- Email: [email protected]
Crystal Structure of the NanB Sialidase from Streptococcus pneumoniae
J. Mol. Biol. (2008) 384, 436–449
doi:10.1016/j.jmb.2008.09.032
Available online at www.sciencedirect.com
Crystal Structure of the NanB Sialidase from Streptococcus pneumoniae Guogang Xu 1 , Jane A. Potter 1 , Rupert J. M. Russell 1 , Marco R. Oggioni 2 , Peter W. Andrew 3 and Garry L. Taylor 1 ⁎ 1
Centre for Biomolecular Sciences, University of St Andrews, St Andrews, Fife, KY16 9ST, UK 2
Lab Microbiologia Molecolare e Biotecnologia, Dipartimento di Biologia Molecolare, Universita' di Siena, Policlinico Le Scotte, 53100 Siena, Italy 3
Department of Infection, Immunity and Inflammation, University of Leicester, Leicester LE1 9HN, UK Received 15 July 2008; received in revised form 7 September 2008; accepted 12 September 2008 Available online 21 September 2008
The Streptococcus pneumoniae genomes encode up to three sialidases (or neuraminidases), NanA, NanB and NanC, which are believed to be involved in removing sialic acid from host cell surface glycans, thereby promoting colonization of the upper respiratory tract. Here, we present the crystal structure of NanB to 1.7 Å resolution derived from a crystal grown in the presence of the buffer Ches (2-N-cyclohexylaminoethanesulfonic acid). Serendipitously, Ches was found bound to NanB at the enzyme active site, and was found to inhibit NanB with a Ki of ∼0.5 mM. In addition, we present the structure to 2.4 Å resolution of NanB in complex with the transition-state analogue Neu5Ac2en (2-deoxy-2,3-dehydro-N-acetyl neuraminic acid), which inhibits NanB with a Ki of ∼ 0.3 mM. The sulphonic acid group of Ches and carboxylic acid group of Neu5Ac2en interact with the arginine triad of the active site. The cyclohexyl group of Ches binds in the hydrophobic pocket of NanB occupied by the acetamidomethyl group of Neu5Ac2en. The topology around the NanB active site suggests that the enzyme would have a preference for α2,3-linked sialoglycoconjugates, which is confirmed by a kinetic analysis of substrate binding. NMR studies also confirm this preference and show that, like the leech sialidase, NanB acts as an intramolecular trans-sialidase releasing Neu2,7-anhydro5Ac. All three pneumoccocal sialidases possess a carbohydrate-binding domain that is predicted to bind sialic acid. These studies provide support for a possible differential role for NanB compared to NanA in pneumococcal virulence. © 2008 Elsevier Ltd. All rights reserved.
Edited by G. Schulz
Keywords: sialidase; neuraminidase; drug design; substrate specificity; crystal structure
Introduction Streptococcus pneumoniae is a major human pathogen responsible for respiratory tract infections, septicemia and meningitis, and continues to produce numerous cases of disease with a relatively high mortality rate. In children, S. penumoniae is the most frequent cause of otitis media,1 and the nasopharynx serves as a reservoir for the bacterium to enter the middle ear via the Eustachian tube, particularly following a respiratory viral infection. Several virulence factors may contribute to colonization and early infection processes.2 Sialidases are one key *Corresponding author. E-mail address: [email protected]. Abbreviations used: BNR, bacterial neuraminidase repeat; CBM, carbohydrate-binding module.
virulence factor, as they remove sialic acid from host cell surface glycans, probably unmasking certain receptors to facilitate bacterial adherence and colonization.3 To date, all S. pneumoniae clinical isolates investigated present prominent sialidase activity. Three distinct sialidases, NanA, NanB and NanC, are encoded in the S. pneumoniae genomes. A screening study of sialidase genes in 342 clinical pneumococcal isolates identified nanA, nanB and nanC to be present in 100%, 96% and 51% of these strains, respectively.4 NanA has been shown to have an important role in host-pneumoccocal interactions in the upper respiratory tract.5,6 It is also known that there is a lethal synergism between the influenza virus and S. pneumoniae, probably as a consequence of the viral sialidase stripping sialic acid from host glycoproteins.7 Experiments with mouse models using isogenic pneumococcal mutants show that both NanA and NanB are essential to S. pneumoniae
0022-2836/$ - see front matter © 2008 Elsevier Ltd. All rights reserved.
Structure of S. Pneumoniae NanB
infection of the respiratory tract and sepsis.8 Finally, the S. pneumoniae sialidases show promise as antigens for vaccine development targeted at pneumoccocal disease.9 There are crystal structures of sialidases from bacteria,10–12 viruses,13–15 trypanosomes,16,17 leech,18 and man.19 The catalytic domains of sialidases have a six-bladed β-propeller topology, with conservation of key catalytic amino acids.20 The non-viral sialidases also have conserved bacterial neuraminidase repeats (BNRs) between one and five times along the sequence. Many sialidases have additional domains described as lectins or carbohydrate-binding modules (CBMs). For example, Vibrio cholerae sialidase has two lectin domains flanking the catalytic domain, one of which binds sialic acid;21 Micromonospora viridifaciens sialidase has a galactose-binding domain C-terminal to the catalytic domain; 22 the leech sialidase has a lectin-like domain N-terminal to the catalytic domain;18 and the trypanosome (trans-) sialidases have a lectin-like domain C-terminal to the catalytic domain.16,17 It has been shown that the presence of these CBMs increases the catalytic efficiency of the sialidases, particularly towards polysaccharide substrates.23 The three pneumoccocal sialidases NanA, NanB and NanC have a molecular mass of 115 kDa, 78 kDa and 82 kDa, respectively. Sequence analysis shows that all three possess a signal sequence, followed by a lectin domain, or CBM, before the catalytic domain, the latter domain containing four BNRs (Fig. 1). In contrast to NanB and NanC, NanA has an additional C-terminal domain containing an LPETG motif that tethers the enzyme to the bacterial surface.24 NanB and NanC share 50% sequence identity, whereas they both share only 25% identity with NanA. NanB is reported to have a pH optimum of 4.5 in contrast to pH 6.5–7.0 for NanA.25 This, together with the inability of the two enzymes to compensate for one another in the gene knockout experiments, suggests that they have separately defined roles in pathogenesis.8 Here, we present the crystal structure of NanB at a resolution of 2.3 Å, and complexes of NanB with Ches and Neu5Ac2en to resolutions of 1.7 Å and 2.4 Å, respectively. The observation of a Ches molecule in the active site led us to measure its ability to inhibit the enzyme, which it does with a Ki of ∼ 0.5 mM comparable to Neu5Ac2en that inhibits with a Ki of ∼ 0.3 mM. The presence of a tryptophan
437 close to the position of the substrate aglycon in NanB is seen also in the leech trans-sialidase and the Trypanosoma cruzi trans-sialidase, which both show a strict preference for α2,3-linked sialosylglycosides. NMR studies show that, like the leech enzyme, NanB is an intramolecular trans-sialidase, releasing Neu2,7-anhydro5Ac, and kinetic studies show that although NanB has a preference for α2,3-linked substrates, it can cleave α2,6 and α2,8 linkages. By homology with the carbohydrate-binding domain of the Clostridium perfringens NanJ sialidase, the similar domain of NanB is likely to bind sialic acid. Finally, Ches provides a framework for the development of a novel class of sialidase inhibitors.
Results Overall structure The crystal structure of NanB spans residues 39 to 696, and is missing the signal peptide (residues 1–38) and the final residue, which is not visible in the electron density maps. NanB is a monomer with three domains: a concanavalin-like lectin domain (residues 39–228) and a canonical β-propeller catalytic domain (residues 229–345 and 459–696) into which is inserted the third domain (residues 346–458) that is formed predominantly from β-strands (Fig. 2a and b). The structure was solved by molecular replacement using the leech trans-sialidase as the search model (Protein Databank code 1sli), which shares 43% sequence identity with NanB, and the two proteins have an rmsd of 1.49 Å over 630 Cα atoms. The six-bladed catalytic domain is formed from a sequential repeat of four-stranded anti-parallel β-sheets and, like the leech enzyme, there is an extended loop between the first and second sheet (residues 297–323) that associates with, and forms part of, the inserted third domain that is itself between the second and third strands of the second sheet. This inserted domain is found only in the three pneumococcal sialidases, NanI from C. perfringens (2bf6) and the leech transsialidase, with no other structurally homologous domain being found in the protein structure databank. The role of this inserted domain is unknown. The electrostatic surface potential of the entire protein shows a distinct asymmetry, typical of secreted bacterial sialidases, with the surface remote from the active site carrying a negative charge (coloured
Fig. 1. A diagram of the three S. pneumoniae sialidases. CBM40 refers to type 40 carbohydrate-binding module (CBM) that recognizes sialic acid. The locations of BNRs are shown that conform to the sequence motif Ser/Thr-x-Asp-(x)-Gly-x-Thr(Trp/Phe), where x is any amino acid. The blue regions, containing the BNRs, comprise the β-propeller catalytic domain.
438
Structure of S. Pneumoniae NanB
Fig. 2. The NanB structure. (a and b) The catalytic β-propeller domain is shown in blue with Neu5Ac2en in the active site, the lectin domain is shown in red and the inserted domain is shown in green. b, the view represents a 90° rotation of the view in a around a horizontal axis. (c and d) The same views as in a and b, respectively, and show a surface representation coloured according to electrostatic potential from −10 kT/e to +10 kT/e, calculated using APBS.53 This and other figures were created with PyMOL [http://pymol.sourceforge.net/].
red in Fig. 2c and d) that may serve to orient the protein towards its negatively charged glycoconjugate substrates.
been retained throughout their evolution. These include three arginines (Arg245, Arg557, Arg619: NanB numbering) that interact with the carboxylate
Lectin or carbohydrate-binding module (CBM) The lectin domain of NanB is structurally homologous to the recently reported family 40 CMB (CBM40) of the C. perfingens NanJ sialidase, which was found to recognize a sialic acid moiety.26 The two domains share 26% sequence identity and superimpose with an rmsd of 1.64 Å for 168 Cα atoms. Several key residues of the sialic acid-binding site are conserved between the two CBMs, suggesting that the CBM of NanB binds to sialic acid containing glycoconjugates (Fig. 3); these include two arginines that interact with the carboxyl group of sialic acid (Arg117 and Arg193), a glutamic acid that interacts with the O4 hydroxyl (Glu115) and a hydrophobic pocket accommodating the methyl of the acetamido group (Leu84, Phe105, Leu126 and Tyr199). Neu5Ac2en binding A complex of NanB with Neu5Ac2en was obtained at 2.4 Å resolution, and the interactions made by the ligand with the active site are shown in Fig. 4a. Although there are a number of significant differences between the active sites of viral, bacterial and eukaryotic sialidases, a number of key features have
Fig. 3. Putative sialic acid-binding site in the NanB lectin domain. The lectin domain of NanB (salmon) is superimposed onto the CBM40 module from Clostridium perfringens NanJ (cyan; PDB code 2v73). Residues are labelled in the order NanB/NanJ, and hydrogen bond interactions in NanJ are shown as dotted lines.
