Podocyte GSK3 is an evolutionarily conserved critical regulator of kidney function

Hurcombe, J. A.; Hartley, P.; Lay, A. C.; Ni, L.; Bedford, J. J.; Leader, J. P.; Singh, S.; Murphy, A.; Scudamore, C. L.; Marquez, E.; Barrington, A. F.; Pinto, V.; Marchetti, M.; Wong, L.-F.; Uney, J.; Saleem, M. A.; Mathieson, P. W.; Patel, S.; Walker, R. J.; Woodgett, J. R.; Quaggin, S. E.; Welsh, G. I.; Coward, R. J. M.

doi:10.1038/s41467-018-08235-1

Download PDF

Article
Open access
Published: 24 January 2019

Podocyte GSK3 is an evolutionarily conserved critical regulator of kidney function

J. A. Hurcombe¹^na1,
P. Hartley²^na1,
A. C. Lay¹,
L. Ni¹,
J. J. Bedford³,
J. P. Leader³,
S. Singh¹,
A. Murphy⁴,
C. L. Scudamore⁵,
E. Marquez¹,
A. F. Barrington¹,
V. Pinto¹,
M. Marchetti¹,
L.-F. Wong ORCID: orcid.org/0000-0003-4348-698X⁶,
J. Uney⁶,
M. A. Saleem¹,
P. W. Mathieson^1,7,
S. Patel^8,9,
R. J. Walker ORCID: orcid.org/0000-0003-3366-0956³,
J. R. Woodgett ORCID: orcid.org/0000-0003-3731-5797⁸,
S. E. Quaggin¹⁰,
G. I. Welsh¹ &
…
R. J. M. Coward¹

Nature Communications volume 10, Article number: 403 (2019) Cite this article

7172 Accesses
48 Citations
137 Altmetric
Metrics details

Subjects

Abstract

Albuminuria affects millions of people, and is an independent risk factor for kidney failure, cardiovascular morbidity and death. The key cell that prevents albuminuria is the terminally differentiated glomerular podocyte. Here we report the evolutionary importance of the enzyme Glycogen Synthase Kinase 3 (GSK3) for maintaining podocyte function in mice and the equivalent nephrocyte cell in Drosophila. Developmental deletion of both GSK3 isoforms (α and β) in murine podocytes causes late neonatal death associated with massive albuminuria and renal failure. Similarly, silencing GSK3 in nephrocytes is developmentally lethal for this cell. Mature genetic or pharmacological podocyte/nephrocyte GSK3 inhibition is also detrimental; producing albuminuric kidney disease in mice and nephrocyte depletion in Drosophila. Mechanistically, GSK3 loss causes differentiated podocytes to re-enter the cell cycle and undergo mitotic catastrophe, modulated via the Hippo pathway but independent of Wnt-β-catenin. This work clearly identifies GSK3 as a critical regulator of podocyte and hence kidney function.

Kif21a deficiency leads to impaired glomerular filtration barrier function

Article Open access 06 November 2023

Hanna Riedmann, Séverine Kayser, … Carsten Bergmann

Distinct Functional Requirements for Podocalyxin in Immature and Mature Podocytes Reveal Mechanisms of Human Kidney Disease

Article Open access 10 June 2020

Ido Refaeli, Michael R. Hughes, … Kelly M. McNagny

Loss of CLDN5 in podocytes deregulates WIF1 to activate WNT signaling and contributes to kidney disease

Article Open access 24 March 2022

Hui Sun, Hui Li, … Yongfeng Gong

Introduction

Nearly 700 million people, representing 10% of the global population, are affected by kidney disease manifesting as albuminuria or a reduction in the glomerular filtration rate (GFR)¹. Even small increases in albuminuria (microalbuminuria) or reductions of GFR increase the relative risk of progressing to end-stage renal failure (ESRF), having a cardiovascular event or dying^2,3. In recent years, it has become clear that a key mammalian kidney cell in preventing albuminuria and preserving renal function is the glomerular podocyte. Podocytes are epithelial-like cells, located on the urinary side of the glomerular filtration barrier. They consist of large cell bodies with radiating major and minor processes that wrap around the filtering glomerular capillaries to form the foot processes with regularly spaced slit diaphragms, that provide the final barrier for glomerular filtration. Importantly, podocytes, similar to neurones⁴, become terminally differentiated soon after birth, with minimal capacity to re-enter the cell cycle and proliferate. Mutations in over 40 human genes have been shown to cause albuminuria⁵, which are expressed in the podocyte and their mutations are deleterious. Therefore, understanding important cellular processes that regulate the unique biology of this cell is of fundamental importance.

Understanding of podocyte biology has been greatly facilitated in recent years by the discovery of equivalent cell types in more tractable organisms that can be rapidly genetically manipulated to study their function. A good example of this is the Drosophila nephrocyte. Nephrocytes have several similarities to podocytes, including expression of many analogues of the critical mammalian slit diaphragm podocyte proteins such as nephrin (stick and stones and hibris), NEPH1 (dumbfounded), podocin (Mec 2) and CD2AP (GC31012). Nephrocytes function as endocytotic filtration cells, maintain adult haemolymph⁶ and are involved in cardiac and immune homeostasis⁷.

Glycogen Synthase Kinase 3 (GSK3) is a multi-functional serine/threonine protein kinase that regulates several distinct biological pathways⁸. It was initially described as a component of glycogen metabolism and was later shown to be downstream of insulin signalling. GSK3 is rapidly phosphorylated and inhibited in response to this hormone through activation of the phosphoinositide 3-kinase (PI3K) pathway, contributing to deposition of glycogen⁹. GSK3 has two major biological actions; as a scaffolding protein and a kinase enzyme to catalyse a variety of down-stream targets¹⁰.

GSK3 is evolutionarily conserved across all eukaryotic species. In Caenorhabditis elegans and Drosophila melanogaster it is encoded by a single gene¹¹. In contrast, in mammals GSK3 exists as two isoforms, GSK3α and GSK3β, encoded by different genes on different chromosomes¹¹. These isoforms have 85% overall structural homology with highly conserved kinase domains (97%), with the differences largely confined to the N and C terminal regions¹². Mammalian GSK3 activity is dynamically regulated through phosphorylation of key residues. Phosphorylation at serine 21 (GSK3α) and serine 9 (GSK3β) results in reduced activity¹³. Although GSK3α and β are structurally similar they also have some distinct functions: GSK3β null mice die during late embryogenesis due to liver apoptosis and defective activation of NF-kappa B¹⁴, together with cardiac abnormalities;¹⁵ in contrast GSK3α null mice are viable, have a normal life span and, interestingly, exhibit enhanced insulin sensitivity when on a susceptible genetic background¹⁶. This suggests that, although the isoforms share structural similarity, they have differing biological functions and are not entirely redundant. Multiple cell-specific GSK3 knockout mouse models have been published that illustrate that the functions of the two mammalian GSK3 isoforms are also cell-type dependent^{17,18,19,20,21}.

Recently it has been reported that inhibiting GSK3 in the podocyte may be therapeutically beneficial for a variety of experimental renal diseases. These studies have focused on the GSK3β isoform with less consideration of the α isoform and have either used specific genetic inhibition of GSK3β exclusively in the podocyte²² or pharmacological inhibitors such as lithium, 6-bromoindirubin-3ʹ-oxime (BIO), and thiadiazolidinone (TDZD-8)^{22,23,24,25,26,27}. The beneficial effects of these agents are postulated to be due to inhibition of GSK3β. However, there are no isoform-specific GSK3 inhibitors currently available, and those that are used inhibit both isoforms similarly. The most common GSK3 inhibitor used in clinical practice is lithium carbonate, in the treatment of bipolar disorders. Intriguingly, lithium can cause glomerulosclerosis and ESRF in some patients given this drug for prolonged periods^28,29 but the reason for this effect is unclear³⁰.

As GSK3 and its isoforms exhibit different roles in different cell types^17,19,20,21, in this study, we investigate GSK3’s importance in the podocytes of mice and in the equivalent nephrocytes of Drosophila using genetic and pharmacological approaches. We find that GSK3 is critically important for the function of these cells both during development and in maturity. Furthermore, the evolutionary segregation of GSK3 into two isoforms (α and β) appears protective as either isoform can fully compensate for the other’s loss. Mechanistically, GSK3 maintains the podocyte in its terminally differentiated form and prevents it from re-entering the cell cycle and undergoing mitotic catastrophe, modulated by Hippo pathway signals.

