Antibiotic susceptibility and genetic relatedness of Shigella species isolated from food and human stool samples in Qazvin, Iran

Objective The aim of this study was to investigate the genetic relatedness and antimicrobial resistance among Shigella species isolated from food and stool samples. Using cross sectional study method, Shigella spp. were isolated from food and clinical samples using culture-based, biochemical and serological methods. Antimicrobial susceptibility and genetic relatedness among the isolates were evaluated using disk diffusion and RAPD-PCR methods respectively. Results The prevalence of Shigella spp. were 4.84 and 7.7% in food and stool samples respectively. All food isolates were Sh. sonnei. 91.42% of the Shigella stool isolates were Sh. sonnei. 62.5% of food isolates were resistant to tetracycline. 46.8, 50 and 65.8% of clinical isolates were resistant to imipenem, amikacin and azithromycin respectively. 50 and 85.7% of the food and clinical isolates respectively were MDR. Dendrogram generated by RAPD-PCR showed that the isolates from food and stool samples were categorized in a same group. Close genetic relatedness between MDR Shigella isolates from food and clinical samples indicate that foods can be considered as one of the main vehicles for transmission of MDR Shigella to human causing acute diseases. Survey of MDR Shigella among food and clinical samples is strongly suggested to be implemented.


Introduction
Shigella are non-motile, gram-negative and rod-shaped bacteria belonging to the Enterobacteriacae family. During an epidemic outbreak, in 1898, Shigella was isolated for the first time from bacillary dysentery cases by Kiyoshi Shiga in Japan [1]. Shigella comprises four species including Sh. dysenteriae, Sh. flexneri, Sh. boydii and Sh. sonnei [2]. Shigella spp. contribute to intestinal disorders including mild to severe bloody diarrhea in children and adults [3]. However, Sh. dysenteriae type 1 cause hemorrhagic uremic syndrome (HUS) as an extraintestinal pathogen via releasing shiga-toxins [4]. Prevalence of Shigella spp. is geographically different, as Sh. sonnei and Sh. flexneri have more been isolated from patients in industrialized and developing countries respectively. Shigellosis have commonly been seen among the children under the age of 5 years old especially in low and middle-income countries. Shigella is an important foodborne pathogen and sometimes is transmitted via animals. Regarding the remarkable reduction in mortality of this pathogen in the recent decades, there are still more than 160,000 deaths annually caused by Shigella worldwide [5]. Drug resistance of human pathogens is one of the main concerns of public health. Multidrug resistant (MDR) is defined when a pathogen is resistant to two or more antibiotics from different groups. Several studies have been highlighting the importance of MDR human and animal pathogens [6][7][8][9][10][11]. Spread of MDR Shigella is associated with severe outcomes and leads to failure in treatment. Foods have been considered as an important vehicle for transmission of pathogens. A few studies demonstrated a close genetic relatedness between the Shigella isolates from food and clinical samples indicating that Shigella may be transmitted through food to human leading to intestinal and extraintestinal infectious diseases [12,13]. Random amplified polymorphic DNA (RAPD) method has widely been used to investigate the genetic relatedness between the pathogenic isolates from different sources [14]. Many researchers used RAPD-PCR to study the genetic diversity among pathogenic isolates [15]. Several studies have been performed to characterize the antibiotic resistance in MDR Shigella isolates from food and clinical samples [13]; however, there are limited literature available that provides the genetic relatedness among MDR Shigella isolates from food and clinical samples. The objective of the present study was to investigate the genetic relatedness among the MDR Shigella spp. isolated from food samples and stool specimens from the patients with diarrhea.

Main text
Bacterial isolates 453 children between the age of 2-5 years old, including 242 males and 211 females, who visited the department of children health and diseases (Qazvin children hospital, Qazvin, Iran) for diarrhea and dysentery from November 2019 to July 2020 were enrolled into the study and forwarded to the clinical central Lab of Qazvin children hospital, Qazvin, Iran. All stool specimens were collected into clean, sterile and disposal containers according to the protocol described by WHO Laboratory investigations, manual of Enteric infections for fecal sample preparation [11]. Totally 165 food samples, consisting of 55 samples of each food item, including minced meat (100 g), vegetable salad (500 g) and raw milk (1 L) were collected from local stores, located in different areas of Qazvin city, from September to December 2019. All clinical and food samples were kept at 4 °C prior to bacterial isolation and were immediately transported to the central Lab. S. sonnei ATCC 25,931; S. boydii ATCC 12,030; and S. flexneri ATCC 12,022 were purchased from Pasteur institute of Iran (Pasteur institute, Tehran, Iran) and used as the reference strains in this study. All reference strains were activated by inoculation into Bovine Heart Infusion (BHI) broth and incubation at 37 °C for 24 h.

