ARTICLE IN PRESS

Microbiological Research 165 (2010) 33—42

0944-5013/$ - sdoi:10.1016/j.

�Correspond+91 942 515532

E-mail addmail.com (A. S

www.elsevier.de/micres

Phenotypic and genotypic characterization ofShigella spp. with reference to its virulence genesand antibiogram analysis from river Narmada

Anjana Sharma�, Susheel Kumar Singh, Divya Bajpai

Bacteriology Laboratory, Department of P.G. Studies and Biological Research, R.D. University, Jabalpur 482001,MP, India

Received 21 December 2006; received in revised form 7 August 2007; accepted 12 February 2008

KEYWORDSRiver water;Shigella;Serotyping;Molecular typing;Virulence gene

ee front matter & 2008micres.2008.02.002

ing author. Tel.:3 (M); fax: +91 761 2603resses: anjoo_1999@yaharma).

SummaryWater samples of the river Narmada from origin to end were analyzed for thepresence of shigellae. Analysis of 40 water samples by biotyping, serotyping, andmolecular typing were done. Out of all 40 isolates, 23 were identified as Shigellaflexneri, 10 as Shigella sonnei, and seven as Shigella dysenteriae. Serotyping wasfound to be a better identification method than biotyping since biotyping was notfound to be very sharp. In the present investigation, amplified ribosomal DNArestriction analysis (ARDRA) with a probe complementary to 16S rRNA wasperformed. Repeated ARDRA analysis establishes the similarities between theisolates and thus suggested ARDRA as authentic and precise detection protocol. Theisolates were also analyzed for the presence of virulence genes including ipaBCD,ipaH, and stx1 and all the 40 isolates of Shigella showed positive result for ipaH genewhile the plasmid encoded invasion-associated genes ipaBCD was present only in S.flexneri and stx1 gene was present only in S. dysenteriae. This study demonstratedthe existence of Shigella in the river Narmada and dispersion of different virulencegenes among these isolates, which appear to constitute an environmental reservoirof Shigella-specific virulence genes.& 2008 Published by Elsevier GmbH.

Published by Elsevier GmbH.

+91 761 2416667 (R);752.hoo.com, anjoo.2007@g-

Introduction

Bacillary dysentery caused by Shigella is endemicthroughout the world and it is among the mostcommon causes of bacterial diarrheal diseases. It isresponsible for approximately 165 million cases

ARTICLE IN PRESS

A. Sharma et al.34

annually, of which 163 million are in developingcountries and 1.5 million in industrialized ones. It isestimated that 1.1 million people die annually fromShigella infection and nearly 580,000 cases ofshigellosis are reported among travelers fromindustrialized countries. The frequency of Shigellaflexneri, Shigella sonnei, Shigella boydii, andShigella dysenteriae were 60%, 15%, 6%, and 6%(30% of S. dysenteriae cases were type 1) indeveloping countries; and 16%, 77%, 2%, and 1% indeveloped ones, respectively. In developing coun-tries, the predominant serotype of S. flexneri is 2a,followed by 1b, 3a, 4a, and 6 (Kotloff et al., 1999;Peirano et al., 2006). Although Shigella-contami-nated food and drinks are often source of epidemicspread and very little is known about its presenceand possible spread through environmental waters.In many developing countries with inadequatesanitation, fecal contaminations of environmentalwaters by enteric pathogens are very common. It istherefore important to understand whether Shigel-la can survive and persist in environmental watersin the absence of primate host and investigate thevirulence characteristics of such environmentalstrains.

Identification of Shigella in environmental sam-ples, where the number of organisms is likely to be

Figure 1. Sampling statio

small, is limited mainly by the lack of a suitableenrichment technique. Although DNA probes or PCRassay directed against the large invasion plasmid orgenes encoding shiga toxin (Venkatesan et al.,1988; Frankel et al., 1990; Oberhelman et al.,1993; Newland and Neill, 1998) can be used todetect the presence of the organism. Isolationof the live bacteria is essential to characterizetheir pathogenic potential as well as their sensi-tivity to antimicrobial agents. Detailed analysisof large number of water samples for the pre-sence of Shigella by conventional culture methodsis impractical, particularly because the numberof non-lactose fermenting colony to be furtheranalyzed may be too high (Frankel et al.,1990). River water is major source of microbialpathogens in developing countries. River Narmadais the largest west flowing river of Indianpeninsula. The present study was designed toisolate Shigella strains from the water samplesof river Narmada in India by combination ofPCR and culture methods and characterize themby appropriate biochemical and serologicaltests. Furthermore, molecular techniques wereused to genetically characterize such environmen-tal strains and compare them with the clinicalstandard.

ns of river Narmada.

