Full text of DOI:10.1111/j.1348-0421.2002.tb02670.x

Microbiol. Immunol., 46(1), 1―10, 2002
FimZ Binds the Salmonella typhimurium fimA
Promoter Region and May Regulate Its Own
Expression with FimY
Kuang-Sheng Yeh*, 1, Juliette K. Tinker2, and Steven Clegg2
2
1Department of Pathobiology, Pig Research Institute Taiwan, P.O. Box 23, Chunan, Miaoli 350, Taiwan, and Department of
Microbiology, College of Medicine, The University of Iowa, 51 Newton Road, Iowa City, IA 52242, U.S.A.
Received June 18, 2001; in revised form, October 10, 2001. Accepted October 15, 2001
Abstract: The FimZ protein, an activator of FimA production in Salmonella typhimurium, acts in conjunction
with FimY to facilitate the expression of type 1 fimbriae. The predicted amino acid sequence of FimZ sug-
gests that this protein may be a DNA-binding protein related to BvgA, a sensory regulator of virulence gene
expression in Bordetella pertussis. Purification of FimZ following overexpression of the protein by a
strong inducible promoter and gel mobility shift assays confirm that FimZ is a 25-kDa polypeptide that binds
to the promoter region of fimA. The region of DNA protected from DNase I digestion by FimZ binding is
located between 47 and 98 nucleotides upstream from the fimA transcription initiation site. This region pos-
sesses a pair of 7-base pair tandem repeats, of which at least one is necessary for FimZ binding. One copy
of the 7-base pair sequence is also located in the fimZ promoter region. In addition, expression from a fimZ-
lacZ reporter construct confirms that FimZ plays a role in its own expression. Both FimZ and FimY are
required for high-level expression of FimZ, which suggests that these two fimbrial proteins are involved in
regulating both FimA and FimZ.
Key words: Salmonella typhimurium, Type 1 fimbriae, Activator
The Salmonella typhimurium fim gene cluster encodes
the expression of the mannose-sensitive type 1 fimbriae
that mediate adherence to a range of eucaryotic cells
and tissues (3, 9, 11, 19). The complete nucleotide
sequence of the gene cluster has been determined and
comprises a number of genes that are responsible for the
ordered assembly and binding specificity of the fimbrial
appendages on the surface of the bacteria (9). Most of
these genes code for proteins that are believed to exhib-
it functions related to those reported in Escherichia coli
fimbrial and pilus systems (12, 17, 18, 21). However, the
regulation of the S. typhimurium major fimbrial subunit
gene (fimA) is controlled by a mechanism different to that
reported for the E. coli fimA gene. Genes homologous to
the E. coli fimB and fimE determinants that control the
inversion of the fimA promoter region are not found
within the S. typhimurium fim gene cluster (6, 7, 13,
24). Also, the expression of an S. typhimurium fimA-lacZ
reporter using E. coli as a host is not observed even
under conditions that result in increased expression of the
E. coli fimA gene. However, in the E. coli host, the S.
typhimurium fimA gene is expressed in the presence of
two Salmonella proteins, FimZ and FimY (33, 40).
The predicted amino acid sequence of FimZ relates to
a family of transcriptional activators that are sensory
regulators of two-component systems (10, 32). However,
examining the fim gene cluster does not indicate the
presence of a gene product that demonstrates homology
to sensory kinases of these systems. Consistent with the
role of FimZ as an activator of S. typhimurium fimA
expression, fimZ mutants are non-fimbriate and exhibit
only low levels of fimA expression (40). Restoration
of FimA production and the fimbriate phenotype can
be achieved by introducing plasmids that carry only the
fimZ gene of the gene cluster into a fimZ mutant (40).
These transformants are fully fimbriate, indicating that
they demonstrate a high level of fimA gene expression.
However, they lack the ability to demonstrate fimbrial
phase variation, the switching phenomenon of individual
cells between a state of expression (fimbriate) and non-
*Address correspondence to Dr. Kuang-Sheng Yeh, Department
of Pathobiology, Pig Research Institute Taiwan, P.O. Box 23,
Abbreviations: BSA, bovine serum albumin; LB, Luria-Bertani;
IPTG, isopropyl-β-D-thiogalactoside; PCR, polymerase chain
reaction; SDS-PAGE, sodium dodecyl sulfate polyacrylamide
gel electrophoresis.
Chunan, Miaoli 350, Taiwan. Fax: 886
ca@mail.prit.org.tw
―37―692820. E-mail:
1

2
K.-S. YEH ET AL
expression (non-fimbriate), or vice versa, to produce
populations containing a mixture of both cell types (13).
This constitutive activity is most likely due to the pres-
ence of relatively large amounts of FimZ in the trans-
formants since fimZ is carried on multi-copy recombinant
plasmids in these strains.
Since FimZ plays a central role in fimA expression by
S. typhimurium, we have identified the specific FimZ
DNA-binding region within fimA promoter. Also, since
FimZ regulates S. typhimurium fimA expression and is
related to a family of sensory regulators, we constructed
a fimZ-lacZ fusion to investigate the expression of this
important regulator of the fimbrial system.
