Notch in the kidney: development and disease

Authors

  • Yasemin Sirin,

    1. Department of Nephrology, Department of Medicine, Albert Einstein College of Medicine, Bronx, New York, USA
    Search for more papers by this author
  • Katalin Susztak

    Corresponding author
    1. Department of Nephrology, Department of Medicine, Albert Einstein College of Medicine, Bronx, New York, USA
    2. Department of Genetics, Albert Einstein College of Medicine, Bronx, New York, USA
    • Division of Nephrology, Albert Einstein College of Medicine, 1300 Morris Park Avenue, Bronx, NY 10461, USA.
    Search for more papers by this author

  • No conflicts of interest were declared.

Abstract

Notch signalling is a highly conserved cell–cell communication mechanism that regulates development, tissue homeostasis, and repair. Within the kidney, Notch has an important function in orchestrating kidney development. Recent studies indicate that Notch plays a key role in establishing proximal epithelial fate during nephron segmentation as well as the differentiation of principal cells in the renal collecting system. Notch signalling is markedly reduced in the adult kidney; however, increased Notch signalling has been noted in both acute and chronic kidney injury. Increased glomerular epithelial Notch signalling has been associated with albuminuria and glomerulosclerosis, while tubular epithelial Notch activation caused fibrosis development most likely inducing an improper epithelial repair pathway. Recent studies thereby indicate that Notch is a key regulator of kidney development, repair, and injury. Copyright © 2011 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd.

The Notch signalling pathway

Normal renal development requires exquisite timing of multiple differentiation programmes. Notch is a well-known master regulator of cell specification, differentiation, and tissue patterning. Lack of or reduced Notch signalling in patients causes renal developmental abnormalities, indicating that Notch signalling plays a role in kidney development. Recently, publications have also surged indicating that Notch is reactivated in the kidney in different disease conditions and plays a role in orchestrating chronic kidney disease development. Understanding how a master regulator like Notch functions in the context of development and disease could further our ability to develop strategies to manipulate Notch in a clinically beneficial manner.

The Notch signalling pathway is an evolutionarily conserved cell–cell communication mechanism. It is present in all metazoans and functions as one of the major pathways that determine cell identity during development 1, 2. There are four receptors in mammals, Notch1–4, and five different ligands, Delta-like1, 3, 4 (Dll1, 3, 4) and Jagged1 and 2 (Jag1, 2) 3. Both receptors and ligands are single-pass type I transmembrane receptor proteins 3, 4. Notch is produced in the endoplasmic reticulum as a 300 kD precursor protein that is cleaved by a furin-like convertase in the trans-Golgi compartment to an intra- and extra-cellular domain 3. The extracellular portion of Notch undergoes extensive N- and O-linked glycosylation during synthesis and secretion 5. The Fringe family of glycosyltransferases catalyses O-fucose elongation by adding N-acethylglucosamines on specific EGF-like repeats. This modification alters the responsiveness of the receptor to its ligands 6. The constitutively cleaved Notch receptor is reassembled into a functional heterodimeric receptor at the cell surface (Figure 1). Notch signalling is initiated by the expression of the ligand on the signal-sending cell. Ligand binding triggers a series of proteolytic cleavages of the receptor (on the signal-receiving cell) by metalloproteases of the ADAM family (S2 site) and finally the γ-secretase complex (S3 site) (Figure 1) 7, 8. The functional γ-secretase complex has four principal transmembrane components: Presenilin, Nicastrin, Aph1 and Pen2. The S3 cleavage induces endocytosis of Notch's extracellular portion along with the ligand by the signal-sending cell. The subsequently released Notch intracellular domain (NICD) then translocates to the nucleus 9.

Figure 1.

