SUCROSE LATE-FERMENTING VARIANTS OF V. CHOLERAE O1 AND O139 BENGAL ISOLATED IN MALAYSIA.

Najmiyatul Fadilah Mohamad and Quazi Manjurul Haque ^A

ABSTRACT

Clinical isolates of 26 strains of Vibrio cholerae O1 and O139 were studied for the demonstration of sucrose fermentation on thiosulfate-citrate-bile salt (TCBS) agar medium. After 24 hours incubation at 37 °C, 15 strains of vibrios were able to ferment sucrose. Another 11 strains failed to show the reaction although they agglutinated with V. cholerae specific monoclonal antibodies. Vibrio cholerae O1 and O139 showed positive results after 48 hours and 72 hours respectively. Polymerase Chain Reaction (PCR) and Enzyme Linked Immunosorbent Assay (ELISA) were used to confirm the identification of these strains.

INTRODUCTION

Vibrio cholerae O1 and O139 are the most important agents that cause cholera. The capability of these microorganisms to ferment sucrose resulting in characteristic yellow colonies on thiosulfate-citrate-bile salt agar (TCBS) is considered the gold standard in its identification. TCBS is a selective medium that suppresses other bacterial species but permits the growth of V. cholerae to grow. This is due to the alkaline pH of the medium that kills most intestinal commensals. The yellow colouration of the colonies is caused by the fermentation of sucrose in the media. The importance of this reaction is to differentiate V. cholerae with other sucrose non-fermentor vibrios such as V. parahaemoliticus. However, there are reports that claim that this microorganism does not give the expected appearance on culture^{1, 2}.

MATERIAL AND METHODS

Bacterial strains

Twenty six clinical isolates of V. cholerae (O1 and O139) were studied. Twenty strains were provided by the Institute of Medical Research, (IMR) Kuala Lumpur and the rest by the Microbiology Laboratory of the Hospital Tengku Ampuan Afzan (HTAA) in Kuantan, Pahang. They are presented in Table 1. Five control strains of V. cholerae O1 and O139 were kindly provided by the Department of Microbiology, School of Medicine, Hospital University Sains Malaysia (HUSM) in Kubang Kerian, Kelantan. Originally these strains were clinical isolates from Japan.

Bacteriology and Serogrouping

Bacteria were grown on TCBS agar plates (OXOID) for colony isolation. However, they were enriched in alkaline peptone water (APW) and incubated at 37 °C for 18 hours before subculture on TCBS agar. The plates were incubated overnight at 37 °C. Serogrouping was performed with polyvalent O1 antiserum, mono-specific Ogawa, Inaba antisera and specific O139 antiserum (Denka Seiken, Tokyo, Japan). The bacteria were also studied in a battery of standard biochemical tests^{3, 4} and also tested in the API 20E biochemical systems (API Systems, BioMerieuxe).

PCR assay for ctxA gene

A PCR assay was performed to detect the ctxA genes with 2 primers:

1.      ctxA 1 (5^¢-CTCAGACGGGATTTGTTAGGCACG-3^¢) and

2.      ctxA 2 (5^¢-TCTATCTCTGTAGCCCCTATTACG-3^¢)

The 2 primers produced an amplicon of 301 base pairs of DNA band^{5, 6}.  The PCR technique was carried out in 0.5 ml microcentrifuge tubes, with 50ml of reaction mixture consisting of 26.75ml of sterile water, 5ml 10X PCR Buffer II, 5ml 25mM MgCl₂ Solution, 1ml 10mM deoxyribonucleotide-phosphate (dNTPs) each, 2ml of each primer (ctx I and ctx II), 0.25ml Amplitaq Gold (Applied Biosystems by Roche Molecular System, Inc) and 5ml of plasmid DNA. The solution mixtures were placed in a DNA thermocycler GeneAmp PCR system 9700 (PE Applied Biosystem, Norwalk, Connecticut, USA). Control reaction of plasmid DNA was replaced by 5ml of ultra pure water and amplification was subjected to V. cholera O1 and O139 strains. The cycling conditions were as follows: initial denaturation at 95°C for 2 min, followed by 30 cycles consisting of 94°C for 1 min for denaturation, 55°C for 1 min for annealing, and 72°C for 1 min for elongation and a final extension step at 72°C for 10 minutes ⁷.

