- Email: [email protected]
Improved diagnosis of melioidosis using a 2-dimensional immunoarray
Diagnostic Microbiology and Infectious Disease 77 (2013) 209–215
Contents lists available at ScienceDirect
Diagnostic Microbiology and Infectious Disease journal homepage: www.elsevier.com/locate/diagmicrobio
Improved diagnosis of melioidosis using a 2-dimensional immunoarray☆ Alanna E. Sorenson a, Natasha L. Williams c, Jodie L. Morris c, Natkunam Ketheesan c, Robert E. Norton d, Patrick M. Schaeffer a, b,⁎ a
School of Pharmacy and Molecular Sciences, James Cook University, Douglas QLD 4811, Australia Comparative Genomics Centre, James Cook University, Douglas QLD 4811, Australia Infectious Disease and Immunology, Australian Institute Tropical Health and Medicine, James Cook University, Douglas QLD 4811, Australia d Queensland Health Pathology Service, The Townsville Hospital, Douglas QLD 4811, Australia b c
a r t i c l e
i n f o
Article history: Received 17 May 2013 Received in revised form 9 July 2013 Accepted 10 July 2013 Available online 14 September 2013 Keywords: Melioidosis Burkholderia pseudomallei Serology Two-dimensional Immunoarray ELISA Diagnostics
a b s t r a c t Melioidosis is caused by the Gram negative bacterium Burkholderia pseudomallei. The gold standard for diagnosis is culture, which requires at least 3–4 days obtaining a result, hindering successful treatment of acute disease. The existing indirect haemagglutination assay (IHA) has several disadvantages, in that approximately half of patients later confirmed culture positive are not diagnosed at presentation and a subset of patients are persistently seronegative. We have developed 2 serological assays, an enzyme-linked immunosorbent assay (ELISA), and a 2-dimensional immunoarray (2DIA), capable of detecting antibodies in patient sera from a greater proportion of IHA-negative patient subsets. The 2DIA format can distinguish between different LPS serotypes. Currently, the 2DIA has a sensitivity and specificity of 100% and 87.1%, respectively, with 100% of culture-positive, IHA-negative samples detected. The ELISA has a sensitivity and specificity of 86.2% and 93.5%, respectively, detecting 67% of culture-positive, IHA-negative samples. The ELISA and 2DIA tests described here are more rapid and reliable for serological testing compared to the existing IHA. © 2013 Elsevier Inc. All rights reserved.
1. Introduction Melioidosis is caused by the Gram-negative bacterium Burkholderia pseudomallei. Classically, melioidosis is characterized by pneumonia and multiple abscesses (Wiersinga et al., 2012). However, presentation varies greatly with severity ranging from subclinical to acute fulminant sepsis or chronic infection that can mimic other diseases. Such protean clinical manifestations complicate diagnosis and contribute to the mortality rate ranging from 16% in northern Australia to 40% in northeast Thailand (Limmathurotsakul et al., 2010b). B. pseudomallei is susceptible to a limited number of antibiotics. In endemic areas, melioidosis is an important cause of morbidity and mortality in humans and animals (Limmathurotsakul and Peacock, 2011). Important risk factors include diabetes mellitus, chronic renal failure, chronic lung disease, and excessive alcohol use. The current gold standard for diagnosis is culture, which often requires enrichment followed by several days of incubation (Limmathurotsakul et al., 2010a). Culture is also an imperfect gold standard due to low sensitivity and negative predictive values (Limmathurotsakul et al., 2010a). In the case of acute infections, bacterial sepsis can develop in a few days and requires immediate treatment with the correct antibiotics as this bacterium is highly drug resistant (White, 2003).
☆ Conflict of interest: The authors declare that they have no conflict of interest. ⁎ Corresponding author. Tel.: +61-(0)7-4781-6388. E-mail address: [email protected] (P.M. Schaeffer). 0732-8893/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.diagmicrobio.2013.07.009
Death usually follows within a few days if appropriate treatment is not rapidly applied. Improving presumptive diagnosis of melioidosis is important as some empirical antibiotic regimens employed in suspected bacterial sepsis do not adequately treat B. pseudomallei infection (Wiersinga et al., 2012). In northern Australia, the most common serological test used is the indirect haemagglutination assay (IHA) (Ashdown, 1987). The IHA has relatively high specificity but poor sensitivity, with values ranging from 92–100% for specificity and 50–85% for sensitivity, respectively (Ashdown et al., 1989; Wongratanacheewi et al., 2001). Approximately half of patients later confirmed to be culture positive are not able to be diagnosed by IHA at presentation, and a subset of these patients are found to be persistently seronegative by IHA (Cheng et al., 2006; Harris et al., 2009). The use of isolates from culture-positive, IHA-negative patients as antigen in IHA has been unsuccessful, indicating these patients do not develop antibodies to the specific epitopes adsorbed onto erythrocytes in the IHA (Harris et al., 2009). However, the same patients have been demonstrated to have responses to antigens from their infecting isolates, indicating antigen display is important in developing effective serological assays. The relatively poor sensitivity of IHA has led to the development of other assays to detect antibodies to B. pseudomallei, most of which are enzyme immunoassays (EIA) based on either crude wholecell preparations (Chantratita et al., 2007), recombinant proteins (Allwood et al., 2008; Felgner et al., 2009; Hara et al., 2013), or lipopolysaccharide (LPS) fractions (Anandan et al., 2010; Thepthai et al., 2005).
210
A.E. Sorenson et al. / Diagnostic Microbiology and Infectious Disease 77 (2013) 209–215
Protein-based EIAs have been found to generally have greater sensitivity than IHA (Chantratita et al., 2007; Felgner et al., 2009). However, the possibility of cross-reactivity is increased in proteinbased assays due to many seroreactive proteins in B. pseudomallei having homologs in other bacteria with similar protein sequences. LPS has been found to be an important virulence factor and antigenic component of B. pseudomallei (Tuanyok et al., 2012). While LPS-based EIAs have high specificity, sensitivity is generally poor due to the inclusion of very few or a single isolate. Studies have demonstrated heterogeneity in LPS among B. pseudomallei isolates with the identification of 4 phenotypes to date, Type A, B, B2, and Rough (Anuntagool et al., 2006; Tuanyok et al., 2012). Type A (typical) is the most common phenotype, and the predominant phenotype found in Thailand and Australia. Types B and B2 (atypical) are less common, with Type B2 only found in Australia and Papua New Guinea to date. The rough variant lacks the O-antigen, a phenotype found only in clinical isolates in Australia thus far (Tuanyok et al., 2012). Failure to use sufficient coverage of LPS phenotypes in assays may result in false negatives, particularly in Australia where atypical LPS phenotypes have been found to be more common than in Southeast Asia (Tuanyok et al., 2012). We have recently developed 2 new LPS-based serological assays capable of detecting antibodies in patient sera from a greater proportion of IHA-negative patient subsets, significantly improving sensitivity compared to the IHA (Cooper et al., 2013). These assays were developed with the most common LPS phenotype found in Australia, i.e., type A. One of these assays used the principle of immuno-PCR (Morin et al., 2010; Morin et al., 2011) to reduce sample volume and improve sensitivity. However, the heterogeneity in LPS phenotypes associated with Australian melioidosis cases necessitates the enhancement of serological assays to include all possible LPS phenotypes to improve diagnostic precision. LPS has been implicated as a major cause of septicaemia, involved in the hyper-inflammatory response (Leon et al., 2008). Septicaemia is a significant cause of death in melioidosis cases, emphasizing the potential importance of LPS phenotype in disease progression and prognosis. This is further highlighted by the demonstration of serum susceptibility in Type B2 and rough phenotype LPS B. pseudomallei isolates (Tuanyok et al., 2012), indicating these genotypes may have lower virulence. Here, we present the development of 2 new serological assays including LPS fractions from various B. pseudomallei isolates representing the 3 most common LPS phenotypes. An enzyme-linked immunosorbent assay (ELISA) incorporating the 3 most common LPS phenotypes has a sensitivity and specificity of 86.2% and 93.5%, respectively, detecting 67% of culture-positive, IHA-negative samples. As the ELISA could not enable serotyping, a further assay was developed, capable of distinguishing between different LPS serotypes in multiplex using a 2-dimensional immunoarray format (2DIA). This assay has a sensitivity and specificity of 100% and 87.1%, respectively, detecting 100% of culture-positive, IHA-negative samples. 2. Methods 2.1. Isolation of secreted LPS Glycerol stocks of K96243, JCUCC152, and JCUJP23 B. pseudomallei isolates were streaked onto separate Luria Bertani (LB) agar plates and incubated at 37 °C for 24 h. Single colonies of each isolate were inoculated into 5-mL LB media and incubated at 37 °C for 18 h with shaking at 150 RPM. Overnight cultures of each isolate were inoculated into separate 1-L flasks with 200-mL sterile LB media and incubated at 37 °C for 30 h with shaking at 150 RPM. Culture supernatants were removed and passed through a 0.2-μm filter. Finely ground ammonium sulphate was added to the filter-sterilized culture supernatants at a concentration of 0.5 mg mL −1 and incubated on ice with shaking for 2 h. Culture supernatants were centrifuged at 18,000
