Rapid and Sensitive Assay for Detection of Enterotoxigenic ...

2 downloads 0 Views 321KB Size Report
Aug 4, 1998 - specific for the lipopolysaccharide of B. fragilis, an assay based on immunomagnetic separation (IMS) in combination with PCR (IMS-PCR) was ...
JOURNAL OF CLINICAL MICROBIOLOGY, Dec. 1998, p. 3545–3548 0095-1137/98/$04.0010 Copyright © 1998, American Society for Microbiology. All Rights Reserved.

Vol. 36, No. 12

Rapid and Sensitive Assay for Detection of Enterotoxigenic Bacteroides fragilis GUANGMING ZHANG

AND

ANDREJ WEINTRAUB*

Department of Immunology, Microbiology, Pathology and Infectious Diseases, Division of Clinical and Oral Bacteriology, Huddinge University Hospital, Karolinska Institute, S-141 86 Huddinge, Sweden Received 30 June 1998/Returned for modification 4 August 1998/Accepted 3 September 1998

Bacteroides fragilis is an obligatory anaerobic, gram-negative bacterium found among the normal intestinal flora of humans. Enterotoxigenic strains of B. fragilis (ETBF) have been associated with diarrheal diseases in humans and animals. The enterotoxin of ETBF induces fluid changes in ligated intestinal segments and cytotoxic response in HT29/C1 cells. By using a pair of monoclonal antibodies (MAbs; MAb C3 and MAb 4H8) specific for the lipopolysaccharide of B. fragilis, an assay based on immunomagnetic separation (IMS) in combination with PCR (IMS-PCR) was developed. After DNA extraction, a 294-bp fragment was amplified. The specificity of the IMS-PCR assay was evaluated by adding previously isolated and confirmed ETBF strains to normal fecal samples. All fecal samples to which ETBF strains were added were positive, showing a 100% specificity. The spiked fecal samples were also used for evaluation of the sensitivity of the assay. The detection limit was found to be ;50 CFU/g of feces. By this method 10 clinical fecal samples (5 from patients with diarrhea and 5 from healthy controls) were examined. The results of PCR were in accordance with the results of the HT29/C1 cell assay for all samples. The minimum time to retrieval of the final result by the IMS-PCR method is 36 h. The proposed IMS-PCR assay is rapid and sensitive for the direct detection of ETBF in stool samples.

Bacteroides fragilis, an obligatory anaerobic bacterium found in high numbers among the normal intestinal flora of humans, is recognized as an important cause of intra-abdominal infections in humans (3). However, neither diarrheal disease nor extracellular toxin production due to B. fragilis was appreciated until 1984, when Myers and coworkers (4, 5) found that some strains of B. fragilis were associated with diarrheal diseases in young farm animals and humans by producing a factor with enterotoxigenic activity. Recent studies have shown that the rate of enterotoxigenic B. fragilis (ETBF) carriage is high both in adults and in children, regardless of whether diarrhea is present (7). The role of ETBF as a diarrheal agent has been described previously (2, 8, 9, 13, 14). It has also been suggested that ETBF may be endemic in communities (10). Therefore, more investigations on the identification, epidemiology, and pathogenic role of ETBF are necessary. PCR has been used to identify ETBF in pure cultures and directly from feces (6, 11). The assays are based on amplification of the enterotoxin gene, which encodes a zinc-binding metalloprotease (2). Pantosti et al. (6) used primers deduced from the sequence of the gene in an assay in which stool samples were inoculated on selective media followed by a sweep of total growth. The sensitivity was 104 to 105 CFU/g of feces (6). Recently, Shetab et al. (11) published a description of a nested PCR assay for the detection of the B. fragilis enterotoxin gene with a detection limit of 100 to 1,000 CFU/g of stool. However, false-negative results may occur if PCR inhibitors, which are commonly present in stool samples, are not removed prior to amplification (16). The aim of the present study was to use well-characterized specific monoclonal antibodies (MAbs) coated on magnetic beads and subsequent spe-

