Development of a psittaci species-specific and genotype-specific real-time PCR Tom Geens, Angelo Dewitte, Nico Boon, Daisy Vanrompay

To cite this version:

Tom Geens, Angelo Dewitte, Nico Boon, Daisy Vanrompay. Development of a Chlamydophila psittaci species-specific and genotype-specific real-time PCR. Veterinary Research, BioMed Central, 2005,36 (5-6), pp.787-797. ￿10.1051/vetres:2005035￿. ￿hal-00903003￿

HAL Id: hal-00903003 https://hal.archives-ouvertes.fr/hal-00903003 Submitted on 1 Jan 2005

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Vet. Res. 36 (2005) 787–797 787 © INRA, EDP Sciences, 2005 DOI: 10.1051/vetres:2005035 Original article

Development of a Chlamydophila psittaci species- specific and genotype-specific real-time PCR

Tom GEENSa*, Angelo DEWITTEa, Nico BOONb, Daisy VANROMPAYa

a Department of Molecular Biotechnology, Faculty of Bioscience Engineering, Ghent University, Coupure Links 653, 9000 Ghent, Belgium b Laboratory of Microbial Ecology and Technology (LabMET), Faculty of Bioscience Engineering, Ghent University, Coupure Links 653, 9000 Ghent, Belgium

(Received 12 October 2004; accepted 22 March 2005)

Abstract – A Chlamydophila psittaci species-specific real-time PCR targeting the rDNA ribosomal spacer was developed as well as a genotype-specific real-time PCR targeting the Cp. psittaci outer membrane A (ompA) . The SYBR Green-based species-specific real-time PCR detected Cp. psittaci genotypes A to F, and the recently discovered E/B genotype. The genotype-specific real-time PCR could easily distinguish genotypes C, D, F by use of TaqMan probes. Genotypes A, B and E could not be distinguished from each other by simply using TaqMan probes. For this purpose, non-fluorescent competitor oligonucleotides, had to be used next to the TaqMan probes. Genotype E/B could only be detected by use of a minor groove binder (MGB) probe. Both real-time PCR assays allowed reproducible, sensitive (10 rDNA or ompA copies/µL DNA extract) and specific detection of Cp. psittaci DNA. The genotype-specific real-time PCR was compared to ompA sequencing and ompA restriction fragment length polymorphism (RFLP) analysis using five Cp. psittaci field isolates (99, 61/8, 7344/2, 8615/1 and 7778B15) each consisting of two different genotypes. The currently developed real-time PCR assays were used in a case study on a veterinary school and a farm. In the veterinary school, Cp. psittaci genotypes D, E/B and F were detected in all five groups of turkeys, and one veterinarian who was taking care of all these turkeys. On the turkey farm, the presence of two Cp. psittaci genotype B infection waves was demonstrated in one randomly selected turkey, the first wave at the age of 6 weeks, and the second at the age of 12 weeks.

Chlamydophila psittaci / real-time PCR / species-specific / genotype-specific / diagnosis

1. INTRODUCTION zoonotic disease in . psittaco- are Gram-negative obli- sis mostly originates from exposure to infected gate intracellular replicating in psittacines, pigeons or poultry, mainly tur- mucosal epithelial cells and macrophages, keys [11] or ducks [14, 18]. Symptoms in causing disease in , humans, other man vary from asymptomatic to severe sys- and marsupials. Chlamydophila temic disease [3]. psittaci (formerly psittaci) is a Cp. psittaci in birds and man still repre- respiratory avian pathogen able to cause sents a diagnostic challenge. Isolation is

* Corresponding author: [email protected]

Article published by EDP Sciences and available at http://www.edpsciences.org/vetres or http://dx.doi.org/10.1051/vetres:2005035 788 T. Geens et al.

Table I. Cp. psittaci strains used for developing species- and genotype-specific primers and probes as well as inhibition control plasmids.