× g for 1 h, the supernatant decanted and retained for further analysis, and the pellet resuspended in 10 mmol/L phosphate buffer, pH 7.4 (2 mL per 25 mL culture supernatant). To retain LPS and remove protein from the antigenic fraction, samples were heated to 95 °C for 15 min, cooled, and then centrifuged at 16,000 × g for 10 min. Supernatants were transferred to fresh tubes and stored at −20 °C for later use. 2.2. LPS quantification LPS content of antigenic fractions was determined using a phenolsulphuric acid total carbohydrate quantification method (Fox and Robyt, 1991). Briefly, triplicate 25-μL samples of LPS at various dilutions were combined with 25 μL 5% w/v phenol and shaken gently for 30 s, along with a standard curve of D-glucose at 0, 50, 75, 100, 150, and 200 μg mL −1. Test plates were placed on ice, and 125-μL concentrated sulphuric acid was added to each reaction. The plates were shaken gently again for 30 s, then heated at 80°C for 30 min, and cooled. Reaction mixture (100 μL) was transferred to a microtitre plate and read at 490 nm. 2.3. Sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) Purified LPS samples were combined 1:1 with 2× Laemmli buffer (50 mmol/L Tris-HCl, pH 6.8, 2% SDS, 10% glycerol, 0.01% bromophenol blue) and heated at 95 °C for 5 min then fractionated on a 10% acrylamide stacking gel at 150 V for 45 min. 2.4. Modified silver staining SDS-PAGE fractionated LPS samples were stained using a modified silver stain to detect LPS (Fomsgaard et al., 1990). To detect protein, the same method was used with 2 modifications. First, periodic acid was omitted from the fixative, and secondly, the gel was incubated with 100 mL 0.02% sodium thiosulphate for 1 min then washed 3 times with MilliQ water for 20 s prior to addition of the silver staining solution. Subsequent to development, the reaction was stopped by incubating the gel with 5% acetic acid for 5 min. 2.5. Confirmation of LPS genotype by quantiative polymerase chain reaction (qPCR) Glycerol stocks of K96243, JCUCC152, and JCUJP23 B. pseudomallei isolates were streaked onto separate Ashdown agar plates and incubated at 37 °C for 24 h. Single colonies of each isolate were streaked onto minimal media and incubated at 37 °C for 72 h. Single colonies of each isolate were selected an a genomic DNA extraction performed using a Roche genomic DNA PCR template kit according to the manufacturer's instructions for bacterial gDNA extraction. Primer sequences for amplification of LPS genotype specific products were obtained from Tuanyok et al. (2012). PCR assays were performed in 20-μL reactions in triplicate for Type A, B, and B2 genotypes on K96243, JCUCC152, and JCUJP23, with all 3 tested with each primer set. The reactions and subsequent melt curve analysis were performed on a Bio-Rad IQ5 thermocycler as described by Tuanyok et al. (2012). 2.6. Sera IHA-positive patients, patients with persistently IHA-nonreactive sera who had culture-proven melioidosis, and healthy controls were requested to provide blood samples. Initial IHAs were performed by Pathology Queensland, The Townsville Hospital (Ashdown, 1987). Ethical approval for collection of sera was obtained from the Townsville Health Service District Ethics Committee (#2502 and #7104). A total of 60 serum samples were tested along with positive and negative controls. Serum samples were initially provided and
A.E. Sorenson et al. / Diagnostic Microbiology and Infectious Disease 77 (2013) 209–215
tested in blind fashion, then identified and confirmed subsequently. A total of 29 melioidosis patient samples were tested, of which 17 were culture positive, IHA positive and 12 culture positive, IHA negative. In addition, 31 samples from healthy controls were tested. 2.7. Immunoblotting Separate SDS-PAGE of each LPS type in quadruplicate was run and transferred onto pre-wet polyvinilidene fluoride (PVDF) membrane (Biorad, Gladesville, Australia) via semi-dry electroblotting at 15 V for 25 min. Following blocking with 5% skim milk in phosphate-buffered saline (PBS), pH 7.4, at room temperature (RT) for 1 h, blots were probed with human serum from culture-confirmed melioidosis patients known to have been infected with B. pseudomallei isolates from each LPS genotype, in addition to human serum from a healthy control. Serum was diluted 1:100 in 1% skim milk in PBS, pH 7.4, and incubated with blots at RT for 1 h. After washing 3 times for 5 min PBS, pH 7.4–0.05% Tween-20 (PBS-T), blots were probed with peroxidaseconjugated protein G (Sigma, Sydney, Australia) diluted 1:5000 in 1% skim milk in PBS, pH 7.4, for 1 h at RT. PVDF membranes were washed again 3 times with PBS-T and developed with 5-mL SIGMAFAST™ 3, 3′-diaminobenzidine/H2O2 solution. 2.8. G-peroxidase ELISA Nunc Maxisorb 96-well round-bottom immunoplates (Thermo Fisher Scientific, Roskilde, Denmark) were coated with 50-μL combined LPS at 2 μg mL −1 each for the K96243, JCUCC152, and JCUJP23 B. pseudomallei isolates, respectively, in carbonate/bicarbonate buffer, pH 9.6, overnight at 4 °C. Wells were blocked with 50-μL 1% bovine serum albumin (BSA) in bind and wash (BW) buffer (20 mmol/L Tris, 150 mmol/L NaCl, 0.005% Tween-20) at RT for 1 h. Human serum (50 μL) was applied at 1:200 in BW and incubated at RT for 1 h. After washing 3 times with BW, 50-μL peroxidase-conjugated protein G (Sigma) diluted 1:5000 in BW was applied for 1 h at RT. Wells were washed again 3 times with BW and developed with tetramethylbenzidine for 5 min (Cooper et al., 2013). Optical densities (ODs) were measured at 450 nm with a Bio-strategy Versa Max microplate reader and corrected by subtracting background values, which were obtained by omitting serum. 2.9. Two-dimensional immunoarray PVDF membrane cut into 12.5 cm 2 squares was wet with methanol for 30 s then rinsed with PBS prior to clamping under a Surf-Blot 10.5 chamber (Idea Scientific, Minneapolis, MN, USA). LPS samples (400 μL) were applied in separate lanes of the chamber in PBS, pH 7.4, and incubated at RT for 1 h with gentle rocking. Optimal dilutions of LPS for each genotype were determined by checkerboard titration. Optimal total carbohydrate content values of 6.25 μg mL −1, 12.5 μg mL −1, and 6.25 μg mL −1 were determined for the K96243, JCUCC152, and JCUJP23 B. pseudomallei isolates, respectively. LPS from Escherichia coli and Pseudomonas aeruginosa at 5 μg mL −1 were included as controls. Samples were decanted off, and PVDF was washed with PBS, pH 7.4–0.05% Tween-20 for 5 min. Membranes were blocked with 5% skim milk in PBS, pH 7.4, at RT for 1 h. After blocking, blots were turned 90° re-clamped under a Surf-Blot 10.5 chamber and probed with 400-μL human sera at a dilution of 1:80 in 1% skim milk in PBS, pH 7.4, for 1 h at RT. After washing 3 times for 5-min PBS, pH 7.4– 0.05% Tween-20 (PBS-T), blots were probed with peroxidaseconjugated protein G (Sigma) diluted 1:5000 in 1% skim milk in PBS, pH 7.4, for 1 h at RT. PVDF membranes were washed again and developed as described above. Intensity of grid squares was determined via integration using Image J software (NIH, Bethesda, MD, USA).
211
2.10. Investigation of common reactivity to crude B. pseudomallei LPS phenotypes Crude LPS samples from each LPS phenotype, E. coli and P. aeruginosa, were combined 1:1 with 2× Laemmli buffer and then fractionated on a 4–15% Mini-PROTEAN® TGX™ Precast acrylamide gel (BioRad) at 150 V for 45 min. Samples were transferred onto nitrocellulose membrane via semi-dry electroblotting at 25 V for 3 min in a Trans-Blot® Turbo™ Transfer System (Biorad). Following blocking with 2.5% BSA in PBS, pH 7.4, at RT for 1 h, blots were probed with human serum from culture-confirmed melioidosis patients demonstrated to have common reactivity to crude samples of Type A, B, and B2 Bps LPS phenotypes. Serum was diluted 1:100 in 1% BSA in PBS, pH 7.4, and incubated with blots at RT for 1 h. After washing 3 times for 5 min with PBS, pH 7.4–0.05% Tween-20 (PBS-T), blots were probed with peroxidase-conjugated protein G (Sigma) diluted 1:5000 in 1% BSA in PBS, pH 7.4, for 1 h at RT. PVDF membranes were washed again 3 times with PBS-T and developed with 5-mL SIGMAFAST™ 3, 3′-diaminobenzidine/H2O2 solution. 2.11. Investigation of common antigen reactivity To determine whether the 45 kDa common antigen was proteinaceous, samples of LPS representative of each phenotype were treated with 100 μg mL −1 proteinase K at 56 °C for 30 min and inactivated at 95 °C for 10 min. Identical samples received the same treatment without the addition of proteinase K. Duplicate untreated and treated samples were fractionated via SDS-PAGE then transferred onto nitrocellulose membrane, as described previously. The membrane was probed with common-reactive sera, as described previously. An identical gel was retained for protein staining with AcquaStain (Acquascience, Uckfield, TN, USA) at RT for 3 h. 3. Results 3.1. LPS genotyping and phenotyping To confirm each B. pseudomallei isolate was representative of each LPS type, genotyping and phenotyping were performed. Culture samples were genotyped for LPS subtypes, as described by Tuanyok et al. (2012). Triplicate samples of DNA from each isolate were analysed by quantitative PCR specific for each LPS genotype. The PCR products were subjected to a standard melt curve from 60 °C to 95 °C. Melting temperatures for Type A (K96243), Type B (JCUCC152), and Type B2 (JCUJP23) were 88 °C, 84 °C, and 89 °C, respectively (Fig. 1A). Amplicons for LPS-specific PCR were only detected in genomic DNA from their respective LPS phenotype-producing isolates, and the melting temperatures were consistent with those obtained by Tuanyok et al. (2012). LPS samples at several concentrations obtained from these isolates were fractionated by SDS-PAGE, then analysed using a modified silver staining protocol to visualise differences in their LPS laddering patterns. The electrophoretic patterns of the LPS phenotypes (Fig. 1B) were found to be consistent with those described in the literature (Anuntagool et al., 2006; Tuanyok et al., 2012). 3.2. Immunoblotting characterisation of LPS genotypes To confirm each LPS phenotype possessed the expected antigenic profile, immunoblotting was performed with serum from cultureconfirmed melioidosis patients known to have been infected with B. pseudomallei isolates representing each LPS phenotype. This immunoblotting demonstrated LPS phenotype-specific antibodies were produced during infection with the respective isolates (Fig. 2). Serum from a patient that had been infected with a Type B LPS isolate demonstrated cross-reactivity with Type B2 LPS. Interestingly, sera
212
A.E. Sorenson et al. / Diagnostic Microbiology and Infectious Disease 77 (2013) 209–215
Fig. 1. Genotyping and phenotyping of B. pseudomallei LPS subtypes. Melt curve analysis of amplicons from each LPS phenotype B. pseudomallei isolate (a). Melting temperatures (Tm) were consistent with the expected Tm for each amplicon. Amplicons were specific to their respective LPS phenotype-producing isolates (b). Modified silver stain for LPS showing characteristic banding pattern for 3 of the known LPS phenotypes in B. pseudomallei. Each phenotype was tested at several dilutions with microgram amounts of LPS indicated below each lane. Image has been inverted for improved contrast between LPS bands and background.
from a patient infected with a B2 genotype isolate did not react against B genotype LPS.