Shigella spp. isolation and identification
After homogenization, stool samples were directly streaked onto MacConkey and Salmonella-Shigella (SS) agar (Promedia, Spain) and incubated anaerobically at 37 °C for 24 h. H 2 S negative and non-lactose fermentative colonies were selected for further biochemical tests. Isolation of Shigella spp. from food samples were conducted using the method described by Mokhtari et al. [13]. First, 25 g or mL of each food sample were homogenized using Stomacher BagMixer Lab blender (InterScience, France) for 1 min. Homogenized samples were mixed and diluted with 225 mL of buffered peptone water broth (BPW, Promedia, Spain) and incubated for 24 h at 37 °C. Diluted samples were inoculated into Shigella-broth (Promedia, Spain) and incubated at 37 °C overnight. Enriched samples were streaked onto the MacConkey and SS agar (Promedia, Spain) plates and incubated anaerobically at 37 °C for 24 h. Urease, motility, triple sugar iron, lysin iron decarboxylase and IMViC tests were performed for biochemical confirmation of the presumptive colonies. Serological assay was used for identification of Shigella genus using Difco Antisera Kit (BD-Difco Co. USA). Different species of Shigella isolates were also identified by serological tests using Shigella species Difco Antisera Kit (BD-Difco Co. USA). All serological tests were conducted according to the kit manufacturers` instruction [12].

Antimicrobial Susceptibility testing
The antimicrobial susceptibility of the isolates was evaluated using the disk diffusion technique which was previously described by Marami et al. [16] in accordance with the guideline of Clinical and Laboratory Standards Institute (CLSI) [17]. Seven antimicrobial disks (Oxoid, Ltd, UK) including chloramphenicol (30 mg), tetracyclin (30 mg), ampicillin (10 mg), amikacin (30 mg), imipenem (10 mg), azithromycin (15 mg) and nalidixic acid (30 mg) were applied. When the isolates were resistant to two or more different antimicrobial agents, multidrug resistance (MDR) was defined.

DNA extraction
All strains were inoculated into trypticase soy broth (TSB, Promedia, Spain), incubated at 37 °C overnight and centrifuged at 3500 rpm for 10 min. After discarding the supernatant, precipitated cell pellets were subjected to DNA extraction using the gram-negative bacterial DNA extraction kit (Sinaclon, Iran) according to the manufacturers` instruction. Quantity and quality of the extracted genome were measured using NanoDrop Spectrophotometer (Thermo Scientific, USA). Concentration of the DNA templates were adjusted to 50 ng. µL −1 .

RAPD-PCR based genotyping and genomic fingerprinting
In this study, primer UBC245: 5ʹ-CGC GTG CCA G-3ʹ [18], previously used for RAPD-genotyping of Shigella, was used. Each reaction tube contained 25 µL of total reaction volume including 12.5 µL of PCR 2X master mix kit (Ampliqon, Denmark), 0.5 µL of primer (50 µM), 1 µL of DNA template (50 ng. µL −1 ) and deionized sterilized water to the final reaction volume. Thermal cycling was run using Bio-rad T100 machine (Bio-rad, USA) in the following program: initial denaturation 5 min at 95 °C; 35 cycles of 95 °C for 1 min, 36 °C for 1 min, 72 °C for 5 min; and 5 min at 72 °C as the final extension. PCR products were characterized using gel electrophoresis for 2 h at 70 v on 1.5% w/v agarose (Merck, Germany) containing DNA safe stain (CinnaGen, Iran) along with 100-bp DNA ladder. Fluorescent banding patterns were recorded using gel documentation system (Mahazma, Iran). For genomic fingerprinting, DNA banding patterns were interpreted using PyElph software version 1.4 [19]. Similarities were calculated and the unweighted pair group method with arithmetic mean (UPGMA) dendrogram was drawn regarding Dice coefficient by NTSYSpc software version 2.20 (Applied Biostatics Inc., USA).