Canonical Notch signalling. The signal-sending cell presents the ligand, belonging to the Delta or Jagged family, to the signal-receiving cell, expressing a Notch receptor on its surface. This leads to a sequential cleavage of Notch through ADAM metalloproteinases at site S2 and the γ-secretase complex at site S3. The released extracellular domain of Notch is endocytosed along with the ligand, whereas the cleaved intracellular Notch (NICD) translocates to the nucleus, where it interacts with its binding partner Rbpj, thus driving target gene (for example, Hes, Hey) transcription

Upon NICD's nuclear translocation, co-repressors associated with Rbpj (also referred to as CSL or CBF1), a binding partner of NICD, are displaced, and a transcriptionally active complex consisting of NICD, Rbpj, and Mastermind (Maml) assembles, thus leading to the transcription of Notch target genes 10. The response to Notch signalling varies greatly between different cell types—for example, Notch promotes cell proliferation in certain cells while inducing apoptosis in others 1, 11–13. Its ability to elicit different responses may be at least partially attributed to a cross-talk with other pathways. For example, several lines of evidence support an interaction between the Notch and NFκβ pathways 14. In addition, the cellular response likely depends on cell type-specific enhancers that are responsive to Notch regulation in a given cell. Interestingly, despite a high homology between the various receptors, the role of each Notch isoform can be markedly different 15–17. The best characterized target genes are the bHLH (basic helix-loop-helix) genes mainly belonging to the Hes and Hey family, which in turn function as transcription factors 18, 19. In the kidney, the regulation of Hes1 and HeyL has been described 20. Signal termination usually occurs by ubiquitin-ligase-mediated NICD degradation, facilitated by Maml, and involves cyclin-dependent kinase-8, which phosphorylates NICD 21. In addition, the ligands are ubiquitinated by an E3 ligase Mindbomb1 (Mib1) and internalized 22, 23. This core signal transduction pathway used in most Notch-dependent processes is known as the canonical Notch signalling pathway.

Notch is involved in the development of diverse tissues in almost all species, where it exhibits a myriad of effects 24, 25. The effect of Notch can be categorized into a few, classic patterns. For example, Notch can provide an ‘inductive signal’ by promoting the development of a given cell type through induction of positively acting regulatory molecules. Thus, the expression of Notch can induce a specific cell fate in neighbouring cells. An example of this occurs in Drosophila wing development, where loss of Notch signalling eliminates wing margin and wing tissue, while ectopic Notch signalling results in extra wing tissue development 26. In this model, loss of Notch signalling leads to a loss of a given cell type, while excessive Notch signalling has the reverse effect. Another pattern of Notch signalling is to restrict a given cell fate, ie ‘inhibitory’ Notch signalling. Initially, cells that share a given cell fate both send and receive Notch signals, known as ‘mutual’ inhibitory Notch signalling. Later, one cell commits to the specialized fate and inhibits surrounding cells from adopting this fate, a phenomenon called ‘lateral’ inhibitory Notch signalling. In this situation, failure of Notch signalling results in cells adopting the special cell fate, while excessive Notch signalling prevents differentiation of these cells. An example of Notch-induced lateral inhibition can be observed in Drosophila neurogenesis. Excessive Notch expression inhibits neuronal differentiation, while its dearth is associated with an excessive neurogenic phenotype 27–30.

Patients with mutations of Notch-related genes

Notch is a very tightly regulated signalling mechanism with multiple special features. For example, Notch activation not only needs cell–cell contact, making it function over short distances, but also occurs without a signal amplification step; ie the cleaved receptor travels to the nucleus without an amplification step. Therefore, it is not surprising that even subtle genetic alterations of the pathway could have significant phenotypic consequences.

Mutations of JAGGED1 or NOTCH231 have been described to be causally linked to Alagille syndrome (AGS). Most cases of AGS are caused by JAGGED1 mutations, whereas only a small subset can be attributed to NOTCH2 mutations. The estimated prevalence of AGS is 1 in every 100 000 live births and it is inherited in an autosomal dominant pattern. AGS is a multi-organ disorder which is mainly diagnosed by liver abnormalities, usually presenting with hepatic bile duct paucity and cholestasis 32. Cardiac, skeletal, and ophthalmological abnormalities are also fairly common in patients with AGS. Renal involvement in patients with AGS is well described in the paediatric literature. The spectrum of renal abnormalities in AGS is wide ranging and includes renovascular disease, renal tubular acidosis, tubulointerstitial nephritis, and renal dysplasia/hypoplasia 33. In addition, a peculiar pattern of glomerular lesions, called ‘mesangiolipidosis’, has been described. This is characterized by the presence of mesangial and extramembranous lipid inclusions 34, 35. The significance and specificity of these deposits are not fully established. It is interesting to note that mice haploinsufficient for Jagged1 show no significant phenotypic abnormalities, indicating that other modifier genes may contribute to the human AGS phenotype 36.

