Antigenic potential of a highly conserved Neisseria meningitidis lipopolysaccharide inner core structure defined by chemical synthesis.

Chemistry & Biology

Article Antigenic Potential of a Highly Conserved Neisseria meningitidis Lipopolysaccharide Inner Core Structure Defined by Chemical Synthesis Anika Reinhardt,1,2,6 You Yang,1,6 Heike Claus,4 Claney L. Pereira,1 Andrew D. Cox,3 Ulrich Vogel,4 Chakkumkal Anish,1,5,* and Peter H. Seeberger1,2,* 1Department

of Biomolecular Systems, Max Planck Institute of Colloids and Interfaces, 14424 Potsdam, Germany of Chemistry and Biochemistry, Freie Universita¨t Berlin, Arnimallee 22, 14195 Berlin, Germany 3Vaccine Program, Human Health Therapeutics Portfolio, National Research Council, 100 Sussex Drive, Ottawa, ON K1A 0R6, Canada 4Institute for Hygiene and Microbiology, University of Wu ¨ rzburg, 97080 Wu¨rzburg, Germany 5Present address: Bacterial Vaccines Discovery and Early Development, Janssen Pharmaceuticals (Johnson & Johnson), Bioscience Park Leiden Zernikedreef 9, 2333 CK Leiden, the Netherlands 6Co-first author *Correspondence: [email protected] (C.A.), [email protected] (P.H.S.) http://dx.doi.org/10.1016/j.chembiol.2014.11.016 2Institute

SUMMARY

Neisseria meningitidis is a leading cause of bacterial meningitis worldwide. We studied the potential of synthetic lipopolysaccharide (LPS) inner core structures as broadly protective antigens against N. meningitidis. Based on the specific reactivity of human serum antibodies to synthetic LPS cores, we selected a highly conserved LPS core tetrasaccharide as a promising antigen. This LPS inner core tetrasaccharide induced a robust IgG response in mice when formulated as an immunogenic glycoconjugate. Binding of raised mouse serum to a broad collection of N. meningitidis strains demonstrated the accessibility of the LPS core on viable bacteria. The distal trisaccharide was identified as the crucial epitope, whereas the proximal Kdo moiety was immunodominant and induced mainly nonprotective antibodies that are responsible for lack of functional protection in polyclonal serum. Our results identified key antigenic determinants of LPS core glycan and, hence, may aid the design of a broadly protective immunization against N. meningitidis.

INTRODUCTION N. meningitidis is an encapsulated human bacterial pathogen that causes meningitis and sepsis worldwide. Vaccines based on capsular polysaccharides are available against invasive meningococcal serogroups A, C, W, and Y, but not for B, as its capsule consists of autoantigenic a-(2/8)-polysialic acid (Finne et al., 1983; Granoff, 2010). Thus, vaccines based on noncapsular antigens are needed to limit N. meningitidis B infections, with globally 20,000 to 80,000 cases per year, especially in developed countries (Hedari et al., 2014). Recently, a multicomponent vaccine 4CMenB, based on recombinant proteins and outer membrane vesicles, was developed by Novartis (Bexsero) that

is protective against 88% of N. meningitidis B strains (Frosi et al., 2013). The composition of 4CMenB was adjusted to target serogroup B strains. The vaccine components of 4CMenB are present in N. meningitidis isolates of other serogroups and bear the potential to protect against non-B serogroups (Hong et al., 2013). Alternatively, lipopolysaccharide (LPS) presents a promising subcapsular antigen that might elicit broader reactivity against all N. meningitidis strains. The LPS is also called lipooligosaccharide because of the absence of O-polysaccharide and is anchored via lipid A in the outer membrane (Nikaido, 2003; Raetz and Whitfield, 2002). The meningococcal LPSs only show a core oligosaccharide that is subdivided into inner and outer cores. The inner core consists of two Kdo (3-deoxya-D-manno-oct-2-ulosonic acid) and two Hep (L-glycero-Dmanno-heptose) residues (Figure 1). The LPS outer core contains mostly of hexoses building up variable a chain extensions from HepI (Figure 1, inset). In addition, b chain and g chain extensions and a phosphoethanolamine (PEtN) modification on HepII are possible. These diverse outer core compositions determine 12 distinct immunotypes of N. meningitidis (L1–12) (Choudhury et al., 2008; Nikaido, 2003; Plested et al., 1999). The immunotype expression is subjected to phase variation (Berrington et al., 2002; Zhu et al., 2002). L1–L8 immunotypes are associated with serogroups B and C, and L9 immunotype is common in serogroups A, B, and C, whereas L10–L12 immunotypes are found in serogroup A (Mistretta et al., 2010). In contrast, the LPS inner core is highly conserved because inner core glycosyltransferase genes are not subjected to phase variation (Kahler et al., 2005). The potential of the LPS inner core as subcapsular antigen was evaluated earlier using purified LPS from a DgalE mutant of the widely studied N. meningitidis B strain MC58 (immunotype L3; Figure 1, inset). The DgalE LPS shows an inner core with an a chain extension of b-(1/4)-linked Glc on HepI, with a-(1/2)linked GlcNAc as a g chain extension and PEtN modification at the 3-position of HepII. A monoclonal antibody (mAb) B5 raised against the purified DgalE LPS bound to 76% of N. meningitidis B strains and to 70% of a collection of N. meningitidis strains of serogroups A, C, W, X, Y, and Z. All these strains share the common DgalE LPS core epitope with PEtN at the 3-position of HepII (Plested et al., 1999). mAb B5 has been shown to promote serum

38 Chemistry & Biology 22, 38–49, January 22, 2015 ª2015 Elsevier Ltd All rights reserved

Chemistry & Biology Defined LPS Inner Core Antigen for N. meningitidis

Figure 1. Library of 12 Synthetic LPS Core Structures Natural LPS structure immunotype L3 of N. meningitidis (inset) and synthetic LPS core structures from N. meningitidis (1), Yersinia pestis (8, 11), Haemophilus influenzae (9), Proteus (10), Chlamydia (12), and their truncated derivatives (2–7).

bactericidal activity (SBA) in a gold standard surrogate assay for in vivo protection against N. meningitidis infections (Plested et al., 2001, 2003). Human antibodies specific for LPS inner cores showed bactericidal and opsonic activity as well (Ja¨kel

et al., 2008; Plested et al., 2001). However, conjugation of detoxified LPS to a carrier protein that is required to induce a robust immune response is not straightforward. Aggregation and low solubility due to the amphiphilic character of purified LPS

Chemistry & Biology 22, 38–49, January 22, 2015 ª2015 Elsevier Ltd All rights reserved 39


