Am J Physiol Renal Physiol 299: F479–F486, 2010.
First published June 30, 2010; doi:10.1152/ajprenal.00585.2009.
Intrarenal urothelium proliferation: an unexpected early event following
ischemic injury
C. Vinsonneau,1,2 A. Girshovich,1 M. Ben M’rad,1 J. Perez,1 L. Mesnard,1,3 S. Vandermersch,1 S. Placier,1
E. Letavernier,1,3 L. Baud,1,3 and J.-P. Haymann1,3
1
INSERM UMRS 702 and Université Pierre et Marie Curie Paris 6, 2Hopital Cochin Saint Vincent de Paul, and 3Hopital
Tenon, Assistance Publique-Hôpitaux de Paris, Paris, France
Submitted 13 October 2009; accepted in final form 27 June 2010
FGFR2; ischemia
RENAL ISCHEMIA IS ONE OF THE leading causes of acute kidney
injury (AKI) (13). AKI results in endothelial and tubular cell
injury, which is a direct consequence of metabolic pathways
activated by ischemia and amplified by inflammation (19).
Tubular cell lesions include initially the loss of brush border
and the disruption of cell polarity, and later on, during the
extension phase, the death of cells by both apoptosis and
necrosis (3, 28). The maintenance phase is characterized by a
balance between tubular cell death and regeneration.
During the repair/regeneration process, the proliferation response may occur as soon as 18 h after ischemic insult and is
reported to be mostly localized in the outer medulla (18). The
origin of proliferating cells is not clearly identified. They could
be bone marrow-derived stem cells that migrate into the injured
kidney and differentiate into mature cells, intrarenal stem cells
Address for reprint requests and other correspondence: J.-P. Haymann,
UPMC Paris 6, INSERM UMR S-702, Hôpital Tenon, 4 rue de la Chine,
75020 Paris , France (e-mail: [email protected]).
http://www.ajprenal.org
that move to the site of repair, and/or surviving tubular cells
that dedifferentiate, proliferate, and eventually differentiate
again (8). Several studies support the role of extrarenal stem
cells which may home on the damaged kidney and engraft onto
tubular epithelium, where they undergo a differentiation process (11, 15, 20). However, the direct contribution of extrarenal
stem cells has been recently challenged, considering the low
frequency of engraftment events (6, 16). Thus bone marrowderived stem cells, especially mesenchymal stem cells, would
participate in renal repair rather through a local paracrine
effect, mainly the production of growth factors (14, 21, 30).
Candidate kidney-resident stem cells have been identified based
upon several characteristics: low cycling, expression of stem cell
markers, and extrusion of Hoechst dye (so-called side population).
They were located in various niches, including renal papilla (23),
glomerular capsule (26), tubule (9, 16), and interstitium (7).
However, whether these cells participate in the repair process,
either directly or through the control of tubule regeneration,
remains to be determined. Finally, several studies support the
view that surviving proximal tubular cells are the main/unique
source of epithelium renewal (6, 9). Thus the origin of cells
proliferating in the kidney early after an ischemic insult is still
undefined and different cellular sources may be at play.
In an attempt to identify the earliest dividing cells following
ischemia-reperfusion injury (IRI), we performed systematic
cutting (with sections along the axial plane) of whole adult
mice kidneys at different time points to establish a precise
cartography. This analysis allowed us to locate and track with
good accuracy and reproducibility the earliest proliferating
cells using 5-deoxyuridine (BrdU) labeling. We observed early
proliferating cells in clusters at the corticomedullary junction
and identified them as renal urothelial cells. Their proliferation
involved the urothelial cell expression of FGF receptor-2
(FGFR2) and the paracrine action of FGF7. These signals
preceded the subsequent proliferation of tubular cells in their
immediate vicinity.
MATERIALS AND METHODS
In vivo experiments. All in vivo experiments were performed with
adult female C57BL/6 mice that weighed 20 –25 g. Animal handling
was performed according to the recommendations of the French
ethical committee and under the supervision of authorized investigators. The experimental protocol in this study involving animal use was
approved by the local ethical committee.
For ischemia-reperfusion (I/R) experiments, surgical procedures
were conducted under general anesthesia (250 mg/kg body wt, Avertine, Sigma), a posterior subcostal incision was made in the left side,
and the renal pedicle was dissected and occluded with a small vascular
clamp. After 45 min, the clamp was removed. Animals were killed at
different time points to collect the kidneys.
0363-6127/10 Copyright © 2010 the American Physiological Society
F479
Downloaded from http://ajprenal.physiology.org/ by 10.220.33.4 on June 17, 2017
Vinsonneau C, Girshovich A, Ben M’Rad M, Perez J, Mesnard L,
Vandermersch S, Placier S, Letavernier E, Baud L, Haymann JP.
