- Email: [email protected]
Import and export of bacterial protein toxins
G Model
ARTICLE IN PRESS
IJMM 50908 1–5
International Journal of Medical Microbiology xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
International Journal of Medical Microbiology journal homepage: www.elsevier.com/locate/ijmm
1
Mini Review
2
Import and export of bacterial protein toxins Q1
3
Volkmar Braun ∗ , Stephanie Helbig, Silke I. Patzer, Avijit Pramanik, Christin Römer Max Planck Institute for Developmental Biology, Department of Protein Evolution, Spemannstrasse 35, 72076 Tübingen, Germany
4 5
6 15
a r t i c l e
i n f o
a b s t r a c t
7
14
Keywords: Colicin M Pesticin Hemolysin Import Export Activity
16
Introduction
8 9 10 11 12 13
The paper provides a short overview of three investigated bacterial protein toxins, colicin M (Cma) of Escherichia coli, pesticin (Pst) of Yersinia pestis and hemolysin (ShlAB) of Serratia marcescens. Cma and Pst are exceptional among colicins in that they kill bacteria by degrading the murein (peptidoglycan). Both are released into the medium and bind to specific receptor proteins in the outer membrane of sensitive E. coli cells. Subsequently they are translocated into the periplasm by an energy-consuming process using the proton motive force. For transmembrane translocation the colicins unfold and refold in the periplasm. In the case of Cma the FkpA peptidyl prolyl cis-trans isomerase/chaperone is required. ShlA is secreted and activated through ShlB in the outer membrane by a type Vb secretion mechanism. © 2014 Published by Elsevier GmbH.
Q2 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41
Bacteria secrete selected proteins by diverse, yet distinct mechanisms. For gram-negative bacteria at least seven protein secretion types and various subtypes are known. In addition, proteins exist which are secreted without a specific secretion mechanism. These are colicins which are released by nearly half of Escherichia coli natural isolates and kill competing sensitive cells. Despite co-synthesis of immunity proteins which confer resistance to the cognate colicins, a small proportion of cells in a population lyses when colicins are overproduced under stress conditions. In contrast to unspecific export highly specific and sophisticated colicin import systems exist in sensitive cells. According to their mode of action colicins enter the periplasm, the cytoplasmic membrane, or the cytoplasm of sensitive cells. For all colicins import starts with binding to specific receptor proteins exposed at the cell surface. Sensitivity of cells is largely determined by those receptors (Braun et al., 2002; Jakes and Cramer, 2012). Here, two colicins from E. coli and Y. pestis and the hemolysin of S. marcescens were investigated by us and will be discussed with respect to their cellular export and import. Colicin M (Cma) and pesticin (Pst) were studied because only preliminary data existed and they differ from other colicins in that they primarily cause lysis of cells. All other colicins either degrade DNA or RNA in the cytoplasm or form pores in the cytoplasmic membrane resulting in collapse of the membrane potential and later cell lysis. Cma inhibits murein biosynthesis by cleavage of the ester bond in the
∗ Corresponding author. E-mail address: [email protected] (V. Braun).
murein precursor resulting in undecaprenol and 1-pyrophosphoMurNAc-(pentapeptide)-GlcNAc (Harkness and Braun, 1989; El Ghachi et al., 2006). These degradation products cannot be used anymore for murein biosynthesis. Pst cleaves the glycan chain of murein between C1 of MurNAc and C4 of GlcNAc (lysozyme activity; Vollmer et al., 1997). Plasmid-encoded Cma is made by E. coli and kills E. coli cells. Plasmid-encoded Pst is made by Yersinia pestis and kills Yersinia strains. It displays all characteristics of colicins and is, therefore, listed here as a colicin. Hemolysin from S. marcescens has been included in our investigation (Braun et al., 1987) because it was at that time one of the few proteins that were secreted by Enterobacteriaceae and had not been investigated. The S. marcescens hemolysin structure and secretion mechanism turned out to be completely different from the well-studied E. coli hemolysin. Domain structure of the E. coli colicin Cma Cma consists of three domains which are characteristic of colicins. The central domain serves to bind Cma to the outer membrane receptor protein FhuA that concomitantly serves as a receptor for several phages and as a transporter for the iron chelator ferrichrome as well as for antibiotics derived from ferrichrome. The N-terminal domain is required for translocation across the outer membrane into the periplasm, and the C-terminal domain encodes the phosphatase activity (Fig. 1). Deletion of the hydrophobic ␣1 helix close to the N-terminus (Fig. 1; gray) abolishes binding of Cma to FhuA. However, phosphatase activity is retained as demonstrated in an assay that bypasses the FhuA function (Helbig and Braun, 2011). Mutations in ␣1 strongly reduce killing of cells. In contrast to wild-type Cma, these mutants do not prevent killing of
http://dx.doi.org/10.1016/j.ijmm.2014.12.006 1438-4221/© 2014 Published by Elsevier GmbH.
Please cite this article in press as: Braun, V., et al., Import and export of bacterial protein toxins. Int. J. Med. Microbiol. (2014), http://dx.doi.org/10.1016/j.ijmm.2014.12.006
42 43 44 45 46 47 48 49 50 51 52 53 54 55 56
57
58 59 60 61 62 63 64 65 66 67 68 69 70
G Model IJMM 50908 1–5 2
ARTICLE IN PRESS V. Braun et al. / International Journal of Medical Microbiology xxx (2014) xxx–xxx
Import of Cma
Fig. 1. Crystal structure (PDB 2XMX) of colicin M (Zeth et al., 2008) in which functionally important residues are indicated (Helbig and Braun, 2011). N-terminal translocation domain (yellow), central receptor binding domain (blue), C-terminal phosphatase domain (magenta),N-terminal end (N), C-terminal end (C) are indicated. Residues identified as important for Cma activity are shown as sticks. ␣1 indicates the region involved in receptor binding. ␣2 contains a strongly polar sequence, 59-EDYIKKH-65, but only the alanine replacement in E59 reduces Cma activity to 10%.
71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94
cells by the antibiotic albomycin, a ferrichrome derivative, which uses FhuA to enter cells. The mutant Cma proteins do not bind to FhuA which indicates that the ␣1 helix mediates binding of Cma to FhuA. Sequence analyses of bacterial genomes reveal cma orthologs in strains of Pectobacteria, Pseudomonas and Burkholderia. The sequences are highly diverged in the receptor binding and translocation domains but similar in the phosphatase domain. The distinct receptor binding and translocation domains reflect individual import proteins in these bacteria. The conserved Cterminal region reflects the common killing mechanism. The three-domain structure also supports our previous conclusion that colicins in different bacteria evolved by horizontal gene transfer via plasmids and an exchange of gene fragments which encode functional domains (Braun et al., 2002; Roos et al., 1989). The phosphatase domain evolved from a single ancestor gene that was fused to distinct receptor binding and translocation genes. Residues 226 DKYDFNASTHR236 are conserved, hydrophilic and exposed at the Cma surface (Fig. 1). Amino acid replacements in this region render Cma inactive. Particularly noteworthy are mutations D226E and D226N which completely inactivate Cma. This region is part of the active center of Cma, and D226 most likely is directly involved in hydrolysis (Pilsl et al., 1993; Helbig and Braun, 2011).
