JOURNAL OF BACTERIOLOGY, Oct. 2007, p. 6816–6823
Copyright © 2007, American Society for Microbiology. All Rights Reserved.
Vol. 189, No. 19
The Amino Terminus of Bacillus subtilis TagB Possesses Separable
Localization and Functional Properties?†
Amit P. Bhavsar,‡ Michael A. D’Elia, Tiffany D. Sahakian, and Eric D. Brown*
Antimicrobial Research Centre and Department of Biochemistry and Biomedical Sciences,
McMaster University, Hamilton, Ontario L8N 3Z5, Canada
Received 10 June 2007/Accepted 20 July 2007
The function(s) of gram-positive wall teichoic acid is emerging with recent findings that it is an important
virulence factor in the pathogen Staphylococcus aureus and that it is crucial to proper rod-shaped cell mor-
phology of Bacillus subtilis. Despite its importance, our understanding of teichoic acid biosynthesis remains
incomplete. The TagB protein has been implicated in the priming step of poly(glycerol phosphate) wall teichoic
acid synthesis in B. subtilis. Work to date indicates that the TagB protein is localized to the membrane, where
it adds a single glycerol phosphate residue to the nonreducing end of the undecaprenol-phosphate-linked
N-acetylmannosamine-?(1,4)-N-acetylglucosamine-1-phosphate. Thus, membrane association is critical to
TagB function. In this work we elucidate the mechanism of TagB membrane localization. We report the
identification of a membrane targeting determinant at the amino terminus of TagB that is necessary and
sufficient for membrane localization. The putative amphipathicity of this membrane targeting determinant was
characterized and shown to be required for TagB function but not localization. This work shows for the first
time that the amino terminus of TagB mediates membrane targeting and protein function.
Wall teichoic acids (WTAs) are phosphate-rich anionic
extracellular polysaccharides found covalently bound to pepti-
doglycan in gram-positive bacteria. In the model gram-positive
bacterium Bacillus subtilis 168, the major WTA is a poly(glyc-
erol phosphate) polymer. Biosynthesis of WTA is carried out
by enzymes encoded by the tag genes (19, 21, 32, 33). The
inability to mutate specific tag genes suggested that WTA was
indispensable for cell viability (1, 2, 5, 6, 28). Remarkably,
while these genes were indispensable in single gene deletion
experiments, the dispensability phenotype was contextual. Re-
cent studies have revealed that the first step (tagO) of teichoic
acid biosynthesis is dispensable in both B. subtilis and Staphy-
lococcus aureus (12, 13, 37). Indeed, we demonstrated that the
late-acting genes tagB, tagD, and tagF could be deleted in both
B. subtilis and S. aureus only in the presence of an accompa-
nying deletion in tagO, encoding the first enzyme in the tei-
choic acid biosynthetic pathway (12, 13). Nevertheless, mutants
with mutations in the first step of teichoic acid synthesis (tagO)
have shown that the polymer is vital for rod shape in B. subtilis
(12) and for virulence in S. aureus infection (37).
Sequence-based homology of Tag proteins coupled with the
known chemical structure of WTA has led to the proposal of a
plausible biosynthetic model (18), whereby WTA synthesis is
initiated on the cytoplasmic face of the membrane on an un-
decaprenol phosphate molecule. Polymer synthesis is carried
out stepwise through addition of N-acetylglucosamine-1-phos-
phate, N-acetylmannosamine, and glycerol phosphate to the
membrane-embedded prenol phosphate substrate. Glycerol phos-
phate addition, from the activated precursor CDP-glycerol, is
proposed to be catalyzed by two enzymes, TagB and TagF,
which share extensive sequence homology. We reported else-
where that TagB catalyzed a reaction that was consistent with
the addition of a single labeled glycerol phosphate to a mem-
brane-bound acceptor (3). Furthermore, mutagenic analysis of
the two proteins indicated a mechanistic similarity between
these enzymes that implicated two conserved histidines as cru-
cial catalytic residues (30). Subsequently TagB was shown to
catalyze glycerol phosphate addition to an acceptor analogue
of undecaprenol-phosphate-linked disaccharide (14). Thus, all
evidence to date supports a role for TagB as a primase whose
product is the substrate for TagF, the teichoic acid polymerase.
We also previously reported that TagB was localized to the
cytoplasmic face of the B. subtilis membrane (3). Interestingly,
we found that this association with the membrane was periph-
eral and not integral as predicted by sequence-based topology
analysis. The same analysis indicated that a predicted amino-
terminal transmembrane helix possessed an amphipathic char-
acter where polar but uncharged residues were oriented to the
same face of the helix. In work reported here we characterize
the role of the amino terminus of TagB in membrane associ-
ation and function. We show by truncation analysis and fusion
to a heterologous carrier protein that this region of TagB
carries a membrane targeting determinant that is necessary
and sufficient for efficient membrane association. In addition,
we characterize the putative amphipathicity of this region and
show that insertional mutagenesis, designed to disrupt the am-
phipathic helical face, has no effect on membrane localization
but a profound impact on TagB function, suggesting that this
region of the protein contains an unusual amphipathic helix.
Taken together, our results suggest that the amino terminus of
TagB mediates membrane targeting and protein function.
* Corresponding author. Mailing address: Department of Biochem-
istry and Biomedical Sciences, McMaster University, 1200 Main Street
West, Hamilton, Ontario L8N 3Z5, Canada. Phone: (905) 525-9140,
ext. 22392. Fax: (905) 522-9033. E-mail: email@example.com.