Structure of S. Pneumoniae NanB
group of Neu5Ac2en. In NanB one of the arginines, Arg557, is 3.6 Å away from the nearest carboxylate oxygen of the ligand, whereas the other two are both within hydrogen bonding distance. The position of the first arginine (Arg245) is stabilised by a conserved glutamic acid (Glu669). The nucleophilic tyrosine (Tyr653) and an accompanying glutamic acid (Glu541) hydrogen bond with each other and sit beneath the C1–C2 bond of the ligand, the hydroxyl of Tyr653 being 3.1 Å from C1 and C2 of the ligand. In common with other bacterial sialidases, the O4 hydroxyl of Neu5Ac2en interacts with an arginine (Arg264) and an aspartic acid (Asp327). A conserved feature of sialidase active sites is the acid/base catalyst, Asp270, situated within a loop above the active site, that here interacts with O7 of the ligand and with a water molecule (W3) that, in turn, interacts with Asn352. This latter residue interacts with the glycerol hydroxyls O7 and O9 of Neu5Ac2en. The presence of Asn352 is one key difference between the active sites of NanB and the leech transsialidase, where the latter has serine at this position, which makes no interaction with Neu5Ac2en. The O8 hydroxyl of Neu5Ac2en does not interact with NanB. All sialidase active sites have a hydrophobic pocket to accommodate the N-acetyl group of the substrate, but the exact residues that form this pocket are generally not conserved. In the NanB structure, this pocket is composed of Ile326, Met346 and Tyr489. In the leech trans-sialidase, the isoleucine (Ile374) and methionine (Met394) are conserved, but the tyrosine is replaced by valine (Val538) with a tryptophan (Trp536) forming part of the pocket. Finally, water molecules W1 and W2 are involved in a network of key amino acids: Asp327, Asn352, Tyr509 and Glu541. Our kinetic data show that Neu5Ac2en inhibits NanB with a Ki of 0.3 mM, and NanA with a Ki of 2 μM. Ches binding Two Ches molecules are bound to NanB in the 1.7 Å resolution structure derived from a crystal grown in the presence of Ches buffer. Both are well ordered and interact with the catalytic domain, one in the active site and the other on the opposite side where it is involved in multiple interactions with the enzyme, either directly or via water molecules and a glycerol molecule. The presence of this latter Ches molecule may well contribute to the increased diffraction quality of crystals grown in the presence of Ches. The interactions made by the Ches molecule in the active site are shown in Fig. 4b. Two oxygens of the sulphonic acid group of Ches interact with all three arginines of the arginine triad. The amino group of Ches interacts with the acid/base Asp270, and via a water molecule (W4) with Asp327. The cyclohexyl ring of Ches occupies the hydrophobic pocket described above for Neu5Ac2en. The key water molecule, W1, that networks Asp327, Glu541 and Tyr509 is conserved in both the Neu5Ac2en and Ches complexes. Superposition of the native, Ne5Ac2en and Ches complexes reveals a very rigid active
439 site (Fig. 4c), with only the side chain of Asn352 moving significantly in order to participate in hydrogen bonding with the glycerol moiety of Neu5Ac2en. Our kinetic data show that Ches inhibits NanB with a Ki of 0.50 mM, and NanA with a Ki of 0.65 mM. Substrate specificity and product formation The leech trans-sialidase has a tryptophan (Trp734) packing against a tyrosine (Tyr643) close to the carboxylate group of Neu5Ac2en, which is suggested to give the leech trans-sialidase strict specificity towards α2,3-linked sialoglycosides. A similar feature is seen in the T. cruzi trans-sialidase, which also has specificity for α2,3-linked sialoglycosides. NanB also has a tryptophan (Trp674) packing against a tyrosine (Tyr589) in the same position (Fig. 5a). Measurement of the release of sialic acid from α2,3-, α2,6- and α2,8-linked substrates shows that NanB has a five- and tenfold preference for α2,3 over α2,6 and α2,8 respectively (Fig. 5a), whereas NanA shows little discrimination between the different linkages (Fig. 5b) and is N 10 times more active against each than NanB. NMR studies carried out at pH 7 confirm the ability of NanB to cleave sialic acid from α2,3 and α2,6 sialyllactose substrates, and show that the product Neu2,7-anhydro5Ac is formed, thus classifying NanB as an intramolecular trans-sialidase (Fig. 6 and Supplementary Data). In the case of α2,3-sialyllactose, the ratio of Neu2,7-anhydro5Ac: Neu5Ac is 9:1, 6:1 and 3:1 after 0.5 h, 24 h and 48 h, respectively. Although the ratio of the two products changes, the amount of Neu2,7-anhydro5Ac remains constant, at least for 48 h. Further experiments are in progress to investigate if both products are formed or whether Neu2,7-anhydro5Ac is first formed and then converted to Neu5Ac. In the case of α-2,6-sialyllactose, the reaction is much slower, with only 30% of the substrate cleaved after 48 h (Supplementary Data). The ratio of Neu2,7-anhydro5Ac:Neu5Ac remains at 1:2 after 24 h and 48 h. These experiments were repeated at pH 5 and gave similar results (Supplementary Data). NMR was used to explore the possibility that NanB might act as an intermolecular trans-sialidase similar to the T. cruzi enzyme, but no such activity was observed (Supplementary Data).
Discussion The crystal structure of NanB was determined using the structure of the leech trans-sialidase as a molecular replacement model, with which it shares 43% sequence identity. The leech trans-sialidase is unusual amongst sialidases in that it is an intramolecular trans-sialidase that cleaves only the Neu5Acα2,3Gal linkage of sialoglycoconjugates, releasing Neu2,7-anhydro5Ac,27,28 and is not inhibited by Neu5Ac2en.28 The selectivity for Neu5Acα2,3Gal in the leech trans-sialidase is dictated by the presence of a tryptophan close to where the aglycon would be
Structure of S. Pneumoniae NanB
440
Fig. 4 (legend on next page)
Structure of S. Pneumoniae NanB
441
Fig. 5. Substrate selectivity of NanB ((a) where ⁎⁎ indicates p b 0.01) and NanA (b).
positioned,29 and the intramolecular formation of Neu2,7-anhydro5Ac is proposed to be aided by the presence of a threonine (Thr593) that may promote an axial positioning of the glycerol group of the substrate.29,30 The NMR results show that NanB acts also as an intramolecular trans-sialidase, releasing Neu2,7-anhydro5Ac from both α2,3 and α2,6 sialyllactose. The critical threonine suggested to help orient the glycerol group is conserved in NanB (Thr539), and the unusual observation, for a sialidase, of a hydrogen bond interaction between the O7 hydroxyl and the acid-base Asp270 in NanB lends further support to the mechanism proposed earlier.30 Neu5Ac2en inhibits NanB with a Ki of 0.3 mM, which is similar to that found for Salmonella typhimurium sialidase (0.38 mM)31 and the Trypanasoma rangeli sialidase (0.14 mM).32 The complex of NanB with Neu5Ac2en clearly shows the inhibitor positioned in the active site (Fig. 4a), in contrast to the Neu5Ac2en leech trans-sialidase complex, which
is reported to not be inhibited by Neu5Ac2en, and which showed very weak electron density for the ligand.29 In NanB, Asn352 hydrogen bonds to both O7 and O9 hydroxyls of the glycerol group of Neu5Ac2en, whereas in the leech trans-sialidase the place of Asn352 in NanB is taken by a serine (Ser400) that does not form hydrogen bonds with the glycerol hydroxyls of Neu5Ac2en,29 and this could certainly lead to a reduced binding affinity for the leech enzyme. Despite the conservation in NanB of the tryptophan that confers strict α2,3 specificity to the leech trans-sialidase, NanB is able to remove sialic acids from α2,6 and α2,8 glycoconjugates, albeit five- to tenfold less efficiently compared to α2,3-linked substrates. A similarly positioned tryptophan is present in the T. cruzi trans-sialidase that also has strict specificity for sialic acid linked α2,3 to a galactose, but NanB has none of the structural features that confer trans-sialidase activity on the T. cruzi enzyme,33 and our NMR results confirm that
Fig. 4. Stereo drawings of the NanB active site. (a), The Neu5Ac2en complex. (b), The Ches complex. In both of these figures, the σa-weighted Fo–Fc difference electron density map is contoured at 3σ. c, Superposition of the native (green), Neu5Ac2en (cyan) and Ches (magenta) structures.
442
Structure of S. Pneumoniae NanB
Fig. 6. NMR spectra showing the time course of product release by NanB with α-2,3-sialyllactose (3SL) as the substrate.
Structure of S. Pneumoniae NanB
NanB does not have intermolecular trans-sialidase activity. It can be concluded that NanB acts as an intramolecular trans-sialidase with a distinct preference for α2,3-linked sialic acids. The amount of Neu2,7-anhydro5Ac produced by NanB remains constant, at least over the 48 h of the NMR experiment, whereas the amount of β-Neu5Ac increases slowly. It has been reported that the leech transsialidase can slowly convert Neu2,7-anhydro5Ac to Neu5Ac,34 and this may be the same for NanB, although further experiments are required to confirm this. Within the sequence encompassed by the structure of NanB reported here, NanB shares 23% sequence identity with NanA, and 50% identity with NanC (Fig. 7). All three sialidases have a lectin domain where the residues predicted to compose the hydrophobic pocket accommodating the acetamido methyl group of sialic acid are conserved as hydrophobic amino acids. In addition, the arginine predicted to interact with the carboxyl group of sialic acid in the CBM (Arg193 in NanB) is conserved across all three sialidases. NanC also conserves a glutamic acid and arginine that may interact with the O4 and carboxyl groups of sialic acid (Glu115 and Arg117 in NanB). However, NanA does not conserve these two amino acids, where they are arginine and serine, respectively. It is probable that the CBM40 domains of NanA and NanC will, like NanB, recognize sialic acid, but the details of binding, particularly for NanA, will require further structural studies. The concave carbohydrate-binding pocket of the CBM is on the same side of NanB as the active site, suggesting that the presence of such domains may help target the enzymes to appropriate substrates, and thereby increase their catalytic efficiency.23 Within the catalytic domain of the three sialidases there is conservation of the four BNR motifs, and the key catalytic amino acids around the active site (Figs. 7 and 8). A major difference around the active site is the tryptophan (Trp674) that confers a preference in NanB for α2,3-linked substrates. This residue is conserved in NanC, which is also predicted to show the same substrate preference as NanB. In NanA, however, the tryptophan is replaced by glycine and there is a large insertion beyond this region (Fig. 7). As shown in Fig. 5b, NanA has no selectivity and can cleave sialic acid equally well from α2,3, α2,6 and α2,8-linked glycoconjugates. Another difference between the three pneumococcal sialidases is the region around Asn352 and Asn353 in NanB, where the first asparagine interacts with the O7 and O8 hydroxyls of Neu5Ac2en. In NanC, these residues are phenylalanine and arginine, whereas in NanA they are isoleucine and phenylalanine: in both cases, presumably providing part of the hydrophobic pocket that accommodates the acetamino group of the substrate, but removing the possibility of hydrogen bonding to the glycerol moiety. Finally, Thr539 that sits beneath the glycerol moiety is serine or asparagine in NanC and NanA, respectively. In the case of NanA, the longer side
443 chain may participate in interactions with the O8 and O9 hydroxyls of the substrate's glycerol group, leading to the lower Ki of 2 μM for Neu5Ac2en, compared to 0.3 mM for NanB. NanA and NanB, which are present in 100% and 96% of clinical isolates of S. pneumoniae,4 have both been shown to participate in the sequential deglycosylation of human glycoconjugates.35,36 It has also been shown that sialic acid can be removed from the Eustachian tube epithelia of chinchillas infected with a NanA-deficient S. pneumoniae strain,37 and that both NanA and NanB are important for infection and sepsis in a mouse model of pneumococcal disease.8 The removal of sialic acids from host glycoconjugates is proposed to expose adherence sites, 6 modify the surfaces of competing bacteria38 and alter the clearance function of host defence proteins.39 In common with many sialidasecontaining bacteria,40 S. pneumoniae can also utilize sialic acid as an energy source,36 as the genome encodes sialic acid transporters, and an N-acetylneuraminate lyase. It would be of interest to discover if Neu2,7-anhydro5Ac is a substrate for the lyase, and hence a source of nutrition. What is the advantage of S. pneumoniae encoding three exported sialidases? NanA contains an anchor motif and remains attached to the bacterial cell surface, has a pH optimum of 7 and is not selective for the sialic acid linakage. NanB, as we have shown, has a substrate preference for α2,3-linked sialic acids, and a previously reported pH optimum of 5.25,41 We would also predict that NanC has properties similar to those of NanB, given its 50% sequence identity and conservation of the tryptophan that confers a selective preference for α2,3-linked sialic acid. Recent studies on the anatomical distribution of sialic acid receptors for the influenza virus has shown that the epithelial cells in the upper respiratory tract of humans possess mainly sialic acid linked to galactose by an α2,6 linkage, but that many cells in the respiratory bronchioles and alveoli possess sialic acid linked α2,3 to galactose.42,43 It may be that NanB, and possibly NanC, provides an advantage for survival of the bacterium in the lungs, whereas NanA would be the dominant sialidase in the upper respiratory tract. Certainly, both NanA and NanB are essential for survival of pneumococci in the nasopharynx in intranasal challenge, and they cannot compensate for one another in single gene knockout experiments.8 Finally, the serendipitous discovery that Ches is an inhibitor of NanB provides the basis for novel drug development against the pneumococcal sialidases. Ches inhibits NanB with a Ki of 0.5 mM, similar to Neu5Ac2en, which inhibits with a Ki of 0.3 mM. Ches also inhibits NanA with a Ki of 0.65 mM. The sulphonic acid group of Ches interacts with all three arginines in the active site, in contrast to the carboxylic acid group of Neu5Ac2en that interacts with only two. In the Ches complex, a water molecule (W4 in Fig. 3) occupies the position of the O4 of Neu5Ac2en. Chemical modification of Ches at its amino group to occupy this water position could provide a ligand with increased binding affinity
Structure of S. Pneumoniae NanB
444
Fig. 7 (legend on next page)
Structure of S. Pneumoniae NanB
445
through both enthalpic and entropic gains. Such modifications based on a Ches framework are currently in progress, and may provide a new class of inhibitors for the pneumococcal, and potentially other sialidases involved in pathogenesis.