Five chromosomally encoded genes are required for Cma killing of sensitive E. coli cells, fhuA, tonB, exbB, exbD, and fkpA. Mutations in any of these genes render cells resistant to high doses of Cma (dilution titers of 105 ). The outer membrane receptor FhuA is functionally coupled to the electrochemical gradient across the cytoplasmic membrane via the proteins TonB, ExbB and ExbD. These proteins are anchored in the cytoplasmic membrane and have distinct periplasmic domains. TonB physically interacts with FhuA, and ExbB and ExbD form a complex for which a stoichiometry of 6:1 was determined (Pramanik et al., 2011). It is hypothesized that FhuA changes its conformation in response to the membrane potential. Such conformational changes are suggested to result in release of high-affinity ligands (Cma, ferrichrome, albomycin, phages T1, T5, and 80) from FhuA into the periplasm. FhuA forms a  barrel composed of 16  strands which is closed by an N-proximal globular segment termed plug or cork. The plug must move to allow ligand passage through FhuA. Currently it is unknown whether FhuA only acts as a primary adsorption site for Cma or also as the import route. Cma uptake is even more complex than the uptake of the other FhuA ligands as TonB must not only interact with FhuA but also with Cma (Pilsl et al., 1993). Cma contains a typical TonB box at the N-terminus. Point mutations in the TonB box abolish Cma uptake. Specific point mutations in TonB are capable to suppress point mutations in the Cma TonB box. This is taken as an indication for a direct interaction between the TonB box of Cma and the periplasmic domain of TonB. Outer membrane receptor proteins and all colicins which require TonB for activity contain an N-terminal consensus sequence, i.e. a TonB box. Suppressor analysis, cysteine cross-linking and crystal structures of TonB fragments bound to receptor proteins demonstrate a physical interaction with TonB (Braun, 2014). Essential role of the periplasmic chaperone FkpA for Cma activity FkpA is a peptidyl prolyl cis-trans isomerase with chaperone activity for which only very recently a control of outer membrane biogenesis under heat shock conditions has been assigned (Ge et al., 2014). We discovered that FkpA is essential for killing cells by imported Cma (Hullmann et al., 2008). Likewise, FkpA is essential for Cma secretion into the periplasm provided an artificial signal sequence has been attached to Cma (Helbig et al., 2011). For both, import and export Cma must unfold for crossing the outer and the cytoplasmic membrane. FkpA is indispensable for refolding and accelerates refolding of denatured Cma in vitro. Upon binding to FhuA Cma becomes sensitive to trypsin digestion. A mutation analysis of the fifteen proline residues in Cma identifies proline P176 as the most likely one which is cis-trans isomerized. Indeed, F175–P176 showed in vitro the fastest isomerization rate assayed with synthetic peptides derived from the Cma sequence. P176 is exposed on the surface of the Cma crystal structure which was determined for wild-type Cma and the inactive P176A mutant (Helbig et al., 2011). Of six inactive FkpA point mutants all are located in the isomerase domain. This data identify FkpA as essential for Cma in vivo activity and suggest that Cma is unfolded during import and refolded by FkpA involving cis-trans isomerization of the F175–P176 bond. Further, the data demonstrate an enzymatic activity for FkpA by which a proline bond is isomerized, and demonstrate for the first time an indispensable function of FkpA as a periplasmic chaperone in protein import. Specific resistance to Cma by CbrA Among Cma tolerant E. coli mutants (tolM), fkpA was identified (tolerance defined as Cma binding and, presumably, uptake, but no
Please cite this article in press as: Braun, V., et al., Import and export of bacterial protein toxins. Int. J. Med. Microbiol. (2014), http://dx.doi.org/10.1016/j.ijmm.2014.12.006
95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126
127 128
129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153
154
155 156
G Model IJMM 50908 1–5
ARTICLE IN PRESS V. Braun et al. / International Journal of Medical Microbiology xxx (2014) xxx–xxx
3
zones on plates seeded with Pst sensitive E. coli that carry the Pst receptor of Y. pestis, termed FyuA. The crystal structure of PstRT -T4L reveals a marked difference to Pst as the two domains are connected by a long ␣-helix in contrast to the short unstructured linker in Pst. Although T4L is structurally very similar to the Pst activity domain, PstRT -T4L is not inhibited by the Pst immunity protein Pim. This indicates that Pim is specific for the Pst activity domain. In contrast to the T4 lysozyme, the lysozyme variant from phage T7 when fused to PstRT is not toxic to cells. This may be due to either failure of uptake into the periplasm or misfolding. PstRT mediated import shows specificity. The activity domain of Pst must unfold to enter the periplasm Fig. 2. Crystal structure (4AQN) of pesticin (Patzer et al., 2012). The N-terminal domain (NT) including the TonB box (TBB) is not visible. The translocation domain (T, red), the receptor binding domain (R, blue) and the activity domain (A, orange) are indicated. CT: C-terminus. See text for details.
173
killing). An additional gene locus was found which upon overexpression rendered cells tolerant to Cma. Unexpectedly, the gene responsible for tolerance was cbrA (Helbig et al., 2012) of which very little is known. The extent of tolerance depends on the level of CbrA expression which is controlled by the transcriptional activator CreB. Expression is particularly high in minimal media which also induce cma transcription. The basal level of CbrA synthesis already confers an 18-fold higher survival rate of Cma-treated cells compared to E. coli devoid of cbrA. Upon high cbrA expression the survival rate is increased to over 1000-fold. Since CbrA is expressed under similar conditions as Cma, CbrA confers an appreciable Cma resistance to Cma producing and Cma sensitive cells. Sequence analysis and biochemical studies assign CbrA to FAD-dependent oxidoreductases. FAD is noncovalently bound to CbrA. We found that CbrA renders cells resistant to osmotic shock suggesting that changes in the membrane structure may cause Cma resistance. The mechanism of CbrA function remains enigmatic.
