To identify the sources of Salmonella contamination, distribution, prevalence and antimicrobial susceptibility patterns, which have significant impact on public and animal health, and international trade.
To identify the sources of Salmonella contamination, distribution, prevalence and antimicrobial susceptibility patterns, which have significant impact on public and animal health, and international trade.
A total of 1888 samples were collected by stratified random sampling from 2009 to 2011 from cattle, camels, poultry, fish, vegetables and humans. All identified Salmonella isolates were serotyped and tested for antimicrobial susceptibility by MIC determinations. A total of 149 Salmonella isolates comprising 17 different serovars were obtained (7·9% prevalence). Salmonella Hadar (37%), S. Eko (17%), S. Enteritidis (10%), S. Kentucky (7%) and S. Uganda (7%) were isolated from different sources. The occurrence of antimicrobial resistance was generally low, but S. Enteritidis and S. Eko showed variable antimicrobial resistance patterns, while all S. Kentucky isolates were resistant to seven of 17 tested antimicrobials, including ciprofloxacin and nalidixic acid. Three S. Hadar isolates revealed reduced susceptibility to ciprofloxacin and susceptibility to nalidixic acid and harboured the plasmid-mediated quinolone resistance gene qnrS1.
Salmonella serovars Hadar, Enteritidis and the previously very rarely reported Eko were the major serovars associated with human infections, animal and environmental contamination in the north-eastern region of Nigeria.
These serovars constitute a health risk to poultry, environment and human population in the region.
Worldwide, Salmonella is estimated to cause 93·8 million human infections and 155 000 deaths annually (Majowicz et al. 2010). Although most of these infections cause mild gastroenteritis, life-threatening disseminated infections are common among children, elderly and immunocompromised patients (Hohmann 2001; Le Hello et al. 2011).
Salmonellosis is considered as one of the most widespread foodborne zoonoses in industrialized as well as developing countries, even though the incidence seems to vary between countries (Molla et al. 2003). Various animals, including poultry, pigs, cattle and reptiles, are reservoirs of Salmonella species, and most human infections are acquired through consumption of undercooked food of animal origin or contaminated water and vegetables (Lynch et al. 2006; Majowicz et al. 2010).
A limited number of the 2579 currently recognized serovars of Salmonella account for the vast majority of human infections (Hendriksen et al. 2011). In most developed countries, Salmonella enterica serovars Typhimurium and Enteritidis are the major reported causes of human salmonellosis; however, other serovars are common in specific geographical regions in association with human salmonellosis, that is, S. Stanley and S. Weltevreden in South-East Asia (Galanis et al. 2006; Hendriksen et al. 2009, 2011, 2012).
It is usually difficult to evaluate the occurrence of Salmonellosis and antimicrobial resistance in developing countries due to lack of coordinated surveillance systems (Acha and Szyfres 2001). This is also the case in Nigeria, where studies conducted so far have only included a limited number of samples/isolates from a single or a few reservoirs (Orji et al. 2005; Akinyemi et al. 2010; Fashae et al. 2010; Muhammad et al. 2010). The comparisons of the Salmonella serovars distribution in different reservoirs, occurrence of antimicrobial resistance in combination with baseline information represent the first necessary step to design an effective Salmonella control programme aiming to limit the occurrence of Salmonella infections in humans in Nigeria.
The purpose of this study was to perform the first comprehensive surveillance of Salmonella in cattle, camel, poultry and poultry-related sources (rodents, litter, water and lizards from poultry houses), fish, vegetables and humans in the north-eastern region of Nigeria. The susceptibility to antimicrobials in the obtained Salmonella isolates was investigated, and isolates displaying a phenotype compatible with the presence of plasmid-mediated quinolone resistance genes were tested for qnr genes.
A total of 1888 samples were collected in 2009–2011 from ruminants (cattle and camel), poultry, poultry-related sources (rodents, litter, water and lizards), catfish (Clarias spp.), vegetables and human. Samples from ruminants were collected from 400 different individuals. From cattle (n = 200), 50 samples each of lymph node, liver, intestine, kidney were collected, while from camel (n = 200), 50 samples each of spleen, faeces, intestine and liver were collected. Samples (16–18 per trip) were taken at Maiduguri central abattoir (Fig. 1; site A), Monday and Gamboru markets in Maiduguri (being the only government owned and certified abattoir and markets patronized by urban dwellers) (Fig. 1; sites D and C), on weekly basis from April to September, 2010 giving an average of 17 samples per sampling trip (Table 1).
