New Integrative Method To Generate Bacillus subtilis Recombinant Strains Free of Selection Markers

Abstract

The novel method described in this paper combines the use of blaI, which encodes a repressor involved in Bacillus licheniformis BlaP β-lactamase regulation, an antibiotic resistance gene, and a B. subtilis strain (BS1541) that is conditionally auxotrophic for lysine. We constructed a BlaI cassette containing blaI and the spectinomycin resistance genes and two short direct repeat DNA sequences, one at each extremity of the cassette.
The BS1541 strain was obtained by replacing the B. subtilis PlysA promoter with that of the PblaP β-lactamase promoter. In the resulting strain, the cloning of the blaI repressor gene confers lysine auxotrophy to BS1541. After integration of the BlaI cassette into the chromosome of a conditionally
lys-auxotrophic (BS1541) strain by homologous recombination and positive selection for spectinomycin resistance, the eviction
of the BlaI cassette was achieved by single crossover between the two short direct repeat sequences. This strategy was successfully
used to inactivate a single gene and to introduce a gene of interest in the Bacillus chromosome. In both cases the resulting strains are free of selection marker. This allows the use of the BlaI cassette to
repeatedly further modify the Bacillus chromosome.

Full text

APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Dec. 2004, p. 7241–7250 Vol. 70, No. 12
0099-2240/04/$08.00⫹0 DOI: 10.1128/AEM.70.12.7241–7250.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.
New Integrative Method To Generate Bacillus subtilis Recombinant
Strains Free of Selection Markers
Alain Brans,† Patrice File´e,† Andy Chevigne´, Aurore Claessens, and Bernard Joris*
Centre for Protein Engineering, Institut de Chimie, Universite´ de Lie`ge, Sart-Tilman, B4000 Lie`ge, Belgium
Received 6 April 2004/Accepted 20 July 2004
The novel method described in this paper combines the use of blaI, which encodes a repressor involved in
Bacillus licheniformis BlaP ␤-lactamase regulation, an antibiotic resistance gene, and a B. subtilis strain
(BS1541) that is conditionally auxotrophic for lysine. We constructed a BlaI cassette containing blaI and the
spectinomycin resistance genes and two short direct repeat DNA sequences, one at each extremity of the
cassette. The BS1541 strain was obtained by replacing the B. subtilis P
lysA
promoter with that of the P
blaP
␤-lactamase promoter. In the resulting strain, the cloning of the blaI repressor gene confers lysine auxotrophy
to BS1541. After integration of the BlaI cassette into the chromosome of a conditionally lys-auxotrophic
(BS1541) strain by homologous recombination and positive selection for spectinomycin resistance, the eviction
of the BlaI cassette was achieved by single crossover between the two short direct repeat sequences. This
strategy was successfully used to inactivate a single gene and to introduce a gene of interest in the Bacillus
chromosome. In both cases the resulting strains are free of selection marker. This allows the use of the BlaI
cassette to repeatedly further modify the Bacillus chromosome.
The completion of the sequencing and annotation of the
Bacillus subtilis 168 genome supply a complete view of the B.
subtilis protein machinery, and this knowledge stimulates new
approaches to analyze biochemical pathways (12, 15). This
postgenomic study requires genetic tools that allow the com-
bination of several gene manipulations in the same strain.
Classically, these chromosomal modifications could be achieved
by a method using a positive selection marker, usually an
antibiotic resistance marker generated by the insertion of a
selection marker gene in the B. subtilis chromosome. In this
strategy, the introduction of a second chromosomal modifica-
tion requires a second resistance gene, or, if the same resis-
tance gene is used, the eviction of this gene by a single cross-
over event prior to further genetic manipulation. In the first
case, the number of chromosomal modifications is limited by
the number of available resistance genes, and, moreover, the
multiantibiotic pressure could modify the physiology of the
manipulated strain. In the second case, selection of the strain
which has lost resistance is tedious due to the relatively low
frequencies and absence of positive selection. In the same way,
the optimization of a recombinant B. subtilis strain for over-
production and secretion of a protein can require chromo-
somal modifications which could involve the integration of
several copies of the gene of interest (7), the construction of
multiprotease- and sporulation-deficient strains (18, 19, 25,
26), and/or the coexpression of chaperones to amplify the
posttranscriptional maturation of the protein (2, 24). Another
example of the usefulness of generating multiple mutants is for
the study of complex physiological pathways, such as sporula-
tion. Indeed, mutations in many sporulation genes regulated by
sporulation-specific sigma factors or found by DNA arrays give
no obvious phenotypes (4, 5). It is thought that the products of
these genes could be largely redundant and that multiple mu-
tants may be needed to unravel gene function.
In terms of routinely carrying out chromosome integrations,
only one method is described for B. subtilis that allows the sub-
sequent excision of the selection marker coupled with positive
selection (6). This method relies on the use of an integrative
cassette containing an antibiotic resistance gene and the upp gene,
encoding uracylphosphoribosyl transferase as a counterselection
marker (the upp cassette). The use of the upp cassette is linked to
aB. subtilis strain deleted of the upp gene conferring resistance to
5-fluorouracyl. Here we describe a conditional auxotrophy-based
method for the eviction of the selection marker. This alternative
to the method of Fabret et al. (6) combines the use of blaI, which
encodes a repressor involved in Bacillus licheniformis BlaP ␤-lac-
tamase regulation (8, 9, 13), an antibiotic resistance gene, and a
conditional lysine-auxotrophic B. subtilis strain. This strategy was
successfully used to inactivate a single gene and to introduce a
gene of interest into the B. subtilis chromosome. In both cases the
resulting strains are free of selection marker, thus allowing the
repeated use of the method for further modifications of the Ba-
cillus chromosome.
