RrgB is the major pilin which forms the pneumococcal pilus backbone. We report the high-resolution crystal structure of the full-length form of RrgB containing the IPQTG sorting motif. The RrgB fold is organized into four distinct domains, D1–D4, each of which is stabilized by an isopeptide bond. Crystal packing revealed a head-to-tail organization involving the interaction of the IPQTG motif into the D1 domain of two successive RrgB monomers. This fibrillar assembly, which fits into the electron microscopy density map of the native pilus, probably induces the formation of the D1 isopeptide bond as observed for the first time in the present study, since neither in published structures nor in soluble RrgB produced in Escherichia coli or in Streptococcus pneumoniae is the D1 bond present. Experiments performed in live bacteria confirmed that the intermolecular bond linking the RrgB subunits takes place between the IPQTG motif of one RrgB subunit and the Lys183 pilin motif residue of an adjacent RrgB subunit. In addition, we present data indicating that the D1 isopeptide bond is involved in RrgB stabilization. In conclusion, the crystal RrgB fibre is a compelling model for deciphering the molecular details required to generate the pneumococcal pilus.

INTRODUCTION

Pili are elongated and flexible appendages associated to the bacterial cell surface and play important roles in host tissue colonization and pathogenesis. Interest in the study of the pilus assembly mechanism of Gram-positive organisms such as Corynebacterium diphtheriae, Streptococcus pyogenes, Streptococcus agalactiae and Streptococcus pneumoniae has been heightened owing to their clear link with bacterial virulence processes and their applicability as potential vaccine antigen candidates [18]. Unlike Gram-negative pili, which are formed through the non-covalent association of protein subunits, Gram-positive pili are assembled by covalent peptidic linkages between individual pilin subunits as a consequence of reactions catalysed by the transpeptidase activity of sortases [9,10].

The pilus shaft in Gram-positive bacteria is formed by the polymerization of the major pilin on to which minor pilins are associated. Minor pilins have been shown to be involved either in the anchoring of the pilus polymer to the peptidoglycan or in adhesion to the host cell. High-resolution crystal structures of major and minor pilins from a number of different species have revealed a common modular topology composed of the association of two domains, CNA-A and CNA-B, originally described in the adhesin CNA, a collagen-binding protein expressed by Staphylococcus aureus [1214]. The crystal structures of the major pilins Spy0128 from S. pyogenes, BcpA from Bacillus cereus and SpaA from C. diphtheriae illustrate a common potential mechanism of pilus polymerization through the formation of intermolecular bonds involving the threonine residue of the C-terminal LPxTG sorting motif from one pilin monomer to an exposed lysine residue in the successive pilin [1517]. This latter residue is present in the YPKN so-called ‘pilin motif’, as observed in the polymerization mechanism of BcpA and SpaA [16,17]. An alternative strategy has been developed by S. pyogenes: in the absence of the pilin motif, the intermolecular bond between Spy0128 monomers recruits Lys161 of the EGSKVPI sequence [15]. Pilins contain one or several self-generated covalent internal bonds, formed by the association of lysine and asparagine residue side chains and stabilized by a carboxylic group of an acidic residue of an E-box or D-box motif [18,19]. The high stability generated by covalent intermolecular bonds linking pilus subunits is reinforced by the presence of such isopeptide bonds, leading to highly stable macromolecular pilus assemblies [1517,20,21].

S. pneumoniae is an important human pathogen that causes a large number of respiratory tract infections, such as pneumonia and sinusitis, as well as invasive diseases such as septicaemia and meningitis. Among the surface-exposed virulence factors, the pneumococcal pilus has been shown to play a role in host–cell adhesion [5,6]. Pilus formation requires the expression of seven genes encoded by the rlrA pathogenicity islet, including a RofA-like transcriptional regulator (rlrA), three sortases (SrtC-1, SrtC-2 and SrtC-3) and three structural proteins, the major pilin RrgB and the accessory pilins RrgA and RrgC [5,22]. EM (electron microscopy) images of native purified pili showed that RrgB is the major pilin subunit forming the elongated fibre shaft and that RrgA and RrgC are present at the distal and the proximal ends respectively, in accordance with the host-adhesion function of RrgA and the putative cell-wall anchoring role of RrgC [23].

Beside their respective and specific functions, the structures of RrgA, RrgB and RrgC share common features, since all three pilins are stabilized by internal lysine–asparagine bonds [19,2427]. In the present study, we report the high-resolution crystal structure of the full-length form of RrgB harbouring the IPQTG sorting motif, hereafter noted as FL-RrgB-IPQTG. RrgB molecules are organized in a head-to-tail fashion involving the interaction of the C-terminal IPQTG sorting motif into a cavity of the D1 domain of the successive RrgB monomer. The structure of RrgB is composed of four domains (D1–D4) each cross-linked by intramolecular isopeptide bonds. The role of these structural features was probed by analysing the consequence on pilus synthesis and stability of RrgB mutants inserted into the infectious pneumococcal strain TIGR4. The present study confirms that RrgB subunit polymerization takes place between the IPQTG motif of one RrgB subunit and the Lys183 pilin motif residue of an adjacent RrgB subunit (as already reported by Gentile et al. [26]). Mutation of Glu143 of the D1 domain isopeptide bond reduces the level of pilus expression supporting a role of this newly discovered bond in pilus genesis and/or stability. The FL-RrgB-IPQTG structure, organized in the crystal unit as a fibre, is unique with respect to previously published RrgB data [2527] and provides details on the mechanism of pilus assembly. Furthermore, it suggests that the formation of the D1 isopeptide bond, observed in the present study for the first time, is dependent on the RrgB fibre assembly.

EXPERIMENTAL

Cloning and site-directed mutagenesis

The amplification of rrgB gene was performed using the chromosomal DNA of the S. pneumoniae TIGR4 strain. The sequence corresponding to the coding region of RrgB (30–633), which lacks the signal peptide, the transmembrane anchor and the C-terminal tail, was amplified and cloned into a pLIM vector using the LIC method (PX′ Therapeutics). Mutations were introduced by PCR-based site-directed mutagenesis (QuikChange® II XL Site-Directed Mutagenesis Kit, Agilent Technologies) and verified by DNA sequencing (Cogenics, Beckman Coulter Genomics).

Expression and purification of RrgB

Escherichia coli BL21/DE3 STAR cells were used for protein expression, which was induced in Terrific Broth (BD Biosciences) with 0.5 mM IPTG (isopropyl β-D-thiogalactopyranoside) at 37°C for 3 h. Cells were harvested by centrifugation and lysed by sonication for 2 min [pulses on 2 s, off 8 s at 4°C using a Sonicator 3000 (Misonix)] in 50 mM Hepes, pH 7.5, 200 mM NaCl, 20 mM imidazole and a protease inhibitor cocktail (Complete EDTA-free, Roche). The lysate was clarified by centrifugation for 20 min at 39000 g and applied on a HisTrap™ HP column (GE Healthcare). Protein elution was performed using a linear gradient of 20–500 mM imidazole over 20 column volumes. Eluted proteins were dialysed and further purified by gel filtration chromatography.

