Skip to main content
NCBI home page
Clinical and Diagnostic Laboratory Immunology logoLink to Clinical and Diagnostic Laboratory Immunology
. 1999 Nov;6(6):838–843. doi: 10.1128/cdli.6.6.838-843.1999

Comparative Analysis of Two Meningococcal Immunotyping Monoclonal Antibodies by Resonant Mirror Biosensor and Antibody Gene Sequencing

Bambos M Charalambous 1, Janet Evans 2, Ian M Feavers 2, Martin C J Maiden 3,*
PMCID: PMC95785  PMID: 10548573

Abstract

Lipooligosaccharide (LOS) is a major surface component of the cell walls of Neisseria meningitidis, which is important for its roles in pathogenesis and antigenic variation, as a target for immunological typing, and as a possible vaccine component. Although the structures of many antigenic variants have been determined, routine immunological typing of these molecules remains problematic. Resonant mirror analysis was combined with gene sequencing to characterize two monoclonal antibodies (MAbs) used in typing panels that were raised against the same LOS immunotype, L3,7,9. The two MAbs (MAb 4A8-B2 and MAb 9-2-L379) were of the same immunoglobulin subtype, but while MAb 9-2-L379 was more than a 1,000-fold more sensitive in immunotyping assays of both whole meningococcal cells and purified LOS, MAb 4A8-B2 was more specific for immunotype L3,7,9. The differences in sensitivity were a consequence of MAb 9-2-L379 having a 44-fold-faster association constant than MAb 4A8-B2. Comparison of the amino acid sequences of the variable chains of the MAbs revealed that they had very similar heavy chains (81% amino acid sequence identity) but diverse light chains (54% sequence identity). The differential binding kinetics and specificities observed with these MAbs were probably due to differences in the epitopes recognized, and these were probably a consequence of the different immunization protocols used in their production.

Neisseria meningitidis, an etiological agent of meningitis and septicemia, is a normally commensal bacterium that nevertheless causes significant morbidity and mortality worldwide (3). Lipooligosaccharide (LOS) is an essential glycolipid component of the meningococcal outer membrane that is equivalent to the longer chained LOSs of the enteric bacteria, which is important in strain identification, vaccine development, pathogenesis and host damage (17). Twelve LOS immunotypes have been described in the literature (20); however, the L3, L7, and L9 immunotypes have an identical carbohydrate structure and have therefore been designated L3,7,9 (8). Specific and cross-reactive epitopes are located on the oligosaccharide part of the LOS molecule. Immunotypes L1 to L9 are associated primarily, but not exclusively, with serogroup B and C meningococci, while immunotypes L10 to L12 are mainly associated with serogroup A isolates (18). Immunotype L3,7,9 is often found in strains thought to be particularly virulent (13, 18) and may contribute to the resistance of these meningococci to complement-mediated lysis (14, 16). This is perhaps due to the fact that the oligosaccharide structure invariably terminates in a moiety that is structurally similar to the terminal sequence of human glycosphingolipids (17). Meningococcal LOS of immunotypes L3,7,9, L2, and L5, in common with that of the related gonococcus, can be further modified in vivo by sialylation or by the addition of cytidine 5′-monophosphate-N-acetylneuraminic acid (7, 15, 17).

Although the complete oligosaccharide structures of LOS molecules corresponding to most immunotypes have been elucidated, making it possible to correlate the immunotype-specific epitopes with defined oligosaccharide structures (20), immunological characterization of the variants for both routine epidemiological and research purposes remains problematic, requiring a relatively complex algorithm based on the reactivity of meningococcal whole cells or purified LOS in enzyme-linked immunosorbent assays (ELISAs) with a panel of monoclonal antibodies (MAbs) (18). For protein antigens it is frequently possible to correlate amino acid sequences of antigenically variable proteins, deduced from gene sequences, with immunological reactivity, and genetic techniques are consequently playing an increasing role in the characterization and study of such molecules. In the case of carbohydrates, including meningococcal LOS, such techniques are unlikely to provide a viable alternative to immunological studies, despite advances in understanding of the biosynthetic genes responsible for their production (10). Consequently, an improved understanding of LOS-antibody interactions is necessary for epidemiological surveillance and studies of the vaccine potential of this antigen.

