Sortase C-Mediated Anchoring of BasI to the Cell Wall Envelope of Bacillus anthracis

Abstract

Vegetative forms of Bacillus anthracis replicate in tissues of an infected host and precipitate lethal anthrax disease. Upon host death, bacilli form dormant spores that contaminate the environment, thereby gaining entry into new hosts where spores germinate and once again replicate as vegetative forms. We show here that sortase C, an enzyme that is required for the formation of infectious spores, anchors BasI polypeptide to the envelope of predivisional sporulating bacilli. BasI anchoring to the cell wall requires the active site cysteine of sortase C and an LPNTA motif sorting signal at the C-terminal end of the BasI precursor. The LPNTA motif of BasI is cleaved between the threonine (T) and the alanine (A) residue; the C-terminal carboxyl group of threonine is subsequently amide linked to the side chain amino group of diaminopimelic acid within the wall peptides of B. anthracis peptidoglycan.

During infection, the cell wall envelope of gram-positive bacteria functions as a surface organelle for microbial interaction with host tissues (34). Many pathogenic strategies of bacteria utilize surface proteins to interact with extracellular matrices, specific host molecules, or target cells and thereby enable bacterial adherence to tissues, target cell invasion, or evasion of immune responses (14, 41). Many, but not all, surface proteins of gram-positive bacteria are anchored to the cell wall envelope by a mechanism requiring a C-terminal sorting signal with an LPXTG motif (10, 46, 47). These surface proteins are synthesized as precursors with N-terminal signal peptides in the bacterial cytoplasm (P1 precursor) (1, 5). After membrane translocation and signal peptide cleavage, the C-terminal sorting signal first retains the polypeptide in the cytoplasmic membrane (P2 precursor) (46). Membrane-anchored sortase A then cleaves the sorting signal between the threonine (T) and the glycine (G) residues (29, 33), generating an acylated enzyme with a thioester bond between the sortase active-site thiol group and the C-terminal carboxyl group of threonine (60, 61). Sortase acyl enzyme intermediates are resolved by nucleophilic attack of the amino group of cross-bridges within wall peptides, thereby anchoring the C terminus of surface proteins to the cell wall envelope (35, 58, 61, 62).

The cell wall of gram-positive bacteria is composed of peptidoglycan (43), a heteropolymeric macromolecule comprised of glycan strands and attached wall peptides (3, 16, 54, 55). In Bacillus species, including B. anthracis, glycan strands consist of the repeating disaccharide N-acetylmuramic acid-(β1-4)-N-acetylglucosamine (MurNAc-GlcNAc) with an average length of 10 to 20 disaccharide subunits (19, 64). Depending on the species, various degrees of hexosamine deacetylation have been observed (4, 17, 49). For example, the genome of B. anthracis encodes six GlcNAc deacetylases that together deacetylate 88% of GlcNAc residues in peptidoglycan to glucosamine (Glc_NH2) (39, 66). In mature peptidoglycan a short peptide component, wall tetrapeptide (l-Ala-d-iGln-m-Dap-d-Ala) or tripeptide (l-Ala-d-iGln-m-Dap), is attached to the glycan strands via an amide bond between l-Ala and the lactyl moiety of MurNAc (49, 64). Some of the wall peptides within assembled peptidoglycan are cross-linked to neighboring wall peptides, i.e., the side chain amino group of meso-diaminopimelic acid (m-Dap) at position 3 of the wall peptide is amide linked to the carboxyl groups of d-Ala at position 4 of the neighboring wall peptide (49, 64). The degree of cross-linking can vary between Bacillus species and growth stages of bacilli. For vegetative forms of B. anthracis, 19% of m-Dap side chains in wall peptides are engaged in cross-linking (49).

B. anthracis encodes three sortases (26, 40). Sortase A cleaves the LPXTG motif sorting signals of seven surface proteins destined for the cell wall envelope of vegetative bacilli (15). Sortase B recognizes the NPKTG motif of BasK (IsdC), a heme-binding protein involved in uptake of iron during infection (24). Sortase C, whose genetic determinant (srtC) is located in the basI-srtC-sctR-sctS operon, is required for the formation of infectious spores in tissues of animals that have succumbed to anthrax disease (28). A hallmark of sporulation in Bacillus spp. is the formation of an asymmetrically positioned (polar) septum that divides the developing cell into dissimilarly sized progeny called the mother cell and the forespore, the smaller cell that develops into a mature spore (22, 36). basI specifies a protein substrate of sortase C with an LPNTA-motif sorting signal, whereas sctR-sctS encodes for a two-component regulatory system (28). A second gene specifying a polypeptide with a LPNTA-motif sorting signal, basH, is located elsewhere on the B. anthracis genome and is expressed by σ^F RNA polymerase only within the developing forespore (28). The srtC operon is regulated by the response regulator sctR, which activates srtC and basI expression 2 h prior to polar septation during the developmental program that leads to spore formation (28). Previous work indicated that srtC is expressed in the predivisional sporulating cell, and its protein product is at least in part inherited by both mother cell and spore envelopes (28). SrtC within forespores is responsible for the anchoring of BasH to the peptidoglycan of this compartment (28). This work left unresolved whether sortase C also anchors BasI to the envelope of forespores. Here we examined the subcellular location and anchor mechanism of BasI. In contrast to BasH, BasI is anchored to the predivisional cell envelope during sporulation and hence is not present in the forespore peptidoglycan. Anchoring of BasI to the cell wall absolutely requires its LPNTA motif sorting signal and the active-site cysteine of SrtC. Sortase C cleaves the LPNTA motif of BasI between the threonine (T) and the alanine (A) residues, and the C-terminal carboxyl group of threonine is amide linked to the side chain amino group of diaminopimelic acid within the wall peptides of B. anthracis peptidoglycan.

MATERIALS AND METHODS

Bacterial strains and plasmids.

B. anthracis strain Sterne (53) and mutant derivatives were used throughout the present study. Mutants were generated with the pLM4 shuttle vector as previously described (28). B. anthracis strain LAM5 contains an in-frame deletion of the srtC gene; its construction was also described previously (28). Strain LAM17 harbors mutations in sequences encoding the BasI LPNTA motif, changing it to TNLAP; pLM224 was used to introduce these modifications. Strain LAM18 harbors a mutation in srtC codon 181 that replaces the wild-type cysteine codon with an alanine codon (C181A); pLM225 was used to introduce the mutation. In strain LAM20, basI was modified by translational fusion with the cherry open reading frame inserted between basI codons 110 and 111, three codons upstream of the sequence that encodes BasI LPNTA motif; pLM232 was used to generate this strain.

B. anthracis chromosomal DNA was extracted with the Wizard Genomic DNA purification kit (Promega). Table 1 lists the primers used for plasmid construction. In order to introduce point mutations in B. anthracis, the basI-srtC-sctR region was amplified from B. anthracis Sterne chromosomal DNA with the P12 and P13 primers and cloned into pLM4 by using EcoRI and XmaI restriction sites. This generated pLM221, which was used as a template to introduce mutations by site-directed mutagenesis. PCR amplification with the primers P46 and P47 using pLM221 as a template introduced mutations in the BasI LPNTA motif, thereby generating pLM224. PCR amplification with the primers P81 and P82 introduced the C181A mutation in srtC, generating pLM225. To introduce a methionyl-hexahistidyl tag into BasI, DNA specifying basI promoter and an open reading frame up to three codons upstream of BasI LPNTA motif was amplified with P3/P4 primers, and the sequences encoding for the basI cell wall sorting signal, srtC and sctR, were amplified with the P5 and P13 primers, where P5 contained an appended sequence encoding for the MH₆ tag (see Table 1). Both PCR products were cloned into pLM4 by using EcoRI, KpnI, and XmaI restriction sites, thereby generating pLM228. To generate a BasI-Cherry fusion, the cherry coding sequence was amplified with the primers P158 and P153 using pRSET-B-mCherry (50) as a template and cloned into pLM228 using KpnI; this generated pLM232. The construction of plasmids pLM218 and pLM226 is described elsewhere (28). PCRs were performed with Pfu DNA polymerase (Stratagene). Ligation products were transformed into E. coli K1077 Δ(dam dcm), and purified (nonmethylated) plasmid DNA was transformed into B. anthracis strains according to a previously developed protocol (20). Growth media were supplemented with kanamycin at 50 or 20 μg/ml to provide for plasmid selection in E. coli and B. anthracis, respectively.