Proteolysis assay and N-terminal sequencing

RrgB wild-type and E143A mutant were diluted in 50 mM Hepes, pH 7.5, and 200 mM NaCl buffer. Subsequently, 20 μg of each protein was mixed with 50 ng of trypsin. The reaction was carried out at 22°C for 2 h. The samples were analysed by SDS/PAGE. The proteins were either stained with a Coomassie Blue solution or transferred to a Problot membrane (Applied Biosystems) and stained according to the supplier's instructions. The protein bands were cut from the membrane and sequenced using automated Edman degradation on an Applied Biosystems gas-phase sequencer model 477A with on-line analysis of the phenyl thiohydantoin-derivatives.

Crystallization and data collection

Crystals of RrgB were obtained by hanging-drop vapour diffusion by mixing 1 μl of protein (10 mg/ml in 50 mM Tris/HCl, pH 8.0, and 100 mM NaCl) and 1 μl of reservoir [100 mM Tris/HCl, pH 7.5, and 23% (w/v) poly(ethylene glycol) 8000 or 100 mM Tris/HCl, pH 7.5, and 18% (w/v) poly(ethylene glycol) 12000]. Crystals were subsequently cryoprotected in mother liquor containing 20% (v/v) glycerol. Crystals were flash-cooled under a liquid nitrogen stream. A highly redundant dataset was collected (Table 1) at the ESRF (European Synchrotron Radiation Facility) ID23-EH1 beamline (Grenoble, France).

Table 1
Data collection, molecular replacement and structure refinement statistics
Parameter Value    
Data collection     
 X-ray source ID23EH1    
 Detector ADSC Quantum Q315    
 Wavelength (Å) 0.97485    
 Scan range (°) 360    
 Oscillation (°) 0.25    
 Space group P32    
a (Å) 74.62    
b (Å) 74.61    
c (Å) 340.53    
 Mosaicity (°) 0.180    
 Resolution (last shell) (Å) 2.39 (2.54–2.39)    
 Completeness (last shell) (%) 97.8 (92.6)    
Rsym (last shell) (%) 5.6 (48.7)    
I/σ(I) (last shell) 25.73 (4.91)    
 Number of observed/unique reflections 653063/164140    
 Wilson B factor (Å247.74    
 Twin fraction 0.392    
Molecular replacement     
 Mol/ASU    
 Phaser     
 RFZ/TFZ 24.0/25.9 21.8/50.3 8.4/49.3  
 LLG 820 3110 4738 4958 
Refinement     
Rwork/Rfree 0.1895/0.2321    
 Number of atoms used in refinement 14138    
 Number of solvent atoms 435    
 RMS deviation, bond lengths (Å) 0.008    
 RMS deviation, bond angles (°) 1.122    
 Mean B factor (Å228.5    
 Residues in most favoured/allowed region of Ramachandran plot (%) 99.9    
 PDB entry 2Y1V    
Parameter Value    
Data collection     
 X-ray source ID23EH1    
 Detector ADSC Quantum Q315    
 Wavelength (Å) 0.97485    
 Scan range (°) 360    
 Oscillation (°) 0.25    
 Space group P32    
a (Å) 74.62    
b (Å) 74.61    
c (Å) 340.53    
 Mosaicity (°) 0.180    
 Resolution (last shell) (Å) 2.39 (2.54–2.39)    
 Completeness (last shell) (%) 97.8 (92.6)    
Rsym (last shell) (%) 5.6 (48.7)    
I/σ(I) (last shell) 25.73 (4.91)    
 Number of observed/unique reflections 653063/164140    
 Wilson B factor (Å247.74    
 Twin fraction 0.392    
Molecular replacement     
 Mol/ASU    
 Phaser     
 RFZ/TFZ 24.0/25.9 21.8/50.3 8.4/49.3  
 LLG 820 3110 4738 4958 
Refinement     
Rwork/Rfree 0.1895/0.2321    
 Number of atoms used in refinement 14138    
 Number of solvent atoms 435    
 RMS deviation, bond lengths (Å) 0.008    
 RMS deviation, bond angles (°) 1.122    
 Mean B factor (Å228.5    
 Residues in most favoured/allowed region of Ramachandran plot (%) 99.9    
 PDB entry 2Y1V    

Structure solution and model refinement

Diffraction images were indexed and scaled with XDS [28], and merged with the CCP4 program suite [29]. Partial and perfect merohedral twinning tests were performed using the Yeates server (http://nihserver.mbi.ucla.edu/Twinning/) [30]. Initial phases were obtained by molecular replacement with PHASER [31], using as a search model the partial structure of RrgB, residues 186–627, PDB code 2X9W [25]. The final, full-length RrgB structure was rebuilt to remove bias from the model by employing ARP/wARP 7.1.1 [32] and Lafire 3.02 [33]. The full RrgB structure was completed by cycles of manual model building with COOT 0.5.2 [34]. In addition, water molecules were added to the residual electron density map using ARP/wARP 7.1.1. Cycles of restrained refinement employing NCS, TLS and amplitude-based twin refinement were performed with REFMAC 5.5 [35] as implemented in the CCP4 program suite (Table 1). Stereochemical verification was performed by PROCHECK [36] and secondary structure assignment by DSSP [37]. Figures were generated using PyMOL (http://www.pymol.org).

EM density map fitting

The rigid body fitting of FL-RrgB crystallographic co-ordinates (PDB code 3RPK) [27] and FL-RrgB-IPQTG (PDB code 2Y1V, the present study) into the pilus EM density map was performed using CHIMERA [38] as detailed previously [25,26].

Construction of TIGR4 RrgB mutant strains

TIGR4 (or T4) pneumococcal strains mutated in the rrgB gene were generated using the Janus cassette [39]. The strains constructed are described in Table 2. The gene rpsL1 was inserted in the locus rpsL conferring resistance to streptomycin (StrR). The T4∇(rpsL1) strain displays the same phenotype as the wild-type T4 strain. The kanamycin-resistance Janus cassette was amplified using primers containing upstream and downstream sequences homologous with regions flanking rrgB. The PCR product was transformed into the TIGR4 strain as previously described [40] and then plated on to Columbia blood agar plates containing kanamycin (250 μg/ml) in order to select for T4∇(rpsL1)Δ(rrgB) strains. The wild-type rrgB gene or genes harbouring the E143A, K162A and K183A point mutations were introduced in replacement of the kan-rpsL Janus cassette, leading to the suppression of the kanamycin resistance phenotype. The sequence of each mutant was verified by sequencing.

Table 2
S. pneumoniae TIGR4 strains

KanR, kanamycin resistance; StrR, streptomycin resistance.