MATERIALS AND METHODS
Bacterial strains, plasmids, and oligonucleotides. The bacterial strains and
plasmids used in this study are listed in Tables 1 and 2, respectively. All B. subtilis
recombinant strains are B. subtilis 168 derivatives. Specific primers used for PCR
amplification were synthesized by Eurogentec (Table 3).
Culture and growth conditions. All organisms were grown in Luria-Bertani
(LB) broth (rich medium) or LB agar supplemented, when required, with the
appropriate antibiotic. Terrific Broth (TB) without glycerol was used as a rich
liquid nutrient broth for ␤-lactamase production. Minimal medium (MM)
(Na
2
HPO
4
·7H
2
O, 12.8 g/liter; KH
2
PO
4
, 3 g/liter; NaCl, 0.5 g/liter; NH
4
Cl, 1
g/liter; MgSO
4
, 1 mM; CaCl
2
, 0.1 mM; glucose, 0.4%; L-tryptophan, 20 mg/liter;
pH 7.4) was used for auxotrophy determination (11). In order to check lysine
auxotrophy, MM was supplemented with lysine at a concentration of 50 ␮g/ml.
The final concentrations of antibiotics were the following: 100 ␮g of ampicillin/
* Corresponding author. Mailing address: Centre for Protein Engi-
neering, Institut de Chimie B6a, Universite´ de Lie`ge, Sart-Tilman,
B4000 Lie`ge, Belgium. Phone: (32) 366 2954. Fax: (32) 366 3364.
E-mail: bjoris@ulg.ac.be.
† A.B. and P.F. contributed equally to this work and are listed in
alphabetical order.
7241

ml, 100 ␮g of spectinomycin/ml, and 50 or 10 ␮g of kanamycin/ml when selecting
for a recombinant in Escherichia coli or Bacillus spp., respectively.
␣-Amylase expression by Bacillus colonies was detected by growing colonies
overnight on an LB plate containing 1% starch and staining the plate with iodine
as described elsewhere (3).
DNA manipulation techniques. The isolation and manipulation of recombi-
nant DNA was performed with standard techniques. Enzymes were commercial
preparations and were used as specified by the suppliers (Gibco, Promega, and
Biolabs). Bacillus chromosomal DNA was prepared with the Wizard Genomic
DNA purification kit (Promega). E. coli transformation was performed as de-
scribed by Sambrook et al. (21). B. subtilis transformation was performed by the
competent cell method (14). In Southern blot experiments, the AlkPhos Direct
Labeling kit (Amersham Pharmacia Biotech) was used to label the DNA probe
with alkaline phosphatase.
Construction of pDML1539. spoVAF and lysA genes were amplified by PCR
from purified chromosomal B. subtilis 168 DNA with the oligonucleotide pairs
spoVAFBglII/spoVAFMluI and lysABamHI/lysAEnd, respectively (Table 3).
The two PCR products were subcloned into pGEM-T-Easy to generate
pDML1534 and pDML1537 (Table 2). pDML1536 was constructed by cloning a
362-bp SacII-ApaI fragment from pDML1570, containing the P
blaP
promoter,
into SacII-ApaI sites of pDML1534. pDML1538 was constructed by cloning the
lysA gene from pDML1537 (on a 1,430-bp BamHI-HincII fragment) into the
BamHI-EcoRV sites of pDML1536. The 1,526-bp XbaI-BglII fragment of
pDG792, containing the gene conferring kanamycin resistance, was inserted into
pDML1538 digested with the same restriction enzymes to generate pDML1539.
Construction of pDML1541. The P
blaP
promoter was amplified by PCR from
pDML995 by using primers BlaIbam⫹and promblaPBglII. The PCR product
was cloned into pCR-Script to generate pDML1570. pDML1540 is a derivative of
pDML1570 in which the 1,244-bp SalI-BglII fragment of pDML1543 (carrying a
truncated 5⬘end of spoVAF) was cloned. The 1,516-bp ApaI-BamHI fragment of
pDML1540, containing the last 1,244 bp of spoVAF and the P
blaP
promoter, was
inserted into pDML1539 digested with the same enzymes to generate pDML1541.
In this construct, the lysA gene is under the control of the P
blaP
promoter.
Construction of the BlaI cassette. The two repeat units (repfront and repback)
corresponding to the last 138 bp of the green fluorescent protein (GFP) gene
were amplified by PCR, using pDML967 as template and two sets of primers,
repGFPfrup/repGFPfrdo and repGFPbackup/repGFPbackdo, respectively. The
PCR-amplified fragments of 155 and 276 bp were cloned in pGEM-T-Easy to
generate pDML1543 and pDML1544, respectively. The blaI gene was amplified
by PCR using pDML995 as template and oligonucleotides pBlaIBglII and stop-
BlaIBamHI as amplimers. The 513-bp amplified fragment was cloned into
pGEM-T-Easy to form pDML1545. The 208-bp EcoRV-SacI fragment of
pDML1544 was inserted into pDML1543, which was digested with EcoRV and
SacI to generate pDML1546. The 2,194-bp BamHI fragment of pIC333 was
recircularized by ligation overnight to yield pDML1566. The 505-bp BglII-
BamHI fragment of pDML1545 was inserted into pDML1566 digested with
BamHI to give pDML1547. The 1,753-bp XbaI-BamHI fragment of pDML1547
was subcloned into the XbaI-BamHI sites of pDML1546. The resulting
pDML1548 plasmid carries the BlaI cassette.