ARTICLE IN PRESS

Phenotypic and genotypic characterization of Shigella spp. 35

Materials and methods

Bacterial strains and water samples

Water samples were collected from 11 stations ofriver Narmada (Amarkantak, Dindori, Mandla,Jabalpur, Narsinghpur, Hoshangabad, Omkareshwar,Koral, Neelkantheshwar, Ankaleshwar, and Dahej),over 1-year period from July 2005 to June 2006(Figure 1). All water samples were collected andbrought to the laboratory under ice-cold conditionsand concentrated on 0.22 mm pore size and 47mmdiameter filters, then enriched in nutrient brothand incubated at 37 1C for 6–8 h under shakingcondition (modified method of APHA given byFaruque et al., 2002). Bacterial colonies wereisolated on MacConkey agar medium, xylose lysinedecarboxylaste agar medium and hektoen entericagar medium (Himedia, India). Non-lactose-fer-menting colonies were picked from culture platesand were subjected to further analysis for theidentification and isolation of possible Shigellacolonies. After the identification, the isolates weremaintained in the Bacteriology lab, Department ofBiological Sciences, R.D. University (MP), India andwere given BGCC (Bacterial Germplasm CollectionCenter) numbers.

Biochemical characteristics

All the isolates were examined for the followingbiochemical characteristics; arginine dihydrolase,lysine decarboxylase, ornithine decarboxylase,citrate, indole, oxidase, urease, MR, VP, gluconate,malonate, acid from arabinose, dulcitol, glucose,lactose, maltose, mannitol, raffinose, rahmnose,sorbitol, sucrose, and xylose (MacFaddin, 1980).The isolates were identified with the help ofBergey’s Manual of Systematic Bacteriology (1984)and probabilistic identification of bacteria (PIB)computer kit (Bryant, 2003).

Serological analysis

Forty isolates presumptively confirmed by bio-chemical tests as Shigella spp. were subjected toserotyping by slide agglutination with poly A, polyB, Cl, C2, C3, and D antigens (Edwards and Ewing,1972).

RAPD

Genomic DNA was extracted as described bySambrook et al. (1989). PCR amplification ofisolated template DNA was carried out with a

primer [(50-CAT TCG ACC 30)] (Silvia et al.,1998). The reaction was performed in a totalvolume of 25 ml per tube containing 2.5 ml of 10�PCR buffer [500mM KCl, 100mM Tris (pH 8.3),25mM MgCl2, 3 ml of template DNA, 2 ml primer,0.3 ml of Taq DNA polymerase (Bangalore Genei,India), 2.5 ml of 2.5mM dNTPs. A negative controlwas maintained containing all components excepttemplate DNA. After initial denaturation at 94 1Cfor 3min, the reaction mixture was run through 45cycles of denaturation at 94 1C for 1min, annealingat 36 1C for 1min and elongation/extension at 72 1Cfor 2min followed by a 10min final extensionperiod at 72 1C and the expected size of theamplicons were ascertained by electrophoresis in1.5% agarose gel with an appropriate molecularsize marker (1 kb DNA ladder, Bangalore Genei,India).

ARDRA

Genomic DNA was extracted as described above.PCR oligonucleotides primers were derived fromconserved region present at the edges of the 16SrDNA. The sequences of primers were F-50 AGA CTGCTA CGG GAG GCA GCA GT 30, R-50 GTT GCG CTCGTT GCG GGA CTT AA 30. In total, 1 ml (around 10 ngDNA) of the template DNA was added to 49 mlaliquots of PCR mixture containing 5 ml of l0�buffer supplied with enzyme (Taq polymerase)2.5 ml of each forward and reverse primer from a10 pmol stock (i.e. 12.5 pmol of each primer), 2.5 mlof 10mM (each) deoxynucleotide triphosphates and0.5 ml of DNA polymerase (3 unit/ml), final volumewas made 50 ml by adding sterile distilled water.After initial denaturation at 94 1C for 3min, thereaction mixture was run through 35 cycles ofdenaturation at 94 1C for 1min, annealing at 57 1Cfor 1min and extension at 72 1C for 1min, finally a10min extension period at 72 1C was carried out.The 16s rRNA type was determined by digestion ofthe amplicons with HindIII (A/AGCTT), AluI (AG/CT)and EcoRI (G/AATTC) and analysis by electrophor-esis on a 1% agarose gel with staining by ethidiumbromide (0.5 mg/ml).