TABLE 1.

Primers used in this study

Primer	Restriction site	Nucleic acid sequence (5′-3′)a
P3	KpnI	aaGGTACCcttatcgtcgtcatccttgtaatcttgtacagctggctttgctac
P4	EcoRI	aaGAATTCcaatgaaatggagtaagtggag
P5	KpnI	aaGGTACCatgcatcaccatcaccatcacggtgaaaaattaccaaacactg
P12	EcoRI	aaGAATTCaggtttgaagaagagacccc
P13	XmaI	aaCCCGGGgatctaatgtcgtaatccacg
P46	None	gctgtacaaggtgaaaaaACaAACCTcGctCcatcaaataatgtagcaatg
P47	None	cattgctacattatttgatgGagCgAGGTTtGTtttttcaccttgtacagc
P81	None	cgatactaacactgacgactGCttacccatttgattatatagg
P82	None	cctatataatcaaatgggtaaGCagtcgtcagtgttagtatcg
P153	KpnI	aaaGGTACCcttgtacagctcgtccatgc
P158	KpnI	aaaGGTACCagcaagggcgaggaggataac
P3784-N	NcoI	aaCCATGGatggaaatcagaaaaaaattagttg
P3784-C	BamHI	aaGGATCCttatttatcatcataccaccaatc

^aCapital letters indicate the sequence of the restriction site inserted. For the sequences with no restriction site, small capital letters indicate nucleotide changes introduced. Underlined nucleotides encode for the MH₆ tag.

pJB1 was used for expression of an untagged version of PlyL amidase (23). It was constructed by cloning the P3784-N/P3784-C PCR product into pET16b by using the NcoI and BamHI restriction sites. Ligation products were used to transform E. coli BL21(DE3) (Novagen) using 100 μg of ampicillin/ml for selection.

Microscopy.

B. anthracis LAM20(pLM218) overnight cultures were grown in brain heart infusion, diluted 1:10 into modG sporulation medium (21), and incubated at 30°C for 30 h. Culture aliquots were removed in hourly intervals and viewed by differential interference contrast or fluorescence microscopy using an Olympus Provis axial microscope (1,000-fold magnification) equipped with a Hamamatsu charge-coupled device camera.

Solubilization and detection of BasI.

B. anthracis strains harboring pLM218 or pLM228 were grown at 30°C up to T₋₁ in modG media (see Results). Strains harboring pLM226 were grown overnight at 30°C in LB broth either in the presence or in the absence of 1 mM IPTG (isopropyl-β-d-thiogalactopyranoside). Cells were collected by centrifugation (1 min, 16,000 × g), washed with 1 ml of sterile double-distilled water (ddH₂O), suspended in 250 μl of UDS buffer (6 M urea, 1 mM dithiothreitol, 1% sodium dodecyl sulfate [SDS], 50 mM Tris-HCl [pH 9.5]) (48), and boiled for 10 min. B. anthracis murein sacculi were sedimented by centrifugation (1 min, 16,000 × g) and separated from the bacterial extract containing membrane and cytoplasmic proteins. Murein sacculi were washed with 1 ml of sterile ddH₂O, suspended in 500 μl of 5% trichloroacetic acid, and boiled for 6 min. Precipitated material was centrifuged (1 min, 16,000 × g) and washed first with 500 μl of 0.5 M Tris-HCl (pH 6.3) and then twice with 50 mM Tris-HCl (pH 6.3)-1.5 mM MgCl₂. Murein sacculi containing peptidoglycan and secondary wall polymers were subjected to mutanolysin digestion in 250 μl of buffer containing 50 mM Tris-HCl (pH 6.3), 1.5 mM MgCl₂, 1 mM phenylmethylsulfonyl fluoride, and 100 U of mutanolysin (Sigma). Digestion was carried out overnight at 37°C, and insoluble material was removed by centrifugation (1 min, 16,000 × g). The supernatant constituted the cell wall fraction and harbored mutanolysin-solubilized proteins. Aliquots (5 μl) of cell wall and bacterial extract samples were subjected to SDS-polyacrylamide gel electrophoresis (PAGE) and electrotransferred to polyvinylidene difluoride membrane. BasI was detected by immunoblotting using a polyclonal αBasI rabbit serum (28).

PlyL digestion of B. anthracis cell walls.

Four liters of B. anthracis Sterne (pLM228) cells were harvested at T₋₁ during sporulation and peptidoglycan was purified as previously described for SEB-MH₆-CWS_IsdC (27). Sedimented murein sacculi were suspended in 30 ml of an E. coli BL21(DE3)(pJB1) lysate. The lysate was obtained by growing cells in 1 liter of LB medium supplemented with 100 μg of ampicillin/ml to an optical density at 600 nm of 0.7, when the expression of amidase was induced by the addition of 1 mM IPTG. Cells were grown for 2 additional hours, collected by centrifugation (5 min, 6,000 × g), suspended in 30 ml of PlyL buffer, and lysed in a French pressure cell at 14,000 lb/in². Digestion was carried out overnight at 37°C, and insoluble material was removed by centrifugation (10 min, 33,000 × g).

Mutanolysin digestion of B. anthracis cell walls.

For mutanolysin digestion, cell wall preparations were obtained according to a previously described protocol for purification of B. subtilis spore peptidoglycan, modified to retain cell wall-anchored proteins (38). A colony of B. anthracis Sterne harboring pLM228 was inoculated into LB broth containing kanamycin at 20 μg/ml and grown overnight at 30°C. Cells were diluted 1:50 into 4 liters of modG sporulation medium, grown until T₋₁ (optical density at 600 nm of 3.0), and harvested by centrifugation at 10,000 × g for 5 min. Sedimented cells were washed once with 100 ml of sterile ddH₂O, suspended in 50 ml of UDS buffer, and boiled for 15 min. B. anthracis murein sacculi were sedimented by centrifugation (10 min, 10,000 × g), washed with 40 ml of sterile ddH₂O, suspended in 10 ml of 5% trichloroacetic acid, and boiled for 6 min. Precipitated murein sacculi were again sedimented by centrifugation (10 min, 10,000 × g) and washed first with 40 ml of 0.5 M Tris-HCl (pH 6.3) and then twice with 50 mM Tris-HCl (pH 6.3)-1.5 mM MgCl₂. After these washes, the murein sacculi were subjected to mutanolysin digestion overnight at 37°C in 30 ml of buffer containing 50 mM Tris-HCl (pH 6.3), 1.5 mM MgCl₂, 1 mM phenylmethylsulfonyl fluoride, and 10,000 U of mutanolysin. Insoluble material was removed by centrifugation (10 min, 33,000 × g), and the soluble supernatant was subjected to Ni-nitrilotriacetic acid (NTA) affinity chromatography.

Purification of BasI_MH6.