3.3. Detection of antibodies against B. pseudomallei LPS antigens by indirect G-peroxidase ELISA Initially, an indirect ELISA was developed with equivalent amounts of each LPS phenotype used as antigen. Serum dilutions 1:200 were found to be optimal for detection of antibodies by G-peroxidase ELISA via checkerboard assays with positive and negative control sera. An
OD450 ratio between the sample and negative control of N1.4 was considered to be seropositive for the G-peroxidase ELISA (Fig. 3A). A summary of results, including IHA results (performed by Pathology Queensland, The Townsville Hospital) for each patient and control, is provided in Table 1. Sensitivity and specificity of the IHA were 58.6% and 100%, respectively, for the samples used in this study. Sensitivity and specificity of the ELISA were 86.2% and 93.5%, respectively. As the ELISA could not enable serotyping, a further assay, the 2DIA, was developed. 3.4. Detection of antibodies against B. pseudomallei LPS antigens by 2DIA To enable serotyping of B. pseudomallei infecting isolates, the 2DIA was developed (Fig. 3B). Serum dilutions of 1:80 were found to be optimal for detection of antibodies by 2DIA via checkerboard assays with positive and negative control sera. ImageJ corrected scores of N4000 were considered to be seropositive for Type A and B LPS and N4500 for Type B2 in the 2DIA (Fig. 3C). Sensitivity and specificity for the 2DIA were 100% and 87.1%, respectively. A total of 13 melioidosis patient serum samples reacted with a common antigen present in all 3 LPS fractions. Of the serotype-specific reacting patient sera, 8 reacted to Type A LPS; 4, to Type B; and 4, to Type B2. Four healthy control serum samples reacted to specific LPS phenotypes; 2, to Type B; and 1, to Type B2 and A, respectively. Of the IHA-positive patients, 88% were identified as having Type A LPS-specific or common antigen-specific antibodies, with 35% having Type A LPS-specific only antibodies. Alternatively, of the IHA-negative patients, 50% were identified as having Type B or B2 LPS-specific antibodies. Scanned images of the 2DIA blots (Figure S1) used for ImageJ analysis (raw data) are included in the supplementary information. 3.5. Investigation of common reactivity to crude Bps LPS phenotypes A series of Western blots were performed to investigate the antigens responsible for common reactivity in the LPS fractions used in the 2DIA. Western blotting with several patient sera, which demonstrated common reactivity in the 2DIA, exhibited reactivity to capsular polysaccharide and/or a band of approximately 45 kDa (Fig. 4). Western blotting with crude LPS samples, pre and postproteinase K digestion confirmed, the antigen of approximately 45 kDa was proteinaceous (Fig. 5A). SDS-PAGE and subsequent incubation with a protein staining reagent also demonstrated the presence of a protein antigen of approximately 45 kDa (Fig. 5B). 4. Discussion
Fig. 2. Reaction of culture-confirmed melioidosis patient sera with LPS from B. pseudomallei isolates K96243 (Type A), JCUCC152 (Type B), and JCUJP23 (Type B2). Reactivity of sera from a patient infected with an A genotype isolate (a). Lack of reactivity of sera from a healthy control (b). Reactivity of sera from patients infected with B (c) and B2 (d) genotype isolates, respectively.
Both new immunoassay platforms developed in this study target the detection of LPS-specific human IgG, with the 2DIA able to confirm the diagnosis of all IHA-negative patients included in this study. These
A.E. Sorenson et al. / Diagnostic Microbiology and Infectious Disease 77 (2013) 209–215
a
213
c
15
10
LPS Fractions
OD450 Ratio
Serum Samples
1
5
PVDF Membrane
2
Surf-Blot
0 POS
b
NEG
Protein GPeroxidase DAB
50K POS NEG
LPS Fractions
Intensity (ImageJ)
3
40K 30K 20K 10K
1 2
3
4 5 6 7 8 9 10 11 12 13 14 15
A B B2
0 POS
NEG
Fig. 3. Comparison between indirect G-peroxidase ELISA (a) and the 2DIA (b) for detection of IgG against Type A, B, and B2 B. pseudomallei LPS phenotypes in melioidosis patient sera. For the ELISA, OD450 ratios between the sample and negative control of N1.25 were considered to be seropositive. For the 2DIA, corrected ImageJ intensity scores N4000 were considered to be seropositive for Type A and B LPS, with N4500 considered to be seropositive for Type B2. Diagram demonstrating the use of the Surf-Blot chamber to perform 2DIA assays (c). LPS samples are blotted onto PVDF membrane in separate channels (Panel 1), then blocked before turning 90° and probing with serum samples in separate channels (Panel 2). Positive and negative control samples are included in each run. The final panel (Panel 3) demonstrates a typical output (where samples 3, 4, 6, 12, 14, and 15 are considered to be positive reactors). ImageJ (NIH) is used to measure the relative intensity of each square.
new assays were able to cover 3 of the 4 known LPS genotypes of B. pseudomallei, which were confirmed by PCR amplification of LPS genotype-specific genes and modified silver staining for LPS. As determined by Anuntagool et al. (2006), atypical LPS genotypes are more common in Australian B. pseudomallei isolates. The failure to include representative isolates from these genotypes in immunoassays for melioidosis may complicate diagnosis, particularly in Australia. The inclusion of representative isolates from the Type B and B2 genotypes in our assays has resulted in the improved detection of specific antibodies in patient sera previously found to be persistently negative by the IHA. The antigen used by the Townsville Hospital in IHAs was prepared using techniques described previously (Ashdown, 1987) from 5 B. pseudomallei isolates of unknown identity. However, it is likely that the isolates used for development of the IHA in Australia are all from the Type A LPS genotype. This is reflected by the high percentage of IHA-positive melioidosis patient sera reacting to Type A LPS and/or common Bps antigen in the 2DIA immunoassay. Additionally, this may explain the previously high proportion of persistently IHA-negative melioidosis patient sera, as 50% of these serum samples reacted specifically to Type B and/or B2 LPS in the 2DIA immunoassay. The addition of antigens from B. pseudomallei isolates representative of the additional LPS phenotypes to the existing IHA antigen complement may improve the sensitivity of this assay. In this study, sensitivity for the IHA, ELISA, and 2DIA was determined to be 58.6%, 86.2%, and 100%, respectively, with specificity for the IHA, ELISA, and 2DIA of 100%, 93.5%, and 87.1%, respectively. The ELISA developed in this study contained 3 B. pseudomallei LPS types (A, B, and B2), resulting in comparable specificity and improved sensitivity compared to the IHA. The reduced sensitivity of the ELISA compared to the 2DIA is most likely due to the increased dilution factor of both the combined LPS fractions and test sera. The 2DIA had slightly reduced specificity and greatly improved sensitivity com-
pared to the IHA, in addition to enabling serotyping of melioidosis patient sera. This distinction may be useful in determining whether LPS genotypes are correlated with disease presentation, prognosis, and isolate virulence. While sensitivity was greatly improved with the ELISA and 2DIA compared to the IHA, specificity was reduced, increasing the potential for false positives, particularly in healthy patients from endemic areas who may have been exposed to B. pseudomallei or closely related species. In routine diagnostics, the ELISA may be of greater utility due to its simplicity and improved sensitivity compared to the IHA. Nevertheless, we expect that further development of the 2DIA into a lateral flow–based assay would provide the best diagnostic solution. Interestingly, sera from the patient infected with a B2 genotype isolate failed to react against B LPS. Serum from melioidosis patients infected with Bps isolates of B genotype was demonstrated to crossreact with B2 genotype LPS by Tuanyok et al. (2012). However, Tuanyok et al. (2012) did not investigate the profile of melioidosis patients infected with Bps isolates of B2 genotype. In the current study, the cross-reactivity of anti-B LPS antibodies with B2 LPS was confirmed, while indicating the reciprocal cross-reactivity does not occur. In this study, several patient sera were found to react to genotype B2 LPS only, indicating this LPS phenotype is antigenically distinct. This finding also demonstrated the importance of the inclusion of all known LPS genotypes in serological assays in order to successfully diagnose patients, especially those infected with Bps isolates with B2 genotype LPS. In some cases, reactivity was common to all 3 LPS genotypes, indicating the presence of a heat-stable antigen remaining in the crude LPS preparations. Western blotting with several patient sera, which demonstrated common reactivity in the 2DIA, exhibited reactivity to capsular polysaccharide and/or a band of approximately 45 kDa. Capsular polysaccharide, in addition to being a virulence factor (Wikraiphat et al., 2009), is known to be antigenic
214
A.E. Sorenson et al. / Diagnostic Microbiology and Infectious Disease 77 (2013) 209–215
Table 1 Serological assay results for culture-confirmed melioidosis patients and controls. Sample no.
IHAa
ELISAb
2DIAc
Typed
Sample no.
IHA
5 16 17 32 49 51 12 19 22 29 48 59 7 31 60 3 30 20 24 2 26 23 33 35 15 8 9 34 10
+ve +ve +ve +ve +ve +ve +ve +ve +ve +ve +ve +ve +ve +ve +ve +ve +ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve
+++ +++ +++ +++ +++ +++ ++ ++ ++ ++ ++ ++ ++ ++ ++ + + + + ++ ++ + + + + − − − −
+++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ ++ ++ ++ ++ ++ +++ +++ ++ ++ ++ ++ ++ + ++ ++ ++ +
C C C C C A A C A A C B C C A B2 A B2 B C A B2 A C B B C C B2
28 14 57 44 1 4 6 11 13 18 21 25 27 36 37 38 39 40 41 42 43 45 46 47 50 52 53 54 55 56 58
Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct
−ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve
ELISA
2DIA
Type
+ + − − − − − − − − − − − − − − − − − − − − − − − − − − −
+ + ++ + − − − − − − − − − − − − − − − − − − − − − − − − − − −
B B A B2 − − − − − − − − − − − − − − − − − − − − − − − − − − −
a +ve designates IHA positive melioidosis patients (IHA N40), −ve designates IHA negative melioidosis patients, Ct designates healthy controls, and each group is distinguished by shading. b ELISA results indicate OD450 ratios between the sample and negative control sera where – is b1.25, + is b1.25b2.0, ++ is N2.0b6.0, and +++ is N6.0. c 2DIA results indicate corrected ImageJ values where – is b4000, + is N4000b7000, ++ is N7000b15,000, and +++ is N15,000. d Type designates LPS genotype where A, B, and B2 indicate reactivity, and C indicates reactivity to antigen common to all 3 isolates.