† Supplemental material for this article may be found at http://jb
‡ Present address: Michael Smith Laboratories, University of British
Columbia, 301-2185 East Mall Road, Vancouver, BC V6T 1Z4, Canada.
?Published ahead of print on 27 July 2007.
MATERIALS AND METHODS
Bacterial strains, reagents, and general methods. The strains and plasmids
and the oligonucleotides used in this study are listed in Tables 1 and 2, respec-
tively. All chemicals, unless otherwise noted, were purchased from Sigma
(Oakville, ON, Canada). General cloning methods for Escherichia coli and B.
subtilis were used according to established protocols (11, 29). Reagents for
molecular cloning were purchased from New England Biolabs (Beverly, MA).
Cultures were grown on Luria-Bertani (LB) medium with antibiotic selection as
follows, where appropriate: 50 ?g/ml ampicillin, 10 ?g/ml chloramphenicol
(CHL), and 25 ?g/ml kanamycin.
Construction of strains used in this study. pRBtagB?N30gfp was constructed by
first PCR amplifying tagB?N30 using tagB?N30for and tagBgfpRImutrev primers.
pRBtagB?N30gfp and transformed into B. subtilis 168 following passage through E.
coli MC1061 (9). Transformants were selected for kanamycin resistance and verified
by diagnostic digestion of plasmid isolated from the transformed cells.
To generate a recombinant construct with the first 30 residues of TagB fused
to a passive carrier protein, the phoA gene (starting at the 64th nucleotide in
order to omit the signal sequence) was amplified from E. coli genomic DNA
using primers phoAfor and phoArev. The amplified product was digested with
TABLE 1. Strains and plasmids used in this study
Strain or plasmid Genotype/descriptionReference or source
hisA1 argC4 metC3
hisA1 argC4 metC3 amyE::cat86 xylR PxylAtagBN30phoA
EB6 with pRBtagB?N30gfp
pheA1 purA16 hisA35 trpC2 tag-1
pheA1 purA16 hisA35 trpC2 tag-1 amyE::cat86 xylR PxylA
pheA1 purA16 hisA35 trpC2 tag-1 amyE::cat86 xylR PxylAtagBphoA
pheA1 purA16 hisA35 trpC2 tag-1 amyE::cat86 xylR PxylAtagB?1phoA
pheA1 purA16 hisA35 trpC2 tag-1 amyE::cat86 xylR PxylAtagB?2phoA
pheA1 purA16 hisA35 trpC2 tag-1 amyE::cat86 xylR PxylAtagB?3phoA
pheA1 purA16 hisA35 trpC2 tag-1 amyE::cat86 xylR PxylAtagB?4phoA
pheA1 purA16 hisA35 trpC2 tag-1 amyE::cat86 xylR PxylAtagBN30?1phoA
pheA1 purA16 hisA35 trpC2 tag-1 amyE::cat86 xylR PxylAtagBN30?2phoA
pheA1 purA16 hisA35 trpC2 tag-1 amyE::cat86 xylR PxylAtagBN30?3phoA
pheA1 purA16 hisA35 trpC2 tag-1 amyE::cat86 xylR PxylAtagBN30?4phoA
Novablue endA1 hsdR17(rK-12
?araD139? ?(araA-leu)7697 ?(codB-lacI)3 galK16 galE15 ??mcrA mutant e14
mutant relA1 rpsL150(Strr) spoT1 mcrB mutant hsdR2
?) supE44 thi-1 recA1 gyrA96 relA1 lac ?F? proA?B?
pBluescript with wild-type tagB from B. subtilis
B. subtilis xylose-inducible ectopic expression vector
pSWEET with tagBphoA insert
pSWEET with tagB?1phoA insert
pSWEET with tagB?2phoA insert
pSWEET with tagB?3phoA insert
pSWEET with tagB?4phoA insert
pSWEET with tagBN30?1phoA insert
pSWEET with tagBN30?2phoA insert
pSWEET with tagBN30?3phoA insert
pSWEET with tagBN30?4phoA insert
pRB374 with tagBgfp fusion insert
pRB374 with tagB?N30gfp fusion insert
TABLE 2. Oligonucleotides used in this study
NameSequence (5? to 3?)
VOL. 189, 2007TagB MEMBRANE TARGETING DETERMINANT CHARACTERIZATION 6817
NheI and BamHI and cloned into a derivative of pBStagB (lacking the stop
codon of tagB) digested with the same sites. The tagBphoA fragment was cloned
into pSWEET-bgaB using PacI and BamHI restriction sites to replace bgaB, and
the plasmid was named pSWEET-tagBphoA. Finally, coding sequence for the
first 30 residues of TagB was amplified from pBStagB using the T7 forward and
tagBN30rev1 primers, digested with PacI and NheI, and cloned into pSWEET-
tagBphoA digested with the same enzymes, essentially replacing the tagB open
reading frame (ORF) with one that coded for only the first 30 residues of TagB.
This plasmid was renamed pSWEET-tagBN30phoA.