Results

Developmental genetic loss of podocyte/nephrocyte GSK3 is catastrophic

To study the developmental importance of GSK3, podocyte-specific GSK3α, GSK3β and combined GSK3 α/β knockout (podGSK3DKO) transgenic mice were generated. This was achieved by crossing floxed GSK3α¹⁶ and/or GSK3β mice¹⁷ with a podocin Cre mouse³¹ (Supplementary Fig. 1a). Mice were genotyped and genomic excision of GSK3α and β DNA verified (Supplementary Fig. 1b). Furthermore, GSK3 isoform protein loss was confirmed using IHC (Supplementary Fig. 1c).

All genotypes were born with normal Mendelian frequency (Supplementary Table 1) indicating that there was no pre-natal lethality. Single isoform podocyte-specific deletion of GSK3α or β, as well as deletion of three out of four of the GSK3 alleles (i.e. podCreGSK3α^fl/fl β^fl/wt or podCreGSK3α^fl/wtβ^fl/fl) caused no discernible phenotypes, with mice surviving normally to two years of age. However, when both isoforms were simultaneously deleted (i.e. all four floxed GSK3 alleles inactivated [podCreGSK3α^fl/flβ^fl/fl or podGSK3DKO]) the mice all died between postnatal day 10 and 16 (p = 0.003 [Fig. 1a]). Prior to death these mice were indistinguishable from their littermates with regard to external appearance, weight and behaviour (Supplementary Fig. 1e). However, at time of death they had pronounced renal involvement with enlarged pale kidneys (Supplementary Fig. 1f), renal failure (Fig. 1b), acidosis (Fig. 1c), and high levels of albuminuria (Fig. 1d; Supplementary Fig. 1g). Renal disease evolved rapidly after birth as evidenced by a 25-fold increase in albuminuria from P2 to P10 (Fig. 1d). Histological examination using light and transmission electron microscopy (TEM) also revealed that glomerular and renal abnormalities were initially subtle in podGSK3DKO mice, but these rapidly progressed over the first 10 days of life to show glomerulosclerosis, multiple tubular protein casts and major disruption of the glomerular filtration barrier on TEM (Fig. 1e). By P10 highly abnormal, cystic and vacuolated, glomeruli were evident in the podGSK3DKO mice (Fig. 1f).

To assess whether GSK3 plays similar roles in Drosophila nephrocytes we used the dKlf15-Gal4 driver to silence the GSK3 gene here, which is called shaggy. This resulted in a total loss of pericardial nephrocytes, indicative of a critical requirement for shaggy during nephrocyte development (Fig. 1g, h). Interestingly, complete absence of nephrocytes was not overtly detrimental to Drosophila.

Mature genetic loss of podocyte/nephrocyte GSK3 is detrimental

After establishing the developmental importance of GSK3 in the podocyte/nephrocyte we evaluated the role of GSK3 in mature cells. Podocin RtTA -tet-o-Cre GSK3α^fl/flβ^fl/fl (mpodGSK3DKO) mice were generated and given doxycycline from 4 weeks of age to specifically delete GSK3 in their podocytes after glomerular development was complete (Supplementary Figs. 2a–d). These mice were compared to Cre negative and non-doxycycline treated control littermates. All but one of the mpodGSK3DKO mice developed albuminuric kidney disease (Fig. 2a, b), but none of the control mice did. This occurred rapidly in the majority of mpodGSK3DKO mice with many developing albuminuria that was detectable within one week of finishing doxycycline (Fig. 2a, b). There was variability in the severity of their renal disease (Fig. 2a; Supplementary Fig. 3a), with some mpodGSK3DKO mice only mildly albuminuric at 22 weeks of age, but in all there was evidence of mesangial hypercellularity and glomerular fibrosis on trichrome staining (Fig. 2c, Supplementary Fig. 3a). In contrast ~30% of animals had raised serum creatinine relative to controls (Supplementary Fig. 3b), associated with global glomerulosclerosis and widespread interstitial fibrosis throughout the kidney (Fig. 2c; Supplementary Fig. 3a). One mouse was found to have severe hypertension (204 mmHg (SEM ± 2.67) / 158 mmHg (SEM ± 2.37) c/w systolic 115 (SEM ± 3.34) and diastolic 85 (SEM ± 2.46) in controls). mpodGSK3DKO mice had fewer podocytes and reduced nephrin expression compared to littermate controls (Supplementary Figs. 3c, d). Fascinatingly, approximately 20% of mpodGSK3DKO mice developed a florid crescentic glomerulopathy (Fig. 2d). To elucidate the cellular origin of the infiltrating crescentic cells we crossed the mpodGSK3DKO mouse with the mT/mG fluorescent reporter mouse to lineage tag the podocytes. This revealed extracapillary cells were of podocyte origin (Fig. 2e).

High-powered TEM demonstrated foot process effacement in mpodGSK3DKO podocytes together with basement membrane thickening in some mice. Low power TEM images confirmed our observations of mesangial hypercellularity and enlarged glomeruli. However, glomerular endothelial appearance and fenestrations were unchanged (Fig. 2f).

We also temporally deleted GSK3 (shaggy) from the mature nephrocyte in adult Drosophila using the dkfl15-TARGET system. This caused cellular dysfunction with reduced nephrocyte number and a significant loss and mislocalisation of the NEPH1 homolog dumfounded expression on the cells (Fig. 2g, h).

Pharmacological suppression of podocyte/nephrocyte GSK3 is detrimental

Complete genetic deletion of GSK3α and β in the podocyte was highly detrimental. However, of more clinical relevance is the pharmacological inhibition of GSK3 as there are multiple pharmacological compounds that can do this, and there is great interest in their potential use in a variety of important diseases including diabetes, cancer and dementia^32,33. The most common agent of use in clinical practice is lithium, which is used for bipolar conditions. It is unknown how lithium precisely inhibits GSK3 activity but it is proposed that it has a dual inhibitory action by interfering with a secondary magnesium binding site as well as increasing inhibitory phosphorylation of both serine 21 for GSK3α, and serine 9 for GSK3β¹⁰. We initially confirmed that lithium caused phosphorylation of these inhibitory sites in conditionally immortalised human and murine podocytes (Supplementary Figs. 4a–d). To assess the functional significance of prolonged podocyte GSK3 inhibition we then studied Wistar rats that were chronically treated with lithium for 6-months. This caused inhibitory phosphorylation of the GSK3 isoforms, together with stabilization of β-catenin specifically in their podocytes within their glomeruli (Fig. 3a). β-catenin activation occurs when GSK3 is sufficiently inhibited and thus unable to regulate the phosphorylation-dependent degradation of β-catenin via its ubiquitinylation and destruction by the 26S proteasome²¹. Lithium-treated rats developed significant proteinuria (Fig. 3b) and glomerulosclerosis (Fig. 3c, d) in comparison to their age and sex matched controls.

We then examined the direct impact of lithium on nephrocyte function in Drosophila using an ex vivo culture system involving semi-intact fly preparations. Within 48 h of lithium treatment dose-dependent nephrocyte hypertrophy was observed (Fig. 3e), coupled with mislocalisation of dumbfounded from uniformly short linear arrays, consistent with previous reports³⁴, and to an accumulation of bright, punctate Duf foci across the nephrocyte surface (Fig. 3f). Furthermore, lithium caused severe functional abnormalities to the nephrocytes’ endocytic function. At lower doses (1 mM), there was an increase in the Dextran signal within nephrocytes (Fig. 3g). However, when the cells were exposed to 20 mM lithium for 48 h it resulted in major issues with their endocytic capabilities with the formation of large lacunae and uptake of large Wheat Germ Agglutinin (WGA) (Fig. 3h). In similar experiments, flies provided with food containing 5 mM lithium for 1 week showed nephrocyte loss, reduced Duf staining and increased large protein WGA accumulation (Fig. 3i). These phenotypes are consistent with progressive loss of diaphragm integrity and comparable to foot process effacement seen in injured mammalian podocytes (Supplementary Fig. 4e). Finally, we examined a kidney biopsy from a patient who had been on long-term lithium therapy and found that they also had inhibitory GSK3 phosphorylation in their podocytes compared to control human kidney samples (Fig. 3j).