(Parthasarathy et al., 2006). Protein staining and Western blotting with crude LPS samples, pre and post-proteinase K digestion, confirmed the antigen of approximately 45 kDa was proteinaceous. The identification of the proteinaceous antigen and performing 2DIA assays with and without proteinase K digestion on LPS fractions would enable improved serotyping. The proteinaceous antigen would
Fig. 5. Western blotting (a) and protein staining (b) with crude LPS samples, pre and post-proteinase K digestion. Loss of band of approximately 45 kDa in proteinase K treated samples (arrow).
be particularly useful in the 2DIA, as it has factors that would contribute to long-term antigen stability in that it is demonstrably either extremely heat-stable or capable of refolding without loss of antigenicity. Indeed, the relative stability of LPS and the 45 kDa proteinaceous antigen lend themselves to the conversion of the 2DIA to a lateral flow–based assay that could be developed as a commercial diagnostic test. Several healthy control sera reacted in both immunoassays, which may reflect the endemic nature of B. pseudomallei in the region. Unfortunately, the use of serological techniques is problematic in some endemic areas due to high background seropositivity in the healthy population (Limmathurotsakul and Peacock, 2011). However, in northern Queensland and the Northern Territory of Australia, the seroprevalence of antibodies to B. pseudomallei is relatively low at approximately 2.5% (Lazzaroni et al., 2008) and 3.0% (James et al., 2013), respectively, making serology a potentially advantageous addition to culture in the diagnosis of melioidosis in this region. Alternatively, this reactivity may be due to prior infection with a near-neighbour Burkholderia species, as it has been demonstrated that some of these species possess common LPS genotypes with B. pseudomallei (Stone et al., 2012). While cross-reactivity between B. pseudomallei and nearneighbour species is possible, the immunoassays developed in this study could improve the rapidity of diagnostic confirmation in cases where melioidosis is suspected and immediate treatment is necessary for patient survival. The use of the assays developed in this study may also improve seroprevalence surveys by detecting antibodies in the healthy population that are unable to be detected using the IHA. In conclusion, the immunoassays developed in this study have comparable specificity and greatly improved sensitivity compared to the IHA and are more rapid than culture, potentially leading to improved diagnosis of melioidosis and patient outcome. These assays would be of particular use in non-endemic areas in suspected melioidosis cases in travelers and military personnel returning from endemic areas. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.diagmicrobio.2013.07.009. Acknowledgments
Fig. 4. Western blotting with several patient sera, which demonstrated common reactivity in the 2DIA. Reactivity exhibited to capsular polysaccharide (N190 kDa) and/ or unidentified proteinaceous antigen (45 kDa).
The authors would like to thank the Smart Futures Fund: National and International Research Alliances Program, the National Health and Medical Research Council, James Cook
A.E. Sorenson et al. / Diagnostic Microbiology and Infectious Disease 77 (2013) 209–215
University, and the Queensland Tropical Health Alliance for financial support. References Allwood EM, Logue CA, Hafner GJ, Ketheesan N, Norton RE, Peak IR, et al. Evaluation of recombinant antigens for diagnosis of melioidosis. FEMS Immunol Med Microbiol 2008;54:144–53. Anandan S, Augustine A, Mathai E, Jesudason MV. Evaluation of IgM ELISA using a sonicate and a lipopolysaccharide antigen for the serodiagnosis of melioidosis. Indian J Med Microbiol 2010;28:158–61. Anuntagool N, Wuthi ekanun V, White NJ, Currie BJ, Sermswan RW, Wongratanacheewin S, et al. Lipopolysaccharide heterogeneity among Burkholderia pseudomallei from different geographic and clinical origins. Am J Trop Med Hyg 2006;74:348–52. Ashdown LR. Indirect haemagglutination test for melioidosis. Med J Aust 1987;147: 364–5. Ashdown LR, Johnson RW, Koehler JM, Cooney CA. Enzyme-linked immunosorbent assay for the diagnosis of clinical and subclinical melioidosis. J Infect Dis 1989;160: 253–60. Chantratita N, Wuthiekanun V, Thanwisai A, Limmathurotsakul D, Cheng AC, Chierakul W, et al. Accuracy of enzyme-linked immunosorbent assay using crude and purified antigens for serodiagnosis of melioidosis. Clin Vaccine Immunol 2007;14: 110–3. Cheng AC, O'Brien M, Freeman K, Lum G, Currie BJ. Indirect hemagglutination assay in patients with melioidosis in northern Australia. Am J Trop Med Hyg 2006;74: 330–4. Cooper A, Williams NL, Morris JL, Norton RE, Ketheesan N, Schaeffer PM. ELISA and immuno-polymerase chain reaction assays for the sensitive detection of melioidosis. Diagn Microbiol Infect Dis 2013;75:135–8. Felgner PL, Kayala MA, Vigil A, Burk C, Nakajima-Sasaki R, Pablo J, et al. A Burkholderia pseudomallei protein microarray reveals serodiagnostic and cross-reactive antigens. Proc Natl Acad Sci USA 2009;106:13499–504. Fomsgaard A, Freudenberg MA, Galanos C. Modification of the silver staining technique to detect lipopolysaccharide in polyacrylamide gels. J Clin Microbiol 1990;28: 2627–31. Fox JD, Robyt JF. Miniaturization of three carbohydrate analyses using a microsample plate reader. Analyt Biochem. 1991;195:93–6. Hara Y, Chin CY, Mohamed R, Puthucheary SD, Nathan S. Multiple-antigen ELISA for melioidosis—a novel approach to the improved serodiagnosis of melioidosis. BMC Infect Dis 2013;13:165. Harris PNA, Ketheesan N, Owens L, Norton RE. Clinical features that affect indirecthemagglutination-assay responses to Burkholderia pseudomallei. Clin Vaccine Immunol 2009;16:924–30.
215
James GL, Delaney B, Ward L, Freeman K, Mayo M, Currie BJ. Surprisingly low seroprevalence for Burkholderia pseudomallei in exposed healthy adults in the highly melioidosis-endemic Darwin region of Tropical Australia. Clin Vacc Immunol 2013;20:759–60. Lazzaroni SM, Barnes JL, Williams NL, Govan BL, Norton RE, LaBrooy JT, et al. Seropositivity to Burkholderia pseudomallei does not reflect the development of cell-mediated immunity. Trans R Soc Trop Med Hyg 2008;102(Suppl 1):S66–70. Leon CG, Tory R, Jia J, Sivak O, Wasan KM. Discovery and development of Toll-like receptor 4 (TLR4) antagonists: a new paradigm for treating sepsis and other diseases. Pharm Res 2008;25:1751–61. Limmathurotsakul D, Wongratanacheewin S, Teerawattanasook N, Wongsuvan G, Chaisuksant S, Chetchotisakd P, et al. Increasing incidence of human melioidosis in Northeast Thailand. Am J Trop Med Hyg 2010a;82:1113–7. Limmathurotsakul D, Jamsen K, Arayawichanont A, Simpson JA, White LJ, Lee SJ, et al. Defining the true sensitivity of culture for the diagnosis of melioidosis using Bayesian latent class models. PLoS One 2010b;5:e12485. Limmathurotsakul D, Peacock SJ. Melioidosis: a clinical overview. British Med Bull. 2011;99:125–39. Morin I, Dixon NE, Schaeffer PM. Ultrasensitive detection of antibodies using a new TusTer-lock immunoPCR system. Mol Biosys. 2010;6:1173–5. Morin I, Askin SP, Schaeffer PM. IgG-detection devices for the Tus-Ter-lock immunoPCR diagnostic platform. Analyst 2011;136:4815–21. Parthasarathy N, DeShazer D, England M, Waag DM. Polysaccharide microarray technology for the detection of Burkholderia pseudomallei and Burkholderia mallei antibodies. Diagn Microbiol Infect Dis 2006;56:329–32. Stone JK, Mayo M, Grasso SA, Ginther JL, Warrington SD, Allender CJ, et al. Detection of Burkholderia pseudomallei O-antigen serotypes in near-neighbor species. BMC Microbiol 2012;12:250. Thepthai C, Smithtikarn S, Suksuwan M, Songsivilai S, Dharakul T. Serodiagnosis of melioidosis by a competitive enzyme-linked immunosorbent assay using a lipopolysaccharide-specific monoclonal antibody. Asian Pac J Allergy Immunol 2005;23:127–32. Tuanyok A, Stone JK, Mayo M, Kaestli M, Gruendike J, Georgia S, et al. The genetic and molecular basis of O-antigenic diversity in Burkholderia pseudomallei lipopolysaccharide. PLoS Neg Trop Dis. 2012;6:e1453. Wiersinga WJ, Currie BJ, Peacock SJ. Melioidosis. New England J Med. 2012;367: 1035–44. White NJ. Melioidosis. Lancet 2003;361:1715–22. Wikraiphat C, Charoensap J, Utaisincharoen P, Wongratanacheewin S, Taweechaisupapong S, Woods DE, et al. Comparative in vivo and in vitro analyses of putative virulence factors of Burkholderia pseudomallei using lipopolysaccharide, capsule and flagellin mutants. FEMS Immunol Med Microbiol 2009;56:253–9. Wongratanacheewi S, Sermswan RW, Anuntagool N, Sirisinha S. Retrospective study on the diagnostic value of IgG ELISA, dot immunoassay and indirect hemagglutination in septicemic melioidosis. Asian Pac J Allergy Immunol 2001;19:129–33.
