Study on the prevalence and genetic diversity of Eimeria species from broilers and free-range chickens in KwaZulu-Natal province, South Africa

This study was conducted from January to October 2018 with the objective to determine the prevalence and genetic diversity of Eimeria species in broiler and free-range chickens in KwaZulu-Natal province, South Africa. A total of 342 faecal samples were collected from 12 randomly selected healthy broiler chicken farms and 40 free-range chickens from 10 different locations. Faecal samples were screened for the presence of Eimeria oocysts using a standard flotation method. The species of Eimeria isolates were confirmed by amplification of the internal transcribed spacer 1 (ITS-1) partial region and sequences analysis. Among broiler and free-ranging chickens, 19 out of 41 pens (46.3%) and 25 out of 42 faecal samples (59.5%) were positive for Eimeria infection. Molecular detection revealed the following species: Eimeria maxima, Eimeria tenella, Eimeria acervulina, Eimeria brunetti and Eimeria mitis in all the samples screened. Similarly, polymerase chain reaction assays specific for three cryptic Eimeria operational taxonomic units were negative for all the samples. Phylogenetic analysis of the ITS-1 sequences supported species identity with the greatest variation detected for E. mitis. This study provides information on the range and identity of Eimeria species, and their genetic relatedness, circulating in commercially reared broilers and free-ranging chickens from different locations in KwaZulu-Natal province.

http://www.ojvr.org Open Access Effective control of coccidiosis in chickens relies on strict management practices, supplemented by timely application of anticoccidial drugs and/or vaccines (Godwin & Morgan 2015) underpinned by proper diagnosis and identification. Traditional diagnostic methods include evaluation of the location and the characteristics of gross pathology (lesion scoring) and microscopic analysis of oocyst morphology ( Kumar et al. 2014). However, the relative complexity and requirement of expertise for these methods necessitated the development of molecular alternatives, including genus-and species-specific polymerase chain reaction (PCR) assays (Lew et al. 2003). The use of nuclear and mitochondrial genetic markers (e.g. internal transcribed spacer [ITS] sequences, 18S ribosomal RNA, cytochrome oxidase subunit I [COI]) has proven effective in the identification and taxonomic classification of protozoan parasites, including Eimeria (Kumar et al. 2015a;Ogedengbe et al. 2018;Tan et al. 2017).
Thus, ITS-1 sequences have served as genetic markers to identify Eimeria species (Cook et al. 2010;Oliveira et al. 2011). Based on the observed diversity, ITS-based speciesspecific primers have been developed for use in the identification of Eimeria species (Lew et al. 2003). However, studies from various countries have reported nucleotide variations in the ITS-1 region within Eimeria species isolates (Bhaskaran et al. 2010;Kumar et al. 2015a;Lew et al. 2003). Genetic diversity among species and strains of Eimeria could pose a major risk to the control of coccidiosis in the future. As such, knowledge defining naturally occurring genetic diversity becomes imperative to understand the pathogenicity and epidemiology of Eimeria that infect chickens (Morris & Gasser 2006).
There is a dearth of information on Eimeria occurrence and diversity in South Africa. As such, reports on circulating Eimeria species in KwaZulu-Natal province together with information on their occurrence in commercial chickens are not available. This study, therefore, aimed to determine prevalence and genetic diversity of Eimeria species in both broiler and free-range chickens in KwaZulu-Natal province.

Study area
KwaZulu-Natal is the second most populous province among the nine provinces in South Africa. It has a population of approximately 10 million people and land size of 94 000 km 2 located between latitude 28°99'S and longitude 30°97'E. The capital city of Pietermaritzburg has a warm and subtropical climate throughout the year, especially around the coastline, but gets colder in the inland areas. The poultry industry in KwaZulu-Natal province is one of the producers of broiler birds in South Africa with a total of 6.7 million broiler birds in 2017, contributing 6.4% to the national broiler production (South African Poultry Association 2017).