Strain Relevant characteristics 
T4 Serotype 4 TIGR4 strain 
T4∇(rpsL1StrR 
T4∇(rpsL1)Δ(rrgB)∇(kan-rpsLKanR, rrgB ::kan-rpsL 
T4∇(rpsL1)Δ(rrgB)∇(rrgBStrR, rrgB 
T4∇(rpsL1)Δ(rrgB)∇(rrgB E143AStrR, rrgB E143A 
T4∇(rpsL1)Δ(rrgB)∇(rrgB K162AStrR, rrgB K162A 
T4∇(rpsL1)Δ(rrgB)∇(rrgB K183AStrR, rrgB K183A 
Strain Relevant characteristics 
T4 Serotype 4 TIGR4 strain 
T4∇(rpsL1StrR 
T4∇(rpsL1)Δ(rrgB)∇(kan-rpsLKanR, rrgB ::kan-rpsL 
T4∇(rpsL1)Δ(rrgB)∇(rrgBStrR, rrgB 
T4∇(rpsL1)Δ(rrgB)∇(rrgB E143AStrR, rrgB E143A 
T4∇(rpsL1)Δ(rrgB)∇(rrgB K162AStrR, rrgB K162A 
T4∇(rpsL1)Δ(rrgB)∇(rrgB K183AStrR, rrgB K183A 

Generation of a TIGR4 strain overexpressing a monomeric RrgB in the culture supernatant

TIGR4Δ(PI-1)/pMU1328 FL-rrgB-Spn was created by gener-ating a TIGR4Δ(PI-1) isogenic mutant, which was then transformed with a pMU1328 plasmid containing the rrgB gene deprived of the region coding for the LPXTG cell sorting signal. Briefly, the TIGR4Δ(PI-1) isogenic mutant was generated as reported previously [26] by using the oligonucleotides listed in Table 3. The FL-RrgB-Spn-expressing plasmid [41] was then obtained by inserting the spr0096 promoter region (intspr0095-spr0096) and the rrgB gene fragment within the pMU1328 vector, by using the primers and restriction enzymes reported in Table 3. Finally, pMU1328 FL-rrgB-Spn was transformed into TIGR4Δ(PI-1) and the transformants were selected on agarose plates supplemented with antibiotics (1μg/ml erythromycin with 500 μg/ml kanamycin).

Table 3
Primers used for the construction of FL-RrgB-Spn

The plasmid used was pMU1328 FL-rrgB-Spn. Underlined sequences correspond to the restriction recognition sites. The bold underlined sequence encodes the 6×His tag protein fragment. Sequences in italics represent the complementary bases allowing the fusion PCR reaction.

Section Amplification primers (5′–3′) 
Promoter region  
 Forward GTGCGTGGATCCGATGATATCAAAGACAGATTGAAA 
 Reverse ATTCGAAAATTCTCCTTCTTTCTA 
Gene  
 Forward TAGAAAGAAGGAGAATTTTCGAATATGAAATCAATCAACAAATTTTTA 
 Reverse CAGCGTGTCGACTTAATGGTGATGGTGATGGTGAGTGATTTTTTTGTTGACTACTTT 
Section Amplification primers (5′–3′) 
Promoter region  
 Forward GTGCGTGGATCCGATGATATCAAAGACAGATTGAAA 
 Reverse ATTCGAAAATTCTCCTTCTTTCTA 
Gene  
 Forward TAGAAAGAAGGAGAATTTTCGAATATGAAATCAATCAACAAATTTTTA 
 Reverse CAGCGTGTCGACTTAATGGTGATGGTGATGGTGAGTGATTTTTTTGTTGACTACTTT 

Expression and purification of FL-RrgB-Spn

TIGR4ΔPI-1/pMU1328 FL-rrgB-Spn strain was grown in Todd–Hewitt broth [0.5% yeast extract supplemented with erythromycin (1 μg/ml) and kanamycin (500 μg/ml)]. Following bacterial harvesting by centrifugation (20 min at 3250 g), the supernatants were filtered (0.2 μm filter pore size), dialysed overnight against a 100 mM sodium phosphate buffer (pH 6.3) and then supplemented with sodium chloride and imidazole (100 mM and 20 mM final concentrations respectively). The FL-RrgB-Spn protein was then purified by metal chelate affinity chromatography on HP Sepharose columns (GE Healthcare), eluting with a 100 mM sodium phosphate buffer (pH 6.3) supplemented with 100 mM sodium chloride and 500 mM imidazole. Protein purity was determined by SDS/PAGE. Fractions containing the purified protein were stored at −80°C until further use.

Western blotting

Wild-type and mutant TIGR4 strains were grown in Todd–Hewitt broth (BD Biosciences) supplemented with 0.5% yeast extract (Sigma–Aldrich) at 37°C with 5% CO2, until a D600 of 0.45 was reached. Cultures were centrifuged (20 min at 3250 g) and the bacterial pellet, after being washed with PBS, was treated with 200 units/ml of mutanolysin (Sigma), 1 mg/ml lysozyme (Sigma) and a protease inhibitor cocktail (Complete, Roche) for 3 h at 37°C. Cellular debris were removed by centrifugation at 39000 g for 15 min. Each sample (20 μl) was mixed with XT Sample Buffer and XT Reducing Agent (Bio-Rad), boiled at 100°C for 10 min and loaded on to 4%–12% Criterion XT Precast gels (Bio-Rad). Gels were run for approximately 3 h and subsequently electrotransferred in Trans-Blot Transfer Medium (Bio-Rad). Incubations of 1 h were successively performed using anti-RrgB polyclonal mouse antibodies (diluted 1:5000) and anti-mouse horseradish peroxidase conjugate antibodies (Sigma; diluted 1:120000) before detection with a chemiluminescent substrate (Pierce).

Accession number

The structural co-ordinates of full-length RrgB reported will appear in the PDB under accession code 2Y1V.

RESULTS AND DISCUSSION

The major pilin RrgB is composed of four structural domains

The FL-RrgB-IPQTG construct used for crystallization lacks the regions corresponding to the signal peptide and the sequence downstream from the IPQTG sorting motif (Figure 1A). The crystal structure of FL-RrgB-IPQTG (residues 30–633) was solved at 2.39 Å (1 Å=0.1 nm) resolution by employing a partial model of RrgB [25] as a search model in a molecular replacement experiment. Data collection, phasing and refinement statistics are included in Table 1. FL-RrgB-IPQTG displays a modular structure comprising four domains: D1, D2 and D4 form ‘beads on a string’, whereas D3 is laterally arranged relatively to D2 (Figure 1B). The arrangement of domains D2–D4 has previously been visualized in the structure of an RrgB fragment which lacked D1 [25]. The cores of all four domains display β-barrel folds, with additional β-strands and one or two helices decorating one side of the β-barrel in the D1 and D4 domains (Figure 1B). Interestingly, once the D3 β-sandwich is formed, its C-terminal region participates in completion of the D2 β-barrel through the contribution of a loop region and a short β-sheet. The D1, D3 and D4 domains of RrgB correspond to the prototype CNA-B fold, whereas the D2 domain emulates the CNA-A-like region from the CNA collagen-binding protein [12,13]. The closest structural homologues of such folds are the CNA-A or CNA-B-type domains of BcpA and SpaA, the major pilins of B. cereus and C. diphtheriae respectively, as well as GBS52 and RrgA, minor pilins of streptococcal species [16,17,24,42]. Details of structural comparisons are given in Supplementary Tables S1 and S2 (at http://www.BiochemJ.org/bj/441/bj4410833add.htm). In addition to the FL-RrgB-IPQTG structure solved in the present study, two other RrgB structures have been solved deleted from the IPQTG sorting motif: a truncated variant lacking the D1 domain [25] and a full-length form [27]. Comparisons of these structures with FL-RrgB-IPQTG are presented (Figures 1C and 1D). Spatial organization, as well as secondary elements of D2, D3 and D4 domains, are common to all three RrgB structures (Figures 1C and 1D). It is of note that the D1 domains in both full-length RrgB proteins are positioned further apart with respect to D2, and their orientation is rotated by approximately 180° with respect to each other (Figure 1C). It was indeed described that the D1 domain in the FL-RrgB structure presents some flexibility owing to the low amount of contacts this domain makes with the D2 domain [27]. On the contrary, the D1 domain of the FL-RrgB-IPQTG is trapped in the RrgB–RrgB complex, which probably constrains the D1 positioning and decreases flexibility.