AACAGTCTGAGG-3' and 5'-GCTGAAGGATCC-
CTCCCGAACGATAATTCGCC-3' containing appro-
priate restriction enzyme sites (underlined, EcoRI and
BamHI respectively), were used to generate a 692-bp
PCR product that was cloned into the expression vector
pT7-7 containing a ϕ10 T7 RNA polymerase promoter
(36). High-level production of FimZ by transformants
was achieved following induction by isopropyl-β-D-
thiogalactoside (IPTG) as previously described (36).
The plasmid vector pMC1403 (8), containing a pro-
moterless lacZ gene, was used to construct the fimZ-
lacZ fusion (pISF239) molecule. The fimZ transcription
initiation site has previously been characterized by
primer extension and was located 227 nucleotides
upstream of the fimZ initiation codon (unpublished data).
A 700-bp DNA fragment carrying the fimZ promoter
region was cloned into the EcoRI site of pMC1403.
The promoter-containing region of fimZ was prepared by
PCR using pISF101 as a template and 5'-GATCTGTG-
GAATTCCGAAGTAACGTTTTGGTG-3' and 5'-CAT-
GTGGTGAATTCCAGGATAAGTGCGCAGAT-3' as
primers (EcoRI sites are underlined). Transformants
possessing a recombinant molecule comprising an in-
frame fusion between the fimZ and lacZ genes were
used to prepare plasmid DNA, and the fidelity of the
Materials and Methods
Bacterial strains, plasmids and media. Table 1 lists
the bacterial strains and plasmids used in this study.
Basically, all the DNA manipulations were performed by
conventional procedures (29). The FimZ expression
plasmid, pKYZ100, was constructed as follows. The
entire fimZ gene was cloned on an EcoRI-BamHI DNA
fragment by conventional polymerase chain reaction
(PCR) techniques using pISF101 as a template. The
primers 5'-GCTAACGGAATTCGGACGCATAACAT-
Table 1. Strains and plasmids used in this study
Strain or plasmid
Genotype or relevant features
Reference or
source
Salmonella typhimurium
LB5010
Wild type; fimbriate with complete fim gene cluster
LB5010 fimZ: : kan Kanr
(34)
(40)
LBZ100
Escherichia coli
BL21 (DE3)
DH5 α
－
－
－
F dcm ompT hsdS (rB mB ) gal λ (DE3)
(36)
(15)
－
＋
deoR endA1 gyrA96 hsdR17 (rk mk ) recA1 relA1 supE44 thi-1 ∆
(lacZYA-argFV169) ϕ80lacZ∆M15 F
－
Plasmids
PT7-7
Cloning vector that contains T7 promoter; Amr
Promoterless lacZ fusion system
(36)
(8)
This study
(9)
pMC 1403
pKYZ
pISF101
629-bp DNA fragment containing fimZ cloned into pT7-7; Amr
12.8-kbp DNA fragment containing fim gene cluster of S. typhimurium cloned
into pACYC184; Cmr Tcr
＋
＋
pISF182
pISF187
pISF189
pISF239
fimZ and fimY cloned into pACYC 184; FimZ FimY ; Cmr
(40)
(40)
(40)
This study
Stratagene
This study
pISF182 with translational terminator inserted into fimY; FimZ FimY ; Cmr
＋
－
pISF182 with translational terminator inserted into fimZ; FimZ FimY ; Cmr
－
＋
fimZ-lacZ reporter fusion
pBluescript II KS (－) Cloning vector; Amr
pKYA452
pKYA390
pKYA241
pKYA211
452-bp DNA fragment containing fimA promoter region cloned into pBluescript
II KS (－); Amr
pKYA452 with the AATAAGA sequence adjacent to 5' end of fimA
This study
This study
This study
promoter deleted; Amr
241-bp BstEII-EcoRV fragment from fimA promoter DNA cloned into
pBluescript II KS (－); Amr
211-bp EcoRV-EcoRI fragment from fimA promoter DNA cloned into
pBluescript II KS (－); Amr

3
FimZ BINDS SALMONELLA TYPHIMURIUM fimA PROMOTER REGION
fusion was determined by DNA sequencing through the
fusion junction. The pKYA452 DNA fragment pos-
sessing the fimA promoter region was prepared from
pISF141 restricted with BstEII and RsaI, and purified
with an agarose gel (14). The ends of the DNA fragment
were made flush by Klenow polymerase (New England
Biolabs, Beverly, Mass., U.S.A.) and this fragment was
subcloned into the EcoRV site of pBluescript II KS (－)
(Stratagene, La Jolla, Calif., U.S.A.). Deletions of the
fimA promoter region were derived from pKYA452 by
PCR or by enzyme restriction. All bacterial strains
were cultured on Luria-Bertani (LB) medium and incu-
bated at 37 C, or 30 C for lysogens, for 24 or 48 hr.