Virulence gene detection

A multiplex PCR assay for the genes encodinginvasive plasmid antigens (ipaBCD) and Shigatoxin (stx1), as well as a second PCR assay foripaH was used to detect the presence of thesevirulence genes in Shigella isolates (Faruque et al.,2002).

ARTICLE IN PRESS

A. Sharma et al.36

Sequences of virulence genes

S. no.

Gene Primer sequence PCR condition Amplicon (bp)

1.

IpaH F 50-GCTGGAAAAACTCAGTGCCT-30 94 1C, 2min 427 55 1C, 2min

R 50-CCAGTCCGTAAATTCATTCT-30

72 1C, 3min 2. ipaBCD F 50-GCTATAGCAGTGACATG-30 94 1C, 2min 500

R 50-ACGAGTTCGAAGCACTC-30

55 1C, 2min 72 1C, 3min

3.

Stx1 F 50-CAACACTGGATGATCTCAG-30 94 1C, 2min 100 R 50-CCCCCTCAACTGCTAATA-30 55 1C, 2min

72 1C, 3min

Antibiogram analysis

Disk diffusion assay was followed to assess theantibiotic resistance/sensitivity pattern as de-scribed by CLSI (2002). Susceptibility to antimicro-bial agents was tested on Mueller–Hinton agar usingcommercially available disks (Himedia, India):amikacin 30mcg, amoxicillin 30mcg, ampicillin10mcg, cephotaxime 30mcg, ceftazidime 30mcg,chloramphenicol 30mcg, ciprofloxacin 30mcg, gen-tamycin 10mcg, nalidixic acid 30mcg, norfloxacin10mcg, streptomycin 25mcg, tetracycline 30mcg,and trimethoprin 10mcg.

Cluster analysis

All binary data were entered and genetic dis-tances were calculated through Numerical Taxon-omy and Multivariate Analysis System (NTSYS-pc),version 2.02 (Exeter Software, New York) andcalculating Euclidean distance and then assemblinga dendrogram using ‘‘unweighted pair group meth-od using arithmetic-average clustering criterion.

Results

Biochemical analysis

In total, 25 biochemical tests were performedand the isolates were identified with the help ofBergey’s manual of Systematic Bacteriology (1984)and PIB computer kit (2003). Among 40 isolates ofShigella spp., seven were identified as S. dysenter-iae, 23 were identified as S. flexneri, and 10 wereidentified as S. sonnei on the basis of morpho-logical, cultural and biochemical characteristics(Table 1).

Serological analysis

All isolates showed different agglutination result.In total, isolates of Shigella showed agglutinationwith polyB antigen and identified as S. flexneri.Seven isolates of Shigella showed agglutination withpolyA antigen and identified as S. dysenteriae and10 isolates of Shigella showed agglutination with Dantigen and identified as S. sonnei.

Random amplified polymorphic DNA analysis(RAPD) of Shigella

Analysis of all 40 isolates revealed that theShigella species produced five different RAPDpattern with number of bands ranging from 5 to11. One common band of 500 bp was found in all theisolates. However, a band of 250 bp was evident inall the isolates except for BGCC# 384, 387, 390,399, 410, 413, and 414. Around 23 isolates ofS. flexneri showed two RAPD profiles with no. ofbands ranging from 9 to 11, 10 isolates of S. sonneishowed two RAPD profiles with no. of bandsranging from 7 to 9 and seven isolates ofS. dysenteriae showed single RAPD profile withsame banding pattern (500, 270, 250, 240, and200 bp) (Figure 2). All isolates were analyzed byUPGMA on the basis of RAPD. Total two clusters(A&B) were found in the dendrogram at a coeffi-cient level of 3.36. Cluster A was represented bynine isolates and B was represented by 31 isolates.Cluster B was further divided into B1 and B2 atcoefficient level of 3.16 (Figure 3).

Restriction fragment length polymorphism ofPCR amplified 16S rRNA gene (ARDRA) ofShigella spp.

