Galactosylation of the Secondary Cell Wall Polysaccharide of Bacillus anthracis and Its Contribution to Anthrax Pathogenesis

ABSTRACT

Bacillus anthracis, the causative agent of anthrax disease, elaborates a secondary cell wall polysaccharide (SCWP) that is essential for bacterial growth and cell division. B. anthracis SCWP is comprised of trisaccharide repeats with the structure, [→4)-β-ManNAc-(1→4)-β-GlcNAc(O3-α-Gal)-(1→6)-α-GlcNAc(O3-α-Gal, O4-β-Gal)-(1→]_6-12. The genes whose products promote the galactosylation of B. anthracis SCWP are not yet known. We show here that the expression of galE1, encoding a UDP-glucose 4-epimerase necessary for the synthesis of UDP-galactose, is required for B. anthracis SCWP galactosylation. The galE1 mutant assembles surface (S) layer and S layer-associated proteins that associate with ketal-pyruvylated SCWP via their S layer homology domains similarly to wild-type B. anthracis, but the mutant displays a defect in γ-phage murein hydrolase binding to SCWP. Furthermore, deletion of galE1 diminishes the capsulation of B. anthracis with poly-d-γ-glutamic acid (PDGA) and causes a reduction in bacterial virulence. These data suggest that SCWP galactosylation is required for the physiologic assembly of the B. anthracis cell wall envelope and for the pathogenesis of anthrax disease.

IMPORTANCE Unlike virulent Bacillus anthracis isolates, B. anthracis strain CDC684 synthesizes secondary cell wall polysaccharide (SCWP) trisaccharide repeats without galactosyl modification, exhibits diminished growth in vitro in broth cultures, and is severely attenuated in an animal model of anthrax. To examine whether SCWP galactosylation is a requirement for anthrax disease, we generated variants of B. anthracis strains Sterne 34F2 and Ames lacking UDP-glucose 4-epimerase by mutating the genes galE1 and galE2. We identified galE1 as necessary for SCWP galactosylation. Deletion of galE1 decreased the poly-d-γ-glutamic acid (PDGA) capsulation of the vegetative form of B. anthracis and increased the bacterial inoculum required to produce lethal disease in mice, indicating that SCWP galactosylation is indeed a determinant of anthrax disease.

INTRODUCTION

Bacillus anthracis is a member of Bacillus cereus sensu lato, a clade of closely related, spore-forming, aerobic, facultative anaerobic bacteria that are generally found in the soil, on plants, or as symbionts of the insect gut (1, 2). However, B. cereus sensu lato also includes pathogenic isolates, such as Bacillus thuringiensis, which acquire virulence plasmids to promote the killing of insect larvae and replication of bacilli in carcass tissues (2). Following the acquisition of two large virulence plasmids, pXO1 and pXO2, B. anthracis evolved as a monomorphic pathogen of mammals, achieving global dissemination (3). B. anthracis spores are ingested or inhaled by mammalian herbivores and invade their gastrointestinal or respiratory tract (4). Spore germination is followed by outgrowth of vegetative bacilli that may disseminate into all tissues, causing potentially fatal infections (4). Vegetative bacilli then sporulate in disintegrating mammalian carcasses and contaminate the environment for subsequent dissemination to other hosts (4).

The pathogenesis of B. anthracis infections requires secreted toxins, i.e., lethal toxin and edema toxin, as well as poly-d-γ-glutamic acid (PDGA) capsulation; the structural genes for the toxin and capsule virulence factors are encoded on the pXO1 and pXO2 virulence plasmid, respectively (5). PDGA capsule is synthesized via the isomeration of l-glutamate to d-glutamate, ATP-dependent ligation of d-glutamate, secretion of poly-d-γ-glutamic acid, and finally, transpeptidation, as well as attachment of PDGA to cell wall peptidoglycan (6–8). Cell wall peptidoglycan is a polymer of glycan strands with a repeating disaccharide, comprised of N-acetylmuramic acid (1→4) linked to N-acetylglucosamine [→4)-β-MurNAc-(1→4)-β-GlcNAc-(1→], and cross-linked wall peptides with the sequence NH₂-l-alanine-d-iso-glutamate-meso-diaminopimelic acid–d-alanine-COOH (l-Ala-d-iso-Glu-m-Dpm-d-Ala) (9). In cross-linked peptidoglycan, the N-terminal amino group of l-Ala in the wall peptide is amide linked to the carboxyl group of MurNAc within glycan strands, whereas the carboxyl group of d-Ala is amide linked to the ε-amino group of m-Dpm in neighboring wall peptides, thereby tethering adjacent glycan strands to generate a rigid macromolecule that protects bacilli from osmotic lysis while also providing a scaffold for the assembly of the secondary cell wall polysaccharide (SCWP) (9).

The structure of the SCWP from B. anthracis is comprised of a repeating trisaccharide with galactosyl modifications, [→4)-β-ManNAc-(1→4)-β-GlcNAc(O3-α-Gal)-(1→6)-α-GlcNAc(O3-α-Gal, O4-β-Gal)-(1→]_6–12 (10). The SCWP from the closely related strain B. cereus ATCC 10987 consists of [→4)-β-ManNAc-(1→4)-β-GlcNAc-(1→6)-α-GalNAc-(1→]_n repeats and β-Gal substitutions at the O3 of α-GalNAc, as well as nonstoichiometrical acetylation at O3 of β-ManNAc (11). B. cereus ATCC 14579 SCWP has the same repeat structure and substitutions of β-GlcNAc at O3 of the β-GlcNAc, in addition to β-Glc at O3 and α-ManNAc at O4 of the α-GalNAc residue (12). SCWP is tethered to murein linkage units, [→4)-β-ManNAc-(1→4)-β-GlcNAc], that are linked via phosphodiester bonds to MurNAc in the bacterial peptidoglycan (13). Some but not all SCWP polymers are modified at O4 and O6 of the terminal ManNAc with ketal pyruvyl and at O3 with acetyl moieties (13–15). The modifications are introduced by products of the csaB (ketal pyruvyl) and patA/patB (acetyl) genes in the B. anthracis surface (S) layer gene cluster and promote the assembly of S layer and S layer-associated proteins, which bind the SCWP via their S layer homology (SLH) domains (13, 14, 16). B. anthracis strain CDC684 is a severely attenuated isolate that lacks all galactosyl modification of the SCWP (15). Although the genome sequence of B. anthracis CDC684 has been determined, the genetic basis for this distinct SCWP phenotype is not known (17).