Preparation of FimZ. FimZ was purified from trans-
formants of E. coli BL21 (pKYZ100) following growth
in 4 liters of LB broth at 30 C to an OD600 of 1.0. IPTG
was added to a final concentration of 1 mM, and cultures
were incubated for an additional 16 hr. The bacteria were
collected by centrifugation and resuspended in 50 ml of
ice-cold Buffer A (20 mM Tris pH 8.0, 1 mM EDTA, 7
mM β-mercaptoethanol, 10% glycerol). The protease
inhibitor phenylmethylsulfonyl fluoride (100 µg/ml) was
added prior to disruption in a French pressure cell at
18,000 psi (1 psi＝8.89 kPa). Following lysis, the sol-
uble cell extract and cell debris were separated by cen-
trifugation at 11,000×g for 30 min at 4 C. Ammonium
sulfate was added to the supernatant to achieve 30%
saturation, and the protein was then collected by cen-
trifugation and redissolved in 30 ml of Buffer A. After
dialysis against Buffer A, the protein extract was applied
to a previously equilibrated Q-Sepharose FF column
(2.6×40 cm). Using an automated FPLC system (Amer-
sham Pharmacia Biotech, Piscataway, N.J., U.S.A.),
tions were carried out at 37 C for 15 min, mixed with 1
µl of loading dye (50% glycerol, 0.1% Xylene Cyanol),
and subsequently applied to a non-denaturing 5% poly-
acrylamide-3% glycine gel. Electrophoresis was per-
formed at 200 volts for 2 hr, after which the gel was dried
and subjected to autoradiography.
DNase I footprint assays. DNase I (Roche Diagnos-
tics GmbH, Mannheim, Germany) was dissolved in 1×
low salt dilution buffer (10 mM Tris-HCl pH 7.5, 10
mM MgCl2, 100 µg/ml BSA) to a final concentration of 5
mg/ml. After the binding reaction, a DNase I (2.5 µg per
assay) digest was performed at 37 C for 30 sec. The
digestion was stopped by adding 100 µl of termination
buffer (1% Sarkosyl, 0.1 M NaCl, 0.1 M Tris-HCl pH 8.0,
10 mM EDTA, 100 µg/ml proteinase K, 25 µg/ml calf
thymus DNA) and the mixture was incubated at 37 C for
15 min. The DNA samples were extracted with phenol:
chloroform (1:1), ethanol-precipitated, and finally resus-
pended in 7 µl of loading buffer (95% formamide, 20 mM
EDTA, 0.05% Bromophenol Blue). The samples were
analyzed by electrophoresis through a 6% polyacryl-
amide/8 M urea sequencing gel, and compared to a
Maxam-Gilbert sequence reaction performed on the
same DNA fragments (23, 30, 38).
Results
Purification of FimZ
Cell extracts from IPTG-induced cultures of E. coli
(pYZ100) were observed by SDS-PAGE to possess large
amounts of a 25-kDa polypeptide that was absent from
un-induced cultures (Fig. 1, Panel A). The size of this
polypeptide was consistent with that (24.8 kDa) pre-
dicted from the nucleotide sequence of fimZ. This
polypeptide was further purified using ion-exchange
chromatography. SDS-PAGE analysis of the column
eluates indicated that the 25-kDa protein was eluted at
FimZ was eluted against a linear gradient of 0
―
1 M
NaCl in Buffer A. Column fractions were subsequently
examined by sodium dodecyl sulfate polyacrylamide
gel electrophoresis (SDS-PAGE) as previously described
(22). The Protein Structure Facility, University of Iowa
determined the N-terminal amino acid sequence of the
purified FimZ.
Gel mobility shift assays. The gel mobility shift assay
was based on the methods previously published (40). For
5'-labeling, DNA fragments were treated with calf intesti-
nal phosphatase (New England Biolabs), and labeled
with [γ-32P]-ATP (Amersham Pharmacia Biotech) using
T4 polynucleotide kinase (Promega, Madison, Wisc.,
U.S.A.). All labeled DNA fragments were purified with
a 5% polyacrylamide gel. The labeled DNA was prein-
cubated in 1×DNA binding buffer (5 mM Tris-HCl pH
7.5, 25 mM KCl, 0.5 mM EDTA, 2.5% glycerol, 0.5 mM
dithiothreitol) (18 µl total volumes) at 37 C for 5 min,
after which 2 µl volumes of protein samples diluted in
2×DNA binding buffer were added. The binding reac-
approximately 400
―
500 mM NaCl in a linear gradient of
0―
1 M (Fig. 1, Panel B). The purified protein was sub-
jected to amino acid sequencing from the N-terminus,
and analysis of this sequence confirmed that the first
10 amino acids of the 25-kDa polypeptide were identical
to those predicted from the nucleotide sequence of the
fimZ gene.
Gel Mobility Shift Assays
Gel mobility shift assays were performed to determine
whether FimZ was able to bind to the target DNA frag-
ments and cause a mobility shift. A representative gel is
shown in Fig. 2. The purified FimZ and FimZ-contain-
ing cell extracts exhibited identical ability to cause a
shift of the DNA fragment possessing the fimA promot-
er (Fig. 2, Lanes 3, 4). Bacterial extracts derived from E.

4
K.-S. YEH ET AL
Fig. 1. SDS-polyacrylamide gel analysis of FimZ. Panel A, Lane
1, molecular weight standards; Lanes 2 and 3, bacterial lysates
from un-induced (Lane 2) and IPTG-induced (Lane 3) cultures of
E. coli BL21 (pKYZ100). Panel B, Lanes 1 and 2, French pres-
sure lysates (Lane 1) and ammonium sulfate precipitates (Lane 2)
from E. coli BL21 transformed with the cloning vector pT7-7.
Lane 3, ammonium sulfate (30% saturation) precipitate from
lysates of E. coli BL21 (pKYZ100). Lanes 4―6, 20 µl fractions
eluted from a Q-sepharose column at 450 mM NaCl (Lane 4), 200
mM (Lane 5) and 800 mM NaCl (Lane 6). Lane 7, molecular
weight standards. The arrow in Panel B represents the size of the
FimZ polypeptide and the source of protein used to determine the
N-terminal amino acid sequence.