Cytosolic extracts of bacilli (obtained during bead-beating, see above) and cell wall lysates (obtained by mutanolysin or PlyL digestion) were adjusted to pH 7.5 with 1 M Na₂HPO₄ prior to purification of BasI_MH6 on Ni-NTA affinity chromatography as previously described for SEB-MH₆-CWS_IsdC (27).

Preparation of C-terminal anchor peptides.

Purified BasI_MH6 was methanol-chloroform precipitated and cleaved with cyanogen bromide, and anchor peptides were purified by a second round of Ni-NTA affinity chromatography and separated by high-pressure liquid chromatography (HPLC) as previously described for SEB-MH₆-CWS_IsdC (27).

MALDI-MS.

Dried HPLC fractions containing peptides of interest were suspended in 15 μl of CH₃CN-water-trifluoroacetic acid (TFA; 30:70:0.1). Matrix-assisted laser desorption ionization-mass spectrometry (MALDI-MS) spectra were obtained on a reflectron time-of-flight (TOF) instrument (ABI Biosystems) in the reflection mode. Samples (0.5 μl) were cospotted with 0.5 μl of matrix (α-cyano-4-hydroxycinnamic acid; Sigma) at 10 mg/ml in CH₃CN-water-trifluoroacetic acid (70:30:0.1). All samples were externally calibrated to a standard of bovine insulin.

RESULTS

BasI is targeted to the envelope of sporulating B. anthracis cells.

In a previous study, we observed that B. anthracis expresses the basI-srtC-sctR-sctS operon during sporulation at T₋₁, 2 h prior to the formation of the polar septum that divides sporulating bacilli into the mother cell (bigger compartment) and forespore (smaller compartment) (T₊₁) (28). After septation, sortase C (the product of srtC) is found in both compartments (28). basH, also encoding a sortase C substrate with LPNTA motif, is, however, expressed in the developing forespore by σ^F RNA polymerase (28). The presence of SrtC in the forespore envelope, observed in fluorescence microscopy experiments using a SrtC-Cherry reporter (28), is likely required for the anchoring of BasH to the spore cell wall. However, the subcellular location of BasI during sporulation remained unknown. To determine whether this polypeptide is targeted to the cell wall envelope of predivisional cells, mother cells, or forespores, we constructed a BasI-Cherry fluorescent reporter protein by the insertion of the cherry coding sequence between basI codons 110 and 111. The mutant strain LAM20 expresses a BasI-Cherry fusion that harbors an N-terminal signal peptide (residues 1 to 24), the Cherry fluorescent protein domain (residues 113 to 346), and a C-terminal LPNTA sorting signal (residues 359 to 393) (Fig. 1A). We then tested whether the fusion protein was targeted to the cell wall compartment of sporulating cells, as is observed for wild-type BasI (28). The murein sacculi of sporulating B. anthracis, strains Sterne (wild type) or LAM20, were treated with mutanolysin, a muramidase that cleaves the repeating disaccharide MurNAc-GlcNAc within peptidoglycan (7, 65). Proteins released by muramidase from cell wall envelopes were separated on SDS-15% PAGE and immunoreactive signals were detected with a BasI-specific antibody (Fig. 1B). Mutanolysin-released BasI species migrated with apparent molecular mass of 17 and 35 kDa, more slowly than the molecular mass of BasI that can be deduced from the mature polypeptide. After signal peptide (37) and sorting signal cleavage (28), the calculated mass of BasI (Asp²⁵-Thr¹¹⁷) is 10,371 Da. Thus, the molecular mass of the more slowly migrating BasI species must be accounted for by the molecular masses of BasI polypeptide and the cell wall fragments to which this polypeptide is linked (see below). Immunoblotting of proteins released from the envelope of B. anthracis LAM20 revealed two predominant BasI-Cherry species that migrated at 20 to 22 kDa and 42 kDa, respectively (Fig. 1B). After signal peptide and sorting signal cleavage, the calculated mass of BasI-Cherry (Asp²⁵-Thr³⁶²) is 39,191 Da. We presume the 42-kDa species represents mature, anchored BasI-Cherry, whereas the faster-migrating species (20 to 22 kDa) are likely caused by the degradation of anchored BasI-Cherry products in the cell wall envelope. Moreover, BasI-Cherry cytoplasmic precursors are also subjected to a similar degradation, as judged from immunoblotting of LAM20 bacterial extracts. To rule out the presence of cytoplasmic contaminating material, cell wall extracts were subjected to immunoblotting with L6-specific antiserum. This experiment demonstrated that BasI-Cherry is targeted to the cell wall compartment of sporulating cells, and therefore the fusion protein is a valid surrogate for the study of BasI localization.

FIG. 1.

BasI is anchored to the envelope of B. anthracis cells during sporulation. (A) Diagram illustrating the primary structure of BasI with a cleavable N-terminal signal peptide and C-terminal sorting signal, encompassing the LPNTA motif, hydrophobic domain (black box), and positively charged tail (+). The sorting signal of BasI is cleaved by sortase C between the threonine (T) and the alanine (A) residues of the LPNTA motif. BasI-Cherry harbors a translational fusion of the Cherry fluorescence protein inserted three residues upstream of the LPNTA sorting signal. The coding sequence for the translational hybrid basI-cherry was recombined into the basI locus of B. anthracis Sterne (wild type), thereby generating strain LAM20. (B) Immunoblotting of proteins in mutanolysin cell wall lysates and bacterial extracts of B. anthracis wild type (BasI) and LAM20 (BasI-Cherry) during sporulation (T₋₁) with BasI-specific antiserum identified anchored products for both BasI and BasI-Cherry in the cell wall envelope, as well as BasI-Cherry precursors. L6 antiserum was used to control cell wall fractions for the presence of contaminating cytoplasmic material. Numbers indicate the mobility of protein molecular mass markers in kilodaltons after SDS-15% PAGE. (C) LAM20(pLM218) cells overexpressing the SctR response regulator were analyzed by fluorescence microscopy during sporulation in modG media. The expression of BasI-Cherry commenced at T₋₁, 2 h prior to formation of the polar septum. BasI-Cherry fluorescence was detected in the envelope of sporulating bacilli. At T₊₁, upon formation of the polar septum, BasI-Cherry was found in the envelope of predivisional cells but not in the forespore. DAPI staining of the chromosomal DNA of mother cells and the condensed chromosomal DNA of forespores (arrowheads) and spores was used to reveal the location of both cellular compartments. Upon disintegration of mother cells and release of endospores at T₊₂₀, BasI-Cherry fluorescence was no longer detected.

To analyze BasI-Cherry localization, cultures of B. anthracis LAM20 (basI-cherry) were subjected to sporulation in modG media. Both differential interference contrast and fluorescence microscopy images were captured to reveal BasI-Cherry or DAPI (4′,6′-diamidino-2-phenylindole)-stained chromosomal DNA of mother cells and the condensed chromosome of forespores and spores (Fig. 1C). BasI-Cherry fluorescent signals were first detected at T₋₁, 2 h prior to the formation of polar septa and the appearance of highly condensed forespore DNA. The fluorescence signal of BasI-Cherry was found in the envelope of sporulating B. anthracis LAM20 cells but not in either the forespore envelope (T₊₁) or the mature spores (T₊₂₀) (Fig. 1C). These observations are consistent with the hypothesis that the reporter protein BasI-Cherry, and likely wild-type BasI, are anchored to the cell wall envelope of predivisional sporulating cells. At the time of polar septation (T₊₁) BasI molecules are presumably anchored and no longer synthesized, thereby preventing BasI anchoring to forespore cell wall.

Sortase C with an active-site cysteine and the LPNTA sorting signal of BasI are required for anchoring of the polypeptide to the cell wall envelope.

