Salmonella Characterization in Poultry Eggs Sold in Farms and Markets in Relation to Handling and Biosecurity Practices in Ogun State, Nigeria

Salmonella remains one of the notable food-borne bacterial pathogens. It is associated with poultry and poultry products including eggs. This study investigated Salmonella distribution in eggshell and content, their antimicrobial resistance pattern, and the possible risk factors driving contamination in Ogun State, Nigeria. A total of 500 eggs (5 eggs pooled into one sample) were collected and culturally examined for the presence of Salmonella serovars. Isolates were further characterized biochemically using Microbact 20E (Oxoid) and Antimicrobial susceptibility determined by the Kirby-Bauer disk diffusion method. A total of 14 Salmonella isolates spread across 10 serovars were recovered from the 100 pooled egg samples; 10 (10%) from the market and 4 (4%) farms, 13(13%) eggshell, and 1(1%) egg content. All tested serovars were susceptible to ampicillin, chloramphenicol, florfenicol, and kanamycin. Resistance was mostly observed in sulfamethoxazole 8 (80%), followed by ciprofloxacin 5 (50%) and tetracycline 3 (30%). Sales of eggs in the market appear to be a strong factor encouraging contamination in addition to poor biosecurity and unhygienic handling of eggs on the farm.


Introduction
Poultry eggs provide significant amount of animal protein in Nigeria and other developing nations Sub-Saharan since they are cheap, available and have little or no limitation in acceptance across socio-cultural and religious divide (1,2). With a poultry population of approximately 180 million, Nigeria produces an average of 3.8 million eggs annually (3). However, safety and wholesomeness of eggs can pose serious public health challenge since when contaminated. Eggs are the major sources of human non-typhoidal (NT) salmonellosis (4,5).
Salmonellosis is an important public health burden in most developing countries and constitute a major food borne pathogen in the developed world (6). It is largely self-limiting but serious consequences may result where infected individuals are immune-compromised or co-infected with malaria or human immunodeficiency virus (HIV) (7,8). In the United States, about 1.4 million people are infected annually, with approximately 15, 000 hospitalizations and 4000 deaths (9).
Consumption of raw and improperly cooked eggs and egg products have been identified as the major route for the transmission of NT Salmonella (10) but cross contamination in the kitchen has also been reported in the outbreaks of Salmonella in humans (11,12). Salmonella contamination of egg impose a public health burden on human population as well as economic losses in the poultry sector.
In Nigeria, NT Salmonella have been mostly characterized in the poultry sector from faeces, dust, environment and poultry meat (13,14,15,16). However, information on the drivers of egg contamination and transmission is scanty due to lack of coordinated national surveillance program.
Also, increasing reports of multi-drug resistance to antimicrobial agent from Salmonella strains isolated from egg shells and contents (17,18) signals a need for similar investigations in the poultry sector in Nigeria. Consequent on the above, this study seeks to establish: a) a baseline survey of Salmonella occurrence in poultry eggs, ii) determine the circulating Salmonella serovars and their antimicrobial resistance profile and iii) determine possible risk factors that may be driving Salmonella contamination of eggs.