We sought to identify the genetic basis for SCWP galactosylation, which is likely dependent on UDP-galactose (UDP-Gal) as the substrate for glucosyltransferases that modify the presumed cytoplasmic precursor of wall polysaccharide synthesis, the undecaprenyl-pyrophosphate-linked trisaccharide C₅₅-(PO₄)₂-ManNAc-(1→4)-β-GlcNAc-(1→6)-α-GlcNAc] (18). Earlier work characterized three gene products of B. anthracis that had been annotated as UDP-glucose 4-epimerase: BAS1093, galE1 (BAS5114), and galE2 (BAS5304) (19). Each of these genes was expressed, and the recombinant products purified from Escherichia coli strain BL21(DE3) and analyzed for epimerase activity. GalE1 and GalE2, but not the BAS1093-encoded protein, catalyzed the reversible interconversion of UDP-Glc and UDP-Gal (19). Furthermore, GalE2, but not GalE1, also converted UDP-GlcNAc to UDP-GalNAc (19). B. anthracis strain Sterne 34F2 lacking galE2, a gene that is only expressed during sporulation, cannot synthesize GalNAc (19). Unlike the spores of wild-type B. anthracis, spores of galE2 mutant bacilli elaborate BclA (Bacillus collagen-like protein of anthracis), the main glycoprotein of the exosporium, harboring GlcNAc, not GalNAc, as the anchoring moiety for two different oligosaccharides, 3-O-Me-Rha(α1–2)Rha(α1–3)GalNAc and Ant(β1–3)Rha(α1–3)Rha(α1–2)Rha(α1–3)GalNAc (19–22).

Here, we show that galE1 mutants of B. anthracis strains Sterne 34F2 and Ames synthesize SCWPs without galactosylation and that this defect can be complemented by the expression of galE1 or galE2 in vegetative bacilli. A lack of SCWP galactosylation does not affect the assembly of S layer and S layer-associated proteins. However, galE1 mutant bacilli display defects in PDGA capsulation and in the pathogenesis of B. anthracis infections.

RESULTS

B. anthracis galE1 is required for SCWP galactosylation.

The genome sequence of B. anthracis Sterne 34F2 contains two genes, designated galE1 (BAS5114) and galE2 (BAS5304), that encode UDP-Glc 4-epimerases (19). The galE1 gene is immediately adjacent to the surface polysaccharide synthesis (sps) locus, which is expressed during B. anthracis vegetative growth and encodes GneZ, the UDP-GlcNAc 2-epimerase for the interconversion of UDP-GlcNAc and UDP-ManNAc (23, 24). galE2 is expressed as a single-gene operon from σ^E- and σ^K-dependent promoters during B. anthracis sporulation (19). BAS1093 is also assigned as a UDP-Glc 4-epimerase-encoding gene; however, the purified recombinant BAS1093 product did not catalyze NAD⁺-dependent interconversion of UDP-Glc and UDP-Gal (19). Earlier work generated bursa aurealis transposon mutants of B. anthracis Sterne 34F2 and determined the location of insertional lesions by DNA sequencing (25). This approach identified insertions in BAS1093 and galE2, suggesting that these genes are not essential for B. anthracis growth. Using allelic replacement, we deleted BAS5114 in B. anthracis Sterne 34F2 to generate the ΔgalE1 strain, which, compared with its B. anthracis Sterne 34F2 parent, did not display a growth defect (data not shown). To analyze SCWP galactosylation in B. anthracis, bacilli were incubated with two reagents. Soybean agglutinin (SBA), conjugated to Alexa Fluor 594, is a lectin that selectively binds terminal α- and β-linked N-acetylgalactosamine and galactose residues (26, 27). The fluorescein isothiocyanate (FITC)-labeled monoclonal antibody (MAb) EAII6G6 binds galactosyl moieties of the B. anthracis SCWP (15, 27, 28). As expected, in fluorescence microcopy experiments, MAb EAII6G6 and SBA lectin bound the SCWP of B. anthracis Sterne 34F2 but not that of B. anthracis CDC684, a strain whose SCWP is devoid of galactosyl residues (Fig. 1) (29). Mutations that abrogated the expression of BAS1093 or galE2 did not affect the binding of MAb EAII6G6 and SBA lectin to the B. anthracis SCWP (Fig. 1). In contrast, deletion of the galE1 gene abolished the binding of MAb EAII6G6 and SBA lectin. This defect was complemented by plasmid-borne expression of galE1 or galE2 from the isopropyl-β-d-thiogalactopyranoside (IPTG)-inducible P_spac promoter of pgalE1 and pgalE2 transformed into B. anthracis ΔgalE1 (Fig. 1). These data indicate that only galE1, and not galE2 or BAS1093, is required for SCWP galactosylation in B. anthracis strain Sterne 34F2 vegetative forms. Nonetheless, the expression of both galE1 and galE2 during vegetative growth can complement the galactosylation defect of the ΔgalE1 mutant strain, in agreement with previous observations that both GalE1 and GalE2 display UDP-Glc 4-epimerase activity (19).

FIG 1.

Deletion of the galE1 gene abolishes SBA lectin and EAII6G6 binding to the galactosylated SCWP of B. anthracis. Vegetative forms of B. anthracis strain Sterne 34F2, its bas1093, galE2, ΔgalE1, ΔgalE1(pgalE1), and ΔgalE1(pgalE2) variants, and B. anthracis CDC684 were stripped of their S layers by treatment with 3 M urea. Bacilli were subsequently fixed with 4% paraformaldehyde and stained with monoclonal antibody EAII6G6-FITC or SBA lectin conjugated to Alexa Fluor 594. Samples were viewed with differential inference contrast (DIC) or fluorescence (EAII6G6 or SBA lectin) microscopy, and images were captured.

SCWP isolated from B. anthracis ΔgalE1.

Earlier work purified SCWPs from B. anthracis Sterne 34F2 and B. anthracis CDC684 and measured the molecular masses of the polysaccharides via size exclusion (SE) chromatography as 10,000 Da and 22,000 Da, respectively (15, 29). We wondered whether or not the lack of galactosylation affects the length of the SCWP. To address this question, SCWPs were released by hydrofluoric acid (HF) treatment of isolated murein sacculi of B. anthracis Sterne 34F2 and its ΔgalE1 and ΔgalE1(pgalE1) variants. HF treatment hydrolyzes the phosphodiester bond between O6 MurNac in peptidoglycan and O1 of the murein linkage unit [GlcNAc-(1→4)-ManNAc] for the SCWP (13). Acid-extracted polysaccharide was precipitated with ethanol and subjected to size exclusion high-performance liquid chromatography (SE-HPLC). SCWPs from Sterne 34F2 and its ΔgalE1 and ΔgalE1(pgalE1) variants eluted with similar retention times (Fig. 2) (30), indicating that the molecular weight of B. anthracis SCWP was not affected by deletion of galE1 and that all three strains produce SCWPs of similar size. As the molecular weight of the SCWP from the galE1 mutant is unchanged, the data suggest further that the overall length, i.e., the number of trisaccharide repeats in each polysaccharide molecule, is increased in the SCWP of the galE1 mutant compared to the numbers of repeats in the SCWPs of B. anthracis Sterne and its ΔgalE1(pgalE1) variant.