Construction of pDML1549. The 5.8-kb BlpI-SphI fragment of pAC7 was
isolated, the cohesive ends were filled in with Klenow fragment, and the DNA
was recircularized by ligation to generate pDML1542. Plasmid pDML1549 was
constructed by inserting the BlaI cassette from pDML1548 (on a 2,077-bp SalI-
NruI fragment) into the SalI and EcoRV sites of pDML1542.
Construction of pDML1567. The 1,282-bp MluI-BamHI fragment from
pDML1515, containing the blaP gene and the P
blaP
promoter, was ligated to
MluI- and BglII-digested pDML1549, resulting in pDML1567.
Electrophoresis, Western blotting, and ␤-lactamase assay. The conditions and
reagents employed for running sodium dodecyl sulfate-polyacrylamide gel electro-
phoresis (SDS-PAGE) gels, agarose gels, and Western blotting were described
elsewhere (17). ␤-Lactamase assay was performed as previously described (8)
RESULTS
Replacement of the B. subtilis 168 P
lysA
promoter with the B.
licheniformis P
blaP
-regulated promoter. In the B. subtilis 168
(BS168) chromosome, the lysA gene, located directly down-
stream of the spoVAF gene, encodes diaminopimelate decar-
boxylase, the last enzyme involved in the lysine biosynthetic
pathway that catalyzes the conversion of mesodiaminopimelate
into lysine. Complementation experiments showed that the
lysA gene is essential when the mutant strain is cultivated on
minimal medium (12, 20). Therefore, to generate a precursor
strain that simplifies the screening for marker replacement
with the P
blaP
promoter, the expression of lysA was interrupted
by replacing the BS168 chromosome region containing the
spoVAF 3⬘end and the intergenic region between spoVAF and
lysA genes with a kanamycin resistance gene. This was accom-
plished by transforming BS168 with the ApaI-linearized E. coli
plasmid pDML1539, a pGEM-T-Easy derivative, carrying the
Kan
r
gene of pDG792 flanked by the spoVAF 5⬘end and the
lysA gene, respectively (Fig. 1). The cells were plated on rich
medium supplemented with 10 ␮g of kanamycin/ml. The se-
lected kanamycin-resistant clones were replicated on minimal
medium supplemented with 10 ␮g of kanamycin/ml and with or
without 50 ␮g of lysine/ml. In 33% of the kanamycin-resistant
colonies, the 752-bp spoVAF 3⬘end, the intergenic region up-
stream of lysA, and the start codon of lysA were exchanged by
double crossover with part of pDML1539 to generate the
BS1539 strain exhibiting a Kan
r
,⌬lysA, and ⌬spoVAF pheno-
type. This strain was unable to grow on minimal medium de-
void of lysine, confirming that absence of the P
LysA
promoter
can provide a reliable lysA mutant phenotype that can be easily
screened. The high levels (67%) of kanamycin-resistant and
lysA
⫹
colonies obtained were the result of a single crossover,
TABLE 1. Strains used in this study
Strain Description
a
Source or reference
E. coli dH5␣F
⫺
␾80dlacZ⌬M15 ⌬(lacZYA-argF)U169 endA1 recA1 hsdR17 (r
k
⫺
,m
k
⫹
)
DeoR thi-1 phoA supE44 ␭
⫺
gyrA96 relA1
Life Technologies
B. subtilis
BS168 trpC2 Bacillus Genetic
Stock Center
BS1539 BS168 spoVAF::Kan
r
::lysA This work
BS1541 BS168 lysA controlled by P
blaP
promoter This work
BS1549 BS1541 amyE::rep Spc
r
blaI rep This work
BS1549S BS1541 amyE::rep This work
BS1567 BS1541 amyE::rep Spc
r
blaI rep blaP This work
BS1567S BS1541 amyE::rep blaP This work
B. licheniformis 749I 16
a
rep, repeat unit corresponding to the last 138 bp of the green fluorescent protein gene.
7242 BRANS ET AL. APPL.ENVIRON.MICROBIOL.

TABLE 2. Plasmids used in this study
Plasmids Description or structure Source or reference
pGEM-T Easy Cloning vector Promega
pPCR-Script SK(⫹) Cloning vector Stratagene
pIC333 Mini-Tn10 delivery vector 23
pAC7 Integrative vector at amyE locus from B. subtilis 168 22
pDML967 pSL1190 derivative with 0.7-kb PCR fragment containing gfp A. Brans
pDG792 Plasmid containing a kanamycin resistance gene 10
pDML995 pMK4 derivative with blaP from B. licheniformis 749I A. Brans
pRB373 B. subtilis/E. coli shuttle plasmid 1
pDML1515 pRB373 with 1.3-kb fragment containing blaP from B. licheniformis 749I A. Brans
pDML1570 This work
pDML1534 This work
pDML1536 This work
pDML1537 This work
pDML1538 This work
pDML1539 This work
pDML1540 This work
Continued on following page
VOL. 70, 2004 CHROMOSOMAL ENGINEERING OF BACILLUS SUBTILIS 7243

TABLE 2—Continued
Plasmids Description or structure Source or reference
pDML1541 This work
pDML1542 This work
pDML1543 This work
pDML1544 This work
pDML1545 This work
pDML1546 This work
pDML1566 This work
pDML1547 This work
pDML1548 This work
Continued on facing page
7244 BRANS ET AL. APPL.ENVIRON.MICROBIOL.

probably due to the partial cleavage of the pDML1539 plasmid
with the ApaI endonuclease.
