The renal (myo-) fibroblast: A heterogeneous group of cells

Abstract

Several studies have demonstrated that mesenchymal stem cells have the capacity to reverse acute and chronic kidney injury in different experimental models by paracrine mechanisms. This paracrine action may be accounted for, at least in part, by microvesicles (MVs) released from mesenchymal stem cells, resulting in a horizontal transfer of mRNA, microRNA and proteins. MVs, released as exosomes from the endosomal compartment, or as shedding vesicles from the cell surface, are now recognized as being an integral component of the intercellular microenvironment. By acting as vehicles for information transfer, MVs play a pivotal role in cell-to-cell communication. This exchange of information between the injured cells and stem cells has the potential to be bi-directional. Thus, MVs may either transfer transcripts from injured cells to stem cells, resulting in reprogramming of their phenotype to acquire specific features of the tissue, or conversely, transcripts could be transferred from stem cells to injured cells, restraining tissue injury and inducing cell cycle re-entry of resident cells, leading to tissue self-repair. Upon administration with a therapeutic regimen, MVs mimic the effect of mesenchymal stem cells in various experimental models by inhibiting apoptosis and stimulating cell proliferation. In this review, we discuss whether MVs released from mesenchymal stem cells have the potential to be exploited in novel therapeutic approaches in regenerative medicine to repair damaged tissues, as an alternative to stem cell-based therapy.

Full text

Editorial Review
The renal (myo-)fibroblast: a heterogeneous group of cells
Peter Boor1,2,3 and Jürgen Floege1
1
Division of Nephrology, RWTH University of Aachen, Aachen, Germany,
2
Institute of Pathology, RWTH University of Aachen,
Aachen, Germany and
3
Institute of Molecular Biomedicine, Comenius University, Bratislava, Slovakia
Correspondence and offprint requests to: Peter Boor; E-mail: boor@email.cz
Abstract
Renal fibrosis is a central pathological process in kidneys
of patients with chronic kidney disease (CKD). Identifi-
cation of effective treatments that halt or reverse fibrosis
would be beneficial for most, if not all, CKD patients.
Key to this is an understanding of fibrogenesis, including
the principal responsible cells, the renal fibroblasts. It is
in part due to their inconspicuous appearance that it was
believed that there might not be much more to a fibroblast
than a simple interstitial mesenchymal cell which makes
up the organ stroma. The so-called ‘renal fibroblasts’are a
heterogeneous population of mesenchymal cells with
various essential functions during kidney development
and in adult life. Still, remarkable uncertainties exist in
the nomenclature of renal mesenchymal cells—or renal
fibroblasts—and molecular characterization remains poor.
The embryonic origin of fibroblasts is unclear as well,
although some studies point to a neural crest origin of
these cells. The renal myofibroblasts appear de novo in
renal fibrosis, originating from renal fibroblasts. Myofibro-
blasts most likely represent a stressed and dedifferentiated
phenotype of fibroblasts. We have only just begun to
appreciate that renal fibroblasts are anything but simple
renal interstitial cells.
Keywords: kidney development; origin of fibroblasts; renal fibrosis;
renal stroma; markers
Chronic kidney disease (CKD), defined as impaired renal
function for 3 months or longer, is often asymptomatic,
which delays the diagnosis of CKD until advanced stages
of disease. The prevalence of CKD has reached pandemic
proportions and some epidemiological studies have shown
that up to 10% of the world’s population is affected [1].
With increases in the aged population, the number of CKD
patients is expected to rise even further in the future. CKD
is a strong and independent risk factor for cardiovascular
morbidity and mortality and the risk increases with each
CKD stage. In so-called developing countries, where
access to renal replacement therapies is limited or does not
exist at all, the final stage of CKD most often equates to
death. It is estimated that >1 million patients die per year
as a result of the final CKD stage, i.e. end-stage renal
disease. In developed countries, both renal replacement
modalities, dialysis and renal transplantation, are available,
but they are extremely expensive and are associated with a
substantial reduction in the quality of life and life expect-
ancy of the patients. The worldwide cost of renal replace-
ment therapy is estimated to be ∼$1 trillion [1].
Essentially, all chronic renal diseases, but also re-
peated or serious acute insults, inevitably lead to renal
fibrosis. The underlying and principal pathological
finding in the kidneys of CKD patients is renal fibrosis
[2–4]. The importance of fibrosis, be it glomerulo-
sclerosis or tubulointerstitial fibrosis, is illustrated by its
close correlation with the decline in renal function. It
has been reported that interstitial fibrosis correlates
more closely with the decline in renal function than
glomerulosclerosis [5], but that may simply reflect our
inability to accurately quantify glomerulosclerosis in a
renal biopsy with limited numbers of glomeruli.
Fibrosis is a fundamental biological process, and an es-
sential and beneficial step in the course of tissue repair
and regeneration. It is possible that focal or initial fibrosis
is a beneficial process in kidney disease, since it might
support the mechanical stability of the injured organ and
encapsulate injured nephrons [2]. Nevertheless, sustained
and uncontrolled fibrosis becomes pathological, since the
functional tissue is ultimately replaced by permanent scar
tissue. Such scars disrupt normal organ structure and
thereby hinder the possibility for regeneration and normal
organ function. The key characteristics of renal fibrosis
are the extensive deposition of the extracellular matrix
(ECM) and expansion of the fibroblast population. Renal
fibrosis is often, if not always, associated with monocytic
inflammatory infiltrates and phenotypic alterations or loss
of resident renal cells, e.g. tubular and capillary endo-
thelial cells. The histological appearance of these changes
is shown in Figure 1.
Given its almost universal appearance in CKD, renal
fibrosis, in particular interstitial fibrosis, is an excellent
treatment target. Within the fibrotic process, a prime
target for therapy is the principal cell responsible for the
exaggerated production and deposition of ECM, i.e. the
renal fibroblast.
© The Author 2012. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.
For Permissions, please e-mail: journals.permissions@oup.com
Nephrol Dial Transplant (2012) 27: 3027–3036
doi: 10.1093/ndt/gfs296
by guest on July 31, 2012http://ndt.oxfordjournals.org/Downloaded from

Fig. 1. Histological features of renal fibrosis in rats. The kidney tubulointerstitium of a healthy rat is depicted in (A). Note the narrow interstitial
space with capillaries and very few inconspicuous cells (arrows). In (B) a rat kidney 5 days after UUO, a widely used model of renal fibrosis is
shown. The altered tubulointerstitium shows all of the typical features observed in renal fibrosis: expansion of ECM (asterisk), altered tubular
phenotype (arrowhead points to a mitosis in a tubular cell of a dilated and partly atrophied tubule) and inflammatory infiltrates (arrow). These changes
closely resemble the fibrosis observed in renal biopsies of patients with kidney diseases. Exaggerated deposition of ECM is shown on the example of
type I collagen immunohistochemistry in healthy (C) and fibrotic rat kidneys (D). In normal kidneys, very fine focal expression is found in the
interstitium (arrowhead in C) and around arterioles (a, arrow in C), whereas its expression is significantly increased during fibrosis (arrow and
arrowhead in D). In UUO, no expression of collagen I is found in the glomeruli (glo). Expression of a widely used marker for myofibroblasts, α-
SMA, is shown for normal (E) and fibrotic kidneys (F). In normal kidneys, α-SMA is expressed only by VSMCs of arteries and arterioles (a). In
fibrosis, a striking de novo expression in the interstitium is found (arrowhead in F). Note that tubular cells do not express α-SMA (arrows in F) and
no expression in glomeruli (glo) is found. A marker for mesenchymal cells, vimentin, is shown for normal (G) and fibrotic kidneys (H). In normal
kidneys, vimentin is strongly expressed in glomeruli (glo) by podocytes, mesangial cells and parietal epithelial cells as well as in arteries by VSMCs
and arterioles (a) and interstitial fibroblasts (arrowhead in G) but not by tubular cells (arrow in G). During fibrosis, expansion of interstitial vimentin-
positive cells is obvious (arrowhead in H). Note the de novo expression of vimentin in single and clusters of injured tubular cells (arrows in H). PAS,
periodic acid-Schiff staining, original magnifications ×400.