Materials and Methods Cloning of NanB The S. pneumoniae nanB gene was supplied in a PGETeasy vector and amplified by polymerase chain reactions (PCR) using the following primers: forward 5′-GTGATTATAGCTAGCATGAATAAAAGAGGTC-3′ reverse 5′-CGATTGGAGGTCTCGAGTTTTGTTAAATC-3′
with the NheI and XhoI restrictions sites underlined. The cleaned PCR product was cloned into the PET23b vector (Novagen) using a Gel Extraction Kit (QIAGEN). Restriction enzyme digestion or colony PCR methods were used to select the positive colonies. All plasmid DNA were extracted with a Mini-Prep Kit (Promega) and stored at −20 °C. The DNA sequence was confirmed by The Sequencing Service, University of Dundee, UK. Expression and purification Recombinant NanB/PET23b (ampicillin-resistant) plasmid was introduced into Escherichia coli BL21 (DE3) expression strain (Novagen) for protein production. Single colonies of the cells containing nanB were picked and put into Luria Bertani (LB) medium with 100 μg/ml ampicillin at 37 °C for inoculation. Isopropyl thio-β-D-galactopyranoside (IPTG) was added (0.5 mM final concentration) to induce the expression of the proteins when the absorbance at 600 nm (A600) of the cultures reached 0.6. The cells continued to be cultured overnight at 25 °C before harvesting by centrifugation at 16,780g for 30 min. The cell pellets were suspended in phosphate-buffered saline (PBS; 20 mM sodium phosphate, 150 mM sodium chloride, pH 7.4) containing 10 mM imidazole and sonicated with five bursts of 30 s each. Protease Inhibitor Cocktail tablets (one tablet per 25 ml extract; Roche Diagnostics) and DNase I (Sigma, final concentration 20 μg/ml) were then added. The crude cell extract was centrifuged at 75,600g for 25 min to remove the cell debris and the supernatant was filtered with a 0.45 μm pore size syringe-driven filter before starting protein purification. The soluble cell extract was first loaded onto a 5 ml HisTrap column (GE Healthcare) equilibrated on a BioCad system (GMI Inc. USA) in PBS containing 10 mM imidazole. The bound protein was eluted with a 30 mM–500 mM gradient of imidazole in PBS. Protein purity was assessed by SDS-PAGE and matrix-assisted laser desorption/ionisation time-of-flight mass spectrometry (MALDI-TOF MS) was used to confirm the protein identity. After overnight dialysis of the collected fractions
Fig. 8. Active site conservation. Residues conserved across NanA, NanB and NanC are coloured red on the NanB surface, with Neu5Ac2en shown with yellow carbons.
in 10 mM Tris-HCl, pH7.5, the protein was run through a 5 ml Q-FF anion-exchange column (GE Healthcare) and eluted with a 0 mM–500 mM gradient of NaCl. Those fractions with relatively high purity protein of expected molecular mass in SDS-PAGE were subjected to sizeexclusion chromatography. A 120 ml Sephacryl S-200 column (GE Healthcare) was used for the last gelfiltration step and high-purity target protein was pooled. The purified NanB was dialysed against 0.01 M sodium phosphate, pH7.0, overnight before concentration and storage. Kinetic characterization of NanB and NanA Sialidase activity was measured by using the fluorogenic substrate 2′-(4-methylumbelliferyl)-α-D-N-acetylneuraminic acid (MUAN; Sigma), which was made as a 10 mM stock in distilled, deionised water and kept at −20 °C. In this assay, appropriately diluted protein samples were assayed in a reaction volume of 1 ml containing 0.1 mM MUAN in 0.1 M sodium acetate pH 5.6. Reactions were incubated at 37 °C for 5 min and then stopped by adding 2ml of 0.25 M glycine buffer, pH 10.5. The relative fluorescence of released methylumbelliferone (MU) was determined at 365 nm for excitation and 450 nm for emission. This method was used also for the kinetic characterization of NanB and NanA. Briefly, all the reactions used 200 ng of NanB, or 35 ng of NanA, various concentrations of MUAN from 10 μM to 100 μM in the presence or in the absence of Ches (1–10 mM) or Neu5Ac2en (0.1–0.8 mM for NanB, 5–20 μM for NanA). The reaction velocity was presented as ΔFluorescence/min, which was determined from the amount of MU released in 5 min.
Fig. 7. Sequence alignment of the three S. pneumoniae sialidases over the region encompassed by the structure of NanB, showing the location of the secondary structural elements. The three domains are shaded red, green and blue. Residues predicted to be involved in sialic acid recognition in the lectin domain are marked + (charged/polar) or o (hydrophobic). Key catalytic residues are marked ⁎. The location of the BNRs are labelled. The figure was produced using ClustalW254 and ESPript.55
Structure of S. Pneumoniae NanB
446 Substrate specificity assay Quantification of N-acetylneuraminic acid released from α2,3, α2,6 or α2,8-linked sialyllactoses (Sigma) was performed with the thiobarbituric acid assay.44 Briefly, after incubating the enzyme with 30 μg of substrates in 100 μl 0.1 M sodium acetate, pH 5.5 at 37 °C for 1 h, the reactions were treated with 0.25 ml of 25 mM H5IO6, 0.125 M H2SO4, pH 1.2 in a 37 °C waterbath for 30 min. Then 0.2 ml of 2% (w/v) NaAsO2, 0.5 M HCl was added to the reactions to remove the excess periodate. The tubes were shaken for 1–2 min until the contents were colourless and then 2 ml of 0.1 M 2-thiobarbituric acid, pH 9 was added and the tube contents boiled for 7.5 min. Afterwards, the cooled reaction was shaken with 3 ml of nbutanol/12 M HCl (19:1, v/v) and centrifuged to separate the layers. The upper butanol layer was removed for the measurement of absorbance at 549 nm. The amount of sialic acid released was calculated as described.45 NMR experiments NanB reactions with α-2,3-sialyllactose, 3SL, or α-2,6sialyllactose, 6SL, were monitored by 1H NMR spectroscopy. In each reaction, 12 μl of 70 μM NanB, 14 μl of 3SL or 6SL (30 mM in 2H2O) and 564 μl of PBS in 2H2O, pH7, were mixed and transferred into a 5 mm NMR tube. Repeats were done in 50 mM sodium acetate in 2H2O at pH 5. For the trans-sialidase assay, a mixture of 25 μl of Nacetyllactosamine (Galβ1-4GlcNAc, 52 mM in 2H2O) and 14 μl of 3SL in 49 μl of PBS in 2H2O were incubated with
NanB for 48 h to assay whether NanB can transfer the sialic acid residues from 3SL to Galβ1-4GlcNAc. After the reactions were incubated at 298 K for 24 h and 48 h, 1H NMR spectra were measured on a 500 MHz Bruker spectrometer at 298 K with 256 scans and a relaxation delay of 2 s. The residual water signal was suppressed by continuous low-power irradiation of the water signal during the relaxation delay. Protein crystallization Results of a pre-crystallization assay kit (Hampton Research) suggested an optimum concentration of purified NanB of 5.3 mg/ml for crystallization trials. All the subsequent crystallization experiments were done at 20 °C by the sitting-drop, vapour-diffusion method. Initially, commercial kits PACT Premier (Molecular Dimensions), Nextal Classic (Qiagen), Index (Hampton Research) and Wizard (Emerald BioSystems) conditions were screened by the high-throughput Rhombix Screen® robot system (Rhombix) for protein crystallization. Those conditions with crystalline materials were selected for crystallization optimization. For manual protein crystallization, each reservoir well of a 96-well protein crystallization tray (Douglas Instruments) was filled with 100 μl of reservoir solution. The crystallization drops were composed of equal amounts of protein solution and reservoir solution (1–2 μl each). After several rounds of optimization, the best conditions for NanB crystallization were: (1) 10% (w/v) polyethylene glycol (PEG) 8 K, 0.15 M NaCl, 0.1 M Ches pH 9.5; and (2) 7% (w/v) PEG 8 K and 0.1 M imidazole
Table 1. X-ray data collection and refinement statistics
A. Data collection Space group X-ray source X-ray wavelength (Å) Resolution (Å) Unit cell dimensions a (Å) b (Å) c (Å) No. observations Redundancy Completeness (%) Rmerge I/σI B. Data refinement No. reflections No. atoms Protein Ligand Water Average B-factor (Å2) Protein Ligands Water R-factor Rfree r.m.s.d from ideal Bond lengths (Å) Bond angles (°) wwPDB code
NanB apo
NanB-DANA complex
NanB-Ches complex
P212121 ID14-1 0.934 30–2.30 (2.38–2.30)
P212121 In-house 1.542 30–2.40 (2.49–2.40)
P212121 In-house 1.542 30–1.70 (1.76–1.70)
76.6 82.7 116.7 211,813 6.1 (5.8) 98.03 (95.0) 0.109 (0.46) 13.9 (4.8)
76.6 82.7 117.4 105,254 5.1 (4.8) 100 (99.6) 0.138 (0.35) 6.3 (2.8)
76.6 82.7 116.7 234,069 3.1 (1.6) 92.1 (62.7) 0.083 (0.36) 8.9 (2.2)
32,865
28,251
71,846
5189 – 451
5189 20 463
5189 26 797
25.2 – 30.0 0.176 0.232 0.012
22.1 20.1 21.8 0.196 0.278 0.012
17.1 20.9 28.1 0.194 0.229 0.008
1.298 2VW0
1.504 2VW1
1.099 2VW2
Numbers in parentheses refer to the highest resolution shell. Rmerge = ∑hkl∑i|Ihkl,i−〈Ihkl〉|/∑hkl〈Ihkl〉. Rcryst and Rfree = (∑||Fo|−|Fc||)/(∑|Fo|).
Structure of S. Pneumoniae NanB pH 8.0. Crystals appeared in about three days and reached maximum size after two weeks. The Neu5Ac2en complex structure was obtained by soaking a NanB crystal (crystallized in 7% (w/v) PEG 8 K, 0.1 M imidazole, pH 8.0) in 100 mM Neu5Ac2en for 15 min. X-ray diffraction data collection and processing Crystals were cryoprotected by transfer for a few minutes to a solution of the crystallization buffer with 10% (v/v) glycerol added before data collection at 100 K. Data were collected in-house (Rigaku-MSC Micromax-007 X-ray generator and R-AXIS detectors) or at the ESRF, Grenoble. MOSFLM46 was used for data integration, the space group is P212121 and data collection statistics are given in Table 1. Structure determination and refinement The leech trans-sialidase structure (PDB code 1SLI) was used to solve the NanB structure by molecular replacement using AMoRe in the CCP4 computer suite.47 The ARPwARP program was used to build the initial structure using the 1.7 Å Ches complex data.48 The protein models were rebuilt with O49 and Coot.50 After further refinement with REFMAC5,51 the structures were inspected and validated with Coot and MolProbity.52 Refinement statistics are summarized in Table 1. Protein Data Bank accession numbers Coordinates and structure factors have been deposited in the Protein Data Bank with accession numbers 2VW0, 2WV1 and 2VW2 for the apo, Neu5Ac2en and Ches complexes, respectively.
Acknowledgements G.X. was supported by a Scottish Government International Scholarship and by Biocryst Pharmaceuticals Inc, Birmingham, Alabama. Resources of the St Andrews-based Scottish Structural Proteomics Facility, funded by the Scottish Funding Council, the Biotechnology and Biological Sciences Research Council (BBSRC) and the University of St Andrews, were used in this project.
Supplementary Data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j. jmb.2008.09.032
References 1. Luotonen, J., Herva, E., Karma, P., Timonen, M., Leinonen, M. & Makela, P. H. (1981). The bacteriology of acute otitis media in children with special reference to Streptococcus pneumoniae as studied by bacteriological and antigen detection methods. Scand. J. Infect. Dis. 13, 177–183.
