Molecular detection of virulence genes in Salmonella spp. isolated from chicken faeces in Mafikeng, South Africa

Chickens have been implicated in most Salmonella disease outbreaks because they act as carriers of the pathogen in their gut. There are over 2500 serotypes of Salmonella that have been reported worldwide and 2000 of these serovars can be found in chickens. The main objective of this study was to determine the Salmonella serotypes found in poultry farms around Mafikeng district, South Africa. Salmonella was identified according to the guidelines of the International Organization for Standardization (ISO) (ISO 6579:2002) standard techniques. Faecal samples were collected and analysed for Salmonella using conventional cultural methods and polymerase chain reaction targeting the 16S Ribosomal Deoxyribonucleic acid (rDNA) gene for Salmonella identification. Out of 130 presumptive Salmonella isolates determined by urease and triple sugar iron tests, only 46 isolates were identified as Salmonella serotypes of which S. Typhimurium was the most frequent with 18 (39.1%), followed by S. Heidelberg with 9 (19.6%), S. bongori with 7 (15.2%), S. Enteritidis with 6 (13.0%) and both S. Paratyphi B and S. Newport with 3 (6.5%) each. Seven virulence genes including invA 100%, spy 39%, hilA 9%, misL 30%, sdfI 13%, orfL 11% and spiC 9% were detected from these Salmonella isolates in this study. The presence of these virulence genes indicates high pathogenicity potential of these isolates which is a serious public health concern because of zoonotic potential of Salmonella.


Introduction
Commercial poultry production is rapidly growing worldwide to meet the needs of the increasing human population (Olobatoke & Mulugeta 2015). Many poor and middle-class farmers in developing countries are taking up poultry to supplement their income, and this industry has become a major income source to them. Therefore, the presence of pathogenic organisms such as Salmonella in chickens, as a major food-borne infection in humans, can have an adverse impact on the production and marketing of poultry (Imanishi et al. 2015). Chickens have been implicated in most of Salmonella outbreaks because they act as carriers of this pathogen in their guts (Black 2008).
Salmonella that can be traced through chickens can be classified into three groups (Hafez 2013). The first group includes highly host adapted and invasive serotypes such as S. typhi in humans and S. gallinarum and S. pullorum in poultry. The second group is non-host-adapted and invasive serotypes such as S. Typhimurium, S. Arizonae and S. Enteritidis. The third group contains nonhost-adapted and non-invasive serotypes, and most of these serotypes are harmless to animals and humans (Andino & Hanning 2015;Umali et al. 2012). Understanding the mechanisms of Salmonella infection, intestinal colonisation, persistence and excretion in poultry are essential to discover appropriate measures to decrease both contamination of flocks and public health risk (Andino & Hanning 2015).
The role played by the virulence genes of Salmonella species, are known based on observations made on epithelial cells. The invA gene for instance affects the host cell by delivery of type III secreted effectors, for mutant phenotype, and is also essential for invasion of epithelial cells (El-Sharkawy et al. 2017;Marcus et al. 2000). The invA gene has been confirmed to be present in Salmonella species only and hence is used in the genetic diagnosis of Salmonella species (Fekry, Ammar & Hussien 2018).
The operon spv (Salmonella plasmid virulence) is considered as one of the virulence plasmids of numerous Salmonella serotypes that generate systemic disease (Castilla et al. 2006). It harbours five genes spvRABCD (Rotger & Casadesús 1999) that have been identified to contribute to its pathogenicity (Card et al. 2016;Haneda et al. 2012). The presence of HilA gene in Salmonella is essential for the expression of the type III secretion system (TTSS) components, and it encodes the central regulator HilA (Borges et al. 2013). This gene (HilA) is required to induce apoptosis of macrophages and invade epithelial cells (Borges et al. 2013). The sipC gene acts as a translocase, mediating bacterial entry into epithelial (Prasad 2012). On the other hand, spiC acts to modulate invasion gene expression (Hayward & Koronakis 1999).
The main objective of this study was to detect prevalent Salmonella serotypes in chicken faeces from poultry farms around Mafikeng in South Africa and assess the presence of virulence genes using conventional polymerase chain reaction (PCR) assays.

Sampling site
The study area covered Mafikeng in North West Province, South Africa, as shown in Figure 1. The province is the second largest chicken producer in South Africa at 21.3% after the Western Cape with 21.9%. The longitude and latitude of the district are 25°50′E and 25°55′N, respectively. Temperatures range from 3 °C to 21 °C in the winter and from 17 °C to 31 °C in the summer. The average rainfall is 360 mm.

