In developing countries like Bangladesh, bacillary dysentery is one of the major causes of death, especially among children. The annual number of Shigella episodes throughout the world is estimated to be 160 million, and 69% of all deaths attributable to shigellosis involve children less than 5 years of age [9]. Shigellosis occurs as an endemic disease in Bangladesh, and at least three large epidemics caused by Shigella dysenteriae type 1 have occurred between 1972 and 1994, causing high morbidity and mortality [3].

Laboratory diagnosis of Shigella is often based on the isolation of the organism from feces and its identification by means of cultural and biochemical characteristics [8]. Serological identification is also indispensable because of the common characteristics of Shigella species with members of the genus Escherichia, from which they cannot even be differentiated by DNA:DNA hybridization [12]. The slide agglutination test using a number of commercially available antisera specific to different groups as well as different types is a common and indispensable test for serological typing of Shigella spp. However, the specificity of commercially produced antisera varies to a significant extent [5, 6], but some of them are useful for routine serological identification purposes [12]. In a previous study, it was found that various types of microorganism such as Hafnia alvei, Plesiomonas shigelloides, Providencia alcalifaciens, and Yersinia enterocolitica serotype O3 and of E. coli (O114:H32, O157:H7, O157:H19, etc.) gave serological cross-reactivity with polyclonal group-specific Shigella antisera [12].

The presence of Shigella in environmental water has already been reported and it can be expected that environmental water plays an important role in spreading the disease throughout the community [7]. There are a number of bacteria present in the environment that have very similar biochemical characteristics to Shigella; thus, it is crucial to understand the serological cross-reactivity of environmental bacteria with Shigella antisera. Previous studies were more or less confined to the serological cross-reactivity of Shigella with clinical isolates using polyclonal group-specific antisera. In the present study, attempts have been taken to investigate the cross-reactivity of Shigella spp. with environmental micro-organisms isolated using standard procedures for isolation of Shigella spp.

Materials and Methods

Sampling, Culturing, and Slide Agglutination Test

Environmental water samples were collected from various Bangladeshi lakes, rivers, and ponds at different places in Dhaka, Gazipur, Karanigonj, Narayan gonj, Mymensingh, and Jamalpur. Fifty millilters of each water sample were filtered through a 0.22-μm membrane filter and the filter was incubated in 50 mL alkaline peptone water (1% peptone, 1%NaCl, pH 8.0) and nutrient broth for 6 h at 37°C. After 6 h, streptomycin was added to the cell suspension to 120 μg/mL final concentration, followed by incubation at 37°C for another 6 h. Aliquots were then streaked onto MacConkey agar plates (OXOID, England) and xylose lysine desoxycholate agar plates (OXOID, England), which were incubated overnight at 37°C. Suspected colonies were subcultered on MacConkey plates and tested by slide agglutination using a commercially available antisera kit (Denka Seiken, Japan) specific for all type- and group-factor antigens of Shigella spp. according to the procedure described previously [12].

PCR and DNA Sequencing

Phylogenetic identification of the strains was based on polymerase chain reaction (PCR) amplification of the highly variable V3 region of the bacterial 16S rRNA gene [14]. The forward primer PRBA338f (5′-ACTCCTACGGGAGGCAGCAG-3′) and the universal reverse primer PRUN518r (5′-ATTACCGCGG CTGCTGG-3′) were used for this purpose. DNA extraction from each isolate was carried out using the GenElute Bacterial Genomic DNA kit (SIGMA, Germany). PCR reactions were carried out with an initial denaturation step at 94°C for 3 min; 30 cycles of 94°C, 1 min; 55°C, 30 s; 72°C, 1 min, followed by a final extension of 6 min at 72°C. The PCR products were subjected to 2% agarose gel electrophoresis, stained with ethidium bromide and visualized on an ultraviolet (UV) transilluminator for the presence of about 200-bp PCR products. The PCR products were purified by using the StrataPrep® DNA purification kit (Stratagene, California). PCR products were then sequenced using the ABI Prism BigDye sequencing kit (Applied Biosystem, UK).

Preparation of Sonicated Whole-Cell Extracts

Forty milliliters of the cell suspension from overnight shake cultures were centrifuged at 10,000g. The pellets were washed with phosphate-buffered saline (PBS) (8.0 g/L NaCl, 0.2 g/L KCl 1.44 g/L Na2HPO4, 0.27 g/L KH2PO4, pH 7.4) and resuspended in 5 mL PBS. The cells were then disrupted by sonication (30 s × 10 times). Sonicates were then centrifuged at 10,000g, filtered, and stored at −20°C.