To generate TagB amino-terminal insertional variants with one, three, and
four amino acid insertions, tagB was amplified using forward primers
tagB?1Afor, tagB?3ALAfor, and tagB?4ALALfor, respectively, with the com-
mon reverse primer tagBgfpRImutrev. This reverse primer incorporates a silent
mutation that disrupts the EcoRI site at the 3? end of tagB. The amplified
product was digested with PacI and XhoI restriction sites and cloned into
pRBtagBgfp digested with the same sites. The tagB insertional variant ORFs were
excised from these constructs using PacI and NheI sites (retaining an additional
methionine residue at the carboxyl terminus of TagB) and cloned into pSWEET-
tagBN30phoA digested with the same sites. These plasmids were named
pSWEET-tagB?1phoA, pSWEET-tagB?3phoA, and pSWEET-tagB?4phoA. To
generate the TagB double amino acid insertional variant, tagB was amplified
using primers tagB?2ALfor and tagBrev2. The product was digested with PacI
and NheI restriction enzymes and cloned into pSWEET-tagBN30phoA digested
with the same enzymes. This plasmid was named pSWEET-tagB?2phoA.
To generate insertional variants in the TagBN30PhoA protein, the
tagBN30phoA ORF was amplified using the insertional variant forward primers
listed above with the phoArev1 reverse primer. The products were digested with
PacI and BamHI and cloned into pSWEET-bgaB digested with the same en-
zymes. These clones were named pSWEET-tagBN30?1phoA, pSWEET-
tagBN30?2phoA, pSWEET-tagBN30?3phoA, and pSWEET-tagBN30?4phoA.
All clones were verified by sequence analysis and ectopically integrated into B.
Immunodetection of TagB in B. subtilis. Detection of TagB derivatives was
performed as previously published (3). Briefly, strains were grown in LB medium
at 30°C until late log phase. Cells were harvested, and pellets were resuspended
in Bacillus lysis buffer and normalized to the same value for optical density at 600
nm. Cells were disrupted by passage through a French pressure cell, and the
ensuing lysate was clarified by differential ultracentrifugation. Immunodetection
was performed using commercial anti-PhoA polyclonal antiserum (Chemicon
International, Temecula, CA), commercial BD Living Colors polyclonal antibody
(BD Biosciences Canada, Mississauga, ON, Canada), anti-FtsY polyclonal anti-
serum, anti-TagD polyclonal antiserum, or anti-EzrA polyclonal antiserum. Don-
key anti-rabbit immunoglobulin G conjugated to horseradish peroxidase (BIO/
CAN Scientific, Mississauga, ON, Canada) was used as secondary antibody.
Immunoblots were detected using the Western Lightning Chemiluminescence
Reagent kit (Perkin-Elmer, Woodbridge, ON, Canada) according to the manu-
Fluorescence microscopy of B. subtilis. Strains EB892 and EB925, which con-
stitutively expressed TagBGFP and TagB?N30GFP, respectively, were fixed as
previously described (3). Fixed samples were visualized using a Leica Upright
model DM6000 motorized microscope with a Semrock green fluorescent protein
(GFP) filter cube, an Exfos X-Cite metal halide fluorescence light source, and a
Leica HCX Plan Apo 63? oil 1.4-numerical-aperture objective. Images were
captured using a Hamamatsu Orca AG monochrome camera and processed
using Volocity Acquisition and Restoration software (Improvision).
Conditional complementation of tag-1 allele. The tagB variants, cloned into the
xylose-inducible pSWEET ectopic integration vector, were transformed into
strain EB486. Transformants were confirmed by starch utilization assay (10) and
PCR. The procedure was performed essentially as previously reported (30).
Briefly, cultures of transformants containing the tag-1 allele and a complement-
ing tagB mutant allele were grown in LB-CHL at 30°C until an optical density at
600 nm of ?0.7 was reached. Samples were normalized for cell density and
serially diluted by 10-fold. Two microliters of each diluted sample was spotted
onto LB-CHL-xylose solid medium and grown at 47°C. To monitor expression of
TagB constructs from the amyE locus, we took advantage of the carboxyl-
terminal PhoA fusion. Strains EB1240 and EB1652 to 1656 were grown at the
permissive temperature of 30°C in the presence of 2% xylose. Cultures were
grown to mid-logarithmic phase, pelleted, and resuspended in Tris-EDTA buffer.
Samples were treated with 1 mg/ml lysozyme for 15 minutes at 37°C and finally
boiled in sodium dodecyl sulfate-polyacrylamide loading buffer. Samples were
separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, trans-
ferred to nitrocellulose, and immunoblotted with anti-PhoA antibody and anti-
FtsY polyclonal antibody as a loading control.
Identification of a membrane targeting determinant at the
amino terminus of TagB. We previously reported that TagB
localized to the cytoplasmic face of the cell membrane via a
peripheral interaction (3). The peripheral interaction was sur-
prising given that sequence-based prediction programs univer-
sally identified a transmembrane helix between residues 6 and
25 of TagB. In the absence of any published data regarding the
mechanism of TagB localization to the cell membrane, we
began by determining the contribution of this sequence to
membrane localization. We generated an amino-terminal trun-
cation variant of TagB that deleted the first 30 residues of
TagB, TagB?N30. However, we were unable to detect this
variant in B. subtilis lysate by Western blotting. Given the ready
detection of full-length TagB from the same expression system,
we hypothesized that this construct was unstable. In an attempt
to increase the stability of this truncation variant we generated
an in-frame carboxyl-terminal fusion to the GFP. Following
expression of this fusion protein in B. subtilis, we were able to
readily detect the fusion protein by immunoblotting, suggest-
ing that the addition of GFP to the carboxyl terminus of this
TagB truncation variant increased the stability of the variant
Lysate from strain EB925 was separated into soluble and
membrane fractions and subjected to immunodetection using
commercial anti-GFP antibody. Compared to the nontrun-
cated fusion protein (3), we observed a profound redistribution
of TagB?N30GFP localization with the majority of the trun-
cated protein found in the soluble lysate (Fig. 1A). To control
for our fractionation procedure, we analyzed the localization
of TagD, the soluble glycerol-3-phosphate cytidylyltransferase
(23), and EzrA, a membrane-bound cell division protein (20).