Notch and kidney development

The renal epithelium is a secondary epithelium formed by reciprocally inductive interactions between two different mesenchymal precursor tissues 37: the ureteric bud (UB), which gives rise to the collecting duct system, and the metanephric mesenchyme (MM), which gives rise to all other renal epithelial cells (from podocytes to distal tubular epithelial cells) 38, 39 (Figure 2). Glial cell line-derived neurotrophic factor might be one of the factors that is secreted by the mesenchyme to attract the UB towards the mesenchyme 40–44. The attracted UB in turn secretes Wnt9b and induces the mesenchyme 45. Upon this induction, mesenchymal cells secrete Wnt4. Wnt4 and fibroblast growth factor play essential roles in mesenchymal-to-epithelial transition (MET) 46. The MM then undergoes condensation followed by segmentation, resulting in structures called comma- and S-shaped bodies 47 (Figure 2). The proximal segments of the S-shaped bodies will become future podocytes (glomerular epithelial cells); the mid-sections give rise to the proximal tubules and loop of Henle; and the distal segments differentiate into distal tubules (Figure 2).

Figure 2.

Notch plays a role in proximal tubule specification and establishment of the cellular composition of the collecting duct. (A) The renal epithelium is formed by reciprocally inductive interactions between two different mesenchymal precursor tissues: the ureteric bud (UB) and the metanephric mesenchyme (MM). Sine oculis homeobox domain 2 (SIX2) suppresses tubulogenesis in the mesenchyme. Glial cell line-derived neurotrophic factor is secreted by the MM, attracting the UB towards the mesenchyme. The UB secretes Wnt9b and induces the mesenchyme-to-epithelial transition (MET) condensation via inducing Fgf8, Lhx1, and Wnt4. After MET, the nascent epithelial renal vesicles (RV) form a proximo-distal axis, where the surface facing the cortex is distal (yellow) and the medullary surface is proximal (green). (B) The proximal segment of the S-shaped body is thought to give rise to the future glomeruli and proximal tubules, Wt1 and Notch2 play a key role in defining these segments. (C) Endothelial cells are attracted by vascular endothelial growth factor (VEGF) and platelet-derived growth factor β (PDGFβ). (D) The mature nephron is further subdivided into glomerulus (green), proximal tubule (red), the loop of Henle (orange), and distal convoluted tubule (yellow), which connects to the collecting duct (blue). The scheme is adapted and modified from Schedl 37

Notch signalling molecules are expressed throughout kidney development; however, the reports addressing the exact location of different Notch ligands and receptors are somewhat controversial. Receptor (Notch1, Notch2) expression has been described in pretubular aggregates as well as in developing glomerular and tubular epithelial cells. The ligands Delta1 and Jagged1 are also expressed in early pretubular aggregates 48–50. Notch2 is expressed in the primitive proximal tubules, with faint co-localization in podocyte progenitors. It is not clear whether Notch1 is expressed in the prospective proximal tubules during development. Subsequently, Jagged1 was localized to the collecting duct and glomerular epithelial cells, while Jagged1 and Dll4 expression was observed on endothelial cells 15, 31, 48, 51–55.