Enzyme-Linked Immunosorbent Assay (ELISA)

The protein concentration of the whole-cell extracts was determined by the Bradford method [2]. One hundred fifty milliliters of each sample were added to the first lane of a microtiter plate at a concentration of 200 μg/125 μL. The protein samples were then subjected to 1:5-fold serial dilutions in the microtiter plate, which were incubated overnight at 4°C in the dark for coating. The procedure was performed as described elsewhere [10] with minor modifications. Primary antisera were from the Denka Seiken antisera kit. Two types of secondary antisera were used: AP conjugate for polyvalent primary antisera and HRP conjugate for monovalent primary antisera. For the HRP conjugate, the substrate was Sigma Fast OPD (O-phenylenediamine dihydrochloride) prepared in citrate buffer (pH 5.2), and for the AP conjugate, the substrate was p- nitrophenyl phosphate prepared in diethylether buffer (pH 9.2).

Western Blot Analysis

Proteins were fractionated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) using 15% polyacrylamide gels following the procedure described elsewhere [11] with a Mini Protean III Cell system (Bio-Rad, USA). The amount of protein loaded was 5 μg per sample. Western blot was performed as described elsewhere [15]. For each strain, the same primary antiserum as indicated in Table 1 was used. HRP-conjugated secondary serum (anti rabbit IgG; Sigma) was used for monovalent primary antisera and AP-conjugated secondary serum (anti-pig IgG; Sigma) was used for polyvalent primary antiserum. For HRP, the substrate was diaminobenzoic acid (DAB) dissolved in citrate buffer (pH 5.2) with H2O2. For AP, substrates were AS-MX napthol phosphate and Fast Red TR (Sigma) dissolved in 50 mM Tris-HCl, pH 9.14.

Table 1. 16S rRNA gene sequence similarities and serological cross-reactivity with Shigella-specific antisera
Full size table

Results

Isolation of Cross-Reacting Strains

The enrichment and isolation procedures were designed in such a way so as to isolate Shigella spp. from the environmental samples. Initially, colonies that showed similarities with characteristic Shigella colonies on selective MacConkey or xylose lysine deoxychocholate agar plates were chosen for further characterization. Eleven isolates were obtained that yielded Shigella-like colonies and strong positive results in the slide agglutination test with type-specific Shigella antisera (Table 1). Six isolates agglutinated with type-specific monovalent S. dysenteriae antiserum, three with type-specific monovalent S. boydii monovalent antiserum, and two isolates with polyvalent S. flexneri antiserum. No isolate agglutinating with polyvalent or monovalent S. sonnei antiserum was obtained.

Phylogenetic Characterization

Partial sequencing of the 16S rRNA genes of all isolates showed that none belonged to the Shigella genus. BLAST searches identified the strains as belonging to a variety of bacterial groups, including the γ-proteobacteria Escherichia, Enterobacter, and Stenotrophomonas, as well as the Gram-positive genus Aerococcus (Table 1). This is the first report of cross-reactivity of type-specific Shigella antisera with members of these genera.

ELISA and Western Blot Analysis

In order to further investigate the serological cross-reaction with Shigella antisera, ELISA and Western blot analyses were performed. Some of the environmental isolates (e.g, PLMB I, IS10B III, and IS11 SD) gave very strong results in the ELISA test and yielded even higher titers than the positive control strains (Table 1). As positive controls, S. flexneri ATCC12022 and S. boydii ATCC12034 were used, both yielding an ELISA titer of 1:55 using S. flexneri type 2 and S. boydii type 15 antisera, respectively. Other Enterobacter strains (LMBI, IS4SD, IS13SF) and I31 E. coli strains gave equal or lower ELISA values than the positive control, demonstrating a strong serological variety among the environmental isolates. The Stenotrophomonas sp. strain RBSD4 and Aerococcus sp. strain NLSD2 gave weaker reactions than the positive control, but still the cross-reactivity was significant.

Western blot analysis of nine of the isolates also demonstrated serological cross-reactivity with Shigella antisera. Six representative Western blot profiles are shown in Fig. 1, in addition to type strains of S. boydii and S. flexneri. IS10BIII and IS11SD strains showed a smear on the Western blot, which might be caused by lipopolysaccharides, as well as a number of protein bands. PLMB1 and LMB1 gave a pattern quite different from the other strains, but they were similar to each other. Stenotrophomonas strain RBSD4 yielded a unique pattern with at least five protein bands, whereas Aerococcus strain NLSD2 only showed two bands (Fig. 1).