Contents lists available at ScienceDirect
Diagnostic Microbiology and Infectious Disease journal homepage: www.elsevier.com/locate/diagmicrobio
Improved diagnosis of melioidosis using a 2-dimensional immunoarray☆ Alanna E. Sorenson a, Natasha L. Williams c, Jodie L. Morris c, Natkunam Ketheesan c, Robert E. Norton d, Patrick M. Schaeffer a, b,⁎ a
School of Pharmacy and Molecular Sciences, James Cook University, Douglas QLD 4811, Australia Comparative Genomics Centre, James Cook University, Douglas QLD 4811, Australia Infectious Disease and Immunology, Australian Institute Tropical Health and Medicine, James Cook University, Douglas QLD 4811, Australia d Queensland Health Pathology Service, The Townsville Hospital, Douglas QLD 4811, Australia b c
a r t i c l e
i n f o
Article history: Received 17 May 2013 Received in revised form 9 July 2013 Accepted 10 July 2013 Available online 14 September 2013 Keywords: Melioidosis Burkholderia pseudomallei Serology Two-dimensional Immunoarray ELISA Diagnostics
a b s t r a c t Melioidosis is caused by the Gram negative bacterium Burkholderia pseudomallei. The gold standard for diagnosis is culture, which requires at least 3–4 days obtaining a result, hindering successful treatment of acute disease. The existing indirect haemagglutination assay (IHA) has several disadvantages, in that approximately half of patients later confirmed culture positive are not diagnosed at presentation and a subset of patients are persistently seronegative. We have developed 2 serological assays, an enzyme-linked immunosorbent assay (ELISA), and a 2-dimensional immunoarray (2DIA), capable of detecting antibodies in patient sera from a greater proportion of IHA-negative patient subsets. The 2DIA format can distinguish between different LPS serotypes. Currently, the 2DIA has a sensitivity and specificity of 100% and 87.1%, respectively, with 100% of culture-positive, IHA-negative samples detected. The ELISA has a sensitivity and specificity of 86.2% and 93.5%, respectively, detecting 67% of culture-positive, IHA-negative samples. The ELISA and 2DIA tests described here are more rapid and reliable for serological testing compared to the existing IHA. © 2013 Elsevier Inc. All rights reserved.
1. Introduction Melioidosis is caused by the Gram-negative bacterium Burkholderia pseudomallei. Classically, melioidosis is characterized by pneumonia and multiple abscesses (Wiersinga et al., 2012). However, presentation varies greatly with severity ranging from subclinical to acute fulminant sepsis or chronic infection that can mimic other diseases. Such protean clinical manifestations complicate diagnosis and contribute to the mortality rate ranging from 16% in northern Australia to 40% in northeast Thailand (Limmathurotsakul et al., 2010b). B. pseudomallei is susceptible to a limited number of antibiotics. In endemic areas, melioidosis is an important cause of morbidity and mortality in humans and animals (Limmathurotsakul and Peacock, 2011). Important risk factors include diabetes mellitus, chronic renal failure, chronic lung disease, and excessive alcohol use. The current gold standard for diagnosis is culture, which often requires enrichment followed by several days of incubation (Limmathurotsakul et al., 2010a). Culture is also an imperfect gold standard due to low sensitivity and negative predictive values (Limmathurotsakul et al., 2010a). In the case of acute infections, bacterial sepsis can develop in a few days and requires immediate treatment with the correct antibiotics as this bacterium is highly drug resistant (White, 2003).
☆ Conflict of interest: The authors declare that they have no conflict of interest. ⁎ Corresponding author. Tel.: +61-(0)7-4781-6388. E-mail address: [email protected] (P.M. Schaeffer). 0732-8893/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.diagmicrobio.2013.07.009
Death usually follows within a few days if appropriate treatment is not rapidly applied. Improving presumptive diagnosis of melioidosis is important as some empirical antibiotic regimens employed in suspected bacterial sepsis do not adequately treat B. pseudomallei infection (Wiersinga et al., 2012). In northern Australia, the most common serological test used is the indirect haemagglutination assay (IHA) (Ashdown, 1987). The IHA has relatively high specificity but poor sensitivity, with values ranging from 92–100% for specificity and 50–85% for sensitivity, respectively (Ashdown et al., 1989; Wongratanacheewi et al., 2001). Approximately half of patients later confirmed to be culture positive are not able to be diagnosed by IHA at presentation, and a subset of these patients are found to be persistently seronegative by IHA (Cheng et al., 2006; Harris et al., 2009). The use of isolates from culture-positive, IHA-negative patients as antigen in IHA has been unsuccessful, indicating these patients do not develop antibodies to the specific epitopes adsorbed onto erythrocytes in the IHA (Harris et al., 2009). However, the same patients have been demonstrated to have responses to antigens from their infecting isolates, indicating antigen display is important in developing effective serological assays. The relatively poor sensitivity of IHA has led to the development of other assays to detect antibodies to B. pseudomallei, most of which are enzyme immunoassays (EIA) based on either crude wholecell preparations (Chantratita et al., 2007), recombinant proteins (Allwood et al., 2008; Felgner et al., 2009; Hara et al., 2013), or lipopolysaccharide (LPS) fractions (Anandan et al., 2010; Thepthai et al., 2005).
210
A.E. Sorenson et al. / Diagnostic Microbiology and Infectious Disease 77 (2013) 209–215
Protein-based EIAs have been found to generally have greater sensitivity than IHA (Chantratita et al., 2007; Felgner et al., 2009). However, the possibility of cross-reactivity is increased in proteinbased assays due to many seroreactive proteins in B. pseudomallei having homologs in other bacteria with similar protein sequences. LPS has been found to be an important virulence factor and antigenic component of B. pseudomallei (Tuanyok et al., 2012). While LPS-based EIAs have high specificity, sensitivity is generally poor due to the inclusion of very few or a single isolate. Studies have demonstrated heterogeneity in LPS among B. pseudomallei isolates with the identification of 4 phenotypes to date, Type A, B, B2, and Rough (Anuntagool et al., 2006; Tuanyok et al., 2012). Type A (typical) is the most common phenotype, and the predominant phenotype found in Thailand and Australia. Types B and B2 (atypical) are less common, with Type B2 only found in Australia and Papua New Guinea to date. The rough variant lacks the O-antigen, a phenotype found only in clinical isolates in Australia thus far (Tuanyok et al., 2012). Failure to use sufficient coverage of LPS phenotypes in assays may result in false negatives, particularly in Australia where atypical LPS phenotypes have been found to be more common than in Southeast Asia (Tuanyok et al., 2012). We have recently developed 2 new LPS-based serological assays capable of detecting antibodies in patient sera from a greater proportion of IHA-negative patient subsets, significantly improving sensitivity compared to the IHA (Cooper et al., 2013). These assays were developed with the most common LPS phenotype found in Australia, i.e., type A. One of these assays used the principle of immuno-PCR (Morin et al., 2010; Morin et al., 2011) to reduce sample volume and improve sensitivity. However, the heterogeneity in LPS phenotypes associated with Australian melioidosis cases necessitates the enhancement of serological assays to include all possible LPS phenotypes to improve diagnostic precision. LPS has been implicated as a major cause of septicaemia, involved in the hyper-inflammatory response (Leon et al., 2008). Septicaemia is a significant cause of death in melioidosis cases, emphasizing the potential importance of LPS phenotype in disease progression and prognosis. This is further highlighted by the demonstration of serum susceptibility in Type B2 and rough phenotype LPS B. pseudomallei isolates (Tuanyok et al., 2012), indicating these genotypes may have lower virulence. Here, we present the development of 2 new serological assays including LPS fractions from various B. pseudomallei isolates representing the 3 most common LPS phenotypes. An enzyme-linked immunosorbent assay (ELISA) incorporating the 3 most common LPS phenotypes has a sensitivity and specificity of 86.2% and 93.5%, respectively, detecting 67% of culture-positive, IHA-negative samples. As the ELISA could not enable serotyping, a further assay was developed, capable of distinguishing between different LPS serotypes in multiplex using a 2-dimensional immunoarray format (2DIA). This assay has a sensitivity and specificity of 100% and 87.1%, respectively, detecting 100% of culture-positive, IHA-negative samples. 2. Methods 2.1. Isolation of secreted LPS Glycerol stocks of K96243, JCUCC152, and JCUJP23 B. pseudomallei isolates were streaked onto separate Luria Bertani (LB) agar plates and incubated at 37 °C for 24 h. Single colonies of each isolate were inoculated into 5-mL LB media and incubated at 37 °C for 18 h with shaking at 150 RPM. Overnight cultures of each isolate were inoculated into separate 1-L flasks with 200-mL sterile LB media and incubated at 37 °C for 30 h with shaking at 150 RPM. Culture supernatants were removed and passed through a 0.2-μm filter. Finely ground ammonium sulphate was added to the filter-sterilized culture supernatants at a concentration of 0.5 mg mL −1 and incubated on ice with shaking for 2 h. Culture supernatants were centrifuged at 18,000
