amyloliquefaciens sipba gene was expressed in an Escherichia coli LepBts mutant and tested for genetic complementation of the temperature sen-sitive TS phenotype as well as pre-OmpA proc
Trang 1Identi®cation and properties of type I-signal peptidases
Hoang Ha Chu, Viet Hoang*, Peter Kreutzmann², Brigitte Hofemeister, Michael Melzer
and JuÈrgen Hofemeister
Institute of Plant Genetics and Crop Plant Research (IPK), Gatersleben, Germany
The use of Bacillus amyloliquefaciens for enzyme production
and its exceptional high protein export capacity initiated this
study where the presence and function of multiple type I
signal peptidase isoforms was investigated In addition to
type I signal peptidases SipS(ba) [Meijer, W.J.J., de Jong, A.,
Bea, G., Wisman, A., Tjalsma, H., Venema, G., Bron, S &
van Dijl, J.M (1995) Mol Microbiol 17, 621±631] and
SipT(ba) [Hoang, V & Hofemeister, J (1995) Biochim
Biophys Acta 1269, 64±68] which were previously identi®ed,
here we present evidence for two other Sip-like genes in
B amyloliquefaciens Same map positions as well as
sequence motifs veri®ed that these genes encode homologues
of Bacillus subtilis SipV and SipW SipU-encoding DNA was
not found in B amyloliquefaciens SipW-encoding DNA
was also found for other Bacillus strains representing
dif-ferent phylogenetic groups, but not for Bacillus
stearother-mophilus and Thermoactinomyces vulgaris The absence of
these genes, however, could have been overlooked due to
sequence diversity Sequence alignments of 23 known
Sip-like proteins from Bacillus origin indicated further branching
of the P-group signal peptidases into clusters represented by
B subtilis SipV, SipS-SipT-SipU and B anthracis Sip3-Sip5 proteins, respectively Each B amyloliquefaciens sip(ba) gene was expressed in an Escherichia coli LepBts mutant and tested for genetic complementation of the temperature sen-sitive (TS) phenotype as well as pre-OmpA processing Although SipS(ba) as well as SipT(ba) eciently restored processing of pre-OmpA in E coli, only SipS(ba) supported growth at TS conditions, indicating functional diversity Changed properties of the sip(ba) gene disruption mutants, including cell autolysis, motility, sporulation, and nuclease activities, seemed to correlate with speci®cities and/or localization of B amyloliquefaciens SipS, SipT and SipV isoforms
Keywords: Signal peptidase I; Bacillus amyloliquefaciens; protein secretion; E coli; genetic complementation
The principles of protein transport through membranes are
basically similar in eukaryotic and prokaryotic organisms
[1], although destinations of proteins are numerous in
eukaryotic cells but only few in bacterial cells, such as the
cytoplasmic membrane, periplasm, outer membrane, cell
wall, spore compartment, or the extracellular environment
[2±4] The majority of export proteins are transported via
the Sec pathway by recognition and site-speci®c processing
[3±5] These export proteins carry a particular N-terminal leader (signal) peptide, which bears distinct domains (N, H and C), that are distinguished by charge and hydrophobicity pro®le [3±5] The N- and H-regions are thought to interact with the translocase machinery and to mediate membrane insertion, whereas the C-region allows sequence-speci®c cleavage by SPases and removal of the signal peptide from the precursor (export) protein [4,6±8] Minor differences between individual signal peptides, speci®c properties of the export protein precursor [5,7,9], as well as the speci®city of distinct SPases [3,10,11] affects the processing of individual
or groups of export proteins The B subtilis genome sequencing project [12] has enabled computer analysis to predict that 166 proteins of the total B subtilis proteome contain a N-terminal signal peptide, characteristic for Sec export protein precursors [4] Several eubacteria and archaebacteria possess only one type I SPase functioning
in Sec export protein processing [13] However, B subtilis contains ®ve chromosomally encoded type I SPases, named SipS, SipT, SipU, SipV, and SipW, respectively [4,5,14,15] Multiple type I SPases were also found in Archaeoglobus fulgidus [16], Streptomyces lividans [17], Bradyrhizobium japonicum [18,19] and Staphylococcus aureus [20] The presence of a unique type I SPase (LepB in E coli) was shown to be essential for cell viability [21,22] In contrast,
B subtilis has ®ve Sip homologues, of which SipS as well as SipT isoforms were shown to be essential for cell viability, and have overlapping processing functions Double mutants
Correspondence to J Hofemeister, Institute of Plant Genetics and
Crop Plant Research (IPK), Corrensstrasse 3, Gatersleben, D-06466,
Germany Fax/Tel.: + 49 394825 138/241,
E-mail: hofemeij@mendel.ipk-gatersleben.de
Abbreviations: Ap, ampicillin; c.f.u., colony forming units; Cm,
chlo-ramphenicol; CWBP, cell wall bound proteins; Em, erythromycin;
pre-OmpA, OmpA precursor protein; Sip, signal peptidase protein;
SPase I, signal peptidase I (leader peptidase I); TS, temperature
sensitivity; IPTG, isopropyl thio-b- D -galactoside.
De®nitions: SipS(ba), SipS(bj), SipT(ba), SipV(ba) and SipW(ba) are
the products of the sipS(ba), sipS(bj), sipT(ba), sipV(ba), and
sipW(ba) genes of Bacillus amyloliquefaciens (ba) or Bradyrhizobium
japonicum (bj), respectively.