Intrarenal urothelium proliferation: an unexpected early event following
ischemic injury. Am J Physiol Renal Physiol 299: F479–F486, 2010. First
published June 30, 2010; doi:10.1152/ajprenal.00585.2009.—Identification
of renal cell progenitors and recognition of the events contributing to
cell regeneration following ischemia-reperfusion injury (IRI) are a
major challenge. In a mouse model of unilateral renal IRI, we
demonstrated that the first cells to proliferate within injured kidneys
were urothelial cells expressing the progenitor cell marker cytokeratin
14. A systematic cutting of the injured kidney revealed that these
urothelial cells were located in the deep cortex at the corticomedullary
junction in the vicinity of lobar vessels. Contrary to multilayered
bladder urothelium, these intrarenal urothelial cells located in the
upper part of the medulla constitute a monolayered barrier and express
among uroplakins only uroplakin III. However, like bladder progenitors, intrarenal urothelial cells proliferated through a FGF receptor-2
(FGFR2)-mediated process. They strongly expressed FGFR2 and
proliferated in vivo after recombinant FGF7 administration to control
mice. In addition, IRI led to FGFR phosphorylation together with the
selective upregulation of FGF7 and FGF2. Conversely, by day 2
following IRI, renal urothelial cell proliferation was significantly
inhibited by FGFR2 antisense oligonucleotide administration into an
intrarenal urinary space. Of notice, no significant migration of these
early dividing urothelial cells was detected in the cortex within 7 days
following IRI. Thus our data show that following IRI, proliferation of
urothelial cells is mediated by the FGFR2 pathway and precedes
tubular cell proliferation, indicating a particular sensitivity of this
structure to changes caused by the ischemic process.
F480
UROTHELIUM PROLIFERATION AFTER ISCHEMIC INJURY
AJP-Renal Physiol • VOL
RESULTS
Distribution of proliferating cells in the mouse kidney early
after IRI. To study early proliferating cells following IRI, we
performed a systematic sagittal cutting of mouse kidneys, from
the convexity to the hilus as shown in Fig. 1, A–C, and two
different techniques, i.e., Ki67 or BrdU staining.
First, using Ki67 staining on a frozen mouse kidney section
(level corresponding to area 1) 18 h after IRI, we detected with
high reproducibility “long chained proliferating cells” with
very rare Ki67-positive cells outside these structures located
mostly in the interstitium (Fig. 1, D and E). As shown, these
proliferating structures were located in the vicinity of lobar
vessels with a peak of proliferation by 24 h (46 ⫾ 8.9%) and
were no longer positive for Ki67 by 60 h following IRI,
whereas some tubular cells in the vicinity were still proliferating (Fig. 1F).
Second, we injected BrdU (ip) at different time points,
euthanized the animals by day 7, and then investigated BrdUpositive cells according to our sectioning. As shown in Fig. 1G,
most of the BrdU-tagged cells detected within whole IRI
kidneys were located in a specific area (area 1 in Fig. 1A)
forming structures similar to previous long chained proliferating cells when BrdU was injected by 16 h. Very few BrdUpositive cells were detected in the cortex, suggesting that these
early proliferating cells were not significantly migrating during
the repairing process and were not accounting significantly for
tubular repair. Moreover, when BrdU was injected by 22 h, the
pattern of BrdU-tagged cells was strikingly different, showing
a massive tubular proliferation in the corticomedullary area
together with the persistence of BrdU-positive cells in the same
structures described previously (Fig. 1H). No significant BrdUtagged cells were detected when BrdU was injected before 16
h after IRI, and only very rare Ki67-positive interstitial cells
could be identified within the first 16 h (following IRI) in all
kidney sections (data not shown).
As assessed by Ki67 staining of paraffin-embedded tissues
by 18 h, proliferating cells were located in area 1 at the
junction of the cortex and the medulla underlined by megalin
and uromodulin staining, respectively. As shown, they were
located in the vicinity of lobar vessels around a urinary space
(Fig. 2, A and C). A higher magnification of this area (Fig. 2,
B and D) shows precisely that Ki67-positive nuclei are lining
the urinary space and do not express megalin or uromodulin.
Very few Ki67-positive cells were detected outside this location, mostly in the interstitium, and virtually no positive tubular
cells were seen. Urothelial Ki67-positive cells were also detected in lower sections, i.e., down to the hilus, on the cortical
side, with rare proliferating cells located in the urothelium
covering the papilla (data not shown). Rare interstitial Ki67positive cells were detected in contralateral or control kidneys,
with no Ki67-positive urothelial cells (data not shown).
Characterization of early proliferating cells. Most renal
urothelial cells stained positive for cytokeratin 14 (Fig. 3, A–C)
within area 1 located within the fornix, i.e., the folding of the
parietal (cortical) and visceral (papillary) urothelium, which is
located at the junction of the cortex and the medulla (Fig. 3, A
and B). Of interest, we described at least two different urothelial cell types: cytokeratin 14-positive cells, located mostly on
the cortical side of the urinary space, and cytokeratin 14negative cells, covering rather the medullary/papillary side
299 • SEPTEMBER 2010 •
www.ajprenal.org
Downloaded from http://ajprenal.physiology.org/ by 10.220.33.4 on June 17, 2017
To explore the role of FGF7 in the proliferative response, we used
a daily intraperitoneal injection of palifermin (60 ␮g/kg, Amgen,
Neuilly-sur-Seine, France) in healthy mice that were harvested at
different time points.
