An arm-swapped dimer of the Streptococcus pyogenes pilin specific assembly factor SipA

YJSBI 6403

No. of Pages 6, Model 5G

12 June 2013 Journal of Structural Biology xxx (2013) xxx–xxx 1

Contents lists available at SciVerse ScienceDirect

Journal of Structural Biology journal homepage: www.elsevier.com/locate/yjsbi

2

Structure Report

6 4 7

An arm-swapped dimer of the Streptococcus pyogenes pilin specific assembly factor SipA

5 8 9

10 1 5 2 2 13 14 15 16 17 18 19 20 21 22 23 24

Q1

Paul G. Young ⇑, Hae Joo Kang 1, Edward N. Baker School of Biological Sciences, University of Auckland, Auckland, New Zealand

a r t i c l e

i n f o

Article history: Received 26 April 2013 Received in revised form 28 May 2013 Accepted 30 May 2013 Available online xxxx Keywords: Streptococcus pyogenes SipA LepA Pilus assembly Signal peptidase

a b s t r a c t Streptococcus pyogenes (group A streptococcus [GAS]) is a major human pathogen. Attachment of GAS to host cells depends in large part on pili. These assemblies are built from multiple covalently linked subunits of a backbone protein (FctA), which forms the shaft of the pilus, and two minor pilin proteins, FctB anchoring the pilus to the cell wall and Cpa functioning as the adhesin at the tip. Polymerisation of the pilin subunits is mediated by a specific sortase, which catalyzes the formation of peptide bonds linking successive subunits. An additional gene, SipA, is also essential for GAS pilus polymerisation, but its function remains undefined. Here we report the crystal structure of a truncated SipA protein from GAS, determined at 1.67 Å resolution. The structure reveals that SipA has the same core fold as the Escherichia coli type-I signal peptidase (SPase-I), but has a much smaller non-catalytic domain. The truncated protein, which lacks 9 N-terminal residues, forms an arm-swapped dimer in which the C-terminal b-strand of each monomer crosses over to interact with an N-terminal strand from the other monomer. In addition, there is no peptide binding cleft and significant differences in the putative membrane association region. Ó 2013 Published by Elsevier Inc.

26 27 28 29 30 31 32 33 34 35 36 37 38 39

40 41

1. Introduction

42

The gram-positive pathogen Streptococcus pyogenes (Group A Streptococcus [GAS]) is a common human pathogen causing a spectrum of diseases ranging from mild localized infections such as strep throat, through to serious illnesses such as scarlet fever, rheumatic fever, and pneumonia (Cunningham, 2000). Recently it has been discovered that GAS produces pili on its surface (Mora et al., 2005). These pili are instrumental in mediating attachment of GAS to host cells and subsequent disease development (Abbot et al., 2007; Manetti et al., 2007). Pili on GAS are composed of multiple covalently-linked subunits of a major backbone pilin (FctA), and two minor pilin proteins, FctB and Cpa (Mora et al., 2005). The major pilin forms the polymeric backbone of the pilus (Kang et al., 2007; Mora et al., 2005), whereas the minor pilin Cpa forms the adhesin at the tip of the pilus (Quigley et al., 2009; Smith et al., 2010) and the basal pilin FctB covalently links the pilus to peptidoglycan of the cell wall (Hendrickx et al., 2011; Linke et al., 2010; Smith et al., 2010). Polymerisation of the pilin subunits is mediated by a specific sortase (SrtC), which catalyzes the formation of an isopeptide bond that joins one subunit to the next (Hendrickx

43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

⇑ Corresponding author. Address: Structural Biology, School of Biological Sciences, University of Auckland, 3a Symonds St., Auckland 1010, New Zealand. E-mail address: [email protected] (P.G. Young). 1 Present address: Membrane Protein Crystallography Group, Division of Molecular Biosciences, Imperial College London, London SW7 2AZ, UK.

et al., 2011; Kang et al., 2007; Mora et al., 2005). A second gene (sipA) has also shown to be essential for pilus polymerisation (Nakata et al., 2009; Zahner and Scott, 2008). The SipA protein (also known as LepA) has significant sequence similarity with type-I signal peptidases, which are membrane-bound serine proteases that cleave the N-terminal signal sequence from secreted proteins (Dalbey et al., 1997; Paetzel et al., 1998). These enzymes have a characteristic serine–lysine catalytic dyad, in which the serine acts as the nucleophile while the amino group of lysine provides the general base that deprotonates the serine hydroxyl group (Paetzel et al., 1998). Most bacteria typically have only one active signal peptidase, which is essential for growth and survival (Dalbey and Wickner, 1985; Inada et al., 1989). However, some gram-positive bacteria have several signal peptidases that appear to have overlapping sequence specificities. In addition, genomic sequencing has identified a growing number of gram-positive signal peptidase-like proteins that are predicted to have the same protein architecture but lack an identifiable catalytic dyad. Here we report the first structure from this family of non-active signal peptidases. This truncated SipA protein, which lacks 9 N-terminal residues that were removed to enhance solubility, shows 23% sequence identity with the Escherichia coli type-I signal peptidase (SPase-I). The structure confirms that SipA shares the core SPase-I fold, but reveals significant differences in the catalytic domain, loss of the peptide binding cleft present in E. coli SPase-I, and concomitant formation of an arm-swapped dimer.