× g for 1 h, the supernatant decanted and retained for further analysis, and the pellet resuspended in 10 mmol/L phosphate buffer, pH 7.4 (2 mL per 25 mL culture supernatant). To retain LPS and remove protein from the antigenic fraction, samples were heated to 95 °C for 15 min, cooled, and then centrifuged at 16,000 × g for 10 min. Supernatants were transferred to fresh tubes and stored at −20 °C for later use. 2.2. LPS quantification LPS content of antigenic fractions was determined using a phenolsulphuric acid total carbohydrate quantification method (Fox and Robyt, 1991). Briefly, triplicate 25-μL samples of LPS at various dilutions were combined with 25 μL 5% w/v phenol and shaken gently for 30 s, along with a standard curve of D-glucose at 0, 50, 75, 100, 150, and 200 μg mL −1. Test plates were placed on ice, and 125-μL concentrated sulphuric acid was added to each reaction. The plates were shaken gently again for 30 s, then heated at 80°C for 30 min, and cooled. Reaction mixture (100 μL) was transferred to a microtitre plate and read at 490 nm. 2.3. Sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) Purified LPS samples were combined 1:1 with 2× Laemmli buffer (50 mmol/L Tris-HCl, pH 6.8, 2% SDS, 10% glycerol, 0.01% bromophenol blue) and heated at 95 °C for 5 min then fractionated on a 10% acrylamide stacking gel at 150 V for 45 min. 2.4. Modified silver staining SDS-PAGE fractionated LPS samples were stained using a modified silver stain to detect LPS (Fomsgaard et al., 1990). To detect protein, the same method was used with 2 modifications. First, periodic acid was omitted from the fixative, and secondly, the gel was incubated with 100 mL 0.02% sodium thiosulphate for 1 min then washed 3 times with MilliQ water for 20 s prior to addition of the silver staining solution. Subsequent to development, the reaction was stopped by incubating the gel with 5% acetic acid for 5 min. 2.5. Confirmation of LPS genotype by quantiative polymerase chain reaction (qPCR) Glycerol stocks of K96243, JCUCC152, and JCUJP23 B. pseudomallei isolates were streaked onto separate Ashdown agar plates and incubated at 37 °C for 24 h. Single colonies of each isolate were streaked onto minimal media and incubated at 37 °C for 72 h. Single colonies of each isolate were selected an a genomic DNA extraction performed using a Roche genomic DNA PCR template kit according to the manufacturer's instructions for bacterial gDNA extraction. Primer sequences for amplification of LPS genotype specific products were obtained from Tuanyok et al. (2012). PCR assays were performed in 20-μL reactions in triplicate for Type A, B, and B2 genotypes on K96243, JCUCC152, and JCUJP23, with all 3 tested with each primer set. The reactions and subsequent melt curve analysis were performed on a Bio-Rad IQ5 thermocycler as described by Tuanyok et al. (2012). 2.6. Sera IHA-positive patients, patients with persistently IHA-nonreactive sera who had culture-proven melioidosis, and healthy controls were requested to provide blood samples. Initial IHAs were performed by Pathology Queensland, The Townsville Hospital (Ashdown, 1987). Ethical approval for collection of sera was obtained from the Townsville Health Service District Ethics Committee (#2502 and #7104). A total of 60 serum samples were tested along with positive and negative controls. Serum samples were initially provided and
A.E. Sorenson et al. / Diagnostic Microbiology and Infectious Disease 77 (2013) 209–215
tested in blind fashion, then identified and confirmed subsequently. A total of 29 melioidosis patient samples were tested, of which 17 were culture positive, IHA positive and 12 culture positive, IHA negative. In addition, 31 samples from healthy controls were tested. 2.7. Immunoblotting Separate SDS-PAGE of each LPS type in quadruplicate was run and transferred onto pre-wet polyvinilidene fluoride (PVDF) membrane (Biorad, Gladesville, Australia) via semi-dry electroblotting at 15 V for 25 min. Following blocking with 5% skim milk in phosphate-buffered saline (PBS), pH 7.4, at room temperature (RT) for 1 h, blots were probed with human serum from culture-confirmed melioidosis patients known to have been infected with B. pseudomallei isolates from each LPS genotype, in addition to human serum from a healthy control. Serum was diluted 1:100 in 1% skim milk in PBS, pH 7.4, and incubated with blots at RT for 1 h. After washing 3 times for 5 min PBS, pH 7.4–0.05% Tween-20 (PBS-T), blots were probed with peroxidaseconjugated protein G (Sigma, Sydney, Australia) diluted 1:5000 in 1% skim milk in PBS, pH 7.4, for 1 h at RT. PVDF membranes were washed again 3 times with PBS-T and developed with 5-mL SIGMAFAST™ 3, 3′-diaminobenzidine/H2O2 solution. 2.8. G-peroxidase ELISA Nunc Maxisorb 96-well round-bottom immunoplates (Thermo Fisher Scientific, Roskilde, Denmark) were coated with 50-μL combined LPS at 2 μg mL −1 each for the K96243, JCUCC152, and JCUJP23 B. pseudomallei isolates, respectively, in carbonate/bicarbonate buffer, pH 9.6, overnight at 4 °C. Wells were blocked with 50-μL 1% bovine serum albumin (BSA) in bind and wash (BW) buffer (20 mmol/L Tris, 150 mmol/L NaCl, 0.005% Tween-20) at RT for 1 h. Human serum (50 μL) was applied at 1:200 in BW and incubated at RT for 1 h. After washing 3 times with BW, 50-μL peroxidase-conjugated protein G (Sigma) diluted 1:5000 in BW was applied for 1 h at RT. Wells were washed again 3 times with BW and developed with tetramethylbenzidine for 5 min (Cooper et al., 2013). Optical densities (ODs) were measured at 450 nm with a Bio-strategy Versa Max microplate reader and corrected by subtracting background values, which were obtained by omitting serum. 2.9. Two-dimensional immunoarray PVDF membrane cut into 12.5 cm 2 squares was wet with methanol for 30 s then rinsed with PBS prior to clamping under a Surf-Blot 10.5 chamber (Idea Scientific, Minneapolis, MN, USA). LPS samples (400 μL) were applied in separate lanes of the chamber in PBS, pH 7.4, and incubated at RT for 1 h with gentle rocking. Optimal dilutions of LPS for each genotype were determined by checkerboard titration. Optimal total carbohydrate content values of 6.25 μg mL −1, 12.5 μg mL −1, and 6.25 μg mL −1 were determined for the K96243, JCUCC152, and JCUJP23 B. pseudomallei isolates, respectively. LPS from Escherichia coli and Pseudomonas aeruginosa at 5 μg mL −1 were included as controls. Samples were decanted off, and PVDF was washed with PBS, pH 7.4–0.05% Tween-20 for 5 min. Membranes were blocked with 5% skim milk in PBS, pH 7.4, at RT for 1 h. After blocking, blots were turned 90° re-clamped under a Surf-Blot 10.5 chamber and probed with 400-μL human sera at a dilution of 1:80 in 1% skim milk in PBS, pH 7.4, for 1 h at RT. After washing 3 times for 5-min PBS, pH 7.4– 0.05% Tween-20 (PBS-T), blots were probed with peroxidaseconjugated protein G (Sigma) diluted 1:5000 in 1% skim milk in PBS, pH 7.4, for 1 h at RT. PVDF membranes were washed again and developed as described above. Intensity of grid squares was determined via integration using Image J software (NIH, Bethesda, MD, USA).
211
2.10. Investigation of common reactivity to crude B. pseudomallei LPS phenotypes Crude LPS samples from each LPS phenotype, E. coli and P. aeruginosa, were combined 1:1 with 2× Laemmli buffer and then fractionated on a 4–15% Mini-PROTEAN® TGX™ Precast acrylamide gel (BioRad) at 150 V for 45 min. Samples were transferred onto nitrocellulose membrane via semi-dry electroblotting at 25 V for 3 min in a Trans-Blot® Turbo™ Transfer System (Biorad). Following blocking with 2.5% BSA in PBS, pH 7.4, at RT for 1 h, blots were probed with human serum from culture-confirmed melioidosis patients demonstrated to have common reactivity to crude samples of Type A, B, and B2 Bps LPS phenotypes. Serum was diluted 1:100 in 1% BSA in PBS, pH 7.4, and incubated with blots at RT for 1 h. After washing 3 times for 5 min with PBS, pH 7.4–0.05% Tween-20 (PBS-T), blots were probed with peroxidase-conjugated protein G (Sigma) diluted 1:5000 in 1% BSA in PBS, pH 7.4, for 1 h at RT. PVDF membranes were washed again 3 times with PBS-T and developed with 5-mL SIGMAFAST™ 3, 3′-diaminobenzidine/H2O2 solution. 2.11. Investigation of common antigen reactivity To determine whether the 45 kDa common antigen was proteinaceous, samples of LPS representative of each phenotype were treated with 100 μg mL −1 proteinase K at 56 °C for 30 min and inactivated at 95 °C for 10 min. Identical samples received the same treatment without the addition of proteinase K. Duplicate untreated and treated samples were fractionated via SDS-PAGE then transferred onto nitrocellulose membrane, as described previously. The membrane was probed with common-reactive sera, as described previously. An identical gel was retained for protein staining with AcquaStain (Acquascience, Uckfield, TN, USA) at RT for 3 h. 3. Results 3.1. LPS genotyping and phenotyping To confirm each B. pseudomallei isolate was representative of each LPS type, genotyping and phenotyping were performed. Culture samples were genotyped for LPS subtypes, as described by Tuanyok et al. (2012). Triplicate samples of DNA from each isolate were analysed by quantitative PCR specific for each LPS genotype. The PCR products were subjected to a standard melt curve from 60 °C to 95 °C. Melting temperatures for Type A (K96243), Type B (JCUCC152), and Type B2 (JCUJP23) were 88 °C, 84 °C, and 89 °C, respectively (Fig. 1A). Amplicons for LPS-specific PCR were only detected in genomic DNA from their respective LPS phenotype-producing isolates, and the melting temperatures were consistent with those obtained by Tuanyok et al. (2012). LPS samples at several concentrations obtained from these isolates were fractionated by SDS-PAGE, then analysed using a modified silver staining protocol to visualise differences in their LPS laddering patterns. The electrophoretic patterns of the LPS phenotypes (Fig. 1B) were found to be consistent with those described in the literature (Anuntagool et al., 2006; Tuanyok et al., 2012). 3.2. Immunoblotting characterisation of LPS genotypes To confirm each LPS phenotype possessed the expected antigenic profile, immunoblotting was performed with serum from cultureconfirmed melioidosis patients known to have been infected with B. pseudomallei isolates representing each LPS phenotype. This immunoblotting demonstrated LPS phenotype-specific antibodies were produced during infection with the respective isolates (Fig. 2). Serum from a patient that had been infected with a Type B LPS isolate demonstrated cross-reactivity with Type B2 LPS. Interestingly, sera
212
A.E. Sorenson et al. / Diagnostic Microbiology and Infectious Disease 77 (2013) 209–215
Fig. 1. Genotyping and phenotyping of B. pseudomallei LPS subtypes. Melt curve analysis of amplicons from each LPS phenotype B. pseudomallei isolate (a). Melting temperatures (Tm) were consistent with the expected Tm for each amplicon. Amplicons were specific to their respective LPS phenotype-producing isolates (b). Modified silver stain for LPS showing characteristic banding pattern for 3 of the known LPS phenotypes in B. pseudomallei. Each phenotype was tested at several dilutions with microgram amounts of LPS indicated below each lane. Image has been inverted for improved contrast between LPS bands and background.
from a patient infected with a B2 genotype isolate did not react against B genotype LPS.