Results
Prevalence and diversity of Salmonella isolates on shells and in contents of eggs Of the 100 samples pooled from 500 eggs, 14 (14%) were positive for Salmonella spread across shell (13/14, 92.9%) and content (1/14; 7.1%) ( Table 1). Salmonella in egg shell was significantly higher (p<0.05) than that of egg content.
Ten (10) Salmonella isolates were obtained from 40 samples (i.e 40x5=200) eggs sold in the markets while the remaining four (4) were from 60 sample (i.e 60 x 5=300) eggs obtained from the farm. The most isolates were from Egba (n=7) followed by Yewa (n=4), Ijebu (n=2) and Remo (n=1). All Salmonella isolates were obtained from egg shell except one (S. Alachua) from Remo zone was obtained from egg contents (Table1). In this study, 10 different Salmonella serovars were identified from 14 Salmonella isolates from eggs predominantly sold in the market and included: Agama, Durham, Bradford, Derby and Kentucky. Only serovars Kentucky was common in samples from the market and farms. Other serovars identified in the farm sourced eggs include Kingston, Colorado, Lattenkamp, Carno and Alachua (Table1).  The antibiotic resistance profile of fourteen Salmonella serovars were subjected to 11 commonly used antimicrobial agents ( Table 2). All tested serovars were susceptible to ampicillin, chloramphenicol, florfenicol and kanamycin. Resistance was most predominantly shown to sulfamethoxazole 8 (80%), followed by ciprofloxacin 5 (50%) and tetracycline 3 (30%). Gentamicin, nalixidic acid and streptomycin showed equal resistance 2 (20%) with the least resistance observed in trimethoprim 1 (10%) (Table2). The most resistant Salmonella serovar to antimicrobials was S. Kentucky (n=6), followed by S. Carno (n=4) and S. Derby (n=3). There was no reissuance shown to any of the antimicrobials by S. Colorado (Table2).  Of the 14 positive Salmonella isolates spread across 10 serovars, five were resistant to two or more antimicrobials and included Kentucky, Bradford, Derby, Carno and Kingston while four including Agama, Lattenkamp, Durham and Alachua showed resistance to single antimicrobials (Table2). For the 7 Salmonella isolates tested from Egba zone, only one multi-resistance pattern (SMX-GEN-TET-STR-CIP-NAL) was observed in comparison to two patterns each from Yewa (SMX-CIP; SMX-TET-CIP) and Ijebu (CIP-NAL; SMX-GEN-STR-TMP) with four and two isolates respectively (Table 3).  (1) Kentucky (1) Durham (1) Bradford (1) SMX ( Derby (1) Kentucky (1) Carno (1) Alachua (1) SMX-TET-CIP (1) CIP-NAL (1) SMX-GEN-STR-TMP (1) CIP (1) Note: All antimicrobials tested were fully sensitive to two S. Agama  Biosecurity practices in the production and handling of eggs The questionnaire results indicated a predominant cage system (73.3%) operations, compared to the deep litter system (26.7%). Eighty percent (48/60) of the farms were less than 500m away from other farms and the tendency for farms to be visited by wild birds. Twenty percent of the farms included antibiotics in their poultry feeds routinely.
A wide biosecurity concerns exist across most farms with only about 21.6% (13/60) of farm operations involve personal protective equipment (PPEs) where necessary. About half of the responding farms share tools with other farms, thereby encouraging pathogen transfer. All respondents (100%) do not clean their eggs in any form before selling. (Table  4). A fifth (12/60) of the respondents include antibiotics in feed as growth promoters or prophylactics while 4168.3 % (41/60) of the respondents allow in-farms sales of eggs (Ta-ble4).

