JOURNAL OF CLINICAL MICROBIOLOGY, Nov. 1999, p. 3662–3667 0095-1137/99/$04.00⫹0 Copyright © 1999, American Society for Microbiology. All Rights Reserved.
Vol. 37, No. 11
Rapid Identification of Burkholderia pseudomallei in Blood Cultures by a Monoclonal Antibody Assay SUPINYA PONGSUNK,1 NITTAYA THIRAWATTANASUK,2 NUANCHAN PIYASANGTHONG,3 4 AND PATTAMA EKPO * Department of Microbiology, Faculty of Medicine, Srinakharinwirot University,1 and Department of Immunology, Faculty of Medicine, Siriraj Hospital, Mahidol University,4 Bangkok, Department of Bacteriology, Sappasitprasong Hospital, Ubon Ratchathani,2 and Department of Clinical Pathology, Khon Kaen Regional Hospital, Khon Kaen,3 Thailand Received 30 November 1998/Returned for modification 7 July 1999/Accepted 12 August 1999
Burkholderia pseudomallei is the causative agent of melioidosis. In northeast Thailand, this gram-negative bacterium is a major cause of mortality from septicemia. The definitive diagnosis of this disease is made by bacterial culture. In this study, we produced a monoclonal antibody (MAb) specific to the 30-kDa protein of B. pseudomallei by in vivo and in vitro immunization of BALB/c mice with a crude culture filtrate antigen. The MAb could directly agglutinate with all 243 clinical isolates of B. pseudomallei but not with other gram-negative bacteria, except for one strain of Burkholderia mallei. However, the MAb cross-reacted with the gram-positive Bacillus sp. and Streptococcus pyogenes. B. pseudomallei in brain heart infusion broth (BHIB) subcultured from a BacT/Alert automated blood culture system could be identified by simple agglutination with this MAb assay. The sensitivity and specificity of direct agglutination compared to the “gold standard,” the culture method, were 94.12 and 98.25%, respectively. However, the MAb adsorbed to polystyrene beads or latex particles directly identified the bacterium in blood culture specimens and in BHIB subcultured from a BacT/Alert automated blood culture system. The sensitivity of the latex agglutination test was 100% for both blood culture and BHIB specimens. The specificity was 85.96 and 96.49% for the blood culture and BHIB specimens, respectively. The specificity could be increased if the nonspecific materials in the blood culture broths were eradicated by centrifugation at low speeds. Thus, a combination of blood culture and the agglutination method could be used for the rapid diagnosis of melioidosis in the routine bacteriological laboratory. This method could speed up detection of the bacterium in blood culture by at least 2 days, compared to the conventional bacterial culture method. In addition, the MAb is stable at room temperature for 2 weeks and at 4, ⴚ20, and ⴚ70°C for at least 1 year. The latex reagent was stable for at least 6 months at 4°C. made at least 2 days sooner than by the conventional bacterial culture method.
Melioidosis is an infection caused by the gram-negative organism Burkholderia pseudomallei. The disease is an important cause of morbidity and mortality in Thailand, Southeast Asia, and northern Australia (3). Its clinical manifestation is associated with chronic localized or acute septicemic infection. The septicemic form of this disease causes the highest mortality. In northeast Thailand, which is an area where melioidosis is endemic, septicemic melioidosis accounts for approximately onefifth of all community-acquired septicemias (1). The clinical manifestations of this disease cannot be differentiated from septicemia caused by other organisms. Early diagnosis and appropriate antibiotic treatment are very important if the mortality rate is to be reduced (12). To speed up