Results
Shigella were detected in 8 food samples (4.84%). All of these isolates were Sh. sonnei and isolated from vegetable salad samples. Shigella spp. were detected in 35 stool samples (7.7%) including 32 Sh. sonnei (91.42%), 2 Sh. flexneri (5.71%) and 1 Sh. boydii (2.85%) isolate. We have not isolated any Sh. dysenteriae. As shown in Table 1, 62.5% of food isolates were resistant to tetracycline and all isolates were sensitive to amikacin. 46.8, 50 and 65.8% of clinical isolates were resistant to imipenem, amikacin and azithromycin respectively. Also, all clinical isolates were sensitive to tetracycline, chloramphenicol and nalidixic acid. 4 (50%) and 30 (85.7%) of the food and clinical isolates respectively were MDR. 16 distinct amplicons ranged in size from 60 to 1100 bp were detected by RAPD-PCR formed 22 different RAPD profiles and 3 main genotypic clusters consisting of groups R1, R2 and R3 (Fig. 1). Genotypic categorization and serogroups of the isolates were demonstrated in Table 2. All Sh. sonnei isolated from food and stool samples were categorized together in a same cluster (Group R2) with a close genetic relatedness.

Discussion
Shigellosis, specially caused by MDR Shigella spp., is now considered as one of the main public health concerns. The prevalence of this pathogen is continually increasing specially in developing countries [20]. Shigella spp. lead to death in children under the age of 5 years old [12]. In addition to person-to-person transmission, Shigella spp. are transmitted via food as a foodborne pathogen to infect human [21]. At the present study, we found higher prevalence rates of MDR Shigella isolates with close genetic relatedness in food and stool samples indicating poor hygiene practices and concern to public health.
The overall prevalence of the Shigella spp. in food and stool samples were 7.7% and 4.84% respectively. The finding of current study is significantly high compared to previous reports. Bakhshi et al. found Shigella spp. in 1.32% of all stool samples, collected from patients with diarrhea in Iran, consisting of Sh. sonnei in 92% of all isolates [22]. Sh. sonnei have predominantly been isolated in developed countries [23]. We also found Sh. sonnei predominantly in 100 and 91.42% of all isolates from food and clinical samples respectively. These findings are comparable to studies conducted before. Ranjbar    Analysis of antimicrobial susceptibility in this study showed that the food isolates were highly resistant to tetracycline and the clinical isolates were strongly resistant to imipenem, amikacin and azithromycin. Higher susceptibility to tetracycline, chloramphenicol and nalidixic acid was also observed in clinical isolates. Shahin et al. isolated Shigella spp. from water samples strongly resistance to nalidixic acid and susceptible to chloramphenicol [28]. Marami et al. isolated MDR Shigella spp. with 75% resistance to tetracycline [16]. Aklilu et al. observed MDR Shigella isolates completely susceptible to amikacin. These differences in antimicrobial susceptibility patterns may be explained by different availability of antibiotics worldwide. We found different antibiotic resistance patterns between food and clinical isolates [29]. Mokhtari et al. also found significantly different antimicrobial susceptibility patterns of Shigella isolates obtained from food and clinical samples [13].
There are only few studies regarding the genetic relatedness among Shigella isolates obtained from food and clinical samples. We found a close genetic relatedness between the food and clinical isolates as they were clustered in a same group. Mokhtari et al. also demonstrated a clonal relationship amongst food and human stool Shigella isolates [13]. These findings indicate that foods are important vehicle of transmission of MDR Shigella to human causing acute intestinal and extraintestinal diseases.

Conclusions
We detected Shigella spp. in 4.84 and 7.7% of food and stool samples respectively. We observed Sh. sonnei as the predominant species in both food and stool samples. We observed food isolates resistant to tetracycline and the clinical isolates strongly resistant to imipenem, amikacin and azithromycin. All clinical isolates were susceptible to tetracycline, chloramphenicol and nalidixic acid. 50% and 85.7% of the food and clinical isolates respectively were MDR. We found a close genetic relatedness between the isolates from food and clinical samples. The present study revealed a high prevalence of MDR Shigella with close genetic relatedness from food and stool samples indicating poor hygiene practices. Our findings also indicate that MDR Shigella can be transmitted through foods and cause infectious disease in human. Performing other studies is of great importance to investigate genetic relatedness between MDR Shigella isolates from food and clinical samples.

Limitations
• 43 Shigella isolates from all food and clinical samples are not sufficient to evaluate the genetic relatedness and the antimicrobial susceptibility of the isolates. • RAPD-PCR method is not adequately precise and reproducible for genotyping of Shigella isolates; other sequence based genotyping techniques have been recommended.