The extracytoplasmic folding factor PrsA is required for protein secretion only in the presence of the cell wall in Bacillus subtilis

Eva Wahlström{dagger}, Marika Vitikainen, Vesa P. Kontinen and Matti Sarvas

Vaccine Development Laboratory, National Public Health Institute, Mannerheimintie 166, FIN-00300, Helsinki, Finland

Correspondence
Vesa P. Kontinen
vesa.kontinen{at}ktl.fi


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Pulse–chase labelling was used to study the role of the cell wall microenvironment in the functioning of Bacillus subtilis PrsA, an extracellular lipoprotein and member of the parvulin family of peptidylprolyl cis/trans-isomerases. It was found that in protoplasts, and thus in the absence of a cell wall matrix, the post-translocational folding, stability and secretion of the AmyQ {alpha}-amylase were independent of PrsA, in contrast to the strict dependency found in rods. The results indicate that PrsA is dedicated to assisting the folding and stability of exported proteins in the particular microenvironment of the cytoplasmic membrane–cell wall interface, possibly as a chaperone preventing unproductive interactions with the wall. The data also provide evidence for a crucial role of the wall in protein secretion. The presence of the wall directly or indirectly facilitates the release of AmyQ from the cell membrane and affects the rate of the signal peptide processing.


Abbreviations: PPIase; peptidylprolyl cis/trans-isomerases; SMS, Spizizen's minimal salts

{dagger}Present address: Institute of Biotechnology, Biocentre 1, PO Box 56, FIN-00014 University of Helsinki, Helsinki, Finland.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
In bacteria, extracytoplasmic proteins that are destined to be secreted into the growth medium as well as components of the cell envelope are exported through the cytoplasmic membrane via a multicomponent protein translocation apparatus (translocase). The principal components of this machinery and their modes of functioning are conserved and similar in both Gram-positive and Gram-negative bacteria (the Escherichia coli translocase has been reviewed by Manting & Driessen, 2000). After translocation and cleavage of a signal peptide, proteins fold into their native conformation. The mechanisms behind this process are now being unravelled. Many folding factors, including chaperones, peptidylprolyl cis/trans-isomerases (PPIase) and thiol-disulphide oxido-reductases, which assist post-translocational folding, have been identified (Tjalsma et al., 2000).

In the Gram-positive Bacillus subtilis, the cell membrane is surrounded by a matrix of cell wall polymers, peptidoglycan and anionic teichoic (or teichuronic) acids (reviewed by Archibald et al., 1993). The matrix polymers form a porous structure that allows macromolecules up to a molecular mass of about 50 kDa (Demchick & Koch, 1996) to pass through the wall. Teichoic acids, which are polymers of glycerol (or ribitol) phosphate units, are attached either to the peptidoglycan (wall teichoic acids) or to the cell membrane (lipoteichoic acids). These anionic polymers are abundant constituents of the wall and provide a high density of negative charge on the cell envelope and a high capacity to bind divalent metal ions and other cationic molecules (Beveridge & Murray, 1980; Peschel et al., 1999; Petit-Glatron et al., 1993). The negative charge as well as the wall's ability to bind cations is reduced by D-alanine ester substitution of teichoic acids (Hyyryläinen et al., 2000; Perego et al., 1995). Since the B. subtilis cell does not have a membrane-enclosed periplasm, secretory proteins move directly through the cell wall into the external medium. However, the pore size and negative charge of the wall matrix can be expected to limit the protein traffic (Demchick & Koch, 1996; Merchante et al., 1995). Some proteins also remain in the matrix of the cell wall. Among them are WprA protease, wall-associated protein WapA and autolysins (Margot & Karamata, 1996; Smith et al., 2000; Yoshida et al., 1995).

Secreted proteins emerge from the translocase to the compartment between the cell wall and the cytoplasmic membrane. This is a demanding environment for protein folding owing to the high density of negative charge, high concentration of cations and low pH immediately outside the membrane. These factors most likely pose stringent requirements for the folding kinetics of secreted proteins. Native proteins compatible with the conditions at the membrane–wall interface fold with fast kinetics into their normal conformation. In contrast, heterologous proteins, produced in B. subtilis for biotechnical applications, may have slow folding kinetics. Furthermore, they are usually more susceptible to proteolytic degradation than native proteins (Stephenson et al., 1998). The incompatibility associated with heterologous proteins may also result in misfolding and aggregation (Bolhuis et al., 1999; Meens et al., 1993). There are regulatory mechanisms (CssRS two-component system) to sense the accumulation of misfolded proteins at the membrane–wall interface and to activate the synthesis of HtrA-type ‘cleaning proteases' (Hyyryläinen et al., 2001). This is thought to be one of the major reasons for the low production levels of heterologous proteins in industrial applications.