cific capture of bacteria by immunomagnetic separation (IMS) in order to concentrate and separate the bacteria. Here, we describe an IMS-PCR assay, based on the use of B. fragilisspecific MAbs, followed by PCR amplification of the enterotoxin gene for the rapid, sensitive, and specific detection of ETBF directly from clinical samples. MATERIALS AND METHODS Bacterial strains. Clinical isolates of 21 ETBF strains were provided by J. M. Albert from the International Centre for Diarrhoeal Disease Research, Bangladesh, in Dhaka, Bangladesh. All the ETBF strains produced enterotoxin, as shown in an assay with the HT29/C1 cell line (see below). The nonenterotoxigenic Bacteroides strains investigated were B. fragilis NCTC 9343 and B. thetaiotaomicron NCTC 10582 (National Collection of Type Cultures, London, United Kingdom); B. fragilis VPI 5631, B. fragilis VPI 4225, B. fragilis VPI 4117, and B. fragilis VPI 2552 (Virginia Polytechnic Institute and State University, Blacksburg); and B. fragilis ATCC 23745, B. ovatus ATCC 8483, and B. vulgatus ATCC 8482 (American Type Culture Collection, Rockville, Md.). In addition, Salmonella typhi IS 501, Shigella flexneri 4bR, Escherichia coli (serotypes O:35, O:104, O:136, and O:138), and Vibrio cholerae O139 strains were from the strain collection at the Division of Clinical and Oral Bacteriology, Huddinge University Hospital. All anaerobic strains were grown on blood agar for 18 to 40 h at 37°C under anaerobic conditions (GasPak; BBL Microbiology Systems, Cockeysville, Md.). Aerobic bacterial strains were cultured at 37°C on blood agar plates. Production and purification of MAbs. The MAbs used in this study were basically prepared as described earlier (1). Two different MAbs were used: MAb 4H8 (immunoglobulin G3 [IgG3]), which binds to an immunodominant epitope in the lipopolysaccharide (LPS) of a majority of B. fragilis isolates (17), and MAb C3 (IgG2b), which most likely binds to a common epitope present in the inner core region of B. fragilis LPS (18). Both MAbs were purified from serum-free medium with a protein A column (Sigma Chemical Company, St. Louis, Mo.). Coagglutination test. Staphylococcus aureus Cowan I bacteria (Sigma) were sensitized individually with 0.1 ml of MAb C3 (1 mg/ml) and 0.1 ml of MAb 4H8 (1 mg/ml). Two to five colonies of the bacterial strains to be investigated were suspended in 0.05 ml of 2% coagglutination test reagent on a glass slide, which was rocked back and forth for 2 min (12, 17). Agglutination was registered during the period and was recorded as 21 when the agglutination was clear by examination with the naked eye and as 11 if a magnifying glass was needed for observation. Readings of both 11 and 21 are regarded as positive results. Preparation of spiked fecal samples and clinical fecal samples for IMS. ETBF-negative stool samples were collected. In order to estimate the sensitivity

* Corresponding author. Mailing address: Division of Clinical and Oral Bacteriology, F82, Huddinge University Hospital, S-141 86 Huddinge, Sweden. Phone: 46-8 5858 7831. Fax: 46-8 711 3918. E-mail: [email protected]. 3545