In the present study, two mouse MAbs that were raised against LOSs of immunotype L3,7,9 were compared. The hybridoma cell lines producing these MAbs were made by immunizing animals with either oligosaccharide-tetanus toxoid conjugate (hybridoma 4A8-B2) (21) or outer membrane complexes (hybridoma 9-2-L379) (24). Unlike MAb 4A8-B2, MAb 9-2-L3,7,9 cross-reacted with the L2, L5, and L8 immunotypes (18), but ascitic fluid produced from this hybridoma appeared to be more sensitive, being usable at a much higher dilution. The use of purified antibodies and LOS in ELISA, together with real-time kinetic analyses with the same reagents, established the relative sensitivities of these MAbs and allowed us to measure their binding kinetics. These data were correlated with the deduced primary structures of the antibodies and known sugar structures of the relevant LOS molecules.

MATERIALS AND METHODS

Preparation of purified L3,7,9 LOS.

Purified immunoreactive LOS for use in ELISA and biosensor analysis was prepared from meningococcal isolate K454 (B:15:P1.7,16:L3,7,9) as described previously (6). The LOS was resuspended in distilled water, dispensed into 0.5-ml samples, vacuum dried, and stored at −20°C until required. Each sample contained 30 to 60 ng of LOS, as estimated by the Limulus amoebocyte lysate (LAL) chromogenic assay (23). The purity, number of species, and Mr of the LOSs were determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and silver staining (6, 19) (data not shown).

Purification of MAbs.

The MAbs were purified from ascites collected from pristane treated mice after intraperitoneal injection with cells of the appropriate hybridoma cell line. The resultant ascitic fluids were buffer exchanged on a Biogel P4 desalting column (Pharmacia, Plc.) with 20 mM sodium phosphate buffer (pH 7.0); the MAbs captured on a protein G column and eluted with 0.1 M glycine (pH 2.7). The peak fractions were pooled, neutralized with Tris (pH 8.8; 65 mM), and buffer exchanged in phosphate-buffered saline (PBS; pH 7.4). The samples were concentrated to ca. 1 μM and tested for immunoglobulin subclass by using an isotyping kit for mouse MAbs (Serotec, Ltd.). The ascites containing MAb 4A8-B2 gave two protein peaks after elution from the protein G column with 0.1 M glycine (pH 2.7). The first peak contained both immunoglobulin G2a (IgG2a) and IgM antibodies, indicating that this sample was contaminated with serum. A second peak contained antibodies of the IgG2a subclass alone, and only the antibodies from this peak were used. Ascites containing MAb 9-2-L379 had a single peak, comprising antibodies of the IgG2a subclass.

PCR cloning and sequencing of the immunoglobulin variable regions.

Hybridoma cell lines were grown to confluence in 25-ml flasks at 37°C in a 5% CO2 atmosphere. Total RNA, prepared with an Isoquick Kit (Orca Research, Inc.), was used to synthesize cDNA by extension of an oligo(dT) primer by using a reverse transcription kit (Clontech UK, Ltd.). The immunoglobulin variable (V) regions encoding each of the MAbs were rescued by PCR from degenerate primers designed from the immunoglobulin framework regions bordering the VHCγ and VL domains as described by Kettleborough et al. (11). The PCR products encoding the VHCγ domains were cloned between the AatII and SalI sites of both the f+ and f− forms of the vector pGEM5Z, while the sequences encoding the VL regions were similarly cloned between the NcoI and XhoI sites of both pGEM5Z vectors. The nucleotide sequences of several independently isolated clones were determined on both strands by “cycle sequencing” with a Taquence kit (Amersham) with M13 forward and reverse primers radiolabelled by T4 polynucleotide kinase with [γ-32P]ATP.

ELISA.

For whole-cell ELISA microtiter plate wells were coated with N. meningitidis K454 (L3,7,9) by the method of Abdillahi and Poolman (1). Coating of plate wells with purified LOS was as described previously (21). To avoid interplate variation, assays on both MAbs were carried out on the same microtiter plate. The secondary antibody was anti-mouse IgG conjugated to horseradish peroxidase. The absorbance was read at 450 nm 30 min after the addition of the chromogenic substrate (0.4 mg of 1,2-phenylenediamine dihydrochloride and 0.4 mg of urea hydrogen peroxide per ml in 0.05 M phosphate citrate buffer, pH 5.0).

Immobilization of L3,7,9 LOS to the resonant mirror biosensor surface.