To block FGFR2 expression, we used a specific antisense oligonucleotide phosphorothioate including 23 bases (TGTTTGGGCAGGACAGTGAGCCA) that was designed to hybridize with the 3=region of the FGFR2 mRNA promoter as previously described by
Villanueva et al. (30). The antisense or scrambled oligonucleotide
(GGCTAGACGTCGAGTGCGTAGAT) used as a control was diluted in 0.9% saline sodium chloride and injected (112 ␮g/kg) in
kidney parenchyme during IRI, just before clamp removal (30).
To evaluate the EGFR pathway, we used an antagonist of the EGFR
(erlotinib, 2 ␮g/g, Roche, Neuilly-sur-Seine, France) 1 day before I/R
procedures with enteral administration on a once daily basis.
To track proliferative cells, mice were injected intraperitoneally
with BrdU (100 mg/kg, Sigma) following different schedules. Animals were harvested at different time points, and kidneys were snap
frozen in liquid nitrogen or paraffin embedded.
Immunodetection. Kidneys were isolated and cut in a sagittal axis
from the convexity to the pedicle. For snap-frozen kidneys, 4-␮m
sections were fixed in 4% paraformaldehyde for 8 min, then washed
in PBS. For paraffin-embedded kidneys, sections were dewaxed and
then incubated with DakoCytomation Target Retrieval Solution
(Dako).
BrdU staining was performed using a monoclonal FITC-labeled
anti-BrdU antibody (RD Systems). The following antibodies were
also used against megalin (kindly provided by Dr. P. Verroust,
INSERM, Paris, France), Ki67 (Abcam), pancytokeratin (SigmaAldrich), cytokeratin 14 (Covance), uromodulin, UPI, UP II, UP III,
FGFR2 (bek), phosphorylated FGFR, FGF7, and FGF 10 (Santa Cruz
Biotechnology), and ␣-smooth muscle actin (Abcam). All secondary
antibodies were ready-to-use peroxydase-labeled antibodies and came
from Histofine (Tokyo, Japan). The staining was revealed by an
avidin-biotin coupling immunoperoxidase technique using a DAB
chromogen (Dako). Quantification analysis of proliferating cells was
performed using a double blind reading under a light microscope
(⫻400). Results were expressed as the percentage of Ki67-positive
urothelial nuclei (mean value from 5 different kidneys and 9 selected
fields/kidney).
Real-time PCR analysis. Total RNA was extracted from kidney using
TRIzol solution (GIBCO BRL). By using reverse transcriptase, cDNA
was obtained and amplified in a thermocycler (ABI Prism 7000) as
follows: 50°C for 2 min, 95°C for 10 min followed by 40 cycles at 95°C
for 45 s and 60°C for 1 min, using a Quantitect Probe PCR Kit (Qiagen),
and TaqMan Gene expression assays (Applied Biosystems): TaqMan
MGB probes, FAM dye-labeled of FGFR2 (Mm01269930_m1), FGF1
(Mm01258325_m1), FGF2 (NM_008006.2), FGF7 (Mm00627025_g1),
FGF10 (Mm01297079_m1), FGFR2IIIb (Mm_010207.2), FGFR2IIIc
(Mm_201601.2), and ␤-actin (Mm02619580_g1) for normalization. Normal kidney cDNA was used as a reference to establish calibration curves
of genes of interest (GOI) and ␤-actin cycle threshold (CT). Results are
expressed as the 2-⌬⌬CT GOI/␤-actin.
Western blot analysis. Protein extraction was performed by using
1-h centrifugation in RIPA buffer and submission to electrophoresis
on a 7.5% polyacrylamide SDS gel. The proteins were then transferred to a nitrocellulose membrane (Immobilon-p, Millipore). A
nonspecific antibody reaction was blocked by incubating the membrane in PBS-Tween solution with milk powder (10%). Detection of
FGF7, FGF10, FGFR2, and phosphorylated FGFR2 was performed
with specific antibodies and peroxidase-labeled secondary antibodies
(Histofine). The membranes were developed with ECL detection
reagent (Amersham Pharmacia Biotech).
Statistical analysis. The differences between means were compared
by Student’s t-test, with P ⬍ 0.05 considered significant.
UROTHELIUM PROLIFERATION AFTER ISCHEMIC INJURY
F481
(Fig. 3C). As shown in Fig. 4B (area 1), all early proliferating
urothelial cells (Ki67 positive) stained for cytokeratin 14,
suggesting that only this contingent would be involved in the
proliferating process. As a matter of fact, this phenotype
characterized also bladder basal urothelial progenitor cells
(Fig. 3D), although renal urothelial cells did not appear as a
multilayer but rather as a monolayer or sometimes a bilayer
(Figs. 3C and 4, A and C). Furthermore, as shown in Fig. 4, A
and C, some of these cytokeratin 14-positive cells also expressed bright staining for a differentiated urothelial marker
Fig. 2. Localization of early proliferating cells.