447 2. Jedrzejas, M. J. (2001). Pneumococcal virulence factors: structure and function. Microbiol. Mol. Biol. Rev. 65, 187–207. 3. Mitchell, T. J. (2000). Virulence factors and the pathogenesis of disease caused by Streptococcus pneumoniae. Res. Microbiol. 151, 413–419. 4. Pettigrew, M. M., Fennie, K. P., York, M. P., Daniels, J. & Ghaffar, F. (2006). Variation in the presence of neuraminidase genes among Streptococcus pneumoniae isolates with identical sequence types. Infect. Immun. 74, 3360–3365. 5. Tong, H. H., Blue, L. E., James, M. A. & DeMaria, T. F. (2000). Evaluation of the virulence of a Streptococcus pneumoniae neuraminidase-deficient mutant in nasopharyngeal colonization and development of otitis media in the chinchilla model. Infect. Immun. 68, 921–924. 6. Tong, H. H., Liu, X., Chen, Y., James, M. & Demaria, T. (2002). Effect of neuraminidase on receptor-mediated adherence of Streptococcus pneumoniae to chinchilla tracheal epithelium. Acta Otolaryngol. 122, 413–419. 7. McCullers, J. A. & Bartmess, K. (2003). Role of Neuraminidase in Lethal Synergism between Influenza Virus and Streptococcus pneumoniae. J. Infect. Dis. 187, 1000–1009. 8. Manco, S., Hernon, F., Yesilkaya, H., Paton, J. C., Andrew, P. W. & Kadioglu, A. (2006). Pneumococcal neuraminidases A and B both have essential roles during infection of the respiratory tract and sepsis. Infect. Immun. 74, 4014–4020. 9. Simell, B., Jaakkola, T., Lahdenkari, M., Briles, D., Hollingshead, S., Kilpi, T. M. & Kayhty, H. (2006). Serum antibodies to pneumococcal neuraminidase NanA in relation to pneumococcal carriage and acute otitis media. Clin. Vaccine Immunol. 13, 1177–1179. 10. Crennell, S. J., Garman, E. F., Laver, W. G., Vimr, E. R. & Taylor, G. L. (1993). Crystal structure of a bacterial sialidase (from Salmonella typhimurium LT2) shows the same fold as an influenza virus neuraminidase. Proc. Natl Acad. Sci. USA, 90, 9852–9856. 11. Crennell, S. J., Garman, E., Laver, G., Vimr, E. R. & Taylor, G. (1994). Crystal structure of Vibrio cholerae neuraminidase reveals dual lectin-like domains in addition to the catalytic domain. Structure, 2, 535–544. 12. Gaskell, A., Crennell, S. J. & Taylor, G. (1995). The three domains of a bacterial sialidase: a beta-propeller, an immunoglobulin module and a galactose-binding jelly-roll. Structure, 3, 1197–1205. 13. Varghese, J. N., Laver, W. G. & Colman, P. M. (1983). Structure of the influenza virus glycoprotein antigen neuraminidase at 2.9 Å resolution. Nature, 303, 35–40. 14. Burmeister, W. P., Henrissat, B., Bosso, C., Cusack, C. & Ruigrok, R. W. H. (1993). Influenza B virus neuraminidase can synthesise its own inhibitor. Structure, 1, 19–26. 15. Crennell, S. J., Takimoto, T., Portner, A. & Taylor, G. (2000). Crystal structure of the multifunctional paramyxovirus hemagglutinin-neuraminidase. Nature Struct. Biol. 7, 1068–1074. 16. Buschiazzo, A., Tavares, G. A., Campetella, O., Spinelli, S., Cremona, M. L., Paris, G. et al. (2000). Structural basis of sialyltransferase activity in trypanosomal sialidases. EMBO J. 19, 16–24. 17. Buschiazzo, A., Amaya, M. F., Cremona, M. L., Frasch, A. C. & Alzari, P. M. (2002). The crystal structure and mode of action of trans-sialidase, a key enzyme in Trypanosoma cruzi pathogenesis. Mol. Cell, 10, 757–768. 18. Lou, Y., Li, S.-C., Chou, M.-Y., Li, Y.-T. & Lou, M.
448
19.
20. 21.
22.
23. 24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
(1998). The crystal structure of an intramolecular trans-sialidase with a NeuAcalpha2 → 3Gal specificity. Structure, (Camb), 6, 521–530. Chavas, L. M., Tringali, C., Fusi, P., Venerando, B., Tettamanti, G., Kato, R., Monti, E. & Wakatsuki, S. (2004). Crystal structure of the human cytosolic Sialidase Neu2: Evidence for the dynamic nature of substrate recognition. J. Biol. Chem, 280, 469–475. Taylor, G. (1996). Sialidases: structures, biological significance and therapeutic potential. Curr. Opin. Struct. Biol. 6, 830–837. Moustafa, I., Connaris, H., Taylor, M., Zaitsev, V., Wilson, J. C., Kiefel, M. J., von Itzstein, M. & Taylor, G. (2004). Sialic acid recognition by Vibrio cholerae neuraminidase. J. Biol. Chem. 279, 40819–40826. Newstead, S. L., Watson, J. N., Bennet, A. J. & Taylor, G. (2005). Galactose recognition by the carbohydratebinding module of a bacterial sialidase. Acta Crystallogr. D, 61, 1483–1491. Thobhani, S., Ember, B., Siriwardena, A. & Boons, G. J. (2003). Multivalency and the mode of action of bacterial sialidases. J. Am. Chem. Soc. 125, 7154–7155. Camara, M., Boulnois, G. J., Andrew, P. W. & Mitchell, T. J. (1994). A neuraminidase from Streptococcus pneumoniae has the features of a surface protein. Infect. Immun. 62, 3688–3695. Berry, A. M., Lock, R. A. & Paton, J. C. (1996). Cloning and characterization of nanB, a second Streptococcus pneumoniae neuraminidase gene, and purification of the NanB enzyme from recombinant Escherichia coli. J. Bacteriol. 178, 4854–4860. Boraston, A. B., Ficko-Blean, E. & Healey, M. (2007). Carbohydrate recognition by a large sialidase toxin from Clostridium perfringens. Biochemistry, 46, 11352–11360. Li, Y. T., Nakagawa, H., Ross, S. A., Hansson, G. C. & Li, S. C. (1990). A novel sialidase which releases 2,7anhydro-alpha-N-acetylneuraminic acid from sialoglycoconjugates. J. Biol. Chem. 265, 21629–21633. Chou, M. Y., Li, S. C., Kiso, M., Hasegawa, A. & Li, Y. T. (1994). Purification and characterization of sialidase L, a NeuAc alpha 2 → 3Gal-specific sialidase. J. Biol. Chem. 269, 18821–18826. Luo, Y., Li, S. C., Chou, M. Y., Li, Y. T. & Luo, M. (1998). The crystal structure of an intramolecular transsialidase with a NeuAc alpha2 → 3Gal specificity. Structure, 6, 521–530. Luo, Y., Li, S. C., Li, Y. T. & Luo, M. (1999). The 1.8 A structures of leech intramolecular trans-sialidase complexes: evidence of its enzymatic mechanism. J. Mol. Biol. 285, 323–332. Hoyer, L. L., Roggentin, P., Schauer, R. & Vimr, E. R. (1991). Purification and properties of cloned Salmonella typhimurium LT2 sialidase with virus-typical kinetic preference for sialyl alpha 2–3 linkages. J. Biochem. 110, 462–467. Amaya, M. F., Buschiazzo, A., Nguyen, T. & Alzari, P. M. (2003). The high resolution structures of free and inhibitor-bound Trypanosoma rangeli sialidase and its comparison with T. cruzi trans-sialidase. J. Mol. Biol. 325, 773–784. Amaya, M. F., Watts, A. G., Damager, I., Wehenkel, A., Nguyen, T., Buschiazzo, A. et al. (2004). Structural insights into the catalytic mechanism of Trypanosoma cruzi trans-sialidase. Structure, 12, 775–784. Lou, Y., Li, S. C., Li, Y. T. & Lou, M. (1998). The 1.8 Å structures of Leech intramolecular trans-sialidase complexes: evidence of its enzymatic mechanism. J. Mol. Biol. 285, 323–332.
Structure of S. Pneumoniae NanB 35. King, S. J., Hippe, K. R. & Weiser, J. N. (2006). Deglycosylation of human glycoconjugates by the sequential activities of exoglycosidases expressed by Streptococcus pneumoniae. Mol. Microbiol. 59, 961–974. 36. Burnaugh, A. M., Frantz, L. J. & King, S. J. (2008). Growth of Streptococcus pneumoniae on human glycoconjugates is dependent upon the sequential activity of bacterial exoglycosidases. J. Bacteriol. 190, 221–230. 37. Tong, H. H., James, M., Grants, I., Liu, X., Shi, G. & DeMaria, T. F. (2001). Comparison of structural changes of cell surface carbohydrates in the eustachian tube epithelium of chinchillas infected with a Streptococcus pneumoniae neuraminidase-deficient mutant or its isogenic parent strain. Microb. Pathog. 31, 309–317. 38. Shakhnovich, E. A., King, S. J. & Weiser, J. N. (2002). Neuraminidase expressed by Streptococcus pneumoniae desialylates the lipopolysaccharide of Neisseria meningitidis and Haemophilus influenzae: a paradigm for interbacterial competition among pathogens of the human respiratory tract. Infect. Immun. 70, 7161–7164. 39. O'Toole, R. D., Goode, L. & Howe, C. (1971). Neuraminidase activity in bacterial meningitis. J. Clin. Invest. 50, 979–985. 40. Corfield, T. (1992). Bacterial sialidases — roles in pathogenicity and nutrition. Glycobiology, 2, 509–521. 41. Berry, A. M., Paton, J. C., Glare, E. M., Hansman, D. & Catcheside, D. E. (1988). Cloning and expression of the pneumococcal neuraminidase gene in Escherichia coli. Gene, 71, 299–305. 42. Shinya, K., Ebina, M., Yamada, S., Ono, M., Kasai, N. & Kawaoka, Y. (2006). Avian flu: influenza virus receptors in the human airway. Nature, 440, 435–436. 43. Chandrasekaran, A., Srinivasan, A., Raman, R., Viswanathan, K., Raguram, S., Tumpey, T. M. et al. (2008). Glycan topology determines human adaptation of avian H5N1 virus hemagglutinin. Nat. Biotechnol. 26, 107–113. 44. Aminoff, D. (1961). Methods for the quantitative estimation of N-acetylneuraminic acid and their application to hydrolysates of sialomucoids. Biochem. J. 81, 384–392. 45. Warren, L. (1959). The thiobarbituric acid assay of sialic acids. J. Biol. Chem. 234, 1971–1975. 46. Leslie, A. G. (2006). The integration of macromolecular diffraction data. Acta Crystallogr. D, 62, 48–57. 47. Collaborative Computational Project, Number 4. (1994). The CCP4 suite: programs for protein crystallography. Acta Crystallogr. D, 50, 760–763. 48. Perrakis, A., Morris, R. & Lamzin, V. S. (1999). Automated protein model building combined with iterative structure refinement. Nature Struct. Biol. 6, 458–463. 49. Jones, T. A., Zou, J. Y., Cowan, S. W. & Kjeldgaard, M. (1991). Improved methods for building protein models in electron density maps and the location of errors in these models. Acta Crystallogr. A, 47, 110–119. 50. Emsley, P. & Cowtan, K. (2004). Coot: model-building tools for molecular graphics. Acta Crystallogr. D, 60, 2126–2132. 51. Murshudov, G. N., Vagin, A. A. & Dodson, E. J. (1997). Refinement of macromolecular structures by the maximum-likelihood method. Acta Crystallogr. D, 53, 240–255. 52. Lovell, S. C., Davis, I. W., Arendall, W. B., de Bakker, P. I., Word, J. M., Prisant, M. G. et al. (2003). Structure
Structure of S. Pneumoniae NanB validation by Cα geometry: ϕ, ψ and Cβ deviation. Proteins: Struct. Funct. Genet. 50, 437–450. 53. Baker, N., Sept, D., Joseph, S., Holst, M. & McCammon, J. (2001). Electrostatics of nanosystems: Application to microtubules and the ribosome. Proc. Natl Acad. Sci. USA, 98, 10037–10041. 54. Larkin, M. A., Blackshields, G., Brown, N. P., Chenna,
449 R., McGettigan, P. A., McWilliam, H. et al. (2007). Clustal W and Clustal X version 2.0. Bioinformatics, 23, 2947–2948. 55. Gouet, P., Robert, X. & Courcelle, E. (2003). ESPript/ ENDscript: Extracting and rendering sequence and 3D information from atomic structures of proteins. Nucleic Acids Res. 31, 3320–3323.