3546

ZHANG AND WEINTRAUB

of the IMS-PCR assay under experimental conditions, serial 10-fold dilutions of the ETBF D-94 strain were prepared in phosphate-buffered saline (PBS), and the mixture was added to a preweighed stool sample in order to obtain concentrations ranging from 106 to 10 CFU/ml. Briefly, 0.1 ml of diluted strain D-94 was added to 0.9 ml of diluted feces (1 g/5 ml in PBS). Similar procedures were also followed with ETBF-negative stool samples spiked with other strains tested in the study (see Table 2). The spiked fecal samples (ca. 1 ml) were inoculated into prereduced peptone yeast glucose (PYG) broth medium (4.5 ml) containing 0.05% (wt/vol) kanamycin (Sigma), and the mixture was incubated at 37°C for 24 to 48 h. After incubation, the broth medium was centrifuged at 600 3 g for 10 min. The supernatant was collected and centrifuged at 3,000 3 g for 10 min, and the pellet was suspended in 80 ml of PBS for incubation with magnetic beads. Clinical fecal samples from patients with diarrhea and from healthy controls were collected and inoculated into tubes containing PYG and kanamycin by the same procedure described above. IMS. Rat anti-mouse IgG2b-coated Dynabeads M450 and sheep anti-mouse IgG-coated Dynabeads M280 (Dynal, Oslo, Norway) were used. The amounts of the MAbs required for coating of the immunomagnetic particles were determined according to the manufacturer’s recommendations, with minor modifications. Briefly, 1 ml of MAb C3 (1 mg/ml) was mixed with 25 ml of rat anti-mouse IgG2b-coated Dynabeads M450 (4 3 108 beads/ml), and 1.5 ml of MAb 4H8 (1 mg/ml) was mixed with 25 ml of sheep anti-mouse IgG-coated Dynabeads M280 (6 3 108 to 7 3 108 beads/ml). The volumes were adjusted to 100 ml with PBS, and the beads were incubated overnight with bidirectional rotation at 4°C. The coated beads were washed three times with PBS containing 0.05% Tween 20 (Merck, Schuchardt, Munich, Germany) and were resuspended in PBS with 0.5% bovine serum albumin (Sigma) at 107 beads/ml. The beads were stored at 4°C until they were used. The coated beads (12.5 ml each) were transferred to a 96-well microtiter plate (Techne, Cambridge, United Kingdom). Spiked fecal samples (80 ml) were added to the wells, and the plate was incubated with shaking for 1 h at room temperature. With the aid of the magnetic separator (Beadprep 96; Techne), the beads with bound bacteria were washed three times with 150 ml of PBS–0.05% Tween 20, for 2 min each time, with gentle shaking. The beads were suspended in 70 ml of Milli Q water. Fifty microliters was transferred to an Eppendorf tube (1.5 ml) and boiled for 10 min, and the tube was centrifuged at 20,000 3 g for 5 min at 4°C (Microcentrifuge 154; Ole Dich, Huidovre, Denmark). The supernatants were collected and kept at 4°C for subsequent PCR analyses. PCR assay. The primers used in the study were those described by Pantosti et al. (6). The forward primer (BF1; 59-dGACGGTGTATGTGATTTGTCTG AGAGA-39) and the reverse primer (BF2; 59-dATCCCTAAGATTTTATTAT TATCCCAAGTA-39) were synthesized by Interactiva Biotechnologie GmbH (Ulm, Germany). The specific amplified 294-bp band of DNA is expected. Twenty microliters of the supernatant containing bacterial DNA was added to a mixture containing 13 PCR buffer with each deoxynucleoside triphosphate (Perkin-Elmer, Norwalk, Conn.) at a concentration of 200 mM, 1.25 U of AmpliTaq DNA polymerase (Perkin-Elmer) per 50 ml, and 20 pmol of each primer in a final volume of 50 ml of enzyme buffer containing 2.5 mM MgCl2 (PerkinElmer). The amplification was subjected to 35 cycles with a GeneAmp PCR System 9600 (Perkin-Elmer), which consisted of 5 min of denaturation at 94°C, 1 min of annealing at 58°C, 1 min of extension at 72°C, and a final 5-min extension at 72°C. For PCR, ETBF D-134 was used as a positive control. Negative controls were (i) water instead of template DNA (reagent control) and (ii) B. fragilis NCTC 9343 (a non-ETBF strain) template DNA (specificity control). PCR products (10 ml) were evaluated by electrophoresis with a 1.5% (wt/vol) agarose gel (Gibco Life Technologies, Paisley, United Kingdom) containing ethidium bromide at 90 mV for 60 min. A molecular mass marker (Marker V; Boehringer Mannheim GmbH, Mannheim, Germany) was run concurrently. The DNA bands were visualized and photographed under UV light. Enterotoxin assay with HT29/C1 cells. To assess whether enterotoxin was produced, after the separation from each fecal sample 20 ml of the magnetic beads was inoculated into brain heart infusion broth (BHI) supplemented with hemin (5 mg/ml), and the mixture was incubated at 37°C for 48 h in an anaerobic cabinet. The broth cultures were centrifuged at 3,000 3 g for 10 min, and the supernatants were filtered through a 0.20-mm-pore-size filter (Sartorius AG, Go ¨ttingen, Germany). The filtered supernatants were frozen immediately and were kept at 220°C until use. The cytotoxicity assay with HT29/C1 cells was performed as described earlier by Weikel et al. (15). Briefly, HT29/C1 cells were grown in Dulbecco’s modified Eagle’s medium (Gibco), trypsinized with 0.005% trypsin–0.05 mM EDTA (Sigma), and resuspended at 200 ml/well in a 96-well microtiter plate (Costar, Cambridge, Mass.). The cells were allowed to grow for 3 days until discrete clusters of cells were visible. The medium was replaced with 200 ml of fresh medium without fetal calf serum. The filtered bacterial culture supernatant (100 ml) was added to the wells in duplicate. The plate was incubated at 37°C in the air with 10% CO2 and was examined after 3 h for the presence of the typical toxininduced cytopathic changes. Strain ETBF D-134 was used as a positive control. Negative controls included medium alone and filtered supernatant from the nonenterotoxigenic strain B. fragilis NCTC 9343.