3028 Nephrol Dial Transplant (2012): Editorial Review
by guest on July 31, 2012http://ndt.oxfordjournals.org/Downloaded from

What are fibroblasts and myofibroblasts?
At first glance, the term fibroblast seems a straightforward
one. Fibroblasts are mesenchymal cells residing and
embedded in the ECM or stroma of connective tissues or
organs [6]. Several general definitions exist for fibroblasts,
e.g. ‘a connective-tissue cell of mesenchymal origin that
secretes proteins and especially molecular collagen from
which the extracellular fibrillar matrix of connective
tissue forms (http://www.merriam-webster.com/medical/
fibroblast)’. Fibroblasts are characterized by light
microscopy as elongated, spindle- or stellate-shaped cells
with rather pale cytoplasm and oval (or round) nuclei.
The first description of these cells dates back to the 1850–
70s by Virchow [7,8]. Ultrastructurally, renal fibroblasts
have an abundant endoplasmatic reticulum, collagen-
secreting granules and processes forming contacts to other
cells and basement membranes, including tubular, endo-
thelial and dendritic cells [9]. These cells can be isolated,
cultured and expanded in vitro, where the cells show a
morphology similar to the one in vivo.
The term fibroblast, although widely used, is not
completely correct. A ‘blast’denotes a cell with stem cell
features or in an activated state. The correct term for a
resting, differentiated fibroblast should be fibrocyte, as is
the case for other mesenchymal cells, e.g. chondrocytes
or osteocytes. However, the term ‘fibrocyte’is not used to
describe resting fibroblasts. Instead, it is used for circulat-
ing progenitors of fibroblasts expressing haematopoietic
markers and collagen (e.g. CD45
+
, CD33
+
, col1
+
)[10,
11]. For clarity, in this review, we will use the term fibro-
blast for both resting and activated fibroblasts, and the
term fibrocyte for the circulating precursor of fibroblasts.
The term myofibroblast was first used some 100 years
after the term fibroblast was introduced, when in 1971
Majno et al.[12] described phenotypic alterations of
fibroblasts towards a smooth muscle-like phenotype in
contracting dermal wound granulation tissue. This also
remains the main definition of a myofibroblast today: ‘a
fibroblast that has developed some of the functional and
structural characteristics (as the presence of myofilaments)
of smooth muscle cells (http://www.merriam-webster.com/
medical/myofibroblast)’. A fully developed myofibroblast
is defined as a cell with a light-microscopic appearance
similar to that of fibroblasts (i.e. spindle- or stellate-
shaped cells with oval nuclei and pale cytoplasm) that
expresses vimentin, α-smooth muscle actin (α-SMA) and
extra-domain A of fibronectin (ED-A fibronectin) but not
smooth muscle myosin or desmin [13,14]. An important
part of the definition, however, is based on the ultrastruc-
ture. A fully differentiated and mature myofibroblast should
have a prominent endoplasmatic reticulum, prominent Golgi
apparatus and collagen-secreting granules, peripheral myofi-
laments and gap and fibronexus junctions (the latter being
connections of intracellular myofilaments with extracellular
fibronectin) [14]. As in most studies of renal fibrosis, in the
following section, we will use the term myofibroblast
for renal interstitial cells positive for α-SMA, i.e. not
necessarily mature myofibroblasts.
Until now, no markers have been found that are
expressed specifically and exclusively on fibroblasts or
that would label all fibroblasts (Table 1). This is not a
specific problem of fibroblasts but of mesenchymal cell
biology in general. For example, a definition of mesench-
ymal stem cells involves the expression and lack of
expression of a number of surface markers (e.g. CD29,
CD44, CD73, CD90, CD105 and CD31, CD33, CD11b,
CD45, respectively) rather than any specific markers [15].
Mesenchymal stem cells, also denoted mesenchymal
stromal cells, indeed bear some similarities to fibroblasts.
Some studies suggested that these cells reside not only in
bone marrow but also in other organs, especially in the
perivascular space (or niche) [16]. Furthermore, fibroblasts
Table 1. Renal expression of some of the most widely used markers
of mesenchymal cells
Marker (human gene
symbol)
Expression Function
Vimentin (Vim) VSMCs Component of
cytoskeleton (class III
intermediate filament of
the desmin group)
Fibroblasts
Myofibroblasts
Pericytes
MSCs
Mesangial cells
Podocytes
Injured tubuli
Desmin (Des) VSMCs Component of
cytoskeleton (muscle-
specific class III
intermediate filament of
the desmin group)
Pericytes
Injured
podocytes
α-SMA α-smooth
muscle actin (Acta2)
VSMCs Major actin isoform of the
contractile cell apparatusMyofibroblasts
Pericytes
Injured
mesangial cells
PDGFR-βplatelet-
derived growth factor
receptor-β(PDGFRβ)
VSMCs Cell membrane receptor
for platelet-derived growth
factor (PDGF)—BB and—
DD
Fibroblasts
Myofibroblasts
Pericytes
MSCs
Mesangial cells
Macrophages
S100A4 S100 calcium-
binding protein A4
(S100A4) also termed
fibroblast-specific
protein 1 (FSP-1)
VSMCs Cytoplasmic and nuclear
protein, member of the
S100 family of proteins
with calcium-binding
motifs and broad cellular
functions
Fibroblasts
Myofibroblasts
Inflammatory
cells
Tubular cells
CD73 ecto-5’-
nucleotidase [5’-NT]
(Nt5e)
Fibroblasts Cell membrane enzyme
involved in conversion of
extracellular nucleotides to
membrane-permeable
nucleosides
Myofibroblasts
Pericytes
T-lymphocytes
Tubular cells
As can be seen, using one single marker does not allow a distinction
between various cell types. It should be mentioned that it is not yet clear
whether these markers are expressed in all cells of a certain type (i.e.
fibroblasts or pericytes) and to what extent the expression might change
during phenotypic alteration. See also Figure 1.
Nephrol Dial Transplant (2012): Editorial Review 3029
by guest on July 31, 2012http://ndt.oxfordjournals.org/Downloaded from

isolated from fibrotic kidneys were shown to be hetero-
geneous and a certain population exhibited some stem cell
features, such as clonal expansion or formation of 3D struc-
tures [17]. It is possible that some of the renal fibroblasts,
in particular perivascular fibroblasts, might be resident me-
senchymal stem cells. To distinguish between fibroblasts,
fibrocytes, monocytes and macrophages in human tissue,
their histological localization together with the analysis of
at least three to four markers was necessary [18].