1047-8477/$ - see front matter Ó 2013 Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jsb.2013.05.021

Please cite this article in press as: Young, P.G., et al. An arm-swapped dimer of the Streptococcus pyogenes pilin specific assembly factor SipA. J. Struct. Biol. (2013), http://dx.doi.org/10.1016/j.jsb.2013.05.021

61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86

YJSBI 6403

No. of Pages 6, Model 5G

12 June 2013 2

P.G. Young et al. / Journal of Structural Biology xxx (2013) xxx–xxx

87

2. Cloning, expression and purification of SipA

88

The sipA gene comprising the entire extracellular region of the protein (SipA36–173) and a truncated version lacking 9 membrane-proximal residues 36–44 (SipA45–173) were PCR-amplified from S. pyogenes strain 90/306S genomic DNA. For simplicity SipA36–173 and the truncated SipA45–173 will be referred to as SipAWT and SipAD9, respectively. SipAWT and SipAD9 were amplified using the gene specific primers SPY0127 F1 50 - AAA GGCGCC CAG TAT GTT TTT GGT GTT ATG ATT A -30 (SipAWT) or SPY0127 F2 50 - AAA GGCGCC AAC ACT AAT GAT ATG AGT CCT GCT TTA AG -30 (SipAD9) and SPY0127 R1 50 - AAA GAATTC TTA AAT TCC TCT CAC TCT TAA TAG AGT TGA G -30 (50 KasI and 30 EcoRI restriction nuclease recognition sites are shown in bold). The amplified fragments were cloned into the vector pProEXHTa (Invitrogen). For recombinant SipA protein expression, E. coli BL21 (kDE3) pRIL cells were transformed with SipA plasmids and grown in Luria–Bertani (LB) media supplemented with the required antibiotics at 310 K until OD600 reached 0.6. The cultures were induced with 0.2 mM IPTG at 301 K or 291 K for 16 h and the cells harvested by centrifugation. Cell pellets were resuspended buffer A (50 mM Tris.Cl pH 8.0, 500 mM NaCl, 2% (v/v) glycerol, 5 mM imidazole) containing Complete Protease Inhibitor Cocktail Mini Tablets EDTA-free (Roche), snap frozen, and stored at 253 K. The recombinant proteins were purified from frozen cells, which were thawed in buffer A with Complete Protease Inhibitor Cocktail Mini Tablets EDTA-free (Roche) and 2 lg/ml DNase I, and then lysed using a cell disruptor (Constant Cell Disruption Systems) at 18 kpsi. Insoluble matter was sedimented by centrifugation and the soluble phase was loaded onto a HiTrap Chelating 5 ml column (GE Healthcare) for purification by IMAC. Bound protein was washed with buffer B (buffer A + 20 mM imidazole) and eluted in a gradient with buffer C (50 mM Tris.Cl pH 8.0, 150 mM NaCl, 2% (v/v) glycerol, 500 mM imidazole). The His6 affinity tag was cleaved from SipAD9 recombinant protein with a 1:50 ratio of rTEV-His6 and concurrently dialyzed against buffer D (25 mM Tris.Cl pH 7.5, 150 mM NaCl, 0.05 mM EDTA) at 277 K for 16 h.

89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122

SipAD9 was separated from the rTEV-His6 protease and uncleaved protein by IMAC. The unbound protein containing SipAD9 was concentrated using a 3 kDa MWCO protein concentrator (VivaScience) and purified by size exclusion chromatography on a Superdex S75 10/300 column (GE healthcare) in crystallization buffer (10 mM Tris.Cl pH 8.0, 100 mM NaCl). SipAD9 eluted in a single peak that corresponds to a dimer of approximately 30 kDa and was 99% pure as indicated by SDS–PAGE. Dynamic light scattering (DLS) data confirmed the protein was monodisperse with a radius of gyration that equates to a molecular weight of 30 kDa, in agreement with the size exclusion chromatography. In contrast, the SipAWT construct produced aggregated protein that could not be cleaved from the His6-tag and predominately eluted in the void volume with size exclusion chromatography.

123

3. Crystallization

137

Vapour diffusion crystallization trials were carried out at 291 K using a Cartesian nanolitre dispensing robot (Genomic systems) and a locally compiled crystallization screen (Moreland et al., 2005). Initial SipAD9 crystals were grown in 0.1 ll format and subsequently optimised in a hanging-drop vapour diffusion format. The crystals used for X-ray data collection grew by mixing 1 ll protein solution (10 mg/ml in 10 mM Tris.Cl pH 8.0, 100 mM NaCl) with 1 ll precipitant (15% ethanol, 0.1 M Tris.Cl pH 8.0) at 291 K.