Fig. 1.
figure 1

Western blot membrane showing immunogenic proteins. Lane A = PLMB1; lane B = LMB1; lane C = IS10BIII; lane D = IS11SD; lane E = RBSD4; lane F = NLSD2; lane G = S. flexneri ATCC 12022, lane H = S. boydii ATCC 12034. Arrows indicate molecular-weight markers.

Full size image

Discussion

Water samples were collected from various ponds, lakes, and rivers to investigate the presence of environmental strains of Shigella. It has previously been found that alkaline peptone water (APW) increases the efficiency for isolation of environmental strains of Entrobacteriaceae like Vibrio cholerae, and even Shigella [1]. Use of streptomycin in nutrient broth has also been recommended for isolation of Shigella [1]. A combination of these two methods was used in the present work in order to improve the recovery rate. After enrichment, samples were cultured overnight on selective agar plates, and Shigella-like colonies were screened serologically using slide agglutination tests with polyvalent Shigella-specific antisera. Most of the isolates used in this study gave serological cross-reactivity with monovalent antisera, but two isolates, I31 and IS13SF, which only reacted with polyvalent antisera, have been included, as they exhibited typical Shigella-like colonies on agar plates and were also oxidase negative.

16S rRNA sequence analysis revealed that PLMB I, IS10B III and I31 gave closest match to E. coli. The V3 region of the 16S rRNA is almost identical for E. coli and Shigella; thus, it is not possible to distinguish these two closely related bacteria by V3 region sequence analysis alone. One of the phenotypical characteristics most often used to distinguish E. coli and Shigella is motility. Shigella spp.are always nonmotile, whereas E. coli is generally motile. IS10B III and I31 were found to be nonmotile, whereas PLMB I was motile. IS10B III demonstrated a very high titer in ELISA (Table 1) and conferred multiple distinguished strong bands in Western blot with the monovalent antisera of S. boydii type 15. This isolate might be the first environmental isolate of S. boydii, but further molecular analyses (e.g., the presence of ipaH, virA, ial, or plasmid profile analysis) is required for confirmation [16]. I31 was found to cross-react only with polyvalent antisera and is probably an inactive E. coli strain.

Isolates LMB I, IS4SD, IS11SD, and IS13SF were identified as Enterobacter spp. according to the 16S rRNA gene sequence analysis. Although these isolates were all Enterobacter spp, they revealed different types of serological cross-reactivity with monovalent or polyvalent antisera (Table 1). RBSD4 and RBSD5 were identified as Stenotrophomonas maltophilia (previously known as Pseudomonas maltophilia or Xanthomonas maltophilia), a nonfermentative, Gram-negative bacterium, which plays an increasingly important role as a nosocomial pathogen in compromised patients [4]. Cross-reactivity between Stenotrophomonas and Shigella spp. has not been reported previously.

Interestingly, NLSD2 and L1SD were found to be Gram-positive bacteria but agglutinated with S. dysenteriae-specific antisera. These strains were isolated following prolonged incubation of MacConkey agar plates. The 16S rRNA sequence analysis of NLSD2 and L1SD revealed that these isolates were most closely related to Aerococcus. This is the first report on the serological cross-reactivity of a Gram- positive bacterium with Shigella.

Sonicated whole-cell extracts of the isolates were subjected to ELISA for quantificationer, even more than that of the positive controls. IS4SD and NLSD2 exhibited relatively low titer obtained with ELISA. The similarities in the surface structure of NLSD2 and IS4SD with those of Shigella might be significantly lower than the other isolates.

Representatives of each group of isolates were subjected to Western blot analysis. The banding pattern and the number of bands varied significantly from organism to organism. A smear of bands was found in the case of IS10B III and IS11SD, indicating that theses isolates possess a huge similarity in the surface structures with that of S. boydii type 15 or S. dysenteriae type 10. Only one band was observed after Western blotting in the case of the NLSD 2 isolate. Normally, cross-reactivity among Enterobacteriaceae is due to similarities in the outer membrane proteins, but cross-reactivity between E. coli O164 and S. dysenteriae type 3 has previously been shown to be due to a similar O-antigen structure [13].

We found that a variety of environmental isolates cross-reacted with different types of Shigella-specific monovalent antisera. Yet, to date, no such findings have been reported. However, based on our results, we cannot identify the major cellular components giving rise to this cross-reactivity. Therefore, further studies should be carried out to confirm this finding. Investigation of proteins and plasmid-encoded antigens of Shigella spp. as candidates for vaccine development against shigellosis is currently a major task. The environmental isolates yielding a high cross-reactivity with various types of Shigella antisera represent a potential for development of a shigellosis vaccine based on live cells because they probably are not virulent, like clinical organisms. This possibility will be further explored.