The major extracytoplasmic folding factor in B. subtilis is the PrsA protein, which belongs to the parvulin family of PPIases (Rahfeld et al., 1994). PrsA is a typical bacterial lipoprotein anchored to the cytoplasmic membrane by an N-terminal diacylglyceryl moiety (Kontinen & Sarvas, 1993; Kontinen et al., 1991; Leskelä et al., 1999b). The hydrophilic, positively charged protein domain is located on the outer surface of the membrane, as indicated by its accessibility to external trypsin in protoplasts (Leskelä et al., 1999b). Our data showed a linear relationship between the amount of cellular PrsA and the secretion of overproduced AmyQ {alpha}-amylase into the culture medium (Vitikainen et al., 2001). The PrsA protein does not influence either the expression or the translocation of secretory proteins, but it is required for their folding and stability in the post-translocational phase of secretion at the membrane–cell wall interface (Hyyryläinen et al., 2000, 2001; Jacobs et al., 1993; Leskelä et al., 1999b; Vitikainen et al., 2001). PrsA is an essential cell component, suggesting that it also affects the folding and stability of some essential proteins involved in the synthesis of the cell wall or the function of the cell membrane (Vitikainen et al., 2001).

This study was designed to elucidate the role of the cell wall for the function of the PrsA folding factor. To accomplish this, we developed a method to pulse–chase label protoplasts. We compared the secretion kinetics and stability of a model protein, AmyQ {alpha}-amylase, in protoplasts and rods. The results indicated that unlike in rods, PrsA is not needed for the folding, stability or secretion of AmyQ in protoplasts. Furthermore, in the absence of the wall, a substantial fraction of AmyQ remains cell-associated.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains, plasmids and growth conditions.
The bacterial strains used in this study were B. subtilis IH6513 (hisA1 thr-5 trpC2, pKTH10), IH6525 (prsA3 hisA1 trpC2, pKTH10) and IH6838 (hisA1 thr-5 trpC2, pKTH10 pKTH277). The pKTH10 plasmid carries the amyQ gene of Bacillus amyloliquefaciens (Palva, 1982), encoding {alpha}-amylase that is secreted at a high level into the culture medium. The pKTH277 plasmid carries the prsA gene of B. subtilis and results in an increased cellular level of the PrsA protein. Bacteria were cultivated in SMSa medium, which is Spizizen's minimal salts medium (Harwood & Cutting, 1990) supplemented with 10 mM potassium glutamate, 1 mM CaCl2, 5 µM FeCl3, 1 µM ZnSO4, 0·5 % (w/v) maltose and 50 µg each of the 20 aa except methionine ml-1. For plasmid maintenance, growth media contained kanamycin (10 mg l-1), chloramphenicol (5 mg l-1) or erythromycin (1 mg l-1), as appropriate.

Preparation of protoplasts.
Cells overproducing B. amyloliquefaciens {alpha}-amylase (AmyQ) were grown in SMSa minimal medium up to the late exponential growth phase. In this growth phase, the culture was typically at a cell density of 50–100 Klett units (OD600 of about 0·5–1·0), and the cells efficiently secreted AmyQ. In a typical experiment, cells from 20 ml culture were harvested and resuspended in 2 ml SMSb [SMSa with 20 % (w/v) sucrose and 10 mM Mg2+ for the stability of protoplast membrane] containing lysozyme (1 mg ml-1) and mutanolysin (100 U ml-1) to remove the cell wall. The lysozyme and mutanolysin were purchased from Sigma. The cell suspension was slowly shaken at 37 °C and the conversion to protoplasts was monitored by phase-contrast microscopy. After incubation for 30 min, all the cells had turned into protoplasts, as revealed by phase-contrast microscopy. Electron microscopy showed that the total surface of the cytoplasmic membrane was exposed, with no visible structures of the cell wall remaining (Fig. 1A). Protoplasts were harvested by centrifugation at 9000 g for 4 min and resuspended in 3 ml SMSb. Electron microscopy was performed as described previously (Lounatmaa, 1985).



View larger version (51K):
[in this window]
[in a new window]
 
Fig. 1. Characterization of B. subtilis protoplasts prepared as described in Methods. (A) Electron micrograph of a protoplast preparation. The arrow indicates a ghost-like structure. Bar, 1 µm. (B) Metabolic labelling and stability of protoplasts. Protoplasts or rods were pulse-labelled with [35S]methionine and chased with non-radioactive methionine as indicated. The cell and supernatant fractions were separated and labelled proteins were analysed by SDS-PAGE and fluorography. The fractions are rod pellet (rp), rod supernatant (rs), protoplast pellet (pp) and protoplast supernatant (ps). A putative intracellular protein (ip), which was quantified by densitometric scanning, is indicated. (C) The effect of 3 mM sodium azide (NaN3) on the processing and secretion kinetics of AmyQ {alpha}-amylase in protoplasts and rods. Secretion was not fully inhibited by this particular concentration of NaN3.