impede reproducible glycoconjugate synthesis. Deacylation and incorporation of functional groups for covalent linkage to carrier proteins are performed on purified LPS prior to conjugation but generate neoepitopes that may elicit nonspecific antibodies (Cox et al., 2010a, 2010b; St Michael et al., 2014). Furthermore, purified LPS offers a limited possibility for antigen refinement due to its restriction to sequentially expressed LPS core structures on engineered knockout mutants. In contrast, chemical synthesis offers virtually unlimited structural variability that allows precise definition and manipulation of the antigen. We used a combination of chemical synthesis and glycan array screening to define a highly conserved portion of the LPS inner core structure as a potential antigen candidate. For that purpose, a library of species-specific LPS inner core oligosaccharides was synthesized (Figure 1). Based on glycan array screening of these synthetic oligosaccharides with human sera, the conserved LPS inner core structure of all N. meningitidis immunotypes a-DGlcNAc-(1/2)-L-a-D-Hep-(1/3)-L-a-D-Hep-(1/5)-a-Kdo (Figure 1) was chosen for further immunological evaluation. Therefore, this tetrasaccharide was conjugated to the carrier protein CRM197, a nontoxic diphtheria toxin variant commonly used for glycoconjugate vaccines. The resulting glycoconjugate generated an antibody response against the tetrasaccharide and a collection of closely related synthetic LPS core analogs in mice. Binding of serum antibodies to a broad collection of viable wild-type (wt) N. meningitidis strains expressing diverse immunotypes demonstrated the accessibility of the LPS inner core on the bacterial cell surface. The structural requirements of the synthetic antigen to elicit protective antibodies were assessed with a combination of glycan array, surface plasmon resonance (SPR), and immunoassays. We report the epitope mapping of mAbs and polyclonal antibodies specific to the LPS core and its correlation with binding to viable N. meningitidis strains. Our approach demonstrates the usefulness of synthetic LPS core structures in identifying crucial epitopes as a basis for the development of an improved synthetic LPS core antigen that may elicit antibodies with broad reactivity to N. meningitidis strains. RESULTS AND DISCUSSION Synthesis of LPS Inner Core Oligosaccharides 5 and 6 The LPS inner core oligosaccharides 1–4, 7–10 (Yang et al., 2012, 2013), 11 (Anish et al., 2013), and 12 (Guo, 2011) were prepared as previously described by relying on de novo and diversity-oriented synthetic strategies (Figure 1). However, synthesis of disaccharide 6 and trisaccharide 5 can be conducted with less sophisticated building blocks. L-a-D-Hep-(1/3)-L-a-D-Hep disaccharide 6 and a-D-GlcNAc-(1/2)-L-a-D-Hep-(1/3)-La-D-Hep trisaccharide 5 were assembled from building blocks 13–15 (Figure 2). Hep building blocks 14 and 15 were prepared by using our previously developed synthetic procedures (Yang et al., 2012, 2013). Hep building block 13 was accessed by introducing an N-benzyl-N-benzyloxycarbonylpentyl linker (Mong et al., 2003; Noti et al., 2006) on the reducing end of Hep intermediate 16 that was, in turn, obtained by de novo synthesis (Ohara et al., 2010). Hep intermediate 16 was converted into the corresponding N-phenyl trifluoroacetimidate (Yu and Tao, 2001, 2002) 17 upon selective anomeric delevulinoylation and N-phenyl trifluoroacetimidate formation (78% yield over two steps). Glyco-

sylation of linker 18 with N-phenyl trifluoroacetimidate 17 under the catalysis of TMSOTf in a mixture of dichloromethane and ether produced the desired a-linked monosaccharide 19 in 72% yield. In the presence of hydrazine and acetic acid, the OLev group in 19 was selectively removed to afford Hep building block 13 in 87% yield. The coupling of Hep 14 and 13 promoted by TMSOTf gave the desired a-(1/3)-linked disaccharide that was treated with HF, pyridine to afford disaccharide 20 in 65% yield over two steps. Saponification of the acetyl groups in 20 with sodium methoxide followed by hydrogenolysis over palladium on carbon (Pd/C) provided L-a-D-Hep-(1/3)-L-a-D-Hep disaccharide 6 in 75% yield over two steps. Coupling of disaccharide 15 and Hep 13 promoted by TMSOTf in a mixture of dichloromethane and ether gave the desired a-(1/3)-linked trisaccharide. Treatment of the newly formed trisaccharide with thioacetic acid in pyridine allowed separation of trisaccharide 21 from the remaining derivative of monosaccharide 13 at this step, thus affording acetamide 21 in 72% yield over two steps. Global deprotection of 21 involved TBDPS cleavage, saponification, and hydrogenolysis over Pd/C to provide a-D-GlcNAc-(1/ 2)-L-a-D-Hep-(1/3)-L-a-D-Hep trisaccharide 5 (33% yield over three steps). Naturally Occurring Human Antibodies Bind Selectively to Synthetic LPS Inner Core Oligosaccharides N. meningitidis colonizes the nasopharyngeal mucosa asymptomatically, a state referred to as carriage occurring in 35% of healthy young adults, and opportunistically invades and infects exclusively humans (Caugant et al., 2007). Mapping binding preferences of natural antibodies raised by asymptomatic carriers and infected patients provides direct evidence for their immunogenicity in vivo that helps to define the antigenic epitopes. Naturally occurring immunoglobulin (Ig)G and IgM antibodies specific for DgalE LPS core glycolipid were earlier detected in healthy and infected persons via ELISA, demonstrating the immunogenicity of the DgalE LPS core (Plested et al., 2000). To examine whether antibodies specific for synthetic LPS core structures are present in humans who have been exposed to N. meningitidis, we performed glycan array screenings of sera from convalescent patients who recovered from N. meningitidis infection (n = 11), asymptomatic N. meningitidis carriers (n = 11), and healthy individuals (n = 15) (Hubert et al., 2013; Figure 3). Glycan array analysis revealed that IgG antibodies in human sera showed exclusive binding to meningococcal tetrasaccharide 1 and trisaccharide 5, while there was only weak binding to trisaccharide 2. Trisaccharide 2 represents a highly conserved LPS inner core of many Gram-negative bacteria. Low binding to 2, compared to 1 and 5, indicates the importance of the distal trisaccharide 5 for antibody recognition of the N. meningitidis LPS core. Weak binding to the LPS cores of Y. pestis 8, H. influenzae 9, and bacteria of genus Proteus 10 revealed the specificity of natural antibodies for N. meningitidis LPS core glycans. The same trend was observed for IgM antibodies (data not shown). No significant differences between healthy individuals and convalescent patients or asymptomatic N. meningitidis carriers were detected. Discrimination between these groups was based on mucosal swabs revealing the current N. meningitidis colonization status of each individual. Possible previous colonization was not considered. Human sera screening revealed the



Figure 2. Synthesis of Disaccharide 6 and Trisaccharide 5 Synthesis of L-a-D-Hep-(1/3)-L-a-D-Hep disaccharide 6 and a-D-GlcNAc-(1/2)-L-a-D-Hep-(1/3)-L-a-D-Hep trisaccharide 5. Pyr, pyridine; Rt, room temperature; MS, molecular sieves.

presence of antibodies specific for synthetic LPS cores 1 and 5 that appear to be highly antigenic. Immunological Evaluation of Synthetic LPS Core Tetrasaccharide 1 from N. meningitidis Based on the results of the human serum antibody screening, the immunogenicity of synthetic LPS core tetrasaccharide 1 was tested in a mouse model. Tetrasaccharide 1 was selected over trisaccharide 5 to elicit antibodies with broader specificities (Evenberg et al., 1992; Pozsgay et al., 1999). In addition, the immunogenic potential of charged motifs in oligosaccharide antigens justified the selection of a synthetic target containing a Kdo moiety (Adamo et al., 2012; Palusiak et al., 2014). Most carbohydrates are poorly immunogenic and fail to induce longlasting memory response. In order to recruit T cell help, we conjugated 1 to carrier protein CRM197 using the previously described di-N-succinimidyl adipate (DSAP) spacer (Bro¨ker et al., 2009, 2011), generating tetrasaccharide 1-CRM197 conjugate 22 (Figures S1A–S1C available online) (Anish et al., 2013; Martin et al., 2013). To investigate the antigenic potential of tetrasaccharide 1, three groups of mice (n = 6) were immunized with different doses of conjugate 22 corresponding to 3 mg, 2 mg, and 1 mg of tetrasaccharide 1 in the presence of a human-approved adjuvant, Alum Alhydrogel (Alum) (Clements and Griffiths, 2002;

Figure 4). Mice (n = 6) immunized with an amount of 22 containing 3 mg antigen 1 formulated in Freund’s adjuvant (FA) were used as the positive control because of the promotion of strong carbohydrate-specific immune responses by this adjuvant (Anish et al., 2013; Fu et al., 1992; Martin et al., 2013). The control group (n = 6) was sham-immunized with an equal volume of sterile PBS and showed no anti-tetrasaccharide 1 response. Each mouse received priming and boosting of 22 formulated with the respective adjuvant. A dose-dependent anti-tetrasaccharide 1 response was observed in all mice immunized with Alumformulated glycoconjugate 22 at day 15 and day 36 (Figure 4). Antibodies against the carrier protein 28 (Figure S1G) and spacer moiety 27 were also observed (Figure 4). Increased fluorescence intensities were detected for BSA-[spacer-GlcNAc]16 27 compared to BSA-[spacer-DiMan]21 34 due to GlcNAc binding antibodies (Figure S1E). A dose of conjugate 22 containing 1 mg tetrasaccharide 1 was sufficient to elicit a carbohydratespecific immune response. We observed increased antibody levels after boosting for all groups immunized with 22, demonstrating successful recruitment of T cell help (day 36). Furthermore, we were interested in the generation of a long-lasting memory response. Therefore, mice were left untreated for 224 days until antibody levels subsided before the animals were reimmunized with one fifth of the priming dose at day



