http://www.kidney-international.

org mini review

& 2009 International Society of Nephrology

Ureteral obstruction as a model of renal interstitial

fibrosis and obstructive nephropathy
1 1 1
Robert L. Chevalier , Michael S. Forbes and Barbara A. Thornhill
1
Department of Pediatrics, University of Virginia, Charlottesville, Virginia, USA

Renal fibrosis is the hallmark of progressive renal disease Renal fibrosis is regarded as the final common pathway for
of virtually any etiology. The model of unilateral ureteral most forms of progressive renal disease, and involves
obstruction (UUO) in the rodent generates progressive glomerular sclerosis and/or interstitial fibrosis. Because most
renal fibrosis. Surgically created UUO can be renal disorders (whether glomerular or interstitial, congenital
experimentally manipulated with respect to timing, severity, or acquired) lead to renal fibrosis, there is great interest in
and duration, while reversal of the obstruction permits the identifying underlying factors to prevent or reverse the
study of recovery. The use of genetically engineered mice changes. For over 50 years, surgical renal ablation in a
has greatly expanded the utility of the model in studying variety of animal species has been used to model progressive
molecular mechanisms underlying the renal response to renal disorders. A seminal paper published in 1981
UUO. Ureteral obstruction results in marked renal emphasized the role of glomerular hyperfiltration in the
hemodynamic and metabolic changes, followed by tubular 1
initiation of lesions in remnant glomeruli in rats.
injury and cell death by apoptosis or necrosis, with Subsequent studies showed that while renal ablation leads
interstitial macrophage infiltration. Proliferation of to glomerulosclerosis in rats, there is little apparent
interstitial fibroblasts with myofibroblast transformation glomerular injury in C57BL/6 mice (the most commonly
leads to excess deposition 2
used ‘wild-type’ experimental mice). However, other mouse
of the extracellular matrix and renal fibrosis. Phenotypic strains (129/Sv and Swiss-Webster mice) develop significant
transition of resident renal tubular cells, endothelial cells, 2
glomerular sclerosis, underscoring the effect of genetic
and pericytes has also been implicated in this process. background on the renal response to renal ablation. While
Technical aspects of the UUO model are discussed in this the focus of investigation remained centered on the
review, including the importance of rodent species or glomerulus for many years, attention turned also to tubular
strain, the age of the animal, surgical procedures, and and interstitial injury, with models of ischemia/reperfusion,
histological methods. The UUO model is likely to reveal 3
immune injury, and nephrotoxins. Clinical studies have also
useful biomarkers of progression of renal disease, as well as demonstrated that progression of renal insufficiency is much
new therapies, which are desperately needed to allow better correlated with renal interstitial fibrosis than with
intervention before the establishment of irreversible renal 4
glomerular pathology.
injury.
In the 1970s, unilateral ureteral obstruction (UUO) in the
Kidney International (2009) 75, 1145–1152; doi:10.1038/ki.2009.86;
rabbit was shown to result in proliferation of renal interstitial
published online 1 April 2009 5
fibroblasts and their transformation into myofibroblasts.
KEYWORDS: apoptosis; fibrosis; macrophages; obstructive nephropathy;
progression of chronic renal failure
Subsequent studies in the rabbit showed increased interstitial
collagens I, III and IV, fibronectin, and heparin sulfate
6
proteoglycan. Since then, animal models of UUO have been
expanded and refined to elucidate the pathogenesis of
obstructive nephropathy as well as mechanisms responsible
7,8
for progressive renal fibrosis. In contrast to adults, in
whom diabetes and hypertension are the major etiologies of
renal failure, congenital urinary tract obstruction is the most
important identifiable cause of renal failure in infants and
9
children.
The development of animal models of renal fibrosis
presumes an understanding of the disorder under investiga-
tion. Unfortunately, there is no general agreement regarding
Correspondence: Robert L. Chevalier, Department of Pediatrics Box a strict definition of renal fibrosis, which underscores the
800386, University of Virginia Health System, Charlottesville, Virginia
need for ongoing investigation of a process central to most
22908, USA. E-mail: [email protected]
progressive renal disease. A number of terms have been
Received 8 January 2009; revised 17 February 2009; accepted 24
February 2009; published online 1 April 2009

11
Kidney International (2009) 75, 1145– Kidney International (2009) 75, 1145–11
 m iChevalier
RL n i r e vet ial.:
e Ureteral obstruction as a model of renal m iofn irenal
RL Chevalier et al.: Ureteral obstruction as a model

applied to the pathological process of renal accumulation of Technical considerations