*Present address: George Beadle Center for Genetics, School of
Bio-logical Sciences, University of Nebraska, Lincoln, USA.
Present address: Lower Saxony Institute for Peptide Research,
Han-nover, Germany.
(Received 29 May 2001, revised 5 November 2001, accepted 13
November 2001)
Trang 2were nonviable, whereas deletion of SipU, SipV and SipW,
even in quadruple mutant derivatives in combination with
either SipS or SipT de®ciency, was without lethal
conse-quences This allows the distinction between Spases of
major and minor importance for cell viability [13,23] These
functional differences are not clearly re¯ected by sequence
motifs, but likely due to unknown activities [15] Only one
minor Bacillus SPase, SipW, differs in several characteristics
from the group of prokaryotic (P-type) SPases, as it has
pronounced similarity to (ER-type) SPases found in archea
and in the ER membrane of eukaryotes [13,24,25] SipW of
B subtilis displays processing speci®city for TasA, an export
protein that acts from the spore membranes on spore coat
assembly This study presents a ®rst example for secretion to
deliver individual proteins to speci®c cellular locations, e.g
during sporulation [25] In spite of this speci®city, other
SPase homologues of B subtilis could apparently substitute
SipW functions, as the sipW gene was dispensable for cell
growth and sporulation [13,25]
B amyloliquefaciens strains have only recently been
ranked from B subtilis subspecies validation to a distinct
taxon [26,27] Although strains of B amyloliquefaciens are
among the most potent producers of industrial enzymes
[28], little is known about physiological and genetic
peculiarities [29] In previous studies, two SipS-like signal
peptidases SipS1(ba) and SipS2(ba) of B amyloliquefaciens
were described [30] and later shown to have the highest
sequence similarity to SipS or SipT of B subtilis,
respect-ively [31] These ®ndings indicated sequence, as well as
mapping speci®city, of type I-SPase homologous of Bacillus
species [4,14] The aim of this study was to isolate additional Sip-like genes in B amyloliquefaciens, and to evaluate differences in functions after genetic complementation
in an E coli LepBts mutant and after construction of
B amyloliquefaciens sip(ba) gene disruption mutants
M A T E R I A L S A N D M E T H O D S
Strains and culture conditions Table 1 lists the strains and plasmids used Bacteria were usually grown in trypton/yeast extract TBY broth or on TBY-agar [32], Spizizen minimal medium (SMM) [33] or Schaeffer's sporulation medium (SSM) [34], respectively Isopropyl thio-b-D-galactoside (IPTG, 1 mM) was added to cultures For antibiotica selection, Bacillus cultures were supplemented with erythromycin (Em, 1 gáL)1) and/or with chloramphenicol (Cm, 5 gáL)1); E coli cultures were sup-plemented with ampicillin (Ap, 50 gáL)1), Cm (10 gáL)1) or kanamycin (20 gáL)1), respectively The spores heat resist-ance test was carried out according to Nicholsen & Setlow [34]
Recombinant DNA techniques Chromosomal DNA from B amyloliquefaciens was pre-pared as described previously [33] Large-scale or mini-preparations of plasmid DNA were made from E coli either by standard methods [35,36] or by using a QIAGEN plasmid isolation kit (Qiagen GmbH, Hilden, FRG) The Table 1 Bacterial strains and plasmids used in this study.
Strains
Escherichia coli XL1blue recA1, endA1, gyrA96, thi-1, hsdR17, supE44, relA1,
lacI, [F proAB, lacI q ZDM15, Tn10(Tc r )] Stratagene
Bacillus amyloliquefaciens
Plasmids
integration plasmid
pEAS*, pEAT*, pEAV*, pEAW* pE194, Em r ::pUC18, Ap r with core-DNA This study
of sipS*(ba),sipT*,sipV*, sipW* genes, respectively
pOpacSh, pOpacTh, pOpacVh, pOpacWh, pDG148 with Sip(ba) expression cassettes
Pspac-ompA- sip(ba)His-tag, respectively. This study
a The symbol:: indicates the insertion of a respective pEA* integration plasmid at a homologous chromosomal sip gene locus.