174
Domain structure of pesticin (Pst)
157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172
175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191
192 193
194 195 196 197 198
Most Y. pestis strains possess a small plasmid (9.5 kb) that encodes pesticin (Pst) together with the adjacent immunity gene pim. In the producer cell the immunity protein is localized in the periplasm where it inactivates Pst and thus prevents killing (Pilsl et al., 1996). Murein hydrolysis does not appear to be the only activity of Pst since cells are not immediately lysed but converted to spheroplasts without osmoprotection. Lysis is slow but killing is fast. Murein hydrolysis by Pst in vitro is also slow (Vollmer et al., 1997). The crystal structure of Pst (Patzer et al., 2012) reveals two distinct structural domains connected by a short peptide linker (Fig. 2). The N-terminal domain represents the receptor binding and translocation domain, the C-terminal domain the activity domain. The structure of the activity domain is similar to phage lysozymes, in particular to that of the T4 phage, albeit sequence identity is only 13%. The near identity of the activity domain fold is particularly noteworthy since activity domains of the other colicins display no overall structural similarity to known enzymes. T4 lysozyme is carried into the periplasm by the receptor and translocation domain of Pst Since the Pst activity domain resembles T4 lysozyme (T4L) it was replaced by T4L to test whether T4L is translocated into the periplasm via Pst receptor-binding and translocation domains (PstRT ) (Patzer et al., 2012). PstRT -T4L kills cells but a 6-fold higher concentration than wild-type Pst is required to yield the same lysis
To test whether PstRT -T4L must unfold for transport cysteine residues were introduced in T4L at sites which should allow formation of disulfide bridges (Patzer et al., 2012). The PstRT -T4L chimera was chosen because it efficiently hydrolyzes murein in vitro in contrast to Pst. Three constructs with PstRT -T4L cross-linked by disulfide bridges were inactive in vivo, yet regained activity upon reduction of the disulfide bonds as they killed cells. The disulfide derivatives of Pst-T4L are active in vitro which indicates that the disulfide linkages do not inactivate the lysozyme function. Rather they prevent unfolding and thus inhibit uptake. To examine whether the activity domain must be specifically translocated unfolded across the outer membrane to be able to fold into an active enzyme within the periplasm, Pst and the activity domain were fused to a MalE fragment that contains a signal sequence for secretion from the cytoplasm into the periplasm. Expression was regulated by an arabinose promoter. Upon synthesis of Mal’-Pst and Mal’-A (A, activity domain) cells lyse. This suggests that the activity domain does not require the RT domains of Pst to fold into an active domain and that there are no specific membrane requirements for Pst translocation and refolding. The active site of Pst was deduced from the crystal structure in comparison with T4L. Amino acid replacements at selected sites reveal that the active site residues are similar but not identical to T4L (Patzer et al., 2012). Secretion of the hemolysin of Serratia marcescens, a model system for type Vb secretion S. marcescens is involved in urinary tract infections, bacteremia, endocarditis, keratitis, arthritis, and meningitis. One of its virulence factors is a hemolysin/cytolysin that lyses erythrocytes and elicits inflammatory response in leukocytes (Hertle et al., 1999; König et al., 1987; Lin et al., 2010; Marre et al., 1989). Hemolysin was the first example of a by now large group of proteins which are secreted by the type Vb or the two-partner secretion system (TPS) (Braun et al., 1987; Poole et al., 1988; Schiebel et al., 1989; Jac¸ob-Dubuisson et al., 2013). The hemolysin is determined by two genes, shlA and shlB, arranged in tandem; shlA encodes hemolysin, and shlB encodes the outer membrane protein responsible for ShlA secretion. Both proteins have N-terminal signal sequences for transport across the cytoplasmic membrane by the Sec system. ShlB not only secretes ShlA but it also activates it; in a shlB deletion mutant, ShlA remains in the periplasm in an inactive non-hemolytic form (Poole et al., 1988; Schiebel and Braun, 1989; Schiebel et al., 1989). Activation can be uncoupled from secretion by in vitro incubation of isolated periplasmic ShlA with purified ShlB resulting in hemolytic ShlA (Hertle et al., 1997; Ondraczek et al., 1992). A ShlA N-terminal fragment of 242 residues (ShlA consists of 1578 residues) activates inactive periplasmic ShlA to hemolytic
Please cite this article in press as: Braun, V., et al., Import and export of bacterial protein toxins. Int. J. Med. Microbiol. (2014), http://dx.doi.org/10.1016/j.ijmm.2014.12.006
199 200 201 202 203 204 205 206 207 208 209
210 211
212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235
236 237
238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259
G Model IJMM 50908 1–5 4
ARTICLE IN PRESS V. Braun et al. / International Journal of Medical Microbiology xxx (2014) xxx–xxx
quantitative assay. The predicted structure of ShlB (Fig. 3) was used to design deletion mutants of H1, L6, and the two POTRA domains to study their effects on secretion and activation. Deletion of H1 does not affect secretion of hemolytic ShlA. H1 has no essential function in ShlB assembly, secretion and activation of ShlA (Pramanik et al., 2014). Deletion of L6 diminishes, but does not fully abolish secretion, but completely compromises ShlA activation. The P1 and P2 single mutants secrete reduced amounts of ShlA of which only ShlA of the P1 mutant shows some hemolytic activity. Thus P1 and P2 mutants discriminate between secretion and activation. H1− , P1− , P2− triple deletion mutants secrete no ShlA. Random mutagenesis yields inactive ShlB derivatives mutated in L6 which supports the importance of the conserved L6 among type Vb transporters (Pramanik et al., 2014; Yang and Braun, 2000). ShlA is secreted and folds into a hemolytic structure through its N-terminus which is secreted first. Secretion of non-hemolytic ShlA is inefficient. Folding into the active conformation probably drives thermodynamically secretion of ShlA through ShlB.
Acknowledgments
Fig. 3. Model of ShlB based on the crystal structure of FhaC (Clantin et al., 2007; Pramanik et al., 2014).
260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298
ShlA provided ShlA242 is secreted by ShlB (Ondraczek et al., 1992; Schönherr et al., 1993). ShlA242 purified from the periplasm of a shlB mutant cannot activate periplasmic ShlA. ShlA242 mimics secretion and activation of ShlA by ShlB but is not hemolytic since the C-terminal region of ShlA is required for pore formation in erythrocyte membranes (Schiebel and Braun, 1989; Schiebel et al., 1989; Schönherr et al., 1993). ShlA and ShlA242 change conformations when they are secreted and activated by ShlB. Secreted ShlA and ShlA242 are largely trypsin resistant. In contrast, cytoplasmic and periplasmic ShlA and ShlA242 are cleaved by trypsin to small fragments (Walker et al., 2004; Yang and Braun, 2000). ShlA242 isolated from the periplasm, the cytoplasm (ShlA242 without a signal sequence), and the culture supernatant of a shlB mutant that secretes but does activate ShlA242 have identical electrophoretic mobilities in native polyacrylamide gels. In contrast, secreted and activated ShlA242 has a much higher mobility. In addition, the CD spectra reveal structural differences (Schiebel et al., 1989; Walker et al., 2004). The data indicate that inactive ShlA and inactive ShlA242 are less structurally constrained than the hemolytic isoforms. Furthermore activation most likely does not involve a chemical modification through interaction with ShlB as activation does not change the molecular weights (Schiebel et al., 1989; Walker et al., 2004). The type Vb secretion system belongs to a transporter superfamily of gram-negative bacteria, mitochondria, and chloroplasts for which the Bam complex of E. coli serves as a model system (Ricci and Silhavy, 2012). Swiss-Model Repository (Kiefer et al., 2009) predicts a ShlB structure similar to that of FhaC, the outer membrane exporter of the Bordetella pertussis filamentous hemagglutinin which consists of an ␣-helix (H1) and a loop (L6), both inside a 16-stranded -barrel, and two polypeptide-transportassociated (POTRA) domains in the periplasm (Fig. 3). The TPS system is simpler than the Bam insertion and secretion machinery. The latter has four additional proteins, BamB–E, and up to seven POTRA domains. Hence, the TPS system is particularly suitable to study the mechanism of protein export across the outer membrane. The soluble N-terminal fragment of ShlA242 contains an export signal, the conformation is susceptible to activation, probably by refolding, which enable hemolysis which is testable by a simple
We thank Andrei Lupas for the generous hospitality in his department and Joachim Schultz and Klaus Hantke for critically reading the manuscript. This work was financed by the Max Planck Q3 Society, the German Science Foundation (BR330/25-1, SFB766) and the Fonds der Chemischen Industrie. Space limitation did not allow citing all of the relevant literature.