collagen or extracellular matrix: fibrosis, sclerosis, and The surgical procedure for creating an animal model of UUO
10
scarring. This process appears to represent a maladaptive is relatively straightforward, if performed as a single
response to injury, which optimally should result in healing operation in an adult rat. Morbidity and mortality will be
of the wound.
3,11
If the insult is prolonged (as with chronic reduced by using a temperature-controlled operating table
UUO), the outcome is irreversible renal injury. If the insult is heated to body temperature, with the animal anesthetized
removed (by relief of obstruction), or if the fibrotic response with isoflurane/oxygen and with the use of a high-quality
is blocked (by inhibition of gene expression, protein binocular microscope to visualize the operating field. This
production, or receptor response), renal injury may be approach is particularly important in mice (especially in the
prevented or reversed. neonatal period), or if the obstruction is to be relieved or
reversed by a subsequent operation. If this be the case,
meticulous technique is essential to avoid adhesions and
Variables in the development of models of UUO tissue damage, and establishment of ureteral patency must be
14,15
As early studies of the renal consequences of UUO were documented at the time of study. Ligation of the ureter is
performed in the rabbit and dog, the preponderance of the technique used most frequently. However, fistulae can
current studies are based on the rat and mouse. Since 1950, form around the ligature, allowing urine to bypass the
there have been over 10,000 publications involving ureteral obstruction (unpublished observations). Adhesions forming
obstruction, nearly 1000 of which address fibrosis (Medline). around the ligature can increase the difficulty in removing
The advantages of using UUO as a model of renal fibrosis the ligature if recovery is to be examined: this is a greater
include the absence of an exogenous toxin, the lack of a problem with adult than neonatal animals. Small vascular
‘uremic’ environment, and the availability of the contralateral clips can be placed around the ureter, but this approach can
kidney as a control. Using the contralateral kidney as a injure the ureter, or allow urine to pass if not closed
control, however, does not take into account its cellular, correctly. A piece of silastic tubing can be folded
metabolic, and functional renal compensatory changes in perpendicularly across the ureter to create an obstruction, or
12
response to UUO. For this reason, comparison of renal the tubing can be slit and fitted around the ureter
changes resulting from UUO should be compared with those longitudinally, forming a sleeve and creating partial
in sham-operated animals (unless comparing a therapeutic obstruction. Partial UUO can also be created by inserting
intervention on the obstructed kidney). Contralateral the ureter in a surgically created slit in the underlying
nephrectomy in animals subjected to partial UUO reveals psoas muscle. However, this technique leads to variable
17
the effects of reduced renal mass superimposed on the (often very mild) degrees of partial obstruction.
obstructive injury. Surgical models of UUO have the added A more reproducible method of creating variable, reversible
15
advantage of allowing variation in the severity, timing, and partial UUO has been developed in the neonatal mouse.
duration of obstruction, as well as the opportunity to study This involves the placement of a fine ligature around the
recovery following relief of the obstruction. ureter and a piece of stainless steel wire of known diameter,
Complete UUO initiates a rapid sequence of events in the which has been placed parallel to the ureter. After ligation,
obstructed kidney, leading within 24 h to reduced renal blood the wire is slipped out, leaving a partial obstruction with the
8
flow and glomerular filtration rate. This is followed within desired luminal diameter. This ligature can be removed at
several days by hydronephrosis, interstitial inflammatory various intervals, allowing the study of recovery of the
infiltration (macrophages), and tubular cell death attribu- lesions.
table to apoptosis and necrosis. Tubular epithelial cell death Glomerular filtrate rate can be measured following relief
is caused by a number of stressors resulting from UUO, of UUO in the rat or mouse, using standard clearance
18,19
including ischemia, hypoxia, oxidant injury, and axial strain techniques. A novel model of UUO has been described in
caused by tubular dilatation. There appears to be a close the adult mouse subjected to contralateral nephrectomy at
association between progressive fibrosis and tubular cell the time of ureteral reimplantation following 10 days of
13 20
death. Following complete UUO in the rat or mouse, the complete UUO. While this model allows serial measure-
progression to a severely hydronephrotic kidney with marked ment of blood urea nitrogen as a reflection of glomerular
loss of renal parenchyma takes place over 1–2 weeks, with filitrate rate in the postobstructed kidney, the development of
more severe fibrosis in the neonate than the adult (JJ Minor, significant proteinuria suggests that secondary glomerular
KA Gordon, MSF, BAT, RLC, unpublished data). injury contributes to the lesions. Tubular function can be
Because most cases of clinical congenital obstructive measured in rats following the release of UUO or bilateral
nephropathy involve partial, rather than complete obstruc- ureteral obstruction, and can be correlated with the
tion, models of partial UUO have been developed in the expression of renal epithelial sodium channel or aquapor-
14,15 21,22
neonatal rat and mouse. These models are also useful in ins. As described below, the UUO model is well suited to
the study of the pathogenesis of renal fibrosis, because the the discovery of new biomarkers which predate irreversible
lesions develop more slowly and may be better suited to the injury and functional changes.
16
study of therapeutic manipulations. The development of reproducible animal models of UUO,
particularly in mice, requires an experienced animal surgeon.

11
Kidney International (2009) 75, 1145– Kidney International (2009) 75, 1145–11
 m iChevalier
RL n i r e vet ial.:
e Ureteral obstruction as a model of renal m iofn irenal
RL Chevalier et al.: Ureteral obstruction as a model

The use of mutant strains of mice increases the possibility of with Mallory trichrome (Figure 1a) or picrosirius red (Figure
greater susceptibility to intraoperative or postoperative 1b). Types I, III, and IV collagen fibrils are identified by
mortality compared with wild-type controls. Operation on 23
these methods, with Type IV constituting the majority of
neonatal animals requires that the pups be accepted by the tubular basement membrane. The aniline blue of Mallory
mother, and nursed successfully. It is important to monitor trichrome may also stain tubular casts, however (Figure 1a),
the daily weight gain of each animal, to assure that overall and picrosirius red is better discriminated by image analysis
growth is adequate; kidneys should also be weighed at the software programs (Figure 1b). Moreover, the correlation
time of harvest, to document parenchymal growth. with tissue hydroxyproline content of collagen identified by
Renal fibrosis can be quantitated by digital morphometry 23
picrosirius is superior to collagen stained with trichrome. It
of renal collagen distribution using tissue sections stained is important to use an unbiased sampling approach: image
analysis software can be used to measure randomly selected
fields on the tissue section. In addition, total renal collagen
content can be quantitated, although this does not directly
address its distribution among renal compartments.
Because of the importance of the number of glomeruli per
kidney as a determinant of progression of renal disease, an
unbiased rigorous approach has been developed: the disector
24
technique. Comparison of this time-consuming approach
with simply counting all glomeruli in a single mid-polar
planar section of a whole kidney showed excellent correlation
25,26
in both rats and mice. The latter technique is therefore
more practical for most rodent models of UUO.