In healthy kidney cortex, apart from vascular smooth
muscle cells (VSMCs), virtually no expression of α-SMA
can be found. In renal fibrosis, a striking up-regulation is
observed (Figure 1). This marker seems to be specific
enough to distinguish myofibroblasts from other mesench-
ymal cells if analysed strictly in the peritubular space
(Table 1and Figure 1). However, α-SMA is not expressed
by all myofibroblasts and some α-SMA-negative cells can
be myofibroblasts [19]. The expression of other markers
of myofibroblasts, e.g. of ED-A fibronectin, has not yet
been used extensively, although it is expressed in an
almost de novo fashion in human interstitial fibrosis [20].
S100A4, also termed fibroblast-specific protein 1 (FSP-1),
was shown to be expressed by leukocytes, which limits its
use as a marker for fibroblasts in renal fibrosis (Table 1)
[9]. Other markers for renal fibroblasts include cadherin 9
[21] while some still need to be examined, e.g. fibroblast
activation protein. The lack of specificmarkersledtoa
definition per exclusion. A renal fibroblast should not
express markers of epithelial (e.g. cytokeratins), endothelial
(e.g. von Willebrand factor), vascular (e.g. SM-22) and
inflammatory cells (e.g. CD45), a definition especially rel-
evant for cell characterization in in vitro studies. Such a
definition obviously does not differentiate between the
various populations of renal mesenchymal cells (Table 1).
The identification and characterization of mesenchymal
cells using immunohistological markers remains poor, as
was shown for myofibroblasts and pericytes [13,22]. At
present, the best way to distinguish between these cell
types is via ultrastructural analyses [13,22]. This is,
however, not feasible in our ‘daily’experimental studies.
Using collagen Iα1 promotor-driven green fluorescent
protein (GFP)-expressing reporter mice, it was found that
in fibrosis induced by unilateral ureter obstruction (UUO)
∼75% of the GFP-positive interstitial cells also expressed
α-SMA [23]. Similarly, other studies showed that,
depending on the stage of experimental fibrosis, renal in-
terstitial fibroblasts vary in their phenotype when assessed
by vimentin and α-SMA (co-) expression [23]. These
studies suggested that, especially in renal fibrosis, various
phenotypes or subtypes of fibroblasts and myofibroblasts
exist. This is well in line with the notion that fibroblasts
are a heterogeneous population of cells not only between
different organs but also within a given organ [24–27].
What do renal fibroblasts do?
The best established and most likely major role of fibro-
blasts is ECM production. Fibroblasts are essential in
shaping the structure of organs, providing structural and
mechanical support. On the basis of localization and ultra-
structure, it seems plausible that renal fibroblasts are
involved in the production of basement membranes of
tubules and capillaries and transmit mechanical and
biochemical signals between these structures. This might
indicate a pericyte-like and neuronal-like function of
fibroblasts (regarding the latter point, see the section con-
cerning the embryonic origin of fibroblasts). A role of
fibroblasts in the control of microcirculation is also sup-
ported by the expression of certain molecules affecting
haemodynamics, such as ecto-5′-nucleotidase (CD73) or
soluble guanylyl cyclase (cGC) [9].
Renal fibroblasts are the major producers of erythro-
poietin (EPO) [28–30]. Supporting the idea of renal fibro-
blasts as a heterogeneous group of cells, a recent study
showed that <20% of renal fibroblasts produce EPO [31].
It is not yet clear why a certain subset of interstitial fibro-
blasts is the major source of EPO and thereby the major
regulator of erythropoiesis for the whole organism. In
renal fibrosis, fibroblasts reduce or even lose their ability
to produce EPO, which is the major cause of renal
anaemia [31]. In vitro fibroblasts from healthy kidneys
express EPO mRNA, whereas myofibroblasts do not.
Stimulation of these myofibroblasts in vitro with various
factors, e.g. neurotrophins, brain-derived neurotrophic
factor, hepatocyte growth factor or lower concentrations
of dexamethasone is able to restore EPO production [31].
The localization of fibroblasts in the interstitium, i.e.
the space in which the transport of fluids and solutes
between tubules and capillaries takes place, implicates a
potential role in this process as well. In particular, this was
suggested for medullary fibroblasts. These cells can con-
tract in response to vasoactive substances and peptides, e.g.
prostaglandin E
2
or atrial natriuretic peptide [32,33].
As in other organs like dermis or lymph nodes
[24,34], fibroblasts interact with immune cells, in particu-
lar with resident dendritic cells and macrophages [35]. In
renal fibrosis, the interaction between the expanded popu-
lation of myofibroblasts and infiltrating inflammatory cells
is most likely reciprocally influenced by each other. For
example, infiltrating macrophages in fibrosis express
platelet-derived growth factor (PDGF)-C, which induces
chemokine expression in fibroblasts, thereby further
augmenting inflammatory infiltrates [36]. To date, there
exist only few data on the interaction of immune cells and
fibroblasts in healthy and diseased kidneys. This field de-
serves further attention, given the potential immunomodu-
latory function of fibroblasts [37].
It is possible that renal fibroblasts also have other func-
tions, which are not yet known. A subset of medullary
interstitial fibroblasts has lipid inclusions (vitamin
A-storing granules). These cells bear some similarities to
hepatic stellate cells, but the relevance of these cells in
the kidney is unclear [38,39]. It has been suggested that
these particular cells express a protein specific for acti-
vated hepatic stellate cells, the cytoglobin/stellate-cell
activation-associated protein (Cypg/STAP) [39]. Cypg/
STAP-expressing cells significantly accumulate in exper-
imental renal fibrosis but do not fully co-localize with
α-SMA [39].
3030 Nephrol Dial Transplant (2012): Editorial Review
by guest on July 31, 2012http://ndt.oxfordjournals.org/Downloaded from

The functions of the various renal fibroblast subsets
still remain to be characterized.
What is the embryonic source of renal fibroblast?
During mammalian embryogenesis, the later kidney,
termed metanephros, is formed from the ureteric bud and
metanephric mesenchyme (also termed metanephrogenic
mesenchyme or metanephrogenic blastema), both deriv-
ing from the intermediate mesoderm [40] (Figure 2). The
structures induce each other, i.e. the ureteric bud induces
the condensation of the metanephric mesenchyme and
induces glomerulo- and tubulogenesis. On the other hand,
the metanephric mesenchyme induces branching of the
ureteric bud and the formation of the collecting ducts.
The origin and development of the renal interstitium,
often called renal stroma, is less well understood. In the
following section, we will often use the term interstitium
(or stroma) referring to mesenchymal stromal cells but not
other stromal elements such as capillaries, vasculature or
resident inflammatory cells [41]. Some uncertainties in
the definitions exist in the embryonic kidney as well; the
precise definition of what all comprises the term ‘renal
stroma’is not completely clear [42].
