Chlamydial species among wild birds and livestock in the foothills of Mt. Afadjato, Ghana.

The members of family Chlamydiaceae have a broad host range and cause many kinds of diseases in humans and animals. Several cases of Chlamydiaceae being detected in atypical hosts have been reported recently. Consequently, cross-species monitoring of Chlamydia in wildlife and livestock is pertinent for public health, animal hygiene and wildlife conservation. In this study, we conducted molecular surveillance of Chlamydia in wild birds and livestock around a small village in the foothills of Mt. Afadjato, Ghana where direct contact between wildlife and livestock occurs. Among 29 captured wild birds and 63 livestock, 5 sheep, 30 goats and 28 chickens, the positive ratios of Chlamydia were 24.1%, 40.0%, 43.3% and 26.9%, respectively. Chlamydia pecorum was detected in wild birds, goats, sheep and chickens. On the basis of the variable domain 2 region of ompA, several samples from different hosts showed identical sequences and were phylogenetically located to the same clusters. In addition, using ompA, C. psittaci, C. abortus and C. gallinacea were also detected in this small habitat. Further genetic and pathogenic analyses of the chlamydial distribution in this area, which represents the interface of wild and domestic animal interactions, may improve our knowledge of their transmission among different hosts.

Chlamydia species are facultative intracellular bacteria widely distributed in various host species, ranging from amphibians to humans. The family Chlamydiaceae currently consists of 11 species, including two recently admitted members, Chlamydia avium and C. gallinacea [26, 27]. From the feral sacred ibis (Threskiornis aethiopicus) a new chlamydial species, named as C. ibidis, has also been isolated [30]. Several species belonging to Chlamydiaceae are recognized as important pathogens in the medical and veterinary fields. C. trachomatis, which is highly adapted to humans, is a serious causative agent of blindness and sexually transmitted diseases worldwide. C. pneumoniae causes community-acquired pneumonia in humans, and it is sometimes detected in other mammals, such as koala. C. psittaci usually infects birds, but occasionally causes psittacosis (ornithosis) in humans. C. abortus is a major cause of abortion in ruminants. C. pecorum may cause severe diseases, such as encephalitis, polyarthritis and enteritis, in ruminants and pigs, but usually results in asymptomatic infections. In addition, C. suis (which mainly infects pigs) and C. felis (which mainly infects cats) are major pathogens in the veterinary field. The recent addition of three new members of Chlamydiaceae in birds, C. avium, C. gallinacea [26, 27] and C. ibidis [30], is indicative of the possibly large diversity of Chlamydia spp. associated with avian chlamydiosis [25].

Generally, Chlamydia has broad host range, but each Chlamydia species tends to be harbored by typical hosts. For example, avian orders Psittaciformes and Columbiformes are the most typical and prominent hosts for C. psittaci [9]. Recently, the detection of Chlamydia spp. in atypical or non-typable hosts has been reported [13, 15, 24, 29, 32]. The harboring of Chlamydia by atypical hosts may aid in their maintenance and circulation in nature. Most Chlamydia species infecting livestock and wild animals can be transmitted to humans as zoonotic agents and vice versa [2]. The prevalence of C. psittaci in birds correlates with the psittacosis prevalence in humans [12], and the psittacosis in humans has rapidly increased recently in Belgium [23]. Thus, the cross-species monitoring of Chlamydia in farm and wild animals is important for public health, animal hygiene and wildlife conservation.

The Republic of Ghana is located along the Gulf of Guinea in the West African sub-region. Ghana is endowed with ecological diversity, ranging from the Sudan Savannah in the north-east to rainforests along the coast, that supports a variety of fauna. Mount Afadjato, at 890 m above sea level, is the highest mountain in Ghana. It lies along the Ghana–Togo border, within the Mount Afadjato and Agumatsa Range Conservation Area (Fig. 1). This area has been designated as an important bird area in Africa [18, 20]. In the foothills of Mt. Afadjato, there are several small villages. In these villages, livestock, like sheep, goats and chickens, are reared under an extensive management system by each family.

Fig. 1.

Map showing the location of the study area, based on shapefiles from GADM database of Global Administrative Areas (http://gadm.org) and from DIVA-GIS (http://www.diva-gis.org). Ghana is located along the Gulf of Guinea. The Afadjato conservation area lies along the Togolese border.