Strain Reference Country Genotype Control plasmid 90/1051 [9] Belgium Amazona sp. A pGemT::CpPsGAS 41A12 [9] Belgium Meleagris gallopavo B pGemT::CpPsGBS GD [15] Gemany Anas platyrhyncos C pGemT::CpPsGCS 7344/2 [9] Italy Columba livia D pGemT::CpPsGDS 3759/2 [9] Italy Columba livia E pGemT::CpPsGES pGemT::CpPsSS 7778B15 [9] Belgium Meleagris gallopavo F pGemT::CpPsGFS WS/RT/E30 [9] Germany Anas platyrhyncos E/B pGemT::CpPsGE/BS labour-intensive, relatively insensitive and tive post PCR detection methods and are not without danger. In birds, serology is not quantitative or only semi-quantitative. used and is nowadays mostly performed by In addition, genotyping still needs to be per- an enzyme linked immunosorbent assay [6, formed by ompA restriction fragment length 21]. However the interpretation of the polymorphism (RFLP) analysis and ompA results is often difficult since most birds sequencing, and both techniques often have pre-existing antibodies from previous require bacterial culture, since amplifica- exposures, and antibodies can persist for up tion of full length ompA can mostly not be to several months. In man, the complement carried out directly from clinical specimens. binding assay (CBA) is often used. How- The present study describes the develop- ever, the CBA cannot distinguish Cp. psit- ment of a Cp. psittaci species-specific real- taci specific antibodies from antibodies time PCR. Additionally, we describe the against other chlamydial human pathogens, development of a genotype-specific real- like Cp. pneumoniae and Chlamydia tra- time PCR allowing the identification of all chomatis. As a result, the CBA is more and avian Cp. psittaci genotypes including the more often replaced by the micro-immun- recently discovered new E/B genotype [9]. ofluorescence (MIF) test. The MIF test dis- The performance of the genotype-specific tinguishes all chlamydial species and real-time PCR was compared to ompA measures IgG as well as IgM titres, allow- ing the detection of recent . How- RFLP analysis and ompA sequencing. The ever, early antibiotic treatment can interfere species- and genotype-specific real-time with antibody formation and in some PCR assays were used in a case study in a patients, antibody responses can develop veterinary school and on a turkey farm. rather slowly. Moreover, obligatory exam- ination of paired sera removes serology from immediate clinical relevance. There- 2. MATERIALS AND METHODS fore, several commercial antigen detection methods were developed for both birds and 2.1. Bacterial cultures man but they are either insensitive and/or less specific [21, 25]. Due to these short- Cp. psittaci genotypes A to F plus E/B comings, nucleic acid amplification methods strains 90/1051, 41A12, GD, 7344/2, 3759/2, have been designed. However, currently 7778B15 and WS/RT/E30 (Tab. I) were described polymerase chain reaction (PCR) used for the development of the species- and assays use labour-intensive and/or insensi- genotype-specific real-time PCR assays. Chlamydophila psittaci species/genotype real-time PCR 789

Bacteria were grown in cycloheximide fied using Qiagen spin columns (Westburg, treated Buffalo Green Monkey (BGM) cells Leusden, The Netherlands) and cloned into as described previously [26]. For each pGem®-T (Promega, Madison, WI, USA) strain, an infected monolayer of 300 cm2 following the manufacturer’s protocol. was disrupted by freezing and thawing, fol- Sequence analyses were performed by the lowed by ultrasonic treatment for 1 min in a VIB Genetic Service Facility (University of tabletop sonicator (Bransonic 12, BIOMEDe- Antwerp, Antwerp, Belgium) using vector vice, San Pablo, CA, USA). A two-hundred associated T7 and SP6 priming sites. millilitre culture harvest was centri- Sequence alignment using ClustalX soft- fuged for 10 min (1 000 × g, 4 °C) and sub- ware [22] allowed us to select a Cp. psittaci sequently concentrated by ultracentrifugation species-specific inhibition control plasmid for 1 h (45 000 × g, 4 °C). Bacteria were (pGemT::CpPsSS, Tab. I). resuspended in 2 mL sucrose phosphate glutamate buffer (SPG, 218 mM sucrose, 2.4. Species-specific real-time PCR 38 mM KH2PO4, 7 mM K2HPO4, 5 mM L-glutamic acid) and stored at –80 °C until use. Real-time PCR was performed with the LightCycler 2.0 Instrument (Roche, Applied 2.2. Preparation of genomic DNA Science, Penzberg, Germany) using the LightCycler FastStart DNA MasterPLUS Genomic DNA for real-time PCR assays SYBR Green I kit and LightCycler Capil- was prepared as described by Wilson et al. laries. The reaction mixture (20 µL) was [34]. DNA samples were further purified by prepared according to the manufacturer’s extracting them twice with 200 µL phenol- protocol: 11 µL PCR grade water, 2 µL of chlorophorm (1:1). Precipitation was per- primer mixture (300 nM CpPsSSfor and formed (1 h, –80 °C) by adding 20 µL sodi- CpPsSSrev), 2 µL 10× Master Mix, 5 µL of umacetate (3M) and 400 µL of 100% DNA template. The cycling conditions ethanol. The pellets obtained following were as follows: 50 cycles of 95 °C for 10 s, centrifugation (20 min, 4 °C, 16 060 × g) 63 °C for 10 s and 72 °C for 8 s. All default were washed for 5 min with 500 µL of 70% program settings were used. Standard graphs ethanol (4 °C, 16 060 × g) and were finally of the Cycle threshold (Ct) values, obtained resuspended in 30 µL bidest. by testing tenfold serial dilutions (108 to 101) of the purified species-specific inhibi- tion control plasmid, were used for quanti- 2.3. Species-specific primers fication. Ct-values were automatically and inhibition control plasmid converted into initial template quantities Published ribosomal spacer sequences (N0) using the LightCycler Software 4.0. of all chlamydial species [7] were aligned DNA from clinical samples was always using ClustalX software (default settings) tested in the presence of control plasmid [22]. Cp. psittaci species-specific forward (50 copies/µL) to check for PCR inhibitors. (CpPsSSfor) and reverse (CpPsSSrev) primers (Isogen Life Sciences, Maarssen, 2.5. Genotype-specific primers, probes The Netherlands) were designed using and inhibition control plasmids Primer Express Software (Applied Biosys- tems, Foster City, CA, USA). Primer spe- The ompA gene of the genotypes A to F cificity was checked by BLAST [1]. and E/B reference strains (Tab. I) was Subsequently, the rDNA of genotypes A amplified as described previously [32]. to E/B (Tab. I) was PCR amplified using OmpA were cloned in pGEM®-T and 35 cycles of 95 °C for 20 s, 63 °C for 20 s sequenced by the VIB Genetic Service Facility and 72 °C for 30 s. PCR products were puri- (Antwerp, Belgium). Plasmids were used as 790 T. Geens et al.

