A Real-Time PCR Assay for the Detection of Atypical Strains of Chlamydiaceae from Pigeons

Recent evidence of the occurrence of atypical Chlamydiaceae strains in pigeons, different from the established Chlamydiaceae, requires the development of a specific and rapid detection tool to investigate their prevalence and significance. Here is described a new real-time PCR assay that allows specific detection of atypical Chlamydiaceae from pigeons. The assay has been used to assess the dissemination of these strains in field samples collected from Parisian pigeon populations in 2009. The results suggest a limited dissemination compared to the usually higher prevalence of Chlamydia psittaci that is the main species associated with avian chlamydiosis.


Introduction
Avian chlamydiosis is a zoonotic disease caused by Chlamydia psittaci, an obligatory intracellular bacterium and member of the family Chlamydiaceae [1]. The infection is usually systemic and occasionally fatal in birds. The clinical signs vary greatly in severity and depend on the species and age of the birds as well as the causative strain [2]. Transmission of C. psittaci from birds to humans is reported regularly, particularly in high-risk individuals such as breeders or slaughterhouse workers [3]. Although the importance of C. psittaci as the causative agent of avian chlamydiosis in birds has been known for decades [4], several studies have recently provided evidence of the occurrence of other chlamydial species in birds such as C. abortus [5,6], C. suis and C. muridarum [7] as well as C. pecorum and C. trachomatis [8]. Additionally, atypical strains of Chlamydiaceae were recently detected in chickens [9,10,11] and pigeons [8,12,13,14].
A better characterisation of these atypical Chlamydiaceae and the availability of fast, sensitive and specific detection tools are needed in order to better understand their epidemiological importance. A specific 16S rDNA-based real-time PCR was developed recently for the specific detection of atypical Chlamydiaceae from chickens (hereafter named ACC) [14].
The work reported in this paper deals with the development and validation for routine diagnostic purposes of a real-time PCR based on the enolase A (enoA) gene target for the specific detection of atypical Chlamydiaceae from pigeons (hereafter named ACP). This assay has been used to investigate the dissemination of this agent in Parisian pigeon populations.

Microorganisms
Chlamydial strains and isolates used for determination of specificity, sensitivity and detection limit of the new ACP-specific real-time PCR assay are listed in Tables 1 and 2. Chlamydiaceae were propagated in the yolk sac of 7 day-old embryonated chicken eggs and stored at 280uC. Non-chlamydial bacterial and fungal strains obtained from the University of Tours (France) and the Anses strain collection (Maisons-Alfort, France) were used for specificity testing.

Preparation of genomic DNA
Chlamydial strains cultivated onto vitellus membranes as well as biological samples were subjected to DNA extraction using the QIAamp DNA Mini Kit (Qiagen, Courteboeuf, France). DNA extracts were stored at 220uC before analysis.
DNA of non-chlamydial samples was extracted using the Instagen Matrix DNA Kit (BioRad, Marnes-la-Coquette, France) according to the manufacturer's instructions. DNA concentration and purity of non-chlamydial extracts were spectrophotometrically assessed by reading A260 and A280 and confirmed by visualization on 0.5% (w/v) agarose gels. Finally, DNAs with a mean concentration of 400 mg/ml were stored at 220uC until required for analysis.

Sequencing of the enoA gene locus
YP_enoA5 and YP_enoA6 used for amplification of an approximately 400-bp segment of the enoA gene were previously published by Pannekoek and colleagues [15]. PCR reactions were performed in a final volume of 25 ml. The mix contained 2 ml of DNA template, 2.5 ml of 106 PCR reaction buffer, 1.25 U of HotStarTaq DNA Polymerase (Qiagen, France), 100 mM of each deoxynucleotide triphosphate and 0.5 mM of each primer. The conditions used for amplification were 95uC for 1 min, 45 cycles of 94uC for 30 s, 48uC for 1 min, 72uC for 1 min and 30 s, 72uC for 10 min (final extension). PCR-amplified segments of enoA were sequenced by Eurofins MWG Operon (Nantes, France). Sequence alignments were conducted using the Bionumerics software package version 4.6 (Applied-Maths, Belgium). A dendrogram including one representative of each enoA sequence type for established Chlamydiaceae strains [15,16], one representative for ACC [14] and one for ACP was constructed using the Unweighted Pair Group Method with Arithmetic Mean algorithm. Nucleotide sequences have been deposited under GenBank accession numbers HE983370 to HE983383.

ACP-specific real-time PCR
Specific primers ACP_Fw (5'-CATGCAAGCTATTGA-GAAAAGTGGT -3') and ACP_Rv (59-CCTTGATATG-TACGTGTTTTCTCG -39) as well as a specific probe ACP_Pr (59-FAM-CACCCCTGGTGAAGATATTTCCTTAGCAT -TAMRA 39) were designed based on enoA and synthesised by Eurofins MWG Operon. Amplification was performed using the 26 TaqMan Universal PCR Master Mix (Applied Biosystems, Courtaboeuf, France). The final volume of the reaction mixture amounted to 20 ml including 10 ml of master mix, 2 ml of DNA sample, 0.6 mM of each primer, 0.1 mM of the probe and sterile PCR water added to reach the final volume. Amplification was carried out at an ABI Prism 7000 thermocycler (Applied Biosystems) using the following cycling parameters: heating at 95 uC for 10 min, 50 cycles of 95uC for 15 s and 60uC for 1 min. For all PCR reactions, DNA of the isolate 10-743/SC13 arbitrarily defined as reference strain for ACP was used as a positive control while sterile water was used as negative control.

