Transcriptional regulators in kidney disease: gatekeepers of ...
Post on 12-May-2015
656 Views
Preview:
Transcript
TIGS-649; No of Pages 11
Transcriptional regulators in kidneydisease: gatekeepers of renalhomeostasisN. Henriette Uhlenhaut and Mathias Treier
Developmental Biology Unit, European Molecular Biology Laboratory, 69117 Heidelberg, Germany
Review
Glossary
Anosmia: absence of the ability to smell.
Autosomal dominant polycystic kidney disease (ADPKD): hereditary disorder
characterized by the presence of multiple fluid-filled cysts inside enlarged kidneys;
cysts arise from all nephron segments.
Autosomal recessive polycystic kidney disease (ARPKD): recessively inherited
disease similar to ADPKD, but cysts derive primarily from dilation of the collecting
tubule.
Bardet-Biedl syndrome (BBS): human genetic disorder including kidney abnorm-
alities, retinal degeneration, mental retardation, obesity, diabetes, polydactyly and
situs inversus.
Chronic kidney disease (CKD): progressive decline of kidney function, leading to
kidney failure.
Collecting duct: epithelial tubes derived from the branched ureteric bud, draining
the urine from the nephrons into the renal papilla.
End-stage renal disease (ESRD): kidney failure (the need for dialysis).
Epithelial-to-mesenchymal transition (EMT): process during which epithelial cells
acquire mesenchymal, fibroblast-like properties (with reduced intercellular adhe-
sion and increased motility); the reverse process is called mesenchymal-to-epithelial transition (MET).Fibrosis: formation of scar (fibrous) tissue.
Glomerulosclerosis: scarring of the glomerulus, which impairs the filtration
process; often observed in chronic kidney disease.
Glomerulus: a small group of looping blood vessels surrounded by Bowman’s
capsule where the blood is filtered and urine is formed; consists of podocytes,
endothelial and mesangial cells; see Figure I.
Lymphedema-Distichiasis syndrome: a condition that affects the function of the
lymphatic system resulting in extra eyelashes and tissue swelling.
Mesangial cells: supportive cells within the glomerulus.
Metanephric mesenchyme (MM): an aggregate of mesenchymal cells in the
embryo from which renal stroma and nephrons originate.
Nephron: the basic functional unit of the kidney, consisting of glomerulus,
proximal tubule, loop of Henle, distal tubule, and collecting duct.
Nephronophthisis (NPHP): autosomal recessive disorder affecting juveniles;
progressive deterioration of kidney function, with medullary cysts, tubular
degeneration and fibrosis. Phthisis (Greek) = a dwindling or wasting away.
Nephropathy: kidney disease.
Podocyte: epithelial cells in the glomerulus that form part of the filtration barrier.
Polyuria: excessive excretion of urine.
Proteinuria: protein in the urine, a sign of renal dysfunction.
Renal agenesis: absence of kidneys caused by a developmental defect.
Renal dysplasia: abnormal development of the kidneys (with regard to size and
shape).
Renal replacement therapy: life-supporting treatments for renal failure, such as
dialysis and kidney transplantation.
Renal tubule: the elongated, tube-like part of the nephron, made up of the promixal
and distal convoluted tubules and the loop of Henle.
Situs inversus: a condition in which the inner organs are arranged in a perfect
mirror image reversal of the normal positioning
Slit diaphragm: transmembrane structure between podocyte foot processes
creating a sieve.
Although we are rapidly gaining a more completeunderstanding of the genes required for kidney function,the molecular pathways that actively maintain organhomeostasis are only beginning to emerge. The studyof the most common genetic cause of renal failure,polycystic kidney disease, has revealed a surprising rolefor primary cilia in controlling nuclear gene expressionand cell division during development as well as main-tenance of kidney architecture. Conditions that disturbkidney integrity seem to be associated with reversal ofdevelopmental processes that ultimately lead to kidneyfibrosis and end-stage renal disease (ESRD). In thisreview, we discuss transcriptional regulators and net-works that are important in kidney disease, focusingon those that mediate cilia function and drive renalfibrosis.
Chronic kidney disease: an emerging pandemic healthproblemOur kidneys serve important physiological functions toexcrete waste products from the body and balance the bodyfluids. Disturbance in the renal filtration process can pose aserious health threat. Currently, 10% of the global adultpopulation, regardless of ethnic origin, are affected bychronic kidney disease. An estimated 1.5 million patientsare dependent on renal replacement therapy [1,2].
For many years, the kidney has been a classical modelfor studying organogenesis (see Boxes 1 and 2 for a briefoverview). Research on human genetic disorders associ-ated with renal malfunction, in conjunction with theanalysis of gene disruption studies inmice and othermodelorganisms, has led to the identification of many transcrip-tion factors required for kidney development and homeo-stasis (Box 2). Despite the wealth of information available,we are still far from an integrative picture of the regulatorynetworks that maintain the integrity of themature kidney.Modulation of signaling pathway strength by extrinsicfactors results in gene expression program changes thatprecede the observed morphological alterations in chronickidney disease. Thus, understanding the transcriptionalnetworks that maintain cell type identity in the maturekidney provides a promising entry point for future thera-peutic interventions [3]. Many forms of chronic kidneydisease (i.e. diabetic nephropathy, nephronophthisis) are
Corresponding author: Treier, M. (treier@embl.de).
0168-9525/$ – see front matter � 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.tig.2008.0
associated with the development of glomerulosclerosisor interstitial fibrosis that result from the reversal ofdevelopmental processes that were originally used to buildthe kidney. Therefore, some of the pathophysiological
Ureteric bud: epithelial protrusion (outpouching) from the Wolffian/nephric duct
that invades the metanephric mesenchyme during kidney development and gives
rise to the collecting duct system.
5.001 Available online xxxxxx 1
Table 1. Transcription factors associated with human renal disease
Gene Murine kidney phenotype Human syndrome OMIM
EYA1 Agenesis, no metanephric mesenchyme specified Bbranchio-otorenal syndrome: hearing loss, ear pits, branchial cysts
or fistulas, kidney anomaliesa
601653
FOXC1 Positioning of ureteric bud affected Axenfeld-Rieger syndrome (congenital ocular disorder) and
congenital anomalies of the kidney and urinary tracta
601090
FOXC2 Kidney hypoplasia, glomerular defects, ureteric bud
positioning affected
Lymphedema-Distichiasis syndrome with renal disease and diabetes
mellitusa
602402
GATA3 No ureteric branching, no metanephric differentiation HDR syndrome: hypoparathyroidism, sensorineural deafness and
renal diseasea
131320
GLI3 Expression of a truncated instead of full-length Gli3
protein mimics Pallister-Hall syndrome; kidney
development is abnormal
Pallister-Hall syndrome: fingers and toes affected, benign
hypothalamic tumors, associated with renal anomalies (and others)a
165240
GLIS2 Kidney degeneration and cysts Nephronophthisis 608539
GLIS3 Polycystic kidneys Diabetes mellitus, hypothyroidism and polycystic kidneys 610192
HNF1b Cystic kidneys Renal cysts and diabetes syndrome 189907
LMX1b Impaired podocyte differentiation Nail Patella syndrome, including nephropathya 602575
NFIA Ureteral and renal defects Central nervous system malformations and urinary tract defectsa 600727
PAX2 Agenesis, nephric duct is not maintained Renal-Coloboma syndrome: optic nerve abnormalities,
vesicoureteral reflux and renal hypoplasiaa
167409
SALL1 Agenesis, no ureteric bud outgrowth Townes-Brocks syndrome: abnormalities in thumbs, feet, heart, ears
(impaired hearing), kidney; imperforate anusa
602218
SALL4 Renal hypoplasia (agenesis in Sall1/Sall 4 compound
heterozygotes)
Duane-Radial Ray (or Okihiro) syndrome associated with kidney
defectsa
607343
SIX1 No ureteric bud invasion, metanephric mesenchyme
is not sustained
Branchio-otorenal) syndromea 601205
SIX5 Unknown Branchio-otorenal) syndromea 600963
WT1 No kidneys Wilm’s tumor, Wilm’s tumor-aniridia-genitourinary anomalies-
mental retardation syndrome, Denys-Drash syndrome, Frasier
syndrome, diffuse mesangial sclerosisa
607102
aIn humans, many of these syndromes are caused by haploinsufficiency, whereas the phenotype of many mouse mutants only manifests itself in the homozygous situation.