inhibition controls in the genotype-specific genotypes present in clinical samples (N0) real-time PCR. Genotypes A to E/B spe- were quantified using the LightCycler 4.0 cific primers (e.g. Cp. psittaci Genotype A- software. Specific forward and reverse primer; CpPs- GASfor and CpPsGASrev), (Invitrogen, 2.7. Analytical sensitivity Carlsbad, CA, USA) as well as genotype- and specificity specific TaqMan probes (e.g. CpPsGAS- pro) (Applied Biosystems, Foster City, CA, The analytical sensitivity of both PCR USA) were designed from aligned variable assays was evaluated using tenfold serial segments of presently sequenced ompA genes dilutions (108 to 101 copies/µL) of all inhi- using Primer Express Software (Applied Bio- bition control plasmids. The specificity was systems). Specificities were checked using evaluated using (1) genomic DNA from BLAST [1]. Cp. pneumoniae, Cp. felis, Cp. caviae, Cp. abortus, Cp. psittaci, C. muridarum and C. trachomatis as well as (2) DNA from 2.6. Genotype-specific real-time PCR other bacterial species commonly found in the avian or human and Genotype-specific probes (Applied Bio- (3) genomic DNA from avian and human systems) were 5’ and 3’ labelled with the respiratory tissue (Tab. II). reporter dye 6-carboxyfluorescein (FAM) and the quencher dye carboxytetramethyl- rhodamine (TAMRA), respectively. Geno- 2.8. Genotype-specific real-time PCR type-specific reactions were performed compared to ompA RFLP analysis with the LightCycler 2.0 instrument in Light- and ompA sequencing Cycler Capillaries, using the LightCycler We previously sequenced the ompA FastStart DNA MasterPLUS Hybridisation genes of 21 European Cp. psittaci field iso- Probes kit, in a total reaction volume of lates, revealing the presence of five mixed- 20 µL (9 µL PCR grade water, 2 µL of primer genotype infections [9]. Five field isolates (300 nM) / probe (300 nM) / competitor (99, 61/8, 7344/2, 8615/1 and 7778B15), (50 nM or 150 nM) mixture, 4 µL 5× Master µ each consisting of two different genotypes, Mix, 5 L of DNA template). Cycling con- were selected in order to compare the per- ditions were as follows: 95 °C for 10 min formances of the genotype-specific real- and subsequently 50 cycles of 95 °C for 10 s time PCR, ompA RFLP analysis and ompA followed by 63 °C (genotype B) or 60 °C sequencing. The latter two tests were per- (all other genotypes), for 20 s. Genotypes formed as described elsewhere [32]. A, B and E could only be distinguished by using competitor oligonucleotides enhanc- ing the TaqMan probe specificity. The 2.9. Case studies detection of the E/B genotype was made 2.9.1. In a veterinary school possible using a minor groove binder (MGB) probe (Applied Biosystems) [17]. Five groups of five conventional turkeys Standard Ct-value graphs obtained from were brought to the Faculty of Veterinary testing serial dilutions of purified control Medicine (Ghent University) to become plasmids (108 to 101) were used for quan- experimentally infected with Ornithobac- tification. Clinical samples from case stud- terium rhinotracheale (ORT). However, ies were tested in the presence of genotype- the birds started to die, even before being specific control plasmids (50 ompA copies/ infected, showing severe respiratory disease. µL) in order to check for PCR inhibitors. Routine bacteriology and virology revealed Ct-values for clinical samples were plotted no pathogens. Therefore, all remaining ani- against standard graphs and Cp. psittaci mals were sampled using one pharyngeal Chlamydophila psittaci species/genotype real-time PCR 791

Table II. Organisms of non-chlamydial origin used for specificity testing.

Human pathogens Avian pathogens Acinetobacter baumannii Proteus mirabilis Acinetobacter sp. Proteus vulgaris Aspergillus flavus Bordetella bronchiseptica Salmonella sp. Candida albicans Citrobacter braakii Serratia marescens Enterococcus faecelis Citrobacter freundii Staphylococcus aureus Escherichia coli Corynebacterium urealyticum Stenotrophomonas Klebsiella sp. Enterobacter aerogenes maltophilia Mycobacterium avium Enterococcus sp. Streptococcus agalactiae Mycoplasma gallisepticum Escherichia coli Streptococcus nilleri Mycoplasma meleagridis Haemophilus influenza Streptococcus pyogenes Ornithobacterium rhinotracheale Hafnia alvei Pasteurella sp. Klebsiella oxytoca Proteus mirabilis Klebsiella pneumoniae Pseudomonas sp. Legionella pneumophila Salmonella enteritidis Moraxella catarrhalis Salmonella gallinarum Morganella morganii Salmonella pullorum Mycoplasma pneumoniae Staphylococcus sp. Pneumococcus sp. Streptococcus sp. Propionibacterium sp. Xanthomonas maltophila swab for each group. Being aware of having the presence of the Cp. psittaci genotype B a possible Cp. psittaci infection, the respon- during both infection waves [9]. Weighted sible veterinarian was sampled twice, pro- pharyngeal swabs from one randomly viding a pharyngeal swab on the same day selected turkey at the age of 3, 6, 8, 12 and (t1) and a second one two weeks later (t2). 15 weeks were used for the genotype B spe- All swabs were used for isolation in BGM cific real-time PCR in order to confirm cells as described previously [32] as well as ompA sequencing results. for the species- and genotype-specific real- time PCR. 3. RESULTS