Podocyte GSK3 loss activates Wnt-β-catenin but this is not responsible for pathology

Genetic deletion of GSK3αβ in the podocytes of podGSK3DKO mice caused substantive activation and nuclear translocation of β-catenin in this cell type. Interestingly β-catenin was also activated throughout the kidney in the tubular compartment (Fig. 4a, b; Supplementary Fig. 5a), which we subsequently identified was probably a consequence of massive albuminuria (Supplementary Fig. 5b). Similarly, pharmacological inhibition of GSK3 also activated β-catenin in the podocyte (Fig. 3a).

As β-catenin activation and transcription of its target genes have been shown to be detrimental in other cell types when GSK3 is lost^35,36, we hypothesized that β-catenin activation was the critical factor causing kidney damage in this model. We therefore generated transgenic mice models in which GSK3α, GSK3β and β-catenin were all contemporaneously deleted specifically from the podocyte both developmentally and in maturity. We confirmed that β-catenin was knocked down in these models using polymerase chain reaction (PCR), immunohistochemistry (IHC) and immunofluorescence (IF) (Supplementary Figs. 5c–e). However, a loss of podocyte β-catenin did not improve survival or kidney involvement in the developmental model (Fig. 4c–e) or renal damage in the mature model (Fig. 4f).

To support these findings, and given the contribution of Wnt/β-catenin along with Notch and Hedgehog pathways to the detrimental phenotype observed in other cells lacking GSK3³⁵ we also used pharmacological inhibitors of these pathways in a genetic GSK3αβ podocyte knockout cell line we developed (described in detail in next section). We inhibited Wnt signalling with iCRT3, NOTCH with DAPT and Hedgehog with GANT 61 but none of these improved cell survival (Fig. 4g–i; Fig. 8c). Interestingly, and unexpectedly, inhibiting Hedgehog signalling significantly increased cell death in this model (Fig. 4i; Fig. 8c).

Podocyte GSK3 loss causes mitotic catastrophe

To elucidate the mechanisms causing cellular damage when GSK3 was lost we generated a conditionally immortalised temperature-sensitive SV40 antigen podocyte cell line from a GSK3α^fl/flβ^fl/fl mouse. This allowed us to initially culture healthy podocytes and then temporally delete GSK3 using lentiviral delivered Cre recombinase to make a GSK3α/β knockout cell line (cipodGSK3DKO) (Supplementary Figs. 6a–c). Three days after Cre transduction there was ~85% reduction of GSK3α and a ~70% reduction of GSK3β (Fig. 5a; Supplementary Fig. 6d) but the podocytes appeared healthy without any detectable cell loss. Therefore, we elected to interrogate this time point to identify early GSK3 driven mechanistic processes. More prolonged loss of GSK3, for 7 days, caused podocytes to be lost and microscopically appear unhealthy (Fig. 5b; Supplementary Fig. 6e).

Non-biased Tandom Mass tagged (TMT) LC-MS/MS proteomic analysis of the cells 3 days after Cre transduction identified 486 proteins that were up-regulated more than 30% in cipodGSK3DKO cells compared to their controls (non-Cre treated GSK3α^fl/flβ^fl/fl podocytes or lentiviral Cre expressing wild-type podocytes) at a statistical significance p < 0.01. STRING (search tool for the retrieval of interacting genes/proteins) was used to visualise the protein–protein interactions and biological processes enriched in this data set. It revealed enrichment of multiple proteins involved in the cell cycle, including many involved in mitosis and mitotic spindle formation (Fig. 5c; Supplementary Fig. 7; Supplementary Table 3). To explore these findings further we performed fixed (Fig. 5d), and live (Supplementary Movie 1 and 2), phase contrast light microscopy of the cipodGSK3DKO cells. These confirmed that the cells were re-entering the cell cycle and attempting to divide as shown by significantly increased numbers of bi-nucleate and multi-nucleate cells in comparison to controls. However, many podocytes were unable to complete cytokinesis and appeared to be undergoing apoptosis and dying (Supplementary Movie 1 and 2). Detailed analysis of ciGSK3DKO cell numbers using an IN-Cell analyser confirmed these observations (Fig. 5b). The presence of multiple nuclei suggested that mitosis was occurring in ciGSK3DKO cells and western blot analysis of ciGSK3DKO cells 5 days after GSK3 knockdown showed a significant increase in phosphorylated Histone H3 (Fig. 5e and Supplementary Fig. 6f), a commonly used marker of mitosis associated with chromatin condensation³⁷. Cyclin B1 is not usually detectable in healthy, mature podocytes³⁸ but our proteomic analysis revealed that it was one of the most up-regulated proteins in ciGSK3DKO cells (Fig. 5c and Supplementary Table 3). This was confirmed by western blotting along with increased phosphorylation of Cdk1 (Fig. 5e; Supplementary Fig. 6f). Levels of the Cyclin B1/Cdk1 complex increase during G2 of the cell cycle with dephosphorylation occurring at the G2/M transition³⁹. The absence of this dephosphorylation event suggests impaired mitotic progression in ciGSK3DKO cells. cipodGSK3DKO cells then appeared to undergo apoptosis as Western blot analysis of cells at 7 days revealed increased cleavage of caspase 3 and PARP (Fig. 5f; Supplementary Fig. 6g). Collectively, these data demonstrate that cipodGSK3DKO cells were undergoing mitotic catastrophe. We also looked at our podGSK3DKO and mpodGSK3DKO models and found evidence of DNA accumulation in their podocytes with nuclear PCNA staining suggesting they were attempting to re-enter the cell cycle (Fig. 5g). Furthermore, in the mature mpodGSK3DKO mice we detected occasional mitotic figures in their differentiated podocytes (Fig. 5h).

Hippo signalling is disrupted in GSK3 deficient podocytes

Proteomic analysis also identified that an important protein in the Hippo-signalling pathway, Ajuba, was significantly increased by 200–300% in the ciGSK3DKO cells (Figs. 5c and 6a). This was validated by western blotting in the ciGSK3DKO cells (Fig. 6b; Supplementary Fig. 8a). Furthermore, an increased level of Ajuba was also detected in podocytes which had GSK3 pharmacologically supressed using either lithium and a more specific GSK3 inhibitor called CHIR99201 (Fig. 6c and d; Supplementary Figs. 8b and 8c). The Hippo pathway leads to phosphorylation of YAP/TAZ proteins and prevents these proteins from translocating from the cytoplasm into the nucleus, where they bind with TEAD transcription co-activators and ultimately promote transcription of a number of genes that are associated with proliferation and apoptosis⁴⁰. Our proteomic analysis revealed significant up-regulation of a range of YAP/TAZ TEAD targets⁴¹ (Supplementary Table 4). Classically, when Ajuba increases it inhibits Hippo signalling by preventing the phosphorylation of YAP/TAZ thereby resulting in the nuclear translocation of these proteins⁴². In support of this mechanism we detected nuclear YAP/TAZ translocation in ciGSK3DKO cells (Fig. 6e), as well as the lithium and CHIR99201 treated podocytes (Fig. 6f and g). This was associated with induction of Cyclin B1 in all these models (Figs. 5e and 6c, d; Supplementary Figs. 6f and 8c). Analysis of our in vivo podGSK3DKO and mpodGSK3DKO mouse models also revealed significant up-regulation of Ajuba and nuclear YAP/TAZ translocation in their podocytes (Fig. 7a–d; Supplementary Fig. 9a). This was associated with an increased number of cells positive for Cyclin A2, a cell cycle protein and a recognised target of YAP/TAZ-TEAD⁴⁰ (Fig. 7e). We also detected an increase of another YAP/TAZ TEAD target, c-myc⁴³ in both podGSK3DKO and mpodGSK3DKO mice (Fig. 7f and Supplementary Figs. 9c, d). Finally, we examined the glomeruli of our lithium treated rats and human biopsy. These both showed increased podocyte Ajuba staining (Fig.7g).

Nuclear translocation of YAP/TAZ also occurred in the podGSK3DKO/β-cateninKO and mpodGSK3DKO/β-cateninKO mice (Supplementary Fig. 9b) indicating that this process is independent of Wnt- β-catenin signalling.

We then investigated if inhibiting YAP/TAZ activity in the nuclei of GSK3α/β knockdown podocytes was beneficial. Treating ciGSK3DKO podocytes with the YAP/TAZ-TEAD inhibitor verteporfin, improved cell survival by approximately 50% demonstrating that at least some of the adverse effects of podocyte GSK3 loss are mediated through the Hippo pathway (Fig. 8a, c; Supplementary Fig. 10a). Western blot analysis showed that verteporfin was also able to attenuate Cyclin B1 accumulation in ciGSK3DKO cells (Fig. 8b; Supplementary Fig. 10b). This was also the case in lithium and CHIR99201 treated podocytes (Fig. 8d, e; Supplementary Figs 10c,10d).

