Advertisement
Research Article

A Protein-Conjugate Approach to Develop a Monoclonal Antibody-Based Antigen Detection Test for the Diagnosis of Human Brucellosis

  • Kailash P. Patra,

    Affiliation: Division of Infectious Diseases, Department of Medicine, University of California San Diego, La Jolla, California, United States of America

    X
  • Mayuko Saito,

    Affiliation: Division of Infectious Diseases, Department of Medicine, University of California San Diego, La Jolla, California, United States of America

    X
  • Vidya L. Atluri,

    Affiliation: Department of Medical Microbiology, University of California Davis, Davis, California, United States of America

    X
  • Hortensia G. Rolán,

    Affiliation: Department of Medical Microbiology, University of California Davis, Davis, California, United States of America

    X
  • Briana Young,

    Affiliation: Department of Medical Microbiology, University of California Davis, Davis, California, United States of America

    X
  • Tobias Kerrinnes,

    Affiliation: Department of Medical Microbiology, University of California Davis, Davis, California, United States of America

    X
  • Henk Smits,

    Affiliation: Department of Biomedical Research, Royal Tropical Institute, Amsterdam, the Netherlands

    X
  • Jessica N. Ricaldi,

    Affiliation: Institute of Tropical Medicine Alexander von Humboldt, Universidad Peruana Cayetano Heredia, Lima, Peru

    X
  • Eduardo Gotuzzo,

    Affiliation: Institute of Tropical Medicine Alexander von Humboldt, Universidad Peruana Cayetano Heredia, Lima, Peru

    X
  • Robert H. Gilman,

    Affiliations: Department of International Health, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, United States of America, Laboratory of Research and Development, and Department of Cellular and Molecular Sciences, Faculty of Sciences, Universidad Peruana Cayetano Heredia, Lima, Peru

    X
  • Renee M. Tsolis mail,

    rmtsolis@ucdavis.edu (RMT); jvinetz@ucsd.edu (JMV)

    Affiliation: Department of Medical Microbiology, University of California Davis, Davis, California, United States of America

    X
  • Joseph M. Vinetz mail

    rmtsolis@ucdavis.edu (RMT); jvinetz@ucsd.edu (JMV)

    Affiliations: Division of Infectious Diseases, Department of Medicine, University of California San Diego, La Jolla, California, United States of America, Institute of Tropical Medicine Alexander von Humboldt, Universidad Peruana Cayetano Heredia, Lima, Peru, Laboratory of Research and Development, and Department of Cellular and Molecular Sciences, Faculty of Sciences, Universidad Peruana Cayetano Heredia, Lima, Peru

    X
  • Published: June 05, 2014
  • DOI: 10.1371/journal.pntd.0002926

Abstract

Human brucellosis is most commonly diagnosed by serology based on agglutination of fixed Brucella abortus as antigen. Nucleic acid amplification techniques have not proven capable of reproducibly and sensitively demonstrating the presence of Brucella DNA in clinical specimens. We sought to optimize a monoclonal antibody-based assay to detect Brucella melitensis lipopolysaccharide in blood by conjugating B. melitensis LPS to keyhole limpet hemocyanin, an immunogenic protein carrier to maximize IgG affinity of monoclonal antibodies. A panel of specific of monoclonal antibodies was obtained that recognized both B. melitensis and B. abortus lipopolysaccharide epitopes. An antigen capture assay was developed that detected B. melitensis in the blood of experimentally infected mice and, in a pilot study, in naturally infected Peruvian subjects. As a proof of principle, a majority (7/10) of the patients with positive blood cultures had B. melitensis lipopolysaccharide detected in the initial blood specimen obtained. One of 10 patients with relapsed brucellosis and negative blood culture had a positive serum antigen test. No seronegative/blood culture negative patients had a positive serum antigen test. Analysis of the pair of monoclonal antibodies (2D1, 2E8) used in the capture ELISA for potential cross-reactivity in the detection of lipopolysaccharides of E. coli O157:H7 and Yersinia enterocolitica O9 showed specificity for Brucella lipopolysaccharide. This new approach to develop antigen-detection monoclonal antibodies against a T cell-independent polysaccharide antigen based on immunogenic protein conjugation may lead to the production of improved rapid point-of-care-deployable assays for the diagnosis of brucellosis and other infectious diseases.

Author Summary

Brucellosis is a OneHealth disease reflecting the risk for human infection by interaction with and relation to affected animal populations. The disease is often difficult to diagnose because of lack of precise or accessible diagnostic reagents, and because culture is complex, hazardous and relatively insensitive. Brucellosis disproportionately affects the poor and dispossessed with human and animal burdens of disease in the Middle East, North Africa, Mongolia and other regions that are simply unknown. The diagnosis of brucellosis most often rests on serological tests—antibody detection—based on agglutination of fixed Brucella abortus. We have developed the basis for developing a new test based on the detection of the B. melitensis lipopolysaccharide, which provides rapid and definitive identification of the presence of the organism in clinically obtainable body fluids. A new approach—protein conjugation to the lipopolysaccharide antigen—was taken to enhance the affinity of the monoclonal antibodies that were generated for the test. These reagents were tested in a mouse model of B. melitensis and in humans from the brucellosis-endemic region of Peru, and provided the data for the basis of further clinical development and clinical trials for the rapid, point-of-care diagnosis of brucellosis that will also provide new tools for assessing the global burden of disease.