3.3. Detection of antibodies against B. pseudomallei LPS antigens by indirect G-peroxidase ELISA Initially, an indirect ELISA was developed with equivalent amounts of each LPS phenotype used as antigen. Serum dilutions 1:200 were found to be optimal for detection of antibodies by G-peroxidase ELISA via checkerboard assays with positive and negative control sera. An
OD450 ratio between the sample and negative control of N1.4 was considered to be seropositive for the G-peroxidase ELISA (Fig. 3A). A summary of results, including IHA results (performed by Pathology Queensland, The Townsville Hospital) for each patient and control, is provided in Table 1. Sensitivity and specificity of the IHA were 58.6% and 100%, respectively, for the samples used in this study. Sensitivity and specificity of the ELISA were 86.2% and 93.5%, respectively. As the ELISA could not enable serotyping, a further assay, the 2DIA, was developed. 3.4. Detection of antibodies against B. pseudomallei LPS antigens by 2DIA To enable serotyping of B. pseudomallei infecting isolates, the 2DIA was developed (Fig. 3B). Serum dilutions of 1:80 were found to be optimal for detection of antibodies by 2DIA via checkerboard assays with positive and negative control sera. ImageJ corrected scores of N4000 were considered to be seropositive for Type A and B LPS and N4500 for Type B2 in the 2DIA (Fig. 3C). Sensitivity and specificity for the 2DIA were 100% and 87.1%, respectively. A total of 13 melioidosis patient serum samples reacted with a common antigen present in all 3 LPS fractions. Of the serotype-specific reacting patient sera, 8 reacted to Type A LPS; 4, to Type B; and 4, to Type B2. Four healthy control serum samples reacted to specific LPS phenotypes; 2, to Type B; and 1, to Type B2 and A, respectively. Of the IHA-positive patients, 88% were identified as having Type A LPS-specific or common antigen-specific antibodies, with 35% having Type A LPS-specific only antibodies. Alternatively, of the IHA-negative patients, 50% were identified as having Type B or B2 LPS-specific antibodies. Scanned images of the 2DIA blots (Figure S1) used for ImageJ analysis (raw data) are included in the supplementary information. 3.5. Investigation of common reactivity to crude Bps LPS phenotypes A series of Western blots were performed to investigate the antigens responsible for common reactivity in the LPS fractions used in the 2DIA. Western blotting with several patient sera, which demonstrated common reactivity in the 2DIA, exhibited reactivity to capsular polysaccharide and/or a band of approximately 45 kDa (Fig. 4). Western blotting with crude LPS samples, pre and postproteinase K digestion confirmed, the antigen of approximately 45 kDa was proteinaceous (Fig. 5A). SDS-PAGE and subsequent incubation with a protein staining reagent also demonstrated the presence of a protein antigen of approximately 45 kDa (Fig. 5B). 4. Discussion
Fig. 2. Reaction of culture-confirmed melioidosis patient sera with LPS from B. pseudomallei isolates K96243 (Type A), JCUCC152 (Type B), and JCUJP23 (Type B2). Reactivity of sera from a patient infected with an A genotype isolate (a). Lack of reactivity of sera from a healthy control (b). Reactivity of sera from patients infected with B (c) and B2 (d) genotype isolates, respectively.
Both new immunoassay platforms developed in this study target the detection of LPS-specific human IgG, with the 2DIA able to confirm the diagnosis of all IHA-negative patients included in this study. These
A.E. Sorenson et al. / Diagnostic Microbiology and Infectious Disease 77 (2013) 209–215
a
213
c
15
10
LPS Fractions
OD450 Ratio
Serum Samples
1
5
PVDF Membrane
2
Surf-Blot
0 POS
b
NEG
Protein GPeroxidase DAB
50K POS NEG
LPS Fractions
Intensity (ImageJ)
3
40K 30K 20K 10K
1 2
3
4 5 6 7 8 9 10 11 12 13 14 15
A B B2
0 POS
NEG
Fig. 3. Comparison between indirect G-peroxidase ELISA (a) and the 2DIA (b) for detection of IgG against Type A, B, and B2 B. pseudomallei LPS phenotypes in melioidosis patient sera. For the ELISA, OD450 ratios between the sample and negative control of N1.25 were considered to be seropositive. For the 2DIA, corrected ImageJ intensity scores N4000 were considered to be seropositive for Type A and B LPS, with N4500 considered to be seropositive for Type B2. Diagram demonstrating the use of the Surf-Blot chamber to perform 2DIA assays (c). LPS samples are blotted onto PVDF membrane in separate channels (Panel 1), then blocked before turning 90° and probing with serum samples in separate channels (Panel 2). Positive and negative control samples are included in each run. The final panel (Panel 3) demonstrates a typical output (where samples 3, 4, 6, 12, 14, and 15 are considered to be positive reactors). ImageJ (NIH) is used to measure the relative intensity of each square.
new assays were able to cover 3 of the 4 known LPS genotypes of B. pseudomallei, which were confirmed by PCR amplification of LPS genotype-specific genes and modified silver staining for LPS. As determined by Anuntagool et al. (2006), atypical LPS genotypes are more common in Australian B. pseudomallei isolates. The failure to include representative isolates from these genotypes in immunoassays for melioidosis may complicate diagnosis, particularly in Australia. The inclusion of representative isolates from the Type B and B2 genotypes in our assays has resulted in the improved detection of specific antibodies in patient sera previously found to be persistently negative by the IHA. The antigen used by the Townsville Hospital in IHAs was prepared using techniques described previously (Ashdown, 1987) from 5 B. pseudomallei isolates of unknown identity. However, it is likely that the isolates used for development of the IHA in Australia are all from the Type A LPS genotype. This is reflected by the high percentage of IHA-positive melioidosis patient sera reacting to Type A LPS and/or common Bps antigen in the 2DIA immunoassay. Additionally, this may explain the previously high proportion of persistently IHA-negative melioidosis patient sera, as 50% of these serum samples reacted specifically to Type B and/or B2 LPS in the 2DIA immunoassay. The addition of antigens from B. pseudomallei isolates representative of the additional LPS phenotypes to the existing IHA antigen complement may improve the sensitivity of this assay. In this study, sensitivity for the IHA, ELISA, and 2DIA was determined to be 58.6%, 86.2%, and 100%, respectively, with specificity for the IHA, ELISA, and 2DIA of 100%, 93.5%, and 87.1%, respectively. The ELISA developed in this study contained 3 B. pseudomallei LPS types (A, B, and B2), resulting in comparable specificity and improved sensitivity compared to the IHA. The reduced sensitivity of the ELISA compared to the 2DIA is most likely due to the increased dilution factor of both the combined LPS fractions and test sera. The 2DIA had slightly reduced specificity and greatly improved sensitivity com-
pared to the IHA, in addition to enabling serotyping of melioidosis patient sera. This distinction may be useful in determining whether LPS genotypes are correlated with disease presentation, prognosis, and isolate virulence. While sensitivity was greatly improved with the ELISA and 2DIA compared to the IHA, specificity was reduced, increasing the potential for false positives, particularly in healthy patients from endemic areas who may have been exposed to B. pseudomallei or closely related species. In routine diagnostics, the ELISA may be of greater utility due to its simplicity and improved sensitivity compared to the IHA. Nevertheless, we expect that further development of the 2DIA into a lateral flow–based assay would provide the best diagnostic solution. Interestingly, sera from the patient infected with a B2 genotype isolate failed to react against B LPS. Serum from melioidosis patients infected with Bps isolates of B genotype was demonstrated to crossreact with B2 genotype LPS by Tuanyok et al. (2012). However, Tuanyok et al. (2012) did not investigate the profile of melioidosis patients infected with Bps isolates of B2 genotype. In the current study, the cross-reactivity of anti-B LPS antibodies with B2 LPS was confirmed, while indicating the reciprocal cross-reactivity does not occur. In this study, several patient sera were found to react to genotype B2 LPS only, indicating this LPS phenotype is antigenically distinct. This finding also demonstrated the importance of the inclusion of all known LPS genotypes in serological assays in order to successfully diagnose patients, especially those infected with Bps isolates with B2 genotype LPS. In some cases, reactivity was common to all 3 LPS genotypes, indicating the presence of a heat-stable antigen remaining in the crude LPS preparations. Western blotting with several patient sera, which demonstrated common reactivity in the 2DIA, exhibited reactivity to capsular polysaccharide and/or a band of approximately 45 kDa. Capsular polysaccharide, in addition to being a virulence factor (Wikraiphat et al., 2009), is known to be antigenic
214
A.E. Sorenson et al. / Diagnostic Microbiology and Infectious Disease 77 (2013) 209–215
Table 1 Serological assay results for culture-confirmed melioidosis patients and controls. Sample no.
IHAa
ELISAb
2DIAc
Typed
Sample no.
IHA
5 16 17 32 49 51 12 19 22 29 48 59 7 31 60 3 30 20 24 2 26 23 33 35 15 8 9 34 10
+ve +ve +ve +ve +ve +ve +ve +ve +ve +ve +ve +ve +ve +ve +ve +ve +ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve
+++ +++ +++ +++ +++ +++ ++ ++ ++ ++ ++ ++ ++ ++ ++ + + + + ++ ++ + + + + − − − −
+++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ ++ ++ ++ ++ ++ +++ +++ ++ ++ ++ ++ ++ + ++ ++ ++ +
C C C C C A A C A A C B C C A B2 A B2 B C A B2 A C B B C C B2
28 14 57 44 1 4 6 11 13 18 21 25 27 36 37 38 39 40 41 42 43 45 46 47 50 52 53 54 55 56 58
Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct Ct
−ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve −ve
ELISA
2DIA
Type
+ + − − − − − − − − − − − − − − − − − − − − − − − − − − −
+ + ++ + − − − − − − − − − − − − − − − − − − − − − − − − − − −
B B A B2 − − − − − − − − − − − − − − − − − − − − − − − − − − −
a +ve designates IHA positive melioidosis patients (IHA N40), −ve designates IHA negative melioidosis patients, Ct designates healthy controls, and each group is distinguished by shading. b ELISA results indicate OD450 ratios between the sample and negative control sera where – is b1.25, + is b1.25b2.0, ++ is N2.0b6.0, and +++ is N6.0. c 2DIA results indicate corrected ImageJ values where – is b4000, + is N4000b7000, ++ is N7000b15,000, and +++ is N15,000. d Type designates LPS genotype where A, B, and B2 indicate reactivity, and C indicates reactivity to antigen common to all 3 isolates.