Discussion
In this study, a prevalence of 14% non-typhoidal Salmonella was detected from pooled egg samples. To the best of our knowledge this study provides the first detailed comparison of Salmonella serovars profile sold on farm and in open market. Our results corroborate Salmonella presence in eggs in Nigeria as previously reported (19). A previous national study reported 24.5% prevalence of NT Salmonella in poultry environments in Ogun State (15). The differences in the prevalence of the two studies may be attributed to the sample types investigated. While the national study employed a matrix of five samples (dust, litter, faeces, feed and water) from poultry environments, the current study focused on pooled poultry egg samples.
Data from this study further highlight the potentials continuous relevance of poultry eggs as important transmission reservoir of Salmonella in humans. Thirteen out of the 14 Salmonella isolates identified in this study were found on the egg shells and may suggest faecal, environmental or handling contamination (20). Only one Salmonella isolate (S. Alachua) was detected in egg content, the route of contamination was not investigated. S. Alachua was recently reported from faecal samples in the Northern part of Nigeria (21). Further study will be required to determine if S. Alachua was an accidental finding in the egg content, vertically transmissible or has the ability to penetrate egg shells into content.
The fourteen Salmonella isolates identified in this study were spread across 10 serovars, which depicts high serovar diversity. Studies across Nigeria have reported similar observations reported (15,16,21). Plausible reasons for this observation is the indiscriminate importation of poultry birds and eggs with no coordinated national screening and control programme in place for salmonellosis. While S. Agama (n=3) was the most occurring in this study, all three isolates were from the same market and zone. It is then possible that all three are clonally related, although clonal relatedness was not explored in this study. Two S. Kentucky serovars were identified, each from a different zone. Fagbamila et al (15) reported S. Kentucky in 11 states out of the 12 that were sampled in Nigeria. Other studies have similarly reported S. Kentucky across Nigeria, thereby suggesting this serovar as widely circulating in Nigeria (13,22,23,21). S. Kentucky has a worldwide distribution and was previously thought to be endemic in Africa with public health significance (24,25).
Notably in this study, Salmonella serovars commonly associated with foodborne infections (S. Enteritidis and S. Typhimurium) (26) were absent. It is possible that S. Gallinarum vaccine commonly used in Nigeria protects against other group D-strains such as S. Enteritidis (27,15). In addition, Fagbamila et al., (2017) and Useh et al., (2016) have suggested that these two serovars likely play minor roles in the Nigerian poultry sector. Put together, serovar diversity may be attributable to a number of reasons but not limited to poor sanitary and biosecurity conditions, indiscriminate importation of poultry chickens and eggs without adequate screening for Salmonella and lack of focused national Salmonella surveillance and control program.
The abuse and misuse of antimicrobial agents in the poultry sector has been linked to increased resistance to antimicrobials (28). In this study, antimicrobial resistance to Salmonella was highest in sulfamethoxazole (SMX), followed by Ciprofloxacin (CIP) and Tetracycline (TET) respectively. A previous study on NT Salmonella occurrence in freshly dressed poultry meat in northern Nigeria reported high Salmonella resistance pattern to Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 5 May 2021 doi:10.20944/preprints202105.0069.v1 SMX, CIP and TET (16). This emerging resistance pattern is corroborated by studies on veterinary students' ranked perception of abused antimicrobials in Africa in which sulphonamides and tetracycline were in the uppermost three ranked antimicrobials (29,30). However, in contrast to our study, earlier investigations in Zimbabwe by Makaya et al. (31) and Adesiyun et al. (32) in the Caribbean region reported no resistance to SXT. It is then possible that increased use and misuse of SXT in the Nigeria poultry sector may be a driving factor in the resistance observed in SXT. In addition, resistance to ciprofloxacin raises concern since this is the drug of choice in the treatment of human invasive salmonellosis. Resistance to tetracycline is not surprising considering it extensive prophylactic usage and as additives in feed and water to enhance performance in Nigeria (33). It may then be inferred that, the frequency of Salmonella spp resistance to these antimicrobials reflect their intense application in the poultry sector in Nigeria. The observed lack of stringent control on the availability and non-prescription use of antibiotics in poultry practice in the study area is concerning. In Nigeria, over the counter availability of most antimicrobials in local drug stores makes the control of antibiotic usage cumbersome (14).
Although not statistically significant, the occurrence of Salmonella in eggs from markets (20%) was higher than farms (6.7%). Contamination of eggs may occur during packing, grading, transporting and sales in the market, as multiple buyers visually inspect, touch and select eggs during sales in the study area (34). In the present study, unhygienic egg handling practices were common in all farms and markets involved. Also, all farms involved in the questionnaire survey have no egg sanitation programs in place.
In this study, data from the questionnaire indicated about 90% of the farms in the study area are accessed by wild birds and rodents. Similar to our results, a study involving three Caribbean countries also reported rodents in 90% of contaminated farms (35). Investigations in Australia have demonstrated the role of environmental vectors in the epidemiology of Salmonella in farm settlements (36,37). High and unchecked rodents population have been associated with increased Salmonella shedding in the environment (38) and are the most effective in the spread of Salmonella pathogen around farms (39). It is then imperative to initiate robust vector prevention programs in farm houses which may include secured access doors and windows, sealing of holes, repair of torn wire net and the use of baits to help control contamination of farms houses (40).
Furthermore, our results revealed certain practices which may encourage Salmonella occurrence and/or persistence in farms. Poor adherence to strict biosecurity measures on farms. The use of protective clothing as barrier to infectious agents was unpopular among majority of the farmers and may contribute to increase chances of contamination. Also, certain high-risk cross-contamination practices such as unhygienic picking of eggs with bare hands, sharing of tools with nearby farms (mostly <500m away) and sales of eggs on the farm were observed. These practices all increase the risk of contamination and transfer of pathogens (41,21,37).
The findings in this study are subject to at least two limitations. First the pooling of samples which ultimately reduced the sample size. While investigating individual egg sample will have provided a more detail data, pooled samples are considered more effective for the successful detection of Salmonella in the context of this study (42). Second, was our in ability to match individual samples collected with corresponding husbandry and biosecurity questionnaires in the data analyses. Considering that the positivity rate of Salmonella was 14%, it was considered that analyzing data as per the region will be more informative than individual sample-farm analysis. .