Introduction

Human brucellosis is most commonly caused by two species of the genus Brucella, typically B. abortus from cattle and B. melitensis from goats and sheep. The definitive diagnosis of brucellosis rests upon demonstration of the causative bacterium in a suspected patient's body fluid, typically by culture isolation [1], [2]. While detection of Brucella nucleic acids [3][13] or antigens [14] would be expected to be diagnostic for new cases of brucellosis, DNA has been reported to persist in blood after successful treatment of solidly diagnosed cases [15], [16]. Therefore PCR amplification-based tests are not useful to confirm brucellosis relapse [15], [16]. Because culture is technically challenging and hazardous in many clinical laboratories, brucellosis is most commonly diagnosed using serological methods that use fixed, whole Brucella abortus as antigen [17][21]. Such methods include the Rose Bengal, slide agglutination, and tube agglutination tests, sometimes accompanied with the use of 2-mercaptoethanol to distinguish IgG from IgM antibodies when determining the presence of active infection requiring antibiotic therapy; newer data obtained using genome-level screens suggest the potential utility of recombinant B. melitensis proteins for characterization of human infection [22][24]. Sometimes, when prozone or other interfering immune phenomena occur where clinical brucellosis may be associated with non-agglutinating antibodies, the Coomb's indirect antibody test or the BrucellaCapt assay can detect anti-Brucella antibodies [19], [25][32]. ELISA to detect IgM or IgG antibodies that react with B. abortus lysates are not recommended for diagnosis because of limited specificity, but a competitive ELISA to detect smooth Brucella LPS [33] and a rapid antibody-detecting test such as the lipopolysaccharide (LPS)-based lateral flow assay has favorable performance characteristics [19], [25][32]. Nonetheless, ELISA tests based on whole cell B. abortus lysates may suffer from false positive results. False positive serological results may be also found with other pathogenic bacteria because of cross-reaction with E. coli O157:H7, Francisella tularensis, Yersinia enterocolica and Salmonella typhi (at low dilutions) can confound serological diagnosis but diseases caused by these agents are rarely confused with brucellosis [34][39]. Nonetheless, serological diagnosis provides only an indirect measure of infection.

The present investigation aimed to develop new monoclonal antibodies against the immunodominant LPS of B. melitensis, towards the development of new tools for the direct detection of Brucella LPS antigen for diagnostic purposes. We adopted a new approach to enhance the affinity of IgG antibodies for the LPS antigen by coupling purified B. melitensis LPS to keyhole limpet hemocyanin (KLH) prior to immunization and boosting, which would be predicted to induce T cell-dependent affinity maturation of the anti-LPS antibody response by B cells. Supernatants from hybridomas which screened positive for anti-B. melitensis LPS by ELISA were further characterized by Western blot and indirect immunofluorescence microscopy. A capture ELISA using the purified monoclonal antibodies was tested for its ability to detect B. melitensis LPS antigen in sera from experimentally infected mice and Peruvian patients diagnosed with brucellosis by blood culture.

Methods

Purification of Brucella melitensis lipopolysaccharide

Brucella melitensis 16M was grown under aeration to stationary phase at 37°C in tryptic soy broth (TSB). Cells were recovered by centrifugation and approximately 10 g of pelleted cells were inactivated by autoclaving (121°C, 40 min). Cell pellets were used to isolate lipopolysaccharide (LPS) using the hot phenol-water method [40]. All manipulations of live Brucella melitensis were performed under Biosafety Level 3 conditions approved by the Select Agent program carried out at the University of California, Davis under conditions established and supervised by the United States Department of Agriculture and the United States Centers for Disease Control and Prevention. The purified LPSs were treated with RNase, DNase and proteinase-K (Sigma Chemicals, St. Louis, MO), and the hot phenol-water treatment was repeated. B. melitensis LPS fractions obtained from both upper phenol saturated aqueous layer (aqueous phase) and lower water saturated phenol layer (phenol phase) were pooled. Purified LPS was analyzed on a 4–12% gradient Tris-Glycine sodium dodecyl sulfate (SDS) polyacrylamide gel (Invitrogen Corp., Carlsbad, CA) under reducing conditions. The presence of LPS in the gels was detected with a periodic acid silver stain [41] and protein with Coomassie blue stain (BioRad, Hercules, CA). B. melitensis LPS was quantified using a colorimetric assay to measure 2-keto-3-deoxyoctonate (KDO) concentration [42]. E. coli 055: B5 LPS (Sigma Chemicals) was used as standard. We also extracted LPS from other Brucella species (B. abortus, B. suis, B. canis, B. ovis), Yersinia enterocolitica 09 and B. melitensis manB mutant by following the earlier mentioned procedure.

Keyhole limpet hemocyanin (KLH) conjugation to B. melitensis LPS and monoclonal antibody development

Three mg of purified B. melitensis LPS was used for conjugation with KLH (SoluLink, Inc., San Diego, CA). Briefly, periodate-treated B. melitensis 16M LPS was linked to succinimidyl 4-hydrazinonicotinate (SANH)-modified KLH. To estimate conjugation efficiency, KLH-conjugated B. melitensis LPS was analyzed on 10% Bis-Tris SDS PAGE gels (Invitrogen Corp., Carlsbad, U.S.A) under reducing conditions and stained with periodic acid silver that detects LPS.

Monoclonal antibodies were raised against KLH-conjugated B. melitensis LPS using standard methods [43], [44]. The primary screen for hybridoma supernatants was an ELISA using wells coated with B. melitensis LPS. Positive hybridoma supernatants were isotyped and IgG isotypes chosen for further study (Monoclonal Antibody Isotyping Kit, Pierce, Rockford, IL).

LPS immunoblot analysis

To determine the reactivity of monoclonal antibodies to B. melitensis LPS, an immunoblotting analysis was performed as described previously [45]. B. melitensis LPS was electrophoresed on a 4–12% gradient Tris-Glycine SDS polyacrylamide gel and transferred to a nitrocellulose membrane using standard methods. The blotted membrane was blocked (Superblock, Thermo Fisher Scientific, Lake Barrington, IL) and strips were incubated individually in neat hybridoma supernatants for 2 h at room temperature. The strips were washed three times in TBS-Tween 20, 0.05% (TBST) and incubated with 1:3000 dilution (1% Superblock in TBST) of goat anti-mouse IgG + IgM phosphatase labeled antibodies (KPL, Maryland, U.S.A) for 1 h and washed four times in TBST. The blots were incubated in substrate solution (BCIP/NBT-1 Phosphatase Substrate, KPL) for the color development.

To evaluate the cross reactivity of monoclonals, LPS of other Brucella species (B. abortus 2308 and B. suis 1330) and mutant (B. melitensis manB), E. coli 0157:H7 and Y. enterocolitica were electrophoresed, transferred and probed with hybridoma supernatants as described above. The dilution of the hybridoma supernatant varies between from 1:50 to 1: 10,000) based on the signal intensity and it's mentioned in the blot for each monoclonal antibodies.