Purified LOS was biotinylated (5), lyophilized and stored in lots of 30 to 50 ng, as estimated by the chromogenic LAL assay, at −20°C. Immobilization of the LOS was carried out in an IAsys resonant mirror biosensor (Affinity Sensors, Cambridge, United Kingdom), essentially according to the manufacturer’s protocol. Streptavidin (Sigma) was captured onto the biotin-coated biosensor cuvette surface in PBST (10 mM sodium phosphate–138 mM NaCl–2.7 mM KCl [pH 7.4] containing 0.05% Tween 20), and unbound streptavidin was removed by washing with PBST after 10 min. Biotinylated LOS (3 to 5 ng) was added and binding was monitored. A response of 100 arc seconds was observed on the addition of LOS to the cuvette. Further additions did not increase the sensitivity of the assay (data not shown). A final bovine serum albumin (BSA) blocking step was performed by reacting the biosensor cuvette with 0.1 mg of BSA per ml in PBST for 5 min. The LOS-coated biosensor surface was treated with 20 mM HCl to remove any weakly bound substances before interaction kinetics were performed and also to regenerate the LOS surface prior to interactions with various MAb concentrations. To obtain comparative kinetic data, the same LOS-coated biosensor cuvette was used with both of the MAbs.

Resonant mirror biosensor analysis.

Real-time kinetic analyses with the IAsys resonant mirror biosensor were undertaken in PBST at 25°C, according to the methods described by the manufacturer. The kinetic data were analyzed by curve-fitting software (FASTfit v2.01), and the binding curves from different MAb concentrations were overlaid and plotted by using FASTplot software (both supplied by Affinity Sensors). Dissociation rates (KOFF) were determined by dilution of unbound MAb in the biosensor cuvette to zero concentration at relatively high concentrations of antibody (∼10 times the dissociation equilibrium constant, KD) and averaged to give the dissociation rate constant Kdiss. Initially, approximate KD values were obtained from concentrations of MAb ranging from nano- to micromolar levels with subsequent, more-accurate KD and affinity constant (KA) values determined from MAb concentrations ranging from 0.01 to 10 times the approximate KD value.

The binding data at high MAb concentrations (∼10 times the KD) fitted a biphasic curve, but only the initial rates were used in the determination of the binding kinetics. The KON rates were calculated from the arc second response over various periods of time and averaged. The gradient of KON rates versus MAb concentration gave the association rate constant Kass. The KA was calculated for each MAb from the Kass/Kdiss ratio, and the KD was calculated from the Kdiss/Kass ratio. The binding data were also reconciled by plotting the total extent of antibody binding in a 10-min interaction period against the MAb concentration and then determining the KD value from nonlinear regression analysis of these binding data (data not shown). The y intercept of the plots of KON versus MAb concentration give approximate KD values, which were in agreement to within 10% of the experimentally derived KD values.

RESULTS

Comparison of MAbs 4A8-B2 and 9-2-L3,7,9 used for immunotyping N. meningitidis by ELISA.

The concentration-dependent binding of 4A8-B2 and 9-2-L379 MAbs to whole cells and to purified LOS demonstrated that MAb 9-2-L379 exhibited approximately a 1,000-fold-greater binding to both whole cells and purified LOS than 4A8-B2 (Fig. 1). The binding observed for both MAbs to whole cells was approximately 10-fold weaker than to the purified LOS.

FIG. 1.

FIG. 1

Open in a new tab

ELISA of MAbs 4A8-B2 and 9-2-L379 against N. meningitidis cells and purified L3,7,9 LOS. The relative binding of MAb 9-2-L379 to purified LOS (○) and whole cells (●) and MAb 4A8-B2 to purified LOS (□) and whole cells (■) of the same meningococcal isolate are shown. Error bars represent the standard deviation of triplicate determinations.

Interaction kinetics of MAbs 4A8-B2 and 9-2-L379 with L3,7,9 LOS.

The real-time binding interactions of 4A8-B2 and 9-2-L379 to immobilized biotinylated LOS, as indicated by arc second response, gave KON rates for MAb 9-2-L379 that were more rapid than that for MAb 4A8-B2, whereas the arc second responses were approximately fourfold greater for MAb 4A8-B2 (Fig. 2). The Kass of MAb 9-2-L379 was 44-fold greater than that of MAb 4A8-B2, whereas the Kdiss were similar for both antibodies. The KA for MAb 9-2-L379 was 31-fold greater than that of MAb 4A8-B2 (Fig. 3; Table 1). Although real-time kinetic data between anti-carbohydrate IgGs and their carbohydrate antigens is limited, the binding kinetic data reported in this study are comparable to the binding kinetics of an anti-carbohydrate IgG against Salmonella serogroup B O-polysaccharide as determined by using surface plasmon resonance (12).