Figure 1 | Representative histologic sections of kidneys from

mice subjected to unilateral ureteral obstruction (UUO). (a, b)
Comparison of serial 3-mm paraffin sections of adult mouse
kidney after 7 days of UUO, showing histochemical stains for
collagen.
((a), Mallory trichrome; (b), picrosirius red). Although there is
general agreement between the aniline blue staining of the
trichrome
and the red of the picrosirius stain, the trichrome may also stain
structures such as cast material (arrow); furthermore, the distinct
color of the picrosirius stain, is better recognized by image-
analysis software. Scale bar ¼ 100 mm for (a) and (b). (c) Adult
mouse kidney
after 7 days of UUO, showing macrophages identified by F4/80
immunostaining, which is restricted to cells in the
interstitium. (d) Adult mouse kidney after 7 days of UUO,
showing TUNEL staining. A degenerating proximal tubule is
undergoing both apoptosis (single arrow) and necrosis
(double-headed arrow); both categories of cell death are
detected by this procedure.
Apoptosis is characterized by cell contraction, blebs, and
condensed nuclear material, while necrotic cells are swollen with
diffuse TUNEL- positive staining. (e) a-Smooth muscle actin (a-
SMA) staining in a section serial to the field shown in panel (c).
Immunostaining is extensive in interstitial cells, and only in a few
instances (arrows) is there evidence of colocalization of
macrophage- and myofibroblast- specific staining. (f) Six-week-old
mouse kidney with partial UUO applied at birth; section stained
for fibroblast-specific protein-1
(FSP-1, also known as S100A4, a protein associated with
intermediate filaments). Although numerous interstitial cells stain
positively (arrows), vascular profiles within glomeruli and in
arterioles (asterisk) are also stained. (g) Three-week-old mouse
kidney with partial UUO
applied at birth. Staining with Lotus tetragonolobus-derived lectin
shows proximal tubules. At this stage of obstruction many
tubules are normal, but some are beginning to attenuate at their
junction with the glomerulus (between arrows). (h) Six-week-old
mouse kidney with partial UUO applied at birth. Crowded
atubular glomeruli surround Lotus lectin-staining fragmented
remnants of the original proximal tubules, which have become
detached from their glomeruli. Scale bar ¼ 100 mm for panels c, e,
f, g, and h. (i, j) Neonatal mouse kidneys after 14 days of UUO (i,
contralateral kidney; j, obstructed kidney). Selective staining of the

11
Kidney International (2009) 75, 1145– Kidney International (2009) 75, 1145–11
m iChevalier
RL n i r e vet ial.:
e Ureteral obstruction as a model of renal vascular endothelium
RL Chevalier et al.:with PECAM-1
Ureteral antibodyasshows
obstruction mdiminished
a model iofn irenal
microvessel density in the obstructed kidney. Scale bar ¼ 100 mm.

11
Kidney International (2009) 75, 1145– Kidney International (2009) 75, 1145–11
 m iChevalier
RL n i r e vet ial.:
e Ureteral obstruction as a model of renal m iofn irenal
RL Chevalier et al.: Ureteral obstruction as a model

As discussed below, animal models have revealed an leading to the formation of atubular glomeruli and tubular
extraordinary complexity in the renal cellular response to atrophy; (3) phenotypic transition of resident renal cells.
UUO, and there is significant disagreement regarding the Chronic UUO activates the renin–angiotensin system, with
pathogenic implications of fibrosis. On one hand, renal production of reactive oxygen species and nuclear factor-k
interstitial matrix accumulation is regarded as a primary B, which promotes macrophage infiltration
28
and renal
cause of peritubular capillary obliteration, tubular atrophy, tubular apoptosis and interstitial fibrosis in rats.
29
27
and progressive renal insufficiency. The alternate view is Classically activated macrophages can generate tumor
that renal inflammation leads to glomerular and tubular necrosis factor- a, which mediates proapoptotic signaling
damage of some nephrons, which in turn leads to injury to and renal tubular cell apoptosis following UUO.
30
By
remaining nephrons, with interstitial fibrosis representing a contrast, alternatively activated macrophages generate anti-
10
secondary manifestation of disease. inflammatory cytokines, and induce cell survival and
Recent studies have revealed major pathways leading to proliferation.
31

the development of renal interstitial fibrosis following UUO In addition to the differentiation of infiltrating hemato-
(Figure 2): (1) interstitial infiltration by macrophages (Figure poietic stem cells and proliferation of resident interstitial
1c), which produce cytokines responsible for tubular fibroblasts, tubular cells can undergo epithelial–
apoptosis and fibroblast proliferation and activation; (2) 32
mesenchymal transition, by which process epithelial cells
tubular cell death by apoptosis and necrosis (Figure 1d), acquire mesenchymal characteristics and invade the
interstitium to contribute to the deposition of extracellular
matrix (Figure 2).