 
Pulse–chase labelling.
Protoplasts in 3 ml SMSb were preincubated at 37 °C for 15 min with shaking (100 r.p.m.) and then radiolabelled with 250 µCi [35S]methionine for 45 s. SMSb was supplemented with NaN3 (3 mM) in experiments to inhibit SecA ATPase. Radiolabelled protoplasts were chased by adding 0·5 ml 1 % (w/v) methionine and 0·5 ml samples were taken at appropriate time intervals into microcentrifuge tubes at 0 °C. When the labelling efficiency or processing kinetics of the AmyQ signal peptide was studied, the tubes contained 50 µl 100 % (w/v) TCA to precipitate either cellular or secreted proteins. Precipitated material was centrifuged, washed once each with acetone and diethylether, resuspended in 50 µl 50 mM Tris/HCl, pH 8·0, 10 mM EDTA and lysozyme (2 mg ml-1), and incubated at 37 °C for 15 min. Then the samples were solubilized with 2·5 µl 20 % (w/v) SDS, neutralized with 1 ml 50 mM Tris/HCl, pH 7·5, 150 mM NaCl, 2 % (w/v) Triton X-100, 1 mM EDTA and 1 mM PMSF, and centrifuged to remove undissolved precipitate. To measure the labelling efficiency of proteins in rods and protoplasts, samples of solubilized material were electrophoresed by SDS-PAGE, fluorographed and scanned (as below) to determine the optical density of the protein bands. AmyQ material in the solubilized samples was immunoprecipitated with specific antibodies and analysed by SDS-PAGE and fluorography as described previously (Leskelä et al., 1999a). The densitometric quantification of AmyQ bands in the fluorographs was performed using Bio Image equipment (Milligen/Biosearch). When the secretion kinetics were studied, the samples were not TCA-precipitated but centrifuged at 12 000 g for 2 min at 4 °C, followed by pipetting of the protoplast supernatants into new Eppendorf tubes containing 10 µl 100 mM PMSF. Both the pellets and protoplast supernatants were rapidly frozen in dry ice/ethanol and stored at -80 °C until immunoprecipitation and analysis of AmyQ. The pellets were resuspended in 50 µl 50 mM Tris/HCl, pH 7·5, followed by solubilization with SDS, neutralization with Triton X-100, immunoprecipitation and analysis of the AmyQ material as described above. The protoplast supernatants were centrifuged at 12 000 g for 10 min to remove possible cell debris and then subjected to immunoprecipitation like the other samples. The pulse–chase labelling of rods was carried out as described previously (Leskelä et al., 1999a).

Other methods.
The trypsin sensitivity of AmyQ secreted from protoplasts was determined as follows. A series of protoplast preparations in 3 ml SMSb each were prepared as described above, followed by incubation at 37 °C for 10 min with shaking (100 r.p.m.) and then continued for 1 h in the presence of different concentrations (0·2, 1, 5, 25 and 100 µg ml-1) of trypsin. After the incubation, samples (0·4 ml) were filtered through Millex-LCR13 filter units (Millipore), trypsin inhibitor was added at a concentration exceeding that of trypsin by at least twofold and the {alpha}-amylase activities of the samples were determined. The trypsin sensitivity of PrsA and PrsA3 in protoplasts was determined with 50 µg ml-1 trypsin in a similar way to that of AmyQ, except that the PrsA content of the protoplast samples was analysed by immunoblotting.

Extracellular {alpha}-amylase activity was determined using Phadebas Amylase Test Tablets (Pharmacia) as described previously (Leskelä et al., 1999a).


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Preparation of metabolically active protoplasts from B. subtilis cells and their pulse–chase labelling
To assess the role of the cell wall matrix in the function of the PrsA folding factor and protein secretion in B. subtilis, we developed a protoplast-based system to pulse–chase-label and study the kinetics of the secretion of B. amyloliquefaciens {alpha}-amylase AmyQ, a PrsA-dependent model exoprotein. Protoplasts were prepared from cells overexpressing amyQ at the late exponential phase of growth, as described in Methods, and suspended in a buffer solution containing 20 % sucrose for osmoprotection and 10 mM Mg2+ for membrane stability. To assess the proportion of protoplasts disrupted during their preparation, the activity of glucose-6-phosphate dehydrogenase, a cytoplasmic enzyme, released into the medium was determined (Puohiniemi et al., 1992). In a typical experiment, about 50 % of the glucose-6-phosphate dehydrogenase activity remained in the supernatant fraction, implying that half of the protoplasts were disrupted. Consistent with this, ghost-like structures were seen in the electron micrograph of the preparation (Fig. 1A) or in phase-contrast microscopy (not shown). Electron microscopy also showed the removal of the cell wall matrix. The total surface of the cytoplasmic membrane was directly exposed to the medium.

We next examined the incorporation of metabolic label into the protoplasts. These were labelled with 50 µCi [35S]methionine for 45 s, then chased with non-radioactive methionine for 10 min, followed by separation of the protoplast and supernatant fractions by centrifugation. About 35 % of the added label was incorporated into the protoplasts, as determined by counting radioactivity of TCA precipitates immediately after labelling. This efficiency of label incorporation was only slightly less than that observed with rods (45 %) labelled under identical conditions. SDS-PAGE and fluorography revealed that the protoplasts were metabolically active: the labelling of proteins was effective and comparable to that in the rods (Fig. 1B). To determine the stability of the protoplasts during the procedure of pulse–chase, we scanned densitometrically one strongly labelled putative intracellular protein (indicated with an arrow in Fig. 1B). In the non-chased sample, about 7 % of this protein resided in the supernatant fraction after pelleting of protoplasts most likely due to breakage of some of the protoplasts during the separation of the protoplast pellet and supernatant fractions subsequent to the pulse–chase labelling (Fig. 1B). This amount did not increase during the chase of up to 40 min (Fig. 1B and data not shown), indicating that the stability of the protoplasts did not deteriorate during the pulse–chase labelling. The pattern and intensity of other protein bands also remained unchanged during the chase. No cellular proteins were found in the rod supernatant.

