| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Division of Immunity and Infection, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK
2 Department of Microbiology and Molecular Genetics, Harvard Medical School, Boston, MA 02115, USA
3 Center for Vaccine Development, University of Maryland, Baltimore, MD 21201, USA
Correspondence
Ian R. Henderson
I.R.Henderson{at}bham.ac.uk
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Present address: Institut National de la Recherche Agronomique (INRA), Centre de Recherche Clermont-Ferrand – Theix – Lyon, UR 454 – Unité de Microbiologie, Site de Theix, F-63122 Saint-Genès Champanelle, France.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Mougous et al., 2006; Thanassi et al., 2005). One of these pathways is the type V secretion system (T5SS), which can be subdivided into: (i) the T5aSS, which includes the classical autotransporters; (ii) the T5bSS, which includes the two-partner secretion pathway; and (iii) the T5cSS, representing the recently described trimeric autotransporters (Henderson et al., 2004).
The T5SS can be briefly defined as a system that permits translocation of proteins, which fold largely into 
-helical structures, across the outer membrane via a -barrel transmembrane pore, with outer membrane translocation occurring in the absence of an apparent energy input, i.e. ATP (Desvaux et al., 2004a, b; Henderson et al., 1998, 2004). Proteins secreted via the T5aSS and T5cSS are synthesized as a single modular polypeptide composed of three main domains: (i) an N-terminal signal peptide mediating inner membrane translocation; (ii) a central passenger domain corresponding to the functional portion of the molecule; and (iii) a C-terminal translocation unit (TU), which forms a -barrel pore within the outer membrane, and permits translocation of the functional passenger domain to the bacterial cell surface and beyond (Henderson et al., 2004). These two systems are differentiated by the fact that members of the T5aSS possess a single intact TU, whereas members of the T5cSS possess only one third of a TU, and are required to trimerize to form a functional TU in the outer membrane (Cotter et al., 2005; Meng et al., 2006; Surana et al., 2004). In contrast, the exoproteins (designated TpsA) secreted via the T5bSS are synthesized separately from their -barrel TU (designated TpsB). Both proteins possess signal sequences and are translocated independently across the inner membrane, and after translocation through the inner membrane, undergo a specific interaction at the outer membrane before the exoprotein is translocated to the external milieu (Henderson et al., 2004; Jacob-Dubuisson et al., 2001, 2004; Newman & Stathopoulos, 2004).
In the current model for T5SS, the signal peptide first targets the protein destined for secretion to the inner membrane Sec translocon (SecYEG), which then exports the protein into the periplasm (Henderson et al., 2004). The majority of proteins secreted via the T5SS have signal sequences that possess the canonical N (charged), H (hydrophobic) and C (signal peptidase cleavage site) domains associated with translocation via the classic posttranslational pathway (Henderson et al., 1998, 2004). However, several members of this family have been noted to have unusual signal peptides that exceed 50 aa in length (Desvaux et al., 2006; Henderson et al., 1998). While these unusual signal peptides do possess the canonical domains associated with signal sequences of the posttranslational pathway (N2, H2 and C), they are distinguished by an N-terminal extension, which possesses an additional charged domain (N1) and an additional hydrophobic domain (H1) designated the extended signal peptide region (ESPR) (Fig. 1). Interestingly, and in contrast to the normal posttranslational signal peptides, which are highly degenerate, the N-terminal extensions of these signal peptides possess a high degree of conservation (Desvaux et al., 2006; Henderson et al., 1998, 2004). Such conservation is often indicative of a conserved function, and suggests that these N-terminal extensions play an important role in the biogenesis of proteins secreted via the T5SS.
|
; Sijbrandi et al., 2003). Further investigations have indicated that deletion of the ESPR inhibits passenger domain translocation across the outer membrane, and that overproduction of proteins harbouring the ESPR affects the translocation efficiency of Sec-dependent proteins (Szabady et al., 2005). In contrast, in vitro and in vivo investigations of filamentous haemagglutinin (FHA), a TpsA protein harbouring an ESPR, have demonstrated that SecA and SecB are required for inner membrane translocation via the posttranslational pathway, and that, despite cross-linking with SRPs, inner membrane translocation is strictly SRP-independent (Chevalier et al., 2004). Moreover, it has been noted that rather than determining a specific mode of membrane targeting and translocation, the extended signal peptide of FHA influences the rate of inner membrane translocation (Chevalier et al., 2004). Thus, the exact nature of the inner membrane translocation event remains controversial.