2.9.2. On a turkey farm 3.1. Species-specific primers The occurrence of Cp. psittaci on a Bel- and inhibition control plasmid gian turkey farm was examined from pro- Cp. psittaci species-specific forward duction onset until slaughter at 15 weeks (CpPsSSfor) and reverse (CpPsSSrev) of age. Seroconversion demonstrated the primers generated a PCR product of 151 bp presence of a first Cp. psittaci infection (Tab. III). Aligment of rDNA of genotypes when the turkeys were 3 to 6 weeks of age, A to E/B allowed us to choose a Cp. psittaci and a second one when they were 8 to species-specific internal inhibition control 12 weeks of age [24]. Sequencing revealed plasmid designated pGemT::CpPsSS (Tab. I). 792 T. Geens et al.

Table III. Primers, probes and competitors.

Oligonucleotide Sequence (5’-3’) Positiona Specificity CpPsSSfor TTATTAAGAGCTATTGGTGGATGCC 1 822 Cp. psittaci CpPsSSrev AACGTATAATGGTAGATGATTAATCTACCG 1 972 CpPsGASfor GGTTTTCAGCTGCAAGCTCAA 488 Genotype A CpPsGASpro CTACCGATCTTCCAACGCAACTTCCTAACG 512 CpPsGAScomB CTACCGATCTTCCAATGCAACTTCCTAACGb 512 CpPsGASrev CCACAACACCTTGGGTAATGC 565 CpPsGBSfor AATAGGGTTTTCAGCTACCAACTCAA 483 Genotype B CpPsGBSpro TCTACCGATCTTCCAATGCAACTTCCTAACGTA 511 CpPsGBScomA TCTACCGATCTTCCAACGCAACTTCCTAACGTA 511 CpPsGBScomE+E/B TCTACCGAGCTTCCAATGCAACTTCCTAACGTA 511 CpPsGBSrev CCACAACACCTTGGGTAATGC 565 CpPsGCSfor GCATCGCTCAACCTAAATTGG 929 Genotype C CpPsGCSpro TCTGCTGTTATGAACTTGACCACATGGAACC 952 CpPsGCSrev ATTGTGGCTTCCCCTAAAAGG 1 009 CpPsGDSfor AACCACTTGGAACCCAACACTTT 969 Genotype D CpPsGDSpro AGGAAAGGCCACAACTGTCGACGG 993 CpPsGDSrev CGAAGCAAGTTGTAAGAAGTCAGAGTAA 1 062 CpPsGESfor CCAAGCCTTCTAGGATCAAGGA 982 Genotype E CpPsGESpro TACTTTGCCCAATAATGGTGGTAAGGATGTTCTATC 1 005 CpPsGEScomA+B TGCTTTGCCCAATAATAGTGGTAAGGATGTTCTATC 1 005 CpPsGEScomE/B TGCTTTGCCCAATAATGCTGGTAAGGATGTTCTATC 1 005 CpPsGESrev CGAAGCAATTTGCAAGACATCA 1 062 CpPsGFSfor GCAACTTTTGATGCTGACTCTATCC 904 Genotype F CpPsGFSpro CATCGCTCAACCTAAATTAGCCGCTGC 930 CpPsGFSrev GTTCCATGTGGTCAAGTTCAAAAC 981 CpPsGE/BSfor CCAAGCCTTCTAGGATCAACCA 982 Genotype E/B CpPsGE/BSpro TGCTTTGCCCAATAATGCTGc 1 005 CpPsGE/BScomA+B TGCTTTGCCCAATAATAGTG 1 005 CpPsGE/BScomE TACTTTGCCCAATAATGGTG 1 005 CpPsGE/BSrev TGCAAGACATCAGATAGAACATCCTT 1 052 a Binding position of the 5’ end on the reference strain. b Nucleotide mismatches between the competitor and the probe are marked in grey. c MGB probe.

3.2. Species-specific PCR all Cp. psittaci genotypes resulted in the expected PCR products with amplification The species-specific PCR detected all curves exceeding the 0.02 threshold before known Cp. psittaci genotypes, including cycle 23. The species-specific PCR was the recently discovered E/B genotype. able to detect 10 rDNA copies/µL DNA Amplification of genomic DNA extracts of extract. Chlamydophila psittaci species/genotype real-time PCR 793