The embryonic origin of the renal stroma, i.e. renal
fibroblasts, is still largely unknown. Several possible
sources exist (reviewed in [33,42,43]): first, it might
arise from the metanephric mesenchyme itself along with
tubulogenesis. Second, it might arise from the uninduced
intermediate mesenchyme in which the developing struc-
tures are embedded. Third, it might be derived from
different structures, e.g. the neural crest. And finally,
given the heterogeneous nature of fibroblasts, it is poss-
ible that these cells might be derived from more than one
of the aforementioned origins (Figure 2). The primary in-
terstitium of the developing kidney was shown to be dis-
tinct from the condensed metanephric mesenchyme (and
of course of the ureteric bud). These differences are both
morphological and molecular. Compared with condensed
metanephric mesenchyme, the primary interstitium was
shown to express tenascin, glycolipid disialoganglioside
G
D3
and, in particular, the winged helix transcription
factor Foxd1 (formerly termed BF-2) [33,41,44,45]. As
all structures develop, the primary interstitium forms
the progenitor population for the mature cortical and
medullary interstitium. The latter is characterized by the
appearance of interstitial cells with lipid inclusions
(vitamin A-storing cells) [33,42].
Descriptive studies showed that embryonic stromal
cells express neuronal markers [46,47]. This suggested
an extrarenal neural crest origin of renal fibroblasts,
recently supported by an intriguing study using reporter
mice [31]. What is the neural crest? Arising from the
neural tube, the neural crest is a multipotent and migratory
cell population that appears transiently during embryogen-
esis. Various cells and tissues develop from the neural
crest, e.g. smooth muscle cells, melanocytes or peripheral
glial cells and neurons. In the aforementioned study, to
mark and follow the cells of neural crest origin, the pro-
moter for myelin protein zero (P0) was used [31]. P0 is
expressed in the neural crest but not in the developing or
adult kidney [31]. The marked cells were interstitial,
showed no expression of inflammatory, dendritic or endo-
thelial cell markers, overlapped nearly completely with
PDGFR-βexpression and they also partly co-expressed
Foxd1, i.e. they appeared to be renal fibroblasts. In renal
fibrosis, these cells were the main, if not the only, source
of myofibroblasts. Using other neural crest reporter mice
(Wnt-1 promotor), migration of neural crest cells to the
developing kidney was also observed. However, these
cells remained in a capsule-like fashion in the early stages
of the metanephric kidney and largely disappeared there-
after with only very few positive intrarenal cells remain-
ing [48]. In mice with a neural crest defect (Splotch-
deficient mice) no expression of neural crest markers in
metanephros was observed and histologically these
kidneys and their interstitium were normal-appearing
[48]. This might suggest that there are various embryonic
sources of renal stroma. It should be mentioned that in
contrast to the aforementioned study [31], the expression
of P0 in mature murine podocytes and interstitial endo-
thelial cells was observed [49]. Taken together, the origin
of renal fibroblasts still remains to be clarified. But the
possible neuronal origin of these cells could facilitate
further studies of biology and present novel roles of renal
fibroblasts.
What is the role of the stroma cells in kidney develop-
ment? Similar to the reciprocal interactions between the
ureteric bud and metanephric mesenchyme, reciprocal
interactions between interstitial cells and the ureteric bud
and evolving tubules have been suggested [33,42,50].
The retinoic acid receptors (RARαand RARβ
2
, involved
in vitamin A signalling), fibroblast growth factor 7 (FGF-7)
and bone morphogenic protein 4 (BMP-4) are all expressed
in embryonic stromal cells and are essential for normal
ureteric bud development and branching [42]. Stromal
cells have also been suggested to play a role in nephro-
genesis [33,42]. Foxd1-deficient mice lacking the stromal
progenitors (Figure 2) develop only very few nephrons
compared with wild-type littermates [45].
The molecules driving the development of renal
stromal cells are largely unknown. To date, only some of
the factors have been described to be important in the
development and survival of embryonic stroma. These
include PDGF receptor αligands, PDGF-A and PDGF-C,
BMP-7 and molecules of the renin-angiotensin system
[33,41,42,51].
Future studies focusing on the origin and clues for the
development of renal stroma, i.e. of fibroblasts, could
introduce new tools for the specific manipulation of this
cell population, such as the Foxd1 mice [52].
What is the source of myofibroblasts in renal
fibrosis?
Fibrosis is characterized by exaggerated ECM deposition
and a striking accumulation of myofibroblasts (Figure 1).
The role and origin of myofibroblasts are therefore of
great interest. This is also one of the most controversially
Nephrol Dial Transplant (2012): Editorial Review 3031
by guest on July 31, 2012http://ndt.oxfordjournals.org/Downloaded from

discussed problems in the nephrological research field, as
obvious from a plethora of critical reviews on this matter
(for example, see [53–58]). We believe in the validity of
Occam’s razor, i.e. that the simplest explanation making
the fewest new assumptions is the most plausible one. In
our opinion, the simplest explanation is that renal fibro-
blasts are the main, if not the only, source of myofibro-
blasts in renal fibrosis. This has indeed been recently
shown in renal fibrosis in mice [31]. By renal fibroblasts,
we mean the broad term encompassing renal interstitial
mesenchymal cells (see above). It is likely that in the
future ‘renal fibroblasts’will comprise several distinct cell
groups, perhaps with different names, such as cells with
pericyte-like functions, perivascular fibroblasts (or even
mesenchymal stem cells), cells with neuronal-like func-
tions, EPO-producing cells and vitamin A-storing cells.
We also believe that if there is any epithelial or endo-
thelial source of myofibroblasts in renal fibrosis, this will
not be a major one. Studies analyzing the epithelial- and
endothelial-to-mesenchymal transition using genetic cell
fate tracking are summarized in Table 2and were re-
viewed in detail elsewhere [53–58]. Most recent studies
from various groups that specifically addressed this issue
using different methodologies did not find any indication
of an epithelial or endothelial source of myofibroblasts
[23,31,52,59–62]. This does not mean that phenotypic
alteration of endothelial or tubular epithelial cells does
not contribute to fibrosis. Phenotypic alteration in the
course of epithelial (or endothelial )-to-mesenchymal
transition or cell-cycle arrest might lead to changes in
their paracrine cell signalling which drives fibroblast pro-
liferation and ECM production and induces a myofibro-
blast phenotype. The role of circulating fibrocytes as a
contributor to renal fibroblast and myofibroblast formation
remains controversial as well (Table 3). We have
previously reviewed these issues in more detail [2].
Some recent studies suggested that the major source of
myofibroblasts in renal fibrosis is pericytes [23,52,55,
63]. By using marker expression such as PDGFR-β, it has
been suggested that, in the peritubular interstitium, the
mesenchymal cells are pericytes and not fibroblasts [23,
52,55,63]. There is currently no single marker that
would be able to discriminate a pericyte from other me-
senchymal cells [22] (Table 1). Other groups using ultra-
structural analysis did not describe pericytes in the renal
interstitium [9]. The simple definition of a pericyte is ‘a
cell of the connective tissue about capillaries or other
small blood vessels (http://www.merriam-webster.com/
medical/pericyte)’. Especially in the healthy renal intersti-
tium, such a definition could easily categorize most cells
Fig. 2. The origin of the renal stroma (interstitium). The nephrogenesis of metanephros is initiated when the metanephric mesenchyme (or
metanephric blastema) induces the outpouching of the ureteric bud from the Wollfian duct (A). The outgrowing ureteric bud induces condensation of
some of the metanephric mesenchyme, forming condensed mesenchyme or cap cells (B). These cells are Pax-2 positive and represent the nephron
progenitor cell population. Surrounding the cap cells are a more loose and morphologically distinct cell population of mesenchymal Foxd1-positive
cells. These cells are the progenitor population for the renal stroma. These cells might originate from metanephric mesenchyme but possibly also
from the neural crest (grey arrows). At later stages, with branching of the ureteric bud (C) the condensed mesenchyme undergoes mesenchymal-to-
epithelial induction to form renal vesicles from which glomeruli and tubules develop. At this stage, the Foxd1 cells migrate along the developing
nephron and form the primary interstitium. The primary interstitium then differentiates to the mature medullary and cortical interstitium. Whether
stromal cells arise directly from the metanephric mesenchyme, uninduced intermediate mesenchyme, neural crest or combination thereof is not clear.