Cell wall matrix influences the kinetics of the signal peptide processing
The AmyQ proteins radiolabelled during a pulse were immunoprecipitated by specific antibodies. The immunoprecipitated preAmyQ and AmyQ polypeptides were separated in SDS-PAGE and visualized by fluorography. In a parallel experiment, rods of the same strain were pulse–chase labelled and analysed similarly. Similar amounts of labelled and immunoprecipitated AmyQ proteins were found in the prechase preparations of rods and protoplasts (Fig. 1C). In rods, the cleavage of the signal peptide was completed within 2 min of the chase. Mature {alpha}-amylase was chased from cells and accumulated in the medium (Fig. 1C). In protoplasts, the signal peptide processing was clearly slower. After 2 min of chasing, less than 50 % of the precursor synthesized during the pulse was chased into the mature form and even after 10 min of the chase, some preAmyQ was still detected (Fig. 1C). Mature AmyQ, but not the precursor, was released into the medium, although the amount was only about 15 % of that released from the rods. The proportions of the intensities of the AmyQ bands varied to some extent from one experiment to another (see Fig. 2B). The processing and export of AmyQ were similarly inhibited in both the rods and protoplasts by sodium azide, an inhibitor of the SecA ATPase of the protein translocator (Fig. 1C). Thus, not only are the protoplasts metabolically active, but also they secrete proteins into the medium through the Sec-dependent pathway.



View larger version (77K):
[in this window]
[in a new window]
 
Fig. 2. Comparison of the PrsA-dependency of the stability of AmyQ in protoplasts and rods. (A) SDS-PAGE separation and fluorographic visualization of cellular proteins pulse–chase labelled in protoplasts and rods of the prsA3 mutant and its wild-type (wt) parent. (B) Immunoprecipitation, SDS-PAGE separation and fluorographic detection of AmyQ material from pulse–chase labelled cells of the experiment in (A).

 
PrsA is needed for the stability of AmyQ in the microenvironment of the cell wall but not in protoplasts
AmyQ is extensively degraded by endogenous proteases in cells depleted of PrsA, just as it is in prsA3 mutants (Hyyryläinen et al., 2000, 2001; Kontinen & Sarvas, 1993; Vitikainen et al., 2001). Degradation in this mutant is most likely a consequence of the post-translocational misfolding of AmyQ at the cytoplasmic membrane–cell wall interface owing to the decreased level of the PrsA foldase, less than 10 % of the wild-type (Hyyryläinen et al., 2001). The level of PrsA3 was low also in protoplasts of the prsA3 mutant; the ratio of PrsA proteins in protoplasts and rods in the prsA3 mutant and a strain with abundant PrsA protein were approximately similar (see Fig. 6a and data not shown). The protoplasts of the prsA3 mutant displayed similar incorporation and labelling efficiencies to those of the wild-type, as evidenced by comparable patterns of labelled proteins (Fig. 2A). Comparison of the fate of cell associated AmyQ in rods of the prsA3 mutant and wild-type strains by pulse–chase labelling revealed a clearly lower amount of AmyQ in the prsA3 (Fig. 2B), consistent with the proteolytic degradation and with previous observations. However, the separation of the cell and medium fractions revealed that once AmyQ was released from the cells and folded into its native conformation, it became highly resistant to proteolysis, as evidenced by the lack of degradation of the exoamylase during 40 min chases in both the wild-type and the prsA3 mutant (Fig. 3). In striking contrast to the low level of mature or secreted AmyQ in prsA3 rods, the prsA3 protoplasts contained twice as much cell-associated AmyQ as the wild-type protoplasts (Figs 2 and 3). This was true both for prechase samples and for samples in the course of the chase for up to 40 min (also see below). Neither in rods nor in protoplasts was the rate of the signal peptide processing affected by the PrsA deficiency.



View larger version (23K):
[in this window]
[in a new window]
 
Fig. 6. Trypsin sensitivity of AmyQ secreted by protoplasts and PrsA bound to the protoplast membrane. (a) Protoplasts with either a low (prsA3 mutant) or high (PrsA overproduction; pKTH277) level of PrsA were incubated while shaken in the absence or presence of trypsin. The immunoblot shows PrsA levels in protoplasts (p) and rods (r) of the prsA3 mutant and the PrsA overproducer strain. The analysed samples were derived from the indicated volumes of culture. (b) Protoplasts of the prsA3 mutant (filled squares) and the PrsA overproducer (open triangles) were incubated in the presence of various concentrations of trypsin. After the incubation and inhibition of the trypsin activity by trypsin inhibitor, {alpha}-amylase activity of the medium was determined. The amylase activity, which was expressed as a percentage of the activity in the absence of trypsin, was plotted against the trypsin concentration.

 


View larger version (45K):
[in this window]
[in a new window]
 
Fig. 3. Secretion kinetics of AmyQ in protoplasts and rods of the prsA3 mutant and wild-type (wt). (a) Pulse–chase labelling of AmyQ in protoplasts and rods. Radiolabelled AmyQ in cell and medium fractions was analysed by immunoprecipitation, SDS-PAGE and fluorography. The analysed samples correspond to 100 µl protoplast preparation. (b) Densitometric quantification of the AmyQ bands in (a). IOD, integrated optical density. Dark grey bars, cells; light grey bars, medium.

