Download Tract Epithelial Barrier of the Infected Urinary Required for

This information is current as
of June 17, 2017.
Macrophage Inflammatory Protein-2 Is
Required for Neutrophil Passage Across the
Epithelial Barrier of the Infected Urinary
Tract
Long Hang, Masashi Haraoka, William W. Agace, Hakon
Leffler, Marie Burdick, Robert Strieter and Catharina
Svanborg
J Immunol 1999; 162:3037-3044; ;
http://www.jimmunol.org/content/162/5/3037
Subscription
Permissions
Email Alerts
This article cites 22 articles, 17 of which you can access for free at:
http://www.jimmunol.org/content/162/5/3037.full#ref-list-1
Information about subscribing to The Journal of Immunology is online at:
http://jimmunol.org/subscription
Submit copyright permission requests at:
http://www.aai.org/About/Publications/JI/copyright.html
Receive free email-alerts when new articles cite this article. Sign up at:
http://jimmunol.org/alerts
The Journal of Immunology is published twice each month by
The American Association of Immunologists, Inc.,
1451 Rockville Pike, Suite 650, Rockville, MD 20852
Copyright © 1999 by The American Association of
Immunologists All rights reserved.
Print ISSN: 0022-1767 Online ISSN: 1550-6606.
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
References
Macrophage Inflammatory Protein-2 Is Required for
Neutrophil Passage Across the Epithelial Barrier of the
Infected Urinary Tract1
Long Hang,* Masashi Haraoka,2* William W. Agace,3* Hakon Leffler,* Marie Burdick,†
Robert Strieter,† and Catharina Svanborg3*
M
ucosal infections trigger local and systemic inflammatory responses (1, 2). Chemotactic substances are released at the site of infection, and inflammatory cells
are recruited. Neutrophils leave the blood stream, migrate through
the tissues, and cross the epithelial barrier into the lumen. Epithelial cells in the mucosal lining are one source of the local chemoattractants, but the role of epithelial chemokines for neutrophil
recruitment in vivo is not well understood.
The epithelial chemokine response to infection has been extensively studied in vitro (3–5) using cultured epithelial cells. Mucosal pathogens stimulate epithelial cells to secrete C-X-C and C-C
chemokines. In the transwell model system, pioneered by Madara
et al. (6), IL-8 has been shown to support neutrophil migration
across uroepithelial cell layers infected with Escherichia coli.
Monoclonal and polyclonal anti-IL-8 Abs completely blocked the
infection-induced increase in neutrophil migration. The effect was
*Department of Medical Microbiology, Division of Clinical Immunology, Lund University, Lund, Sweden; and †Department of Internal Medicine, Division of Pulmonary
and Critical Care Medicine, University of Michigan Medical Center, Ann Arbor,
Michigan, 48109
Received for publication July 10, 1998. Accepted for publication December 3, 1998.
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
due to cell-bound rather than soluble IL-8 and was made possible
by a parallel increase in IL-8 receptor expression by the infected
cells.
Urinary tract infections (UTI)5 are accompanied by a neutrophil
response. Indeed, “pyuria” is one of the classical diagnostic tools
in UTI. IL-8 has also been implicated as a major neutrophil chemoattractant in the human urinary tract (1, 3). Recent studies have
shown that urine IL-8 levels increase in patients with UTI, especially during acute pyelonephritis, and that urine IL-8 can support
neutrophil migration in vitro (7, 8). The kinetics of the mucosal
IL-8 response were studied in human volunteers after deliberate
intravesical inoculation with bacteria. The urine IL-8 concentrations increased rapidly and correlated with urine neutrophil numbers in individual patients (3). These in vitro and in vivo observations strongly suggested that IL-8 is involved in the mucosal
neutrophil responses and that IL-8 of epithelial origin directs neutrophil migration across epithelial barriers.
Macrophage inflammatory protein (MIP)-2 is one of the IL-8
homologues in the mouse. The aim of this study was to examine
the C-X-C chemokines and especially the MIP-2 response to experimental UTI and the contribution of MIP-2 to the neutrophil
recruitment during UTI.
Materials and Methods
Bacteria
1
This study was supported by the Medical Faculty, University of Lund, the Swedish
Medical Research Council (Grants No. 7934, 10852), the Österlund, Crawford, Wallenberg, and Lundberg Foundations, and the Royal Physiographic Society.
2
Present address: Department of Urology, Children’s Hospital Medical Center
Fukuoka, 2-5-1 Toujin-mati, Chyuo-ku, Fukuoka, 810, Japan.
3
Present address: Division of Rheumatology, Immunology and Allergy, Lymphocyte
Biology Section, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA 02115.
4
Address correspondence and reprint requests to Dr. Catharina Svanborg, Department of Medical Microbiology, Division of Clinical Immunology, Lund University,
Sàlvegatan 23, S-223-62, Lund, Sweden. E-mail address: Catharina.Svanborg@
mmb.lu.se
Copyright © 1999 by The American Association of Immunologists
E. coli 1177 of serotype O1:K1:H7, was isolated from a child with acute
pyelonephritis (9). The strain was previously shown to cause infection in
the mouse UTI model and to evoke a strong inflammatory response (10).
