Occurrence and Enumeration of Multiple Bacterial Pathogens in Edible Snails from South East Nigeria

Edible snails are usually obtained from the forest and in high demand among consumers. Data on the level of contamination of edible snails with bacterial pathogens are needed for making legislations that will improve food safety and protect public health. This study aimed to determine the occurrence and distribution of counts of selected bacterial pathogens in Achatina achatina from major markets within South East Nigeria. A total of 300 samples of A. achatina were examined for occurrence and counts of Citrobacter, Shigella, Escherichia coli, Staphylococcus, Aeromonas and Bacillus cereus using enrichment broth, differential and selective media. Snail samples from Ogbete market had the highest mean aerobic plate count (9.32 ± 0.308 Log CFU/g), while Abakaliki market samples had highest mean count of coliforms (7.63 ± 0.389 Log CFU/g). Among pathogens, highest counts were observed for Citrobacter and E. coli which ranged from 6.0 to 8.0 Log CFU/g in 300 (100%) and 180 (60%) samples respectively. Significant differences were observed among the locations (p < 0.01). Our findings highlight the need for formulation and implementation of strategies for the reduction of bacterial pathogens in edible snails along the value chain.


Introduction
Foodborne diseases cause significant burden of disability and mortality in most countries. Diarrhoeal diseases are the most common illnesses resulting from the consumption of contaminated foods, causing 550 million people to fall ill and 230,000 deaths annually (WHO, 2017). Currently, Nigeria's diarrhoea prevalence is 18.8% according to the joint report from the federal ministries of Agriculture, Environment and Health (FMAEH, 2017).
Food poisoning and diarrhoea caused by foods contaminated by Citrobacter have been reported (Doulgeraki et al., 2011). Currently, the incidence of shigellosis worldwide is highest among children less than five years of age (Taneja and Mewara, 2016). E. coli is one of the major foodborne pathogens of foods of animal origin with wide variability of virulence (Kobayashi et al., 2002;Johnson et al., 2005). Aeromonas have been associated with several food-borne outbreaks and are progressively being isolated from patients with traveler's diarrhoea (Von Graevenitz, 2007). The true burden of illnesses caused by B. cereus is unknown probably because they commonly occur as sporadic cases, rather than in major outbreaks (Logan et al., 2011). All food groups contribute to the burden of foodborne diseases, and foods of animal origin cause the highest burden (Havelaar, 2016). Carriage and shedding of zoonotic pathogens contaminate the environment and eventually enters the food chain during processing and post processing procedures (Coker et al., 2000). However, most people are inadvertently exposed to microbial hazards from several other sources (Okafor et al., 2017;Amini et al., 2012), which can cause diseases that go unreported. The proportion of the national burden of disease linked to the environment in Nigeria is 29% (WHO, 2009). Snails are consumed by the people of both rural and urban communities (Efuntoye et al., 2011). The most common species in West Africa is Achatina achatina (Hodasi, 1984). Consumption level of snail meat among people of Bori (a southern city in Nigeria) is as high as 70% (Nodu et al., 2003). The global snail market has recorded a turnover of 10 billion Euros per year with consumption of about 400,000 tons of snails, of which only 50,000 tons were produced in snail farms (Toader, 2012).
However, mild gastroenteritis has been reported to be common among people that consume snails regularly (Serrano et al., 2004). Bacterial pathogens have been detected from varieties of snails in previous studies (Adegoke et al., 2010;Omenewa et al., 2011, Adagbada et al., 2011Ebenso et al., 2012;Nyoagbe et al., 2016). However, there is limited data on the comparative 24 Occurrence and Enumeration of Multiple Bacterial Pathogens in Edible Snails from South East Nigeria distribution of viable counts of multiple bacterial pathogens among snails displayed for sale in markets in more than two states in Nigeria. Such data is important for making legislations that will improve food safety.
The objective of this study was to determine the occurrence and distribution of the viable counts of Citrobacter, Shigella, E. coli, Staphylococcus, Aeromonas and Bacillus cereus in A. achatina from major markets in three states within South east geopolitical zone of Nigeria.