138

4. Data collection and structure determination

146

Crystals of SipAD9 were transferred to cryoprotectant (100 mM Tris.Cl pH 8, 50 mM NaCl, 15% ethanol, 20% ethylene glycol) prior to flash-freezing in liquid nitrogen. For phase determination experiments crystals were soaked in cryoprotectant supplemented with 500 mM NaI for 3 min before freezing. X-ray diffraction data were collected in-house (Micromax007HF, Rigaku; MAR345DTB, MAR Research) at 110 K. All datasets were integrated using XDS (Kabsch, 1993), reindexed using POINTLESS (Evans, 2006) and scaled using SCALA (Evans, 2006). The

147

Table 1 Data collection and refinement statistics.

  *

Data collection

Native

NaI

Wavelength (Å) Resolution range (Å)* Space group Unit cell axial lengths (Å) Angles (°) Total No. of observations* Unique reflections* Redundancy* Completeness (%)* Mean I/r(I)* Rmerge (%)*,  Anomalous completeness* Anomalous multiplicity* DelAnom correlation between half-sets Mid-slope of anom normal probability No. of heavy atoms FOM (phaser EP acentric/centric) Refinement Resolution range (Å) Rwork/Rfree (%) No. atoms per AU Average B-factors (Å2) RMS deviations Bond lengths Bond angles Ramachandran plot Residues in most favored regions, allowed, disallowed (%)

1.54179 19.31–1.67 (1.76–1.67) P41212 a = b = 53.28, c = 131.75 a = b = c = 90 808,555 (95,927) 22,923 (3178) 35.3 (30.2) 99.4 (96.9) 60.8 (13.3) 4.3 (26.2)

1.54179 26.3–1.86 (1.96–1.86) P41212 a = b = 53.79, c = 131.27 a = b = c = 90 402,042 (47,171) 16,684 (2049) 24.1 (23.0) 97.7 (85.3) 35.1 (6.5) 8.0 (57.9) 97.6 (84.2) 13.4 (12.5) 0.83 0.61 20 0.43/0.14

19.3–1.67 18.5/21.4 1061 20.2 0.010 1.35 99.1, 0.9, 0

Rmerge = Rhkl Ri | Ii(hkl)  hI (hkl)i | /Rhkl Ri Ii(hkl). Values in parentheses are for the outermost resolution shell.

Please cite this article in press as: Young, P.G., et al. An arm-swapped dimer of the Streptococcus pyogenes pilin specific assembly factor SipA. J. Struct. Biol. (2013), http://dx.doi.org/10.1016/j.jsb.2013.05.021

124 125 126 127 128 129 130 131 132 133 134 135 136

139 140 141 142 143 144 145

148 149 150 151 152 153 154 155

YJSBI 6403

No. of Pages 6, Model 5G

12 June 2013 P.G. Young et al. / Journal of Structural Biology xxx (2013) xxx–xxx 156 157 158 159 160 161 162 163 164 165 166 167 168 169

structure of SipAD9 was determined by single-wavelength anomalous diffraction (SAD) phasing from the NaI-soaked data, using SHELX (Sheldrick, 1998) and Phaser-EP (Mccoy et al., 2007) to identify and refine iodine sites. This was followed by density modification in Dm (Cowtan and Zhang, 1999), and auto-building using Arp-Warp (Perrakis et al., 1999) which built 80% of the model. This model was used for molecular replacement into the higher resolution native data, which was subsequently refined using iterative cycles of manual building in COOT (Emsley et al., 2010), and refinement with REFMAC (Murshudov et al., 2011). The quality of the SipAD9 model was inspected using the program PROCHECK (Laskowski et al., 1993). Data collection and refinement statistics are shown in Table 1. All figures were generated using PyMOL (DeLano, 2009).