(Parthasarathy et al., 2006). Protein staining and Western blotting with crude LPS samples, pre and post-proteinase K digestion, confirmed the antigen of approximately 45 kDa was proteinaceous. The identification of the proteinaceous antigen and performing 2DIA assays with and without proteinase K digestion on LPS fractions would enable improved serotyping. The proteinaceous antigen would
Fig. 5. Western blotting (a) and protein staining (b) with crude LPS samples, pre and post-proteinase K digestion. Loss of band of approximately 45 kDa in proteinase K treated samples (arrow).
be particularly useful in the 2DIA, as it has factors that would contribute to long-term antigen stability in that it is demonstrably either extremely heat-stable or capable of refolding without loss of antigenicity. Indeed, the relative stability of LPS and the 45 kDa proteinaceous antigen lend themselves to the conversion of the 2DIA to a lateral flow–based assay that could be developed as a commercial diagnostic test. Several healthy control sera reacted in both immunoassays, which may reflect the endemic nature of B. pseudomallei in the region. Unfortunately, the use of serological techniques is problematic in some endemic areas due to high background seropositivity in the healthy population (Limmathurotsakul and Peacock, 2011). However, in northern Queensland and the Northern Territory of Australia, the seroprevalence of antibodies to B. pseudomallei is relatively low at approximately 2.5% (Lazzaroni et al., 2008) and 3.0% (James et al., 2013), respectively, making serology a potentially advantageous addition to culture in the diagnosis of melioidosis in this region. Alternatively, this reactivity may be due to prior infection with a near-neighbour Burkholderia species, as it has been demonstrated that some of these species possess common LPS genotypes with B. pseudomallei (Stone et al., 2012). While cross-reactivity between B. pseudomallei and nearneighbour species is possible, the immunoassays developed in this study could improve the rapidity of diagnostic confirmation in cases where melioidosis is suspected and immediate treatment is necessary for patient survival. The use of the assays developed in this study may also improve seroprevalence surveys by detecting antibodies in the healthy population that are unable to be detected using the IHA. In conclusion, the immunoassays developed in this study have comparable specificity and greatly improved sensitivity compared to the IHA and are more rapid than culture, potentially leading to improved diagnosis of melioidosis and patient outcome. These assays would be of particular use in non-endemic areas in suspected melioidosis cases in travelers and military personnel returning from endemic areas. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.diagmicrobio.2013.07.009. Acknowledgments
Fig. 4. Western blotting with several patient sera, which demonstrated common reactivity in the 2DIA. Reactivity exhibited to capsular polysaccharide (N190 kDa) and/ or unidentified proteinaceous antigen (45 kDa).
The authors would like to thank the Smart Futures Fund: National and International Research Alliances Program, the National Health and Medical Research Council, James Cook
A.E. Sorenson et al. / Diagnostic Microbiology and Infectious Disease 77 (2013) 209–215
University, and the Queensland Tropical Health Alliance for financial support. References Allwood EM, Logue CA, Hafner GJ, Ketheesan N, Norton RE, Peak IR, et al. Evaluation of recombinant antigens for diagnosis of melioidosis. FEMS Immunol Med Microbiol 2008;54:144–53. Anandan S, Augustine A, Mathai E, Jesudason MV. Evaluation of IgM ELISA using a sonicate and a lipopolysaccharide antigen for the serodiagnosis of melioidosis. Indian J Med Microbiol 2010;28:158–61. Anuntagool N, Wuthi ekanun V, White NJ, Currie BJ, Sermswan RW, Wongratanacheewin S, et al. Lipopolysaccharide heterogeneity among Burkholderia pseudomallei from different geographic and clinical origins. Am J Trop Med Hyg 2006;74:348–52. Ashdown LR. Indirect haemagglutination test for melioidosis. Med J Aust 1987;147: 364–5. Ashdown LR, Johnson RW, Koehler JM, Cooney CA. Enzyme-linked immunosorbent assay for the diagnosis of clinical and subclinical melioidosis. J Infect Dis 1989;160: 253–60. Chantratita N, Wuthiekanun V, Thanwisai A, Limmathurotsakul D, Cheng AC, Chierakul W, et al. Accuracy of enzyme-linked immunosorbent assay using crude and purified antigens for serodiagnosis of melioidosis. Clin Vaccine Immunol 2007;14: 110–3. Cheng AC, O'Brien M, Freeman K, Lum G, Currie BJ. Indirect hemagglutination assay in patients with melioidosis in northern Australia. Am J Trop Med Hyg 2006;74: 330–4. Cooper A, Williams NL, Morris JL, Norton RE, Ketheesan N, Schaeffer PM. ELISA and immuno-polymerase chain reaction assays for the sensitive detection of melioidosis. Diagn Microbiol Infect Dis 2013;75:135–8. Felgner PL, Kayala MA, Vigil A, Burk C, Nakajima-Sasaki R, Pablo J, et al. A Burkholderia pseudomallei protein microarray reveals serodiagnostic and cross-reactive antigens. Proc Natl Acad Sci USA 2009;106:13499–504. Fomsgaard A, Freudenberg MA, Galanos C. Modification of the silver staining technique to detect lipopolysaccharide in polyacrylamide gels. J Clin Microbiol 1990;28: 2627–31. Fox JD, Robyt JF. Miniaturization of three carbohydrate analyses using a microsample plate reader. Analyt Biochem. 1991;195:93–6. Hara Y, Chin CY, Mohamed R, Puthucheary SD, Nathan S. Multiple-antigen ELISA for melioidosis—a novel approach to the improved serodiagnosis of melioidosis. BMC Infect Dis 2013;13:165. Harris PNA, Ketheesan N, Owens L, Norton RE. Clinical features that affect indirecthemagglutination-assay responses to Burkholderia pseudomallei. Clin Vaccine Immunol 2009;16:924–30.
215
James GL, Delaney B, Ward L, Freeman K, Mayo M, Currie BJ. Surprisingly low seroprevalence for Burkholderia pseudomallei in exposed healthy adults in the highly melioidosis-endemic Darwin region of Tropical Australia. Clin Vacc Immunol 2013;20:759–60. Lazzaroni SM, Barnes JL, Williams NL, Govan BL, Norton RE, LaBrooy JT, et al. Seropositivity to Burkholderia pseudomallei does not reflect the development of cell-mediated immunity. Trans R Soc Trop Med Hyg 2008;102(Suppl 1):S66–70. Leon CG, Tory R, Jia J, Sivak O, Wasan KM. Discovery and development of Toll-like receptor 4 (TLR4) antagonists: a new paradigm for treating sepsis and other diseases. Pharm Res 2008;25:1751–61. Limmathurotsakul D, Wongratanacheewin S, Teerawattanasook N, Wongsuvan G, Chaisuksant S, Chetchotisakd P, et al. Increasing incidence of human melioidosis in Northeast Thailand. Am J Trop Med Hyg 2010a;82:1113–7. Limmathurotsakul D, Jamsen K, Arayawichanont A, Simpson JA, White LJ, Lee SJ, et al. Defining the true sensitivity of culture for the diagnosis of melioidosis using Bayesian latent class models. PLoS One 2010b;5:e12485. Limmathurotsakul D, Peacock SJ. Melioidosis: a clinical overview. British Med Bull. 2011;99:125–39. Morin I, Dixon NE, Schaeffer PM. Ultrasensitive detection of antibodies using a new TusTer-lock immunoPCR system. Mol Biosys. 2010;6:1173–5. Morin I, Askin SP, Schaeffer PM. IgG-detection devices for the Tus-Ter-lock immunoPCR diagnostic platform. Analyst 2011;136:4815–21. Parthasarathy N, DeShazer D, England M, Waag DM. Polysaccharide microarray technology for the detection of Burkholderia pseudomallei and Burkholderia mallei antibodies. Diagn Microbiol Infect Dis 2006;56:329–32. Stone JK, Mayo M, Grasso SA, Ginther JL, Warrington SD, Allender CJ, et al. Detection of Burkholderia pseudomallei O-antigen serotypes in near-neighbor species. BMC Microbiol 2012;12:250. Thepthai C, Smithtikarn S, Suksuwan M, Songsivilai S, Dharakul T. Serodiagnosis of melioidosis by a competitive enzyme-linked immunosorbent assay using a lipopolysaccharide-specific monoclonal antibody. Asian Pac J Allergy Immunol 2005;23:127–32. Tuanyok A, Stone JK, Mayo M, Kaestli M, Gruendike J, Georgia S, et al. The genetic and molecular basis of O-antigenic diversity in Burkholderia pseudomallei lipopolysaccharide. PLoS Neg Trop Dis. 2012;6:e1453. Wiersinga WJ, Currie BJ, Peacock SJ. Melioidosis. New England J Med. 2012;367: 1035–44. White NJ. Melioidosis. Lancet 2003;361:1715–22. Wikraiphat C, Charoensap J, Utaisincharoen P, Wongratanacheewin S, Taweechaisupapong S, Woods DE, et al. Comparative in vivo and in vitro analyses of putative virulence factors of Burkholderia pseudomallei using lipopolysaccharide, capsule and flagellin mutants. FEMS Immunol Med Microbiol 2009;56:253–9. Wongratanacheewi S, Sermswan RW, Anuntagool N, Sirisinha S. Retrospective study on the diagnostic value of IgG ELISA, dot immunoassay and indirect hemagglutination in septicemic melioidosis. Asian Pac J Allergy Immunol 2001;19:129–33.