Trang 3same procedures were applied for B subtilis or B
amylo-liquefaciens except that for plasmid isolation, cells were
prepared in lysis buffer with lysozyme (4 gáL)1) at 37 °C,
incubated for 5 min The digestion and ligation of DNA
were performed with various restriction enzymes and T4
DNA ligase, following the supplier's instructions The
general molecular cloning techniques and DNA
electro-phoresis were carried out essentially as described by
Sambrook et al [36] and Ausubel et al [35] DNA
fragments were prepared from agarose gels using the
QIAEX gel elution kit (Qiagen) E coli was transformed
with competence treatment [36] B subtilis was transformed
either after competence treatment [33] or protoplast
forma-tion [37] The latter method was also used to transform
B amyloliquefaciens, except that prior to transformation,
the DNA was occasionally treated with BamHI methylase
Alternatively, plasmid DNA was transformed into
B amyloliquefaciens by electroporation [38]
DNA sequencing and sequence analysis
DNA sequencing was performed by an automated system
(A.L.F express, Pharmacia), using the recommended
primers for the pGEM-T and pUC18 vector, with the
AutoRead sequencing kit (Pharmacia) Sequence analysis
was performed with the PC/GENE software from
Intelli-Genetics, Inc (Mountain View, Calif.) and DNA-STAR
software from Lasergene Inc (Madison, WI, USA)
Information, Bethesda, MD, USA) was used for online
database scanning Phenetic and cladistic analyses of the
amino-acid alignment were performed in PAUP* 4.0b8
[39] Mean character differences were used to calculate
pairwise distances, which were clustered with the
Neighbor-Joining algorithm Fitch parsimony analysis was
con-ducted with ACCTRAN character optimization, with the
gaps treated as missing data and the heuristic search
algorithm with 100 random sequence additions To test
the statistical support of the branches in phenetic and
cladistic analyses, bootstrap resamples were conducted
with 5000 and 500 replicates, respectively Parsimony
analysis resulted in two equally parsimonious trees of
1388 steps length (CI 0.7177, RI 0.6631), which are
compatible with the tree topology obtained by the
Neighbor-Joining analysis
Plasmid and mutant constructions
For Sip protein expression and processing studies plasmids
pOpacSh, pOpacTh, pOpacVh, and pOpacWh were
con-structed These plasmids contain within the vector pDG148
[40], the ompA as well as the respective sip(ba)gene for
IPTG-inducible expression of pre-OmpA and Sip proteins
in one single cassette, essentially as follows:
Pspac-ompA-sipHis-tag-Ppen-lacI A PCR fragment carrying the ompA
gene was ampli®ed from E coli chromosomal DNA using
primers Omp1 and Omp2 (Table 2) This ompA DNA was
cloned into vector pUC18 and after digestion with HindIII
and EcoRI, cloned into pDG148 vector to obtain the
plasmid pOpac (Table 1) The respective sip(ba) genes
were isolated from initial cloning vectors and cloned into
the vector pQE16 in frame with the His-tag-encoding
sequence The lepB gene was PCR ampli®ed from E coli
DNA using the primers Lep1 and Lep2 (Table 2), and then cloned into the pQE16 vector After restriction enzyme digestion of pQE derivated plasmids (pQSh, pQTh, pQVh, pQWh, pQBh), the His-tagged genes (sipSh, sipTh, sipVh, sipWh, and lepBh) were isolated and each cassette was cloned into the pOpac vector to obtain the OmpA-Sip/Lep expression plasmids (pOpacSh, pOpacTh, pOpacVh, pOp-acWh, and pOpacBh) The integrative plasmids pEAS*, pEAT*, pEAV* and pEAW* used for construction of the
B amyloliquefaciens sip gene disruption mutants were formed as follows: a core DNA fragment covering an internal portion of the respective sip gene was ampli®ed using chromosomal DNA of B amyloliquefaciens and the respective gene-speci®c, internal primers (Table 2) These PCR fragments were cloned into the pUC18 vector and the resulting pUC18-sip construct ligated into the PstI site of the temperature sensitive (TS) plasmid pE194 After the transformation of ALKO2718 cells with one respective integrative plasmid (see above), integration mutants were isolated by Em selection at 42 °C [41] Several Em and heat resistant colonies of each mutant progeny were isolated and tested for integration of the sip::pE* cassette after PCR ampli®cation The construction scheme of mutant strains GBA13 (sipS::pEAS*), GBA14 (sipT::pEAT*), GBA15 (sipV:: pEAV*) and GBA16 (sipW::pEAW*) is shown in Fig 4
Pulse-chase protein labelling, immunoprecipitation, SDS/PAGE and ¯uorography
The pulse-chase labelling method was carried out as described by Edens [42] The preculture of each strain was grown at 30 °C to D600 0.5, and then divided into cultures A and B, each of 5 mL After incubation for
10 min at either 30 °C or 42 °C, the two cultures were labelled for 1 min at the indicated temperature by the addition of 35S[methionine] (50 mCiáL)1) and chased by the addition of nonradioactive methionine and cysteine (2.5 gáL)1) Samples were collected at intervals and after the addition of 1 mL trichloroacetic acid, kept on ice for
30 min Polyclonal anti-OmpA Ig was used to precipitate the protein
Assay for cell autolysis Cultures of wild-type and mutant strains of B amylolique-faciens were grown in TBY medium to D600 0.6 After addition of 0.05M sodium azide, cell lysis was followed spectrophotometrically while continuing incubation at