Fig. 3. DNase I footprint analysis of the fimA promoter region by
FimZ. Lanes 1 and 2, the A＋G and C＋T chemical sequence
analysis of the promoter region. Lane 3, DNA fragments incu-
bated with no protein. Lanes 4―7, DNA incubated with 500,
250, 125, and 62.5 ng, respectively, of protein prepared from bac-
terial extracts of E. coli transformed with the expression vector
pT7-7. Lanes 8―11, DNA incubated with 500, 250, 125, and
62.5 ng, respectively, of protein from cell extract prepared from
E. coli (pKYZ100) containing FimZ. The region of DNA pro-
tected from DNase I digestion is indicated by the brackets. The
numbers indicate the location of the protected region relative to the
transcription initiation site of fimA (Ti).
Fig. 2. Gel mobility shift assay. The 452-bp fimA promoter DNA
was used in Lanes 1
―
4; the 390-bp derivative DNA is in Lanes 5
―
8; a 746-bp metF promoter DNA is in Lanes 9
―12; the 241-bp
derivative DNA is in Lanes 13, 14; the 211-bp derivative DNA is
in Lanes 15, 16. Lanes 1, 5, and 9, DNA incubated with no pro-
tein; Lanes 2, 6, 10, 13, and 15, DNA incubated with 5 µg of cell
extract possessing no FimZ; Lanes 3, 7, and 11, DNA incubated
with 5 µg of FimZ-containing cell extract; Lanes 4, 8, 12, 14, and
16, DNA incubated with 1 µg of purified FimZ. The arrow indi-
cates the DNA-protein complex.
DNA binding activity for the fimA promoter region (Fig.
2, Lanes 2, 6, 10). FimZ did not affect the mobility of a
S. typhimurium-derived DNA fragment possessing the
non-fim promoter metF that is controlled by the het-
erologous Salmonella DNA binding protein (Fig. 2,
coli BL21 (pT7-7) carrying the cloning vector alone,
and used at protein concentrations equivalent to those
employed for FimZ binding, did not demonstrate any

5
FimZ BINDS SALMONELLA TYPHIMURIUM fimA PROMOTER REGION
Fig. 4. Sequence analysis of fimA and fimZ promoter regions. Panel A, the underlined region shows the FimZ-protected area within fimA
promoter. A pair of direct repeat AATAAGA are in boldface. Transcription initiation site of fimA is previously determined by nuclease
S1 protection assay (9) to be located 195-bp upstream from the fimA initiation codon and denoted by Ti. The nucleotide sequences of
N-terminal parts of tetrahydrofolate dehydrogenase (folD) and fimA gene products are indicated by bent arrows. Panel B, transcription
initiation site of fimZ is determined by primer extension to be located 227-bp upstream from the fimZ initiation codon (unpublished data)
and is denoted by Ti. AATAAGA sequence is shown in boldface. The nucleotide sequences of N-terminal part of fimZ product and C-
terminal part of fimY product are indicated by bent arrows. Nucleotide marked with an asterisk is the stop codon of fimY.

6
K.-S. YEH ET AL
Lanes 11, 12). However, not only was a shift in the
electrophoretic mobility of the DNA fragment possess-
ing the fimA promoter observed, but also the use of
FimZ resulted in aggregation of this DNA region such
that some labeled DNA did not migrate into the gel and
remained within the wells of the gel (Fig. 2, top portions
of Lanes 3, 4, 11, and 12).
tected region. In plasmid pKY390, 12 nucleotides of the
52-bp FimZ-protected region, including one of the 7-bp
direct repeat sequences, were deleted. As indicated in
Fig. 2, Lane 14, pKY241 was also retarded in gel mobil-
ity shift assays following incubation with FimZ. The
DNA fragments from pKY390 and pKY211 that lacked
the intact 52-bp FimZ-protected region indicated by
footprint analysis were not retarded in gel shift assays
(Fig. 2, Lanes 7, 8, and 16). Aggregation of labeled
DNA fragments that remained within the wells of the gel
were also observed as described above when using FimZ
protein (Fig. 2, top portions of Lanes 7, 8, 14, and 16).
Figure 5 summarizes the results of gel mobility shift
assays.
Location of FimZ-Binding Region on the fimA Promoter
A DNase I footprint assay was performed to determine
the specific region of the fimA promoter to which FimZ
binds. As shown in Fig. 3, when a DNA fragment was
incubated with FimZ-containing extracts from E. coli
BL21 (pKYZ100), a 52-bp region of the fragment was
protected from DNase I digestion. This region was not
protected when cell extracts from E. coli BL21 (pT7-7)
were used. Figure 4 shows the nucleotide sequences of
the protected region, and this DNA region extends from
47-bp to 98-bp upstream from the transcription initiation
site previously reported for fimA (33). A pair of 7-bp
direct repeats (5'-AATAAGA-3') were found within the
protected region (Fig. 4, Panel A).
Since a 452-bp DNA fragment carrying the fimA pro-
moter region was employed in both the gel mobility
shift and DNase I footprinting assays, deletions within
this fragment, carried on plasmid pKY452, were con-
structed. Plasmid pKY241 possessed Salmonella-derived
DNA in which the 52-bp-protected region was present,
but deletions downstream from this region were made.
