![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Vol. 12, No. 15, pp. 2318-2331, August 1, 1998
1 Department of Genetics, Groningen Biomolecular Sciences and Biotechnology Institute, University of Groningen, 9751 NN Haren, The Netherlands; 2 Institute of Genetics, University of Bayreuth, D-95440 Bayreuth, Germany; 3 Genencor International, 2288 GJ Rijswijk, The Netherlands; 4 Department of Pharmaceutical Biology, University of Groningen, 9713 AV Groningen, The Netherlands
 |
Abstract |
|---|
Top
Abstract
Introduction
Results
Discussion
Materials & Methods
References
|
|---|
Approximately 47% of the genes of the Gram-positive bacterium Bacillus subtilis belong to paralogous gene families. The present studies were aimed at the functional analysis of the sip gene family of B. subtilis, consisting of five chromosomal genes, denoted sipS, sipT, sipU, sipV, and sipW. All five sip genes specify type I signal peptidases (SPases), which are actively involved in the processing of secretory preproteins. Interestingly, strains lacking as many as four of these SPases could be obtained. As shown with a temperature-sensitive SipS variant, only cells lacking both SipS and SipT were not viable, which may be caused by jamming of the secretion machinery with secretory preproteins. Thus, SipS and SipT are of major importance for protein secretion. This conclusion is underscored by the observation that only the transcription of the sipS and sipT genes is temporally controlled via the DegS-DegU regulatory system, in concert with the transcription of most genes for secretory preproteins. Notably, the newly identified SPase SipW is highly similar to SPases from archaea and the ER membrane of eukaryotes, suggesting that these enzymes form a subfamily of the type I SPases, which is conserved in the three domains of life.
[Key Words: Bacillus subtilis; paralogous gene family; protein secretion; signal peptidase; leader peptidase]
| |
Introduction |
|---|
|
Top
Abstract
Introduction
Results
Discussion
Materials & Methods
References
|
|---|
Since 1995, at least 12 microbial genomes have been completely
sequenced and annotated (for a recent update, see the microbial database at
http://www.tigr.org/tdb/mdb/mdb.html).
A remarkable common property of these genomes is that many genes are
present in multiple (paralogous) copies. A good example is the genome of Bacillus subtilis, which contains ~4200 genes, 47% of
which belong to paralogous gene families (Kunst et al. 1997
). Close examination of the corresponding proteins of B. subtilis has
indicated that many of these are involved in transcription regulation,
and the transport of compounds into and out of the cell. Interestingly, most genes for components of the protein secretion machinery are present in only one copy, despite the fact that B. subtilis
has a large capacity for protein secretion. The only known exception concerns the genes for type I signal peptidases (SPases; Tjalsma et al.
1997). These SPases remove amino-terminal signal peptides from
secretory preproteins during, or shortly after their translocation across the cytoplasmic membrane, to release these proteins from the
trans side of this membrane (for review, see Dalbey et al. 1997).
Homologous type I SPases have been identified in archaea,
Gram-positive, and Gram-negative eubacteria, the inner membrane of
yeast mitochondria, the thylakoid membrane of chloroplasts, and the
endoplasmic reticular (ER) membranes of yeast and higher eukaryotes.
Despite the fact that considerable similarities exist between the known
type I SPases when amino acid sequences are compared pairwise, only few
residues are strictly conserved in all known enzymes of this family
(Dalbey et al. 1997). In particular, the type I SPases of eubacteria,
mitochondria, and chloroplasts differ considerably from their homologs
in archaea and the ER membrane, indicating that these enzymes belong to
distinct subfamilies of SPases. Below, we have referred to these
subfamilies as the P (prokaryotic)- and ER-type SPases,
respectively. Most P-type SPases contain conserved serine and lysine
residues that are essential for enzymatic activity, possibly by forming
a catalytic dyad (Black 1993; Tschantz et al. 1993; van Dijl et al.
1995; Paetzel and Dalbey 1997). The putative active site serine residue
of the P-type SPases is also conserved in the ER-type SPases. In
contrast, the putative active site lysine residue of the P-type SPases
is replaced with a histidine residue in the ER-type SPases (Dalbey and
von Heijne 1992; van Dijl et al. 1992).
In many eubacteria, such as Escherichia coli (Blattner et al.
1997), Haemophilus influenzae (Fleischmann et al. 1995),
Helicobacter pylori (Tomb et al. 1997), and Mycobacterium
tuberculosis (Philipp et al. 1996), one P-type SPase seems to be
sufficient for the processing of secretory pre-proteins. Similarly, the
archaea Methanococcus jannaschii (Bult et al. 1996) and
Methanobacterium thermoauthotrophicum (Smith et al 1997), and
the yeast Saccharomyces cerevisiae (Goffeau et al. 1997)
contain only one ER-type SPase. As shown for E. coli (Dalbey
and Wickner 1985; van Dijl et al. 1988) and yeast (Böhni et al.