170

5. Cloning, expression and purification of FctA and SrtC

171

205

The gene for FctA(21–328), encompassing the entire major pilin extracellular domain (including both the signal peptide and sortase motif sequences) was amplified by PCR from S. pyogenes strain 90/ 306S genomic DNA using the gene specific primers PYT9 FctA F1 50 AAA GGATCC ATG AGT CAA AAC GTG AAG GCG -30 and PYT9 FctA R1 50 - TTT GAATTC TTA TGG AGC AAG GGT CCC TAC AAC ACC AGT TG -30 . The amplified products were cloned into the BamHI and EcoRI restriction sites of the pGEX-3c expression vector and overexpressed in E. coli BL21 (kDE3) pRIL cells. Cells were grown in LB media supplemented with the required antibiotics at 310 K until OD600 reached 0.6. The cultures were induced with 0.2 mM IPTG at 291 K for 16 h and the cells harvested by centrifugation. Cells were resuspended in a buffer containing 50 mM Tris.Cl pH 8.0, 50 mM NaCl (buffer A) and lysed using a cell disruptor (Constant Systems). The lysate was clarified by centrifugation and loaded onto a Glutathione Sepharose 4B (GE Healthcare). The column was extensively washed with buffer A containing 500 mM NaCl and GST-FctA was eluted in buffer A with 15 mM GSH. The GST-tag was cleaved from FctA by incubating overnight in buffer A containing 400 lg of recombinant picornavirus 3C protease and 2 mM DTT at 277 K, with concurrent dialysis in buffer B (50 mM Tris.Cl pH 8, 150 mM NaCl) to remove GSH. The untagged FctA was separated from GST protein and un-cleaved fusion protein by passage through Glutathione Sepharose 4B. FctA was further purified by size exclusion chromatography using a Superdex S200 10/300 column (GE Healthcare) in buffer B. SrtC was PCR-amplified from S. pyogenes strain 90/306S genomic DNA using the gene specific primers SrtC F1-50 TTCCAAGGTCCG GAT TCT TAT CAT CTC TAT CAA 30 and SrtC R1-50 GAAAGCTGGGTG TTA TTC TTG AAT AGT ACC GAC 30 , and cloned into the Gateway vector pDEST-17. For recombinant SrtC protein expression, E. coli BL21 (kDE3) pRIL cells were transformed with the sequence-verified plasmid and protein expression was induced with IPTG (0.2 mM) at 310 K for 16 h. Recombinant SrtC was purified as described for SipA protein purification.

206

6. Pulldown assays

207

FctA (20 lg), SrtC (15 lg) and SipA (15 lg) were mixed to a total volume of 50 ll in 50 mM Tris.Cl pH 8.0 and 150 mM NaCl, with or without 5 mM b-mercaptoethanol, and incubated for 60 min at 310 K. A sample was taken as a control and the remaining volume passed through a His-SpinTrap column (GE Healthcare). The flowthrough was collected, and the beads washed three times with buffer containing 20 mM imidazole. Bound proteins were eluted with 500 mM imidazole and analyzed on 12% SDS–PAGE gels electrophoresis. Experiments were performed with either SipA or SrtC as the His-tagged target proteins.

172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204

208 209 210 211 212 213 214 215 216

3

7. Polymerisation assay

217

FctA (20 lg), SrtC (15 lg) and SipA (15 lg) in 50 mM Tris.Cl pH 8.0 and 150 mM NaCl were mixed with or without 5 mM bmercaptoethanol to a total volume of 50 ll and incubated for 20 h at 310 K. The reactions were analyzed on 12% SDS–PAGE gels electrophoresis, and examined for evidence of FctA polymerisation with silver-staining.

218

8. Results and discussion

224

SipA is a membrane-anchored extracellular protein that is predicted to contain a single transmembrane domain at the N-terminus of the protein. We produced two N-terminal deletion constructs to aid soluble expression. The first construct produced (SipAWT) has the first 35 N-terminal residues including the transmembrane anchor deleted. This construct, which encompasses the entire extracellular domain, produced aggregates and was unsuitable for crystallization experiments. To increase SipA solubility an additional construct was made. In the construct presented here, SipAD9, an additional nine mostly hydrophobic N-terminal residues were deleted. This construct closely resembles the product of type-I signal peptidase self-cleavage that occurs in many gram-positive bacteria (Rao et al., 2009; van Roosmalen et al., 2000; Zheng et al., 2002). The SipAD9 construct was expressed in E. coli, purified to homogeneity and crystallized using the vapor diffusion technique. The crystals belong to the primitive tetragonal space group P41212. The structure of SipA was determined by SAD phasing from NaIsoaked crystals and the resulting model was then used for molecular replacement into the higher resolution native data, giving a model that was refined at 1.67 Å resolution (Rwork = 18.3%, Rfree = 21.4%). Data collection and refinement statistics are shown in Table 1. There is one molecule in the asymmetric unit, with clearly interpretable electron density for all but the first nine Nterminal residues, 45–53, and the last three C-terminal residues of the SipAD9 construct. The structure reveals that SipAD9 is a largely b-strand protein whose core fold is typical of the SPase-I or S26A signal peptidase family, of which E. coli SPase-I (PDB: 1B12) (Paetzel et al., 1998) is the only structurally-characterized example. SipAD9 and E. coli SPase-I share a sequence identity of 23% and can be superimposed with a root-mean-square difference in Ca atom positions of 2.16 Å over 120 aligned residues (Fig. 1). There are two major differences evident. Firstly, the E. coli SPase-I non-catalytic domain is substantially larger than the corresponding domain of SipAD9, which is minimally decorated and similar in size to that predicted for many gram-positive signal peptidases, including Bacillus subtilis SipS and Staphylococcus aureus SpsB. The extra size of the E. coli SPase-I noncatalytic domain originates from a four-stranded b-sheet insertion that follows b-strand 8 of SPase-I (b-strand 6 of SipAD9) (Fig. 1). The function of this domain is largely unknown, although modelling studies in E. coli suggest that the non-catalytic domain in SPase-I is involved in the recognition of residues in the mature region of the pre-protein immediately after the Ala-X-Ala motif, designated +1 to +6 (Choo et al., 2008). The second major difference between SipA and E. coli SPase-I involves the core domain that contains the catalytic site of SPase-I. This domain contains all the conserved sequence motifs of the S26A signal peptidase family (Boxes B–E) (Paetzel et al., 2002) (Fig. 2). In SPase-I, a three-stranded anti-parallel b-sheet comprising b-strands 1, 2, and 17 (the C-terminal strand) is postulated to sit either on or in the membrane (Paetzel et al., 2002). Beta-strands 1 and 2 form a b-hairpin with an extended loop that positions the nucleophilic serine of the catalytic dyad correctly in the active site,