37 °C and agitation at 200 r.p.m [43]
Cell-wall-bound protein (CWBP) extraction, and autolysin detection after SDS/PAGE
Cell wall substrate was isolated from exponential growing
B amyloliquefaciens cells according to Harwood et al [44] The CWBP extract was prepared from vegetative B amy-loliquefaciens cells according to Blackman et al [43], except that cells were desintegrated by ultrasonication Autolysin activities were performed after SDS/PAGE, and enzymo-graphy was assayed after renaturation of gels as described
by Foster [45] using B amyloliquefaciens vegetative cell wall
as the substrate
Trang 4DNase detection after SDS/PAGE
Supernatant proteins of respective cultures were
trichloro-acetic acid-precipitated, collected, washed and separated by
12% SDS/PAGE containing calf thymus DNA (10 mgáL)1)
according to the method described by Rosenthal & Lacks
[46]
Electron microscopy
For the primary ®xation, cells of B amyloliquefaciens were
kept for 1 h at room temperature in 50 mM cacodylate
buffer (pH 7.2), containing 0.5% (v/v) glutaraldehyde and
2.0% (v/v) formaldehyde After washing, the samples were
subjucted to a secondary ®xation [1 h in a solution of 1.0%
(w/v) OsO4in 50 mMcacodylate buffer] Prior to
dehydra-tion, the cells were washed and transferred into 1.5% agar
Dehydration, embedding and cutting of 1 mm3agar blocks
was performed as previously described [47] The
ultrathin-sections were contrasted with a saturated methanolic
solution of uranyl acetate and lead citrate prior to
exam-ination in a Zeiss CEM 920 A transmission electron microscope at 80 kV
R E S U L T S
Cloning of asipV-like gene PCR reactions with genomic DNA of B amyloliquefaciens and primers V1, V2, and V3 (Table 2) for regions MKKRFWFLA, VFIDYKVEG, and IVGVISDAE of the B subtilis SipV protein were chosen and found to generate PCR fragments of about 0.5 and 0.4 kb, respec-tively (Fig 1,A1) Moreover, Southern hybridization with the 0.5 kb-PCR fragment as a probe indicated that the genomic DNA of B amyloliquefaciens contains speci®cally hybridizing DNA The previously described RAGE proto-col [30] was used to PCR amplify and clone a corresponding DNA region (Fig 1,A2) ABLASTsearch of the nucleotide sequence of the ampli®ed DNA fragment revealed the presence of three ORFs, the deduced proteins having 70, 77, and 67% identity to proteins encoded by the yhjE-sipV-yhjG
Table 2 Oligonucleotide primers used for PCR.
a The IUPAC-code was used; N denotes an inosine residue.
Trang 5genes of B subtilis, respectively (Fig 1,A3) Thus, ¯anking
ORFs indicated the B amyloliquefaciens sipV(ba) gene to
occupy a similar genomic position as compared to sipV(bs)
of B subtilis [12]
Search forsipU
Several attempts were made, but failed to identify a sipU
homologous gene in B amyloliquefaciens The degenerate
primers U1, CH5, CH6 and HV11, CH7, and CH8
(Table 2) based on several regions (MNAKTITLKK,
MIAALIFTI, FKPFLIEG, YFVMGDN,
NGMGMP-SED, PFGEMRQAK) were chosen, which were most
speci®c for the B subtilis SipU, when compared to other Sip proteins In nine independent primer combinations, parallel PCR ampli®cation reactions were carried out with genomic DNA from B subtilis 168 or B amyloliquefac-iens Although the former template always resulted in the generation of a DNA fragment of the expected size, no ampli®cation products were obtained with B amylolique-faciens DNA (data not shown) In extension, the forward primers CH1 and CH2 were designed for conserved regions HFGAGNIG and HGSAPDIAG of the genes mtlD and ycsA, which map in B subtilis upstream of the searched sipU, and used in combination with reverse primers HV11, CH7 and CH8 Speci®c ampli®cation
Fig 1 Identi®cation of the sipV(ba) and sipW(ba) gene regions of B amyloliquefaciens (A1/B1) PCR reactions with genomic DNA of B amy-loliquefaciens and the degenerative primers V1/V3 and V2/V3 or W1 and W2, led to the isolation of core DNA fragments of about 0.5 kb (lane b) and 0.4 kb (lane c), or 0.2 kb, respectively (A2/B2) The respective 0.5 or 0.2 kb-DNA fragments were used as a probe for Southern hybridization of either PstI (lane a) or EcoRI (lane b) digested chromosomal DNA in case of sipV or HindIII (lane a) or EcoRI-SacI (lane b) digested genomic DNA
in case of sipW Hybridization indicated DNA fragments of about 2.5 and 1.6 kb or 0.8 and 1.2 kb, respectively Each digest indicated one speci®c signal and suggested the existence of sipV or sipW like genes in B amyloliquefaciens (A3/B3) The RAGE protocol [30] was used for PCR ampli®cation as follows: The DNA of B amyloliquefaciens was cut with either PstI and EcoRI in case of sipV or EcoRI-SacI in case of sipW and ligated into corresponding sites of pUC18 DNA The ligation mixes were used for PCR with oligo nucleotides Uni1 or Uni2 and pairs of primers V6/V7 and V4/V5 or W9/W10 and W11/W12 (Table 2) for forward or reverse reactions, respectively The latter were chosen according the indicated regions within the 0.5 kb- or 0.2 kb-PCR fragments from step A1 or step B1 The PCR fragments were cloned and sequenced Ultimate PCR led to fragments covering 1.2 or 1.3 kb of DNA, respectively The nucleotide sequence was submitted to GenBank and given the accession number AF085497 or AF084950, respectively The detected open reading frames are indicated.