Functional studies performed by Cheng et al55 showed a key role for Notch signalling during nephrogenesis. Mouse metanephroi cultured in the presence of a γ-secretase inhibitor (Figure 1), a blocker of Notch receptor cleavage that precludes its activation, caused a severe deficiency in proximal tubules and glomeruli (Figure 2). Further substantiating these findings, Wang et al showed that mice with genetic deletion for subunits of the Notch processing γ-secretase complex (Presenilin 1, 2) have severe defects in nephrogenesis and virtually no comma- and S-shaped bodies or mature glomeruli 53. Mice with a hypomorphic allele for Notch2, Notch2equation image, or mice with a targeted deletion of Notch2 from the MM showed a similar phenotype 56. Early kidney development, ureteric bud migration, branching, and mesenchymal aggregate formation occurred normally; however, glomeruli did not form, nor did podocytes or proximal tubules appear. Interestingly, the glomerular vascularization defect in the Notch2equation image mice phenocopied glomerular abnormalities observed in mice with PDGFβ and VEGF mutations 56. The development of distal tubules and collecting ducts was not compromised in these animals (Figure 2). While Notch1 ectopic expression was able to rescue the Notch2 null phenotype, Notch1 deletion from the MM did not cause developmental abnormalities 15. Later studies indicated that although the Notch2 receptor is dominant (its loss in the MM prior to the onset of kidney development results in complete nephron loss), Notch1 may also contribute to proximal kidney development and it becomes more apparent when Notch2 levels are limited 57. Studies by Cheng et al suggest that Notch1 may be a weaker activator of the target promoters than Notch2, possibly owing to tertiary structure differences, and that its endogenous levels could be below the threshold required to activate Notch2 target 15. These studies suggest a Notch-mediated proximal fate determination (proximal tubular epithelium and podocyte), an attractive idea considering the classic lateral inhibition mechanisms of Notch in a variety of biological processes.

Recent studies further clarified the role of Notch in kidney development and somewhat contradict the original observations. First of all, it appears that the initial proximo-distal axis is still established even when Notch2 has been deleted, and it is likely mediated by an Lhx1-dependent mechanism (Figure 2). In addition, Bonegio et al showed that cells can adopt a podocyte fate in the absence of Notch signalling as well and that the lack of podocyte development might have been observed secondary to a lack of proximal tubule specification 58. Furthermore, experiments from the Nishinakamura group showed that genetic overexpression of Notch2 in the embryonic kidney did not induce proximal fate specification, but led to the depletion of the Six2 positive progenitor pool through premature differentiation towards renal tubules via ectopic Wnt4 activation 59, 60. The result could be consistent with a model where Notch2 activation does not directly lead to proximal fate specification, but rather is necessary for the stabilization of the proximal fate, and by a Notch-controlled progenitor cell maintenance 59.

One can presume that Jagged1 is the Notch2 ligand during development as mice with compound heterozygous mutations for Jagged1 and Notch2 phenocopy the homozygous Notch2equation image mice. In addition to the role of Notch2 in proximal epithelial cell development 52, several studies indicate a role for Notch in the development of the collecting duct. In this nephron segment (which develops from the UB), two different cell types are intertwined: principal cells that are mainly involved in solute and water absorption, and intercalated cells that regulate acid–base homeostasis 61, 62. Jeong et al very elegantly showed that Notch signalling in the developing collecting duct is involved in determining the ratio of principal and intercalated cells 63. They demonstrated that inactivation of Mind-bomb1, an E3 ligase that is expressed in the ligand-expressing cells and is required for efficient Notch activation, led to an increase in intercalated cell number and a decrease in principal cell number in the developing collecting ducts. Given that intercalated cells are involved in regulating the acid–base homeostasis and principal cells in the salt and water homeostasis, Mind-bomb1-deficient mice presented with increased urinary production, decreased urinary osmolality, sodium wasting, and a severe urinary concentrating defect, similar to nephrogenic diabetes insipidus 63.

Moreover, studies performed in the zebrafish pronephros system 64, 65 showed that single or multi-ciliated cell (MCC) fate is controlled by Notch signalling as well. In the zebrafish pronephros, single ciliated cells possess the morphology of typical transport epithelial cells, and MCCs, by several criteria, appear specialized in fluid propulsion. These cell types are distributed in a ‘salt-and-pepper’ fashion in the pronephros, showing some similarities to principal and intercalated cells in the mammalian kidney. Jagged2 is expressed in MCCs and Notch3 is expressed in pronephric epithelial cells. Morpholino knockdown of either Jagged2 or Notch3 dramatically expanded ciliogenic gene expression, whereas ion transporter expression was lost, indicating that pronephric cells had been transfated to MCCs. The reciprocal relationship between ligand and receptor-expressing cells during development gives rise to two different cell types next to each other and drives the ciliary programme in the zebrafish pronephros. It is not clear whether Notch plays any role in regulating ciliary gene expression in the mammalian kidney.

