INFECTION AND IMMUNITY,
Copyright © 1998, American Society for Microbiology
Jan. 1998, p. 330–335Vol. 66, No. 1
Use of Green Fluorescent Protein To Assess Urease Gene
Expression by Uropathogenic Proteus mirabilis during
Experimental Ascending Urinary Tract Infection
HUI ZHAO,1RICHARD B. THOMPSON,2VIRGINIA LOCKATELL,3DAVID E. JOHNSON,3
AND HARRY L. T. MOBLEY1*
Department of Microbiology and Immunology,1Department of Biochemistry and Molecular Biology,2and Division of
Infectious Diseases,3University of Maryland School of Medicine, Baltimore, Maryland 21201
Received 7 August 1997/Returned for modification 24 September 1997/Accepted 7 October 1997
Proteus mirabilis, a cause of complicated urinary tract infection, expresses urease when exposed to urea.
While it is recognized that the positive transcriptional activator UreR induces gene expression, the levels of
expression of the enzyme during experimental infection are not known. To investigate in vivo expression of
P. mirabilis urease, the gene encoding green fluorescent protein (GFP) was used to construct reporter fusions.
Translational fusions of urease accessory gene ureD, which is preceded by a urea-inducible promoter, were
made with gfp (modified to express S65T/V68L/S72A [B. P. Cormack et al. Gene 173:33–38, 1996]). Constructs
were confirmed by sequencing of the fusion junctions. UreD-GFP fusion protein was induced by urea in both
Escherichia coli DH5? and P. mirabilis HI4320. By using Western blotting with antiserum raised against GFP,
expression level was shown to correlate with urea concentration (tested from 0 to 500 mM), with highest
induction at 200 to 500 mM urea. Fluorescent E. coli and P. mirabilis bacteria were observed by fluorescence
microscopy following urea induction, and the fluorescence intensity of GFP in cell lysates was measured by
spectrophotofluorimetry. P. mirabilis HI4320 carrying the UreD-GFP fusion plasmid was transurethrally
inoculated into the bladders of CBA mice. One week postchallenge, fluorescent bacteria were detected in thin
sections of both bladder and kidney samples; the fluorescence intensity of bacteria in bladder tissue was higher
than that in the kidney. Kidneys were primarily infected with single-cell-form fluorescent bacteria, while
aggregated bacterial clusters were observed in the bladder. Elongated swarmer cells were only rarely observed.
These observations demonstrate that urease is expressed in vivo and that using GFP as a reporter protein is
a viable approach to investigate in vivo expression of P. mirabilis virulence genes in experimental urinary tract
Urinary tract infection (UTI) with Proteus mirabilis may lead
to serious complications that include renal stones, acute pye-
lonephritis, catheter obstruction, and bacteremia (20, 35, 39).
Urease is recognized as a major virulence factor for P. mirabilis
by virtue of its ability to rapidly generate ammonia from the
hydrolysis of urea present at 400 to 500 mM in urine (13).
Elevated pH results in ion precipitation in the form of struvite
or carbonate-apatite kidney or bladder stones. Ammonia may
also have a direct cytotoxic effect upon kidney cells in cultures
(13, 23, 30, 32–34). Production of urease appears to be one
reason that Proteus infections cause more severe histological
damage than Escherichia coli infections (20, 35, 40).
P. mirabilis urease, a nickel-metalloenzyme, resides in the
cytosol of the bacterium (21). The urease gene cluster is com-
prised of eight contiguous genes. The structural genes, ure-
ABC, which encode subunits of the enzyme, are flanked im-
mediately upstream by the ureD and downstream by the
ureEFG genes. These seven genes are transcribed on the same
mRNA transcript (18, 21, 22, 37). The four accessory genes
(ureDEFG) are necessary for the insertion of nickel ions into
the apoenzyme and required for assembly of a catalytically
active urease. The ureR gene lies 400 bp upstream of ureD, is
oriented opposite the other seven genes, and acts as a positive
regulator in the presence of urea to activate transcription of
urease structural and accessory genes via sequences upstream
of ureD (36).
To evaluate the contribution of urease to virulence, a ure-
ase-negative mutant was previously constructed and the viru-
lence was analyzed by our lab, using the CBA mouse model of
ascending UTI (20, 23). After 48 h of infection, the number of
mutant bacteria recovered from urine, bladder, and both kid-
neys was significantly (100-fold) lower than that of the parent
strain. After 1 week of infection, the mutant concentration was
1 million times less than that of the parent, which produced
significantly more severe pathology than the mutant. The ure-
ase-negative mutant had a 50% infective dose of ?2.7 ? 109
CFU, a value more than 1,000-fold greater than that of the
parent strain (2.2 ? 106CFU).