In this study, we assessed the prevalence of Chlamydia among wild birds and livestock in a small village in the foothills of Mt. Afadjato, Ghana.

MATERIALS AND METHODS

Specimens

All the specimens from wild birds and livestock were collected in the Gbledi Village situated in the foothill of Mt. Afadjato (7°2’6”N, 0°33’42”E) in March 2014 (Fig. 1). Wild birds (n=29) were captured using a mist net set approximately 200 m from the village with a license from the Ghana Ringing Scheme. Samples were also taken from domesticated animals reared by villagers, including 5 sheep (Ovis aries; the Forest-type breed), 30 goats (Capra hircus; the West-African Dwarf breed) and 28 chickens (Gallus gallus; the Forest Ecotype). The collection of specimens from wild birds and livestock was approved by the Gifu University Animal Experiment Committee (Approval number: 17070). Cloacal or rectal swabs from wild birds or livestock were collected into sucrose–phosphate–glutamate transportation medium for Chlamydia [28]. The specimens were kept at 4°C for approximately 2 days during transportation to the laboratory at the University of Ghana. DNAs from the specimens were extracted using a PureLink Genomic DNA kit (Thermo Fisher Scientific, Waltham, MA, USA) following the manufacturer’s instructions.

Screening for chlamydial DNA in specimens

All the specimens were screened for chlamydial DNA using TaqMan and SYBR Green real-time PCR. The TaqMan real-time PCR amplified 23S rRNA as described elsewhere [5] using the following primers: TQF (5′-GAAAAGAACCCTTGTTAAGGGAG-3′) and TQR (5′-CTTAACTCCCTGGCTCATCATG-3′), with a probe (FAM-CAAAAGGCACGCCGTCAAC-TAMRA). The reaction was performed with the Premix Ex-Taq (TaKaRa Bio, Kusatsu, Japan) on a Thermal Cycler Dice TP800 (TaKaRa Bio) in accordance with the manufacturer’s instructions. The SYBR Green real-time PCR targeted envB as described previously [19] using the following primers: Env-F (5′-AACCTCGGATAGCAAATTAATCTGG-3′) and Env-R (5′-ATTTGGTATAAGAGCGAAGTTCTGG-3′). The reaction was performed with the SYBR Premix Ex-Taq (TaKaRa Bio) on a Thermal Cycler TP800 (TaKaRa Bio) in accordance with the manufacturer’s instructions. The TaqMan real-time PCR detected a broad range of Chlamydiaceae, including C. psittaci, C. abortus, C. felis, C. caviae, C. pneumoniae, C. pecorum, C. trachomatis, C. suis and C. muridarum [5], whereas the SYBR Green real-time PCR detected C. psittaci, C. abortus, C. felis, C. caviae and C. pecorum with a high sensitivity and a detection limit of 10 inclusion-forming units (IFUs) [19].

Phylogenetic analyses of detected species of family Chlamydiaceae

Among positive samples for chlamydial DNA, the ompA variable domain (VD) 2 region that encompasses portions of outer membrane protein A, 16S rRNA and enoA, encoding enolase A, were sequenced for phylogenetic analyses. We first attempted to sequence the ompA gene because it has been used to differentiate chlamydial strains and species [11]. However, it was difficult to produce the full-length ompA sequence data for each positive sample; consequently, we attempted to amplify and analyze the VD2 region of the ompA gene, which contains chlamydial species/strain-specific nucleotide motifs, instead of the full-length ompA. The partial VDs of the ompA gene, including the VD2 region, were also analyzed to identify chlamydial species and strains [4, 8, 10].