To investigate the structure–function relationship of the unusual signal peptide, and in particular the ESPR region, we studied Pet, a prototypical member of the SPATE subfamily (Eslava et al., 1998; Henderson & Nataro, 2001; Henderson et al., 2004). In the current study, we provide evidence that an ESPR-containing signal sequence directs protein translocation across the inner membrane in a novel posttranslational fashion, and that the presence of the ESPR strongly affects the efficiency of inner membrane translocation.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Strains were passaged routinely on LB broth or agar with the following supplements where appropriate: ampicillin (100 µg ml–1), kanamycin (50 µg ml–1), nalidixic acid (50 µg ml–1), 5-bromo-4-chloro-3-indolyl phosphate (XP; 40 µg ml–1) and IPTG (0.5 mM). The plasmid pCEFN1 overexpressing Pet has been described previously (Eslava et al., 1998). pQUANTagen(kx) (Quantum Appligene) was used to construct PhoA reporter fusions.
|
). Protein concentrations were calculated using the Bradford dye method (Bradford, 1976). 1D SDS-PAGE was performed by using 12.5 % (w/v) acrylamide resolving gels and 4.5 % (w/v) acrylamide stacking gels (Laemmli, 1970). Proteins were detected by staining with Coomassie brilliant blue R250 or the SilverQuest Silver Staining kit (Invitrogen), as described by the supplier. For Western immunoblotting, horseradish peroxidase-conjugated goat anti-rabbit IgG (Sigma) was used as the localizing reagent, and reacting bands were visualized using the SuperSignal West Pico Chemiluminescent Substrate according to the manufacturer's instructions (Pierce).
DNA manipulations and sequencing.
Standard molecular biology procedures and DNA manipulations were used (Sambrook & Russell, 2001). Plasmids were purified using the QIAprep Spin Miniprep kit (Qiagen). DNA purification from PCR/enzymic reactions and agarose gels were performed using the QIAquick PCR Purification and Gel Extraction kits (Qiagen), respectively. Restriction enzymes and T4 DNA ligase from Invitrogen were used. PCR experiments were performed using Platinum Pfx DNA Polymerase (Invitrogen). Site-directed mutations were constructed with the QuikChange II kit (Stratagene). Error-prone PCR (EP-PCR) was performed using the GeneMorph II Random Mutagenesis kit (Stratagene), to achieve 1–3 mutations per kilobase. All of the above manipulations were carried out in accordance with the instructions of the respective manufacturers. PCR and sequencing primers are listed in Table 2.
|
*) was constructed in which the putative initiation codon was converted to a stop codon (TAG). Similarly, pCEFN1-N2S2 (primers IRHN2F and IRHN2R), pCEFN1-Y5R5 (primers IRHY5F and IRHY5R), pCEFN1-Y9R9 (primers IRHY9F and IRHY9R) and pCEFN1-S20ELR20DR (primers IRHS20F and IRHS20R) were constructed using the designated primers. Plasmids were transformed into Escherichia coli HB101.
Pulse–chase labelling.
Pulse–chase labelling was performed as described by Ulbrandt et al. (1997). Briefly, E. coli DO251 and its isogenic secA mutant BA13 were transformed with pQMDSSPet, and grown overnight at 30 °C in M9 medium containing ampicillin, 0.4 % (w/v) glucose, and 40 µg L-amino acids ml–1, excluding methionine and cysteine. Cells were resuspended in fresh medium at OD550 0.02, and grown for 3 h. Overexpression of PhoA fusion proteins was induced by the addition of IPTG 30 min prior to labelling. Aliquots were removed and cells were pulse-labelled for 30 s with 30 µCi ml–1 (1.1 MBq ml–1) Tran35S-label (ICN Biochemicals). Cold methionine and cysteine (1 mM) were then added for the chase. Cold TCA was added at various times to a final concentration of 10 %. Precipitates were collected by centrifugation (16 000 g, for 10 min), and pellets were redissolved in solubilization buffer. Samples were diluted 1 : 20 into radioimmunoprecipitation (RIPA) buffer and aliquots were withdrawn for immunoprecipitation by anti-PhoA antibodies. After washing, immunoprecipitated proteins were resolved by SDS-PAGE and visualized using a Fuji BAS2000 phosphorimager.
Construction and analysis of thioredoxin reporter constructs.
The pet signal sequence, incorporating three residues of mature Pet in order to maintain the signal peptidase cleavage site, was amplified using primers AST1 and AST2, incorporating BspHI and NcoI restriction sites at the 5' and 3' ends, respectively. To create pPetSSTrxA, the digested PCR product was ligated to a digested inverse PCR product of the backbone of plasmid pCFS123, generated using primers AST3 and AST4, such that the plasmid expressed an in-frame fusion between the Pet signal sequence and thioredoxin (PetSS–TrxA). TrxA expression was monitored as previously described (Huber et al., 2005).
Construction and analysis of PhoA fusions.