1988), these SPases are likely to be essential for cell viability. In
contrast, two paralogous P-type SPases are found in
Synechocystis PCC 6803 (Kaneko et al. 1996) and the inner
membrane of yeast mitochondria (Behrens et al. 1991; Nunnari et al.
1993), whereas the presence of two paralogous ER-type SPases appears to
be characteristic for most eukaryotic species (Dalbey et al. 1997).
Finally, the largest numbers of paralogous SPases appear to be present
in the archaeon Archaeoglobus fulgidus, which contains three
genes for ER-type SPases (Klenk et al. 1997), and the Gram-positive bacterium B. subtilis in which seven sip genes for
type I SPases have been identified. Five of the sip genes of
B. subtilis (denoted sipS, sipT, sipU, sipV, and
sipW) are located on the chromosome (van Dijl et al. 1992;
Tjalsma et al. 1997); two additional sip genes (denoted
sipP) are located on plasmids which were, thus far, only found
in natto-producing strains of B. subtilis (Meijer et al.
1995).
Thus far, the presence of both P- and ER-type SPases in one organism
was only evident for eukaryotes, which contain P-type SPases in their
organelles and ER-type SPases in the ER (Dalbey et al. 1997). In this
paper we document our surprising observation that one of the seven type
I SPases of B. subtilis (ie. SipW) belongs to the ER-type
subfamily of SPases. Thus, B. subtilis is the first organism
known to contain SPases of the P- and the ER-type in one membrane.
Furthermore, our studies, aimed at the functional analysis of the type
I SPases of B. subtilis, show that SipS and SipT are the most
important SPases of B. subtilis, whereas SipU, SipV, and SipW
appear to have a minor role in precursor processing.
| |
Results |
|---|
|
Top
Abstract
Introduction
Results
Discussion
Materials & Methods
References
|
|---|
SipW belongs to the subfamily of ER-type SPases
To make an inventory of the known type I SPases, database searches
were performed with various P- and ER-type SPases. Unexpectedly, similarity searches with the ER-type SPase of M. jannaschii
showed that the yqhE gene of B. subtilis (GenBank
accession no. D84432) specifies a protein with a high degree of
sequence similarity to the known ER-type SPases from archaea and
eukaryotes. One of the SPases of A. fulgidus (GenBank locus
AF1655) exhibited the highest degree of similarity with the deduced
YqhE protein sequence (57% identical residues and conservative
replacements in 131 residues). This observation strongly suggested that
the yqhE gene specifies an ER-type SPase of B. subtilis. By analogy to other SPase-encoding genes of bacilli, we
therefore renamed the yqhE gene sipW (Fig. 1A). As shown previously for the ER-type SPases of
yeast (i.e., Sec11p) and Canis familiaris (i.e., SPC18 and
SPC21; see Dalbey et al. 1997), SipW also shows amino acid sequence
similarity with P-type SPases, but this similarity is mainly limited to
the conserved domains B-E, which are present in all known SPases (Fig.
1B). Like the other known ER-type SPases, SipW contains a conserved histidine residue in domain D instead of the conserved lysine residue
in domain D of the P-type SPases (Fig. 1A, B). Finally, SipW seems to
contain one amino-terminal (AI), and one carboxy-terminal (AII)
membrane anchor (Fig. 1C), as suggested previously for the P-type
SPases of Rhodobacter capsulatus and yeast mitochondria (Dalbey et al. 1997).
|
The availability of the sequences of SipW of B. subtilis and
recently identified SPases from archaea allowed the detailed comparison
of the ER-type SPases, which is documented in Figure 1A. The comparison
shows that these enzymes belong to a heterogeneous group of proteins.
As documented previously for the P-type SPases, only the domains B-E
are conserved in the ER-type SPases (Fig. 1A,B). A striking difference
with the P-type SPases is that the domains B and C are separated by
only one residue, and the domains D and E are separated by only 2-11
residues (Fig. 1A,C). The corresponding domains of the P-type SPases
are separated by 19-42, and 23-118 residues, respectively (see Dalbey
et al. 1997). Finally, the number of putative membrane spanning domains
seems to vary in the ER-type SPases. All known enzymes of this
subfamily have one amino-terminal membrane anchor (AI, Fig. 1A). In
addition, some of these SPases, such as SipW of B. subtilis
and SPC21 of A. fulgidus, seem to have a carboxy-terminal
membrane anchor (AII, Fig. 1A,C), and a second SPase of A. fulgidus [Sip (Afu)] seems to have even three carboxy-terminal
membrane anchors (data not shown).