Schematic arrangement and tertiary fold of full-length RrgB

Figure 1
Schematic arrangement and tertiary fold of full-length RrgB

(A) RrgB is composed of four domains. (B) Modular organization of the three-dimensional FL-RrgB-IPQTG structure. (C) Superimposition of the structures of FL-RrgB-IPQTG (blue) with FL-RrgB (green) [27], views at 0° and 90° orientations. (D) Superimposition of the structures of FL-RrgB-IPQTG (blue) with the D2-D4 RrgB truncated form (orange) [25].

Figure 1
Schematic arrangement and tertiary fold of full-length RrgB

(A) RrgB is composed of four domains. (B) Modular organization of the three-dimensional FL-RrgB-IPQTG structure. (C) Superimposition of the structures of FL-RrgB-IPQTG (blue) with FL-RrgB (green) [27], views at 0° and 90° orientations. (D) Superimposition of the structures of FL-RrgB-IPQTG (blue) with the D2-D4 RrgB truncated form (orange) [25].

Fibrillar organization of FL-RrgB-IPQTG molecules in the crystal packing: a model for pneumococcal pilus polymerization

The asymmetric unit of the crystals contains three RrgB molecules that generate columns of monomers that align with the z axis (Figure 2A). Molecules stack head-to-tail and a rotation angle of approximately 120° is observed between two successive RrgB monomers (Figure 2A). The head-to-tail organization is due to the packing of the I628PQTG632 sorting recognition motif located at the C-terminus of the D4 domain of one RrgB molecule into a cavity of the D1 domain of the successive RrgB monomer (Figure 2). Within the IPQTG sequence, the side-chain of Ile628 is buried in a hydrophobic cavity in proximity to the D1 residues Leu172, Ile168, Val180 and Ile166 (Figure 2B). Sortase enzymatic activity cleaves the Thr631–Gly632 bond, after which peptidic bond formation between the new C-terminus and a juxtaposed lysine residue covalently links adjacent RrgB subunits. As observed in the crystal structure accounting for an RrgB-fibre-like complex, the most favourable candidates for this intermolecular link are either Lys162 or Lys183, which are present within the D1 cavity (Figure 2B and see Supplementary Figure S1 at http://www.BiochemJ.org/bj/441/bj4410833add.htm). Lys162 is part of the EGSKVPI sequence and Lys183 is part of the YPKN pilin motif, both of which are potentially involved in RrgB intermolecular bonds, as described previously in BcpA, SpaA and Spy0128 [1517].

Crystal packing of FL-RrgB-IPQTG molecules

Figure 2
Crystal packing of FL-RrgB-IPQTG molecules

(A) Head-to-tail stacking of two successive RrgB molecules as observed in the asymmetric unit: the C-terminal IPQTG motifs are represented by spheres, in blue for molecule 1 and in cyan for molecule 2. Molecules stack head-to-tail and a rotation angle of approximately 120° is observed between two successive RrgB monomers. The encircled structure of the IPQTG subunit 1 docked into the D1 domain of subunit 2, in which Lys183 is represented by sticks and coloured in magenta, is magnified in the inset. (B) Surface representation of the RrgB complex. The IPQTG residues (from molecule 1) are represented in ball-and-sticks. In the D1 domain (from molecule 2), the hydrophobic cavity is highlighted in orange, and Lys183 and Lys162 are represented in violet and cyan respectively.

Figure 2
Crystal packing of FL-RrgB-IPQTG molecules

(A) Head-to-tail stacking of two successive RrgB molecules as observed in the asymmetric unit: the C-terminal IPQTG motifs are represented by spheres, in blue for molecule 1 and in cyan for molecule 2. Molecules stack head-to-tail and a rotation angle of approximately 120° is observed between two successive RrgB monomers. The encircled structure of the IPQTG subunit 1 docked into the D1 domain of subunit 2, in which Lys183 is represented by sticks and coloured in magenta, is magnified in the inset. (B) Surface representation of the RrgB complex. The IPQTG residues (from molecule 1) are represented in ball-and-sticks. In the D1 domain (from molecule 2), the hydrophobic cavity is highlighted in orange, and Lys183 and Lys162 are represented in violet and cyan respectively.

To validate the relevance of these structural observations in the native pilus, we constructed a series of S. pneumoniae TIGR4 strain variants in which RrgB expression is suppressed by gene deletion and complemented by insertion of wild-type or mutated forms (K162A and K183A) of RrgB. The effects of these mutations on RrgB polymerization were analysed by Western blotting on cell wall extracts (Figure 3). The wild-type pattern corresponds to a ladder of high-molecular-mass bands, which reflects the covalent nature of the association between RrgB subunits and the size distribution of RrgB polymers (Figure 3, lane 1). The deletion of rrgB totally abolishes fibre formation (Figure 3, lane 2), which is fully restored once the strain is complemented with the wild-type rrgB gene (Figure 3, lane 6). Complementation with the K162A mutant leads to the formation of a wild-type RrgB polymer, indicating that this residue does not affect RrgB polymerization (Figure 3, lane 4), whereas mutation of Lys183 inhibits the formation of the polymer, leading to the accumulation of RrgB monomers (Figure 3, lane 5), as recently reported by Gentile et al. [26]. This result indicates that Lys183, which is located within the pilin motif in the D1 domain, is involved in forming a covalent bond with the IPQTG motif of a neighbouring RrgB, leading to RrgB polymerization. These findings strongly suggest that the fibrillar packing of FL-RrgB-IPQTG monomers observed in the crystal lattice is representative of the transpeptidation reaction involved in the RrgB pilus formation. It is of note that the FL-RrgB-IPQTG structure reported in the present study is the first description of a complete soluble major pilin, since structures of Spy0128, BcpA, SpaA and RrgB lack their respective LPXTG motifs [1517,25,27].

In vivo RrgB polymerization

Figure 3
In vivo RrgB polymerization

Mutanolysin extracts from TIGR4 strains were loaded on to 4%–12% Criterion XT Precast gels (Bio-Rad) and Western blot analysis was performed using polyclonal rabbit antisera against RrgB. Lane 1, T4∇(rpsL1); lane 2, T4∇(rpsL1)Δ(rrgB)∇(kan-rpsL); lane 3, T4∇(rpsL1)Δ(rrgB) ∇(rrgB E143A); lane 4, T4∇(rpsL1)Δ(rrgB)∇(rrgB K162A); lane 5, T4∇(rpsL1)Δ(rrgB)∇(rrgB K183A); and lane 6, T4∇(rpsL1)Δ(rrgB)∇(rrgB). The RrgB monomer is indicated by the arrow.