Representative PAS-stained section of human embryonic kidney at 20 weeks of gestation is shown in D (original magnification ×200). The
schematics (A–C) were adopted from [42,43].
3032 Nephrol Dial Transplant (2012): Editorial Review
by guest on July 31, 2012http://ndt.oxfordjournals.org/Downloaded from

as pericytes. However, the true identification of pericytes
is currently possible only using ultrastructural analyses
[22], in which a mature pericyte is a cell completely or
partially embedded within the vascular basement mem-
brane [22,64]. Pericytes have processes that form contacts
with endothelial cells. These are of different types, e.g.
peg and socket, adhesion plaques and gap junctions
[22,64]. As described earlier, renal fibroblasts have been
described to form connections with endothelial basement
membrane and endothelial cells. To date, it is unknown
how many of the ‘renal interstitial cells’or ‘renal fibro-
blasts’might indeed be mature pericytes. Detailed ultra-
structural and immunogold studies that assess marker
panels to characterize and quantify various phenotypes of
renal mesenchymal cells are eagerly awaited and might
solve this question [65].
According to the ultrastructural definition, mature myo-
fibroblasts are tissue-contracting cells. They have impor-
tant functions in wound closure but also in the initiation
of revascularization and regeneration [66]. In organs such
as the kidney, such functions are not entirely clear. It is
possible, but not proven, that myofibroblasts normalize
biomechanical forces in injured organs thereby supporting
normal function in the uninjured areas and reducing
further organ damage. It has been shown that biomechani-
cal forces are crucial for correct cell differentiation and a
major trigger to induce a myofibroblast phenotype [19].
In experimental renal fibrosis, alteration of renal intratubu-
lar hydrodynamic forces leads to phenotypic alterations of
tubular cells towards a more profibrotic phenotype [67,
68]. Histology of human or experimental renal fibrosis
shows tubular atrophy, dilatation and loss (Figure 1). It is
well known that advanced fibrosis is characterized by
shrinkage of the kidneys. These data suggest that in renal
Table 2. Overview of studies using genetic cell fate tracking to analyze epithelial and endothelial source of renal (myo-)fibroblasts
Genetic approach Determination of transition
(colocalization with)
Animal model Study
Result
Tracing tubular cells β-galactosidase/anti-β-galactosidase
IHC with
FSP-1, HSP-47 IHC
UUO, bone marrow chimeras [80]
(β-glutamyl-transferase-driven Cre/
ROSA26LacZ)
Fifteen per cent of fibroblasts in UUO are bone
marrow derived
Thirty-six per cent of fibroblasts in UUO are derived
from tubular cells EMT)
Tracing tubular cells
(Pax8rtTA/LC1/ROSA26LacZ or Z/Red)
β-galactosidase/X-Gal with
Collagen 1 IHC
Overexpression of active TGF-βin tubular cells
(tetracyclin-induced)
[81]
No indication of tubular source of fibroblasts/EMT
Tracing tubular cells EYFP with UUO [62]
(Ksp-Cre/ROSA26-EYFP) FSP-1, α-SMA, entactin IHC No indication of tubular source of fibroblasts/EMT
Tracing tubular cells UUO, unilateral I/R [52]
(Six2-GFP-Cre/ROSA26LacZ or Z/Red) β-galactosidase/X-Gal, RFP with No indication of tubular source of fibroblasts/EMT in
any of the approaches(Hoxb7-Cre/ROSA26LacZ or Z/Red) FSP-1, α-SMA IHC
Tracing interstitial mesenchymal cells β-galactosidase/X-Gal, GFP with
(FoxD1-GFP-Cre-ER/ROSA26LacZ or Z/Red) PDGFR-β, CD73, α-SMA IHC Foxd1-traced interstitial cells form the majority of
myofibroblasts
Tracing interstitial mesenchymal cells ECFP with UUO, folic acid nephropathy, I/R [31]
(P0-Cre/ROSA26ECFP or ROSA26-tdRFP) PDGFR-β, CD73, α-SMA IHC No indication of tubular source of fibroblasts/EMT
P0-traced interstitial cells form the majority of
myofibroblasts
Tracing endothelial cells EYFP with UUO [82]
(Tie2-Cre/ROSA26-stop-EYFP) FSP-1, α-SMA IHC Some FSP-1 and fewer α-SMA-positive cells derived
from Tie2 traced cells
Tracing endothelial cells EGFP with STZ-induced diabetic nephropathy [83]
(Tie2-Cre/EGFP) α-SMA IHC Tie2-traced cells comprised 10% (at 1 month) and
24% (at 6 months of diabetic nephropathy) of
myofibroblasts
α-SMA, α-smooth muscle actin; EMT, epithelial-to-mesenchymal transition; FSP-1, fibroblast specific protein 1 (S100A4); HSP-47, heat shock protein
47; IHC, immunohistochemistry; I/R, ischemia/reperfusion injury model; STZ, streptozotocin; UUO, unilateral ureteral obstruction.
Table 3. Summary of the main conclusions of this review
Renal fibroblast is not a ‘simple mesenchymal spindle-shaped cell’, but
rather a term including heterogeneous population of renal interstitial
mesenchymal cells with various functions, e.g. EPO production.
The nomenclature of mesenchymal cells in general is not clearly defined,
mostly because an ultrastructural analysis is currently the only means of
identification and differentiation between various mesenchymal cells and
their transitional phenotypes.
We lack specific molecular markers that could specifically differentiate
renal mesenchymal cells and their different phenotypic subgroups.
The embryonic source of renal fibroblasts is yet unclear, some studies
suggesting a neural crest origin.
Renal fibroblasts are the major, if not the only, source of renal
myofibroblasts in renal fibrosis.
Most studies refer to the appearance of the so-called renal myofibroblasts
found in renal fibrosis, which in most studies is described solely by α-
SMA expression, as ‘differentiated’or ‘transformed’cells. However, it is
likely that these cells rather represent a ‘stressed’or ‘dedifferentiated’
mesenchymal cells that have lost some of their (differentiated) functions,
e.g. EPO production.