Alternatively activated
macrophage
Monocyte

Capillary
loss
ANG II
Apoptosis
ROS NFκB
Cell Ischemia
Classically
survival hypoxia
activated
TGF- 1 macrophage TGF-β1
TNF-α
Endothelial
cell
Renal EndMT Pericyte
Apoptosis tubular
epithelial cell TGF-β1
EMT
TNF-α
Resident interstitial
Dead
Renal fibroblast
epithelial
cell interstitial fibroblast

ECM
ECM Hematopoietic
deposition
degradation stem cell

Myofibroblast
PAI-1 tPA stress fibers: α smooth muscle actin

Figure 2 | Renal cellular interactions in the rodent kidney subjected to unilateral ureteral obstruction (UUO). The interstitium is
infiltrated by monocytes, which are ‘classically’ activated to macrophages that release cytokines such as TGF-b1 and tumor necrosis factor-a
(TNF-a). In turn, TGFb1 promotes a phenotypic response of tubular epithelial cells either to undergo apoptosis (leading to tubular atrophy)
or to undergo epithelial–mesenchymal transition (EMT), becoming fibroblasts that migrate to the interstitium. Angiotensin II (ANG II),
produced by the activation of monocytes, stimulates the production of nuclear factor-k B (NF-kB), which leads to the recruitment of more
macrophages, as well as to the production of reactive oxygen species (ROS), which aggravates renal tubular injury. In contrast, alternatively
activated macrophages can enhance tubular cell survival and proliferation. Endothelial cells can undergo endothelial–mesenchymal
transition (EndMT) or apoptosis, which leads to capillary loss and secondary renal ischemia and hypoxia. Resident pericytes and infiltrating
hematopoietic stem cells can also differentiate into fibroblasts. Under the stimulus of cytokines, such as TGF-b1 produced by macrophages
or other cells, fibroblasts synthesize stress fibers and undergo further differentiation to become myofibroblasts. The myofibroblasts are
contractile and augment the deposition of the extracellular matrix (ECM), leading to progressive interstitial fibrosis. This process is
augmented by a decrease in ECM degradation, mediated by plasminogen-activator inhibitor-1 (PAI-1) and tissue-type plasminogen
activator (tPA).

11
Kidney International (2009) 75, 1145– Kidney International (2009) 75, 1145–11
 m iChevalier
RL n i r e vet ial.:
e Ureteral obstruction as a model of renal m iofn irenal
RL Chevalier et al.: Ureteral obstruction as a model

Similarly, endothelial cells can undergo endothelial–mesen- clinical therapeutic approach to slowing or preventing the
33
chymal transition, or can undergo apoptosis, leading to progression of most forms of renal disorders.
34
capillary loss and consequent renal ischemia and hypoxia. For each of these pathways, factors contributing to
In addition to tubular cells and endothelial cells, pericytes progressive renal fibrosis are balanced by counteracting
35
can also differentiate into myofibroblasts, which express factors. Thus, heme oxygenase-1 suppresses monocyte
42
a-smooth muscle actin (Figure 1e), and are major contri- chemoattractant protein-1, a stimulator of interstitial
butors to interstitial extracellular matrix. The importance of macrophage accumulation; the oncoprotein bcl-2 counters
43
endothelial cells and pericytes as a major source of renal caspase-induced tubular apoptosis; and bone morphoge-
collagen-producing cells following UUO has only recently netic protein-7 opposes epithelial–mesenchymal transition
been recognized, which shifts attention from the tubular stimulated by transforming growth factor-b1.
39,44
Similarly,
epithelial cell to the renal vasculature as a focus for renal factors promoting collagen degradation offset those promot-
33,35
fibrotic injury. Significant advances in understanding the ing its synthesis and deposition (Figure 2).
45
Improved
process of epithelial–mesenchymal transition followed the understanding of these interrelationships should lead to new
development of an antibody to fibroblast-specific protein-1 13
methods of limiting or reversing fibrotic injury, as well as to
(Figure 1f ), which permitted the tracking of the lineage of the identification of urinary biomarkers that result from renal
36
cells undergoing phenotypic transition. However, it should 39
injury or response to therapy. Several important lessons
be recognized that this antibody binds also to a number of have been learned from these studies, underscoring the
cell types, including endothelial cells (Figure 1f ) and complexity of the interactions. While as described above,
35
macrophages. This problem can be addressed with the angiotensin II plays a central role in the increase of fibrosis
33,35
application of molecular lineage tracing techniques. The due to UUO, these effects are driven by stimulation of the
use of immunohistochemistry to localize individual cell types renal angiotensin Type 1 receptors: stimulation of Type 2
can be combined with digital morphometry to quantitate the 46
receptors actually suppresses fibrosis. Also of note, while
parameters and to make statistical comparisons. In addition, most macrophages infiltrating the obstructed kidney appear
lectins (or other segment-specific markers) can be used to to be injurious (‘classical activation’), other macrophage
identify tubular segments, permitting further refinement of populations may attenuate inflammation (‘alternative activa-
localization of cell types. Certain lectins, such as derived tion’).
31,47
Furthermore, factors having a salutary effect on
from Lotus tetragonolobus (which labels mouse proximal one renal compartment can also have an injurious effect on
tubules) (Figure 1g), permit identification of tubular cell 48
another. While the glycoprotein osteopontin stimulates the
remnants after significant tubular injury and fragmentation
15 accumulation of macrophages in the obstructed kidney and
resulting from UUO (Figure 1h). However, Lotus lectin can contributes to renal fibrosis, osteopontin also suppresses
also bind to intercalated cells of collecting ducts and elastic 49,50
37 renal tubular apoptosis, which reduces tubular injury. In
tissue of arterioles. Thus, as with the tissue identification of contrast, death-associated protein kinase contributes to
collagen, apoptosis, and fibroblasts, the accurate tubular apoptosis in mice with UUO, but attenuates the
identification of tubular segments requires an experienced 51
progression of fibrotic injury. These findings highlight the
microscopist to interpret the localization of the marker in the need for caution in the interpretation of experimental results.
context of renal morphology. Peritubular capillaries can be
In addition to the known counteracting factors described
identified by immunohistochemistry using antibody to
above, unidentified counteracting factors can confound
platelet endothelial cell adhesion molecule-1. As shown in
interpretation in studies of gene knockout mice that have
Figures 1i and j, UUO in the neonatal mouse leads to
been subjected to UUO. An effect of deletion of a gene seen
decreased peritubular capillary density in the renal
at an early time point following UUO may disappear
medulla. 38
later.
Use of genetically engineered mice subjected to UUO Thus, suppression of a specific molecule may impair or retard
In the past decade, many important pathways responsible for fibrosis early in the process, but overlapping pathways can
renal fibrosis have been elucidated, using mice subjected to erase the effects over time. As renal fibrosis may not develop
gene deletion or increased expression (either transgenic or until late in the disease process, these other pathways may
delivery of the gene). These have been reviewed recently, actually be of greater therapeutic importance. For these
summarizing the data leading to the elucidation of molecular reasons, both early and late observation points should be
38,39 included in such studies. The selection of time points of
pathways. A number of studies point to endogenous
renal angiotensin II as a central mediator of the renal study using models of UUO depends largely on the model
response to UUO, including inflammation, apoptosis, and and the question being addressed. Complete UUO leads to
40 rapid destruction of the renal parenchyma, such that most of
interstitial fibrosis. In neonatal mice with 0, 1, 2, or 4
the renal cellular changes have taken place by 2–3 weeks.
functional copies of the angiotensinogen gene subjected to
Partial UUO has the advantage of better reflecting clinical
UUO, a linear relationship existed between the number of
41 obstructive nephropathy, as renal cellular responses progress
gene copies and the severity of renal fibrosis. This effect is over a number of weeks. Relief of either complete or partial
independent of systemic hemodynamic changes. At the UUO, furthermore, allows the study of recovery at various
present time, angiotensin inhibition represents the principal intervals. In a report of temporary complete UUO in the
11
Kidney International (2009) 75, 1145– Kidney International (2009) 75, 1145–11
 m iChevalier
RL n i r e vet ial.:
e Ureteral obstruction as a model of renal m iofn irenal
RL Chevalier et al.: Ureteral obstruction as a model

