Search tips
Search criteria 


Logo of glycobLink to Publisher's site
Glycobiology. 2009 June; 19(6): 665–673.
Published online 2009 March 6. doi:  10.1093/glycob/cwp036
PMCID: PMC2682610

Secondary cell wall polysaccharides of Bacillus anthracis are antigens that contain specific epitopes which cross-react with three pathogenic Bacillus cereus strains that caused severe disease, and other epitopes common to all the Bacillus cereus strains tested


The immunoreactivities of hydrogen fluoride (HF)-released cell wall polysaccharides (HF-PSs) from selected Bacillus anthracis and Bacillus cereus strains were compared using antisera against live and killed B. anthracis spores. These antisera bound to the HF-PSs from B. anthracis and from three clinical B. cereus isolates (G9241, 03BB87, and 03BB102) obtained from cases of severe or fatal human pneumonia but did not bind to the HF-PSs from the closely related B. cereus ATCC 10987 or from B. cereus type strain ATCC 14579. Antiserum against a keyhole limpet hemocyanin conjugate of the B. anthracis HF-PS (HF-PS-KLH) also bound to HF-PSs and cell walls from B. anthracis and the three clinical B. cereus isolates, and B. anthracis spores. These results indicate that the B. anthracis HF-PS is an antigen in both B. anthracis cell walls and spores, and that it shares cross-reactive, and possibly pathogenicity-related, epitopes with three clinical B. cereus isolates that caused severe disease. The anti-HF-PS-KLH antiserum cross-reacted with the bovine serum albumin (BSA)-conjugates of all B. anthracis and all B. cereus HF-PSs tested, including those from nonclinical B. cereus ATCC 10987 and ATCC 14579 strains. Finally, the serum of vaccinated (anthrax vaccine adsorbed (AVA)) Rhesus macaques that survived inhalation anthrax contained IgG antibodies that bound the B. anthracis HF-PS-KLH conjugate. These data indicate that HF-PSs from the cell walls of the bacilli tested here are (i) antigens that contain (ii) a potentially virulence-associated carbohydrate antigen motif, and (iii) another antigenic determinant that is common to B. cereus strains.

Keywords: antigens, Bacillus anthracis, Bacillus cereus, polysaccharides, specificity


Anthrax is primarily a disease of herbivores although humans can also be infected. The etiologic agent of anthrax is Bacillus anthracis. Systemic anthrax, secondary to any of its associated routes of entry – cutaneous, gastrointestinal, and inhalation – is, if untreated, potentially fatal. The potential for using B. anthracis as a weapon has been widely reported (Hilleman 2002; Baillie 2005). Since the anthrax bioterrorism events in 2001, there has been a renewed interest in effective diagnostic tools and medical countermeasures. The carbohydrate antigens of B. anthracis have not been extensively investigated.

In general, Gram-positive bacteria have a cell surface comprising several classes of polysaccharides. These include teichoic acid, a polysaccharide that consists of repetitive sugar–phosphate residues that can have noncarbohydrate substituents such as d-alanine; lipoteichoic acid, a type of teichoic acid that is anchored to the membrane via a glycolipid; and teichuronic acid; a polymer similar to teichoic acid except that one or more of the saccharide moieties consist of glycuronosyl residues. Gram-positive bacteria also often contain additional neutral and acidic polysaccharides in their cell walls, including polysaccharide capsules that are not teichoic acid, lipoteichoic acid, or teichuronic acid. Many of these polysaccharides, as well as teichoic acid and teichuronic acid, are linked to the cell wall peptidoglycan (PG). In an attempt to classify the polysaccharides in Gram-positive bacteria, Schaffer and Messner (2005) termed polysaccharides of the teichoic acid and teichuronic acid type as “classical” secondary cell wall polymers (SCWPs), while the others were grouped as “nonclassical” SCWPs.

Our objective is to determine whether carbohydrates either on B. anthracis spores or on vegetative cells are antigenic and have structural or immunochemical properties that may make them suitable for the development of improved diagnostic methods and new or improved vaccines. Recently, two B. anthracis carbohydrate antigens have been identified that show this potential (Daubenspeck et al. 2004; Choudhury et al. 2006; Mehta et al. 2006). One of these carbohydrates is an oligosaccharide that is part of the collagen-like protein, BclA, on the spore exosporium (Daubenspeck et al. 2004; Mehta et al. 2006), and the second is a nonclassical secondary cell wall polysaccharide found in the vegetative cell wall (Choudhury et al. 2006).

The research described in this report focuses on the secondary cell wall polysaccharide that is released from the B. anthracis cell wall by aqueous hydrogen fluoride (HF-PS). For B. anthracis, it was shown that the HF-PS anchors cell surface proteins, such as S-layer proteins, to the peptidoglycan (Mesnage et al. 2000). It is thought that the HF-PS is the ligand for the carbohydrate-binding SLH domain of the surface protein while a HF-labile phosphate bond anchors the PS to the peptidoglycan (Mesnage et al. 2000). A recent report identified 23 B. anthracis genes that encode proteins with SLH domains and, further, demonstrated that one of these genes, bslA, is present on the pXO1 pathogenicity island and that its product is necessary for adherence of B. anthracis to host cells (Kern and Schneewind 2008). We have previously shown, by examining the cell walls of B. anthracis and related Bacillus cereus strains, that B. anthracis produces a specific HF-PS structure that is identical in the investigated B. anthracis strains, i.e. Ames, Sterne, and Pasteur, but different from that of B. cereus cell walls (Choudhury et al. 2006; Leoff, Choudhury, et al. 2008, Leoff, Saile, et al. 2008). As shown in Figure Figure1,1, our structural investigations showed that the B. anthracis HF-PS comprises an amino sugar backbone of →6)-α-GlcNAc-(1→4)-β-ManNAc-(1→4)-β-GlcNAc-(1→ in which the α-GlcNAc residue is substituted with α-Gal and β-Gal at O3 and O4, respectively, and the β-GlcNAc substituted with α-Gal at O3 (Choudhury et al. 2006). In comparison, the HF-PS from the closely related B. cereus ATCC 10987 consists of a →6)-α-GalNAc-(1→4)-β-ManNAc-(1→4)-β-GlcNAc-(1→ backbone in which the α-GalNAc is substituted at O3 with a β-Gal residue and the β-ManNAc is acetylated at O3 (Leoff, Choudhury, et al. 2008). To date, our structural investigations into the B. cereus HF-PSs from B. cereus ATCC 10987 and from the more distantly related B. cereus type strain ATCC 14579 revealed a common structural theme (see Figure Figure1)1) consisting of a HexNAc-ManNAc-GlcNAc backbone that is substituted with terminal galactosyl (Gal) or glucosyl (Glc) residues or noncarbohydrate substituents such as acetyl groups (Leoff, Choudhury, et al. 2008).