The 16S rRNA gene (16S rDNA) was enzymaticallyamplified for all the 40 isolates belonging to three

ARTIC

LEIN

PRES

S

Table 1. Biochemical characteristics of Shigella spp. isolated from the river Narmada

Isolates no. Identification parameters Identified spp.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

BGCC382 � � � + � � � � � + � � � � � � � + + + � � + + + S. flexneriBGCC383 � � � + � � � � � + � � � � � � � + + + + � + + + S. flexneriBGCC384 � � + + � � � � � + � � � � � � + � + � � � + � S. dysenteriaeBGCC385 � � � + � � � � � + � � � � � � � + + + + � + + + S. flexneriBGCC386 � � � + � � � � � + � � + � � � � + + + � � � + � S. sonneiBGCC387 � � + + � � � � � + � � � � � � � + � + � � � + � S. dysenteriaeiBGCC388 � � + + � � � � � + � � + � � � � + + + � � � + � S. sonneiBGCC389 � � + + � � � � � + � � � � � � � + + + � � � + + S. flexneriBGCC390 � � + + � � � � � + � � � � � � � + � + + � + + � S. dysenteriaeBGCC391 � � � + � � � � � + � � � � � � � + + + � � � + � S. flexneriBGCC392 � � � + � � � � � + � � � � � � � + + + + � + + + S. flexneriBGCC393 � � � + � � � � � + � � + � � � � + + + � � � + � S. sonneiBGCC394 � � � + � � � � � + � � � � � � � + + + + � � + � S .flexneriBGCC395 � � + + � � � � � + � � � � � � � + + + + � + + + S. flexneriBGCC396 � � + + � � � � � + � � � � � � � + + + � � + + + S. flexneriBGCC397 � � + + � � � � � + � � � � � � � + + + + � + + + S. flexneriBGCC398 � � + + � � � � � + � � � � � � � + + + + � + + + S. flexneriBGCC399 � � + + � � � � � + � � � � � � � + � + � � � + � S. dysenteriaeiBGCC400 � � � + � � � � � + � � + � � � � + + + � � � + � S. sooneiBGCC401 � � � + � � � � � + � � + � � � � + + + � � � + � S. sonneiBGCC402 � � � + � � � � � + � � + � � � � + + + � � � + � S. flexneriBGCC403 � � � + � � � � � + � � � � � � � + + + � � � + � S. sonneiBGCC404 � � � + � � � � � + � � � � � � � + + + + � � + + S. flexneriBGCC405 � � � + � � � � � + � � + � � � � + + + � � � + � S. sonneiiBGCC406 � � + + � � � � � + � � � � � � � + + + � � � + � S. flexneriBGCC407 � � � + � � � � � + � � � � � � � + + + + � � + + S. dysenteriaeBGCC408 � � � + � � � � � + � � � � � � � + + + � � � + � S. flexneriBGCC409 � � � + � � � � � + � � � � � � � + + + � � � + � S. flexneriBGCC410 � � � + � � � � � + � � � � � � � + + + + � � + + S. sonneiBGCC411 � � � + � � � � � + � � � � � � � + + + � � + + + S. flexneriBGCC412 � � � + � � � � � + � � � � � � � + + + � � + + + S. flexneriBGCC413 � � + + � � � � � + � � � � � � � + � + � � � + � S.dysenteriaeBGCC414 � � + + � � � � � + � � � � � � � + � + � � � + � S. dysenteriaeBGCC415 � � � + � � � � � + � � + � � � � + + + � � � + � S. sonneiBGCC416 � � � + � � � � � + � � � � � � � + + + � � + + + S. flexneriBGCC417 � � � + � � � � � + � � + � � � � + + + � � � + � S. sonneiBGCC418 � � � + � � � � � + � � � � � � � + + + � � + + + S. flexneriBGCC419 � � � + � � � � � + � � � � � � � + + + � � + + + S. flexneriBGCC420 � � � + � � � � � + � � + � � � � + + + � � � + � S. sonneiBGCC421 � � � + � � � � � + � � � � � � � + + + � � + + + S. flexneriBLSG/001 � � + + � � � � � + � � � � � � � + � + � � � + � S. dysenteriae type1

1 – motility, 2 - oxidase, 3 - Indole, 4 – MR, 5 – VP, 6 – Citrate, 7 – Urease, 8 – Pb (TSI) 9 – Gas(TSI), 10 – Glucose, 11 – Lactose/Sucrose, 12 – Arginine hydrolysis, 13 – Ornithine decorboxylase,14 – lysine decorboxylase, 15 – Gluconate, 16 – Malonate, Acid production (17 – Dulcitol, 18 – Arabinose, 19 – Mannitol, 20 – Trehalose, 21 –Sorbitol, 22 – Raffinose, 23 – Rahmnose, 24 –

Maltose, 25 – Xylose).Note: +: Positive, �: Negative, BGCC: Bacterial Germplasm Collection Centre.