doi:10.1016/j.jmb.2008.09.032
Available online at www.sciencedirect.com
Crystal Structure of the NanB Sialidase from Streptococcus pneumoniae Guogang Xu 1 , Jane A. Potter 1 , Rupert J. M. Russell 1 , Marco R. Oggioni 2 , Peter W. Andrew 3 and Garry L. Taylor 1 ⁎ 1
Centre for Biomolecular Sciences, University of St Andrews, St Andrews, Fife, KY16 9ST, UK 2
Lab Microbiologia Molecolare e Biotecnologia, Dipartimento di Biologia Molecolare, Universita' di Siena, Policlinico Le Scotte, 53100 Siena, Italy 3
Department of Infection, Immunity and Inflammation, University of Leicester, Leicester LE1 9HN, UK Received 15 July 2008; received in revised form 7 September 2008; accepted 12 September 2008 Available online 21 September 2008
The Streptococcus pneumoniae genomes encode up to three sialidases (or neuraminidases), NanA, NanB and NanC, which are believed to be involved in removing sialic acid from host cell surface glycans, thereby promoting colonization of the upper respiratory tract. Here, we present the crystal structure of NanB to 1.7 Å resolution derived from a crystal grown in the presence of the buffer Ches (2-N-cyclohexylaminoethanesulfonic acid). Serendipitously, Ches was found bound to NanB at the enzyme active site, and was found to inhibit NanB with a Ki of ∼0.5 mM. In addition, we present the structure to 2.4 Å resolution of NanB in complex with the transition-state analogue Neu5Ac2en (2-deoxy-2,3-dehydro-N-acetyl neuraminic acid), which inhibits NanB with a Ki of ∼ 0.3 mM. The sulphonic acid group of Ches and carboxylic acid group of Neu5Ac2en interact with the arginine triad of the active site. The cyclohexyl group of Ches binds in the hydrophobic pocket of NanB occupied by the acetamidomethyl group of Neu5Ac2en. The topology around the NanB active site suggests that the enzyme would have a preference for α2,3-linked sialoglycoconjugates, which is confirmed by a kinetic analysis of substrate binding. NMR studies also confirm this preference and show that, like the leech sialidase, NanB acts as an intramolecular trans-sialidase releasing Neu2,7-anhydro5Ac. All three pneumoccocal sialidases possess a carbohydrate-binding domain that is predicted to bind sialic acid. These studies provide support for a possible differential role for NanB compared to NanA in pneumococcal virulence. © 2008 Elsevier Ltd. All rights reserved.
Edited by G. Schulz
Keywords: sialidase; neuraminidase; drug design; substrate specificity; crystal structure
Introduction Streptococcus pneumoniae is a major human pathogen responsible for respiratory tract infections, septicemia and meningitis, and continues to produce numerous cases of disease with a relatively high mortality rate. In children, S. penumoniae is the most frequent cause of otitis media,1 and the nasopharynx serves as a reservoir for the bacterium to enter the middle ear via the Eustachian tube, particularly following a respiratory viral infection. Several virulence factors may contribute to colonization and early infection processes.2 Sialidases are one key *Corresponding author. E-mail address: [email protected]. Abbreviations used: BNR, bacterial neuraminidase repeat; CBM, carbohydrate-binding module.
virulence factor, as they remove sialic acid from host cell surface glycans, probably unmasking certain receptors to facilitate bacterial adherence and colonization.3 To date, all S. pneumoniae clinical isolates investigated present prominent sialidase activity. Three distinct sialidases, NanA, NanB and NanC, are encoded in the S. pneumoniae genomes. A screening study of sialidase genes in 342 clinical pneumococcal isolates identified nanA, nanB and nanC to be present in 100%, 96% and 51% of these strains, respectively.4 NanA has been shown to have an important role in host-pneumoccocal interactions in the upper respiratory tract.5,6 It is also known that there is a lethal synergism between the influenza virus and S. pneumoniae, probably as a consequence of the viral sialidase stripping sialic acid from host glycoproteins.7 Experiments with mouse models using isogenic pneumococcal mutants show that both NanA and NanB are essential to S. pneumoniae
0022-2836/$ - see front matter © 2008 Elsevier Ltd. All rights reserved.
Structure of S. Pneumoniae NanB
infection of the respiratory tract and sepsis.8 Finally, the S. pneumoniae sialidases show promise as antigens for vaccine development targeted at pneumoccocal disease.9 There are crystal structures of sialidases from bacteria,10–12 viruses,13–15 trypanosomes,16,17 leech,18 and man.19 The catalytic domains of sialidases have a six-bladed β-propeller topology, with conservation of key catalytic amino acids.20 The non-viral sialidases also have conserved bacterial neuraminidase repeats (BNRs) between one and five times along the sequence. Many sialidases have additional domains described as lectins or carbohydrate-binding modules (CBMs). For example, Vibrio cholerae sialidase has two lectin domains flanking the catalytic domain, one of which binds sialic acid;21 Micromonospora viridifaciens sialidase has a galactose-binding domain C-terminal to the catalytic domain; 22 the leech sialidase has a lectin-like domain N-terminal to the catalytic domain;18 and the trypanosome (trans-) sialidases have a lectin-like domain C-terminal to the catalytic domain.16,17 It has been shown that the presence of these CBMs increases the catalytic efficiency of the sialidases, particularly towards polysaccharide substrates.23 The three pneumoccocal sialidases NanA, NanB and NanC have a molecular mass of 115 kDa, 78 kDa and 82 kDa, respectively. Sequence analysis shows that all three possess a signal sequence, followed by a lectin domain, or CBM, before the catalytic domain, the latter domain containing four BNRs (Fig. 1). In contrast to NanB and NanC, NanA has an additional C-terminal domain containing an LPETG motif that tethers the enzyme to the bacterial surface.24 NanB and NanC share 50% sequence identity, whereas they both share only 25% identity with NanA. NanB is reported to have a pH optimum of 4.5 in contrast to pH 6.5–7.0 for NanA.25 This, together with the inability of the two enzymes to compensate for one another in the gene knockout experiments, suggests that they have separately defined roles in pathogenesis.8 Here, we present the crystal structure of NanB at a resolution of 2.3 Å, and complexes of NanB with Ches and Neu5Ac2en to resolutions of 1.7 Å and 2.4 Å, respectively. The observation of a Ches molecule in the active site led us to measure its ability to inhibit the enzyme, which it does with a Ki of ∼ 0.5 mM comparable to Neu5Ac2en that inhibits with a Ki of ∼ 0.3 mM. The presence of a tryptophan
437 close to the position of the substrate aglycon in NanB is seen also in the leech trans-sialidase and the Trypanosoma cruzi trans-sialidase, which both show a strict preference for α2,3-linked sialosylglycosides. NMR studies show that, like the leech enzyme, NanB is an intramolecular trans-sialidase, releasing Neu2,7-anhydro5Ac, and kinetic studies show that although NanB has a preference for α2,3-linked substrates, it can cleave α2,6 and α2,8 linkages. By homology with the carbohydrate-binding domain of the Clostridium perfringens NanJ sialidase, the similar domain of NanB is likely to bind sialic acid. Finally, Ches provides a framework for the development of a novel class of sialidase inhibitors.
Results Overall structure The crystal structure of NanB spans residues 39 to 696, and is missing the signal peptide (residues 1–38) and the final residue, which is not visible in the electron density maps. NanB is a monomer with three domains: a concanavalin-like lectin domain (residues 39–228) and a canonical β-propeller catalytic domain (residues 229–345 and 459–696) into which is inserted the third domain (residues 346–458) that is formed predominantly from β-strands (Fig. 2a and b). The structure was solved by molecular replacement using the leech trans-sialidase as the search model (Protein Databank code 1sli), which shares 43% sequence identity with NanB, and the two proteins have an rmsd of 1.49 Å over 630 Cα atoms. The six-bladed catalytic domain is formed from a sequential repeat of four-stranded anti-parallel β-sheets and, like the leech enzyme, there is an extended loop between the first and second sheet (residues 297–323) that associates with, and forms part of, the inserted third domain that is itself between the second and third strands of the second sheet. This inserted domain is found only in the three pneumococcal sialidases, NanI from C. perfringens (2bf6) and the leech transsialidase, with no other structurally homologous domain being found in the protein structure databank. The role of this inserted domain is unknown. The electrostatic surface potential of the entire protein shows a distinct asymmetry, typical of secreted bacterial sialidases, with the surface remote from the active site carrying a negative charge (coloured
Fig. 1. A diagram of the three S. pneumoniae sialidases. CBM40 refers to type 40 carbohydrate-binding module (CBM) that recognizes sialic acid. The locations of BNRs are shown that conform to the sequence motif Ser/Thr-x-Asp-(x)-Gly-x-Thr(Trp/Phe), where x is any amino acid. The blue regions, containing the BNRs, comprise the β-propeller catalytic domain.
438
Structure of S. Pneumoniae NanB
Fig. 2. The NanB structure. (a and b) The catalytic β-propeller domain is shown in blue with Neu5Ac2en in the active site, the lectin domain is shown in red and the inserted domain is shown in green. b, the view represents a 90° rotation of the view in a around a horizontal axis. (c and d) The same views as in a and b, respectively, and show a surface representation coloured according to electrostatic potential from −10 kT/e to +10 kT/e, calculated using APBS.53 This and other figures were created with PyMOL [http://pymol.sourceforge.net/].
red in Fig. 2c and d) that may serve to orient the protein towards its negatively charged glycoconjugate substrates.
been retained throughout their evolution. These include three arginines (Arg245, Arg557, Arg619: NanB numbering) that interact with the carboxylate
Lectin or carbohydrate-binding module (CBM) The lectin domain of NanB is structurally homologous to the recently reported family 40 CMB (CBM40) of the C. perfingens NanJ sialidase, which was found to recognize a sialic acid moiety.26 The two domains share 26% sequence identity and superimpose with an rmsd of 1.64 Å for 168 Cα atoms. Several key residues of the sialic acid-binding site are conserved between the two CBMs, suggesting that the CBM of NanB binds to sialic acid containing glycoconjugates (Fig. 3); these include two arginines that interact with the carboxyl group of sialic acid (Arg117 and Arg193), a glutamic acid that interacts with the O4 hydroxyl (Glu115) and a hydrophobic pocket accommodating the methyl of the acetamido group (Leu84, Phe105, Leu126 and Tyr199). Neu5Ac2en binding A complex of NanB with Neu5Ac2en was obtained at 2.4 Å resolution, and the interactions made by the ligand with the active site are shown in Fig. 4a. Although there are a number of significant differences between the active sites of viral, bacterial and eukaryotic sialidases, a number of key features have
Fig. 3. Putative sialic acid-binding site in the NanB lectin domain. The lectin domain of NanB (salmon) is superimposed onto the CBM40 module from Clostridium perfringens NanJ (cyan; PDB code 2v73). Residues are labelled in the order NanB/NanJ, and hydrogen bond interactions in NanJ are shown as dotted lines.