Fig. 1
The repeating unit structure of the HF-PS from B. anthracis Sterne, Pasteur, and Ames as previously reported (top) and the consensus structure that is indicated for the HF-PS from members of the B. cereus group (bottom). This consensus repeating unit ...

The presence of strain-specific structural features as well as a general common structural theme in these HF-PSs prompted us to further investigate whether these polysaccharides might be (i) antigenic and, if so, (ii) to characterize their immunochemical specificities. In this paper, we show that the HF-PS from B. anthracis is antigenic in that anti-HF-PS IgG antibodies are found in the antisera from rabbits inoculated with B. anthracis live or killed spores. In addition, we demonstrate that HF-PSs from pathogenic B. cereus clinical isolates of human patients suffering from severe or fatal pneumonia (Hoffmaster et al. 2004, 2006; Avashia et al. 2007), i.e. B. cereus strains G9241, 03BB87, and 03BB102, share carbohydrate antigen epitopes with B. anthracis and that these epitopes are not found on the nonpathogenic B. cereus ATCC 10987 or the B. cereus type strain ATCC 14579. We show, using antisera against a keyhole limpet hemocyanin (KLH) conjugate of B. anthracis HF-PS, that the five B. cereus and three B. anthracis strains tested share a common epitope in their HF-PS-BSA conjugates. Finally, using antisera from Rhesus macaques that survived inhalation anthrax, we demonstrate that the HF-PS antigen is expressed during B. anthracis infection in vivo.


Reactivity of antispore antisera with HF-PS from B. anthracis and B. cereus strains

Immunoreactivity of HF-PS extracts from selected B. anthracis and B. cereus strains was evaluated by enzyme-linked immunosorbent assay (ELISA). Antiserum to both live and killed B. anthracis spores contained IgG antibodies that bound conjugates of the HF-PSs from B. anthracis and the B. cereus clinical isolates, G9241, 03BB87, and 03BB102, isolated from cases of severe or fatal pneumonia (Figure (Figure2A2A and B). In contrast however, these antisera did not bind the B. cereus ATCC 14579 HF-PS-BSA conjugate. The binding of anti-B. anthracis spore antisera to the synthetic AntRha2-BSA conjugate was also observed, as previously reported (Mehta et al. 2006). There was no detectable binding of these antisera to the negative control bovine serum albumin (BSA) or maltoheptaose-BSA conjugate. Furthermore, anti-B. cereus ATCC 14579 spore antiserum bound to a B. cereus ATCC 14579 HF-PS-BSA conjugate but not to the HF-PS-BSA conjugates from B. anthracis, or the B. cereus G9241, 03BB87, and 03BB102 clinical isolates (Figure (Figure2C).2C). These data support the conclusion that the B. anthracis HF-PS epitopes are present on B. anthracis spores and that cross-reactive epitopes exist between B. anthracis spore HF-PS antigen and the HF-PSs from the three clinical B. cereus isolates that caused severe or fatal pneumonia.

Fig. 2
Immunoreactivity of the HF-PSs to antisera raised in rabbits against B. anthracis Sterne (A) live spores and (B) killed spores, and to (C) B. cereus ATCC 14579 spores. The ELISA microtiter plate wells were coated with B. anthracis Pasteur HF-PS-BSA ([filled triangle]), ...

The specificity of the antibody binding to the HF-PSs from B. anthracis and B. cereus was further evaluated using inhibition ELISAs where a B. anthracis Pasteur HF-PS-BSA conjugate was used as the capture antigen and unconjugated HF-PSs from B. anthracis Pasteur, B. anthracis Ames, B. cereus ATCC 10987, B. cereus ATCC 14579, and B. cereus G9241 were used as the inhibitors (Figure (Figure3).3). The data show that the HF-PSs from B. anthracis Pasteur and B. anthracis Ames were effective inhibitors and that the B. cereus G9241 HF-PS was able to inhibit binding but to a lesser extent; i.e., 50% inhibition required a 10-fold greater concentration of B. cereus G9241 HF-PS compared to B. anthracis HF-PS. In contrast to these HF-PSs, the HF-PSs from B. cereus ATCC 14579 and from B. cereus ATCC 10987 were not effective inhibitors, even when presented at 50-fold excess (wt/wt). Also, no inhibition was observed when using the chemically synthesized spore AntRha2 trisaccharide indicating that the reactivity to HF-PS is due to epitopes different from those on AntRha2.

Fig. 3
An ELISA inhibition assay showing the ability of various HF-PSs to inhibit the binding of antiserum to live B. anthracis spores to the HF-PS-BSA conjugate of B. anthracis Pasteur. The microtiter plates were coated with the HF-PS-BSA and the ability to ...

Reactivity of anti-B. anthracis HF-PS-KLH conjugate antisera with HF-PS from B. anthracis and B. cereus

Rabbit anti-B. anthracis HF-PS-KLH conjugate antiserum reacted to similar levels in ELISA with HF-PS-BSA conjugates of B. anthracis and B. cereus ATCC 14579 extracts indicating the presence of common cross-reactive epitopes in these HF-PS-protein conjugates (Figure (Figure4).4). The presence of these common cross-reactive epitopes in the HF-PS-protein conjugates of all the B. anthracis and B. cereus strains was further examined by immuno-dot blot assays which are described below. A low level of binding to a maltoheptaose-BSA conjugate was observed but this occurred only at the highest serum concentration and was at the assay threshold of 0.5 OD units. In contrast, no binding to the AntRha2-BSA conjugate was observed, indicating, again, that the cross-reactive epitopes to the HF-PS are not present on this spore carbohydrate.