Finally, we assessed the role of Hippo signalling in the nephrocytes of Drosophila. Verteporfin protected inducible genetic shaggy knockout nephrocytes from the loss and mis-localisation of dumfounded (Fig. 9a–c) and also lithium induced nephrocyte damage (Fig. 9d, e). Collectively this data supports a key role of dysregulated Hippo signalling when podocyte or nephrocyte GSK3 activity is lost (Fig. 9f).

Discussion

Here, we have comprehensively assessed the evolutionary importance of GSK3 in podocytes and nephrocytes, which are key cells that control excretory function in mammals and Drosophila. Our findings show that GSK3 is critically important in these cells both developmentally and in maturity, as sufficient loss of this enzyme results in severe phenotypes in both distinct species.

We initially assessed the developmental importance of GSK3 using a genetic approach to study the mammalian podocyte in transgenic mice. This revealed a high level of compensation between the GSK3 isoforms in the podocyte. Indeed, possessing a single allele of GSK3α or β was sufficient to maintain normal function of this cell. In contrast, simultaneous loss of both isoforms caused an extreme phenotype with all mice dying in renal failure in the neonatal period. This severe phenotype was mimicked in our Drosophila model, where nephrocyte-specific silencing of shaggy resulted in the complete loss of this cell. This shows firstly that GSK3’s action in these excretory cell types is fundamentally important and has likely been preserved throughout evolution. Secondly, the evolutionary duplication of GSK3 into two mammalian isoforms is beneficial for the podocyte as it allows a high level of compensation within this system should a single isoform be lost. This contrasts to some other cell types where the β isoform appears to be the predominant mammalian isoform that maintains normal homeostatic cellular function^44,45,46.

We then investigated GSK3’s importance in maturity. We found that mpodGSK3DKO mice developed a spectrum of kidney disease, ranging from albuminuric mesangial hyper-cellularity to glomerulosclerosis, severe hypertension and renal failure. In some mice a proliferative crescentic glomerulopathy ensued, suggesting that the terminally differentiated podocyte cells had re-entered the cell cycle. We suspect the variety of kidney disease we observed was due to variable degrees of efficiency of excision of GSK3 in this model, and because the mice were on mixed genetic backgrounds. The crescentic glomerulopathy in 20% of mice is interesting as the capacity of the mature podocyte to re-enter the cell cycle and proliferate is widely debated and controversial, but our studies support previous work suggesting that terminally differentiated podocytes can re-enter the cell cycle, divide and form crescents in inflammatory glomerulonephritis⁴⁷ and in HIV associated nephropathy⁴⁸. Conceivably, GSK3αβ loss in the podocyte could be a common final pathway in a variety of crescentic conditions and this is currently under investigation. Our mature Drosophila studies confirmed that GSK3 is also important for maintaining nephrocyte function after development in this species, showing this is also evolutionarily conserved.

After evaluating the key role of podocyte/nephrocyte GSK3 in our genetic models we investigated whether in vivo pharmacological inhibition of GSK3 was also detrimental to excretory function, which we found to be the case both in rodents and Drosophila. We did this for three reasons: firstly because recent experimental work has suggested that pharmacologically inhibiting podocyte GSK3 activity could be beneficial in treating a variety of renal diseases, predominantly through suppression of the β isoform;^22,25,26 secondly because there is great therapeutic interest in pharmacologically inhibiting GSK3 in a number of important disease processes including type 2-diabetes⁴⁹, bipolar disorders²¹, Alzheimer’s disease²¹, cardiovascular disease⁵⁰ and several cancers⁵¹, thereby making it likely that new GSK3 inhibitors will be trialled in patients in the future; and thirdly because some pharmacological GSK3 inhibitors are currently already widely used in clinical practice. The commonest of these is the mood-stabilizer lithium⁵². This drug is normally well tolerated, and when administered at doses that achieve therapeutic levels, only suppresses cellular GSK3 activity by approximately 25%²¹. Unfortunately some patients taking lithium develop renal side effects, the most common being diabetes insipidus, but the most severe is ESRF, which is 6–8 times more common in these patients, especially when they have taken the medication for a prolonged period of time^53,54,55. It has previously been postulated that lithium-associated ESRF is due to interstitial fibrosis⁵⁶, however, there are multiple reports showing that glomerulosclerosis also commonly co-exists as renal function deteriorates^28,29,57. We suspect that in some patients receiving lithium there is excessive suppression of GSK3α/β activity in their podocytes and this leads to glomerular and kidney damage. Going forward it will be desirable to define precisely which downstream pathways are differentially regulated by GSK3α and GSK3β in the podocyte as this could elucidate both the beneficial and detrimental pathways that are regulated by this multi-functional enzyme. Our work, and the recent studies proposing the potential therapeutic benefit of suppressing the β isoform alone^22,25,26, also support the development of isoform-specific GSK3 inhibitors as it is possible that these could be therapeutically beneficial, without the unwanted side-effects associated with excessive suppression of both GSK3α and β. Finally, it may be prudent to monitor the levels of albuminuria in patients who have been on long-term or high-dose GSK3 inhibitors, such as lithium. Detecting albuminuria would suggest excessive GSK3 inhibition in the podocyte and may result in reduction of the drug dose or change of therapeutic approach. This could conceivably prevent the patient from developing chronic or end-stage renal failure.

Mechanistically, we initially hypothesized that the activation of β-catenin within the podocyte was the major pathway causing cellular damage when GSK3 activity was lost here, as this is the case in other cell types lacking GSK3, including neurones³⁵ and mammary cells³⁶. Furthermore, podocyte β-catenin deletion has been shown to be protective in a variety of experimental glomerular diseases⁵⁸, suggesting that its activation is detrimental. However, we did not find that β-catenin activation was responsible for the detrimental pathology we observed. Instead we discovered that a lack of podocyte GSK3αβ caused these terminally differentiated cells to re-enter, but not exit, the cell cycle and to undergo mitotic catastrophe. We were surprised that this differed from neural progenitors lacking GSK3α/β, as these cells have several similarities to podocytes including being terminally differentiated⁴. Neural progenitors lacking GSK3α and β proliferate uncontrollably due to activation of Wnt-β-catenin, Notch, and Hedgehog pathways³⁵, which is not the case in podocytes. Rather, podocytes enter, but fail to exit, the cell cycle and then undergo apoptosis as a result of mitotic catastrophe in a manner similar to mature cardiomyocytes that completely lack GSK3⁵⁹. We suspect that the necessity of GSK3 in controlling cell cycling in the podocyte explains why our podGSK3DKO mice were initially unaffected at birth but then developed severe kidney disease and subsequently died after 10–16 days. This indicates that GSK3 activity is important in the later stages of podocyte development, facilitating transition from a proliferating to a functional, terminally differentiated cell. We do not know of any other gene that has been developmentally knocked out in the podocyte that causes a similar phenotype.

Sufficient genetic or pharmacological suppression of GSK3 in the podocyte caused the Hippo pathway to be switched off, preventing the phosphorylation of YAP/TAZ, resulting in its subsequent nuclear translocation. Importantly, inhibiting the effects of nuclear YAP/TAZ translocation with verteporfin improved podocyte cell survival in ciGSK3DKO cells, showing that this is an important pathway modulated by this enzyme. This was also the case in genetic or pharmacologically GSK3-suppressed nephrocytes showing again that this is an evolutionarily conserved pathway in these cell types.

Interestingly, our results show that nuclear YAP/TAZ translocation caused by GSK3 loss in podocytes is independent of Wnt/β-catenin signaling in contrast to observations in other cell types⁶⁰. We note that inhibiting nuclear YAP/TAZ actions only rescued survival in ciGSK3DKO podocytes by 50% and that inhibiting Notch, Wnt-β catenin, or Hedgehog did not enhance the survival of the cells. It will therefore be interesting to define the other critical pathways regulated by GSK3 in the podocyte.

In conclusion, this study has shown that GSK3 is a critical regulator of cellular function in podocytes and nephrocytes. Mechanistically it is important for maintaining the differentiated phenotype of these cells and preventing them from re-entering the cell cycle resulting in apoptosis through mitotic catastrophe. If treating patients with GSK3 inhibitors, care needs to be taken to ensure adequate GSK3 function remains in the podocyte in order to protect against the development of severe kidney disease.