neonatal rat, renal functional recovery was complete after 1 nephrons are functional at birth in the rat or mouse, with the
month, but after 1 year 80% of function was lost, at which remainder
time fibrotic changes also appeared in the contralateral
52,53
kidney. The latter are presumably the result of gradual
deterioration of the postobstructed kidney, and hyperfiltra-
tion by the contralateral kidney.

Effect of species
The use of animal models of human disease should take into
account the possibility of species-specific signaling
pathways. A number of studies have shown a reduction in
the renal production of epidermal growth factor following
54–56
UUO in the rat or mouse, as well as a reduction in
urinary epi- dermal growth factor in human obstructive
57,58
nephropathy.
Administration of exogenous epidermal growth factor to
neonatal rats with UUO reduces tubular apoptosis and inter-
stitial fibrosis, and has a salutary effect 1 month following
59
release of obstruction. This effect is mediated in the rat by
maintaining phosphorylation of BAD, a proapoptotic mole-
60
cule. However, administration of epidermal growth factor
potentiates renal injury due to UUO in the neonatal mouse, a
response mediated by elevated Src activity in mouse tubular
56,61
cells, and not detected in rat or human tubular cells.

Effect of age
As noted above, the study of neonatal models of UUO is
important because obstructive nephropathy is a clinically
important cause of renal failure in infants and children. In
contrast to the consequences of UUO in the adult animal,
obstruction to urine flow in the embryonic kidney can
interfere with morphogenesis of kidneys and urinary tract,
although obstruction later in fetal life or in the neonate can
prevent renal growth and maturation. A number of animal
models of congenital obstructive nephropathy have been
developed, including the chick embryo, fetal rabbit, and
62
opossum. Use of the opossum represents a novel approach
to fetal research, as the postnatal marsupial is essentially an
extrauterine fetus, in which ureteral obstruction can be
accomplished by withdrawing the animal from its mother’s
63
pouch. Renal maturation in the fetal sheep is more similar
to that of the human, in which nephrogenesis is complete
before birth. A number of studies have been performed in
this model, showing that recovery following relief of
obstruction is directly proportional to the duration of
intrauterine decompression, and inversely proportional to
64
the duration of obstruction. Notably, UUO early in
gestation leads to compensatory growth of the contralateral
62
kidney in the fetal lamb. This indicates that compensatory
growth is not necessarily dependent on functional demand,
as excretory function in the fetus is accomplished by the
placenta. Experimental UUO in the fetal monkey reveals
glomerular as well as tubulointerstitial changes, similar to
65,66
those found in human fetal obstruction.
In contrast to the primate, sheep, and guinea pig, in
which nephrogenesis is complete before birth, only 10% of