References Braun, V., 2014. Energy-coupled transport across the outer membrane of gramnegative bacteria. In: Remaut, H., Fronzes, R. (Eds.), Bacterial Membranes. Structure and Molecular Biology. Caister Academic Press, Norfolk, pp. 249–291. Braun, V., Neuss, B., Ruan, Y., Schiebel, E., Schöffler, H., Jander, G., 1987. Identification of the Serratia marcescens hemolysin determinant by cloning into Escherichia coli. J. Bacteriol. 169, 2113–2120. Braun, V., Patzer, S.I., Hantke, K., 2002. Ton-dependent colicins and microcins: modular design and evolution. Biochimie 84, 365–380. Clantin, B., Delattre, A.-S., Rucktooa, P., Saint, N., Méli, A.C., Locht, C., Jac¸obDubuisson, F., Villeret, V., 2007. Structure of the membrane protein FhaC: a member of the Omp85-TpsB transporter superfamily. Science 317, 957–961. El Ghachi, M., Bouhss, A., Barreteau, H., Touzé, T., Auger, G., Blanot, D., MenginLecreulx, D., 2006. Colicin M exerts its bacteriolytic effect via enzymatic degradation of undecaprenyl phosphate-linked peptidoglycan precursors. J. Biol. Chem. 281, 22761–22772. Ge, X., Lyu, Z.X., Liu, Y., Wang, R., Zhao, X.S., Fu, X., Chang, Z., 2014. Identification of FkpA as a key quality control factor for the biogenesis of outer membrane proteins under heat shock conditions. J. Bacteriol. 196, 672–680. Harkness, R.E., Braun, V., 1989. Colicin M inhibits peptidoglycan biosynthesis by interfering with lipid carrier recycling. J. Biol. Chem. 264, 6177–6182. Helbig, S., Braun, V., 2011. Mapping functional domains of colicin M. J. Bacteriol. 193, 815–821. Helbig, S., Hantke, K., Ammelburg, M., Braun, V., 2012. CbrA is a flavin adenine dinucleotide protein that modifies the Escherichia coli outer membrane and confers specific resistance to colicin M. J. Bacteriol. 194, 4894–4903. Helbig, S., Patzer, S.I., Schiene-Fischer, C., Zeth, K., Braun, V., 2011. Activation of colicin M by the FkpA prolyl cis-trans isomerase/chaperone. J. Biol. Chem. 286, 6280–6290. Hertle, R., Hilger, M., Weingardt-Kocher, S., Walev, I., 1999. Cytotoxic action of Serratia marcescens hemolysin on human epithelial cells. Infect. Immun. 67, 817–825. Hertle, R., Brutsche, S., Groeger, W., Hobbie, S., Koch, W., Könninger, U., Braun, V., 1997. Specific phosphatidylethanolamine dependence of Serratia marcescens cytotoxin activity. Mol. Microbiol. 26, 853–865. Hullmann, J., Patzer, S.I., Römer, C., Hantke, K., Braun, V., 2008. Periplasmic chaperone FkpA is essential for imported colicin M toxicity. Mol. Microbiol. 69, 926–937. Jac¸ob-Dubuisson, F., Guérin, J., Baelen, S., Clantin, B., 2013. Two-partner secretion: as simple as it sounds? Res. Microbiol. 164, 583–595. Jakes, K.S., Cramer, W.A., 2012. Border crossings: colicins and transporters. Annu. Rev. Genet. 46, 209–231. Kiefer, F., Arnold, K., Künzli, M., Bordoli, L., Schwede, T., 2009. The SWISS-MODEL repository and associated resources. Nucleic Acids Res. 37, D387–D392. König, W., Faltin, Y., Scheffer, J., Schöffler, H., Braun, V., 1987. Role of cell-bound hemolysin as a pathogenicity factor for Serratia infections. Infect. Immun. 55, 2554–2561.
Please cite this article in press as: Braun, V., et al., Import and export of bacterial protein toxins. Int. J. Med. Microbiol. (2014), http://dx.doi.org/10.1016/j.ijmm.2014.12.006
299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317
318
319 320 321 322 323 324
325
326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369
G Model IJMM 50908 1–5
ARTICLE IN PRESS V. Braun et al. / International Journal of Medical Microbiology xxx (2014) xxx–xxx
370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392
Lin, C.S., Horng, J.T., Yang, C.H., Tsai, Y.H., Su, L.H., Wei, C.F., Chen, C.C., Hsieh, S.C., Lu, C.C., Lai, H.C., 2010. RssAB-FlhDC-ShlBA as a major pathogenesis pathway in Serratia marcescens. Infect. Immun. 78, 4870–4881. Marre, R., Hacker, J., Braun, V., 1989. The cell-bound hemolysin of Serratia marcescens contributes to uropathogenicity. Microb. Pathog. 7, 153–156. Ondraczek, R., Hobbie, S., Braun, V., 1992. In vitro activation of the Serratia marcescens hemolysin through modification and complementation. J. Bacteriol. 174, 5086–5094. Patzer, S.I., Albrecht, R., Braun, V., Zeth, K., 2012. Structure and mechanistic studies of pesticin, a bacterial homolog of phage lysozymes. J. Biol. Chem. 287, 23381–23396. Pilsl, H., Glaser, C., Groß, P., Killmann, H., Ölschläger, T., Braun, V., 1993. Domains of colicin M involved in uptake and activity. Mol. Gen. Genet. 240, 103–112. Pilsl, H., Killmann, H., Hantke, K., Braun, V., 1996. Periplasmic location of the pesticin immunity protein suggests inactivation of pesticin in the periplasm. J. Bacteriol. 178, 2431–2435. Poole, K., Schiebel, E., Braun, V., 1988. Molecular characterization of the hemolysin determinant of Serratia marcescens. J. Bacteriol. 170, 3177–3188. Pramanik, A., Hauf, W., Hoffmann, J., Cernescu, M., Brutschy, B., Braun, V., 2011. Oliogomeric structure of ExbB and ExbB-ExbD isolated from Escherichia coli as revealed by LILBID mass spectrometry. Biochemistry 50, 8950–8956. Pramanik, A., Könninger, U., Selvam, A., Braun, V., 2014. Secretion and activation of the Serratia marcescens hemolysin by structurally defined ShlB mutants. Int. J. Med. Microbiol. 304, 351–359.
5
Ricci, D.P., Silhavy, T.J., 2012. The Bam machine: a molecular cooper. Biochim. Biophys. Acta 1818, 1067–1084. Roos, U., Harkness, R.E., Braun, V., 1989. Assembly of colicin genes from a few DNA fragments. Nucleotide sequence of colicin D. Mol. Microbiol. 3, 891–902. Schiebel, E., Braun, V., 1989. Integration of the Serratia marcescens haemolysin into human erythrocyte membranes. Mol. Microbiol. 3, 445–453. Schiebel, E., Schwarz, H., Braun, V., 1989. Subcellular location and unique secretion of the hemolysin of Serratia marcescens. J. Biol. Chem. 264, 16311–16320. Schönherr, R., Tsolis, R., Focareta, T., Braun, V., 1993. Amino acid replacements in the Serratia marcescens haemolysin ShlA define sites involved in activation and secretion. Mol. Microbiol. 9, 1229–1237. Vollmer, W., Pilsl, H., Hantke, K., Höltje, J.-V., Braun, V., 1997. Pesticin displays muramidase activity. J. Bacteriol. 179, 1580–1583. Walker, G., Hertle, R., Braun, V., 2004. Activation of Serratia marcescens hemolysin through a conformational change. Infect. Immun. 72, 611–614. Yang, F.-L., Braun, V., 2000. ShlB mutants of Serratia marcescens allow uncoupling of activation and secretion of the ShlA hemolysin. Int. J. Med. Microbiol. 290, 529–538. Zeth, K., Römer, C., Patzer, S.I., Braun, V., 2008. Crystal structure of colicin M, a novel phosphatase specifically imported by Escherichia coli. J. Biol. Chem. 283, 25324–25331.