It expressed P and type 1 fimbriae but was hemolysin negative. Bacteria
were maintained in deep agar stabs sealed with sterile paraffin. For infection, bacteria were grown overnight in static Luria broth and harvested by
5
Abbreviations used in this paper: UTI, urinary tract infection; MIP, macrophage
inflammatory protein; MPO, myeloperoxidase; KC, keratinocyte; JE, monocyte chemoattractant protein-1, MCP-1; ENA, epithelial neutrophil activating peptide; Sap,
saponin; TBS, Tris-buffered saline.
0022-1767/99/$02.00
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
IL-8 is a major human neutrophil chemoattractant at mucosal infection sites. This study examined the C-X-C chemokine response
to mucosal infection, and, specifically, the role of macrophage inflammatory protein (MIP)-2, one of the mouse IL-8 equivalents,
for neutrophil-epithelial interactions. Following intravesical Escherichia coli infection, several C-X-C chemokines were secreted
into the urine, but only MIP-2 concentrations correlated to neutrophil numbers. Tissue quantitation demonstrated that kidney
MIP-2 production was triggered by infection, and immunohistochemistry identified the kidney epithelium as a main source of
MIP-2. Treatment with anti-MIP-2 Ab reduced the urine neutrophil numbers, but the mice had normal tissue neutrophil levels.
By immunohistochemistry, the neutrophils were found in aggregates under the pelvic epithelium, but in control mice the neutrophils crossed the urothelium into the urine. The results demonstrate that different chemokines direct neutrophil migration from
the bloodstream to the lamina propria and across the epithelium and that MIP-2 serves the latter function. These findings suggest
that neutrophils cross epithelial cell barriers in a highly regulated manner in response to chemokines elaborated at this site. This
is yet another mechanism that defines the mucosal compartment and differentiates the local from the systemic host response. The
Journal of Immunology, 1999, 162: 3037–3044.
3038
MIP-2 REQUIREMENT FOR NEUTROPHIL PASS ACROSS EPITHELIA
centrifugation at 4000 rpm for 10 min. The pellet was resuspended in 0.01
M of PBS, pH 7.2, at a concentration of 1–2 3 109 CFU/ml. The bacterial
concentration was confirmed by viable counts.
Experimental UTI
Chemokine responses
Murine MIP-2 (12), keratinocyte (KC) (13), MIP-1a (14) monocyte chemoattractant protein-1, MCP-1 (JE) (15), eotaxin (16), and human epithelial neutrophil activating peptide-78 (ENA-78) (17, 18) recombinant chemokines were used for the generation of Abs. Standards for ELISA were
purchased from R & D Systems (Minneapolis, MN). Polyclonal antimurine and anti-human chemokine antisera were produced by immunization of rabbits with chemokines at multiple intradermal sites with CFA.
Polyclonal anti-human ENA antiserum cross-reacts with the murine
equivalent (17).
Murine MIP-2, KC, ENA, MIP-1a, JE, and eotaxin in the urine and
MIP-2 in the kidneys and bladders were quantitated by a modification of a
double ligand method (19). Urine samples were centrifuged and the supernatant stored at 220°C for chemokine analysis. Kidneys and bladders were
removed, immediately snap frozen, and stored at 270°C. Kidneys or bladders were homogenized in 3 ml of lysis buffer containing 0.5% Triton
X-100, 150 mM NaCl, 15 mM Tris, 1 mM CaCl2, and 1 mM MgCl2, pH
7.40, using a tissue homogenizer (Dremel, Racine, WI). Homogenates were
incubated on ice for 30 min, then centrifuged at 1500 3 g for 10 min.
Supernatants were collected, passed through a 0.45-mm filter (Gelman Sciences, Ann Arbor, MI), and used in the ELISA.
Assessment of the neutrophil response
The number of neutrophils migrating across the mucosa into the urine was
quantified in uncentrifuged urine, using a hemocytometer chamber. Earlier
studies have shown that 99% of the inflammatory cell infiltrate consists of
neutrophils (1).
Tissue neutrophils were quantitated using the myeloperoxidase (MPO)
assay (20). Kidneys and bladders were homogenized in 2 ml of 50 mM
potassium phosphate, pH 6.0, with 5% hexadecyl-trimethylammonium
bromide and 5 mM EDTA. The homogenate was sonicated and centrifuged
at 12,000 3 g for 15 min. The supernatant was mixed 1:15 with assay
buffer and read at 490 nm. MPO units were calculated as the change in
absorbance over time.