Phenotypic

andgenotyp

iccharacterization

ofShigella

spp.

37

ARTICLE IN PRESS

Lane 1 to 41

3000

500

100

3000

500

100

M MM

Figure 2. RAPD profile of Shigella spp. isolated from river Narmada. Lane – 1, 4, 6, 7, 9, 10, 12, 15, 16, 17, 22, 24, 25,26, 29, 30, 31, 32, 36, 38, 39, 40, 41, (S. flexneri), Lane- 3, 5, 8, 11, 13, 18, 19, 28, 33, 34, (S. sonnei), Lane- 2, 14, 20,23, 27, 35, 37, (S. dysenteriae), and Lane- 21 (control, S. dysenteriae type 1) M – Marker DNA Ladder 1 kb.

Coefficient

0.00 0.84 1.68 2.52 3.36

BGCC382 BGCC383 BGCC389 BGCC392 BGCC396 BGCC397 BGCC404 BGCC406 BGCC408 BGCC409 BGCC411 BGCC418 BGCC421 BGCC401 BGCC405 BGCC420 BGCC386 BGCC388 BGCC393 BGCC400 BGCC403 BGCC415 BGCC417 BGCC384 BGCC387 BGCC390 BGCC399 BGCC410 BGCC413 BGCC414 BGCC419 BLSG/001 BGCC385 BGCC391 BGCC394 BGCC395 BGCC398 BGCC402 BGCC407 BGCC412 BGCC416

Figure 3. Dendrogram analysis of Shigella spp. isolated from river Narmada on the basis of RAPD patterns. BGCC – 382,383, 385, 389, 391, 392, 394, 395, 396, 397, 398, 402, 404, 406, 407, 408, 409, 411, 412, 416, 418, 419, 421(S. flexneri), BGCC – 386, 388, 393, 400, 401, 403, 405, 415, 417, 420 (S. sonnei), BGCC – 384, 387, 390, 410, 413, 414(S. dysenteriae) and BLSG/001 – (control – S. dysenteriae type 1).

A. Sharma et al.38

described genomic species of the genus Shigella,and the amplicon was restricted independentlywith three different restriction enzymes (EcoRI,HindIII, and AluI). All the three enzymes produced2–5 fragments on digestion (Figure 4). However,

EcoRI resulted in overall higher dissimilarity amongthe isolates. Based on EcoRI digestion with dissim-ilarity coefficient of 2.30, two major divisionsdesignated as A and B were found. The vastmajority of strains of Shigella were found to be in

ARTICLE IN PRESS

Figure 5. Dendrogram analysis of Shigella spp. isolated from river Narmada on the basis of ARDRA patterns. BGCC –

382, 383, 385, 389, 391, 392, 394, 395, 396, 397, 398, 402, 404, 406, 407, 408, 409, 411, 412, 416, 418, 419, 421(S. flexneri), BGCC – 386, 388, 393, 400, 401, 403, 405, 415, 417, 420 (S. sonnei), BGCC – 384, 387, 390, 410, 413, 414(S. dysenteriae) and BLSG/001 – (control- S. dysenteriae type 1).

ARDRA Type 3

3000 3000

500500

100 100

Lane 1 to 41

M2 2 1 2 3 1 3 2 1 2 2 3 2 2 2 2 2 1 3 3 1M2 3 2 3 2 2 2 2 1 2 2 1 1 3 2 M2322

Figure 4. ARDRA profile of Shigella spp. isolated from river Narmada. Lane – 1, 2, 4, 8, 10, 11, 13, 14, 15, 16, 17, 22,24, 26, 27, 28, 29, 31, 32, 36, 38, 39, 41 (S. flexneri), Lane – 5, 7, 12, 19, 20, 23, 25, 35, 37, 40 (S. sonnei), Lane – 3, 6,9, 18, 30, 33, 34 (S. dysenteriae), and Lane – 21 (Control – S. dysenteriae type 1), M- Marker DNA Ladder 1 kb.