For the detection and confirmation of PCR products by gel electrophoresis, 10 ml of the amplified product mixture was subjected to electrophoresis through 1% agarose gel (Nacalai Tesque, Inc. Kyoto, Japan), stained with ethidium bromide and photographed under UV fluorescence (Polaroid. Gel Cam. Standard Electrophoresis Hood. BIO-RAD). Control stains used were V. cholerae O1 and O139 (positive for CT) and reaction mixtures containing distilled water and other reagents but not template for negative control.

Enzyme Linked Immunosorbent Assay (ELISA)

Cholera toxin (CT) was detected from V. cholerae using Protein Detector™ ELISA Kit, HRP, ABTS System (Kirkegaard Y Perry Laboratories, Inc. USA). Culture filtrates of all isolates were prepared as described by Sanyal et al. 1984⁸ with some modification. The culture filtrates was made from AKI and APW broths with and without addition of polymixin B⁹. Serial of 10-folds dilution of the culture filtrates was prepared. The presence of CT was detected by the following procedure: wells of microtitre plates (Thermo Labsystems, Dynex) were coated with 200 ml prepared antibody consisting of 1ml monoclonal antibody of cholera toxin, ab8248-100 (Abcam Limited, Cambridge, UK), diluted in 1/10 coating buffer solution which had undergone 1 hour incubation at room temperature before overnight incubation at 4°C. The solution was decanted before 300 ml diluted Blocking Solution (BSA 1% in PBS) was added to each well at room temperature. After 5 minutes reaction, residual liquid was tapped out and 2 ml of vibrios filtrates were added in each well (5 fold dilutions). This was followed by the addition of 100 ml Antibody Conjugate Solution (HRP Anti-Mouse IgG, H+L) to each well and incubation for 1 hour at room temperature. After all liquid were tapped out, each well was washed 4 times with wash solution (100 ml), each for 4 minutes. Enzyme substrate (100 ml) was added to each well for 15 minutes before the reaction was stopped using 100 ml peroxidase stop solution. The tests were done in duplicate

A positive result for CT would have an absorbance value greater than or equivalent to 0.2 above the mean background value of 405 nm and could be seen with the naked eyes by the appearance of a blue-green colour. Control stains used were V. cholerae O1 and O139 (positive for CT) and reaction mixtures containing distilled water and other reagents but not vibrios filtrate for negative control. Medium assay controls were done using APW and AKI medium to ensure that no result was influenced by them. The toxin titer was defined as the interpolated dilution giving an enzyme-mediated light absorption (A_405nm) of 0.20 above background with substrate.

RESULTS

We used TCBS as a selective media for culturing the stool specimens. After 24 hours incubation at 37 °C, 15 colonies appeared yellow and 11 colonies appeared green (Table 1 and Figure 1). All yellow and green colonies were studied in a battery of standard biochemical tests^{3, 4} and also tested in the API 20E biochemical systems (API Systems, BioMerieuxe). All the results showed that these specimens were V. cholerae.

Further observation was done on the green colonies which appeared on TCBS. The colonies remained green after 24 hours but turned yellows after 48 hours for V. cholerae O1 and 72 hours for V. cholerae O139 when they were again cultivated in APW medium. The appearance of the yellow colonies could be seen on both TCBS agar and Triple Sugar Ion (TSI) slant agar.    

Slide agglutinations were performed using V. cholerae O1 and V. cholerae O139 antiserum (Denka Seiken, Tokyo, Japan). Both yellow and green colonies agglutinated with the V. cholerae specific monoclonal antibody. Further characterization by antigenic serotyping on V. cholerae O1 showed 4 Inaba, 10 Ogawa and 1 Hikojima serotypes (Table II).

The yellow and green colonies were analyzed by polymerase chain reaction (PCR). Both generated the expected 301-base pair’s amplicon by PCR with two primers, specific for the ctxA gene of the ctx operon of V. cholerae ⁵ (Table I). This result suggests that both yellow and green colonies were V. cholerae although only 15 strains could be amplified by PCR¹⁹ including the control strains. Cholera toxin did not show up 100% in PCR because we took plasmid DNA (6279.63 base pairs) that also encoded the gene¹⁹.

All 26 strains were tested for reactivity with monoclonal antibody specificity for cholera enterotoxin, ab8248-100 (Abcam Limited, Cambridge, UK) by ELISA method. All strains of V. cholerae O1 and O139 were positive for cholera enterotoxin (CT) until 1:100 dilution of cholera culture supernatants.