Specificity of monoclonal antibodies against B. melitensis LPS by indirect immunofluorescence microscopy

ELISA-positive hybridoma supernatants were further screened by immunofluorescence against acetone-fixed B. melitensis 16 M, E. coli 0157:H7 strain EDL933 and E. coli strain DH10B expressing Yersinia enterocolitica 09 LPS [46]. Hybridoma supernatants were diluted 2-fold from 1:50 to 1:64,000; 50 µl of diluted supernatant was added to the acetone-fixed specimen and incubated for 2 h in a humidified chamber at 22°C. After three washes, FITC-labeled rabbit anti-mouse IgG and IgM (Jackson ImmunoResearch Laboratories, Inc.) was added at a 1:100 dilution and incubated for 1 h at room temperature. After additional washes, the specimens were mounted with mounting medium (Invitrogen Corp., Carlsbad, CA), counterstained with DAPI (0.5 µg/ml; Molecular Probes, Inc.) and slides were examined using an Axiovert M200 microscope (Carl Zeiss, Germany).

ELISA with whole cell antigen of B. abortus and B. melitensis

To determine the relative reactivity of mAbs for B. abortus and B. melitensis, a native whole cell antigen ELISA was used. Whole cell antigen was prepared by growth of B. abortus 2308 and B. melitensis 16M in tryptic soy broth with aeration at 37C for 16 h. Bacteria were diluted to an OD600 of 0.75, and 0.1 ml of antigen was incubated in a U-bottom microtiter plate with serial dilutions of mAbs for 1.5 h with rocking at 37°C. Three washes were performed by centrifugation (1000×g, 15 min) and resuspension in 0.1 ml PBS. After the last wash, bacterial pellets were suspended in a 1:250 dilution of goat anti-mouse HRP conjugate and incubated 1.5 h with rocking at 37°C. After three washes, pellets were resuspended in HRP substrate and incubated at room temperature to allow for color development. After stopping the enzyme reaction, samples were filtered through 0.2 uM filters (MultiScreen HTS, Millipore) to eliminate bacteria and read at 450 nm. Data shown are from a single experiment that was replicated once with a similar result.

Reactivity of B. melitensis LPS-reactive monoclonal antibodies by ELISA

An ELISA method was used to compare binding efficiency of mAbs according to a published method [47]. LPS from E. coli 055:B5 (Sigma) and Leptospira licerasiae [48] were used as specificity controls. B. melitensis LPS was dissolved (2 µg/ml) in 0.06 M sodium carbonate buffer (pH 9.6); 100 µl was used to coat wells by incubation overnight at 4°C. For the assay, Superblock blocking buffer (Thermo Fisher Scientific, IL, and USA) was used for blocking, TBST was used for washing the plate and 1% Superblock blocking buffer in TBST was used to dilute antibodies and conjugates. LPS-coated plates were blocked (for 1 hr at 22°C followed by three washes. mAbs were diluted (2 µg–0.02/ml) and added 200 µl per well in duplicate and incubated overnight at 4°C. The wells were washed three times with TBST and 200 µl of diluted (1:5,000) goat anti-mouse immunoglobulin G (IgG) antibody conjugated to horseradish peroxidase was added to each well and incubated for 1 hr at room temperature. The plate was washed four times and 100 µl of chromogenic substrate (TMB Microwell Peroxidase Substrate system, KPL, Gaithersburg, MD) was added to each well and the reaction was stopped by the addition of 100 µl of 2N H2SO4. The plate was read using a microplate reader (Spectramax Plus, Molecular Devices, Sunnyvale, CA) at wavelength 450 nm.

B. melitensis LPS antigen capture ELISA development in sera from experimentally infected mice

A total of 15 C57BL/6/c mice (Jackson Laboratories, Bar Harbor, Maine) were inoculated intraperitoneally with 2.8×105 CFU of B. melitensis 16M. At weeks 1, 3 and 8 after infection groups of 5 mice were euthanized and blood collected at necropsy. Colonization of liver and spleen was determined by serial dilution of tissue homogenates and plating on tryptic soy agar (TSA). These experiments were carried out under ABSL-3 containment at UC Davis and approved by the UC Davis Institutional Animal Care and Use Committee. As negative control, sera from uninfected mice (N = 3) were obtained. Serum samples were diluted 1:5 in phosphate-buffered saline and were passed twice through 0.22 µM filters for biological safety reasons. All samples were streaked on TSA + blood plates to confirm a lack of viable bacteria before testing for antigen by ELISA.

B. melitensis LPS antigen capture ELISA testing on human sera

Two B. melitensis LPS immuno-reactive mAbs were used for assay development: 2D1 mAb as capture antibody (200 ng/well) and biotinylated 2E8 (0.5 mg/ml) for detection. Biotinylation of 2E8 mAb was carried using standard methods according to manufacturer's instructions (EZ-Link Biotinylation, Thermo Scientific, Ill, U.S.A). B. melitensis LPS antigen detection was carried out using: a) B. melitensis LPS spiked into the human brucellosis case and control sera, b) B. melitensis infected and control mice sera and c) Sera from patients with brucellosis confirmed by blood culture to have B. melitensis infection and sera from control patients from Peru who were referred to the study for suspected brucellosis but were found to be blood culture and serologically negative for brucellosis [23], [24]. The available serological standard tests (Rose Bengal test, lateral flow assay [29] were used to screen patients for brucellosis. Routine diagnostic tests for brucellosis confirmation included the standard tube agglutination assay and 2-mercaptoethanol test (to distinguish IgM and IgG agglutinating antibodies) and were also carried out on this set of patient samples, as previously described [23], [24], [32]. A commercially available ELISA test kit, based on a B. abortus whole cell lysate putatively used for brucellosis, was also used (Genway IgG Brucellosis ELISA, Catalog GWB-3C0D26, San Diego, CA).

Normal human sera (N = 10) purchased from a commercial blood bank in USA (Interstate Blood Bank, Memphis, TN) were spiked with measured quantities of B. melitensis LPS for testing in the capture assay; unspiked sera served as negative controls. Sera from human brucellosis cases were obtained at the time of blood culture positivity; 10 negative control sera from Peru (culture negative and Rose Bengal Test negative) were used.

The antigen-detection ELISA was developed by coating plates with capture mAb 2D1 (200 ng/well) in 0.06 M sodium bicarbonate buffer (pH 9.6) incubated overnight at 4°C in a humid chamber. Blocking and washing steps, antibody dilution buffer and color development were done as described above. Test serum samples were diluted (1:25 dilution) in 1% Superblock TBST and 200 µl of the diluted sample was added per well in duplicates and incubated over night at 4°C. For human serum samples from Peru, the dilution was done in TBST and heat inactivated (56°C for 15 minutes) before adding to the wells. The plates were washed after incubation in test serum samples. For detection, 200 µl of biotinylated mAb 2E8 (1:5,000 dilution, 0.5 mg/ml) was added per well and incubated for 2 hr. Plates were washed 4 times and 200 µl of 1:8,000 dilution of peroxidase-conjugated streptavidin (Thermo Scientific, Ill, USA) was added and incubated for one hr. The plates were washed and colorimetric assay developed as described above.