Phenotypic and genotypic characterization of Shigella spp. 39

ARTICLE IN PRESS

Lane 1 to 41

100

500

100

3000

500

3000

MM

Figure 7. Multiplex PCR analysis of IpaBCD and Stx 1 gene of Shigella spp. isolated from river Narmada. Lane – 1, 2, 4,8, 10, 11, 13, 14, 15, 16, 17, 22, 24, 26, 27, 28, 29, 31, 32, 36, 38, 39, 41 (S. flexneri), Lane – 5, 7, 12, 19, 20, 23, 25,35, 37, 40 (S. sonnei), Lane – 3, 6, 9, 18, 30, 33, 34 (S. dysenteriae), and Lane – 21 (Control – S. dysenteriae type 1),M – Marker DNA Ladder 1 kb.

Lane 1 to 41

3000

500

100 100

500

3000

M M

Figure 6. Uniplex PCR analysis of IpaH gene of Shigella spp. isolated from river Narmada. Lane – 1, 2, 4, 8, 10, 11, 13,14, 15, 16, 17, 22, 24, 26, 27, 28, 29, 31, 32, 36, 38, 39, 41 (S. flexneri), Lane – 5, 7, 12, 19, 20, 23, 25, 35, 37, 40(S. sonnei), Lane – 3, 6, 9, 18, 30, 33, 34 (S. dysenteriae), and Lane – 21 (Control – S. dysenteriae type 1), M – MarkerDNA Ladder 1 kb.

A. Sharma et al.40

group B, which were further divided into B1 and B2(Figure 5).

PCR analysis of virulence gene

All the 40 isolates were subjected to PCR for thedetection of IpaH gene whose protein product isnecessary for invasion of colonic epithelial cells andalso for the detection of Shigella in environment,IpaBCD gene the invasion plasmid antigen and shigatoxin (stx1) gene. The invasion gene ipaH (427 bp)was present in all the 40 isolates (Figure 6), whichconfirmed the presence of Shigella in river waterand showed 100% specificity for all Shigellaisolates. The ipaBCD gene (500 bp) was present inall the 23 isolates of S. flexneri spp. while ipaBCDgene was absent in all the isolates of S. dysenteriaeand S. sonnei, respectively. The shiga toxin (stx1)

gene (100 bp) was present in all the seven isolatesof S. dysenteriae and absent in all the isolates ofS. flexneri and S. sonnei (Figure 7).

Antibiogram analysis of Shigella spp. isolatedfrom the river Narmada

The Antibiogram of the isolated Shigella spp. ispresented in the Table 2. Out of seven isolates ofS. dysenteriae, most of the strains showed resis-tance against ampicillin, amikacin, amoxycillin,streptomycin, tetracycline, cephotaxime, genta-mycin, and nalidixic acid. All S. dysenteriae strainswere susceptible to chloramphenicol and trimetho-prim. Likewise, out of 10 isolates of S. sonnei,most of the strains showed resistance againstamoxycillin and ampicillin and were susceptible toamikacin, chloramphenicol, streptomycin, norfloxacin,

ARTICLE IN PRESS

Table 2. Antibiogram analysis of Shigella spp. isolatedfrom different stations of river Narmada

BGCC No. Species Antibiotic resistanceprofile

BGCC382 S. flexneri A, AmBGCC383 S. flexneri A, AmBGCC384 S. dysenteriae A, Am, Cae, Na, Ce,

G, Ak, Cf, T, SBGCC385 S. flexneri TBGCC386 S. sonnei A, AmBGCC387 S. dysenteriae Am, GBGCC388 S. sonnei A, AmBGCC389 S. flexneri S, TBGCC390 S. dysenteriae Na, Ak, NxBGCC391 S. flexneri C, TBGCC392 S. flexneri C, TBGCC393 S. sonnei Am, A, Na, TrBGCC394 S. flexneri A, AmBGCC395 S. flexneri A, TrBGCC396 S. flexneri TrBGCC397 S. flexneri Tr, TBGCC398 S. flexneri A, TrBGCC399 S. dysenteriae A, Am, AkBGCC400 S. sonnei A, AmBGCC401 S. sonnei A, AmBGCC402 S. flexneri A, Am, TBGCC403 S. sonnei A, AmBGCC404 S. flexneri A, Am, TrBGCC405 S. sonnei NaBGCC406 S. flexneri A, Am, Cae, Na, TrBGCC407 S. flexneri Cae, NaBGCC408 S. flexneri CaeBGCC409 S. flexneri Cae, Na, TBGCC410 S. dysenteriae Na, Ce, SBGCC411 S. flexneri A, Am, GBGCC412 S. flexneri A, Am, Cae, NaBGCC413 S. dysenteriae Am, G, Ce,BGCC414 S. dysenteriae Am, G, Ce, TBGCC415 S. sonnei Am, T, GBGCC416 S. flexneri A, T, GBGCC417 S. sonnei G, T, NaBGCC418 S. flexneri Cae, Na, SBGCC419 S. flexneri Cae, Na, SBGCC420 S. sonnei G, T, NaBGCC421 S. flexneri A, Am, Cae, S