Figure 3. Human IgG Specific to Synthetic LPS Core 1 and 5 on Glycan Array Glycan array analysis of IgG antibodies in sera from N. meningitidis-infected convalescent patients (n = 11), asymptomatic N. meningitidis carriers (n = 11), and healthy individuals (n = 15) against 12 synthetic LPS inner core oligosaccharides. (A) Representative array scan of patient sera (left) and printing pattern (right). (B) Quantification of mean fluorescence intensity (MFI) values for selected LPS cores of N. meningitidis 1, Y. pestis 8, H. influenzae 9 and Proteus 10, and truncated LPS core trisaccharides 5 and 2. Error bars represent SEM of two spots of two separate arrays. MFI values were compared with two-tailed unpaired t tests.

260 (Figure 4). A robust recall response after reimmunization was observed at day 267. Overall, the immunization with tetrasaccharide 1-based glycoconjugate 22 administered with Alum elicited robust primary, secondary, and long-lasting memory IgG responses and revealed the immunogenicity of this glycan. Accessibility of Conserved LPS Core Tetrasaccharide 1 on Viable N. meningitidis On the cell surface of N. meningitidis, tetrasaccharide 1 is presented as a subcapsular antigen that is modified by several side chains (Figure 1, inset). To investigate whether antibodies generated against synthetic tetrasaccharide 1 are able to recognize the natural LPS core, we performed immunolabeling of heat-inactivated N. meningitidis (27 different strains) with postimmune sera generated against conjugate 22 administered with FA (hereinafter referred to as the ‘‘FA group’’), as these mice produced the highest IgG levels against 1. Preimmune sera served as the control. Confocal fluorescence microscopy showed weak IgG binding for preimmune sera, while immunolabeling with postimmune sera revealed strong IgG binding to N. meningitidis bacteria visualized by bright fluorescein isothiocyanate (FITC) fluorescence (Figure 5A). Fluorescence signals observed on the bacterial surfaces indicated that antibodies generated against synthetic tetrasaccharide 1 recognized the LPS core on heat-inactivated N. meningitidis. Immunolabeling of the encapsulated N. meningitidis B wt strain MC58 (used here as a positive control), and its LPS-free DlpxA mutant was quantified by flow cytometry (Figure S2A). Postimmune sera generated against 22 containing 3 mg of antigen 1 formulated with Alum (hereinafter referred to as the ‘‘Alum group’’), and postimmune sera of the FA group showed significant binding to MC58 wt compared to MC58 LPS-free DlpxA mutant. This demonstrates that generated serum antibodies bind to the LPS core of heat-inactivated N. meningitidis bacteria. As inactivation may alter the cell-surface structures and expose buried inner core epitopes, we immunolabeled viable, noninactivated N. meningitidis strains in an ELISA-based assay. As observed for heat-inactivated strains, postimmune sera of the Alum group

showed significantly higher binding to viable MC58 wt compared to LPS-free DlpxA mutant (Figure 5B) (p = 0.0000028). This observation demonstrates that the LPS inner core on noninactivated strains is accessible to antibodies generated against synthetic tetrasaccharide 1. Binding to representative strains of all 12 immunotypes was observed, revealing the presence and accessibility of conserved tetrasaccharide 1 by postimmune sera of the Alum group (Figure 5B). LPS immunotypes differ in length of a chain as well as modification of HepII with PEtN (3- or 6-position) and/ or Glc. In an earlier immunization study with DgalE mutant expressing LPS with a PEtN at the 3-position of HepII, the murine mAb B5 was elicited, which recognized only strains expressing this particular PEtN modification (Plested et al., 1999). However, the synthetic tetrasaccharide 1 used in our approach lacks PEtN modification but elicited antibodies that bound to immunotypes expressing PEtN modification at the 3-position (immunotype L1, L3, L7, or L8) (Verheul et al., 1994) or 6-position (immunotype L2, L4, or L6) (St Michael et al., 2009; Weynants et al., 2009) (Figure 5B). Furthermore, the immunotype L5, which is free of PEtN modification, was also bound by postimmune sera of the Alum group. We systematically mapped the epitope recognized by antibodies on the bacterial surface by testing a collection of mutated N. meningitidis MC58 strains expressing sequentially truncated LPS structures (Figure 6A) (Jennings et al., 1995; Kurzai et al., 2005; Rahman et al., 2001). Postimmune sera of the Alum group showed significant binding to full-length and to all truncated LPSs. It is interesting that the Dpgm and DgalE mutants showed the highest binding compared to other mutants. The LPS core Dpgm mutant closely resembles the synthetic tetrasaccharide 1 (Figure 1). Increased antibody binding to these mutants is in accordance with observations made in earlier immunization studies with detoxified LPS core glycoconjugates that revealed strongest reactivity to homologous strains (Cox et al., 2010a; St Michael et al., 2014). The enhanced recognition of Dpgm and DgalE mutants might be the result of optimal exposure and accessibility of the a-D-GlcNAc-(1/2)-[PEtN-3]-L-a-D-Hep-(1/3)-La-D-Hep motif. Furthermore, binding analysis of natural human



Figure 4. Anti-tetrasaccharide 1-Specific Antibodies in Boosting and Memory Response of Immunized Mice Glycan array analysis of antibodies in mouse sera collected at days 0, 15, and 36, representing the initial immune responses against tetrasaccharide 1. The memory response specific for tetrasaccharide 1 was assessed at days 260 and 267. Mice immunized with equal volumes of sterile PBS were used as the control group. (A) Representative array scans of IgG antibodies in preimmune sera (upper panel) and postimmune sera (lower panel) of a mouse immunized with an amount of 22 containing 3 mg tetrasaccharide 1 (3 mg 1 in 22 + Alum). The printing pattern is shown at the right of the scans. (B) Quantification of mean fluorescence intensity of IgG antibodies in mouse sera specific for tetrasaccharide 1. Error bars represent SEM (n = 6). Relative response units represent mean fluorescence intensity of each group normalized to background fluorescence. See also Figure S1D.

antibodies against synthetic LPS cores showed a binding preference to distal trisaccharide 5 (Figure 3). The importance of a-GlcNAc was proven by the significantly decreased binding (p = 0.00019) to the DrfaK mutant that lacks this unit (Figure 6A). For postimmune sera of the FA group, weak but significant binding to viable MC58 wt, truncated Dlst, and DlgtA LPS mutants was observed (Figure S2C). Similarly to sera of the Alum group, sera of FA group showed increased binding to Dpgm and DgalE mutants due to expression of closely related LPS core structures to tetrasaccharide 1. However, FA group sera did not recognize further truncated LPS core of DrfaK and DrfaF mutants. Overall binding to viable N. meningitidis by FA group sera was lower compared to that of the Alum group. Therefore, FA group sera were not further considered in the following studies with viable bacteria. Postimmune sera of the Alum group showed significant binding to N. meningitidis strains of serogroups A, B, C, W, Y, X, and E and a capsule null locus (cnl), a strain lacking the capsule, demonstrating that LPS core recognition is not influenced by capsule (Figure S2B). Taken together, the immunolabeling studies with postimmune sera of the Alum group raised with tetrasaccharide 1 revealed broad binding to various viable N. meningitidis strains. Binding of Anti-tetrasaccharide 1 Antibodies to LPS Inner Core of Closely Related Gram-Negative Bacteria Since most Gram-negative bacteria express LPSs that contain core structure motifs shared with tetrasaccharide 1, we determined the specificity of antibody recognition of postimmune sera to a panel of Gram-negative bacteria, while LPS-free Gram-positive bacteria served as negative controls (Figure 6B). No cross-reactivity was observed for closely related N. mucosa or other Gram-negative and Gram-positive bacteria. Significant cross-reactivity only occurred to N. lactamica. This finding is in