Inhibition of neutrophil recruitment
RB6-8C5, a rat IgG2b mAb specific for murine neutrophils and eosinophils
(21–23) was a kind gift from Dr. A. Sjöstedt (Umea University, Sweden),
Dr. W. Conlan (Trudeau Institute, Saranac Lake, NY), and Dr. R. Coffman
(DNAX Research Institute, Palo Alto, CA). mAb RB6-8C5 was purified
from hybridoma supernatants by ammonium sulfate precipitation or by
protein G-Sepharose (Pharmacia Biotech, Uppsala, Sweden). The Ig concentration of the purified mAb was determined by ELISA. The mAb or
control rat IgG (0.25 mg diluted in 0.5 ml of pyrogen-free saline) were
injected i.p. into mice 24 h and again 2 h before bacterial inoculation.
Polyclonal rabbit anti-murine MIP-2 Abs were generated by immunization of rabbits with murine MIP-2 (R & D Systems) (19, 12). Mice were
injected i.p. with anti-MIP-2 antiserum or preimmune serum, (500 ml/
mouse), 2 h before intravesical infection.
Histology and immunohistochemical analysis was performed on kidneys
and bladders obtained from mice sacrificed at 0, 2, 6, and 24 h after
inoculation (24). Tissues were cut to 3 3 4 3 5 mm pieces, embedded in
OCT compound (Tissue Tek, Miles, Elkhart IN), rapidly frozen in liquid
nitrogen, and kept at 280°C until examined. Sections were cut with a steel
knife (6 mm) and mounted on poly-L-lysine-coated glass slides. Urine samples were harvested at 6 h postinfection and were spun onto poly-L-lysinecoated glass slides in a Cytospin 11 cytocentrifuge (Shandon Southern
Product, Chesire, U.K.) at 300 rpm for 5 min. Samples were fixed in freshly
prepared 4% paraformaldehyde in PBS, pH 7.4, for 15 min, rinsed in PBS,
and air dried. Sections were stained with hematoxylin and eosin for histology observation. The samples were treated with 0.1% saponin (Sigma,
St. Louis, MO) in PBS (PBS-Sap) containing 5% normal mouse serum for
60 min in room temperature to reduce nonspecific binding. After washing
in PBS-Sap three times, the samples were incubated with a 1:100 dilution
of rabbit anti-mouse MIP-2 antiserum or preimmune serum overnight at
4°C. After washing in 0.1% saponin in Tris-buffered saline (TBS-Sap),
alkaline phosphate-conjugated goat anti-rabbit Igs (Dako, Copenhagen,
Denmark) were added at a 1:50 dilution’s in TBS-Sap and left to incubate
for 60 min at room temperature in a moist chamber. After washes in TBSSap, the Fast-Red substrate containing levamisole (Dako) was prepared
according to the manufacturer’s recommendations, added to the samples,
and left to incubate for 20 min at room temperature. Thereafter, the samples were washed in TBS, pH 7.6, and counterstained with Mayer hematoxylin (Kebo Laboratories, Stockholm, Sweden) for a few seconds,
washed in distilled water, and mounted with Mount-quick “AQUEOUS”
(Daido Sangyo, Tokyo, Japan). The samples were investigated under a
light microscope (Microphot, Nikon, LRI Instruments AB, Tokyo, Japan).
Statistical analysis
Differences between samples were analyzed with the Mann-Whitney U test
(two-tailed) and the Spearman rank correlation test. Differences were considered significant for values of p , 0.05.
Results
Chemokine response to UTI
The C-C and C-X-C chemokine response to intravesical E. coli
1177 infection was examined in urine samples obtained 2, 6, and
24 h after infection (Fig. 1). KC and JE increased rapidly with peak
levels after 2– 6 h. MIP-2 and eotaxin had increased by 2 h but
reached peak levels after 6 h. ENA-78 increased slowly, with the
highest concentrations after 24 h. MIP-1a was also elevated before
infection and showed no significant change over time. These results demonstrated that E. coli elicits a diverse chemokine response
involving both the C-C and the C-X-C chemokine families.
Neutrophil recruitment to the urinary tract
The neutrophil response to intravesical E. coli 1177 infection was
quantitated in urine samples obtained at various times after infection (Fig. 2). Urine neutrophil numbers had increased after 2 h
(66 6 12 cells/ml), reached a peak by 6 h (238 6 31 cells/ml), and
remained elevated at 24 h after infection (129 6 22 cells/ml) (Fig.
2A).
Tissue sections showed an influx of neutrophils into kidneys and
bladders with a peak after 6 h (Fig. 3). Neutrophils were scattered
throughout the renal tissue, adjacent to the pelvic epithelium, and
were seen in the lumen of the renal pelvis. The density of neutrophils in bladder tissue was lower than in the kidneys.
Urine MIP-2 concentrations correlate with neutrophil
recruitment
Urine chemokine concentrations were examined for possible correlations with neutrophil numbers in individual urine samples.
Samples in which the neutrophil numbers and the chemokine concentrations had been determined were included in the analysis.