Figure 3
In vivo RrgB polymerization

Mutanolysin extracts from TIGR4 strains were loaded on to 4%–12% Criterion XT Precast gels (Bio-Rad) and Western blot analysis was performed using polyclonal rabbit antisera against RrgB. Lane 1, T4∇(rpsL1); lane 2, T4∇(rpsL1)Δ(rrgB)∇(kan-rpsL); lane 3, T4∇(rpsL1)Δ(rrgB) ∇(rrgB E143A); lane 4, T4∇(rpsL1)Δ(rrgB)∇(rrgB K162A); lane 5, T4∇(rpsL1)Δ(rrgB)∇(rrgB K183A); and lane 6, T4∇(rpsL1)Δ(rrgB)∇(rrgB). The RrgB monomer is indicated by the arrow.

In order to understand better the disposition of the single RrgB monomers into the pilus scaffold, the two FL-RrgB X-ray co-ordinates (FL-RrgB and FL-RrgB-IPQTG) were fitted in the EM density map obtained from purified preparations of pilus by a single particle reconstruction method [25]. As shown in Figure 4(A), the rigid body fitting of both structures resulted in the exposure outside the density of the same residues (Asn87–Ala117), corresponding to a loop region. Interestingly, this loop results in twisting by 180° of the two fittings due to the 180° flip of the D1 domain in the two FL-RrgB structures (Figure 4B). In addition, this loop, which by definition is a very flexible and mobile region, is exposed outside the EM density due to the orientation of D1 with respect to D2, which, in turn, is probably depending on the high rotational flexibility of the short loop (Tyr181–Pro189) connecting D1 to D2, as already reported by Gentile et al. [26], in the NMR solution structure of the single D1 domain. Notably, the leftover volume after the fitting of the FL-RrgB-IPQTG structure in the EM density map could be fitted by the same residues (Asn87–Ala117) if differently folded. Finally, the RrgB-fibre molecules, as organized into the FL-RrgB-IPQTG crystal packing, can be fitted into the EM density map of the native pneumoccoccus pilus (Figure 4C).

FL-RrgB crystal structures fit into EM density map of native pneumococcal pili

Figure 4
FL-RrgB crystal structures fit into EM density map of native pneumococcal pili

(A) X-ray co-ordinates of FL-RrgB (yellow, [27]) and FL-RrgB-IPQTG (pink, the present study) were fitted into the EM density map; the loops that do not fit are coloured in orange and green respectively. (B) Superimposition of the D1 domains, as orientated into the crystal structures of FL-RrgB and FL-RrgB-IPQTG. The same residues (Asn87–Ala117) form the loops exposed outside the EM map, which are flipped 180° with respect to each other. (C) Fit of two associated FL-RrgB-IPQTG molecules into the EM map.

Figure 4
FL-RrgB crystal structures fit into EM density map of native pneumococcal pili

(A) X-ray co-ordinates of FL-RrgB (yellow, [27]) and FL-RrgB-IPQTG (pink, the present study) were fitted into the EM density map; the loops that do not fit are coloured in orange and green respectively. (B) Superimposition of the D1 domains, as orientated into the crystal structures of FL-RrgB and FL-RrgB-IPQTG. The same residues (Asn87–Ala117) form the loops exposed outside the EM map, which are flipped 180° with respect to each other. (C) Fit of two associated FL-RrgB-IPQTG molecules into the EM map.

The FL-RrgB-IPQTG structure includes four isopeptide bonds

The resistance and stability of Gram-positive pili has been shown to require the formation of isopeptide amide bonds within the pilin monomers themselves [19,20,43]. These bonds are the product of an intramolecular reaction between the side chains of lysine and asparagine residues, the resulting link being stabilized by a hydrogen bond with an acidic residue in the immediate vicinity. The structure of the FL-RrgB-IPQTG reveals the presence of four intramolecular isopeptide bonds, one in each of the D1, D2, D3 and D4 domains (Figure 1A). The bond within D1 is formed between the side chains of Lys41 and Asn184, and is stabilized by the side chain of Glu143, which is part of an E-box motif. The isopeptide bonds in the D3 (Lys349–Asn428) and D4 (Lys453–Asn623) domains are stabilized by residues Glu405 and Glu577 respectively, which belong to E-box motifs, whereas the intramolecular link in the D2 domain (Lys193–Asn318) is coupled to Asp241, located within a D-box.

Although the intramolecular isopeptide bonds in D2, D3 and D4 were present in the already solved RrgB crystal structures including these domains [25,27], the presence of a D1 isopeptide bond in the crystal structure of FL-RrgB-IPQTG was unexpected. In fact, this bond is absent in the FL-RrgB-IPQTG protein in solution [19], in the NMR solution structure of the isolated D1 domain [26], and also in the D1 domain included in the crystal structure of the FL-RrgB [27]. Uniquely in the FL-RrgB-IPQTG structure, the D1 domain structure shows the presence of two secondary elements, β-strand-10 (Ala156–Val164) and β-strand-13 (Pro182–Lys186) (Figure 5, and Supplementary Figures S2 and S3 at http://www.BiochemJ.org/bj/441/bj4410833add.htm). β-strand-13 harbours the residue Lys183 from the pilin motif, as well as residue Asn184, both involved in the isopeptide bond formation. A detailed view of the IPQTG sequence docked in the D1 domain is shown in Figure 5(A). The Lys183 from the pilin motif and Lys162 are orientated in the same direction. The Asn184 position is constrained by β-strand-13, which allows interaction with Lys41 to form the isopeptide bond. In order to appreciate the structural modifications induced by the sorting sequence docked into the D1 domain, the corresponding view of the FL-RrgB structure is shown, in which the D1 isopeptide bond is not formed (Figure 5B) [27]. The most striking features are the modifications of the β structures, the rotation by approximately 90° left of the Lys162 position and the observation that Asn184 is no longer held by a β-strand, which alters its orientation and impairs its contact with Lys41. This suggests that the β-strand structure is influenced by the interaction of the RrgB subunits provided by the fibrillar crystal packing and/or the D1 isopeptide bond formation. In both NMR and crystallographic structures of FL-RrgB, Asn184 is distant from the Lys41 side chain, whereas in the FL-RrgB-IPQTG structure, both residues are joined to form a covalent bond (Supplementary Figure S2). It is of interest that the position of Glu143 is common to all three structures, independently of the lysine–asparagine bond formation (Supplementary Figure S2). Since all of the RrgB constructs were produced in E. coli expression systems, we wondered if the production of FL-RrgB in a monomeric secreted form in pneumococcal cells (FL-RrgB-Spn) would allow efficient conditions for the formation of the D1 isopeptide bond. For this purpose, FL-RrgB-Spn was overexpressed and purified from pneumococcal growth supernatants, and its molecular mass was determined by ESI–MS (electrospray ionization MS). The molecular mass was 65216.82 Da, corresponding to the presence of three isopeptide bonds (calculated mass of 65215.74 Da).