Sample collection
A total of 342 chicken faecal samples were collected from 12 broiler farms consisting of 41 pens (1-5 pens per farm) and free-range chickens. The age of broiler chickens at the time of sampling ranged from 3 to 10 weeks, with the exception of a single farm consisting of 12-week-old chickens. In addition, 42 faecal samples of 40 free-ranging 3-week-old village chickens were randomly collected from four localities. The 342 samples were collected randomly once from the following locations: Pietermaritzburg, Phoenix, Scottburg, Stanger, Chatsworth, Westville, Maphumulo, Umvoti, Port Sherpstone and Shongweni of KwaZulu-Natal province from January to October 2018. Detailed information on the number of pens per farm, number of samples per pen, number of farms per location and number of chickens per location is shown in Appendix 1 Tables 1-A1 and 2-A1. There were no clinical signs of coccidiosis among the chickens on any of the farms sampled. Samples were collected following the procedure described by Kumar et al. (2014). Briefly, in the broiler farms, faecal samples were collected following a pre-determined 'W' pathway in each pen to allow random sampling. Fifty-millilitre conical tubes containing 10 mL of 2% potassium dichromate were used to collect faeces up to 20 mL of the tube and stored at 4 °C until further use. Depending on the size of the pen, four to eight 50-mL conical tubes of faecal samples were collected per pen and the content was mixed together vigorously.

Sample processing and microscopic oocyst identification
Samples were processed based on the procedures described by Kumar et al. (2014), with minor modifications. Two grams of faecal samples were weighed into a beaker and mixed with 100 mL of distilled water. This was stirred with a glass rod and later filtered through a gauze. The filtrate was transferred into a new 50-mL conical tube and filled to the brim with saturated salt solution. This was then centrifuged at 800 × g for 10 minutes. The supernatant was decanted and the sediment was transferred into a new 50-mL tube and then later pelleted at 14 000 × g for 3 min. Oocysts per gram (OPG) were counted using a McMaster counting chamber following a standard protocol (Haug, Williams & Larsen 2006). Samples with OPG greater or equal to 250 OPG were selected for deoxyribonucleic acid (DNA) extraction. Photomicrograph images of unsporulated oocysts were taken randomly from each farm sampled using an OMAX compound microscope containing a 5 MP camera at 400×.

Polymerase chain reaction amplification
A nested PCR protocol targeting the genomic ITS-1 region was used to detect each Eimeria species. Genus-and speciesspecific primers were used as described by Lew et al. (2003). Each 25 µL PCR contained 12.5 µL 2X DreamTaq Green PCR Master Mix (Thermo Scientific, US), 1 µL of each forward and reverse primer (10 µM of stock solution; . Similarly, the samples were also screened for the presence of three cryptic Eimeria OTUs by targeting the ITS-2 genomic region using the primers and thermal cycling procedure described by Fornace et al. (2013), as shown in Table 2. The PCR products were sent for sequencing at Inqaba Biotech (South Africa). Sequencing was done with both forward and reverse primers using Big Dye chemistries in an ABI 3500XL Genetic Analyzer, POP-7 TM (Thermo Scientific, US).

Sequence analysis
A total of 28 ITS-1 sequences were viewed, edited and trimmed. Consensus sequences were generated from both forward and reverse sequences using BioEdit version 7.0.5.3 software (Hall 1999). The sequences were submitted to National Center Biotechnology Information and assigned accession numbers (Appendix 1 Table 3-A1). Also, the sequences were compared with selected published sequences from the GenBank. Sequence alignment was performed using the ClustalW programme. Pairwise percentage identity (Appendix 1 Figure 1-A1) was carried using Sequence Demarcation Tool (SDT) version 1.2 software (Muhire, Varsani & Martin 2014). Genetic distance within Eimeria species isolates from this study was calculated with MEGA version 6.0 (Tamura et al. 2013) using the Tamura 3-parameter model.