Fig. 4
Reactivity of rabbit anti-B. anthracis HF-PS-KLH antiserum with HF-PS from other species. ELISA microtiter plates were coated with the various HF-PS-BSA conjugates and the ability of the HF-PS-KLH antiserum to bind these conjugates was determined as described ...

The ability of the rabbit anti-B. anthracis HF-PS-KLH conjugate to bind B. cereus and B. anthracis HF-PSs, HF-PS-BSA conjugates, cells and cell walls, and to B. anthracis spores was explored further using an immuno-dot blot assay. In this assay, at a serum dilution of 1:1600, antibody binding was observed to B. anthracis Sterne spores, but not to the AntRha3 trisaccharide or to the maltoheptaose or BSA controls (Figure (Figure5,5, Panel A). This result indicates that the AntRha2 trisaccharide is a distinct antigen from the HF-PS and, also, that one or more structural motifs of HF-PS are present as antigens in spores. Panel B of Figure Figure55 shows that the binding of rabbit anti-B. anthracis HF-PS-KLH antiserum to unconjugated HF-PSs could be observed down to a threshold level of 0.1 μg for the B. anthracis HF-PS, 1–3 μg for the HF-PSs from the three B. cereus clinical isolates G9241, 03BB87, and 03BB102, and no detectable binding for up to 5 μg of the HF-PS from nonclinical B. cereus ATCC 14579-type strain. However, a different reactivity pattern was observed with the HF-PS-BSA conjugate antigens for which antiserum against B. anthracis HF-PS-KLH reacted strongly to the HF-PS-BSA conjugates from all species and strains, including that from B. cereus ATCC 14579 (Figure (Figure5,5, Panel B). This latter result indicates that conjugation to protein produces or exposes an antigenic determinant that is common to protein conjugates of the HF-PSs from all of the B. cereus strains used in this study.

Fig. 5
An immuno-dot blot assay shows the binding of the antiserum to B. anthracis HF-PS-KLH conjugate to the indicated amount (μg) of (A) BSA, maltoheptaose, chemically synthesized AntRha2 trisaccharide (labeled as Anthrose), and B. anthracis Sterne ...

Panel C of Figure Figure55 shows that rabbit anti-B. anthracis HF-PS-KLH antiserum was reactive against whole cells and cell walls of all B. anthracis strains used in this assay, as well as whole cells and cell wall extracts from B. cereus clinical isolates G9241, 03BB87, and 03BB102. The detection threshold limit for binding the cell walls of the B. anthracis strains was 0.1 μg. In comparison, this threshold limit for binding the cell walls from the B. cereus clinical isolates G9241 and 03BB87 was increased to about 1.0 μg, and even greater for the clinical isolate 03BB102 requiring 10 μg. No binding of this antiserum to the cells and cell walls of B. cereus ATCC 14579 and B. cereus ATCC 10987 was observed, a result which is consistent with the data described above showing that anti-B. anthracis spore antiserum binds the HF-PSs from B. anthracis and the three B. cereus clinical isolates, but not the HF-PS from these latter two B. cereus strains. This cross-reactivity of the B. anthracis HF-PS with the HF-PS from B. cereus clinical isolates that caused fatal pneumonia is intriguing and indicative of a shared structural epitope among these pathogenic bacilli, a conclusion that is consistent with the similar glycosyl compositions of these HF-PSs.

Reactivity of Rhesus macaque (Macacca mulata) anti-AVA and postinfection sera with B. anthracis HF-PS

The presence of IgG antibodies that bind the B. anthracis HF-PS in animals inoculated with B. anthracis spores prompted an examination of available antiserum from naïve and anthrax vaccine adsorbed (AVA) vaccinated Rhesus macaques that had survived aerosol challenge with B. anthracis. Prechallenge and convalescent sera were obtained from five anthrax-vaccinated (RM1, RM3, RM4, RM5, RM6) and three naïve (RM8, RM9, and RM10) Rhesus macaques. Vaccinated animals had received three doses (week 0, 4, 26) of a 1:10 (RM4), 1:20 (RM1, RM3, RM6), or 1:40 (RM5) dilution of AVA and survived aerosol challenge with 20–422 LD50 equivalents (7 × 105–4 × 106 CFUs) of B. anthracis Ames strain given at week 52. Sera were evaluated by ELISA using the B. anthracis Pasteur HF-PS-KLH conjugate as the capture antigen (Figure (Figure6).6). None of the animals showed a prevaccination response (week 0). Three of the five vaccinated animals (RM3, RM5, RM6) had an above-threshold response at week 30 indicating that AVA may contain HF-PS and all of the vaccinated animals responded above the threshold on day 14 postexposure at levels much greater than those in naïve animals (Figure (Figure6A).6A). None of the naïve animals had a detectable preexposure response above the threshold and only one of the three unvaccinated animals (RM10) mounted an immune response above the threshold on day 14 postexposure (Figure (Figure6B).6B). All animals mounted an antiprotective antigen (PA) IgG response postexposure, confirming that they had been infected with B. anthracis (data not shown).

Fig. 6
Reactivity of Rhesus macaque sera with B. anthracis HF-PS. Pre- and postexposure sera from five AVA-vaccinated (A) and three naïve (B) animals were tested with ELISA for the presence of anti-HF-PS IgG on plates coated with the B. anthracis Pasteur ...


The data presented in this report demonstrate that the major polysaccharides released from the cell walls of a selection of B. anthracis and B. cereus strains by aqueous HF are antigenic and animals exposed to spores of these strains generated antipolysaccharide IgG antibodies to B. anthracis and B. cereus, respectively. Postinfection Rhesus macaque serum also reacted to B. anthracis HF-PS indicating that this antigen is expressed during infection, and the presence of anti-HF-PS antibodies in the serum from vaccinated animals prior to spore exposure indicated that HF-PS is likely present in the AVA. Further, immunochemical analysis of these polysaccharide antigens showed that they contain both common and strain-specific epitopes depending on the antiserum–antigen combination used for investigation.