Two quadruple SPase mutants of B. subtilis are viable
To determine whether SipW is required for growth or viability of
the cell, an internal SacI-NdeI fragment of the
chromosomal sipW gene, specifying the conserved domains C-E,
was replaced with a tetracycline resistance (Tcr) marker
(schematically shown in Fig. 2). As shown previously for strains lacking intact sipS, sipT, sipU, or sipV
genes (Bolhuis et al. 1996; Tjalsma et al. 1997), the disruption of the
sipW gene had no detectable effect on cell growth or viability
(data not shown).
|
To investigate which SPases are needed for growth and cell viability,
various chromosomal sip gene deletions or disruptions were
combined. The double, triple, and quadruple sip mutants
obtained are listed in Table 1, which
shows that most sip gene mutations could be combined, with one
exception: It was not possible to construct strains lacking both
sipS and sipT. Interestingly, the quadruple mutants
jointly lacking sipS, sipU, sipV, and sipW (B. subtilis 
SUVW), or sipT, sipU, sipV, and
sipW (B. subtilis TUVW) showed similar growth
rates as the parental strain B. subtilis 8G5, irrespective of
the growth medium used (data not shown). Thus, it seems that the
secretory precursor processing machinery is functionally redundant, and
that one SPase (i.e., SipS or SipT) is sufficient for growth and cell
viability of B. subtilis.
|
Functional SipS is essential for viability of a B. subtilis strain lacking SipT
To evaluate the importance of SipS and SipT, a strain was
constructed that produces a temperature-sensitive SipS mutant protein, but lacks wild-type SipS and SipT. To this purpose, we first selected temperature-sensitive variants from a large collection of previously constructed site-specific mutants of SipS (van Dijl et al. 1995). By
use of a plate assay for SipS activity (van Dijl et al. 1992), five
temperature-sensitive mutants were identified, which are active at
30°C, but not at 48°C (Table 2). These contain
the L74A, Y81A, R84A, R84H, or D146A mutations, respectively. In a second step, B. subtilis 8G5 sipS, which lacks the
chromosomal sipS gene (Bolhuis et al. 1996), was transformed
with plasmids specifying the L74A, Y81A, R84A, R84H, or D146A SipS
mutant proteins. In a third and final step, the sipT gene of
the latter transformants was disrupted with a chloramphenicol
resistance (Cmr) marker, as described previously (Tjalsma et
al. 1997). The resulting mutant strains (ST
sipSts) showed no growth defects at 30°C,
37°C, and 42°C (data not shown). However, after a temperature
shift to 48°C during the exponential growth phase, all five ST
sipSts strains stopped growing when they reached the
transition phase between exponential and postexponential growth and, on
continued incubation at 48°C, they even started to lyse. In
contrast, the control strain ST (pGDL41), producing wild-type
SipS, continued to grow in the postexponential growth phase (Fig.
3A). Because all ST sipSts
mutant strains showed similar characteristics with respect to growth
and cell viability, only one of these strains [i.e., ST (pS-D146A)] was selected for further analyses. As shown by Western blotting, the cellular levels of wild-type SipS were not affected by
the temperature shift from 37°C to 48°C. In contrast, the
cellular levels of SipS-D146A, which were already reduced a 37°C,
were decreased to levels below detection at 48°C (Fig. 3B). In
summary, these observations show that for growth and cell viability of B. subtilis, a functional SipS is required in the absence of
SipT.
|
|
Accumulation of secretory preproteins in the absence of functional SipS and SipT
As a first approach to monitor the effects of the absence of
functional SipS and SipT on the processing of secretory preproteins, we
analyzed the processing of pre(A13i)-
-lactamase. This hybrid precursor is most efficiently processed by SipS and SipT (van Dijl et
al. 1992; Tjalsma et al. 1997). As shown in Figure 3, B and C, the
disappearance of the SipS-D146A mutant protein in B. subtilis
ST (pS-D146A) at 48°C was paralleled by the accumulation of
pre(A13i)--lactamase. In contrast, cells of B. subtilis
ST (pGDL41; wild-type SipS) contained much smaller amounts of this precursor and, in addition, they contained the corresponding mature protein (Fig. 3C). These findings show that processing of
pre(A13i)--lactamase is strongly affected in the absence of
functional SipS and SipT.
To investigate the effects of depletion of SipS in cells lacking SipT,
two strains with a disrupted sipT gene were constructed in
which the transcription of sipS or sipS-D146A is
controlled by the xylose-inducible xylA promoter. In these
strains, denoted STxS and STxS-D146A, respectively, the
wild-type sipS gene was removed from its original locus and
the xylA-sipS or xylA-sipS-D146A constructs were
integrated into the amyE locus (Fig. 4A).