The species-specificity test revealed a other genotypes. Adding 50 mM of com- Ct-value of 20 for Cp. psittaci with a prod- petitors to the PCR reaction mixture in uct melting temperature of 79.3 °C. Ampli- combination with the use of a cycling tem- fication of Cp. pneumoniae, Cp. felis, Cp. perature of 63 °C for the genotype B spe- abortus, C. muridarum and C. trachomatis cific reaction, could solve the problem rDNA generated no reaction products. (Tab. III). For the genotype B specific reac- However, amplification of Cp. caviae tion competitor CpPsGBScomA had to be rDNA generated a fluorescent signal at a added. Genotype A specific detection required Ct-value above 37. The amplified rDNA CpPsGAScomB, while specific detection showed a melting temperature of 79.0 °C of genotype E required CpPsGEScomA, instead of 79.3 °C. Sequencing revealed a CpPsGAScomB and CpPsGEScomE/B. Cp. caviae-specific rDNA sequence of The detection limit of the genotype-specific 151 bp. The Cp. psittaci species-specific real-time PCR was 10 ompA copies/µL primers did not react with other pathogens DNA extract. Standard curves made using commonly found in the avian and human 108 to 101 copies/µL showed almost ideal respiratory tract (Tab. II), neither with slopes around –3.3 with correlation coeffi- DNA from avian or human respiratory tissue. cients > 98.5%. The Cp. psittaci genotype- specific primers did not react with other pathogens commonly found in the avian 3.3. Genotype-specific primers, probes and human respiratory tract (Tab. II) neither and inhibition control plasmids with DNA from avian or human respiratory tissue. The cloned ompA genes from Cp. psittaci genotypes A to E/B revealed sequences of 1 065 to 1 098 bp depending on the genotype 3.5. Genotype-specific real-time PCR (GenBank accession numbers AY762608, compared to ompA RFLP AY762609, AY762610, AY762611, and ompA sequencing AY762612 and AY762613). Sequences of the determined genotype-specific primers Five Cp. psittaci field isolates, each con- generating amplicons of 71 to 85 bp as well sisting of a mixture of 2 different genotypes as from the genotype-specific TaqMan were used to compare the developed geno- probes are listed in Table III. The detection type-specific real-time PCR with the over- ompA of the E/B genotype was only possible all used RFLP genotyping method ompA using a minor groove binding (MGB) probe and sequencing. All 10 genotypes, present in those 5 field isolates, could be and genotypes A, B and E could only be dis- detected by the genotype-specific real-time tinguished from each other using competi- PCR, including the recently discovered tor oligonucleotides next to the TaqMan E/B genotype (Tab. IV). OmpA RFLP probes (Tab. III). incorrectly characterised E/B genotype as genotype E. 3.4. Genotype-specific PCR We developed a genotype-specific real- 3.6. Case studies time PCR detecting all 7 known avian Cp. psittaci genotypes. Genotype C-, D-, F- and 3.6.1. In a veterinary school E/B- specific probes specifically reacted with the corresponding genotypes and were Cp. psittaci was isolated from all exam- therefore easily distinguishable. However, ined samples, which was confirmed by the the genotype A-, B- and E- specific Taq- species-specific real-time PCR detecting Man probes not only reacted with the cor- Cp. psittaci rDNA in all pharyngeal swabs. responding genotype, but also with the two Ribosomal DNA copies/µL DNA extract in 794 T. Geens et al.