Plasmid pDML1541 is a derivative of the E. coli pCR-Script
plasmid that contains the 1,244-bp spoVAF 3⬘end and the lysA
genes, between which a 362-bp fragment was inserted. This
fragment contains the regulated B. licheniformis P
blaP
pro-
moter of the ␤-lactamase blaP gene (for details see Materials
and Methods). This plasmid, previously linearized by the ac-
tion of the ApaI restriction endonuclease, was introduced into
BS1539 to achieve a double-crossover event. In the resulting
BS1541 strain, selected on minimal medium (Fig. 1), the ho-
mologous recombination restores the spoVAF gene, the Kan
r
gene is removed, and the lysA gene is under the control of the
P
blaP
promoter (for this experiment the yield of double cross-
over was around 1%). Consequently, the BS1541 strain could
be switched to lysine auxotrophy if the blaI gene that negatively
controls the P
blaP
promoter is added in this strain. For this
reason, the BS1541 strain is a conditionally auxotrophic strain
for lysine. The double-crossover events in the selected BS1539
and BS1541 strains were confirmed by Southern blot analysis
(Fig. 1).
Construction of the selection-eviction BlaI cassette. To
achieve positive selection for genetic modification of BS1541
and the eviction of the gene encoding antibiotic resistance used
as a selection marker, the 2,029-bp BlaI cassette illustrated in
Fig. 2 was constructed. In this cassette, the Spc
r
gene respon-
sible for spectinomycin resistance in Bacillus spp. and the blaI
gene encoding a DNA-binding protein are flanked by two
TABLE 3. Synthetic primers for PCR amplification
Name Length (bp) Sequence (5⬘33⬘)
BlaIbam⫹30 ATATGGATCCATCAAAATCGTCTCCCTCCG
promblaPBglII 32 ATCCAGATCTTCCCTCCGTTCATTTGTCCCCG
lysABamHI 34 ATGGATCCACACGGCACAAGCAGACAAAATCAAC
lysAend 25 CATTGATTTCTTCGTATCTATCTGG
repGFPbackup 33 GATATCGGATCCTTTTACCAGACAACCATTACC
repGFPbackdo 27 AGATCTATGACCATGATTACGCCAAGC
repGFPfrup 30 GTCGACTCCTTTTACCAGACAACCATTACC
repGFPfrdo 33 GATATCTCTAGATTTGTATAGTTCATCCATGCC
spoVAFBglI 31 AGATCTTGCGATTATGAATTGGTAGGCTGCC
spoVAFMluI 27 ACGCGTATGCCGGACCACAAGGAAGAG
pBlalBglII 30 CAGATCTAAGTGATGGAATTAAAAATGCAG
stopBlaIBamHI 31 CGGATCCTCATTCCTTCTTTCTGTTCTTATG
Spc⫹31 GGAAGTTCAATACTTGGAGTATATCTATTTG
Spc⫺21 CTTATCATCACACTCTCCCCG
BlaP⫹33 CGCTTCGATATAGTGACAATGCGGCACAGAATC
BlaP⫺31 GCCGTTCATGTTTAAGGCTTTCATTACCACC
BlaINdeI 30 ATACATATGAAAAAAATACCTCAAATCTCTG
BlaIEcoRI 37 ATAGAATTCATTTCATTCCTTCTTTCTGTTCTTATG
kanaup 22 ATGGCTAAAATGAGAATATCAC
kanadown 21 CTAAAACAATTCATCCAGTAA
amyEfront 25 TTTATTGCTGTTTCATTTGGTTCTG
amyEback 24 GATGGTGTATGTTTTGCCAAATTG
TABLE 2—Continued
Plasmids Description or structure Source or reference
pDML1549 This work
pDML1567 This work
VOL. 70, 2004 CHROMOSOMAL ENGINEERING OF BACILLUS SUBTILIS 7245

FIG. 1. (A) Construction of BS1539 (⌬spoVAF::Kan
r
⌬lysA) and BS1541 (P
blaP
lysA). pDML1539 and pDML1541 are pGEM-T-Easy and
pCR-Script derivatives, respectively, constructed as described in Materials and Methods. P
lysA
and P
blaP
are B. subtilis and B. licheniformis
promoters. The operator DNA sequences recognized by BlaI are indicated by shaded boxes. The Kan
r
gene was used for kanamycin selection. An
X indicates the recombination events leading to the chromosomal constructions of BS1539 and BS1541. In the latter strain, the B. licheniformis
P
blaP
promoter replaces the B. subtilis P
lysA
promoter. (B) Southern Blot analyses of the BglII-digested chromosomal DNA from chromosomal
BS168 (lanes 1, 4, and 7), BS1539 (lanes 2, 5, and 8), and BS1541 (lanes 3, 6, and 9) DNA. Panels I, II, and III contain the same patterns of DNA
digests hybridized to lysA (I), P
blap
(II), and Kan
r
(III) probes. lysA,P
blaP
, and Kan
r
probes were generated by PCR with the following pairs of
primers as amplimers: lysABamHI/lysAEnd, BlaIBam⫹/promblaPBglII, and kanaup/kanadown. pDML1539 or pDML1541 was the DNA template.