Ethics statement

Human subjects participating in this study and providing clinical samples (blood only) for routine and new experimental approaches to brucellosis diagnosis were part of a prospective study to develop brucellosis diagnostics based in Lima, Peru from 2005–2010. The enrollment of human subjects in this study was carried out under a protocol approved by the Human Subjects Protections Program, University of California San Diego and the Ethical Committee of Universidad Peruana Cayetano Heredia, Lima, Peru. The use of the samples for the purposes described in the present experiments was approved by both Institutional Review Boards. All samples were anonymized before sending for analysis. All adults provided written informed consent to participate in this study; all minors involved in the study provided verbal assent and written informed consent was provided by each minor's parent/guardian.

Data analysis

Data analysis and graphs were done with Prism 5.0 software (GraphPad Prism, San Diego, CA). For statistical significance between positive and control groups, Student's t test was performed with p<0.05 considered statistically significant. ELISA results were considered positive when the O.D. was ≥3.0 standard deviations above the mean of the negative control (in triplicate).

Results

Production and selection of hybridomas producing anti-B. melitensis monoclonal antibodies

Mice vaccinated with purified, protein-free B. melitensis LPS (Fig. 1A) conjugated to keyhole limpet hemocyanin (KLH) (Fig. 1B) were sacrificed, splenocytes fused to non-secreting myeloma cells and hybridomas screened by ELISA for reactivity against B. melitensis LPS. Eight hybridomas screening positive by ELISA were confirmed to recognized B. melitensis LPS by Western immunoblot (Fig. 1C).

thumbnail

Figure 1. Sodium dodecyl sulfate polyacrylamide gel electrophoresis and western immunoblot analysis of purified and conjugated Brucella melitensis lipopolysaccharide.

A. B. melitensis LPS was separated on a 4–12% Tris Glycine SDS-polyacrylamide gradient gel. The gel was stained for protein contamination of the lipopolysaccharide preparation with Coomassie Blue and periodic acid silver stain for LPS detection. Molecular mass (kDa) indicated at right of the panel. B. Keyhole Limpet hemocyanin (KLH) conjugated to B. melitensis LPS. B. melitenis LPS was oxidized with sodium periodate and linked to succinimidyl 4-hydrazinonicotinate (SANH)-modified KLH. Escherichia coli O55:B5 LPS (control), B. melitensis LPS alone and KLH-conjugated- KLH-conjugated B. melitensis LPS. The samples were electrophoresed on a 10% Bis-Tris polyacrylamide gel and stained with silver stain for LPS detection. Molecular mass (kDa) indicated at left of panel. C. Western immunoblot of anti-LPS monoclonal antibodies. Purified B. melitensis LPS was subjected to SDS-PAGE, transferred to nitrocellulose, and strips were probed individually with hybridoma supernatants (mAbs: 5D1, 2F1, 5H1, 2D2, 2D1 and 5E1).

doi:10.1371/journal.pntd.0002926.g001

Specificity of anti-B. melitensis 16M LPS monoclonal antibodies using western immunoblot, ELISA and indirect immunofluorescence assay

The reactivity of six monoclonal antibodies was determined against LPS antigens from B. melitensis strain 16M (genome strain), a mutant LPS-deficient B. melitensis 16M manB mutant [49], B. abortus, B. suis, B. canis, B. ovis, E. coli O157:H7, and Yersinia enterocolitica. LPS extracts were electrophoresed on a 10% Bis-Tris polyacrylamide gel and stained with silver stain for LPS detection (LPS)(Fig. 2, upper left panel). Other gels were transferred to nitrocellulose, and strips were probed individually with dilutions of hybridoma supernatants as indicated: 2D1 (1:1,000), 2E8 (1:10,000), 2F1 (1:100), 5D1 (1:500) and 5H1 (1:5,000). The mAbs reacted strongly with B. melitensis 16M (but not the LPS-deficient manB mutant), and with B. abortus, B. suis, and, faintly, with B. ovis. No cross reactivity was observed by Western immunoblot with E. coli O157:H7 or Y. enterocolitica.

thumbnail

Figure 2. LPS immunoblot probed with monoclonal antibody-containing supernatants to determine cross reactivity to the main pathogenic species of Brucella and potentially cross-reactive organisms.

Upper left, silver stain for LPS. Remaining panels reacted with mAbs as indicated.

doi:10.1371/journal.pntd.0002926.g002

To further investigate potential cross-reactivity using a different immunochemical technique, soluble antigens from B. melitensis 16M and B. abortus 2308 using an ELISA (Fig. 3A and B). Five of the 6 demonstrated reactivity against both Brucella spp.; one, 5D11 produced a detectable reaction with B. melitensis 16M but not with B. abortus 2308 (bottom right panel, Fig. 3). To characterize further the binding specificity of these mAbs using native (not heated, not chemically extracted) LPS, indirect immunofluorescence microscopy (IFA) was carried out on B. melitensis and B. abortus fixed directly onto slides (Table 1); in addition, E. coli O157:H7 strain EDL933 and E. coli strain DH108 expressing the Yersinia enterocolitica O9 antigen previously reported to cross-react with Brucella LPS [37], [38] were used as antigen for IFA (Table 1). Based on shared patterns of reactivity for both B. melitensis 16M and B. abortus 2308, mAbs 2D1 and 2E8 were chosen for further development of a Brucella LPS antigen capture ELISA.

thumbnail

Figure 3. ELISA-determined reactivity of a panel of monoclonal antibodies raised against KLH-conjugated B. melitensis 16M lipopolysaccharide against B. melitensis 16M and B. abortus 2308 native whole cell antigen.

doi:10.1371/journal.pntd.0002926.g003
thumbnail

Table 1. Specificity of anti-Brucella melitensis 16M lipopolysaccharide monoclonal antibodies by indirect immunofluorescence assay.

doi:10.1371/journal.pntd.0002926.t001

Development of an antigen capture ELISA using anti-Brucella melitensis 16M LPS monoclonal antibodies

MAbs 2D1 (capture antibody) and 2E8 (biotinylated for detection) were tested in a checkerboard assay for optimum concentrations for the development of a capture ELISA to detect B. melitensis 16M LPS (data not shown). A capture ELISA using these mAbs specifically detected B. melitensis 16M LPS spiked into normal human serum (Fig. 4), not producing a signal with E. coli or Leptospira licerasiae LPS.

thumbnail

Figure 4. Sensitivity of B. melitensis LPS antigen detection by capture ELISA.