accordance with earlier reports wherein mAb B5 generated against DgalE LPS also cross-reacted with N. lactamica (Plested et al., 1999). It was shown that the LPS of N. lactamica contains antigens cross-reacting to meningococcal immunotypes L3, L7, and L9 (Braun et al., 2004). These results indicate the high specificity of the anti-tetrasaccharide 1 antibodies to N. meningitidis. Immunoprotective Effects of Anti-tetrasaccharide 1 Antibodies Besides their binding to the pathogen, functional antibodies are required to eliminate bacteria and protect from infection. Two in vitro assays are currently used to assess the functional activity of antibodies: the opsonophagocytic activity (OPA) (Ja¨kel et al., 2008; Plested et al., 2001) and SBA assays. The gold standard test to detect N. meningitidis lysis in the presence of complement is the SBA assay that correlates with immunoprotection (Plested et al., 2001). For the SBA measurement of N. meningitidis B, a human complement source, typically serum, should be used because increased bactericidal titers were observed in the presence of rabbit complement (Santos et al., 2001; Zollinger and Mandrell, 1983). In contrast, SBA of N. meningitidis serogroup C can be conducted with rabbit complement source. Thus, we performed our SBA assay with N. meningitidis serogroup C expressing homologous immunotype L3 wt (WUE2120) and the corresponding truncated Dpgm mutant (WUE2420). Despite the broad and specific recognition of N. meningitidis strains by serum antibodies, no serum-dependent bactericidal activity was detected (Figures S3A–S3C). The OPA measures antibody opsonization of N. meningitidis that triggers uptake into phagocytic cells such as neutrophils, resulting in protection against bacteremia. The postimmune sera of the Alum group showed a tendency toward higher OPA titers for MC58 wt (p = 0.08; Figure S3D).



Figure 5. Immunolabeling of Heat-Inactivated and Viable N. meningitidis with Tetrasaccharide 1-Specific Mouse Sera (A) Confocal fluorescence microscopy images of heat-inactivated N. meningitidis B strain DE8759 with pre- and postimmune sera of the FA group. Bacteriabound antibodies were detected with secondary anti-mouse IgG-FITC antibody (green). DNA staining with DAPI (blue) was used as the counterstain. Differential interference contrast (DIC) images are shown in the upper right. (B) Binding of postimmune sera of the Alum group and control group to viable N. meningitidis strains detected by an ELISA-based assay. One representative strain of each of the 12 immunotypes was immunolabeled. The N. meningitidis DlpxA mutant that lacks LPS was used as negative control. Bacteria-bound antibodies were detected with a secondary anti-mouse IgG-HRPO antibody. Fold changes in mean optical density 450 nm of bacteria incubated with pooled postimmune serum from the Alum group, with respect to control group, are shown as bars. Representative results from three independent experiments with mean of three replicates are shown. Error bars represent SEM. *p < 0.05, inferred by two-tailed unpaired t tests against the DlpxA strain.

Further analysis of tetrasaccharide 1-specific serum antibodies revealed IgG1 as the main isotype in Alum serum, followed by a small IgG2a portion (Figure S4A), in accordance with the fact that Alum is mainly a Th2-skewing adjuvant that primarily induces IgG1 antibodies in mice (Bungener et al., 2008). Isotypes IgG3 and IgG2a are described to promote SBA and OPA (Michaelsen et al., 2004). The absence of isotype IgG3 and the low amount of isotype IgG2a may explain the lack of SBA and low OPA. Improved formulation strategies using approved Th1-skewing adjuvants that force a class switch toward IgG2a and IgG3 may elicit protective antibodies against N. meningitidis infection (Baudner et al., 2009; Lin et al., 2004). Determination of Conserved Immunogenic Epitope of Anti-LPS Inner Core Antibodies Although we observed broad binding of the serum antibodies to viable N. meningitidis strains, the absence of functional antibodies suggests that presentation of the LPS core requires improvement. We aimed to determine the essential structural requirements of the LPS core antigen that may elicit protection using glycan array analysis and SPR. Postimmune sera of the Alum and FA groups showed similar binding preferences against 12 synthetic LPS core structures presented on a glycan array (Figures 7A and 7B). Three truncated versions of tetrasaccharide 1, compounds 2–4 were recognized, whereas no cross-reactivity for conserved LPS core of genus Chlamydia 12 was observed, demonstrating specific binding to mono-a-Kdo 4 and L-a-DHep-(1/5)-a-Kdo components. This observation explains the cross-reactivity to the related LPS cores of Y. pestis 8, H. influenzae 9, and Proteus 10, representing variants of structure 2, a conserved LPS core of many Gram-negative bacteria. In contrast, LPS core only consisting of Hep moieties such as di-Hep 6, mono-Hep 7, and tri-Hep 11 were not recognized by the serum antibodies, indicating a prominent role of the charged Kdo in antibody-glycan interactions. Unlike the natural human serum antibodies that bind preferentially to the distal trisaccha-

ride 5 (Figure 3), the mouse serum antibodies generated against tetrasaccharide 1 predominantly recognize the Kdo-containing structures (Figure 7B). Earlier reports on anti-LPS core mAbs showed that mouse germline antibodies harbor an inherited binding pocket specific for Kdo (Nguyen et al., 2003). In addition, we have observed a specific antibody response in an immunization study with the unconjugated tetrasaccharide 1 (Figure S5). Furthermore, an immunization approach with detoxified LPS conjugated to CRM197 showed that Kdo disaccharide-b-GlcN of the conjugate seems to be somewhat immunodominant (Cox et al., 2010a). In summary, glycan array studies identified Kdo as the immunodominant residue in antigen 1. Binding specificities of polyclonal sera may be due to contribution of multiple antibodies. To deduce antibody specificities at the molecular level, we generated tetrasaccharide 1-specific mAb from a mouse via hybridoma technique (Anish et al., 2013; Ko¨hler and Milstein, 1975). Two mAbs, termed 1A5 1G1 and 1B6 4E1, were purified to homogeneity (Figure S6A). In glycan array analysis, mAb 1A5 1G1 recognized all oligosaccharides interacting with polyclonal sera except trisaccharide 5 (Figures 7C and 7D). mAb 1A5 1G1 has a binding preference for Kdocontaining epitopes explaining the cross-reactivity to related LPS cores 8–10 and truncated LPS cores 2–4. mAb 1B6 4E1 predominantly binds to tetrasaccharide 1, and thereafter to trisaccharide 2, causing cross-reactivity to related LPS cores 8–10. To further elucidate binding profiles, we determined the binding affinities to the LPS cores 1–5 by SPR (Figures S6B–S6G). For mAb 1A5 1G1, dissociation constant (KD) values in the low micromolar range for LPS cores 1–4 were detected with decreasing affinity to truncated LPS cores 2–4 (Table S1). No detectable binding was seen for distal trisaccharide 5 (data not shown). Binding curves of mAb 1A5 1G1 reached steady state only for a-Kdo 4 (Figure S6E). SPR analysis confirmed Kdo as the predominant epitope component of mAb 1A5 1G1. For the interaction of mAb 1B6 4E1 with tetrasaccharide 1, KD values in the high micromolar range were observed (Figures S6F and S6G). No