diagnosis, a number of serological tests as well as DNA detection have been developed (4, 7, 8). However, blood culture and antibacterial sensitivity tests are still required because of the problem of antibiotic resistance. The disadvantage of the blood culture method is that it is time-consuming. It takes 2 to 6 days before a definite result is obtained, quite often after the patient has died. In this study, we tried to shorten the diagnosis time by using a monoclonal antibody (MAb) specific to the 30-kDa protein of B. pseudomallei to identify the bacterium in blood culture. By this method, identification of B. pseudomallei could be
MATERIALS AND METHODS Bacteria. B. pseudomallei and other bacteria used in this study, including Acinetobacter anitratus, Acinetobacter baumannii, Acinetobacter calcoaceticus, Acinetobacter lwoffii, Aeromonas hydrophila, Aeromonas sobria, Bacillus spp., Burkholderia cepacia, Citrobacter diversus, Citrobacter freundii, Enterobacter cloacae, Escherichia coli, Flavimonas oryzihabitans, Klebsiella pneumoniae, Proteus mirabilis, Proteus vulgaris, Providencia stuartii, Pseudomonas spp., Pseudomonas aeruginosa, Pseudomonas putida, Pseudomonas stutzeri, Salmonella spp., group B Salmonella, group D Salmonella, coagulase-negative Staphylococcus, Staphylococcus aureus, Streptococcus pyogenes, group B Streptococcus, group D (nonenterococcus) Streptococcus, Streptococcus pneumoniae, Vibrio cholerae biotype Ogawa, Vibrio parahaemolyticus, and Xanthomonas maltophilia, were isolated from clinical specimens from patients admitted to two hospitals in northeast Thailand, Sappasitprasong Hospital (Ubon Ratchathani) and Khon Kaen Regional Hospital (Khon Kaen), and a hospital in central Thailand, Siriraj Hospital (Bangkok). Two strains of Burkholderia mallei were kindly provided by T. Ezaki, Gifu Medical School, Gifu, Japan. The type strain of B. mallei, ATCC 23344, was isolated from a patient. The other strain, ATCC 10399, was isolated from the lung and nose of a horse. Preparation of antigens. Crude culture filtrate (CCF) antigen was prepared as previously described (6). Briefly, the bacteria were cultured in protein-free medium at 37°C for 2 weeks. The culture supernatant was filter sterilized and concentrated by ultrafiltration with a membrane with a molecular weight (MW) of 10,000. The outer membrane protein (OMP) antigen was prepared by culturing the bacteria in brain heart infusion broth (BHIB). The bacterial cells were collected by centrifugation, washed in phosphate-buffered saline, sonicated, and then centrifuged at low speed to remove intact cells. The cell envelope was obtained by centrifugation of the supernatant at 80,000 rpm (Beckman NVT90 rotor) for 1 h at 4°C. After the inner membrane in the pellet was solubilized by 1% sodium lauroyl sarcosinate, the OMP antigen was obtained by centrifugation at 80,000 rpm (Beckman NVT90 rotor) for 2 h at 4°C and then resuspended in
* Corresponding author. Mailing address: Department of Immunology, Faculty of Medicine, Siriraj Hospital, Mahidol University, Bangkok 10700, Thailand. Phone: 662-418-0569. Fax: 662-418-1636. E-mail:
[email protected]. 3662
VOL. 37, 1999
RAPID IDENTIFICATION OF BURKHOLDERIA PSEUDOMALLEI
3663
FIG. 1. Schematic diagram of the routine blood culture procedure by the BacT/Alert blood automatic culture system. The culture method usually requires 1 to 4 days for primary plating and 1 to 2 days for biochemical identification. The identification of the bacteria in an incubated blood culture, in subcultured BHIB, or on a primary plate by direct agglutination or the latex agglutination test takes 2 days less than the conventional blood culture method.