Purified SrtC_ΔN, with a six-histidyl replacing the N-terminal signal peptide of sortase C, cleaves LPNTA motif peptide between threonine (T) and alanine (A) (28). Cleavage is blocked by methyl-methane thiosulfonate, which reacts with the thiol group to form a disulfide bond (51), suggesting the cysteine residue of sortase C (Cys¹⁸¹) functions as an active site in vitro, similar to the cysteine (Cys¹⁸⁴) of S. aureus sortase A (28, 60, 63) (Fig. 2A). To examine the requirement of SrtC Cys¹⁸¹ for in vivo anchoring of BasI, B. anthracis strain LAM18, carrying an alanine substitution at codon 181 of srtC, was generated (C181A, Fig. 2B). basI-srtC expression was achieved by IPTG induction of sctR, encoding the response regulator and transcriptional activator of basI-srtC, via the P_spac promoter on plasmid pLM226. BasI immunoreactive species were not detected in cell wall lysates of B. anthracis LAM18 (SrtC_C181A); however, immunoblotting revealed accumulation of the BasI precursor in bacterial extracts of this strain (Fig. 2C). In contrast, BasI was found in cell wall envelope lysates, but not in bacterial extracts, of wild-type bacilli (Fig. 2C). Both wild-type bacilli and LAM18 carrying the srtC_C181A substitution expressed sortase C, which could be detected with SrtC-specific antibody in bacterial extracts (Fig. 2C). B. anthracis LAM5, with a deletion of the srtC gene, also failed to anchor BasI to the cell wall envelope and accumulated BasI precursor in bacteria. As expected, SrtC immunoreactive species were not detected in bacterial extracts of strain LAM5. Immunoblotting with antibodies to ribosomal protein L6 was used as a control for proper sample loading. To determine whether sortase C-mediated anchoring of BasI to the cell wall envelope also requires the C-terminal sorting signal of the BasI precursor, we generated strain LAM17, in which the basI coding sequence for the LPNTA motif was scrambled to TNLAP. Immunoblotting experiments revealed precursor species in bacterial extracts; however, BasI_TNLAP was not found in the cell wall envelope of sporulating bacilli (Fig. 2C). In sum, BasI anchoring to the cell envelope of sporulating bacilli requires both the active-site cysteine of sortase C and the C-terminal sorting signal of BasI precursor.

FIG. 2.

srtC and basI requirements for the anchoring of BasI. (A) Alignment to reveal the conservation of cysteine in the active site of Staphylococcus aureus sortase A (Sa-SrtA), B. anthracis sortase A (Ba-SrtA), B. anthracis sortase B (Ba-SrtB), and B. anthracis sortase C (Ba-SrtC). In Ba-SrtC, the conserved cysteine occupies residue 181 (Cys¹⁸¹). (B) B. anthracis wild-type (WT) and three mutant strains were analyzed for the anchoring of BasI to the cell wall envelope of sporulating cells. Strain LAM5 harbors an in-frame deletion of the srtC gene (ΔsrtC), strain LAM17 encodes a mutant basI in which the coding sequence for its LPNTA motif has been scrambled to TNLAP, and strain LAM18 harbors a srtC_C181A allele in which the active-site Cys¹⁸¹ codon (UGU) has been replaced with an alanine codon (GCU). All B. anthracis strains were transformed with plasmid pLM226, which provides for IPTG-inducible expression of the SctR response regulator and then expression of basI-srtC. (C) Where indicated, IPTG was added to sporulating bacilli in order to induce the expression of basI-srtC. Immunoblotting with antisera specific for BasI, SrtC, or ribosomal protein L6 was used to detect polypeptides in mutanolysin-degraded murein sacculi (cell wall lysate) or in bacterial extracts after separation of proteins on SDS-PAGE and electroblotting of samples to polyvinylidene difluoride membranes. The numbers indicate the mobility of protein molecular mass markers in kilodaltons after SDS-15% PAGE.

Purification of BasI_MH6 from the cell wall of sporulating bacilli.

To analyze the cell wall anchor structure of BasI, we engineered BasI_MH6 with an insertion of methionyl-six histidyl (MH₆) upstream of the LPNTA motif sorting signal of BasI. To achieve high anchoring efficiency, we overexpressed BasI_MH6, along with sortase C and the SctR response regulator, using plasmid pLM228 (Fig. 3A). To test whether B. anthracis anchors BasI_MH6 similarly to wild-type BasI, the cell walls of bacilli harboring pLM218 or pLM228 were analyzed by immunoblotting to compare the anchoring pattern of both proteins. By overexpressing the SctR response regulator, pLM218 provides for expression of chromosomal wild-type basI, whereas pLM228 expresses chromosomal wild-type basI and plasmid-encoded basI_MH6. Cells were grown in sporulating media until T₋₁, sedimented by centrifugation, and digested with mutanolysin (Fig. 3B). Cell wall lysates were subjected to immunoblotting with BasI-specific antibodies, which revealed similar SDS-PAGE mobility patterns for immunoreactive BasI and BasI_MH6 species (Fig. 3B). Detection of the engineered protein with nickel-horseradish peroxidase demonstrated that BasI_MH6 is anchored to the envelope of sporulating cells in a fashion similar to that of BasI and can therefore be used as a reporter to study the cell wall anchor structure of BasI.

FIG. 3.

Purification of BasI_MH6 from the cell wall envelope of sporulating B. anthracis. (A) In order to purify anchored BasI, a methionyl-six-histidyl tag was inserted three residues upstream of the BasI LPNTA motif and cell wall sorting signal. The BasI_MH6 variant is encoded by plasmid pLM228, which also includes srtC and sctR, the latter providing for the expression of both basI_MH6 and srtC during sporulation. (B) To validate the use of BasI_MH6 as a surrogate for the study of BasI anchoring, cell wall fractions of wild-type bacilli carrying pLM218 and pLM228 were compared for the presence of BasI and BasI_MH6 by Western blotting with anti-BasI, which detects both versions of the surface protein, and nickel-horseradish peroxidase, which only detects BasI_MH6. The pattern of mutanolysin-released anchored species was similar for both BasI variants. (C) Cell wall anchor structure of B. anthracis BasI and cleavage sites of two murein hydrolases used in the present study: mutanolysin and PlyL amidase. (D) Coomassie blue-stained SDS-PAGE of Ni-NTA-purified BasI_MH6 released from murein sacculi by digestion with mutanolysin (Mtln) or PlyL amidase. As a control, cytoplasmic precursors of BasI_MH6 were purified from bacterial extracts. The numbers indicate the mobility of protein molecular mass markers in kilodaltons after SDS-15% PAGE.