Unexpectedly, even in the absence of xylose, both strains showed normal
growth at 37°C and 42°C, suggesting that the xylA
promoter was not completely repressed in the absence of xylose (data
not shown). In fact, the inhibition of growth and subsequent cell lysis
was only observed for B. subtilis STxS-D146A on
incubation at 48°C in the absence of xylose (data not shown),
similar to B. subtilis ST (pS-D146A) (Fig. 3A). As shown
by Western blotting experiments with cells in the postexponential
growth phase, the cellular levels of SipS or SipS-D146A in B. subtilis STxS and STxS-D146A, respectively, depended on
the presence or absence of xylose. Compared with the parental strain
B. subtilis 8G5, B. subtilis STxS contained
slightly reduced levels of SipS in the absence of xylose, whereas SipS was about five-fold overproduced in the presence of xylose. In contrast, in B. subtilis STxS-D146A, only very low
levels of SipS-D146A were detectable in the presence of xylose,
whereas SipS-D146A was not detectable in the absence of xylose (Fig.
4B). The latter observation suggests that very small amounts of
SipS-D146A are sufficient for growth and cell viability.
|
To determine the effects of various levels of SipS (-D146A)
activity on the accumulation of secretory precursor proteins in cells
lacking SipT, B. subtilis STxS and STxS-D146A were
transformed with plasmid pKTH10 that specifies the Bacillus
amyloliquefaciens 
-amylase AmyQ. The presence of pKTH10
results in the accumulation of pre-AmyQ, even in wild-type cells (Palva
1982; Kontinen and Sarvas 1988). As shown by Western blotting, cells of
B. subtilis STxS-D146A, which were incubated at 48°C
in the absence of xylose, accumulated strongly increased amounts of
pre-AmyQ (Fig. 4C). In contrast, at 37°C, cells of the latter strain
accumulated similar amounts of pre-AmyQ as B. subtilis
STxS, producing close to wild-type levels of SipS (37°C and
48°C; Fig. 4C), or overproducing SipS (data not shown). Taken
together, our observations show that the presence of SipS and SipT is
required for efficient processing of pre-AmyQ, and that in cells
lacking SipT, very small amounts of SipS, which are below the level of
detection, are sufficient to prevent the accumulation of pre-AmyQ at
high levels.
Impaired translocation of pre-AmyQ in cells lacking functional SipS and SipT
To investigate whether SipS and SipT depletion affects the
translocation of pre-AmyQ, the sequences specifying the
biotin-accepting domain of a transcarboxylase from
Propionibacterium shermannii (PSBT; Jander et al. 1996) were
fused to the 3' end of the amyQ gene on pKTH10. The
resulting plasmid pKTH10-BT was used to transform B. subtilis
STxS-D146A and, subsequently, the accumulation of biotinylated
(pre-)AmyQ-PSBT was determined. The rationale of this experiment is
that pre-AmyQ-PSBT can only be biotinylated by the cytoplasmic
biotin-ligase if the PSBT domain folds into its native
three-dimensional structure in the cytoplasm. This will only happen if
the translocation of pre-AmyQ-PSBT across the membrane is impaired. As
shown in Figure 5A, processing of pre-AmyQ-PSBT
occurred efficiently in cells of B. subtilis STxS-D146A growing at 37°C in the absence of xylose, as only small amounts of
pre-AmyQ-PSBT were detectable. Under these conditions, mature nonbiotinylated AmyQ-PSBT was secreted into the growth medium (data
not shown). Similar results were obtained with the parental strain 8G5,
both at 37°C and 48°C (data not shown). In contrast, high levels
of pre-AmyQ-PSBT accumulated in cells of B. subtilis STxS-D146A incubated in the absence of xylose at 42°C
(permissive temperature for growth) or 48°C (nonpermissive
temperature) (Fig. 5B, top panel). As shown with a
streptavidin-horseradish peroxidase (HRP) conjugate, significant
amounts of the accumulating pre-AmyQ-PSBT were biotinylated, showing
that the translocation of these precursor molecules was impaired (Fig.
5B, bottom panel). Compared with cells of B. subtilis
STxS-D146A, which were grown in the absence of xylose at 37°C,
cells grown at 42°C, or 48°C accumulated significantly reduced
levels of mature AmyQ-PSBT; although mature AmyQ-PSBT was produced in
the absence of xylose at 42°C or 48°C, it was not detectably
biotinylated (Fig 5A,B).