Review Trends in Genetics Vol.xxx No.x
TIGS-649; No of Pages 11
processes observed in chronic kidney disease are betterunderstood in light of normal kidney development. Sur-prisingly, many transcriptional regulators that whenmutated cause congenital abnormalities of the kidney, suchas renal agenesis, renal hypoplasia, dysplasia or ureticmalformations, are reactivated during chronic kidney dis-eases (summarized in Table 1 and Box 2; for a review, seeRef. [4]).
Here we focus on emerging evidence that suggests thatdevelopmental gene expression programs are reactivatedin chronic renal disease, pointing toward an underappre-ciated cellular plasticity of renal cells. Furthermore, wediscuss recent findings that localization and processing oftranscriptional regulators at the cilium are crucial for themaintenance of kidney homeostasis.
Cystic kidney diseases are ciliopathiesCystic renal diseases are the most common genetic cause ofend-stage renal disease (ESRD) [2]. Among children,nephronophthisis (NPHP) and autosomal recessive polycys-tic kidney disease (ARPKD) are predominant, whereasautosomal dominant polycystic kidney disease (ADPKD)prevails among adults. Mutations in the polycystin-1(PKD1) and polycystin-2 (PKD2) genes are responsible formost ADPKD cases, whereas PKHD1 mutations accountpredominantly for ARPKD. The major steps of early kidneydevelopment proceed normally in patients with polycystickidney disease, but terminal differentiation (i.e. the epi-thelial morphology of renal cells) cannot be sustained,resulting in the functional tubular architecture being dis-rupted by fluid-filled cysts. Several genes responsible forcyst development have been identified by positional cloningin human patients, and the subsequent study of these
2
proteinshas led to theunifying theory thatpolycystic kidneydiseases are associated with a defect in primary cilia func-tion [5,6].
Nearly all nondividing cells of the body extend a singleprimary, nonmotile cilium into the extracellular space.Primary cilia are sensory organelles involved in photore-ception, olfaction and mechanosensation. They consist of acentral axoneme built of microtubules, covered by aspecialized plasma membrane. Ciliary function dependson intraflagellar transport (IFT) along their microtubules,because there is no protein synthesis within the ciliumitself [7,8] (Figure 1).
The membrane proteins encoded by PKD1, PKD2 andPKHD1 as well as the NPHP and Bardet-Biedl-syndrome(BBS) proteins localize to primary cilia and/or basal bodiesor centrosomes. Consequently, mislocalisation of theseproteins caused by absence of renal cilia or disruption ofcilia function causes cyst formation [5]. Elegant geneticexperiments in adult mice using a tamoxifen-inducible Crerecombinase system have shown that loss of cilia throughdisruption of IFT leads to slow onset cystic kidney disease[9]. A similarly designed study that induced PKD1 inacti-vation in the mature kidney reported the same findings[10]. Although the progression of the disease phenotype inboth cases was less aggressive than that observed afterdisruption of IFT or PKD1 during development, theseresults nevertheless suggest that cilia function is requiredto actively maintain kidney architecture throughout life.
In addition to polycystic kidney disease, people withcilia-associated genetic disorders often present withretinal degeneration, liver cysts and fibrosis, anosmia,situs inversus and systemic phenotypes such as obesity,diabetes, heart defects, hypertension, as well as skeletal
Review Trends in Genetics Vol.xxx No.x
TIGS-649; No of Pages 11
and neurological defects [11]. Dysfunction of motile ciliasuch as the sperm’s flagellum or the ones lining our air-ways and fallopian tubes can lead to laterality defects,respiratory problems, infertility and hydrocephalus [6].
The function of the cilium in hedgehog signalingMouse genetic data have provided compelling evidencethat apart from their role as sensory organelles, cilia areessential for Hedgehog (Hh) signal transduction. Mam-malian Hh signaling is mediated by three Gli-Kruppeltranscription factor family members (Gli1–3). Without aHh signal, Patched (the Hedgehog receptor) inhibitsSmoothened, and the Gli transcription factors are cleavedto repressor forms, whereas presence of a Hh signalrelieves this inhibition, resulting in full-length Gliproteins entering the nucleus and serving as transcrip-tional activators (reviewed in Refs. [12,13]). Evidence for arole of Sonic hedgehog (Shh) in kidney function has comefrom the analysis of human patients with renal malfor-mations associated with Pallister-Hall syndrome(Table 1), which is caused by mutations in the GLI3 gene.Thesemutations are thought to result in a truncatedGLI3protein, and mice expressing only this shortened formhave defective kidney development [12,14].
Mice with homozygous mutations for Shh or thosetreated with cyclopamine (a Hedgehog antagonist), showsevere disruption of renal organogenesis, in theworst casesresulting in a complete lack of kidneys. In the absence ofHh signaling, reduced expression of Pax2, Sall1, cell cycleregulators (CyclinD1 and N-Myc), Gli1 and Gli2 wasobserved. Intriguingly, Gli3 expression was increased inthe absence of Shh, and deletion of Gli3 rescued the renalphenotype of Shh�/� mice, restoring expression of Pax2,Sall1, CyclinD1, N-Myc, Gli1 and Gli2 [15]. These resultstogether strongly suggest that these genes are direct tar-gets of the Gli transcription factors, whereby Gli1 and 2 actas transcriptional activators and Gli3 predominantly func-tions as a transcriptional repressor in the absence of Shh.
Box 1. An overview of kidney development
Mammalian kidney development proceeds through three successive
steps, termed the pronephros, the mesonephros and the metane-
phros. The pronephros is formed first and most rostrally (at E8.5 in
the mouse), followed by a more caudal formation of the mesonephros
and finally the metanephros (at E10.5). The first two are transient
structures in mammals, and only the metanephros will become the
definite adult kidney [74,75] (see Figure I).
Kidney development is characterized by sequential reciprocal
inductive interactions and mesenchymal-to-epithelial transforma-
tions. It is initiated with the formation of the nephric ducts, or
Wolffian ducts (at E8 in the mouse), two epithelial tubes that originate
from the intermediate mesoderm and extend toward the posterior of
the embryo. Metanephros development depends on interactions
between the metanephric mesenchyme (MM), a specialized group
of kidney precursor cells, and the nephric duct. At the level of the
hindlimb bud, signaling from the MM induces evagination of the
ureteric bud (UB) from the nephric duct, which will invade the MM
and form multiple branches. These branches will later form the
collecting duct system that funnels the urine into the bladder. At the
tips of the branching ureter, the surrounding mesenchyme is induced
to condense, epithelialize and differentiate into mature nephrons. The
differentiation of nephrons occurs through morphogenetic stages that
are referred to as renal vesicles and comma- and S-shaped bodies
These studies have illuminated themolecular mechanismsdownstream of Hh signaling in kidney development andcan explain the malformations found in Pallister-Hallsyndrome that are thought to be caused by expression ofa dominant-negative GLI3 repressor.