Please cite this article in press as: Braun, V., et al., Import and export of bacterial protein toxins. Int. J. Med. Microbiol. (2014), http://dx.doi.org/10.1016/j.ijmm.2014.12.006
393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414
ARTICLE IN PRESS
IJMM 50908 1–5
International Journal of Medical Microbiology xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
International Journal of Medical Microbiology journal homepage: www.elsevier.com/locate/ijmm
1
Mini Review
2
Import and export of bacterial protein toxins Q1
3
Volkmar Braun ∗ , Stephanie Helbig, Silke I. Patzer, Avijit Pramanik, Christin Römer Max Planck Institute for Developmental Biology, Department of Protein Evolution, Spemannstrasse 35, 72076 Tübingen, Germany
4 5
6 15
a r t i c l e
i n f o
a b s t r a c t
7
14
Keywords: Colicin M Pesticin Hemolysin Import Export Activity
16
Introduction
8 9 10 11 12 13
The paper provides a short overview of three investigated bacterial protein toxins, colicin M (Cma) of Escherichia coli, pesticin (Pst) of Yersinia pestis and hemolysin (ShlAB) of Serratia marcescens. Cma and Pst are exceptional among colicins in that they kill bacteria by degrading the murein (peptidoglycan). Both are released into the medium and bind to specific receptor proteins in the outer membrane of sensitive E. coli cells. Subsequently they are translocated into the periplasm by an energy-consuming process using the proton motive force. For transmembrane translocation the colicins unfold and refold in the periplasm. In the case of Cma the FkpA peptidyl prolyl cis-trans isomerase/chaperone is required. ShlA is secreted and activated through ShlB in the outer membrane by a type Vb secretion mechanism. © 2014 Published by Elsevier GmbH.
Q2 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41
Bacteria secrete selected proteins by diverse, yet distinct mechanisms. For gram-negative bacteria at least seven protein secretion types and various subtypes are known. In addition, proteins exist which are secreted without a specific secretion mechanism. These are colicins which are released by nearly half of Escherichia coli natural isolates and kill competing sensitive cells. Despite co-synthesis of immunity proteins which confer resistance to the cognate colicins, a small proportion of cells in a population lyses when colicins are overproduced under stress conditions. In contrast to unspecific export highly specific and sophisticated colicin import systems exist in sensitive cells. According to their mode of action colicins enter the periplasm, the cytoplasmic membrane, or the cytoplasm of sensitive cells. For all colicins import starts with binding to specific receptor proteins exposed at the cell surface. Sensitivity of cells is largely determined by those receptors (Braun et al., 2002; Jakes and Cramer, 2012). Here, two colicins from E. coli and Y. pestis and the hemolysin of S. marcescens were investigated by us and will be discussed with respect to their cellular export and import. Colicin M (Cma) and pesticin (Pst) were studied because only preliminary data existed and they differ from other colicins in that they primarily cause lysis of cells. All other colicins either degrade DNA or RNA in the cytoplasm or form pores in the cytoplasmic membrane resulting in collapse of the membrane potential and later cell lysis. Cma inhibits murein biosynthesis by cleavage of the ester bond in the
∗ Corresponding author. E-mail address: [email protected] (V. Braun).
murein precursor resulting in undecaprenol and 1-pyrophosphoMurNAc-(pentapeptide)-GlcNAc (Harkness and Braun, 1989; El Ghachi et al., 2006). These degradation products cannot be used anymore for murein biosynthesis. Pst cleaves the glycan chain of murein between C1 of MurNAc and C4 of GlcNAc (lysozyme activity; Vollmer et al., 1997). Plasmid-encoded Cma is made by E. coli and kills E. coli cells. Plasmid-encoded Pst is made by Yersinia pestis and kills Yersinia strains. It displays all characteristics of colicins and is, therefore, listed here as a colicin. Hemolysin from S. marcescens has been included in our investigation (Braun et al., 1987) because it was at that time one of the few proteins that were secreted by Enterobacteriaceae and had not been investigated. The S. marcescens hemolysin structure and secretion mechanism turned out to be completely different from the well-studied E. coli hemolysin. Domain structure of the E. coli colicin Cma Cma consists of three domains which are characteristic of colicins. The central domain serves to bind Cma to the outer membrane receptor protein FhuA that concomitantly serves as a receptor for several phages and as a transporter for the iron chelator ferrichrome as well as for antibiotics derived from ferrichrome. The N-terminal domain is required for translocation across the outer membrane into the periplasm, and the C-terminal domain encodes the phosphatase activity (Fig. 1). Deletion of the hydrophobic ␣1 helix close to the N-terminus (Fig. 1; gray) abolishes binding of Cma to FhuA. However, phosphatase activity is retained as demonstrated in an assay that bypasses the FhuA function (Helbig and Braun, 2011). Mutations in ␣1 strongly reduce killing of cells. In contrast to wild-type Cma, these mutants do not prevent killing of
http://dx.doi.org/10.1016/j.ijmm.2014.12.006 1438-4221/© 2014 Published by Elsevier GmbH.
Please cite this article in press as: Braun, V., et al., Import and export of bacterial protein toxins. Int. J. Med. Microbiol. (2014), http://dx.doi.org/10.1016/j.ijmm.2014.12.006
42 43 44 45 46 47 48 49 50 51 52 53 54 55 56
57
58 59 60 61 62 63 64 65 66 67 68 69 70
G Model IJMM 50908 1–5 2
ARTICLE IN PRESS V. Braun et al. / International Journal of Medical Microbiology xxx (2014) xxx–xxx
Import of Cma
Fig. 1. Crystal structure (PDB 2XMX) of colicin M (Zeth et al., 2008) in which functionally important residues are indicated (Helbig and Braun, 2011). N-terminal translocation domain (yellow), central receptor binding domain (blue), C-terminal phosphatase domain (magenta),N-terminal end (N), C-terminal end (C) are indicated. Residues identified as important for Cma activity are shown as sticks. ␣1 indicates the region involved in receptor binding. ␣2 contains a strongly polar sequence, 59-EDYIKKH-65, but only the alanine replacement in E59 reduces Cma activity to 10%.
71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94
cells by the antibiotic albomycin, a ferrichrome derivative, which uses FhuA to enter cells. The mutant Cma proteins do not bind to FhuA which indicates that the ␣1 helix mediates binding of Cma to FhuA. Sequence analyses of bacterial genomes reveal cma orthologs in strains of Pectobacteria, Pseudomonas and Burkholderia. The sequences are highly diverged in the receptor binding and translocation domains but similar in the phosphatase domain. The distinct receptor binding and translocation domains reflect individual import proteins in these bacteria. The conserved Cterminal region reflects the common killing mechanism. The three-domain structure also supports our previous conclusion that colicins in different bacteria evolved by horizontal gene transfer via plasmids and an exchange of gene fragments which encode functional domains (Braun et al., 2002; Roos et al., 1989). The phosphatase domain evolved from a single ancestor gene that was fused to distinct receptor binding and translocation genes. Residues 226 DKYDFNASTHR236 are conserved, hydrophilic and exposed at the Cma surface (Fig. 1). Amino acid replacements in this region render Cma inactive. Particularly noteworthy are mutations D226E and D226N which completely inactivate Cma. This region is part of the active center of Cma, and D226 most likely is directly involved in hydrolysis (Pilsl et al., 1993; Helbig and Braun, 2011).