Collection of Snail Samples
A total of 300 samples of live edible snails (Achatina achatina) were randomly collected from markets in south east, Nigeria and analysed: comprising of hundred samples each from three states in south east geopolitical zone of Nigeria, namely Anambra, Ebonyi and Enugu. Central markets serving as the largest platforms for sale of live edible snails in these states were selected for this study. They were: Ogbete main market at Enugu State, Abakaliki meat market at Ebonyi State and Nkwo Igboukwu market at Anambra State. Samples were collected from July 2016-December 2016 and April 2017-June 2017.
Edible Snails (A. achatina) were identified according to their shape, size, markings, colour, spire angle, sculpture and aperture form (Igbinosa et al., 2016;Raut and Barker, 2002). Edible snails displayed on the tables for sale were aseptically collected in plastic containers sterilized with 70% alcohol and dried with commercially available sterile paper towel. Samples were quickly transported to the laboratory for analysis.

Sample Preparation
The shells of the snails were surface sterilized with 70% ethanol before being aseptically shucked with a sterile iron rod to extract the meat. The lab blender was sterilized with 70% ethanol. The sample (50 g) was homogenized in 450 ml of Ringers solution (Oxoid) using the lab blender for 2 mins at medium speed. The homogenate was used for serial dilution (1:10). Aliquot (1 ml) of appropriate dilution factor was used for determination of bacterial counts

Determination of Total Aerobic Plate Count
Plate count agar (Oxoid) was prepared according to manufacturer's instructions and maintained at 45 0 C. Aliquot (1 ml) of appropriate dilution factor (10 -6 -10 -8 ) of the homogenate was pipetted into sterile petri dish and the molten agar media was poured into the petri dish. The plate was swirled to mix the homogenate with the agar media. It was done in triplicates for each sample. Plates were incubated aerobically at 37 0 C for 24 hours after which colonies were counted and recorded.

Determination of Coliform Count
MacConkey agar (Titan) was prepared according to manufacturer's instructions. Aliquot (0.1 ml) of appropriate dilution factor was plated out on the agar medium. It was done in triplicates for each sample. Plates were incubated aerobically at 37 0 C for 24 hours after which pink colonies were counted and recorded.

Determination of Viable Counts of Bacterial Pathogens
All agar media namely: Salmonella-Shigella agar (Biotech), Eosin Methylene Blue agar (Oxoid), Mannitol Salt agar (Oxoid), Thiosulfate Citrate Bile salt Sucrose agar (Titan) and Brain Heart Infusion agar (Titan) were prepared according to manufacturer's instructions.
Aliquot (0.1 ml) of appropriate dilution factors was directly plated out on appropriate agar media, except Brain Heart Infusion agar, specific for each pathogen and incubated aerobically at 37 0 C for 24 hours. It was done in triplicates for each sample. Typical colonies were counted and recorded after 24 hours. However, appropriate dilution factors were heated in a water bath at 80 0 C for 10 mins before being plated on Brain Heart Infusion agar. It was done in triplicates for each sample. Plates were incubated aerobically at 37 0 C for 24 hours after which typical colonies of Bacillus cereus were counted and recorded.

Isolation and Identification of Selected Bacterial Pathogens
The procedure for identification of the six genera of bacterial pathogens was based on the United Kingdom Standards for Microbiology Investigations as published by Public Health England (2014 and and is discussed in the following sections.

Isolation and Identification of Citrobacter
The procedure used was tailored towards isolation of Salmonella. Briefly, aliquot (5 ml) of the homogenate was enriched in Selenite fluid (Tulip) (45 ml) for 24 hours, after which a loopful was streaked on Salmonella-Shigella agar (Biotech) and aerobically incubated at 37 0 C for 24 hours. Presumptive colonies (white colonies with black centres) were subcultured in Tryptose Soya agar (Oxoid) and subjected to further tests such as Gram staining, catalase test, motility test, oxidase test, indole test, urease test and triple sugar iron test. Representative isolates were forwarded to International Institute of Tropical Agriculture, Ibadan for confirmation of identity using 16S rRNA gene sequencing technique.

