome of dogs. Other primary causes of CIRD include canine parainfluenzavirus, canine respiratory coronavirus (Mitchell et al., 2013), as well as other viruses (Buonavoglia and Martella, 2007; Mitchell et al., 2013, Schulz et al., 2014). Healthy dogs can harbor B. bronchiseptica, and following clinical signs of CIRD caused by B. bronchiseptica, dogs may shed the bacteria for up to three months, which makes it difficult to eradicate (Bemis et al., 1977). Clinical signs include dry cough, tracheal mucus accumulation, vomiting and pneumonia, and usually last for 1-3 weeks after which a temporary immunity develops (Goodnow, 1980; Greene, 2006). The traditional diagnostic method consists of culturing nasopharyngeal swabs on non-selective and differentiating growth media, which may take several days and depends on large amounts of bacteria for successful isolation, followed by biochemical testing. Real-time PCR is a fast and sensitive method suitable for detection of specific pathogens. Previous work by others have aimed at developing methods that could distinguish several different Bordetella-species, targeting multi-copy insertion sequences present in more than one species (Roorda et al., 2011; Tatti et al., 2011). These methods demand running several assays per sample (Tizolova et al., 2014) and the use of algorithms to deduce proper results. Others have developed similar assays based on SYBR Green detection and melt curve analysis and thorough evaluation of any assay has only been done using human clinical isolates (Koidl et al., 2007; Roorda et al., 2011; Tatti et al., 2011; Tizolova et al., 2014). A SYBR Green real-time PCR was developed recently for detection of B. bronchiseptica in dogs (Schulz et al., 2014), but this assay was deemed unspecific as it cross-reacts with sequences of B. parapertussis when tested in silico. Depending on the purpose of diagnostics this might not be a problem. However, the objective of the present study was to develop a sensitive and specific Taqman based real-time PCR for direct detection of B. bronchiseptica in canine samples, suitable for combination with real-time RT-PCR for other canine respiratory pathogens.

Material and Methods

DNA extraction

DNA was extracted from bacterial cultures by suspension of roughly 1 µL colony material in 500 µL nuclease free water followed by boiling at 98-100°C for ten minutes. The lysates were rapidly chilled in ice and then centrifuged at 13.000 × g for five minutes. The supernatant was used as PCR template. ESwabs with Amie’s medium (Copan diagnostics), used to collect clinical specimens, were shaken thoroughly and 90 µL of the liquid content was mixed with 10 µL proteinase K, 20 mg/mL (Sigma-Aldrich) and incubated at room temperature for 5 minutes prior to DNA extraction using Magnatrix 8000+ automatic platform and Vet Viral NA kit (NorDiag AB) according to the manufacturer’s instructions.

Real-time PCR assay design

To find potential PCR targets within the B. bronchiseptica genome that would amplify B. bronchiseptica, but not other Bordetellae, Gegenees software V2.2.1 for comparative genome analysis was used (available from http://www.gegenees.org/). The following whole genome sequences were used for assay design: B. bronchiseptica 253 (NC_019382.1), B. bronchiseptica MO149 (NC_018829.1), B. bronchiseptica RB50 (NC_002927.3), B. bronchiseptica KM22 (JMHR01000000), B. bronchiseptica 1289 (NZ_CAKS00000000.1), B. parapertussis Bpp5 (NC_018828.1), B. parapertussis 12822 (NC_002928.3), B. parapertussis 18323 (NC_018518.1), B. pertussis Tohama I (NC_002929.2), B. pertussis CS (NC_017223.1), B. avium 197N (NC_010645.1) and B. petrii ASM6720v1 (NC_010170.1). A real-time PCR assay was subsequently designed targeting the bfrZ gene, present in one copy in the B. bronchiseptica genome, using a primer set suggested by the software and a DNA probe that hybridizes to the PCR product which was labeled with carboxyfluorescein (FAM) at the 5’-end and a non-fluorescent quencher (BHQ1) at the 3’-end (Table 1). Sequence specificity and sensitivity of the assay were evaluated in silico by basic local alignment search tool, Primer-BLAST (http://www.ncbi.nlm.nih.gov/). The PCR assay was tested and evaluated in 15 µL PCR reactions containing nuclease free water, PerfeCTa qPCR Toughmix with low ROX reference dye (Quanta Biosciences), 400 nM of each primer, 133 nM FAM-labeled probe and 2 µL template DNA. A 96-well format thermocycler,

Table 1: Oligonucleotides used for real-time PCR and sequencing in the present study.

Table 2: Evaluation of the bfrZ PCR assay on clinical isolates and reference strains of Bordetella. Also shown are the Maldi-TOF results, appearance on Smith-Baskerville agar and oxidase reaction with tetramethyl-p-phenylenediamine (TMPD).