Monoclonal antibody 2D1 to capture antigen and biotinylated mAb 2E8 used to detect LPS in the human serum samples that were spiked with known amount of B. melitensis LPS. The average absorption value obtained from the duplicate wells were plotted in the Y-axis, and the mean value along with standard deviation in for each groups were represented as bar. Positive result defined as OD450 ≥3.0 standard deviations from mean value of negative control group (0 ng LPS). Capture ELISA using Escherichia coli 055:B5 LPS (10 ng) and Leptospira licerasiae LPS (10 ng) did not yield a signal above background and not shown in the figure.

doi:10.1371/journal.pntd.0002926.g004

Determination of B. melitensis LPS antigenemia in experimentally infected mice

To determine whether the capture assay would be able to detect B. melitensis LPS in serum of experimentally infected mice, C57BL/6 mice were injected with 106 CFU of a low passage strain of B. melitensis 16M, and serum, liver and spleens obtained at 1,3 and 8 weeks after inoculation. B. melitensis LPS was detected at a significantly higher level in serum at 1 and 3 weeks post-infection but was not detectable at 8 weeks, despite persisting viable B. melitensis observable in liver and spleen as determined by CFU counts (Fig.5, bottom).

thumbnail

Figure 5. Detection of B. melitensis LPS in serum of experimentally infected mice.

A. BALB/c mice were infected with B. melitensis 16M. Sera were tested for the presence of LPS at weeks 1, 3 and 8 using the 2D1-2E8 capture ELISA. B. melitensis LPS was detected above background at weeks 1 (p≤0.003) and 3 (p≤0.004) but not at week 8. B. Colony forming units (CFU) in liver and spleen at each time point are indicated.

doi:10.1371/journal.pntd.0002926.g005

Determination of B. melitensis LPS antigenemia in Peruvian brucellosis patients

Ultimately, the capture ELISA to detect B. melitensis LPS was developed with the goal of having a culture-independent method of rapidly diagnosing brucellosis. In a pilot study of 30 patients referred for suspected brucellosis, we studied three groups of 10 subjects each (Table 2): 1) with definitive brucellosis (blood culture positive); 2) Rose Bengal screen positive, and various titers of standard tube agglutination positive patients but negative blood cultures; and 3) serologically and blood culture negative. A majority (7/10) of the patients with positive blood cultures had B. melitensis LPS detected in the initial blood specimen obtained (Table 2, Figure 6). One of ten patients (with relapsed brucellosis) with a negative blood culture had a positive serum antigen test. No seronegative and blood culture negative patient had a positive serum antigen test.

thumbnail

Figure 6. Detection of B. melitensis LPS in human brucellosis patients from Peru.

Sera from patients presenting with a suspicion of brucellosis were either confirmed to have B. melitensis brucellosis by blood culture and serological positive (C+RB+); serologically positive but blood culture negative (C- RB+) and both serologically and blood culture negative samples (C- and RB-). Three negative control serum obtained from US blood bank (-ve USA) were also included in the assay. The average absorption value of the samples obtained by capture ELISA were plotted, and the mean value in each group are shown as bar. The signal was significantly higher in the blood culture positive than in the blood culture negative/serologically negative group (p≤0.0004) and blood culture negative/seropositive group (p≤0.001). The lower table shows the comparative result by serological tests (Rose Bengal test and Brucella IgG ELISA Kit, GenWay Biotech Inc, San Diego, CA) and the capture ELISA, which shows 70% (7/10) of the culture positive samples were detected with higher LPS antigenemia (OD450≥3.0 standard deviations from the mean value of six Peruvian CS samples).

doi:10.1371/journal.pntd.0002926.g006
thumbnail

Table 2. Clinical features and diagnostic results of Peruvian patient groups.

doi:10.1371/journal.pntd.0002926.t002

Discussion

Here we report the development of a monoclonal antibody-based Brucella melitensis antigen detection assay based on new monoclonal antibodies developed using a novel LPS-protein conjugation approach. Here we made B. melitensis LPS into a T-dependent antigen by chemical conjugation to a carrier protein, keyhole limpet hemocyanin (KLH) to enhance immunogenicity by inducing helper T cell function. Such an approach has been demonstrated to increase titers and affinities of antibody responses against T cell-independent antigens such as bacterial capsular polysaccharides [50][55] and the potential for enhancing affinity maturation of poorly antigenic proteins [56], [57] as well as T-independent antigens [58] such as LPS. The immunization strategy used here yielded a panel of monoclonal antibodies with varying properties. Hybridomas producing mAbs that recognized both B. melitensis and B. abortus LPS and one that produced a mAb that selectively recognized B. melitensis were obtained. An antigen capture assay was developed that was able to detect B. melitensis LPS in the serum of experimentally-infected mice and blood culture-positive brucellosis cases in Peru.

The present work is consistent with previous reports of mAb development against Brucella LPS in detecting both shared and unique epitopes (O-antigen, A- and M- epitopes) within Brucella spp. LPS [14], [59][63]. Using the new panel of mAbs, we found that in the mouse model of brucellosis that LPS antigen was detectable at 1 and 3 weeks after infection. However, while colony counts of B. melitensis remained robust in liver and spleen eight weeks after infection, LPS antigen was no longer detectable. This discrepancy may relate to the human data reported here in which we found only a moderate sensitivity (70%) of LPS antigen detection at clinical presentation and poor sensitivity (10%) when in seropositive, blood culture negative patients. This observation is consistent with the known imperfect sensitivity of blood culture for the diagnosis of brucellosis, and suggests the compartmentalization of B. melitensis during human infection into sites where neither bacteria nor LPS are released into blood after acute infection. We do not believe that the performance of the capture ELISA reported here is due to lack of performance of the mAbs for two reasons: 1) the mAbs used were determined to detect both B. melitensis and B. abortus LPS; and 2) human brucellosis in Peru is caused exclusively by B. melitensis, so that non-B. melitensis brucellosis cases are unlikely to have occurred in our patient population. It is possible that the intracellular location of Brucella may reduce the sensitivity of LPS antigen detection. Regarding the specificity of the mAb detection ELISA described here, the pair of mAbs—2D1 and 2E8—were chosen specifically not to detect either E. coli O157:H7 or Yersinia enterocolitica, which was confirmed by Western immunoblot. Regardless, these infections present quite differently than brucellosis and would not be confused by an experienced clinician; hence antigen detection assays for brucellosis would not likely yield false positive results in these disease. Indeed, it may be possible to use the relevant mAbs (Table 1) to detect infections by either of these organisms in the appropriate clinical context.