Representative microphotographs of paraffinembedded mouse kidney sagittal section corresponding to area 1 stained for Ki67 (brown) and
megalin (A and C) or uromodulin (B and D;
blue). The asterisk denotes urinary space. The
cortex is located at the top of the picture, and the
inner medulla at the bottom. Ki67-positive nuclei are lining the urinary space (C and D).
Magnification ⫻200 (A and B) and ⫻600 (C
and D).
AJP-Renal Physiol • VOL
299 • SEPTEMBER 2010 •
www.ajprenal.org
Downloaded from http://ajprenal.physiology.org/ by 10.220.33.4 on June 17, 2017
Fig. 1. Localization of early proliferating cells in the kidney following ischemia-reperfusion injury (IRI). Representative microphotographs are shown of normal
mouse kidney sagittal sections (hematoxylin and eosin staining) at 3 different levels from the deep cortex (A) down to the papilla (B and C). Area 1 represents
the fornix area. The asterisk denotes urinary space. Also shown are representative microphotographs of mouse frozen kidney sections collected at 18 (D and E)
or 60 h (F) after IRI corresponding to area 1, stained for Ki67. G and H: representative microphotographs of mouse frozen kidney sections stained for
5-deoxyuridine (BrdU; green). Mice were injected with BrdU (ip) respectively 16 and 22 h after IRI and euthanized at day 7. a, Artery; v, vein.
F482
UROTHELIUM PROLIFERATION AFTER ISCHEMIC INJURY
such as uroplakin 3 but did not stain for uroplakin 1 and 2 (data
not shown). Of notice, both cytokeratin 14 and uroplakin 3
staining disappeared in urothelial cells covering the papilla
down the tip (data not shown). As confirmed by scanning
electron microscopy (Fig. 4D), these “cortical” renal urothelial
cells have a polygonal shape and form a monolayered barrier,
most of the time. Moreover, like bladder urothelium, these
cells are located near ␣-smooth muscle actin (␣-SMA) cells from
the hilus to the fornix but only on the cortical side (Fig. 5, A–D).
Altogether, our results show that the early cells undergoing
proliferation after I/R are intrarenal urothelial cells that express
cytokeratin 14, a marker of bladder urothelial progenitors, and
are located mostly near the fornix at the corticomedullary
junction, near the vascular axis, in contact with ␣-SMApositive cells.
Identification of the signaling pathway involved in proliferation of intrarenal urothelial cells. Among the signaling pathways involved in renal urothelial cell growth, the FGFR2
pathway was a potential candidate, as previously shown in
bladder urothelial progenitor proliferation (34). Immunochem-
Fig. 4. Localization and characterization of
early proliferating cells. Shown are representative microphotographs of paraffin-embedded mouse kidney sagittal sections collected
at 24 h corresponding to area 1 stained for
Ki67 (brown) and uroplakin 3 (blue; A) or
cytokeratin 14 (brown) and uroplakin 3
(blue; C). Magnification ⫻600. B: fluorescent double staining of a frozen mouse kidney section (level corresponding to area 1)
18 h after IRI using anti-cytokeratin 14 (red)
and Ki67 antibodies (green). D: scanning
electron microscopy of monolayered urothelial cells located in the upper medulla (area
2). The asterisk denotes urinary space. v,
Vein.
AJP-Renal Physiol • VOL
299 • SEPTEMBER 2010 •
www.ajprenal.org
Downloaded from http://ajprenal.physiology.org/ by 10.220.33.4 on June 17, 2017
Fig. 3. Localization and characterization of
renal urothelial cells. A and B: representative
microphotographs of a normal paraffin-embedded mouse kidney sagittal sections corresponding to area 1 using anti-cytokeratin 14
(brown) and, respectively, anti-megalin (C)
or -uromodulin (D) antibodies (blue). Cytokeratin 14 staining is seen in most monolayered urothelial cells and basal cells (C). Cytokeratin 14-negative cells are located mostly
on visceral urothelium as shown (A–C). Magnification ⫻600. D: normal paraffin-embedded mouse bladder stained for cytokeratin 14
alone. Magnification ⫻600.
UROTHELIUM PROLIFERATION AFTER ISCHEMIC INJURY
F483
istry of a normal mouse kidney revealed that FGFR2 was
expressed strongly in urothelial cells (Fig. 6A) and was scattered in rare tubule cells throughout the outer medulla. Following IRI, FGFR2 expression was decreased in urothelial
cells (data not shown), suggesting its internalization and degradation after FGFR2 ligand binding (2) with no significant
increase in the two FGFR2 mRNA variants (FGFR2IIIb and
IIIc) (Fig. 6, C and D). Consistent with FGFR2 engagement,
the level of phosphorylated FGFR increased markedly from
baseline within the first 2 days after IRI (Fig. 6B). Moreover,
injection of specific FGFR2 antisense oligodeoxynucleotides
into the kidney urinary space at the time of IRI significantly
inhibited FGFR phosphorylation at day 2 after I/R compared
with control animals receiving a scrambled probe (Fig. 6B).
Furthermore, as illustrated Fig. 7, A–C, urothelial cell proliferation was significantly inhibited at day 2 when the animals
were receiving antisense compared with scrambled probe (14.9
vs. 26.6%, P ⫽ 0.007).