7246 BRANS ET AL. APPL.ENVIRON.MICROBIOL.

direct repeat units corresponding to the last 138 bp of the
Aequorea victoria gfp gene (for details see Materials and Meth-
ods). The pDML1548 plasmid, a pGEM-T-Easy derivative,
carries the BlaI cassette, which can be easily excised or fused
with another sequence of interest by the presence of flanking
unique restriction sites.
Integration and eviction of the BlaI cassette in BS1541. To
probe the feasibility of the eviction of the BlaI cassette after its
integration into the BS1541 chromosome, the BlaI cassette was
subcloned in the E. coli pAC7 plasmid. In the resulting
pDML1549 plasmid, the BlaI cassette is inserted between two
DNA fragments corresponding to the 5⬘and 3⬘ends (amy-
Efront and amyEback) of the nonessential B. subtilis amyE
gene, which encodes an ␣-amylase. Competent BS1541 cells
were transformed with linearized pDML1549 and were plated
on rich medium supplemented with 100 ␮g of spectinomycin/
ml. The selected spectinomycin-resistant BS1549 strain (Fig. 3)
is auxotrophic for lysine when the strain is replicated on min-
imal medium supplemented with spectinomycin (100 ␮g/ml),
and the double-crossover event was confirmed by the lack of
␣-amylase activity visualized on starch plate assay and by
Southern blot hybridization (data not shown). In BS1549, the
lysA gene under the control of the P
blaP
promoter is more
tightly repressed by BlaI than the blaI gene itself, because the
two BlaI operators located in the P
blaP
promoter are recog-
nized by BlaI with a higher affinity than the operator present in
the P
blaI
promoter (9) (Fig. 4). The auxotrophy of the BS1549
strain for lysine confirms that BlaI tightly represses the lysA
gene. The fact that this strain is unable to grow on minimal
medium without lysine suggests that it can be used as a coun-
terselection marker during the eviction of the BlaI cassette.
The eviction of the BlaI cassette by a single crossover event
(Fig. 3) was achieved by growing BS1549 for 24 h in 10 ml of
antibiotic-free LB medium. After dilution of the culture, the
clones isolated on minimal medium were analyzed for their
ability to grow on minimal medium alone or supplemented
with spectinomycin. It appeared that 50% of the isolated
clones that had recovered the ability to grow on minimal me-
dium without lysine (lysA
⫹
) were spectinomycin sensitive. The
BlaI cassette had thus been excised from these clones. To
increase the yield of the lysA
⫹
and spectinomycin-sensitive
recombinant strain, a 24-h culture of BS1549 cultivated in
FIG. 2. Schematic representation of the BlaI cassette. The blaI
gene is under the control of its own promoter (P
blaI
). Spc
r
indicates the
spectinomycin resistance gene from pIC333. Rectangles represent the
operator DNA sequence recognized by the BlaI repressor. repfront and
repback contain direct repeat unit sequences (138 bp) used for the
eviction of the BlaI cassette. For details see the text.
FIG. 3. Eviction of the BlaI cassette by single crossover between the two direct repeat unit sequences (repfront and repback) to generate the
BS1549S (⌬amyE) strain. The lysine auxotrophy of the BS1549 strain, linked to the presence of the blaI gene (conditional auxotrophy), was used
to select the recombinant strain that had excised the BlaI cassette.
VOL. 70, 2004 CHROMOSOMAL ENGINEERING OF BACILLUS SUBTILIS 7247

FIG. 4. Construction and characterization of BS1567 and BS1567S strains by PCR and Southern blot analyses. (A) Linearized pDML1567
plasmid carrying blaP, the BlaI cassette, and amyEfront and amyEback sequences. P
blaI
and P
blaP
are the native promoters of the B. licheniformis
749/I blaI and blaP genes, respectively. Rectangles correspond to the operator DNA sequences recognized by the BlaI repressor. (B) The
pDML1567 insert was introduced by double crossover into the BS1541 strain to generate the BS1567 strain. This strain exhibits lysine auxotrophy
7248 BRANS ET AL. APPL.ENVIRON.MICROBIOL.

antibiotic-free LB medium was diluted 1,000-fold into minimal
medium and was further cultivated for 24 h. This latter step
was repeated once, and the cells were spread on minimal agar
plates. After analysis, 95% of the clones in the 72-h culture
exhibited the lysA
⫹
and spectinomycin-sensitive phenotype.
This result highlights that after 72 h of culture, the selection
cassette had been excised in the majority of the lysA
⫹
clones.