FIG. 2.

FIG. 2

Open in a new tab

Real-time interaction curves for MAbs 4A8-B2 and 9-2-L379 against immobilized L3,7,9. Real-time binding interaction curves for various concentrations of each MAb with purified LOS of immunotype L3,7,9 are shown. (A) Overlay of the binding curves of MAb 4A8-B2 at 0.07, 0.17, 0.35, 0.52, 0.70, and 1.05 μM. (B) Overlay of the binding curves of the MAb 9-2-L379 at 0.005, 0.01, 0.02, 0.03, 0.10, and 0.25 μM. For clarity, not all of the binding data collected are shown, and the dissociation curves have been omitted.

FIG. 3.

FIG. 3

Open in a new tab

Binding kinetics for MAbs 4A8-B2 and 9-2-L379 against immobilized L3,7,9 LOS. The association rates (KON) of MAb 4A8-B2 (●) and MAb 9-2-L379 (○) with purified LOS at 25°C, plotted against MAb concentration, are shown. The Kass constant of each MAb was derived from the gradient of the slope.

TABLE 1.

Binding kinetics of MAb 4A8-B2 and 9-2-L379a

MAb Kass (M−1 s−1) Kdiss (s−1) KA (M−1) KD (M)
9-2-L379 1.33 (±0.01) × 106 (n = 7) 0.010 (±0.005) (n = 10) 133 (±67) × 106 7.52 (±3.8) × 10−9
4A8-B2 0.03 (±0.002) × 106 (n = 8) 0.007 (±0.004) (n = 8) 4.29 (±2.47) × 106 233.3 (±134.1) × 10−9
Open in a new tab
a

KA is calculated from the Kass/Kdiss ratio. KD is calculated from the Kdiss/Kass ratio. Figures in parentheses are the standard deviation of the given measurement, and n is the number of replicate KON or KOFF rates used. 

DNA sequence analysis and primary sequence comparison of the V regions of MAbs 9-2-L379 and 4A8-B2.

The VHV-D-J region of MAbs 4A8-B2 and 9-2-L379 shared 81% overall identity, both being derived from the J558 V region family. The MAb 4A8-B2 and MAb 9-2-L379 heavy chains showed similar levels of identity to the germ line genes VMU3.2 and 186-2 and used JH2 and JH3, respectively. The deduced amino acid sequences of their CDR1s were identical, their CDR2s were 88% identical, and their CDR3s had no sequence identity.

The amino acid sequences of the VL regions of the MAbs were 54% identical overall, the CDR1s, CDR2s, and CDR3s exhibiting 47, 14, and 11% identity, respectively (Fig. 4). The VL regions of the MAbs were encoded by different Vκ gene families, Vκ 8 in MAb 4A8-B2 and Vκ ARS in MAb 9-2-L379. Comparison of the nucleotide sequences of MAb 4A8-B2 and MAb 9-2-L379 with other murine antibody genes revealed that MAb 4A8-B2 κ light chain was 82% identical to the D23 germline and used the Jκ1 J segment, whereas the MAb 9-2-L379 κ chain was 84% identical to germ line gene 28.4.10A(κ) and used Jκ2. In addition to the high degree of identity to κ light chains, both MAb 4A8-B2 and MAb 9-2-L379 had the highly conserved residues phenylalanine at position 71 and glutamine at positions 90 for MAb 9-2-L379 and 89 for MAb 4A8-B2 (Fig. 4), confirming the VL regions as κ light chains. Significant differences were found between the two MAbs in the VL CDR1, where MAb 9-2-L379 had four positively charged residues and MAb 4A8-B2 had one, and in the VH CDR3, where MAb 9-2-L379 had no charged residues but three hydrophilic residues and MAb 4A8-B2 had two negatively charged residues and one positively charged residue, with only one hydrophilic residue (Fig. 4).

FIG. 4.