FIG 2.

Size exclusion chromatography of SCWPs from B. anthracis strains. Cultures of B. anthracis Sterne 34F2 and its ΔgalE1 and ΔgalE1(pgalE1) variants were used to isolate the murein sacculi of vegetative forms. Murein sacculi were treated with hydrofluoric acid to release SCWP for analysis by size exclusion high-performance liquid chromatography. Absorbance was measured at 206 nm (mAu, milli-absorbance units). Data are representative of two independent experiments.

Galactosylation of SCWP from B. anthracis.

The SCWPs of B. anthracis Sterne 34F2 and its ΔgalE1 and ΔgalE1(pgalE1) variants were analyzed by matrix-assisted laser desorption ionization (MALDI) mass spectrometry (MS). Ion signals m/z 1,136.6 and m/z 2,233.0 represent sodiated adducts of one or two SCWP repeating units, which were abundantly present in SCWP preparations from B. anthracis Sterne 34F2 and its ΔgalE1(pgalE1) variant but not in the SCWP preparation from the ΔgalE1 variant. Table 1 summarizes ion signals obtained from the SCWP of each strain. Numerous sodiated ion signals conforming to the variable lengths of the polysaccharide repeating units were detected in SCWP samples from B. anthracis Sterne 34F2 and ΔgalE1(pgalE1) strains. Less-abundant ions differed from those of the repeating units, in agreement with the general hypothesis of SCWP heterogeneity in galactosyl modification and acetylation. For example, m/z 1,500.55 conforms to a compound structure of 4 acetylated amino sugars and 3 decorating galactosyl sugars with one deacetylated amino sugar in the repeating sequence of the SCWP (Table 1). The ΔgalE1 strain SCWP generated very few ion signals during MALDI-time of flight (TOF) MS, with the predominant signal being m/z 649.9, which represents the sodiated ion of the nongalactosylated trisaccharide [ManNAc-GlcNAc₂]. These data are consistent with a model whereby ΔgalE1 mutant B. anthracis cannot synthesize UDP-Gal and is unable to promote galactosylation of the SCWP. In B. anthracis, only some SCWP molecules are acetylated and ketal-pyruvylated at their terminal repeat units (15). Our experiments did not detect acetylated and ketal-pyruvylated species in SCWP samples from the galE1 mutant strain. This was attributed to the increased size of the galE1 mutant SCWP, as only small polysaccharide fragments can be ionized in MALDI mass spectrometry experiments (13, 14). Nonetheless, nongalactosylated SCWP from the galE1 mutant must be acetylated and ketal-pyruvylated, as bacterial surface (S) layer assembly is not affected by the galE1 mutation (see below).

TABLE 1.

MALDI-TOF mass spectrometry of hydrofluoric acid-released SCWPs from B. anthracis Sterne 34F2 and its ΔgalE1 and ΔgalE1(pgalE1) variants^a

Proposed composition of SCWPb	Observed m/z in:			Theoretical m/z (monoisotopic)	Δm/z in:
	Sterne 34F2	ΔgalE1 variant	ΔgalE1(pgalE1) variant		Sterne 34F2	ΔgalE1 variant	ΔgalE1(pgalE1) variant
HexNAc3Na+	649.90	649.93	649.98	649.97	0.08	−0.04	−0.01
HexNAc3-Gal3Na+	1,137.07		1,136.56	1,136.40	−0.67		−0.16
HexNAc4-Gal3Na+	1,339.15			1,339.48	0.33
HexNAc3-GlcN-Gal4Na+	1,459.03		1,459.25	1,459.52	0.49		0.27
HexNAc4-GlcN-Gal3Na+	1,500.39c		1,500.37c	1,500.55	0.16		0.18
HexNAc5-Gal3Na+	1,542.84		1,542.74	1,542.55	−0.29		−0.19
HexNAc5-Gal32Na+	1,565.53		1,565.83	1,565.54	0.01		−0.28
HexNAc4-Gal52Na+	1,686.87		1,686.51	1,686.57	−0.30		0.06
HexNAc4-GlcN-Gal5Na+	1,824.19		1,824.23	1,824.65	0.46		0.42
HexNAc4-Gal6 2Na+	1,848.68		1,848.68	1,848.62	−0.06		−0.06
HexNAc5-Gal6Na+	2,028.76		2,028.42	2,028.71	−0.05		0.29
HexNAc6-Gal5Na+	2,070.29		2,070.24	2,069.74	−0.55		−0.50
HexNAc5-GlcN-Gal6Na+	2,190.23		2,190.36	2,189.78	−0.45		−0.58
HexNAc6-Gal6Na+	2,233.15c		2,233.57c	2,231.79	−1.36		−1.78
HexNAc9-Gal8Na+	3,165.93			3,165.13	−0.80
HexNAc9-Gal9Na+	3,327.97			3,327.18	−0.79

^aIon signals and proposed composition of compounds from MALDI-TOF mass spectra of RP-HPLC-purified SCWPs from B. anthracis Sterne 34F2 and its ΔgalE1 and complemented variants.

^bHexNAc, ManNAc (N-acetylmannosaminyl) and GlcNAc (N-acetylglucosaminyl); GlcN, glucosaminyl; Gal, galactosyl. The SCWPs isolated from the Sterne 34F2 and ΔgalE1(pgalE1) strains generated nearly identical sets of ions, similar to the spectra characterizing the SCWP structure. Of note, additional compositional explanations for the observed masses exist; however, they are not listed here.

^cObserved ion signal was accompanied by a less intense ion signal with an additional m/z of 15.9999 and was interpreted as an oxygen adduct of the parent ion.

S layer assembly in wild-type and ΔgalE1 B. anthracis.