Plasmid pKY211 comprised DNA lacking the 52-bp-pro-
Expression of β-Galactosidase from the fimZ-lacZ
Reporter in E. coli and S. typhimurium
Examination of the nucleotide sequence of the fimZ
promoter region indicated one copy of the sequence,
5'-AATAAGA-3' was identical to the directly repeated
sequence found in the FimZ binding region of the fimA
promoter (Fig. 4, Panel B). Accordingly, the ability of
FimZ to regulate its own expression was analyzed. The
expression of FimZ was investigated in vivo using the
fimZ-lacZ fusion carried on plasmid pISF239. In an E.
coli host previously shown not to express the S.
typhimurium fimA subunit gene in the absence of FimZ
and FimY (33, 40), a relatively low level of FimZ expres-
sion was detected (Table 2) even when the fusion was
carried on a medium copy-number vector. As indicated
Fig. 5. Summary of the gel mobility shift assays. The ability of FimZ to bind to the fimA promoter DNA and its derivatives is indicat-
ed as positive, while negative indicates that FimZ does not cause a mobility shift of these DNA fragments. The solid black boxes denote
the FimZ protection region determined by the DNase I footprint assay. The sequences represented by the solid areas are as indicated. A
pair of direct repeat AATAAGA are in boldface. Transcription initiation site of fimA is designated by arrow.

7
FimZ BINDS SALMONELLA TYPHIMURIUM fimA PROMOTER REGION
Table 2. Expression of β-galactosidase by fimZ-lacZ reporter constructs in E. coli and S.
typhimurium
Strain
Plasmid (relevant genotypes)
β-Galactosidase expressiona)
E. coli
DH5 α
DH5 α
DH5 α
DH5 α
pISF239 (fimZ-lacZ)
24
3468
36
＋
＋
＋pISF182 (fimZ fimY )b)
＋
－
＋pISF187 (fimZ fimY )
－
＋
＋pISF189 (fimZ fimY )
65
S. typhimurium
LB5010
pISF239 (fimZ-lacZ)
pISF239 (fimZ-lacZ)
603
196
1088
755
LBZ100
LBZ100
LBZ100
LBZ100
＋
＋
＋pISF182 (fimZ fimY )
＋
－
＋pISF187 (fimZ fimY )
－ ＋
＋pISF189 (fimZ fimY )
771
a) β-Galactosidase activity is reported in Miller (25) units.
b) pISF239 ＋ pISF182.
in Table 2, E. coli (pISF239) transformed with pISF187
or pISF189 merely exhibited small increases in FimZ
expression even though these plasmids encode func-
tional FimZ (pISF187) or FimY (pISF189) polypep-
tides (33, 40). However, more than a 100-fold increase
in FimZ expression was detected in the presence of
pISF182 that produced both FimZ and FimY.
The fimZ mutant S. typhimurium LBZ100 did not
produce type 1 fimbriae on its surface and demonstrated
low levels of fimA expression (40). Transformants of this
strain possessing pISF239 exhibited detectable levels
of β-galactosidase activity (Table 2). However, fimZ-lacZ
transformants of the parental strain S. typhimurium
LB5010, bearing an intact fimZ on the chromosome,
resulted in higher levels of FimZ expression than did the
fimZ mutant. Introduction of cloned copies of fimZ or
fimY into S. typhimurium LBZ100 resulted with FimZ
expression slightly higher than that produced by S.
typhimurium LB5010. The presence of many copies
of both fimZ and fimY was responsible for maximal
expression of FimZ (Table 2).
FimZ to the fimA promoter could be demonstrated in vitro
and the location of binding was precisely detected by
DNase I footprint assays.
The FimZ-protected region extends from approxi-
mately 47 to 98 base pairs upstream from the transcrip-
tion initiation site of fimA. This region shares charac-
teristics with the activator binding site for class I tran-
scription factors (20). Most activator-binding sites of this
class are a short distance upstream from the relevant
transcription initiation site, and protein-protein contact
between RNA polymerase and the activator has been
suggested to increase the rate of isomerization of the
polymerase-promoter complex from the closed to the
open state (20). Transcriptional factors belonging to
this class include Ada, OxyR, CRP, OmpR, TrpI, and
IHF (20). Removal of the 52-bp FimZ-binding region
from fimA promoter resulted in abrogation of FimZ
binding but deletion of nucleotides outside of the pro-
tected region did not affect this binding. A pair of 7-bp
direct repeats, predicted to be separated by two helical
turns, are present within the binding region. Removing
one of these repeats causes the FimZ binding to be lost,
suggesting that these sequences participate in FimZ
binding.
Discussion
The predicted amino acid sequence of the S.
typhimurium fimbrial regulator, FimZ, exhibits a high
degree of homology to a variety of bacterial transcrip-
tional activators (10, 32). The N-terminal domain of
FimZ is similar in sequence to a family of response reg-
ulators including BvgA. BvgA is a DNA-binding protein
involved in virulence gene expression in Bordetella per-
tussis (10, 31). Examination of the amino acid sequence
of the C-terminal domain of FimZ reveals a helix-turn-
helix motif and this, along with our previous reports of
FimZ affecting fimA expression (33, 40), indicates that
FimZ can bind to the fimA promoter region. Binding of
We have previously reported that both FimZ and
FimY are necessary for fimA expression in vivo (9, 33).