To assess urease expression in situ, green fluorescent protein
(GFP) from Aequorea victoria was used as a reporter protein in
this study. In comparison with products of other reporter genes
(e.g., lacZ, lux, or cat), GFP does not require addition of a
cofactor or substrate to permit observation of its expression,
merely excitation with UV light. GFP is stable in bacterial cells,
is not photobleached by prolonged exposure to UV light, and
does not require lysis of bacterial or host cell for accurate
detection (7, 11, 24, 38). The GFP variant used in these studies
[GFP(S65T/V68L/S72A)] has its peak excitation band shifted
from the wild-type position of 395 nm (470-nm shoulder) to
481 nm, as well as improved folding efficiency (9). Conse-
quently, it exhibits enhanced brightness when expressed in
* Corresponding author. Mailing address: Department of Microbi-
ology and Immunology, University of Maryland School of Medicine,
655 W. Baltimore St., Baltimore, MD 21201. Phone: (410) 706-0466.
Fax: (410) 706-2129. E-mail: email@example.com.
bacteria compared to the wild-type protein (9). The emission
maximum of the mutant GFP is 507 nm.
For studies of pathogenic bacteria, GFP has thus far been
used to assess whether promoters are active in mycobacteria
(11, 25) or Salmonella typhimurium (41, 43) within macro-
phages. S. typhimurium and Yersinia pseudotuberculosis ex-
pressing GFP have been sorted by fluorescence-activated cell
sorting (42, 43). As well, bacterium-plant interactions (12),
aquatic survival (26), and other applications (28) have been
studied by using GFP expression. In our study, an in-frame
UreD-GFP translational fusion was constructed and fluores-
cent bacteria from both in vitro cultures and tissue from ex-
perimentally challenged mice were detected by fluorescence
microscopy. We examined the feasibility of studying gene ex-
pression by using GFP in an experimental infection model of
MATERIALS AND METHODS
Bacterial strains, plasmids, and media. P. mirabilis HI4320 (urease positive;
produces MR/P, Pmf, and ATF fimbriae; hemolysin positive) was isolated from
an elderly woman with urinary catheter-associated bacteriuria (34). E. coli DH5?
[supE44 ?lacU169 (?80lacZ?M15) hsdR17 recA1 endA1 gyrA96 thi-1 relA1] was
used as a recipient for transformations. Luria broth (10 g of tryptone, 5 g of yeast
extract, and 10 g of NaCl per liter) and L agar (Luria broth containing 1.5% agar)
were used as culture media. Nonswarming agar (10 g of tryptone, 5 g of yeast
extract, 5 ml of glycerol, 0.4 g of NaCl, and 29 g of agar per liter) was used to
prevent swarming of P. mirabilis (6). The mutant gfp gene was kindly provided by
B. Cormack (Stanford University).
Recombinant DNA techniques. Chromosomal DNA was isolated by the
method of Maniatis et al. (27). Plasmid DNA was isolated by using Qiagen
columns as specified by the manufacturer (Qiagen, Inc.). Electroporation, trans-
formation, and other genetic techniques were performed by standard methods
(4, 27) or according to manufacturers’ instructions.
Nucleotide sequencing. Sequencing was performed by the dideoxy-chain ter-
mination method (4, 27) with double-stranded DNA as the template. Reagents
from the Prism Ready Reaction Dye Deoxy Termination kit (Applied Biosys-
tems) were used in conjunction with Taq polymerase (Boehringer Mannheim
Corporation). Reaction were run on a model 373 DNA sequencer (Applied
PCR. PCR was used to amplify gfp sequence from plasmid pKEN-GFPMut2
(9). Primers were synthesized by the phosphorimidite method on Applied Bio-
systems automated DNA synthesizer (model 380B). Reactions were carried out
in a thermocycler (The Minicycler, model PTC-150-16; MJ Research, Inc.), using
Vent DNA polymerase (Biolabs) or Taq DNA polymerase (Boehringer Mann-
heim). The thermocycler was programmed for 30 cycles of 94°C for 45 s, 52°C for
45 s, and 72°C for 45 s. An upstream primer (5? AGGATCCCTGCAGGTAA
AGGAGAAGAACTT 3?) contains a BamHI site and covers sequence encoding
the second, third, and fourth amino acid residues of the N terminus of GFP. The
downstream primer (5? TTGGAATTCTTATTTGTATAGTTCATCC 3?) con-
tains an EcoRI site and includes the sequence encoding the last three residues of
the C terminus of GFP. Amplification resulted a 736-bp PCR product which was
ligated into the pCRScript SrfI site.