FIG. 4

Open in a new tab

Primary sequence comparison of the variable regions within MAbs 4A8-B2 and 9-2-L379. The aligned, deduced primary sequences of the variable regions from MAbs 4A8-B2 and 9-2-L379 are shown. The amino acid sequences corresponding to MAb 9-2-L379 are compared with those of MAb 4A8-B2. Amino acid identities are represented with a period, and differences are indicated by the appropriate letter; hyphens represent deletions relative to the sequence of MAb 4A8-B2. The sequences are numbered according to the Kabat database (9), with lowercase letters representing variable-length insertion sequences. CDRs are indicated by white text on a black background.

DISCUSSION

N. meningitidis presents carbohydrate structures to its human host that mimic self-antigens and are poor immunogens. During colonization of the nasopharynx, switching between the capsulate and acapsulate forms occurs, exposing the capsule and the outer membrane LOS sequentially (10). Consequently, the interactions between host defences and the bacterial carbohydrate are of central importance in understanding the pathogenicity of, and in the development of vaccines against, the meningococcus. Further, mouse MAbs are important reagents in the immunotyping of this organism (18). The measurement of the binding properties of antibodies to antigens by real-time binding kinetic analysis therefore has potential applications in both the standardization of immunotyping reagents and assays and in the investigation of human responses to bacterial antigens.

The two antibodies investigated in the present work, although originally raised against the same meningococcal LOS immunotype, were produced by distinct immunization protocols and exhibited different apparent sensitivities and specificities in routine immunotyping ELISAs (18). The results obtained here with purified reagents confirmed that the less-specific MAb 9-2-L379, raised against outer membrane complexes, was one 1,000-fold more sensitive than the more-specific MAb 4A8-B2, which had been raised against tetanus-toxoid conjugate. Real-time kinetic analysis by using a resonant mirror biosensor revealed that the more sensitive reagent, MAb 9-2-L379, had a 44-fold-faster Kass and a 31-fold-higher KA than MAb 4A8-B2, while the Kdiss values of the two antibodies were similar. Interestingly, the arc second response seen with MAb 4A8-B2 was fourfold greater than that observed with MAb 9-2-L379, implying that in the biosensor assays MAb 4A8-B2 bound in greater quantities to the immobilized LOS. The latter observation contradicted the results of the ELISAs, which indicated that this MAb bound relatively poorly to LOS. As the arc second response recorded in the biosensor is dependent upon both the quantity of antibody bound and the distance of the antibody from the resonant surface (4), this apparent discrepancy between assays may be the result of differences in the proximity of the bound antibodies to the resonant mirror.

Comparisons of the deduced primary structures of the two antibodies revealed the differences responsible for their distinct binding activities. Notwithstanding the different protocols used for their production in different laboratories, the two antibodies shared practically identical heavy chains, with the exception of their CDR3s, but possessed diverse light chains. This observation presents the interesting prospect of using antibody engineering techniques (2) to produce anti-L3,7,9 MAbs with different properties by using different combinations of the various complementarity-determining regions (CDRs) reported here. An MAb with the sensitivity of MAb 9-2-L379 but the specificity of MAb 4A8-B2 would be a particularly useful reagent. Such constructs would also be potentially valuable in improving our understanding of the immunology of LOS immunotypes.

The differences of the specificity and binding kinetics of these two MAbs, together with the differences in their sequences and the differences in the arc second response observed, implied that they recognized distinct epitopes within the LOS of immunotype L3,7,9. Previous studies of the specificity of anti-LOS immunotyping MAbs in whole-cell ELISA demonstrated that MAb 9-2-L379 cross reacted with LOS of immunotypes L2, L3,7,9, L5, and L8, whereas MAb 4A8-B2 was specific for LOS of immunotype L3,7,9 (18). These observations, in combination with the data reported here, suggest that the specificity of MAb 4A8-B2 may be the result of interactions with both the phosphoethanolamine (PEA) group 1-3 linked to the second heptose of the L3,7,9 structure and the terminal disaccharide of the lacto-N-neotetraose moiety. Conversely, MAb 9-2-L379 may interact with the galactose residue, located between the N-acetylglucosamine and glucose residues in the lacto-N-neotetraose moiety, in combination with the same PEA group which are present in both the L3 and L8 immunotypes. The relatively weak cross-reactivity of MAb 9-2-379 with LOS of immunotypes L2 and L5 suggests that the interaction with this PEA group can be replaced, to a limited extent, by interactions with a glucose moiety in the same position (Fig. 5).

FIG. 5.