Positive significant correlations were found for MIP-2 (r 5
0.6183, p , 0.005) but not for KC (r 5 0.1615, p 5 0.4867),
ENA-78 (r 5 0.0803, p 5 0.7365), MIP-1a (r 5 0.2200,
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
C3H/HeN mice were bred in the animal facilities of the Department of
Medical Microbiology, University of Lund (Lund, Sweden). Female mice
were used at 9 –13 wk of age. Mice were anesthetized, and 0.1 ml of an E.
coli 1177 suspension was injected into the bladder through a soft polyethylene catheter (outer diameter 0.61 mm; Clay Adams, Parsippany, NJ)
(11). The catheter was immediately removed, and the mice were allowed
food and water ad libitum.
Animals were sacrificed after 2, 6, and 24 h under anesthesia by cervical
dislocation. Kidneys and bladders were removed and homogenized in sterile disposable plastic bags using a laboratory blender (Stomacher 80 homogenizer, Seward Medical, UAC House, London, U.K.). The homogenates were diluted in sterile PBS, and 0.1 ml of each dilution was plated on
tryptic soy agar. The number of colonies was scored after overnight culture
at 37°C.
Urination was induced by gentle pressure on the mouse abdomen, and
urine was collected at the urethral orifice into sterile tubes. Urine samples
collected before infection were cultured to ensure that the mice were uninfected and were examined for a preexisting neutrophil response. Urine
samples collected at 2, 6, and 24 h postinfection were used for neutrophil
counts and to quantify the local chemokine response.
Histology and immunohistochemistry
The Journal of Immunology
3039
p , 0.5588), JE (r 5 0.0847, p 5 0.7302), or eotaxin (r 5 0.0102,
p 5 0.9660). This suggests that MIP-2 was involved in the infection-induced neutrophil recruitment.
MIP-2 is produced in the kidney and by recruited PMNs
FIGURE 2. Neutrophil response to E. coli infection and inhibition by
RB6-8C5 and anti-MIP-2 Ab. A, A rapid neutrophil response to E. coli
1177 was observed in untreated mice. Numbers represent means 6 SE of
urine samples from 8 –10 mice per group. B, Urine neutrophil numbers
were reduced in mice treated with anti-MIP-2 antiserum (black bars). Preimmune serum (hatched bars) had no inhibition effect compared with the
PBS control. Numbers represent means 6 SE of urine samples from 8 –10
mice per group. #, p , 0.01; ##, p , 0.003; compared with mice treated
with preimmune serum (hatched bars). C, Neutrophil recruitment was completely blocked after pretreatment of the mice with RB6-8C5 Ab. Monoclonal irrelevant control Ab had no inhibitory effect. Numbers represent
means 6 SE of urine samples from 8 –10 mice per group. p, p 5 0.007; pp,
p , 0.001; and ppp, p 5 0.0286; compared with controls receiving irrelevant mAb (hatched bars).
were similar at 2 h in both groups of mice, suggesting that the early
response was independent of neutrophils. At 6 h, the neutrophildepleted mice had significantly lower tissue MIP-2 levels than
control mice, suggesting that the recruited neutrophils accounted
for the increase at this time. At 24 h after infection, similar MIP-2
levels were observed in both neutrophil-depleted and control animals, suggesting that MIP-2 originated from the tissues. MIP-2
concentrations in bladder tissue were low (Fig. 4A).
Inhibition of neutrophil recruitment into the urine following antiMIP-2 Ab treatment
FIGURE 1. Chemokine response to intravesical E. coli infection. Chemokine concentrations in urine at different times following intravesical E.
coli 1177 infection (mean 6 SE, 8 –10 mice per point). C-X-C chemokines
are shown in black, and C-C chemokines in are shown in gray.
Pretreatment of the mice with polyclonal anti-MIP-2 Abs inhibited
infection-induced neutrophil migration into the urine (Fig. 2B).
Intraperitoneal injection with 500 ml of anti-MIP-2 antiserum 2 h
before intravesical infection with E. coli 1177 caused a decrease in
urine neutrophil numbers to background levels after 2 h, and neutrophil numbers remained lower than the control after 6 and 24 h.
Preimmune serum had no effect. These levels confirmed that
MIP-2 was involved in infection-induced neutrophil recruitment.
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
The site of MIP-2 production in the urinary tract was examined
using kidney and bladder homogenates from infected and control
mice (Fig. 4). Kidney MIP-2 levels increased to about 40 ng/ml,
while bladder levels remained below 10 ng/ml. Kidney MIP-2 levels increased before a detectable neutrophil response but the later
kinetics followed the neutrophil influx (Fig. 4A).
The local site of MIP-2 production was further examined by
immunohistochemistry (Fig. 5). Renal pelvic mucosal cells
showed strong intracellular and surface staining for MIP-2 after
infection (Fig. 5b), but no staining before infection (Fig. 5a). There
was no evidence of MIP-2 staining in the bladder epithelium, but
neutrophils on their way through the epithelial barrier stained
brightly (Fig. 5d). MIP-2 was also detected in urine neutrophils
(Fig. 5f).