phosphate-buffered saline. The protein concentration was determined by a commercial protein assay kit (Bio-Rad Laboratories, Richmond, Calif.). Production and characterization of MAb. Ten micrograms of CCF antigen was homogenized in complete Freund’s adjuvant and then injected intraperitoneally into BALB/c mice. Two weeks after immunization, the spleens of the mice were removed. The spleen cells were immunized in vitro with 1 g of CCF antigen per ml. After 2 days, the cells were then fused with a myeloma cell line (P3 ⫻ 63 Ag 8.653). The fusion procedure was carried out as previously described (6). Hybridoma cells producing antibodies were screened by indirect enzyme-linked immunosorbent assay (ELISA) with CCF antigen (3 g of protein/ml) and direct agglutination with B. pseudomallei (6). The cells in positive wells were cloned by limiting dilution. The specificity of the MAb was tested by indirect ELISA and immunoblotting with a panel of CCF antigens. The hybridoma cells producing specific MAbs were cloned three more times and then expanded in a petri dish for bulk production. The isotypes of the MAb were determined by indirect ELISA with CCF antigen from B. pseudomallei (6). The MAb was further tested by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and an immunoblotting method with OMP antigen. The partially purified MAb was prepared from the culture supernatants by precipitation with 50% ammonium sulfate and kept in small aliquots at room temperature and at 4, ⫺20, and ⫺70°C. SDS-PAGE and immunoblotting. The protein profiles of CCF and OMP antigens were examined by SDS-PAGE in a minigel apparatus (Bio-Rad Laboratories). A 3.5% stacking gel and a 12% separating acrylamide gel were used. Samples were solubilized under denaturing condition and heated in a boilingwater bath for 5 min before being loaded onto the gel. After the proteins were separated at a constant current of 170 mA, the protein bands were visualized by staining with Coomassie blue R-250. Electrophoretic transfer of the separated proteins to nitrocellulose membranes (Bio-Rad Laboratories) was performed with a minigel apparatus (BioRad Laboratories) at a constant voltage of 80 V for 2 h at 4°C. The immunoblot reaction with the MAb was done as previously described (6). A direct agglutination test with bacterial colonies in primary culture. The MAb assay was used for identifying bacteria grown on either blood, MacConkey, or chocolate agar plates initially seeded with clinical specimens, blood culture
broth, or bacteria from stock cultures. A colony was picked and suspended in 20 l of normal saline solution on a glass slide and mixed with an equal volume of 0.1 mg of MAb per ml. The slide was rotated for at least 2 min. The result was read visually. Aliquots of the MAb kept at different temperatures were agglutinated at different times with a B. pseudomallei colony and B. pseudomallei in BHIB cultured from the stock culture. MAb-sensitized latex particles. Five hundred microliters of 0.793-m sulfonated latex beads (Interfacial Dynamics Corporation, Portland, Oreg.) was washed three times with 0.17 M glycine-buffered saline (GBS) (pH 7.3). After washing, 750 g of the ammonium sulfate-precipitated MAb in GBS was added to the washed latex beads and gently mixed overnight at room temperature. One milliliter of 1% bovine serum albumin (BSA) in GBS was added to the latex beads and gently mixed at room temperature for 30 min. After incubation, the beads were washed twice with 1% BSA in GBS, resuspended to a 0.5% concentration in storage buffer (0.02 M phosphate buffer [pH 7.4], 0.15 M NaCl, 1% BSA, 5% glycerol, and 0.1% NaN3), and stored at 4°C until used. Sensitivity of direct agglutination with the MAb assay and the latex agglutination test. To determine the sensitivity of the agglutination tests, 20 l of B. pseudomallei in BHIB, in a concentration that varied from 103 to 109 CFU/ml, was agglutinated with an equal volume of the MAb or the latex solution on a glass slide. The slide was rotated for at least 2 min. The result was read visually. An aliquot of the MAb kept at different temperatures and an aliquot of the latex reagent kept at 4°C were also used at different times to evaluate the sensitivity of the agglutination test. Identification of B. pseudomallei in blood culture. The procedure is outlined in Fig. 1. Blood culture broths from the BacT/Alert automated system (Organon Teknika Corporation, Durham, N.C.) were Gram stained. Aliquots of the broths that were positive for gram-negative bacteria were directly mixed with an equal volume (20 l) of 1 mg of the MAb per ml on glass slides and then rotated for at least 2 min. The blood culture broths that were positive for Gram staining (either Gram positive or Gram negative) were given biochemical tests. The remaining blood culture broths were routinely processed by subculture onto MacConkey, blood, and chocolate agar plates and simultaneously subcultured
3664
PONGSUNK ET AL.