Structural rearrangement induced by the sorting sequence docked in the D1 domain

Figure 5
Structural rearrangement induced by the sorting sequence docked in the D1 domain

Lys183 and Lys162 are represented in green and magenta respectively, and Asn184, which forms the isopeptide bond, is coloured in orange. (A) Structure of FL-RrgB-IPQTG (the present study). The IPQTG peptide is represented in ball-and-sticks and coloured in blue. A stereoview of this image is shown in Supplementary Figure S3 (at http://www.BiochemJ.org/bj/441/bj4410833add.htm). (B) Structure of FL-RrgB [27].

Figure 5
Structural rearrangement induced by the sorting sequence docked in the D1 domain

Lys183 and Lys162 are represented in green and magenta respectively, and Asn184, which forms the isopeptide bond, is coloured in orange. (A) Structure of FL-RrgB-IPQTG (the present study). The IPQTG peptide is represented in ball-and-sticks and coloured in blue. A stereoview of this image is shown in Supplementary Figure S3 (at http://www.BiochemJ.org/bj/441/bj4410833add.htm). (B) Structure of FL-RrgB [27].

Taken together, the results from the present study indicate that lack of the D1 isopeptide bond in the monomeric soluble RrgB protein and in the crystals obtained so far is not related to the expression context, and suggest that the D1 isopeptide bond formation relies mainly on the RrgB molecules association in a fibre-like arrangement, even though not necessarily covalently linked. However, the presence of the D1 isopeptide bond in the native pneumococcal pilus remains to be verified. Related observations have been reported in the B. cereus major pilin BcpA: the mature polymerized protein forming the pilus fibre shaft was shown to contain an intramolecular bond in the CNA1 domain (Lys37–Asn163), which was absent in the soluble recombinant protein [16]. The comparison of RrgB D1 domain structures, in the absence or presence of the covalent intramolecular bond, suggests that the interaction between the IPQTG motif and Lys183, as observed in the FL-RrgB-IPQTG crystal lattice, may induce a local rearrangement of the Asn184 position, allowing it to be positioned slightly closer to Lys41 and facilitating covalent bonding. A recent publication reports that two fragments engineered from the major pilin Spy0128, a large region (residues 18–299, containing Lys179 and Glu258) and a peptide (residues 293–308, containing Asn303), associate spontaneously and covalently to each other through the formation of the isopeptide bond Lys179–Asn303 [44]. This suggests that the structural environment of the triad ‘lysine–asparagine–acidic residue’ influences the formation of the isopeptide bond, which, in addition, can be formed following folding of its parent protein.

Role of the D1 isopeptide bond in RrgB stabilization

The presence of isopeptide bonds within D2, D3 and D4 domains was shown to be important for protein stabilization as measured by analysing the proteolytic susceptibility and the thermal stability of RrgB variants [19]. Disruption of the D2 isopeptide bond induced the most important tryptic degradation when compared with D3 and D4 bonds [19]. D2, D3 and D4 isopeptide bonds contributed equally and significantly to the RrgB thermal stability. Although the native RrgB did not show any thermal denaturation, the Tm values of D241A, E405A and E577A were 58.5°C, 58.9°C and 48.6°C respectively [19]. The triple RrgB mutant that does not contain any intramolecular bonds has a deeply decreased thermal stability reflected by the low Tm value of 43.1°C [19]. The observation of an additional bond in the D1 domain, as revealed in the crystal structure described in the present study, led us to investigate its potential role in RrgB stabilization. The residues Lys41, Glu143 and Asn184, which are directly involved in the D1 isopeptide bond formation, were individually mutated into an alanine residue in the FL-RrgB-IPQTG construct. The typical signature of the formation of the isopeptide bond, the loss of 17 Da, was not detected by ESI–MS [19], confirming that the D1 isopeptide bond is not formed in RrgB in solution in either the wild-type or mutated proteins (Table 4). The RrgB E143A mutant is more susceptible to trypsin proteolysis than wild-type RrgB (Figure 6). Although trypsin treatment of both RrgB forms yielded a fragment with an apparent molecular mass of approximately 35 kDa, proteolysis of RrgB-E143A was almost complete after 120 min, when approximately 50% of wild-type RrgB appeared unaffected (Figure 6). ESI–MS and N-terminal sequencing analyses revealed that the fragment corresponds to a molecular mass of 48638.64 Da and the N-terminal sequence N184TEAKPK (Table 5). The molecular mass discrepancy determined by ESI–MS and SDS/PAGE probably originates from the highly compact fold of RrgB induced by the three isopeptide bonds [19]. Trypsin cleavage occurs at Lys183, the conserved residue of the pilin motif, adjacent to Asn184, which participates in isopeptide bond formation. The E143A mutation, introduced in the native RrgB protein expressed in the TIGR4 strain, lowers the amounts of monomeric and polymeric RrgB (Figure 3, lanes 1 and 6). This latter observation suggests that the absence of the D1 isopeptide bond in the native pilus induces a substantial destabilization of the RrgB molecule, which leads to a less compact RrgB form more susceptible to in vivo degradation, thus impairing efficient RrgB polymerization.

Table 4
ESI–TOF (time-of-flight) MS analyses of the native and mutant RrgB proteins
 Maverage* (Da) 
RrgB Expected Observed by ESI–TOF Δ(MexpectedMobserved) (Da) NH3 units lost 
Wild-type 67453 67399 54 
K41A 67343 67396 52 
E143A 67395 67341 54 
N184A 67357 67410 52 
 Maverage* (Da) 
RrgB Expected Observed by ESI–TOF Δ(MexpectedMobserved) (Da) NH3 units lost 
Wild-type 67453 67399 54 
K41A 67343 67396 52 
E143A 67395 67341 54 
N184A 67357 67410 52 
*

Average molecular mass.

Difference between expected and observed molecular mass.

Table 5
Molecular masses and N-terminal sequences of the trypsin digestion of wild-type and mutant RrgB variants
RrgB Undigested measured mass Undigested N-terminal sequence Digested measured mass Digested N-terminal sequence Ratio 
Wild-type 67399 AHHHHH 67399 AHHHHH 63% 
   48639 NTEAK 37% 
E143A 67341 AHHHHH 67341 AHHHH 20% 
   48639 NTEAK 80% 
RrgB Undigested measured mass Undigested N-terminal sequence Digested measured mass Digested N-terminal sequence Ratio 
Wild-type 67399 AHHHHH 67399 AHHHHH 63% 
   48639 NTEAK 37% 
E143A 67341 AHHHHH 67341 AHHHH 20% 
   48639 NTEAK 80% 

Proteolysis stability of RrgB

Figure 6
Proteolysis stability of RrgB

Trypsin digestion products were separated by SDS/PAGE (12.5% gels). Molecular mass markers are shown on the left-hand side of the gel in kDa. Trypsin addition and incubation times are noted above the gel.

Figure 6
Proteolysis stability of RrgB

Trypsin digestion products were separated by SDS/PAGE (12.5% gels). Molecular mass markers are shown on the left-hand side of the gel in kDa. Trypsin addition and incubation times are noted above the gel.