Three poultry slaughter sites (Fig. 1; sites C, D and E) processing intensively and extensively reared chickens were visited 10 times between September and October, 2010. A total of 228 samples, including faecal droppings and liver (n = 50 each), small intestine (n = 53) and spleen (n = 75), from different animals were collected with an average of 23 samples per visit. Five intensively managed poultry farms (Fig. 1; sites G, D and F) were visited 10 times between November and February, 2011. A total of 270 samples were collected (26–28 samples per trip) from litter (n = 70), feed (n = 50), faeces from rodents (n = 50), water from drinkers (n = 50) and faeces from lizards (n = 50), with an average of 27 samples per visit (Table 1).
A total of 490 faecal samples from human patients (men and women) with symptoms of gastroenteritis attending UMTH hospital (Fig. 1; site F) and Ben medical and diagnostic laboratory (Fig. 1; site B), Maiduguri were collected during the months of October–November, February–April and January–March, respectively, in years 2009, 2010 and 2011 (Table 1).
Two hundred samples of catfish (Clarias spp.) intestines were collected during eight visits to two fish farms (Fig. 1; sites G, H) and Gamboru market (Fig. 1; site C) during the months of May–August in 2009 (Table 1).
Vegetable samples were collected during ten visits to five different farms (Fig. 1; sites H and C) and Maiduguri central market (Fig. 1; sites D and E) in May–August 2009. A total of 300 samples (average of 30 samples per visit) divided equally among five species of vegetables, including spinach (Amaranthus hybridus spp.), Corchorus Olitorus spp., sorrel (Hibiscus sabdariffa), bitter leaf (Vernonia Amygadalina spp.) and water leaf (Talinum Triangulares spp), were collected (Table 1).
The samples were analysed at the microbiology laboratory of the Faculty of Veterinary Medicine, University of Maiduguri, Nigeria, on the same day as they were collected.
Twenty-five grams of each meat sample and 5 g of each faecal, litter and vegetable leave sample were enriched in 10 ml and 25 ml of Selenite-F broth, respectively (Laboratarios Britania, Buenos Aires, Argentina). The samples were incubated at 37°C for 18–24 h. Following enrichment, cultures were inoculated onto desoxycholate citrate agar (Park Scuntif, Northampton, UK) and incubated at 37°C for 18–24 h. Presumptive nonlactose fermenting dark centre colonies were subcultured onto xylosine lysine desoxycholate agar (Oxoid Ltd., Hampshire, UK) and incubated for 18–24 h at 37°C. Finally, presumptive positive Salmonella isolates were inoculated onto nutrient agar (Fluka Biochemika, Steinheim, Germany) slants and incubated at 37°C for 24–48 h. Salmonella isolates were presumptively identified using biochemical characterization according to the standard techniques recommended (Cowan and Steel 1974).
All the isolates were serotyped at the WHO National Salmonella and Shigella Center, Bangkok, Thailand, by slide agglutination, O and H antigens were characterized by agglutination with hyperimmune sera (S & A Reagents Laboratory, Ltd., Bangkok, Thailand), and serotypes were assigned according to the Kauffmann–White scheme (Popoff and Minor 2007). MIC determinations were performed on all isolates at DTU-Food, Denmark, using a commercially prepared, dehydrated panel (Sensititre; TREK Diagnostic Systems Ltd., East Grinstead, England). Based on MIC results, Salmonella serovars exhibiting a phenotype compatible with the presence of PMQR genes were selected for PCR amplification of qnrA, qnrB, qnrC, qnrD and qnrS, genes, respectively (Hendriksen et al. 2009). PCR products were purified (GFX PCR DNA kit; Amersham Biosciences, Little Chalfont, Buckinghamshire, UK) and submitted to Macrogen Inc. for sequencing. Sequence analysis and alignment were performed using Vecton NTI suite 9 (InforMax, Inc., Bethesda, MD, USA). Resulting nucleotide sequences were compared by use of the BLAST computer program (National Center for Biotechnology Information).