DISCUSSION

Vibrios are well defined on the basis of biochemical test activities and DNA homology, but the species are not homogeneous with regard to pathogenic potential^{10, 11}. Among the 206 currently known serogroups¹², only O1 and O139 strains are capable of causing epidemic cholera, but these two serogroups are rarely found in the environment except during large outbreaks. In this study, V. cholerae O1 and O139 were isolated from patients with diarrhoea during outbreaks of cholera, from a different geographical area in Malaysia.

Cultures on TCBS medium remain the ‘gold standard’ for cholera diagnosis and confirmation. However, non-sucrose fermenting strains that would not give the expected appearance on culture have been reported from Brazil¹. Ansaruzzaman et al. 1995² showed that sucrose late-fermenting and nonfermenting variants of V. cholerae O139 Bengal were found in Bangladesh. The incapacity of V. cholerae O1 and O139 to break down sucrose on TCBS agar plates as shown in this study was in agreement with this previous report (Table I and Figure I).

The most important test for identification of V. cholerae O1 or O139 is agglutination in antisera raised against them. Differences in the sugar composition of the heat-stable surface somatic “O” antigen are the basis of the serological classification of V. cholerae and in this study both yellow and green colonies agglutinated with V. cholerae O1 and O139 specific monoclonal antibodies, confirming the earlier diagnosis of V. cholerae. There were 15 and 11 strains of O1 and O139 vibrios respectively, with 4 Inaba, 10 Ogawa and 1 Hikojima serotypes (Table II) on further characterization done by antigenic serotyping onto V. cholerae O1.

Although demonstration of typical agglutination essentially confirmed the diagnosis, additional tests such as oxidase reaction, indole reaction, fermentation reaction, lysine, arginine and ornithine decarboxylase reaction also helps in the confirmation of V. cholerae. Inoculation of both yellow and green colonies in a battery of biochemical tests produced an identical biochemical reaction except for the green colonies which did not ferment sucrose (Figure I). The strains of V. cholerae O1 showed delayed sugar utilization after 48 hours, and 72 hours for O139 vibrios when they were again cultivated in APW medium. The yellow colonies can be seen both on TCBS agar and on TSI agar slant. APW is an enrichment broth, from which maximal recovery of V. cholerae can be obtained^{13, 14, 15}. V. cholerae O1 and O139 were tested in the API 20E System, a standard series of commercially available enteric identification tests and this resulted in identification code of 7166124 and 7162124, which corresponded to V. cholerae.

In the present study, biochemical tests did not show many discrepancies, but it is better to include as many tests as possible to confirm and identify an organism, For this reason, although the isolates were earlier identified as V. cholerae, reconfirmation by biochemical tests was performed which gave satisfactory results.

Polymerase Chain Reaction (PCR) refers to a process of amplifying one or more specific DNA sequences contained in a nucleic acid or mixture of nucleic acids of any origin (bacteria, virus, plant or human), up to 10⁶ fold or hundreds of millions of times. This requires a matter of hours to produce enough DNA to be adequately tested. PCR was invented by Kary Mullis in 1985 at Berkeley. It utilizes two primers which are complementary to the ends of each specific sequence to be amplified. Extension of each primer creates a DNA strand including a sequence complementary to the opposite primer as it is directed enzymatically to amplify the specific DNA sequence of interest¹⁶. Characterization of the yellow and green colonies was done using Cholera-toxin (CT) detection primers in PCR.

The results in this study showed that both colonies generated 301 base pairs amplicon by PCR, with two primers specific for the ctxA gene of the ctx operon of V. cholerae O1 and O139 ^{5, 6} including control strains. Amplification of the single plasmid DNA band by PCR confirmed the presence of CT-encoded genes in these bacteria. The bacteria were shown to be in association of the ctx gene, and were confirmed as V. cholerae.

However, 9 strains of V. cholerae O1 and 7 strains of V. cholerae O139 failed the PCR amplification in the present study. One explanation for the poor or no amplification of DNA could be that these primers may not have the complementary sequence on the template DNA to serve as binding sites. Secondly, they may have special requirements for amplification with respect to properties of the primers such as the G+C content¹⁷. Ramser et al. 1996¹⁸ suggested that lack of proper binding sites or poor quality of template DNA may also be one of the contributing factors. The failure of characterization of those single bands of plasmid in this study suggests that it should be amplified using another set of ctx primer. This is because an important factor in evaluating any DNA-based test is the specificity of the DNA sequences chosen for the gene and the strain of interest.