Structure of S. Pneumoniae NanB
group of Neu5Ac2en. In NanB one of the arginines, Arg557, is 3.6 Å away from the nearest carboxylate oxygen of the ligand, whereas the other two are both within hydrogen bonding distance. The position of the first arginine (Arg245) is stabilised by a conserved glutamic acid (Glu669). The nucleophilic tyrosine (Tyr653) and an accompanying glutamic acid (Glu541) hydrogen bond with each other and sit beneath the C1–C2 bond of the ligand, the hydroxyl of Tyr653 being 3.1 Å from C1 and C2 of the ligand. In common with other bacterial sialidases, the O4 hydroxyl of Neu5Ac2en interacts with an arginine (Arg264) and an aspartic acid (Asp327). A conserved feature of sialidase active sites is the acid/base catalyst, Asp270, situated within a loop above the active site, that here interacts with O7 of the ligand and with a water molecule (W3) that, in turn, interacts with Asn352. This latter residue interacts with the glycerol hydroxyls O7 and O9 of Neu5Ac2en. The presence of Asn352 is one key difference between the active sites of NanB and the leech transsialidase, where the latter has serine at this position, which makes no interaction with Neu5Ac2en. The O8 hydroxyl of Neu5Ac2en does not interact with NanB. All sialidase active sites have a hydrophobic pocket to accommodate the N-acetyl group of the substrate, but the exact residues that form this pocket are generally not conserved. In the NanB structure, this pocket is composed of Ile326, Met346 and Tyr489. In the leech trans-sialidase, the isoleucine (Ile374) and methionine (Met394) are conserved, but the tyrosine is replaced by valine (Val538) with a tryptophan (Trp536) forming part of the pocket. Finally, water molecules W1 and W2 are involved in a network of key amino acids: Asp327, Asn352, Tyr509 and Glu541. Our kinetic data show that Neu5Ac2en inhibits NanB with a Ki of 0.3 mM, and NanA with a Ki of 2 μM. Ches binding Two Ches molecules are bound to NanB in the 1.7 Å resolution structure derived from a crystal grown in the presence of Ches buffer. Both are well ordered and interact with the catalytic domain, one in the active site and the other on the opposite side where it is involved in multiple interactions with the enzyme, either directly or via water molecules and a glycerol molecule. The presence of this latter Ches molecule may well contribute to the increased diffraction quality of crystals grown in the presence of Ches. The interactions made by the Ches molecule in the active site are shown in Fig. 4b. Two oxygens of the sulphonic acid group of Ches interact with all three arginines of the arginine triad. The amino group of Ches interacts with the acid/base Asp270, and via a water molecule (W4) with Asp327. The cyclohexyl ring of Ches occupies the hydrophobic pocket described above for Neu5Ac2en. The key water molecule, W1, that networks Asp327, Glu541 and Tyr509 is conserved in both the Neu5Ac2en and Ches complexes. Superposition of the native, Ne5Ac2en and Ches complexes reveals a very rigid active
439 site (Fig. 4c), with only the side chain of Asn352 moving significantly in order to participate in hydrogen bonding with the glycerol moiety of Neu5Ac2en. Our kinetic data show that Ches inhibits NanB with a Ki of 0.50 mM, and NanA with a Ki of 0.65 mM. Substrate specificity and product formation The leech trans-sialidase has a tryptophan (Trp734) packing against a tyrosine (Tyr643) close to the carboxylate group of Neu5Ac2en, which is suggested to give the leech trans-sialidase strict specificity towards α2,3-linked sialoglycosides. A similar feature is seen in the T. cruzi trans-sialidase, which also has specificity for α2,3-linked sialoglycosides. NanB also has a tryptophan (Trp674) packing against a tyrosine (Tyr589) in the same position (Fig. 5a). Measurement of the release of sialic acid from α2,3-, α2,6- and α2,8-linked substrates shows that NanB has a five- and tenfold preference for α2,3 over α2,6 and α2,8 respectively (Fig. 5a), whereas NanA shows little discrimination between the different linkages (Fig. 5b) and is N 10 times more active against each than NanB. NMR studies carried out at pH 7 confirm the ability of NanB to cleave sialic acid from α2,3 and α2,6 sialyllactose substrates, and show that the product Neu2,7-anhydro5Ac is formed, thus classifying NanB as an intramolecular trans-sialidase (Fig. 6 and Supplementary Data). In the case of α2,3-sialyllactose, the ratio of Neu2,7-anhydro5Ac: Neu5Ac is 9:1, 6:1 and 3:1 after 0.5 h, 24 h and 48 h, respectively. Although the ratio of the two products changes, the amount of Neu2,7-anhydro5Ac remains constant, at least for 48 h. Further experiments are in progress to investigate if both products are formed or whether Neu2,7-anhydro5Ac is first formed and then converted to Neu5Ac. In the case of α-2,6-sialyllactose, the reaction is much slower, with only 30% of the substrate cleaved after 48 h (Supplementary Data). The ratio of Neu2,7-anhydro5Ac:Neu5Ac remains at 1:2 after 24 h and 48 h. These experiments were repeated at pH 5 and gave similar results (Supplementary Data). NMR was used to explore the possibility that NanB might act as an intermolecular trans-sialidase similar to the T. cruzi enzyme, but no such activity was observed (Supplementary Data).
Discussion The crystal structure of NanB was determined using the structure of the leech trans-sialidase as a molecular replacement model, with which it shares 43% sequence identity. The leech trans-sialidase is unusual amongst sialidases in that it is an intramolecular trans-sialidase that cleaves only the Neu5Acα2,3Gal linkage of sialoglycoconjugates, releasing Neu2,7-anhydro5Ac,27,28 and is not inhibited by Neu5Ac2en.28 The selectivity for Neu5Acα2,3Gal in the leech trans-sialidase is dictated by the presence of a tryptophan close to where the aglycon would be
Structure of S. Pneumoniae NanB
440
Fig. 4 (legend on next page)
Structure of S. Pneumoniae NanB
441
Fig. 5. Substrate selectivity of NanB ((a) where ⁎⁎ indicates p b 0.01) and NanA (b).
positioned,29 and the intramolecular formation of Neu2,7-anhydro5Ac is proposed to be aided by the presence of a threonine (Thr593) that may promote an axial positioning of the glycerol group of the substrate.29,30 The NMR results show that NanB acts also as an intramolecular trans-sialidase, releasing Neu2,7-anhydro5Ac from both α2,3 and α2,6 sialyllactose. The critical threonine suggested to help orient the glycerol group is conserved in NanB (Thr539), and the unusual observation, for a sialidase, of a hydrogen bond interaction between the O7 hydroxyl and the acid-base Asp270 in NanB lends further support to the mechanism proposed earlier.30 Neu5Ac2en inhibits NanB with a Ki of 0.3 mM, which is similar to that found for Salmonella typhimurium sialidase (0.38 mM)31 and the Trypanasoma rangeli sialidase (0.14 mM).32 The complex of NanB with Neu5Ac2en clearly shows the inhibitor positioned in the active site (Fig. 4a), in contrast to the Neu5Ac2en leech trans-sialidase complex, which
is reported to not be inhibited by Neu5Ac2en, and which showed very weak electron density for the ligand.29 In NanB, Asn352 hydrogen bonds to both O7 and O9 hydroxyls of the glycerol group of Neu5Ac2en, whereas in the leech trans-sialidase the place of Asn352 in NanB is taken by a serine (Ser400) that does not form hydrogen bonds with the glycerol hydroxyls of Neu5Ac2en,29 and this could certainly lead to a reduced binding affinity for the leech enzyme. Despite the conservation in NanB of the tryptophan that confers strict α2,3 specificity to the leech trans-sialidase, NanB is able to remove sialic acids from α2,6 and α2,8 glycoconjugates, albeit five- to tenfold less efficiently compared to α2,3-linked substrates. A similarly positioned tryptophan is present in the T. cruzi trans-sialidase that also has strict specificity for sialic acid linked α2,3 to a galactose, but NanB has none of the structural features that confer trans-sialidase activity on the T. cruzi enzyme,33 and our NMR results confirm that
Fig. 4. Stereo drawings of the NanB active site. (a), The Neu5Ac2en complex. (b), The Ches complex. In both of these figures, the σa-weighted Fo–Fc difference electron density map is contoured at 3σ. c, Superposition of the native (green), Neu5Ac2en (cyan) and Ches (magenta) structures.
442
Structure of S. Pneumoniae NanB
Fig. 6. NMR spectra showing the time course of product release by NanB with α-2,3-sialyllactose (3SL) as the substrate.
Structure of S. Pneumoniae NanB
NanB does not have intermolecular trans-sialidase activity. It can be concluded that NanB acts as an intramolecular trans-sialidase with a distinct preference for α2,3-linked sialic acids. The amount of Neu2,7-anhydro5Ac produced by NanB remains constant, at least over the 48 h of the NMR experiment, whereas the amount of β-Neu5Ac increases slowly. It has been reported that the leech transsialidase can slowly convert Neu2,7-anhydro5Ac to Neu5Ac,34 and this may be the same for NanB, although further experiments are required to confirm this. Within the sequence encompassed by the structure of NanB reported here, NanB shares 23% sequence identity with NanA, and 50% identity with NanC (Fig. 7). All three sialidases have a lectin domain where the residues predicted to compose the hydrophobic pocket accommodating the acetamido methyl group of sialic acid are conserved as hydrophobic amino acids. In addition, the arginine predicted to interact with the carboxyl group of sialic acid in the CBM (Arg193 in NanB) is conserved across all three sialidases. NanC also conserves a glutamic acid and arginine that may interact with the O4 and carboxyl groups of sialic acid (Glu115 and Arg117 in NanB). However, NanA does not conserve these two amino acids, where they are arginine and serine, respectively. It is probable that the CBM40 domains of NanA and NanC will, like NanB, recognize sialic acid, but the details of binding, particularly for NanA, will require further structural studies. The concave carbohydrate-binding pocket of the CBM is on the same side of NanB as the active site, suggesting that the presence of such domains may help target the enzymes to appropriate substrates, and thereby increase their catalytic efficiency.23 Within the catalytic domain of the three sialidases there is conservation of the four BNR motifs, and the key catalytic amino acids around the active site (Figs. 7 and 8). A major difference around the active site is the tryptophan (Trp674) that confers a preference in NanB for α2,3-linked substrates. This residue is conserved in NanC, which is also predicted to show the same substrate preference as NanB. In NanA, however, the tryptophan is replaced by glycine and there is a large insertion beyond this region (Fig. 7). As shown in Fig. 5b, NanA has no selectivity and can cleave sialic acid equally well from α2,3, α2,6 and α2,8-linked glycoconjugates. Another difference between the three pneumococcal sialidases is the region around Asn352 and Asn353 in NanB, where the first asparagine interacts with the O7 and O8 hydroxyls of Neu5Ac2en. In NanC, these residues are phenylalanine and arginine, whereas in NanA they are isoleucine and phenylalanine: in both cases, presumably providing part of the hydrophobic pocket that accommodates the acetamino group of the substrate, but removing the possibility of hydrogen bonding to the glycerol moiety. Finally, Thr539 that sits beneath the glycerol moiety is serine or asparagine in NanC and NanA, respectively. In the case of NanA, the longer side
443 chain may participate in interactions with the O8 and O9 hydroxyls of the substrate's glycerol group, leading to the lower Ki of 2 μM for Neu5Ac2en, compared to 0.3 mM for NanB. NanA and NanB, which are present in 100% and 96% of clinical isolates of S. pneumoniae,4 have both been shown to participate in the sequential deglycosylation of human glycoconjugates.35,36 It has also been shown that sialic acid can be removed from the Eustachian tube epithelia of chinchillas infected with a NanA-deficient S. pneumoniae strain,37 and that both NanA and NanB are important for infection and sepsis in a mouse model of pneumococcal disease.8 The removal of sialic acids from host glycoconjugates is proposed to expose adherence sites, 6 modify the surfaces of competing bacteria38 and alter the clearance function of host defence proteins.39 In common with many sialidasecontaining bacteria,40 S. pneumoniae can also utilize sialic acid as an energy source,36 as the genome encodes sialic acid transporters, and an N-acetylneuraminate lyase. It would be of interest to discover if Neu2,7-anhydro5Ac is a substrate for the lyase, and hence a source of nutrition. What is the advantage of S. pneumoniae encoding three exported sialidases? NanA contains an anchor motif and remains attached to the bacterial cell surface, has a pH optimum of 7 and is not selective for the sialic acid linakage. NanB, as we have shown, has a substrate preference for α2,3-linked sialic acids, and a previously reported pH optimum of 5.25,41 We would also predict that NanC has properties similar to those of NanB, given its 50% sequence identity and conservation of the tryptophan that confers a selective preference for α2,3-linked sialic acid. Recent studies on the anatomical distribution of sialic acid receptors for the influenza virus has shown that the epithelial cells in the upper respiratory tract of humans possess mainly sialic acid linked to galactose by an α2,6 linkage, but that many cells in the respiratory bronchioles and alveoli possess sialic acid linked α2,3 to galactose.42,43 It may be that NanB, and possibly NanC, provides an advantage for survival of the bacterium in the lungs, whereas NanA would be the dominant sialidase in the upper respiratory tract. Certainly, both NanA and NanB are essential for survival of pneumococci in the nasopharynx in intranasal challenge, and they cannot compensate for one another in single gene knockout experiments.8 Finally, the serendipitous discovery that Ches is an inhibitor of NanB provides the basis for novel drug development against the pneumococcal sialidases. Ches inhibits NanB with a Ki of 0.5 mM, similar to Neu5Ac2en, which inhibits with a Ki of 0.3 mM. Ches also inhibits NanA with a Ki of 0.65 mM. The sulphonic acid group of Ches interacts with all three arginines in the active site, in contrast to the carboxylic acid group of Neu5Ac2en that interacts with only two. In the Ches complex, a water molecule (W4 in Fig. 3) occupies the position of the O4 of Neu5Ac2en. Chemical modification of Ches at its amino group to occupy this water position could provide a ligand with increased binding affinity
Structure of S. Pneumoniae NanB
444
Fig. 7 (legend on next page)
Structure of S. Pneumoniae NanB
445
through both enthalpic and entropic gains. Such modifications based on a Ches framework are currently in progress, and may provide a new class of inhibitors for the pneumococcal, and potentially other sialidases involved in pathogenesis.