Nephrol Dial Transplant (2012): Editorial Review 3033
by guest on July 31, 2012http://ndt.oxfordjournals.org/Downloaded from

fibrosis significant alterations of biomechanical forces
take place, which might be a major signal for myofibro-
blast formation. This is supported by the observation that
in reversible acute kidney injury myofibroblasts form only
transiently during the period of tubular injury and dilata-
tion [69]. Another major signal for the induction of a
myofibroblast phenotype is paracrine signalling, including
the profibrotic molecules like TGF-β
1
[70]. Other signals
that may drive the myofibroblast induction in renal fibro-
sis are as yet unclear. The majority of interventions shown
to be antifibrotic were also associated with a reduction of
the number of α-SMA-expressing cells [3,36,70–73].
The molecules governing the phenotypic switch of fibro-
blasts are not yet well defined. In vitro, a wide variety of
molecules were found to be involved in proliferation of
(myo-)fibroblasts. These include TGF-β, PDGF, FGF-2,
connective tissue growth factor (CTGF), tissue plasmino-
gen activator (tPA), fibrinogen, potassium channel
KCa3.1, high glucose or the anti-mitogenic EP
4
and were
reviewed previously in more detail [2–4,70,71,74–76].
Some of the molecules involved in renal fibroblasts pro-
liferation might also be involved in their phenotypic
switch, and these processes might even be closely linked.
One could argue that, from a clinical point of view, the
source and precise phenotype of the expanded pool of in-
terstitial mesenchymal cells in renal fibrosis is irrelevant
as long as our intervention counteracts this process.
However, we believe that a better characterization und un-
derstanding of cell biology of the various renal fibroblast
subtypes might lead to more specific treatments or even
targeting approaches for certain subtypes of these cells.
What does the myofibroblast phenotype
in fibrotic kidneys represent?
Most of the studies of renal fibrosis describe the change
of a renal fibroblast into a myofibroblast phenotype as
‘differentiation’,‘transdifferentiation’,‘transformation’or
‘activation’. We hypothesize that most of the so-called
myofibroblasts actually represent stressed cells that have
lost their mature, differentiated phenotype. These cells
lose their functions, alter their phenotype and proliferate,
thereby resembling a non-mature state, i.e. they dediffer-
entiate to some degree. Such changes are well described
in other mesenchymal cell types, e.g. the VSMCs but also
in pericytes [77–79]. During injury, VSMCs lose the
expression of proteins characteristic of their contractile
function, thereby losing the ability to contract. Stress- or
injury-induced alterations of cell phenotype that resem-
bles the embryonic precursor were also proposed for renal
tubular cells. Tubular cells develop embryonically via me-
senchymal-to-epithelial transition from metanephric me-
senchyme (Figure 2). In renal fibrosis re-expression of
some mesenchymal markers in tubular cells has been ob-
served and termed epithelial-to-mesenchymal transition
(Figure 1)[2].
The same process most likely occurs in renal fibro-
blasts. In renal fibrosis, these cells lose their ability to
produce EPO [31]. Using detailed ultrastructural analyses
from nine patients with renal fibrosis, only one case pre-
sented with cells with features of mature myofibroblasts.
The other patients showed many different phenotypes of
myofibroblast-like cells [65]. Importantly, no typical
mature fibroblasts were found in these biopsies either
[65]. During embryonic development, the typical markers
of myofibroblasts, including α–SMA, vimentin or fibro-
nectin ectodomains, are expressed, but at later stages, they
are down-regulated or lost as the cells differentiate
[9,19]. Furthermore, myofibroblasts, and especially their
precursor forms, have also been characterized as ’rather
poor construction workers’[19].
Thus, at least part of what we term renal myofibro-
blasts, i.e. α-SMA-expressing interstitial cells, might be
dedifferentiated renal fibroblasts. Vice versa,only a frac-
tion of so-called myofibroblast indeed meets the criteria
of mature myofibroblasts. At first glance, this difference
might be purely semantic. We think that by acknowledging
this fact, we might move the research towards approaches
that might translate into novel therapeutic options for renal
fibrosis. For example, treatment that would re-establish
mature fibroblast phenotype could also lead to re-expression
of EPO, as was achieved in vitro [31].
What do we conclude and where do we go from
here?
The simple, inconspicuous light-microscopic appearance
of fibroblasts together with a lack of specific markers is
most likely responsible for the current uncertainties in the
nomenclature of renal fibroblasts, or more generally, of
renal interstitial mesenchymal cells. We are beginning to
understand that the term ‘fibroblast’, as currently used in
the literature, comprises a large and functionally impor-
tant population of differentiated and diverse mesenchymal
cells.
Identification of specific markers for fibroblasts, which
will allow their specific isolation and in vitro and in vivo
characterization, are one of the essential goals for future
research. Only a combination of anatomical localization,
shape and protein expression together with ultrastructural
analyses can distinguish between various mesenchymal
cell types and their various phenotypes. Specific markers
could also facilitate genetic animal studies allowing the
specific targeting of these cells. The obvious question is
whether there are such markers at all. Characterization of
the embryonic origin and the molecules that drive their
development are other issues deserving further studies.
The world of renal fibroblasts is open to be conquered.
Acknowledgements. We apologize to all authors whose important work
we could not cite because of space restrictions. We thank our colleagues
for helpful discussions and for critically reviewing the manuscript. This
study was supported by a financial research grant from the Deutsche For-
schungsgemeinschaft to P.B. (BO 3755/1-1) and J.F. (SFB TRR57, P17
and P19).
Conflict of interest statement. None declared.
3034 Nephrol Dial Transplant (2012): Editorial Review
by guest on July 31, 2012http://ndt.oxfordjournals.org/Downloaded from

References
1. Floege J, Johnson RJ, Feehally J. Comprehensive Clinical Nephrol-
ogy. St. Louis, MO: Elsevier Saunders, 2010
2. Boor P, Ostendorf T, Floege J. Renal fibrosis: novel insights into mech-
anisms and therapeutic targets. Nat Rev Nephrol 2010; 6: 643–656
3. Boor P, Sebekova K, Ostendorf T et al. Treatment targets in renal
fibrosis. Nephrol Dial Transplant 2007; 22: 3391–3407
4. Boor P, Floege J. Chronic kidney disease growth factors in renal fi-
brosis. Clin Exp Pharmacol Physiol 2011; 38: 391–400
5. Vleming LJ, de Fijter JW, Westendorp RG et al. Histomorphometric
correlates of renal failure in IgA nephropathy. Clin Nephrol 1998;
49: 337–344
6. Kalluri R, Zeisberg M. Fibroblasts in cancer. Nat Rev Cancer 2006;
6: 392–401
7. Virchow R. Die Cellularpathologie in Ihrer Begruendung auf Phy-
siologische und Pathologische Gewebelehre. Berlin, Germany:
Hirschwald, A, 1858
8. Duvall M. Atlas d’Embryologie. Paris, France: Masson, 1879
9. Kaissling B, Le Hir M. The renal cortical interstitium: morphological
and functional a spects. Histochem Cell Biol 2008; 130: 247–262
10. Niedermeier M, Reich B, Rodriguez Gomez M et al. CD4+ T cells
control the differentiation of Gr1+ monocytes into fibrocytes. Proc
Natl Acad Sci USA 2009; 106: 17892–17897
11. Keeley EC, Mehrad B, Strieter RM. Fibrocytes: bringing new in-
sights into mechanisms of inflammation and fibrosis. Int J Biochem
Cell Biol 2010; 42: 535–542
12. Majno G, Gabbiani G, Hirschel BJ et al. Contraction of granulation tissue
in vitro: similarity to smooth muscle. Science 1971; 173: 548–550
13. Eyden B. The myofibroblast: an assessment of controversial issues
and a definition useful in diagnosis and research. Ultrastruct Pathol
2001; 25: 39–50
14. Eyden B. The myofibroblast: phenotypic characterization as a prere-
quisite to understanding its functions in translational medicine.