The dependence of Hh signaling on ciliary localizationleads to the question as to how the signal is transmittedfrom the cilium to the nucleus. In a surprising twist to theHh story, Patched and Gli transcription factors were foundto localize in the cilium, and Smoothened was shown tomove into the cilium in response to theHh ligand, resultingin processing of the Gli transcription factors within thecilium. Consequently, defects in IFT, or other ciliary func-tions, can lead to disruption of Hh signaling [13,16].
Glis factors: novel players in renal diseaseFurther clues as to how signaling within and from theprimary cilium funnels into nuclear gene expression tomaintain polarity of renal epithelial cells in the maturekidney are now emerging. Recent evidence suggests thatloss of function mutations in the murine zinc finger tran-scription factor Gli-similar 2 (Glis2), as well as in humanGLIS2, lead to nephronophthisis [17,18]. Glis2 is one ofthree Glis transcription factor family members whose zincfingers share high homology with the Gli family. Like theGli proteins, Glis2 is detected in the primary cilium, but itis not required for proper cilia formation. Interestingly,Glis2 seems to act as a transcriptional repressor by bindingto Gli-consensus DNA binding sites and suppressing devel-opmental gene expression programs once kidney architec-ture is established. Notably, all three Glis family membersare expressed in overlapping patterns in the adult mousekidney. Furthermore, mutations in GLIS3 have beenshown to cause polycystic kidneys in humans [19] andmice (Uhlenhaut and Treier, unpublished observations).Thus, Glis proteins in the adult kidney might be part of atranscriptional network downstream of cilia signalingrequired for kidney integrity. It will be important to deter-
until they finally connect with the collecting duct. This system of
branching and differentiation is reiterated until nephrogenesis is
completed shortly after birth in mice (reviewed in Refs. [74,75]).
A mature nephron consists of highly specialized cell types carrying
out various physiological functions associated with waste excretion
from the blood. One end of the nephron is formed by the glomerulus,
followed by the proximal convoluted tubule, Henle’s loop and the
distal convoluted tubule, which inserts into the collecting duct. The
initial filtration of the blood occurs inside the glomerulus, where
fenestrated endothelial capillary cells and glomerular podocytes
create a filter (reviewed in Ref. [41]). This initial filtrate is concentrated
by selective reabsorption along the different segments of the renal
tubule. Glucose, amino acids, electrolytes and peptides are reab-
sorbed in the proximal convoluted tubule, whereas water and
electrolytes are taken up by Henle’s loop and the distal convoluted
tubule, because of the segmental expression of distinct sets of solute
transporters [75]. Defects in the podocyte foot processes enveloping
the glomerular capillaries or in the basement membrane between the
two cell types usually lead to protein loss into the urine (proteinuria).
Accordingly, malfunction of the above mentioned tubular segments
generally results in abnormally high levels of the corresponding
molecules (e.g. glucose or electrolytes) within the urine, polyuria and
dehydration.
3
Figure I. Schematic overview of mammalian kidney development and morphology. Kidney development proceeds through three stages: pronephros, mesonephros and
metanephros. (a) The metanephros becomes the permanent kidney and is formed by invasion and branching of the ureteric bud from the nephric duct into the
metanephric mesenchyme in a process of reciprocal inductive interactions. (b) The ureteric branches give rise to the collecting duct system and induce epithelialization
of the surrounding mesenchyme, first creating renal vesicles and then comma- and S-shaped bodies and finally nephrons that join the collecting duct. (c) The nephron
Review Trends in Genetics Vol.xxx No.x
TIGS-649; No of Pages 11
4
Box 2. Transcription factors in early kidney development
The transcriptional networks governing early kidney development have
been reviewed extensively elsewhere [4,74–76]. To summarize briefly,
many of the early specification events center around the activation of
glial cell line-derived neurotrophic factor (GDNF), a crucial signaling
molecule that is secreted by the metanephric mesenchyme (MM) to
induce budding and branching of the ureter. GDNF signals through its
receptor tyrosine kinase Ret and its coreceptor Gfra1, both of which are
expressed in the nephric duct [77].
The evolutionarily conserved transcriptional hierarchy or circuitry of
Eya/Pax/Six protein families are fundamental regulators of early kidney
development [78]. Mutations in human EYA1, SIX1 and SIX5 genes
have been shown to underlie the branchial arch defects, hearing loss
and renal anomalies associated with BOR (branchio-oto-renal) syn-
drome [78,79] (see Table 1). Pax2 and Pax8 cooperate in nephric lineage
specification. Mice homozygous mutant for the Pax2 gene alone fail to
induce the metanephric kidney, and Pax2/Pax8 double mutant mice fail
to form the pronephros. Likewise, in Eya1 homozygous mutant mice,
the metanephric mesenchyme fails to be specified, which results in
renal agenesis.
In the absence of the homeobox transcription factor Six1, which is
expressed in the uninduced metanephric mesenchyme, ureteric bud
invasion is abolished, and the MM subsequently undergoes apoptosis
[80]. Similar to the observed synergy between Pax2 and Pax8, Six1/Six4
double mutants display a complete loss of a Pax2-expressing
metanephric mesenchyme [81]. Six2 homozygous mutant mice have
a unique phenotype of kidney hypoplasia that is caused by early
depletion of the mesenchymal precursor pool and premature differ-
entiation of the mesenchyme into nephrons, which also results in fewer
ureteric branches [82].
Not surprisingly, the Hox11 cluster, which is crucial for embryonic
patterning along the anterioposterior axis at the level of the emerging
kidney, is also required for renal development. Triple mutants for
Hoxa11/Hoxc11/Hoxd11 show a complete loss of metanephric kidney
formation [83].
Pax2, Six1, Six2 and Six4 have been implicated in regulating the
expression of Gdnf to stimulate ureteric bud outgrowth [77], whereas
Gata3 has been suggested as a transcriptional regulator of Ret
expression downstream of Pax2/8 in the developing nephric duct [84].
Interestingly, the Hox11 paralogous proteins have been shown to
form a complex with Pax2 and Eya1 to directly activate expression of
Gndf and Six2 [85], so these genes are all part of a common pathway
regulating early kidney development. Lineage specification events
require the establishment of gene expression patterns that is most
likely achieved through epigenetic modifications of local chromatin
structures. Pax2 has been reported to recruit the assembly of a histone
H3 lysine 4 methyltransferase complex through interaction with Pax
transcription activating domain interacting protein (PTIP) to activate
target promoters [86], which might be a hint at a general mechanism of
cell fate determination via this regulatory network.
Apart from the regulation of Gdnf/Ret expression, other pathways
parallel to the aforementioned Eya1 and Pax2/8 are required for proper
induction and specification of the metanephric field along the
rostrocaudal axis. Early kidney development also requires expression
of Odd1 and Lim1 [74], two transcription factors expressed in the
intermediate mesoderm. Interestingly, the zinc-finger transcription
factor Odd1 is downregulated on tubule differentiation, and persistent
expression of Odd1 in the chick prevents differentiation [87].
Other genes required for renal branching morphogenesis include
Sall1 and Sall4, zinc finger proteins homologous to the Drosophila
gene spalt [88], the homeobox factor Emx2, which is required for
ureteric bud functions after Pax2 [75], and WT1. WT1 mutations
underlie the urogenital malformations in WAGR, Denys-Drash, Frasier
syndromes and Wilm’s tumor (reviewed in Ref. [43]).