Materials and Methods Cloning of NanB The S. pneumoniae nanB gene was supplied in a PGETeasy vector and amplified by polymerase chain reactions (PCR) using the following primers: forward 5′-GTGATTATAGCTAGCATGAATAAAAGAGGTC-3′ reverse 5′-CGATTGGAGGTCTCGAGTTTTGTTAAATC-3′
with the NheI and XhoI restrictions sites underlined. The cleaned PCR product was cloned into the PET23b vector (Novagen) using a Gel Extraction Kit (QIAGEN). Restriction enzyme digestion or colony PCR methods were used to select the positive colonies. All plasmid DNA were extracted with a Mini-Prep Kit (Promega) and stored at −20 °C. The DNA sequence was confirmed by The Sequencing Service, University of Dundee, UK. Expression and purification Recombinant NanB/PET23b (ampicillin-resistant) plasmid was introduced into Escherichia coli BL21 (DE3) expression strain (Novagen) for protein production. Single colonies of the cells containing nanB were picked and put into Luria Bertani (LB) medium with 100 μg/ml ampicillin at 37 °C for inoculation. Isopropyl thio-β-D-galactopyranoside (IPTG) was added (0.5 mM final concentration) to induce the expression of the proteins when the absorbance at 600 nm (A600) of the cultures reached 0.6. The cells continued to be cultured overnight at 25 °C before harvesting by centrifugation at 16,780g for 30 min. The cell pellets were suspended in phosphate-buffered saline (PBS; 20 mM sodium phosphate, 150 mM sodium chloride, pH 7.4) containing 10 mM imidazole and sonicated with five bursts of 30 s each. Protease Inhibitor Cocktail tablets (one tablet per 25 ml extract; Roche Diagnostics) and DNase I (Sigma, final concentration 20 μg/ml) were then added. The crude cell extract was centrifuged at 75,600g for 25 min to remove the cell debris and the supernatant was filtered with a 0.45 μm pore size syringe-driven filter before starting protein purification. The soluble cell extract was first loaded onto a 5 ml HisTrap column (GE Healthcare) equilibrated on a BioCad system (GMI Inc. USA) in PBS containing 10 mM imidazole. The bound protein was eluted with a 30 mM–500 mM gradient of imidazole in PBS. Protein purity was assessed by SDS-PAGE and matrix-assisted laser desorption/ionisation time-of-flight mass spectrometry (MALDI-TOF MS) was used to confirm the protein identity. After overnight dialysis of the collected fractions
Fig. 8. Active site conservation. Residues conserved across NanA, NanB and NanC are coloured red on the NanB surface, with Neu5Ac2en shown with yellow carbons.
in 10 mM Tris-HCl, pH7.5, the protein was run through a 5 ml Q-FF anion-exchange column (GE Healthcare) and eluted with a 0 mM–500 mM gradient of NaCl. Those fractions with relatively high purity protein of expected molecular mass in SDS-PAGE were subjected to sizeexclusion chromatography. A 120 ml Sephacryl S-200 column (GE Healthcare) was used for the last gelfiltration step and high-purity target protein was pooled. The purified NanB was dialysed against 0.01 M sodium phosphate, pH7.0, overnight before concentration and storage. Kinetic characterization of NanB and NanA Sialidase activity was measured by using the fluorogenic substrate 2′-(4-methylumbelliferyl)-α-D-N-acetylneuraminic acid (MUAN; Sigma), which was made as a 10 mM stock in distilled, deionised water and kept at −20 °C. In this assay, appropriately diluted protein samples were assayed in a reaction volume of 1 ml containing 0.1 mM MUAN in 0.1 M sodium acetate pH 5.6. Reactions were incubated at 37 °C for 5 min and then stopped by adding 2ml of 0.25 M glycine buffer, pH 10.5. The relative fluorescence of released methylumbelliferone (MU) was determined at 365 nm for excitation and 450 nm for emission. This method was used also for the kinetic characterization of NanB and NanA. Briefly, all the reactions used 200 ng of NanB, or 35 ng of NanA, various concentrations of MUAN from 10 μM to 100 μM in the presence or in the absence of Ches (1–10 mM) or Neu5Ac2en (0.1–0.8 mM for NanB, 5–20 μM for NanA). The reaction velocity was presented as ΔFluorescence/min, which was determined from the amount of MU released in 5 min.
Fig. 7. Sequence alignment of the three S. pneumoniae sialidases over the region encompassed by the structure of NanB, showing the location of the secondary structural elements. The three domains are shaded red, green and blue. Residues predicted to be involved in sialic acid recognition in the lectin domain are marked + (charged/polar) or o (hydrophobic). Key catalytic residues are marked ⁎. The location of the BNRs are labelled. The figure was produced using ClustalW254 and ESPript.55
Structure of S. Pneumoniae NanB
446 Substrate specificity assay Quantification of N-acetylneuraminic acid released from α2,3, α2,6 or α2,8-linked sialyllactoses (Sigma) was performed with the thiobarbituric acid assay.44 Briefly, after incubating the enzyme with 30 μg of substrates in 100 μl 0.1 M sodium acetate, pH 5.5 at 37 °C for 1 h, the reactions were treated with 0.25 ml of 25 mM H5IO6, 0.125 M H2SO4, pH 1.2 in a 37 °C waterbath for 30 min. Then 0.2 ml of 2% (w/v) NaAsO2, 0.5 M HCl was added to the reactions to remove the excess periodate. The tubes were shaken for 1–2 min until the contents were colourless and then 2 ml of 0.1 M 2-thiobarbituric acid, pH 9 was added and the tube contents boiled for 7.5 min. Afterwards, the cooled reaction was shaken with 3 ml of nbutanol/12 M HCl (19:1, v/v) and centrifuged to separate the layers. The upper butanol layer was removed for the measurement of absorbance at 549 nm. The amount of sialic acid released was calculated as described.45 NMR experiments NanB reactions with α-2,3-sialyllactose, 3SL, or α-2,6sialyllactose, 6SL, were monitored by 1H NMR spectroscopy. In each reaction, 12 μl of 70 μM NanB, 14 μl of 3SL or 6SL (30 mM in 2H2O) and 564 μl of PBS in 2H2O, pH7, were mixed and transferred into a 5 mm NMR tube. Repeats were done in 50 mM sodium acetate in 2H2O at pH 5. For the trans-sialidase assay, a mixture of 25 μl of Nacetyllactosamine (Galβ1-4GlcNAc, 52 mM in 2H2O) and 14 μl of 3SL in 49 μl of PBS in 2H2O were incubated with
NanB for 48 h to assay whether NanB can transfer the sialic acid residues from 3SL to Galβ1-4GlcNAc. After the reactions were incubated at 298 K for 24 h and 48 h, 1H NMR spectra were measured on a 500 MHz Bruker spectrometer at 298 K with 256 scans and a relaxation delay of 2 s. The residual water signal was suppressed by continuous low-power irradiation of the water signal during the relaxation delay. Protein crystallization Results of a pre-crystallization assay kit (Hampton Research) suggested an optimum concentration of purified NanB of 5.3 mg/ml for crystallization trials. All the subsequent crystallization experiments were done at 20 °C by the sitting-drop, vapour-diffusion method. Initially, commercial kits PACT Premier (Molecular Dimensions), Nextal Classic (Qiagen), Index (Hampton Research) and Wizard (Emerald BioSystems) conditions were screened by the high-throughput Rhombix Screen® robot system (Rhombix) for protein crystallization. Those conditions with crystalline materials were selected for crystallization optimization. For manual protein crystallization, each reservoir well of a 96-well protein crystallization tray (Douglas Instruments) was filled with 100 μl of reservoir solution. The crystallization drops were composed of equal amounts of protein solution and reservoir solution (1–2 μl each). After several rounds of optimization, the best conditions for NanB crystallization were: (1) 10% (w/v) polyethylene glycol (PEG) 8 K, 0.15 M NaCl, 0.1 M Ches pH 9.5; and (2) 7% (w/v) PEG 8 K and 0.1 M imidazole
Table 1. X-ray data collection and refinement statistics
A. Data collection Space group X-ray source X-ray wavelength (Å) Resolution (Å) Unit cell dimensions a (Å) b (Å) c (Å) No. observations Redundancy Completeness (%) Rmerge I/σI B. Data refinement No. reflections No. atoms Protein Ligand Water Average B-factor (Å2) Protein Ligands Water R-factor Rfree r.m.s.d from ideal Bond lengths (Å) Bond angles (°) wwPDB code
NanB apo
NanB-DANA complex
NanB-Ches complex
P212121 ID14-1 0.934 30–2.30 (2.38–2.30)
P212121 In-house 1.542 30–2.40 (2.49–2.40)
P212121 In-house 1.542 30–1.70 (1.76–1.70)
76.6 82.7 116.7 211,813 6.1 (5.8) 98.03 (95.0) 0.109 (0.46) 13.9 (4.8)
76.6 82.7 117.4 105,254 5.1 (4.8) 100 (99.6) 0.138 (0.35) 6.3 (2.8)
76.6 82.7 116.7 234,069 3.1 (1.6) 92.1 (62.7) 0.083 (0.36) 8.9 (2.2)
32,865
28,251
71,846
5189 – 451
5189 20 463
5189 26 797
25.2 – 30.0 0.176 0.232 0.012
22.1 20.1 21.8 0.196 0.278 0.012
17.1 20.9 28.1 0.194 0.229 0.008
1.298 2VW0
1.504 2VW1
1.099 2VW2
Numbers in parentheses refer to the highest resolution shell. Rmerge = ∑hkl∑i|Ihkl,i−〈Ihkl〉|/∑hkl〈Ihkl〉. Rcryst and Rfree = (∑||Fo|−|Fc||)/(∑|Fo|).
Structure of S. Pneumoniae NanB pH 8.0. Crystals appeared in about three days and reached maximum size after two weeks. The Neu5Ac2en complex structure was obtained by soaking a NanB crystal (crystallized in 7% (w/v) PEG 8 K, 0.1 M imidazole, pH 8.0) in 100 mM Neu5Ac2en for 15 min. X-ray diffraction data collection and processing Crystals were cryoprotected by transfer for a few minutes to a solution of the crystallization buffer with 10% (v/v) glycerol added before data collection at 100 K. Data were collected in-house (Rigaku-MSC Micromax-007 X-ray generator and R-AXIS detectors) or at the ESRF, Grenoble. MOSFLM46 was used for data integration, the space group is P212121 and data collection statistics are given in Table 1. Structure determination and refinement The leech trans-sialidase structure (PDB code 1SLI) was used to solve the NanB structure by molecular replacement using AMoRe in the CCP4 computer suite.47 The ARPwARP program was used to build the initial structure using the 1.7 Å Ches complex data.48 The protein models were rebuilt with O49 and Coot.50 After further refinement with REFMAC5,51 the structures were inspected and validated with Coot and MolProbity.52 Refinement statistics are summarized in Table 1. Protein Data Bank accession numbers Coordinates and structure factors have been deposited in the Protein Data Bank with accession numbers 2VW0, 2WV1 and 2VW2 for the apo, Neu5Ac2en and Ches complexes, respectively.
Acknowledgements G.X. was supported by a Scottish Government International Scholarship and by Biocryst Pharmaceuticals Inc, Birmingham, Alabama. Resources of the St Andrews-based Scottish Structural Proteomics Facility, funded by the Scottish Funding Council, the Biotechnology and Biological Sciences Research Council (BBSRC) and the University of St Andrews, were used in this project.
Supplementary Data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j. jmb.2008.09.032
References 1. Luotonen, J., Herva, E., Karma, P., Timonen, M., Leinonen, M. & Makela, P. H. (1981). The bacteriology of acute otitis media in children with special reference to Streptococcus pneumoniae as studied by bacteriological and antigen detection methods. Scand. J. Infect. Dis. 13, 177–183.