 
Release of AmyQ from protoplasts is retarded in the absence of a cell wall and independent of PrsA
The pulse–chase labelling of rods revealed that practically all labelled AmyQ was chased into the medium (wild-type) or degraded (prsA3 mutant) (Fig. 3). In contrast, AmyQ was inefficiently chased from protoplasts, even after the processing of the signal peptide. About 50 % of the synthesized AmyQ remained associated with the wild-type protoplasts after a 40 min chase (Fig. 3). The retention was associated with significantly less accumulation of AmyQ in the growth medium of protoplasts than rods, between 45 and 15 %, as determined by pulse–chase labelling (Fig. 3b) or enzymically (Fig. 4). Also, in prsA3 protoplasts, there was a similar high level of AmyQ retention as in wild-type protoplasts, up to 60 % (Fig. 3b). Furthermore, despite the large differences in the amount of PrsA protein they contain, the prsA3 and wild-type protoplasts secreted similar amounts of mature AmyQ into the medium, despite the much lower amount secreted by prsA3 rods in comparison with wild-type rods (Figs 3b and 4). Thus, PrsA seems to have only a minor, if any, role in the release mechanism of AmyQ from the cell membrane, whereas the cell wall is either directly or indirectly involved in the release.



View larger version (14K):
[in this window]
[in a new window]
 
Fig. 4. Secretion of AmyQ by protoplasts (open symbols) and rods (filled symbols) of the prsA3 mutant (triangles) and wild-type (squares). The accumulation of AmyQ in the medium was determined enzymically.

 
Ca2+ ions are a component of the cell wall and are thought to catalyse the release of secretory proteins from the cytoplasmic membrane (Leloup et al., 1997). We therefore examined the secretion of AmyQ from protoplasts in the absence and presence of 10 mM CaCl2 by pulse–chase labelling. As shown in Fig. 5, protoplasts of both the prsA3 mutant and wild-type secreted AmyQ into the medium at similar levels and kinetics.



View larger version (29K):
[in this window]
[in a new window]
 
Fig. 5. Effect of calcium on the secretion of AmyQ by protoplasts. Protoplasts were prepared from either the prsA3 mutant or wild-type and subjected to pulse–chase labelling in the presence or absence of 10 mM CaCl2. The medium was SMSb devoid of Ca2+ and with 10 mM Mg2+ to maintain the stability of protoplasts.

 
PrsA is not needed for the folding of AmyQ in the absence of the cell wall
It has been reported (Sanders & May, 1975) that in protoplasts, emerging nascent {alpha}-amylase polypeptide is sensitive to exogenous trypsin but becomes resistant after completion of the folding. To examine the putative role of PrsA in the folding of nascent AmyQ in protoplasts, we determined the trypsin sensitivity of AmyQ secreted by the protoplasts of a strain containing either a low (prsA3 mutant) or high (wild-type overexpressing prsA) level of PrsA protein. Western blotting of the protoplasts showed about a 50-fold difference in the amount of PrsA between the strains (Fig. 6a). The activity of {alpha}-amylase secreted by the protoplasts in the presence of various concentrations of trypsin was determined (see Methods). A significant amount of PrsA was removed from protoplasts in the course of trypsin treatment (Fig. 6a). However, the trypsin sensitivity of the mutant PrsA3 and wild-type PrsA proteins was similar, and the 50-fold difference in the foldase level remained after the treatment. As shown in Fig. 6(b), the presence of trypsin decreased the amount of enzymically active AmyQ in a concentration-dependent manner. The level of PrsA (PrsA3 vs PrsA overproduction) in the protoplast membrane did not affect the profile of AmyQ sensitivity to trypsin, suggesting that the folding of nascent secreted proteins was not enhanced by PrsA in the absence of the cell wall.


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
In the microenvironment of the cell wall matrix, the post-translocational folding of AmyQ is disturbed in the absence of appropriate folding assistance from PrsA (Hyyryläinen et al., 2001; Vitikainen et al., 2001). The results of this study showed that in protoplasts, AmyQ folding into the native, enzymically active, protease-resistant conformation was not disturbed by PrsA depletion. AmyQ was remarkably stable irrespective of the level of PrsA, suggesting that in the absence of the wall, AmyQ folding is independent of PrsA. Unpublished results from our laboratory also inferred that in vitro AmyQ, which was denatured in either 6 M urea or 8 M guanidium hydrochloride and then diluted with a buffer solution, rapidly and fully refolded into the enzymically active conformation. This indicates that, as such, AmyQ is able to fold correctly without any folding assistance, but spontaneous folding is hampered in the microenvironment of the membrane–wall interface and the cell wall. Alternatively, we have considered the possibility that the folding of AmyQ is impaired in prsA3 protoplasts, but it is stabilized by removal of the cell-wall-bound protease(s) that degrade unfolded or misfolded AmyQ along with the cell wall. However, nascent AmyQ was not more sensitive to an exogenous protease, trypsin, in protoplasts with a low level of PrsA (prsA3 mutant) than in those with a high level of PrsA. A PrsA defect thus does not result in significant protein misfolding or slow kinetics of folding in the absence of the wall. This result strongly suggests that the post-translocational folding of AmyQ is PrsA-independent in the absence of the wall.