Western blotting. Soluble protein from whole-cell French press lysates of E.
coli DH5? or P. mirabilis HI4320 containing plasmids was electrophoresed on a
sodium dodecyl sulfate (SDS)–12% polyacrylamide gel and transferred to a
polyvinylidene difluoride membrane (Immobilon-P; Millipore). Western blots
were incubated with polyclonal antiserum to GFP (1:10,000; Clontech) raised in
rabbits against recombinant GFP isolated from transformed E. coli; this was
followed by incubation with a goat anti-rabbit immunoglobulin G (1:2,000;
Sigma) coupled to alkaline phosphatase; the blot was developed with 5-bromo-
4-chloro-3-indolylphosphate–nitroblue tetrazolium (Sigma) as a chromogenic
substrate for alkaline phosphatase.
Fluorescence microscopy. An overnight Luria broth culture of strains carrying
the gfp fusion plasmid was diluted 1:100 in Luria broth containing ampicillin (50
?g/?l) and grown to an optical density at 600 nm of 0.1. Bacteria were induced
with urea (0 to 500 mM) and harvested after 3 h of additional growth.
For fluorescence microscopy, induced bacteria were washed twice in phos-
phate-buffered saline (8 g of NaCl, 0.2 g of KCl, 1.44 g of Na2HPO4, and 0.24 g
of KH2PO4per liter [pH 7.4]). A slide was prepared by air drying a drop of
culture on the surface. A Zeiss Axiophot epifluorescence microscope with filter
sets for fluorescein isothiocyanate fluorescence was used. Images were recorded
on Ektachrome color slide film (ASA 400; Kodak).
Spectrophotofluorimetry. Bacteria from induced cultures (100 ml) were
washed twice in 4 ml of 10 mM Tris-HCl (pH 7.4)–100 mM NaCl–1 mM
MgCl2–10 mM dithiothreitol and suspended in 4 ml of the buffer. Cells were
lysed in a French pressure cell (20,000 lb/in2), and the lysate was centrifuged
(5,000 ? g, 5 min, 4°C). The supernatant was collected and centrifuged (27,000 ?
g, 15 min, 4°C) (7). The emission (470-nm excitation wavelength, emission at 490
to 590 nm, 2-nm slits) and corrected excitation (330 to 530-nm excitation wave-
length, emission at 550 nm) spectra of the supernatant from the high-speed
centrifugation were obtained on a Spectronics AB-2 spectrophotofluorimeter
(Spectronics, Inc., Rochester, N.Y.).
CBA mouse model of ascending UTI. A modification (19) of the mouse model
of ascending UTI originally developed by Hagberg et al. (14) was used. Female
mice (20 to 22 g, 6 to 8 weeks old; Jackson Laboratory, Bar Harbor, Maine)
tested for the absence of bacteriuria were anesthetized with methoxyflurane and
inoculated with P. mirabilis HI4320 (107CFU suspended in 0.05 ml of phosphate-
buffered saline) through a sterile polyethylene catheter inserted into the bladder
through the urethra. Mice were provided with drinking water containing ampi-
cillin (250 ?g/ml). After 1 week, the mice were sacrificed by administration of an
overdose of methoxyflurane. Urine was collected, and the bladder and both
kidneys were removed. Half of the bladder or kidney samples were embedded in
OCT (Tissue-Tek; Miles Inc.), frozen on dry ice, and cryosectioned into 5- to
10-?m sections for fluorescence microscopic analysis. The remaining half of each
sample was quantitatively cultured, and viable counts were determined as CFU/
milliliter of urine or CFU/gram of tissue.
Construction of ureD-gfp translational fusion. Urease ex-
pression is regulated at the transcriptional level. ureD is up-
stream of structural subunits ureABC and is transcribed on the
same mRNA as ureABC, under control of the same promoter
(18, 22, 37) (Fig. 1). Thus, the level of UreD-GFP expression
reflects urease apoenzyme expression. Full-length ureR and the
first 108 bp of ureD were cloned on an EcoRI/BamHI fragment
from pMIR10DZ (18), which is a subclone of pMID1010, into
pBluescript to form pURE-RD. A fragment carrying the gfp
open reading frame amplified by PCR from pGFPmut2 (9) was
cloned into pCRScript; the BamHI fragment from this plasmid
was isolated and ligated into BamHI-digested pURE-RD. The
resultant plasmid, designated pURE-RD-GFP (Fig. 1), was
isolated, and insertion of the proper fragment was confirmed
by restriction enzyme digestion. The in-frame translational fu-
sion was confirmed by nucleotide sequencing of the junction
(data not shown).