BGCC: Bacterial Germplasm Collection Centre.A – ampicillin, Am – amoxycillin, Ak – amikacin, Ce –

cephotaxime, Cae – ceftazidime, C – chloramphenicol, Cf –

ciprofloxacin, G – gentamycin, Nx, norfloxacin, Na – nalidixicacid, S – streptomycin, T – tetracycline, and Tr – trimethoprim.

Phenotypic and genotypic characterization of Shigella spp. 41

ceftazidime, and cephotaxime. Out of 23 isolates,most of the S. flexneri showed resistance toamoxycillin, ampicillin, ceftazidime, trimethoprim,nalidixic acid, tetracycline, and were susceptibleto cephotaxime, amikacin, ciprofloxacin, and nor-floxacin.

Discussion

Understanding the ecology of Shigella had beenlimited mainly due to the lack of suitable techni-ques to detect the presence of Shigella in environ-ment samples (Faruque et al., 2002). In the presentstudy, we used molecular techniques as well asconventional typing methods including biotyping,serotyping to detect Shigella in river waters withspecial reference to its virulence genes andantibiogram analysis.

We standardized the assay by culturing theenvironmental water samples and simultaneouslyconducting PCR tests. In a similar study by Faruqueet al. (2002), ipaH gene was used as an indicatortool to detect the presence of Shigella in environ-mental waters. In our study, the isolates weresubjected to detect the presence of ipaH gene andalso for the plasmid-associated virulence genesincluding ipaBCD and stx1 genes.

Fresh contamination of surface water by fecalmaterial of dysentery patients is a possibility indeveloping countries where sanitation is poorresulting in the presence of Shigella in surfacewater. Several previous studies have also detectedShigella in surface waters or sewage samples(Faruque et al., 2002; Obi et al., 2004) and haveindicated that Shigella strains can possibly betransported by surface waters (Mathan et al.,1984). However, in the present study the demon-stration of absence or deletion of virulence genes inenvironmental isolates of Shigella suggested thatthe presence of these strains were not possible dueto fresh fecal contamination of river water butinstead, these strains may have survived andpersisted in the aquatic environment. It appearsthat since in the environment the invasion asso-ciated gene did not have any known function, thebacteria apparently lost some of the plasmid-encoded genes. However, a chromosomally locatedmulticopy virulence genes, ipaH, which is alsoknown to have a role in producing invasivecharacteristic (Faruque et al., 2002) was found tobe more stable and was present in all theenvironmental strains analyzed. Therefore, PCRscreening of these strains for ipaH genes shouldprovide a better indicator of possible presence ofShigella than screening for the plasmid caringipaBCD genes.

The presence of Shigella in river waters, asdemonstrated in the present study, may have publichealth implications. Although most of the environ-mental strains lacked one or more of the knownvirulence genes, the environmental strains sharedan amplified ribosomal DNA restriction analysis(ARDRA) type with standard strains and belonged

ARTICLE IN PRESS

A. Sharma et al.42

to the three species, which are associated with themost cases of Shigellosis in India and many otherdeveloping countries. The detection of Shigellastrains in the river waters thus appears significant.In the present investigation, ARDRA with a probecomplementary to 16S rRNA was performed. Unlikethe ARDRA, where the differentiation was speciesspecific, in RAPD the isolates were found to bedistributed in all the clusters, depending on thebanding pattern. This observation establishes thesuperiority of ARDRA over RAPD in establishing thegenetic differentiation among the fresh waterisolates of the river Narmada.

Antimicrobial resistance in enteric pathogens isof great significance in the developing world,where the rate of diarrheal diseases is highest.The progressive increase in antimicrobial resistanceamong enteric pathogens in developing countries isbecoming a critical area of concern. The acutediarrheal diseases for which antimicrobial therapyis clearly effective include shigellosis, cholera, andcampylobacteriosis.