The SCWP retains two S layer proteins, Sap and EA1, as well as 22 S layer-associated proteins (bacillus surface layer proteins [BSLs]) whose S layer homology (SLH) domains associate with ketal-pyruvylated and acetylated SCWP. Earlier work demonstrated the deposition of the EA1 S layer protein in the envelope of B. anthracis CDC684 (15) but left unanswered whether a lack of SCWP galactosylation affects the assembly of the Sap S layer and the deposition of S layer-associated proteins into the S layer compartment. Cultures of B. anthracis Sterne 34F2 and its ΔgalE1 and ΔgalE1(pgalE1) variants were fractionated into the extracellular medium (M), S layer (S), and cellular (C) compartments. Proteins in all fractions were separated by SDS-PAGE and analyzed by Coomassie blue staining or immunoblotting with rabbit antibodies raised against purified recombinant proteins (Fig. 3). As observed for wild-type B. anthracis Sterne 34F2, the S layer proteins were found in the extracellular medium and S layer fraction of the ΔgalE1 and ΔgalE1(pgalE1) variants, suggesting that S layer assembly was not affected by mutation in the galE1 gene. This conjecture was confirmed by immunoblotting for S layer proteins Sap and EA1, as well as S layer-associated proteins BslO, BslS, and BslT, whose distribution into the medium and S layer fractions was indistinguishable between wild-type and mutant strains. As a control, the lipoprotein PrsA fractionated with B. anthracis vegetative cells and was found neither in the medium nor in the S layer fractions (Fig. 3).

FIG 3.

Assembly of S layer and S layer-associated proteins in B. anthracis. Spores (1 × 10⁷/ml) derived from B. anthracis Sterne 34F2 and its ΔgalE1 and ΔgalE1(pgalE1) variants were germinated and grown in BHI broth with 0.8% bicarbonate at 37°C for 5.5 h. Cultures were fractionated into medium (M), S layer (S), and cellular (C) fractions, and samples were analyzed with Coomassie-stained 10% SDS-PAGE gel (top) or immunoblotting with rabbit antisera raised against purified Sap, EA1, BslA, BslO, BslS, BslT, and PrsA (bottom). Numbers to the left indicate the migratory positions of molecular mass markers in kilodaltons.

SCWP galactosylation affects murein hydrolase binding to B. anthracis.

Previous work demonstrated that bacteriophage lysins PlyG and PlyL associate with the B. anthracis SCWP through their C-terminal cell wall binding domain (CBD) (31, 32). Furthermore, SCWP galactosylation impacts the binding of PlyG and PlyL as measured with purified SCWPs from B. anthracis Sterne 34F2 and CDC684 strains and with in vitro-synthesized SCWP trisaccharides (33, 34). PlyG_CBD-mCherry, encompassing the CBD of PlyG fused to mCherry fluorescent reporter protein, was used to test whether SCWP galactosylation affects murein hydrolase binding to the envelope of B. anthracis (35). S layer-stripped vegetative forms of B. anthracis Sterne 34F2 and its ΔgalE1 and ΔgalE1(pgalE1) variants were incubated with purified PlyG_CBD-mCherry and analyzed by fluorescence microscopy or fluorescence quantification of PlyG_CBD-mCherry-decorated bacteria. Compared to the result for B. anthracis Sterne 34F2 (arbitrarily set as 100%), ΔgalE1 mutant bacilli bound fewer PlyG_CBD-mCherry molecules; this phenotypic defect was fully restored by transformation of the ΔgalE1 mutant with the galE1 expression plasmid in the ΔgalE1(pgalE1) strain (Fig. 4). As a control, B. cereus ATCC 14579, which synthesizes SCWP with a different structure, [→4)-β-ManNAc-(1→4)-β-GlcNAc(O3-β-GlcNAc)-(1→6)-α-GalNAc(O3-β-Glc, O4-α-ManNAc)-(1→] (12), does not bind to PlyG_CBD-mCherry, which is attributed to the altered repeat unit and the lack of galactosylation (Fig. 4) (33).

FIG 4.

PlyG_CBD-mCherry binding to the envelope of B. anthracis. Vegetative forms of B. anthracis Sterne 34F2, its ΔgalE1 and ΔgalE1(pgalE1) variants, and B. cereus ATCC 14579 were stripped of their S layers with 3 M urea. Stripped bacilli were then incubated with purified PlyG_CBD-mCherry, sedimented by centrifugation, and washed with buffer. (A) Bright-field (BF) and fluorescence (PlyG_CBD-mCherry) microscopy images were acquired. (B) Binding of fluorescent protein was assessed by fluorescence measurements. Arbitrary units of fluorescence were converted to percentages of binding. PlyG_CBD-mCherry binding to B. anthracis Sterne 34F2 was set as 100%. Mean values and associated standard errors were derived from 7 independent experiments, and statistical analysis was performed by one-way ANOVA and Tukey's post hoc analysis. *, P < 0.05; ***, P < 0.001.

galE1 or galE2 expression does not restore SCWP galactosylation in B. anthracis CDC684.

Genome sequencing revealed that B. anthracis CDC684 harbors a large chromosomal inversion and 51 single-nucleotide polymorphisms (SNPs) compared with the genome of its closest relative, the fully virulent B. anthracis strain Vollum (17). However, none of the 51 SNPs are located in galE1 or galE2. Nevertheless, we wondered whether the chromosomal inversion in B. anthracis CDC684 may affect the expression of galE1 and galE2 via an unknown mechanism and asked whether episomal expression of UDP-Glc 4-epimerase may restore SCWP galactosylation. The B. anthracis CDC684 strain was transformed with pgalE1 or pgalE2. The expression of galE1 and galE2 during vegetative growth was induced via the addition of IPTG, and fluorescence microscopy was used to detect SCWP galactosylation. However, compared to the results for the fully virulent isolate B. anthracis Ames strain, SBA lectin and the antibody EAII6G6 did not bind to B. anthracis CDC684, CDC684(pgalE1), or CDC684(pgalE2) (Fig. 5). Thus, the defect in SCWP galactosylation of B. anthracis CDC684 is not caused by a lack of functional UDP-Glc 4-epimerase expression.

FIG 5.

Plasmid-borne expression of galE1 or galE2 cannot restore SCWP galactosylation in B. anthracis CDC684. Vegetative forms of B. anthracis strains Ames, CDC684, CDC684(pgalE1), and CDC684(pgalE2) were stripped of their S layers with 3 M urea. Bacilli were fixed with 4% paraformaldehyde and incubated with monoclonal antibody EAII6G6-FITC or SBA lectin conjugated to Alexa Fluor 594. Samples were subjected to bright-field (BF) or fluorescence (EAII6G6 or SBA lectin) microscopy, and images were captured.

galE1 expression contributes to PDGA capsulation in B. anthracis Ames.