The relative contribution of the kidney epithelium and the recruited PMNs to MIP-2 production was examined by comparing
neutrophil depleted and control mice. Peripheral neutrophil depletion was achieved by i.p. injection of RB6-8C5 (250 mg/mouse)
with isotope-matched irrelevant Ab (rat IgG2b, 250 mg/mouse) as
a control. Pretreatment with RB6-8C5 Ab at 24 h and 2 h before
intravesical infection completely inhibited the neutrophil influx
into the urine (Fig. 2C). Control Ab had no inhibitory effect.
MIP-2 concentrations in kidneys and bladders of control mice
and RB6-8C5 treated mice are shown in Fig. 4B. MIP-2 levels
3040
MIP-2 REQUIREMENT FOR NEUTROPHIL PASS ACROSS EPITHELIA
Difference in localization of tissue neutrophils between RB68C5- and anti-MIP-2-treated mice
The neutrophil infiltrate into the tissues of RB6-8C5- or antiMIP-2 Ab-treated mice is shown in Fig. 6, A and B, respectively,
using histology and immunohistochemistry. Pretreatment with
RB6-8C5 Ab completely inhibited the neutrophil influx into the
kidney tissue and into the urine (Fig. 6A). Pretreatment with antiMIP-2 Ab did not decrease neutrophil recruitment into the kidneys.
In contrast, the density of neutrophils in the kidneys of anti-MIP2-treated mice was higher than that in mice treated with the preimmune serum. The neutrophils accumulated under the mucosal
barrier, but did not cross the urothelium into the urine (Fig. 6B).
To confirm this difference in neutrophil localization, kidney MPO
was quantitated in control or Ab-treated mice (Fig. 7). Mice pretreated
with RB6-8C5 Ab had low kidney MPO activity at 6 and 24 h, confirming the inhibition of neutrophil recruitment. Anti-MIP-2 Abtreated mice showed no difference in tissue MPO levels compared
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
FIGURE 3. Neutrophil recruitment
to the urinary tract following E. coli infection. Neutrophil response to intravesical infection with E. coli 1177. Tissue
sections were obtained at 0, 6, and 24 h
after infection and stained with hematoxylin and eosin. Neutrophils are indicated by the arrow. Magnification,
3200.
with controls, suggesting that neutrophil recruitment into the kidney
was unchanged. The bladder MPO activity remained at background
levels at 6 and 24 h after infection in all groups.
These results demonstrated that the low urine neutrophil levels
in anti-MIP-2-treated mice were due to blocked transepithelial migration and not to defective neutrophil recruitment from the bloodstream into the tissues.
Discussion
This study demonstrated that specific chemokines direct neutrophil
migration across the epithelial barrier at infected mucosal sites. A
rapid mucosal chemokine response was observed following experimental UTI, and neutrophils were recruited to the urinary tract. MIP-2
showed the strongest correlation with urine neutrophil numbers, suggesting a direct involvement of this chemokine in neutrophil migration. Tissue quantitation demonstrated that kidney MIP-2 production
was triggered by infection, and immunohistochemistry identified the
The Journal of Immunology
3041
FIGURE 4. Tissue sources of MIP-2. A,
MIP-2 concentrations in kidney (–E–) and
bladder (–‚–) homogenates from mice infected
with E. coli 1177. Urine neutrophil numbers
(–M–) are shown for comparison. Numbers are
means 6 SE of 5– 8 mice per time point. B,
MIP-2 levels in urine of mice treated with mAb
RB6-8C5 (–M–) or control Ab (–l–). Numbers are means 6 SE of 5– 8 mice per time
point.
FIGURE 5. Localization of MIP2-producing cells in kidney and
bladder sections. Tissues were obtained from uninfected mice (a) or
from mice sacrificed at 6 h after infection and were stained with preimmune serum (c and e) or anti-MIP-2
Ab (b, d, and f). Strong MIP-2 staining was found in (b) epithelial cells
lining of the renal pelvis and (f)
urine neutrophils. d, There was no
evidence of MIP-2 staining in the
bladder epithelium. The MIP-2
staining was due to neutrophils located at the surface of the bladder.
The arrow indicates the bladder lumen. Sections were counterstained
with hematoxylin. Magnification,
3200 (a, b, c, and d); 3800 (e and f).
the anti-MIP-2-treated mice. In control mice, the neutrophils passed
across the mucosa and into the urine. The results suggest that different
chemokines direct neutrophil migration from the bloodstream into the
kidney or across the epithelium into the urine and that MIP-2 serves
the latter function.
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
renal epithelium as the main source of MIP-2. Treatment of the mice
with anti-MIP-2 Ab caused a drastic reduction in urine neutrophil
counts, but there was no effect on kidney neutrophil numbers or in
kidney MPO activity. Immunohistochemical analysis showed neutrophil accumulation on the renal tissue side of the pelvic epithelium of
3042
MIP-2 REQUIREMENT FOR NEUTROPHIL PASS ACROSS EPITHELIA
Chemokines are commonly assigned to two subgroups depending on the position of the first two cysteines and disulfide bridges.