Phylogenetic analysis of internal transcribed spacer-1 sequences
The genetic diversity that exists between the ITS-1 sequences generated in this study (n = 28) and those of American, Chinese, Indian, Australian, Egypt, Sudan and Swedish  Eimeria species isolates published in GenBank (Appendix 1  Table 4-A1) were analysed. Phylogenetic analyses using the maximum likelihood (ML) method were carried out with MEGA version 6.0 (Tamura et al. 2013). The nucleotide substitution model that best fitted the data set was identified using Model-Test in MEGA6. Based on the Akaike Information Criterion, the Jukes-Cantor model was identified as the best model. Gaps in the alignment were treated as missing characters. Bootstrap iteration was based on 1000 replicates and the percentage value was indicated at each node. Neospora caninum (GenBank accession number: AF038860.1) and Toxoplasma gondii (EU025025.1) were used as out-group species to root the tree.

Statistical analysis
Data generated were analysed using the Statistical Package for the Social Sciences (SPSS) software version 25.0. Descriptive statistics were used to determine the prevalence of detected Eimeria species.

Ethical consideration
The protocol for this study was approved by the University of KwaZulu-Natal Animal Research Ethics Committee and assigned the reference number AREC/058/017D.

Polymerase chain reaction amplification and microscopic unsporulated oocyst detection
Among broiler and free-ranging chickens, 19 out of 41 pens (46.3%) and 25 out of 42 samples (59.5%) were positive for Eimeria infection (Figure 1). The highest level of Eimeria infection was observed in the following locations in both broiler and free-ranging chickens as shown in Figure 2: Phoenix (

DNA amplification of Eimeria species
The most common mixed species combinations detected in broiler and free-ranging chicken faecal samples were E. tenella + E. maxima (

Phylogenetic analysis of internal transcribed spacer-1 sequences
Maximum likelihood with the Jukes-Cantor model was used to create the phylogenetic tree ( Figure 5) of the 28 ITS-1 sequences generated in this study, together with reference India, Sudan and Sweden with a very strong support. All the E. maxima sequences from this study clustered with E. maxima sequences from America and India with low support. Within the E. brunetti clade, all the three sequences from this study clustered with E. brunetti sequences from India and Australia with a very strong support. Genetic distances between ITS-1 sequences of Eimeria isolates in this study and those of a public database were as follow: E. mitis (0.12 ± 0.013), E. acervulina (0.02 ± 0.005), E. maxima (0.07 ± 0.016), E. tenella (0.07 ± 0.010) and E. brunetti (0.01 ± 0.004).