Trang 6products were obtained with DNA of B subtilis, but not
with DNA of B amyloliquefaciens (data not shown)
Southern hybridization experiments with a 0.4-kb DNA
fragment for a sipU-speci®c probe, that had been PCR ampli®ed with primers CH5 and CH7 from B subtilis DNA (Table 2) were carried out with B subtilis as well as
B amyloliquefaciens genomic DNA Even at low strin-gency, B amyloliquefaciens yielded no hybridizing band, but with DNA from B subtilis 168, a positive band appeared (data not shown)
Cloning of asipW-like gene
In order to isolate a sipW-like gene from B amylolique-faciens a similar strategy as outlined in Fig 1 was followed PCR ampli®cation was done with degenerative primers W1 and W2 according to conserved regions VLSGSMEPEFNTG and TKGDNNAAAD of B subtilis SipW (Fig 1,B1) The core DNA fragment of about 0.2 kb, was in Southern hybridization experiments found to hybridize with B amyloliquefaciens DNA (Fig 1,B2) Sub-sequent RAGE ampli®cation with B amyloliquefaciens genomic DNA and primers W7, W8 and W9, W10 (Table 2), as well as a ®nal ampli®cation step with terminal primers yielded a DNA fragment of 1.3 kb The nucleotide sequence demonstrated the presence of three ORFs, and the deduced proteins to have 42, 73, and 82% of identity to
Fig 2 Abundance of sipW- like DNA in several Bacillus species representing dierent 16S rRNA phylogenetic groups [26,48] Group 1, B subtilis (13), B amyloliquefaciens (1), B circulans (2), B lentus (3), B licheniformis (4), B megaterium (5), B thuringiensis (6); From group 2, B sphaericus (7); From group 3, B macerans (8), B polymyxa (9); From group 4, B brevis (10); From group 5, B stearothermophilus (11); Thermoactinomyces vulgaris (12) PCR was under standard conditions using about 2 lg of genomic DNA and primers W1 and W2 (Table 2) The brightness of DNA bands correlates with the amount of PCR product per run.
Fig 3 E coli LepBts complementation after Sip(ba) protein expression (A) Growth of E.coli IT41 transformants containing following plasmids: (s), pTK100; (r), pOpacBh; (.), pOpacSh; pTK99; (d), pOpacTh; (j), pOpacVh; (h), pOpacWh The cultures were grown in TBY medium at 42 °C without IPTG Dierent transformant colonies were used in repeated experiments (B) Processing of pre-OmpA in
E coli IT41 was analysed after pulse-chase labelling, immunoprecipi-tation, SDS/PAGE and ¯uorography Samples were withdrawn at the intervals indicated p, precursor; m, mature protein (a), IT41/pOpac; (b), IT41/pTK100; (c), pOpacSh; (d), pOpacTh; (e), pOpacVh or (f), pOpacWh, respectively (C) Expression of His-tagged Bacillus SipS(ba) proteins in E coli IT41 was detected by Western blotting Lanes 1/2, 3/4, 5/6 and 7/8 refer to His-tagged Sip(ba) protein detection in cells with SipS(ba), SipT(ba), SipV(ba) or SipW(ba) expression either grown without or with the addition of IPTG at 30 °C.
Trang 7proteins encoded by yqxM-sipW-tasA genes of B subtilis,
respectively (Fig 1,B3) This map position also indicated
the B amyloliquefaciens sipW(ba) gene to be similar to
sipW(bs) of B subtilis [12]
Abundance of sipW-encoding DNA in diverse
Bacillus groups
After successful application of primers W1 and W2 for PCR
ampli®cation of a sipW gene homologue from B
amyloliq-uefaciens (Fig 1B), the same strategy was applied to search
for the abundance of similar genes in other, distantly related
Bacillus species Genomic DNA of several species, including
at least one strain of each Bacillus 16S rRNA-phylogentic
group [26,48], was used to carry out the above mentioned
PCR approach The abundance of the sipW-like genes was
indeed con®rmed for distantly related Bacillus species, but
not found in DNA of B stearothermophilus and
Thermo-actinomyces vulgaris (Fig 2) The latter might have been
overlooked due to primer speci®city
Genetic complementation of anE coli lepBts mutant
E coli strain IT41, a lep-9 TetrP1 transductant of E coli
W3110, contains an amber mutation in the lepB gene Due
to heat sensitivity of its type I-signal peptidase LepB [22], the
mutant stops growth and accumulates precursor proteins,
i.e of pre-OmpA, at non permissive temperature conditions
It has been repeatedly shown that growth of this mutant at
high temperature can be restored after transformation with
lepB-like genes from negative as well as
Gram-positive bacterial origin, i.e of Bradyrhizobium japonicum
[18,19], Staphylococcus aureus [20], Streptococcus
pneumo-niae [49] and Streptomyces lividans TK21 [17] In order to
verify the functionality of the four Sip(ba) proteins, the
SPases and OmpA were coexpressed in a single expression
cassette in this E coli mutant The expression cassette of
plasmids pOpacSh, pOpacTh, pOpacVh, and pOpacWh, was basically as follows: Pspac-ompA-sipHis-tag-Ppen-lacI, whereby the C-terminal His-tag enabled immunodetection
of Sip protein expression (Fig 3) For a control, the LepB protein of E coli was His-tagged and the gene likewise cloned on plasmid pOpacBh The plasmids pTK100 and pTK99, carrying a sipS-like gene of Bradyrhizobium japon-icum in either sense or antisense orientation [19] served for
an additional control E coli IT41 transformants were maintained at 30 °C After growth in TBY with or without IPTG expression of Sip(ba)- and LepB- proteins were in respective transformants con®rmed by immunodetection using His-tag antibodies (Fig 3) Without IPTG induction, IT41 transformants with expression of LepB, SipS(ba), or SipS(bj) of B japonicum revealed growth at 42 °C, but not with SipT(ba), SipV(ba) or SipW(ba) (Fig 3) Moreover, all Sip(ba) expressing IT41 cultures, grow signi®cantly slowed after IPTG addition indicating overexpression lethality (data not shown) Processing of pre-OmpA was thus studied without IPTG induction at the non permissive temperature and found in SipS(ba) and SipT(ba) expressing IT41 cultures, but not with SipV(ba) or SipW(ba) expression (Fig 3)