In addition, all V. cholerae strains were tested for cholera toxin production with monoclonal antibody specific for cholera enterotoxin, ab8248-100 (Abcam Limited, Cambridge, UK) using the ELISA method. All strains of V. cholerae O1 and O139 were positive for cholera enterotoxin (CT) up to 1:100 dilutions of cholera culture supernatants. This result supports the others results that both yellow and green colonies are V. cholerae.

The data suggest that it is possible to encounter sucrose non-fermenting and late-fermenting strains of V. cholerae O1 and O139 in clinical specimens because these strains appear as green colonies on TCBS agar. The failure of these V. cholerae O1 and O139 to ferment sucrose during routine plating on TCBS agar raised a new problem in laboratory diagnosis of V. cholerae. Reliance on appearance of only yellow colonies will miss these strains on TCBS and cause misinterpretation and confusion with V. parahaemolyticus (a sucrose nonfermenter).

We suggest that the morphology and size of the colonies of V. cholerae should be considered even when green colonies are obtained. The incubation time should be prolonged if the stool specimens are from suspected cholera cases. Taurocholate-tellurite-gelatin agar might be a better medium for V. cholerae because the strains will not be missed on it ². Other methods which can be used are ELISA and PCR.

The use of PCR is becoming increasingly important in the diagnosis of infectious disease. It is not only a rapid test but has the potential of being more specific and possibly more sensitive then currently available tests for the detection and identification of causative agents of infectious disease. However it is costly and not yet available in most hospitals. Furthermore, a critical factor in evaluating any DNA-based test is the specificity of the DNA sequence chosen for the gene and the strains of interest.

Another suggestion is that specific bands of plasmid of ctx toxin can be detected using Perfectprep Plasmid Mini Kit (eppendorf) with modification¹⁹. The detection of a single band of plasmids (100%) which can be amplified by PCR can give added information as 48.39% of the single band of plasmid has CT-encoded toxin. There is high uniformity of the plasmid profiles from the Malaysian isolates as shown in the study. However further studies are needed to confirm this finding.

CONCLUSION

Our data suggest that it is possible to encounter sucrose non-fermenting and late-fermenting strains of V. cholerae O1 and O139 in clinical specimens. These strains appear as green colonies on TCBS agar. The failure of these V. cholerae O1 and O139 to ferment sucrose during routine plating on TCBS agar raises a new problem in laboratory diagnosis of V. cholerae. Reliance on only the appearance of yellow colonies will cause these strains to be missed on TCBS and misinterpretation with V. parahaemolyticus (a sucrose non-fermenter). This problem can be overcome by performing as many tests as possible on specimens taken from suspected cholera patients, especially when the isolated colony on TCBS does not give the expected appearance.

ACKNOWLEDGEMENTS

We appreciate the assistance of Dr. M. Ravichandran from the Department of Microbiology, School of Medicine, Hospital University Science Malaysia (HUSM), Kubang Kerian, Kelantan for providing the control strains of vibrios.

Table I

V. cholera growth on TCBS

* (+) Positive reaction to PCR

 All 5 control strains were positive to PCR (USM I: V. cholerae O139; USM II: V. cholerae O1 El Tor [Ogawa]; USM III: V. cholerae O1 Classical [Inaba]; USM IV: V. cholerae O1 Classical [Ogawa]; USM V: V. cholerae O1 El-Tor [Inaba].

Table II

V. cholerae serotyping

                                                                                        ^* (+) Positive reaction to antisera

Figure I

Late fermentation occurred on TCBS plate representing V. cholerae

REFERENCES:

1.      Ramos FL, Lins-Lainson ZC, Da Silva EL, Proietti Junior AA, Mareco ML, and            Lamarao ML, 1997. Cholera in North Brazil: On the Occurrence of Strains of V. cholerae O1 Which Fail to Ferment Sucrose During Routine Plating on Thiosulphate-Citrate-Bile Salt-Sucrose Agar (TCBS). A new problem in diagnosis and Control? Revista Latinoamericana de Microbiologia. 39 (3-4):141-4.

2.      Ansaruzzaman M, Rahman M, Kibriya K.M.G, Bhuiyan N.A, Islam M.S, and Albert M.J. 1995. Isolation of sucrose late-fermentating and non-fermentating variants of V. cholerae O139 Bengal: implications for diagnosis of cholera. Journal Clinical Microbiology. Volume 33: 1339-40.

