Molecular Approaches for Isolation of Shigella Species from Culture Negative Diarrheal Samples: A Hospital Based Study in Bangladesh

Diarrhoeal infection is defined as a significant reason of morbidity as well as mortality all over the world. Isolation of Shigella, the most significant etiological agents of diarrhea in developing countries, was missing occasionally when patients feces are cultured by conventional procedures. The collected 200 samples from hospitalized diarrhoeal patients in Tangail district, Bangladesh during 2010-2011, Shigella were isolated from 24 (12%) samples utilizing conventional culture methods. Among176 culture negative stool samples,ipa H gene, specific marker to identify Shigella species was detected from 14 (8%) by PCR. Furthermore, these samples which possess the ipa H gene were cultured and identified as Shigella-like organism (s) by utilizing various phenotypic as well as molecular tools. Of 24 Shigella spp. which was identified using conventional culture methods, S. flexneri were dominant 41.7%, followed by S. boydii 29.2%, S. sonnei 25%, and S. dysenteriae 4.2%. Among 14 of 176 culture negative ipaH positive samples, 7 were identified as Shigella sspp., of which, 3 were S. flexneri, 1 was S. boydii, 2 were S. sonnei and 1 identified as a novel group of S. dysenteriae designated as KIVI 156. Additionally, all Shigella strains were resistant to ampicillin, sulfamethaxol-trimethoprime and were susceptible to ciprofloxacin and mecillinam. This study demonstrated that the conventional culture method cannot be referred as an adequate method to focus the actual disease burden caused by Shigella.


Introduction
Shigella species are referred to be the causative agent for shigellosis. This disorder has emerged as an indispensable public health crisis in countries like Bangladesh where hygiene is limited. In every year, around 165 million cases of reports demonstrating the Shigella infection have been investigated all over the world [1]. Furthermore, it appears that Shigella spp. frequently escape diagnosis by established culture procedure [2]. In addition, an extensive variety of antigenic specificity exists among various Shigella species [3]. For these consequences, isolation by conventional culture methods has always been suspected. Previous studies demonstrated that no Shigella spp. were isolated from 42% clinically Shigellosis patients [4]. Moreover, numerous plasmid-encoded virulence genes like invasion antigen loci (ial) and invasion plasmid antigen H gene (ipa H), can be utilized for identification of Shigella by Polymerase chain reaction (PCR). The gene ial is located on the large virulence 40 Tarana Sharmin et al.: Molecular Approaches for Isolation of Shigella Species from Culture Negative Diarrheal Samples: A Hospital Based Study in Bangladesh plasmid (140 MDa) while the ipa H is encoded multiple copies of the plasmid as well as the chromosome [5]. This study was designed to analyze the undetected cases (clinically shigellosis cases, but Shigella was not detected by conventional culture methods) by using molecular fingerprinting tools, PCR to identify the causative agents, especially Shigella species and their prevalence, which will be conducted for proper treatment as well as further improved prevention. Therefore, the endeavor of this investigation was to identify Shigella and Shigella-like organism (s) by PCR using ipa H gene from the stools of clinically Shigellosis patients from whom no Shigella species were detected by conventional culture methods.

Materials and Methods
A total of 200 stool samples was collected from diarrheal patients irrespective of age, sex and disease severity attended in the Health and Demographic Surveillance System (DSS) area in Kumudini Hospital, Mirzapur, the field site of icddr,b between December 2010 and February 2011. Of these samples, Shigella species were isolated from 24 (12%) samples.

Sampling Procedure
At least 3 ml/grams of fresh stool sample was collected between December 2010 and February 2011 into sterile, screw-capped containers containing Carry-Blair media maintaining cool temperature. All the specimens were cultured on MacConkey and Salmonella-Shigella (SS) Agar (Oxoid,UK) media plate within 6 hours of collection and incubated overnight at 37°C. Furthermore, non-lactose fermenting Shigella like colonies were isolated and characterized using standard methods. All these 200 samples were further enriched in Trypticase Soy Broth (TSB) (Oxoid,UK) with 0.3% yeast extract (YE) for overnight at 37°C. Of these enriched broth cultures from whom no Shigella species could be detected,were further administered to detect the presence of Shigella species by PCR using ipaH gene. Moreover, ipa H positive samples were further analyzed to detect the presence of Shigella species by using different phenotypic and molecular tools [6].

Strains
YSH6000, S. flexneri 2a and an Escherichia coli strain (ATCC 25922) susceptible to all antibiotics were used as positive and negative controls, respectively, to test for detection of ipa H gene and the antibiotic susceptibility test [6].