However, our in vitro results indicate that FimZ alone
will bind to the fimA promoter. Such binding leads to an
electrophoretic profile consistent with a DNA fragment
which exhibits very slow mobility with a large aggrega-
tion of FimZ-DNA complexes. This profile is dissimilar
to that which would be predicted by an increase in mo-
lecular size when one, two, or three protein molecules
bind to the DNA. The wells of the gel loaded with
FimZ/DNA mixtures frequently possessed labeled DNA
fragments that did not migrate into the gel. FimZ/DNA

8
K.-S. YEH ET AL
with FimZ-binding region mixtures had more protein-
DNA complex within the wells than those of FimZ/DNA
with no FimZ-binding region. It is possible that the
FimZ prepared in our study has some non-specific DNA-
binding activity. However, when FimZ binds in vitro to
the fimA promoter, many molecules of FimZ are associ-
ated with the DNA fragments, creating a large DNA-pro-
tein complex that has difficulty entering the gel. The
observed aggregation of FimZ on the fimA promoter in
vitro is not likely to occur where additional regulatory
proteins may play a role in fimA expression. In addition,
this aggregation is not seen when crude bacterial lysates
possessing FimZ are used in the gel-shift assays (40), and
is therefore a result of using over-expressed or purified
protein in these assays. The predicted amino acid
sequence of FimZ and its ability to bind to the fimA-pro-
moter region suggest that this protein may be a response
regulator; it is unknown whether FimZ needs to be phos-
phorylated in order to bind. It has been demonstrated that
response regulators can be phosphorylated in vitro by
small molecule phosphate donors such as phospho-
ramide and acetyl phosphate (16). No differences in
the binding activity of FimZ to the fimA promoter were
observed when acetyl phosphate was used as a donor
(data not shown).
chromosomal DNA exhibit almost the same level of
FimZ expression. There is no result demonstrating that
many copies of fimZ alone can activate FimZ expression;
that many copies of fimY alone can activate FimZ expres-
sion could be explained by gene dosage effect. Notably,
the in vivo expression of FimZ is highest in the presence
of both FimZ and FimY. It is possible that FimZ and
FimY, in correct stoichiometric ratios, are required for
optimal expression of FimZ. This result corresponds to
the observations previously reported for the expression of
fimA in vivo (33). It was also demonstrated that tran-
scription from the cloned fimY promoter was not detect-
ed in the E. coli host unless both FimY and FimZ were
present (37). Consequently, we propose that both FimZ
and FimY act cooperatively to regulate their target genes.
However, in contrast to the case of the previously report-
ed fimA expression (33, 40), a relatively low level of
FimZ expression occurs in the absence of the FimZ pro-
tein. The sequence of the FimZ polypeptide is 71%
identical to the predicted amino acid sequence of an E.
coli polypeptide, and can be produced by an open read-
ing frame located 155-bp upstream from the transcrip-
tional start of the argU gene (26). The presence of this S.
typhimurium FimZ counterpart in E. coli may explain the
FimZ expression level in the E. coli host. Our data
indicate that FimZ affects FimZ expression in vivo, and
that FimZ may, therefore, be autoregulatory. However,
this activation is maximally achieved in the presence
of FimY. The molecular interaction of both FimZ and
FimY proteins must be determined to understand the
role of these two gene products in Salmonella fimbrial
expression.
The precise role of the 7-bp repeat AATAAGA
sequence, which is 352-bp upstream from the fimZ tran-
scription initiation site, was not elucidated in this study.
This sequence is at a large distance from the fimZ-pro-
moter region. If this sequence is critical for fimZ acti-
vation, it could be an enhancer-like element as docu-
mented in other bacterial systems (39). Nonetheless, fur-
ther study is required to explore the function of this
AATAAGA sequence.
Although FimZ is closely related to the response reg-
ulator BvgA and possesses the highly conserved motifs
associated with this class of regulator, the fimZ gene is not
immediately adjacent to a determinant that is likely to
encode a sensory kinase associated with two-compo-
nent regulatory systems. Upstream from fimZ lies fimY
(9), and an examination of the amino acid sequence of
the FimY does not reveal the presence of highly con-
served amino acids found within sensory kinases. Also,
FimY is a relatively small polypeptide (27 kDa) with no
apparent membrane-spanning regions. Hence, if FimZ is
a component of a sensory regulatory system, it is not
Analysis of the nucleotide sequences upstream from
fimZ indicated the presence of a single 7-bp sequence that
is between 358 and 352 base pairs upstream from the
fimZ transcription initiation site and is identical to the
direct repeats found within the fimA promoter. Since we
could neither demonstrate the binding of FimZ to its
promoter DNA nor identify the precise FimZ site of
binding to its promoter region in vitro by gel mobility
shift or DNase I footprint assays (data not shown), we
decided to determine whether FimZ could regulate its
own expression by using a fimZ-lacZ reporter construct.
The observation that putative DNA-binding proteins can
be shown to control their target gene expression without
demonstrable binding in vitro has been reported for sev-
eral systems (1, 2, 5). In fact, by using a fimZ-lacZ
reporter construct that possesses an intact fimZ promot-
er region, we were able to demonstrate that FimZ expres-
sion is significantly reduced in a fimZ mutant, S.
typhimurium LBZ100. The Miller units seen with the
fimZ mutants introduced with cloned copies of fimZ or
fimY are higher than those seen with the fimZ mutant, but
are only slightly higher than those seen with the wild type
S. typhimurium LB5010. Many copies of fimY alone
restore the FimZ expression in the fimZ mutant back to
the wild type strain since there is no functional FimZ in
either chromosomal DNA or the introduced plasmid.