3.      Centres for Disease Control and Prevention. Laboratory methods for the diagnosis of Vibrio cholerae. Atlanta, Georgia: CDC 1994.

4.      Barrow GI, Feltham RKA. In Cowan and Steel’s. Manual for identification of medical bacteria (3^rd ed). Cambridge, 1993; p: 42.

5.      Shirai H., Nishibuchi M., Ramamurthy T., Bhattacharya S. K., Pal S. C., Takeda Y. 1991. Polymerase Chain Reaction for the Detection of the Cholera Enterotoxin Operon of Vibrio cholerae. Journal of Clinical Microbiology. Volume 29:2517-2521.

6.      Tsai SC, Noda M, Adamik R, Moss J, and Vaughan M. Proc. Natl. Acad. Sci. USA 1987. Volume 84: p: 5139-5142.

7.      Keasler SP, Hall RH. 1993. Detecting and Biotyping Vibrio cholerae O1 with Multiplex Polymerase Chain reaction. Lancet. Volume 341: 1661.

8.       Sanyal, S. C., P. K. B. Neogi, K. Alam, M. I. Huq, and K. A. Al-Mahmud, 1984. A new enterotoxin Produced by V. cholerae O1. Journal Diarrh. Dis. Res. Volume 2:3-12

9.      Quazi Manjurul Haque, Akira Sugiyama, Yoshito Iwade, Yutaka Midorikawa, Toru Yamauchi, 1996. Diarrheal and Environmental Isolates of Aeromonas spp. Produce a Toxin Similar to Shiga-Like Toxin 1. Current Microbiology. Volume 32: 239-245.

10.  J. M. Janda 1998. Vibrio, Aeromonas and Plesiomonas. In L. Collier (ed.). Topley & Wilson’s Microbiology & Microbial Infections. 9^th Edition. Oxford University Press, New York, N.Y.

11.  David L. Tison. 1999. Vibrio. Bergeys Manual of Microbiolgy. 17^th Edition. In Murray PR (ed). Washington DC, ASM Press. 497-506

12.  Shimada TE, et al. 1994. Extended serotyping scheme for Vibrio cholerae. Current Microbiology. Volume 28:175-178.

13.  Murad Lesmana, Emily Richie, Decy Subekti, Cyrus Simanjuntak, Stephen E. Walz 1997. Comparison of Direct-Plating and Enrichment Methods for Isolation of Vibrio cholerae from Diarrhea Patients. Journal of Clinical Microbiology. Volume 37(7): 1856-1858.

14.  Murad Lesmana, Rockhill RC, Sutanti D, Sumoto A. 1985. An Evaluation of Alkaline Peptone Water for Enrichment of Vibrio cholerae in Feces. Southeast Asian Journal Tropica Medicine Public Health. Volume 16(2):265-7.

15.  Renels MB, Levine MM, Daya V, Angle P, Young C. 1980. Selective vs. Nonselective Media and Direct Plating vs. Enrichment Technique in Isolation of Vibrio cholerae: Recommendations for Clinical Laboratories. Journal Infections Diseases. Volume 142(3):328-31.

16.  Laimore, T. C. 1990. Method: Polymerase Chain Reaction.  http://hdklab.wusb.edu/lab_manual/pcr/pcrl.html

17.  Fritsh, P., M. A. Hanson, C. D. Spore, P. E. Pack, L. H. Rieseberg 1993. Constancy of RAPD Primer Amplification Strength Among Distantly Related Taxa of Flowering Plants. Plant Molecular Biology Rep. Volume 11: 10-20.

18.  Ramser, J., C., Lopez-Peraita, R. Wetzel, K. Weising, G. Kahl 1996. Genomic Variation and Relationships in Aerial Yam (Disocorea bulbifera L.) Detected by Random Amplified Polymorphic DNA. Genome. Volume 39:17-25.

19.  Najmiyatul Fadilah Mohamad, Quazi Manjurul Haque 2002. Detection of Plasmids from Vibrio cholerae Strains Isolated in Malaysia using The Perfectprep Plasmid  Mini Kit^Ò. The International Medical Journal. Volume 2(1).

! Department of Basic Medical Science, Faculty of Medicine, International Islamic University Malaysia.