The C-X-C group contains a single amino acid insert between the
first two cysteines, which are adjacent in the C-C group (25). Several members of both groups have been identified in the mouse;
MIP-2, KC, ENA-78 (C-X-C), JE, MIP-1a, and eotaxin (C-C). In
this study, experimental UTI was shown to trigger a broad mucosal
chemokine response including members of both the C-C and CX-C families. The chemokines showed slightly different response
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
FIGURE 6. Effect of RB6-8C5
or anti-MIP-2 Ab treatment on neutrophil recruitment 6 h into kidney
and bladder tissue. A, Sections from
RB6-8C5 Ab-treated mice were
stained with hematoxylin-eosin (a
and b) or anti-neutrophil Ab (c and
d). a, Neutrophils were absent from
kidneys. b, Neutrophils were absent
from bladders. c, Same section as a
with anti-neutrophil Ab. d, Same
section as b with anti-neutrophil Ab.
Magnification, 3400. B, Sections
from anti-MIP-2 Ab-treated mice
were stained with hematoxylin-eosin (a and b) or anti-neutrophil Ab
(c and d). a, Neutrophils accumulated under the uroepithelial lining
in the kidneys. b, Neutrophil accumulation was not seen in the bladder. c, Same section as a stained
with anti-neutrophil Ab. d, Same
section as b stained with antineutrophil
Ab.
Magnification,
3400.
kinetics, with KC and JE increasing most rapidly and ENA-78
most slowly. These results confirm in vitro studies of epithelial
chemokine responses to different infectious agents. Kagnoff et al.
showed that human colonic epithelial cells constitutively express
IL-8, growth related oncogena (GROa), GROb, GROg, ENA-78,
MCP-1, MCP-1a, MIP-1b, and RANTES (26). Uropathogenic E.
coli triggers the production of IL-8, GROa, GROb, GROg, ENA78, inflammatory protein 10, Mig, monocyte chemoattractant protein-1, RANTES, MIP-1a, and MIP-1b in human kidney cell lines
The Journal of Immunology
(G. Godaly et al., unpublished observations). This broad mucosal
chemokine repertoir in both mice and humans suggests that there
is a need for mucosal chemokine diversity. We propose that this
diverse chemokine repertoir provides a basis for specialized recruitment of diverse cell populations to mucosal sites.
The cellular origin of MIP-2 was examined by quantitation in tissue
homogenates and by immunohistochemistry using the MIP-2 antiserum. MIP-2 levels in kidney homogenates increased by 2 h after
infection and reached a peak after 6 h. Immunohistochemistry showed
MIP-2 staining localized to epithelial cells lining the renal pelvis.
These results are consistent with the localization of IL-8 in human
kidney biopsies. The epithelial lining of the human urinary tract is rich
in IL-8 (24), and human epithelial cells in culture respond to bacterial
stimulation and secrete IL-8 both apically and basolaterally. Apical
secretion may contribute to the increase in urine IL-8 concentrations
during infection. Basolaterally secreted IL-8 is likely to bind the epithelial cells and support neutrophil recruitment. Cell-bound IL-8 of
epithelial origin has been shown to support neutrophil passage across
the epithelial layer (27). This study showed a similar role of epithelial
MIP-2 for neutrophil passage across the epithelium in the mouse urinary tract mucosa.
MIP-2 staining was also detected in the neutrophils that migrated
into renal tissue and into the urine. The relative contribution to the
MIP-2 response of kidney epithelial cells and neutrophils, respectively, was analyzed in mice after depletion of neutrophils by pretreatment with RB6-8C5. The kidneys were also shown to produce
significant amounts of MIP-2 in the depleted mice. Kinetics suggested
that the kidney epithelium was the main source of MIP-2 chemokine
during the early stages of infection, but that peak 6-h levels were
produced mainly by recruited neutrophils. These observations illustrated the sequential nature of the early inflammatory response, with
a first “wave” of mediators produced by the local tissue in response to
infection and a second “wave” following the recruitment and activation of specialized inflammatory cells.
The results in the bladder mucosa differed between human and
mouse. While IL-8 staining is found in human bladder epithelium, no
intracellular epithelial MIP-2 staining was seen in the mouse bladder.
The cells with strong MIP-2 staining in the bladder lumen were neutrophils. Tissue sections did not show large neutrophil numbers at any
time point, and bladder MPO levels were low. After treatment with
anti-MIP-2 Ab, there was no build up of neutrophils under the bladder
epithelium. These observations suggested that most of the neutrophils
in the infected urinary tract were recruited through the kidneys rather
than the bladder or that chemokines other than MIP-2 exert this function in the bladder mucosa.
Important differences were noted between the two Abs used to
inhibit neutrophil migration. Both Abs caused a significant reduction in urine neutrophil numbers, but only the RB6-8C5 Ab
blocked neutrophil migration into the kidneys. Histology revealed
the virtual absence of neutrophils in RB6-8C5-treated mice, and
kidney MPO levels were low. MIP-2 Ab treatment blocked neutrophil migration into the lumen, but had no effect on the neutrophil recruitment into the kidneys. Mice pretreated with anti-MIP-2
Ab had adequate tissue neutrophil levels but kidney MPO-levels
increased after infection to the same levels as in the control mice.