＋
However, in S. typhimurium LBZ100＋pISF187 (fimZ
－
fimY ), many copies of fimZ and one copy of fimY in

9
FimZ BINDS SALMONELLA TYPHIMURIUM fimA PROMOTER REGION
gene fusions that join an enzymatically active β-galactosidase
segment to amino-terminal fragments of exogenous pro-
teins: Escherichia coli plasmid vectors for the detection and
cloning of translational initiation signals. J. Bacteriol. 143:
closely linked to its cognate sensory kinase. Examples of
other proteins that are related to response regulators and
also are not contiguous with identifiable kinases have
been reported in Salmonella (4, 35). In addition, the
dependence of fimA activation on both FimZ and FimY
in vivo indicates that if FimZ is a sensory regulator,
FimY may act as a third component in such a system
along with an as-yet-unidentified sensory kinase. The
influence of cultural conditions on the expression of
type 1 fimbriae by S. typhimurium has been compre-
hensively described by Duguid and coworkers (11, 27,
28). Therefore, environmental factors probably play a
role in affecting Salmonella fimbrial expression, and
FimZ is a putative candidate in this process.
971
9) Clegg, S., and Swenson, D.L. 1994. Salmonella fimbriae,
p.105 114. In Klemm, P. (ed), Fimbriae: adhesion, genetics,
―980.
―
biogenesis, and vaccines, CRC Press, Boca Raton, Florida.
10) Cotter, P.A., and Miller, J.F. 1994. BvgAS-mediated signal
transduction: analysis of phase locked regulatory mutants of
Bordetella bronciseptica in a rabbit model. Infect. Immun.
62: 3381―3390.
11) Duguid, J.P., Anderson, E.S., and Campbell, I. 1966. Fim-
briae and adhesive properties in Salmonellae. J. Pathol. Bac-
teriol. 92: 107―138.
12) Eisenstein, B.I. 1988. Type 1 fimbriae of Escherichia coli:
genetic regulation, morphogenesis, and role in pathogenesis.
Rev. Infect. Dis. 10 (Suppl. 2): S341―S344.
13) Gally, D.L., Leathart, J., and Blomfield, I.C. 1996. Interaction
of FimB and FimE with the fim switch that controls the
phase variation of type 1 fimbriae in Escherichia coli K-12.
This work was supported by a grant from the National
Research Initiative of the USDA, U.S.A. (97-35204-4616) and in
part by a grant from the National Science Council, Taiwan (88-
2313-B-059-017). JKT was supported by a predoctoral fellowship
from a National Institutes of Health Parasitism Training Grant,
U.S.A. (TEA107511). The authors wish to thank Peggy Wang and
Ted Knoy for editorial suggestions for this article.
Mol. Microbiol. 21: 725―738.
14) Gerlach, G.F., Clegg, S., Ness, N.J., Swenson, D.L., Allen,
B.L., and Nichols, W.A. 1989. Expression of type 1 fimbri-
ae and mannose-sensitive hemagglutinin by recombinant
References
plasmids. Infect. Immun. 57: 764
15) Hanahan, D. 1983. Studies on transformation of Escherichia
coli with plasmids. J. Mol. Biol. 166: 557 580.
16) Hoch, J.A., and Silhavy, T.J. (eds), 1995. Two-component
signal transduction, ASM Press, Washington, D.C.
―770.
1) Ahmer, B.M., Reeuwijk, J.V., Timmers, C.D., Valentine,
P.J., and Heffron, F. 1998. Salmonella typhimurium encodes
an SdiA homolog, a putative quorum sensor of the LuxR
family, that regulates genes on the virulence plasmid. J.
―
17) Hultgren, S.J., and Normark, S. 1991. Biogenesis of the
Bacteriol. 180: 1185―1193.
bacterial pilus. Curr. Opin. Gen. Dev. 1: 313―318.
2) Bajaj, V., Hwang, C., and Lee, C.A. 1995. hilA is a novel
ompR/toxR family member that activates the expression of
Salmonella typhimurium invasion genes. Mol. Microbiol.
18) Hultgren, S.J., Normark, S., and Abraham, S.N. 1991. Chap-
erone-assisted assembly and molecular architecture of adhe-
sive pili. Annu. Rev. Microbiol. 45: 383―415.
18: 715―727.
19) Isaacson, R.E., and Kinsel, M. 1992. Adhesion of Salmonella
typhimurium to porcine intestinal epithelial surfaces: iden-
tification and characterization of two phenotypes. Infect.
3) Baumler, A.J., Tsolis, R.M., and Heffron, F. 1997. Fimbrial
adhesions of Salmonella typhimurium. Role in bacterial
interactions with epithelial cells. Adv. Exp. Med. Biol. 412:
Immun. 60: 3193
20) Ishihama, A. 1993. Protein-protein communication within the
transcription apparatus. J. Bacteriol. 175: 2483 2489.
21) Klemm, P., and Krogfelt, K.A. 1994. Type 1 fimbriae of
Escherichia coli, p. 9 26. In Klemm, P. (ed), Fimbriae:
―3200.
149―158.