Of the 1888 samples analysed, 149 (7·9%) were positive for Salmonella. A total of 17 serovars were identified, of which seven, nine, twenty, seven, three and three were distributed among cattle, camel, poultry/poultry sources, human faeces, fish and vegetables, respectively (Table 1, Fig. 2).
In cattle, the highest occurrence of Salmonella-positive samples was recorded for lymph nodes for which 8 (16%) samples resulted in Salmonella growth, and the lowest occurrence of Salmonella-positive samples was shown by kidney samples being only 1 (2%) sample positive. In camels, Salmonella was isolated from 14% (n = 7) and 12% (n = 6) spleen and small intestine samples, respectively, and from 2% (n = 1) liver samples. In poultry, the highest occurrence of Salmonella-positive samples was found in spleen and intestine samples being 11% (n = 8) and 11% (n = 6), respectively. In this animal species, the lowest occurrence of Salmonella-positive samples was found in liver (2%, n = 1). The occurrence of Salmonella-positive samples recovered from poultry environment was 16% (n = 8) both in feed and in rodents, and 10% (n = 7) in litter. Occurrence of 6% (n = 27) Salmonella-positive samples was found in humans accounting for the highest number of different serovars. Salmonella isolation occurred in 12% (n = 24) catfish and 6% (n = 19) vegetable samples.
Overall, S. Hadar was the most prevalent serovar (n = 55, 37%) followed by S. Eko (n = 26, 17%), S. Enteritidis (n = 15, 10%), S. Kentucky (n = 11, 7%), S. Uganda (n = 10, 7%) and S. enterica ser. 47:mt:-(n = 9, 6%) (Table 1). In addition, a few rarely encountered serovars were observed such as S. Ank, S. Bonnariensis and S. Verviers. Several of the most prevalent serovars were observed in multiple reservoirs. S. Hadar was obtained from poultry (small intestine, spleen), litter/poultry droppings, fish, vegetables and human. Similarly, S. Eko was isolated from cattle, camel, fish and human. In contrast to the Salmonella serovars Hadar, Eko, Uganda, and S. enterica ser. 47:mt:- that were all found in both humans and multiple reservoirs, S. Kentucky was not detected in humans but in poultry, rodents, water and lizards (Table 1).
In general, a low-level prevalence of resistance to the 17 tested antimicrobials was observed across all serovars included in this study (Table 2). No resistance was observed to apramycin (APR), cefotaxime (FOT) and ceftiofur (XNL).
|Salmonella Serovar||Origin||No. of isolates||No. (%) of isolates resistant to various antimicrobial agents at the indicated breakpoints (μg ml−1)a|
|Hadar||Poultry/poultry environment||12||0||0||0||0||2 (17)||3 (25)||0||0||2 (17)||0||0||0||1 (8)||5 (42)||1 (8)||0||1 (8)||0|
|Vegetable||13||0||1 (8)||0||0||0||0||0||0||0||0||1 (8)||2 (15)||0||5 (38)||3 (23)||1 (8)||4 (31)||0|
|Human||10||0||2 (20)||0||0||0||0||0||0||0||0||0||0||1 (10)||3 (30)||1 (10)||0||1 (10)||0|
|Fish||20||0||0||0||1 (5)||0||0||0||0||0||0||0||2 (10)||1 (5)||9 (45)||7 (35)||0||4 (20)||0|
|Eko||Fish||1||0||1 (100)||0||0||0||0||0||0||0||0||0||0||0||0||1 (100)||1 (100)||1 (100)||0|
|Cattle||12||0||0||0||0||0||0||0||0||0||0||0||0||0||1 (8)||2 (17)||0||0|
|Human||6||0||4 (67)||0||0||0||0||0||0||0||0||0||0||0||0||5 (83)||5 (83)||5 (83)||0|
|Enteritidis||Poultry/poultry environment||10||0||1 (100)||0||0||1 (10)||0||0||0||0||0||0||0||0||1 (10)||1 (10)||0||1 (10)||0|
|Human||5||0||4 (80)||0||0||4 (80)||0||0||0||0||0||0||0||0||4 (80)||4 (80)||4 (80)||4 (80)||0|
|Kentucky||Poultry/poultry environment||2||0||0||0||0||0||0||2 (100)||0||0||2 (100)||2 (100)||0||2 (100)||2 (100)||2 (100)||2 (100)||0||0|
|Rodent||6||0||0||0||0||0||0||6 (100)||0||0||6 (100)||6 (100)||0||6 (100)||6 (100)||6 (100)||6 (100)||0||0|