Role of Notch beyond development: acute kidney injury, repair, and regeneration

Compared with development, the expression of Notch receptors and ligands is much decreased in the mature kidney 66. A few tubular epithelial and most probably endothelial cells, however, remain positive for Notch ligand or receptor expression 67. Further studies will be needed to better characterize the exact expression pattern and location of these cells.

Adult organ injury induces regeneration and repair activity, leading to the expression of many different developmental signalling pathways 68. The kidney has a tremendous capacity for repair following different types of acute injury. Experimentally, acute kidney injury (AKI) can be induced by short-term hypoxia (by clamping the renal arteries) or injection of compounds that are directly toxic to tubular epithelial cells (for example, cisplatin, gentamycin or mercury chloride). Proximal tubular epithelial cells are particularly susceptible to ischaemic injury as they have the highest metabolic rate. In the case of AKI, after an initial phase of cell death and de-differentiation, tubular epithelial cells proliferate and re-differentiate, thus replacing cells that underwent apoptosis during injury 69. Hence it is not surprising that in the setting of injury, Notch 50 is reactivated. Studies performed by Kobayashi et al were the first to observe increased expression of Delta1, Notch2, and the downstream target Hes1 in a rat ischaemia–reperfusion injury model 70. Co-culture of Delta1-expressing stromal cells and renal tubular epithelial cells led to an increased proliferation rate of the latter. From this study, the authors concluded that Notch2 expression might play a role in tubular epithelial cell regeneration during AKI. Studies by the Rosenberg group using the same rat model of ischaemia–reperfusion indicate that Dll4 blockage, perhaps via influencing endothelial cells, interferes with repair after ischaemia–reperfusion injury 71. Jagged1/Notch1/HeyL genes were also activated in the folic acid precipitation-induced acute kidney injury model 20. There was some increase in the expression of Notch2 and Hes1 in this model, but the authors failed to see regulation of Dll molecules. Treatment of animals with the pharmacological blocker of the γ-secretase complex had no effect on the rise and decline of serum creatinine and blood urea nitrogen induced by folic acid 20. However, detailed investigation into Notch's functional role using mice with targeted genetic deletion has not been performed in AKI models. In summary, while Notch activation is observed in different models of acute kidney injury, its functional contribution to injury, repair, and regeneration requires further elucidation.

Chronic kidney disease and fibrosis

Kidney fibrosis is the final common pathway leading to chronic kidney disease (CKD). Proper kidney function depends on the correct cellular interaction of over 20 different cell types 2. In fibrosis, this complex architecture is changed and cell–cell interaction balance is altered. It is characterized by the accumulation of collagen, activated myofibroblasts, and inflammatory cells, as well as loss of microvasculature and epithelial cells.

The first evidence suggesting the regulation of Jagged1/Notch1 in CKD came from early gene expression profiling studies performed on mouse models of fibrosis using the unilateral ureteral obstruction model 72. Gene array studies performed on control and diseased human kidneys 73 confirmed the regulation of Notch pathway-related genes in CKD. Using in situ hybridization, Walsh and co-workers 81 showed an increase of Jagged1 and Hes1 expression, as well as their co-expression, in tubules from kidneys of diabetic nephropathy patients. In this dataset, Notch pathway-based gene clustering was able to distinguish diabetic nephropathy samples from control living donor samples and kidneys with minimal change disease. A recent study by Murea and co-workers showed increased levels of Notch pathway protein expression not only in diabetic nephropathy samples, but also in other chronic kidney disease samples associated with kidney fibrosis (Figure 2) 74. The severity of fibrosis correlated with the expression level of cleaved Notch1 in the tubulointerstitium, suggesting that Notch pathway activation might be a common downstream mechanism in a range of chronic kidney diseases (Figure 3).

Figure 3.