In rods, nascent AmyQ most likely interacts with the negatively charged polymers in the cell wall and on the cytoplasmic membrane, consequently resulting in AmyQ misfolding that renders it susceptible to the extracytoplasmic quality-control proteases activated by the ensuing folding stress (Hyyryläinen et al., 2001). A likely role of PrsA is to prevent such interactions and thus facilitate the folding. This function may be a chaperone-like function rather than being due to PPIase activity. There is no clear saturation of the PrsA-enhanced AmyQ secretion, when PrsA is produced in high excess as compared with AmyQ (Vitikainen et al., 2001), suggesting that the crucial mechanism of PrsA in AmyQ secretion is a chaperone-like function. Furthermore, the B. licheniformis {alpha}-amylase AmyL, which is like AmyQ secreted in a PrsA-dependent manner (Kontinen & Sarvas, 1993 and unpublished results) does not contain any prolines in the cis conformation (PDB code: 1BLI), again pointing to a non-PPIase mechanism in the PrsA-catalysed {alpha}-amylase secretion.

Many PPIases have, in addition to their prolyl isomerase activity, a chaperone-like activity. This has been shown with all three classes of PPIases, cyclophilins, FK506-binding proteins and parvulins (Arié et al., 2001; Behrens et al., 2001; Bose et al., 1996; Freeman et al., 1996; Freskgard et al., 1992; Ramm & Plückthun, 2000; Scholz et al., 1997). The PPIase activity-independent chaperone activity is located either in the PPIase domain, as in the E. coli FkpA protein (Ramm & Plückthun, 2001), or outside it, as in the E. coli SurA protein (Behrens et al., 2001).

A major effect of the removal of the cell wall was to impair drastically the overall secretion of AmyQ. Significantly, the pulse–chase labelling revealed that a substantial fraction of mature AmyQ (~50 %) remained firmly associated with the protoplast membrane, whereas in rods, AmyQ was completely released into the external medium. We can envisage several possible mechanisms for the defective release of AmyQ from the protoplast membrane. There may be in the cell wall a ‘releasing’ factor that facilitates AmyQ release from the membrane. Its absence in protoplasts could be responsible for the observed retention of AmyQ in the membrane. This operational factor might be a cell-wall-associated protein(s), wall-bound divalent metal ion(s) or the cell wall matrix itself. Retardation in protoplasts of another secreted protein, AmyE of B. subtilis, has been described previously, and a mechanism coupled with folding specifically dependent on calcium ions was suggested (Haddaoui et al., 1997; Leloup et al., 1999). However, the calcium-enhanced release is most likely not a general release mechanism in the secretion of proteins since, in our study, the secretion of AmyQ from protoplasts was found to be independent of calcium. Consistent with this conclusion, it has been shown previously that divalent cations facilitate the production of enzymically active protease by protoplasts of B. amyloliquefaciens, whereas they have no effect on the production of {alpha}-amylase (Sanders & May, 1975). We could not determine whether Mg2+ could replace Ca2+, because of the fragility of the protoplasts in the absence of Mg2+.

As a negatively charged matrix, the cell wall may facilitate protein release from the membrane by electronegative interactions. Alternatively, removal of the cell wall may perturb functions of the cell membrane such as proton motive force, protein translocation, signal peptide cleavage or expression of proteases associated with the wall and wall–membrane interface and thereby indirectly affect the AmyQ release from the membrane. We observed a moderate defect of the signal peptide processing, but this is hardly the reason for the AmyQ retention, since mutants that are defective in the signal peptide processing such as ecs (Pummi et al., 2002) do not cause retention of mature AmyQ in the cell membrane. The high metabolic activity of the protoplasts in terms of protein synthesis excludes major dysfunctions of the membrane. Our results also indicated that PrsA is not involved in the release mechanism.

The pulse–chase-labelling method of protoplasts described above can be used to study late stages of protein secretion in B. subtilis by making the components involved accessible to external manipulation and addressing the role of the cell wall matrix. The main problem of the method is the considerable lysis of protoplasts during their preparation and handling. This is, however, mitigated by their stability during pulse–chase labelling. Another limitation is the inability of lysozyme to remove all components of the cell wall; the protoplasts most likely still contain membrane-bound lipoteichoic acids and precursors of the peptidoglycan, which, in the absence of the rest of the wall matrix, may present non-physiological functions. Lysozyme may partially neutralize the negative charge on the protoplast surface and thereby influence the properties of the protoplasts.


   ACKNOWLEDGEMENTS
 
This work was supported by the Biotech grant Bio-CT96-0097 from the Commission of the European Union (CEU). We thank D. Smart for correction of the English.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Archibald, A. R., Hancock, I. C. & Harwood, C. R. (1993). Cell wall structure, synthesis and turnover. In Bacillus subtilis and Other Gram-positive Bacteria, pp. 381–410. Edited by A.L. Sonenshein, J.A. Hoch & R. Losick. Washington, DC: American Society for Microbiology.