447 2. Jedrzejas, M. J. (2001). Pneumococcal virulence factors: structure and function. Microbiol. Mol. Biol. Rev. 65, 187–207. 3. Mitchell, T. J. (2000). Virulence factors and the pathogenesis of disease caused by Streptococcus pneumoniae. Res. Microbiol. 151, 413–419. 4. Pettigrew, M. M., Fennie, K. P., York, M. P., Daniels, J. & Ghaffar, F. (2006). Variation in the presence of neuraminidase genes among Streptococcus pneumoniae isolates with identical sequence types. Infect. Immun. 74, 3360–3365. 5. Tong, H. H., Blue, L. E., James, M. A. & DeMaria, T. F. (2000). Evaluation of the virulence of a Streptococcus pneumoniae neuraminidase-deficient mutant in nasopharyngeal colonization and development of otitis media in the chinchilla model. Infect. Immun. 68, 921–924. 6. Tong, H. H., Liu, X., Chen, Y., James, M. & Demaria, T. (2002). Effect of neuraminidase on receptor-mediated adherence of Streptococcus pneumoniae to chinchilla tracheal epithelium. Acta Otolaryngol. 122, 413–419. 7. McCullers, J. A. & Bartmess, K. (2003). Role of Neuraminidase in Lethal Synergism between Influenza Virus and Streptococcus pneumoniae. J. Infect. Dis. 187, 1000–1009. 8. Manco, S., Hernon, F., Yesilkaya, H., Paton, J. C., Andrew, P. W. & Kadioglu, A. (2006). Pneumococcal neuraminidases A and B both have essential roles during infection of the respiratory tract and sepsis. Infect. Immun. 74, 4014–4020. 9. Simell, B., Jaakkola, T., Lahdenkari, M., Briles, D., Hollingshead, S., Kilpi, T. M. & Kayhty, H. (2006). Serum antibodies to pneumococcal neuraminidase NanA in relation to pneumococcal carriage and acute otitis media. Clin. Vaccine Immunol. 13, 1177–1179. 10. Crennell, S. J., Garman, E. F., Laver, W. G., Vimr, E. R. & Taylor, G. L. (1993). Crystal structure of a bacterial sialidase (from Salmonella typhimurium LT2) shows the same fold as an influenza virus neuraminidase. Proc. Natl Acad. Sci. USA, 90, 9852–9856. 11. Crennell, S. J., Garman, E., Laver, G., Vimr, E. R. & Taylor, G. (1994). Crystal structure of Vibrio cholerae neuraminidase reveals dual lectin-like domains in addition to the catalytic domain. Structure, 2, 535–544. 12. Gaskell, A., Crennell, S. J. & Taylor, G. (1995). The three domains of a bacterial sialidase: a beta-propeller, an immunoglobulin module and a galactose-binding jelly-roll. Structure, 3, 1197–1205. 13. Varghese, J. N., Laver, W. G. & Colman, P. M. (1983). Structure of the influenza virus glycoprotein antigen neuraminidase at 2.9 Å resolution. Nature, 303, 35–40. 14. Burmeister, W. P., Henrissat, B., Bosso, C., Cusack, C. & Ruigrok, R. W. H. (1993). Influenza B virus neuraminidase can synthesise its own inhibitor. Structure, 1, 19–26. 15. Crennell, S. J., Takimoto, T., Portner, A. & Taylor, G. (2000). Crystal structure of the multifunctional paramyxovirus hemagglutinin-neuraminidase. Nature Struct. Biol. 7, 1068–1074. 16. Buschiazzo, A., Tavares, G. A., Campetella, O., Spinelli, S., Cremona, M. L., Paris, G. et al. (2000). Structural basis of sialyltransferase activity in trypanosomal sialidases. EMBO J. 19, 16–24. 17. Buschiazzo, A., Amaya, M. F., Cremona, M. L., Frasch, A. C. & Alzari, P. M. (2002). The crystal structure and mode of action of trans-sialidase, a key enzyme in Trypanosoma cruzi pathogenesis. Mol. Cell, 10, 757–768. 18. Lou, Y., Li, S.-C., Chou, M.-Y., Li, Y.-T. & Lou, M.
448
19.
20. 21.
22.
23. 24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
(1998). The crystal structure of an intramolecular trans-sialidase with a NeuAcalpha2 → 3Gal specificity. Structure, (Camb), 6, 521–530. Chavas, L. M., Tringali, C., Fusi, P., Venerando, B., Tettamanti, G., Kato, R., Monti, E. & Wakatsuki, S. (2004). Crystal structure of the human cytosolic Sialidase Neu2: Evidence for the dynamic nature of substrate recognition. J. Biol. Chem, 280, 469–475. Taylor, G. (1996). Sialidases: structures, biological significance and therapeutic potential. Curr. Opin. Struct. Biol. 6, 830–837. Moustafa, I., Connaris, H., Taylor, M., Zaitsev, V., Wilson, J. C., Kiefel, M. J., von Itzstein, M. & Taylor, G. (2004). Sialic acid recognition by Vibrio cholerae neuraminidase. J. Biol. Chem. 279, 40819–40826. Newstead, S. L., Watson, J. N., Bennet, A. J. & Taylor, G. (2005). Galactose recognition by the carbohydratebinding module of a bacterial sialidase. Acta Crystallogr. D, 61, 1483–1491. Thobhani, S., Ember, B., Siriwardena, A. & Boons, G. J. (2003). Multivalency and the mode of action of bacterial sialidases. J. Am. Chem. Soc. 125, 7154–7155. Camara, M., Boulnois, G. J., Andrew, P. W. & Mitchell, T. J. (1994). A neuraminidase from Streptococcus pneumoniae has the features of a surface protein. Infect. Immun. 62, 3688–3695. Berry, A. M., Lock, R. A. & Paton, J. C. (1996). Cloning and characterization of nanB, a second Streptococcus pneumoniae neuraminidase gene, and purification of the NanB enzyme from recombinant Escherichia coli. J. Bacteriol. 178, 4854–4860. Boraston, A. B., Ficko-Blean, E. & Healey, M. (2007). Carbohydrate recognition by a large sialidase toxin from Clostridium perfringens. Biochemistry, 46, 11352–11360. Li, Y. T., Nakagawa, H., Ross, S. A., Hansson, G. C. & Li, S. C. (1990). A novel sialidase which releases 2,7anhydro-alpha-N-acetylneuraminic acid from sialoglycoconjugates. J. Biol. Chem. 265, 21629–21633. Chou, M. Y., Li, S. C., Kiso, M., Hasegawa, A. & Li, Y. T. (1994). Purification and characterization of sialidase L, a NeuAc alpha 2 → 3Gal-specific sialidase. J. Biol. Chem. 269, 18821–18826. Luo, Y., Li, S. C., Chou, M. Y., Li, Y. T. & Luo, M. (1998). The crystal structure of an intramolecular transsialidase with a NeuAc alpha2 → 3Gal specificity. Structure, 6, 521–530. Luo, Y., Li, S. C., Li, Y. T. & Luo, M. (1999). The 1.8 A structures of leech intramolecular trans-sialidase complexes: evidence of its enzymatic mechanism. J. Mol. Biol. 285, 323–332. Hoyer, L. L., Roggentin, P., Schauer, R. & Vimr, E. R. (1991). Purification and properties of cloned Salmonella typhimurium LT2 sialidase with virus-typical kinetic preference for sialyl alpha 2–3 linkages. J. Biochem. 110, 462–467. Amaya, M. F., Buschiazzo, A., Nguyen, T. & Alzari, P. M. (2003). The high resolution structures of free and inhibitor-bound Trypanosoma rangeli sialidase and its comparison with T. cruzi trans-sialidase. J. Mol. Biol. 325, 773–784. Amaya, M. F., Watts, A. G., Damager, I., Wehenkel, A., Nguyen, T., Buschiazzo, A. et al. (2004). Structural insights into the catalytic mechanism of Trypanosoma cruzi trans-sialidase. Structure, 12, 775–784. Lou, Y., Li, S. C., Li, Y. T. & Lou, M. (1998). The 1.8 Å structures of Leech intramolecular trans-sialidase complexes: evidence of its enzymatic mechanism. J. Mol. Biol. 285, 323–332.
Structure of S. Pneumoniae NanB 35. King, S. J., Hippe, K. R. & Weiser, J. N. (2006). Deglycosylation of human glycoconjugates by the sequential activities of exoglycosidases expressed by Streptococcus pneumoniae. Mol. Microbiol. 59, 961–974. 36. Burnaugh, A. M., Frantz, L. J. & King, S. J. (2008). Growth of Streptococcus pneumoniae on human glycoconjugates is dependent upon the sequential activity of bacterial exoglycosidases. J. Bacteriol. 190, 221–230. 37. Tong, H. H., James, M., Grants, I., Liu, X., Shi, G. & DeMaria, T. F. (2001). Comparison of structural changes of cell surface carbohydrates in the eustachian tube epithelium of chinchillas infected with a Streptococcus pneumoniae neuraminidase-deficient mutant or its isogenic parent strain. Microb. Pathog. 31, 309–317. 38. Shakhnovich, E. A., King, S. J. & Weiser, J. N. (2002). Neuraminidase expressed by Streptococcus pneumoniae desialylates the lipopolysaccharide of Neisseria meningitidis and Haemophilus influenzae: a paradigm for interbacterial competition among pathogens of the human respiratory tract. Infect. Immun. 70, 7161–7164. 39. O'Toole, R. D., Goode, L. & Howe, C. (1971). Neuraminidase activity in bacterial meningitis. J. Clin. Invest. 50, 979–985. 40. Corfield, T. (1992). Bacterial sialidases — roles in pathogenicity and nutrition. Glycobiology, 2, 509–521. 41. Berry, A. M., Paton, J. C., Glare, E. M., Hansman, D. & Catcheside, D. E. (1988). Cloning and expression of the pneumococcal neuraminidase gene in Escherichia coli. Gene, 71, 299–305. 42. Shinya, K., Ebina, M., Yamada, S., Ono, M., Kasai, N. & Kawaoka, Y. (2006). Avian flu: influenza virus receptors in the human airway. Nature, 440, 435–436. 43. Chandrasekaran, A., Srinivasan, A., Raman, R., Viswanathan, K., Raguram, S., Tumpey, T. M. et al. (2008). Glycan topology determines human adaptation of avian H5N1 virus hemagglutinin. Nat. Biotechnol. 26, 107–113. 44. Aminoff, D. (1961). Methods for the quantitative estimation of N-acetylneuraminic acid and their application to hydrolysates of sialomucoids. Biochem. J. 81, 384–392. 45. Warren, L. (1959). The thiobarbituric acid assay of sialic acids. J. Biol. Chem. 234, 1971–1975. 46. Leslie, A. G. (2006). The integration of macromolecular diffraction data. Acta Crystallogr. D, 62, 48–57. 47. Collaborative Computational Project, Number 4. (1994). The CCP4 suite: programs for protein crystallography. Acta Crystallogr. D, 50, 760–763. 48. Perrakis, A., Morris, R. & Lamzin, V. S. (1999). Automated protein model building combined with iterative structure refinement. Nature Struct. Biol. 6, 458–463. 49. Jones, T. A., Zou, J. Y., Cowan, S. W. & Kjeldgaard, M. (1991). Improved methods for building protein models in electron density maps and the location of errors in these models. Acta Crystallogr. A, 47, 110–119. 50. Emsley, P. & Cowtan, K. (2004). Coot: model-building tools for molecular graphics. Acta Crystallogr. D, 60, 2126–2132. 51. Murshudov, G. N., Vagin, A. A. & Dodson, E. J. (1997). Refinement of macromolecular structures by the maximum-likelihood method. Acta Crystallogr. D, 53, 240–255. 52. Lovell, S. C., Davis, I. W., Arendall, W. B., de Bakker, P. I., Word, J. M., Prisant, M. G. et al. (2003). Structure
Structure of S. Pneumoniae NanB validation by Cα geometry: ϕ, ψ and Cβ deviation. Proteins: Struct. Funct. Genet. 50, 437–450. 53. Baker, N., Sept, D., Joseph, S., Holst, M. & McCammon, J. (2001). Electrostatics of nanosystems: Application to microtubules and the ribosome. Proc. Natl Acad. Sci. USA, 98, 10037–10041. 54. Larkin, M. A., Blackshields, G., Brown, N. P., Chenna,
449 R., McGettigan, P. A., McWilliam, H. et al. (2007). Clustal W and Clustal X version 2.0. Bioinformatics, 23, 2947–2948. 55. Gouet, P., Robert, X. & Courcelle, E. (2003). ESPript/ ENDscript: Extracting and rendering sequence and 3D information from atomic structures of proteins. Nucleic Acids Res. 31, 3320–3323.