ND=not done; QA=isolate from quality assurance testing

ABI 7500 Fast (Applied Biosystems), was used for amplification and detection of the target sequence. Cycling conditions were as follows: 45ºC for 10 minutes, 95ºC for 10 minutes, 48 cycles of 95ºC for 15 seconds and 60ºC for 45 seconds during which fluorescence intensity was measured. Ramp rate was set to 1.6 degrees per second.

Assay validation

A total of 23 clinical isolates and one reference strain of B. bronchiseptica were used in the present evaluation (Table 2). Clinical strains had previously been isolated from various animal species during routine diagnostic work at the National Veterinary Institute, Sweden. All clinical isolates were typed as B. bronchiseptica by Maldi-TOF (Bruker Daltonik GmbH, Germany) according to the manufacturer’s instructions (Table 2). Amplification efficiency and limit of detection were determined by six replicates of ten-fold and two-fold serial dilutions of B. bronchiseptica strain CCUG 219 DNA. DNA concentration was measured by Qubit fluorometer using the QuantIT HS kit (Invitrogen) and number of genome copies was calculated as described previously (Mendoza-Cano and Sánchez-Paz 2013), assuming a genome size of 5.26 Mb. Limit of detection in clinical samples was tested by adding reference strain B. bronchiseptica CCUG 219 in serial dilutions to ESwabs (COPAN Diagnostics) collected from the nasopharynx of healthy dogs. Swabs were then processed as described above. Viable count was determined by plating dilution series of B. bronchiseptica on blood agar plates incubated at 37°C for 48 h, whereupon colonies were counted. Plates containing between 5-250 colonies were included in final

Table 3: Exclusion panel for the testing of specificity. All isolates tested negative with the bfrZ real-time PCR

Table 4: Similarity between the 16S rDNA sequence of isolate 13-VLT003007 and strains of B. bronchiseptica and related species based on BLAST search against the Greengenes and GenBank databases

calculations of CFU (colony forming unit) per mL of sample. To test assay specificity a broad-range exclusion panel of 45 bacterial isolates (Table 3), plus reference strains of B. pertussis and B. parapertussis (Table 2), were analyzed. These bacteria were chosen as they either represent the normal flora of canine upper respiratory tract (Bailie et al., 1978; Elliott et al., 2005) or are genetically closely related to B. bronchiseptica. To evaluate the PCR assay in clinical samples from dogs, 74 nasopharyngeal swabs collected from 57 dogs with clinical signs of CIRD and 17 dogs without signs of CIRD were used. All samples had been culture negative for B. bronchiseptica in routine diagnostic investigations.

16S rDNA sequencing

Amplification of 16S rDNA was done as described by Weisburg et al. (1991), with slight modifications; PCR reaction contained 2 µL template DNA, SsoFast™ EvaGreen® Supermix with Low ROX (Bio-Rad) and 250 nM of each primer (Table 1) with following cycling conditions: 95°C for 3 minutes followed by 35 cycles of 95°C for 30 seconds, 42°C for 30 seconds and 72°C for 4 minutes, followed by 72°C for 20 minutes. PCR products were verified by melt curve analysis: 95°C for 15 seconds, 60°C for 60 seconds, temperature increase from 60°C to 95°C by a 1% ramp rate, 95°C for 15 seconds and finally 60°C for 15 seconds. PCR product purification was performed by incubating 10 µL PCR product with 0.05 µL ExoI and 0.5 µL FastAP (Thermo Scientific) for 15 minutes at 37ºC followed by inactivation for 15 minutes at 85ºC. A mixture of 5 µL purified PCR product and 50 pmol of either forward or reverse primer was sent for Sanger sequencing (Macrogen, The Netherlands). Sequences were assembled using CLC Main Workbench V6.7 and compared to the NCBI 16S rDNA sequence database (nucleotide BLAST, http://blast.ncbi.nlm.nih.gov/Blast.cgi).

Results and Discussion

Assay design

For initial design of a real-time PCR assay, whole genome sequences of five B. bronchiseptica strains were used. These strains represented all four clusters of flaA types (1a, 1b, 2 and 3) described by Khayer et al., (Khayer et al., 2014) and have been isolated from different host species (dog, human, rabbit, pig and monkey). Typing of the flaA gene, which encodes a subunit of the flagellar filaments, has been used to detect genetic diversity in a number of bacterial species, for instance Campylobacter spp., Legionella pneumophila and Bordetella spp. (Fitch et al., 2005; Edwards et al., 2008; Khayer et al., 2014). The aim was to find a potential PCR target region that was sufficiently conserved even among a diverse set of isolates within the species. Related Bordetella species were used to exclude unspecific regions. The specificity and sensitivity of the potential target region was then verified in silico by Primer-BLAST before continuing on with laboratory testing.