J. CLIN. MICROBIOL. TABLE 1. Agglutination of the MAb against B. pseudomallei with gram-negative bacterial colonies cultured from clinical specimens Microorganism
FIG. 2. SDS-PAGE profiles of CCF and OMP antigens prepared from B. pseudomallei (A) and the immunoblot patterns of B. pseudomallei-specific MAb against CCF and OMP antigens (B). The arrow indicates the 30-kDa position at which the MAb reacted with the CCF and OMP antigens.
into BHIB for about 3 h. The bacteria grown on the plates were assessed for colony morphology, purity, and Gram-stain reaction. The bacteria in the BHIB were also assessed for Gram-stain reaction and then subjected to an antibiotic sensitivity test. The MAb was used for agglutination with the gram-negative bacteria subcultured in BHIB by the method described above. The results were compared to those obtained by the conventional method, bacterial culture and biochemical tests. The blood culture and BHIB specimens tested for direct agglutination were kept frozen at ⫺20°C. The latex agglutination test was performed with these frozen specimens. Equal volumes (20 l each) of the specimens and the latex reagent were mixed, rotated for 2 min, and read visually.
RESULTS Characterization of MAbs. A number of hybridomas, secreting antibody reactive with CCF and whole-cell antigens of B. pseudomallei, were obtained. These MAbs were analyzed for specificity by indirect ELISA and immunoblotting with a panel of bacterial antigens prepared from B. pseudomallei, B. cepacia, P. aeruginosa, P. putida, and E. coli. One MAb was secreted from a stable clone; it reacted only with the B. pseudomallei antigen. The isotype of the MAb was IgM. The immunoreactive component that reacted with the MAb was shown by SDS-PAGE to use OMP and CCF antigens from B. pseudomallei, blotted onto nitrocellulose membrane, and then probed with the MAb. The results (Fig. 2) showed that the MAb reacted with the CCF and OMP antigens at a 30-kDa position. Identification of B. pseudomallei colonies with the MAb assay. The microorganisms from the primary plates were sampled and agglutinated with the MAb. The agglutination results compared with the bacteriological and biochemical identifications are shown in Table 1 and Table 2. The results in Table 1 show that the MAb reacted with all 243 isolates of B. pseudomallei but not with 27 other species of gram-negative bacteria. However, it cross-reacted with the type strain of B. mallei (the gram-negative strain ATCC 23344, isolated from a patient) but it did not react with ATCC 10399, which had been isolated from a horse. Of seven species of gram-positive bacteria (Table 2), the MAb reacted with two species, a Bacillus sp. (1 of 4) and S. pyogenes (4 of 4). Sensitivity of the direct agglutination and the latex agglutination assays. The MAb could agglutinate with B. pseudomallei grown in BHIB at a minimum concentration of 4.5 ⫻ 108 CFU/ml. However, the sensitivity of the latex agglutination assay was about 100 times more sensitive (5.2 ⫻ 106 CFU/ml) than direct agglutination with the same MAb assay.
No. positive/no. tested
B. pseudomallei ................................................................... 243/243 A. anitratus .......................................................................... 0/1 A. baumannii ....................................................................... 0/9 A. calcoaceticus ................................................................... 0/1 A. lwoffii ............................................................................... 0/3 A. hydrophila ....................................................................... 0/1 A. sobria ............................................................................... 0/2 B. cepacia............................................................................. 0/1 B. mallei ............................................................................... 1/2a C. diversus............................................................................ 0/3 C. freundii ............................................................................ 0/3 E. cloacae............................................................................. 0/9 E. coli ................................................................................... 0/15 F. oryzihabitans.................................................................... 0/1 K. pneumoniae..................................................................... 0/13 P. mirabilis ........................................................................... 0/4 P. vulgaris............................................................................. 0/4 P. stuartii .............................................................................. 0/1 Pseudomonas spp................................................................ 0/2 P. aeruginosa........................................................................ 0/11 P. putida............................................................................... 0/1 P. stutzeri.............................................................................. 0/2 Salmonella spp. ................................................................... 0/10 Group B Salmonella........................................................... 0/2 Group D Salmonella .......................................................... 0/4 V. cholerae biotype Ogawa ................................................ 0/1 V. parahaemolyticus ............................................................ 0/1 X. maltophilia ...................................................................... 0/6 a Two strains of B. mallei were cultured from stock bacterial cultures. One is the type strain, isolated from a patient. The type strain agglutinated with the MAb. However, the other strain (isolated from a horse) did not agglutinate with the MAb.