4) Benjamin, W.H., Jr., Wu, X., and Swords, W.E. 1996. The
predicted amino acid sequence of the Salmonella typhimuri-
um virulence gene mviA(＋) strongly indicates that MviA is
a regulatory protein of a previously unknown S. typhimurium
―
―
adhesion, genetics, biogenesis, and vaccines, CRC Press,
Boca Raton, Florida.
response regulator family. Infect. Immun. 64: 2365―2367.
5) Benjamin, W.H., Jr., Yother, J., Hall, P., and Briles, D.E.
1991. The Salmonella typhimurium locus mviA regulates
virulence in Itys but not Ityr mice: functional mviA results in
avirulence; mutant (nonfunctional) mviA results in viru-
22) Laemmli, M.K. 1970. Cleavage of structural proteins during
the assembly of the head of bacteriophage T4. Nature 227:
680―685.
23) Maxam, A.M., and Gilbert, W. 1980. Sequencing end-
lence. J. Exp. Med. 174: 1073―1083.
labeled DNA with base-specific chemical cleavages. Meth-
6) Blomfield, I.C., Kulasekara, D.H., and Eisenstein, B.I. 1997.
Integration host factor stimulates both FimB- and FimE-
mediated site-specific DNA inversion that controls phase
variation of type 1 fimbriae expression in Escherichia coli.
ods Enzymol. 65: 499―560.
24) McClain, M.S., Blomfield, I.C., Eberhardt, K.J., and Eisen-
stein, B.I. 1993. Inversion-independent phase variation of
type 1 fimbriae in Escherichia coli. J. Bacteriol. 175: 4335
―
Mol. Microbiol. 23: 705―717.
4344.
7) Blomfield, I.C., McClain, M.S., Princ, J.A., Calie, P.J., and
Eisenstein, B.I. 1991. Type 1 fimbriation and fimE mutants of
25) Miller, J.H. (ed), 1972. Experiments in molecular genetics,
Cold Spring Harbor Laboratory, Cold Spring Harbor, New
York.
Escherichia coli K-12. J. Bacteriol. 173: 5298
―5307.
8) Casadaban, M.J., Chou, J., and Cohen, S.N. 1980. In vitro
26) Muramatsu, S., and Mizuno, T. 1990. Nucleotide sequence of

10
K.-S. YEH ET AL
the region encompassing the int gene of a cryptic prophage
and the dnaY gene flanked by a curved DNA sequence of
Escherichia coli K12. Mol. Gen. Genet. 220: 325 328.
34) Swenson, D.L., Clegg, S., and Old, D.C. 1991. The fre-
quency of fim genes among Salmonella serovars. Microb.
Pathog. 10: 487―492.
―
27) Old, D.C., and Duguid, J.P. 1971. Selection of fimbriate
transductants of Salmonella typhimurium dependent on
35) Swords, W.E., and Benjamin, W.H., Jr. 1994. Mouse viru-
＋
lence gene A (mviA ) is a pleiotropic regulator of gene
motility. J. Bacteriol. 107: 655
―658.
expression in Salmonella typhimurium. Ann. N.Y. Acad.
28) Old, D.C., and Duguid, J.P. 1970. Selective outgrowth of fim-
briate bacteria in static liquid medium. J. Bacteriol. 103:
Sci. 730: 295―296.
36) Tabor, S., and Richardson, C.C. 1985. A bacteriophage T7
RNA polymerase/promoter system for controlled exclusive
expression of specific genes. Proc. Natl. Acad. Sci. U.S.A.
447―456.
29) Sambrook, J., Fritsch, E.F., and Maniatis, T. (eds), 1989. Mo-
lecular cloning; a laboratory manual, Cold Spring Harbor
Laboratory Press, Cold Spring Harbor, New York.
82: 1074―1078.
37) Tinker, J.K., and Clegg, S. 2000. Characterization of FimY
30) Schmitz, A., and Galas, D.J. 1979. The interaction of RNA
polymerase and lac repressor with the lac control region.
as a coactivator of type 1 fimbrial expression in Salmonella
enterica serovar Typhimurium. Infect. Immun. 68: 3305
―
Nucleic Acids Res. 6: 111
―117.
3313.
31) Stibitz, S. 1994. Mutations in the bvgA gene of Bordetella
pertussis that differentially effect regulation of virulence
38) Urbanowski, M.L., and Stauffer, G.V. 1989. Genetic and
biochemical analysis of the MetR activator-binding site in the
metE metR control region of Salmonella typhimurium. J.
determinants. J. Bacteriol. 176: 5615―5621.
32) Stock, J.B., Ninfa, A.J., and Stock, A.M. 1989. Protein
phosphorylation and regulation of adaptive responses in
Bacteriol. 171: 5620
39) Xu, H., and Hoover, T.R. 2001. Transcriptional regulation at
a distance in bacteria. Curr. Opin. Microbiol. 4: 138 144.
―5629.
bacteria. Microbiol. Rev. 53: 450
―490.
―
33) Swenson, D.L., and Clegg, S. 1992. Identification of ancillary
fim genes affecting fimA expression in Salmonella typhimuri-
40) Yeh, K.S., Hancox, L.S., and Clegg, S. 1995. Construction
and characterization of a fimZ mutant of Salmonella
um. J. Bacteriol. 174: 7697
―7704.
typhimurium. J. Bacteriol. 177: 6861―6865.