Figure 6. Immunolabeling of Viable N. meningitidis LPS Core Mutants, and Gram-Negative and Gram-Positive Bacteria, with Tetrasaccharide 1-Specific Mouse Sera of the Alum Group (A) Binding of postimmune sera of the Alum group and control group to viable N. meningitidis (immunotype L3) MC58 mutants expressing truncated LPS measured by ELISA-based assay. (B) Binding of postimmune sera of the Alum and control groups to viable related Neisseria species, Gram-negative and Gram-positive bacteria measured by an ELISA-based assay. Detection of bound antibodies was performed with secondary anti-mouse IgG-HRPO antibody. Fold change in mean optical density 450 nm of the Alum group, with respect to control group, is plotted. Average of three independent experiments with mean of three replicates is shown. Error bars represent SEM. *p < 0.05, two-sided unpaired t tests against DlpxA.

concentration-dependent binding was observed for trisaccharide 2 and a-Kdo 4, disclosing selective binding of 1B6 4E1 to tetrasaccharide 1 (data not shown). The reactivity of these

mAbs suggests that immunization with tetrasaccharide 1 elicited antibodies strongly binding to Kdo-containing structures with insignificant binding to the crucial distal trisaccharide 5. In



Figure 7. Specificity of Mouse Sera of the Alum and FA Groups and of Tetrasaccharide 1-Specific mAbs 1A5 1G1 and 1B6 4E1 toward 12 Synthetic LPS Core Structures Inferred by Glycan Array (A–D) Representative array scan of sera from the Alum group (A, left) of mAbs 1A5 1G1 (C, left) and 1B6 4E1 (C, right) and the printing pattern (A, right) are depicted. Quantification of mean fluorescence intensity of postimmune sera (B) and mAbs (D) are also shown. Error bars represent SD of two spots of two separate arrays, respectively.

contrast to the postimmune sera, both mAbs did not recognize the LPS core on viable N. meningitidis strains. The broad cross-reactivity to synthetic LPS cores observed on the glycan arrays may be caused by the presentation of Kdo residues via their pentyl amino linker on the array surface. On the contrary, accessibility to Kdo may be restricted on natural bacterial surfaces due to close proximity to the outer membrane. Moreover, postimmune serum antibodies bound to compound 5, whereas the mAbs did not (Figure 7). Thus, exposure of distal trisaccharide 5 on the surface of N. meningitidis most likely permits binding of postimmune sera to the LPS core. In another immunization approach, BSA conjugates with different antigen loadings of the synthetic oligosaccharides from Vibrio cholerae O1 Inaba Opolysaccharide elicited a comparable level of antibodies with limited protection (Meeks et al., 2004). To further refine the structural requirements for LPS recognition by antibodies, we tested mAb B5 raised against LPS of DgalE mutant (a kind gift from Dr. A.D. Cox). mAb B5 showed SBA and reacted against several N. meningitidis strains expressing LPS with PEtN at 3-position of HepII (Plested et al., 1999, 2003). In a glycan array screening, mAb B5 did not bind to any synthetic LPS core, probably due to the absence of a PEtN modification (Figure S7A). To further delineate the immunodominance of Kdo, we tested the binding of our antibodies against isolated and delipidated DgalE LPS that was immobilized on an array slide via an adipic acid dihydrazide (ADH) spacer. This immobilization destroyed the a-Kdo motif due to derivatization at its

anomeric position by hydrazide group with predominant b-configuration (Lee and Shin, 2005). mAb B5 bound to immobilized DgalE LPS core, whereas our mAbs did not (Figures S7B and S7C). mAb B5 binds to the LPS core via the trisaccharide 5 derivative with an additional PEtN on HepII and Glc on HepI (Plested et al., 1999), independent of the a-Kdo motif. While postimmune sera of the FA group showed weak reactivity to immobilized DgalE LPS core compared to mAb B5, postimmune sera from the Alum group did not show any binding to DgalE LPS core (Figures S7B and S7C). This effect may be explained by higher avidity of tetrasaccharide 1-specific antibodies raised by the Th1-skewing FA compared to Th2-skewing Alum (Figures S4B and S4C). Accordingly, higher signal intensities were gained for FA sera compared to Alum in glycan array screening on synthetic LPS core structures (Figures 7A and 7B). Based on glycan array screening, SPR, and bacterial immunolabeling, we infer that the trisaccharide motif represented by 5 is the crucial component of an LPS core antigen against N. meningitidis. This distal LPS core 5 was recognized by postimmune sera produced against synthetic tetrasaccharide 1, but antibody levels were low compared to that raised against Kdocontaining structures 1–4 and 8–10. Moreover, human sera predominantly recognize structure 5 (Figure 3). Furthermore, bactericidal mAb B5 bound only weakly to the N. meningitidis DicsB mutant expressing tetrasaccharide 1 derivative with an additional PEtN on HepII (identical phenotype to Dpgm) (Plested et al., 1999), substantiating the hypothesis that distal parts of the LPS



core are important in eliciting SBA. Due to the predominant reactivity of mAb B5 against PEtN modification on HepII, it only bound to 76% of N. meningitidis B strains expressing this specific PEtN but not to strains with alternative modifications (Plested et al., 1999). Avoidance of LPS core variants such as PEtN modification and Glc residue on HepII may recruit more broadly reactive antibodies. Thus, by leaving out Kdo, the highly conserved LPS core trisaccharide 5 might already be sufficient to raise a broadly protective immune response. Furthermore, via addition of conserved Glc to HepI in 5, an enlarged, branched, and more surface-exposed LPS core a-D-GlcNAc-(1/2)-L-a-D-Hep-(1/3)[b-D-Glc-(1/4)]-L-a-D-Hep is formed that may elicit both broad cross-reactivity to N. meningitidis and bactericidal activity. This assumption is substantiated by the broadly cross-protective mAb WN1 222-5, which binds to the LPS inner cores of E. coli and Salmonella. The 4-phosphate groups on the branched HepII and the HepIII side chains were identified as the minimal structural determinants for high-affinity binding to these LPS inner core structures. This finding demonstrates the relevance of branched structures (Mu¨ller-Loennies et al., 2007). Currently, efforts are underway to redesign the synthetic antigen to generate adequate antibody levels to distal parts of the N. meningitidis LPS inner core with functional activity. Here, we have shown the power of synthetic LPS core oligosaccharides in defining a universal antigen candidate for N. meningitidis. A T cell-dependent antibody response specific to conserved LPS core 1 was elicited in mice. The immunodominant a-Kdo residue accounted for the lack of functional antibodies. In contrast, the distal trisaccharide 5 mediates specific recognition of the natural LPS core on N. meningitidis. Based on our findings, an improved synthetic LPS core compared to compound 5 is currently being explored as a broadly protective antigen against N. meningitidis. SIGNIFICANCE Designing a synthetic carbohydrate antigen candidate involves the choice of the optimal structure to confer protection. Today, antigen identification is an iterative process involving trial-and-error experiments. Here, we report progress in this process using a highly conserved LPS core structure of N. meningitidis as an example. Antigenicity and immunogenicity of synthetic LPS core oligosaccharides were evaluated using glycan arrays, in vivo immunogenicity studies, SPR, and in vitro surrogate assays for functional bacteria-killing antibodies. In combination with the epitope mapping of earlier described protective mAb B5, structural information of a potential, broadly reactive LPS core epitope was inferred. Glycan array screening of patient sera revealed antibodies specific to different LPS inner core structures. Based on these screenings, a tetrasaccharide was selected for immunological evaluation in mice and, therefore, conjugated to carrier protein CRM197. This glycoconjugate yielded LPS core-specific robust IgG and memory responses. Serum antibodies bound to numerous viable N. meningitidis strains. The immunodominant Kdo moiety contained in this tetrasaccharide elicited highly affine antibodies against Kdo-containing oligosaccharide but failed to bind to N. meningitidis bacteria. We identified the distal part of the LPS core as the crucial unit to recruit antibodies that are