|
Relative contributions of SipU, SipV, and SipW to the processing of pre-AmyQ
We have reported previously that the processing of pre-AmyQ is
stimulated in the absence of SipS, and retarded in the absence of SipT,
indicating that the production of SipS interferes with pre-AmyQ
processing, and that this precursor is a preferred substrate of SipT
(Tjalsma et al. 1997). To examine the relative contributions of SipU,
SipV, and SipW to the processing of pre-AmyQ in the absence of SipS or
SipT, Western blotting and pulse-chase labeling experiments were
performed with strains lacking multiple SPases. The parental strain 8G5
was used as a control. As the results obtained by pulse-chase labeling
confirmed those obtained by Western blotting, only the latter are
documented. As shown in Figure 6A, cells of B. subtilis S, or SU, accumulated only very low levels of
pre-AmyQ. Similar results were obtained with cells of B. subtilis SV and SW (data not shown). The level of
pre-AmyQ was slightly increased in cells of B. subtilis
SUV, but remained lower than in the parental strain (Fig. 6A). The
levels of pre-AmyQ accumulation in B. subtilis SUW and
SVW were similar to those in B. subtilis SUV (data not shown). Compared with B. subtilis T, which
accumulates more pre-AmyQ than the parental strain, cells of B. subtilis TU and TUV showed no additional accumulation of
pre-AmyQ (Fig. 6A), and similar results were obtained with B. subtilis TV, TW, and TUW (data not shown). The
levels of pre-AmyQ in cells of B. subtilis SUV or
TUV could only be increased further by disruption of the
sipW gene, showing that SipW is actively involved in the
processing of pre-AmyQ (Fig. 6B). In conclusion, these observations
show that the accumulation of pre-AmyQ in cells lacking either SipS or
SipT was significantly exacerbated only by the simultaneous removal of
SipU, SipV, and SipW.
|
Transcription of only sipS and sipT is temporally regulated by the DegS-DegU system
The transcription of the sipS and sipT genes is
temporally controlled, whereas the sipU and sipV
genes are constitutively transcribed (Bolhuis et al. 1996; Tjalsma et
al. 1997). To analyze the transcription of the sipW gene, a
transcriptional sipW-lacZ fusion was constructed, and
introduced into the chromosome of B. subtilis 8G5
(schematically shown in Fig. 7A). Cells of the resulting strain were grown in TY or minimal medium, and samples withdrawn at hourly intervals were assayed for -galactosidase activity. The results show that, irrespective of the growth medium used, sipW is constitutively transcribed at a low level (only the results for minimal medium are shown in Fig. 7B, 
).
|
Because it was shown previously that the temporal control of
sipS transcription is exerted by the DegS-DegU two-component regulatory system (Bolhuis et al. 1996), which is also required for the
transcription of most genes for secreted degradative enzymes (for
review, see Ferrari et al. 1993), we investigated whether this
regulatory system is also involved in the transcription of the
sipT, sipU, sipV, and sipW genes. To this purpose,
the degU32(Hy) mutation, which leads to increased
transcription of genes for degradative enzymes (Ferrari et al. 1993;
Msadek et al. 1993) and sipS (Bolhuis et al. 1996), was
introduced in strains carrying chromosomal sipS-lacZ, sipT-lacZ,
sipU-lacZ, sipV-lacZ, or sipW-lacZ gene fusions. The
resulting strains were grown in minimal medium and samples, withdrawn
at hourly intervals, were assayed for -galactosidase activity. The
results show that only the transcription of sipS and
sipT was stimulated in strains containing the
degU32(Hy) mutation (Fig. 7B,C; indicated with 
and 
respectively). In contrast, the degU32(Hy) mutation had no
effect on the transcription of sipU, sipV, and sipW
(Fig. 7B,C; indicated with 
, 
, , respectively).
To verify the involvement of the DegS-DegU system in the transcription
of the sipT gene, Northern hybridization experiments were
performed with total RNA from the B. subtilis strains 8G5, 8G5
degU32(Hy), and 8G5 degS-degU. In contrast to
sipS, for which two transcripts were detectable (Bolhuis et
al. 1996), only one, monocistronic, sipT-specific transcript
was detected. Compared with the parental strain 8G5, the amount of the
sipT-specific transcript was about twofold increased in cells
with the degU32(Hy) mutation, although it was hardly
detectable in cells containing the degS-degU deletion (Fig.
7D). These findings show that the transcription of the sipT
gene is controlled by the DegS-DegU system. Consistent with this
conclusion, we identified the sequence: 5'-TATGATAAAGTA-3' at a
position 46 nucleotides upstream of the translational start of
sipT. This sequence contains one mismatch and one base
insertion compared with a consensus sequence (AMTGAAMAKWW in which
K = G or T; M = A or C; W = A or T), which has been
implicated in the DegU-dependent control of gene expression (Jacobs
1995). Taken together, these results show that the transcription of
sipS and sipT is concerted with that of the genes for
secreted degradative enzymes. In contrast, the sipU, sipV, and
sipW genes are constitutively transcribed at a low level.
| |
Discussion |
|---|
|
Top
Abstract
Introduction
Results
Discussion
Materials & Methods
References
|
|---|
Various microorganisms contain paralogous type I SPases.