TagD was predominantly distributed in the soluble lysate,
whereas EzrA was predominantly distributed in the membrane
fraction, verifying that membranes were separated from solu-
ble lysate under our differential ultracentrifugation conditions.
The increased distribution of TagB?N30GFP in the soluble
lysate was consistent with fluorescence microscopy examina-
tion of strain EB925. We note that in contrast to the peripheral
fluorescence signal, normally observed with TagBGFP, the
TagB?N30GFP-expressing strain had a diffuse fluorescence
signal throughout the cell (see Fig. S1 in the supplemental
material). These data suggest that the amino terminus of TagB
is necessary for efficient membrane localization in B. subtilis.
Characterization of heterologous protein targeting by the
amino terminus of TagB. We next tested if the membrane
targeting function of the TagB amino terminus was indepen-
dent of the remainder of the TagB protein. In this regard we
attempted to create hybrid proteins that fused the first 30
amino acids of TagB to a heterologous passive carrier protein.
We first tried to create a fusion of these residues to GFP.
Intriguingly, we were unable to generate this recombinant
clone in the pRB374 vector in E. coli. It has previously been
reported that the vegII promoter, found on this plasmid, is
capable of transcription in both E. coli and B. subtilis (7),
suggesting that the TagBN30GFP ORF was toxic to E. coli. As
an alternative we employed a gram-positive specific xylose-
inducible expression system that allowed ectopic integration
into B. subtilis at the amyE locus (4). We created a fusion of
6818BHAVSAR ET AL.J. BACTERIOL.
residues 1 to 30 of TagB to the amino terminus of the mature
form of E. coli alkaline phosphatase (PhoA), e.g., PhoA lack-
ing the first 21 residues that comprise its signal sequence,
under the transcriptional control of the xylA promoter. E. coli
PhoA was chosen because we were able to detect expression of
PhoA in B. subtilis by immunoblot analysis.
B. subtilis lysate from strains expressing TagBN30PhoA
(EB921) or the unfused PhoA ORF was generated by lysis of
cells grown in the presence of xylose. Differential ultracentrif-
ugation was used to generate soluble lysate and membrane
fractions from these strains. Immunodetection analyses using
commercial anti-PhoA antibody showed that the unfused
PhoA control (lacking its signal sequence) was localized in the
soluble lysate (data not shown). In contrast, the TagBN30PhoA
ORF was predominantly localized to the membrane fraction
(Fig. 1B). This suggested that the TagB amino terminus is
sufficient for membrane localization in B. subtilis.
Disruption of TagB amino-terminal amphipathicity. We
previously noted that the putative transmembrane helix at the
amino terminus of TagB possessed an unusual amphipathicity
in which polar but uncharged residues were situated on one
face of the helix when examined on a helical wheel (3). To
examine the contribution of amphipathicity to TagB function
and localization, we created insertional variants of TagB that
were predicted by helical wheel analysis to disrupt the amphi-
pathic face of the TagB helix (Fig. 2). We created insertional
variants that introduced one, two, three, and four amino acids
at position 10 of TagB. Insertion of up to three residues was
predicted to disrupt the amphipathic face to various extents. In
contrast, the insertion of four amino acids was predicted to
restore amphipathicity to the putative helix, in essence adding
a full turn to the helix. Given that localization of TagB to the
membrane allows the enzyme to be situated in close proximity
to its membrane-embedded substrate, we were interested in
FIG. 1. Localization of TagB?N30GFP and TagBN30PhoA in B. subtilis 168. Whole-cell lysates (Lys.) of EB925 (tagB?N30gfp) (A) and EB921
(tagBN30phoA) (B) were fractionated into soluble lysate (Sol.) and membrane samples (Mem.) by differential ultracentrifugation. Samples were
subjected to immunodetection analysis using anti-GFP and anti-PhoA antisera, respectively. Samples were also probed with anti-TagD and
anti-EzrA antisera to assess the quality of fractionation.
FIG. 2. Predicted impact of insertional mutagenesis on helical amphipathicity of TagB. Residues 6 to 25 of TagB (WT) are shown in helical
wheel analysis. The sequence of each residue is indicated, and polar residues are shaded gray. Helical wheel analyses of TagB variants containing
insertions of up to four residues at position 10 are also depicted (denoted ?1, ?2, ?3, and ?4, respectively). The specific amino acids inserted
at position 10 are indicated by black circles. Note the change in the distribution of polar residues (shaded gray) resulting from amino acid insertion
such that the amphipathic face of the helix is disrupted by insertion of up to three residues but is restored by insertion of four residues.
VOL. 189, 2007 TagB MEMBRANE TARGETING DETERMINANT CHARACTERIZATION6819
learning if amino-terminal amphipathicity was a mechanism
for TagB localization or function.