Table IV. Comparison of the genotype-specific 4. DISCUSSION real-time PCR with ompA RFLP analysis using 5 isolates each consisting of 2 different geno- Real-time technology offers the possi- types as determined with ompA sequencing. bility to automatically combine amplifica- tion, specific hybridisation and detection in Isolate OmpA OmpA Real-time PCR one single test, allowing specific and sen- sequencing RFLP No. sitive gene quantification with a minimal A A 918 contamination risk. The technique has been 99 E/B E 4020 used to detect Cp. pneumoniae [10, 16, 23, A A 111 000 33], C. trachomatis [19], Cp. felis [12, 13, 61/8 E/B E 422 000 20] and Cp. pecorum [5]. The present study is the first to demonstrate the use of real- B B 5160 7344/2 time PCR to specifically detect Cp. psittaci D D 1840 as well as its presently known genotypes. In B B 9000 8615/1 the currently developed real-time PCR, kits E/B E 11 200 with the AmpErase® UNG system incorpo- B B 3660 7778B15 rated were used to prevent post PCR carry F F 2900 over, and constructed species- or genotype- specific plasmids were applied in the assays as internal inhibition controls. The present turkey groups taken at t1 varied from 790 data show our species-specific and geno- to 1 772 while the human sample taken at type-specific real-time PCR assays to be t1 and t2 showed 2 160 and 710 rDNA cop- potentially useful for testing both avian ies/µL DNA extract, respectively. Geno- (chlamydiosis) and human () spec- type-specific real-time PCR revealed the imens. The species-specific real-time PCR presence of genotypes D, F and E/B in all could be suitable for diagnosing psittacosis turkey groups as well as in the veterinarian. in man, by distinguishing the human patho- Turkeys as well as the veterinarian showed gens C. trachomatis and Cp. pneumoniae from the highest amount of ompA copies/µL the zoonotic agents Cp. felis, Cp. abortus DNA extract for genotype E/B (up to 13 000 and Cp. psittaci. The genotype-specific for both turkeys and the veterinarian) fol- PCR could be used for epidemiological lowed by genotype F (up to 7 000 for tur- research in both birds and humans. How- keys and up to 2 000 for the veterinarian) ever, further research is needed. and genotype D (up to 1 000 for turkeys and The specificity of all our primers was up to 2 000 for the veterinarian). The veter- checked by alignment with sequences inarian showed no clinical signs at both deposited in public databases revealing only sampling time points. perfect matches with Cp. psittaci DNA. Only avian chlamydial DNA could be 3.6.2. On a turkey farm amplified with the exception of Cp. caviae DNA, generating a Cp. caviae-specific ampli- DNA extracts from five different pha- con in the Cp. psittaci species-specific real- ryngeal swabs (weeks 3, 6, 8, 12 and 15) of time PCR. However, the Cp. caviae ampli- one randomly selected turkey, shown pre- con could easily be distinguished from the viously to be infected with genotype B, Cp. psittaci species-specific amplicon were tested in the genotype-specific real- since the Ct-value of the latter amplification time PCR. Seroconversions, demonstrated was remarkably lower (20 against > 37) and elsewhere when turkeys were 6 weeks of the melting temperature of the non-specific age and 12 weeks of age, were correlated to amplicon was 0.3 °C higher than for the a two- to threefold increase in Cp. psittaci Cp. psittaci specific amplicon. Moreover, genotype B ompA DNA/µL DNA extract. Cp. caviae is normally not present in birds Chlamydophila psittaci species/genotype real-time PCR 795 and humans since it only infects guinea 21 examined birds, which is remarkably and therefore does not compromise the use high as compared to serotyping, indicating of the currently developed Cp. psittaci spe- that genotype-specific real-time PCR is more cies-specific real-time PCR in birds and suitable for epidemiological examinations. humans. Relative copy numbers could be deter- Cp. psittaci comprises 9 known ompA mined in both PCR assays by using stand- genotypes (A to F, E/B, WC and M56). All ard curves for the target of interest. The of them, except for genotype E/B can be analytical sensitivity of the species-specific identified using ompA RFLP analysis [30] and genotype-specific real-time PCR was or ompA sequencing [4]. Genotype E/B can 10 rDNA or ompA copies/µL DNA extract. only be detected using ompA sequencing The species- and genotype-specific real- since it generates the same restriction pat- time PCR assays were successfully used in tern as genotype E [9]. Genotypes A, B and two different case studies. In the first case D differ in pathogenicity for turkeys [28, study, turkeys were infected with three dif- 29] but the pathogenic significance of avian ferent genotypes showing the highest amount genotypes for different host species, includ- of genotype E/B organisms in each swab ing man has to be investigated further. The followed by genotype F and D, respec- presently developed genotype-specific real- tively. Genotype E/B has mainly been time PCR, allowing the detection of the detected in ducks but also in two pigeons newly discovered E/B genotype by the use and one . Clinical data on the ducks of an MGB probe, could contribute to this research. MGB probes are shorter than nor- and the parrot were not available. The mal TaqMan probes and show a higher spe- pigeons showed no clinical signs [9]. This cificity for single base mismatches. Moreover, is the first time genotype E/B has been dis- their fluorescence quenching is higher, covered in turkeys. So far, genotype F has resulting in increased sensitivity [17]. We only been discovered in one American par- used competing oligonucleotides to distin- akeet (strain VS225) and one Belgian tur- guish genotypes A, B and E. Competitors key (strain 7778B15) [2, 9]. Genotype D have already been used in Fluorescence In has often been isolated from turkeys being Situ Hybridisation (FISH), blocking non- responsible for either severe or mild respi- specific probe binding sites on contaminating ratory signs [31]. As for the turkeys, the vet- DNA and as