Five chromosomally encoded genes are required for Cma killing of sensitive E. coli cells, fhuA, tonB, exbB, exbD, and fkpA. Mutations in any of these genes render cells resistant to high doses of Cma (dilution titers of 105 ). The outer membrane receptor FhuA is functionally coupled to the electrochemical gradient across the cytoplasmic membrane via the proteins TonB, ExbB and ExbD. These proteins are anchored in the cytoplasmic membrane and have distinct periplasmic domains. TonB physically interacts with FhuA, and ExbB and ExbD form a complex for which a stoichiometry of 6:1 was determined (Pramanik et al., 2011). It is hypothesized that FhuA changes its conformation in response to the membrane potential. Such conformational changes are suggested to result in release of high-affinity ligands (Cma, ferrichrome, albomycin, phages T1, T5, and 80) from FhuA into the periplasm. FhuA forms a  barrel composed of 16  strands which is closed by an N-proximal globular segment termed plug or cork. The plug must move to allow ligand passage through FhuA. Currently it is unknown whether FhuA only acts as a primary adsorption site for Cma or also as the import route. Cma uptake is even more complex than the uptake of the other FhuA ligands as TonB must not only interact with FhuA but also with Cma (Pilsl et al., 1993). Cma contains a typical TonB box at the N-terminus. Point mutations in the TonB box abolish Cma uptake. Specific point mutations in TonB are capable to suppress point mutations in the Cma TonB box. This is taken as an indication for a direct interaction between the TonB box of Cma and the periplasmic domain of TonB. Outer membrane receptor proteins and all colicins which require TonB for activity contain an N-terminal consensus sequence, i.e. a TonB box. Suppressor analysis, cysteine cross-linking and crystal structures of TonB fragments bound to receptor proteins demonstrate a physical interaction with TonB (Braun, 2014). Essential role of the periplasmic chaperone FkpA for Cma activity FkpA is a peptidyl prolyl cis-trans isomerase with chaperone activity for which only very recently a control of outer membrane biogenesis under heat shock conditions has been assigned (Ge et al., 2014). We discovered that FkpA is essential for killing cells by imported Cma (Hullmann et al., 2008). Likewise, FkpA is essential for Cma secretion into the periplasm provided an artificial signal sequence has been attached to Cma (Helbig et al., 2011). For both, import and export Cma must unfold for crossing the outer and the cytoplasmic membrane. FkpA is indispensable for refolding and accelerates refolding of denatured Cma in vitro. Upon binding to FhuA Cma becomes sensitive to trypsin digestion. A mutation analysis of the fifteen proline residues in Cma identifies proline P176 as the most likely one which is cis-trans isomerized. Indeed, F175–P176 showed in vitro the fastest isomerization rate assayed with synthetic peptides derived from the Cma sequence. P176 is exposed on the surface of the Cma crystal structure which was determined for wild-type Cma and the inactive P176A mutant (Helbig et al., 2011). Of six inactive FkpA point mutants all are located in the isomerase domain. This data identify FkpA as essential for Cma in vivo activity and suggest that Cma is unfolded during import and refolded by FkpA involving cis-trans isomerization of the F175–P176 bond. Further, the data demonstrate an enzymatic activity for FkpA by which a proline bond is isomerized, and demonstrate for the first time an indispensable function of FkpA as a periplasmic chaperone in protein import. Specific resistance to Cma by CbrA Among Cma tolerant E. coli mutants (tolM), fkpA was identified (tolerance defined as Cma binding and, presumably, uptake, but no
Please cite this article in press as: Braun, V., et al., Import and export of bacterial protein toxins. Int. J. Med. Microbiol. (2014), http://dx.doi.org/10.1016/j.ijmm.2014.12.006
95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126
127 128
129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153
154
155 156
G Model IJMM 50908 1–5
ARTICLE IN PRESS V. Braun et al. / International Journal of Medical Microbiology xxx (2014) xxx–xxx
3
zones on plates seeded with Pst sensitive E. coli that carry the Pst receptor of Y. pestis, termed FyuA. The crystal structure of PstRT -T4L reveals a marked difference to Pst as the two domains are connected by a long ␣-helix in contrast to the short unstructured linker in Pst. Although T4L is structurally very similar to the Pst activity domain, PstRT -T4L is not inhibited by the Pst immunity protein Pim. This indicates that Pim is specific for the Pst activity domain. In contrast to the T4 lysozyme, the lysozyme variant from phage T7 when fused to PstRT is not toxic to cells. This may be due to either failure of uptake into the periplasm or misfolding. PstRT mediated import shows specificity. The activity domain of Pst must unfold to enter the periplasm Fig. 2. Crystal structure (4AQN) of pesticin (Patzer et al., 2012). The N-terminal domain (NT) including the TonB box (TBB) is not visible. The translocation domain (T, red), the receptor binding domain (R, blue) and the activity domain (A, orange) are indicated. CT: C-terminus. See text for details.
173
killing). An additional gene locus was found which upon overexpression rendered cells tolerant to Cma. Unexpectedly, the gene responsible for tolerance was cbrA (Helbig et al., 2012) of which very little is known. The extent of tolerance depends on the level of CbrA expression which is controlled by the transcriptional activator CreB. Expression is particularly high in minimal media which also induce cma transcription. The basal level of CbrA synthesis already confers an 18-fold higher survival rate of Cma-treated cells compared to E. coli devoid of cbrA. Upon high cbrA expression the survival rate is increased to over 1000-fold. Since CbrA is expressed under similar conditions as Cma, CbrA confers an appreciable Cma resistance to Cma producing and Cma sensitive cells. Sequence analysis and biochemical studies assign CbrA to FAD-dependent oxidoreductases. FAD is noncovalently bound to CbrA. We found that CbrA renders cells resistant to osmotic shock suggesting that changes in the membrane structure may cause Cma resistance. The mechanism of CbrA function remains enigmatic.