In vivo activity of TagB insertional variants. To examine
TagB activity, we employed a previously characterized in vivo
activity assay that is based upon the conditional complemen-
tation, by xylose-inducible TagB variants, of the temperature-
sensitive tag-1 allele, identified as a TagBG188D mutation
(30). An in vivo functional assay was employed because previ-
ous work has shown that this assay faithfully recapitulates the
results of the in vitro functional assay of the TagF-like family
of enzymes and, additionally, the TagB in vivo assay is much
more robust than its in vitro counterpart (3, 30). We appended
E. coli PhoA to the carboxyl termini of the TagB variants to
allow verification of variant protein expression by immunode-
tection analysis. For these studies we examined the effect of
insertion of up to four amino acids at position 10 of TagB. The
xylose-inducible variant constructs were ectopically introduced
at the amyE locus of strain EB486 and verified by PCR anal-
ysis. Strains harboring the tag-1 allele at the tag locus and
either empty vector, wild-type tagB, or variants of tagB con-
taining the alanine (single), alanine/leucine (double), alanine/
leucine/alanine (triple), and alanine/leucine/alanine/leucine
(quadruple) residue insertions integrated at the amyE locus
were initially grown at the permissive temperature, normalized
for growth, and grown at the restrictive temperature (to inac-
tivate endogenous TagB) under conditions that induced vari-
ant protein expression.
As shown in Fig. 3A, the absence of a xylose-inducible copy
of tagB yielded little growth at the nonpermissive temperature
in the presence of xylose. We noted that the carboxyl-terminal
fusion of PhoA to TagB did not affect TagB activity in the in
vivo assay to an appreciable extent because the wild-type copy
of tagB fused to phoA under the transcriptional control of PxylA
(EB1652) gave rise to robust growth at the restrictive temper-
ature in the presence of xylose. We did not observe any ap-
preciable difference in growth between wild-type TagB
(EB1652) and a variant of TagB that contained a single alanine
insertion at position 10 (TagB?1PhoA; EB1653) where both
strains grew to dilutions of 10?5. In contrast, insertion of either
alanine/leucine (EB1654) or alanine/leucine/alanine (EB1655)
at position 10 had a marked effect on TagB function. For both
these protein variants growth was only 10-fold better than that
of the negative control (EB1240) and approximately 2 to 3
orders of magnitude impaired with respect to the wild type
(EB1652). Interestingly, the impairment of TagB function due
to insertion of two or three amino acids at position 10 could be
suppressed by the addition of four amino acids at the same
position. The growth of this strain (EB1656) was indistinguish-
able from that of the positive control (EB1652).
To ensure that growth impairment in the TagB in vivo assay
was not due to poor expression of TagB variants, whole-cell
lysate of each strain (EB1240 and EB1652-EB1656) was pre-
pared under inducing conditions and subjected to immunode-
tection analyses with commercial anti-PhoA antibody. As
shown in Fig. 3B we could clearly detect TagB and TagB
variant expression from the xylose-inducible promoter. This
analysis suggests that all TagB variants were expressed to sim-
ilar levels, implying that protein expression was not the cause
of impaired activity in the in vivo activity assay. To ensure
equal loading of cell lysates, we analyzed levels of FtsY pro-
Localization of TagB insertional variants. To ascertain
whether the impaired in vivo activity of TagB?2PhoA and
TagB?3PhoA variants was due to altered subcellular localiza-
tion of TagB, cultures expressing the wild type and all four
variant proteins were grown at the permissive temperature
under inducing conditions. Whole-cell lysates from each cul-
ture were fractionated by differential ultracentrifugation into
soluble lysate and membrane samples. The distribution of
TagB within these fractions was examined by immunodetection
using commercial PhoA antibody (Fig. 4A). Again, the distri-
bution of soluble TagD and membrane-bound EzrA proteins
was also examined to validate the fractionation procedure.
These analyses verified separation of soluble lysate and mem-
branes under these conditions as TagD exclusively localized in
the soluble lysate and EzrA predominantly localized to the
membrane. As shown in Fig. 4A, the insertional variants of
TagB all appeared membrane localized, suggesting that altered
subcellular distribution of the insertional variants could not
account for the diminished in vivo activity of TagB?2PhoA
Although there did not appear to be an impact on the lo-
calization of the TagB insertional variants, we were mindful
that a minor amount of TagB remained localized to the mem-
brane in the absence of its first 30 residues (Fig. 1A). To
determine if another membrane targeting determinant en-
coded in the remaining TagB sequence was responsible for the
membrane localization of the insertional variants, the same
FIG. 3. In vivo functional analyses of TagBPhoA insertional vari-
ants. (A) TagB activity was monitored by complementation of the tag-1
allele at the restrictive temperature via xylose-inducible insertional
variants of TagB. Strains EB1240 (empty vector; -ve), EB1652 (tagB-
phoA; wt), EB1653 (tagB?1phoA; ?1), EB1654 (tagB?2phoA; ?2),
EB1655 (tagB?3phoA; ?3), and EB1656 (tagB?4phoA; ?4) were
grown at the permissive temperature, diluted as indicated, spotted on
LB-CHL-xylose solid medium, and grown at 47°C. (B) Immunodetec-
tion analysis of TagB variant expression from cell lysate using anti-
PhoA antiserum. Immunodetection of FtsY levels was used as a load-
ing control. The asterisk denotes a cross-reactive band.
6820 BHAVSAR ET AL. J. BACTERIOL.
insertions were generated in the TagBN30PhoA background.
We reasoned that any impact on localization could be unam-
biguously attributed to insertion in the membrane targeting
determinant since in the TagBN30PhoA construct only the first
30 residues of TagB directed membrane targeting (Fig. 1B).
Strains carrying a xylose-inducible copy of TagBN30PhoA with
insertions of one to four amino acids at position 10 (EB1696 to
EB1699, respectively) were grown at the permissive tempera-
ture under inducing conditions. Lysates were generated and
fractionated as indicated above. Immunodetection analysis us-
ing commercial anti-PhoA antibody again showed no appre-
ciable difference in distribution of the TagBN30PhoA variants,
with the bulk of the proteins localizing to the membrane (Fig.