Following spore challenge via subcutaneous injection into guinea pigs, B. anthracis CDC684 is severely attenuated in its ability to cause anthrax disease (17). Nonetheless, strain CDC684 secretes anthrax toxins (lethal toxin and edema toxin) and elaborates poly-γ-d-glutamic acid (PDGA) capsule, a known antiphagocytosis factor of B. anthracis vegetative forms during replication in mammalian hosts (36, 37). B. anthracis capsulation requires the expression of the capBCADE operon on the pXO2 virulence plasmid, whose products promote the synthesis, transport, and assembly of PDGA (7). Toxin production is dependent on pXO1-carried genetic determinants (pagA, lef, and cya) for three secreted polypeptides (protective antigen [PA], lethal factor [LF] and edema factor [EF]) and on atxA, whose product functions as a transcriptional activator responding to environmental CO₂ signals (38, 39). However, B. anthracis CDC684 is endowed with wild-type pXO1 and pXO2 plasmids and its virulence defect is therefore not caused by mutational lesions in toxin and capsulation genes (17). We wondered whether a genetic lesion in galE1, which abrogates SCWP galactosylation, may also affect virulence.

The galE1 gene of B. anthracis Ames, a fully virulent isolate, was deleted and the ΔgalE1 mutant strain was transformed with pgalE1 as well as pgalE2. Fluorescence microscopy experiments revealed that the binding of SBA lectin and MAb EAII6G6 to the SCWP of B. anthracis Ames was abolished in the ΔgalE1 mutant strain. SCWP galactosylation in B. anthracis Ames ΔgalE1 was restored to wild-type levels, as indicated by SBA lectin and MAb EAII6G6 binding, following transformation with pgalE1 and pgalE2 (Fig. 6A). B. anthracis cultures were separated into medium, S layer, and cellular compartments, and the proteins in each compartment were analyzed by Coomassie-stained SDS-PAGE and immunoblotting. As observed for B. anthracis Sterne 34F2, the deletion of galE1 did not affect S layer and S layer-associated protein secretion and assembly. Furthermore, the production and secretion of anthrax toxin components (PA, LF, and EF) were not affected by the ΔgalE1 mutation (Fig. 6B). When observed via microscopy of India ink-stained samples, the capsule size of the ΔgalE1 variant grown on agar plates with bicarbonate appeared smaller than that of its B. anthracis Ames parent (Fig. 6C). Of note, the capsule size of B. anthracis CDC684 also appeared much smaller than that of B. anthracis Ames (Fig. 6C). When measured 3 and 5 h after germination of spores in capsule-inducing medium, the ΔgalE1 variant displayed diminished PDGA capsulation compared with that of wild-type B. anthracis Ames (Fig. 6D and E). The capsulation defect of the ΔgalE1 mutant was ameliorated by plasmid-borne expression of galE1 in the ΔgalE1(pgalE1) strain (Fig. 6D and E). Surprisingly, capsule production was not observed for B. anthracis CDC684 at the 3-, 5-, and 7-h time intervals after germination (Fig. 6D and data not shown). These data suggest that deletion of the galE1 gene in B. anthracis Ames negatively impacts PDGA capsulation. Furthermore, B. anthracis CDC684 elaborates less capsular material and is delayed in the production of PDGA capsule following spore germination compared to B. anthracis Ames.

FIG 6.

SCWP galactosylation impacts PDGA capsulation in B. anthracis Ames. (A) Vegetative forms of B. anthracis strain Ames and its ΔgalE1, ΔgalE1(pgalE1), and ΔgalE1(pgalE2) variants were stripped of their S layers with 3 M urea. Bacilli were fixed with 4% paraformaldehyde and incubated with monoclonal antibody EAII6G6-FITC or SBA lectin conjugated to Alexa Fluor 594. Samples were visualized by bright-field (BF) or fluorescence (EAII6G6 or SBA lectin) microscopy. (B) Vegetative forms of B. anthracis Ames and its ΔgalE1 mutant were grown in BHI broth with 0.8% bicarbonate at 37°C for 5.5 h. Cultures were fractionated into medium (M), S layer (S), and cellular (C) fractions, and samples were analyzed with Coomassie-stained 10% SDS-PAGE gel (top) or immunoblotting with rabbit antisera raised against purified PA, LF, EF, Sap, EA1, BslA, BslO, and BslS (bottom). Numbers to the left indicate the migratory positions of molecular mass markers in kilodaltons. (C) Light microscopy images of formalin-fixed B. anthracis Ames, its ΔgalE1 variant, and B. anthracis CDC684 stained with India ink after overnight growth on agar plates with 5% CO₂ atmosphere. (D, E) Capsule sizes of B. anthracis Ames, its ΔgalE1 and ΔgalE1(pgalE1) variants, and B. anthracis CDC684 at 3 and 5 h postgermination were measured from bright-field micrographs of vegetative bacilli (n = 100). Representative images (D) and corresponding box-and-whisker plots of the capsule widths (E) are shown. Statistical analysis was performed by one-way ANOVA and Tukey's post hoc analysis. ***, P < 0.001. Data are representative of two independent experiments.

galE1 expression contributes to B. anthracis virulence in a mouse model of anthrax disease.

Cohorts of C57BL/6 mice (n = 10) were injected subcutaneously with 10² and 10³ spores derived from either wild-type B. anthracis Ames, i.e., approximately 3 and 30 lethal dose equivalents (40), or the B. anthracis Ames ΔgalE1 variant or with B. anthracis CDC684 (only 10³ spores). All animals challenged with 10³ wild-type spores succumbed to infection within 28 to 64 h (median time of survival, 45 h), whereas animals challenged with 10³ ΔgalE1 spores succumbed to infection within 39 to 64 h (median time of survival, 64 h) (Fig. 7A). When these data were analyzed with the log-rank test, a significant difference in survival between wild-type and galE1 variant challenge groups was recorded (P = 0.0325). All animals challenged with 10² Ames wild-type spores succumbed to infection within 47 to 67 h (median time of survival, 64 h), whereas only 50% of animals challenged with 10² Ames ΔgalE1 spores succumbed to infection within 64 to 87 h (median time of survival, 175 h; P < 0.0001 for wild-type versus galE1) (Fig. 7A). All animals that were infected with spores of B. anthracis CDC684 strain survived the challenge.

FIG 7.

Attenuation of ΔgalE1 mutant of B. anthracis Ames in a mouse model of anthrax. (A) Survival of cohorts (n = 10) of C57BL/6 mice following subcutaneous inoculation of 10² or 10³ spores derived from B. anthracis Ames, its ΔgalE1 mutant, or B. anthracis CDC684. B. anthracis CDC684 was avirulent at a challenge dose of 10³ spores (wild type versus CDC684, log rank test, P < 0.0001). B. anthracis ΔgalE1 virulence was attenuated at challenge doses of 10² and 10³ spores (10³ spores, wild type versus ΔgalE1, log rank test, P = 0.0325; 10² spores, wild type versus ΔgalE1, log rank test, P < 0.0001). Data are representative of two independent experiments. (B) Light microscopy images of formalin-fixed and India ink-stained spleen tissues from mice infected with B. anthracis Ames or its ΔgalE1 mutant.