11
Kidney International (2009) 75, 1145– Kidney International (2009) 75, 1145–11
 m iChevalier
maturing
RL n i r postnatally.
e vet ial.:
e Ureteral obstruction
67
The earlyas apostnatal period in
model of renal m iofn irenal
RL Chevalier et al.: Ureteral obstruction as a model
these species parallels that of the midtrimester human
fetus, and allows the study of the effects of UUO during
the period of most rapid nephrogenesis. Chronic UUO in the
neonatal rat increases renal interstitial accumulation of
collagen types I, III, and V, with increased type IV in tubular
68
basement membrane.
Although surgical models of UUO in the neonatal mouse
reveal the effects of obstruction without intrinsic defects in
renal development, the use of genetically altered strains reveals
mole- cular mechanisms involved in the response of the
38
developing kidney to surgical obstructive injury.
In human studies and animal models, fetal and neonatal
UUO results in reduced nephron number, with the
magnitude of reduction being dependent on the severity
14,25
and duration of obstruction. The mechanisms of
nephron loss include glomerulosclerosis, phenotypic transi-
tion of glomerular cells with disappearance of glomeruli, and
glomerulotubular disconnection leading to the formation of
14,15,53
atubular glomeruli. Chronic UUO in the neonatal rat
markedly increases renal renin production, with persistence
of the fetal renal pattern of renin distribution along afferent
55 52
arterioles. This is reversed by relief of obstruction.
Similarly, markers of renal tubular and interstitial maturation
retain a fetal pattern in neonatal rats subjected to chronic
55
UUO. The use of the neonatal UUO model can shed light
on the obstructive lesion in the adult: neonatal renal pericytes
express NG2 and a-smooth muscle actin (pericyte markers),
which disappear with normal maturation, but persist in
35
response to neonatal UUO.
In investigating mechanisms of obstructive injury in the
immature kidney, it is important to recognize the role(s), in
normal renal development, of factors deemed injurious to the
obstructed adult kidney. Thus, while angiotensin inhibition
has been shown to ameliorate renal lesions due to UUO in
69
adult animals, the administration of angiotensin-converting
enzyme inhibitors to neonatal rats with UUO can actually
16
exacerbate the renal lesions and augment renal fibrosis.
Similar considerations arise in the examination of other
factors, such as transforming growth factor-b1 or endothelial
70,71
nitric oxide synthase.
In conclusion, the use of animal models of UUO has pro-
vided many new insights into the pathogenesis of obstructive
nephropathy, and of progressive renal fibrosis in general.
The promising biomarkers of disease severity, progression,
and response to therapy have already been revealed by the
39
model, and it is likely that application of emerging tech-
nologies will ensure its usefulness well into the future.

DISCLOSURE
All the authors declared no competing interests.

REFERENCES
1. Hostetter TH, Olson JL, Rennke HG et al. Hyperfiltration in remnant
nephrons: a potentially adverse response to renal ablation. Am J Physiol
1981; 241: F85–F93.
2. Ma LJ, Fogo AB. Model of robust induction of glomerulosclerosis in mice:
Importance of genetic background. Kidney Int 2003; 64: 350–355.

11
Kidney International (2009) 75, 1145– Kidney International (2009) 75, 1145–11
 m iChevalier
RL n i r e vet ial.:
e Ureteral obstruction as a model of renal m iofn irenal
RL Chevalier et al.: Ureteral obstruction as a model