Arié, J. P., Sassoon, N. & Betton, J. M. (2001). Chaperone function of FkpA, a heat shock prolyl isomerase, in the periplasm of Escherichia coli. Mol Microbiol 39, 199–210.[CrossRef][Medline]

Behrens, S., Maier, R., de Cock, H., Schmid, F. X. & Gross, C. A. (2001). The SurA periplasmic PPIase lacking its parvulin domains functions in vivo and has chaperone activity. EMBO J 20, 285–294.[Abstract/Free Full Text]

Beveridge, T. J. & Murray, R. G. (1980). Sites of metal deposition in the cell wall of Bacillus subtilis. J Bacteriol 141, 876–887.[Medline]

Bolhuis, A., Tjalsma, H., Smith, H. E., de Jong, A., Meima, R., Venema, G., Bron, S. & van Dijl, J. M. (1999). Evaluation of bottlenecks in the late stages of protein secretion in Bacillus subtilis. Appl Environ Microbiol 65, 2934–2941.[Abstract/Free Full Text]

Bose, S., Weikl, T., Bugl, H. & Buchner, J. (1996). Chaperone function of Hsp90-associated proteins. Science 274, 1715–1717.[Abstract/Free Full Text]

Demchick, P. & Koch, A. L. (1996). The permeability of the wall fabric of Escherichia coli and Bacillus subtilis. J Bacteriol 178, 768–773.[Abstract]

Freeman, B. C., Toft, D. O. & Morimoto, R. I. (1996). Molecular chaperone machines: chaperone activities of the cyclophilin Cyp-40 and the steroid aporeceptor-associated protein p23. Science 274, 1718–1720.[Abstract/Free Full Text]

Freskgard, P. O., Bergenhem, N., Jonsson, B. H., Svensson, M. & Carlsson, U. (1992). Isomerase and chaperone activity of prolyl isomerase in the folding of carbonic anhydrase. Science 258, 466–468.[Medline]

Haddaoui, E. A., Leloup, L., Petit-Glatron, M. F. & Chambert, R. (1997). Characterization of a stable intermediate trapped during reversible refolding of Bacillus subtilis {alpha}-amylase. Eur J Biochem 249, 505–509.[Abstract]

Harwood, C. R. & Cutting, S. M. (1990). Molecular Biological Methods for Bacillus. New York: Wiley.

Hyyryläinen, H.-L., Vitikainen, M., Thwaite, J., Wu, H., Sarvas, M., Harwood, C., Kontinen, V. & Spehenson, K. (2000). D-Alanine substitution of teichoic acids as a modulator of protein folding and stability at the cytoplasmic membrane/cell wall interface of Bacillus subtilis. J Biol Chem 275, 26696–26703.[Abstract/Free Full Text]

Hyyryläinen, H. L., Bolhuis, A., Darmon, E. & 8 other authors (2001). A novel two-component regulatory system in Bacillus subtilis for the survival of severe secretion stress. Mol Microbiol 41, 1159–1172.[CrossRef][Medline]

Jacobs, M., Andersen, J. B., Kontinen, V. & Sarvas, M. (1993). Bacillus subtilis PrsA is required in vivo as an extracytoplasmic chaperone for secretion of active enzymes synthesized either with or without pro-sequences. Mol Microbiol 8, 957–966.[Medline]

Kontinen, V. & Sarvas, M. (1993). The PrsA lipoprotein is essential for protein secretion in Bacillus subtilis and sets a limit for high-level secretion. Mol Microbiol 8, 727–737.[Medline]

Kontinen, V. P., Saris, P. & Sarvas, M. (1991). A gene (prsA) of Bacillus subtilis involved in a novel, late stage of protein export. Mol Microbiol 5, 1273–1283.[Medline]

Leloup, L., el Haddaoui, A., Chambert, R. & Petit-Glatron, M. F. (1997). Characterization of the rate-limiting step of the secretion of Bacillus subtilis {alpha}-amylase overproduced during the exponential phase of growth. Microbiology 143, 3295–3303.[Abstract]

Leloup, L., Le Saux, J., Petit-Glatron, M.-F. & Chambert, R. (1999). Kinetics of the secretion of Bacillus subtilis levanase overproduced during the exponential phase of growth. Microbiology 145, 613–619.[Abstract]

Leskelä, S., Wahlström, E., Hyyryläinen, H. L., Jacobs, M., Palva, A., Sarvas, M. & Kontinen, V. P. (1999a). Ecs, an ABC transporter of Bacillus subtilis: dual signal transduction functions affecting expression of secreted proteins as well as their secretion. Mol Microbiol 31, 533–543.[CrossRef][Medline]

Leskelä, S., Wahlström, E., Kontinen, V. P. & Sarvas, M. (1999b). Lipid modification of prelipoproteins is dispensable for growth but essential for efficient protein secretion in Bacillus subtilis: characterization of the lgt gene. Mol Microbiol 31, 1075–1085.[CrossRef][Medline]

Lounatmaa, K. (1985). Electron microscopic methods for the study of bacterial surface structures. In Enterobacterial Surface Antigens. Methods for Molecular Characterisation, pp. 243–261. Edited by T. K. Korhonen, E. A. Dawes & P. H. Mäkelä. Amsterdam: Elsevier.