225

Please cite this article in press as: Young, P.G., et al. An arm-swapped dimer of the Streptococcus pyogenes pilin specific assembly factor SipA. J. Struct. Biol. (2013), http://dx.doi.org/10.1016/j.jsb.2013.05.021

219 220 221 222 223

226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278

YJSBI 6403

No. of Pages 6, Model 5G

12 June 2013 4

P.G. Young et al. / Journal of Structural Biology xxx (2013) xxx–xxx

Fig.1. Structures of SipAD9 and E. coli SPase-I. (A) Stereo-view of a structural alignment between the extracellular domains of SipA and SPase-I. The conserved catalytic domains of SipA and SPase-I are shown in blue and green, respectively. The ‘non-catalytic’ domains are shown in purple (SipA) and yellow (SPase-I). Regions of SPase-I nonhomologous with SipA are shown in grey to highlight the core similarity between the two proteins. SipA b-strands 2 and 10b are involved in intermolecular interactions. Shown in stick form are the SPase-I catalytic dyad residues Ser 90 and Lys 145 and the corresponding residues in SipA (Ser 50 and 54, and Lys 83). The structurally conserved residues, SipA Asp 148 and Arg 150, and SPase-I Asp 280 and Arg 282 are shown to highlight the conservation of the S26 protein fold in SipA. The asterisk shows the position of a b-hairpin extension, strands 3 and 4, in the SPase-I catalytic domain. Relevant b-strands are labeled black for SipA and red for SPase-I. Dashed lines represent regions not visible in the electron density. N = N-terminus, C = C-terminus. (B) Topology diagrams of SipA and E. coli SPase-I. Catalytic domains are shown in blue and green and the noncatalytic domains in purple and yellow for SipA and SPase-I, respectively. Regions of SPase-I non-homologous with SipA are shown in grey. SipA b-strands 2 and 10b are involved in intermolecular interactions. Dashed lines represent regions not visible in the electron density. The positions of key catalytic residues are shown in circles. N = Nterminus, C = C-terminus. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

279 280 281 282

and forms an integral part of the peptide binding cleft (Fig. 1). This cleft is composed predominantly of residues from b-strands 1, 2, 5, and 6 and contains two shallow hydrophobic pockets predicted to accommodate the P1 and P3 residues (Ala-X-Ala) of signal peptides

(Choo et al., 2008; Paetzel et al., 1998; Paetzel et al., 2000). In the SipAD9 structure there is no three-stranded anti-parallel b-sheet at the membrane surface of the molecule (Fig. 1). This results from the loss of b-strand 1, arising from the deletion of 9 N-terminal res-

Please cite this article in press as: Young, P.G., et al. An arm-swapped dimer of the Streptococcus pyogenes pilin specific assembly factor SipA. J. Struct. Biol. (2013), http://dx.doi.org/10.1016/j.jsb.2013.05.021

283 284 285 286

YJSBI 6403

No. of Pages 6, Model 5G

12 June 2013 P.G. Young et al. / Journal of Structural Biology xxx (2013) xxx–xxx

5

Fig. 2. Sequence alignment of the extracellular domains of SipA and E. coli SPase-I. Regions termed Box B, C, D, and E are in grey boxes. Residues forming b-strands 1, 2, and 17 in SPase-I and corresponding residues in SipA are shown in bold. SipA b-strand 1, truncated in this construct, is underlined (dashes) in italics. The arrow indicates the Nterminus of the SipAD9 construct. Key residues in SPase-I and the corresponding SipA residues are shown as a larger font in bold.