To monitor the ability of the Pet signal sequence and mutant derivatives to mediate inner membrane translocation, a reporter system, under the control of an IPTG-inducible ptac promoter, was constructed using pQUANTagen(kx). Chloramphenicol amplification was used for pQUANTagen(kx) isolation (Sambrook & Russell, 2001). Fractions were harvested from mid-exponential-phase cultures after treatment with 100 mM iodoacetamide (DeLisa et al., 2003) and combined lysozyme/osmotic shock treatment (French et al., 1996). Alkaline phosphatase activity of PhoA was assayed spectrophotometrically, utilizing p-nitrophenyl phosphate as a substrate at 25 °C, by monitoring the release of p-nitrophenolate at 410 nm (Walter & Schutt, 1976). Specific activity was determined in a range in which a linear relationship with protein concentration was established. Results were expressed in µmol min–1 (mg protein)–1, using a molar extinction coefficient of 18 300 M cm–1 for p-nitrophenolate. Data were derived from three independent experiments, and were analysed by Student's t test for unpaired data, to determine statistically significant differences (P <0.05).
The Pet signal peptide was amplified from pCEFN1 using primers SSPetBamFw and SSPetSacRv, and then cloned into pQUANTagen(kx), resulting in plasmid pQMDSSPet. Sequential deletions from the N-terminal end of the Pet signal peptide were achieved by PCR from pQMDSSPet, using the reverse primer SSPetSacRv and the forward primers SSPetDeletN2-I7BamFw, SSPetDeletN2-I17BamFw, SSPetDeletN2-A23BamFw and SSPetDeletN2-I27BamFw. The respective plasmid constructs pQMDSSPet
N2-I7, pQMDSSPetN2-I17, pQMDSSPetN2-A23 and pQMDSSPetN2-I27 were obtained after cloning of the products into pQUANTagen(kx). pQMDSSPetI34-I49 was obtained after cloning PCR products from pQMDSSPet, using the forward primer SSPetBamFw and the reverse primer SSPetDeletI34-I49SacRv, into pQUANTagen(kx).
Internal deletions were constructed by inverse PCR using pQMDSSPet as a template. Primers IPCRSSPetDeletA11-A23NdeFw and IPCRSSPetDeletA11-A23NdeRv were used to create a signal sequence–PhoA construct (pQMDSSPetA11-A23) lacking the H1 domain, and primers IPCRSSPetDeletK24-K33SfiFw and IPCRSSPetDeletK24-K33SfiRv were used to create a signal sequence–PhoA construct (pQMDSSPetK24-K33) lacking the N2 domain. Linear DNA products from the inverse PCR reactions were digested with the appropriate restriction enzyme (either NdeI or SfiI) before ligation.
To construct pQMDSSPetM1-A52, dsDNA was obtained by mixing equimolar amounts of 5'-phosphorylated primers 5'pNoSSFw and 5'pNoSSRv dissolved in annealing buffer (10 mM Tris, pH 8.0, 50 mM NaCl, 1 mM EDTA), and then heating at 95 °C and gradually cooling to room temperature before cloning into pQUANTagen(kx) pre-digested with BamHI/SacI. The -lactamase (SSBlaBamFw and SSBlaSacRv) and MBP*1 (SSMBP*1BamFw and SSMBP*1SacRv) signal peptides were synthesized without a DNA template, following PCR using the indicated primer pairs, which resulted in pQMDSSBla and pQMDSSMBP*1 after cloning in pQUANTagen(kx). The TorA signal peptide was similarly synthesized using primers SSTorABamFw and SSTorARv1, but with a nested PCR using primers SSTorABamFw and SSTorASacRv2, before cloning to create pQMDSSTorA.
Random and site-directed mutagenesis of the Pet signal sequence was performed using pQMDSSPet as a template. For random mutagenesis, a 1112 bp fragment was amplified by EP-PCR using primers EPPCRFw and EPPCRRv, and the products were cloned into pQUANTagen(kx). Mutagenesis of the positions (i) M1 into a stop codon (ATGTAG), (ii) N2 into A2 (AATGCT), (iii) Y9 into D9 (TATGAT), and (iv) E21 into A21 (GAAGCA) was performed by PCR with the reverse primer SSPetSacRv and the forward primers SSPetM1*1BamFw, SSPetN2A2BamFw, SSPetY9D9BamFw and SSPetE21A21BamFw, respectively (Table 2
). The resulting PCR products were cloned into pQUANTagen(kx), resulting in pQMDSSPetM1*, pQMDSSPetN2A2, pQMDSSPetY9D9 and pQMDSSPetE21A21, respectively.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
). To assess the ability of this putative extended signal sequence to translocate proteins across the inner membrane, a heterologous reporter construct (pQMDSSPet) was created in which the signal sequence of PhoA, whose expression was governed by an IPTG-inducible ptac promoter, was replaced with the 52 aa of the putative Pet signal sequence (Fig. 1). Since PhoA is active only after it has been translocated to the periplasm, it provides a method of determining the efficiency of inner membrane translocation. After induction with IPTG, PhoA activity could be detected from E. coli Star pQMDSSPet, indicating that the ESPR-containing signal sequence acted as a signal peptide (Fig. 1). In contrast, E. coli Star carrying a construct pQMDSSPetN2-A52 lacking residues N2–A52 of the Pet signal sequence demonstrated no PhoA activity or detectable levels of PhoA, as assessed by specific-activity measurements and Western immunoblotting, presumably due to the rapid degradation of the mistargeted protein (Fig. 1).