The reciprocal inductive interactions between the ureteric bud (UB)
and the MM suggest the involvement of secreted signaling molecules
on top of GDNF. One important player that has surfaced in this process
is the canonical Wnt pathway: Analysis of mutant mice has divulged
that Wnt9b is a primary signal necessary for renal vesicle induction,
including Wnt4 production [89]. Wnt4 itself seems to be an autocrine
factor necessary for differentiation of the MM into epithelial nephrons.
In fact, Wnt4 was recently shown to be directly regulated by Pax2
during renal vesicle formation [90]. Interestingly, subsequent renal
tubule differentiation is not compatible with activated, stabilized b-
catenin, implying the need for a downregulation of Wnt signaling in
later stages [32].
In conclusion, the specification along the anterioposterior and
mediolateral axes to induce formation of MM and UB and to drive
kidney morphogenesis requires the activation of multiple pathways.
The relationship between these proteins and their interactions are
complex and still incompletely understood.
Review Trends in Genetics Vol.xxx No.x
TIGS-649; No of Pages 11
mine the peptide motifs and protein machinery that directGli and Glis proteins to the cilia to fully understand thesignal-dependent modifications of these proteins.
Connecting the cilium with the nucleusLocalization of transcriptional regulators to cilia and thecell membrane is not limited to the Gli and Glis families.Additional pathways allow the cilium to influence nucleargene transcription: the proteins encoded by PKD1 andPKD2, polycystin-1 and polycystin-2, are transmembraneglycoproteins that interact with one another andmediate avariety of complex formations. PKD2 binds to and seques-ters Id2, a basic helix–loop–helix (bHLH) transcriptionfactor that regulates cell proliferation by suppressingp21, a CDK inhibitor. By preventing nuclear localizationof Id2, PKD2 inhibits cell cycle progression by upregulatingp21 [20]. Interestingly, this interaction requires the phos-phorylation of polycystin-2. This phosphorylation reactionis dependent on PKD1, which can itself be proteolyticallycleaved, and its processed C-terminal tail can enter the
represents the functional unit of the kidney and consists of a glomerulus, a proximal tubu
blood filtration occurs inside the glomerulus, which harbors a dense capillary tuft, podo
plasma passes through the fenestrated endothelium, the glomerular basement membran
processes, which acts as a macromolecular filter, into the urinary space.
nucleus, where it activates activator protein 1 (AP-1)–de-pendent transcriptional pathways [21,22]. Cell cycle regu-lation via p21 induction can also occur by direct activationof JAK-STAT signaling by PKD1 and 2 [23]. In addition,processing of the C-terminal tail of PKD1 might be trig-gered in response to mechanosensation of fluid flow inrenal tubules, resulting in ciliary-nuclear translocationtogether with its interactors STAT6 and P100 [24].Because no kidney malfunction has been reported in micewith mutated Id2 or Stat6 genes, the physiological import-ance of these proteins in renal disease is not clear. Further-more, Pkd1-mutant mice show ectopic Pax2 expression,and deletion of Pax2 reduces cyst growth in Pkd1-deficientmice, suggesting that PKD1 represses Pax2 in an, as yet,unknown manner [25].
Similarly, polyductin (also known as fibrocystin)encoded byPKHD1 is subject to proteolytic cleavage, whichis in this case dependent on Ca2+- signaling [26]. Its largeextracellular domain is shed from the cilium, whereas theintracellular part has been suggested to enter the nucleus
le, the loop of Henle and a distal tubule connected to the collecting duct. The initial
cytes and supporting mesangial cells, surrounded by Bowman’s capsule. (d) Blood
e, and ultimately through the slit diaphragm between interdigitating podocyte foot
5
Figure 1. Schematic representation of a primary cilium. A primary cilium consists of a central axoneme made of microtubules enclosed by a distinct cell membrane. Several
structural elements such as the periciliary membrane, the transition fibers and basal bodies form a selective barrier at the entrance of the cilium and create a unique
environment that allows for compartmentalization. Protein products of genes implicated in polycystic kidney disease (‘cystic proteins’) localize to the basal body,
membrane or axoneme of the cilium. Cilia are sensory organelles that can probe the extracellular environment, but they also act as signaling centers. Bending of the cilium
by renal tubular flow causes an intracellular Ca2+ influx that sets off a variety of signaling cascades. Components of the Hedgehog (and other) signal transduction
pathway(s) have been shown to depend on ciliary localization. The binding of the secreted Sonic hedgehog protein (Shh) to its receptor Patched, and the Smoothened-
mediated processing of the Gli transcription factors has been detected in the cilium. The Gli transcription factors can move from there to the nucleus to regulate gene
expression in response to Hedgehog signaling, either activating or repressing transcription (GliA or GliR). Other transcription factors, such as Stat6 and Glis2, and
fragments of ‘cystic proteins’ have also been reported to localize to the cilium and the nucleus.
Review Trends in Genetics Vol.xxx No.x
TIGS-649; No of Pages 11
after cleavage by gamma-secretase, in amanner analogousto Notch processing [27]. Likewise, inversin, a proteinimplicated in nephronophthisis, has been proposed tolocalize to the nucleus [28,29]. Moreover, the ‘cysticprotein’ CEP290 binds to and modulates the activity ofthe transcription factor ATF4 [30]. These results togethersuggest that the cleaved fragments of various ciliaryproteins can have a co-activator/co-repressor-like role intranscriptional regulation. However, it is not yet clear ifthese events occur in vivo and any functional consequencesfor kidney development and homeostasis remain to beelucidated.
6
Maintenance of renal epithelial morphologyIn contrast to the putative physiological role of nuclearciliary protein fragments, the function of b-catenin intranscriptional regulation downstream of canonical Wntsignaling is firmly established [31]. Canonical Wnt sig-naling is required during early kidney development,whereas the noncanonical planar cell polarity (PCP) path-way is required for the proper alignment of cell divisionsduring epithelial tubule elongation [32–34]. NoncanonicalWnt signaling is independent of b-catenin, but sharesseveral other pathway components such as Frizzled andDishevelled with the canonical Wnt signaling pathway.
Review Trends in Genetics Vol.xxx No.x
TIGS-649; No of Pages 11
Furthermore, it has been suggested that the ciliary proteininversin triggers the switch between canonical and non-canonical Wnt signaling, which could occur in response tourine flow [28,35]. In this respect, it is noteworthy thatinversin interacts with b-catenin [28]. It is conceivable thatadult kidney homeostasis requires a ‘shut-down’ of thecanonical Wnt pathway after organogenesis is completed,whereas the PCP pathway confers the necessary spatialinformation for appropriate mitotic spindle orientation inrenal tubules. Consequently, overexpression of a constitu-tively active form of b-catenin in mice causes cystic renaldisease [31]. Although there is evidence for disruptedplanar cell polarity in polycystic kidneys [33], a require-ment for ciliary localization has not been demonstrated forcomponents of the Wnt signaling pathway [11].
We are beginning to gain a clearer insight into thetranscriptional regulation of the ‘cystic genes’ themselves,in particular Pkd2 and Pkhd1. Recent evidence from miceshows that the homeobox transcription factor hepatocytenuclear factor-1b (Hnf1b), also known as transcriptionfactor 2 (Tcf2), regulates Pkd2 (but not Pkd1) and Pkhd1gene expression by binding directly to their promoters.Consequently, deletion of HNF1b in murine kidneys leadsto cyst formation and the downregulation of Pkd2, Pkhd1,Ift88 (also known asTg737 orPolaris) andUmod; the lattertwo genes are involved in the development of renal cysts[36]. In humans, mutations in HNF1b cause the RCADsyndrome (renal cysts and diabetes) [37]. A similar phe-notype was observed inmice with a targeted inactivation ofthe transcriptional coactivator Wwtr1/TAZ [38].