Fig. 6. Regulation of FGF receptor-2 (FGFR2)
in mouse kidney following IRI. A: representative microphotographs of a frozen normal
mouse kidney sagittal section corresponding to
area 1 stained for FGFR2 (brown). Magnification ⫻400. The asterisk denotes urinary space.
a, Artery; v, vein. B: phosphorylation of FGFR
by Western blot analysis of kidney extracts
obtained at days 0, 1, and 2 following IRI as
indicated (n ⫽ 2 animals for each time point)
and at day 2 following injection of scrambled
(Sc) or antisense probe (AS) in renal urinary
space at the time of IRI (n ⫽ 3 animals for each
time point). Western blotting for actin was used
as a control. C–F: quantitative RT-PCR from
kidney cDNA obtained at days 0, 1, and 2
following IRI (n ⫽ 6 for each time point) for
FGFR2IIIb (C), FGFR2IIIc (D), FGF7 (E), and
FGF2 (F).
AJP-Renal Physiol • VOL
299 • SEPTEMBER 2010 •
www.ajprenal.org
Downloaded from http://ajprenal.physiology.org/ by 10.220.33.4 on June 17, 2017
Fig. 5. Localization of smooth muscle actinpositive cells in the vicinity of urothelial cells.
Shown are representative microphotographs
of paraffin-embedded normal mouse kidney
sagittal sections stained for ␣-smooth muscle
actin antibodies (brown) corresponding to
area 3 (A and B), area 2 (C), or area 1 (D).
Magnification ⫻100 (A), ⫻600 (B), ⫻200 (C
and D). The asterisk denotes urinary space. a,
Artery; v, vein.
F484
UROTHELIUM PROLIFERATION AFTER ISCHEMIC INJURY
As previously shown (32), FGF2 mRNA was upregulated
(Fig. 6F); however, FGFR2IIIb (known to be present on
bladder urothelial cells) does not efficiently bind to FGF2 (24).
Among the three known ligands of FGFR2-IIIb, only FGF7
RNA was enhanced in kidney extracts as soon as day 1 (Fig.
6E), whereas FGF1 and FGF10 mRNA were unaffected (data
not shown).
Moreover, administration in control mice of a single dose of
FGF7 (palifermin) induced a proliferation of intrarenal urothelial cells, mimicking the regeneration process occurring at day
1 after IRI (Fig. 7D). Finally, to identify the importance of the
EGFR signaling pathway, mice were given an EGFR tyrosine
kinase inhibitor (erlotinib) at the time of IRI. Under these
conditions, urothelial cell proliferation was completely preserved after IRI (Fig. 7E). Moreover, when control mice were
given a single dose of EGF, no proliferation of intrarenal
urothelial cells was detected (data not shown). In sum, proliferation of intrarenal urothelial cells early after IRI requires the
engagement of the FGFR2 pathway, most likely through FGF7
paracrine secretion and binding, while independent of the
EGFR signaling pathway.
DISCUSSION
We show here that the first cells to proliferate following
renal ischemia were a contingent of urothelial cells mostly
located in the corticomedullary area. Whereas the urothelium is
directly contiguous to the cortex at the base of the kidney (near
the hilus), the fornix is located at the corticomedullary junction
in close contact with the vascular axis (as shown in Fig. 1). Our
data point out the fornix (area 1 in Fig. 1A) as the site where
the urothelial cell proliferation index is the highest, with a
marked decrease in the proliferation down the hilus and no
AJP-Renal Physiol • VOL
detectable proliferation in the urothelial papilla. Of notice, the
urothelial cells located near the fornix are easily identified in
paraffin-embedded tissues (Fig. 3, C and D), whereas without
one’s skilled experience they are difficult to detect on frozen
sections using our sectioning and impossible to detect when a
random sectioning is performed. However, when these urothelial cells divide they appear as “long chain-like structures”
lining an open urinary space in paraffin-embedded tissues and
a virtual urinary space in frozen sections by virtue of tissue
processing. Renal urothelial cells appear not to be the labelretaining cells recently described by Oliver et al. (22, 23), as
they do not originate from the papilla. Of notice, these proliferating urothelial structures were not previously reported following IRI despite similar BrdU pulse procedures (17, 18),
probably because of the narrow time window of proliferation
with no cell proliferation detected before 16 h, and a peak
occurring only between 16 and 22 h following IRI. Indeed,
after a 2-h BrdU pulse, the number of urothelial BrdU-positive
cells is higher at 18 than at 24 h following IRI (Fig. 1, D and
E), suggesting that a renal urothelial proliferation peak occurs
within this time window with no longer significant proliferation by day 3 (we also ruled out a dye dilution bias by using
different reading frames, data not shown).