Methods

Rodent models

Mice in which exon two of GSK3α¹⁶ or GSK3β¹⁷ has been flanked by loxP sites were crossed with podocin-Cre mice³¹ to generate podocyte-specific GSK3α and/or GSK3β knockout animals from embryonic day 12 (podGSK3DKO mice) (Supplementary Fig. 1a for breeding scheme). Cre negative mice, or mice with 2 or 3 of the 4 GSK3 alleles inactivated, served as controls. For the maturity model podocin rtTA and tet-o-Cre⁶¹ mice were crossed with GSK3α/β floxed mice to generate doxycycline inducible podocyte specific GSK3α/β knockout mice (mpodGSK3DKO mice) (Supplementary Fig. 2a for breeding scheme). To induce GSK3 deletion, 4-week old mice were given doxycycline (Sigma) via their drinking water (2 mg/ml doxycycline in 5% sucrose) for 3 weeks. Controls were podocin rtTA or tet-o-Cre negative mice together with podocin rtTA and tet-o-cre positive but without floxed GSK3 genes who also received doxycycline. We confirmed the cellular specificity of the Cre drivers by crossing the podocin-Cre mouse and the pod rtTA tet-o-cre mouse with a membrane-targeted dimer Tomato/ Cre mediated membrane-targeted Green Fluorescent Protein reporter mouse (mT/mG)⁶² (Supplementary Figs. 1d and 2b).

For the triple GSK3α /GSK3β /β-catenin podocyte specific transgenic mice we also crossed in the floxed β-catenin mouse floxed Ctnnb1gene encoding β-catenin⁶³ (The Jackson Laboratory).

All mice were on a mixed genetic background with contributions from 129/SV, FVB, DBA2J and C57BL/6. Both sexes were studied, and no phenotypic differences observed.

Transgenic mouse work was carried out in accordance with the University of Bristol’s institutional guidelines and procedures approved by the United Kingdom (UK) Home Office in accordance with UK legislation.

For the rat experiments seven male Wistar rats were given lithium at an initial dose of 40 mmol lithium/kg dry food for 7 days and then 60 mmol lithium/kg dry food until 6 months of age. Rats given lithium had access to a salt block to maintain sodium balance and prevent lithium intoxication. This resulted in the rats having lithium levels similar to those found in humans of 0.3 to 1.3 mmol/l. These rats were compared to 6 age-matched control male Wistar rats given standard rodent diet (Speciality Foods, Perth, Australia) for 6 months. This study had ethical approval from the University of Otago Animals Ethics committee under New Zealand National Animal Welfare guidelines.

Drosophila husbandry

The Oregon R wild type Drosophila stock, tub-Gal80^ts and shaggy RNAi lines were ordered from Bloomington. dKlf15-Gal4 flies and dKlf15-TARGET flies were used as previously described⁶⁴. For temporal shaggy knockout, adults were reared at 18 °C until eclosure, then transferred to 29 °C. This temperature shift permits the expression of RNAi to shaggy specifically in nephrocytes at the higher but not the lower temperature. Driver and RNAi over-expression combinations were generated by standard crosses. Flies were maintained under standard laboratory conditions (12 h: 12 h light dark, @ 25 °C, with ad lib access to food).

Cell lines

Conditionally immortilsed podocyte cell lines were cultured in RPMI1640 supplemented with 10% Fetal Bovine Serum and 1% penicillin/streptomycin. Cells were initially grown at 33 °C to allow proliferation before switching to 37 °C for 10–14 days to allow cells to differentiate. Cell lines were authenticated by western blotting for typical podocyte marker proteins.

End point PCR

Genomic DNA was isolated from mouse tissues using a blood and tissue DNA extraction kit (Qiagen). 500–1000 ng DNA was used in end point PCR reactions using hotmaster taq polymerase (5 Prime).

Urine and blood analyses

Gross albuminuria was detected using Coomassie staining of 2 μl of urine run on an SDS-PAGE gel and comparing to bovine specific albumin controls of 1 μg and 5 μg run on the same gel. Quantitative analysis was also performed using a mouse-specific albumin ELISA (Bethyl) and creatinine companion kit (Exocell), following the manufacturers methodology. Serum creatinine, urea and CO₂ was measured using the Roche Cobas system with reagents and protocols supplied by the manufacturer.

Histology and electron microscopy

Tissues were fixed in 10% buffered neutral formalin, further processed and paraffin embedded. 3 μm sections were stained using Periodic acid Schiff (PAS) and Masson’s trichrome staining kits (Sigma) or by haematoxylin and eosin (H&E) using standard techniques. Tissues were imaged using a Leica DM2000 microscope and micrographs taken with Leica Application Suite. Image analysis was performed with ImageJ; all images were contrast enhanced using the same parameters.

Histological abnormalities in mpodGSK3DKO mice were assessed by an experienced pathologist and scored as follows: 0 = absent; 1 = mild; 2 = moderate; 3 = severe.

Tissues for electron microscopy were fixed in 0.1 M sodium cocodylate, 2% glutaraldehyde, and imaged on a Technai 12 transmission electron microscope.

Imaging nephrocytes

The abdomen of adult flies was dissected open and eviscerated to reveal the heart and adjacent pericardial nephrocytes. Under phase optics, nephrocytes are readily identifiable by their characteristic size and anatomical location. Specific markers were generated by raising antisera to a peptide corresponding to amino acids 32–46 of dumbfounded, the Drosophila orthologue of human KIRREL/NEPH1. Tissues were counterstained with WGA and DRAQ5 to see nuclei. Fixed tissues were imaged using a Leica SP8 confocal microscope and micrographs taken with Leica Application Suite. Image analysis was performed with ImageJ; all images were contrast enhanced using the same parameters.

Immunohistochemistry

Tissues were fixed in 10% buffered neutral formalin, processed and paraffin embedded. Three-micrometre sections were deparaffinised in Histo-Clear and rehydrated through a graded alcohol series. Antigen retrieval was in 10 mM citrate buffer, pH6 for either 1 h at 65 °C (activated β-catenin) or by boiling for 5 min (pGSK3α/β, YAP/TAZ). Sections were quenched using 3% H₂O_2, followed by a blocking step in 1% BSA in PBS (activated β-catenin and pGSK3α/β) or tris buffered saline with 1% tween 20 (TBST) and 5% normal goat serum (Ajuba; YAP/TAZ; pYAPs127) for 30–45 min. Sections were incubated with primary antibodies overnight at 4 °C (activated β-catenin 1:500; pGSK3α/β 1:20; Ajuba 1:100; YAP/TAZ 1:100; pYAPs127 1:100). Sections were washed then incubated with Signalstain Boost detection reagent (Cell Signalling Technology) for 30 min at room temperature. Signalstain DAB substrate kit (Cell Signalling Technology) was applied for 1–2 min and the sections dehydrated and mounted in DPX (Sigma). Tissues were imaged using a Leica DM2000 microscope and micrographs taken with Leica Application Suite. Image analysis was performed with ImageJ; all images were contrast enhanced using the same parameters.

IF staining of kidney sections

Frozen kidneys were sectioned at 5 μm. Sections were blocked in phosphate buffered saline (PBS) containing 3% Bovine Serum Albumin (BSA) and 0.3% triton X-100 for 1 h, then incubated with primary antibodies overnight at 4 °C (nephrin 1:300; total β-catenin 1:100; c-myc 1:100). Following 3 phosphate buffered saline (PBS) rinses, sections were incubated with fluorophore-conjugated secondary antibodies (Fisher) for 1 h at room temperature. Tissues were imaged using a Leica DM2000 microscope and micrographs taken with Leica Application Suite. Image analysis was performed with ImageJ; all images were contrast enhanced using the same parameters.

Stimulation of cultured podocytes with LiCl and CHIR99201

Conditionally immortalised mouse and human podocytes were incubated with LiCl (Sigma) at 20 mM or CHIR99201 (Sigma) at 3 μM for the times indicated.