Isolation and Identification of Shigella
Presumptive colonies (white colonies without black centres) on Salmonella-shigella agar (Biotech) were subcultured in Tryptose Soya agar (Oxoid) and subjected to further tests such as Gram staining, catalase test, motility test, oxidase test, urease test and glucose fermentation test.

Isolation and Identification of Escherichia coli
Aliquot (5 ml) of the homogenate was enriched in lactose bile broth (20 ml) for 18 hours, after which a loopful was streaked on Eosin Methylene Blue (EMB) agar (Oxoid) and aerobically incubated at 37 0 C for 24 hours. Presumptive colonies (blue-black colonies with green metallic sheen and dark centres) on EMB agar were streaked on sorbitol MacConkey agar (Titan) and subcultured in Tryptose Soya agar (Oxoid) and subjected to further tests such as Gram staining, catalase test, indole test, urease test and citrate test, haemolysis test.

Isolation and Identification of Staphylococcus
Aliquot (5 ml) of the homogenate was enriched in Nutrient broth (Oxoid) containing 3% NaCl (20 ml) for 24 hours, after which a loopful was streaked on Mannitol Salt agar (Oxoid) and aerobically incubated at 37 0 C for 24 hours. Presumptive colonies (yellow colonies) were subcultured in Tryptose Soya agar (Oxoid) and subjected to further tests such as Gram staining, catalase test, coagulase test and haemolysis test.

Isolation and Identification of Aeromonas
Briefly, aliquot (5 ml) of the homogenate was enriched in Nutrient broth (Oxoid) containing 3% NaCl (20 ml) for 24 hours, after which a loopful was streaked on TCBS agar and aerobically incubated at 37 0 C for 24 hours. Presumptive colonies (yellow colonies) on Thiosulfate Citrate Bile salt Sucrose agar (TCBS) agar (Titan) were subcultured in Tryptose Soya agar (Oxoid) and subjected to further tests such as Gram staining, oxidase test, motility test, haemolysis test, lecithinase test and gelatinase test.

Isolation and Identification of Bacillus cereus
Aliquot (5 ml) of the homogenate was heated in a water bath at 80 0 C for 10 mins and enriched in Brain Heart Infusion broth (Titan) (20 ml) for 24 hours, after which a loopful was streaked on agar and aerobically incubated at 37 0 C for 24 hours. Presumptive colonies (raised grey colonies) on Brain Heart Infusion agar (Titan) were subcultured in Tryptose Soya agar and subjected to further tests such as Gram staining, spore staining, motility test, haemolysis test, lecithinase test and gelatinase test.

Data Analysis
Descriptive statistics such as means and frequencies were used to present some of the findings. All data on plate counts were converted to logarithmic value. Analysis of variance (ANOVA) was performed using statistical software available in Vassarstat website.

Results
The Mean bacterial loads in 300 snails analysed in this study are presented in Table 1. The mean aerobic plate count of samples ranged from 8.43 -9.61 Log CFU/g. Samples from Ogbete market had the highest mean total aerobic plate count (9.32 ± 0.308 Log CFU/g) while the lowest mean count was found in Igboukwu samples (8.74 ± 0.312 Log CFU/g). There were significant differences between total aerobic plate counts of all groups of samples analysed (p < 0.01). Samples from Abakaliki market had the highest mean count of coliforms (7.63 ± 0.389 Log CFU/g) while Igboukwu samples (7.41 ± 0.191 Log CFU/g) had the least counts and there were no significant differences between Igboukwu and Ogbete samples. The highest mean Citrobacter counts was found in Abakaliki samples (7.24 ± 0.210 Log CFU/g) and there were no significant differences between Igboukwu and Ogbete samples (p < 0.01). Abakaliki samples were found to contain the highest mean Shigella counts (4.61 ± 0.354 Log CFU/g). There were significant differences between Shigella counts of all groups of samples analysed (p < 0.01). Igboukwu samples contained the highest mean counts of E. coli (7.14 ± 0.170 Log CFU/g) and there were significant differences between all groups of samples analysed (p < 0.01). Staphylococcus was not detected in Igboukwu samples. The highest mean count of Aeromonas was found in Ogbete samples (4.80 ± 0.473 Log CFU/g). Abakaliki samples had the highest mean counts of Bacillus cereus (4.50 ± 0.136 Log CFU/g).
Samples were found to contain different levels of bacterial loads ( Table 2). All samples had total aerobic plate counts >10 8 CFU/g. Most samples (86.7%) had coliform counts ranging from >10 6 -10 8 CFU/g. While Citrobacter counts ranged from >10 6 -10 8 CFU/g in all samples, Shigella counts were <10 4 CFU/g in 35% of the samples. E. coli counts were >10 6 -10 8 CFU/g in 60% of the samples. The staphylococci counts were >10 4 -10 6 CFU/g in 26.7% of the samples. While Aeromonas counts were <10 4 CFU/g in 43.3%, Bacillus cereus counts were <10 4 CFU/g in 60% of the samples.   All 300 samples of market snails analysed in this study were found to be contaminated with pathogens irrespective of the source of the samples (Table 3). Citrobacter was detected in all samples across the three sources. Shigella was recovered from 60% of all samples analysed, with the highest prevalence in Ogbete samples (80%). E. coli, Staphylococcus, Aeromonas, and B. cereus were recovered from 90%, 36.7%, 76.6% and 80% of all samples respectively.