To determine the efficiency of the Pet signal sequence in directing inner membrane translocation of proteins, PhoA reporter constructs were created with signal peptides representative of different targeting pathways. Thus, the signal peptides from -lactamase (as a model for protein routing towards a Sec-dependent, but SRP-independent pathway), MBP*1 (a triple-point mutant of the MBP signal peptide, used as a model of SRP-dependent signal peptides), and TorA (a model of a strictly Tat-dependent pathway) were fused to PhoA, and the specific activity of PhoA was calculated for each construct (Beha et al., 2003; Cristobal et al., 1999; DeLisa et al., 2003; Lee & Bernstein, 2001) (Fig. 1). After the addition of IPTG, the Pet signal sequence fusion (PetSS–PhoA) demonstrated PhoA activity approximately 2.5-fold higher than that of the TorA signal sequence fusion (TorASS–PhoA). However, Western immunoblotting revealed that the presence of detectable PhoA was severely diminished in strains possessing the TorASS–PhoA fusion (Fig. 1). This phenomenon has previously been observed, and the presence of active PhoA is attributed to leakage of Tat signal sequence fusions of Sec substrates through the Sec pathway, and to folding quality control of the Tat pathway (Richter & Bruser, 2005). In contrast, the MBP*1 and -lactamase signal sequences resulted in higher levels of PhoA activity, even though similar levels of PhoA were present, indicating that the differences in PhoA activity were due to differences in the amount of active translocated enzyme (Fig. 1).
ESPR-containing signal sequences direct inner membrane translocation posttranslationally
As we determined that the unusual signal peptide predicted for Pet was capable of directing inner membrane translocation, we wished to determine whether Pet was translocated via the SecYEG complex. The E. coli strains DO251 (secA+ supFts) and BA13 (secAam supFts) were transformed with pQMDSSPet and grown in minimal medium. Cells were subjected to pulse–chase labelling after the addition of IPTG and a shift to 41 °C, which switches off expression of SecA in strain BA13. Consistent with the hypothesis that the signal sequence directs inner membrane translocation via the SecYEG translocon, depletion of SecA had a pronounced effect on the accumulation of mature PhoA by inhibiting the cleavage of the Pet signal sequence from PhoA (Fig. 2A). In contrast, in the control strain DO251, the signal sequence was rapidly cleaved and mature PhoA was detected (Fig. 2A).
|
). To determine if the Tat pathway played a role in inner membrane translocation, E. coli MC4100 and its isogenic tatC mutant E. coli B1LKO were transformed with either pQMDSSPet or pCEFN1 expressing native Pet. A significant role for the Tat pathway could be ruled out, as the levels of Pet secretion and PhoA activity were unaltered in the tatC mutant compared with those in the parent strain (Fig. 2B).
In contrast, SRP has been proposed to mediate targeting and inner membrane translocation of ESPR-containing homologues of Pet, namely, the SPATE autotransporters Hbp and EspP (Peterson et al., 2003; Sijbrandi et al., 2003). However, controversy has arisen during investigation of FHA, a TpsA protein harbouring an ESPR-containing signal peptide, demonstrating that despite cross-linking transiently with SRP, inner membrane translocation is trigger factor- and SecB-dependent, and strictly SRP-independent (Chevalier et al., 2004). Thus, it remains to be unequivocally determined whether ESPR-containing signal peptides direct posttranslational or cotranslational translocation across the inner membrane. Due to problems associated with depletion experiments, which have given rise to the above controversy, we sought to investigate the role of the ESPR in inner membrane targeting and translocation by using a reporter fusion based on the cytoplasmic protein thioredoxin. It has previously been demonstrated that thioredoxin is a rapidly folding protein which can only be efficiently translocated to the periplasm in the presence of a cotranslational signal peptide; posttranslational translocation is prevented by the rapid folding of thioredoxin in the cytoplasm, and the inability of the SecYEG translocon to accommodate folded proteins (Huber et al., 2005; Schierle et al., 2003). The DNA encoding the Pet signal sequence was ligated in-frame to trxA, the gene encoding thioredoxin, to create pPetSSTrxA, which expressed a Pet signal sequence–TrxA fusion protein under the control of an IPTG-inducible trc promoter. Western blotting with anti-thioredoxin antibodies of various cellular fractions from E. coli WP570, expressing a DsbA signal sequence–TrxA fusion (DsbASS–TrxA), revealed that the cotranslational SRP-dependent DsbA signal sequence could efficiently export thioredoxin to the periplasm, as observed previously (Fig. 3). In contrast, neither the PhoA signal sequence (PhoASS–TrxA) nor the Pet signal sequence (PetSS–TrxA) fusion could direct inner membrane translocation of thioredoxin (Fig. 3). These data strongly suggest that the ESPR-containing signal peptide does not direct cotranslational export via the classical SRP-dependent method, as previously described (Peterson et al., 2003; Sijbrandi et al., 2003), but that export more closely resembles the posttranslational pathway.