In summary, the mechanisms controlling proliferation,differentiation and apoptosis of renal epithelial cells seemto entail complex interactions between ciliary proteins andtranscriptional regulators with cross-talk on multiplelevels. However, the functional significance this cross-talkhas for kidney physiology needs further analysis.
Glomerular diseases: the podocyte takes center stageGlomerular diseases encompass a wide range of pathologi-cally defined syndromes that account for most cases ofESRD [39]. The glomerulus serves a primary function infiltering the urine from the blood, and failure in thisprocess might result in proteinuria and kidney failure(see Figure I in Box 1). Common diseases, such as diabetes,hypertension, autoimmune diseases and toxic drug intakecan cause glomerular insults that mainly affect the podo-cyte, a highly specialized visceral epithelial cell type.Mutations in genes encoding structural components ofthe podocyte foot processes, slit diaphragms or glomerularbasement membrane (such as nephrin, podocin, collagentype IV and b2-laminin) have long been known to be causesof glomerulopathies [40,41], but the transcriptional regu-lation of these genes in renal physiology and disease is justbeginning to become clear [42].
Wilms’ tumor protein 1 (WT1) is a zinc-finger transcrip-tion factor with several roles in kidney development andfunction [43]. The human Wilm’s tumor-aniridia-genitour-inary anomalies-mental retardation syndrome (WAGR)and theDenys-Drash and Frasier syndromes are all causedby different mutations in WT1. The two main WT1 spliceisoforms referred to as �KTS and +KTS, have overlapping
but distinct functions in renal and gonadal development[43]. In particular, the absence of the +KTS form in Frasiersyndrome causes defects in podocyte function [43]. At leastthree ‘podocyte genes’ are regulated by WT1 directly:podocalyxin, nephrin and Pax2. The relationship betweenPax2 and WT1 is complex, and Pax2 expression in podo-cytes has been linked to repression of WT1. Interestingly,ectopic Pax2 expression in glomerular podocytes has beenfound in several pathological conditions. Persistent trans-genic expression of Pax2, which is normally repressedduring terminal differentiation of renal epithelial cells,has been shown to disrupt kidney function andmorphology(cyst formation and absence of foot processes). These find-ings have been extended by another mouse model thatallows inducible expression of Pax2 in fully mature podo-cytes of adult kidneys, which leads to dedifferentiation andestablishes a causal relationship between ectopic Pax2activation and glomerular disease [44].
The human nail-patella syndrome, which is often associ-ated with renal disease, is caused by mutations in the LIMhomeodomain transcription factor LMX1B [45]. LMX1Bhas been shown to regulate the expression of a3(IV) anda4(IV) collagen chains directly; these are essential com-ponents of the glomerular basement membrane and ofpodocin, a nephrin-binding membrane protein implicatedin steroid-resistant nephrotic syndrome [45]. A recentcomprehensive study addressed the role of Lmx1b in glo-merular disease. Lmx1b-deficient podocytes fail to formfoot processes and slit diaphragms, which leads to protei-nuria. Moreover, podocyte-specific inactivation of Ldb1,which interacts with Lmx1b biochemically, also resultsin a podocyte phenotype with gradual loss of foot processes[46].
Furthermore, a subset of mutations in the forkheadtranscription factor FOXC2 underlying the human lym-phedema-distichiasis syndrome have been reported to beassociated with renal malfunction [47]. Indeed, knockoutstudies in mice have confirmed a role for Foxc2 in podocytedevelopment [47,48]. Promising candidate genes for unex-plained forms of human inherited glomerular disease in-clude MAFB, which encodes a basic domain leucine zippertranscription factor, and transcription factor 21 (TCF21)also known as Podocyte-expressed 1 (POD1), whichencodes a bHLH transcription factor. Mouse models withMafB orTcf21/Pod1 gene disruptions affect podocyte func-tion because of diminished expression of podocin andnephrin [49–51]. Moreover, the Notch pathway has beenshown to be essential for podocyte development [52]. Thetranscriptional regulator RBPjk, which can form hetero-dimeric complexes with other bHLH transcription factors,functions downstream of Notch signaling. Interestingly,genetic deletion of RBPjk in podocytes delays the pro-gression of glomerular disease, providing another examplethat termination of the developmental program is import-ant to prevent renal disease [53]. Additional candidategenes for the development of glomerular disease encodethe zinc fingers and homeoboxes (ZHX) transcription fac-tors, which were reported to be expressed in podocytes andto regulate numerous functionally important podocytegenes [42]. In the future, more genes will join the ranksof regulators of podocyte function, as efforts are currently
7
Review Trends in Genetics Vol.xxx No.x
TIGS-649; No of Pages 11
being made to identify glomerulus-specific transcripts on alarge scale [48]. Like all other renal epithelial cells, podo-cytes possess cilia. However the importance of cilia sig-naling for podocyte function remains to be studied.
As alluded to earlier, the most common cause of kidneyfailure is diabetic nephropathy. High glucose concen-trations in the blood induce proliferation of another glo-merular cell type, themesangial cell (see Figure I in Box 1).Mesangial cell proliferation is a hallmark feature of glo-merulosclerosis, and has been correlated with productionof angiotensin II, transforming growth factor b (TGF-b),and platelet-derived growth factors (PDGFs) [54]. Inparticular, the JAK–STAT pathway mediates a significantpart of the proliferative effects of PDGF and ANG II. JAKproteins are cytoplasmic tyrosine kinases, which onceactivated phosphorylate STAT transcription factors,resulting in their nuclear translocation and the activationof responsive promoters [54]. Furthermore, Y-Box protein1, which has transcriptional regulatory activity, was impli-cated as a mediator of PDGF-B–induced mesangial cellproliferation [55]. Inhibition of this pathway might thusprovide an interesting target for therapeutic intervention.
Renal fibrosis: the final stage of chronic kidney diseaseRegardless of the initial cause, virtually all types ofchronic kidney disease are complicated by the histologicalappearance of glomerulosclerosis and tubulointerstitial
Figure 2. Signaling pathways in renal fibrosis. Renal epithelial tubules are formed by
during development, whereas the reverse process, epithelial-to-mesenchymal transitio
fibrosis. During EMT, epithelial cells lose apical-basal characteristics and expression of c
morphology. Factors such as TGFb, USAG-1, Snail, Gli1, KAP1 and CBF-A promote EMT
genes such as FSP1 and a-SMA (pink). In healthy renal tubules, proteins such as BMP7, K
the epithelial phenotype and inhibiting ectopic Gli1 or Snail activation (purple).