All colonies that had the lysA
⫹
and spectinomycin-sensitive
phenotype (BS1549S) were tested for ␣-amylase production on
a starch plate assay and were all amyE mutants, in agreement
with the fact that, after the eviction of the counterselection and
selection markers, one repeat unit remains in the chromosome
and interrupts the amyE gene (Fig. 3). The reason why 50% of
BS1549 cells have not excised the BlaI cassette after 24 h of
culture in LB medium followed by selection on solid minimal
medium remains unexplained. One hypothesis is that the small
leakage of the P
blaP
promoter repressed by BlaI combined with
the exogenous lysine coming from the LB medium used to
inoculate the minimal medium could allow lysA mutant cells to
survive. On the other hand, a time course of the BlaI cassette
eviction followed by the appearance of reporter ␤-lactamase
activity showed that the BS1567 strain (see the next paragraph)
cultivated in minimal medium grew very slowly over the first
24h(A
600
varied from 0.094 to 0.165) and that the enzymatic
activity was only detected after 24 h of culture. Thereafter the
cells grew normally and reached a cellular density similar to
that of a B. subtilis lysA
⫹
strain.
Integration of a gene of interest in BS1541 and subsequent
eviction of the BlaI cassette. The B. licheniformis 749/I blaP
gene, encoding a class A ␤-lactamase, was used as a gene of
interest and was inserted between amyEback and the second
repeat of the BlaI cassette in the pDML1541 plasmid to gen-
erate pDML1567 (Fig. 4A). The chromosomal integration of
the blaP gene and the BlaI cassette into the BS1541 amyE gene
and the eviction of the BlaI cassette from the chromosome
were carried out as described above. This generated BS1567
(Spc
r
) and the desired spectinomycin-sensitive BS1567S strain
(Fig. 4B). These two strains were characterized by PCR and
Southern blot experiments, and their ability to express the blaP
gene was estimated by measuring the ␤-lactamase activity in
the culture medium. The PCR amplifications obtained by using
complementary amplimers for amyEfront and amyEback and
BS1567 or BS1567S chromosomal DNA as template is shown
in Fig. 4C. The amplified fragment in BS1567 (4.5 kb) was the
same length as that obtained when pDML1567 was used as
DNA template, and it corresponds to the amyE gene in which
blaP and the BlaI cassette have been inserted. By contrast,
when the BS1567S chromosome was used as DNA template,
the length of the amplified fragment was shorter by about 1.9
kb. This difference corresponds to the expected amplified frag-
ment resulting from the excision of a fragment containing one
repeat, blaI and the spectinomycin resistance gene, from the
BS1567 chromosome. Southern blot analysis supported this
conclusion (Fig. 4D). Indeed, BglII digestion of BS1567S chro-
mosomal DNA yielded no detectable signal when blaI and Spc
r
probes were used, whereas the presence of the blaP gene was
detected (Fig. 4D). The analysis of the protein content of
BS1567 and BS1567S culture media was carried out by SDS-
PAGE, and, as expected, the excision of the counterselection
and selection markers is correlated with overexpression of a
protein that exhibits an apparent molecular size of 31 kDa,
corresponding to that of the BlaP ␤-lactamase. The absence of
this band in the culture medium of BS1567 and the determi-
nation of the ␤-lactamase activity in the two supernatants high-
lights the repression of the ␤-lactamase synthesis when the
product of the blaI gene is present (Fig. 5). A 1,000-fold in-
crease in ␤-lactamase activity was observed for BS1567S (BlaP
specific activity per cell density, 12,000 ⫾300 nmol of nitro-
cefin hydrolyzed/min 䡠A
600
) compared to that of BS1567 (BlaP
specific activity per cell density, 15 ⫾1 nmol of nitrocefin
hydrolyzed/min 䡠A
600
). These experiments show that the re-
pression mediated by BlaI is very efficient and that the basal
expression of the gene of interest is maintained at a low level
until excision of the repressor. This result pinpoints another
interesting feature of the proposed method, i.e., the use of the
BlaI repressor to control the gene of interest. In this case, the
strong repression of the gene of interest until the excision of
the BlaI cassette allows one to determine the permissive in-
sertion sites in the Bacillus chromosome. Indeed, if no recom-
binant strain carrying the BlaI cassette is obtained for a specific
chromosome insertion, it can be concluded that this insertion
site is not permissive, because the lethal phenotype does not
result in the overexpression of the gene of interest but is linked
to the inactivation of the gene used for the target insertion site.
FIG. 5. ␤-Lactamase production in BS1567 and BS1567S. SDS-
PAGE analysis of BS1567 and BS1567S culture supernatants is shown.
MM, molecular mass marker.
due to the presence of the BlaI repressor that negatively controls the expression of the lysA gene. For the same reason, the expression of the blaP
gene is very low. The eviction of the BlaI cassette by single crossover between the two direct repeat unit sequences (repfront and repback) was
achieved as described in the legend to Fig. 3 to generate the BS1567S (⌬amyE blaP) strain. (C) PCR amplifications of amyE. The amplified
fragments were generated with primers amyEfront and amyEback and were analyzed by agarose gel electrophoresis. Lanes 1, 2, and 3 correspond
to PCR experiments carried out with BS1567 chromosomal DNA, plasmid pDML1567, and BS1567S chromosomal DNA as template, respectively.