Sample collection
A list of poultry farms in the Mafikeng area was compiled using the records of the provincial Department of Agriculture of North West provincial government. A few farms in the south, east, north and west were randomly selected, the farmers were approached and those who agreed were included in the study. Sampling sites were therefore conveniently selected. Sterile spatulas were used to collect samples of freshly passed poultry droppings in sterile universal sampling bottles.
Samples were collected at three different points in each poultry farm once a week. This was performed to have a good representation and distribution of the organisms. After collection, samples were packed in properly labelled sterile polyethylene bags and transported in a sterilised icebox and processed immediately upon arrival to the laboratory.

Isolation and identification of Salmonella
Each faecal sample (1 g) was weighed and transferred into a sterile container. A volume of 10 mL of buffered peptone water (BPW Oxoid, Biolab, South Africa) was added into each sample and then homogenised by vortexing for about 2 minutes followed by incubation at 37 °C ± 1 °C for 18 hours -24 hours Thereafter, 1 mL of the sample was transferred to 10 mL of Mueller-Kauffmann Tetrathionate Novobiocin (MKTTn) broth (Sigma-Aldrich, S.A. Barcelona, Spain) which was incubated at 4.5 °C for 6 h. Then, 1 mL was transferred from MKTTnB to 10 mL of Rappaport-Vassiliadis medium with soya (RVS) broth (Sigma-Aldrich, S.A. India) and incubated at 36 °C for 24 h. A 1 mL aliquot from the RVS broth was then transferred to a 30% glycerol solution (EMD chemicals, United States [US]) and stored at −20 °C for later use.

Culturing of Salmonella on selective agar plates
After incubation, a loopful of the enriched cultures of RVS broth was streaked separately onto two selective agar plates: Xylose Lysine Deoxycholate agar (XLD) (Merck, Wadeville, South Africa) and Brilliant Green Agar (BGA) (Scharlau Chemie S.A. Barcelona). These plates were incubated in an overturned position at 37 °C ± 1 °C for 18 h -24 h. Following incubation, the black and pink colonies with or without black centre on XLD agar, the colourless or opaque white colonies surrounded by pink or red zone and the red colonies on BGA were identified as suspected Salmonella. Suspected colonies of Salmonella spp. were then confirmed according to the guidelines of ISO 6579: 2002. Such colonies were picked out and streaked on Nutrient agar (NA) (Merck, Wadeville, South Africa) and incubated at 37 °C ± 1 °C for 18 h -24 h.

Biochemical identification
All gram-negative, rod-shaped bacteria were subjected to preliminary biochemical tests. The presumptive identification of Salmonella colonies was carried out using urease and triple sugar iron (TSI) tests. Two to three purified colonies of the presumptive Salmonella isolates from the 24-h NA plates were subjected to the urea broth and incubated at 37 °C for 18 h -24 h. Presumptive Salmonella colonies were also carried out by stabbing the butt of TSI slants, and the slants were incubated at 37 °C and examined after 18 h -24 h for gas production, hydrogen sulphide production and carbohydrate fermentation. The analytical profile index (API) 20E (BioMerieux, Marcy l'Etoile, France) was used for identification, and indices were generated for the diverse isolates and used to verify their identities using the API web TM identification software. Biochemical identification was performed in order to get presumptive positives and as part of a large set of other experiments that were being conducted on the samples.

Genomic deoxyribonucleic acid extraction
Genomic deoxyribonucleic acid (DNA) was extracted using Fungal/Bacterial Soil Microbe DNA Mini Prep kit according to the manufacturer's instructions (Zymo-Research, US). Extracted DNA was eluted with 100 µL of DNA elution buffer into clean 1.5 mL micro-centrifuge tube and stored at −20 °C for molecular confirmation of the Salmonella species and detection of virulence genes.

Molecular identification of Salmonella serovars using 16S rDNA gene
Polymerase chain reactions were conducted using the Universal primers: forward (5'-AGA GTT TGA TCC TGG CTC AG-3') and the Reverse (5'-ACG GCT ACC TTG TTA CGA CTT-3'), with the reaction volume of 25 µL containing 12.5 µL PCR Master Mix, 2 µL template DNA, 8.5 µL nuclease free water and 1 µL of each oligonucleotide primer using an Engine T100 Thermal TM cycler (Bio-Rad, Singapore). The thermo cycling conditions consisted of an initial denaturation step at 95 °C for 5 min followed by 35 cycles of denaturation at 94 °C for 30 s, annealing at 61 °C for 30 s and extension at 72 °C for 5 min, and finally a single and final extension step at 72 °C for 7 min. Polymerase chain reaction products were identified by electrophoresis on 2% (weight per volumne [w/v]) agarose gel stained with ethidium and visualised under ultraviolet light on a ChemiDoc Imaging System (Bio-Rad ChemiDocTM MP Imaging System, Hertfordshire, United Kingdom).