Study Location
The cross sectional study was carried out in selected poultry farms within the four zones in Ogun State, Nigeria. Ogun State consists of 20 Local Government areas divided into four zones namely Egba, Ijebu, Yewa and Remo. It lies between latitude 6.2oN and 7.8oN and longitude 3. square kilometers and 4,054,272 populations (43). A stratified probability random sampling design was adopted for this study such that poultry farms and markets from the four zones of Ogun state were evenly represented in the final sample.

Sample Collection
Egg samples from poultry farms and markets were used for Salmonella determination. A total of five hundred (500) eggs were collected representing 125 eggs per each of the four zones. From each zone, 75 eggs were obtained from 3 poultry farms and 50 eggs from 2 markets. Samples were analyzed in pools of 5 making a total of 25 samples units per zone (15 from farms and 10 from markets). Samples were collected into sterile bags using sterile nylon gloves and transported to the laboratory at 37oC.

Isolation of Salmonella
The egg surface was disinfected with 70% alcohol and alcohol residue removed by flaming. A sterile thumb forceps was used to aseptically separate the shell from the interior content. Five ml of homogenized egg was poured into 45 ml of buffered peptone water and incubated at 37 oC for 18 h.
The pre-enrichment broth culture was mixed and 1 ml was transferred with a sterile pipette into a tube containing sterile 9 ml of Mueller-Kaufmann Tetrathionate novobiocin (MKTTn) broth. Moreover, 0.1 ml of the pre-enrichment broth culture was transferred in drops onto 9ml of sterile Modified semi solid Rappaport Vassiliadis agar (MSRV). Inoculated MKTTn broth was incubated at 37oC for 24 h and MSRV was incubated at 41.5 ˚C for 24 h.
After incubation for 24 h, a loopful each of MRVS and MKTTn cultures was streaked onto the surface of XLD agar (Xylose lysine desoxycholate agar) and Mac-Conkey agar (MAC). The plates were incubated at 37oC for 24 h. After incubation, the plates were checked for growth of typical Salmonella colonies. Typical colonies of Salmonella grow on XLD agar with a black centre and a lightly transparent zone of reddish colour due to the colour change of indicator, while typical colonies of Salmonella grows on MAC agar as pale/colourless translucent colonies and slightly convex (ISO, 2002).

Biochemical Identification of Salmonella
Suspected colonies were subjected to catalase and oxidase tests. For further identification different biochemical tests were carried out using MICROBACT TM GNB 24E KIT (OXOID) for Gram negative bacteria and result was interpreted using the computer software package (Oxoid Microbact® 2000 version 2.03) according to (Cheesbrough, 2006).

Procedure:
Overnight culture of test isolate in peptone water broth was adjusted to an equivalent of 0.5 McFarland concentration (approximately 1x108cfu/mL). This was evenly spread on the surface of freshly prepared Mueller Hinton agar (MHA) plate. Antimicrobial discs were carefully and firmly placed at equidistance. This was incubated at 35±2˚C for 18 h. The diameter of zone of inhibition around each disc was measured and result interpreted according to CLSI, 2015. A test questionnaire was distributed among 15 farms not included in the final data collection to determine clarity of questionnaire. Based on the feedbacks from the respondents, adjustments were made in the final questionnaire.

Determination of biosecurity practices on poultry farms
Sixty (60) well-structured questionnaire were administered to farmers and farms spread across the four zones of Ogun State to evaluate the husbandry practices and adherence to biosecurity measures on farms (Appendix I). Fifteen (15) questionnaires were distributed per zone and included farms from which egg samples were previously taken for Salmonella detection. Farmers were informed of their rights to discontinue participation at any stage of the project and anonymity and confidentiality of data was stressed.

Data Management and Statistical Analysis
Data collation and management was computed using Microsoft Excel. The responses to questionnaire were presented in percentages; descriptive statistics were use to describe the prevalence analysis. The Pearson's chi square value was determined for the data generated using Statistical Package for Social Sciences (SPSS) software, version 20.

Conclusions
This study demonstrated the presence of diverse NT Salmonella serovars in eggs with potential antimicrobial resistant traits. Sales of eggs in the market seems to promote the risk of Salmonella contamination as well as other unhygienic biosecurity practices in the farm.

Funding:
No funding was received for conducting this study

Institutional Review Board Statement:
Not applicable

Data Availability Statement:
Data set used during study are available from corresponding author on reasonable request.