Identification of B. pseudomallei in blood culture. To speed up the identification of B. pseudomallei in blood culture, a combination of bacterial culture and agglutination with the MAb assay was performed in the bacteriological laboratory at Sappasitprasong Hospital, Ubon Ratchathani, Thailand. The experiment was done between September and November 1997 and June and July 1998. Blood culture broths grown with the automated BacT/Alert system and the corresponding blood culture broths subcultured in BHIB for about 3 h were Gram stained. Samples that contained gram-negative bacteria were mixed with an equal volume of 1 mg of MAb per ml on glass slides and rotated for at least 2 min. In addition, the agglutination procedure for identification of the bacteria grown on the primary plate for gram-negative bacteria (on a MacConkey or eosin-methylene blue agar plate) was also performed. The agglutination results (Table 3) were compared with the bacterial culture and biochemical identification. Of 165 blood culTABLE 2. Agglutination of the MAb against B. pseudomallei with gram-positive bacterial colonies cultured from clinical specimens Microorganism
No. positive/no. tested
Bacillus spp. ........................................................................... Coagulase-negative Staphylococcus ..................................... S. aureus.................................................................................. S. pyogenes .............................................................................. Group B Streptococcus.......................................................... Group D (nonenterococcus) Streptococcus........................ S. pneumoniae ........................................................................ a
1/4a 0/1 0/3 4/4a 0/1 0/2 0/1
Shows cross-reaction of the MAb with gram-positive bacteria.
RAPID IDENTIFICATION OF BURKHOLDERIA PSEUDOMALLEI
VOL. 37, 1999
3665
TABLE 3. Direct and indirect agglutination of the MAba No. positive/no. tested Microorganism(s) cultured
Burkholderia pseudomalleic Acinetobacter spp. Aeromonas spp. Burkholderia cepacia Citrobacter diversus Enterobacter spp. Escherichia coli Haemophilus influenzae Klebsiella ozaenae Klebsiella pneumoniae Plesiomonas shigelloides Pseudomonas spp. Pseudomonas aeruginosa Proteus mirabilis Salmonella spp. Salmonella paratyphi A Group B Salmonella Group D Salmonella Salmonella typhi Vibrio parahaemolyticus E. coli and K. pneumoniae
Blood culture
BHIB
Direct agglutination
Latex agglutinationb
Direct agglutination
Latex agglutination
Primary plate and direct agglutination
24/51 0/9 1/3 0/3 0/2 0/2 1/36 0/1 0/1 0/18 0/1 1/8 1/12 0/3 0/1 0/1 0/2 1/8 0/1 0/1 0/1
51/51 3/9 0/3 0/3 0/2 0/2 3/36 0/1 0/1 4/18 0/1 2/8 4/12 0/3 0/1 0/1 0/2 0/8 0/1 0/1 0/1
48/51 1/9 0/3 0/3 0/2 0/2 1/36 0/1 0/1 0/18 0/1 0/8 0/12 0/3 0/1 0/1 0/2 0/8 0/1 0/1 0/1
51/51 1/9 0/3 0/3 0/2 0/2 0/36 0/1 0/1 0/18 0/1 1/8 0/12 2/3 0/1 0/1 0/2 0/8 0/1 0/1 0/1
51/51 0/9 0/3 0/3 0/2 0/2 0/36 0/1 0/1 0/18 0/1 0/8 0/12 0/3 0/1 0/1 0/2 0/8 0/1 0/1 0/1
a Direct agglutination against B. pseudomallei was measured. In indirect agglutination, the MAb adsorbed to latex particles with bacteria in blood culture broth, in subcultured BHIB, or on a primary plate is also shown. b The false positive results in some specimens could be eradicated by centrifugation at 1,000 ⫻ g for 5 min before the agglutination was performed. c The sensitivity (%) and specificity (%) of agglutination of the MAb with B. pseudomallei are as follows. Values listed after culture method and assay indicate sensitivity, followed in parentheses by specificity, for 51 specimens tested. Blood culture (direct agglutination), 47.06 (95.61); blood culture (latex agglutination), 100 (85.96); BHIB (direct agglutination), 94.12 (98.25); BHIB (latex agglutination), 100 (96.49); primary plate (direct agglutination), 100 (100).