Discussion
A. achatina was the focus of this study because of its conspicuous presence in the southern part of Nigeria. It is consumed by the people of both rural and urban communities (Okafor, 1989;Efuntoye et al., 2011). Land snails aestivate from December of a year to March of the next year, and re-surface during the rainy season (Fagbuaro et al., 2006) which explains why this study was conducted during the rainy season.
The findings of this study demonstrate that most snails sold in markets in Anambra, Ebonyi and Enugu states contain various levels of high loads of bacterial indicators and pathogens. The mean aerobic plate count of samples analysed in this study ranged from 8.43 -9.61 Log CFU/g. These data appear close to the findings of other related studies: Adegoke et al. (2010) reported total aerobic bacterial count in market snails at Akwa Ibom state was 8.0 Log CFU/g. In Ghana, Nyoagbe et al. (2016) reported that total viable count ranged from 6.61 to 8.29 Log CFU/g. However, Temelli et al. (2006) found the average total aerobic bacterial count in live snails in Turkey to be 6.85 Log CFU/g. Also, mean aerobic counts varied significantly (p < 0.01) between the three states from which samples were collected, probably because of the difference in the nature of soil and debris present in the natural habitats of these snails across these locations.
Aerobic plate count is generally used as a means of assessing the overall microbial quality of raw ingredients (Siragusa et al., 1998). According to ICMSF (1986), the acceptable upper limit of total aerobic bacterial load for seafoods is 5.0 Log CFU/g and this limit has been cited in most research articles till date. It is important to note that all snail samples analysed in this study had total aerobic plate counts >10 8 CFU/g. This implies that 100% of market snails analysed pose microbiological risk to handlers and consumers. However, the use of the aerobic plate count as an indicator for the presence of specific pathogens is generally not satisfactory (Miskimin et al., 1976;Siragusa et al., 1998).
Coliform counts are used for assessing the amount of contamination on meat arising from gut contents and are the most frequently studied indicators (Wu et al., 2011). The acceptable upper limit of total coliform is 2.0 Log CFU/g (ICMSF, 1986). In this study, the coliform counts were >2.0 Log CFU/g in all samples analysed. Similar coliform count in snails has been previously reported (Adegoke et al., 2010;Nyoagbe et al., 2016), even though the prevalence of levels of concentrations of coliforms in snails has not been reported in any study. It is appropriate to note that snails discharge their faeces within their habitat (Ibom et al., 2012) and may explain the high loads of coliforms observed in this study.
Citrobacter was detected in all samples across the three sources. This observation is supported by another study in India on the bacterial diversity of the gastrointestinal tract of A. fulica using culture-independent and culture-dependent methods. The study also concluded that an apparent feature of bacterial communities in snails' gastrointestinal tract was the abundance of members of the genus Citrobacter (Pawar et al., 2012). The highest mean Citrobacter count was found in Abakaliki samples (7.24 ± 0.210 Log CFU/g). Citrobacter is classically considered a resident commensal of the intestinal tracts of both humans and animals (Guerrant et al., 1976). It is also prevalent in soil and water through contamination from the waste materials of animals. A study concluded that healthy pet turtles are a potential carrier of C. freundii (Sabrina-Hossain et al., 2017). Therefore, animals are the probable source of Citrobacter around swampy environments where most snails are collected.