Fig.3. Ribbon diagram of the SipA dimer. Two molecules of SipA form a dimer (green and blue) with inter-molecular b-strand interactions between b-strands 2 and 10b. Dashed lines depict an additional nine N-terminal residues that have no interpretable electron density. Shown in stick form are residues Asp 148 and Arg 150 that form a structurally conserved salt bridge, and residues Ser 54 and Lys 83. The position of the second potential nucleophile (Ser 50), disordered in this structure, is depicted by a circle. N = N-terminus, C = C-terminus. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309

idues from this construct, and from the arm-swapped dimer that is formed in the crystal structure. Thus, although the construct does retain the residues corresponding to the extended loop of b-hairpin 1 and b-strand 2, this N-terminal section diverges from the core fold (Fig. 1) to engage in intermolecular parallel b-strand interactions with an adjacent molecule in the crystal (Fig. 3). As a consequence, SipAD9 has no peptide binding cleft equivalent to that in SPase-I. In the intermolecular interaction, b-strand two pairs with the C-terminal b-strand 10 from the adjacent molecule, which has a sharp bend in it that enables its C-terminal half (b-strand 10b) to project out as an extended arm (Fig. 3). This inter-molecular arm exchange generates a dimer in the crystal lattice, with 26 hydrogen bonds and a total of 30% of solvent accessible surface buried in the dimer interface. SipAD9 also forms a dimer in solution as judged by both dynamic light scattering and size exclusion experiments, although we cannot conclude that the crystallographic and solution dimers are the same. Whether this arm exchange is a feature of SipAWT or an artefact of the truncation mutation is still unclear. We are continuing to work on the expression of the SipAWT construct that retains b-strand 1. This construct forms large aggregates. It is possible that these aggregates are caused by similar domain or strand swapping involving b-strand 1 and the additional nine N-terminal

residues unaccounted for in the electron density of our SipAD9 structure, thus potentially forming an array or multimeric structure at the membrane surface. Alternatively, the deletion of bstrand 1 could have had a profound effect on the folding of the N-terminal region of SipA, resulting in the arm-swapped dimer seen in this SipAD9 construct. The homology SipA shares with type-I signal peptidases suggests it has some role in peptide recognition. However, SipA appears to have atypical sequences in the boxes B and D that carry the catalytic serine and lysine residues in signal peptidases (Zahner and Scott, 2008). In SipA, boxes B and D contain Ser 50 and Lys 83 respectively, but these do not match the catalytic residues of SPase-I when the sequences are aligned according to the other key conserved residues in these motifs (Fig. 2). There are two potential serine nucleophile residues in SipA, Ser 50 and 54. Serine 50 sits within box B and Ser 54 is immediately adjacent to box B. Serine 54 is a less likely candidate for the catalytic dyad due to its position outside box B, but is included in figures as it is the first N-terminal residue with interpretable electron density in our structure. In SipAD9 neither Ser 50 nor Ser 54 is in a position to act as a nucleophile for potential peptidase activity. Recombinant SipA also fails to show any detectable peptidase activity against pre-Cpa or synthetic peptides encompassing the potential Cpa signal peptide cleavage site (Nakata et al., 2009). Our structure of SipAD9 confirms that Lys 83 is solvent exposed and not buried within the enzyme, as is the case for Lys 145 in E. coli SPase-I. The hydrophobic environment surrounding Lys 145 in SPase-I is thought to facilitate the lowering of its pKa so that it exists in the deprotonated state necessary for both the acylation and deacylation steps of catalysis (Paetzel et al., 1998). Calculations of the pKa values of ionizable groups in SipAD9 predict that Lys 83 is fully protonated (Gordon et al., 2005; Myers et al., 2006). The side chain amino-group of Lys 83 is hydrogen-bonded to the main chain oxygen of Leu 147, the side chain oxygen (OE2) of Glu 145, and an ethanol molecule. These considerations suggest that Lys 83 would be unable to function as a general base for any potential nucleophile. Taken together with mutational studies in serotype M3, which showed the serine residues homologous to Ser 50 and 54 were both non-essential for pilus polymerisation, this suggests that SipA does not have peptidase activity. Indeed, our SipAD9 structure does not have any recognizable peptide binding groove. It is perhaps more likely that SipA may form a scaffold at the cell surface on which various components of pilus polymerisation are positioned. We have, however, been unable to detect any interaction between SipAD9 and the pilus shaft protein (FctA). Our pull-

Please cite this article in press as: Young, P.G., et al. An arm-swapped dimer of the Streptococcus pyogenes pilin specific assembly factor SipA. J. Struct. Biol. (2013), http://dx.doi.org/10.1016/j.jsb.2013.05.021

310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355

YJSBI 6403

No. of Pages 6, Model 5G

12 June 2013 6

P.G. Young et al. / Journal of Structural Biology xxx (2013) xxx–xxx

368

down assays with SipAD9 failed to show any interaction in vitro with either the transpeptidase (SrtC) or pre-FctA, which retains the entire extracellular region of the protein including both the signal peptide and sortase motif regions. Furthermore incubation of SipAD9 with both SrtC and pre-FctA in vitro does not produce polymers of FctA (data not shown). This suggests a number of possibilities: (i) that SipA may need to be in a multimeric form to interact with other pilin proteins; (ii) that our construct has an alternative topology that results from the deletion of b-strand 1, and that SipA may need to form a structure more like E. coli SPase-I to be able to function as a pilus accessory factor; or (iii) that SipA needs to be anchored in the membrane, along with the other pilin proteins to function correctly.