Common cross-reactive epitopes were demonstrated by the reaction of rabbit anti-B. anthracis HF-PS-KLH antiserum with the HF-PS-BSA conjugate antigens from all B. anthracis and B. cereus strains investigated. This antiserum reacted strongly with the BSA conjugates of the HF-PSs from B. cereus strains ATCC 10987 and ATCC 14579 as well as with these same antigens from B. anthracis and the three clinical B. cereus isolates that caused severe or fatal pneumonia. The identity of the structural features in the HF-PSs responsible for the observed common cross-reactive epitopes is unknown, but this cross-reactivity depended on conjugation of the isolated HF-PSs to a protein. This dependence suggests that the common cross-reactive epitopes are normally cryptic and not exposed in the cells, cell walls, or unconjugated HF-PSs. One possible explanation is that the combination of releasing the HF-PS from the cell wall with conjugation to a protein exposes a common structural feature that becomes immunoreactive. Data from this laboratory suggest that the HF-PS from B. anthracis and all of the B. cereus strains examined here have a backbone repeating unit structure that is rich in aminoglycosyl residues (Figure (Figure1),1), of which two residues are GlcNAc and ManNAc with another being either GlcNAc or GalNAc, and that this backbone structure is substituted with Gal or Glc residues or noncarbohydrate groups such as acetyl substituents (Leoff, Choudhury, et al. 2008). It may be that conjugation to proteins involves a ManNAc-GlcNAc- common structural motif in these HF-PSs that, when conjugated to protein, becomes a more accessible epitope for the host's immune response and for antibody binding. A second possible explanation is that a common structural motif may be present in the form of a highly conserved linkage group between these HF-PSs and the PG, e.g., if the HF-PSs of all of these B. anthracis and B. cereus strains were attached to the PG via the same -HexNAc-P(P)-PG glycosyl-phosphate (or pyrophosphate) bridge. In the cell wall, such a common -HexNAc-P(P)-PG region in each of the polysaccharides would be in the innermost portion of the cell wall and not directly accessible to the host's immune system while the structurally variable portion of the polysaccharide is more exposed and accessible. However, when the polysaccharides are released by HF cleavage of the phosphate bridge, the common structural region that was linked to the PG is “uncovered” and, therefore, more accessible to the host's immune system. Conjugation of the isolated HF-PS to the protein may enhance this accessibility and result in the observed cross-reactivity between anti-B. anthracis HF-PS-KLH antiserum and all of the HF-PS-BSA conjugates. At this time, it is not known if all of these HF-PSs have a common structural region at their reducing ends (i.e., the end that would have been attached to the PG via a phosphate bridge). There is evidence, however, that cell wall teichoic acid polymers of certain bacilli are linked to the peptidoglycan through a common -ManNAc-GlcNAc-P(P)-PG linkage (Bhavsar et al. 2004; Freymond et al. 2006; Ginsberg et al. 2006). It has also been shown that other secondary cell wall polysaccharides from several bacilli are linked from a GlcNAc residue to the PG muramic acid residue via phosphate or pyrophosphate (Schaffer et al. 1999, 2000; Steindl et al. 2005). Investigation into the existence and structures of the PG linkage region of the B. anthracis and B. cereus HF-PSs is underway.

Specific epitopes were demonstrated by the reaction of antiserum raised against live or killed B. anthracis spores with the isolated HF-PS or HF-PS-BSA conjugate antigens from B. anthracis strains. Also, these antisera reacted, at a reduced level, with HF-PS-BSA conjugate antigens from the clinical B. cereus isolates that caused fatal or severe pneumonia. However, no reaction was observed with the HF-PS-BSA conjugate from the B. cereus type strain ATCC 14579. Likewise, antiserum to the spores from B. cereus ATCC 14579 only reacted with the HF-PS-BSA conjugate of that strain. We also demonstrated the existence of specific epitopes in cells, cell walls, and isolated but unconjugated HF-PSs from B. anthracis strains and from the three clinical B. cereus isolates through their reactivity with an antiserum raised against the B. anthracis HF-PS-KLH conjugate. This antiserum did not react with the same extracts from B. cereus strains ATCC 14579 or ATCC 10987. It was also observed that this anti-B. anthracis HF-PS antiserum reacted with B. anthracis spores. Thus, in addition to its specificity, the reactivity of the anti-HF-PS-KLH antiserum with B. anthracis spores, as well as the presence of anti-HF-PS IgG antibodies in antiserum generated against B. anthracis killed spores, supports the conclusion that this HF-PS structure is a spore antigen or a component of these spore preparations, as well as a vegetative cell wall antigen.

As stated above, we observed that cross-reactive epitopes that bound B. anthracis spore antiserum were present in the HF-PSs from three clinical isolates of B. cereus that caused severe or fatal pneumonia, G9241, 03BB87, and 03BB102 (Hoffmaster et al. 2004, 2006; Avashia et al. 2007), indicating structural conservation or relatedness in the HF-PS antigens of these strains to that from B. anthracis. The cross-reactive epitopes were not observed in the HF-PSs from the closely related B. cereus ATCC 10987 strain or the B. cereus ATCC 14579 type strain. The lack of cross-reactive epitopes on these latter two B. cereus HF-PSs is likely due to the fact that the structures of these molecules differ significantly from the B. anthracis HF-PS (Choudhury et al. 2006; Leoff, Choudhury, et al. 2008; Leoff, Saile, et al. 2008). On the other hand, the cross-reactive epitopes on the HF-PSs from the three clinical B. cereus isolates are most likely due to the similarity in their structures to that of the B. anthracis HF-PSs. These HF-PSs are very similar in glycosyl residue composition to the B. anthracis HF-PSs (Leoff, Saile, et al. 2008), and recent structural analysis indicates that they all have the same aminoglycosyl trisaccharide backbone structure as the B. anthracis HF-PS but with more extensive substitution by Gal residues (Choudhury et al., in preparation). We hypothesize, therefore, that these results indicate the existence of pathogenicity-related conserved structural elements in these cell wall antigens. If these hypothesized cross-reactive structural features in the HF-PSs are confirmed, they could be particularly useful for the development of multivalent vaccines that would be effective against both B. anthracis as well as against B. cereus strains that cause severe illness.