174
Domain structure of pesticin (Pst)
157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172
175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191
192 193
194 195 196 197 198
Most Y. pestis strains possess a small plasmid (9.5 kb) that encodes pesticin (Pst) together with the adjacent immunity gene pim. In the producer cell the immunity protein is localized in the periplasm where it inactivates Pst and thus prevents killing (Pilsl et al., 1996). Murein hydrolysis does not appear to be the only activity of Pst since cells are not immediately lysed but converted to spheroplasts without osmoprotection. Lysis is slow but killing is fast. Murein hydrolysis by Pst in vitro is also slow (Vollmer et al., 1997). The crystal structure of Pst (Patzer et al., 2012) reveals two distinct structural domains connected by a short peptide linker (Fig. 2). The N-terminal domain represents the receptor binding and translocation domain, the C-terminal domain the activity domain. The structure of the activity domain is similar to phage lysozymes, in particular to that of the T4 phage, albeit sequence identity is only 13%. The near identity of the activity domain fold is particularly noteworthy since activity domains of the other colicins display no overall structural similarity to known enzymes. T4 lysozyme is carried into the periplasm by the receptor and translocation domain of Pst Since the Pst activity domain resembles T4 lysozyme (T4L) it was replaced by T4L to test whether T4L is translocated into the periplasm via Pst receptor-binding and translocation domains (PstRT ) (Patzer et al., 2012). PstRT -T4L kills cells but a 6-fold higher concentration than wild-type Pst is required to yield the same lysis
To test whether PstRT -T4L must unfold for transport cysteine residues were introduced in T4L at sites which should allow formation of disulfide bridges (Patzer et al., 2012). The PstRT -T4L chimera was chosen because it efficiently hydrolyzes murein in vitro in contrast to Pst. Three constructs with PstRT -T4L cross-linked by disulfide bridges were inactive in vivo, yet regained activity upon reduction of the disulfide bonds as they killed cells. The disulfide derivatives of Pst-T4L are active in vitro which indicates that the disulfide linkages do not inactivate the lysozyme function. Rather they prevent unfolding and thus inhibit uptake. To examine whether the activity domain must be specifically translocated unfolded across the outer membrane to be able to fold into an active enzyme within the periplasm, Pst and the activity domain were fused to a MalE fragment that contains a signal sequence for secretion from the cytoplasm into the periplasm. Expression was regulated by an arabinose promoter. Upon synthesis of Mal’-Pst and Mal’-A (A, activity domain) cells lyse. This suggests that the activity domain does not require the RT domains of Pst to fold into an active domain and that there are no specific membrane requirements for Pst translocation and refolding. The active site of Pst was deduced from the crystal structure in comparison with T4L. Amino acid replacements at selected sites reveal that the active site residues are similar but not identical to T4L (Patzer et al., 2012). Secretion of the hemolysin of Serratia marcescens, a model system for type Vb secretion S. marcescens is involved in urinary tract infections, bacteremia, endocarditis, keratitis, arthritis, and meningitis. One of its virulence factors is a hemolysin/cytolysin that lyses erythrocytes and elicits inflammatory response in leukocytes (Hertle et al., 1999; König et al., 1987; Lin et al., 2010; Marre et al., 1989). Hemolysin was the first example of a by now large group of proteins which are secreted by the type Vb or the two-partner secretion system (TPS) (Braun et al., 1987; Poole et al., 1988; Schiebel et al., 1989; Jac¸ob-Dubuisson et al., 2013). The hemolysin is determined by two genes, shlA and shlB, arranged in tandem; shlA encodes hemolysin, and shlB encodes the outer membrane protein responsible for ShlA secretion. Both proteins have N-terminal signal sequences for transport across the cytoplasmic membrane by the Sec system. ShlB not only secretes ShlA but it also activates it; in a shlB deletion mutant, ShlA remains in the periplasm in an inactive non-hemolytic form (Poole et al., 1988; Schiebel and Braun, 1989; Schiebel et al., 1989). Activation can be uncoupled from secretion by in vitro incubation of isolated periplasmic ShlA with purified ShlB resulting in hemolytic ShlA (Hertle et al., 1997; Ondraczek et al., 1992). A ShlA N-terminal fragment of 242 residues (ShlA consists of 1578 residues) activates inactive periplasmic ShlA to hemolytic
Please cite this article in press as: Braun, V., et al., Import and export of bacterial protein toxins. Int. J. Med. Microbiol. (2014), http://dx.doi.org/10.1016/j.ijmm.2014.12.006
199 200 201 202 203 204 205 206 207 208 209
210 211
212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235
236 237
238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259
G Model IJMM 50908 1–5 4
ARTICLE IN PRESS V. Braun et al. / International Journal of Medical Microbiology xxx (2014) xxx–xxx
quantitative assay. The predicted structure of ShlB (Fig. 3) was used to design deletion mutants of H1, L6, and the two POTRA domains to study their effects on secretion and activation. Deletion of H1 does not affect secretion of hemolytic ShlA. H1 has no essential function in ShlB assembly, secretion and activation of ShlA (Pramanik et al., 2014). Deletion of L6 diminishes, but does not fully abolish secretion, but completely compromises ShlA activation. The P1 and P2 single mutants secrete reduced amounts of ShlA of which only ShlA of the P1 mutant shows some hemolytic activity. Thus P1 and P2 mutants discriminate between secretion and activation. H1− , P1− , P2− triple deletion mutants secrete no ShlA. Random mutagenesis yields inactive ShlB derivatives mutated in L6 which supports the importance of the conserved L6 among type Vb transporters (Pramanik et al., 2014; Yang and Braun, 2000). ShlA is secreted and folds into a hemolytic structure through its N-terminus which is secreted first. Secretion of non-hemolytic ShlA is inefficient. Folding into the active conformation probably drives thermodynamically secretion of ShlA through ShlB.
Acknowledgments
Fig. 3. Model of ShlB based on the crystal structure of FhaC (Clantin et al., 2007; Pramanik et al., 2014).
260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298
ShlA provided ShlA242 is secreted by ShlB (Ondraczek et al., 1992; Schönherr et al., 1993). ShlA242 purified from the periplasm of a shlB mutant cannot activate periplasmic ShlA. ShlA242 mimics secretion and activation of ShlA by ShlB but is not hemolytic since the C-terminal region of ShlA is required for pore formation in erythrocyte membranes (Schiebel and Braun, 1989; Schiebel et al., 1989; Schönherr et al., 1993). ShlA and ShlA242 change conformations when they are secreted and activated by ShlB. Secreted ShlA and ShlA242 are largely trypsin resistant. In contrast, cytoplasmic and periplasmic ShlA and ShlA242 are cleaved by trypsin to small fragments (Walker et al., 2004; Yang and Braun, 2000). ShlA242 isolated from the periplasm, the cytoplasm (ShlA242 without a signal sequence), and the culture supernatant of a shlB mutant that secretes but does activate ShlA242 have identical electrophoretic mobilities in native polyacrylamide gels. In contrast, secreted and activated ShlA242 has a much higher mobility. In addition, the CD spectra reveal structural differences (Schiebel et al., 1989; Walker et al., 2004). The data indicate that inactive ShlA and inactive ShlA242 are less structurally constrained than the hemolytic isoforms. Furthermore activation most likely does not involve a chemical modification through interaction with ShlB as activation does not change the molecular weights (Schiebel et al., 1989; Walker et al., 2004). The type Vb secretion system belongs to a transporter superfamily of gram-negative bacteria, mitochondria, and chloroplasts for which the Bam complex of E. coli serves as a model system (Ricci and Silhavy, 2012). Swiss-Model Repository (Kiefer et al., 2009) predicts a ShlB structure similar to that of FhaC, the outer membrane exporter of the Bordetella pertussis filamentous hemagglutinin which consists of an ␣-helix (H1) and a loop (L6), both inside a 16-stranded -barrel, and two polypeptide-transportassociated (POTRA) domains in the periplasm (Fig. 3). The TPS system is simpler than the Bam insertion and secretion machinery. The latter has four additional proteins, BamB–E, and up to seven POTRA domains. Hence, the TPS system is particularly suitable to study the mechanism of protein export across the outer membrane. The soluble N-terminal fragment of ShlA242 contains an export signal, the conformation is susceptible to activation, probably by refolding, which enable hemolysis which is testable by a simple
We thank Andrei Lupas for the generous hospitality in his department and Joachim Schultz and Klaus Hantke for critically reading the manuscript. This work was financed by the Max Planck Q3 Society, the German Science Foundation (BR330/25-1, SFB766) and the Fonds der Chemischen Industrie. Space limitation did not allow citing all of the relevant literature.