3. Schnaper HW, Kopp JB. Why kidneys fail: Report from an American 30. Misseri R, Meldrum DR, Dinarello CA et al. TNF-alpha mediates
Society of Nephrology Advances in Research Conference. J Am Soc obstruction-induced renal tubular cell apoptosis and proapoptotic
Nephol 2006; 17: 1777–1781. signaling. Am J Physiol 2004; 288: F406–F411.
4. Schainuck LI, Striker GE, Cutler RE et al. Structural-functional correlations 31. Duffield JS. The inflammatory macrophage: a story of Jekyll and Hyde.
in renal disease. Part II: The correlations. Hum Pathol 1970; 1: 631–641. Clin Sci 2003; 104: 27–38.
5. Nagle RB, Bulger RE, Cutler RE et al. Unilateral obstructive nephropathy in 32. Iwano M, Plieth D, Danoff TM et al. Evidence that fibroblasts derive from
the rabbit:I. Early morphologic, physiologic, and histochemical changes. epithelium during tissue fibrosis. J Clin Invest 2002; 110: 341–350.
Lab Invest 1973; 28: 456–467. 33. Zeisberg EM, Potenta SE, Sugimoto H et al. Fibroblasts in kidney fibrosis
6. Sharma AK, Mauer SM, Kim Y et al. Interstitial fibrosis in obstructive emerge via endothelial-to-mesenchymal transition. J Am Soc Nephrol
nephropathy. Kidney Int 1993; 44: 774–788. 2008; 19: 2282–2287.
7. Klahr S, Morrissey J. Obstructive nephropathy and renal fibrosis. Am J 34. Ohashi R, Shimizu A, Masuda Y et al. Peritubular capillary regression
Physiol 2002; 283: F861–F875. during the progression of experimental obstructive nephropathy. J Am
8. Vaughan ED, Marion D, Poppas DP et al. Pathophysiology of unilateral Soc Nephrol 2002; 13: 1795–1805.
ureteral obstruction: Studies from Charlottesville to New York. J Urol 35. Lin SL, Kisseleva T, Brenner DA et al. Pericytes and perivascular fibroblasts
2004; 172: 2563–2569. are the primary source of collagen-producing cells in obstructive fibrosis
9. Seikaly MG, Ho PL, Emmett L et al. Chronic renal insufficiency in children: of the kidney. Am J Pathol 2008; 173: 1617–1627.
The 2001 annual report of the NAPRTCS. Pediatr Nephrol 2003; 18: 36. Strutz F, Okada H, Lo CW et al. Identification and characterization of a
796–804. fibroblast marker: FSP1. J Cell Biol 1995; 130: 393–405.
10. Kriz W, LeHir M. Pathways to nephron loss starting from glomerular 37. Schulte BA, Spicer SS. Histochemical evaluation of mouse and rat kidneys
diseases–Insights from animal models. Kidney Int 2005; 67: 404–419. with lectin-horseradish peroxidase conjugates. Am J Anat 1983; 168:
11. Eddy AA. Can renal fibrosis be reversed? Pediatr Nephrol 2005; 20: 345–362.
1369–1375. 38. Bascands JL, Schanstra JP. Obstructive nephropathy: insights from
12. Chevalier RL. Counterbalance in functional adaptation to ureteral genetically engineered animals. Kidney Int 2005; 68: 925–937.
obstruction during development. Pediatr Nephrol 1990; 4: 442–444. 39. Chevalier RL. Obstructive nephropathy: towards biomarker discovery and
13. Docherty NG, O’Sullivan OE, Healy DA et al. Evidence that inhibition of gene therapy. Nat Clin Prac Nephrol 2006; 2: 157–168.
tubular cell apoptosis protects against renal damage and development of 40. Chevalier RL, Cachat F. Role of angiotensin II in chronic ureteral
fibrosis following ureteric obstruction. Am J Physiol Renal Physiol 2006; obstruction. In: Wolf G (ed). The Renin-Angiotensin System and Progression
290: F4–F13, Jan.F4-13, /1. of Renal Diseases. 1 ed. Karger: Basel, 2001 pp 250–260.
14. Thornhill BA, Burt LA, Chen C et al. Variable chronic partial ureteral 41. Fern RJ, Yesko CM, Thornhill BA et al. Reduced angiotensinogen
obstruction in the neonatal rat: A new model of ureteropelvic junction expression attenuates renal interstitial fibrosis in obstructive
obstruction. Kidney Int 2005; 67: 42–52. nephropathy in mice. J Clin Invest 1999; 103: 39–46.
15. Thornhill BA, Forbes MS, Marcinko ES et al. Glomerulotubular 42. Pittock ST, Norby SM, Grande JP et al. MCP-1 is up-regulated in
disconnection in neonatal mice after relief of partial ureteral obstruction. unstressed and stressed HO-1 knockout mice: Pathophysiologic
Kidney Int 2007; 72: 1103–1112. correlates. Kidney Int 2005; 68: 611–622.
16. Chen CO, Park MH, Forbes MS et al. Angiotensin converting enzyme 43. Miyajima A, Chen J, Lawrence C et al. Antibody to transforming growth
inhibition aggravates renal interstitial injury resulting from partial factor-b ameliorates tubular apoptosis in unilateral ureteral obstruction.
unilateral ureteral obstruction in the neonatal rat. Am J Physiol Kidney Int 2000; 58: 2301–2313.
Renal Physiol 2007; 292: F946–F955. 44. Morrissey J, Hurska K, Guo G et al. Bone morphogenetic protein-7
17. Josephson S, Robertson B, Claesson G et al. Experimental obstructive improves renal fibrosis and accelerates the return of renal function. J Am
hydronephrosis in newborn rats. I. Surgical technique and long-term Soc Nephrol 2002; 13: S14–S21.
morphologic effects. Invest Urol 1980; 17: 478–483. 45. Zhang G, Kim H, Cai X et al. Urokinase receptor deficiency accelerates
18. Shi YM, Li CL, Thomsen K et al. Neonatal ureteral obstruction alters renal fibrosis in obstructive nephropathy. J Am Soc Nephrol 2005; 14:
expression of renal sodium transporters and aquaporin water channels. 1254–1271.
Kidney Int 2004; 66: 203–215. 46. Morrissey JJ, Klahr S. Effect of AT2 receptor blockade on the pathogenesis
19. Cochrane AL, Kett MM, Samuel CS et al. Renal structural and functional of renal fibrosis. Am J Physiol Renal Physiol 1999; 276: F39–F45.
repair in a mouse model of reversal of ureteral obstruction. J Am Soc 47. Eardley KS, Cockwell P. Macrophages and progressive tubulointerstitial
Nephrol 2005; 16: 3623–3630. disease. Kidney Int 2005; 68: 437–455.
20. Tapmeier TT, Brown KL, Tang Z et al. Reimplantation of the ureter after 48. Chevalier RL. Pathogenesis of renal injury in obstructive uropathy. Curr
unilateral ureteral obstruction provides a model that allows functional Opin Pediatr 2006; 18: 153–160.
evaluation. Kidney Int 2008; 73: 885–889. 49. Yoo KH, Thornhill BA, Forbes MS et al. Osteopontin regulates renal
21. Li C, Wang W, Norregaard R et al. Altered expression of epithelial sodium apoptosis and interstitial fibrosis in neonatal chronic unilateral ureteral
channel in rats with bilateral or unilateral ureteral obstruction. Am J obstruction. Kidney Int 2006; 70: 1735–1741.
Physiol 2007; 293: F333–F341. 50. Rouschop KMA, Sewnath ME, Claessen N et al. CD44 deficiency increases
22. Li C, Wang W, Kwon TH et al. Downregulation of AQP1, -2, and -3 after tubular damage but reduces renal fibrosis in obstructive nephropathy.
ureteral obstruction is associated with a long-term urine-concentrating J Am Soc Nephrol 2004; 15: 674–686.
defect. Am J Physiol 2001; 281: F163–F171. 51. Yukawa K, Hoshino K, Kishino M et al. Deletion of the kinase domain
23. Hironaka K, Sakaida I, Uchida K et al. Correlation between stellate cell in death-associated protein kinase attenuates renal tubular cell
activation and serum fibrosis markers in choline-deficient L-amino apoptosis in chronic obstructive uropathy. Int J Mol Med 2004; 13:
acid-defined diet-induced rat liver fibrosis. Dig Dis Sci 2000; 45: 515–520.
1935–1943. 52. Chevalier RL, Kim A, Thornhill BA et al. Recovery following relief of
24. Nyengaard JR. Stereological methods and their application in kidney unilateral ureteral obstruction in the neonatal rat. Kidney Int 1999; 55:
research. J Am Soc Nephol 1999; 10: 1100–1123. 793–807.
25. Chevalier RL, Thornhill BA, Chang AY et al. Recovery from release of 53. Chevalier RL, Thornhill BA, Chang AY. Unilateral ureteral obstruction in
ureteral obstruction in the rat: relationship to nephrogenesis. Kidney Int neonatal rats leads to renal insufficiency in adulthood. Kidney Int 2000;
2002; 61: 2033–2043. 58: 1987–1995.
26. Dziarmaga A, Eccles M, Goodyer P. Suppression of ureteric bud apoptosis 54. Storch S, Saggi S, Megyesi J et al. Ureteral obstruction decreases renal
rescues nephron endowment and adult renal function in Pax2 mutant prepro-epidermal growth factor and Tamm-Horsfall expression. Kidney Int
mice. J Am Soc Nephol 2006; 17: 1568–1575. 1992; 42: 89–94.
27. Eddy AA. Molecular basis of renal fibrosis. Pediatr Nephrol 2000; 15:
55. Chung KH, Chevalier RL. Arrested development of the neonatal kidney
290–301.
following chronic ureteral obstruction. J Urol 1996; 155: 1139–1144.
28. Tashiro K, Tamada S, Kuwabara N et al. Attenuation of renal fibrosis by 56. Kiley SC, Thornhill BA, Belyea BC et al. Epidermal growth factor
proteasome inhibition in rat obstructive nephropathy: possible role of potentiates renal cell death in hydronephrotic neonatal mice, but cell
nuclear factor kappaB. Int J Mol Med 2003; 12: 587–592.
survival in rats. Kidney Int 2005; 68: 504–514.
29. Miyajima A, Kosaka T, Seta K et al. Novel nuclear factor kappaB activation
57. Grandaliano G, Gesualdo L, Bartoli F et al. MCP-1 and EGF renal
inhibitor prevents inflammatory injury in unilateral ureteral obstruction.
expression and urine excretion in human congenital obstructive
J Urol 2003; 169: 1559–1563.
nephropathy. Kidney Int 2000; 58: 182–192.