J Cell Mol Med 2008; 12: 22–37
15. Phinney DG, Prockop DJ. Concise review: mesenchymal stem/mul-
tipotent stromal cells: the state of transdifferentiation and modes of
tissue repair—current views. Stem Cells 2007; 25: 2896–2902
16. Crisan M, Yap S, Casteilla L et al. A perivascular origin for me-
senchymal stem cells in multiple human organs. Cell Stem Cell
2008; 3: 301–313
17. Sommer M, Schaller R, Funfstuck R et al. Abnormal growth and
clonal proliferation of fibroblasts in an animal model of unilateral
ureteral obstruction. Nephron 1999; 82: 39–50
18. Pilling D, Fan T, Huang D et al. Identification of markers that dis-
tinguish monocyte-derived fibrocytes from monocytes, macro-
phages, and fibroblasts. PLoS ONE 2009; 4: e7475
19. Hinz B. The myofibroblast: paradigm for a mechanically active cell.
J Biomech 2010; 43: 146–155
20. Van Vliet A, Baelde HJ, Vleming LJ et al. Distribution of fibronec-
tin isoforms in human renal disease. J Pathol 2001; 193: 256–62
21. Thedieck C, Kalbacher H, Kuczyk M et al. Cadherin-9 is a novel
cell surface marker for the heterogeneous pool of renal fibroblasts.
PLoS ONE 2007; 2: e657
22. Armulik A, Genove G, Betsholtz C. Pericytes: developmental, phys-
iological, and pathological perspectives, problems, and promises.
Dev Cell 2011; 21: 193–215
23. Lin SL, Kisseleva T, Brenner DA et al. Pericytes and perivascular
fibroblasts are the primary source of collagen-producing cells in ob-
structive fibrosis of the kidney. Am J Pathol 2008; 173: 1617–1627
24. Sorrell JM, Caplan AI. Fibroblasts-a diverse population at the center
of it all. Int Rev Cell Mol Biol 2009; 276: 161–214
25. Schmitt-Graff A, Desmouliere A, Gabbiani G. Heterogeneity of
myofibroblast phenotypic features: an example of fibroblastic cell
plasticity. Virchows Arch 1994; 425: 3–24
26. Sorrell JM, Caplan AI. Fibroblast heterogeneity: more than skin
deep. J Cell Sci 2004; 117(Pt 5): 667–675
27. Chang HY, Chi JT, Dudoit S et al. Diversity, topographic differen-
tiation, and positional memory in human fibroblasts. Proc Natl Acad
Sci USA 2002; 99: 12877–12882
28. Bachmann S, Le Hir M, Eckardt KU. Co-localization of erythropoie-
tin mRNA and ecto-5′-nucleotidase immunoreactivity in peritubular
cells of rat renal cortex indicates that fibroblasts produce erythro-
poietin. J Histochem Cytochem 1993; 41: 335–341
29. Maxwell PH, Osmond MK, Pugh CW et al. Identification of the
renal erythropoietin-producing cells using transgenic mice. Kidney
Int 1993; 44: 1149–1162
30. Paliege A, Rosenberger C, Bondke A et al. Hypoxia-inducible
factor-2alpha-expressing interstitial fibroblasts are the only renal
cells that express erythropoietin under hypoxia-inducible factor
stabilization. Kidney Int 2010; 77: 312–318
31. Asada N, Takase M, Nakamura J et al. Dysfunction of fibroblasts of
extrarenal origin underlies renal fibrosis and renal anemia in mice.
J Clin Invest 2011; 121: 3981–3990
32. Zhuo JL. Renomedullary interstitial cells: a target for endocrine and
paracrine actions of vasoactive peptides in the renal medulla.
Clin Exp Pharmacol Physiol 2000; 27: 465–473
33. Alcorn D, Maric C, McCausland J. Development of the renal inter-
stitium. Pediatr Nephrol 1999; 13: 347–354
34. Lukacs-Kornek V, Malhotra D, Fletcher AL et al. Regulated release
of nitric oxide by nonhematopoietic stroma controls expansion of
the activated T cell pool in lymph nodes. Nat Immunol 2011; 12:
1096–1104
35. Fine LG, Norman JT, Ong A. Cell-cell cross-talk in the
pathogenesis of renal interstitial fibrosis. Kidney Int Suppl 1995; 49:
S48–S50
36. Eitner F, Bucher E, van Roeyen C et al. PDGF-C is a proinflamma-
tory cytokine that mediates renal interstitial fibrosis. J Am Soc
Nephrol 2008; 19: 281–289
37. Cappellesso-Fleury S, Puissant-Lubrano B, Apoil PA et al.
Human fibroblasts share immunosuppressive properties with bone
marrow mesenchymal stem cells. J Clin Immunol 2010; 30:
607–619
38. Yamamoto M, Enzan H, Hara H et al. Fluorescence and electron
microscopic studies on the perivascular mesenchymal cells and
fibroblasts after vitamin A administration. Acta Pathol Jpn 1978;
28: 513–521
39. Kida Y, Asahina K, Inoue K et al. Characterization of vitamin A-
storing cells in mouse fibrous kidneys using Cygb/STAP as a marker
of activated stellate cells. Arch Histol Cytol 2007; 70: 95–106
40. Saxén L. Organogenesis of the Kidney. Cambridge, Cambridge:
University Press, 1987
41. Sequeira Lopez ML, Gomez RA. Development of the renal arter-
ioles. J Am Soc Nephrol 2011; 22: 2156–2165
42. Cullen-McEwen LA, Caruana G, Bertram JF. The where, what and
why of the developing renal stroma. Nephron Exp Nephrol 2005;
99: e1–e8
43. Dressler GR. The cellular basis of kidney development. Annu Rev
Cell Dev Biol 2006; 22: 509–529
44. Levinson RS, Batourina E, Choi C et al. Foxd1-dependent signals
control cellularity in the renal capsule, a structure required for
normal renal development. Development 2005; 132: 529–539
45. Hatini V, Huh SO, Herzlinger D et al. Essential role of stromal me-
senchyme in kidney morphogenesis revealed by targeted disruption
of Winged Helix transcription factor BF-2. Genes Dev 1996; 10:
1467–1478
46. Sainio K, Nonclercq D, Saarma M et al. Neuronal characteristics in
embryonic renal stroma. Int J Dev Biol 1994; 38: 77–84
47. Sariola H, Holm K, Henke-Fahle S. Early innervation of the meta-
nephric kidney. Development 1988; 104: 589–599
48. Itaranta P, Viiri K, Kaartinen V et al. Lumbo-sacral neural crest
derivatives fate mapped with the aid of Wnt-1 promoter integrate
but are not essential to kidney development. Differentiation 2009;
77: 199–208
49. Plaisier E, Mougenot B, Verpont MC et al. Glomerular permeability
is altered by loss of P0, a myelin protein expressed in glomerular
epithelial cells. J Am Soc Nephrol 2005; 16: 3350–3356
50. Sariola H, Sainio K. Cell lineages in the embryonic kidney: their