Deoxyribonucleic acid sequencing
Amplified PCR products were sequenced using an ABI PRISM® 3500XL DNA Sequencer (Applied Biosystems) at Inqaba Biotechnical Industrial (Pty) Ltd. (Pretoria, South Africa). The acquired sequences were aligned on GenBank database using Basic Local Alignment Search Tool (BLAST) (www.ncbi.nlm.nih.gov/BLAST) from the National Center for Biotechnology Information (NCBI) to identify sequences with high similarity.

Detection of virulence genes using polymerase chain reaction
Seven pairs of published oligonucleotide primers were used to detect the virulence genes using PCR. The individual PCRs, for each virulent gene, were set up in a 25 µL, which consisted of 12.5 µL AmpliTaq Gold® 360 PCR Master Mix (AmpliTaq Gold® DNA Polymerase 0.05 units/µL, Gold buffer 930 mM Tris/HCl pH 8.05, 100 mM KClO, 400 mM of each dNTP and 5 mM MgCl 2 ) (Applied Biosystems, California, US). Then, 2.5 mM of each primer, 2 µL of template DNA and ddH 2 O were added to make the final volume. Test DNA was replaced with 5 µL of sterile nuclease-free water as negative control. Cycling conditions for PCR as well as the information of the published primers used are detailed in Table 1.

Phylogenetic analysis
All confirmed sequencing results were edited by using Bio-Edit software (Hall 1999) and saved as FASTA format. The sequences were used to search the GenBank database with the BLASTn algorithm to find out the relative Phylogenetic positions. The sequences were aligned by using the multiple alignment fast Fourier transform (MAFFT) programme 6.8464 to conduct multiple and pairwise sequence alignments against corresponding nucleotide sequences retrieved from GenBank. Evolutionary distance matrices were generated as described previously by Jukes and Cantor (1969). Phylogenetic analysis was performed using MEGA version 7 (Kumar, Stecher & Tamura 2016), and neighbour-joining (NJ), maximumparsimony, and maximum likelihood methods were used for the construction of the trees. Bootstrap analyses were performed using 1000 replications for NJ, maximumparsimony, maximum likelihood. Recognised chimeric sequences were identified using the Chimera Buster 1.0 software. Manipulation and tree editing were carried out using Tree View (Timme et al. 2013).

Statistical analysis
Descriptive statistics, particularly proportions, percentages, means and standard deviations, were determined using Excel software (2013).

Ethical considerations
Prior to the commencement of the study, the research proposal was approved based on Animal Research Ethics Committee (NWU-00274-18-A5) guidelines by North-West University Research Ethics Regulatory Committee (NWU-RERC).

Phylogenetic analysis of the Salmonella isolates from chickens
A phylogenetic tree for Salmonella isolates from chickens was constructed to understand the genetic closeness of Salmonella strains with other related strains from different countries inside and outside the African continent. Shigella fleneri (KY199565) from the family Enterobacteriaceae was used as an out-group for the 16S-Ribosomal Deoxyribonucleic acid gene. The resulting NJ revealed that strain S. Heidelberg clustered closely with sequences of the following S. Heidelberg: CP004086.1, originating from retail meats, humans and animals; CP016586.1, originating from food sources, human and animal; S. Tennessee: CP024164, originating from laboratory control strains and CP025217.1, originating from the food industry; S. bongori: MF289161.1, originating from the crops, and MG663480.1, originating from chicken faeces; S. enterica Paratyphi; CP006575.1, originating from bacteria strain obtained from source laboratory of Zoonosis; S. Typhimurium: CP014971.2, originating from humans and cattle; S. Enteritidis: CP018651.1, originating from the historical S. Enteritidis isolated between the 1940s and 1990s. The phylogenetic tree is shown in Figure 2.

Distribution of virulence genes among Salmonella isolates
A representative detection of Salmonella virulence genes is shown in Table 3. The results revealed that 100% isolates were harbouring invA gene, 39% harboured spy gene, 9%