Cell wall-anchored BasI_MH6 was solubilized from purified peptidoglycan of B. anthracis Sterne (pLM228) cells with two different murein hydrolases and the released polypeptides were subjected to Ni-NTA affinity chromatography. Mutanolysin cleavage of the glycan strands (Fig. 3C) released a spectrum of BasI_MH6 species that migrated as two clusters (17 to 25 kDa and 36 to 45 kDa) on Coomassie-stained SDS-PAGE, a finding in agreement with the migration patterns of BasI reported in Fig. 1 (Mtln, Fig. 3D). Since both of the spectra of BasI_MH6 species copurified with peptidoglycan and require a murein hydrolase for solubility, we conclude that they represent anchored products that may or may not harbor peptidoglycan with higher degrees of cross-linking attached to its C-terminal end. Previous work determined that 19% of wall peptides within B. anthracis peptidoglycan are cross-linked to neighboring wall peptides (49). PlyL, a presumed amidase that is thought to cleave the amide bond between MurNAc-l-Ala (23), was expressed in E. coli and bacterial lysates harboring the enzyme were used to cleave the cell wall envelope of sporulating bacilli. Purified BasI_MH6 species that had been liberated by PlyL digestion from the cell wall envelope of bacilli migrated mostly as a single discrete species with mobility of 19 kDa on SDS-PAGE (PlyL, Fig. 3D). To determine the effect of attached peptidoglycan on the electrophoretic mobility of mutanolysin- and PlyL-released BasI_MH6, we purified BasI_MH6 precursors from bacterial extracts and analyzed their mobility by gel electrophoresis. Coomassie blue-stained SDS-PAGE gels revealed two BasI_MH6 species with apparent molecular masses of 19 and 17 kDa (extract, Fig. 3D), whose mobility on SDS-PAGE corresponded to the P1 precursor (BasI_MH6 with N-terminal signal peptide and C-terminal sorting signal [18,155 Da]) and the P2 precursor (BasI_MH6 with only the C-terminal sorting signal [15,695 Da]), respectively (26). Cleavage of the glycan strands or wall peptides within cross-linked peptidoglycan is known to release distinct spectra of anchored proteins with different mobility on SDS-PAGE (45). In agreement with earlier observations, the P1 and P2 precursors, PlyL- and mutanolysin-solubilized BasI_MH6, each display different migration patterns (masses) on SDS-PAGE (29). We conclude that the different migration patterns must be due to different amounts of peptidoglycan attached to the C terminus of BasI_MH6 after the polypeptide has been solubilized by PlyL or mutanolysin (see below).

Cell wall anchor structure of PlyL-solubilized BasI_MH6.

BasI_MH6 was solubilized by PlyL digestion from the cell wall envelope of sporulating bacilli and purified by affinity chromatography on Ni-NTA. After cleavage of purified BasI_MH6 at methionine residues with CNBr (52), C-terminal peptides harboring the cell wall anchor structures were purified by affinity chromatography on Ni-NTA and subjected to reversed-phase (RP)-HPLC on a C₁₈ column. PlyL-released anchor peptides eluted at 28% CH₃CN-0.01% TFA and were subjected to MALDI-TOF MS (Fig. 4A). A complex set of ion signals was observed. Four predominant ion signals—m/z 1,751.98, 1,779.98, 1,794.98, and 1,807.98—were explained as C-terminal anchor peptides linked to diaminopimelic acid (H₆GEKLPNT-m-Dap). meso-Diaminopimelic acid (m-Dap) is a diamino acid at position 3 of the Bacillus cell wall pentapeptide precursor (l-Ala-d-iGln-m-Dap-d-Ala-d-Ala), and its side chain amino group is also involved in peptidoglycan cross-linking by forming an amide bond with d-Ala at position four of neighboring wall peptides [l-Ala-d-iGln-(l-Ala-d-iGln-m-Dap-d-Ala-)m-Dap-d-Ala-d-Ala] (44). The predicted m/z of the anchor peptide H₆GEKLPNT-m-Dap is 1,752.84, and its formylated, diformylated, and carbamylated forms are calculated to be 1,780.85, 1,808.86, and 1,795.87 (Table 2). Formylation and carbamylation of α-NH₂ groups in cell wall anchor peptides are known to occur during CNBr cleavage (70% formic acid) and affinity chromatography (urea buffer) (27). To examine the validity of our prediction, the parent ion at m/z 1,779.98 was subjected to collisionally induced dissociation (CID) in a MALDI-TOF tandem MS experiment, and daughter ion spectra were collected (Fig. 4C). Table 3 summarizes the observed ions, their structural interpretation, and fragmentation patterns, which are in agreement with the peptide structure H₆GEKLPNT-m-Dap. The presence of a single Dap linked to the C terminus of the BasI_MH6 anchor peptide is, however, not in agreement with the predicted amidase function of PlyL. We would have expected to find anchor peptides encompassing tri- or tetrapeptides [l-Ala-d-iGln-(H₆GEKLPNT-)m-Dap and l-Ala-d-iGln-(H₆GEKLPNT-)m-Dap-d-Ala)]. It should be noted that our protocol used recombinant PlyL from crude E. coli lysates as an enzyme preparation. Thus, cleavage of wall peptides between d-iGln-m-Dap may be a feature of contaminating E. coli enzymes that can act on the primary cleavage products of PlyL. In agreement with this view, several less-abundant ion signals of PlyL-released BasI_MH6 anchor peptides were identified and interpreted to represent cell wall tetrapeptides linked to the C-terminal end of the polypeptide chain (Table 2).

FIG. 4.

MS of PlyL-released BasI_MH6 anchor peptides. (A) MALDI-TOF spectrum of BasI_MH6 anchor peptides released from B. anthracis murein sacculi. Isolated cell wall preparations were treated with extracts of recombinant E. coli harboring PlyL amidase, and cell wall lysates were subjected to Ni-NTA affinity chromatography. Purified BasI_MH6 was cleaved with CNBr at methionine residues to generate C-terminal anchor peptides, which were isolated by a second chromatography step on Ni-NTA and further purified by RP-HPLC on a C₁₈ column. Anchor peptides eluted at 30% CH₃CN and were subjected to MALDI-TOF. Major ion signals are labeled; please see Table 2 for a listing of all ions. (B) Deduced anchor structures and calculated m/z values for species released by PlyL digestion (see Table 2 for mass measurements). Most ions detected by MALDI-TOF were modified by formyl or carbamyl at their NH₂ groups (arrows), and the calculated mass of formylated (the most common modification) ion is indicated. Cell wall tri- or tetrapeptides lacking or containing a terminal d-Ala (between parentheses), respectively, were identified in BasI anchor structures. (C) CID and MALDI-TOF/TOF analysis of the m/z 1,779.98 identified in panel A. Please see Table 3 for a listing of daughter ions and their structural interpretation.

TABLE 2.