MM, molecular size marker (Smart ladder; Eurogentec). (D) Southern Blot analysis of the BglII-digested chromosomal DNA from BS1567 (lanes
2, 5, and 8) and BS1567S (lanes 3, 6, and 9). Linearized pDML1567 (lanes 1, 4, and 7) was used as positive control. blaP,blaI, and Spc
r
probes
were generated by PCR with the following primers pairs as amplimers: BlaP⫹/BlaP⫺, BlaINdeI/BlaIEcoRI, and Spc⫹/Spc⫺. pDML1567 was the
DNA template. For more details see Materials and Methods.
VOL. 70, 2004 CHROMOSOMAL ENGINEERING OF BACILLUS SUBTILIS 7249

On the contrary, the absence of a recombinant that had excised
the BlaI cassette during selection on minimal medium suggests
that inappropriate expression of the protein of interest is le-
thal.
DISCUSSION
In the present study, we developed a method for directed
genetic manipulation of BS168 combined with eviction of the
antibiotic resistance gene used as a selection marker. This
method relies on the use of the B. licheniformis ␤-lactamase
P
blaP
promoter, regulated by the BlaI repressor, to confer
conditional lysine auxotrophy in a BS1541 strain in which the
endogenous P
lysA
promoter has been replaced with the B. li-
cheniformis P
blaP
promoter.
We used this method both to inactivate a specific gene and
to introduce a gene of interest into the BS1541 strain. This
strategy can also be used to replace one promoter with an-
other, to generate a large chromosomal deletion, or to deliver
a point mutation. In the latter case, the point mutation must be
present in the two direct repeat sequences flanking the BlaI
cassette. In addition, these genetic manipulations can be com-
bined in the same strain, because the strategy can be repeated
sequentially after eviction of the BlaI cassette. The direct re-
peat sequences flanking the BlaI cassette can be replaced by
any other nucleotide sequence; for example, by a portion of the
target sequence itself. This flexibility in choice of direct repeat
sequences offers many potential applications, among which are
(i) the selective deletion of gene(s) inserted in an operon
without alteration of the gene(s) downstream of the deleted
gene(s); (ii) the engineering of an artificial operon; and (iii) the
replacement of a promoter or the replacement of a nonessen-
tial endogenous gene with an exogenous gene without modi-
fying the native promoter and terminator of the replaced gene.
Compared to the delivery system described by Fabret et al. (6),
based on the use of the upp cassette and a B. subtilis upp mutant,
our approach gives similar results and offers the following two
advantages. First, in our strategy the selection of the lysA
⫹
and
spectinomycin-sensitive B. subtilis cells that have excised the BlaI
cassette can be achieved by a simple transfer of the selected lysA
⫹
cells onto a spectinomycin-containing medium. That is not the
case with the upp cassette strategy, for which PCR selection fol-
lowed by strain isolation is required. Second, in our case, except
for the introduced modification, the excised strain possesses the
same phenotype as the wild-type B. subtilis isolate and is not a upp
mutant. In conclusion, the strategy described in this paper is very
efficient and can be used as a tool to manipulate the B. subtilis
chromosome.
ACKNOWLEDGMENTS
This work was supported by the Belgian program on Interuniversity
Poles of Attraction, initiated by the Federal Office for Scientific, Tech-
nical, and Cultural Affairs (PAI no. P5/33), the Fond National de la
Recherche Scientifique (FNRS; cre´dit aux chercheurs no. 1.5201.02,
credit FRFC 2.4530.03), la Re´gion Wallonne (Bioval 3822–007), and
the Communaute´ Franc¸aise de Belgique (Actions de Recherche Con-
certe´es 2003–2008). B.J. is a FNRS associate researcher. P.F. was a
fellow of the Fonds pour la Formation a` la Recherche dans l’Industrie
et l’Agriculture (FRIA; Brussels, Belgium).
REFERENCES
1. Bruckner, R. 1992. A series of shuttle vectors for Bacillus subtilis and Esch-
erichia coli. Gene 122:187–192.
2. Collier, D. N. 1994. Expression of Escherichia coli SecB in Bacillus subtilis
facilitates secretion of the SecB-dependent maltose-binding protein of E.
coli. J. Bacteriol. 176:4937–4940.
3. Dahl, M. K., and C. G. Meinhof. 1994. A series of integrative plasmids for
Bacillus subtilis containing unique cloning sites in all three open reading
frames for translational lacZ fusions. Gene 145:151–152.
4. Driks, A. 1999. Bacillus subtilis spore coat. Microbiol. Mol. Biol. Rev. 63:1–
20.
5. Eichenberger, P., S. T. Jensen, E. M. Conlon, C. van Ooij, J. Silvaggi, J. E.
Gonzalez-Pastor, M. Fujita, S. Ben-Yehuda, P. Stragier, J. S. Liu, and R.
Losick. 2003. The ␴E regulon and the identification of additional sporulation
genes in Bacillus subtilis. J. Mol. Biol. 327:945–972.
6. Fabret, C., S. D. Ehrlich, and P. Noirot. 2002. A new mutation delivery
system for genome-scale approaches in Bacillus subtilis. Mol. Microbiol.
46:25–36.
7. Fahnestock, S. R., and K. E. Fisher. 1986. Expression of the staphylococcal
protein A gene in Bacillus subtilis by gene fusions utilizing the promoter from
aBacillus amyloliquefaciens ␣-amylase gene. J. Bacteriol. 165:796–804.