The whole-genome amplification of positive samples was conducted using an illustra GenomiPhi DNA Amplification kit (GE healthcare, Chicago, IL, USA) to enhance the PCR efficiency. The ompA VD2 region was amplified using nested PCR with an outer pair of primers (CMGP-1F: 5′-CCTTGTGATCCTTGCGCTACTTG-3′/CMGP-1R: 5′-GTGAGCAGCTCTTCGTTGAT-3′) and an inner pair of primers (CMGP-2F: 5′-GCCTTAAACATCTGGGATCG-3′/CMGP-2R: 5′-GCACAACCACATTCCCATAAAG-3′) and previously described conditions [3] on a PCR Thermal Cycler Dice (TaKaRa Bio). The nested PCR reaction was performed with BIOTAQ HS DNA Polymerase (Bioline Reagents, London, UK) in Ampdirect Plus buffer (Shimadzu Corporation, Kyoto, Japan) instead of Taq polymerase as reported previously [3, 4] because of better recorded amplification results. The 16S rRNA was amplified with the pan-Chlamydiales primers panCh 16F2: 5′-CCGCCAACACTGGGACT-3′ and panCh16R2: 5′-GGAGTTAGCCGGTGCTTCTTTAC-3′ that could amplify a broad range of Chlamydiales, including parachlamydia, with high sensitivity and a detection limit of 5 copies [1, 16]. The enoA gene was amplified with the primers YPenoA5: 5′-CCWATGATGAAYCTYATTAACGG-3′ and YPenoA6: 5′-TCTTCYTCAGGRAGGCCATCT-3′, (W=A or T; Y=C or T), following a method described elsewhere [22]. The 57-, 169- and 381-nt sequences were determined for ompA, 16S rRNA and enoA, respectively. Amino acid sequences deduced from the nucleotide sequences were aligned using ClustalW (https://www.genome.jp/tools-bin/clustalw) and phylogenetically analyzed using Mega 7 [14]. The independent dendrograms of ompA VD2 and 16S rRNA were constructed using the Maximum-likelihood method based on the Le_Gascuel_2008 model. A dendrogram of the enoA gene was constructed by the Maximum-likelihood method based on the Kimura 2-parameter model.

Nucleotide sequence accession numbers

Nucleotide sequences obtained in this study have been deposited at the DNA Data Bank of Japan (DDBJ) under accession numbers LC271681 to LC271692 for the VD2 region of ompA, LC484863 to LC484866 for partial 16S rRNA and LC271693 for the partial enoA gene.

RESULTS

The wild birds (n=29) captured and screened in the present study belonged to two orders (Passeriformes and Columbiformes) and comprised 10 species (Table 1). Livestock samples (n=63) were from sheep (the Forest-type breed), goats (the West-African Dwarf breed) and chicken (the Forest Ecotype). Using real-time PCR, three out of nine species of Passeriformes tested positive for chlamydial DNA, namely Phyllastrephus icterinus (33.3%), Cyanomitra olivacea (33.3%) and Alethe castanea (100%). In all, 24.1% of wild birds were positive for Chlamydia spp. (Table 1). For livestock, DNA samples from 43.3% of goats, 40.0% of sheep and 26.9% of chicken were positive for Chlamydia spp. (Table 2).

Table 1.

Prevalence of Chlamydia spp. in the wild birds captured at the foothill of Mt. Afadjato, Ghana (Nomenclature follows Borrow and Demey, 2010)

Order	Species		Number of specimens	Number of positive specimens	Percentage of positive specimens
	Latin name	English name
Passeriformes	Stiphrornis erythrothorax	Forest Robin	2	0	0
	Terpsiphone rufiventer	Red-bellied Paradise Flycatcher	1	0	0
	Bleda canicapillus	Grey-headed Bristlebill	5	0	0
	Bleda eximius	Green-tailed Bristlebill	2	0	0
	Eurillas latirostris	Yellow-whiskered Greenbul	2	0	0
	Phyllastrephus icterinus	Icterine Greenbul	6	2	33.3
	Cyanomitra olivacea	Olive sunbird	6	2	33.3
	Alethe diademata	White-tailed Alethe	3	3	100
	Hylia prasina	Green Hylia	1	0	0
Columbiformes	Tutur tympanistria	Tambourine Dove	1	0	0
Total			29	7	24.1

Table 2.

