Prevalence of Virulence Genes Associated with Diarrheagenic Pathotypes of Escherichia coli Isolates from Water, Sediment, Fish, and Crab in Aby Lagoon, Côte d'Ivoire

This study was conducted to characterize virulence genes of Escherichia coli isolates from water, sediment, fish, and crab in Aby Lagoon. Serogrouping was performed by EPEC antisera in 113 E. coli strains. The presence of diarrhea-associated genes (eae, stx, AggR, elt, and est) was assessed by multiplex PCR using specific primers. Based on the multiplex PCR, sixty-two isolates (42 from water, 19 from sediment, and 1 from crab) were positive for virulence genes, including 34 positive for elt (ETEC), 46 positive for est (ETEC), 24 positive for both elt and est, 6 positive for stx (EHEC), 1 positive for both stx + est, and 1 positive for both stx + elt. Genes eae (EPEC) and AggR (EAEC) were not detected. Nine serogroups (O114, O127, O55, O111, O86, O119, O126, O128, and O142) were identified. This study revealed the presence of diarrheagenic and nondiarrheagenic E. coli and potential public health risks if fishery products are not appropriately cooked.


Introduction
Most Escherichia coli strains are a normal inhabitant of the intestinal tract of humans and warm-blooded animals. Despite being usually harmless, various E. coli strains have acquired genetic determinants (virulence genes) giving them the capacity to cause illness for both humans and animals. Some strains of E. coli are now seen as pathogenic species with remarkable versatility in their ability to cause disease in humans and animals [1]. E. coli is one of the most frequent causes of diarrhea in children in developing countries [2]. According to Grasso et al. [3] and Tumwine et al. [4], infectious pathotypes of E. coli are related to the lack of sanitation and personal hygiene but also the consumption of well water, river water, and other contaminated surface waters.
Among the E. coli pathogenic strains, in most developing countries, EPEC, ETEC, and EAEC are the most common cause of infectious diarrhea in young children [10,11]. Research into EPEC is intense and provides a good virulence model of other E. coli infections as well as other pathogenic bacteria [12]. According to the World Health Organization (WHO) in 1987, most EPEC strains belonged to a series of O antigenic groups known as EPEC serogroups which included O26, O55, O86, O111, O114, O119, O125, O126, O127, O128, O142, and O158 [13]. Serogrouping of E. coli based on somatic O antigen used for differentiating diarrheagenic E. coli is costly and time-consuming and poorly correlates with the presence of virulence factors. So, in the last decade, modern molecular detection methods were reported in the literature for rapid identification of E. coli pathotypes including PCR and multiplex PCR. In spite of increasing evidence that E. coli strains originating from human and animal feces contain several virulence genes, only a few studies have investigated the presence of E. coli pathotype in environmental waters [14][15][16][17][18]. The presence of E. coli strains with virulence genes profiles similar to EHEC, EPEC, and ETEC in environmental waters has already been reported. To the best of our knowledge, no investigation on E. coli pathotypes distribution has been carried out on the estuarine water environments of Côte d'Ivoire. Yet, these environments that receive frequently domestic wastewater and mammalian feces provide important fishery resources. The production of fish and shellfish is estimated, respectively, to be 6.000 and 7.000 tons per year in the Aby Lagoon [19]. Contamination of lagoon waters by E. coli pathotypes could have a negative impact on fish, crabs, and other animals in this environment. Thus, a potential public health risk exists if these fishery products were contaminated by these pathotypes on the one hand and on the other hand if the hygiene measures are faulty during cooking. According to Rangel et al. [20], exposure to recreational waters has been linked to high numbers (21 out of 31) of reported E. coli O157:H7 disease outbreaks in the United States from 1982 to 2002. In addition, direct ingestion or aerosols of contaminated water during spray irrigation and contaminated vegetable could cause infection.
The aim of this study was to use PCR method to detect four pathotypes of E. coli (ETEC, EPEC, EAEC, and EHEC) from water, sediment, fish, and crab samples. During the study, both PCR and culture-based methods were used.