This apparent paradox was explained by immunohistochemistry.
In anti-MIP-2 Ab-treated mice, neutrophils accumulated on the
tissue side of the pelvic epithelium, suggesting that MIP-2 is required to support neutrophil migration across the urothelium into
the urine. This is consistent with in vitro studies showing that IL-8
supports neutrophil migration across E. coli-stimulated human uroepithelial cells layers and that Abs to IL-8 block tissue epithelial
neutrophil migration (28).
Neutrophils are essential effector cells of the antimicrobial host
defense at systemic infection sites (21, 24, 27). Defects that influence the inflammatory response to infection impair bacterial clearance. In a parallel study, we examined the role of the neutrophils
for bacterial clearance from the urinary tract. Mice depleted of
neutrophils with the RB6-8C5 Ab were highly susceptible to infection with 1000-fold more bacteria in the kidneys than control
mice after 24 h. In contrast, anti-MIP-2 Ab-treated mice were able
to clear bacteria from the kidneys. This suggested that the neutrophils that accumulate under the epithelium remain fully functional
as antibacterial effector cells, despite their inability to cross the
epithelial barrier. While these studies clearly demonstrate the importance of neutrophils for host resistance to mucosal infection,
they also provide evidence for complexity and redundancy in the
mechanism of neutrophil recruitment. Our results suggest that neutrophils cross epithelial cell barriers in a highly regulated manner
in response to chemotactic gradients elaborated at this site.
Epithelial cell chemokine production may have evolved to serve
physiologic functions unrelated to the antimicrobial defense. Aging neutrophil cells need to be eliminated from the circulation and
the tissues. If transported to the mucosa in response to a suitable
gradient, they can leave the body with their granular content intact
and be destroyed in the lumen, where the toxic products do little
damage to the host. It may be speculated that the diversity and
apparent hierarchy of mucosal chemokine responses have evolved
to support such functions and not only for the defense against
mucosal pathogens.
Acknowledgments
We thank Drs. A Sjöstedt, W. Conlan, and R. Coffman for their generous
gift of the RB6-8C5 hybridoma.
References
1. Shahin, R. D., I. Engberg, L. Hagberg, and C. Svanborg-Eden. 1987. Neutrophil
recruitment and bacterial clearance correlated with LPS responsiveness in local
gram-negative infection. J. Immunol. 138:3475.
2. Linder, H., I. Engberg, I. M. Baltzer, K. Jann, and C. Svanborg-Edén. 1988.
Induction of inflammation by Escherichia coli on the mucosal level: requirement
for adherence and endotoxin. Infect. Immun. 56:1309.
3. Agace, W., S. Hedges, M. Ceska, and C. Svanborg. 1993. IL-8 and the neutrophil
response to mucosal Gram negative infection. J. Clin. Invest. 92:780.
4. Eckmann, L., M. F. Kagnoff, and J. Fierer. 1993. Epithelial cells secrete the
chemokine interleukin-8 in response to bacterial entry. Infect. Immun. 61:4569.
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
FIGURE 7. MPO activity in renal tissue. Kidney MPO activity was
quantitated at 6 and 24 h after intravesical infection with E. coli 1177. Mice
were pretreated with mAb RB6-8C5 (white bars), anti-MIP-2 antiserum
(black bars), or pyrogen-free saline (gray bars) i.p. as described. Numbers
denote means 6 SE of tissues from four mice per time point. p, p 5
0.0281; pp, p , 0.0286.
3043
3044
MIP-2 REQUIREMENT FOR NEUTROPHIL PASS ACROSS EPITHELIA
17. Strieter, R. M., S. L. Kunkel, M. D. Burdick, P. M. Lincoln, and A. Walz. 1992.
The detection of a novel neutrophil-activating peptide (ENA-78) using a sensitive
ELISA. Immunol. Invest. 21:589.
18. Colletti, L. M., S. L. Kunkel, A. Walz, M. D. Burdick, R. G. Kunkel, C. A. Wilke,
and R. M. Strieter. 1995. Chemokine expression during hepatic ischemia/reperfusion-induced lung injury in the rat: the role of epithelial neutrophil activating
protein. J. Clin. Invest. 95:134.
19. Marc, J. G., R. M. Strieter, S. L. Kunkel, J. M. Danforth, L. L. Laichalk,
D. C. McGillicuddy, and T. J. Standiford. 1996. Neutralization of macrophage
protein-2 attenuates neutrophil recruitment and bacterial clearance in murine
Klebsiella pneumonia. J. Infect. Dis. 173:159.
20. Kasama, T., R. M. Strieter, N. W. Lukacs, P. M. Lincoln, M. D. Burdick, and
S. L. Kunkel. 1995. Interleukin-10 expression and chemokine regulation during
the evolution of murine type II collagen-induced arthritis. J. Clin. Invest. 6:2868.