11
Kidney International (2009) 75, 1145– Kidney International (2009) 75, 1145–11
 m iChevalier
RL n i r e vet ial.:
e Ureteral obstruction as a model of renal m iofn irenal
RL Chevalier et al.: Ureteral obstruction as a model

58. Yang Y, Hou Y, Wang CL et al. Renal expression of epidermal growth 65. Tarantal AF, Han VKM, Cochrum KC et al. Fetal rhesus monkey model of
factor and transforming growth factor-beta1 in children with congenital obstructive renal dysplasia. Kidney Int 2001; 59: 446–456.
hydronephrosis. Urology 2004; 67: 817–821. 66. Zhang PL, Peters CA, Rosen S. Ureteropelvic junction obstruction:
59. Chevalier RL, Goyal S, Thornhill BA. EGF improves recovery following relief morphological and clinical studies. Pediatr Nephrol 2000; 14:
of unilateral ureteral obstruction in the neonatal rat. J Urol 1999; 162: 820–826.
1532–1536. 67. Merlet-Benichou C, Gilbert T, Muffat-Joly M et al. Intrauterine growth
60. Kiley SC, Thornhill BA, Tang SS et al. Growth factor-mediated retardation leads to a permanent nephron deficit in the rat. Pediatr
phosphorylation of proapoptotic BAD reduces tubule cell death in vitro Nephrol 1994; 8: 175–180.
and in vivo. Kidney Int 2003; 63: 33–42. 68. Haralambous-Gasser A, Chan D, Walker RG et al. Collagen studies in
61. Kiley SC, Chevalier RL. Species differences in renal Src activity direct EGF newborn rat kidneys with incomplete ureteric obstruction. Kidney Int
receptor regulation in life or death response to EGF. Am J Physiol 2007; 1993; 44: 593–605.
293: F895–F903. 69. Klahr S, Ishidoya S, Morrissey J. Role of angiotensin II in the
62. Peters CA. Obstruction of the fetal urinary tract. J Am Soc Nephrol 1997; 8: tubulointerstitial fibrosis of obstructive nephropathy. Am J Kidney Dis
653–663. 1995; 26: 141–146.
63. Steinhardt G, Salinas-Madrigal L, Farber R et al. Experimental ureteral 70. Liu AL, Ballermann BJ. TGF-Beta type II receptor in rat renal vascular
obstruction in the fetal opossum. I. Renal functional assessment. J Urol development: Localization to juxtaglomerular cells. Kidney Int 1998; 53:
1990; 144: 564–566. 716–725.
64. Glick PL, Harrison MR, Adzick NS et al. Correction of congenital 71. Forbes MS, Thornhill BA, Park MH et al. Lack of endothelial nitric-oxide
hydronephrosis in utero IV: In utero decompression prevents renal synthase leads to progressive focal renal injury. Am J Pathol 2007; 170:
dysplasia. J Ped Surg 1984; 19: 649–657. 87–99.

11
Kidney International (2009) 75, 1145– Kidney International (2009) 75, 1145–11