Sip disruption mutants
In order to study the phenotype of sipS(ba), sipT(ba), sipV(ba) and sipW(ba) mutants, B amyloliquefaciens strains GB13, GBA14, GBA15 and GBA16 were grown and tested under certain conditions Inspecting the changed characters of the mutants, it should be stressed that secondary mutant allels (see Material and methods), translation of front portions of each sip gene as well as promoter activities on downstream genes from the large DNA insert (Fig 4), are unlikely but ®nally not excluded
Growth and protein secretion The growth of sip(ba) mutants was compared at either 37 or
45 °C in TBY and SMM medium Under each condition, the sipV(ba) mutant exhibited slower growth rate, com-pared to the wild type and other sip(ba) mutant strain (data not shown) The yields of protein secreted after 24 h of growth in TBY medium as well as the protein changed banding pattern were compared after SDS/PAGE (data not shown) Although the total protein of sipS(ba) and sipT(ba) mutants, was about 30% lower compared to wild type cultures, slab gel SDS/PAGE was not suf®cient to demonstrate more than vague differences which might correlate with a distinct sip(ba) gene de®ciency (data not shown, but see Fig 6)
Sporulation
In order to study the effect of sip(ba) gene disruption on sporulation, strains were grown in SSM for 8±48 h at 37 °C and respective samples tested for heat resistant c.f.u Under these conditions, wild-type cultures contained after 24 and
48 h about 25 and 43% of spores, respectively No signi®cant differences were observed for sipS(ba), sipV(ba) and sipW(ba) mutant cultures, whereas spores were rarely found at frequencies of about 0.001% in the sipT(ba)
Fig 4 Scheme of construction of sip(ba) gene disruption mutants of
B amyloliquefaciens The integrative plasmids pEAS*, pEAT*,
pEAV* and pEAW* were transformed into ALKO2718 cells, and
integration was achieved after several rounds of cultivation with Em
selection at 42 °C [41] Campbell-type integration of the 6.3 kb-DNA
cassette disrupted the respective sip(ba) gene Integration of the
sip::pE* cassette at the desired sip(ba) gene locus was con®rmed after
PCR ampli®cation using primers with speci®city for either
chromo-somal DNA outside of the integration cassette or pUC18 primer Uni1
(Table 2) The relative positions and orientation of open reading
frames (amp and ery stand for antibiotic resistance genes of the
plas-mids) as well as proposed tanscription terminator elements (t) are
shown.
Trang 8mutant cultures These experiments were several times repeated This distinct gene disruption was always found to correlate with low spore frequencies and cell lysis of SSM cultures after 8 h of growth (data not shown) The few sporulating cells from sipT(ba) mutant cultures in this incubation period exhibited the structure of stage III forespores and exhibited obvious abnormalities in either coat or cortex structures (Fig 5) The progeny from SMM cultures without antibiotica selection, was Em sensitive to about 80%, and exhibited restored spore frequencies (data not shown) This correlation underlined sipT(ba) gene disruption to correlate with spore formation de®ciency Autolysis and cell motility
Microscopical inspection of stationary cultures indicated cells of the sipV(ba) mutant to grow in TBY as ®laments, while cultures of the wild-type and other mutants grow as rod shaped cells This observation indicated a de®ciency in either cell division or cell wall formation We therefore compared the mutants for cell autolysis, cell motility as well
as autolysin activities of puri®ed CWBP fractions by SDS/PAGE After the addition of sodium azide (0.05M) cultures of the sipV(ba) mutant were exceptionally less affected by autolysis, compared to wild-type and the other sip(ba) mutants (Fig 6) As changed cell autolysis was expected to correlate with changed cell motility [43], the halo diameter of colonies of wild type and mutant cultures was compared after plating on soft agar and growth at 25 or
37 °C The sipV(ba) mutant colonies in the average had swarming halo diameters of 49% compared to the diameter of wild type as well as other mutant colonies, except that of the sipW(ba) mutant, which was reduced to about 70% (data not shown) CWBP preparations of wild type and mutant cells were analysed for autolysin activities after SDS/PAGE and enzymography (Fig 6) The CWBP pattern of sipV(ba) mutant cells was indeed found to differ with respect to the presence and the relative amount of several CWBP's However, at least one out of about ®ve major autolysin activities of wild type cells, which runs in a double band of proteins of about 35 kDa was missing Nuclease activities
Occasionally we observed that after plasmid isolation higher yields of DNA could be obtained from distinct mutants The respective sipS(ba) and sipT(ba) mutants were there-fore suspected to have either changed content of plasmid DNA or reduced DNA degradation due to loss or reduction
of nuclease activities A pronounced zone of nuclease activity was after SDS/PAGE and enzymography found in
Fig 5 Electron microscopy of B
amylolique-faciens wild type (A) and SipT(ba) mutant
forespore structures (B) Cells were collected
after 8 h of growth in SSM and processed for
ultrathin-section electron micrography as
described in Materials and methods.