such increasing the accessibil- erinarian also had the highest amount of ity of the target DNA for the labelled probes genotype E/B organisms present in each [8]. However, the use of these non-fluores- pharyngeal swab. Accurate quantification cent helper oligonucleotides is innovative in swabs taken at different time points was in real-time technology. The genotype-spe- rather difficult since the swabs were not cific real-time PCR was certainly more weighted as in the case study on the turkey rapid, easier to perform and more sensitive farm and primers for human house keeping than ompA RFLP and ompA sequencing genes were not included. The veterinarian since the latter two typing methods often showed no clinical signs at sampling times, required bacterial culture. Moreover, all which can be normal regarding the incuba- genotypes present in the 5 “mixed infec- tion time of approximately two weeks and tion” samples could be detected. Infections in the following weeks and months, the vet- with two different Cp. psittaci strains have erinarian experienced only the usual “com- already been demonstrated in the past by mon cold”. This is the first report on Vanrompay et al., detecting two different Cp. psittaci genotype D, E/B and F infec- ompA serovars in 4 of 105 examined tion in man. In the second case study, the infections [27, 30]. The genotype-specific genotype B infection on the farm could be real-time PCR demonstrated the presence proven and the number of ompA copies/µL of two different genotypes in 5 of the of DNA extract corresponded remarkably 796 T. Geens et al. well to the observed seroconversion, indic- eration of protein database search programs, ative of the presence of replicating bacteria. Nucleic Acids Res. 25 (1997) 3389–3402. The present results show the potential [2] Andersen A.A., Two new serovars of Chlamy- dia psittaci from North American birds, J. usefulness of the Cp. psittaci species-spe- Vet. Diagn. Invest. 9 (1997) 159–164. cific and genotype-specific real-time PCR [3] Andersen A.A., Vanrompay D., Avian chlamy- in both the veterinary and human clinic. diosis, Rev. Sci. Tech. 19 (2000) 396–404. However, technical performances of both [4] Bush R.M., Everett K.D., Molecular evolu- assays can be further improved by including tion of the Chlamydiaceae, Int. J. Syst. Evol. primers for avian and human house keeping Microbiol. 51 (2001) 203–220. genes allowing a more accurate quantifica- [5] DeGraves F.J., Gao D., Hehnen H.R., Schlapp tion of bacterial DNA. Quantification is T., Kaltenboeck B., Quantitative detection of especially useful when monitoring an Chlamydia psittaci and C. pecorum by high- sensitivity real-time PCR reveals high prev- infection or antibiotic treatment. In addi- alence of vaginal infection in , J. Clin. tion, the genotype-specific real-time PCR Microbiol. 41 (2003) 1726–1729. could be adapted to a more cost-effective [6] Dorrestein G.M., Wiegman L.J., Inventory of multiplex assay. the shedding of Chlamydia psittaci by para- keets in the Utrecht area using ELISA, Tijd- schr. Diergeneeskd. 114 (1989) 1227–1236. ACKNOWLEDGEMENTS [7] Everett K.D., Bush R.M., Andersen A.A., Emended description of the order Chlamy- We would like to thank E.F. Kaleta (Institut diales, proposal of Parachlamydiaceae fam. für Geflügelkrankheiten, Justus-Liebig-Univer- nov. and Simkaniaceae fam. nov., each con- sität, Giessen, Germany), S. Magnino (Istituto taining one monotypic genus, revised taxon- Zooprofilattico Sperimentale della Lombardia e omy of the family Chlamydiaceae, including dell’ Emilia Romagna, Sezione di Pavia, Italy), a new genus and five new species, and stand- ards for the identification of organisms, Int. A.A. Andersen (United States Department of J. Syst. Bacteriol. 49 (1999) 415–440. Agriculture, National Animal Disease Center, Ames, United States) and M. Mariën (Department [8] Fuchs B.M., Glockner F.O., Wulf J., Amann of Pathology, Bacteriology and Poultry Diseases, R., Unlabeled helper oligonucleotides increase the in situ accessibility to 16S rRNA of flu- UGent, Belgium) for providing Cp. psittaci iso- orescently labeled oligonucleotide probes, lates or clinical specimens. E. De Graef (Depart- Appl. Environ. Microbiol. 66 (2000) 3603– ment of Pathology, Bacteriology and Poultry 3607. Diseases, UGent, Belgium) and G. Claeys (Depart- [9] Geens T., Desplanques A., Van Loock M., ment of Clinical Biology, Microbiology and Bönner B.M., Kaleta E.F., Magnino S., Immunology, UGent, Belgium) are acknowledged Andersen A.A., Everett K.D.E., Vanrompay for providing human or avian respiratory patho- D., Sequencing of the Chlamydophila psit- gens. J. Vandesompele (Center for Medical Genet- taci ompA gene reveals a new genotype, E/B, ics, UGent, Belgium) is acknowledged for helpful and the need for a rapid discriminatory gen- discussions. J. Masschelein (Roche Diagnostics), otyping method, J. Clin. Microbiol. 43 (2005) K. Verminnen (Department of Molecular Biotech- 2456–2461. nology, UGent, Belgium) and A. Desplanques [10] Hardick J., Maldeis N., Theodore M., Wood (Department of Virology, Parasitology and Immu- B.J., Yang S., Lin S., Quinn T., Gaydos C., nology, UGent, Belgium) are acknowledged for Real-time PCR for excellent assistance during the experiments. utilizing the Roche Lightcycler and a 16S The Flemish Institute for Scientific Research rRNA gene target, J. Mol. Diagn. 6 (2004) (Grant S2/5DPE174) is acknowledged for finan- 132–136. cial support. Ghent University is acknowledged [11] Hedberg K., White K.E., Forfang J.C., Korlath for providing a grant (Grant 01113401) to T. Geens. J.A., Friendshuh K.A., Hedberg C.W., MacDonald K.L., Osterholm M.T., An out- break of psittacosis in Minnesota turkey REFERENCES industry workers: implications for modes of transmission and control, Am. J. Epidemiol. [1] Altschul S.F., Madden T.L., Schaffer A.A., 130 (1989) 569–577. Zhang J., Zhang Z., Miller W., Lipman D.J., [12] Helps C., Reeves N., Tasker S., Harbour D., Gapped BLAST and PSI-BLAST: a new gen- Use of real-time quantitative PCR to detect Chlamydophila psittaci species/genotype real-time PCR 797