Western blotting

Cultured cells or tissue samples were lysed in radioimmunoprecipitation assay (RIPA) buffer supplemented with protease and phosphatase inhibitors (Sigma). Ten to 30 μg of protein was resolved by electrophoresis then transferred to a polyvinylidene difluoride (PVDF) membrane (Millipore). Membranes were blocked in TRIS-buffered saline with 0.1% tween 20 and 5% BSA for 1 h then incubated overnight with primary antibody at a dilution of 1:1000. Membranes were washed before incubation with horseradish peroxidase conjugated secondary antibody (Sigma). Immunoreactive bands were visualised using Clarity ECL Western blotting substrate (Biorad) on a GE AI600 imager. Densitometry was performed using ImageJ software. All uncropped blots can be found in Supplementary Figs. 11–13.

Lentiviral transduction of GSK3αβ floxed podocyte cell line

Kidneys were isolated from GSK3αfl/flβfl/fl mice and used to make a temperature-sensitive SV40 conditionally immortalised podocyte cell line as described previously⁶⁵. Cells were cultured at 33 °C and when 50% confluent were transduced with a lentivirus expressing Cre recombinase. Transduction was in RPMI media with hexadimethrine bromide (Sigma) at 4 μg/ml and the virus used at a multiplicity of infection of 1. Following a 24-h incubation, the lentivirus was removed and replaced with fresh media. Cells were thermo-switched to 37 °C and incubated for a further 3–7 days before imaging and protein extraction.

Proteomics

GSK3α/β floxed podocytes were incubated at 37 °C for 3 days following overnight transduction with Cre expressing lentivirus. Untransduced cells and wild-type mouse podocytes transduced with the lentivirus served as controls. Cells were lysed in RIPA buffer and subjected to LC-MS/MS using isobaric TMT labelling using methodology as previously described⁶⁶.

Fixed and live imaging of cipodGSK3DKO podocytes

The signalling pathway agonists/inhibitors Verteporfin (Tocris); iCRT3 (Sigma); DAPT (Merck Millipore) and GANT61(R and D Systems) were applied at the concentrations indicated during a 7-day incubation following 24-hour transduction of GSK3α/β floxed cells with Cre recombinase expressing lentivirus. Cells were stained with Hoechst (Sigma) at 1 µg/ml, imaged, using an IN Cell analyser 2200 (GE Healthcare) and cell number determined using IN Cell analyser workstation software (GE Healthcare).

Live cell imaging was carried out 5–7 days after viral transduction using an IncuCyte ZOOM cell imaging system (Essen Biosciences).

YAP/TAZ nuclear translocation assay

ciGSK3DKO podocytes or wild-type podocytes treated with LiCl or CHIR99201 were cultured in 96 well plates (Corning). Cells were washed three times in PBS, blocked using 3% BSA, 0.3% triton X-100 for 1 h before overnight incubation at 4 °C with a YAP/TAZ antibody at 1:100. Following 3 PBS washes, cells were incubated for 1 h with fluorophore-conjugated secondary antibody. Nuclei were visualised using Hoescht at 1 μg/ml and cells imaged using an IN Cell analyser 2200 (GE Healthcare). Nuclear/cytoplasmic fluorescence intensity was determined using In Cell analyser workstation software (GE Healthcare).

Antibodies

Antibodies were obtained from Cell Signaling Technology (pGSK3α (Ser21) #9316; pGSK3β(Ser9) #9323; pGSK3α/β(Ser21/9) #9331; p27Kip1 #2552; total β-catenin #8480; Cyclin B1 #12231; pcdk1(Tyr15)#4539; phosphoHistone3(ser10)#3377; cleaved Caspase 3 #9664; PARP #9542; YAP/TAZ #8418; pYAP (ser127) #4911; Ajuba #4897), Life Technologies (total GSK3 clone 21 A); Millipore (activated β-catenin clone 8E7), Sigma (β-actin; Podocin #0372; GAPDH), Acris (nephrin), Santa Cruz Biotechnology (Synaptopodin H-140; p27kip1), and Genetex (SV40 T antigen).

Statistics

Statistical analysis was performed using GraphPad Prism software. When comparing two groups t-tests were used. When comparing more than two groups ANOVA was used with appropriate post hoc analysis. Statistical tests used, and n numbers are shown in figure legends. Data are presented as the mean and error bars represent standard error of the mean. For survival characteristics Kaplan–Meier survival plots were generated and log-rank analysis performed. P values less than 0.05 were deemed statistically significant.

Data availability

The authors declare that all data supporting the findings of this study are available within the article and its supplementary information files or from the corresponding author upon reasonable request.The mass spectrometry proteomics data have been deposited in the PRIDE repository as part of the ProteomeXchange Consortium⁶⁷ under the dataset identifier PXD012145.