Interestingly, this seems to be a common feature of Gram-positive eubacteria in particular. For example, B. amyloliquefaciens
(Hoang and Hofemeister 1995; Meijer et al. 1995), Staphylococcus
aureus (Cregg et al. 1996) and Streptomyces lividans (S. Schacht and J. Anné, pers. comm.) contain at least two paralogous
SPases of the P-type, whereas Deinococcus radiodurans seems to
contain at least three of these enzymes (sequences are available via
ftp://ftp.tigr.org/pub/data/d_radiodurans). Notably, we have identified one ER-type and six P-type
SPases in B. subtilis. In the present studies, which were
aimed at the functional analysis of the chromosomally encoded SPases of
B. subtilis, we show that, in principle, the presence of one
SPase (i.e., SipS or SipT) is sufficient for precursor processing,
growth, and cell viability. Thus, the secretory precursor processing
machinery of B. subtilis is functionally redundant, ensuring
that this organism can avail of backup SPases. Under natural
conditions, this may be of particular importance for the fitness of
B. subtilis and other Gram-positive eubacteria, which can
secrete large amounts of proteins into the medium as an adaptive
response to changes in the environment.
Our results show that SipS and SipT play a major role in the processing
of secretory preproteins and that, under the conditions tested, the
three other chromosomally encoded SPases (SipU, SipV, and SipW)
contribute only to a minor extent to this process. Interestingly, the
transcription of sipS (Bolhuis et al. 1996) and sipT
(this study) is temporally controlled via the DegS-DegU two-component regulatory system, in concert with the transcription of the genes for
most secretory proteins. Thus, B. subtilis can increase its preprotein processing capacity under conditions of high-level protein
secretion in the postexponential growth phase. Our observation that
cells lacking SipS and SipT stop growing, and die in the transition
phase between exponential and postexponential growth, indicates that
the potential to increase the processing capacity is important for cell
growth and viability. Nevertheless, the production of very small
amounts of SipS-D146A, a temperature-sensitive variant of SipS, is
sufficient for growth and cell viability. This suggests that the
increased processing capacity provided by SipS and SipT relates to, as
yet undefined, differences between the specificities of these two
SPases and those of SipU, SipV, and/or SipW. This view is
supported by the observation that SipU, SipV, and SipW cannot
complement for the absence of SipS and SipT, even if the corresponding
genes are placed on multi-copy plasmids (H. Tjalsma et al., unpubl.).
In this respect, it is important to note that we have shown previously
that SipS, SipT, SipU, and SipV have overlapping substrate
specificities (Tjalsma et al. 1997). At present, we do not know the
precise function of each of the SPases of B. subtilis, but it
seems likely that SipS and SipT are required, in particular, for the
processing of preproteins that are synthesized in the postexponential
growth phase. In contrast, as the sipU, sipV, and
sipW genes are transcribed constitutively, the corresponding
SPases are most likely ivolved in the processing of preproteins which
are synthesized during all growth phases.
The previously reported presence of SPase-encoding genes on endogenous
plasmids (Meijer et al. 1995) seems to give certain strains of B. subtilis an additional possibility to increase their capacity for
preprotein processing. The latter hypothesis is supported by our
observation that the SPase SipP, specified by plasmid pTA1015, can
complement for the absence of SipS and SipT (H. Tjalsma et al.,
unpubl.). Thus, it seems that SipS, SipT, and SipP have similar substrate specificities.
Dalbey and Wickner (1985) have shown that E. coli cells
depleted of their type I SPase (i.e., leader peptidase) accumulate preproteins on the outer surface of the cytoplasmic membrane. Thus,
processing by SPase was shown to be a prerequisite for the release of
translocated proteins from the membrane. However, the question whether
the accumulation of secretory preproteins in cells depleted of SPase
would affect protein translocation across the membrane remained
unresolved. Our observation that biotinylated pre-AmyQ-PSBT
accumulated in B. subtilis cells lacking functional SipS and
SipT shows that this is the case. Thus, the decreased viability of
these cells may be due to impaired protein translocation across the
membrane. However, the alternative explanation that certain proteins,
which are essential for cell growth and viability, do not reach their
correct cellular localization (e.g., the cell wall) cannot be excluded.
Interestingly, no biotinylated mature AmyQ-PSBT was produced by cells
lacking functional SipS and SipT, suggesting that biotinylated
pre-AmyQ-PSBT is not translocated across the membrane of B. subtilis. As biotinylated proteins can be transported by the
preprotein translocase of E. coli (Jander et al. 1996), it
seems that the preprotein translocase of B. subtilis is more
restrictive with respect to (partially) folded, biotinylated preproteins than that of E. coli.
An intriguing result of the present studies is that B. subtilis contains an ER-type SPase (SipW). Thus, B. subtilis is the first known eubacterium containing SPases of both
the P- and the ER-type. This may not be a peculiarity of B. subtilis, as we identified the partial sequence of a
SipW/SPC21-like protein from Clostridium perfringens by similarity searches in GenBank (accession no.
X86488). Furthermore, the identification of SipW of B. subtilis has at least three interesting implications. First, the
presence of ER-type SPases in all three domains of life suggests that
the P-type SPases, which are only present in eubacteria and the
organelles of eukaryotes, have evolved from the ER-type SPases.
However, the alternative possibilities of convergent evolution of P-
and ER-type SPases and/or horizontal transfer of ER-type
SPases from archaea or eukaryotes to eubacteria cannot be excluded.