Chlamydophila felis infection, J. Clin. Chlamydia pneumoniae, J. Clin. Microbiol. Microbiol. 39 (2001) 2675–2676. 40 (2002) 575–583. [13] Helps C., Reeves N., Egan K., Howard P., [24] Van Loock M., Geens T., de Smit L., Nauwynck Harbour D., Detection of Chlamydophila H., Van Empel P., Naylor C., Hafez H.M., felis and feline herpesvirus by multiplex real- Goddeeris B., Vanrompay D., Key role of time PCR analysis, J. Clin. Microbiol. 41 Chlamydophila psittaci on Belgian turkey (2003) 2734–2736. farms in association with other respiratory [14] Hinton D.G., Shipley A., Galvin J.W., Harkin pathogens, Vet. Microbiol. 107 (2005) 91–101. J.T., Brunton R.A., Chlamydiosis in workers [25] Vanrompay D., Laboratory Medicine Avian at a duck farm and processing plant, Aust. and Exotic Pets, in: Fudge A.M. (Ed.), Avian Vet. J. 70 (1993) 174–176. Chlamydial Diagnostics, W.B. Saunders [15] Illner V.F., Zur Frage der Ubertragung des Company, Philadelphia, 2000, pp. 99–110. Ornithosevirus durch das Ei, Monatsh. Vet- [26] Vanrompay D., Ducatelle R., Haesebrouck erinärmed. 17 (1960) 116–117. F., Diagnosis of avian chlamydiosis: specif- icity of the modified Gimenez staining on [16] Kuoppa Y., Boman J., Scott L., Kumlin U., smears and comparison of the sensitivity of Eriksson I., Allard A., Quantitative detection isolation in eggs and three different cell cul- of respiratory Chlamydia pneumoniae infec- tures, J. Vet. Med. 39 (1992) 105–112. tion by real-time PCR, J. Clin. Microbiol. 40 (2002) 2273–2274. [27] Vanrompay D., Andersen A.A., Ducatelle R., Haesebrouck F., Serotyping of European iso- [17] Kutyavin I.V., Afonina I.A., Mills A., Gorn lates of Chlamydia psittaci from poultry and V.V., Lukhtanov E.A., Belousov E.S., Singer other birds, J. Clin. Microbiol. 31 (1993) M.J., Walburger D.K., Lokhov S.G., Gall 134–137. A.A., Dempcy R., Reed M.W., Meyer R.B., Hedgpeth J., 3’-minor groove binder-DNA [28] Vanrompay D., Ducatelle R., Haesebrouck probes increase sequence specificity at PCR F., Pathogenecity for turkeys of Chlamydia extension temperatures, Nucleic Acids Res. psittaci strains belonging to the avian serovars 28 (2000) 655–661. A, B and D, Avian Pathol. 23 (1994) 247–262. [18] Newman C.P., Palmer S.R., Kirby F.D., Caul [29] Vanrompay D., Mast J., Ducatelle R., E.O., A prolonged outbreak of ornithosis in Haesebrouck F., Goddeeris B., Chlamydia psittaci duck processors, Epidemiol. Infect. 108 in turkeys: pathogenesis of infections in avian serovars A, B and D, Vet. Microbiol. (1992) 203–210. 47 (1995) 245–256. [19] Solomon A.W., Holland M.J., Burton M.J., [30] Vanrompay D., Butaye P., Sayada C., Ducatelle West S.K., Alexander N.D., Aguirre A., Massae R., Haesebrouck F., Characterization of P.A., Mkocha H., Munoz B., Johnson G.J., avian Chlamydia psittaci strains using omp1 Peeling R.W., Bailey R.L., Foster A., Mabey restriction mapping and serovar-specific D.C., Strategies for control of : monoclonal antibodies, Res. Microbiol. 148 observational study with quantitative PCR, (1997) 327–333. Lancet 362 (2003) 198–204. [31] Vanrompay D., Butaye P., Van Nerom A., [20] Sykes J.E., Allen J.L., Studdert V.P., Browning Ducatelle R., Haesebrouck F., The preva- G.F., Detection of feline calicivirus, feline lence of Chlamydia psittaci infections in Belgian herpesvirus 1 and Chlamydia psittaci commercial turkey poults, Vet. Microbiol. 54 mucosal swabs by multiplex RT-PCR/PCR, (1997) 85–93. Vet. Microbiol. 81 (2001) 95–108. [32] Vanrompay D., Cox E., Mast J., Goddeeris B., [21] Thiele D., Karo M., Krauss H., Monoclonal Volckaert G., High-level expression of Chlamy- antibody based capture ELISA/ELIFA for dia psittaci major outer membrane protein in the detection of Chlamydia psittaci in veter- COS cells and in skeletal muscles of turkeys, inary clinical specimens, Zentralbl. Bakte- Infect. Immun. 66 (1998) 5494–5500. riol. 277 (1992) 39–48. [33] Welti M., Jaton K., Altwegg M., Sahli R., [22] Thompson J.D., Gibson T.J., Plewniak F., Wenger A., Bille J., Development of a mul- Jeanmougin F., Higgins D.G., The tiplex real-time quantitative PCR assay to CLUSTAL_X windows interface: flexible detect Chlamydia pneumoniae, Legionella strategies for multiple sequence alignment pneumophila and Mycoplasma pneumoniae aided by quality analysis tools, Nucleic Acids in respiratory tract secretions, Diagn. Micro- Res. 25 (1997) 4876–4882. biol. Infect. Dis. 45 (2003) 85–95. [23] Tondella M.L., Talkington D.F., Holloway [34] Wilson P.A., Phipps J., Samuel D., Saunders B.P., Dowell S.F., Cowley K., Soriano- N.A., Development of a simplified polymer- Gabarro M., Elkind M.S., Fields B.S., Devel- ase chain reaction-enzyme immunoassay for opment and evaluation of real-time PCR- the detection of Chlamydia pneumoniae, J. based fluorescence assays for detection of Appl. Bacteriol. 80 (1996) 431–438.