Urea induction and Western blotting. To demonstrate that
synthesis of UreD-GFP could be induced by urea, E. coli
DH5?(pURE-RD-GFP) was grown in Luria broth at 37°C.
When exponentially growing cultures reached an optical den-
sity at 600 nm of 0.1, urea (0 to 500 mM) was added. After 2 h
of induction, bacteria were collected and lysed by passage
through a French pressure cell. Western blots of soluble pro-
tein were used to assess expression of the UreD-GFP fusion.
FIG. 1. Construction of a ureD-gfp fusion. Intact ureR and part of ureD were
cloned as an EcoRI/BamHI fragment from pMIR10DZ (18), which is a subclone
of pMID1010, into pBluescript to form pURE-RD. A fragment carrying the gfp
open reading frame (see text) amplified by PCR from pGFPmut2 (9) was cloned
into pCRScript; the BamHI fragment from this plasmid was isolated and ligated
into BamHI-digested pURE-RD. The final construct was designated pURE-RD-
VOL. 66, 1998GFP REPORTER IN EXPERIMENTAL UTI 331
Soluble protein from either E. coli DH5? or P. mirabilis
HI4320 transformed with pURE-RD-GFP, induced or unin-
duced by urea, was electrophoresed on an SDS–12% polyacryl-
amide gel. Proteins were transferred to nitrocellulose and re-
acted with rabbit anti-GFP. The UreD-GFP fusion protein was
predicted to contain the first 37 amino acids of UreD and all
but the first two of a total of 238 amino acids of GFP. The
fusion protein, therefore, was predicted to contain 275 amino
acids and have a molecular size of 30.0 kDa. Western blot
analysis showed that urea-induced E. coli(pURE-RD-GFP)
cells produced a polypeptide of 30 kDa, consistent with the
predicted size (Fig. 2A). The expression level correlated with
the urea concentration, with maximal induction at 200 and 500
mM. The band was missing from the uninduced sample, indi-
cating that the induction of fusion GFP synthesis occurs only in
the presence of urea in a E. coli background. Although P.
mirabilis(pURE-RD-GFP) exhibited similar urea concentra-
tion-dependent expression, it also produced a low level of
UreD-GFP fusion protein in the absence of urea (Fig. 2B).
Fluorescence determination of GFP expression. The level of
UreD-GFP fusion protein expression was also quantitated by
spectrophotofluorimetry. Soluble cell extracts were prepared
from uninduced or urea-induced (50 and 250 mM) bacterial
cultures. Both fluorescence excitation (data not shown) and
emission spectra of the induced soluble cell extracts from E.
coli (Fig. 3A) and P. mirabilis (Fig. 3B) were very similar to
those observed previously for this GFP variant (9), suggesting
that GFP had indeed been expressed. Upon induction, both
species exhibited dramatic, urea-dependent increases in fluo-
rescence compared with the corresponding uninduced strains.
In particular, E. coli displayed a 33-fold increase in intensity at
509 nm when induced with 50 mM urea and a 50-fold increase
when induced with 250 mM urea (Fig. 3A). Similarly, P. mira-
bilis exhibited a fourfold increase in emission at 509 nm when
induced with 250 mM urea and a negligible change in emission
when induced with 50 mM urea (Fig. 3B). The differences in
the degree of fluorescence enhancement are attributable to
both lower fluorescence background in E. coli and more sub-
stantial expression of GFP.
Visible fluorescence of bacteria after urea induction. Fol-
lowing urea induction in vitro, E. coli(pURED-GFP) and P.
mirabilis(pURED-GFP) were also quantitatively assessed by
fluorescence microscopy. For E. coli containing the construct,
without induction there were no visible fluorescent bacteria. At
10 mM urea, all bacteria were very weakly fluorescent; at ?100
mM, all bacteria were brightly fluorescent. For P. mirabilis
containing the construct, without induction, the vast majority
of bacteria showed no fluorescence; a very small percentage of
bacteria, however, were brightly fluorescent. At 10 mM urea,
all of the bacteria were at least weakly fluorescent and some
were brightly fluorescent; at ?100 mM, all bacteria were
brightly fluorescent but did not display the same intensity as
did the transformed E. coli.
In vivo expression of GFP in experimental ascending UTI.