The antigen-detection mAbs described here also have a One Health application, and could readily be adapted for screening herds of livestock for active infection that present either animal or human health risk. The ability of this panel of mAbs to identify B. suis antigen was not assessed here; it would be unlikely that B. canis infections would be identified given the lack of LPS in this species. The assay here only was tested with B. melitensis infection, and further work needs to be done to assess which of the mAbs would be most effective to identify B. abortus infection.

Here we did not assess the ability of the mAb capture assay to detect persistent antigenemia to suggest persistent infection, which has been suggested by persistent DNAemia after B. melitensis infection [15]. We believe, however, that our data are consistent with the detection of active bacteremia and disease, which persistent DNAemia does not necessarily indicate. Chronic and occult brucellosis is difficult to diagnosis, likely due to low organism burden, hence we would not expect the mAb capture ELISA described here to be sensitive in diagnosing this manifestation of brucellosis. Nonetheless, the utility of antigen detection to diagnosis manifestations of brucellosis such as focal or chronic forms remains to be determined. Finally, antigen detection in specimens other than serum needs to be further assessed in other endemic settings and other clinical specimens such whole blot, blood clot, bone marrow, urine, abscess aspirates and cerebrospinal fluid.

Acknowledgments

We thank Paula Maguina, Staff Research Associate at UC San Diego for her research contributions that were essential to assure regulatory compliance, ethics and logistical aspects of this project.

Author Contributions

Conceived and designed the experiments: KPP MS VLA HGR HS JNR EG RHG RMT JMV. Performed the experiments: KPP VLA HGR BY TK JNR. Analyzed the data: KPP MS HS JNR EG RHG JMV. Contributed reagents/materials/analysis tools: KPP VLA RMT HS. Wrote the paper: KPP MS VLA JNR EG RHG RMT JMV.