4B). Again, the observed distributions of TagD and EzrA
served to ensure adequate separation of membranes and sol-
uble lysate. The membrane localization of the TagB insertional
variants suggests that insertional disruption of the amino-ter-
minal membrane targeting determinant does not impact on
membrane targeting even though it results in a profound de-
crease of TagB function. This further suggests that the impact
on TagB function is manifested at the level of biochemical
Teichoic acid synthesis has been the subject of considerable
study genetically (26) and biochemically. The latter encom-
passes vintage work with crude cell fractions (36) and more-
recent studies with pure recombinant proteins (3, 14, 23, 24, 31,
33). While these studies have provided important insights into
teichoic acid biogenesis, we still have a limited structure-func-
tion understanding of the enzymes directly involved in polymer
synthesis. Indeed, biosynthesis models place teichoic acid syn-
thesis at the cytoplasmic membrane, and yet how Tag proteins
mediate this localization is unclear. In this work we have iden-
tified a membrane targeting determinant at the amino termi-
nus of TagB. This region of the protein, encompassing the first
30 residues of TagB, is necessary and sufficient for membrane
localization. We have also identified an unusual amphipathicity
in this determinant, the disruption of which impacted on pro-
tein function but not localization.
The proposed TagB amino-terminal membrane targeting
determinant identified in this work is often identified as a
transmembrane helix using sequence-based prediction pro-
grams. On the basis of extraction data via chaotropic agent or
alkali treatment, we previously concluded that TagB, in fact,
interacted peripherally with the cell membrane (3). Further-
more, we predicted that this region of TagB, putatively encod-
ing an ?-helix, had an amphipathic character, in which polar
but uncharged amino acids were distributed in a similar orien-
tation about the helical face (3). This is reminiscent of the
peripheral membrane proteins MinD and FtsA, which use am-
phipathic helices consisting of positively charged and hydro-
phobic amino acids to mediate interaction with the cytoplasmic
face of the bacterial membrane (16, 25, 34). It is believed that
the positively charged residues of the amphipathic helix inter-
act electrostatically with phospholipid head groups to provide
the initial recruitment to the membrane (17). In the case of
MinD it is hypothesized that the hydrophobic residues of the
amphipathic helix insert into the membrane directly upon ATP
binding (38). However, the lack of charged residues in the
putative helical region of this determinant (residues 6 to 25)
suggests that TagB may use a mechanism different than either
FtsA or MinD for membrane interaction. In this respect it is
noteworthy that high ionic strength could only partially extract
TagB from B. subtilis membranes (A. P. Bhavsar and E. D.
Brown, unpublished observations). This suggests that hydro-
FIG. 4. Localization analyses of TagBPhoA insertional variants. (A) Strains EB1653 (tagB?1phoA; ?1), EB1654 (tagB?2phoA; ?2), EB1655
(tagB?3phoA; ?3), and EB1656 (tagB?4phoA; ?4) were grown under inducing conditions, harvested, and lysed. Whole-cell lysate (Lys.) was
fractionated into soluble lysate (Sol.) and membrane samples (Mem.) by differential ultracentrifugation. Samples were subjected to immunode-
tection with anti-PhoA antiserum. Samples were also probed with anti-TagD and anti-EzrA antisera to assess the quality of the fractionation. The
asterisk denotes a cross-reactive band. (B) Samples from strains EB1696 (tagBN30?1phoA; ?1), EB1697 (tagBN30?2phoA; ?2), EB1698
(tagBN30?3phoA; ?3), and EB1699 (tagBN30?4phoA; ?4) were fractionated as outlined above and subjected to the same immunodetection
VOL. 189, 2007 TagB MEMBRANE TARGETING DETERMINANT CHARACTERIZATION6821
phobicity plays an important role in membrane localization,
since high ionic strength augments hydrophobic interactions.
Our experimental results are the first to suggest that the
amino terminus of TagB is folded into an amphipathic helix.
Disruption of TagB function by the insertion of two and three
amino acids at position 10, and the subsequent restoration of
function upon insertion of four amino acids at the same posi-
tion, supports this assertion given that ?-helices contain 3.6
residues per turn. This further implies that the amino terminus
of TagB contains not only membrane targeting information but
also structural content that is crucial for TagB function. In-
triguingly, we note that to date there is a complete lack of
structural information for the TagF-like enzymes. Thus, while
our data support the assertion that the amphipathicity of the
TagB amino-terminal helix plays an important role in TagB
function, we cannot rule out the possibility that the defect in
TagB function is due to specific amino acid insertion.
Although it is not yet understood how the amino terminus of
TagB directs membrane targeting, e.g., via direct or indirect
interactions, it is clear that this can be done in a manner that
is independent of TagB activity, as exemplified by the identi-
fication of insertional variants that uncouple TagB localization
and function. One plausible mechanism to reconcile these ob-
servations is that TagB might independently localize to the
membrane but then be recruited into a multiprotein complex
whose formation is requisite for glycerophosphotransfer to an
undecaprenol-pyrophosphoryl-disaccharide moiety. It is inter-
esting to note that the prospect of a teichoic acid synthesome
has been previously suggested (21), although there is not yet
any evidence to prove or disprove the hypothesis.