Discussion
Coccidiosis is an enteric disease that poses a threat to efficient poultry production (Ogedengbe, Hanner & Barta 2011), compromising economic productivity and chicken welfare. For effective diagnosis, control and epidemiology of the disease, the identification of specific species of Eimeria is essential. Understanding the occurrence of genetic diversity and regional population structure are important (Hamza, Al-Massodi & Jeddoa 2015;Morris & Gasser 2006).
In this study, Eimeria infection had an overall prevalence of 46.3% (19 out of 41 pens) and 59.5% (25 out of 42 samples) across different farms and locations, which was higher than the 29.4% found among Eimeria parasites from KwaZulu-Natal and Limpopo (Malatji et al. 2016). However, it was lower than previous reports from other regions including Ethiopia (56%; Luu et al. 2013 Molecular diagnosis using nested species-specific ITS-1 primers was used to identify five species of Eimeria (E. tenella, E.maxima, E. acervulina, E. brunetti and E. mitis) circulating in both commercial broiler and free-range chickens in KwaZulu-Natal province. This is similar to the study of Debbou-Iouknane, Benbarek and Ayad (2018), who reported the same five species of Eimeria among broilers farms in Bejaia region of Algeria. The prevalence of one or more species of Eimeria in broiler farms in this study could be influenced by the different anticoccidial used in various farms (Carvalho et al. 2011), although our study did not document anticoccidial use in the farms.
The most prevalent species among broiler farms in this study was E. tenella (68.4%), which is in agreement with other studies that have reported a high prevalence that ranges from 80.67% to 100% in Anhui Province, China, Trinadad and Indonesia (Brown et al. 2018;Hamid et al. 2018;Huang et al. 2017). The high prevalence of E. tenella poses a major concern to the health status of chickens because it is associated with caecal lesions causing haemorrhage, oedema and anaemia (Iacob & Duma 2009). However, E. mitis (96%) had the highest prevalence among free-ranging village chickens in this study. The reason for this is unclear as it is contrary to reports of most studies where E. acervulina and E. tenella are known to be highly prevalent in most farms because of their high reproductive potentials (Williams 2001).
Co-infection with multiple Eimeria species is a common finding in many poultry farms (Aarthi et al. 2010;Haug et al. 2008). We also found multiple infections (57.9% and 100%) to be common in both chicken types, with two or more species among the samples examined. Eimeria tenella + E. maxima (21.1%) and E. mitis + E. maxima + E. acervulina (44%) were the most common co-infections. This is in line with different studies which reported the frequency of E. maxima in most mixed species infection (Kaboudi, Umar & Munir, 2016).
Mixed infections among Eimeria species poses a challenge to the control of coccidiosis in chickens as it can increase pathogenicity of the disease among birds (Jekins et al. 2008). It could also serve as a potential threat to the effectiveness of anticoccidial vaccine, and this has warranted the combination of different Eimeria strains in some species, such as E. maxima, in the design of anticoccidial vaccines.
The efficacy of anticoccidial vaccines is under threat, especially with the recent upsurge of new Eimeria variants (OTUs), which was first detected circulating among commercial birds in Australia (Cantacessi et al. 2008). The presence of these OTUs (OTUx, OTUy and OTUz) has also been reported across much of the Southern Hemisphere (Clark et al. 2016;Fornace et al. 2013;Jatau et al. 2016). In this study, none of the samples was positive for any of the three OTUs. This could be because of the geographical location of our study sample, which is on latitude 28°S. Although a study has reported the distribution of these cryptic species (OTUs) in the northern hemisphere (Jatau et al. 2016), a more elaborate study by Clark et al. (2016) in 20 different countries from five continents has opined that these OTUs are distributed towards the south of the 30°N latitude. The study reported eight different countries to be populated with OTUs with the following distribution: OTUz was found in all the eight countries south of the 30°N latitude and OTUx was detected south of 30°N in six out of the eight countries, whilst OTUx, OTUy and OTUz were only detected in Nigeria among all the African countries at the same geographical location (Clark et al. 2016).
Similarly, ITS-1 sequences belonging to five different Eimeria species were generated in this study. The similarity of the sequences generated in this study when compared with published Eimeria species sequences ranged from 90% to 93% in E. mitis, 99.31% in E. maxima, 99% to 100% in E. tenella, 100% in E. brunetti and 99.38% in E. acervulina. Although the ML tree, as shown in Figure 5, grouped all five species of Eimeria into five distinct clades, some level of variation existed within species of Eimeria in this study and that of the public database, as indicated by their mean genetic distances. The lowest genetic distance of 0.01 was observed among E. brunetti isolates. Similar ITS-1 sequence variations among E. mitis, E. tenella and E. maxima have also been reported by different authors (Bhaskaran et al. 2010;Kumar et al. 2015a;Lew et al. 2003;Thenmozhi, Veerakumari & Raman 2014).
In conclusion, this study characterised Eimeria species in broiler and free-range chickens based on molecular diagnostic techniques and determined their diversity in KwaZulu-Natal province. The study reports the presence of five Eimeria species (E. tenella, E. maxima, E. acervulina, E. brunetti and E. mitis), all of which are regarded as pathogenic. Although none of the chickens showed clinical signs of coccidiosis during sampling, the high prevalence of these pathogenic parasites in the study area suggests that subclinical infection is common in all infected chickens. Thus, effective control strategies remain imperative to curtail coccidial infection in poultry farms in the study areas. A survey on the types of anticoccidial used among commercial farms and their efficacy should be conducted to understand the impact of this disease. This will also help in the implementation of policies for the control of this disease in KwaZulu-Natal province.