To determine whether GFP could be used as a reporter for
urease expression in vivo and whether the expression level was
sufficient during infection to visualize bacteria, CBA mice were
inoculated transurethrally with 107CFU of P. mirabilis
HI4320(pURED-GFP). After 1 week, the geometric means of
log10concentrations of bacteria in urine, bladder, and kidney
were determined and found to be typical of previous experi-
mental infections (5): urine, 7.79 CFU/ml; bladder, 6.22
CFU/g; and kidneys, 5.15 CFU/g. Thus, GFP expression did
not compromise survival of the challenge strains.
To search for fluorescent bacteria, frozen thin-sectioned
bladder (Fig. 4A to H) or kidney (Fig. 4I to K) samples from
infected animals were observed by fluorescence microscopy.
Numerous green fluorescent bacteria were detected in both
bladder and kidney samples. The fluorescence intensity of bac-
teria from bladder appeared qualitatively higher than that
from kidney, suggesting a higher level urease expression in
bladder than in kidney. While some single bacteria attached to
the surface of bladder epithelium by the pole of the organism
(Fig. 4G), other aggregated bacteria, apparently covered in
FIG. 2. Western blot analysis of UreD-GFP fusion protein induction by urea.
E. coli (A) and P. mirabilis (B) carrying pURE-RD-GFP were uninduced or
induced by urea (10, 50, 100, 200, and 500 mM). Soluble protein from these
strains was electrophoresed on an SDS–12% polyacrylamide gel. A polyclonal
antiserum raised in rabbits against recombinant GFP was used for Western
FIG. 3. Fluorescence emission spectra of soluble extracts from E. coli
DH5?(pURE-RD-GFP) (A) and P. mirabilis HI4320(pURE-RD-GFP) (B).
Bacteria were induced with 0 (——), 50 (— — —), and 250 (– – –) mM urea.
332ZHAO ET AL.INFECT. IMMUN.
FIG. 4. In vivo expression of UreD-GFP by P. mirabilis infecting the bladders and kidneys of CBA mice. Thin sections of bladders (A to H) and kidneys (I to K)
obtained from CBA mice infected with P. mirabilis HI4320(pURE-RD-GFP) were observed by fluorescence microscopy. Fluorescent bacteria are identified by arrows.
(F and H) Phase-contrast images of panels E and G, respectively.
biofilm, were loosely attached to the bladder tissue (Fig. 4D
and E). Such large aggregates were commonly observed in the
bladder. While an occasional elongated swarmer cell was ob-
served in the bladder sections (Fig. 4B and C), only single
vegetative forms of bacteria were found scattered around the
kidney tissue samples. The results clearly indicate that the urea
concentration was high enough to induce the GFP-UreD fu-
sion protein levels sufficient to make individual bacteria easily
visible by fluorescence microscopy.
An in-frame translational fusion of ureD with gfp (S65T/
V68L/S72A) was successfully constructed and confirmed by
nucleotide sequencing of the fusion junction. Using Western
blotting and spectrophotofluorimetry, we found expression of
UreD-GFP only in the samples with urea induction in E. coli;
the expression level was correlated with urea concentration,
with highest induction at 200 to 500 mM, an observation con-
sistent with previous reports (37). We have learned that urease
genes are indeed expressed in vivo by P. mirabilis during UTI,
confirming an assertion for which significant circumstantial
evidence exists. More importantly, however, we have demon-
strated that GFP can be used successfully to study expression
of virulence genes in an experimental model of ascending UTI.
While we are confident that expression of the fusion protein
represents an accurate proxy for measurement of urease activ-
ity, it should be stressed that we are measuring expression of
ureD (encodes an accessory protein that is not part of the
enzyme), a gene directly upstream of ureA (encodes the small-
est subunit of the apoenzyme) (Fig. 1). It has been determined
previously that ureD and ureA are transcribed on the same
urea-inducible mRNA (18, 21, 22, 37). Nevertheless, expres-
sion of the UreD-GFP fusion is an indirect measurement of
urease enzyme expression.
In these studies, we noted that uninduced cultures of P.
mirabilis(pURED-GFP) produced a low level of the fusion
protein whereas E. coli carrying the same multicopy plasmid
maintains tight regulation of enzyme synthesis in the absence
of urea. This was observed both on Western blots (Fig. 2) and
by spectrophotofluorimetry (Fig. 3) (compare levels of unin-
duced production of fusion proteins and fluorescence intensity
for P. mirabilis and E. coli). This finding is consistent with the
fact that uninduced P. mirabilis produces a low level of urease
(21, 31). These observations suggest that P. mirabilis may have
an additional tier of regulation beyond UreR-mediated tran-
scriptional activation. Allison et al. (3) provided evidence for
this reporting that expression of urease-specific mRNA is in-
creased during swarming, suggesting that expression of urease
goes beyond simple urea induction. Indeed, it is logical to
always produce some enzyme; a low level of urease may be
necessary for adequate nitrogen metabolism in P. mirabilis in
the bowel or outside the host. In some bacterial species, like
Morganella morganii, urease is synthesized constitutively to en-
sure that some enzyme is always produced (17).