broadly reactive to N. meningitidis expressing diverse LPS immunotypes. Based on our results, we propose an improved, highly conserved LPS core tetrasaccharide a-DGlcNAc-(1/2)-L-a-D-Hep-(1/3)-[b-D-Glc-(1/4)]-L-a-DHep to generate antibodies that may unify broad reactivity and bactericidal activity against N. meningitidis. EXPERIMENTAL PROCEDURES Chemical Synthesis A detailed description of experimental procedures and characterization of synthetic compounds are available in Supplemental Experimental Procedures. Human Sera Convalescent sera used have been published recently (Hubert et al., 2013). Sera from healthy individuals carrying or not carrying N. meningitidis were obtained from human volunteers, with permission of the Ethics Committee of the Medical Faculty of the Julius-Maximilians-University. Ethics Statement Animal experiments were performed in accordance with the German regulations of the Society for Laboratory Animal Science and the European Health Law of the Federation of Laboratory Animal Science Associations. All efforts were made to minimize the suffering of the animals. Immunization Studies and Evaluation of Immune Response Using SPR and Microarray Glycoconjugate of CRM197 and amino-linked antigen was produced, characterized, and immunized in BALB/c mice in a dose- and formulation-dependent manner (Martin et al., 2013). Splenocytes of a mouse with the highest antibody titer (FA group) were used to generate mAbs via the hybridoma technique (Anish et al., 2013; Ko¨hler and Milstein, 1975). Specificities of serum antibodies and purified mAbs were analyzed by glycan array screening. A glycan array of synthetic compounds was prepared, and antibody samples were incubated as described elsewhere (Martin et al., 2013). For immobilization of purified glycans, slides were modified as reported elsewhere (Lee and Shin, 2005). Affinity of direct immobilized mAbs to solubilized glycans and binding of mouse sera to immobilized glycans was analyzed by SPR. Immunolabeling Studies of Bacteria and Functional Activity Test Staining of bacteria with mAb and polyclonal sera was performed as described elsewhere (Anish et al., 2013). Antibodies bound to viable bacteria were detected via secondary goat anti-mouse IgG (H+L)-HRPO antibody, and readout was accomplished at 450 nm in a microplate reader. Antibodies bound to inactivated bacteria were detected by appropriated fluorescentlabeled secondary antibody and measured by flow cytometry or confocal fluorescence microscopy. SBA (Elias et al., 2013) and OPA (Hazenbos et al., 1994; Rodrı´guez et al., 2001) of sera were tested as described elsewhere. For a detailed description, please refer to the Supplemental Experimental Procedures. SUPPLEMENTAL INFORMATION Supplemental Information including Supplemental Experimental Procedures, seven figures, and one table and can be found with this article online at http://dx.doi.org/10.1016/j.chembiol.2014.11.016. AUTHOR CONTRIBUTIONS A.R. performed animal studies, generation of mAbs, immunolabeling, and functional activity assays. Y.Y. synthesized LPS core structures with the help of C.L.P., who also initiated the LPS core screening project. A.D.C. provided mAb B5 and detoxified LPS core structure. H.C. performed mutagenesis, bacterial culture, and initial immunolabeling experiments with the help of U.V. C.A. and P.H.S. designed and initiated this study. A.R., C.A., and P.H.S. wrote the manuscript with input from all authors.



ACKNOWLEDGMENTS We thank the Max Planck Society and the Ko¨rber Foundation for generous financial support (Ko¨rber Prize to P.H.S.) and the Studienstiftung des Deutschen Volkes for a graduate fellowship (A.R.). C.A. acknowledges support by German Federal Ministry of Education and Research (grant 0315447). We acknowledge Pfeńex, Inc. (Reagent proteins) for providing CRM197 at a discounted price. We acknowledge Mr. A. Geissner and Mr. F. Broecker for valuable discussions and careful review of the manuscript. We acknowledge the advice of Dr. B. Lepenies on institutional animal ethics policy. Received: July 30, 2014 Revised: November 11, 2014 Accepted: November 16, 2014 Published: January 15, 2015 REFERENCES Adamo, R., Romano, M.R., Berti, F., Leuzzi, R., Tontini, M., Danieli, E., Cappelletti, E., Cakici, O.S., Swennen, E., Pinto, V., et al. (2012). Phosphorylation of the synthetic hexasaccharide repeating unit is essential for the induction of antibodies to Clostridium difficile PSII cell wall polysaccharide. ACS Chem. Biol. 7, 1420–1428. Anish, C., Guo, X., Wahlbrink, A., and Seeberger, P.H. (2013). Plague detection by anti-carbohydrate antibodies. Angew. Chem. Int. Ed. Engl. 52, 9524–9528. Baudner, B.C., Ronconi, V., Casini, D., Tortoli, M., Kazzaz, J., Singh, M., Hawkins, L.D., Wack, A., and O’Hagan, D.T. (2009). MF59 emulsion is an effective delivery system for a synthetic TLR4 agonist (E6020). Pharm. Res. 26, 1477–1485. Berrington, A.W., Tan, Y.C., Srikhanta, Y., Kuipers, B., van der Ley, P., Peak, I.R., and Jennings, M.P. (2002). Phase variation in meningococcal lipooligosaccharide biosynthesis genes. FEMS Immunol. Med. Microbiol. 34, 267–275. Braun, J.M., Beuth, J., Blackwell, C.C., Giersen, S., Higgins, P.G., Tzanakaki, G., Unverhau, H., and Weir, D.M. (2004). Neisseria meningitidis, Neisseria lactamica and Moraxella catarrhalis share cross-reactive carbohydrate antigens. Vaccine 22, 898–908. Bro¨ker, M., Dull, P.M., Rappuoli, R., and Costantino, P. (2009). Chemistry of a new investigational quadrivalent meningococcal conjugate vaccine that is immunogenic at all ages. Vaccine 27, 5574–5580. Bro¨ker, M., Cooper, B., Detora, L.M., and Stoddard, J.J. (2011). Critical appraisal of a quadrivalent CRM(197) conjugate vaccine against meningococcal serogroups A, C W-135 and Y (Menveo) in the context of treatment and prevention of invasive disease. Infect. Drug Resist. 4, 137–147. Bungener, L., Geeraedts, F., Ter Veer, W., Medema, J., Wilschut, J., and Huckriede, A. (2008). Alum boosts TH2-type antibody responses to whole-inactivated virus influenza vaccine in mice but does not confer superior protection. Vaccine 26, 2350–2359. Caugant, D.A., Tzanakaki, G., and Kriz, P. (2007). Lessons from meningococcal carriage studies. FEMS Microbiol. Rev. 31, 52–63. Choudhury, B., Kahler, C.M., Datta, A., Stephens, D.S., and Carlson, R.W. (2008). The structure of the L9 immunotype lipooligosaccharide from Neisseria meningitidis NMA Z2491. Carbohydr. Res. 343, 2971–2979. Clements, C.J., and Griffiths, E. (2002). The global impact of vaccines containing aluminium adjuvants. Vaccine 20 (Suppl 3 ), S24–S33. Cox, A.D., St Michael, F., Neelamegan, D., Lacelle, S., Cairns, C.M., Giuliani, M.M., Biolchi, A., Hoe, J.C., Moxon, E.R., and Richards, J.C. (2010a). Investigating the candidacy of LPS-based glycoconjugates to prevent invasive meningococcal disease: immunology of glycoconjugates with high carbohydrate loading. Glycoconj. J. 27, 643–648. Cox, A.D., St Michael, F., Neelamegan, D., Lacelle, S., Cairns, C., and Richards, J.C. (2010b). Investigating the candidacy of LPS-based glycoconjugates to prevent invasive meningococcal disease: chemical strategies to prepare glycoconjugates with good carbohydrate loading. Glycoconj. J. 27, 401–417.