tures that were positive for gram-negative bacteria, 51 were positive for B. pseudomallei and 114 were positive for other gram-negative bacteria. When compared with the standard method of bacterial culture and biochemical identification, the direct agglutination results of the bacteria on the primary plates showed 100% specificity and sensitivity. The sensitivity and specificity of direct agglutination with bacteria grown in blood culture and subcultured into BHIB for about 3 h were 94.12 and 98.25%, respectively. The direct agglutination with the samples taken directly from the blood culture showed the lowest sensitivity and specificity, i.e., 47.06 and 95.61%, respectively. The latex agglutination procedure with the same blood culture and BHIB specimens frozen at ⫺20°C was performed. The results (Table 3) show that the latex agglutination test directly identified the bacteria in blood culture specimens with sensitivity and specificity of 100 and 85.96%, respectively. When the latex agglutination assay was performed with the subcultured specimens in BHIB, the sensitivity and specificity of the test were 100 and 96.49%, respectively. However, after 16 blood culture broths which gave false positive results in the latex agglutination assay were centrifuged at 1,000 ⫻ g for 5 min and the supernatants were reacted with the latex reagent, negative agglutination results were obtained. Therefore, the specificity of the latex agglutination test with blood culture broths was increased to 100%. The stability of the MAb and latex reagent. The MAb kept at room temperature or at 4, ⫺20, or ⫺70°C for either 1, 2, or 4 weeks; 2, 3, 4, 5, or 6 months; or a year was compared to the agglutination assay. The MAb kept at 4, ⫺20, or ⫺70°C for at least 1 year or for 2 weeks at room temperature could agglutinate with bacteria grown on primary plates and with bacteria grown in BHIB at a minimal concentration of 4.5 ⫻ 108 CFU/
ml. The latex reagent could be kept at 4°C for at least 6 months without losing specificity and sensitivity. DISCUSSION The clinical course of melioidosis varies from acute sepsis to a chronic localized infection. The septicemic form occurs in 60% of patients with melioidosis (13). The mortality rate of the septicemic form is very high. If the bacteria are detected in blood cultures from patients within 24 h of showing symptoms, the patients are more likely to die (10). In addition, most patients will be dead within 2 to 3 days if they are not properly diagnosed and treated. In northeast Thailand where the disease is endemic, B. pseudomallei is the major bacterium in blood cultures from patients with the disease. The “gold standard” for the laboratory diagnosis of septicemic melioidosis is a combination of blood culture and biochemical identification of the cultured colonies. The specific protein component of B. pseudomallei, a 30-kDa protein, was characterized by SDS-PAGE and immunoblotting. The SDS-PAGE profiles of the CCF antigen (Fig. 2A) showed the major protein to have a molecular mass of 30 kDa; this protein component was the specific epitope which reacted with the monoclonal antibody (Fig. 2B). The specific MAb also reacted with the OMP antigen at a molecular weight of 30 kDa (Fig. 2B). This reactive component was on the bacterial surface and it was also secreted into the culture medium. Therefore, the MAb could directly agglutinate with B. pseudomallei and react with the component in the CCF antigen. The SDS-PAGE profiles of the OMP antigen prepared from B. pseudomallei (Fig. 2A) were similar to the profiles reported by Gotoh and coworkers (5). This specific component corresponded to one of
3666
PONGSUNK ET AL.
