First description of class 1 integron in typhoidal Salmonella isolated from clinical and lettuce samples in Burkina Faso.

Background: In Burkina Faso, dirty water, in particular those of the stoppings and the gutter are used for irrigation of vegetables. The aim of this study is to contribute to the knowledge on the molecular level of Salmonella circulating in the hospitals and environment next to hospitals in Burkina Faso. Methods: Salmonella isolated from diarrhea stools between 2009 to 2015 and lettuce samples isolated in 2014 in Burkina Faso were characterized by simple Polymerase Chain Reaction using specific primers. Sequencing was performed using the BigDye terminator protocol on the ABI PRISM 310. The sequences obtained corresponding were searched using the BlastN program. Results: Out of 100 Salmonella isolated, 53% were from human and 47% from lettuce samples. Globally, the highest prevalence was observed with inv A, mis L, pip D, orf L and spv R genes in 97%, 96%; 74%; and 21%. Forty of these isolates carried class 1 integron, 31 from clinical samples and 9 from lettuce samples. Sequencing showed different gene with aadA1 in 13/15 strains, aadA7 and aac(3)-Id in 2/15 strains. Eight percent (8/100) of Salmonella harbored gyrB and parE genes. Sequencing showed no mutation in these genes. Three distinct Pulsed-Field Gel Electrophoresis (PFGE) types were observed from clinical samples with 90-95% similarity in each case. All Salmonella from lettuce had similar pulsotypes. Conclusion: This study showed that lettuce is a potential source of transmission of Salmonella causing diarrhea among human in Burkina Faso.


Introduction
Salmonella entrica are common causes of human foodborne outbreaks and diseases in worldwide [1].
Their transmission to human is principally by ingesting contaminated water or food [2].
In Burkina Faso, rains shortage leads to the practice of farming irrigated by dam or untreated wastewater. Wastewater of the stoppings and the gutter are often used for irrigation of vegetables could contain pathogenic germs, such as Salmonella. Consumption of vegetables contaminated results in diarrheal diseases. To treat these infections in developing countries, fluoroquinolones cephalosporins have been considered as first-line drugs [2]. In recent years, with multidrug resistance, another trend has arisen: the emergence of virulence-resistance plasmids; these are hybrid plasmids that harbor resistance and virulence determinants [3]. The appearance of these plasmids is of concern because they could lead to the co-selection of virulence through the use of antimicrobial drugs [4]. It could result not only an increase virulence and resistance genes in some strains but also promote the transfer of genes virulence and resistance between strains of the same species or different species. This study contributes to the knowledge at the molecular level of epidemiology, virulence and resistance genes of Salmonella circulating in the hospital and environmental circles in BF.
The aims of this study were (i) to characterize the virulence and resistance genes of Salmonella Typhi and Paratyphi isolated from stool samples and lettuce in BF, (ii) to determine conjugation transfer of resistance genes and (iii) the molecular strain typing. -Gourcy (District Sanitaire de Gourcy (DSG)) and -Boromo (District Sanitaire de Boromo (DSB)), which are rural areas located in northern and western parts of the country.

Collection of bacterial strains
All Salmonella strains were stored at -30°C for molecular biology analysis to Institut Pasteur of Dakar (IPD), Senegal.Antimicrobial susceptibility testAll strains were subjected to a set of 14 antibiotics to study their antibiogram by using Kirby-Bauer disk diffusion method.

DNA extraction
DNA extraction was done as described by Madadgar et al. (2008) [5] PCR amplification and sequencing Virulence genes detection: five genes (invA,spvR,orfL, pipD and misL) associated with the virulence of Salmonella was determined. PCRs were carried out with theirs specific primers.
Sequencing: It was performed using the BigDye terminator protocol on the ABI PRISM 310. The sequences obtained corresponding were searched using the BlastN program available in http://www.nbci.nlm.nih.gov.

Bacterial conjugation
A total 21 Salmonella Typhi and Paratyphi A, B, C (18 for clinical and 3 for lettuce samples) carrying class 1 integrons and resistant to nalidixic acid were used to bacterial conjugation. The recipients (E. coli NalR) were selected on Luria-Bertani (LB) medium containing streptomycin (25μg/ml), nalidixic acid (50μg/ml), trimethoprim (5μg/ml) and ampicillin (100μg/ml). The donors were selected on LB containing only streptomycin and trimethoprim (BioMérieux, France). A total of 5 ml of fresh LB broth was inoculated with either recipient or donor bacteria and incubated for 24 h at 37°C. After, cultures were diluted at 1:50 and incubated at 37° C with strong agitation for 4 hours. Donor and recipient strains were then mixed at the following ratios 1:1; 1:2; 1:10 and incubated at 37°C for 3 hours.

Pulsed-Field Gel Electrophoresis (PFGE)
A total 29 Salmonella Paratyphi B (21 from clinical and 8 from lettuce) were typed. DNA from all Salmonella isolated was prepared in agarose plugs [7]. Digested DNA from S. Braenderup H9812 was loaded every five lanes as the molecular marker, as recommended by PulseNet [8].
Antimicrobial susceptibility testAmong 100 isolates of S. Typhi and Paratyphi collected from human and lettuce samples, 40% were resistant at least one antibiotic. Detection of resistance genes Forty percent of these isolates carried class 1 integron, with 31 (58.5%) from clinical and 9 (19.15%) from lettuce samples (figure 1). Out of these 40 isolates, 15 had cassettes genes.
PFGE The genetic relatedness among the S. Paratyphi B strains was investigated by XbaImacrorestriction analysis. Three distinct PFGE types were observed in each case from clinical samples with 90-95% similarity. All pulsotypes from lettuce sample were similar to 95%. Eighty to ninety (80-90%) similarity was observed with clusters from clinical and lettuce samples.