The reason urothelial cells proliferate shortly after renal IRI
is not known (no urothelial proliferation was detected in
control kidneys, and the proliferation index in normal kidneys
is very low; data not shown). There was no evidence of cell
death by apoptosis (data not shown). However, urothelium
which forms a physical barrier with the environment, as epithelium lining the gastrointestinal tract or the airway epithelium, is able to sense changes in its environment and to actively
respond to these changes. Thus it is likely that renal urothelial
299 • SEPTEMBER 2010 •
www.ajprenal.org
Downloaded from http://ajprenal.physiology.org/ by 10.220.33.4 on June 17, 2017
Fig. 7. Commitment of FGFR2 and EGFR pathways in the proliferation of urothelial cells following IRI. A and B: representative microphotographs of Ki67
staining at day 2 (frozen kidney sagittal sections corresponding to area 1) when a scrambled probe (A) or FGFR2 antisense probe (B) was injected in the renal
urinary space at the time of IRI. C: Ki67-positive urothelial nuclei count expressed as a percentage of total urothelial cells. The average of 5 determinations was
calculated (n ⫽ 5 mice for each time point). D: immunofluorescence staining of a mouse kidney 48 h after administration (ip) of FGF7 using anti-BrdU antibody
(administration of BrdU in drinking water from day 0). E: immunofluorescence staining of a kidney from a mouse 24 h after IRI and erlotinib treatment, using
anti BrdU antibody (administration of BrdU ip by 22 h). The asterisk denotes urothelial structures. a, Artery; v, vein.
UROTHELIUM PROLIFERATION AFTER ISCHEMIC INJURY
AJP-Renal Physiol • VOL
tion of FGFR pathways in the proliferative response following
IRI is in agreement with previous studies (10, 30, 31). Indeed,
Villanueva et al. (30) have shown that FGFR2 antisense
oligonucleotide injection shortly after IRI decreases proliferation and worsens renal function, thus suggesting that tubule
repair is FGFR2 dependent. Our findings using an FGFR2
antisense probe support this view and show that FGFR2 activation is critical for the subsequent phosphorylation of all FGF
receptors (Fig. 6B) and also for the proliferation of urothelial
cells together with some tubules (Fig. 7, A and B). As a matter
of fact, FGFR2 activation was reported to be a critical event
during kidney organogenesis, as shown in FGF7 and FGFR2
knockout mice (1, 35). Hence, the issue raised is whether
FGFR2-positive urothelial and (rare) tubular cells following
IRI would trigger the proliferation of a second wave of
FGFR2-negative tubular cells located in their vicinity, thus
mimicking the ureteral bud, which triggers the metanephric
blastema proliferative response during embryogenesis. Ongoing study aim at the selective deletion of urothelial cells and the
identification of growth factors secreted by urothelial cells
early after IRI could give some clues.
To conclude, our data lay stress upon the topological links
between the corticomedullary junction and urothelial structures
in mice. Intrarenal urothelium proliferation is an unexpected
early event following IRI and raises the issue of its precise role
during kidney repair besides barrier self-renewal.
ACKNOWLEDGMENTS
We thank Dr. Isabel le Disquet (Jussieu, IFR 83, Paris) for the scanning
electron microscopy data, Chantal Jouanneau (INSERM UMRS 702) for great
help in tissue processing, and Dr. Jean-Pierre Levraud (Institut Pasteur, Paris)
for critical comments.
GRANTS
This work was supported by INSERM UMRS 702 and Université Pierre et
Marie Curie Paris 6.
DISCLOSURES
No conflicts of interest, financial or otherwise, are declared by the authors.
REFERENCES
1. Bates CM. Role of fibroblast growth factor receptor signaling in kidney
development. Pediatr Nephrol 22: 343–9, 2007.
2. Belleudi F, Leone L, Nobili V, Raffa S, Francescangeli F, Maggio M,
Morrone S, Marchese C, Torrisi MR. Keratinocyte growth factor
receptor ligands target the receptor to different intracellular pathways.
Traffic 8: 1854 –1872, 2007.
3. Bonventre JV, Weinberg JM. Recent advances in the pathophysiology of
ischemic acute renal failure. J Am Soc Nephrol 14: 2199 –2210, 2003.
4. Celli G, Larochelle WJ, Mackem S, Sharp R, Merlino G. Soluble
dominant-negative receptors uncover essential roles for fibroblast growth
factors in multiorgan induction, and patterning. EMBO J 17: 1642–1655,
1998.
5. Dixon JS, Gosling JA. The musculature of the human renal calices,
pelvis, and upper ureter. J Anat 135: 129 –137, 1982.
6. Duffield JS, Park KM, Hsiao LH, Kelley VR, Scadden DT, Ichimura
T, Bonventre JV. Restoration of tubular epithelial cells during repair of
the post ischemic kidney occurs independently of bone-marrow derived
stem cells. J Clin Invest 115: 1743–1755, 2005.
7. Ghielli M, Verstrepen W, Nouwen E, De Broe ME. Regeneration
processes in the kidney after acute injury: role of infiltrating cells. Exp
Nephrol 6: 502–507, 1998.
8. Humphreys BD, Bonventre JV. The contribution of adult stem cells to
renal repair. Néphrol Thér 3: 3–10, 2007.
299 • SEPTEMBER 2010 •
www.ajprenal.org
Downloaded from http://ajprenal.physiology.org/ by 10.220.33.4 on June 17, 2017
cells express pattern-recognizing receptors that bind endogenous molecules released by ischemic tissue, including high
mobility box group 1 and heat shock proteins (27). Testing this
hypothesis would require additional studies.