Expression of Notch1 in healthy and diseased human kidney. Representative immunostaining images showing cleaved Notch1 staining on control and chronic kidney diseased (CKD) human kidneys, indicating increased expression of cleaved Notch1 in patient samples with fibrosis. Scale bars = 50 µm

The evidence for the functional role of Notch in fibrosis is emerging from a recently performed study by Bielesz et al. In this study, the authors performed comprehensive pharmacological, genetic in vivo and in vitro experiments to analyse the role of Notch in renal tubules during kidney fibrosis 20. They reported that γ-secretase inhibitor treatment significantly alleviated fibrosis development in two different mouse models of fibrosis, induced either by injection of folic acid (FA) or by unilateral ureteral obstruction (UUO). As mice with systemic deletion of Notch-related genes die early during development, the authors had to use a mouse model where Notch was deleted in a cell type-specific manner. Mice in which canonical Notch signalling (using Rbpjfloxed mice) is deleted specifically from proximal tubular epithelial cells (using PEPCKcre mice) did not show any significant developmental phenotype. Using the FA model of fibrosis, Bielesz et al also demonstrated that proximal tubular epithelial cell-specific genetic deletion of Rbpj (PEPCKequation image/Rbpjequation image mice) led to decreased expression of fibrosis markers, such as collagen1a1, smooth muscle actin, and vimentin. In concordance with these findings, conditional inducible expression of the Notch1 intracellular domain (NICD) in tubular epithelial cells induced rapid tubulointerstitial fibrosis (TIF) development in mice (Figure 4). Notch induced collagen and smooth muscle actin transcripts and promoted macrophage influx similar to human CKD and fibrosis. In cultured renal epithelial cells, transforming growth factor beta (TGFβ) is a strong regulator of Notch ligand Jagged1 expression 20, 67, 75, 76 and Notch is an inducer of epithelial-to-mesenchmyal transition (EMT) 75. However, in vivo expression of NICD did not regulate EMT-related transcriptional profiles, indicating that the Notch-induced fibrosis development is independent of EMT.

Figure 4.

Increased expression of Notch in the kidney induces fibrosis development.(A) In the adult kidney, very few tubular epithelial cells express Notch1 and Jagged1. In the setting of injury, expression of Notch1 and Jagged1 is increased. Notch activation induces proliferation and failed differentiation of tubular epithelial cells. In addition, epithelial Notch activation induces fibroblast proliferation and myofibroblast activation. (B) Representative histological images of control mouse kidneys and from mice with tubule-specific Notch1 expression (Pax8rtTA/TREICNotch1)

Similar to Notch's developmental role in mature tubular epithelial cells (TECs), Notch strongly induced expression of the proximal tubule marker cadherin 6 54, 59 and another developmental target gene, Wnt4. On the other hand, markers of terminal differentiation such as solute carriers were decreased, suggesting a loss or altered differentiation of TECs in the presence of Notch. The authors concluded that sustained Notch activation induces a failed or incomplete differentiation of epithelial cells and along with prolonged proliferation and growth factor production, results in a vicious circle, activating other pathways, which in turn perpetuate the process and eventually lead to TIF development (Figure 4) 69.

Recent studies also indicate a broader role for Notch in regulating fibrogenesis. Increased expression of Notch was noted in other models of fibrosis as well, including peritoneal fibrosis and patients and animal models of systemic sclerosis 77–79. Overexpression of Notch induced fibrosis development, while inhibition of Notch using genetic (anti-sense method) or pharmacological inhibitors (GSI) reduced fibrogenesis. In these models, inhibition of Notch was not only able to protect animals from sclerosis, but could also revert established fibrotic lesions.

Role of Notch in glomerular disease

Whereas Notch activity is indispensable for glomerular development, Notch is almost completely absent from healthy adult glomeruli. Recent gene expression studies performed in our and other laboratories showed increased levels of Notch receptors and ligands in different glomerular diseases 73, 74. Antibody-based expression analysis using kidney samples from ten different disease groups showed increased expression of cleaved Notch1, Notch2, and Jagged1 in the glomerulus in different proteinuric nephropathies 80. More importantly, statistical analysis showed that the degree of kidney disease, measured by the amount of albuminuria and glomerulosclerosis, correlated with the level of Notch1, 2, and Jagged1 expression.