FIG. 5

Open in a new tab

Sugar structures of LOS molecules known to interact with MAbs 9-2-L379 and 4A8-B2. The structures of LOS molecules associated with the immunotypes L2, L3,7,9, L5, and L8 are given (7, 15, 18, 22), with the sugar moieties that are likely to be involved in the binding interactions that distinguish the two MAbs highlighted. Those moieties putatively contributing to the epitope recognized by MAb 9-2-L379 are shown with boldface text, while those that might play a role in the more specific interaction of MAb 4A8-B2 with L3,7,9 LOS are shown boxed. The PEA group that is potentially involved in binding of both antibodies to LOS immunotype L3,7,9 is shown boxed in boldface text.

While much remains to be learned concerning antibody-LOS interactions, and particularly human antibody-LOS interactions, this analysis of two mouse MAbs by a combination of resonant mirror and antibody sequencing technologies shows the potential of these techniques in enhancing our understanding of antibody-carbohydrate interactions for both routine typing and research purposes.

ACKNOWLEDGMENTS

We thank Jan Poolman and Wendell Zollinger for providing the hybridomas used in this study.

This work was funded by grant number G9202857SB from the UK Medical Research Council. M.C.J.M. is a Wellcome Trust Senior Research Fellow in Biodiversity.

REFERENCES

  • 1.Abdillahi H, Poolman J T. Whole-cell ELISA for typing Neisseria meningitidis with monoclonal antibodies. FEMS Microbiol Lett. 1987;48:367–371. [PubMed] [Google Scholar]
  • 2.Alfthan K. Surface plasmon resonance biosensors as a tool in antibody engineering. Biosens Bioelectron. 1998;13:653–663. doi: 10.1016/s0956-5663(98)00020-7. [DOI] [PubMed] [Google Scholar]
  • 3.Cartwright K A V. Meningococcal disease. Chichester, England: Wiley; 1995. [Google Scholar]
  • 4.Davies R J, Pollard Knight D. An optical biosensor system for molecular interaction studies. Am Biotechnol Lab. 1993;11:52–54. [PubMed] [Google Scholar]
  • 5.Diaz-Romero J, Outschoorn I. Selective biotinylation of Neisseria meningitidis group B capsular polysaccharide and application in an improved ELISA for the detection of specific antibodies. J Immunol Methods. 1993;160:35–47. doi: 10.1016/0022-1759(93)90006-s. [DOI] [PubMed] [Google Scholar]
  • 6.Evans J S, Maiden M C J. Purification of meningococcal lipooligosaccharide by FPLC techniques. Microbiology. 1996;142:57–62. doi: 10.1099/13500872-142-1-57. [DOI] [PubMed] [Google Scholar]
  • 7.Gamian A, Beurret M, Michon F, Brisson J R, Jennings H J. Structure of the L2 lipopolysaccharide core oligosaccharides of Neisseria meningitidis. J Biol Chem. 1992;267:922–925. [PubMed] [Google Scholar]
  • 8.Jennings H J, Johnson K G, Kenne L. The structure of an R-type oligosaccharide core obtained from some lipopolysaccharides of Neisseria meningitidis. Carbohydr Res. 1983;121:233–241. doi: 10.1016/0008-6215(83)84020-8. [DOI] [PubMed] [Google Scholar]
  • 9.Johnson G, Kabat E A, Wu T T. Kabat database of sequences of proteins of immunological interest. In: Weir W M, Herzenberg L A, Blackwell C C, editors. Weir’s handbook of experimental immunology I. Immunochemistry and molecular immunology. 5th ed. Cambridge, Mass: Blackwell Science, Inc.; 1996. pp. 6.1–6.21. [Google Scholar]
  • 10.Kahler C M, Stephens D S. Genetic basis for biosynthesis, structure, and function of meningococcal lipooligosaccharide. Crit Rev Microbiol. 1998;24:281–334. doi: 10.1080/10408419891294216. [DOI] [PubMed] [Google Scholar]
  • 11.Kettleborough C A, Saldanha J, Ansell K H, Bendig M M. Optimization of primers for cloning libraries of mouse immunoglobulin genes using the polymerase chain reaction. Eur J Immunol. 1993;23:206–211. doi: 10.1002/eji.1830230132. [DOI] [PubMed] [Google Scholar]