J. CLIN. MICROBIOL. TABLE 1. Results of coagglutination tests with ETBF strains and specific anti-B. fragilis MAbs absorbed to protein Acontaining staphylococci ETBF strain

D-2 D-50 D-62 D-64 D-73 D-82 D-87 D-93 D-94 D-99 D-100 D-104 D-107 D-111 D-115 D-117 D-128 D-133 D-134 D-149 D-159

Result with the following MAb: MAb C3

MAb 4H8

1 1 1 1 1 1 1 2 1 1 1 1 1 1 2 1 1 1 1 1 1

1 1 2 1 2 1 1 1 1 2 1 1 2 1 1 1 2 1 1 1 1

RESULTS Coagglutination tests of ETBF strains with two MAbs. Both coagglutination test reagents were used for all 21 ETBF strains (Table 1). All strains except ETBF D-93 and ETBF D-115 were positive by tests with MAb C3. However, these two strains were positive when MAb 4H8 was used. By tests with both MAbs, all 21 strains were correctly identified as B. fragilis. Fourteen strains reacted with both coagglutination test reagents (Table 1). Sensitivity of IMS-PCR for detection of ETBF. The sensitivity of the IMS-PCR assay was investigated with fecal samples spiked with 10-fold dilutions of ETBF D-94. The DNA from bacterial cells isolated by IMS was subjected to PCR. As determined by electrophoresis, the lowest original dilution of ETBF bacteria that could be detected by IMS-PCR was 102 CFU/ml (Fig. 1). This corresponds to ;50 CFU per g of fecal sample. No amplification of DNA from nonspiked fecal samples could be seen. Detection of ETBF in spiked fecal samples by IMS-PCR. Fecal samples spiked with different strains were tested. PCR

FIG. 1. IMS-PCR products from spiked fecal samples prepared with serial dilutions of ETBF D-94. Lanes 1 to 6, dilutions from 106 to 10 CFU/ml, respectively; lane 7, positive control; lane 8, negative control; lanes M, DNA molecular mass markers (Marker V; Boehringer Mannheim). The 294-bp product correlates with the amplified portion of the enterotoxin gene.

DETECTION OF ENTEROTOXIGENIC B. FRAGILIS

VOL. 36, 1998

3547

TABLE 2. Specificity of IMS-PCR assay with MAb C3 and MAb 4H8 for detection of ETBF in spiked fecal samples No. of isolates or samples tested

Strain or samples tested

ETBF Nonenterotoxigenic B. fragilis Other Bacteroides species Salmonella typhi IS 501 Shigella flexneri 4bR Escherichia coli Vibrio cholerae O139 Fecal samples (patients) Fecal samples (controls)

21 6 3 1 1 4 1 5 5

followed by electrophoresis showed that only in tests with the ETBF strains listed in Table 2 was the amplification product detected (294 bp) (Fig. 2). None of the fecal samples spiked with other bacterial strains was positive. Nonspiked fecal samples were also negative. Detection of ETBF in fecal samples by IMS-PCR and HT29/ C1 cell assay. Five fecal samples from patients with diarrhea and five fecal samples from healthy controls were tested by the IMS-PCR assay. Amplification products were found in tests with three of five samples from patients with diarrhea (Fig. 2). The samples from healthy controls were negative. The clinical samples were also tested by the HT29/C1 cell assay after growth in brain heart infusion medium. All three samples from the patients positive for the amplification product by IMS-PCR caused alterations in the morphologies of HT29/C1 cells. Both the IMS-PCR-negative samples from the patients with diarrhea and those from the healthy controls were negative by the HT29/C1 cell assay. DISCUSSION ETBF has been considered to be an important cause of diarrheal disease in humans as a result of the production of an enterotoxin (5, 8). The epidemiology of ETBF or its pathogenic role is not fully understood. No final proof has yet been provided to define the pathogenic importance of this microorganism in diarrheal diseases. To evaluate the association between diarrhea and the presence of ETBF, it is necessary to detect the microorganism from among the complex mixture of bacterial flora in the feces and to investigate its ability to