At present, we do not know the details of the relationship between pathogenicity and HF-PS structures. However, it is likely that the HF-PS has important functions for growth and/or pathogenicity, e.g., involving the carbohydrate binding domain (CBD) of cell surface proteins. It is known that surface proteins in B. anthracis, S-layer proteins and others, have a CBD. The CBD, e.g., in the B. anthracis S-layer proteins Sap and EA1, is a protein domain that normally comprises three short amino acid stretches with a motif known as the SLH motif (for S-layer homology) (Zona and Janećek 2005). In the case of Sap and EA1 from B. anthracis, their export and anchoring to the cell wall are mediated by the SLH domain to form a crystalline array in the surface of the cell (Mesnage et al. 2000). It is thought that the SLH protein domain binds to the HF-PS which, in turn, is covalently bound via a phosphate bridge to the PG of the cell wall (Mesnage et al. 2000). In addition to Sap and EA1, it was recently reported that another surface protein, BslA, that is encoded on the pXO1 plasmid contains a SLH domain and is responsible for adherence of B. anthracis to host cells (Kern and Schneewind 2008). Thus, the HF-PS of pathogenic strains could be involved in exporting/anchoring proteins, such as BslA, that are necessary for virulence. However, to date no proof has emerged that shows directly this protein's exporting/anchoring function or its involvement in host–cell interactions and other functions that may be required for growth. We have embarked on testing this hypothesis by preparing and characterizing the phenotypes (i.e., growth and virulence properties, HF-PS structure, and binding affinities to surface proteins) of B. anthracis mutants carrying mutations in genes thought to encode enzymes required for the synthesis of the HF-PS.

Finally, we showed that sera from all vaccinated Rhesus macaques that were exposed to B. anthracis spores contain IgG antibodies that bind the B. anthracis HF-PS. This result supports further investigation into the potential use of the HF-PS conjugates to detect exposure of primates to B. anthracis, and for use as an alternative antigen component for the development or improvement of anthrax vaccines. These investigations as well as investigations into the HF-PS structures from the pathogenic B. cereus strains are in progress.

Material and methods

Bacterial strains and culture conditions

The strains/isolates used in this work and their phylogenetic relatedness are listed in Table TableI.I. All B. anthracis strains were obtained from the CDC culture collection. Cells cultured overnight in the brain heart infusion medium (BHI) (BD BBL, Sparks, MD) containing 0.5% glycerol were used to inoculate four 250 mL volumes of BHI medium in 2 L Erlenmeyer flasks the next morning. Cultures were grown at 37°C (B. anthracis) or 30°C (B. cereus) with shaking at 200 rpm. Growth was monitored by measuring the optical density of the cultures at 600 nm. In the mid-log phase, cells were harvested by centrifugation (8000 × g, 4°C, 15 min), washed two times in sterile saline, enumerated by dilution plating on BHI agar plates, and then autoclaved for 1 h at 121°C before further processing.

Table I
Bacillus anthracis and B. cereus group strains used in this study

Preparation of bacterial cell walls

Bacterial cell walls were prepared from previously enumerated autoclaved bacterial cells (3 × 108 to 3 × 109 CFU/mL) that were disrupted in 40 mL sterile saline on ice by four 10 min sonication cycles. The complete or near complete disruption of cells was checked microscopically. Unbroken cells were removed by a low speed centrifugation run (8000 × g, 4°C, 15 min). The separated pellet and supernatant fractions were stored at −70°C. The cell walls were separated from the low speed supernatants by ultracentrifugation at 100,000 × g, 4°C for 4 h. The resulting cell wall pellets were washed by suspension in cold, deionized water followed by an additional ultracentrifugation at 100,000 × g, 4°C for 4 h, and lyophilized.

Release of phosphate-bound polysaccharides from the cell wall

Phosphate-bound polysaccharides were released from the cell walls by treatment with aqueous HF according to the modification of the procedure described by Ekwunife et al. (1991). Briefly, the cell walls were subjected to 47% HF under stirring at 4°C for 48 h. The reaction mixture was neutralized with NH4OH, subjected to a 10 min low speed centrifugation, and the supernatant with the released polysaccharides lyophilized, redissolved in deionized water, and subjected to a chromatographic size separation on a BioGel P2 column (Bio-Rad). The fractions eluting from the Bio-Gel P2 column were monitored using a refractive index detector. Polysaccharide-containing fractions were pooled, lyophilized, and analyzed by gas chromatography-mass spectrometry as previously described (Choudhury et al. 2006).

Preparation of spores

Spores of B. anthracis were prepared from liquid cultures of the phage assay (PA) medium (Green et al. 1985) grown at 37°C, 200 rpm for 6 days. Spores of B. cereus ATCC 14579 were prepared from liquid cultures of the PA medium grown at 30°C, 200 rpm for 6 days. Spores were harvested by centrifugation and washed two times by suspension in cold (4°C) sterile deionized water followed by centrifugation at 10,000 × g. They were then purified in a 50% Reno-60 (Bracco Diagnostics Inc., Princeton, NJ) gradient (10,000 × g, 30 min, 4°C) and washed a further four times in cold sterile deionized water. After suspension in sterile deionized water, spores were quantified by surface spreading on BHI agar plates (BD BBL, Sparks, MD) and counting the colony forming units (cfu). Spore suspensions were stored in water at −80°C.

For the preparation of killed spores, 500 μL aliquots of spore suspensions in water, prepared as described above and containing approximately 3 × 108 CFU/mL, were irradiated in 2 mL Sarstedt freezer tubes (Sarstedt, Newton, NC) in a gamma cell irradiator with an absorbed dose of 2 million rads. Sterility after irradiation was confirmed by spread-plating 10 μL aliquots of irradiated spore suspension on BHI agar plates. The plates were incubated for 72 h at 37°C and monitored for colony growth. The absence of growth was taken as an indicator of sterility.