Urothelial cells express cytokeratin 14, a known marker of
epithelial progenitors (33), and thus share striking similarities
with bladder urothelial progenitors (12). Of notice, they have a
low proliferation index in physiological conditions and smooth
muscle cells are located in close contact, as previously shown
(5). In the bladder, smooth muscle cells were previously shown
to synthesize and secrete FGF7 and FGF10 that bind to
urothelial cell membrane FGFRIIb, thereby inducing proliferation (29, 34). Several lines of evidence strengthen the view
that FGF7-FGFR2 binding is also involved in renal urothelial
cell proliferation following IRI: 1) FGF7 synthesis is markedly
upregulated in the kidney following IRI; 2) FGFR2 staining
decreased after IRI, a process reported after FGF7 binding (and
not FGF10 binding) (2); 3) injection of FGF7 in mice (without
IRI) induces a proliferation response in the urothelium (and
also in some tubules located in the vicinity) very similar to the
IRI proliferation pattern; and 4) conversely, FGFR2 antisense
oligonucleotide injection within the renal urinary space shortly
after IRI significantly decreases the urothelial cell proliferation
index.
FGF2 may also be involved in early FGFR2-induced proliferation (31), as suggested by its upregulation within a similar
time frame as FGF7. However, FGF2 binds with a 10- to
15-fold lower affinity than FGF7 to FGFR2IIIb whereas FGF2
but not FGF7 binds with a high affinity to FGFR2IIIc present
on mesenchymal-derived cells (29, 31). Altogether, these data
suggest that FGFR2IIIb and not FGFR2IIIc are expressed in
urothelial cells and that among the potential FGFR2IIIb ligands, FGF7 is the most likely candidate to account for early
time frame renal urothelial proliferation.
As shown in Fig. 1E, proliferation of tubular cells is delayed
compared with urothelial cells with a lag time ⬍6 h (initiation
of the proliferation of most tubules located in the corticomedullary area begins by 22 h after IRI). This sequence of events
within a narrow time frame raises two main issues: 1) the
potential of urothelial cells to migrate and differentiate into
tubular cells; and 2) a potential paracrine effect of dividing
urothelial cells triggering the proliferation of tubular cells
located in the vicinity (i.e., the corticomedullary area).
First, following a BrdU pulse at 16 h, renal urothelial cells
proliferate and remain at the same location (i.e., urinary space)
and although some rare BrdU-positive cells can be detected
within tubules in the vicinity of urinary and vascular structures,
they are very likely to originate from some early dividing
tubules tagged by BrdU rather than dividing urothelial cells.
Thus we can consider that early proliferating cells do not
migrate and thus do not account for tubule cell proliferation
through a hypothetical differentiation (we ruled out a dye
dilution as we found a similar pattern when the animals were
euthanized by days 2 and 3; data not shown).
Second, rare constitutive FGFR2-positive tubular cells are
present within the normal kidney, located mostly in some
collecting ducts near urothelial structures as previously shown
(4, 25). They proliferate after FGF7 injection (Fig. 7C) and
probably account for some of the proliferating tubular cells
detected at this early time frame, giving rise to outer medullary
tubular regeneration. The recognition of the critical contribu-
F485
F486
UROTHELIUM PROLIFERATION AFTER ISCHEMIC INJURY
AJP-Renal Physiol • VOL
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
migration of label-retaining cells of the kidney papilla. JASN 20: 2315–
2327, 2009.
Oliver JA, Maarouf O, Cheema FH, Martens TP, Al-Awqati Q. The
renal papilla is a niche for adult kidney stem cells. J Clin Invest 114:
795–804, 2004.
Ornitz DM, Xu J, Colvin JS, McEwen DG, MacArthur CA, Coulier F,
Gao G, Goldfarb M. Receptor specificity of the fibroblast growth factor
family. J Biol Chem 271: 15292–15297, 1996.
Powers CJ, McLesley SW, Wellstein A. Fibroblast growth factors, their
receptor, and signaling. Endocr Relat Cancer 7: 165–197, 2000.
Sagrinati C, Netti GS, Mazzinghi B, Lazzeri E, Liotta F, Frosali F,
Ronconi E, Meini C, Gacci M, Squecco R, Carini M, Gesualdo L,
Francini F, Maggi E, Annunziato F, Lasagni L, Serio M, Romagnani
S, Romagnani P. Isolation, and characterization of multipotent progenitor
cells from the Bowman’s capsule of adult human kidneys. J Am Soc
Nephrol 17: 2443–2456, 2006.
Shirali AC, Goldstein DR. Tracking the toll of kidney disease. J Am Soc
Nephrol 19: 1444 –1450, 2008.
Sutton TA, Molitoris BA. Mechanisms of cellular injury in ischemic
acute renal failure. Semin Nephrol 18: 490 –497, 1998.
Tash JA, David SG, Vaughan ED ED, Herzlinger DA. Fibroblast
growth factor-7 regulates stratification of the bladder urothelium. J Urol
66: 2536 –2541, 2001.