References

Eckardt, K. U. et al. Evolving importance of kidney disease: from subspecialty to global health burden. Lancet 382, 158–169 (2013).
Article Google Scholar
Tonelli, M. et al. Chronic kidney disease and mortality risk: a systematic review. J. Am. Soc. Nephrol. 17, 2034–2047 (2006).
Article Google Scholar
Perkovic, V. et al. The relationship between proteinuria and coronary risk: a systematic review and meta-analysis. PLoS Med. 5, e207 (2008).
Article Google Scholar
Sun, Y. et al. The expression and significance of neuronal iconic proteins in podocytes. PLoS ONE 9, e93999 (2014).
Article ADS Google Scholar
Bierzynska, A., Soderquest, K. & Koziell, A. Genes and podocytes - new insights into mechanisms of podocytopathy. Front Endocrinol. 5, 226 (2014).
Google Scholar
Denholm, B. & Skaer, H. Bringing together components of the fly renal system. Curr. Opin. Genet. Dev. 19, 526–532 (2009).
Article CAS Google Scholar
Soukup, S. F., Culi, J. & Gubb, D. Uptake of the necrotic serpin in Drosophila melanogaster via the lipophorin receptor-1. PLoS Genet. 5, e1000532 (2009).
Article Google Scholar
Jope, R. S. & Johnson, G. V. The glamour and gloom of glycogen synthase kinase-3. Trends Biochem. Sci. 29, 95–102 (2004).
Article CAS Google Scholar
Roach, P. J. Glycogen and its metabolism. Curr. Mol. Med. 2, 101–120 (2002).
Article CAS Google Scholar
Jope, R. S. Lithium and GSK-3: one inhibitor, two inhibitory actions, multiple outcomes. Trends Pharmacol. Sci. 24, 441–443 (2003).
Article CAS Google Scholar
Saidi, Y., Hearn, T. J. & Coates, J. C. Function and evolution of ‘green’ GSK3/Shaggy-like kinases. Trends Plant. Sci. 17, 39–46 (2012).
Article CAS Google Scholar
Woodgett, J. R. Molecular cloning and expression of glycogen synthase kinase-3/factor A. EMBO J. 9, 2431–2438 (1990).
Article CAS Google Scholar
Frame, S., Cohen, P. & Biondi, R. M. A common phosphate binding site explains the unique substrate specificity of GSK3 and its inactivation by phosphorylation. Mol. Cell 7, 1321–1327 (2001).
Article CAS Google Scholar
Hoeflich, K. P. et al. Requirement for glycogen synthase kinase-3beta in cell survival and NF-kappaB activation. Nature 406, 86–90 (2000).
Article ADS CAS Google Scholar
Kerkela, R. et al. Deletion of GSK-3beta in mice leads to hypertrophic cardiomyopathy secondary to cardiomyoblast hyperproliferation. J. Clin. Invest. 118, 3609–3618 (2008).
Article CAS Google Scholar
MacAulay, K. et al. Glycogen synthase kinase 3alpha-specific regulation of murine hepatic glycogen metabolism. Cell Metab. 6, 329–337 (2007).
Article CAS Google Scholar
Patel, S. et al. Tissue-specific role of glycogen synthase kinase 3beta in glucose homeostasis and insulin action. Mol. Cell. Biol. 28, 6314–6328 (2008).
Article CAS Google Scholar
Gillespie, J. R. et al. GSK-3beta function in bone regulates skeletal development, whole-body metabolism, and male life span. Endocrinology 154, 3702–3718 (2013).
Article CAS Google Scholar
Zhou, W. et al. The effects of glycogen synthase kinase-3beta in serotonin neurons. PLoS ONE 7, e43262 (2012).
Article ADS CAS Google Scholar
Hurtado, D. E. et al. Selectively silencing GSK-3 isoforms reduces plaques and tangles in mouse models of Alzheimer’s disease. J. Neurosci. 32, 7392–7402 (2012).
Article CAS Google Scholar
Kaidanovich-Beilin, O. & Woodgett, J. R. GSK-3: Functional Insights from Cell Biology and Animal Models. Front. Mol. Neurosci. 4, 40 (2011).
Article CAS Google Scholar
Zhou, S. et al. Genetic and pharmacologic targeting of glycogen synthase kinase 3beta reinforces the Nrf2 antioxidant defense against podocytopathy. J. Am. Soc. Nephrol. 27, 2289–2308 (2016).
George, B. et al. GSK3beta inactivation in podocytes results in decreased phosphorylation of p70S6K accompanied by cytoskeletal rearrangements and inhibited motility. Am. J. Physiol. Ren. Physiol. 300, F1152–F1162 (2011).
Article CAS Google Scholar
Guo, J. et al. GSK-3beta and vitamin D receptor are involved in beta-catenin and snail signaling in high glucose-induced epithelial-mesenchymal transition of mouse podocytes. Cell. Physiol. Biochem. 33, 1087–1096 (2014).
Article CAS Google Scholar
Wang, Z. et al. Pharmacological targeting of GSK3beta confers protection against podocytopathy and proteinuria by desensitizing mitochondrial permeability transition. Br. J. Pharmacol. 172, 895–909 (2015).
Article CAS Google Scholar
Xu, W., Ge, Y., Liu, Z. & Gong, R. Glycogen synthase kinase 3beta orchestrates microtubule remodeling in compensatory glomerular adaptation to podocyte depletion. J. Biol. Chem. 290, 1348–1363 (2015).
Article Google Scholar
Li, C., Ge, Y., Peng, A. & Gong, R. The redox sensitive glycogen synthase kinase 3beta suppresses the self-protective antioxidant response in podocytes upon oxidative glomerular injury. Oncotarget 6, 39493–39506 (2015).
PubMed PubMed Central Google Scholar
Markowitz, G. S. et al. Lithium nephrotoxicity: a progressive combined glomerular and tubulointerstitial nephropathy. J. Am. Soc. Nephrol. 11, 1439–1448 (2000).
CAS PubMed Google Scholar
Hestbech, J., Hansen, H. E., Amdisen, A. & Olsen, S. Chronic renal lesions following long-term treatment with lithium. Kidney Int. 12, 205–213 (1977).
Article CAS Google Scholar
Alsady, M., Baumgarten, R., Deen, P. M. & de Groot, T. Lithium in the kidney: friend and foe? J. Am. Soc. Nephrol. 27, 1587–1595 (2016).
Article CAS Google Scholar
Eremina, V., Wong, M. A., Cui, S., Schwartz, L. & Quaggin, S. E. Glomerular-specific gene excision in vivo. J. Am. Soc. Nephrol. 13, 788–793 (2002).
CAS PubMed Google Scholar
Martinez, A., Castro, A., Dorronsoro, I. & Alonso, M. Glycogen synthase kinase 3 (GSK-3) inhibitors as new promising drugs for diabetes, neurodegeneration, cancer, and inflammation. Med. Res. Rev. 22, 373–384 (2002).
Article CAS Google Scholar
Cohen, P. & Goedert, M. GSK3 inhibitors: development and therapeutic potential. Nat. Rev. Drug. Discov. 3, 479–487 (2004).
Article CAS Google Scholar
Weavers, H. et al. The insect nephrocyte is a podocyte-like cell with a filtration slit diaphragm. Nature 457, 322–326 (2009).
Article ADS CAS Google Scholar
Kim, W. Y. et al. GSK-3 is a master regulator of neural progenitor homeostasis. Nat. Neurosci. 12, 1390–1397 (2009).
Article CAS Google Scholar
Dembowy, J., Adissu, H. A., Liu, J. C., Zacksenhaus, E. & Woodgett, J. R. Effect of glycogen synthase kinase-3 inactivation on mouse mammary gland development and oncogenesis. Oncogene 34, 3514–3526 (2015).
Article CAS Google Scholar
Wei, Y., Yu, L., Bowen, J., Gorovsky, M. A. & Allis, C. D. Phosphorylation of histone H3 is required for proper chromosome condensation and segregation. Cell 97, 99–109 (1999).
Article CAS Google Scholar
Thomasova, D. & Anders, H. J. Cell cycle control in the kidney. Nephrol. Dial. Transplant. 30, 1622–1630 (2015).
Article CAS Google Scholar
Porter, L. A. & Donoghue, D. J. Cyclin B1 and CDK1: nuclear localization and upstream regulators. Prog. Cell Cycle Res. 5, 335–347 (2003).
PubMed Google Scholar
Badouel, C. & McNeill, H. SnapShot: The hippo signaling pathway. Cell 145, 484–484 e481 (2011).
Article CAS Google Scholar
Zanconato, F. et al. Genome-wide association between YAP/TAZ/TEAD and AP-1 at enhancers drives oncogenic growth. Nat. Cell Biol. 17, 1218–1227 (2015).
Article CAS Google Scholar
Varelas, X. The Hippo pathway effectors TAZ and YAP in development, homeostasis and disease. Development 141, 1614–1626 (2014).
Article CAS Google Scholar
Chen, X. et al. C-MYC and BCL-2 mediate YAP-regulated tumorigenesis in OSCC. Oncotarget 9, 668–679 (2018).
PubMed Google Scholar
Yook, J. I. et al. A Wnt-Axin2-GSK3beta cascade regulates Snail1 activity in breast cancer cells. Nat. Cell Biol. 8, 1398–1406 (2006).
Article CAS Google Scholar
Emamian, E. S., Hall, D., Birnbaum, M. J., Karayiorgou, M. & Gogos, J. A. Convergent evidence for impaired AKT1-GSK3beta signaling in schizophrenia. Nat. Genet. 36, 131–137 (2004).
Article CAS Google Scholar
Trivedi, C. M. et al. Hdac2 regulates the cardiac hypertrophic response by modulating Gsk3 beta activity. Nat. Med. 13, 324–331 (2007).
Article CAS Google Scholar
Moeller, M. J. et al. Podocytes populate cellular crescents in a murine model of inflammatory glomerulonephritis. J. Am. Soc. Nephrol. 15, 61–67 (2004).
Article Google Scholar
Shkreli, M. et al. Reversible cell-cycle entry in adult kidney podocytes through regulated control of telomerase and Wnt signaling. Nat. Med. 18, 111–119 (2011).
Article Google Scholar
MacAulay, K. & Woodgett, J. R. Targeting glycogen synthase kinase-3 (GSK-3) in the treatment of Type 2 diabetes. Expert. Opin. Ther. Targets 12, 1265–1274 (2008).
Article CAS Google Scholar
Lal, H., Ahmad, F., Woodgett, J. & Force, T. The GSK-3 family as therapeutic target for myocardial diseases. Circ. Res. 116, 138–149 (2015).
Article CAS Google Scholar
Beurel, E., Grieco, S. F. & Jope, R. S. Glycogen synthase kinase-3 (GSK3): regulation, actions, and diseases. Pharmacol. Ther. 148, 114–131 (2015).
Article CAS Google Scholar
Freland, L. & Beaulieu, J. M. Inhibition of GSK3 by lithium, from single molecules to signaling networks. Front. Mol. Neurosci. 5, 14 (2012).
Article CAS Google Scholar
Dols, A. et al. The prevalence and management of side effects of lithium and anticonvulsants as mood stabilizers in bipolar disorder from a clinical perspective: a review. Int. Clin. Psychopharmacol. 28, 287–296 (2013).
Article Google Scholar
Aiff, H. et al. Effects of 10 to 30 years of lithium treatment on kidney function. J. Psychopharmacol. 29, 608–614 (2015).
Article CAS Google Scholar
Bendz, H., Schon, S., Attman, P. O. & Aurell, M. Renal failure occurs in chronic lithium treatment but is uncommon. Kidney Int. 77, 219–224 (2010).
Article CAS Google Scholar
Hansen, H. E. et al. Chronic interstitial nephropathy in patients on long-term lithium treatment. Q. J. Med. 48, 577–591 (1979).
CAS PubMed Google Scholar
Aurell, M., Svalander, C., Wallin, L. & Alling, C. Renal function and biopsy findings in patients on long-term lithium treatment. Kidney Int. 20, 663–670 (1981).
Article CAS Google Scholar
Zhou, L. & Liu, Y. Wnt/beta-catenin signalling and podocyte dysfunction in proteinuric kidney disease. Nat. Rev. Nephrol. 11, 535–545 (2015).
Article CAS Google Scholar
Zhou, J. et al. Loss of adult cardiac myocyte GSK-3 leads to mitotic catastrophe resulting in fatal dilated cardiomyopathy. Circ. Res. 118, 1208–1222 (2016).
Article CAS Google Scholar
Azzolin, L. et al. YAP/TAZ incorporation in the beta-catenin destruction complex orchestrates the Wnt response. Cell 158, 157–170 (2014).
Article CAS Google Scholar
Eremina, V. et al. VEGF inhibition and renal thrombotic microangiopathy. N. Engl. J. Med. 358, 1129–1136 (2008).
Article CAS Google Scholar
Muzumdar, M. D., Tasic, B., Miyamichi, K., Li, L. & Luo, L. A global double-fluorescent Cre reporter mouse. Genesis 45, 593–605 (2007).
Article CAS Google Scholar
Brault, V. et al. Inactivation of the beta-catenin gene by Wnt1-Cre-mediated deletion results in dramatic brain malformation and failure of craniofacial development. Development 128, 1253–1264 (2001).
CAS PubMed Google Scholar
Ivy, J. R. et al. Klf15 is critical for the development and differentiation of drosophila nephrocytes. PLoS ONE 10, e0134620 (2015).
Article Google Scholar
Saleem, M. A. et al. A conditionally immortalized human podocyte cell line demonstrating nephrin and podocin expression. J. Am. Soc. Nephrol. 13, 630–638 (2002).
CAS PubMed Google Scholar
Wilson, M. C. et al. Comparison of the proteome of adult and cord erythroid cells, and changes in the proteome following reticulocyte maturation. Mol. Cell. Proteom. 15, 1938–1946 (2016).
Article CAS Google Scholar
Vizcaino, J. A. et al. ProteomeXchange provides globally coordinated proteomics data submission and dissemination. Nat. Biotechnol. 32, 223–226 (2014).
Article CAS Google Scholar