Walsh et al demonstrated that in mice, the expression of Notch transcriptional targets decreases in developing podocytes after cell fate determination is complete, suggesting that podocyte development and differentiation are dependent on a temporal fine-tuning of Notch pathway activity 81. The idea was further supported by the fact that ectopic expression of NICD in developing podocytes, using a transgenic mouse model, caused glomerulosclerosis along with proteinuria. Disease development was paralleled by a loss of mature podocyte features. Simultaneous conditional inactivation of Rbpj, the downstream transcriptional partner of Notch (Figure 1), rescued this phenotype, highlighting the role of Notch silencing in the podocyte maturation process and maintenance of a functional glomerular filtration barrier.

In a recent study, Lasagni et al looked at the role of Notch activation in putative renal progenitors in the setting of glomerular disease 82. Performing in vitro experiments, they first showed that Notch activation in renal progenitors induced cell proliferation, whereas Notch pathway silencing was required for progenitor cell differentiation towards the podocyte lineage. Prolonged activation of Notch signalling in these cells induced mitotic catastrophe and cell death. Accordingly, Notch pathway inhibition using a γ-secretase inhibitor in the adriamycin-induced mouse model of glomerulosclerosis alleviated proteinuria and diminished podocyte loss. However, prolonged Notch pathway inhibition throughout the regenerative phases of glomerular injury resulted in increased proteinuria and more severe glomerulosclerosis, suggesting a beneficial role of Notch during later stages of renal regeneration. This model would be consistent with a Notch-regulated stem cell niche, which has been shown in other organs; however, resident renal stem cells have not been conclusively demonstrated in the kidney 83, 84.

Functional studies performed by Niranjan et al using transgenic and knockout animal models showed the deleterious effects of sustained Notch signalling in podocytes 67. Transgenic inducible expression of NICD in podocytes induced rapid development of albuminuria, followed by glomerulosclerosis and death of the animals. Mechanistically, inducible podocyte-specific NICD expression led to p53 expression and subsequent apoptosis. In line with these findings, podocyte-specific genetic deletion of Rbpj, the transcription partner of Notch, protected mice from proteinuria and podocyte loss in the streptozotocin mouse model of diabetes. Moreover, pharmacological inhibition of Notch signalling through the application of a γ-secretase inhibitor ameliorated proteinuria in the puromycin aminonucleoside-induced rat glomerular disease model, indicating that expression of Notch in podocytes is responsible for albuminuria. Similarly, studies performed in a diabetic rat model replicated these findings, as long-term use of a γ-secretase inhibitor protected rats from diabetes-induced albuminuria and glomerulosclerosis development 85. Mechanistic studies indicate that Notch interacts with both the VEGF and the TGFβ pathways in the glomerulus 67, 85, 86. The application of γ-secretase inhibitors ameliorated renal damage in a model of lupus nephritis as well 87. However, the authors speculated that the main effect in this model was most likely attributable to an interference with the immune system rather than to a direct renal effect. In summary, evidence is mounting that there is increased expression of Notch in different glomerular disorders. Sustained activation of Notch in podocytes induced albuminuria and glomerulosclerosis, while inhibition of Notch signalling was beneficial in multiple glomerular disease models.

Renal cell cancer with more or less Notch

There are several sub-types of renal cell cancer (RCC); the majority of RCC cases (83%) are of the clear cell type and about 11–18% belong to the group of chromophil or papillary RCCs. Chromophobe is the least common RCC and seems to have a better prognosis compared with other types. The most common genetic mutations that occur in RCC are mutations in the tumour suppressor gene VHL (von Hippel-Lindau protein). A rare form of RCC is associated with alterations of TSC (tuberous sclerosis complex) 88, 89. As Notch interacts both with the HIF/VHL 90 and with the TSC 91, 92 pathways and is a strong inducer of proliferation in tubular epithelial cells, a role for Notch in RCC may not be very surprising 20.