inductive interactions and signalling molecules. Biochem Cell Biol
1998; 76: 1009–1016
Nephrol Dial Transplant (2012): Editorial Review 3035
by guest on July 31, 2012http://ndt.oxfordjournals.org/Downloaded from

51. Ding H, Wu X, Bostrom H et al. A specific requirement for PDGF-
C in palate formation and PDGFR-alpha signaling. Nat Genet 2004;
36: 1111–111 6
52. Humphreys BD, Lin SL, Kobayashi A et al. Fate tracing reveals the
pericyte and not epithelial origin of myofibroblasts in kidney fibro-
sis. Am J Pathol 2010; 176: 85–97
53. Kriz W, Kaissling B, Le Hir M. Epithelial-mesenchymal transition
(EMT) in kidney fibrosis: fact or fantasy? J Clin Invest 2011; 121:
468–474
54. Quaggin SE, Kapus A. Scar wars: mapping the fate of epithelial-
mesenchymal-myofibroblast transition. Kidney Int 2011; 80: 41–50
55. Grgic I, Duffield JS, Humphreys BD. The origin of interstitial myo-
fibroblasts in chronic kidney disease. Pediatr Nephrol 2012; 27:
183–193
56. Liu Y. New insights into epithelial-mesenchymal transition in
kidney fibrosis. J Am Soc Nephrol 2010; 21: 212–222
57. Fragiadaki M, Mason RM. Epithelial-mesenchymal transition in
renal fibrosis—evidence for and against. Int J Exp Pathol 2011; 92:
143–150
58. Zeisberg M, Duffield JS. Resolved: EMT produces fibroblasts in the
kidney. J Am Soc Nephrol 2010; 21: 1247–1253
59. Picard N, Baum O, Vogetseder A et al. Origin of renal myofibro-
blasts in the model of unilateral ureter obstruction in the rat. Histo-
chem Cell Biol 2008; 130: 141–155
60. Humphreys BD, Valerius MT, Kobayashi A et al. Intrinsic
epithelial cells repair the kidney after injury. Cell Stem Cell 2008; 2:
284–291
61. Faulkner JL, Szcykalski LM, Springer F et al.Origin of interstitial
fibroblasts in an accelerated model of angiotensin II-induced renal
fibrosis. Am J Pathol 2005; 167: 1193–1205
62. Li L, Zepeda-Orozco D, Black R et al.. Autophagy is a component
of epithelial cell fate in obstructive uropathy. Am J Pathol 2010;
176: 1767–1778
63. Schrimpf C, Duffield JS. Mechanisms of fibrosis: the role of the
pericyte. Curr Opin Nephrol Hypertens 2011; 20: 297–305
64. Diaz-Flores L, Gutierrez R, Varela H et al. Microvascular pericytes:
a review of their morphological and functional characteristics. Histol
Histopathol 1991; 6: 269–286
65. Ru Y, Eyden B, Curry A et al. Actin filaments in human renal
tubulo-interstitial fibrosis: significance for the concept of epithelial-
myofibroblast transformation. J Submicrosc Cytol Pathol 2003; 35:
221–233
66. Kilarski WW, Samolov B, Petersson L et al. Biomechanical regu-
lation of blood vessel growth during tissue vascularization. Nat Med
2009; 15: 657–664
67. Rohatgi R, Flores D. Intratubular hydrodynamic forces influence tu-
bulointerstitial fibrosis in the kidney. Curr Opin Nephrol Hypertens
2010; 19: 65–71
68. Li L, Zepeda-Orozco D, Patel V et al. Aberrant planar cell polarity
induced by urinary tract obstruction. Am J Physiol Renal Physiol
2009; 297: F1526–F1533
69. Fujigaki Y, Muranaka Y, Sun D et al. Transient myofibroblast differ-
entiation of interstitial fibroblastic cells relevant to tubular dilatation
in uranyl acetate-induced acute renal failure in rats. Virchows Arch
2005; 446: 164–176
70. Boor P, Celec P, Martin IV et al. The peroxisome proliferator-acti-
vated receptor-alpha agonist, BAY PP1, attenuates renal fibrosis in
rats. Kidney Int 2011; 80: 1182–1197
71. Boor P, Konieczny A, Villa L et al. PDGF-D inhibition by CR002
ameliorates tubulointerstitial fibrosis following experimental glomer-
ulonephritis. Nephrol Dial Transplant 2007; 22: 1323–1331
72. Boor P, Konieczny A, Villa L et al. Complement C5 mediates
experimental tubulointerstitial fibrosis. J Am Soc Nephrol 2007; 18:
1508–1515
73. Ostendorf T, Rong S, Boor P et al. Antagonism of PDGF-D by
human antibody CR002 prevents renal scarring in experimental glo-
merulonephritis. J Am Soc Nephrol 2006; 17: 1054–1062
74. Grgic I, Kiss E, Kaistha BP et al. Renal fibrosis is attenuated by
targeted disruption of KCa3.1 potassium channels. Proc Natl Acad
Sci USA 2009; 106: 14518–14523
75. Sorensen I, Susnik N, Inhester T et al. Fibrinogen, acting as a
mitogen for tubulointerstitial fibroblasts, promotes renal fibrosis.
Kidney Int 2011; 80: 1035–1044
76. Nakagawa N, Yuhki KI, Kawabe JI et al. The intrinsic prostaglandin
E(2)-EP(4) system of the renal tubular epithelium limits the develop-
ment of tubulointerstitial fibrosis in mice. Kidney Int 2012 doi:
10.1038/ki.2012.115
77. Owens GK. Regulation of differentiation of vascular smooth muscle
cells. Physiol Rev 1995; 75: 487–517
78. Song Z, Li G. Role of specific microRNAs in regulation of vascular
smooth muscle cell differentiation and the response to injury. J Car-
diovasc Transl Res 2010; 3: 246–250
79. Olson LE, Soriano P. PDGFRbeta signaling regulates mural
cell plasticity and inhibits fat development. Dev Cell 2011; 20:
815–826
80. Iwano M, Plieth D, Danoff TM et al. Evidence that fibroblasts
derive from epithelium during tissue fibrosis. J Clin Invest 2002;
110: 341–350
81. Koesters R, Kaissling B, Lehir M et al. Tubular overexpression of
transforming growth factor-beta1 induces autophagy and fibrosis but
not mesenchymal transition of renal epithelial cells. Am J Pathol
2010; 177: 632–643
82. Zeisberg EM, Potenta SE, Sugimoto H et al. Fibroblasts in kidney
fibrosis emerge via endothelial-to-mesenchymal transition. J Am Soc
Nephrol 2008; 19: 2282–2287
83. Li J, Qu X, Bertram JF. Endothelial-myofibroblast transition contrib-
utes to the early development of diabetic renal interstitial fibrosis in
streptozotocin-induced diabetic mice. Am J Pathol 2009; 175:
1380–1388
Received for publication: 22.2.2012; Accepted in revised form:
22.4.2012
3036 Nephrol Dial Transplant (2012): Editorial Review
by guest on July 31, 2012http://ndt.oxfordjournals.org/Downloaded from