8
fibrosis [56,57]. Fibrosis is characterized by an excessiveaccumulation and deposition of extracellular matrix(ECM) that progressively leads to the destruction of func-tional nephrons. The majority of ECM is produced bya-smooth muscle actin expressing myofibroblasts [58].Myofibroblasts are believed to form through phenotypictransition of existing interstitial fibroblasts, mesangialcells, or migration of more distant cells into the kidney[59]. However, there is increasing evidence further sup-ported by the analysis of Glis2 mutant kidneys that myo-fibroblasts can also originate from renal epithelial tubulesthrough epithelial-to-mesenchymal transition (EMT)[60,61] (Figure 2). Although EMT is an essential processrequired for metazoan embryogenesis, it is responsiblefor the detrimental effects seen under pathophysiologicalconditions of organ fibrosis and cancer metastasis [62,63].Loss ofGlis2 or other fibrotic responses induce expressionof the zinc finger transcription factor Snail, which in turnrepresses E-cadherin and Cadherin-16 expression. Down-regulation of cadherin leads to a loss of cell adhesion andcell polarity. Interestingly, Snail does not directly repressCadherin-16, but rather reduces expression of its activatorHNF1b. Ectopic activation of Snail in adult mice has beenshown to be sufficient to drive EMT and cause renalfibrosis. Moreover, Snail upregulation can be observedin patients with fibrotic kidneys [64]. In addition, Gli1was recently shown to regulate the expression of Snail,
mesenchymal-to-epithelial transition (MET) from the metanephric mesenchyme
n (EMT), is associated with the loss of cell polarity and adhesion leading to kidney
ell adhesion molecules like E-cadherin, acquire mobility and adopt a fibroblast-like
and the appearance of myofibroblasts that are characterized by the expression of
CP, Glis2 and HNF1b prevent fibrosis by stabilizing the expression of cadherins and
Box 3. Outstanding questions and future directions
What are the gene expression signatures of the specialized renal cell
types in the adult kidney, and what is their relationship to renal
physiology and tissue homeostasis?
What is the lineage relationship of the differentiated renal cell types
and what is the extent of renal cellular plasticity in the adult organ?
What can be learned from the identification of the transcriptional
networks downstream of cilia signaling in the adult kidney?
Why is only part of the developmental program reactivated in renal
injury and repair and how do these processes differ from renal
fibrosis?
What are the pathways that terminate mesenchymal-to-epithelial
transition during renal development and is epithelial-to mesench-
ymal transition in kidney fibrosis reversible?
Do true adult renal stem cells exist which could be utilized for renal
replacement and repair therapy after injury?
What are the renal disease-associated transcriptional programs
resulting from ageing, diabetes, obesity and hypertension?
Review Trends in Genetics Vol.xxx No.x
TIGS-649; No of Pages 11
and Gli1-mediated Snail activation (followed by E-cad-herin downregulation) drives transformation of epithelialcells in tumor progression [65]. These results emphasizethe requirement for repression of Gli-responsive promo-ters to prevent EMT. Fibroblasts created through EMT inthe kidney express theFSP1 gene (knownasS100a4 in thecancer literature), which encodes fibroblast-specificprotein 1. The transcriptional regulators, CArG box-bind-ing factor-A (CBF-A) and KRAB-associated protein 1(KAP-1) bind to distinct DNA motifs in the promoter ofFSP1 and activate its transcription. Similar bindingsites can be found in the promoters of several otherEMT-associated transcripts such as Twist, Snail, E-cad-herin, vimentin, a1(I)collagen and a-smooth muscle actin,all of which are activated by the CBF-A/KAP-1 complex[66,67] (Figure 2).
Our understanding of the transcriptional network con-trollingEMT in different organ systems is improving fast. Itis generally thought that TGF-b and its downstreamSMADsignaling play an essential role in most forms of chronickidney disease. Expression of TGF-b in transgenic micepromotes EMT and fibrosis, whereas inhibition of TGF-bby different approaches (including overexpression of theinhibitory Smad7) prevents it [60,68]. Another member ofthe TGF-b superfamily, BMP7, which is required for earlykidney development, has the opposite effect: it counteractsEMT and prevents fibrosis [68,69]. Furthermore, modifiersof BMP signaling, such as the enhancer KCP and the BMPantagonist USAG-1, have been shown to improve or worsenthe formation of fibrous tissue, respectively [69,70].
Thus, molecularmechanisms underlying kidney fibrosissuggest that renal fibrosis can be thought of as reversal ofearly kidney development. However, we have much tolearn about the transcriptional network that drives theformation of renal epithelial cells through mesenchymal-to-epithelial transition (MET) during development. Thisinformation will be useful in designing novel therapeuticstrategies to reverse kidney fibrosis (Figure 2).
Concluding remarks and open questionsTranscription factors not only serve as genetic markers forspecific cell populations but more importantly orchestratethe genetic program within each cell. Therefore, theyprovide useful entry points to decipher the cis-regulatorynetworks that underlie the coordinated expression ofspecific sets of genes to create the various renal cell typesrequired for kidney function.
Over recent years, many transcriptional regulators ofkidney development have been identified, but most of thetarget genes controlled by these transcription factorsremain elusive (with few exceptions). Likewise, the hier-archical relationships between the individual regulatorsare thus far difficult to assess. In the future, it will beimportant to move beyond descriptive approaches of loss-of-function phenotypes for single genes. Time courses andhigh-throughput validation of the expression profilesobtained from loss-of-function studies should help toidentify gene expression signatures that are importantfor kidney development and disease [48,71,72]. In vivochromatin immunoprecipitation (ChIP) studies, in combi-nation with genomics, bioinformatics and proteomics
technologies, will be useful to elucidate transcriptionalregulatory networks governing kidney physiology. In thisrespect, it is reassuring to see the publications of the firstrenal studies driven by a systems biology approach [72,73].
Although we have learned much about early eventsduring metanephric development, there is still a lack ofknowledge regarding later events such as patterning ofnephric tubule segments, and self-renewal of the mesench-ymal progenitor pool. Furthermore, it is becoming increas-ingly clear that ‘terminally’ differentiated cell types in themature kidney might not exist. This is best underscored byrecent observations of the reactivation of developmentaltranscription factors in renal disease progression, asshown for Pax2 and RBPjk in polycystic kidney and glo-merular disease and for Gli1 and Snail in renal fibrosis.Thus, adult kidney homeostasis seems to rely on activesilencing of developmental gene expression programs, asexemplified by the analysis of Glis2 function. The under-appreciated cellular plasticity in the kidney, however,might open up untapped sources for the treatment ofchronic kidney disease.
In this respect, it will be of tremendous interest to eluci-date the molecular mechanisms of cilia signaling beyond itsproposed function as amechanosensor of renal tubular flow.Furthermore, studying the regulation and necessity of cili-ary localization for various signaling pathways and its func-tional importance in kidney development and disease hasalready begun to open up a new field of investigation. Formost transcription factors that are important during earlykidney organ induction and patterning, their role duringlater morphogenetic stages is still not clarified. Thus, itwill be important to systematically take advantage ofconditionally and temporally controlled gene disruptionapproaches in mice to illuminate their specific function inmature nephrons and during kidney disease. Finally, con-ditionalmutagenesisandgenetic screens inhumanpatients,zebrafish or other model organisms will continue to yieldnovel genes required for kidney organ development.
Clearly a lot still needs to be learned about organ de-velopment and physiology in this fascinatingmodel systemto open new avenues for therapeutic interventions in thebattle against the increasing pandemic of chronic kidneydisease (Box 3).
9
Review Trends in Genetics Vol.xxx No.x
TIGS-649; No of Pages 11
AcknowledgementsThe authors apologize to all colleagues whose excellent work could not becited because of space constraints. The authors thank Petra Riedinger forhelp with figures.