Discussion
The genome of Salmonella enterica possesses multiple pathogenicity islands (PIs), which are genetic elements within the bacterial genome that harbor genes associated with virulence [9]. For our experiments, five virulence genes for PCR amplification from the Salmonella serovar were selected.
The invA genes contain sequences that are unique to Salmonella spp. and have been shown to be suitable for specific targets in different diagnostic and research laboratories [10,11]. Several studies in Africa have shown similar results [12]. In South Africa, Afema et al. (2016) found similar results from treated wastewater used for the irrigation of vegetables.
The misL gene encodes an auto-transporter protein involved in intestinal colonization and essential for survival in macrophages [13]. The majority of Salmonella isolates in this study carried this gene.
Similar results were found in most studies in South Africa and Colombia [14,15].
Recent studies have shown that caspase-1 apparently has a protective role for the host during systemic Salmonella infection, suggesting that caspase-1 activation by Salmonella would be detrimental to the organism in disseminated disease [16,17].
The spvR locus is strongly associated with strains that cause non-typhoid bacteremia, but are not present in typhoid strains [13]. SpvR is a positive regulator gene of four effector genes, spvA, spvB, spvC and spvD. These virulence genes are found in the most frequently isolated non-typhoid serotypes of S. enterica. Indeed only Salmonella Typhimirium and Enteritidis could contain the spv plasmids of virulence, which explains why bacteria carrying this plasmids can't cause gastroenteritis in people [13]. However, this study showed spvR in 61.9% of Salmonella Typhi and Paratyphi isolates from human diarrheal samples. The presence of these virulence genes in Salmonella Typhi and Paratyphi isolated from lettuce and clinical samples indicate the capabilities of these isolates in causing infections in susceptible hosts.
Resistances related to class 1 integrons were found in 40%. Class 2 and 3 integrons were absents.
This predominance of class 1 integrons in Salmonella was previously described by others authors trimethoprim/sulfamexazole is strongly associated with the presence of class 1 integrons, due to the frequent presence of a dfrA1 cassette and sul1 gene in the 3' region [26].
Integrons are potentially capable to transmit drug resistance to other S. enterica isolates or to other bacteria. Since integron represents the main vehicle of antibiotic resistance, their presence in S. Typhi indicates uninterrupted transfer of drug resistance genes from one organism to another irrespective of their species which are worrisome with respect to the spread of AMR [27].
Class 1 integrons are increasingly described in environmental and animal bacteria. The release of hospital and municipal effluents is the main way to integrate integrons into the environment [22]. In this study according to our methodology, transfer experiments were not successful. It is possible that these class 1 integron which carried resistance genes contained in Salmonella isolated were not a transferable elements.
The main mechanism of resistance to quinolones is linked to chromosomal mutations especially in the gyrA or parC genes and more rarely at the gyrB and parE genes [28,29]. This study showed only 8 isolates could be amplified both the gyrB and parE genes. Previous studies from Brazil and Senegal were showed similar results with our results [28][29][30][31].
Extrachromosomal genes qnr, aac(6')-Ib-cr and qepA were discovered since 2002 and are carried by conjugative plasmids [32]. Since then, several types of qnr (qnrA, qnrB, qnrS) have been described [32]. These genes were not detected in this study. Several studies shown that these plasmids were most found in non-typhoidal Salmonella and absent in Salmonella Typhi. This was similar with ours results. Others studies shown that plasmids carrying genes encoding ESBLs generally carry genes encoding quinolone resistance [33,34]. This study showed only one ESBL-producing strain (blaCTX-M-14-like). This results were similar to [35] in Sub-Saharan African from blood samples to febrile patients.
Pulsed-field gel electrophoresis has been widely used to determine strain relatedness, to confirm bacterial disease outbreaks, and to identify the sources of strains [36,37]. In the present study, three distinct PFGE types (pulsotypes) were observed from clinical samples with 90-95% similarity in each case. The diversity of pulsotypes could explained by the fact that the samples come from different zones and the fact that the clinical isolates were collected from much longer time span, so the PFGE patterns would change more. These zones differ from their climatic, socio-cultural and even demographic conditions. However all Salmonella Paratyphi B from lettuce samples indicated indistinguishable pulsotypes. This is not surprising since these isolates originate from the same sampling site. In addition, 80-90% similarity was observed between these pulsotypes and clinical ones from Ouagadougou. This could be due to the fact that our strains from lettuce were collected in gardens close to a health center.

Conclusion
This study showed that the isolated Salmonella strains had several virulence factors that play a role at different stages of the infectious process. The study showed the emergence of Salmonella resistant to fluoroquinolone which was previously considered as a hospital problem. These pathogens affect humans through contaminated water and/or food. The foods like lettuce might the route of transmissions of Salmonella disease in Burkina. It is necessary to have a transversal monitoring by pooling the efforts of the actors of the environment, human and veterinary medicine.

Limitations
We were considered strains from clinical and lettuce samples in this study. For few study we will consider strains from soil and wastewater used to irrigate vegetables.

Ethics approval and consent to participate
The study protocol was approved by the Ethical Committees from Health Research of Burkina Faso N°2 019-39, 2009 July, 17. Permission to conduct this study was obtained from hospital authorities and verbal consent was obtained from parents/guardians of every child before taking the stool samples.

Availability of data and materials
The finding of this study is generated from the data collected and analyzed based on the stated methods and materials. All generated data are included in the manuscript.

Competing interests:
The authors declare that they have no competing interests