Summary of ions produced during MS of PlyL-released BasI_MH6 anchor peptides

m/z		Δcalc-obsb	Proposed structure	Modificationc
Calculateda	Observed
1,752.84	1,751.98	0.86	HHHHHHGEKLPNT-m-Dap	None
1,780.85	1,779.98	0.87	HHHHHHGEKLPNT-m-Dap	HC(O)-
1,795.87	1,794.98	0.89	HHHHHHGEKLPNT-m-Dap	NH2-C(O)-
1,808.86	1,807.98	0.88	HHHHHHGEKLPNT-m-Dap	2 × HC(O)-
1,823.88	1,823.00	0.88	HHHHHHGEKLPNT-m-Dap-d-Ala	None
1,851.89	1,851.03	0.86	HHHHHHGEKLPNT-m-Dap-d-Ala	HC(O)-
1,866.91	1,866.04	0.87	HHHHHHGEKLPNT-m-Dap-d-Ala	NH2-C(O)-
1,979.95	1,980.09	−0.14	d-iGln(HHHHHHGEKLPNT-)m-Dap-d-Ala	HC(O)-
1,994.97	1,995.09	−0.12	d-iGln(HHHHHHGEKLPNT-)m-Dap-d-Ala	NH2-C(O)-
2,022.98	2,022.14	0.84	l-Ala-d-iGln(HHHHHHGEKLPNT-)m-Dap-d-Ala	None
2,050.99	2,050.14	0.85	l-Ala-d-iGln(HHHHHHGEKLPNT-)m-Dap-d-Ala	HC(O)-
2,066.01	2,065.16	0.85	l-Ala-d-iGln(HHHHHHGEKLPNT-)m-Dap-d-Ala	NH2-C(O)-
2,095.01	2,094.25	0.76	[HHHHHHGEKLPNT-m-Dap-d-Ala]m-Dap-d-Ala	HC(O)-
2,123.02	2,122.19	0.83	[HHHHHHGEKLPNT-m-Dap-d-Ala]m-Dap-d-Ala	2 × HC(O)-
2,195.06	2,195.23	−0.17	[l-Ala-d-iGln(HHHHHHGEKLPNT-)m-Dap-d-Ala]m-Dap	None
2,223.07	2,222.24	0.83	[l-Ala-d-iGln(HHHHHHGEKLPNT-)m-Dap-d-Ala]m-Dap	HC(O)-
2,238.09	2,237.23	0.86	[l-Ala-d-iGln(HHHHHHGEKLPNT-)m-Dap-d-Ala]m-Dap	NH2-C(O)-
2,251.08	2,250.24	0.84	[l-Ala-d-iGln(HHHHHHGEKLPNT-)m-Dap-d-Ala]m-Dap	2 × HC(O)-
2,266.10	2,265.25	0.85	[l-Ala-d-iGln(HHHHHHGEKLPNT-)m-Dap-d-Ala]m-Dap-d-Ala	None
2,294.11	2,294.25	−0.14	[l-Ala-d-iGln(HHHHHHGEKLPNT-)m-Dap-d-Ala]m-Dap-d-Ala	HC(O)-
2,309.13	2,309.28	−0.15	[l-Ala-d-iGln(HHHHHHGEKLPNT-)m-Dap-d-Ala]m-Dap-d-Ala	NH2-C(O)-
2,322.12	2,322.29	−0.17	[l-Ala-d-iGln(HHHHHHGEKLPNT-)m-Dap-d-Ala]m-Dap-d-Ala	2 × HC(O)-
2,422.17	2,422.34	−0.17	l-Ala-d-iGln[l-Ala-d-iGln(HHHHHHGEKLPNT-)m-Dap-d-Ala]m-Dap	HC(O)-
2,450.18	2,450.32	−0.14	l-Ala-d-iGln[l-Ala-d-iGln(HHHHHHGEKLPNT-)m-Dap-d-Ala]m-Dap	2 × HC(O)-

^aCalculations are based on masses obtained with the MS-Product in the ProteinProspector 4.0.5 internet tool (http://prospector.ucsf.edu/ucsfhtml4.0/msprod.htm) (9).

^bThat is, the difference between the calculated and observed ion masses.

^cModification of αNH₂ groups. Formylation, HC(O)-; carbamylation, NH₂-C(O)-.

TABLE 3.

Summary of daughter ions produced during MS/MS of the PlyL-released m/z 1,779.98 parent ion

m/z		Δcalc-obsb	Proposed structurec	Ion typed
Calculateda	Observed
110.07	110.06	0.01	H	ie
275.13	275.19	−0.06	HH	a2
303.12	303.21	−0.09	HH	b2
412.18	412.27	−0.09	HHH	a3
440.18	440.29	−0.11	HHH	b3
452.23	452.36	−0.13	HGEK	i
503.25	502.37	0.88	PNT-m-Dap	y4
549.24	549.36	−0.12	HHHH	a4
565.31	565.42	−0.11	HGEKL	i
577.24	577.34	−0.10	HHHH	b4
589.28	589.41	−0.13	HHGEK	i
686.30	686.39	−0.09	HHHHH	a5
702.37	702.46	−0.09	HHGEKL	i
714.30	714.37	−0.07	HHHHH	b5
726.34	726.43	−0.09	HHHGEK	i
823.36	823.40	−0.04	HHHHHH	a6
851.36	851.37	−0.01	HHHHHH	b6
908.34	908.38	−0.04	HHHHHHG	b7
930.49	929.49	1.00	GEKLPNT-m-Dap	y8
1,009.42	1,009.31	0.11	HHHHHHGE	a8
1,037.42	1,037.30	0.12	HHHHHHGE	b8
1,054.45	1,054.32	0.13	HHHHHHGE	c8
1,067.55	1,067.42	0.13	HGEKLPNT-m-Dap	y9
1,137.52	1,137.29	0.23	HHHHHHGEK	a9
1,165.51	1,165.29	0.22	HHHHHHGEK	b9
1,204.61	1,204.35	0.26	HHGEKLPNT-m-Dap	y10
1,250.60	1,250.30	0.30	HHHHHHGEKL	a10
1,278.60	1,278.24	0.36	HHHHHHGEKL	b10
1,341.67	1,341.25	0.42	HHHGEKLPNT-m-Dap	y11
1,392.68	1,392.21	0.47	HHHHHHGEKLP	c11
1,478.72	1,478.14	0.58	HHHHGEKLPNT-m-Dap	y12
1,489.69	1,489.07	0.62	HHHHHHGEKLPN	b12
1,607.77	1,606.90	0.87	HHHHHHGEKLPNT	c13
1,615.78	1,614.87	0.91	HHHHHGEKLPNT-m-Dap	y13

^aCalculations are based on average masses obtained with the MS-Product in the ProteinProspector 4.0.5 internet tool (http://prospector.ucsf.edu/ucsfhtml4.0/msprod.htm) (9).

^bThate is, the difference between the calculated and observed masses of daughter ions.

^cN-terminal daughter ions contain a formylated N-terminal histidyl residue.

^dThe nomenclature used refers to the NH₂-and CO₂H-terminal cleavage fragments according to Biemann (6).

^eInternal ion; the calculated mass corresponds to the sum of the residue masses (M+H⁺).

Ion signals indicating cross-linking between anchor peptides and neighboring cell wall peptides were also detected (Table 2). Anchor peptides with ion signals at m/z 2,195.23, 2,222.24, 2,237.23, and 2,250.24 were explained as anchor peptides linked to cell wall tetrapeptides that were in turn tethered to diaminopimelic acid within another wall peptide ([l-Ala-d-iGln(H₆GEKLPNT-)m-Dap-d-Ala]m-Dap, calculated m/z = 2,195.06) and its formylated and carbamylated forms (Table 2). The predicted structure of this compound was also validated in a tandem MS experiment (data not shown). Degradation products of cross-linked wall peptides tethered to the C-terminal end of BasI_MH6 were also detected (Table 2). Taken together, these results demonstrate that during anchoring of BasI_MH6, the polypeptide chain is cleaved by sortase C between threonine and alanine of the LPNTA motif and that the carboxyl group of threonine is amide linked to the side chain amino group of diaminopimelic acid within wall peptides that may or may not be cross-linked to neighboring wall peptides in the murein sacculus of the B. anthracis envelope.

Cell wall anchor structure of muramidase-solubilized BasI_MH6.

The glycan strands of isolated peptidoglycan were cut with mutanolysin at the β(1-4) glycosidic bond between the repeating disaccharide MurNAc-GlcNAc (Fig. 5B) and anchor peptides were purified. The amino sugars of C-terminal anchor peptides were reduced by sodium borohydride treatment, and glycopeptides were separated by RP-HPLC on a C₁₈ column. Mutanolysin-released anchor peptides eluted at 28% CH₃CN-0.01% TFA and were subjected to MALDI-TOF MS. Two sets of ions were detected (Fig. 5A). The compound at m/z 2,418.38 represented the most abundant ion of the first cluster of compounds, and its structure was explained as a formylated C-terminal anchor peptide linked to murein disaccharide tripeptide:Glc_NH2-(β1-4)-MurNAc-[L-Ala-d-iGln-(H₆GEKLPNT-)m-Dap], calculated m/z = 2,418.54) (Fig. 5B and Table 4). To test this prediction, the compound was subjected to CID in a MALDI-TOF/TOF experiment. Consistent with previous reports using MS to analyze the structure of peptidoglycan fragments (2, 27), the most abundant CID daughter ions resulted from the breakage of the β(1-4) glycosidic bond between Glc_NH2 and MurNAc (Table 5, observed m/z = 2,255.91, calculated m/z = 2,255.93). However, many other, less-abundant ions were also detected and could be structurally assigned to N- or C-terminal CID fragments of the predicted structure (Table 5). The lack of amino acetylation of glucosamine (Glc_NH2) residues and the abundance of wall tripeptides corroborate previous work on the structure of peptidoglycan for members of the B. cereus family (49, 66). Nevertheless, anchor peptides containing C-terminal tetrapeptides were found as well (Fig. 5A and Table 4).