|
). The H2 domain was shown to be important, as PhoA activity could not be detected from the pQMDSSPetI34-I49 construct. In contrast, deletion of the N1 (pQMDSSPetN2-I7), H1 (pQMDSSPetA11-A23) or N2 (pQMDSSPetK24-K33) domains demonstrated that none of these individual domains was absolutely required for PhoA activity, although in the N2 construct, PhoA activity was severely diminished, indicating an important role for this region in inner membrane translocation. Surprisingly, however, protein export was drastically increased when the N1H1 regions were deleted, either separately or together (pQMDSSPetN2-I7, pQMDSSPetN2-A23, pQMDSSPetN2-I27 and pQMDSSPetA11-A23), even though the total cellular levels of PhoA were approximately equivalent, as assessed by Western blotting with anti-PhoA antibodies. Indeed, the efficiency of inner membrane translocation was comparable to that obtained with pQMDSSMBP*1, an SRP-dependent signal peptide (Fig. 1).
|
I1, M1L1 and M1K1) were identified (Fig. 4). When examining the whole Pet signal peptide, the only other amino acid identified that was non-permissive for inner membrane translocation was L41 within the H2 domain; L41P41 and L41Q41 mutations abolished PhoA activity completely. Kyte and Doolittle hydropathy profiling of the signal sequences possessing such mutations demonstrated that these mutations decreased the level of hydrophobicity in the H2 domain (data not shown).
To ensure that we had not missed any of the highly conserved residues during our EP-PCR experiments, we used site-directed mutagenesis to make non-conservative mutations within the ESPR region of the Pet signal sequence–PhoA fusion. As expected, mutation of the N-terminal M1 into a stop codon abolished PhoA activity (Fig. 4). Surprisingly, none of the non-conservative mutations, i.e. N2A2, Y9D9 or E21A21, significantly affected the efficiency of PhoA translocation (Fig. 4).
Since the system we were using is a heterologous system, we could not rule out the possibility that the ESPR conserved residues only play a role in biogenesis of proteins secreted via the T5SS. Thus, we mutated non-conservatively all of the highly conserved residues within the native Pet signal sequence, and monitored by SDS-PAGE analysis of supernatant fractions the ability of these constructs to produce Pet (Fig. 5). Surprisingly, and as with similar mutations within the conserved residues of the signal sequence–PhoA fusions, none of the mutations within the highly conserved residues of the ESPR appeared to significantly affect Pet secretion. However, it should be noted that upon repeated fractionation, the level of Pet released into the culture supernatant was slightly diminished relative to that of the wild-type, when the conserved tyrosine residues were mutated to charged residues (Y5–R5 and Y9–R9). Previous investigations have suggested that the conserved residues of the ESPR may play a role in the correct folding of the passenger domain. As a marker of correct biogenesis, we tested the functional activity of the protein derived from each site-directed mutant. All mutant forms of Pet demonstrated similar levels of toxicity in the standard HEp-2 cell assay described previously (Eslava et al., 1998; data not shown).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 2004).
The question of which inner membrane translocation mechanism is utilized by polypeptides possessing an ESPR-containing signal sequence has been the subject of much recent debate. The expression analysis presented in this paper appears to rule out any significant role for the Tat pathway, despite the presence of a motif similar to the Tat motif within some of the signal sequences. In contrast, initial experiments investigating Hbp, a homologue of Pet, indicate that the unusual signal peptide mediates targeting and inner membrane translocation in an SRP-dependent cotranslational manner (Sijbrandi et al., 2003). Further studies using the autotransporter EspP suggest that SRP dependence is due to the inherent hydrophobicity of the signal peptide, and appears not to be associated with the presence of the ESPR per se (Lee & Bernstein, 2001; Peterson et al., 2003). However, Hbp and EspP contain helix-breaking residues in their H2-domain, which is generally considered to be a discriminatory factor excluding SRP recognition of signal peptides (Adams et al., 2002). Aware of the bias resulting from the study of SRP dependence using in vivo approaches in which depletion of SRP results in the depletion of SecY, an essential component of the Sec translocon (Beha et al., 2003; Koch et al., 1999; Seluanov & Bibi, 1997), Chevalier et al. (2004) have investigated the inner membrane translocation of FHA by coupling in vivo and in vitro approaches. Despite transient cross-linking to SRP, it has been concluded that FHA is exported via SecYEG in a posttranslational manner. Indeed, the most recent work from Bernstein's group indicates that the ESPR-containing signal sequence of EspP mediates posttranslational export in a trigger factor-independent fashion (Peterson et al., 2006). The data derived from the TrxA fusion experiments presented here are consistent with a posttranslational route for inner membrane translocation of Pet, and reinforce the recent data generated for FHA and EspP. However, it remains possible that SRP binds the signal sequence and targets the protein to the inner membrane Sec translocon, but that translocation ultimately occurs in a posttranslational fashion. Nevertheless, what remains ultimately clear is that if SRP is involved, translocation does not occur in the classical SRP-dependent cotranslational fashion.