Fig 6 Cell autolysis, CWBP pattern and autolysin activities of
B amyloliquefaciens and of sip(ba) mutants (A) Sodium azide (0.05 M )
was added to exponential-phase TBY-cultures (D 600 0.5±0.6) Cell
lysis was followed spectrophotometrically at 600 nm (d) wild type
GBA12; (s), sipS(ba) mutant GBA13; (.), sipT(ba) GBA14; (,),
sipV(ba) mutant GBA 15; (j) sipW(ba) mutant GBA16 (B) SDS/
PAGE separation of the CWBP fraction of B amyloliquefaciens
GBA12 and sipV(ba) mutant GBA15 (a) and enzymography of
autolysin activities after renaturing SDS/PAGE of gels containing
puri®ed B amyloliquefaciens cell wall material as substrate (b) Sample
preparation is described in Methods and renaturation of the SDS/
PAGE gel was according to Foster [45] Lane 1, GBA15; lane 2,
GBA12 The arrows indicate protein bands reduced or lacking in the
mutant The labels a 1 and a 2 point to autolysins which are most
sig-ni®cantly aected The data are from one representative experiment
after three times of repetition.
Trang 9the area of 30 kDa-proteins of supernatant fractions of wild
type and several sip(ba) mutant cultures This nuclease
activities were strongly reduced in the sipT(ba), about nearly
lacking in the supernatant of sipS(ba) mutant cultures
(Fig 7) In spite of numerous attempts to purify that
suspected nuclease, likely due to low protein concentration,
it could not be isolated from wild type cultures
Conse-quently, the identity of the suspected nuclease activity, even
of its export protein character has not been veri®ed
D I S C U S S I O N
Here we present evidence for the existence of sipV- and
sipW-like genes in chromosomal DNA of B
amylolique-faciens as compared to B subtilis [12] The similarity of the
deduced proteins, conserved sequence motifs, the protein
length, as well as same neighbourhood of genes, also
con®rm gene homology with sipV and sipW of B subtilis as
previously shown for SipS(ba) and SipT(ba) of B
amylo-liquefaciens [12,13,23,30,31] In contrast, sipU-encoding
DNA was not found Although, its absence could have
been overlooked due to sequence diversity, one of these two
species might have gained or lost a sipU-type gene paralogue
after evolutionary constrains [1] Variation of the sip gene
multiplicity was already indicated by the presence of
additional plasmidal sipP genes in Bacillus strains [5,31]
In this study, SipW-encoding DNA was ampli®ed from
10 out of 12 strains representing four distant 16S
rRNA-groups of the phylogenetic Bacillus tree [26,48], but not
found in group 5 strains B stearothermophilus and
Ther-moactinomyces vulgaris As indicated for SipU, the PCR
approach could have also been failed due to DNA sequence
diversity of these remote Gram-positive spore- forming bacilli [27] Diversity of paralogous Sip proteins is also indicated from phenetic distance analysis as illustrated in Fig 8 As much as 23 Bacillus Sip proteins known from
B subtilis, B amyloliquefaciens, B licheniformis, B stearo-thermophilus, B caldolyticus, B halodurans and B anthra-cis were included The Neighbor-Joining algorism was used
to compare the Bacillus sequences with E coli LepB and yeast Sec11 ER-type SPase The phylogenetic tree
strength-en the distinction betwestrength-en P- and ER-type SPases as previously proposed [13], but also the clustering of P-type Sip proteins into at least three subgroups represented by
B subtilis SipV-, SipS,T,U- and B anthracis Sip3,5-like SPases, respectively This analysis showed close relationship between Sip proteins of B amyloliquefaciens and B subtilis
as well as their relatedness to other SPases, where the Sip-isoforms of these two Bacillus species Basically these data are similar to those of van Roosmalen et al [15], where 15 different SPases were included and the authors claimed the distinction between major and minor SPases upon similar phylogenetic analyses According to our data, which include additional SPases from B halodurans, as well as from
B anthracis, the given criteria for major and minor SPases might differ from one species to another For instance, SipV(Bha) of B halodurans, apparently plays the role of a major SPase, but according to its phylogenetic character would not belong to the group of major SPases
With respect to their group character of Sip isoforms, it was asked, whether SPases of one group genetically complement each other more likely, than SPases from another group Each of the four Sip(ba) proteins was in a LepBts mutant of E coli tested for its ef®ciency to restore de®ciencies of LepB, i.e the complementation of the mutant
TS phenotype as well as pre-OmpA processing Only SipS(ba) and SipT(ba) were active in processing pre-OmpA, while SipV(ba) and SipW(ba) failed The lack of processing activities of the latter correlates with enhanced degradation,
as indicated by degradation products of the SipV(ba) as well
as SipW(ba) protein from E coli after immunodetection (Fig 3) These observation could re¯ect inactivation by self-cleavage of these SPases in E coli, as it was shown for a truncated SipS(ba) protein lacking its unique N-terminal membrane anchor [50] The processing data might be compared to the processing activities of their B subtilis homologues in E coli, tested with the pre(A13i)-b-lactam-ase precursor, where SipV was also inactive [14]
Moreover, the growth of the LepBts mutant was only restored after SipS(ba) but not after SipT(ba) expression These likely re¯ects differences in the speci®city or capacity
to cope with the range of growth-limiting LepB processing functions These differences for the ®rst time demonstrated differences between SipS- and SipT-like SPases with respect
to their activities in E coli and could be due to their different mode of membrane insertion as well as enzyme activities [15,50]
Mutant studies in E coli implemented LepB to be essential [21,22], while in B subtilis, heat inactivation of SipSts in a SipT mutant background, i.e a SipS/SipT double de®ciency, had lethal consequences [5,13] So far, no distinct phenotype was found to distinguish B subtilis sip mutants, although SipW activities in pre-TasA processing and transport into B subtilis endospores provided a ®rst example of speci®cation of this SPase isoform for
spore-Fig 7 Enzymogram of nuclease activity of culture supernatant of
B amyloliquefaciens sip(ba) mutants Aliquots (50 lL) of cell-free
supernatant of 24 h TBY-cultures of wild type GBA12 (1) mutant
strains sipS(ba) GBA13 (2), sipT(ba) GBA14 (3), sipV(ba) GBA 15 (4)
and sipW(ba) GBA 16 (5) were separated after 12% SDS/PAGE
containing calf thymus DNA (10 mgáL )1 ) The gel was renatured and
stained with ethidium bromide The bright areas indicate zones of
DNA hydrolysis.