FIG. 5.

MS of mutanolysin-released BasI_MH6 anchor peptides. (A) MALDI-TOF spectrum of BasI_MH6 anchor peptides released from B. anthracis murein sacculi. Cell wall preparations were treated with mutanolysin, and lysates were subjected to Ni-NTA affinity chromatography. Purified BasI_MH6 was cleaved with CNBr at methionyl to generate C-terminal anchor peptides, which were isolated by a second chromatography step on Ni-NTA and further purified by RP-HPLC. Anchor peptides eluted at 28% CH₃CN and were subjected to MALDI-TOF. (B) Proposed structures and their calculated m/z values for species released by mutanolysin digestion. Major ion signals are labeled; please see Table 4 for a listing of all detected ions. Ion signals detected by MALDI-TOF in panel A correspond to formylated or carbamylated species with modified glucosamine- and/or alpha-amino groups (arrows), and the calculated mass of formylated (most common modification) ion is indicated. Glucosamine residues in BasI_MH6 anchor peptides were deacetylated (Glc_NH2). Both cell wall tri- or tetrapeptides, with or without the terminal d-Ala (between parentheses), were found to be linked to the C-terminal end of BasI_MH6. (C) CID and MALDI-TOF/TOF of the m/z 2,418.38 identified in panel A. Please see Table 5 for a listing of daughter ions and their structural interpretation.

TABLE 4.

Summary of ions produced during MS of mutanolysin-released BasI_MH6 anchor peptides

m/z		Δcalc-obsb	Proposed structure	Modificationc
Calculateda	Observed
2,390.53	2,390.37	0.16	GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-(HHHHHHGEKLPNT-)m-Dap]	None
2,418.54	2,418.38	0.16	GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-(HHHHHHGEKLPNT-)m-Dap]	HC(O)-
2,433.56	2,433.41	0.15	GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-(HHHHHHGEKLPNT-)m-Dap]	NH2-C(O)-
2,446.55	2,446.38	0.17	GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-(HHHHHHGEKLPNT-)m-Dap]	2 × HC(O)-
2,461.61	2,461.40	0.21	GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-(HHHHHHGEKLPNT-)m-Dap-d-Ala]	None
2,489.62	2,489.41	0.22	GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-(HHHHHHGEKLPNT-)m-Dap-d-Ala]	HC(O)-
2,517.63	2,518.43	−0.80	GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-(HHHHHHGEKLPNT-)m-Dap-d-Ala]	2 × HC(O)-
2,532.61	2,532.41	−0.20	GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-(HHHHHHGEKLPNT-)m-Dap-d-Ala]	2 × HC(O)-; NH2-C(O)-
3,270.43	3,270.90	−0.47	GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-(HHHHHHGEKLPNT-)m-Dap-{GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-m-Dap-}d-Ala]	None
3,298.44	3,298.90	−0.46	GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-(HHHHHHGEKLPNT-)m-Dap-{GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-m-Dap-}d-Ala]	HC(O)-
3,326.45	3,326.83	−0.38	GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-(HHHHHHGEKLPNT-)m-Dap-{GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-m-Dap-}d-Ala]	2 × HC(O)-
3,341.51	3,341.97	−0.46	GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-(HHHHHHGEKLPNT-)m-Dap-{GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-m-Dap-d-Ala}d-Ala]	None
3,354.46	3,354.92	−0.46	GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-(HHHHHHGEKLPNT-)m-Dap-{GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-m-Dap-}d-Ala]	3 × HC(O)-
3,369.52	3,368.85	0.67	GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-(HHHHHHGEKLPNT-)m-Dap-{GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-m-Dap-}d-Ala]	HC(O)-
3,384.54	3,383.91	0.63	GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-(HHHHHHGEKLPNT-)m-Dap-{GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-m-Dap-}d-Ala]	NH2-C(O)-
3,397.53	3,397.89	−0.36	GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-(HHHHHHGEKLPNT-)m-Dap-{GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-m-Dap-}d-Ala]	2 × HC(O)-

^aCalculations are based on masses obtained with the MS-Product in the ProteinProspector 4.0.5 internet tool (http://prospector.ucsf.edu/ucsfhtml4.0/msprod.htm) (9).

^bThat is, the difference between the calculated and observed ion masses.

^cModification of NH₂ groups, see Fig. 5B: formylation, HC(O)-; and carbamylation, NH₂-C(O)-.

TABLE 5.

Summary of daughter ions produced during MS/MS of the mutanolysin-released m/z 2,418.38 parent ion

m/z		Δcalc-obsb	Proposed structurec	Ion typed
Calculateda	Observed
275.13	275.23	−0.10	HH	a2
412.18	412.24	−0.06	HHH	a3
440.18	440.38	−0.20	HHH	b3
549.24	549.28	−0.04	HHHH	a4
577.24	577.52	−0.28	HHHH	b4
714.30	714.56	−0.26	HHHHH	b5
726.34	726.69	−0.35	HHHGEK	ie
823.36	823.63	−0.27	HHHHHH	a6
851.36	851.61	−0.25	HHHHHH	b6
908.38	908.61	−0.23	HHHHHHG	b7
1,009.42	1,009.67	−0.25	HHHHHHGE	a8
1,037.42	1,037.65	−0.23	HHHHHHGE	b8
1,137.52	1,137.73	−0.21	HHHHHHGEK	a9
1,165.51	1,165.69	−0.18	HHHHHHGEK	b9
1,233.58	1,233.39	0.19	HHHHHHGEKL(-NH3)	i
1,252.60	1,251.73	0.87	GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-(LPNT-)m-Dap]	y8
1,261.57	1,261.57	0.00	HHHHHHGEKL(-NH3)	b10-NH3
1,278.60	1,278.73	−0.13	HHHHHHGEKL	b10
1,489.69	1,490.69	−1.00	HHHHHHGEKLPN	b12
1,544.74	1,544.66	0.08	HHHHHHGEKLPNT(-H2O)	i
1,703.82	1,704.63	−0.81	GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-(HGEKLPNT-)m-Dap]	y12
1,716.82	1,716.50	0.32	HHHHHHGEKLPNT-m-Dap-(-H2O)	i
1,779.84	1,779.53	0.31	HHHHHHGEKLPNT-m-Dap	c14
1,840.88	1,841.50	−0.62	GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-(HHGEKLPNT-)m-Dap]	y13
1,890.89	1,890.41	0.48	d-iGln-(HHHHHHGEKLPNT)m-Dap	b15
1,908.90	1,908.41	0.59	l-Ala-d-iGln-(HHHHHHGEKLPNT)m-Dap	c15
1,977.94	1,978.42	−0.48	GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-(HHHGEKLPNT-)m-Dap]	y14
2,052.00	2,051.34	0.66	lactoyl-[l-Ala-d-iGln-(HHHHHHGEKLPNT-)m-Dap]	-f
2,115.00	2,115.23	−0.23	GlcNH2-(β1-4)-MurNAc-[l-Ala-d-iGln-(HHHHGEKLPNT-)m-Dap]	y15
2,255.91	2,255.93	−0.02	MurNAc-[l-Ala-d-iGln-(HHHHHHGEKLPNT-)m-Dap]h	-g

^aCalculations are based on masses obtained with the MS-Product in the ProteinProspector 4.0.5 internet tool (http://prospector.ucsf.edu/ucsfhtml4.0/msprod.htm) (9).