Symptomatic infected animals were euthanized and necropsied, and their spleens homogenized, serially diluted, and plated on agar medium for colony formation to enumerate bacterial loads. Animals infected with wild-type or B. anthracis Ames ΔgalE1 strains contained similar numbers of vegetative bacilli in their spleens: ≥8 × 10⁶. Bacilli from spleen homogenates were also visualized after India ink staining (Fig. 7B). Compared to the capsulation in the wild-type parent B. anthracis Ames, the capsule size in the ΔgalE1 mutant strain appeared diminished, similar to what was observed during in vitro growth.

DISCUSSION

When comparing B. cereus sensu lato members, it is notable that the SCWPs vary between species in the amino sugars in the trisaccharide repeats, the modifying hexoses and hexosamines, and the glycosidic linkage bonds and attachment sites (11, 12, 29, 41). Using comparative genome analysis, Schuch and colleagues identified the sps locus, whose content of 15 to 20 genes varies among members of the B. cereus sensu lato group (23). The DNA sequence of the sps locus displays G+C content that differs from conserved genome sequences of B. cereus species, suggesting that sps genes have been acquired via horizontal transfer, presumably enabling adaptive evolution of bacilli to fit specific ecological niches (23). The gneZ gene is located within the sps locus and encodes UDP-GlcNAc 2-epimerase, catalyzing the reversible conversion of UDP-GlcNAc and UDP-ManNAc and providing ManNAc substrate for the synthesis of both murein linkage units and SCWP (24). Another gene, gneY, also encodes functional UDP-GlcNAc 2-epimerase and is located within a locus that provides enzymes for the synthesis of murein linkage units (24, 35). However, during vegetative growth, only gneZ is expressed and gneZ depletion abolishes SCWP synthesis and stalls bacterial cell division (42). Similarly, depletion of tagO, a gene responsible for the synthesis of the SCWP linkage unit, halts bacterial FtsZ Z ring assembly and subsequent cell division plane formation and cellular replication (35). Thus, SCWP synthesis in B. anthracis is essential for cell division and vegetative growth (30).

Similar to UDP-GlcNAc 2-epimerase, the genome of B. anthracis also includes two genes, galE1 and galE2, for UDP-Glc 4-epimerase, an enzyme that catalyzes the reversible interconversion of UDP-Glc and UDP-Gal (19). One of these two genes, galE2, is expressed only during B. anthracis sporulation, providing UDP-Gal and UDP-GalNAc substrates for the glycosylation of BclA, an exosporium protein in the outer spore nap that is produced by all members of the B. cereus sensu lato group (21, 43–45). In contrast, galE1 is only expressed during vegetative growth and provides UDP-Gal for the galactosylation of the SCWP in B. anthracis. Unlike genes located within the sps locus, galE1 is conserved among all members of the B. cereus sensu lato group whose SCWPs contain galactosyl modifications (18). The lcpC gene, which is located immediately adjacent to galE1 within the sps locus, is conserved in a manner similar to galE1. The product of lcpC is thought to attach the murein linkage units of SCWP to bacterial peptidoglycan (46, 47).

Unlike the SCWP trisaccharide repeat structure or its linkage to peptidoglycan, the galactosyl modifications of the polysaccharide are not required for bacterial cell division or vegetative growth of B. anthracis. Furthermore, SCWP galactosylation is dispensable for S layer assembly and function, which includes S layer protein-mediated control of cell-cell separation and cell chaining of vegetative forms (48, 49), but it is required for the proper functioning of PlyG and PlyL murein hydrolases, which are dependent on the β-Gal modification at O4 of α-GlcNAc in the repeating trisaccharide (34). In agreement with this conjecture, PlyG, which is encoded by the γ-phage, and PlyG_CBD-mCherry, a hybrid of the SCWP binding domain of PlyG and mCherry, bind to SCWPs from B. anthracis strains Sterne 34F2 and 7702 and B. cereus G9241 but not to SCWPs from B. cereus ATCC 10987 and ATCC 14579 (33).

Surprisingly, the galE1 mutant displayed diminished PDGA capsulation, a key virulence trait of B. anthracis (50). The molecular basis for this phenotypic defect is not yet resolved. PDGA is attached to the ε-amino group of m-Dpm within the wall peptide of the bacterial peptidoglycan and threaded across the SCWP and the S layer of B. anthracis (6, 51, 52). In contrast, SCWP is attached via murein linkage units and phosphodiester bonds to the O6 of MurNAc within the glycan strands of peptidoglycan (46, 47). It seems plausible that changes in the chemical nature of the SCWP, i.e., lack of galactosylation, may either impact PDGA's attachment to peptidoglycan or affect the threading of PDGA strains across the multiply layered envelope of B. anthracis for functional display on the bacterial surface. We favor the latter hypothesis for the following reason.

The SCWPs from B. anthracis Ames, Pasteur, and Sterne 34F2 strains, as well as SCWPs from closely related B. cereus G9241, 03BB87, and 03BB102 strains that are pathogenic for mammals, represent short polysaccharides that are uniform in size, about 10 repeating trisaccharides in length (10 to 12 kDa) (29, 53, 54). Size exclusion chromatography experiments measured 10 kDa for nongalactosylated SCWP from ΔgalE1 mutant B. anthracis, a length of 16 trisaccharide repeats (Fig. 2) (15). The SCWP backbone of B. anthracis CDC684 (22 kDa) is even longer, 33 trisaccharide repeats (15). These data suggest that galactosylation of trisaccharide precursors may be a length determinant during the synthesis of SCWP. An increased length of SCWP may impact the synthesis or the function of PDGA capsule strands that are threaded through the SCWP and the S layer. PDGA requires surface display to prevent complement deposition via lectin binding to bacterial surfaces, which otherwise triggers opsonophagocytosis and phagocyte clearance of bacteria (55, 56). Increased SCWP chain length may diminish the synthesis or the functional assembly of PDGA capsule strands and may thereby weaken the protective effects of PDGA capsule strands when bacilli reside in host tissues. Whatever the mechanism, deletion of the galE1 gene diminished PDGA capsulation and the virulence of B. anthracis Ames in a mouse model of anthrax disease. The ΔgalE1 mutant of B. anthracis Ames was more virulent than B. anthracis CDC684, which failed to cause anthrax disease in mice and, as had already been reported, is also severely attenuated in guinea pigs (17). We presume the lack of virulence of B. anthracis CDC684 may not be based solely on its defect in galactosylation of the SCWP and may also be caused by the slowed growth phenotype observed in broth cultures of this strain, which has been attributed to its large chromosomal inversion (17).