Trang 10speci®c protein sorting [13,24,25] Inactivation of either SipS
or SipT in B subtilis just decreased the total yields of export
proteins compared to the wild type, as it was also found in
B amyloliquefaciens to 30% (data not shown) All of the
B amyloliquefaciens Sip(ba) disruption mutants were
via-ble, but some had impaired growth, sporulation and cell
division properties Strict correlation of a distinct mutant
phenotype with that distinct sip gene disruption, as well as
restoration of the mutant phenotype after spontaneous
excision of the insertion cassette from B amyloliquefaciens
mutants, strongly indicated gene disruption to correlate
with the distinct mutant de®ciencies, which were preliminary
analysed
Disruption of sipT(ba) in B amyloliquefaciens correlated
with a drastic reduction of sporulation and rare forespores
stalled in stage III development with apparently changed
cortex or coat structures [45] Similar, sporulation de®ciency
of B subtilis sipT-sipV double-deletion mutants have been
reported [51] These two ®ndings would suggest a distinct
role of those SPases in the processing and export of
sporulation-related proteins in both species
The same might be true for export of a not yet de®ned
nuclease in B amyloliquefaciens, which was most affected
by sipS(ba), and to a lesser extend also by sipT(ba) gene
disruption The respective nuclease of B amyloliquefaciens
is apparently not a homologue of the 12 kDa-B subtilis
extracellular NucB [52], as the size of the protein was about
30 kDa Its nature remains unknown, as any attempt to
isolate the protein from B amyloliquefaciens failed (data not shown)
Moreover, impaired growth, inhibited cell autolysis and reduced motillity speci®ed sipV(ba) mutants Indeed, changed pattern of CWBP's as well as loss of at least one (major) 35-kDa autolysin correlated with SipV(ba) de®ciency In B subtilis, LytC (50 kDa amidase), LytD (90 kDa glucosaminidase) or a sigD controlled (minor) 49-kDa autolysin are shown to change cell wall turnover, septation, cell lysis as well as swarming motility [43] An autolysin of B amyloliquefaciens might correlate with that distinct SipV(ba) mutant phenotype which seems to differ from known autolysins of B subtilis [12,45]
In summary, more detailed studies are required to explain the mystery of multiple Sip proteins with more or less different characters in various Bacillus species As B amyl-oliquefaciens strains have only recently ranked to a distinct taxon [26,27], the indicated differences of B amyloliquefac-iens Sip candidates compared to B subtilis homologues could indicate these species have slightly changed characters
of their processing apparatus with respect to the presence and specialization of Sip protein homologous
A C K N O W L E D G E M E N T S
We thank Susanne KoÈnig and Christian Horstmann for nucleotide and amino acid sequencing, and Renate Manteuel for preparation of antibodies, respectively Peter MuÈller kindly provided plasmids pKT99
Fig 8 Unrooted phenogram of the Neighbor±
Joining analysis of known Bacillus Sip proteins
including Saccharomyces cerevisiae Sec11_Sce
(NP012288) The Bacillus Sip proteins
ana-lysed are: B amyloliquefaciens SipS_Bam
(P41026), SipT_Bam (P41025), SipV_Bam
(AAF02219), SipW_Bam (AAF02220);
B subtilis SipS_Bsu (P28628), SipT_Bsu
(G69707), SipU_Bsu (I39890), SipV_Bsu
(A69708), SipW_Bsu (B69708), pTA1015
(I40470), pTA1040 (I40552); B halodurans
SipV_Bha (BAB04749), SipW_Bha
(BAB05849); B licheniformis Sip_Bli
(CAA53272); B caldolyticus Sip_Bca
(I40175); B anthracis Sip1_Ban, Sip2_Ban,
Sip3_Ban, Sip4_Ban, Sip5_Ban, SipW_Ban;
B stearothermophilus Sip1_Bst, Sip2_Bst
(preliminary sequence data from the website
http://www.tigr.org) The length of each pair
of branches represents the distance between
sequence pairs and bootstrap values are given.
The cluster of major SPases as de®ned by
Roosmalen et al [15] is circled.