Validation

For validation of the bfrZ real-time PCR we tested detection limit, amplification efficiency, intermediate precision, sensitivity and specificity. Limit of detection was determined by PCR of serial dilutions of purified DNA of Bordetella bronchiseptica CCUG 219 and lowest amount detected in six replicates was 8 copies/reaction. After plotting concentrations against CT-values, PCR amplification efficiency was calculated to 95% over 7 log (Figure 1).

Figure 1: Amplification efficiency of the bfrZ real-time PCR

A standard curve was generated by 10-fold serial dilutions of B. bronchiseptica CCUG 219 DNA. PCR in six replicates revealed a good linear relationship (E=95%) and a wide dynamic range.

Intermediate precision was determined by analyzing diluted DNA of CCUG 219 in three replicates in five different PCR runs at different days and coefficient of variance was calculated to 1.6%. In absence of culture positive clinical specimens, diagnostic sensitivity could not be properly evaluated; however, limit of detection in clinical samples was determined by a spike-in experiment. Lowest amount of bacteria detected in clinical samples spiked with Bordetella bronchiseptica CCUG 219 in three consecutive experiments was 4×103 CFU/mL of sample on average, which corresponds to a theoretical value of less than 10 CFU/PCR reaction, assuming a 100% DNA yield during sample preparation and extraction. Inclusion and exclusion was evaluated by a set of bacterial isolates as well as in silico. None of the bacterial isolates in the exclusion panel tested positive, and all but one isolate of B. bronchiseptica tested positive by the newly developed PCR assay, which resulted in a sensitivity (inclusion) of 96% and a specificity (exclusion) of 100%. The temperature profile included a reverse transcription step at 45°C to enable simultaneous amplification and detection of CIRD associated RNA viruses. However, the assay was also tested without this step, with shorter denaturation and elongation times and ramp rates as fast as 3.5 degrees per second, with similar amplification efficiency and detection limit (data not shown). Next, the assay was evaluated using clinical samples collected from dogs with (n=57) and without CIRD (n=17). Two dogs with signs of CIRD and one dog with no signs of CIRD were found positive for B. bronchiseptica. The PCR-positive healthy control dog had been vaccinated with an intra-nasal live attenuated B. bronchiseptica vaccine (Nobivac KC vet., Intervet, Sweden) nine days prior to sampling and most likely the PCR assay detected the vaccine strain. It has been shown that the live attenuated vaccine strain of B. bronchiseptica is shed during the first month after vaccination (Iemura et al., 2009).

16S rDNA sequencing

All but one isolate were amplified by the bfrZ assay (Table 2). Isolate 13-VLT003007, which derived from a hedgehog, was identified as B. bronchiseptica by Maldi-TOF but could not be detected with the bfrZ PCR and was therefore subjected to 16S rDNA sequencing in order to make a taxonomic analysis. BLAST analysis against 16S sequences at GenBank and Greengenes (http://greengenes.lbl.gov/cgi-bin/nph-index.cgi) showed that 13-VLT003007 was identical to previously sequenced strains of both B. bronchiseptica and B. parapertussis but also showed differences to several strains of B. bronchiseptica (Table 4). This indicates that 13-VLT003007 is closely related to both species but the 16S rDNA sequence probably does not offer sufficient resolution to determine exact taxonomic relationship of such closely related species as B. bronchiseptica and B. parapertussis.

Conclusions

Here we describe development and validation of a probe-based real-time PCR assay to detect the bfrZ gene of Bordetella bronchiseptica for diagnosis of respiratory infections in dogs. The assay was evaluated with nasopharyngeal swabs from dogs with and without clinical signs of CIRD. All swab samples had been tested negative for B. bronchiseptica with current standard culture methods; however, by the newly developed real-time PCR assay, three of the samples showed positive reaction for B. bronchiseptica. Specificity of the assay was evaluated both in silico and with an exclusion panel of bacterial isolates and no false positives were detected. There was one false negative among the B. bronchiseptica isolates but it was considered to be of minor importance for diagnostics of dogs. In conclusion, the presented real-time PCR assay for B. bronchiseptica provides a sensitive and specific detection method, suitable for combination with viral molecular diagnostic assays for determining the etiological cause of CIRD in clinical samples.

Acknowledgements

The authors would like to thank Sofia Hanås, Bodil Ström-Holst, Camilla Wikström, Erik Eriksson and Elisabeth Bagge for providing clinical samples. This study was funded by the Research Foundation of Agria Animal Insurances and Swedish Kennel Club, the Sveland Foundation for Animal Health and Welfare, Jan Skogsborg’s Foundation and MSD Animal Health. The funders did not have any influence of the contents of this study or on the decision to publish.