Taken together, this study shows that galactosylation of the SCWP contributes to regulating the length of the wall polysaccharide, i.e., the number of trisaccharide repeats in each molecule. Furthermore, the length of SCWP molecules appears to impact PDGA capsulation in B. anthracis vegetative forms, which is a key virulence factor during the pathogenesis of anthrax disease. Future work will need to unravel the molecular link between SCWP galactosylation and PDGA capsulation in B. anthracis.

MATERIALS AND METHODS

Bacterial growth and reagents.

B. anthracis strains Sterne 34F2, Ames, and CDC684 and their variants were grown in brain heart infusion (BHI; BD) broth or agar at 30 to 40°C. E. coli was grown in Luria-Bertani (LB; BD) broth or agar at 37°C. Where necessary, kanamycin (Kan; Fisher Scientific) was added at a concentration of 20 μg ml⁻¹ for B. anthracis and at 50 μg ml⁻¹ for E. coli. B. anthracis strains were sporulated in modified G medium (57). Spore preparations were heat treated at 68°C for 2 h to kill vegetative bacilli and then stored at 4°C. To monitor bacterial growth, spores were germinated by inoculation into BHI at the temperatures indicated, with or without appropriate antibiotics. Bacterial growth was monitored at timed intervals by recording the absorbance at 600 nm (A₆₀₀). To induce capsule production, bacteria were grown on nutrient broth yeast extract (NBY) agar supplemented with 0.8% sodium bicarbonate and 10% heat-inactivated horse serum (Gibco). Alternatively, spores were germinated in NBY medium supplemented with 0.6% sodium bicarbonate and 10% heat-inactivated horse serum at 37°C in 5% CO₂. To visualize capsulation, India ink was added to samples prior to microscopy imaging.

B. anthracis strains and plasmids.

The alleles of BAS1093::spec and galE2::spec were obtained by transposon mutagenesis with bursa aurealis (25). Bacteriophage CP51 was used to transduce the spectinomycin-marked mutations into B. anthracis Sterne 34F2 (50). Deletion of galE1 was achieved by unmarked allelic replacement using the temperature-sensitive vector pLM4 (58). Briefly, 1-kbp flanking DNA sequences upstream and downstream from the galE1 gene were amplified with oligonucleotide primers (Table 2); PCR products were cut with restriction enzymes and ligated into pLM4 cut with the same enzymes. The recombinant plasmid was electroporated into B. anthracis strains Sterne 34F2 and Ames as previously described (58). Allelic replacement was induced via changes in incubation temperature and verified by PCR using specific oligonucleotide primers (Table 2). For complementation studies, pgalE1 and pgalE2 were constructed in pJK4 from PCR products (primer sequences are in Table 2); the expression of galE1 and galE2 from the P_spac promoter of pgalE1 and pgalE2 was induced with 0.1 mM isopropyl-β-d-thiogalactopyranoside (IPTG) (16).

TABLE 2.

Oligonucleotides used in this study

Oligonucleotide	Sequence
BAS5114-1000kb-upFW	AAAGAGCTCGGGTACTCGGTATCCTTGG
BAS5114-1000kb-upRV	AAAGCTAGCCGCCGCAGATTAGAATTGAATTC
BAS5114-1000kb-dnFW	AAAGCTAGCGGCATCAGAAGCAACCTAATG
BAS5114-1000kb-dnRV	AAACCCGGGCAAAAGAATTGGAGCCGGG
BAS5114BF	GTGCTAAAGGTCACGAAGTTTGG
BAS5114AR	TCTCCTCCTTCCATCCCGTA
5114-pLM5-F	AAATCTAGAATGAATTCAATTCTAATCTGCGGCGGA
5114-pLM5-R	AAAGGTACCTTATTTCTCATACCCATTAGGTTGCTTC
BAS5304-pJK4F	AAATCTAGAATGGCGATACTTATAACGGGTGGA
BAS5304-pJK4R	AAAGGTACCTAAATCATTTGATAGCCATTTTTATTATTTACTTGCCATCTCC

Immunofluorescence labeling of B. anthracis.

B. anthracis bacteria grown overnight on BHI plates at 37°C were scraped off the agar surface, suspended and washed once in phosphate-buffered saline (PBS), and boiled at 95°C for 10 min in PBS–3 M urea to extract the S layer. The bacteria were then sedimented by centrifugation for 1 min at 16,000 × g, and the sediment washed twice with PBS and fixed with formalin. Bacteria were again washed 3 times with PBS and stained as follows. Four microliters of bacterial suspension was incubated for 30 min with 1 μl of FITC-labeled monoclonal antibody EAII6G6 or lectin soy bean agglutinin (SBA) conjugated to Alexa Fluor 594 (50 μg/ml) (Thermo Fisher Scientific). Two-microliter samples were placed on a slide with SlowFade gold (Thermo Fisher Scientific), and a coverslip was added. Images were captured with a charge-coupled device (CCD) camera on an Olympus IX2-UCB microscope using a 100× objective.

Analysis of SCWPs.

Bacilli were grown overnight at 37°C on BHI agar, scraped off the agar surface, suspended in 25 ml water, and sedimented by centrifugation for 10 min at 6,000 × g. Bacteria were washed once in water, suspended in 400 ml 4% SDS solution, boiled for 30 min, washed, and suspended in water. The bacteria were mechanically lysed with 0.1-mm glass beads. The resulting murein sacculi were sedimented by centrifugation at 17,000 × g for 15 min, suspended in 100 mM Tris-HCl (pH 7.5), and incubated for 4 h at 37°C with 10 μg/ml DNase and 10 μg/ml RNase supplemented with 20 mM MgSO₄. Samples were incubated for 16 h at 37°C with 10 μM trypsin supplemented with 10 mM CaCl₂. Enzymes were inactivated by boiling for 30 min in a water bath in 1% SDS. The SDS was removed by 5 cycles of centrifugation and washing in water. The murein sacculi were then washed with water, 100 mM Tris-HCl (pH 8.0), water, 0.1 M EDTA (pH 8.0), water, acetone, and finally twice with water. Murein sacculi were suspended in 5 ml of water, and 25 ml of 48% hydrofluoric acid (HF) was added. Samples were incubated overnight on ice with shaking. The acid-extracted murein sacculi were sedimented by centrifugation at 17,000 × g for 15 min. SCWP-containing supernatant was mixed with ice-cold ethanol in a 1:5 ratio, causing SCWP precipitation. The polysaccharide was sedimented by centrifugation at 17,000 × g at 4°C for 15 min, and the sediment was washed extensively with ice-cold ethanol. The SCWP was suspended in water at a concentration of 100 mg/ml, and 100 μl of this solution was subjected to reversed-phase high-performance liquid chromatography (RP-HPLC) analysis.