Before the S65T/V68L/S72A variant of gfp was available,
both wild-type gfp and gfp (S65T) (8, 15) were fused by us to
ureD (data not shown). In both cases, however, no strong
fluorescence emitted from either E. coli or P. mirabilis carrying
the pURED-GFP plasmid could be observed by fluorescence
microscopy or spectrophotofluorimetry. Therefore, we looked
for expression of the GFP fusion protein in whole-cell extract
by Western blotting using polyclonal anti-GFP. Western blot-
ting demonstrated that the fusion protein was expressed and
that levels of induction correlated with urea concentration.
However, by separating inclusion bodies from whole-cell ex-
tract, we noted that most of the fusion protein partitioned with
the inclusion bodies. This finding is consistent with what has
been reported by several groups, specifically that overex-
pressed GFP in inclusion bodies of bacteria does not generate
the internal chromophore and is therefore nonfluorescent (16).
The newest variant of gfp (S65T/V68L/S72A) appears to over-
come the folding problem in bacteria and also has increased
fluorescence intensity. Therefore, this version of GFP, unlike
previous versions, is suitable for in vivo studies in the urinary
In vivo expression of UreD-GFP was assessed in experimen-
tal ascending UTI. Urea output in mouse urine (24.3 mg/24 h;
range of volume, 0.9 to 2.9 ml; therefore, the urea concentra-
tion range is 140 to 450 mM ) is similar to that of humans
and is high enough to fully induce GFP in P. mirabilis trans-
formed with pURED-GFP encoding the translational fusion.
Fluorescent bacteria were detected as single cells in both blad-
der and kidney, indicating that the P. mirabilis urease gene was
induced in both tissues. In the bladder, some interesting phe-
nomena were observed. First, adherence of single bacteria to
the bladder epithelium, which may be the first step of coloni-
zation of the host, was mediated by one end of the cell, sug-
gesting the polarized distribution of the adhesin structures or
an intimate attachment by the bacterium. Second, aggregates
and multiple layers of bacteria appeared to be embedded in
biofilm (polysaccharide matrix) that was loosely attached to
surface of the bladder epithelium. Interestingly, bacteria clus-
tered inside a protective biofilm have been implicated in
chronic bacterial UTI and bladder stone formation and may
have the advantage of being more resistant to host defenses
and antibiotic therapy (29, 40). Third, vegetative forms of P.
mirabilis (single bacterial cells as opposed to elongated swarm-
ing cells) were most often observed in the bladder tissue.
Fourth, elongated swarming cells (5 to 10 cell lengths) were
occasionally found in bladder; the role of these cells, however,
is unclear. In the kidneys, extracellular bacteria were fluores-
cent and tended to remain as single cells in a vegetative form.
Since the fluorescence was qualitatively weaker in the kidney
than in the bladder, this finding suggested that either the urea
concentration was lower in the kidney or that bacteria in the
kidney were less accessible to urine because they had invaded
more deeply into tissue. The fact that the elongated swarming
cells were not observed in kidneys does not necessarily mean
that swarming cells do not play a role in infection. These cells
may have invaded kidney cells where the urea concentration
was too low to induce the fusion protein (1, 2) and thus make
themselves visible. Nevertheless, we have demonstrated for the
first time that the enhanced GFP can be used to study expres-
sion of virulence genes by P. mirabilis in a mouse model of
This work was supported in part by Public Health Service grants
AI23328 and DK47920 from the National Institutes of Health.
1. Allison, C., N. Coleman, P. L. Johns, and C. Hughes. 1992. Ability of Proteus
mirabilis to invade human urothelial cells is coupled to motility and swarming
differentiation. Infect. Immun. 60:4740–4746.
2. Allison, C., L. Emody, N. Coleman, and C. Hughes. 1994. The role of swarm
cell differentiation and multicellular migration in the uropathogenicity of
Proteus mirabilis. J. Infect. Dis. 169:1155–1158.
3. Allison, C., H.-C. Lai, and C. Hughes. 1992. Co-ordinate expression of
virulence genes during swarm cell differentiation and population migration
of Proteus mirabilis. Mol. Microbiol. 6:1583–1591.
4. Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A.
Smith, and K. Struhl (ed.). 1995. Current protocols in molecular biology.
John Wiley & Sons, Inc., New York, N.Y.
334 ZHAO ET AL.INFECT. IMMUN.
5. Bahrani, F. K., G. Massad, C. V. Lockatell, D. E. Johnson, R. Russell, J. W. Download full-text
Warren, and H. L. T. Mobley. 1994. Construction of an MR/P fimbrial
mutant of Proteus mirabilis: role in virulence in a mouse model of ascending
urinary tract infection. Infect. Immun. 62:3363–3371.
6. Belas, R., D. Erskine, and D. Flaherty. 1991. Transposon mutagenesis in
Proteus mirabilis. J. Bacteriol. 173:6289–6293.
7. Chalfie, M., T. Yuan, G. Euskirchen, W. W. Ward, and D. C. Prasher. 1994.
Green fluorescent protein as a marker for gene expression. Science 263:802–
8. Clontech. 1996. Living colors™ enhanced GFP vectors. CLONTECHniques
9. Cormack, B. P., R. H. Valdivia, and S. Falkow. 1996. FACS-optimized
mutants of the green fluorescent protein (GFP). Gene 173:33–38.
10. Crispens, C. G. Jr. 1975. Handbook on the laboratory mouse, p. 130. Charles
C Thomas, Springfield, Ill.
11. Dhandayuthapani, S., L. E. Via, C. A. Thomas, P. M. Horowitz, D. Deretic,
and V. Deretic. 1995. Green fluorescent protein as a marker for gene ex-
pression and cell biology of mycobacterial interaction with macrophages.
Mol. Microbiol. 17:901–912.
12. Gage, D. J., T. Bobo, and S. R. Long. 1996. Use of green fluorescence protein
to visualize the early events of symbiosis between Rhizobium meliloti and
alfalfa. J. Bacteriol. 178:7159–7166.
13. Griffith, D. P., D. M. Musher, and C. Itin. 1976. Urease: the primary cause
of infection-induced urinary stones. Invest. Urol. 13:346–350.
14. Hagberg, L., I. Engberg, R. Freter, J. Lam, S. Olling, and C. Svanborg-Eden.
1983. Ascending unobstructed urinary tract infection in mice cause by pyelone-
phritogenic Escherichia coli of human origin. Infect. Immun. 40:273–283.
15. Heim, R., A. B. Cubitt, and R. Y. Tsien. 1995. Improved green fluorescence.
16. Heim, R., D. C. Prasher, and R. Y. Tsien. 1994. Wavelength mutations and
posttranslational autoxidation of green fluorescent protein. Proc. Natl. Acad.
Sci. USA 91:12501–12504.
17. Hu, L.-T., E. B. Nicholson, B. D. Jones, M. J. Lynch, and H. L. T. Mobley.
1990. Morganella morganii urease: purification, characterization, and isola-
tion of gene sequences. J. Bacteriol. 172:3073–3080.
18. Island, M., and H. L. T. Mobley. 1995. Proteus mirabilis urease: operon
fusion and linker insertion analysis of ure gene organization, regulation, and
function. J. Bacteriol. 177:5653–5600.
19. Johnson, D. E., C. V. Lockatell, M. Hall-Craigs, H. L. T. Mobley, and J. W.
Warren. 1987. Uropathogenicity in rats and mice of Providencia stuartii from
long-term catheterized patients. J. Urol. 138:632–635.
20. Johnson, D. E., R. G. Russell, C. V. Lockatell, J. C. Zulty, J. W. Warren, and
H. L. T. Mobley. 1993. Contribution of Proteus mirabilis urease to persis-
tence, urolithiasis, and acute pyelonephritis in a mouse model of ascending
urinary tract infection. Infect. Immun. 61:2748–2754.
21. Jones, B. D., and H. L. T. Mobley. 1988. Proteus mirabilis urease: genetic
organization, regulation, and expression of structural genes. J. Bacteriol.
22. Jones, B. D., and H. L. T. Mobley. 1989. Proteus mirabilis urease: nucleotide
sequence determination and comparison with jack bean urease. J. Bacteriol.
23. Jones, B. D., C. V. Lockatell, D. E. Johnson, J. W. Warren, and H. L. T.
Mobley. 1990. Construction of a urease-negative mutant of Proteus mirabilis:
analysis of virulence in a mouse model of ascending urinary tract infection.
Infect. Immun. 58:1120–1123.