Discussion
The primary objective of this study was to isolate and detect the types of Salmonella spp. from chicken droppings around Mafikeng, South Africa. Six Salmonella serovars were isolated, namely S. Typhimurium, S. Heidelberg, S. bongori, S. Enteritidis, S. Paratyphi B and S. Newport. All these serovars are known to be pathogenic in humans and cause salmonellosis (Fagbamila et al. 2017;Orji, Onuigbo & Mbata 2005;Tejada et al. 2016). Among the Salmonella isolates detected from this study, S. Typhimurium, S. Enteritidis and S. Newport have previously been reported in raw broiler samples by another study here in Mafikeng, South Africa (Olobatoke & Mulugeta 2015). The presence of these isolates, especially S. Typhimurium and S. Enteritidis as determined by this study, is a concern because they are known to cause serious human illness (Alvarez et al. 2004;Lapuz et al. 2008).
A previous study of Salmonella isolates in South Africa undertaken by Kidanemariam, Engelbrecht and Picard (2010) between 1999 and 2006 revealed the 10 common Salmonella serotypes isolated: S. Newport, S. Typhimurium, S. Dublin, S. Enteritidis, S. Muenchen, S. Chester, S. Heidelberg, S. Hadar, S. Schwarzengrund and S. Mbandaka. Our study did not isolate all these Salmonella species in the Mafikeng chicken samples but agrees with that study which identifies S. Typhimurium as the most prevalent isolate in poultry (Kidanemariam et al. 2010).
Many poor and middle-class farmers in developing countries are taking up poultry to supplement their income, and this industry has become a major income source to them.
Unfortunately, there has also been an increase in human salmonellosis cases, and these have been linked with consumption of contaminated chicken products (Imanishi et al. 2015;Lebert et al. 2018;Olobatoke & Mulugeta 2015).  Freitas Neto et al. 2010). Additionally, the majority of Salmonella cases worldwide are caused by S. serovar Enteritidis, from eggs and poultry meat (Backhans & Fellström 2012). Salmonella enterica serovar Paratyphi B has been linked with human outbreaks in the different countries like US (Harris et al. 2009), Australia (Levings et al. 2006), Canada (Stratton et al. 2001) and European countries (Miko et al. 2002). This is noteworthy as it highlights the possibility that the detected strains from this study may also play a significant role in human disease here in the province if proper sanitary measures are not applied. The phylogenetic analysis also revealed that the Salmonella isolates found in this study are very similar to other genotypes previously identified in several other sources, including humans, milk, chickens and other animals. This therefore indicates that the strains that were found in chickens could easily be circulating in other poultry, livestock, animal products and/or humans.
The disease causing potential of Salmonella isolates is derived from the virulent genes that they may carry. So, this study undertook to determine the presence of these virulent genes to understand how potentially pathogenic the strains were (Ekwanzala et al. 2017;Sunar et al. 2014). Forty-six (n = 46) Salmonella isolates were analysed for the presence of seven known virulence genes, namely invA, SdfI, hilA, misL, Spy, orfL and spiC. It was interesting to note that all the tested virulence genes were detected by this study in differing proportions. All 46 (n = 46) isolates were positive for invA, indicating that all of them have the ability to invade and to cause gastroenteritis (Ekwanzala et al. 2017;Hu et al. 2008;Lan et al. 2018;Odjadjare & Olaniran 2015;Sunar et al. 2014) and can survive in macrophages (Goodman et al. 2017). The inner membrane of Salmonella contains protein coded for by invA, which is vital for invasion into epithelial cells (Salehi, Mahzounieh & Saeedzadeh 2005). This gene (invA) is found in Pathogenicity island-1 and it is a TTSS apparatus, which secretes invasion effectors like invasion factor A   (Sabbagh et al. 2010). Eighteen isolates harboured spy, and these isolates were identified as those of S. Typhimurium. This gene has been used to identify and confirm S. Typhimurium (Can et al. 2014). The spy gene appears to exclusively act as a molecular chaperone (Wells 2015) and thus confers pathogenicity to this strain. The Sdf I gene has been used to identify S. Enteritidis (Mohd Afendy & Son 2015). Salmonella serovars encoding spy and Sdf I (S. Typhimurium and S. Enteritidis) are known to be associated with human illness (Odjadjare & Olaniran 2015).
Out of 46 Salmonella isolates, 14 were identified as harbouring the misL gene. This gene aids virulence by being involved in the intra-macrophage survival of the pathogen (Hughes et al. 2008). The misL gene is located in the SPI-3 of Salmonella (Zishiri et al. 2016). The orfL and spiC genes were positive from five and four isolates, respectively. The spiC gene is one of the virulence factors of Salmonella which is found in island 2 and associated with type 3 effector protein.
It normally disrupts the vesicular transport of the host cell (Kaur & Jain 2012). The orfL gene, on the other hand, has a secretion system that mediates the secretion of toxins and is necessary for macrophages survival (Odjadjare & Olaniran 2015).
The presence of these genes in the isolates from Mafikeng therefore shows that these strains are of public health importance and can cause disease when and if contamination is not properly managed.

Conclusion
This study has identified Salmonella isolates that have traditionally been associated with disease not only in South Africa but elsewhere in the world. These isolates have the virulent genes that are important for pathogenesis and therefore are of serious public health concern and measures should be put to control disease outbreaks in Mafikeng specifically and South Africa in general.