FIG. 2. IMS-PCR products from spiked fecal samples and clinical samples. Lanes: 1, ETBF D-100 (104 CFU/ml); 2, ETBF D-104 (104 CFU/ml); 3, nonenterotoxigenic B. fragilis NCTC 9343; 4, E. coli O:104, 5, B. vulgatus; 6, fecal sample a; 7, fecal sample b; 8, control sample; 9, positive control; 10, negative control; M, DNA molecular mass markers. The 294-bp product correlates with the amplified portion of the enterotoxin gene.

No. of isolates or samples with the indicated result by the following assay: IMS-PCR

HT29/C1 cell assay

Positive

Negative

Positive

Negative

21 0 0 0 0 0 0 3 0

0 6 3 1 1 4 1 2 5

21 0 0 0 0 0 0 3 0

0 6 3 1 1 4 1 2 5

produce enterotoxin. A direct PCR assay has been used to detect ETBF from fecal samples (6, 11). However, the PCR inhibitors present in feces seem to affect the sensitivity of this method and can cause false-negative results (16). Therefore, we developed a combined IMS and PCR assay for the specific, sensitive, and rapid detection of ETBF in feces. The MAbs used in this study (MAb 4H8 and MAb C3) were described earlier: MAb 4H8, which specifically binds to an immunodominant epitope of LPSs from B. fragilis (17), and MAb C3, which most likely recognizes an epitope in the inner core region of LPSs from B. fragilis (18). Mixtures of magnetic beads individually coated with one of the two MAbs were used in order to increase the capacity to capture B. fragilis in the IMS procedure. Before applying the fecal samples to IMS, the samples were inoculated into PYG broth containing kanamycin. An antibiotic agent was used to inhibit aerobic bacteria and to select for B. fragilis. With the aid of magnetic beads, the captured B. fragilis strains are concentrated, and the substances which may be inhibitory to PCR analyses are effectively removed. Combined with PCR, the assay provided a very sensitive method for the detection of ETBF in fecal samples, with a detection limit of ;50 CFU/g of feces. Although the time required for analysis is extended, we believe that incubation of the captured B. fragilis in PYG medium is necessary. The minimum time from receipt of the sample to the final PCR result was 36 h if the incubation in PYG medium was carried out for 24 h. This can be compared to the time required to perform the conventional method, i.e., the cell culture cytotoxicity assay, which takes between 48 and 60 h. The fecal samples spiked with ETBF were positive by IMSPCR, and the amplified 294-bp product was detected. None of the samples spiked with other (nonenterotoxigenic) B. fragilis isolates was positive. In addition, three of five clinical fecal samples from patients with diarrhea yielded the amplification products, and these samples were also tested by the HT29/C1 cell assay, which confirmed the enterotoxin production, thus showing that the IMS-PCR assay is specific for the detection of ETBF in clinical samples. In order to fully evaluate the sensitivity and specificity of the method described here, a study with larger numbers of samples must be conducted. We are currently evaluating this method by analyzing a large collection of fecal samples from patients with diarrhea and control subjects. ACKNOWLEDGMENTS This study was supported by the Ruth and Rickard Julins Foundation. We are grateful to Kerstin Bergman for skillful technical assistance in this study.