Abbreviations

     
  • EM

    electron microscopy

  •  
  • ESI–MS

    electrospray ionization MS

  •  
  • TOF

    time-of-flight

AUTHOR CONTRIBUTION

Lamya El Mortaji carried out the in vivo and biochemical experiments, performed data analysis and helped with Figure preparation. Carlos Contreras-Martel solved the crystal structure, analysed the data and helped with Figure preparation. Monica Moschioni and Ilaria Ferlenghi provided the EM data and Figures, as well as RrgB produced in S. pneumoniae. Clothilde Manzano performed crystallization assays. Thierry Vernet and Andrea Dessen helped with experiment interpretation and edited the paper prior to submission. Anne Marie Di Guilmi designed the research, analysed the data and wrote the paper.

We thank Rémi Terrasse for technical help, Izabel Bérard and Luca Signor (LSMP, IBS) for MS measurements, J.-P. Andrieu (LEM, IBS) for N-terminal sequencing, J. Marquez and the HTX Lab team (HTXLab, Partnership for Structural Biology) for access to high-throughput crystallization robotics, and the staff of ESRF beamline ID23 for help with data collection. We are greatly indebted to J.P. Claverys (Paul Sabatier University, Toulouse, France) and C. Grangeasse (Institute of the Biology and Chemistry of Proteins, Lyon University, Lyon, France) for providing the Janus cassette, the R800 and R1226 pneumococcal strains and for helpful discussions.

FUNDING

This work was partly supported by the European Commission [grant number LSHM-CT-2004-512138] and the Agence Nationale de la Recherche [grant number 05-JC-JC-0049].

References

References
1
Mandlik
A.
Swierczynski
A.
Das
A.
Ton-That
H.
Corynebacterium diphtheriae employs specific minor pilins to target human pharyngeal epithelial cells
Mol. Microbiol.
2007
, vol. 
64
 (pg. 
111
-
124
)
2
Abbot
E. L.
Smith
W. D.
Siou
G. P. S.
Chiriboga
C.
Smith
R. J.
Wilson
J. A.
Hirst
B. H.
Kehoe
M. A.
Pili mediate specific adhesion of Streptococcus pyogenes to human tonsil and skin. Cell
Microbiol.
2007
, vol. 
9
 (pg. 
1822
-
1833
)
3
Dramsi
S.
Caliot
E.
Bonne
I.
Guadagnini
S.
Prévost
M. C.
Kojadinovic
M.
Lalioui
L.
Poyart
C.
Trieu-Cuot
P.
Assembly and role of pile in group B streptococci
Mol. Microbiol.
2006
, vol. 
60
 (pg. 
1401
-
1413
)
4
Konto-Ghiorghi
Y.
Mairey
E.
Mallet
A.
Duménil
G.
Caliot
E.
Trieu-Cuot
P.
Dramsi
S.
Dual role for pilus in adherence to epithelial cells and biofilm formation in Streptococcus agalactiae
PLOS Pathogens
2009
, vol. 
5
 pg. 
e1000422
 
5
Barocchi
M. A.
Ries
J.
Zogaj
X.
Hemsley
C.
Albiger
B.
Kanth
A.
Dahlberg
S.
Fernebro
J.
Moschioni
M.
Masignani
V.
, et al. 
A pneumococcal pilus influences virulence and host inflammatory responses
Proc. Natl. Acad. Sci. U.S.A.
2006
, vol. 
103
 (pg. 
2857
-
2862
)
6
Nelson
A. L.
Ries
J.
Bagnoli
F.
Dahlberg
S.
Falkner
S.
Rounioja
S.
Tschop
J.
Morfeldt
E.
Ferlenghi
I.
Hilleringmann
M.
, et al. 
RrgA is a pilus-associated adhesin in Streptococcus pneumoniae
Mol. Microbiol.
2007
, vol. 
66
 (pg. 
329
-
340
)
7
Mora
M.
Bensi
G.
Capo
S.
Falugi
F.
Zingaretti
C.
Manetti
A. G.
Maggi
T.
Taddei
A. R.
Grandi
G.
Telford
J. L.
Group A Streptococcus produce pilus-like structures containing protective antigens and Lancefield T antigens
Proc. Natl. Acad. Sci. U.S.A.
2005
, vol. 
102
 (pg. 
15641
-
15646
)
8
Gianfaldoni
C.
Censini
S.
Hilleringmann
M.
Moschioni
M.
Facciotti
C.
Pansegrau
W.
Masignani
V.
Covacci
A.
Rappuoli
R.
Barocchi
M. A.
Ruggiero
P.
Streptococcus pneumoniae pilus subunits protect mice against lethal challenge
Infect. Immun.
2007
, vol. 
75
 (pg. 
1059
-
1062
)
9
Dramsi
S.
Magnet
S.
Davison
S.
Arthur
M.
Covalent attachment of proteins to peptidoglycan
FEMS Microbiol. Rev.
2008
, vol. 
32
 (pg. 
307
-
320
)
10
Clancy
K. C.
Melvin
J. A.
McCafferty
D. G.
Sortase transpeptidases: insights into mechanism, substrate specificity, and inhibition
Biopolymers
2010
, vol. 
94
 (pg. 
385
-
396
)
11
Reference deleted
12
Deivanayagam
C. C.
Rich
R. L.
Carson
M.
Owens
R. T.
Danthuluri
S.
Bice
T.
Höök
M.
Narayana
S. V.
Novel fold and assembly of the repetitive B region of the Staphylococcus aureus collagen-binding surface protein
Structure
2000
, vol. 
8
 (pg. 
67
-
78
)
13
Zong
Y.
Xu
Y.
Liang
X.
Keene
D. R.
Höök
A.
Gurusiddappa
S.
Höök
M.
Narayana
S. V.
A ‘Collagen Hug’ model for Staphylococcus aureus CNA binding to collagen
EMBO J.
2005
, vol. 
24
 (pg. 
4224
-
4236
)
14
Vengadesan
K.
Narayana
S. V.
Structural biology of gram-positive bacterial adhesins
Protein Sci.
2011
, vol. 
20
 (pg. 
759
-
772
)
15
Kang
H. J.
Coulibaly
F.
Clow
F.
Proft
T.
Baker
E. N.
Stabilizing isopeptide bonds revealed in Gram-positive bacterial pilus structure
Science
2007
, vol. 
318
 (pg. 
1625
-
1628
)
16
Budzik
J. M.
Poor
C. B.
Faull
K. F.
Whitelegge
J. P.
He
C.
Schneewind
O.
Intramolecular amide bonds stabilize pili on the surface of bacilli
Proc. Natl. Acad. Sci. U.S.A.
2009
, vol. 
106
 (pg. 
19992
-
19997
)
17
Kang
H. J.
Paterson
N. G.
Gaspar
A. H.
Ton-That
H.
Baker
E. N.
The Corynebacterium diphtheriae shaft pilin SpaA is built of tandem Ig-like modules with stabilizing isopeptide and disulfide bonds
Proc. Natl. Acad. Sci. U.S.A.
2009
, vol. 
106
 (pg. 
16967
-
16971
)
18
Ton-That
H.
Schneewind
O.
Assembly of pili on the surface of Corynebacterium diphtheriae
Mol. Microbiol.
2003
, vol. 
50
 (pg. 
1429
-
1438
)
19
El Mortaji
L.
Terrasse
R.
Dessen
A.
Vernet
T.
Di Guilmi
A. M.
Stability and assembly of pilus subunits of Streptococcus pneumoniae
J. Biol. Chem.
2010
, vol. 
285
 (pg. 
12405
-
12415
)
20
Kang
H. J.
Baker
E. N.
Intramolecular isopeptide bonds: protein crosslinks built for stress?
Trends Biochem. Sci.
2010
, vol. 
815
 (pg. 
1
-
9
)
21
Hendrickx
A. P.
Budzik
J. M.
Oh
S. Y.
Schneewind
O.
Architects at the bacterial surface – sortases and the assembly of pili with isopeptide bonds
Nat. Rev. Microbiol.
2011
, vol. 
9
 (pg. 
166
-
176
)
22
LeMieux
J.
Hava
D. L.
Basset
A.
Camilli
A.
RrgA and RrgB are components of a multisubunit pilus encoded by the Streptococcus pneumoniae rlrA pathogenicity islet
Infect. Immun.
2006
, vol. 
74
 (pg. 
2453
-
2456
)
23
Hilleringmann
M.
Ringler
P.
Müller
S. A.
De Angelis
G.
Rappuoli
R.
Ferlenghi
I.
Engel
A.
Molecular architecture of Streptococcus pneumoniae TIGR4 pili
EMBO J.
2009
, vol. 
29
 (pg. 
3921
-
3930
)
24
Izoré
T.
Contreras-Martel
C.
El-Mortaji
L.
Manzano
C.
Terrasse
R.
Vernet
T.
Di Guilmi
A. M.
Dessen
A.
Structural basis of host cell recognition by the pilus adhesin from Streptococcus pneumoniae
Structure
2009
, vol. 
18
 (pg. 
106
-
115
)
25
Spraggon
G.
Koesema
E.
Scarselli
M.
Malito
E.
Biagini
M.
Norais
N.
Emolo
C.
Barocchi
M. A.
Giusti
F.
Hilleringmann
M.
, et al. 
Supramolecular organization of the repetitive backbone unit of the Streptococcus pneumoniae pilus
PLoS ONE
2010
, vol. 
5
 pg. 
e10919
 