If the ESPR region is not involved in export via the SRP cotranslational pathway, then what is the role for this conserved region in inner membrane translocation? Our results for the Pet signal sequence demonstrated that translocation was less efficient than that associated with signal sequences mediating posttranslational or cotranslational translocation. These data agree well with initial observations for FHA, in which functional investigations have revealed that the unusual signal peptide is a poor mediator of inner membrane translocation (Jacob-Dubuisson et al., 1996). However, data arising from mutational analysis of the Pet signal sequence are striking, revealing that deletion of the ESPR results in a significant increase in the level of PhoA activity. Such a delay in maturation has previously been observed for EspP containing a deletion of the ESPR region (Szabady et al., 2005). However, these authors have speculated that the ESPR might permit temporary inner membrane anchoring, such that the -domain of the EspP autotransporter can fold into a correct configuration prior to insertion into the outer membrane (Szabady et al., 2005). Indeed, these investigators have specifically noted that the presence of the ESPR affects neither the rate nor the efficiency of inner membrane translocation (Szabady et al., 2005). However, the theory proposed by Szabady et al. (2005) is difficult to reconcile with the presence of the ESPR in FHA and the other TpsA proteins, where the TpsB outer membrane -barrel is secreted separately across the inner membrane, and thus folds independently of the secreted protein (Jacob-Dubuisson et al., 2004). While both studies indicate a delay in maturation, we favour a hypothesis whereby the process of inner membrane translocation is delayed, rather than release of the mature protein from the inner membrane. Previous investigations have demonstrated that PhoA is active within the periplasm, even when tethered to the inner membrane (Manoil & Beckwith, 1986), and thus it appears that the difference in PhoA activity can only be due to the efficiency of translocation through the SecYEG translocon. In agreement with this finding, recent studies of the FHA ESPR suggest that, in the absence of the ESPR, the secreted protein remains functional, but the presence of the ESPR affects the rate and/or efficiency of inner membrane translocation (Chevalier et al., 2004). Indeed, based on work with EspP, Peterson et al. (2006) now propose that the ESPR mediates an interaction with a cytoplasmic factor, or induces the formation of an unusual signal peptide conformation which delays the onset of protein translocation. Thus, it appears that, rather than having an involvement in a specific mode of inner membrane targeting, the ESPR may simply play a role in regulating the rate of protein export.
The question then arises as to why a specific polypeptide might regulate the rate of inner membrane translocation. In this respect, decreasing the rate of export allows the molecules to fold into a quasi-competent state for subsequent insertion into, and translocation across, the outer membrane. Indeed, the delay observed elsewhere in outer membrane translocation of EspP derivatives lacking the ESPR can alternatively be explained by aggregation of periplasmic intermediates that are translocated across the inner membrane at a high rate in the absence of ESPR (Szabady et al., 2005). However, if the ESPR is absolutely crucial for folding of secreted T5SS proteins, it is difficult to explain why it is not systematically present in all such proteins. On the other hand, overexpression of autotransporters is often detrimental to bacterial cells (Henderson & Desvaux, 2004), and the presence of the ESPR may prevent rapid accumulation of periplasmic intermediates to a toxic level.
Nevertheless, these data raise several possibilities for the role of the ESPR in inner membrane translocation. One such possibility is that the ESPR-containing signal sequences target proteins posttranslationally, but delay interaction of the signal peptide with the SecYEG translocon by adopting a specific conformation. The importance of secondary structure within signal sequences is well established and again demonstrated here, where a mutation disrupting the secondary structure, such as mutation of L41 or deletion of any domain, greatly influences translocation.
Alternatively, the ESPR may delay translocation of the polypeptide through the SecYEG translocon. This observation is supported by the delay in accumulation of active PhoA in the periplasm, and by other data which demonstrate that overproduction of a protein containing an ESPR-type signal sequence delays export of other proteins when expressed in trans (Szabady et al., 2005). However, if the mechanism behind decreased protein export depends only on the presence of a transmembrane helix within ESPR, it is difficult to explain why this domain is so conserved, and why mutation of the conserved residues does not appear to affect inner membrane translocation.
A third possibility is that the presence of the ESPR permits recognition by a specific cytoplasmic or inner-membrane protein(s) which slows down protein translocation. While debate still continues over SRP involvement, there remains the possibility that other cytoplasmic or inner-membrane factors are also involved. In fact, global mutagenesis strategies employed by us and others have identified several cytoplasmic and inner membrane proteins which appear to play a role in biogenesis of proteins secreted via the T5SS (Hardie et al., 2004; M. Desvaux and others, unpublished results).