8. File´e, P., K. Benlafya, M. Delmarcelle, G. Moutzourelis, J. M. Fre`re, and B.
Joris. 2002. The fate of BlaI repressor during the induction of the Bacillus
licheniformis BlaP ␤-lactamase. Mol. Microbiol. 44:685–694.
9. File´e, P., C. Vreuls, R. Herman, I. Thamm, T. Aerts, P. P. De Deyn, J. M.
Fre`re, and B. Joris. 2003. Dimerization and DNA binding properties of the
Bacillus licheniformis 749/I BlaI repressor. J. Biol. Chem. 278:16482–16487.
10. Guerout-Fleury, A. M., K. Shazand, N. Frandsen, and P. Stragier. 1995.
Antibiotic-resistance cassettes for Bacillus subtilis. Gene 167:335–336.
11. Harwood, C. R., and S. M. Cutting (ed.). 1990. Molecular methods for
Bacillus. John Wiley & Sons, Chichester, United Kingdom.
12. Kobayashi, K., S. D. Ehrlich, A. Albertini, G. Amati, K. K. Andersen, M.
Arnaud, K. Asai, S. Ashikaga, S. Aymerich, P. Bessieres, F. Boland, S. C.
Brignell, S. Bron, K. Bunai, J. Chapuis, L. C. Christiansen, A. Danchin, M.
Debarbouille, E. Dervyn, E. Deuerling, K. Devine, S. K. Devine, O. Dreesen,
et al. 2003. Essential Bacillus subtilis genes. Proc. Natl. Acad. Sci. USA
100:4678–4683.
13. Kobayashi, T., Y. F. Zhu, N. J. Nicholls, and J. O. Lampen. 1987. A second
regulatory gene, blaR1, encoding a potential penicillin-binding protein re-
quired for induction of ␤-lactamase in Bacillus licheniformis. J. Bacteriol.
169:3873–3878.
14. Kunst, F., T. Msadek, and G. Rapoport. 1994. Signal transduction network
controlling degradative enzyme synthesis and competence in Bacillus subtilis,
p. 1–20. In P. J. Piggot, C. P. Moran, and P. Youngman (ed.), Regulation of
bacterial differentiation. American Society for Microbiology, Washington,
D.C.
15. Kunst, F., N. Ogasawara, I. Moszer, A. M. Albertini, G. Alloni, V. Azevedo,
M. G. Bertero, P. Bessieres, A. Bolotin, S. Borchert, R. Borriss, L. Boursier,
A. Brans, M. Braun, S. C. Brignell, S. Bron, S. Brouillet, C. V. Bruschi, B.
Caldwell, V. Capuano, N. M. Carter, S. K. Choi, J. J. Codani, I. F. Conner-
ton, A. Danchin, et al. 1997. The complete genome sequence of the gram-
positive bacterium Bacillus subtilis. Nature 390:249–256.
16. Lampen, J. O., J. B. Nielsen, K. Izui, and M. P. Caulfield. 1980. Bacillus
licheniformis beta-lactamases: multiple forms and their roles. Philos. Trans.
R. Soc. Lond. B Biol. Sci. 289:345–348.
17. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecular cloning: a
laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring Har-
bor, N.Y.
18. Olmos-Soto, J., and R. Contreras-Flores. 2003. Genetic system constructed
to overproduce and secrete proinsulin in Bacillus subtilis. Appl. Microbiol.
Biotechnol. 62:369–373.
19. Pierce, J. A., C. R. Robertson, and T. J. Leighton. 1992. Physiological and
genetic strategies for enhanced subtilisin production by Bacillus subtilis.
Biotechnol. Prog. 8:211–218.
20. Rosner, A. 1975. Control of lysine biosynthesis in Bacillus subtilis: inhibition
of diaminopimelate decarboxylase by lysine. J. Bacteriol. 121:20–28.
21. Sambrook, J. E. F. Fritsch, and T. Maniatis.1989. Molecular cloning: a
laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold
Spring Harbor, N.Y.
22. Shimotsu, H., and D. J. Henner. 1986. Construction of a single-copy inte-
gration vector and its use in analysis of regulation of the trp operon of
Bacillus subtilis. Gene 43:85–94.
23. Steinmetz, M., and R. Richter. 1994. Easy cloning of mini-Tn10 insertions
from the Bacillus subtilis chromosome. J. Bacteriol. 176:1761–1763.
24. Williams, R. C., M. L. Rees, M. F. Jacobs, Z. Pragai, J. E. Thwaite, L. W.
Baillie, P. T. Emmerson, and C. R. Harwood. 2003. Production of Bacillus
anthracis protective antigen is dependent on the extracellular chaperone,
PrsA. J. Biol. Chem. 278:18056–18062.
25. Wu, S. C., M. Hassan Qureshi, and S. L. Wong. 2002. Secretory production
and purification of functional full-length streptavidin from Bacillus subtilis.
Protein Expr. Purif. 24:348–356.
26. Wu, S. C., J. C. Yeung, Y. Duan, R. Ye, S. J. Szarka, H. R. Habibi, and S. L.
Wong. 2002. Functional production and characterization of a fibrin-specific
single-chain antibody fragment from Bacillus subtilis: effects of molecular
chaperones and a wall-bound protease on antibody fragment production.
Appl. Environ. Microbiol. 68:3261–3269.
7250 BRANS ET AL. APPL.ENVIRON.MICROBIOL.