References

  1. 1. Ruiz Castaneda M (1961) Laboratory diagnosis of brucellosis in man. Bull World Health Organ 24: 73–84.
  2. 2. Pappas G, Akritidis N, Bosilkovski M, Tsianos E (2005) Brucellosis. N Engl J Med 352: 2325–2336. doi: 10.1056/nejmra050570
  3. 3. Baddour MM, Alkhalifa DH (2008) Evaluation of three polymerase chain reaction techniques for detection of Brucella DNA in peripheral human blood. Can J Microbiol 54: 352–357. doi: 10.1139/w08-017
  4. 4. Bricker BJ (2002) PCR as a diagnostic tool for brucellosis. Vet Microbiol 90: 435–446. doi: 10.1016/s0378-1135(02)00228-6
  5. 5. Castano MJ, Solera J (2009) Chronic brucellosis and persistence of Brucella melitensis DNA. J Clin Microbiol 47: 2084–2089. doi: 10.1128/jcm.02159-08
  6. 6. Debeaumont C, Falconnet PA, Maurin M (2005) Real-time PCR for detection of Brucella spp. DNA in human serum samples. Eur J Clin Microbiol Infect Dis 24: 842–845. doi: 10.1007/s10096-005-0064-0
  7. 7. Ilhan Z, Aksakal A, Ekin IH, Gulhan T, Solmaz H, et al. (2008) Comparison of culture and PCR for the detection of Brucella melitensis in blood and lymphoid tissues of serologically positive and negative slaughtered sheep. Lett Appl Microbiol 46: 301–306. doi: 10.1111/j.1472-765x.2007.02309.x
  8. 8. Imaoka K, Kimura M, Suzuki M, Kamiyama T, Yamada A (2007) Simultaneous detection of the genus Brucella by combinatorial PCR. Jpn J Infect Dis 60: 137–139.
  9. 9. Leal-Klevezas DS, Lopez-Merino A, Martinez-Soriano JP (1995) Molecular detection of Brucella spp.: rapid identification of B. abortus biovar I using PCR. Arch Med Res 26: 263–267.
  10. 10. Mitka S, Anetakis C, Souliou E, Diza E, Kansouzidou A (2007) Evaluation of different PCR assays for early detection of acute and relapsing brucellosis in humans in comparison with conventional methods. J Clin Microbiol 45: 1211–1218. doi: 10.1128/jcm.00010-06
  11. 11. Navarro E, Casao MA, Solera J (2004) Diagnosis of human brucellosis using PCR. Expert Rev Mol Diagn 4: 115–123. doi: 10.1586/14737159.4.1.115
  12. 12. Navarro E, Escribano J, Fernandez J, Solera J (2002) Comparison of three different PCR methods for detection of Brucella spp in human blood samples. FEMS Immunol Med Microbiol 34: 147–151. doi: 10.1111/j.1574-695x.2002.tb00616.x
  13. 13. Navarro E, Segura JC, Castano MJ, Solera J (2006) Use of real-time quantitative polymerase chain reaction to monitor the evolution of Brucella melitensis DNA load during therapy and post-therapy follow-up in patients with brucellosis. Clin Infect Dis 42: 1266–1273. doi: 10.1086/503035
  14. 14. Al-Shamahy HA, Wright SG (1998) Enzyme-linked immunosorbent assay for brucella antigen detection in human sera. J Med Microbiol 47: 169–172. doi: 10.1099/00222615-47-2-169
  15. 15. Vrioni G, Pappas G, Priavali E, Gartzonika C, Levidiotou S (2008) An eternal microbe: Brucella DNA load persists for years after clinical cure. Clin Infect Dis 46: e131–136. doi: 10.1086/588482
  16. 16. Maas KS, Mendez M, Zavaleta M, Manrique J, Franco MP, et al. (2007) Evaluation of brucellosis by PCR and persistence after treatment in patients returning to the hospital for follow-up. Am J Trop Med Hyg 76: 698–702.
  17. 17. Al Dahouk S, Tomaso H, Nockler K, Neubauer H, Frangoulidis D (2003) Laboratory-based diagnosis of brucellosis—a review of the literature. Part II: serological tests for brucellosis. Clin Lab 49: 577–589.
  18. 18. Diaz R, Casanova A, Ariza J, Moriyon I (2011) The Rose Bengal Test in human brucellosis: a neglected test for the diagnosis of a neglected disease. PLoS Negl Trop Dis 5: e950. doi: 10.1371/journal.pntd.0000950
  19. 19. Gomez MC, Nieto JA, Rosa C, Geijo P, Escribano MA, et al. (2008) Evaluation of seven tests for diagnosis of human brucellosis in an area where the disease is endemic. Clin Vaccine Immunol 15: 1031–1033. doi: 10.1128/cvi.00424-07
  20. 20. Ruiz-Mesa JD, Sanchez-Gonzalez J, Reguera JM, Martin L, Lopez-Palmero S, et al. (2005) Rose Bengal test: diagnostic yield and use for the rapid diagnosis of human brucellosis in emergency departments in endemic areas. Clin Microbiol Infect 11: 221–225. doi: 10.1111/j.1469-0691.2004.01063.x
  21. 21. Dabdoob WA, Abdulla ZA (2000) A panel of eight tests in the serodiagnosis and immunological evaluation of acute brucellosis. East Mediterr Health J 6: 304–312.
  22. 22. Cannella AP, Lin JC, Liang L, Atluri V, Gotuzzo E, et al. (2012) Serial kinetics of the antibody response against the complete Brucella melitensis ORFeome in focal vertebral brucellosis. J Clin Microbiol 50: 922–926. doi: 10.1128/jcm.05298-11
  23. 23. Liang L, Leng D, Burk C, Nakajima-Sasaki R, Kayala MA, et al. (2010) Large scale immune profiling of infected humans and goats reveals differential recognition of Brucella melitensis antigens. PLoS Negl Trop Dis 4: e673. doi: 10.1371/journal.pntd.0000673
  24. 24. Liang L, Tan X, Juarez S, Villaverde H, Pablo J, et al. (2011) Systems biology approach predicts antibody signature associated with Brucella melitensis infection in humans. J Proteome Res 10: 4813–4824. doi: 10.1021/pr200619r
  25. 25. Bosilkovski M, Katerina S, Zaklina S, Ivan V (2009) The role of Brucellacapt test for follow-up patients with brucellosis. Comp Immunol Microbiol Infect Dis 33: 435–42. doi: 10.1016/j.cimid.2009.06.001
  26. 26. Mantecon MA, Gutierrez P, del Pilar Zarzosa M, Duenas AI, Solera J, et al. (2006) Utility of an immunocapture-agglutination test and an enzyme-linked immunosorbent assay test against cytosolic proteins from Brucella melitensis B115 in the diagnosis and follow-up of human acute brucellosis. Diagn Microbiol Infect Dis 55: 27–35. doi: 10.1016/j.diagmicrobio.2005.11.003
  27. 27. Casao MA, Navarro E, Solera J (2004) Evaluation of Brucellacapt for the diagnosis of human brucellosis. J Infect 49: 102–108. doi: 10.1016/j.jinf.2004.03.004
  28. 28. Orduna A, Almaraz A, Prado A, Gutierrez MP, Garcia-Pascual A, et al. (2000) Evaluation of an immunocapture-agglutination test (Brucellacapt) for serodiagnosis of human brucellosis. J Clin Microbiol 38: 4000–4005.
  29. 29. Smits HL, Abdoel TH, Solera J, Clavijo E, Diaz R (2003) Immunochromatographic Brucella-specific immunoglobulin M and G lateral flow assays for rapid serodiagnosis of human brucellosis. Clin Diagn Lab Immunol 10: 1141–1146. doi: 10.1128/cdli.10.6.1141-1146.2003
  30. 30. Irmak H, Buzgan T, Evirgen O, Akdeniz H, Demiroz AP, et al. (2004) Use of the Brucella IgM and IgG flow assays in the serodiagnosis of human brucellosis in an area endemic for brucellosis. Am J Trop Med Hyg 70: 688–694.
  31. 31. Kose S, Smits HL, Abdoel TH, Ozbel Y (2006) Prevalence of Brucella antibodies in rural and suburban communities in three provinces of Turkey: need for improved diagnosis and prevention. J Infect 53: 308–314. doi: 10.1016/j.jinf.2005.12.022