Villanueva S, Cespedes C, Gonzales AA, Roessler E, Vio CP. Inhibition
of bFGF-receptor type 2 increases kidney damage, and suppresses nephrogenic protein expression after ischemic renal failure. Am J Physiol Regul
Integr Comp Physiol 294: R819 –R828, 2008.
Villanueva S, Cespedes C, Gonzalez A, Vio C. bFGF induces an earlier
expression of nephrogenic proteins after ischemic acute renal failure. Am
J Physiol Regul Integr Comp Physiol 291: R1677–R1687, 2006.
Villanueva S, Céspedes C, Vio CP. Ischemic acute renal failure induces
the expression of a wide range of nephrogenic proteins. Am J Physiol
Regul Integr Comp Physiol 290: R861–R870, 2006.
Yano S, Ito Y, Fujimoto M, Hamazaki TS, Tamaki K, Okochi H.
Characterization, and localization of side population cells in mouse skin.
Stem Cells 23: 834 –841, 2005.
Zhang D, Kosman J, Carmean N, Grady R, Bassuk JA. FGF10 and its
receptor exhibit bidirectional paracrine targeting to urothelial and smooth
muscle cells in the lower urinary tract. Am J Physiol Renal Physiol 291:
F481–F494, 2006.
Zhao H, Kegg H, Grady S, Truong HT, Robinson ML, Baum M, Bates
CM. Role of fibroblast growth factor receptors 1, and 2 in the ureteric bud.
Dev Biol 276: 403–15, 2004.
299 • SEPTEMBER 2010 •
www.ajprenal.org
Downloaded from http://ajprenal.physiology.org/ by 10.220.33.4 on June 17, 2017
9. Humphreys BD, Valerius MT, Kobayashi A, Mugford JW, Soeung S,
Duffield JS, Mc Mahon AP, Bonventre JV. Intrinsic epithelial cells
repair the kidney after injury. Cell Stem Cell 2: 284 –291, 2008.
10. Ichimura T, Finch P, Zhang G, Kan M, Stevens J. Induction of FGF7
after kidney damage: a possible paracrine mechanism for tubule repair. Am
J Physiol Renal Fluid Electrolyte Physiol 271: F967–F976, 1996.
11. Kale S, Karihaloo A, Clark PR, Kashgarian M, Krause DS, Cantley
LG. Bone marrow stem cells contribute to repair of the ischemically
injured renal tubule. J Clin Invest 112: 42–49, 2003.
12. Kurzrock EA, Lieu DK, Degraffenried LA, Chan CW, Isseroff RR.
Label-retaining cells of the bladder: candidate urothelial stem cells. Am J
Physiol Renal Physiol 294: F1415–F1421, 2008.
13. Lameire N, Van Biesen W, Vanholder R. Acute renal failure. Lancet
365: 417–430, 2005.
14. Lange C, Tögel F, Ittrich H, Clayton F, Nolte-Ernsting C, Zander AR,
Westenfelder C. Administered mesenchymal stem cells enhance recovery
from ischemia reperfusion induced acute renal failure in rats. Kidney Int
68: 1613–1617, 2005.
15. Lin F, Cordes K, Li L, Hood L, Couser WG, Shankland SJ, Igarashi
P. Hematopoietic stem cells contribute to the regeneration of renal tubules
after renal ischemia-reperfusion injury. J Am Soc Nephrol 14: 1188 –1199,
2003.
16. Lin F, Moran A, Igarashi P. Intrarenal cells not bone marrow-derived
cells are the major source of regeneration in postischemic kidney. J Clin
Invest 115: 1756 –1764, 2005.
17. Maeshima A, Yamashita S, Nojima Y. Identification of renal progenitorlike tubular cells that participate in the regeneration processes of the
kidney. J Am Soc Nephrol 14: 3138 –3146, 2003.
18. Maeshima A, Yamashita S, Nojima Y. Involvement of Pax-2 in the
action of activin A on tubular cell regeneration. J Am Soc Nephrol 14:
3138 –3146, 2003.
19. Molitoris BA, Sutton TA. Endothelial injury, and dysfunction: role in the
extension phase of acute renal failure. Kidney Int 66: 496 –499, 2004.
20. Morigi M, Imberti B, Zoja C, Corna D, Tomasoni S, Abbate M,
Rottoli D, Angioletti S, Benigni A, Perico N, Alison M, Remuzzi G.
Mesenchymal stem cells are renotropic, helping to repair the kidney, and
improve function in acute renal failure. J Am Soc Nephrol 15: 1794 –1804,
2004.
21. Morigi M, Introna M, Imberti B, Corna D, Abbate M, Rota C, Rottoli
D, Benigni A, Perico N, Zoja C, Rambaldi A, Remuzzi A, Remuzzi G.
Human bone marrow mesenchymal stem cells accelerate recovery of acute
renal injury, and prolong survival in mice. Stem Cells 26: 2075–2082,
2008.
22. Oliver JA, Klinakis A, Cheema FH, Friedlander CJ, Sampogna RV,
Martens TP, Liu C, Efstratiadis A, Al-Awqati Q. Proliferation and