369

9. Conclusion

370

387

While the structure of our SipAD9 construct lacks a peptide binding cleft analogous to E. coli SPase-I, we have shown that SipA and SPase-I are structurally homologous. This implies some aspect of shared function, and certainly a shared ancestry. Our structure suggests that the solvent-exposed Lys 83 in SipA would be fully protonated, and together with previous mutational analysis of the potential serine nucleophiles (Zahner and Scott, 2008), is consistent with SipA lacking the peptidase activity of SPases. The most likely function of SipA is an ability to bind to signal peptides, or other peptide sequences in the pilus proteins. This implies that the different structures formed by the N- and C-terminal b-strands in our structure may be an artifact of either the loss of b-strand 1 or the crystallization conditions, or both. This being the case, we would expect the SipAWT construct to more closely match the SPase-I structure and contain a peptide binding cleft. It is anticipated that the structure of the SipAWT construct will help establish whether SipA has a role as a structural scaffold or has an as yet undefined enzymatic function in pilus formation.

388

Accession number

389 391

The coordinates and structure factors of SipAD9 have been deposited in the Protein Data Bank under the accession code of 4K8W.

392

Acknowledgments

393 394

This work was supported by the Health Research Council of New Zealand.

395

References

396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412

Abbot, E.L., Smith, W.D., Siou, G.P., Chiriboga, C., Smith, R.J., Wilson, J.A., Hirst, B.H., Kehoe, M.A., 2007. Pili mediate specific adhesion of Streptococcus pyogenes to human tonsil and skin. Cell Microbiol. 9, 1822–1833. Choo, K.H., Tong, J.C., Ranganathan, S., 2008. Modeling Escherichia coli signal peptidase complex with bound substrate: determinants in the mature peptide influencing signal peptide cleavage. BMC Bioinf. 9 (Suppl. 1), S15. Cowtan, K.D., Zhang, K.Y.J., 1999. Density modification for macromolecular phase improvement. Prog. Biophys. Mol. Bio. 72, 245–270. Cunningham, M.W., 2000. Pathogenesis of group A streptococcal infections. Clin. Microbiol. Rev. 13, 470–511. Dalbey, R.E., Wickner, W., 1985. Leader peptidase catalyzes the release of exported proteins from the outer surface of the Escherichia coli plasma membrane. J. Biol. Chem. 260, 15925–15931. Dalbey, R.E., Lively, M.O., Bron, S., van Dijl, J.M., 1997. The chemistry and enzymology of the type I signal peptidases. Protein Sci. 6, 1129–1138. DeLano, W.L., 2009. PyMOL molecular viewer: updates and refinements. Abstr. Pap. Am. Chem. Soc. 238.