It is important to note that most of the strainsmay thus also serve as reservoirs of drug resistancegenes. It is also worth noting that none of the S.dysenteriae strains were resistance to chloramphe-nicol and trimethoprim, and none of the S. sonneiwere resistance to amikacin, chloramphenicol,streptomycin, norfloxacin, ceftazidime, and cepho-taxime and none of S. flexneri were resistance tocephotaxime, amikacin, ciprofloxacin, and norflox-acin which can be used for the treatment ofshigellosis.

Moreover, many studies have been carried outalmost exclusively with clinical strains; while therehave been few studies on the pathogenic aspects ofenvironmental strains. Since the aquatic environ-ment is implicated as the reservoir for thesemicroorganisms, and consequently responsible fortheir transmission in humans, it is obvious thatdetailed studies on the pathogenic potential of theenvironmental Shigella strains will certainly con-tribute to understanding the virulence properties ofthese bacteria and to establish the importance ofthese significant pathogens of aquatic systems.

Acknowledgment

Authors are thankful to the Head, Department ofBiosciences, R.D. University, Jabalpur, for providinglab facilities and MEF, New Delhi, for financialassistance.

References

Bryant TN. Probablistics identification of bacteria. PIBcomputer kit, Medical statistics and computing,University of Southampton, Southampton GeneralHospital, Southampton, 5094 XYUX, 2003.

CLSI. Clinical Laboratory Standard Institute. NCCC, vol.9(22), 2002. p. 1.

Edwards PR, Ewing WH. Identification of the Enterobac-teriaceae; Burgess Publication Co. Minneaposis. Epi-demiology 1972;2:26–30.

Faruque SM, Khan R, Kammuzzaman M, Tamasaki S,Ahmad QS, Azim T, et al. Isolation of Shigelladysenteriae type 1 and S. flexneri strains from surfacewater in Bangladesh; comparative molecular Anlysis ofEnvironmental Shigella versus clinical strains. ApplEnviron Microbiol 2002;68(8):3908–13.

Frankel G, Riley L, Giron JA, Valmassoi J, Friedmann A,Strockbine N, et al. Detection of Shigella in fecesusing DNA amplification. J Infect Dis 1990;161:1252–6.

Kotloff KL, Winickoff JP, Ivanoff B, Clemens JD, SwerdlowDL, Sansonetti PJ, et al. Global burden of Shigellainfection: implications for vaccine development andimplementation of control stretegies. Bull WHO1999;77:651–66.

MacFaddin FJ. Biochemical tests for identification ofmedical bacteria, vol. 1. Baltimore, USA: Williams andWilkins Co.; 1980.

Mathan VI, Bhat Kapadia CR, Panniah J, Baker SJ.Epidemic dysenteriae caused by Shiga bacillus in asouthern Indian village. J Diarr Dis Res 1984;2:27–32.

Newland JW, Neill RJ. DNA probes for Shiga like toxin Iand II and for toxin-converting bacteriophages. J ClinMicrobiol 1998;26:1292–7.

Oberhelman RA, Kopecko DJ, Venkatesan MM, Salazar-Lindo E, Gotuzzo E, Yi A, et al. Evaluation of alkalinephosphatase-labelled ipaH probe for diagnosis ofShigella infections. J Clin Microbiol1993;31(8):2101–4.

Obi CL, Bessong PO, Momba MNB, Postgieter N, Samie A,Igumbor EO. Profiles of antibiotic susceptibilities ofbacterial isolates and physico-chemical quality ofwater supply in rural Venda communities, SouthAfrica. Water SA 2004;30(4):4378–738.

Peirano G, Souza FS, Rodrigues DP. Frequency of seroversand antimicrobial resistance in Shigella spp. fromBrazil. Mem Inst Oswaldo Cruz Rio de Janeiro2006;101(3):245–50.

Sambrook J, Fritcsh EF, Maniatis T. Molecular cloning.(2nd ed.). Cold Spring Harbor, NY: Cord Spring HarbourLaboratoty Press; 1989. p. 18.47–18.59.

Silvia YB, Gloria RF, Martinez MB, Trabulsi LR, Moreira–

Filho CA. Characterization of enteroinvasive Escher-ichia coli and Shigella strains by RAPD analysis. FEMSMicrobiol Lett 1998;165:159–65.

Venkatesan MM, Buysse JM, Dennis JK. Chracterization ofinvasion plasmid antigen genes (ipaBCD) from Shigellaflexneri. Proc Natl Acad Sci USA 1988;85:9317–21.