Prevalence of Chlamydia spp. in livestock reared in a village at the foothill of Mt. Afadjato, Ghana

Livestock (breed)	Number of specimens	Number of positive specimens	Percentage of positive samples
Goat (- Forest type)	30	13	43.3
Sheep (- West-African Dwarf)	5	2	40.0
Chicken (- Forest Ecotype)	28	7	26.9
Total	63	22	34.9

We succeeded in sequencing the VD2 region of the ompA gene from 12 of 29 positive samples (7 wild bird and 22 livestock). Among the 12 samples, 9 were located on clusters near C. pecorum, which strongly indicates that these samples belonged to C. pecorum. Additionally, C. pecorum positive samples were obtained from two wild birds (P. icterinus), three goats, two sheep and two chickens (Fig. 2). C. pecorum strains can be classified into two major clusters based on the ompA genotype [17]. In the present study, at least three clusters of C. pecorum were identified. The first cluster consisted of samples 14–69, 14–60 and 14–05 from wild bird, goat and chicken, respectively. The second cluster contained samples 14–70 (wild bird), 14–34 (goat) and 14–35 (goat), whereas the third cluster contained samples 14–14 (chicken), 14–59 (sheep) and 14–64 (sheep). The partial VD2 ompA sequences were identical in all the samples in the first and second clusters, and identical for 14–14 and 14–64 in the third cluster. Representative samples of each cluster, 14–69, 14–34 and 14–64, respectively, were subjected to a phylogenetic analysis of their 16S rRNAs and classified within the cluster of C. pecorum, which confirmed that these samples belonged to C. pecorum (Fig. 3).

Fig. 2.

Phylogenetic analysis of the ompA variable domain 2 (VD2) region using the by Maximum-likelihood method. Some of the Chlamydiaceae positive samples in this study from which the ompA VD2 region (dotted squares) was sequenced successfully were compared with those of other reported Chlamydiaceae strains. The tree with the greatest log likelihood is shown. Branch lengths are measured in amino acid substitutions. Numbers indicate percentages of the bootstrap replicates (n=1,000). The scale bar represents the percentage of sequence diversity. The sources of the collected samples are shown, and the ID number of each nucleotide or protein in GenBank is provided.

Fig. 3.

Phylogenetic analysis of the 16S rRNA gene using the Maximum-likelihood method. Some representative Chlamydia pecorum and C. gallinacea positive samples from which the ompA variable domain 2 region was sequenced successfully were subjected to 16S rRNA (dotted squares) sequencing and compared with those of other reported Chlamydiaceae strains. The tree with the greatest log likelihood is shown. Branch lengths are measured in nucleotide substitutions. Numbers indicate percentages of the bootstrap replicates (n=1,000). The scale bar represents the percentage of sequence diversity. The ID number of each nucleotide or protein in GenBank is provided.

We also detected strains as C. abortus (sample 14–49) and as C. psittaci (sample 14–33) in goats (Fig. 2). C. psittaci can be divided into 10 clusters (I–X) of the VD2 variant, which is correlated with the C. psittaci genotype (equivalent to C. psittaci serotype determined with anti-major outer membrane protein antibodies) as described previously [3]. The sample 14–33 was located on the same cluster as C. psittaci 6BC. C. psittaci 6BC belongs to genotype A, which is the most prevalent genotype, and is usually harbored by psittacine birds [7]. C. psittaci 6BC also belongs to cluster I as determined by analyzing the VD2 region [3], suggesting that the 14–33 sample from goat belongs to genotype A.

In addition to C. pecorum, C. abortus and C. psittaci, we detected another Chlamydia sp. in this area from a chicken sample, 14–06, and identified it as C. gallinacean on the basis of the enoA gene (Fig. 4), as well as ompA VD2 (Fig. 2) and 16S rRNA (Fig. 3) analyses.

Fig. 4.

Phylogenetic analysis of the enoA gene using the Maximum-likelihood method. The enoA gene of sample 14-06 (square) from a chicken that tested positive for Chlamydiaceae was compared with those of other reported Chlamydiaceae strains. The tree with the greatest log likelihood is shown. Branch lengths are measured in nucleotide substitutions. Numbers indicate percentages of the bootstrap replicates (n=1,000). The scale bar represents the percentage of sequence diversity. The ID number of each nucleotide or protein in GenBank is provided.

DISCUSSION

Generally, the members of family Chlamydiaceae have a broad host range and cause many kinds of diseases in humans and other animals. In this study, we conducted molecular surveillance of Chlamydia among livestock and wild birds around a small village, where direct contact between wildlife and livestock occurs, in the foothills of Mt. Afadjato, Ghana. We detected a variety of chlamydial species in different hosts in this area. Our major results were as follows: (i) at least, four kinds of Chlamydia (C. pecorum, C. psittaci, C. abortus and C. gallinacea) were harbored asymptomatically in livestock and wild birds near the village and (ii) according to the phylogenetic analysis, C. pecorum detected in wild bird, goat and chicken located to the same cluster.