The concept of a multifunctional peptide sequence is not
unique to TagB. Gram-negative lipoproteins contain an amino-
terminal signal sequence that directs the nascent protein for
secretion through the sec system, directs cleavage and acylation
of the translocated protein, and directs final localization to
either the outer leaflet of the inner membrane or the inner
leaflet of the outer membrane (15). Mutations in this lipopro-
tein signal sequence impact on the proper localization of the
lipoprotein, and this may lead to functional impairment of the
protein. Similarly, it was recently demonstrated that the signal
sequences of Streptococcus pyogenes PrtF and M6 proteins
direct not only their secretion but also the spatial restriction of
secretion (8). However, the bifunctional amino-terminal pep-
tide sequence of TagB differs in that the information contained
therein directs both localization and biochemical activity, with
the latter separable from the former. This scenario is reminis-
cent of the 26-amino-acid B. subtilis sporulation protein
SpoVM. This small membrane-binding protein is predicted to
adopt an amphipathic ?-helical fold, and most intriguingly,
site-directed mutagenesis identified residues that disrupted
SpoVM function without affecting its membrane targeting
(35). Subsequent study localized these residues to the mem-
brane interaction face of the amphipathic helix, the face op-
posite which a binding site for the sporulation protein SpoIVA
was identified (27). Although the hydrophilic face of the
SpoVM ?-helix consists of six charged amino acids, in contrast
to the polar but uncharged residues in the TagB amino termi-
nus, the SpoVM studies suggest that an amphipathic helix can
mediate both membrane binding and protein-protein interac-
tion, which is an attractive model for the role of the amino
terminus of TagB.
This work was funded by an operating grant (MOP-15496) and a
Canada Research Chair in Chemical Biology from the Canadian In-
stitutes of Health Research to E.D.B.
We acknowledge the generosity of the following: Petra Levin, who
provided pPL51 and anti-EzrA antibody; David Andrews, who pro-
vided anti-FtsY antibody; and members of the laboratory, for helpful
1. Bhavsar, A. P., T. J. Beveridge, and E. D. Brown. 2001. Precise deletion of
tagD and controlled depletion of its product, glycerol 3-phosphate cytidylyl-
transferase, leads to irregular morphology and lysis of Bacillus subtilis grown
at physiological temperature. J. Bacteriol. 183:6688–6693.
2. Bhavsar, A. P., L. K. Erdman, J. W. Schertzer, and E. D. Brown. 2004.
Teichoic acid is an essential polymer in Bacillus subtilis that is functionally
distinct from teichuronic acid. J. Bacteriol. 186:7865–7873.
3. Bhavsar, A. P., R. Truant, and E. D. Brown. 2005. The TagB protein in
Bacillus subtilis 168 is an intracellular peripheral membrane protein that can
incorporate glycerol phosphate onto a membrane-bound acceptor in vitro.
J. Biol. Chem. 280:36691–36700.
4. Bhavsar, A. P., X. Zhao, and E. D. Brown. 2001. Development and charac-
terization of a xylose-dependent system for expression of cloned genes in
Bacillus subtilis: conditional complementation of a teichoic acid mutant.
Appl. Environ. Microbiol. 67:403–410.
5. Boylan, R. J., and N. H. Mendelson. 1969. Initial characterization of a
temperature-sensitive Rod?mutant of Bacillus subtilis. J. Bacteriol. 100:
6. Briehl, M., H. M. Pooley, and D. Karamata. 1989. Mutants of Bacillus subtilis
168 thermosensitive for growth and wall teichoic acid biosynthesis. J. Gen.
7. Bruckner, R. 1992. A series of shuttle vectors for Bacillus subtilis and Esch-
erichia coli. Gene 122:187–192.
8. Carlsson, F., M. Stalhammar-Carlemalm, K. Flardh, C. Sandin, E.
Carlemalm, and G. Lindahl. 2006. Signal sequence directs localized se-
cretion of bacterial surface proteins. Nature 442:943–946.
9. Casadaban, M. J., and S. N. Cohen. 1980. Analysis of gene control signals by
DNA fusion and cloning in Escherichia coli. J. Mol. Biol. 138:179–207.
10. Cutting, S. M., and P. B. Vander Horn. 1990. Genetic analysis, p. 175–209.
In C. R. Harwood and S. M. Cutting (ed.), Molecular biological methods for
Bacillus. John Wiley & Sons, New York, NY.
11. Cutting, S. M., and P. Youngman. 1994. Gene transfer in Gram-positive
bacteria, p. 348–364. In R. G. E. Murray, N. R. Krieg, W. A. Wood, and P.
Gerhardt (ed.), Methods for general and molecular bacteriology. American
Society for Microbiology, Washington, DC.
12. D’Elia, M. A., K. E. Millar, T. J. Beveridge, and E. D. Brown. 2006. Wall
teichoic acid polymers are dispensable for cell viability in Bacillus subtilis. J.
13. D’Elia, M. A., M. P. Pereira, Y. S. Chung, W. Zhao, A. Chau, T. J. Kenney,
M. C. Sulavik, T. A. Black, and E. D. Brown. 2006. Lesions in teichoic acid
biosynthesis in Staphylococcus aureus lead to a lethal gain of function in the
otherwise dispensable pathway. J. Bacteriol. 188:4183–4189.
14. Ginsberg, C., Y. Zhang, Y. Yuan, and S. Walker. 2006. In vitro reconstitution
of two essential steps in wall teichoic acid biosynthesis. ACS Chem. Biol.
15. Hayashi, S., and H. C. Wu. 1990. Lipoproteins in bacteria. J. Bioenerg.
16. Hu, Z., and J. Lutkenhaus. 2003. A conserved sequence at the C-terminus of
MinD is required for binding to the membrane and targeting MinC to the
septum. Mol. Microbiol. 47:345–355.