|Water||1||0||0||0||0||0||0||1 (100)||0||0||1 (100)||1 (100)||0||1 (100)||1 (100)||1 (100)||1 (100)||0||0|
|Lizard||2||0||0||0||0||0||0||2 (100)||0||0||2 (100)||2 (100)||0||2 (100)||2 (100)||2 (100)||2 (100)||0||0|
|Poultry/poultry environment||3||0||0||0||0||0||0||0||0||0||0||0||0||0||0||2 (67)||0||0||0|
|Fish||2||0||0||0||0||0||0||0||0||0||0||0||0||0||1 (50)||0||1 (50)||0||0|
|Poultry/poultry environment||1||0||0||0||0||0||0||0||1 (100)||0||0||0||0||0||0||0||0||0||0|
|Wilhemsburg||Poultry/poultry environment||3||0||0||0||0||0||0||0||0||0||0||0||0||0||0||2 (67)||0||0||0|
|Total||149||0||13 (9)||0||1 (1)||7 (5)||3 (2)||14 (9)||1 (1)||2 (1)||11 (7)||13 (9)||4 (3)||14 (9)||42 (28)||4||23 (15)||21 (14)||0|
All the S. Kentucky isolates, irrespective of their source of isolation (poultry and poultry-related samples), showed a high level resistance (>0·06 mg l−1) to ciprofloxacin (CIP) and nalidixic acid (NAL) indicating multiple mutations in gyrA and/or parC as mechanisms of resistance. All these isolates were additionally resistant to gentamicin (GEN), spectinomycin (SPE), streptomycin (STR), sulfamethoxazole (SMX) and tetracycline (TET). All S. Hadar isolates were susceptible to amoxicillin + clavulanic acid (AUG), apramycin (APR), colistin (COL) and GEN, while the level of resistance to the remaining antimicrobials varied according to the origin of the isolates (Table 2). It is noteworthy to mention that three S. Hadar isolates from poultry showed low-level resistance to CIP (MIC = 0·064–1) and susceptibility to NAL (MIC ≤ 8).
The occurrence of antimicrobial resistance was low in S. enterica serovar 47:mt:-, S. Uganda, S. Amager, S. Ank and S. Wilhelmsburg, and the remaining eight serovars comprising S. Elisabethville. S. Mbadaka, S. Verviers, S. Vinohrady, S. Colindale, S. Westhampton, S. Give and S. Bonnariensis were pansusceptible to all the 17 antimicrobials tested (Table 2).
The three S. Hadar isolates obtained from poultry sources and exhibiting low-level resistance to CIP and susceptibility to NAL harboured qnrS1 genes.
In this study, the presence of multiple Salmonella serovars may be related to the geographical location of the area, is bordered in the north by Niger Republic, north-east by Chad Republic and east by Cameron Republic, coupled with transhumance across the borders into Nigeria, during the dry season (Braukamer 1996). Furthermore, farmers utilize cattle and poultry faeces as fertilizer on farmlands situated close to the river; during the rainy season, the topsoil is washed into the river resulting in environmental and food (vegetable and fish) contamination (Johannessen et al. 2004; Brooks et al. 2012). These result in a complex epidemiology which need to be investigated, and preventive and appropriate control measures implemented.
The major reservoirs contributing to human infections were cattle, camel poultry and environmental samples, fish and vegetables (Fig. 2). They are important sources of protein and vitamins for the population (Gambo et al. 2010) and constitute the basic food products consumed by the inhabitants.
Salmonella Hadar serovar is frequently isolated from chickens, is equally one of the serovars commonly recovered from human in Europe and Africa (Nzouankeu et al. 2010; Hendriksen et al. 2011), possibly due to international food and livestock trade from endemic areas and human migration (Le Hello et al. 2011). With this study, it might be due to environmental contamination by peasant farmers; this might be a significant contributory factor for the multi-host dissemination of the serovars, thus revealing the widespread cross-contamination in the country agricultural production system.