References1 Weening, J.J. (2004) Advancing nephrology around the globe: an
invitation to contribute. J. Am. Soc. Nephrol. 15, 2761–27622 Hallan, S.I. et al. (2006) International comparison of the relationship of
chronic kidney disease prevalence and ESRD risk. J. Am. Soc. Nephrol.17, 2275–2284
3 Chatziantoniou, C. and Dussaule, J.C. (2008) Is kidney injury areversible process? Curr. Opin. Nephrol. Hypertens. 17, 76–81
4 Schedl, A. (2007) Renal abnormalities and their developmental origin.Nat. Rev. Genet. 8, 791–802
5 Bisgrove, B.W. and Yost, H.J. (2006) The roles of cilia in developmentaldisorders and disease. Development 133, 4131–4143
6 Hildebrandt, F. and Otto, E. (2005) Cilia and centrosomes: a unifyingpathogenic concept for cystic kidney disease? Nat. Rev. Genet. 6, 928–940
7 Satir, P. and Christensen, S.T. (2007) Overview of structure andfunction of mammalian cilia. Annu. Rev. Physiol. 69, 377–400
8 Reiter, J.F. and Mostov, K. (2006) Vesicle transport, cilium formation,and membrane specialization: the origins of a sensory organelle. Proc.Natl. Acad. Sci. U. S. A. 103, 18383–18384
9 Davenport, J.R. et al. (2007) Disruption of intraflagellar transport inadult mice leads to obesity and slow-onset cystic kidney disease. Curr.Biol. 17, 1586–1594
10 Piontek, K. et al. (2007) A critical developmental switch defines thekinetics of kidney cyst formation after loss of Pkd1.Nat.Med. 13, 1490–1495
11 Hildebrandt, F. and Zhou, W. (2007) Nephronophthisis-associatedciliopathies. J. Am. Soc. Nephrol. 18, 1855–1871
12 Gill, P.S. and Rosenblum, N.D. (2006) Control of murine kidneydevelopment by sonic hedgehog and its GLI effectors. Cell Cycle 5,1426–1430
13 Eggenschwiler, J.T. and Anderson, K.V. (2007) Cilia and developmentalsignaling. Annu. Rev. Cell Dev. Biol. 23, 345–373
14 Bose, J. et al. (2002) Pallister-Hall syndrome phenotype inmicemutantfor Gli3. Hum. Mol. Genet. 11, 1129–1135
15 Hu, M.C. et al. (2006) GLI3-dependent transcriptional repression ofGli1, Gli2 and kidney patterning genes disrupts renal morphogenesis.Development 133, 569–578
16 Rohatgi, R. et al. (2007) Patched1 regulates hedgehog signaling at theprimary cilium. Science 317, 372–376
17 Attanasio, M. et al. (2007) Loss of GLIS2 causes nephronophthisis inhumans and mice by increased apoptosis and fibrosis. Nat. Genet. 39,1018–1024
18 Kim, Y.S. et al. (2008) Kruppel-like zinc finger protein glis2 is essentialfor the maintenance of normal renal functions. Mol. Cell. Biol. 28,2358–2367
19 Senee, V. et al. (2006) Mutations in GLIS3 are responsible for arare syndrome with neonatal diabetes mellitus and congenitalhypothyroidism. Nat. Genet. 38, 682–687
20 Li, X. et al. (2005) Polycystin-1 and polycystin-2 regulate the cell cyclethrough the helix-loop-helix inhibitor Id2. Nat. Cell Biol. 7, 1202–1212
21 Chauvet, V. et al. (2004) Mechanical stimuli induce cleavage andnuclear translocation of the polycystin-1 C terminus. J. Clin. Invest.114, 1433–1443
22 Guay-Woodford, L.M. (2004) RIP-ed and ready to dance: newmechanisms for polycystin-1 signaling. J. Clin. Invest. 114, 1404–1406
23 Bhunia, A.K. et al. (2002) PKD1 induces p21(waf1) and regulation ofthe cell cycle via direct activation of the JAK-STAT signaling pathwayin a process requiring PKD2. Cell 109, 157–168
24 Low, S.H. et al. (2006) Polycystin-1, STAT6, and P100 function in apathway that transduces ciliary mechanosensation and is activated inpolycystic kidney disease. Dev. Cell 10, 57–69
25 Stayner, C. et al. (2006) Pax2 gene dosage influences cystogenesis inautosomal dominant polycystic kidney disease. Hum. Mol. Genet. 15,3520–3528
26 Hiesberger, T. et al. (2006) Proteolytic cleavage and nucleartranslocation of fibrocystin is regulated by intracellular Ca2+ andactivation of protein kinase C. J. Biol. Chem. 281, 34357–34364
10
27 Kaimori, J.Y. et al. (2007) Polyductin undergoes notch-like processingand regulated release from primary cilia. Hum. Mol. Genet. 16, 942–956
28 Nurnberger, J. et al. (2002) Inversin forms a complex with catenins andN-cadherin in polarized epithelial cells. Mol. Biol. Cell 13, 3096–3106
29 Otto, E.A. et al. (2003) Mutations in INVS encoding inversin causenephronophthisis type 2, linking renal cystic disease to the function ofprimary cilia and left-right axis determination.Nat. Genet. 34, 413–420
30 Sayer, J.A. et al. (2006) The centrosomal protein nephrocystin-6 ismutated in Joubert syndrome and activates transcription factor ATF4.Nat. Genet. 38, 674–681
31 Benzing, T. et al. (2007) Wnt signaling in polycystic kidney disease. J.Am. Soc. Nephrol. 18, 1389–1398
32 Park, J.S. et al. (2007) Wnt/beta-catenin signaling regulates nephroninduction during mouse kidney development. Development 134, 2533–2539
33 Fischer, E. et al. (2006) Defective planar cell polarity in polycystickidney disease. Nat. Genet. 38, 21–23
34 Schmidt-Ott, K.M. et al. (2007) beta-catenin/TCF/Lef controls adifferentiation-associated transcriptional program in renal epithelialprogenitors. Development 134, 3177–3190
35 Simons, M. et al. (2005) Inversin, the gene product mutated innephronophthisis type II, functions as a molecular switch betweenWnt signaling pathways. Nat. Genet. 37, 537–543
36 Gresh, L. et al. (2004) A transcriptional network in polycystic kidneydisease. EMBO J. 23, 1657–1668
37 Igarashi, P. et al. (2005) Roles of HNF-1beta in kidney developmentand congenital cystic diseases. Kidney Int. 68, 1944–1947
38 Hossain, Z. et al. (2007) Glomerulocystic kidney disease in mice with atargeted inactivation of Wwtr1. Proc. Natl. Acad. Sci. U. S. A. 104,1631–1636
39 Wiggins, R.C. (2007) The spectrum of podocytopathies: a unifying viewof glomerular diseases. Kidney Int. 71, 1205–1214
40 Michaud, J.L. and Kennedy, C.R. (2007) The podocyte in health anddisease: insights from the mouse. Clin. Sci. (Lond.) 112, 325–335
41 Quaggin, S.E. and Kreidberg, J.A. (2008) Development of the renalglomerulus: good neighbors and good fences.Development 135, 609–620
42 Chugh, S.S. (2007) Transcriptional regulation of podocyte disease.Transl. Res. 149, 237–242
43 Rivera, M.N. and Haber, D.A. (2005) Wilms’ tumour: connectingtumorigenesis and organ development in the kidney. Nat. Rev.Cancer 5, 699–712
44 Wagner, K.D. et al. (2006) An inducible mouse model for PAX2-dependent glomerular disease: insights into a complex pathogenesis.Curr. Biol. 16, 793–800
45 Morello, R. et al. (2001) Regulation of glomerular basement membranecollagen expression by LMX1B contributes to renal disease in nailpatella syndrome. Nat. Genet. 27, 205–208