Second, recent studies indicate that the P-type SPases make use of a
serine-lysine catalytic dyad (see Dalbey et al. 1997). In contrast,
the catalytic mechanism of the ER-type SPases is presently unclear,
because the conserved lysine residue in domain D, which is invoked in
the catalytic mechanism of the P-type SPases, is not conserved in the
ER-type SPases; instead the ER-type SPases contain a conserved
histidine residue at this position (Fig. 1B). This could mean that the
ER-type SPases use the latter histidine residue for catalysis, for
example in a serine-histidine-aspartic acid catalytic triad, like in
the classical serine proteases. In addition to the conserved histidine residue in domain D, such a catalytic triad could involve the strictly
conserved serine residue in domain B, which is essential for the
activity of the P-type SPases, and the strictly conserved aspartic acid
residue in domain E. Alternatively, the ER-type SPases could make use
of a serine-lysine catalytic dyad involving the strictly conserved
serine and lysine residues in domains B and E, respectively. Third, the
ER-type SPase of yeast and higher eukaryotes are part of larger
complexes, containing three additional subunits of which the SPC3 or
SPC22/23 subunits, respectively, are essential for SPase
activity (Newsome et al. 1992; Fang et al. 1997; Meyer and Hartmann
1997). None of these additional subunits is conserved in eubacteria or
archaea (H. Tjalsma et al., unpubl.), suggesting that the ER-type
SPases of the latter groups of organisms require no, or very different
additional subunits for activity. The identification of the catalytic
mechanism of the ER-type SPases, and the identification of their
putative partner proteins in eubacteria and archaea, are major
challenges for future research.
Finally, the present studies on the sip gene family of B. subtilis have important implications for projects on the functional analysis of the genomes of B. subtilis and other organisms, such as E. coli and yeast, because the presence of paralogous gene families is a common feature of these microbial genomes. As examplified by the lack of effect of the single disruption of the sipW gene, strategies based on single gene disruptions will probably be inadequate to assign functions to many multiplied genes with thus far unidentified functions.
| |
Materials and methods |
|---|
|
Top
Abstract
Introduction
Results
Discussion
Materials & Methods
References
|
|---|
Plasmids, bacterial strains, and media
Table 1 lists the plasmids and bacterial strains used. TY medium
(tryptone/yeast extract) contained Bacto-tryptone (1%), Bacto yeast extract (0.5%), and NaCl (1%). Minimal medium for B. subtilis contained: 100 mM potassium phosphate buffer at
pH 7.0, glucose (1%), trisodium citrate (3 mM),
MgSO4 (3 mM), casamino acids (0.1%), potassium
glutamate (0.2%), tryptophan (20 µg/ml), histidine
(20 µg/ml), methionine (20 µg/ml),
tyrosine (20 µg/ml), adenine (20 µg/ml), uracil (20 µg/ml),
nicotinic acid (0.4 µg/ml), riboflavin (0.4 µg/ml), and Fe-ammonium citrate (1.1 µg/ml). S7 media 1 and 3, used for labeling of
B. subtilis proteins with [35S]methionine
(Amersham), were prepared as described by van Dijl et al. (1990a, b).
If required, medium for E. coli was supplemented with
ampicillin (Ap, 50 µg/ml), erythromycin (Em, 100 µg/ml), or kanamycin (Km, 40 µg/ml); media for B. subtilis were
supplemented with Cm (5 µg/ml), Em (2 µg/ml), Km (10 µg/ml), or
spectinomycin (Sp, 100 µg/ml).
DNA and RNA techniques
Procedures for DNA purification, restriction, ligation, agarose
gel electrophoresis, and transformation of E. coli were
carried out as described by Sambrook et al. (1989). Enzymes were from Boehringer Mannheim. B. subtilis was transformed by growth in minimal medium until an OD600 of ±1, the subsequent
addition of plasmid or chromosomal DNA to the culture, and continued
incubation for at least 4 hr. Multiple sip mutants were
constructed by transforming single mutants with chromosomal or
linearized plasmid DNA containing a disrupted sip gene. To
minimize the number of antibiotic resistance markers in the cells,
triple sip-mutants were prepared from double sip-mutants containing no, or only one, antibiotic resistance marker. Correct integration of linearized DNA fragments, or plasmids in
the chromosome of B. subtilis was verified by Southern
hybridization. PCR was carried out with Vent DNA polymerase (New
England Biolabs) as described by van Dijl et al. (1995). DNA and
protein sequences were analyzed by use of version 6.7 of the PCGene
Analysis Program (Intelligenetics Inc.). The BLAST algorithm (Altschul
et al. 1997) was used for protein comparisons in GenBank. RNA was
isolated with the RNeasy total RNA kit from Qiagen. RNA reference
markers were from GIBCO BRL Life Technologies Inc. Northern
hybridizations were performed as described by Bolhuis et al. (1996).