3548

ZHANG AND WEINTRAUB

J. CLIN. MICROBIOL.

REFERENCES

10. Sears, C. L., L. L. Myers, A. Lazenby, and R. L. Van Tassell. 1995. Enterotoxigenic Bacteroides fragilis. Clin. Infect. Dis. 20:S142–S148. 11. Shetab, R., S. H. Cohen, T. Prindiville, Y. J. Tang, M. Cantrell, D. Rahmani, and J. J. Silva. 1998. Detection of Bacteroides fragilis enterotoxin gene by PCR. J. Clin. Microbiol. 36:1729–1732. 12. Svenungsson, B., and A. A. Lindberg. 1978. Identification of Salmonella bacteria by co-agglutination, using antibodies against synthetic disaccharideprotein antigens O2, O4, and O9, adsorbed to protein A-containing staphylococci. Acta Pathol. Microbiol. Scand. Sect. B 86:283–290. 13. Van Tassell, R., D. Lyerly, and T. Wilkins. 1994. Characterization of enterotoxigenic Bacteroides fragilis by a toxin-specific enzyme-linked immunosorbent assay. Clin. Diagn. Lab. Immunol. 1:1–7. 14. Van Tassell, R., D. Lyerly, and T. Wilkins. 1992. Purification and characterization of an enterotoxigenic Bacteroides fragilis. Infect. Immun. 60:1343– 1350. 15. Weikel, C. S., F. D. Grieco, J. Reuben, L. L. Myers, and R. B. Sack. 1992. Human colonic epithelial cells, HT29/C1, treated with crude Bacteroides fragilis enterotoxin dramatically alter their morphology. Infect. Immun. 60: 321–327. 16. Wilde, J., J. Eiden, and R. Yolken. 1990. Removal of inhibitory substances from human fecal specimens for detection of group A rotaviruses by reverse transcriptase and polymerase chain reactions. J. Clin. Microbiol. 28:1300– 1307. 17. Zhang, G., Q. Pan, and A. Weintraub. Identification of Bacteroides fragilis by modified immuno polymerase chain reaction (mIPCR), using a specific monoclonal antibody. Anaerobe, in press. 18. Zhang, G., and A. Weintraub. 1997. Identification of Bacteroides fragilis by coagglutination, using a specific monoclonal antibody. Anaerobe 3:295–300.

1. de St. Groth, S. F., and D. Scheidegger. 1980. Production of monoclonal antibodies: strategy and tactics. J. Immunol. Methods 39:369–375. 2. Moncrief, J. S., R. Obiso, L. A. Barroso, J. J. Kling, R. H. Wright, R. L. Van Tassell, D. M. Lyerly, and T. D. Wilkins. 1995. The enterotoxin of Bacteroides fragilis is a metalloprotease. Infect. Immun. 63:175–181. 3. Moore, W., and L. Holdeman. 1974. Human fecal flora: the normal flora of 20 Japanese-Hawaiians. Appl. Microbiol. 27:961–979. 4. Myers, L. L., B. D. Firehammer, D. S. Shoop, and M. M. Border. 1984. Bacteroides fragilis: a possible cause of acute diarrheal disease in newborn lambs. Infect. Immun. 44:241–244. 5. Myers, L. L., D. S. Shoop, B. D. Firehammer, and M. M. Border. 1985. Association of enterotoxigenic Bacteroides fragilis with diarrheal disease in calves. J. Infect. Dis. 152:1344–1347. 6. Pantosti, A., M. Malpeli, M. Wilks, M. Grazia Menozzi, and F. D’Ambrosio. 1997. Detection of enterotoxigenic Bacteroides fragilis by PCR. J. Clin. Microbiol. 35:2482–2486. 7. Pantosti, A., M. G. Menozzi, A. Frate, L. Sanfilippo, F. D’Ambrosio, and M. Malpeli. 1997. Detection of enterotoxigenic Bacteroides fragilis and its toxin in stool samples from adults and children in Italy. Clin. Infect. Dis. 24:12–16. 8. Sack, R. B., L. L. Myers, J. Almeido-Hill, D. S. Shoop, W. C. Bradbury, R. Reid, and M. Santosham. 1992. Enterotoxigenic Bacteroides fragilis: epidemiologic studies of its role as a human diarrhoeal pathogen. J. Diarrhoeal Dis. Res. 10:4–9. 9. Sears, C., S. Koshy, R. Van Tassell, and F. Chambers. 1994. Pathophysiology of human intestinal epithelial monolayers (T84) treated with the toxin of enterotoxigenic Bacteroides fragilis. In Program and abstracts of the Congress on Anaerobic Bacteria and Anaerobic Infections. Anaerobic Society of the Americas, Los Angeles, Calif.