Elias, J., Findlow, J., Borrow, R., Tremmel, A., Frosch, M., and Vogel, U. (2013). Persistence of antibodies in laboratory staff immunized with quadrivalent meningococcal polysaccharide vaccine. J. Occup. Med. Toxicol. 8, 1–49. Evenberg, D., Hoogerhout, P., van Boeckel, C.A.A., Rijkers, G.T., Beuvery, E.C., van Boom, J.H., and Poolman, J.T. (1992). Preparation, antigenicity, and immunogenicity of synthetic ribosylribitol phosphate oligomer-protein conjugates and their potential use for vaccination against Haemophilus influenzae type b disease. J. Infect. Dis. 165 (Suppl 1 ), S152–S155. Finne, J., Leinonen, M., and Ma¨kela¨, P.H. (1983). Antigenic similarities between brain components and bacteria causing meningitis. Implications for vaccine development and pathogenesis. Lancet 2, 355–357. Frosi, G., Biolchi, A., Lo Sapio, M., Rigat, F., Gilchrist, S., Lucidarme, J., Findlow, J., Borrow, R., Pizza, M., Giuliani, M.M., and Medini, D. (2013). Bactericidal antibody against a representative epidemiological meningococcal serogroup B panel confirms that MATS underestimates 4CMenB vaccine strain coverage. Vaccine 31, 4968–4974. Fu, Y., Baumann, M., Kosma, P., Brade, L., and Brade, H. (1992). A synthetic glycoconjugate representing the genus-specific epitope of chlamydial lipopolysaccharide exhibits the same specificity as its natural counterpart. Infect. Immun. 60, 1314–1321. Granoff, D.M. (2010). Review of meningococcal group B vaccines. Clin. Infect. Dis. 50 (Suppl 2 ), S54–S65. Guo, X. (2011). Synthetic bacterial lipopolysaccharide core structures as vaccine candidates against Clamydia trachomatis and Yersinia pestis. PhD thesis (Zurich, Switzerland: ETH Zurich). Hazenbos, W.L., van den Berg, B.M., van’t Wout, J.W., Mooi, F.R., and van Furth, R. (1994). Virulence factors determine attachment and ingestion of nonopsonized and opsonized Bordetella pertussis by human monocytes. Infect. Immun. 62, 4818–4824. Hedari, C.P., Khinkarly, R.W., and Dbaibo, G.S. (2014). Meningococcal serogroups A, C, W-135, and Y tetanus toxoid conjugate vaccine: a new conjugate vaccine against invasive meningococcal disease. Infect. Drug Resist. 7, 85–99. Hong, E., Giuliani, M.M., Deghmane, A.E., Comanducci, M., Brunelli, B., Dull, P., Pizza, M., and Taha, M.K. (2013). Could the multicomponent meningococcal serogroup B vaccine (4CMenB) control Neisseria meningitidis capsular group X outbreaks in Africa? Vaccine 31, 1113–1116. Hubert, K., Devos, N., Mordhorst, I., Tans, C., Baudoux, G., Feron, C., Goraj, K., Tommassen, J., Vogel, U., Poolman, J.T., and Weynants, V. (2013). ZnuD, a potential candidate for a simple and universal Neisseria meningitidis vaccine. Infect. Immun. 81, 1915–1927. Ja¨kel, A., Plested, J.S., Hoe, J.C., Makepeace, K., Gidney, M.-A.J., Lacelle, S., St Michael, F., Cox, A.D., Richards, J.C., and Moxon, E.R. (2008). Naturallyoccurring human serum antibodies to inner core lipopolysaccharide epitopes of Neisseria meningitidis protect against invasive meningococcal disease caused by isolates displaying homologous inner core structures. Vaccine 26, 6655–6663. Jennings, M.P., Bisercic, M., Dunn, K.L.R., Virji, M., Martin, A., Wilks, K.E., Richards, J.C., and Moxon, E.R. (1995). Cloning and molecular analysis of the Isi1 (rfaF) gene of Neisseria meningitidis which encodes a heptosyl-2-transferase involved in LPS biosynthesis: evaluation of surface exposed carbohydrates in LPS mediated toxicity for human endothelial cells. Microb. Pathog. 19, 391–407. Kahler, C.M., Datta, A., Tzeng, Y.L., Carlson, R.W., and Stephens, D.S. (2005). Inner core assembly and structure of the lipooligosaccharide of Neisseria meningitidis: capacity of strain NMB to express all known immunotype epitopes. Glycobiology 15, 409–419. Ko¨hler, G., and Milstein, C. (1975). Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 256, 495–497. Kurzai, O., Schmitt, C., Claus, H., Vogel, U., Frosch, M., and Kolb-Maürer, A. (2005). Carbohydrate composition of meningococcal lipopolysaccharide modulates the interaction of Neisseria meningitidis with human dendritic cells. Cell. Microbiol. 7, 1319–1334.



Lee, M.R., and Shin, I. (2005). Facile preparation of carbohydrate microarrays by site-specific, covalent immobilization of unmodified carbohydrates on hydrazide-coated glass slides. Org. Lett. 7, 4269–4272. Lin, L., Gerth, A.J., and Peng, S.L. (2004). CpG DNA redirects class-switching towards ‘‘Th1-like’’ Ig isotype production via TLR9 and MyD88. Eur. J. Immunol. 34, 1483–1487. Martin, C.E., Broecker, F., Oberli, M.A., Komor, J., Mattner, J., Anish, C., and Seeberger, P.H. (2013). Immunological evaluation of a synthetic Clostridium difficile oligosaccharide conjugate vaccine candidate and identification of a minimal epitope. J. Am. Chem. Soc. 135, 9713–9722. Meeks, M.D., Saksena, R., Ma, X., Wade, T.K., Taylor, R.K., Kovać, P., and Wade, W.F. (2004). Synthetic fragments of Vibrio cholerae O1 Inaba O-specific polysaccharide bound to a protein carrier are immunogenic in mice but do not induce protective antibodies. Infect. Immun. 72, 4090–4101. Michaelsen, T.E., Kolberg, J., Aase, A., Herstad, T.K., and Høiby, E.A. (2004). The four mouse IgG isotypes differ extensively in bactericidal and opsonophagocytic activity when reacting with the P1.16 epitope on the outer membrane PorA protein of Neisseria meningitidis. Scand. J. Immunol. 59, 34–39. Mistretta, N., Seguin, D., Thiebaud, J., Vialle, S., Blanc, F., Brossaud, M., Talaga, P., Norheim, G., Moreau, M., and Rokbi, B. (2010). Genetic and structural characterization of L11 lipooligosaccharide from Neisseria meningitidis serogroup A strains. J. Biol. Chem. 285, 19874–19883. Mong, T.K.-K., Lee, H.-K., Duroń, S.G., and Wong, C.-H. (2003). Reactivitybased one-pot total synthesis of fucose GM1 oligosaccharide: a sialylated antigenic epitope of small-cell lung cancer. Proc. Natl. Acad. Sci. USA 100, 797–802. Mu¨ller-Loennies, S., Brade, L., and Brade, H. (2007). Neutralizing and crossreactive antibodies against enterobacterial lipopolysaccharide. Int. J. Med. Microbiol. 297, 321–340. Nguyen, H.P., Seto, N.O.L., MacKenzie, C.R., Brade, L., Kosma, P., Brade, H., and Evans, S.V. (2003). Germline antibody recognition of distinct carbohydrate epitopes. Nat. Struct. Biol. 10, 1019–1025. Nikaido, H. (2003). Molecular basis of bacterial outer membrane permeability revisited. Microbiol. Mol. Biol. Rev. 67, 593–656. Noti, C., de Paz, J.L., Polito, L., and Seeberger, P.H. (2006). Preparation and use of microarrays containing synthetic heparin oligosaccharides for the rapid analysis of heparin-protein interactions. Chemistry 12, 8664–8686. Ohara, T., Adibekian, A., Esposito, D., Stallforth, P., and Seeberger, P.H. (2010). Towards the synthesis of a Yersinia pestis cell wall polysaccharide: enantioselective synthesis of an L-glycero-D-manno-heptose building block. Chem. Commun. (Camb.) 46, 4106–4108. _ Palusiak, A., Maciejewska, A., qugowski, C., and Ro´zalski, A. (2014). The amide of galacturonic acid with lysine as an immunodominant component of the lipopolysaccharide core region from Proteus penneri 42 strain. Acta Biochim. Pol. 61, 129–132. Plested, J.S., Makepeace, K., Jennings, M.P., Gidney, M.A.J., Lacelle, S., Brisson, J., Cox, A.D., Martin, A., Bird, A.G., Tang, C.M., et al. (1999). Conservation and accessibility of an inner core lipopolysaccharide epitope of Neisseria meningitidis. Infect. Immun. 67, 5417–5426. Plested, J.S., Gidney, M.A.J., Coull, P.A., Griffiths, H.G., Herbert, M.A., Bird, A.G., Richards, J.C., and Moxon, E.R. (2000). Enzyme linked immunosorbent assay (ELISA) for the detection of serum antibodies to the inner core lipopolysaccharide of Neisseria meningitidis group B. J. Immunol. Methods 237, 73–84. Plested, J.S., Ferry, B.L., Coull, P.A., Makepeace, K., Lehmann, A.K., MacKinnon, F.G., Griffiths, H.G., Herbert, M.A., Richards, J.C., and Moxon, E.R. (2001). Functional opsonic activity of human serum antibodies to inner core lipopolysaccharide (galE) of serogroup B meningococci measured by flow cytometry. Infect. Immun. 69, 3203–3213.