Sampling Sites.
The Aby Lagoon is located between 2 ∘ 51 and 3 ∘ 21 eastern longitude and 5 ∘ 05 and 5 ∘ 22 northern latitude southeast. The two main tributaries (Bia and Tanoe) are escape routes from anthropogenic and mining operations within Aby Lagoon's watershed in Côte d'Ivoire and Ghana ( Figure 1). Six sampling stations spread throughout the Aby Lagoon were selected in view of the fact that these stations were subject to various discharges (wastewater, excreta). Station 1 is located near an urban area. Swimming and fishing are practiced here. Station 2 located at the mouth of the river Bia is a fishing zone. Stations 3 and 6, located, respectively, near the latrine on the pile of the Aby and Assomlan villages, are sites where recreational activities are constantly practiced. Stations 4 and 5 are fishing zones.

Sampling.
Six campaigns were carried out from June 2010 to March 2011 for the collection of water, sediment, fish, and Table 1: Primers used for PCR in this study [23].

Genes Sequence (5 to 3 ) Size (bp) References
Eae F CCC GAA TTC GGC ACA AGC ATA AGC R CCC GGA TCC GTC TCG CCA GTA TTC G 881 [24] Stx F GAG CGA AAT AAT TTATAT GTG R TGA TGA TGG CAA TTC AGT AT 518 [25] AggR F GTA TAC ACA AAA GAA GGA AGC R ACA GAA TCG TCA GCA TCA GC 254 [26] Elt F TCTCTATGTGCATACGGAGC R CCATACTGATTGCCGCAAT 322 [27] Est F TTAATAGCACCCGGTACAAGCAGG R CCTGACTCTTCAAAAGAGAAAATTAC 147 [28] crab samples. These six campaigns are distributed as follows: two campaigns for the rainy season (June-July), two for the flood season (September-October), and two for the dry season (February-March

Results
Sixty-two strains (55%) of the 113 strains tested were positive for virulence genes. Pathogenic strains of E. coli were more isolated in the sediment with a frequency of 70% of the cases, followed by the strains from water (68%). Virulence strains were least observed with the crabs (9%). No pathogenic strain of E. coli was detected in fish samples ( Table 2).   The four pathotypes of E. coli in this study according to the nature of the samples analyzed are shown in Table 3. Two E. coli pathotypes were identified, namely, enterotoxigenic E. coli (ETEC) with a percentage of 90% and enterohemorrhagic E. coli (EHEC) with a prevalence of 10%. These two pathotypes were observed in the samples of water, sediments, and crab. In water samples, 8% and 60% of the pathogenic strains belonged to EHEC and ETEC, respectively. For sediment samples, 2% of the cases of the virulent strains belonged to EHEC and 29% to ETEC. No strains of enteropathogenic E. coli (EPEC) and enteroaggregative E. coli (EAEC) were identified. The only pathogenic strain identified in the crab samples belonged to ETEC. Table 4 shows the prevalence of virulence genes according to the nature of the samples examined. The genes belonging to ETEC were the most detected with a frequency of 74% and 55% of the cases for the genes "est" and "elt," respectively. These genes were identified in strains isolated from water, sediment, and crabs with the most important prevalence from the water samples (32% for "elt" gene and 50% for "est" gene). The ETEC strains harboring "est" gene were the most identified (74%). A prevalence of 35% of these strains possessed both the heat-labile toxin gene (elt) and the heatstable toxin gene (est). About 10% of enterohemorrhagic E. coli (EHEC) harbored "stx" gene. The simultaneous presence of genes stx + est and elt + stx was also identified in some strains with a prevalence of 2% for each combination. Figure 2 shows the PCR amplification products of the target genes studies.
The various serogroups of potential pathogenic E. coli according to the nature of samples are shown in Table 5. The results of the serogrouping by antisera showed that 37% of the 62 pathogenic E. coli isolates were typeable with the used antisera. Nine serogroups, namely, O114 (14%), O127 (6%), O55 (5%), and 2% for O111, O86, O119, O126, O128, and O142, were identified. The O114 serogroup was the most detected. Different serogroups identified are not specific to each group of pathotype (Table 6).