  32. 32. Franco MP, Mulder M, Gilman RH, Smits HL (2007) Human brucellosis. Lancet Infect Dis 7: 775–786. doi: 10.1016/s1473-3099(07)70286-4
  33. 33. Lucero NE, Foglia L, Ayala SM, Gall D, Nielsen K (1999) Competitive enzyme immunoassay for diagnosis of human brucellosis. J Clin Microbiol 37: 3245–3248.
  34. 34. Shemesh AA, Yagupsky P (2011) Limitations of the standard agglutination test for detecting patients with Brucella melitensis bacteremia. Vector Borne Zoonotic Dis 11: 1599–1601. doi: 10.1089/vbz.2011.0704
  35. 35. Somily AM, Adam MH, Gad El Rab MO, Morshed MG, Shakoor Z (2011) Detection of Salmonella typhi agglutinins in sera of patients with other febrile illnesses and healthy individuals. Ann Afr Med 10: 41–44. doi: 10.4103/1596-3519.76584
  36. 36. Gutierrez MP, Bratos MA, Garrote JI, Duenas A, Almaraz A, et al. (2003) Serologic evidence of human infection by Francisella tularensis in the population of Castilla y Leon (Spain) prior to 1997. FEMS Immunol Med Microbiol 35: 165–169. doi: 10.1016/s0928-8244(03)00002-6
  37. 37. Chart H, Okubadejo OA, Rowe B (1992) The serological relationship between Escherichia coli O157 and Yersinia enterocolitica O9 using sera from patients with brucellosis. Epidemiol Infect 108: 77–85. doi: 10.1017/s0950268800049529
  38. 38. Corbel MJ, Stuart FA, Brewer RA (1984) Observations on serological cross-reactions between smooth Brucella species and organisms of other genera. Dev Biol Stand 56: 341–348.
  39. 39. Westerman RB, He Y, Keen JE, Littledike ET, Kwang J (1997) Production and characterization of monoclonal antibodies specific for the lipopolysaccharide of Escherichia coli O157. J Clin Microbiol 35: 679–684.
  40. 40. Baker PJ, Wilson JB (1965) Hypoferremia in mice and its application to the bioassay of endotoxin. J Bacteriol 90: 903–910.
  41. 41. Kittelberger R, Hilbink F (1993) Sensitive silver-staining detection of bacterial lipopolysaccharides in polyacrylamide gels. J Biochem Biophys Methods 26: 81–86. doi: 10.1016/0165-022x(93)90024-i
  42. 42. Karkhanis YD, Zeltner JY, Jackson JJ, Carlo DJ (1978) A new and improved microassay to determine 2-keto-3-deoxyoctonate in lipopolysaccharide of Gram-negative bacteria. Anal Biochem 85: 595–601. doi: 10.1016/0003-2697(78)90260-9
  43. 43. de StGroth SF, Scheidegger D (1980) Production of monoclonal antibodies: strategy and tactics. J Immunol Methods 35: 1–21. doi: 10.1016/0022-1759(80)90146-5
  44. 44. Harlow E, Lane D (1988) Antibodies: a laboratory manual. Cold Spring Harbor, N.Y.: Cold Spring Harbor Laboratory.
  45. 45. Towbin H, Staehelin T, Gordon J (1992) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. 1979. Biotechnology 24: 145–149. doi: 10.1073/pnas.76.9.4350
  46. 46. Skurnik M, Biedzka-Sarek M, Lubeck PS, Blom T, Bengoechea JA, et al. (2007) Characterization and biological role of the O-polysaccharide gene cluster of Yersinia enterocolitica serotype O:9. J Bacteriol 189: 7244–7253. doi: 10.1128/jb.00605-07
  47. 47. Alonso-Urmeneta B, Marin C, Aragon V, Blasco JM, Diaz R, et al. (1998) Evaluation of lipopolysaccharides and polysaccharides of different epitopic structures in the indirect enzyme-linked immunosorbent assay for diagnosis of brucellosis in small ruminants and cattle. Clin Diagn Lab Immunol 5: 749–754.
  48. 48. Matthias MA, Ricaldi JN, Cespedes M, Diaz MM, Galloway RL, et al. (2008) Human leptospirosis caused by a new, antigenically unique Leptospira associated with a Rattus species reservoir in the Peruvian Amazon. PLoS Negl Trop Dis 2: e213. doi: 10.1371/journal.pntd.0000213
  49. 49. Kahl-McDonagh MM, Ficht TA (2006) Evaluation of protection afforded by Brucella abortus and Brucella melitensis unmarked deletion mutants exhibiting different rates of clearance in BALB/c mice. Infect Immun 74: 4048–4057. doi: 10.1128/iai.01787-05
  50. 50. Lee CJ, Banks SD, Li JP (1991) Virulence, immunity, and vaccine related to Streptococcus pneumoniae. Crit Rev Microbiol 18: 89–114. doi: 10.3109/10408419109113510
  51. 51. Mulholland K (1999) Strategies for the control of pneumococcal diseases. Vaccine 17 Suppl 1S79–84. doi: 10.1016/s0264-410x(99)00112-7
  52. 52. Robbins JB, Schneerson R (1987) Haemophilus influenzae type b: the search for a vaccine. Pediatr Infect Dis J 6: 791–794. doi: 10.1097/00006454-198708000-00039
  53. 53. Goldblatt D (1998) Immunisation and the maturation of infant immune responses. Dev Biol Stand 95: 125–132.
  54. 54. Barbour ML (1996) Conjugate vaccines and the carriage of Haemophilus influenzae type b. Emerg Infect Dis 2: 176–182. doi: 10.3201/eid0203.960303
  55. 55. Dull PM, McIntosh ED (2012) Meningococcal vaccine development—from glycoconjugates against MenACWY to proteins against MenB—potential for broad protection against meningococcal disease. Vaccine 30 Suppl 2B18–25. doi: 10.1016/j.vaccine.2012.01.062
  56. 56. Kubler-Kielb J, Majadly F, Biesova Z, Mocca CP, Guo C, et al. (2010) A bicomponent Plasmodium falciparum investigational vaccine composed of protein-peptide conjugates. Proc Natl Acad Sci U S A 107: 1172–1177. doi: 10.1073/pnas.0913374107
  57. 57. Vincent AA, Fanning S, Caira FC, Williamson KC (1999) Immunogenicity of malaria transmission-blocking vaccine candidate, y230.CA14 following crosslinking in the presence of tetanus toxoid. Parasite Immunol 21: 573–581. doi: 10.1046/j.1365-3024.1999.00255.x
  58. 58. Roulland S, Suarez F, Hermine O, Nadel B (2008) Pathophysiological aspects of memory B-cell development. Trends Immunol 29: 25–33. doi: 10.1016/j.it.2007.10.005
  59. 59. Vizcaino N, Chordi A, Fernandez-Lago L (1991) Characterization of smooth Brucella lipopolysaccharides and polysaccharides by monoclonal antibodies. Res Microbiol 142: 971–978. doi: 10.1016/0923-2508(91)90007-w
  60. 60. Laurent TC, Mertens P, Dierick JF, Letesson JJ, Lambert C, et al. (2004) Functional, molecular and structural characterisation of five anti-Brucella LPS mAb. Mol Immunol 40: 1237–1247. doi: 10.1016/j.molimm.2003.11.037
  61. 61. Cloeckaert A, Weynants V, Godfroid J, Verger JM, Grayon M, et al. (1998) O-Polysaccharide epitopic heterogeneity at the surface of Brucella spp. studied by enzyme-linked immunosorbent assay and flow cytometry. Clin Diagn Lab Immunol 5: 862–870.
  62. 62. Kittelberger R, Bundesen PG, Cloeckaert A, Greiser-Wilke I, Letesson JJ (1998) Serological cross-reactivity between Brucella abortus and yersinia enterocolitica 0:9: IV. Evaluation of the M- and C-epitope antibody response for the specific detection of B. abortus infections. Vet Microbiol 60: 45–57. doi: 10.1016/s0378-1135(97)00202-2
  63. 63. Weynants V, Gilson D, Cloeckaert A, Tibor A, Denoel PA, et al. (1997) Characterization of smooth lipopolysaccharides and O polysaccharides of Brucella species by competition binding assays with monoclonal antibodies. Infect Immun 65: 1939–1943.