References Braun, V., 2014. Energy-coupled transport across the outer membrane of gramnegative bacteria. In: Remaut, H., Fronzes, R. (Eds.), Bacterial Membranes. Structure and Molecular Biology. Caister Academic Press, Norfolk, pp. 249–291. Braun, V., Neuss, B., Ruan, Y., Schiebel, E., Schöffler, H., Jander, G., 1987. Identification of the Serratia marcescens hemolysin determinant by cloning into Escherichia coli. J. Bacteriol. 169, 2113–2120. Braun, V., Patzer, S.I., Hantke, K., 2002. Ton-dependent colicins and microcins: modular design and evolution. Biochimie 84, 365–380. Clantin, B., Delattre, A.-S., Rucktooa, P., Saint, N., Méli, A.C., Locht, C., Jac¸obDubuisson, F., Villeret, V., 2007. Structure of the membrane protein FhaC: a member of the Omp85-TpsB transporter superfamily. Science 317, 957–961. El Ghachi, M., Bouhss, A., Barreteau, H., Touzé, T., Auger, G., Blanot, D., MenginLecreulx, D., 2006. Colicin M exerts its bacteriolytic effect via enzymatic degradation of undecaprenyl phosphate-linked peptidoglycan precursors. J. Biol. Chem. 281, 22761–22772. Ge, X., Lyu, Z.X., Liu, Y., Wang, R., Zhao, X.S., Fu, X., Chang, Z., 2014. Identification of FkpA as a key quality control factor for the biogenesis of outer membrane proteins under heat shock conditions. J. Bacteriol. 196, 672–680. Harkness, R.E., Braun, V., 1989. Colicin M inhibits peptidoglycan biosynthesis by interfering with lipid carrier recycling. J. Biol. Chem. 264, 6177–6182. Helbig, S., Braun, V., 2011. Mapping functional domains of colicin M. J. Bacteriol. 193, 815–821. Helbig, S., Hantke, K., Ammelburg, M., Braun, V., 2012. CbrA is a flavin adenine dinucleotide protein that modifies the Escherichia coli outer membrane and confers specific resistance to colicin M. J. Bacteriol. 194, 4894–4903. Helbig, S., Patzer, S.I., Schiene-Fischer, C., Zeth, K., Braun, V., 2011. Activation of colicin M by the FkpA prolyl cis-trans isomerase/chaperone. J. Biol. Chem. 286, 6280–6290. Hertle, R., Hilger, M., Weingardt-Kocher, S., Walev, I., 1999. Cytotoxic action of Serratia marcescens hemolysin on human epithelial cells. Infect. Immun. 67, 817–825. Hertle, R., Brutsche, S., Groeger, W., Hobbie, S., Koch, W., Könninger, U., Braun, V., 1997. Specific phosphatidylethanolamine dependence of Serratia marcescens cytotoxin activity. Mol. Microbiol. 26, 853–865. Hullmann, J., Patzer, S.I., Römer, C., Hantke, K., Braun, V., 2008. Periplasmic chaperone FkpA is essential for imported colicin M toxicity. Mol. Microbiol. 69, 926–937. Jac¸ob-Dubuisson, F., Guérin, J., Baelen, S., Clantin, B., 2013. Two-partner secretion: as simple as it sounds? Res. Microbiol. 164, 583–595. Jakes, K.S., Cramer, W.A., 2012. Border crossings: colicins and transporters. Annu. Rev. Genet. 46, 209–231. Kiefer, F., Arnold, K., Künzli, M., Bordoli, L., Schwede, T., 2009. The SWISS-MODEL repository and associated resources. Nucleic Acids Res. 37, D387–D392. König, W., Faltin, Y., Scheffer, J., Schöffler, H., Braun, V., 1987. Role of cell-bound hemolysin as a pathogenicity factor for Serratia infections. Infect. Immun. 55, 2554–2561.
Please cite this article in press as: Braun, V., et al., Import and export of bacterial protein toxins. Int. J. Med. Microbiol. (2014), http://dx.doi.org/10.1016/j.ijmm.2014.12.006
299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317
318
319 320 321 322 323 324
325
326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369
G Model IJMM 50908 1–5
ARTICLE IN PRESS V. Braun et al. / International Journal of Medical Microbiology xxx (2014) xxx–xxx
370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392
Lin, C.S., Horng, J.T., Yang, C.H., Tsai, Y.H., Su, L.H., Wei, C.F., Chen, C.C., Hsieh, S.C., Lu, C.C., Lai, H.C., 2010. RssAB-FlhDC-ShlBA as a major pathogenesis pathway in Serratia marcescens. Infect. Immun. 78, 4870–4881. Marre, R., Hacker, J., Braun, V., 1989. The cell-bound hemolysin of Serratia marcescens contributes to uropathogenicity. Microb. Pathog. 7, 153–156. Ondraczek, R., Hobbie, S., Braun, V., 1992. In vitro activation of the Serratia marcescens hemolysin through modification and complementation. J. Bacteriol. 174, 5086–5094. Patzer, S.I., Albrecht, R., Braun, V., Zeth, K., 2012. Structure and mechanistic studies of pesticin, a bacterial homolog of phage lysozymes. J. Biol. Chem. 287, 23381–23396. Pilsl, H., Glaser, C., Groß, P., Killmann, H., Ölschläger, T., Braun, V., 1993. Domains of colicin M involved in uptake and activity. Mol. Gen. Genet. 240, 103–112. Pilsl, H., Killmann, H., Hantke, K., Braun, V., 1996. Periplasmic location of the pesticin immunity protein suggests inactivation of pesticin in the periplasm. J. Bacteriol. 178, 2431–2435. Poole, K., Schiebel, E., Braun, V., 1988. Molecular characterization of the hemolysin determinant of Serratia marcescens. J. Bacteriol. 170, 3177–3188. Pramanik, A., Hauf, W., Hoffmann, J., Cernescu, M., Brutschy, B., Braun, V., 2011. Oliogomeric structure of ExbB and ExbB-ExbD isolated from Escherichia coli as revealed by LILBID mass spectrometry. Biochemistry 50, 8950–8956. Pramanik, A., Könninger, U., Selvam, A., Braun, V., 2014. Secretion and activation of the Serratia marcescens hemolysin by structurally defined ShlB mutants. Int. J. Med. Microbiol. 304, 351–359.
5
Ricci, D.P., Silhavy, T.J., 2012. The Bam machine: a molecular cooper. Biochim. Biophys. Acta 1818, 1067–1084. Roos, U., Harkness, R.E., Braun, V., 1989. Assembly of colicin genes from a few DNA fragments. Nucleotide sequence of colicin D. Mol. Microbiol. 3, 891–902. Schiebel, E., Braun, V., 1989. Integration of the Serratia marcescens haemolysin into human erythrocyte membranes. Mol. Microbiol. 3, 445–453. Schiebel, E., Schwarz, H., Braun, V., 1989. Subcellular location and unique secretion of the hemolysin of Serratia marcescens. J. Biol. Chem. 264, 16311–16320. Schönherr, R., Tsolis, R., Focareta, T., Braun, V., 1993. Amino acid replacements in the Serratia marcescens haemolysin ShlA define sites involved in activation and secretion. Mol. Microbiol. 9, 1229–1237. Vollmer, W., Pilsl, H., Hantke, K., Höltje, J.-V., Braun, V., 1997. Pesticin displays muramidase activity. J. Bacteriol. 179, 1580–1583. Walker, G., Hertle, R., Braun, V., 2004. Activation of Serratia marcescens hemolysin through a conformational change. Infect. Immun. 72, 611–614. Yang, F.-L., Braun, V., 2000. ShlB mutants of Serratia marcescens allow uncoupling of activation and secretion of the ShlA hemolysin. Int. J. Med. Microbiol. 290, 529–538. Zeth, K., Römer, C., Patzer, S.I., Braun, V., 2008. Crystal structure of colicin M, a novel phosphatase specifically imported by Escherichia coli. J. Biol. Chem. 283, 25324–25331.
Please cite this article in press as: Braun, V., et al., Import and export of bacterial protein toxins. Int. J. Med. Microbiol. (2014), http://dx.doi.org/10.1016/j.ijmm.2014.12.006
393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414