Plasmid pKTH10-BT was constructed in three steps. First,
the amyQ gene was amplified by PCR with the primers
amyQ5'
(5'-G GC CT AG GAT C CTA AGAGAG GGAGAGGA A ACA T GAT-3')
and amyQ3' (5'-GGCCATGTCGACAGTACTGCATGCTTATCTAGATTTCTGAACATAAATGGAGAC-3') using pKTH10 as a template. The amplified fragment was cleaved with
BamHI and SalI, and cloned into the corresponding
sites of pBR322, resulting in pBR322amyQ. Second, the
sequences specifying the biotinylation domain (PSBT) of the P. shermanii transcarboxylase 1.3 S subunit were amplified by PCR with
the primers PSBT1
(5'-GGCCATGAGCTCTCTAGAGTCGCCGGTAAGGCCGGAGAG-3') and PSBT2
(5'-GAATTCGCATGCGCGTCAGCCGATCTTGATGAGACC-3') using plasmid pCY66 (Cronan 1990) as a template. The amplified fragment was cleaved
with XbaI and SphI, and cloned into the corresponding sites of pBR322amyQ, resulting in pBR322amyQ-BT.
Third, the amyQ-PSBT gene fusion was amplified by PCR with
the primers amyQ-BT5'
(5'-GGCCATAAGCTTTAATCAATCCGTGTTTGATGTTCC-3') and
amyQ-BT3'
(5'-GGCCATGGATCCGCGTCAGCCGATCTTGATGAGACC-3'), using pBR322amyQ-BT as a template. The amplified fragment was
cleaved with HindIII and BamHI, and ligated into the
corresponding sites of pKTH10, resulting in pKTH10-BT.
Plate assay for the processing of the
(A13i)--lactamase precursor
The plate assay for the processing of the hybrid precursor
pre(A13i)--lactamase was carried out as described by van Dijl et
al. (1992).
Pulse-chase protein labeling, immunoprecipitation, SDS-PAGE, and fluorography
Pulse-chase labeling of B. subtilis,
immunoprecipitation, SDS-PAGE, and fluorography were performed as
described previously (van Dijl et al. 1991a, b).
14C-Methylated molecular weight markers were from Amersham
International. Relative amounts of precursor and mature forms of
secreted proteins were estimated by film scanning with an LKB ultroscan
XL laser densitometer.
Western blot analysis
Western blotting was performed as described by Kyhse-Andersen
(1984). After separation by SDS-PAGE, proteins were transferred to
Immobilon-PVDF membranes (Millipore Corporation). To detect the
-amylase AmyQ, (A13i)--lactamase, or SipS, B. subtilis cells were separated from the growth medium by
centrifugation (5 min, 12.000 rpm, room temperature), and samples for
SDS-PAGE were prepared as described by van Dijl et al. (1991a). AmyQ,
(A13i)--lactamase, or SipS were visualized with specific
antibodies and HRP anti-rabbit IgG conjugates (Amersham International).
A streptavidin-HRP conjugate (Amersham International) was used to
detect biotinylated AmyQ-PSBT.
-Galactosidase activity assay
Overnight cultures were diluted 100-fold in fresh medium and
samples were taken at hourly intervals for OD600 readings and -galactosidase activity determinations. The assay and the
calculation of -galactosidase units (expressed as units per
OD600) were carried out as described by Miller et al. (1982).
| |
Acknowledgments |
|---|
We thank Dr. M. Sarvas for providing plasmid pKTH10 and specific sera against AmyQ, Dr. J. Cronan for providing plasmid pCY66, H. Mulder for preparing figures, and Dr. J. Jongbloed for useful discussions. H.T. was supported by Genencor International (Rijswijk, the Netherlands) and Gist brocades B.V. (Delft, the Netherlands), A.B., S.B., C.J.B., J.M.v.D., W.J.Q., W.S., and T.W. were supported by Biotechnology Grants (Bio2-CT93-0254 and Bio4-CT96-0097) from the European Union, M.L.v.R was supported by the Dutch Ministry of Economic Affairs through ABON (Associatie Biologische Onderzoeksscholen Nederland).
The publication costs of this article were defrayed in part by payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 USC section 1734 solely to indicate this fact.
| |
Footnotes |
|---|
Received March 23, 1998; revised version accepted June 8, 1998.
5 Corresponding author.
E-MAIL J.M.VAN.DIJL{at}BIOL.RUG.NL; FAX 31503632348.
| |
References |
|---|
|
Top
Abstract
Introduction
Results
Discussion
Materials & Methods
References
|
|---|
) B. subtilis 
) of 1% xylose, and sample for
SDS-PAGE and Western blotting were withdrawn 4 hr after the transition
from exponential to postexponential growth. (C) Accumulation
of pre-AmyQ in cells of B. subtilis 