Mean Shigella counts exceeded 4.00 Log CFU/g in all samples examined. Most studies on snails have reported the presence of Shigella without indicating its level of concentration (Adagbada et al., 2011). Several aquatic bodies have been found to contain Shigella and aquatic foods may play a role in its transmission if such foods are harvested from sewage-contaminated water (Iwamoto et al., 2010). The number of Shigella cells required to initiate infection ranges from 10 1 -10 4 cells/person (Dupont et al., 1989;Heymann, 2004). Since 60% of snails in our study exceeded 10 CFU/g, it implies such percentage of snails represent health threat to handlers and consumers, especially children less than five years of age (Taneja and Mewara, 2016).
For E. coli, the highest mean counts were found in Igboukwu samples (7.14 ± 0.170 Log CFU/g), while Ogbete samples had the lowest mean counts (5.65 ± 0.239 Log CFU/g). Sixty percent of the snails in this study had E. coli counts > 6.0 Log CFU/g which is within the range of counts prominent for resulting in diarrhoeal diseases 6.0 -9.0 Log CFU/g (Kornacki and Marth, 1982).
The mean staphylococcal count in samples analysed in this study ranged from 4.66 -4.74 Log CFU/g. The only study that quantified the level of Staphylococcus in snails reported a range between 2.66 and 7.68 Log CFU/g (Nyoagbe et al., 2016). Diagnosis of staphylococcal food poisoning is generally confirmed by the recovery of at least 5.0 Log CFU/g from food (Halpin-Dohnalek and Marth, 1989;Hennekinne et al., 2012). It is suggested that since Staphylococcus is also present in intestinal tract, meat may contain Staphylococcus resulting from contamination with intestinal content during evisceration (Bhalla et al., 2007). It is not clear why Staphylococcus was not detected in Igboukwu samples, but may have been present at very low counts.
The mean count of Aeromonas was found to range from 3.10 -4.80 Log CFU/g. The infectious dose of Aeromonas species in foods is not known (Isonhood and Drake, 2012). In another study, mesophilic aeromonads were isolated from 26% of vegetable samples, 70% of meat and poultry samples, and from 72% of fish and shrimps. Numbers of motile aeromonads present in these samples varied from < 2.0 to > 5.0 Log CFU/g (Neyts et al., 2000). While in our present study, 76.6% of the snail samples contained Aeromonas. This is important because snails feed on assortment of plant and animal species including algae (Okafor, 1989). Formulated feeds for snails are not available in the market. Therefore, it has become common practice for snail rearers to use vegetables, plant leaves and kitchen wastes to feed snails (Chah and Inegbedion, 2013). Aeromonas species are widely distributed in the aquatic environment (Palumbo, 1996, Neyts et al., 2000 and their prevalence in various water and food sources represents a significant public health threat (Wu et al., 2007). Edible snails sold in the markets are usually obtained from the forest and are in high demand among consumers (Nyoagbe et al., 2016). Also, they are usually purchased alive in the market by consumers and brought into homes where they are handled and prepared in the domestic kitchens.
The results of this study indicate that Bacillus counts for  (Harmon et al., 1992).

Conclusions
Snails displayed for sale in markets at Anambra, Ebonyi and Enugu states contain high loads of bacterial indicators and pathogens at high prevalence rate. Among the pathogens studied, highest counts were observed for Citrobacter and E. coli in samples. Edible snails may play a role in the transmission of foodborne bacterial pathogens in the food chain. Sources of edible snails should be monitored and protected from routes of bacterial contamination. There is need to collaborate with environmental professionals. Our findings highlight the need for formulation and implementation of strategies for the reduction of bacterial pathogens in edible snails along the value chain.