  • 12.MacKenzie C R, Hirama T, Deng S J, Bundle D R, Narang S A, Young N M. Analysis by surface plasmon resonance of the influence of valence on the ligand binding affinity and kinetics of an anti-carbohydrate antibody. J Biol Chem. 1996;271:1527–1533. doi: 10.1074/jbc.271.3.1527. [DOI] [PubMed] [Google Scholar]
  • 13.Mackinnon F G, Borrow R, Gorringe A R, Fox A J, Jones D M, Robinson A. Demonstration of lipooligosaccharide immunotype and capsule as virulence factors for Neisseria meningitidis using an infant mouse intranasal infection model. Microb Pathog. 1993;15:359–366. doi: 10.1006/mpat.1993.1085. [DOI] [PubMed] [Google Scholar]
  • 14.Mandrell R E, Apicella M A. Lipo-oligosaccharides (LOS) of mucosal pathogens: molecular mimicry and host-modification of LOS. Immunobiology. 1993;187:382–402. doi: 10.1016/S0171-2985(11)80352-9. [DOI] [PubMed] [Google Scholar]
  • 15.Michon F, Beurret M, Gamian A, Brisson J R, Jennings H J. Structure of the L5 lipopolysaccharide core oligosaccharides of Neisseria meningitidis. J Biol Chem. 1990;265:7243–7247. [PubMed] [Google Scholar]
  • 16.Moran A P, Prendergast M M, Appelmelk B J. Molecular mimicry of host structures by bacterial lipopolysaccharides and its contribution to disease. FEMS Immunol Med Microbiol. 1996;16:105–115. doi: 10.1111/j.1574-695X.1996.tb00127.x. [DOI] [PubMed] [Google Scholar]
  • 17.Preston A, Mandrell R E, Gibson B W, Apicella M A. The lipooligosaccharides of pathogenic gram-negative bacteria. Crit Rev Microbiol. 1996;22:139–180. doi: 10.3109/10408419609106458. [DOI] [PubMed] [Google Scholar]
  • 18.Scholten R J, Kuipers B, Valkenburg H A, Dankert J, Zollinger W D, Poolman J T. Lipo-oligosaccharide immunotyping of Neisseria meningitidis by a whole-cell ELISA with monoclonal antibodies. J Med Microbiol. 1994;41:236–243. doi: 10.1099/00222615-41-4-236. [DOI] [PubMed] [Google Scholar]
  • 19.Tsai C-M, Frasch C E. A sensitive silver stain for detecting lipopolysaccharides in polyacrylamide gels. Anal Biochem. 1982;119:115–119. doi: 10.1016/0003-2697(82)90673-x. [DOI] [PubMed] [Google Scholar]
  • 20.Verheul A F, Boons G J, Van der Marel G A, Van Boom J H, Jennings H J, Snippe H, Verhoef J, Hoogerhout P, Poolman J T. Minimal oligosaccharide structures required for induction of immune responses against meningococcal immunotype L1, L2, and L3,7,9 lipopolysaccharides determined by using synthetic oligosaccharide-protein conjugates. Infect Immun. 1991;59:3566–3573. doi: 10.1128/iai.59.10.3566-3573.1991. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Verheul A F, Braat A K, Leenhouts J M, Hoogerhout P, Poolman J T, Snippe H, Verhoef J. Preparation, characterization, and immunogenicity of meningococcal immunotype L2 and L3,7,9 phosphoethanolamine group-containing oligosaccharide-protein conjugates. Infect Immun. 1991;59:843–851. doi: 10.1128/iai.59.3.843-851.1991. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Verheul A F, Poolman J T, Snippe H, Verhoef J. The influence of the adjuvant Quil A on the epitope specificity of meningococcal lipopolysaccharide anti-carbohydrate antibodies. Mol Immunol. 1991;28:1193–1200. doi: 10.1016/0161-5890(91)90005-5. [DOI] [PubMed] [Google Scholar]
  • 23.Young N S, Levin J, Prendergast R A. An invertebrate coagulation system activated by endotoxin: evidence for enzymatic mediation. J Clin Investig. 1972;51:1790–1797. doi: 10.1172/JCI106980. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Zollinger W D, Mandrell R E. Importance of complement source in bactericidal activity of human antibody and murine monoclonal antibody to meningococcal group B polysaccharide. Infect Immun. 1983;40:257–264. doi: 10.1128/iai.40.1.257-264.1983. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Clinical and Diagnostic Laboratory Immunology are provided here courtesy of American Society for Microbiology (ASM)

RESOURCES