Performance evaluation of the ACP-specific real-time PCR
The detection limit of the real-time PCR was determined in triplicate (independent runs) using decimal serial dilutions of purified genomic DNA from 10-743/SC13 suspensions equivalent to 3610 21 to 3610 5 inclusion-forming units (IFUs) per PCR reaction. The concentration of this DNA was previously determined by comparison to a DNA purified from cell culture containing defined numbers of IFUs of C. psittaci. The standard curve and the number of IFUs were automatically determined by the software using manually entered IFU standard concentrations. The standard curve was generated by plotting Ct against the logarithm of IFU per PCR reaction mix. The Ct, at which the fluorescence crosses the threshold value, and the lower limit of detection of the assay were calculated automatically.
The specificity of the new real-time PCR was evaluated on 22 Chlamydiaceae DNA samples ( Table 2) yielding Ct values between 22 and 30 when tested with the 23S rRNA-based Chlamydiaceaespecific real-time PCR (Chlamydiaceae 23S-rtPCR) [17], and 22 non-chlamydial microorganisms.
DNA of strain 10-743/SC13 was amplified 30 times in the same run to determine intra-assay repeatability, and 11 times in separate runs on different days for inter-assay reproducibility. The respective CVs were then calculated.

Chlamydiaceae field samples
A panel of 125 DNAs extracted from cloacal swab samples collected from pigeons in a previous study [13] was used to evaluate the dissemination of the ACP. These samples, which contained an internal positive control (Exogenous internal positive control, TaqMan, Life technologies), were previously tested positive in the Chlamydiaceae 23S-rtPCR [17].

Results and Discussion
Recent evidence of the occurrence of atypical Chlamydiaceae strains in chickens and pigeons highlighted the necessity of specific and fast detection tools for the investigation of their epidemiological and clinical importance. Thirteen ACP strains have been isolated up to now in France, Germany and Italy ( Table 1). A similar approach to the one used for ACC [14] has been adopted in this study, except that the enolase A (enoA) gene has been selected as target. This gene locus had been previously included in a Multi-Locus Sequence Typing scheme for Chlamydia spp. [15,16]. All enoA sequences of ACP strains in Table 1 proved identical. The consensus sequence was compared in silico to those of the nine established Chlamydiaceae species [15,16] and also to ACC sequences [14]. Alignment ( Figure S1) and phylogenetic analysis ( Figure 1) revealed that this sequence was specific to ACP and could be a convenient target for their unambiguous identification by real-time PCR.
In order to determine the detection limit of the new real-time PCR, a decimal dilution series of the titrated 10-743/SC13 DNA was tested. The results showed that the assay was able to detect the DNA from 3 IFUs of the respective pathogen per PCR reaction with an efficiency of 93% ( Figure 2). The intra-assay and interassay CVs were 2.5% and 1.9% respectively. Therefore, the realtime PCR assay presented here has proved to be sensitive, specific and reproducible and offers the possibility of assessing the epidemiological importance of ACP infections in pigeon populations.
To evaluate the suitability of the assay for this purpose, a retrospective examination of previously collected field samples in Parisian feral pigeon populations [13] has been carried out. Out of the 125 Chlamydiaceae-positive samples, 10 were positive (8%) with the new ACP-specific real-time PCR (Table 3), including samples 09-489/LP23 and 09-589/S46 that had previously been identified as ACP by partial ompA and 16S rRNA sequencing [13]. These findings suggest a low dissemination of ACP in comparison with C. psittaci, which prevalence of 68% had been determined ( Table 3) using the incA-based C. psittaci-specific real-time PCR (incA-rtPCR C.psittaci) according to Menard and colleagues [18]. A mixed infection involving an ACP strain and C. psittaci was also found in two samples from Genevilliers ( Table 3).
Given the low number of samples tested and the limited geographical range covered, the results of the present study can only be regarded as preliminary. Atypical pigeon strains have been isolated in the laboratories of the authors from the intestinal content or organs of necropsied urban pigeons. Their pathogenicity for pigeons is unknown. Another open issue is the host range. A recently isolated atypical parrot strain from Germany (designated 10DC88) was found to possess an enoA sequence identical to those of ACP, so that it was positive with the ACPspecific real-time PCR assay. This indicates that ACP seem not to be restricted to pigeons and could possibly be circulating in other avian hosts.
All in all, as a rapid and specific diagnostic tool is now available, it seems straightforward to include ACP investigation to analyses of diagnostic samples from birds and to undertake systematic studies to characterize the epidemiology and etiology of ACP infections in avian populations.

Conclusions
A sensitive and specific real-time PCR assay for the detection of atypical Chlamydiaceae from pigeons has been developed and preliminary validated. The use of this new tool could contribute to a better understanding of the importance of these bacterial agents.  [15,16]

Supporting Information
Figure S1 Alignment of partial enoA sequence for ACC (08-1274/3 reference), ACP (10-743/SC13 reference) and various established Chlamydiaceae strains (one representative of each enoA sequence type; [15,16]). Nucleotide homologies are represented by dots. The numbers refer to alignment positions. ACP-specific primer sequences are boxed and the probe sequence is underlined. (TIF)

Acknowledgments
We are grateful to S. Thierry, D. Albert and G. Le Carrou (Anses Maisons-Alfort, France) for providing non-chlamydial materials and for their precious advice.