five major OMP components, a 31-kDa protein, prepared from 12 different strains (5). The evidence that the MAb could directly agglutinate all 243 clinical isolates of B. pseudomallei also supported the view that the specific epitope was common in this bacterium. However, the 30-kDa epitope was not present in the arabinose assimilators (Ara⫹) of B. pseudomallei isolated from soil (data not shown). Since the specific 30-kDa protein of B. pseudomallei was a common major OMP of the clinical isolates of B. pseudomallei and it was not present in the very closely related bacterium B. cepacia (9), it was a good candidate antigen for identification of the bacterium in clinical specimens. The automatic method for blood culture in the routine bacteriological laboratory in Sappasitprasong Hospital, Ubon Ratchathani, Thailand, is outlined in Fig. 1. The blood culture broths which were consecutively taken from the BacT/Alert automated blood culture system were Gram stained. Gram staining can detect bacteria at a minimum concentration of 103 cells/ml. Although the blood culture broths containing B. pseudomallei were usually not turbid after an overnight incubation, the presence of the bacterium can be demonstrated by Gram staining (13). The MAb cross-reacted with some grampositive bacteria but not with other gram-negative bacteria, except for B. mallei, a bacterium which has never been reported in Thailand. False-positive reactions are possible with one strain of Bacillus sp. that sometimes stains gram negative, especially in an old culture, and can lead to a false identification. However, the bacilli are large (up to 10 m) and commonly adhere in chains. In contrast, B. pseudomallei is a small rod about 3 by 0.5 m (2) and tends to stain in a bipolar, safety pin fashion. Therefore, Gram staining is very useful for screening out those cross-reactive organisms. Only specimens that contained gram-negative bacteria were further agglutinated with the MAb and the latex reagent. Cross-reaction of the MAb with all isolates (4 of 4) of S. pyogenes was observed. To identify the cross-reacting antigens, we used the MAb assay to monitor cloning and expression of the gene from B. pseudomallei genomic libraries. The gene and the purified 30-kDa protein will be analyzed and the crossreacting antigens should be identified. As shown in Table 3, the agglutination of the MAb directly with the bacteria in blood culture broths is less sensitive and less specific than the agglutination of the MAb with the same broths subcultured for 3 h into BHIB. The concentration of B. pseudomallei in 53% of cases of septicemic melioidosis has been estimated to be less than 10 CFU/ml (11). After incubation, it was possible that the concentration of the bacteria in most blood culture specimens was still less than 4.5 ⫻ 108 CFU/ml, and the bacterium could not be detected by the direct agglutination method. The evidence confirmed that the latex agglutination assay, which was 100 times more sensitive (5.2 ⫻ 106 CFU/ml) than direct agglutination, could readily identify the bacterium in blood culture specimens. In addition, direct agglutination could agglutinate only the specific component on the bacterial surface. The soluble form of antigen, the secreted component, could not be directly agglutinated with the MAb. When the MAb was adsorbed to the latex particle, it could agglutinate both forms of antigen. However, there might be some nonspecific substances, i.e., erythrocytes, antibody in the specimens, or unknown chemicals in the commercial broths, that interfered with the latex agglutination reaction. After the blood culture broths were subcultured into BHIB, the nonspecific substances were diluted and the number of the bacteria was increased. Therefore, the specificity of the latex agglutination in the subcultured BHIB specimens (96.49%) was higher than in the blood culture specimens (85.96%). However, the
J. CLIN. MICROBIOL.
nonspecific materials in the blood culture broths could also be eradicated by centrifugation at low speeds. It was also expected that direct agglutination would identify the bacteria in the subcultured BHIB with lower sensitivity than the latex agglutination test. The sensitivity of the latex agglutination test for identification of B. pseudomallei either in blood culture or in the subcultured BHIB specimens taken from the BacT/Alert automatic blood culture system was 100%. The bacterial detection by the BacT/Alert system was faster than the manual method (10, 13). Our new agglutination procedure with this MAb assay can be carried out before growing the bacteria on the primary plates. The procedure used previously needed 1 to 4 days for incubation. There is no requirement for biochemical identification, which takes 1 to 2 days. Therefore, the