In conclusion, a consensus opinion is now emerging that inner membrane translocation of proteins possessing an ESPR-type signal peptide occurs in a posttranslational fashion, and the data presented here argue against the proposition that these proteins must be translocated through the SecYEG translocon in a cotranslational fashion to avoid degradation or folding within the cytoplasm. Furthermore, these data indicate that the ESPR delays inner membrane translocation by hindering interaction with the SecYEG translocon, or by delaying passage through the translocon in a novel fashion.
|
ACKNOWLEDGEMENTS |
|---|
Edited by: S. MacIntyre
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Beha, D., Deitermann, S., Muller, M. & Koch, H. G. (2003). Export of beta-lactamase is independent of the signal recognition particle. J Biol Chem 278, 22161–22167.
Bogsch, E. G., Sargent, F., Stanley, N. R., Berks, B. C., Robinson, C. & Palmer, T. (1998). An essential component of a novel bacterial protein export system with homologues in plastids and mitochondria. J Biol Chem 273, 18003–18006.
Bradford, M. M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72, 248–254.[CrossRef][Medline]
Chevalier, N., Moser, M., Koch, H. G., Schimz, K. L., Willery, E., Locht, C., Jacob-Dubuisson, F. & Muller, M. (2004). Membrane targeting of a bacterial virulence factor harbouring an extended signal peptide. J Mol Microbiol Biotechnol 8, 7–18.[CrossRef][Medline]
Cotter, S. E., Surana, N. K. & St Geme, J. W., 3rd (2005). Trimeric autotransporters: a distinct subfamily of autotransporter proteins. Trends Microbiol 13, 199–205.[CrossRef][Medline]
Cristobal, S., de Gier, J. W., Nielsen, H. & von Heijne, G. (1999). Competition between Sec- and TAT-dependent protein translocation in Escherichia coli. EMBO J 18, 2982–2990.[CrossRef][Medline]
DeLisa, M. P., Tullman, D. & Georgiou, G. (2003). Folding quality control in the export of proteins by the bacterial twin-arginine translocation pathway. Proc Natl Acad Sci U S A 100, 6115–6120.
Desvaux, M., Parham, N. J. & Henderson, I. R. (2004a). Type V protein secretion: simplicity gone awry? Curr Issues Mol Biol 6, 111–124.[Medline]
Desvaux, M., Parham, N. J. & Henderson, I. R. (2004b). The autotransporter secretion system. Res Microbiol 155, 53–60.[Medline]
Desvaux, M., Cooper, L. C., Filenko, N. A., Scott-Tucker, A., Turner, S. M., Cole, J. A. & Henderson, I. R. (2006). The unusual signal sequence of the Type V secretion is phylogenetically restricted. FEMS Microbiol Lett 264, 22–30.[CrossRef][Medline]
Eslava, C., Navarro-Garcia, F., Czeczulin, J. R., Henderson, I. R., Cravioto, A. & Nataro, J. P. (1998). Pet, an autotransporter enterotoxin from enteroaggregative Escherichia coli. Infect Immun 66, 3155–3163.
French, C., Keshavarz-Moore, E. & Ward, J. M. (1996). Development of a simple method for the recovery of recombinant proteins from the Escherichia coli periplasm. Enzyme Microb Technol 19, 332–338.[CrossRef]
Hardie, K. R., Arnold, L., Dodson, S. & Baldwin, T. (2004). Autotransporters do not reach their extracellular location alone. In Abstracts of the 155th Meeting of the Society for General Microbiology, 6–9 September 2004, Trinity College Dublin, abstract no. CCS 28. Reading, UK: Society for General Microbiology. http://www.sgm.ac.uk/meetings/pdfabstracts/tcd2004abs.pdf
Henderson, I. R. & Desvaux, M. (2004). The type V secretion pathway: a premium source of virulence factors? Drug Discov Today 9, 240–243.[CrossRef]
Henderson, I. R. & Nataro, J. P. (2001). Virulence functions of autotransporter proteins. Infect Immun 69, 1231–1243.
Henderson, I. R., Navarro-Garcia, F. & Nataro, J. P. (1998). The great escape: structure and function of the autotransporter proteins. Trends Microbiol 6, 370–378.[CrossRef][Medline]
Henderson, I. R., Navarro-Garcia, F., Desvaux, M., Fernandez, R. C. & Ala'Aldeen, D. (2004). Type V protein secretion pathway: the autotransporter story. Microbiol Mol Biol Rev 68, 692–744.
Huber, D., Boyd, D., Xia, Y., Olma, M. H., Gerstein, M. & Beckwith, J. (2005). Use of thioredoxin as a reporter to identify a subset of Escherichia coli signal sequences that promote signal recognition particle-dependent translocation. J Bacteriol 187, 2983–2991.