RP-HPLC analysis was performed over a 250-mm by 4.6-mm C₁₈ octadecylsilyl (ODS) Hypersil column with a 3-μm-particle-size guard column (Thermos) using a water-acetonitrile gradient where the mobile phase was supplemented with 0.1% trifluoroacetic acid (TFA). The gradient was derived from buffer A (water–0.1% TFA) and buffer B (acetonitrile–0.1% TFA), with a continuous flow rate of 0.5 ml per minute and the following parameters: 0 to 10 min, 0.1% buffer B; 11 to 15 min, linear gradient of 0.1 to 10% buffer B; 16 to 35 min, linear gradient of 10 to 20% buffer B; 36 to 100 min, linear gradient of 20 to 99.9% buffer B; and 101 to 110 min, 99.9% buffer B. The eluate was monitored for absorbance at 206 nm, and 0.5-ml fractions were collected. SCWP was also subjected to size exclusion high-performance liquid chromatography (SE-HPLC) on a 300-mm by 7.8-mm BioBasic SEC 300 (Thermo Fisher Scientific) column equilibrated with 50 mM sodium phosphate buffer (pH 7.5) at a 1-ml/min flow rate. The absorbance at 206 nm was monitored to assess the retention time of SCWP material.

MALDI-TOF mass spectrometry.

RP-HPLC fractions were spotted onto a Prespotted AnchorChip II (PAC II) plate (Bruker) that contained an α-cyano-4-hydroxycinnamic acid (HCCA) matrix. Calibrants were used directly from the PAC II plate. The samples were subjected to matrix-assisted laser desorption ionization–time of flight (MALDI-TOF) mass spectrometry using an autoflex speed Bruker MALDI instrument in positive-reflectron mode. The predicted molecular weights of the various saccharides were calculated using the following average incremental values based on the atomic weights of the elements: hexose, 162.142; 2-N-acetamido-2-deoxyhexose, 203.195; and free reducing end, 18.0153.

S layer fractionation.

B. anthracis spores were germinated in BHI broth in the presence or absence of 0.8% bicarbonate. Cultures were centrifuged for 5 min at 16,000 × g and separated into medium (supernatant) and pellet fractions. Proteins in the medium were precipitated with 10% (vol/vol) trichloroacetic acid (TCA) for 30 min on ice and centrifuged at 16,000 × g for 10 min. Bacterial sediments (pellets) were washed twice with PBS and boiled at 95°C for 10 min in 100 μl of PBS–3 M urea to extract S layer and S layer-associated proteins. Extracts were centrifuged at 16,000 × g, and the S layer extract was separated with the supernatant from the bacterial sediment. S layer extract was added to an equal volume of sample buffer (4% SDS, 1% β-mercaptoethanol, 10% glycerol, 50 mM Tris-HCl [pH 7.5], 0.01% bromophenol blue). The bacterial sediment was washed twice with PBS and mechanically lysed by silica bead beating for 5 min (FastPrep-24; MP Biomedical), except for B. anthracis Ames and variants, whose bacterial sediments were directly precipitated with TCA. After sedimentation of the beads, proteins in the cell lysates were precipitated with TCA and sedimented by centrifugation at 16,000 × g for 10 min. All TCA precipitates were washed with ice-cold acetone and dried. Samples were suspended in 100 μl of 1 M Tris-HCl (pH 8.0)–4% SDS and mixed with an equal volume of sample buffer. Aliquots (10 μl) of each sample were separated by 10% SDS-PAGE and analyzed by Coomassie staining or electrotransferred to polyvinylidene difluoride (PVDF) membranes for immunoblot analysis. Proteins were detected with rabbit antiserum raised against purified antigens. Immunoreactive products were revealed by chemiluminescent detection after incubation with horseradish peroxidase (HRP)-conjugated secondary antibody (Cell Signaling Technology).

PlyG_CBD-mCherry.

B. anthracis bacteria grown overnight on BHI plates at 37°C were scraped off the agar surface, suspended and washed once in PBS, and boiled at 95°C for 10 min in PBS–3 M urea to extract the proteins in the S layer. The bacteria were then centrifuged for 1 min at 16,000 × g, washed twice with PBS, and normalized by optical density (A₆₀₀). One-hundred-microliter amounts of bacterial cell suspensions were incubated with 1 μl of purified PlyG_CBD-mCherry overnight at 4°C as previously described (35). Bacteria were sedimented by centrifugation, washed two times with PBS, and either imaged by fluorescence microscopy or had their fluorescence intensity measured in 96-well plates with a BioTek Synergy HT microplate reader. The excitation and emission wavelengths were converted to percentage of binding; data obtained with wild-type B. anthracis were arbitrarily set at 100%. Statistical analysis of data was performed by one-way analysis of variance (ANOVA) and Tukey's post hoc analysis.

Animal challenge.

Protocols for animal experiments were reviewed, approved, and supervised by the Institutional Animal Care and Use Committee (IACUC) at the University of Chicago. All experiments involving B. anthracis Ames, B. anthracis CDC684, and their variants, including animal infections, were carried out in biological safety level 3 and animal biological safety level 3 containment laboratories at the Howard Taylor Ricketts Laboratory. The University of Chicago Select Agent Program is approved and routinely inspected by both the Institutional Biosafety Committee and Centers for Disease Control and Prevention (CDC) officials. Six-week-old female C57BL/6 mice (Jackson Laboratory), in cohorts of 10 animals, were injected subcutaneously with 100 μl of a suspension of 10² or 10³ spores diluted in PBS. Animal health was subsequently monitored at 8-h intervals. Moribund animals were euthanized by CO₂ asphyxiation and cervical dislocation. Statistical analyses between surviving groups were performed with the log-rank test. Spleen samples isolated during necropsy were homogenized in PBS, serially diluted, and plated in duplicate on agar medium to enumerate bacterial loads in CFU. Alternatively, samples were fixed with neutral buffered formalin and stained with India ink to visualize capsulation of bacilli.

B. anthracis capsulation.

Capsule measurements were determined from micrographs using ImageJ 1.40g software (https://imagej.nih.gov/ij/). A total of 100 individual cells were measured for each bacterial strain analyzed. Four measurements were made per cell and averaged to generate a single average capsule width, which was converted to length in micrometers with objective micrometer reference images. Statistical analysis of data was performed by one-way ANOVA and Tukey's post hoc analysis.