Preparation of rabbit anti-spore antiserum and Rhesus macaque infection sera

Anti-spore antiserum against spores of B. anthracis Sterne and B. cereus ATCC 14579 were prepared in female New Zealand White rabbits (2.0–3.5 kg) purchased from Myrtle's Rabbitry (Thompson Station, TN). Each of two rabbits was inoculated intramuscularly at two sites in the dorsal hind quarters with 0.5 mL of washed live-spore or killed-spore inoculum (3 × 106 total spores). Rabbits were vaccinated at 0, 14, 28, and 42 days. Serum was collected prior to the first immunization (preimmune serum) and at day 7 and day 14 after each injection of antigen. Terminal bleeds were collected on day 14 after the last immunization. All animal protocols were approved by the CDC Animal Care and Use Committee (ACUC) and implemented under the direction of the CDC attending veterinarian. Rhesus macaque sera were made available from anthrax correlates of protection studies at CDC (C.P. Quinn, unpublished data).

Conjugation of HF-PS to BSA or KLH

Conjugation was performed by modification of a previously described method (Bystricky et al. 2000; Shafer et al. 2000). Approximately 1 mg of freeze-dried polysaccharide was dissolved in 90 μL of 0.15 M HEPES buffer, pH 7.4. While stirring, 4 mg of 1-cyano-4-dimethylaminopyridinium tetrafluoroborate (CDAP) in acetonitrile (90 μL) was slowly added to a solution of the polysaccharide to avoid precipitation. After activation of the polysaccharide (30 s), aqueous triethylamine (120 μL of 0.3 M triethylamine) was added and stirred for 2 min. Finally, 4 mg of BSA (Sigma, St. Louis, MO) or keyhole limpet hemocyanin (KLH; Sigma) were dissolved in 348 μL 0.01 M phosphate buffered saline (PBS), pH 7.4, and added to the reaction mixture. After stirring for 18 h at 4°C, the reaction mixture was quenched with the addition of 120 μL of 0.5 M ethanolamine in the 0.75 M HEPES buffer, pH 7.4. After 15–20 min of stirring, the unconjugated sugars in the mixture were separated from the protein–polysaccharide conjugate by centrifugation at 3200 × g using a centrifugal filter device (Centriplus YM-10, Millipore, Billerica, MA). The conjugate was lyophilized and stored at room temperature. The percentage of sugars in the conjugates was determined by the preparation and GC-MS analysis of trimethylsilyl methyl glycosides (York et al. 1985). Briefly, 200 μg of the HF-PS-KLH or -BSA conjugate was methanolyzed in methanolic 1 M HCl, derivatized into trimethylsilyl ethers, and analyzed by GC-MS. Using this procedure, the percent mass of hexose and the amount of carbohydrate in the HF-PS-protein conjugates was determined from the known hexose percent present in the unconjugated HF-PS, e.g., based on Gal for B. anthracis and on Glc for B. cereus ATCC 14579 HF-PS-protein conjugates.

Preparation of antiserum to the B. anthracis HF-PS-KLH conjugates—B. anthracis

B. anthracis Pasteur HF-PS was conjugated to KLH as described above and used for the preparation of anti-HF-PS antiserum. For antiserum production each of two female (2.0–3.5 kg) New Zealand White rabbits (Myrtle's Rabbitry, Thompson Station, TN) were inoculated intramuscularly at two sites in the dorsal hind quarters. For the primary injection, 1.0 mL of MPL + TDM + CWS Adjuvant System (Sigma) with 500 μg of the HF-PS-KLH conjugate was divided into two injections per rabbit. For the booster shots, 1.0 mL of MPL + TDM + CWS Adjuvant System with 250 μg of the HF-PS-KLH conjugate was used. Rabbits were immunized at 0, 14, 28, and 42 days. Serum was collected prior to the first immunization (preimmune serum) and at day 7 and day 14 after each injection of antigen. Terminal bleeds were collected 14 days after the last immunization.

Enzyme-linked immunosorbent assay determination of IgG binding to B. anthracis and B. cereus HF-PS-protein conjugates

The immunochemical reactivity of serum from rabbits inoculated with B. anthracis spores and that of serum from Rhesus macaques that survived inhalation anthrax were tested against protein-conjugated HF-PS extracts from B. anthracis Ames and B. cereus ATCC 14579 by ELISA. Slightly different protocols were used to examine these antisera.

The rabbit anti-B. anthracis spore antisera were assayed using the wells of a 96-well microtiter plate (Immulon II-HB, Thermo Labsystems, Franklin, MA) in which each well was coated with the 100 μL of a 5 μg/mL solution of HF-PS-BSA conjugate in 100 μL of 0.01 M PBS, pH 7.4, and incubated over night at 4C. The next day, the plates were washed three times with the wash buffer (0.01 M PBS, pH 7.4, 0.1% Tween-20) followed by the blocking buffer (5% nonfat dry milk in 0.01 M PBS, pH 7.4, 0.5% Tween-20) for 1 h at room temperature. The plates were then washed again, and serial dilutions (100 μL per well) of spore rabbit antiserum in the blocking buffer were added and the plates incubated for 1 h at room temperature. The plates were then washed three times with the wash buffer. Horseradish peroxidase (HRPO)-labeled goat anti-rabbit IgG, 1:5000 dilution, was added (100 μL/well) and incubated for 1 h at room temperature. Plates were washed five times with the wash buffer before adding 100 μL of ABTS/H2O2 peroxidase substrate (KPL, Gaithersburg, MD) for 10 min. The color development was stopped with the addition of 100 μL of ABTS peroxidase stopping solution (KPL, Gaithersburg, MD) and the optical density of each well was read at a wavelength of 405 nm with a microtiter plate reader (Bio-Rad Laboratories, Hercules, CA).

The Rhesus macaque sera were assayed as described above with the exception that the B. anthracis HF-PS-KLH conjugate rather than the BSA conjugate was used to coat the microtiter plates. Samples were tested three times and average OD and standard deviation were calculated. Anti-HF-PS IgG responses were expressed as a “fold response” over a reactivity threshold (RT) value. The RT was determined from the average OD value plus two standard deviations (SD) from the sera of 88 true negative Rhesus macaques tested against HF-PS-KLH by ELISA. Each sample was tested twice at a 1:100 dilution in the dilution buffer. The RT was calculated as an OD value of 0.22.