21. Rogers, H., and E. Unanue. 1993. Neutrophils are involved in acute, nonspecific
resistant to Listeria monocytogenes in mice. Infect. Immun. 61:5090.
22. Conlan, J., and R. North. 1994. Neutrophils are essential for early anti-Listeria
defense in the liver, but not in the spleen or peritoneal cavity, as revealed by a
granulocyte-depleting monoclonal antibody. J. Exp. Med. 179:259.
23. Tepper, R. I., R. L. Coffman, and P. Leder. 1992. An eosinophil-dependent mechanism for the antitumor effect of interleukin-4. Science 257:548.
24. Hang, L., B. Wullt, Z. X. Shen, D. Karpman, and C. Svanborg. 1998. Cytokine
repertoire of epithelial cells lining the human urinary tract. J. Urol. 159:2185.
25. Conlan, J. W. 1996. Neutrophils prevent extracellular colonization of liver microvasculature by Salmonella typhimurium. Infect. Immun. 64:1043.
26. Rowell, D. L., L. Eckmann, M. B. Dwinell, S. P. Carpenter, J. L. Raucy,
S. K. Yang, and M. F. Kagnoff. 1997. Human hepatocytes express an array of
proinflammatory cytokines after agonist stimulation or bacterial invasion.
Am. J. Physiol. 273:322.
27. Conlan, J. W., and R. J. North. 1994. Neutrophils are critical for host defense
against primary infection with the facultative intracellular bacterium Francisella
tularensis in mice and participate in defense against reinfection. Infect. Immun.
62:2779.
28. Godaly, G., A. E. I. Proudfoot, R. E. Offord, C. Svanborg, and W. Agace. 1997.
Role of epithelial interleukin-8 and neutrophil IL-8 receptor A in Escherichia
coli-induced transuroepithelial neutrophil migration. Infect. Immun. 65:3451.
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
5. Jung, H., L. Eckmann, S. Yang, A. Panja, J. Fierer, E. Wroblewska, and
M. Kagnoff. 1995. A distinct array of proinflammatory cytokines is expressed in
human colon epithelial cells in response to bacterial invasion. J. Clin. Invest.
95:55.
6. Parkos, C., C. Delp, M. Armin Arnout, and J. Madara. 1991. Neutrophil migration across cultured intestinal epithelium. J. Clin. Invest. 88:1605.
7. Ko, Y. C., N. Mukaida, S. Ishiyama, A. Tokue, T. Kawai, K. Matsushima, and
T. Kasahara. 1993. Elevated interleukin-8 levels in the urine of patients with
urinary tract infections. Infect. Immun. 61:1307.
8. Ko, Y., and T. Kawai. 1995. (Interleukin-8) (Abstract). Jpn. J. Clin. Pathol.
43:329 (Abstr.).
9. Mårild, S., U. Jodal, I. Ørskov, F. Ørskov, and C. Svanborg-Edén. 1989. Special
virulence of the Escherichia coli O1:K1:H7 clone in acute pyelonephritis. J. Pediatr. 115:40.
10. Connell, H., W. Agace, P. Klemm, M. Schembri, S. Mårild, and C. Svanborg.
1996. Type 1 fimbrial adhesion enhances Escherichia coli virulence for the urinary tract. Proc. Natl. Acad. Sci. USA 93:9827.
11. Hagberg, L., I. Engberg, R. Freter, J. Lam, S. Olling, and C. Svanborg-Edén.
1983. Ascending, unobstructed urinary tract infection in mice caused by pyelonephritogenic Escherichia coli of human origin. Infect. Immun. 40:273.
12. Wolpe, S. D., B. Sherry, D. Juers, G. Davatelis, R. W. Yurt, and A. Cerami. 1989.
Identification and characterization of macrophage inflammatory protein-2. Proc.
Natl. Acad. Sci. USA 86:612.
13. Bozic, C. R., L. F. Kolakowski, N. P. Gerard, C. Garcia-Rodrigues, and
C. von Uxekull-Guldenband. 1995. Expression and biologic characterization of
the murine chemokine KC. J. Immunol. 154:6048.
14. Kasama, T., R. M. Strieter, T. J. Standiford, M. D. Burdick, and S. L. Kunkel.
1993. Expression and regulation of human neutrophil-derived macrophage inflammatory protein-1a. J. Exp. Med. 178:63.
15. Rollins, R. J., E. D. Morrision, and C. D. Stiles. 1988. Cloning and expression of
JE, a gene inducible by platelet-derived growth factor and whose product has
cytokine-like properties. Proc. Natl. Acad. Sci. USA 85:3738.
16. Ganzalo, J. A., G. Q. Jia, V. Aguirre, D. Friend, A. J. Coyle, N. A. Jenkins,
G. S. Lin, H. Katz, A. Lichtman, N. Copeland, M. Kopf, and
J. C. Gutierrez-Ramos. 1996. Mouse eotaxin expression parallels eosinophil accumulation during lung allergic inflammation but it is not restricted to a Th2-type
response. Immunity 4:1.