46 Suleiman, H. et al. (2007) The podocyte-specific inactivation of Lmx1b,Ldb1 and E2a yields new insight into a transcriptional network inpodocytes. Dev. Biol. 304, 701–712
47 Yildirim-Toruner, C. et al. (2004) A novel frameshift mutation ofFOXC2 gene in a family with hereditary lymphedema-distichiasissyndrome associated with renal disease and diabetes mellitus. Am.J. Med. Genet. A. 131, 281–286
48 Takemoto, M. et al. (2006) Large-scale identification of genesimplicated in kidney glomerulus development and function. EMBOJ. 25, 1160–1174
49 Moriguchi, T. et al. (2006) MafB is essential for renal development andF4/80 expression in macrophages. Mol. Cell. Biol. 26, 5715–5727
50 Sadl, V. et al. (2002) The mouse Kreisler (Krml1/MafB) segmentationgene is required for differentiation of glomerular visceral epithelialcells. Dev. Biol. 249, 16–29
51 Quaggin, S.E. et al. (1999) The basic-helix-loop-helix protein pod1 iscritically important for kidney and lung organogenesis. Development126, 5771–5783
52 Cheng, H.T. et al. (2007) Notch2, but not Notch1, is required forproximal fate acquisition in the mammalian nephron. Development134, 801–811
53 Niranjan, T. et al. (2008) The Notch pathway in podocytes plays a rolein the development of glomerular disease. Nat. Med. 14, 290–298
54 Marrero, M.B. et al. (2006) Role of the JAK/STAT signaling pathwayin diabetic nephropathy.Am. J. Physiol. Renal Physiol. 290, F762–F768
Review Trends in Genetics Vol.xxx No.x
TIGS-649; No of Pages 11
55 van Roeyen, C.R. et al. (2005) Y-box protein 1 mediates PDGF-B effectsin mesangioproliferative glomerular disease. J. Am. Soc. Nephrol. 16,2985–2996
56 Fogo, A.B. (2007) Mechanisms of progression of chronic kidney disease.Pediatr. Nephrol. 22, 2011–2022
57 Khwaja, A. et al. (2007) The management of CKD: a look into thefuture. Kidney Int. 72, 1316–1323
58 Simonson, M.S. (2007) Phenotypic transitions and fibrosis in diabeticnephropathy. Kidney Int. 71, 846–854
59 Wada, T. et al. (2007) Fibrocytes: a new insight into kidney fibrosis.Kidney Int. 72, 269–273
60 Liu, Y. (2006) Renal fibrosis: new insights into the pathogenesis andtherapeutics. Kidney Int. 69, 213–217
61 Kalluri, R. andNeilson, E.G. (2003) Epithelial-mesenchymal transitionand its implications for fibrosis. J. Clin. Invest. 112, 1776–1784
62 Shook, D. and Keller, R. (2003) Mechanisms, mechanics and function ofepithelial-mesenchymal transitions in early development. Mech. Dev.120, 1351–1383
63 Radisky, D.C. et al. (2007) Fibrosis and cancer: do myofibroblastscome also from epithelial cells via EMT? J. Cell. Biochem. 101, 830–839
64 Boutet, A. et al. (2006) Snail activation disrupts tissue homeostasis andinduces fibrosis in the adult kidney. EMBO J. 25, 5603–5613
65 Li, X. et al. (2007) Gli1 acts through Snail and E-cadherin to promotenuclear signaling by beta-catenin. Oncogene 26, 4489–4498
66 Venkov, C.D. et al. (2007) A proximal activator of transcription inepithelial-mesenchymal transition. J. Clin. Invest. 117, 482–491
67 Teng, Y. et al. (2007) Transcriptional regulation of epithelial-mesenchymal transition. J. Clin. Invest. 117, 304–306
68 Wang, W. et al. (2005) Transforming growth factor-beta and Smadsignalling in kidney diseases. Nephrology (Carlton) 10, 48–56
69 Yanagita, M. (2006) Modulator of bone morphogenetic proteinactivity in the progression of kidney diseases. Kidney Int. 70, 989–993
70 Lin, J. et al. (2005) Kielin/chordin-like protein, a novel enhancer of BMPsignaling, attenuates renal fibrotic disease. Nat. Med. 11, 387–393
71 Schwab, K. et al. (2006) Comprehensivemicroarray analysis of Hoxa11/Hoxd11 mutant kidney development. Dev. Biol. 293, 540–554
72 McMahon,A.P. et al. (2008)GUDMAP: the genitourinarydevelopmentalmolecular anatomy project. J. Am. Soc. Nephrol. 19, 667–671
73 He, L. et al. (2008) The glomerular transcriptome and a predictedprotein-protein interaction network. J. Am. Soc. Nephrol. 19, 260–268
74 Boyle, S. and de Caestecker, M. (2006) Role of transcriptional networksin coordinating early events during kidney development. Am. J.Physiol. Renal Physiol. 291, F1–F8
75 Bouchard, M. (2004) Transcriptional control of kidney development.Differentiation 72, 295–306
76 Dressler, G.R. (2006) The cellular basis of kidney development. Annu.Rev. Cell Dev. Biol. 22, 509–529
77 Costantini, F. and Shakya, R. (2006) GDNF/Ret signaling and thedevelopment of the kidney. Bioessays 28, 117–127
78 Brodbeck, S. and Englert, C. (2004) Genetic determination ofnephrogenesis: the Pax/Eya/Six gene network. Pediatr. Nephrol. 19,249–255
79 Hoskins, B.E. et al. (2007) Transcription factor SIX5 is mutated inpatients with branchio-oto-renal syndrome. Am. J. Hum. Genet. 80,800–804
80 Xu, P.X. et al. (2003) Six1 is required for the early organogenesis ofmammalian kidney. Development 130, 3085–3094
81 Kobayashi, H. et al. (2007) Six1 and Six4 are essential for Gdnfexpression in the metanephric mesenchyme and ureteric budformation, while Six1 deficiency alone causes mesonephric-tubuledefects. Mech. Dev. 124, 290–303
82 Self, M. et al. (2006) Six2 is required for suppression of nephrogenesisand progenitor renewal in the developing kidney. EMBO J. 25, 5214–5228
83 Wellik, D.M. et al. (2002) Hox11 paralogous genes are essential formetanephric kidney induction. Genes Dev. 16, 1423–1432
84 Grote, D. et al. (2006) Pax 2/8-regulated Gata 3 expression is necessaryfor morphogenesis and guidance of the nephric duct in the developingkidney. Development 133, 53–61
85 Gong, K.Q. et al. (2007) A Hox-Eya-Pax complex regulates early kidneydevelopmental gene expression. Mol. Cell. Biol. 27, 7661–7668
86 Patel, S.R. et al. (2007) The BRCT-domain containing protein PTIPlinks PAX2 to a histone H3, lysine 4 methyltransferase complex. Dev.Cell 13, 580–592
87 James, R.G. et al. (2006) Odd-skipped related 1 is required fordevelopment of the metanephric kidney and regulates formationand differentiation of kidney precursor cells. Development 133,2995–3004
88 Nishinakamura, R. and Osafune, K. (2006) Essential roles of Sallfamily genes in kidney development. J. Physiol. Sci. 56, 131–136
89 Carroll, T.J. et al. (2005) Wnt9b plays a central role in the regulation ofmesenchymal to epithelial transitions underlying organogenesis of themammalian urogenital system. Dev. Cell 9, 283–292
90 Torban, E. et al. (2006) PAX2 activates WNT4 expressionduring mammalian kidney development. J. Biol. Chem. 281, 12705–12712
11
top related