Specificity of antispore antisera for HF-PS

Specificity analyses were done by inhibition ELISAs using various unconjugated HF-PSs and evaluating their ability to block the binding of anti-B. anthracis spore antiserum IgG to the B. anthracis Pasteur HF-PS-BSA conjugate. The HF-PS samples tested for inhibition were as follows: B. anthracis Pasteur HF-PS, B. anthracis Ames HF-PS, B. cereus ATCC 10987 HF-PS, B. cereus ATCC 14579 HF-PS, and B. cereus G9241 HF-PS. BSA was used as the inhibition negative control. For analysis, rabbit anti-live spore serum was diluted 1:1600 in the ELISA blocking buffer to obtain an OD of approximately 1.0 for the positive control HF-PS-BSA conjugate (Figure (Figure3).3). Subsequently, 100 μL of diluted serum was added to the coated microtiter plate wells together with 0-, 5-, 10-, 25-, or 50-fold excess unconjugated HF-PS (i.e., fold excess relative to the 0.35 μg of carbohydrate equivalent to the B. anthracis HF-PS-BSA conjugate coating each well of the microtiter plate). Each HF-PS inhibitor was diluted in the blocking buffer. Inhibitor and serum were briefly mixed in an uncoated microtiter plate followed by immediate transfer to the coated plate. Plates were incubated for 1 h at room temperature followed by washing with the wash buffer three times. The microtiter plates were incubated with horseradish peroxidase-labeled anti-rabbit IgG and developed as described above.

Reactivity of anti-HF-PS-KLH antiserum with cells, cell walls, and spores of B. anthracis and B. cereus strains

Immuno-dot blot assays were used to measure the binding of various antiserum preparations to cells, cell walls, and spores. Cells, cell walls, or spores were suspended in distilled water and blotted onto a nitrocellulose membrane. The spore suspension had an optical density of 0.56 at 600 nm. Samples with a volume >5 μL were taken from 1 mg/mL of cell or cell wall stock preparations, dried in a speed-vac, and re-dissolved in 3 μL of distilled water before they were blotted onto the membrane. Samples with volumes <5 μL were taken from the above-mentioned stock preparations and directly blotted onto the membrane without prior reduction of the volume. BSA, maltoheptaose, and chemically synthesized B. anthracis BclA AntRha2 trisaccharide were blotted as controls. The membrane was allowed to dry overnight before blocking with the blocking buffer for 1 h. The membrane was then incubated at room temperature for 1 h with the antiserum to B. anthracis HF-PS-KLH conjugate that had been diluted 1:1600 in the blocking buffer. After washing three times with the wash buffer, the membrane was incubated with a 1:1000 dilution of mouse anti-rabbit IgG linked to alkaline phosphatase in the 0.01 M PBS buffer, pH 7.4, for 1 h at room temperature. After washing five times, the membrane was developed using Nitro Blue Tetrazolium (0.3 mg/mL in 0.1 M NaCl, 0.1 M trishydroxymethylaminomethane (Tris), 5mM MgCl2, of 0.15 mg/mL of 5-bromo-4-chloro-indolyl-phosphate, pH 9.0). The reaction was stopped by washing with tap water.


NIAID (R21 AI059577 to R.W.C.) and DOE (DE-FG02-93ER20097 to C.C.R.C.).


The findings and conclusions in this report are those of the author(s) and do not necessarily represent the views of the Centers for Disease Control and Prevention. Patent Pending – University of Georgia Research Foundation, Inc. and US Centers for Disease Control and Prevention. A portion of the results presented in this paper were obtained as part of Christine Leoff's doctoral thesis work (

Conflict of interest statement

None declared.


anthrose (4-N-β-hydroxyisovalyerl-4,6-dideoxy-d-glucose)
anthrax vaccine adsorbed
bovine serum albumin
carbohydrate binding domain
keyhole limpet hemocyanin