Categorization of Biochemical Features
The biochemical characters of the strains were investigated by standard traditional procedure [7]. In brief, non-lactose fermenting strains characteristically resembling Shigella were separated from other nonlactose-fermenting strains by inoculating into Kligler's iron agar for characteristic reaction to differentiate organisms by demonstrating hydrogen sulfide production and the fermentation of dextrose and lactose. The motility indole urea test was administered to differentiate the organisms on the basis of motility, urease and indole production and Cimon citrate for citrate utilization according to the procedure described elsewhere. After incubation for 18-24 h at 37ºC, the media were analyzed for the distinctive features of Shigella infection.

Investigation of ipaH Gene by PCR
Exploration of the ipa H gene was performed by amplification by PCR with primers according to previously described procedures [8]. Primers (forward, 5´TGGAAAAACTCAGTGCCTCT3´; reverse, 5´ CCAGTCCGTAAATTCATTCT3´) were utilized using an Oligo 1000 DNA synthesizer (Beckman), available in our laboratory at ICDDR, B. PCR reactions were carried out in a DNA thermal cycler (model 480; Perkin-Elmer Cetus, Emeryville, USA). All PCR reactions followed the same fundamental procedures: 94°C for 90 sec followed by 30 cycles of 45 sec at 94ºC, at T annealing at 60°C for 45 sec and Ti elongate at 72°C for 90 sec, where T annealing was the specific annealing temperature and Ti elongate was the specific elongation time for each reaction, with a final extension at 72°C for 3 min. DNA obtained from E. coli ATCC 25922 and YSH6000, S. flexneri 2a strains were used as negative and positive controls. Every isolate containing one single colony was suspended in 25 µl of reaction mixer containing 3.0 µl of 10x PCR master mix, 1.5 µl of 50 mM MgCl 2 , 2.5 µl of 2.5 mM dNTP, 0.5 µl of primer (forward and reverse) together with 0.2µl of Taq DNA polymerase (5U/µl). Amplified PCR products were taken into horizontal gel electrophoresis in 1% agarose gel in TBE (Tris-borate EDTA) buffer at room temperature at 100 volts (50 mA) for 1h. Briefly, 10 µl of amplified DNA for each sample was mixed with 1 µl of tracking dye and loaded into an individual well of the gel (approx. 5 mm thick). Obtained DNA bands were analyzed by staining the gel with ethidium bromide (0.5 µg/ml) for 30 minutes at room temperature and photographs were taken. In order to measure the molecular size of the amplified products, 1kb DNA size standard (Bio-Rad) was utilized as a marker.

Investigation of ipaH Positive Colony by Colony Patch PCR Technique
176 enriched broth cultures from whom no Shigella species could be detected were further used to detect presence of Shigella species by PCR using ipaH gene following the procedure described elsewhere [8]. Of them, ipaH positive samples were further analyzed to detect presence of Shigella species by using various phenotypic and molecular tools [6]. Different non-lactose fermenting single colonies grown in MacConkey agar plate from ipaH positive broth to create a grid pattern of 36 colonies (6x6) using colony patch technique. The grids were marked off on the bottom of the plate with equally spaced vertical and horizontal lines. The resulting numbered squares demonstrated the identification of individual colonies. The overnight incubation was carried out with the media plates at 37 . This method is significant as well as advantageous as it permits a relatively large number of colonies to be screened simultaneously. In this technique, PCR reactions were condensed into 20 by taking the colonies per row and per column in 5ml sterile distilled water. Additionally, DNA extraction was carried out by boiling the suspensions for 10 min and 5µl of the supernatant was utilized in the PCR reactions. However, the positive colonies were further confirmed by biochemical, serological and API 20E test (bio Me P rieux).

Analytical Profile Index (API 20E)
The API 20E strips comprise of 20 Microtubes which contain dehydrated substrates. These tubes were inoculated with a bacterial suspension that reconstitutes the media. In the course of incubation, color changes occur due to the metabolic reactions that are either spontaneous or exposed by the adding up of reagents. The reactions were further investigated in accordance to the reading table. Finally, the detection is obtained by referring to the analytical profile index.

Serotyping
Biochemically identified Shigella species were serologically confirmed by agglutination reaction with commercially available polyvalent somatic (O) antigen grouping sera, mentioned by testing with monovalent antisera for specific serotype identification (antisera from Denka Seiken, Co. Ltd, Japan). Isolates were sub-cultured on MacConkey agar (Oxoid, UK) plates and after the incubation for overnight, serological reaction tests were performed by the agglutination reaction carried out in the glass slides [6]. These strains were then tested with antiserum raised in rabbits against a heat-killed suspension of the selected strains, a novel group of S. dysenteriae designated as KIVI 162 [9] and the other one KIVI 156. Antisera were prepared according to the protocol [10].