356 357 358 359 360 361 362 363 364 365 366 367

371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386

390

Q2

Emsley, P., Lohkamp, B., Scott, W.G., Cowtan, K., 2010. Features and development of coot. Acta Crystallogr. D 66, 486–501. Evans, P., 2006. Scaling and assessment of data quality. Acta Crystallogr. D Biol. Crystallogr. 62, 72–82. Gordon, J.C., Myers, J.B., Folta, T., Shoja, V., Heath, L.S., Onufriev, A., 2005. H++: a server for estimating pK(a)s and adding missing hydrogens to macromolecules. Nucleic Acids Res. 33, W368–W371. Hendrickx, A.P., Budzik, J.M., Oh, S.Y., Schneewind, O., 2011. Architects at the bacterial surface – sortases and the assembly of pili with isopeptide bonds. Nat. Rev. Microbiol. 9, 166–176. Inada, T., Court, D.L., Ito, K., Nakamura, Y., 1989. Conditionally lethal amber mutations in the leader peptidase gene of Escherichia coli. J. Bacteriol. 171, 585– 587. Kabsch, W., 1993. Automatic processing of rotation diffraction data from crystals of initially unknown symmetry and cell constants. J. Appl. Crystallogr. 26, 795– 800. Kang, H.J., Coulibaly, F., Clow, F., Proft, T., Baker, E.N., 2007. Stabilizing isopeptide bonds revealed in gram-positive bacterial pilus structure. Science 318, 1625– 1628. Laskowski, R.A., Macarthur, M.W., Moss, D.S., Thornton, J.M., 1993. Procheck – a program to check the stereochemical quality of protein structures. J. Appl. Crystallogr. 26, 283–291. Linke, C., Young, P.G., Kang, H.J., Bunker, R.D., Middleditch, M.J., Caradoc-Davies, T.T., Proft, T., Baker, E.N., 2010. Crystal structure of the minor pilin FctB reveals determinants of group A streptococcal pilus anchoring. J. Biol. Chem. 285, 20381–20389. Manetti, A.G., Zingaretti, C., Falugi, F., Capo, S., Bombaci, M., Bagnoli, F., Gambellini, G., Bensi, G., Mora, M., Edwards, A.M., Musser, J.M., Graviss, E.A., Telford, J.L., Grandi, G., Margarit, I., 2007. Streptococcus pyogenes pili promote pharyngeal cell adhesion and biofilm formation. Mol. Microbiol. 64, 968–983. Mccoy, A.J., Grosse-Kunstleve, R.W., Adams, P.D., Winn, M.D., Storoni, L.C., Read, R.J., 2007. Phaser crystallographic software. J. Appl. Crystallogr. 40, 658–674. Mora, M., Bensi, G., Capo, S., Falugi, F., Zingaretti, C., Manetti, A.G., Maggi, T., Taddei, A.R., Grandi, G., Telford, J.L., 2005. Group A Streptococcus produce pilus-like structures containing protective antigens and Lancefield T antigens. Proc. Natl. Acad. Sci. USA 102, 15641–15646. Moreland, N., Ashton, R., Baker, H.M., Ivanovic, I., Patterson, S., Arcus, V.L., Baker, E.N., Lott, J.S., 2005. A flexible and economical medium-throughput strategy for protein production and crystallization. Acta Crystallogr. D Biol. Crystallogr. 61, 1378–1385. Murshudov, G.N., Skubak, P., Lebedev, A.A., Pannu, N.S., Steiner, R.A., Nicholls, R.A., Winn, M.D., Long, F., Vagin, A.A., 2011. REFMAC5 for the refinement of macromolecular crystal structures. Acta Crystallogr. D 67, 355–367. Myers, J., Grothaus, G., Narayanan, S., Onufriev, A., 2006. A simple clustering algorithm can be accurate enough for use in calculations of pKs in macromolecules. Proteins 63, 928–938. Nakata, M., Koller, T., Moritz, K., Ribardo, D., Jonas, L., McIver, K.S., Sumitomo, T., Terao, Y., Kawabata, S., Podbielski, A., Kreikemeyer, B., 2009. Mode of expression and functional characterization of FCT-3 pilus region-encoded proteins in Streptococcus pyogenes serotype M49. Infect. Immun. 77, 32–44. Paetzel, M., Dalbey, R.E., Strynadka, N.C., 1998. Crystal structure of a bacterial signal peptidase in complex with a beta-lactam inhibitor. Nature 396, 186–190. Paetzel, M., Dalbey, R.E., Strynadka, N.C.J., 2000. The structure and mechanism of bacterial type I signal peptidases: A novel antibiotic target. Pharmacol. Therapeut. 87, 27–49. Paetzel, M., Dalbey, R.E., Strynadka, N.C., 2002. Crystal structure of a bacterial signal peptidase apoenzyme: implications for signal peptide binding and the Ser-Lys dyad mechanism. J. Biol. Chem. 277, 9512–9519. Perrakis, A., Morris, R., Lamzin, V.S., 1999. Automated protein model building combined with iterative structure refinement. Nat. Struct. Biol. 6, 458–463. Quigley, B.R., Zahner, D., Hatkoff, M., Thanassi, D.G., Scott, J.R., 2009. Linkage of T3 and Cpa pilins in the Streptococcus pyogenes M3 pilus. Mol. Microbiol. 72, 1379– 1394. Rao, S.C.V., Bockstael, K., Nath, S., Engelborghs, Y., Anne, J., Geukens, N., 2009. Enzymatic investigation of the Staphylococcus aureus type I signal peptidase SpsB – implications for the search for novel antibiotics. FEBS J. 276, 3222–3234. Sheldrick, G.M., 1998. SHELX applications to macromolecules. NATO Adv. Sci. ICMat. 507, 401–411. Smith, W.D., Pointon, J.A., Abbot, E., Kang, H.J., Baker, E.N., Hirst, B.H., Wilson, J.A., Banfield, M.J., Kehoe, M.A., 2010. Roles of minor pilin subunits Spy0125 and Spy0130 in the serotype M1 Streptococcus pyogenes strain SF370. J. Bacteriol. 192, 4651–4659. van Roosmalen, M.L., Jongbloed, J.D.H., Kuipers, A., Venema, G., Bron, S., van Dijl, J.M., 2000. A truncated soluble Bacillus signal peptidase produced in Escherichia coli is subject to self-cleavage at its active site. J. Bacteriol. 182, 5765–5770. Zahner, D., Scott, J.R., 2008. SipA is required for pilus formation in Streptococcus pyogenes serotype M3. J. Bacteriol. 190, 527–535. Zheng, F., Angleton, E.L., Lu, J., Peng, S.B., 2002. In vitro and in vivo self-cleavage of Streptococcus pneumoniae signal peptidase I. Eur. J. Biochem. 269, 3969–3977.

Please cite this article in press as: Young, P.G., et al. An arm-swapped dimer of the Streptococcus pyogenes pilin specific assembly factor SipA. J. Struct. Biol. (2013), http://dx.doi.org/10.1016/j.jsb.2013.05.021

413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492