26
Gentile
M. A.
Melchiorre
S.
Emolo
C.
Moschioni
M.
Gianfaldoni
C.
Pancotto
L.
Ferlenghi
I.
Scarselli
M.
Pansegrau
W.
Veggi
D.
, et al. 
Structural and functional characterization of the Streptococcus pneumoniae RrgB pilus backbone D1 domain
J. Biol. Chem.
2011
, vol. 
286
 (pg. 
14588
-
14597
)
27
Paterson
N. G.
Baker
E. N.
Structure of the full-length major pilin from Streptococcus pneumoniae: implications for isopeptide bond formation in Gram-positive bacterial pile
PLoS ONE
2011
, vol. 
6
 pg. 
e22095
 
28
Kabsch
W.
Automatic processing of rotation diffraction data from crystals of initially unknown symmetry and cell constant
J. Appl. Cryst.
1993
, vol. 
26
 (pg. 
795
-
800
)
29
Collaborative Computational Project, Number 4
The CCP4 suite: programs for protein crystallography
Acta Crystallogr. Sect. D Biol. Crystallogr.
1994
, vol. 
50
 (pg. 
760
-
763
)
30
Yeates
T.
Detecting and overcoming crystal twinning
Methods Enymol
1997
, vol. 
276
 (pg. 
344
-
358
)
31
Storoni
L.
McCoy
A.
Read
R.
Likelihood-enhanced fast rotation functions
Acta Crystallogr. Sect. D Biol. Crystallogr.
2004
, vol. 
57
 (pg. 
1373
-
1382
)
32
Cohen
S. X.
Ben Jelloul
M.
Long
F.
Vagin
A.
Knipscheer
P.
Lebbink
J.
Sixma
T. K.
Lamzin
V. S.
Murshudov
G. N.
Perrakis
A.
ARP/wARP and molecular replacement: the next generation
Acta Crystallogr. Sect. D Biol. Crystallogr.
2008
, vol. 
64
 (pg. 
49
-
60
)
33
Yao
M.
Zhou
Y.
Tanaka
I.
LAFIRE: software for automating the refinement process of protein-structure analysis
Acta Crystallogr. Sect. D Biol. Crystallogr.
2006
, vol. 
62
 (pg. 
189
-
196
)
34
Emsley
P.
Cowtan
K.
Coot: model-building tools for molecular graphics
Acta Crystallogr. D Biol. Crystallogr.
2004
, vol. 
60
 (pg. 
2126
-
2132
)
35
Murshudov
G.
Vagin
A.
Dodson
E.
Refinement of macromolecular structures by the maximum-likelihood method
Acta Crystallogr. Sect. D Biol. Crystallogr.
1997
, vol. 
53
 (pg. 
240
-
255
)
36
Laskowski
R. A.
MacArthur
M. W.
Moss
D. S.
Thornton
J. M.
PROCHECK: a program to check the stereochemical quality of protein structures
J. App. Cryst.
1993
, vol. 
26
 (pg. 
283
-
291
)
37
Kabsch
W.
Sander
C.
Dictionary of protein secondary structure: pattern recognition of hydrogen-bonded and geometrical features
Biopolymers
1983
, vol. 
22
 (pg. 
2577
-
2637
)
38
Pettersen
E. F.
Goddard
T. D.
Huang
C. C.
Couch
G. S.
Greenblatt
D. M.
Meng
E. C.
Ferrin
T. E.
UCSF Chimera–a visualization system for exploratory research and analysis
J. Comput. Chem.
2004
, vol. 
25
 (pg. 
1605
-
1612
)
39
Sung
C. K.
Li
H.
Claverys
J. P.
Morrison
D. A.
An rpsL cassette, Janus, for gene replacement through negative selection in Streptococcus pneumoniae
Appl. Environ. Microbiol.
2001
, vol. 
67
 (pg. 
5190
-
5196
)
40
Bricker
A. L.
Camilli
A.
Transformation of a type 4 encapsulated strain of Streptococcus pneumoniae
FEMS Microbiol. Lett.
1999
, vol. 
172
 (pg. 
131
-
135
)
41
Achen
M. G.
Davidson
B. E.
Hillier
A. J.
Construction of plasmid vectors for the detection of streptococcal promoters
Gene
1986
, vol. 
45
 (pg. 
45
-
49
)
42
Krishnan
V.
Gaspar
A. H.
Ye
N.
Mandlik
A.
Ton-That
H.
Narayana
S. V.
An IgG-like domain in the minor pilin GBS52 of Streptococcus agalactiae mediates lung epithelial cell adhesion
Structure
2007
, vol. 
15
 (pg. 
893
-
903
)
43
Kang
H. J.
Baker
E. N.
Intramolecular isopeptide bonds give thermodynamic and proteolytic stability to the major pilin protein of Streptococcus pyogenes
J. Biol. Chem.
2009
, vol. 
284
 (pg. 
20729
-
20737
)
44
Zakeri
B.
Howarth
M.
Spontaneous intermolecular amide bond formation between side chains for irreversible peptide targeting
J. Am. Chem. Soc.
2010
, vol. 
132
 (pg. 
4526
-
4527
)

Author notes

1

These authors contributed equally to this work.

The structural co-ordinates of full-length RrgB reported will appear in the PDB under accession code 2Y1V.

Supplementary data