Results
About 200 stool samples were collected between December 2010 and February 2011, whereas Shigella spp. were isolated from 24 samples (12%) by conventional culture methods. All these 24 Shigella spp. were also ipaH positive. No Shigella species were isolated from the rest 88% (n=176) of these 200 stool samples by conventional culture methods. These 176 samples were enriched to test for the existence of ipaH gene by PCR. Of them, 14 (8%) of 176 samples were positive for ipaH, which is specific for Shigella and EIEC. These 14 ipaH positive samples were further analyzed for the detection of specific ipaH positive Shigella-like colonies (non-lactose fermenting colonies) by colony patch technique using PCR (Figure 1).

Plasmid Profile Analysis
Heterogenous plasmid profile was found in all of these culture negative ipaH positive samples. (Figure 2)

Prevalence of Shigella from ipaH Positive Culture Negative Stool Sample
About 14 ipaH positive samples, 6 (42.86%) were identified as Shigella and 1 (7.14%) as SLOs (Table 1) and no Shigella were isolated from the rest 7 (50%). Of Shigella spp., 3were identified as S. flexneri, 1 was as S. boydii, 2 were as S. sonnei and the other one SLO could not be serotyped referring the existing serotyping methods. This strain was then reacted with antiserum raised in rabbits against a heat-killed suspension of a representative strain, a novel group of S. dysenteriae designated as KIVI 156.

Investigation of Biochemical Properties
All isolated strains demonstrated biochemical reactions resembles Shigella species (Table 1). They were oxidase as well as catalase-negative, and did not ferment D-mannitol, sucrose, lactose, rhamnose, raffinose, xylose, trehalose or dulcitol. All strains utilized arginine and were lysine-and ornithine-negative. API 20E tests identified all isolates as Shigella species. (Table 1) Note: M=Motility, I=Indole, U=Urea.

Antibiotic Susceptibility Test
All 7 strains of Shigella were susceptible to ciprofloxacin and mecillam, both of them were commonly administered for treating shigellosis, each strains were resistant to ampicillin and sulfamethaxol trimethoprim ( Table 2).

Discussion
This study has investigated the effectiveness of identification of Shigella species between conventional culture methods and PCR based methods. In order to perform a rapid and definite diagnosis of Shigella infection in patients, PCR assay is highly efficient and specific for identification of Shigella spp. using ipaH gene from fecal samples [2]. Therefore, the present study suggested that the detection of Shigella spp. by PCR using ipaH gene was higher than the existing culture plate method as well as specifically and exclusively identifies organisms that possess this gene. However, sometimes PCR can identify the genetic material of deceased organisms as well. However, in this investigation, 14 culture negative and ipaH positive stool samples were analysed using colony patch technique. Among them, 6 samples were identified as Shigella species and one as novel S. Dysenteriae KIVI 156. Unfortunately, no Shigella species could be identified from the remaining 7 samples of ipaH positive colonies. The possible reason might be accidental contamination or may be due to the presence of lactose fermenting EIEC, since lactose fermenting EIEC also possess ipaH gene. It is significant to note that we used only non-lactose-fermenting colonies incase of colony patch technique. In some cases, DNA from dead bacteria present in stool eventually renders a false positive result for the detection of Shigella. It is also evident that plasmid analysis (core plasmid) can be used as a diagnostic tool as well as 140MDa plasmid could be a diagnostic marker (in case of invasive organism) for the detection of existing Shigella as well as their new variants [12]. Several studies revealed that plasmid profile analysis is considered to differentiate strains to strains as an efficient tool in epidemiological studies dealing with enteric infections [12]. Therefore, further study will be required in details regarding plasmid extraction and analysis of the organisms isolated from the enriched culture negative and ipaH positive stool samples. All these strains were further characterized utilizing various phenotypic tools, such as, morphological characteristics, biochemical properties and antigenic features. All strains were non-lactose fermenting colonies resembling Shigella, which were discriminated from other non-lactose fermenting enteropathogenic strains by typical reaction to differentiate by demonstrating hydrogen sulfide production and fermentation of dextrose and lactose, on the basis of motility, urease and indole production and for citrate utilization (Table  1) according to the procedure described elsewhere [7]. All strains showed typical biochemical characteristics of Shigella species and confirmed by serotyping. Antibiogram analysis revealed that all strains were sensitive to ciprofloxacin as well as mecillinam, where both of the antibiotics were commonly prescribed for treating shigellosis. Therefore, this study also suggested that shigellosis patients could be treated with ciprofloxacin and mecillinam.

Conclusion
This study was designed to analyze the hidden cases (disease positive, but Shigella negative by conventional culture methods) by using molecular fingerprinting tool as PCR. Furthermore, the combination of traditional culture method with molecular fingerprinting tools will be beneficial to identify the actual disease burden due to the infection with causative agents of Shigellosis, which will be conducted for proper treatment along with further protection through improved precaution.