Manting, E. H. & Driessen, A. J. M. (2000). Escherichia coli translocase: the unravelling of a molecular machine. Mol Microbiol 37, 226–238.[CrossRef][Medline]

Margot, P. & Karamata, D. (1996). The wprA gene of Bacillus subtilis 168, expressed during exponential growth, encodes a cell-wall-associated protease. Microbiology 142, 3437–3444.[Abstract]

Meens, J., Frings, E., Klose, M. & Freudl, R. (1993). An outer membrane protein (OmpA) of Escherichia coli can be translocated across the cytoplasmic membrane of Bacillus subtilis. Mol Microbiol 9, 847–855.[Medline]

Merchante, R., Pooley, H. M. & Karamata, D. (1995). A periplasm in Bacillus subtilis. J Bacteriol 177, 6176–6183.[Abstract]

Palva, I. (1982). Molecular cloning of {alpha}-amylase gene from Bacillus amyloliquefaciens and its expression in B. subtilis. Gene 19, 81–87.[CrossRef][Medline]

Perego, M., Glaser, P., Minutello, A., Strauch, M. A., Leopold, K. & Fischer, W. (1995). Incorporation of D-alanine into lipoteichoic acid and wall teichoic acid in Bacillus subtilis. Identification of genes and regulation. J Biol Chem 270, 15598–15606.[Abstract/Free Full Text]

Peschel, A., Otto, M., Jack, R. W., Kalbacher, H., Jung, G. & Götz, F. (1999). Inactivation of the dlt operon in Staphylococcus aureus confers sensitivity to defensins, protegrins and other antimicrobial peptides. J Biol Chem 274, 8405–8410.[Abstract/Free Full Text]

Petit-Glatron, M.-F., Grajcar, L., Munz, A. & Chambert, R. (1993). The contribution of the cell wall to a transmembrane calcium gradient could play a key role in Bacillus subtilis protein secretion. Mol Microbiol 9, 1097–1106.[Medline]

Pummi, T., Leskelä, S., Wahlström, E., Gerth, U., Tjalsma, H., Hecker, M., Sarvas, M. & Kontinen, V. P. (2002). ClpXP protease regulates the signal peptide cleavage of secretory preproteins in Bacillus subtilis with a mechanism distinct from that of the Ecs ABC transporter. J Bacteriol 184, 1010–1018.[Abstract/Free Full Text]

Puohiniemi, R., Simonen, M., Muttilainen, S., Himanen, J. P. & Sarvas, M. (1992). Secretion of the Escherichia coli outer membrane proteins OmpA and OmpF in Bacillus subtilis is blocked at an early intracellular step. Mol Microbiol 6, 981–990.[Medline]

Rahfeld, J.-U., Rücknagel, K. P., Schelbert, B., Ludvig, B., Hacker, J., Mann, K. & Fischer, G. (1994). Confirmation of the existence of a third family among peptidyl-prolyl cis/trans isomerases. Amino acid sequence and recombinant production of parvulin. FEBS Lett 352, 180–184.[CrossRef][Medline]

Ramm, K. & Plückthun, A. (2000). The periplasmic Escherichia coli peptidyl-prolyl cis/trans-isomerase FkpA. II. Isomerase-independent chaperone activity in vitro. J Biol Chem 275, 17106–17113.[Abstract/Free Full Text]

Ramm, K. & Plückthun, A. (2001). High enzymatic activity and chaperone function are mechanistically related features of the dimeric E. coli peptidyl-prolyl-isomerase FkpA. J Mol Biol 310, 485–498.[CrossRef][Medline]

Sanders, R. L. & May, B. K. (1975). Evidence for extrusion of unfolded extracellular enzyme polypeptide chains through membranes of Bacillus amyloliquefaciens. J Bacteriol 123, 806–814.[Medline]

Scholz, C., Stoller, G., Zarnt, T., Fischer, G. & Schmid, F. X. (1997). Cooperation of enzymatic and chaperone functions of trigger factor in the catalysis of protein folding. EMBO J 16, 54–58.[Abstract/Free Full Text]

Smith, T. J., Blackman, S. A. & Foster, S. J. (2000). Autolysins of Bacillus subtilis: multiple enzymes with multiple functions. Microbiology 146, 249–262.[Free Full Text]

Stephenson, K., Carter, N. M., Harwood, C. R., Petit-Glatron, M. F. & Chambert, R. (1998). The influence of protein folding on late stages of the secretion of {alpha}-amylases from Bacillus subtilis. FEBS Lett 430, 385–389.[CrossRef][Medline]

Tjalsma, H., Bolhuis, A., Jongbloed, J. D., Bron, S. & van Dijl, J. M. (2000). Signal peptide-dependent protein transport in Bacillus subtilis: a genome-based survey of the secretome. Microbiol Mol Biol Rev 64, 515–547.[Abstract/Free Full Text]

Vitikainen, M., Pummi, T., Airaksinen, U., Wu, H., Sarvas, M. & Kontinen, V. P. (2001). Quantitation of the capacity of the secretion apparatus and requirement for PrsA in growth and secretion of {alpha}-amylase in Bacillus subtilis. J Bacteriol 183, 1881–1890.[Abstract/Free Full Text]

Yoshida, K., Sano, H., Seki, S., Oda, M., Fujimura, M. & Fujita, Y. (1995). Cloning and sequencing of a 29 kb region of the Bacillus subtilis genome containing the hut and wapA loci. Microbiology 141, 337–343.[Abstract]

Received 7 February 2002; revised 27 September 2002; accepted 18 November 2002.