17. Johnson, J. E., and R. B. Cornell. 1999. Amphitropic proteins: regulation by
reversible membrane interactions. Mol. Membr. Biol. 16:217–235.
18. Lazarevic, V., F. X. Abellan, S. B. Moller, D. Karamata, and C. Mauel. 2002.
Comparison of ribitol and glycerol teichoic acid genes in Bacillus subtilis W23
and 168: identical function, similar divergent organization, but different
regulation. Microbiology 148:815–824.
19. Lazarevic, V., and D. Karamata. 1995. The tagGH operon of Bacillus subtilis
168 encodes a two-component ABC transporter involved in the metabolism
of two wall teichoic acids. Mol. Microbiol. 16:345–355.
20. Levin, P. A., I. G. Kurtser, and A. D. Grossman. 1999. Identification and
characterization of a negative regulator of FtsZ ring formation in Bacillus
subtilis. Proc. Natl. Acad. Sci. USA 96:9642–9647.
21. Mauel, C., M. Young, and D. Karamata. 1991. Genes concerned with syn-
thesis of poly(glycerol phosphate), the essential teichoic acid in Bacillus
subtilis strain 168, are organized in two divergent transcription units. J. Gen.
6822BHAVSAR ET AL. J. BACTERIOL.
22. Mauel, C., M. Young, A. Monsutti-Grecescu, S. A. Marriott, and D. Karamata. Download full-text
1994. Analysis of Bacillus subtilis tag gene expression using transcriptional fu-
sions. Microbiology 140:2279–2288.
23. Park, Y. S., T. D. Sweitzer, J. E. Dixon, and C. Kent. 1993. Expression,
purification, and characterization of CTP:glycerol-3-phosphate cytidylyl-
transferase from Bacillus subtilis. J. Biol. Chem. 268:16648–16654.
24. Pereira, M. P., and E. D. Brown. 2004. Bifunctional catalysis by CDP-ribitol
synthase: convergent recruitment of reductase and cytidylyltransferase activ-
ities in Haemophilus influenzae and Staphylococcus aureus. Biochemistry 43:
25. Pichoff, S., and J. Lutkenhaus. 2005. Tethering the Z ring to the membrane
through a conserved membrane targeting sequence in FtsA. Mol. Microbiol.
26. Pooley, H. M., and D. Karamata. 1994. Teichoic acid synthesis in Bacillus
subtilis: genetic organization and biological roles, p. 187–198. In J. M.
Ghuysen and R. Hakenbeck (ed.), Bacterial cell wall. Elsevier, Amsterdam,
27. Ramamurthi, K. S., K. R. Clapham, and R. Losick. 2006. Peptide anchoring
spore coat assembly to the outer forespore membrane in Bacillus subtilis.
Mol. Microbiol. 62:1547–1557.
28. Rogers, H. J., M. McConnell, and I. D. Burdett. 1970. The isolation and
characterization of mutants of Bacillus subtilis and Bacillus licheniformis with
disturbed morphology and cell division. J. Gen. Microbiol. 61:155–171.
29. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a
laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold
Spring Harbor, NY.
30. Schertzer, J. W., A. P. Bhavsar, and E. D. Brown. 2005. Two conserved
histidine residues are critical to the function of the TagF-like family of
enzymes. J. Biol. Chem. 280:36683–36690.
31. Schertzer, J. W., and E. D. Brown. 2003. Purified, recombinant TagF protein
from Bacillus subtilis 168 catalyzes the polymerization of glycerol phosphate
onto a membrane acceptor in vitro. J. Biol. Chem. 278:18002–18007.
32. Soldo, B., V. Lazarevic, and D. Karamata. 2002. tagO is involved in the
synthesis of all anionic cell-wall polymers in Bacillus subtilis 168. Microbiol-
33. Soldo, B., V. Lazarevic, H. M. Pooley, and D. Karamata. 2002. Character-
ization of a Bacillus subtilis thermosensitive teichoic acid-deficient mutant:
gene mnaA (yvyH) encodes the UDP-N-acetylglucosamine 2-epimerase. J.
34. Szeto, T. H., S. L. Rowland, L. I. Rothfield, and G. F. King. 2002. Membrane
localization of MinD is mediated by a C-terminal motif that is conserved
across eubacteria, archaea, and chloroplasts. Proc. Natl. Acad. Sci. USA
35. van Ooij, C., and R. Losick. 2003. Subcellular localization of a small sporu-
lation protein in Bacillus subtilis. J. Bacteriol. 185:1391–1398.
36. Ward, J. B. 1981. Teichoic and teichuronic acids: biosynthesis, assembly, and
location. Microbiol. Rev. 45:211–243.
37. Weidenmaier, C., J. F. Kokai-Kun, S. A. Kristian, T. Chanturiya, H.
Kalbacher, M. Gross, G. Nicholson, B. Neumeister, J. J. Mond, and A.
Peschel. 2004. Role of teichoic acids in Staphylococcus aureus nasal col-
onization, a major risk factor in nosocomial infections. Nat. Med. 10:243–
38. Zhou, H., and J. Lutkenhaus. 2003. Membrane binding by MinD involves
insertion of hydrophobic residues within the C-terminal amphipathic helix
into the bilayer. J. Bacteriol. 185:4326–4335.
VOL. 189, 2007TagB MEMBRANE TARGETING DETERMINANT CHARACTERIZATION 6823