24. Kain, S. R., M. Adams, A. Kondepudi, T. T. Yang, W. W. Ward, and P. Kitts.
1995. Green fluorescent protein as a reporter of gene expression and protein
localization. BioTechniques 19:650–655.
25. Kremer, L., A. Baulard, J. Estaquier, O. Poulain-Godefroy, and C. Locht.
1995. Green fluorescence protein as a new expression marker in mycobac-
teria. Mol. Microbiol. 17:913–922.
26. Leff, L. G., and A. A. Leff. 1996. Use of green fluorescent protein to monitor
survival of genetically engineered bacteria in aquatic environments. Appl.
Environ. Microbiol. 62:3486–3488.
27. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecular cloning: a
laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor,
28. Matthysse, A. G., S. Stretton, C. Dandie, N. C. McClure, and A. E. Goodman.
1996. Construction of GFP vectors for use in gram-negative bacteria other
than E. coli. FEMS Microbiol. Lett. 145:87–94.
29. McLean, R. J. C., J. A. Downey, A. L. Lablans, J. M. Clark, A. J. Dumanski,
and J. C. Nickel. 1992. Modeling biofilm-associated urinary tract infections
in animals. Int. J. Biodeterior. Biodegrad. 30:201–216.
30. Mobley, H. L. T., and G. R. Chippendale. 1990. Hemagglutinin, urease, and
hemolysin production by Proteus mirabilis from clinical sources. J. Infect. Dis.
31. Mobley, H. L. T., G. R. Chippendale, K. G. Swihart, and R. Welch. 1991.
Cytotoxicity of the HpmA hemolysin and urease of Proteus mirabilis and
Proteus vulgaris against cultured human renal proximal tubular epithelial
cells. Infect. Immun. 59:2036–2042.
32. Mobley, H. L. T., M. D. Island, and R. P. Hausinger. 1995. Molecular biology
of microbial urease. Microbiol. Rev. 59:451–480.
33. Mobley, H. L. T., and R. P. Hausinger. 1989. Microbial ureases: significance,
regulation, and molecular characterization. Microbiol. Rev. 53:85–108.
34. Mobley, H. L. T., and J. W. Warren. 1987. Urease positive bacteriuria and
obstruction of long term urinary catheters. J. Clin. Microbiol. 25:2216–2219.
35. Mobley, H. L. T. 1996. Virulence of Proteus mirabilis, p. 245–271. In H. L. T.
Mobley and J. W. Warren (ed.), Urinary tract infection: molecular patho-
genesis and clinical management. ASM Press, Washington, D.C.
36. Nicholson, E. B., E. A. Concaugh, P. A. Foxall, M. D. Island, and H. L. T.
Mobley. 1993. Proteus mirabilis urease: transcriptional regulation by ureR. J.
37. Nicholson, E. B., E. A. Concaugh, and H. L. T. Mobley. 1991. Proteus
mirabilis urease: use of a ureA-lacZ fusion demonstrates that induction is
highly specific for urea. Infect. Immun. 64:5332–5340.
38. Prasher, D. C., V. K. Eckenrode, W. W. Ward, and F. G. Prendergast. 1992.
Primary structure of the Aequorea victoria green fluorescent protein. Gene
39. Rubin, R., N. Tolkoff-Rubin, and R. Cotran. 1986. Urinary tract infection,
pyelonephrititis and reflux nephropathy, p. 1085–1141. In G. Brenner and F.
Rector, Jr. (ed.), The kidney, 3rd ed. The W. B. Saunders Co., Philadelphia,
40. Salyers, A. A., and D. D. Whitt. 1994. Host defenses against bacterial-
pathogens: defenses of body surfaces, p. 3–15. In A. A. Salyers and D. D.
Whitt (ed.), Bacterial pathogenesis: a molecular approach. ASM Press,
41. Valdivia, R. H., and S. Falkow. 1997. Fluorescence-based isolation of bac-
terial genes expressed within host cells. Science 277:2007–2011.
42. Valdivia, R. H., A. E. Hromockyj, D. Monack, L. Ramakrishnan, and S.
Falkow. 1996. Applications for green fluorescent protein (GFP) in the study
of host-pathogen interaction. Gene 173:47–52.
43. Valdivia, R. H., and S. Falkow. 1996. Bacterial genetics by flow cytometry:
rapid isolation of Salmonella typhimurium acid-inducible promoters by dif-
ferential fluorescence induction. Mol. Microbiol. 22:367–378.
Editor: J. T. Barbieri
VOL. 66, 1998GFP REPORTER IN EXPERIMENTAL UTI335