In this study, we were unable to obtain full-length ompA sequences and other sequences, such as 16S/23S rRNA, owing to limitations of the target DNA available and perhaps the presence of PCR inhibitors [31] in field fecal samples. This inherent limitation in molecular epidemiological studies prevented us from carrying out further genetic analyses, such as multi-locus sequence typing, which is an appropriate way to analyze the genetic diversity of Chlamydia [21]. For the purpose of achieving a high sensitivity of Chlamydia detection by PCR for nucleic acid sequencing, we amplified total DNA using several DNA amplification kits, as well as PCR enzymes and buffers, that suppressive inhibitors in biological samples. Finally, we succeeded in obtaining the partial VD2 ompA sequences in 12 samples and constructed a phylogenetic tree. In this phylogenetic tree, three clusters of C. pecorum were identified in this narrow sampling area (Fig. 2). Each cluster included at least three samples from the different animal species sampled, goat, sheep, chicken and wild birds. C. pecorum is common in ruminants but has been rarely identified in wild birds [15, 17]. Although previous studies reported the presence of C. pecorum in birds, most reports are from captive birds [6, 24]. Therefore, to confirm the presence of C. pecorum in the wild birds of the Afadjato area, we attempted to analyze the 16S rRNA region using pan-Chlamydiales PCR that could detect a broad range of Chlamydiales, including parachlamydia, with a high sensitivity and a detection limit of 5 copies as described previously [1, 16]. In this pan-Chlamydiales PCR, we used locked nucleic acids in the PCR primers to improve PCR performance [1, 16]. With this improved PCR method, we successfully constructed a phylogenetic tree of partial 16S rRNA sequences. The 14–69, 14–64 and 14–34 samples were within the C. pecorum cluster in the resulting phylogenetic tree, which corroborated the results of the tree constructed using partial VD2 ompA sequences.

Furthermore, C. pecorum samples belonging to the same clusters as determined by the ompA VD2 region analysis were not necessarily from the same hosts. For example, the cluster containing C. pecorum E58 included the 14–69 sample from wild bird, the 14–60 sample from goat and the 14–05 sample from chicken. In each cluster, the partial VD2 ompA sequences from different host were identical for 14–69 (wild bird), 14–60 (goat) and 14–05 (chicken); identical for 14–34 (goat), 14–70 (wild bird) and 14–35 (goat); and identical for 14–14 (chicken), 14–64 (sheep) and 14–59 (sheep). This revealed the existence of C. pecorum among wild birds, goats and chickens in this small area.

The detection of Chlamydia in atypical hosts has been reported. In these reports, samples were collected from different countries over a long period [32] or from different provinces/regions for several months in each country [15, 29]. In contrast, we collected samples in a small village (consisting of approximately 10 families), where direct contact between farm and wild animals occurs, over 2 days. All the wild birds examined in this study were native birds. Because they were captured near the village, contact between wild birds and livestock could be frequently occurring. In fact, livestock reared in the village was always free range and grazed and fed in the foothills, which ensured that they were in contact with wild birds. In addition, all the Chlamydia were detected in healthy wild birds and livestock. Therefore, Chlamydiaceae bacteria may infect asymptomatically, resulting in its circulation and maintenance among the wild birds and livestock in this area. Here, the existence of C. pecorum among different hosts, such as wild birds, livestock and chickens, in the small area might imply the circulation of Chlamydia. Certainly, multi-locus sequence typing or whole-genome sequencing analyses of detected strains and a fixed-point sampling are necessary for discussing the circulation of pathogens in a specific environment, like those conducted in urban pigeons in one German town [24].

This is the first report that examined the prevalence and diversity of Chlamydiaceae in several kinds of mammals and birds in the Afadjato and Agumatsa area of Ghana, West Africa. It is imperative to further investigate the genetic diversity of harbored by wild birds and livestock in this area to explore their evolutionary relationships, host specificity and cross-species transmission mechanisms. These results will help to evaluate the impact of Chlamydiaceae on public health, livestock hygiene and wildlife conservation.

POTENTIAL CONFLICTS OF INTEREST

No author has a potential conflict of interest relevant to this article.