Discussion
Results of the prevalence of potential pathogenic E. coli strains found in water (68%) and sediment (70%) samples were similar to those reported by Obi et al. [14] from water and sediment of six rivers in South Africa. These results could be explained by the fact that this lagoon received all effluents. Indeed, several effluents are released often without any treatment in the lagoon. Kambiré et al. [29] showed that the Aby Lagoon was influenced by continental waters. In addition, these authors indicated that most of the household members (93%) living in places without latrines defecated directly into the lagoon. The prevalence of nonpathogenic E. coli was 45%. According to Bekal et al. [30], Escherichia coli is a normal inhabitant of the intestinal tract of humans and warmblooded animals. Despite being usually harmless, various International Journal of Microbiology 5 E. coli strains have acquired genetic determinants (virulence genes) rendering them pathogenic for both humans and animals.
The pathogenic E. coli strains found in this study belong to two different pathotypes: ETEC and EHEC. ETEC (90%) represents the most frequent pathotype. This result is similar to those reported by Salem et al. [6]. ETEC was identified as the common cause of infections among tourists visiting Asia, Africa, and South America and also as a common diarrheal pathogen in children in many developing countries of Asia, Africa, and South America [31,32].
The prevalence of heat-stable toxin gene (est) of ETEC was 74% of the strains tested compared to the heat-labile toxin gene (lt), 55%. Other studies showed predominance of "est" gene [33,34]. Several authors have also reported the simultaneous presence of the genes est and lt in ETEC [32,35,36] like in this study. According to Munshi et al. [37], the genes encoding LT (elt or etx) reside on plasmids that also may contain genes (est) encoding ST.
The prevalence of EHEC pathotype was 10%. This frequency is lower than that obtained by Ndlovu et al. [15] which was 15% in their study on the characterization of E. coli isolated from surface water sources. However, frequency in this study is higher than that obtained by Obi et al. [14] which was 2% in South African rivers. Our prevalence is approximately similar to those reported by Dadié et al. [38] in 1780 samples of food (meat and dairy products) and 1416 patients in Côte d'Ivoire. One isolate harbored the combination of stx and elt genes and another stx and est genes. An association gene was also observed by Moalic and Guennec [39] from E. coli strain causing diarrhea in pigs in France. According to Titilawo et al. [16], the lower prevalence of the EHEC pathotype compared to other pathotypes suggests that human fecal contamination is the main source of diarrheagenic E. coli pathotypes in the surface water as opposed to contamination from animals. Contrary to the studies of Sidhu et al. [40] and Titilawo et al. [16] in the characterization of E. coli from surface water and rivers in Southwestern Nigeria, respectively, genes for EPEC (eae) and EAEC (AggR) were not detected in this study.
Phenotype assays such as serogrouping with traditional antisera are the routine methods that have been widely used in clinical laboratories [41]. Serogrouping has been shown to be insufficient for the identification of a particular pathotype group. The 12 antisera specific for EPEC group according to the WHO are permitted to detect other pathotype E. coli groups like ETEC and EHEC in this study. Nine serogroups were identified in this study. Among the identified serogroups, the O114 serogroup was the most isolated. This serogroup has been the cause of an epidemic of infantile gastroenteritis in England [42].

Conclusion
This study shows the presence of pathotypes of E. coli in water, sediment, and crab. The pathogenic E. coli belongs to two different pathotypes: ETEC and EHEC. ETEC represented the most frequent pathotype. Nonpathogenic strains of E. coli were also identified in all samples analyzed, especially in fish samples. Nine serogroups have been identified with O114 as majority group. This study shows the importance of controlling sources of human fecal pollution, such as municipal wastewater management, to reduce potential risks to human health. In this sense, all latrines built on pile should be suppressed. The domestic water must also be treated before being discharged into the lagoon.

6
International Journal of Microbiology

Conflicts of Interest
The authors declare that there are no conflicts of interest regarding the publication of this paper.