Jacob-Dubuisson, F., Locht, C. & Antoine, R. (2001). Two-partner secretion in Gram-negative bacteria: a thrifty, specific pathway for large virulence proteins. Mol Microbiol 40, 306–313.[CrossRef][Medline]
Jacob-Dubuisson, F., Buisine, C., Mielcarek, N., Clement, E., Menozzi, F. D. & Locht, C. (1996). Amino-terminal maturation of the Bordetella pertussis filamentous haemagglutinin. Mol Microbiol 19, 65–78.[CrossRef][Medline]
Jacob-Dubuisson, F., Fernandez, R. & Coutte, L. (2004). Protein secretion through autotransporter and two-partner pathways. Biochim Biophys Acta 1694, 235–257.[Medline]
Koch, H. G., Hengelage, T., Neumann-Haefelin, C., MacFarlane, J., Hoffschulte, H. K., Schimz, K. L., Mechler, B. & Muller, M. (1999). In vitro studies with purified components reveal signal recognition particle (SRP) and SecA/SecB as constituents of two independent protein-targeting pathways of Escherichia coli. Mol Biol Cell 10, 2163–2173.
Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685.[CrossRef][Medline]
Lee, H. C. & Bernstein, H. D. (2001). The targeting pathway of Escherichia coli presecretory and integral membrane proteins is specified by the hydrophobicity of the targeting signal. Proc Natl Acad Sci U S A 98, 3471–3476.
Manoil, C. & Beckwith, J. (1986). A genetic approach to analyzing membrane protein topology. Science 233, 1403–1408.
Meng, G., Surana, N. K., St Geme, J. W., III & Waksman, G. (2006). Structure of the outer membrane translocator domain of the Haemophilus influenzae Hia trimeric autotransporter. EMBO J 25, 2297–2304.[CrossRef][Medline]
Mougous, J. D., Cuff, M. E., Raunser, S., Shen, A., Zhou, M., Gifford, C. A., Goodman, A. L., Joachimiak, G., Ordonez, C. L. & other authors. (2006). A virulence locus of Pseudomonas aeruginosa encodes a protein secretion apparatus. Science 312, 1526–1530.
Newman, C. L. & Stathopoulos, C. (2004). Autotransporter and two-partner secretion: delivery of large-size virulence factors by gram-negative bacterial pathogens. Crit Rev Microbiol 30, 275–286.[CrossRef][Medline]
Peterson, J. H., Woolhead, C. A. & Bernstein, H. D. (2003). Basic amino acids in a distinct subset of signal peptides promote interaction with the signal recognition particle. J Biol Chem 278, 46155–46162.
Peterson, J. H., Szabady, R. L. & Bernstein, H. D. (2006). An unusual signal peptide extension inhibits the binding of bacterial presecretory proteins to the signal recognition particle, trigger factor and the SecYEG complex. J Biol Chem 281, 9038–9048.
Richter, S. & Bruser, T. (2005). Targeting of unfolded PhoA to the Tat translocon of Escherichia coli. J Biol Chem 280, 42723–42730.
Sambrook, J. & Russell, D. W. (2001). Molecular Cloning: a Laboratory Manual, 3rd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
Schierle, C. F., Berkmen, M., Huber, D., Kumamoto, C., Boyd, D. & Beckwith, J. (2003). The DsbA signal sequence directs efficient, cotranslational export of passenger proteins to the Escherichia coli periplasm via the signal recognition particle pathway. J Bacteriol 185, 5706–5713.
Seluanov, A. & Bibi, E. (1997). FtsY, the prokaryotic signal recognition particle receptor homologue, is essential for biogenesis of membrane proteins. J Biol Chem 272, 2053–2055.
Sijbrandi, R., Urbanus, M. L., ten Hagen-Jongman, C. M., Bernstein, H. D., Oudega, B., Otto, B. R. & Luirink, J. (2003). Signal recognition particle (SRP)-mediated targeting and Sec-dependent translocation of an extracellular Escherichia coli protein. J Biol Chem 278, 4654–4659.
Surana, N. K., Cutter, D., Barenkamp, S. J. & St Geme, J. W., III (2004). The Haemophilus influenzae Hia autotransporter contains an unusually short trimeric translocator domain. J Biol Chem 279, 14679–14685.
Szabady, R. L., Peterson, J. H., Skillman, K. M. & Bernstein, H. D. (2005). An unusual signal peptide facilitates late steps in the biogenesis of a bacterial autotransporter. Proc Natl Acad Sci U S A 102, 221–226.
Thanassi, D. G., Stathopoulos, C., Karkal, A. & Li, H. (2005). Protein secretion in the absence of ATP: the autotransporter, two-partner secretion and chaperone/usher pathways of Gram-negative bacteria (review). Mol Membr Biol 22, 63–72.[Medline]
Ulbrandt, N. D., Newitt, J. A. & Bernstein, H. D. (1997). The E. coli signal recognition particle is required for the insertion of a subset of inner membrane proteins. Cell 88, 187–196.[CrossRef][Medline]
Walter, K. & Schutt, C. (1976). Methods of Enzymatic Analysis, 2nd edn. New York: Academic Press.
Received 24 April 2006;
revised 24 September 2006;
accepted 25 September 2006.
| ||||||||||||||||||||||||||||||||||||||||||||||||