  • Avashia SB, Riggins WS, Lindley C, Hoffmaster AR, Drumgoole R, Nekomoto T, Jackson PJ, Hill KK, Williams K, Lehman L, et al. Fatal pneumonia among metalworkers due to inhalation exposure to Bacillus cereus containing Bacillus anthracis toxin genes. Clin Infect Dis. 2007;44:414. [PubMed]
  • Baillie LW. Bacillus anthracis, a story of nature subverted by man. Lett Appl Microbiol. 2005;41:227–229. [PubMed]
  • Bhavsar AP, Erdman LK, Schertzer JW, Brown ED. Teichoic acid is an essential polymer in Bacillus subtilis that is functionally distinct from teichuronic acid. J Bacteriol. 2004;186:7865–7873. [PMC free article] [PubMed]
  • Bystricky S, Machova E, Bartek P, Kolarova N, Kogan G. Conjugation of yeast mannans with protein employing cyanopyridinium agent (CDAP) – An effective route of antifungal vaccine preparation. Glycoconj J. 2000;17:677–680. [PubMed]
  • Choudhury B, Leoff C, Saile E, Wilkins P, Quinn CP, Kannenberg EL, Carlson RW. The structure of the major cell wall polysaccharide of Bacillus anthracis is species specific. J Biol Chem. 2006;281:27932–27941. [PubMed]
  • Daubenspeck JM, Zeng H, Chen P, Dong S, Steichen CT, Krishna NR, Pritchard DG, Turnbough CL., Jr Novel oligosaccharide side chains of the collagen-like region of BclA, the major glycoprotein of the Bacillus anthracis exosporium. J Biol Chem. 2004;279:30945–30953. [PubMed]
  • Ekwunife FS, Singh J, Taylor KG, Doyle RJ. Isolation and purification of cell wall polysaccharide of Bacillus anthracis (delta Sterne) FEMS Microbiol Lett. 1991;82:257–262. [PubMed]
  • Ford WW, Lawrence JS. Studies on aerobic spore bearing non-pathogenic bacteria in milk. J Bacteriol. 1916;1:277–320. [PMC free article] [PubMed]
  • Freymond PP, Lazarevic V, Soldo B, Karamata D. Poly(glucosyl-N-acetylgalactosamine 1-phosphate), a wall teichoic acid of Bacillus subtilis 168: Its biosynthetic pathway and mode of attachment to peptidoglycan. Microbiology. 2006;152:1709–1718. [PubMed]
  • Ginsberg C, Zhang YH, Yuan Y, Walker S. In vitro reconstitution of two essential steps in wall teichoic acid biosynthesis. ACS Chem Biol. 2006;1:25–28. [PubMed]
  • Green BD, Battisti L, Koehler TM, Thorne CB, Ivins BE. Demonstration of a capsule plasmid in Bacillus anthracis. Infect Immun. 1985;49:291–297. [PMC free article] [PubMed]
  • Hilleman MR. Overview: Cause and prevention in biowarfare and bioterrorism. Vaccine. 2002;20:3055–3067. [PubMed]
  • Hoffmaster A, Novak R, Marston C, Gee J, Helsel L, Pruckler J, Wilkins P. Genetic diversity of clinical isolates of Bacillus cereus using multilocus sequence typing. BMC Microbiol. 2008;8:191. [PMC free article] [PubMed]
  • Hoffmaster AR, Hill KK, Gee JR, Marston CK, De BK, Popovic T, Sue D, Wilkins PP, Avashia SB, Drumgoole R, et al. Characterization of Bacillus cereus isolates associated with fatal pneumonias: Isolates are closely related to Bacillus anthracis and harbor B. anthracis virulence genes. J Clin Microbiol. 2006;44:3352–3360. [PMC free article] [PubMed]
  • Hoffmaster AR, Ravel J, Rasko DA, Chapman GD, Chute MD, Marston CK, De BK, Sacchi CT, Fitzgerald C, Mayer LW, et al. Identification of anthrax toxin genes in a Bacillus cereus associated with an illness resembling inhalation anthrax. Proc Natl Acad Sci USA. 2004;101:8449–8454. [PubMed]
  • Kern J, Schneewind O. BslA, a pXO1-encoded adhesin of Bacillus anthracis. Mol Microbiol. 2008;68:504–515. [PubMed]
  • Leoff C, Choudhury B, Saile E, Quinn CP, Carlson RW, Kannenberg EL. Structural Elucidation of the non-classical secondary cell wall polysaccharide from Bacillus cereus ATCC 10987: Comparison with the polysaccharides from Bacillus anthracis and B. cereus type strain ATCC 14579 reveals both unique and common structural features. J Biol Chem. 2008;283:29812–29821. [PubMed]
  • Leoff C, Saile E, Sue D, Wilkins P, Quinn CP, Carlson RW, Kannenberg EL. Cell wall carbohydrate composition of strains from Bacillus cereus group of species correlate with phylogenetic relatedness. J Bacteriol. 2008;190:112–121. [PMC free article] [PubMed]
  • Mehta AS, Saile E, Zhong W, Buskas T, Carlson RW, Kannenberg EL, Reed Y, Quinn CP, Boons GJ. Synthesis and antigenic analysis of the BclA glycoprotein oligosaccharide from Bacillus anthracis exosporium. Chemistry. 2006;12:9136–9149. [PubMed]
  • Mesnage S, Fontaine T, Mignot T, Delepierre M, Mock M, Fouet A. Bacterial SLH domain proteins are non-covalently anchored to the cell surface via a conserved mechanism involving wall polysaccharide pyruvylation. EMBO J. 2000;19:4473–4484. [PubMed]
  • Priest FG, Barker M, Baillie LWJ, Holmes EC, Maiden MCJ. Population structure and evolution of the Bacillus cereus group. J Bacteriol. 2004;186:7959–7970. [PMC free article] [PubMed]
  • Schaffer C, Kahlig H, Christian R, Schulz G, Zayni S, Messner P. The diacetamidodideoxyuronic acid containing glycan chain of Bacillus stearothermophilus NRS 2004/3a represents the secondary cell wall polymer of wild type B. stearothermophilus strains. Microbiology. 1999;145:1575–1583. [PubMed]
  • Schaffer C, Messner P. The structure of secondary cell wall polymers: How Gram-positive bacteria stick their cell walls together. Microbiology. 2005;151:643–651. [PubMed]
  • Schaffer C, Muller N, Mandal PK, Christian R, Zayni S, Messner P. A pyrophosphate bridge links pyruvate containing secondary cell wall polymer of Paenibacillus alvei CCM2051 to muramic acid. Glycoconj J. 2000;17:681–690. [PubMed]
  • Shafer DE, Toll B, Schuman RE, Nelson BL, Mond JJ, Lees A. Activation of soluble polysaccharides with 1-cyano-4-dimethylaminopyridinium tetrafluoroborate (CDAP) for use protein-polysaccharide conjugate vaccines and immunological reagent: II. Selective crosslinking of proteins to CDAP-activated polysaccharides. Vaccine. 2000;18:1273–1281. [PubMed]
  • Smith NH. Aerobic spore forming bacteria. US Dep. Agric. Monogr. 1952;16:1–148.
  • Steindl C, Schaffer C, Smrecki V, Messner P, Muller N. The secondary cell wall polymer of Geobacillus tepidamans GS5-97T structure of different glycoforms. Carbohydr Res. 2005;340:2290–2296. [PubMed]
  • Sterne M. The effects of different carbon dioxide concentrations on the growth of virulent anthrax. Onderstepoort J Vet Sci Anim Ind. 1937;9:49–67.
  • Van Ert MN, Easterday WR, Simonson TS, U’Ren JM, Pearson T, Kenefic LJ, Busch JD, Huynh LY, Dukerich M, Trim CB, et al. Strain-specific single-nucleotide polymorphism assays for the Bacillus anthracis Ames strain. J Clin Microbiol. 2007;45:47–53. [PMC free article] [PubMed]
  • York WS, Darvill AG, McNeil M, Stevenson TT, Albersheim P. Isolation and characterization of plant cell wall components. Meth Enzymol. 1985;118:3–40.
  • Zona R, Janeček S. Relationship between SLH motifs from different glycoside hydrolase families. Biologia (Bratisl) 2005;60:115–121.

Articles from Glycobiology are provided here courtesy of Oxford University Press