Download references

Acknowledgements

This work was funded by Kidney Research UK, the Medical Research Council, which funds RJMC with a Senior Research Fellowship (MR/K010492/1), European Renal Association- European Dialysis Transplantation Association (ERA-EDTA), Canadian Institutes of Health Research and the National Institute of Health. Latterly it was supported by funding from the Innovative Medicines Initiative 2 Joint Undertaking under grant agreement No 115974. This Joint Undertaking receives support from the European Union’s Horizon 2020 research and innovation programme and EFPIA with JDRF.

Author information

These authors contributed equally: J. A. Hurcombe, P. Hartley.

Authors and Affiliations

Bristol Renal, Dorothy Hodgkin Building, University of Bristol, Bristol, BS1 3NY, UK
J. A. Hurcombe, A. C. Lay, L. Ni, S. Singh, E. Marquez, A. F. Barrington, V. Pinto, M. Marchetti, M. A. Saleem, P. W. Mathieson, G. I. Welsh & R. J. M. Coward
Bournemouth University, Bournemouth, BH12 5BB, UK
P. Hartley
Dunedin School of Medicine, University of Otago, Dunedin, 9016, New Zealand
J. J. Bedford, J. P. Leader & R. J. Walker
Department of Pathology, Southern General Hospital, Glasgow, G51 4TF, UK
A. Murphy
Mary Lyon Centre, MRC Harwell, Didcot, Oxford, OX11 0RD, UK
C. L. Scudamore
Translational Health Sciences, University of Bristol, Bristol, BS2 8DZ, UK
L.-F. Wong & J. Uney
The University of Hong Kong, Pokfulam, Hong Kong
P. W. Mathieson
Lunenfeld-Tanenbaum Research Institute, Sinai Health System & University of Toronto, Toronto, M5G 1X5, Canada
S. Patel & J. R. Woodgett
Institute of Metabolic Science, University of Cambridge, Cambridge, CB2 0QQ, UK
S. Patel
Feinberg Cardiovascular Research Institute, Northwestern University Feinberg School of Medicine, Chicago, 60611, IL, USA
S. E. Quaggin

Authors

J. A. Hurcombe
View author publications
You can also search for this author in PubMed Google Scholar
P. Hartley
View author publications
You can also search for this author in PubMed Google Scholar
A. C. Lay
View author publications
You can also search for this author in PubMed Google Scholar
L. Ni
View author publications
You can also search for this author in PubMed Google Scholar
J. J. Bedford
View author publications
You can also search for this author in PubMed Google Scholar
J. P. Leader
View author publications
You can also search for this author in PubMed Google Scholar
S. Singh
View author publications
You can also search for this author in PubMed Google Scholar
A. Murphy
View author publications
You can also search for this author in PubMed Google Scholar
C. L. Scudamore
View author publications
You can also search for this author in PubMed Google Scholar
E. Marquez
View author publications
You can also search for this author in PubMed Google Scholar
A. F. Barrington
View author publications
You can also search for this author in PubMed Google Scholar
V. Pinto
View author publications
You can also search for this author in PubMed Google Scholar
M. Marchetti
View author publications
You can also search for this author in PubMed Google Scholar
L.-F. Wong
View author publications
You can also search for this author in PubMed Google Scholar
J. Uney
View author publications
You can also search for this author in PubMed Google Scholar
M. A. Saleem
View author publications
You can also search for this author in PubMed Google Scholar
P. W. Mathieson
View author publications
You can also search for this author in PubMed Google Scholar
S. Patel
View author publications
You can also search for this author in PubMed Google Scholar
R. J. Walker
View author publications
You can also search for this author in PubMed Google Scholar
J. R. Woodgett
View author publications
You can also search for this author in PubMed Google Scholar
S. E. Quaggin
View author publications
You can also search for this author in PubMed Google Scholar
G. I. Welsh
View author publications
You can also search for this author in PubMed Google Scholar
R. J. M. Coward
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

The study was conceived by R.J.C., S.E.Q. and J.R.W. with expert input from P.W.M., M.A.S., G.I.W., R.J.W., P.H., J.A.H. and S.P. The in vitro and in vivo mouse experiments were performed by J.A.H., P.H., A.C.L., S.S., L.N., E.M., A.F.B., V.P. and M.M. Lithium Rat studies and the study of human kidney biopsies of patients taking lithium was performed by J.J.B. and J.P.L. under the supervision of R.J.W. The immortalised GSK3 knockout cell line was made by L.N., L.F.W. and J.U. The Drosophila studies were conceived and performed by P.H., viability quantification being performed by Ms Charlotte Clarke. Expert histopathology support was given by C.L.S. and A.M. The paper was written by J.A.H. and R.J.C. and then all authors read and intellectually commented on the paper. We thank Professors Jeremy Tavare and Julian Hamilton-Shields for their helpful opinions on this manuscript.

Corresponding author

Correspondence to R. J. M. Coward.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Journal peer review information: Nature Communications thanks Ricardo Biondi and the other anonymous reviewers for their contribution to the peer review of this work. Peer reviewer reports are available.

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Supplementary Movie 1

Supplementary Movie 2

Peer Review File

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Hurcombe, J.A., Hartley, P., Lay, A.C. et al. Podocyte GSK3 is an evolutionarily conserved critical regulator of kidney function. Nat Commun 10, 403 (2019). https://doi.org/10.1038/s41467-018-08235-1

Download citation

Received: 10 May 2018
Accepted: 21 December 2018
Published: 24 January 2019
DOI: https://doi.org/10.1038/s41467-018-08235-1

This article is cited by

AZD5438 a GSK-3a/b and CDK inhibitor is antiapoptotic modulates mitochondrial activity and protects human neurons from mitochondrial toxins
- Gongyu Shi
- Helen Scott
- James B. Uney
Scientific Reports (2023)
Long non-coding RNA DLX6-AS1 is the key mediator of glomerular podocyte injury and albuminuria in diabetic nephropathy by targeting the miR-346/GSK-3β signaling pathway
- Jia Guo
- Wen Zheng
- Zhangsuo Liu
Cell Death & Disease (2023)
Insights into human kidney function from the study of Drosophila
- Sybille Koehler
- Tobias B. Huber
Pediatric Nephrology (2023)
Propofol suppresses adipose-derived stem cell progression via PI3K/AKT-Wnt signaling pathway
- Guoping Yin
- Jia Wang
- Weidong Wu
BMC Anesthesiology (2022)
Drosophila melanogaster: a simple genetic model of kidney structure, function and disease
- Julian A. T. Dow
- Matias Simons
- Michael F. Romero
Nature Reviews Nephrology (2022)

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.