Author’s Contributions

TJ was responsible for assay design, experimental design, collection and analysis of data and drafting. KM was involved in data generation and collection. EE was involved in providing clinical specimens and strains and was revising the intellectual content of the manuscript. JJW was involved in providing clinical specimens, revising intellectual content and gave the final approval of the manuscript.

Conflict of Interest

The authors declare that there is no conflict of interests regarding the publication of this article.

References

• Bailie, W.E., Stowe, E.C., Schmitt, A.M. (1978) Aerobic bacterial flora of oral and nasal fluids of canines with reference to bacteria associated with bites. Journal of Clinical Microbiology 7, 223–231
• Bemis, D.A., Carmichael, L.E., Appel, M.J. (1977) Naturally occurring respiratory disease in a kennel caused by Bordetella bronchiseptica. The Cornell Veterinarian 67, 282–293
• Buonavoglia, C., Martella, V. (2007) Canine respiratory viruses. Veterinary Research 38, 355–373
• Edwards, M.T., Fry, N.K., Harrison, T.G. (2008) Clonal population structure of Legionella pneumophila inferred from allelic profiling. Microbiology 154, 852–864
• Elliott, D.R., Wilson, M., Buckley, C.M.F., Spratt, D.A. (2005) Cultivable Oral Microbiota of Domestic Dogs. Journal of Clinical Microbiology 43, 5470–5476
• Fitch, B.R., Sachen, K.L., Wilder, S.R., Burg, M.A., Lacher, D.W., Khalife, W.T., Whittam, T.S., Young, V.B. (2005) Genetic Diversity of Campylobacter sp. Isolates from Retail Chicken Products and Humans with Gastroenteritis in Central Michigan. Journal of Clinical Microbiology 43, 4221–4224
• Goodnow, R.A. (1980) Biology of Bordetella bronchiseptica. Microbiological Reviews 44, 722–738
• Greene, C.E. (2006) Infectious diseases of the dog and cat. 1387 pages, 3rd ed., Saunders/Elsevier, St. Louis, Mo.
• Iemura, R., Tsukatani, R., Micallef, M.J., Taneno, A. (2009) Simultaneous analysis of the nasal shedding kinetics of field and vaccine strains of Bordetella bronchiseptica. Veterinary Record 165, 747–751
• Khayer, B., Magyar, T., Wehmann, E. (2014) Flagellin typing of Bordetella bronchiseptica strains originating from different host species. Veterinary Microbiology 173, 270–278
• Koidl, C., Bozic, M., Burmeister, A., Hess, M., Marth, E., Kessler, H.H. (2007) Detection and Differentiation of Bordetella spp. by Real-Time PCR. Journal of Clinical Microbiology 45, 347–350
• Mendoza-Cano, F., Sánchez-Paz, A. (2013) Development and validation of a quantitative real-time polymerase chain assay for universal detection of the White Spot Syndrome Virus in marine crustaceans. Virology Journal 10, 186
• Mitchell, J.A., Brooks, H.W., Szladovits, B., Erles, K., Gibbons, R., Shields, S., Brownlie, J. (2013) Tropism and pathological findings associated with canine respiratory coronavirus (CRCoV). Veterinary Microbiology 162, 582–594
• Mitchell, J.A., Cardwell, J.M., Renshaw, R.W., Dubovi, E.J., Brownlie, J. (2013) Detection of Canine Pneumovirus in Dogs with Canine Infectious Respiratory Disease. Journal of Clinical Microbiology 51, 4112–4119
• Roorda, L., Buitenwerf, J., Ossewaarde, J.M., Van der Zee, A. (2011) A real-time PCR assay with improved specificity for detection and discrimination of all clinically relevant Bordetella species by the presence and distribution of three Insertion Sequence elements. BMC Research Notes 4, 11
• Schulz, B.S., Kurz, S., Weber, K., Balzer, H.-J., Hartmann, K. (2014) Detection of respiratory viruses and Bordetella bronchiseptica in dogs with acute respiratory tract infections. The Veterinary Journal 201, 365–369
• Tatti, K.M., Sparks, K.N., Boney, K.O., Tondella, M.L. (2011) Novel multitarget real-time PCR assay for rapid detection of Bordetella species in clinical specimens. Journal of clinical microbiology 49, 4059–4066
• Tizolova, A., Brun, D., Guiso, N., Guillot, S. (2014) Development of real-time PCR assay for differential detection of Bordetella bronchiseptica and Bordetella parapertussis. Diagnostic Microbiology and Infectious Disease 78, 347–351
• Weisburg, W.G., Barns, S.M., Pelletier, D.A., Lane, D.J. (1991) 16S ribosomal DNA amplification for phylogenetic study. Journal of Bacteriology 173, 697–703