combination of blood culture and agglutination test could speed up the identification of the bacterium by at least 2 days, compared with the conventional blood culture method. The faster the results are obtained, the quicker the appropriate antibiotic treatment can be administered. The combination of blood culture and latex agglutination test for identification of the bacterium in blood culture and in BHIB would be very useful for routine bacteriological laboratories in an area where melioidosis is endemic. The agglutination method is very simple and cheap and requires no complicated reagents or instruments. In addition, the MAb is very stable and can be produced in large amounts without variations from lot to lot. Although the MAb could agglutinate the bacterial colonies with 100% sensitivity and 100% specificity (Table 3), bacterial culture plates required incubation at least overnight before the agglutination test could be performed. However, the identification of the bacterial colonies cultured from specimens other than blood samples (i.e., pus and urine) will be very useful especially in bacteriological laboratories that are not familiar with the colony morphology and the biochemical tests for identification of this bacterium. In a further study with the collaboration of the Ministry of Public Health, Nonthaburi, Thailand, we will evaluate the combination of the blood culture method and the latex agglutination assay for the identification of B. pseudomallei in five hospitals that use the BacT/Alert automatic system and nine hospitals that use the manual system of identification. ACKNOWLEDGMENTS This work was partly supported by research grants from Srinakharinwirot University and the Siriraj-China Medical Board, Faculty of Medicine, Siriraj Hospital, Mahidol University. We are very grateful to Suttipant Sarasombath, head of the Immunology Department, Faculty of Medicine, Siriraj Hospital, Mahidol University, and Robert Cowan, who kindly criticized and reviewed the manuscript. We thank Utane Rungpanich of the Immunology Department, Faculty of Medicine, Siriraj Hospital, Mahidol University, for performing the latex agglutination test. REFERENCES 1. Chaowagul, W., N. J. White, D. A. B. Dance, Y. Wattanagoon, P. Naigowit, T. M. Davis, S. Looareesuwan, and N. Pitakwatchara. 1989. Melioidosis: a major cause of community acquired septicemia in northeastern Thailand. J. Infect. Dis. 159:890–899. 2. Collins, C. H., P. M. Lyne, and J. M. Grange. 1995. Microbiological methods, 7th ed., p. 282 and 391. Butterworth-Heinemann Ltd., London, England. 3. Dance, D. A. B. 1991. Melioidosis: the tip of the iceberg? Clin. Microbiol. Rev. 4:52–60. 4. Desakorn, V., M. D. Smith, V. Wuthiekanun, D. A. B. Dance, H. Aucken, P. Suntharasamai, A. Rajchanuwong, and N. J. White. 1994. Detection of Pseudomonas pseudomallei antigen in urine for the diagnosis of melioidosis. Am. J. Trop. Med. Hyg. 51:627–633. 5. Gotoh, N., N. J. White, W. Chaowagul, and D. E. Woods. 1994. Isolation and characterization of the outer membrane proteins of Burkholderia (Pseudo-
VOL. 37, 1999
RAPID IDENTIFICATION OF BURKHOLDERIA PSEUDOMALLEI
monas) pseudomallei. Microbiology 140:797–805. 6. Pongsunk, S., P. Ekpo, and T. Dharakul. 1996. Production of specific monoclonal antibodies to Burkholderia pseudomallei and their diagnostic application. Asian Pac. J. Allergy Immunol. 14:43–47. 7. Rattanathongkom, A., R. W. Sermswan, and S. Wongratanacheewin. 1997. Detection of Burkholderia pseudomallei in blood samples using polymerase chain reaction. Mol. Cell. Probes 11:25–31. 8. Rugdech, P., N. Anuntagool, and S. Sirisinha. 1995. Monoclonal antibodies to Pseudomonas pseudomallei and their potential for diagnosis of melioidosis. Am. J. Trop. Med. Hyg. 52:231–235. 9. Thomas, A. D. 1983. Evaluation of the API 20E and Microbact 24E system for the identification of Pseudomonas pseudomallei. Vet. Microbiol. 8:611– 615.
3667
10. Tiangpitayakorn, C., S. Songsivilai, N. Piyasangthong, and T. Dharakul. 1997. Speed of detection of Burkholderia pseudomallei in blood cultures and its correlation with the clinical outcome. Am. J. Trop. Med. Hyg. 57:96–99. 11. Walsh, A. L., S. D. Smith, W. Wuthiekanun, Y. Suputtamongkol, W. Chaowagul, D. A. B. Dance, B. Angus, and N. J. White. 1995. Prognostic significance of quantitative bacteremia in septicemic melioidosis. Clin. Infect. Dis. 21:1498–1500. 12. White, N. J., D. A. B. Dance, W. Chaowagul, Y. Wattanagoon, V. Wuthiekanun, and N. Pitakwatchara. 1989. Halving of mortality of severe melioidosis by ceftazidime. Lancet ii:697–701. 13. Wuthiekanun, V., D. A. B. Dance, W. Chaowagul, Y. Suputtamongkol, Y. Wattanagoon, and N. J. White. 1990. Blood culture technique for the diagnosis of melioidosis. Eur. J. Clin. Microbiol. Infect. Dis. 9:654–658.