Plested, J.S., Harris, S.L., Wright, J.C., Coull, P.A., Makepeace, K., Gidney, M.-A.J., Brisson, J.-R., Richards, J.C., Granoff, D.M., and Moxon, E.R. (2003). Highly conserved Neisseria meningitidis inner-core lipopolysaccharide epitope confers protection against experimental meningococcal bacteremia. J. Infect. Dis. 187, 1223–1234. Pozsgay, V., Chu, C., Pannell, L., Wolfe, J., Robbins, J.B., and Schneerson, R. (1999). Protein conjugates of synthetic saccharides elicit higher levels of serum IgG lipopolysaccharide antibodies in mice than do those of the O-specific polysaccharide from Shigella dysenteriae type 1. Proc. Natl. Acad. Sci. USA 96, 5194–5197. Raetz, C.R., and Whitfield, C. (2002). Lipopolysaccharide endotoxins. Annu. Rev. Biochem. 71, 635–700. Rahman, M.M., Kahler, C.M., Stephens, D.S., and Carlson, R.W. (2001). The structure of the lipooligosaccharide (LOS) from the a-1,2-N-acetyl glucosamine transferase (rfaK(NMB)) mutant strain CMK1 of Neisseria meningitidis: implications for LOS inner core assembly and LOS-based vaccines. Glycobiology 11, 703–709. Rodrı´guez, M.E., Van der Pol, W.L., and Van de Winkel, J.G.J. (2001). Flow cytometry-based phagocytosis assay for sensitive detection of opsonic activity of pneumococcal capsular polysaccharide antibodies in human sera. J. Immunol. Methods 252, 33–44. Santos, G.F., Deck, R.R., Donnelly, J., Blackwelder, W., and Granoff, D.M. (2001). Importance of complement source in measuring meningococcal bactericidal titers. Clin. Diagn. Lab. Immunol. 8, 616–623. St Michael, F., Vinogradov, E., Wenzel, C.Q., McIntosh, B., Li, J., Hoe, J.C., Richards, J.C., and Cox, A.D. (2009). Phosphoethanolamine is located at the 6-position and not at the 7-position of the distal heptose residue in the lipopolysaccharide from Neisseria meningitidis. Glycobiology 19, 1436–1445. St Michael, F., Cairns, C.M., Filion, A.L., Biolchi, A., Brunelli, B., Giuliani, M., Richards, J.C., and Cox, A.D. (2014). Investigating the candidacy of LPSbased glycoconjugates to prevent invasive meningococcal disease: conjugates based on core oligosaccharides. Glycoconj. J. 31, 25–39. Verheul, A.F., Kuipers, A.J., Braat, A.K., Dekker, H.A., Peeters, C.C., Snippe, H., and Poolman, J.T. (1994). Development, characterization, and biological properties of meningococcal immunotype L3,7,(8),9-specific monoclonal antibodies. Clin. Diagn. Lab. Immunol. 1, 729–736. Weynants, V., Denoe¨l, P., Devos, N., Janssens, D., Feron, C., Goraj, K., Momin, P., Monnom, D., Tans, C., Vandercammen, A., et al. (2009). Genetically modified L3,7 and L2 lipooligosaccharides from Neisseria meningitidis serogroup B confer a broad cross-bactericidal response. Infect. Immun. 77, 2084–2093. Yang, Y., Martin, C.E., and Seeberger, P.H. (2012). Total synthesis of the core tetrasaccharide of Neisseria meningitidis lipopolysaccharide, a potential vaccine candidate for meningococcal diseases. Chem. Sci. 3, 896–899. Yang, Y., Oishi, S., Martin, C.E., and Seeberger, P.H. (2013). Diversity-oriented synthesis of inner core oligosaccharides of the lipopolysaccharide of pathogenic Gram-negative bacteria. J. Am. Chem. Soc. 135, 6262–6271. Yu, B., and Tao, H. (2001). Glycosyl trifluoroacetimidates. Part 1: Preparation and application as new glycosyl donors. Tetrahedron Lett. 42, 2405–2407. Yu, B., and Tao, H. (2002). Glycosyl trifluoroacetimidates. 2. Synthesis of dioscin and xiebai saponin I. J. Org. Chem. 67, 9099–9102. Zhu, P., Tsai, C.M., and Frasch, C.E. (2002). Immunologic and genetic characterization of lipooligosaccharide variants in a Neisseria meningitidis serogroup C strain. FEMS Immunol. Med. Microbiol. 34, 193–200. Zollinger, W.D., and Mandrell, R.E. (1983). Importance of complement source in bactericidal activity of human antibody and murine monoclonal antibody to meningococcal group B polysaccharide. Infect. Immun. 40, 257–264.


Meningitis caused by a lipopolysaccharide deficient Neisseria meningitidis.

Lipopolysaccharide serotyping of Neisseria meningitidis by hemagglutination inhibition.

Toxicity and immunogenicity of Neisseria meningitidis lipopolysaccharide incorporated into liposomes.

Outer-membrane protein and lipopolysaccharide serotyping of Neisseria meningitidis by inhibition of a solid-phase radioimmunoassay.

Identification of some basic extractable compounds produced by Neisseria gonorrhoeae and Neisseria meningitidis in a defined medium.

Endophthalmitis caused by Neisseria meningitidis.

Characterization of the virulence and antigenic structure of Salmonella typhimurium strains with lipopolysaccharide core defects.

Immunogenicity and antigenic heterogeneity of a human transferrin-binding protein in Neisseria meningitidis.

Common antigenic domains in transferrin-binding protein 2 of Neisseria meningitidis, Neisseria gonorrhoeae, and Haemophilus influenzae type b.

Electron capture gas-liquid chromatographic-mass spectral identification of acids produced by Neisseria meningitidis in a defined medium.

Neisseria meningitidis: a cause of nosocomial pneumonia.

Conserved outer membrane protein of Neisseria meningitidis involved in capsule expression.

Neisseria meningitidis: pathogenesis and immunity.

[A hemorrhagic stroke revealing Neisseria meningitidis meningitis].

Potential impact of vaccination against Neisseria meningitidis on Neisseria gonorrhoeae in the United States: results from a decision-analysis model.

Penicillin resistance in Neisseria meningitidis.

Frequency of simultaneous carriage of Neisseria gonorrhoeae and Neisseria meningitidis.

The structure of a contact-dependent growth-inhibition (CDI) immunity protein from Neisseria meningitidis MC58.

Genetic diversity of penicillin-resistant Neisseria meningitidis.

Rapid identification of Neisseria gonorrhoeae and Neisseria meningitidis by using enzymatic profiles.

Bacteraemic pneumonia caused by Neisseria meningitidis serogroup Y.

Regulation of capsule in Neisseria meningitidis.

Characterization of the Neisseria meningitidis Helicase RecG.

Structure and mechanism of a molecular rheostat, an RNA thermometer that modulates immune evasion by Neisseria meningitidis.