The study revealed low S. Hadar positive samples when compared with 2·2% obtained by Muhammad et al. (2010) in Jos, central Nigeria. The isolation of S. Hadar from vegetables and fish illustrates the possibility of cross-contamination as previously described (Cardinale et al. 2005; Bouchrif et al. 2009; Muhammad et al. 2010; Marin et al. 2011).
Three of the S. Hadar isolates harboured plasmid-mediated quinolone resistance gene qnrS1 located on mobile genetic elements; lack of effective policy on the use of antibiotics in poultry, including fluoroquinolones, may facilitate selection and spread of plasmids to other serovars (Parry and Threlfall 2008; Garcia-Fernandez et al. 2009) ; this is a cause of concern because resistance to fluoroquinolones may compromise effective treatment for Salmonella infections in humans.
Salmonella Eko is rarely reported previously; to our knowledge, it has only been reported from poultry sources, in Cameroun in 2006–2007 (Nzouankeu et al. 2010). The emergence of this serovar is a cause for concern especially its presence in multiple reservoirs (cattle, camel, fish and human). Based on the very low frequency and limited geographical distribution, this serovar could be considered as being unique to Africa. This might be due to the extensive methods of livestock and poultry production practiced in most developing countries especially in tropical regions where native animals are raised and consumed locally. Those locally produced animals may harbour a greater diversity of less commonly reported serovars and tend to remain within the region. Similarly, some serovars may possess a genetic background that adapts them to the harsh tropical environment (Joerger et al. 2009; Hendriksen et al. 2011).
The association of S. Kentucky with poultry dates back to 1937 in United States where it was isolated from chickens, the patterns of resistance to antimicrobials including CIP and NAL, reaffirmed the initial speculation of the existence of a prevailing circulating poultry-associated STI98-X1 CIP R S. Kentucky clone in Africa, including Nigeria (Le Hello et al. 2012). The lack of policy to control the use of antimicrobials in poultry especially fluoroquinolones, including ciprofloxacin, enrofloxacin and ofloxacin in Nigeria (Adetosoye, 1980), may have contributed to its rapid spread in poultry farms.
The majority of the infrequent serovars were generally pansusceptible. These serovars were distributed in multiple reservoirs. Previous studies showed that reptiles might be reservoirs for infrequent Salmonella serovars which are able to infect humans, thereby constituting a significant public health problem (Mascher et al. 1988; Wells et al. 2004). In Nigeria, the transmission routes may be associated with household lizards, which scavenge on foodstuffs and farm produces spread in the open surfaces to be sundried resulting in contamination of the food/farm produces.
The limitation of this study includes a limited geographical coverage and the convenience sampling method employed; the combined effect of these limitations is reflected in the research output; it is therefore suggested that future researcher should take these limitation into consideration.
It is recommended that action is taken to diminish the level of cross-contamination in agricultural food production systems, thereby lowering the burden of Salmonella in the different reservoirs; this will equally boost the export trade. It is also suggested that a source attribution programme be created to reveal the true link between the different Salmonella serovars. The information obtained from this attribution study can then be used to control infections by launching targeted, serovar-specific interventions against the true reservoirs.
This study reports the relative importance of S. Hadar, S. Eko and S. Enteritidis as major pathogens associated with human infections, animals and environmental contamination in the north-eastern region of Nigeria. In addition, this study reported the occurrence of rare/infrequent serovars which can be of global importance because of travels, transhumance, animals and food products trade. It equally emphasized a need for heightened awareness by national and international health, food and agricultural authorities to implement measures to monitor and limit the spread of Salmonella by Nigeria government. This study also highlighted relatively diffuse occurrence of resistance to critically important antimicrobials like fluoroquinolones. Because antimicrobials are easily obtainable over the counter, there is a need for the authorities in Nigeria to promulgate policy to regulate antimicrobial use in agriculture and in clinical settings to limit the spread of multidrug-resistant Salmonella isolates and plasmids among humans and animals.
We are grateful to Miss Maria Louise Johannsen for her outstanding technical assistance. This work was supported by the Center for Genomic Epidemiology (www.genomicepidemiology.org) at the Technical University of Denmark funded by Grant 09-067103/DSF from the Danish Council for Strategic Research, as well as by Grant 3304-FVFP-08 from the Danish Food Industry Agency and by the World Health Organization Global Foodborne Infections Network (www.who.int/gfn).
No conflict of interest to be declared.