^bThat is, the difference between the calculated and observed masses of daughter ions.

^cN-terminal histidyl residues are formylated.

^dThe nomenclature used refers to the NH₂-and CO₂H-terminal cleavage fragments according to Biemann (6); exceptions are indicated below. Ion number is calculated considering the histidine tail as the main NH₂ terminus.

^eInternal ion.

^fThe y-type ion originated from the cleavage of the ether bond between N-acetylglucosamine and lactoyl groups of N-acetylmuramitol (56).

^gThe y-type ion originated from the cleavage of the β1-4 O-glycosidic bond between N-acetylglucosamine and N-acetylmuramitol (56).

^hMain peak.

The second cluster of ions comprised BasI_MH6 anchor species that were linked to cross-linked peptidoglycan, i.e., murein tetrapeptide-tripeptide and tetrapeptide-tetrapeptide (Fig. 5A and Table 4). The most abundant compound of this cluster, m/z 3,341.97, could be assigned the structure Glc_NH2-(β1-4)-MurNAc-[l-Ala-d-iGln-(H₆GEKLPNT-)m-Dap-{Glc_NH2-(β1-4)-MurNAc-[l-Ala-d-iGln-m-Dap-d-Ala-d-Ala]} (calculated m/z = 3,341.43). MALDI-TOF/TOF analysis of this ion corroborated our hypothesis (data not shown). Anchor peptides linked to murein subunits with higher degrees of cross-linking could not be detected. In all cases examined, anchor species modified by formylation and/or carbamylation on either αNH₂ or glucosamine NH₂ groups (Fig. 5B) were observed. In summary, the cell wall anchor structure of BasI_MH6 revealed that the polypeptide is linked to B. anthracis peptidoglycan subunits through the formation of an amide bond between its C-terminal threonine and the side chain amino group of diaminopimelic acid. The low degree of cross-linking observed for BasI_MH6 anchor peptides is in agreement with the overall low degree of cross-linking in the peptidoglycan of B. anthracis.

DISCUSSION

The function of many sortases is the anchoring of surface proteins to the peptidoglycan cell wall envelope of gram-positive bacteria (26). Three classes of sortases catalyze such reactions. Sortase A recognizes its substrates by the presence of an LPXTG motif sorting signal, cleaving polypeptides between the conserved threonine (T) and glycine (G) residues (60). Sortase A performs a transpeptidation reaction whereby the C-terminal threonine of the LPXTG bearing polypeptide is amide linked to cell wall cross-bridges (61). Cross-bridges are heterogeneous in structure, and surface protein anchoring has been described for the pentaglycine cross-bridges of staphylococci (27, 35, 45, 58, 59), as well as for the listerial diaminopimelic acid cross-bridge (11). In spite of such structural heterogeneity, the availability of the cross-bridge amino group is a conserved prerequisite feature of the sorting pathway (57); the amino group functions as a nucleophile and resolves the thioester linked acyl enzyme intermediate between the cysteine thiol of sortase and the carboxyl group of threonine at the C-terminal end of the polypeptide chain (31). Sortase B anchors proteins to cell wall cross-bridges under iron starvation conditions and cleaves the NPQTN or NPKTG motif of S. aureus or B. anthracis IsdC, respectively (24, 27, 30, 32). Sortase B anchored products appear to assume a unique location in murein sacculi, enabling passage of heme-iron from the surface across the cell wall envelope layer for delivery to membrane transporters, a distance of up to 100 nm (25). Sortase C has been shown to specifically cleave the LPNTA motif in B. anthracis BasI and BasH, presumably anchoring these polypeptides during sporulation (28). Here we reveal the BasI anchor structure and demonstrate for the first time that sortase C links the C-terminal threonine (T) of BasI to the amino group of diaminopimelic acid cross-bridges in the cell wall envelope of B. anthracis sporulating cells. This report presents the first surface protein anchor structure for Bacillus species. Our findings further corroborate the universality of the sortase transpeptidation mechanism. However, the data also identify two unique features of sortase C. This enzyme not only cleaves polypeptides between threonine and alanine (instead of threonine and glycine) but sortase C also acts on two different substrates in two different subcellular compartments: the envelope of predivisional sporulating cells (BasI) and the peptidoglycan of developing forespores (BasH).

A second gene encoding a sortase C substrate with an LPNTA motif sorting signal, basH, is expressed in the B. anthracis forespore after formation of the polar septum (28). Its product, BasH, appears to be targeted exclusively to the forespore envelope, since fluorescent reporter BasH-Cherry can be found only in the spore envelope and not in mother cell envelopes (L. A. Marraffini and O. Schneewind, unpublished data). In contrast, BasI-Cherry fluorescence is only observed in the predivisional cell envelope and not in the envelope of forespores. Fluorescence microscopy experiments with SrtC-Cherry detected sortase C first in the envelope of predivisional sporulating cells at T_-1 (28). Upon the formation of polar septa at T₊₁, SrtC-Cherry fluorescence was also observed in the forespore envelope and even in mature spores that have been released after mother cell disintegration at T₊₂₀. We hypothesize that BasI is expressed and anchored at T₋₁ and that anchoring reaches completion sometime after T₋₁. During a later stage of this developmental program, at T₊₁, BasI molecules that have been previously anchored remain in the envelope of sporulating cells. However, since BasI may no longer be synthesized and forespore peptidoglycan turned over, the polypeptide is not found in the forespore envelope. These results support the notion that different surface proteins decorate different compartments of the sporulating cell, with BasI present in predivisional cells and BasH present in forespores. On the other hand, sortase C, which is responsible for the targeting of BasI and BasH, is present in both predivisional cells and forespores.

Sortase C is a member of a distinct class of sortases (class D) that is found in spore-forming bacteria (bacilli, clostridia, and actinomycetales) (12). Previous work examined the role of class D sortases in Streptomyces coelicolor, a gram-positive, filamentous soil bacterium. S. coelicolor forms mycelia that submerge in a liquid or even in a semiliquid environment. The mycelia eventually differentiate to generate aerial hyphae that septate and then develop chains of spores for bacterial dissemination in air. Hyphal surfaces are highly hydrophobic, a property that promotes outgrowth into the air and dispersion of spores, which eventually give rise to new mycelia (18). Six surface proteins named chaplins are involved in aerial hyphae formation, and three of these are synthesized as precursors carrying N-terminal signal peptides and C-terminal sorting signals that require a sortase C homolog for proper anchoring to the bacterial envelope (8, 13). Presumably, these polypeptides function to lower aqueous surface tension for the emergence of aerial hyphae into air (8, 13). In this regard, it seems noteworthy that B. anthracis sortase C is also required for spore formation in infected guinea pig tissues or sheep blood under conditions of low oxygen concentration (28). Perhaps one or both of the sortase C anchored products of B. anthracis, BasI in the envelope of predivisional cells or BasH in the forespore envelope, fulfill functions analogous to those of chaplins. If acting as surfactants, this may allow sporulating bacilli access to air-tissue interfaces in anthrax carcasses or access to oxygen, which is essential for the completion of spore development (42). Our ongoing experiments are designed to examine basI and basH contributions to spore formation in carcasses.