Tissue microarray analyses showed that there is significantly increased Notch1 expression in chromophobe renal cell carcinoma 93–95. In another study, high Jagged1 expression in RCC samples was statistically linked to reduced overall and disease-free survival, especially at the early stage 93, 96. The Axelsson group showed that the Notch signalling cascade is constitutively active in human clear cell-type cancer cell lines, independently of the VHL/HIF pathway 97. Notch signalling blockade resulted in attenuated proliferation and restrained anchorage-independent growth of RCC cell lines. Treatment of nude mice with an inhibitor of Notch signalling potently inhibited the growth of xenotransplanted RCC cells. Moreover, the growth of primary RCC cells was attenuated upon Notch inhibition. It is interesting to note that transgenic expression of ICNotch1 in tubular epithelial cells, however, did not directly induce RCC development in mice 20. Hence, it is possible that Notch plays a role in RCC progression rather than initiation.

Analysis of a rare form of renal cell cancer—the human type 1 papillary renal cell cancer—indicated reduced Notch signalling in this type of cancer 98. Gene expression analysis indicated that at least one Notch target (Hey1) was reduced, while the expression and nuclear localization of KyoT3/FHL1B were increased. KyoT/FHL1B has been reported to act as an inhibitor of the canonical Notch signalling pathway 98. Consistently, mice with developmental deletion of Notch signalling (Notch1 and 2 or Rbpj) presented with papillary microadenoma formation 66.

Therapeutic potential

Notch signalling is known to be involved in many different oncological diseases such as T-cell lymphoblastic leukaemia 99, colon cancer 93, as well as tumour angiogenesis 100. Therefore, Notch gained significance as a therapeutic target in the field of oncology. The unique circuitry makes Notch signalling an attractive therapeutic target. In this context, ligand binding followed by γ-secretase-mediated cleavage is an important step at which Notch pathway activity can be pharmacologically modified. In addition, due to its ability to cleave amyloid-β precursors, the γ-secretase complex received attention as a potential therapeutic target for Alzheimer's disease 101, 102. Many small molecular compounds targeting the γ-secretase complex have been developed 102, 103 and are being tested in phase I–III human trials. Although very promising, pharmacological Notch pathway inhibition through application of non-specific γ-secretase inhibitors showed many undesired side effects. Most importantly, non-selective γ-secretase inhibitors cause intestinal goblet cell hyperplasia, as inhibition of Notch in intestinal progenitor cells drives differentiation towards goblet cell fate and inhibits enterocyte cell specification 104. Recent work, however, suggests that selective blocking of Notch1 or Notch2 by therapeutic antibody targeting can circumvent gut toxicity 105. Similarly, concomitant administration of glucocorticoids might also alleviate GI toxicity 106. In human studies, an intermittent dosing strategy has been developed that allows gut cells to differentiate and reduces GI side effects. The availability of γ-secretase inhibitors makes Notch an attractive therapeutic candidate; however, further studies will be needed to determine whether these are clinically applicable strategies.

Conclusions and future directions

In conclusion, several studies indicate a key role of Notch in determining proximal epithelial fate in the kidney. Evidence is emerging indicating a reactivation of Notch signalling in the kidney in different disease conditions, specifically in proximal tubular and glomerular epithelial cells. Increased Notch activation in adult renal epithelial cells induced kidney fibrosis and end-stage kidney disease development. Pharmacological and genetic blockade of Notch ameliorated kidney damage in different disease models. The studies reviewed here suggest that Notch signalling in the adult kidney displays a wealth of contextual diversity that we are only beginning to understand. Notch is an important master regulator; therefore, further studies will be needed to better understand the context dependency of Notch signalling. Specifically understanding the activation and the effect of specific ligands and receptors in development and disease would be important. In addition, we need to understand the cross-talk with other pathways to efficiently manipulate Notch signalling. It is clear, however, that Notch is a key player in the developing and adult kidney and will be the focus of many future studies towards understanding the function and dysfunction of the kidney.

Acknowledgements

This work was supported by grants from the National Institute of Health to KS (R01DK076077) and by the Juvenile Diabetes Research Foundation.

Author contribution statement

KS and YS jointly wrote and edited the manuscript and produced the accompanying figures.

Teaching materials

Powerpoint slides of the figures from this review are supplied as supporting information in the online version of this article.

Ancillary