Epithelial TGF-β/β-Catenin Axis in Proximal Tubule Response to Chronic Kidney Disease

Abstract

Chronic kidney disease (CKD) affects 10% of humans and increases the risk of cardiovascular diseases. Regardless of the etiology, tubulointerstitial fibrosis (TIF) is the histopathological feature of CKD that correlates with the loss of renal function, and excessive growth factor (GF) activation is a common mechanism in CKD. Among several GF pathways, the TGF-β/β-catenin axis plays a crucial role in the pathophysiology of CKD. Most compelling studies reported the pivotal role of the proximal tubule (PT), the most metabolic and vulnerable renal segment, in the post-injury response and the pathogenesis of CKD. Though the systemic activation of the TGF-β/β-catenin axis is detrimental in CKD, recent studies have reported the beneficial effects of the TGF-β/β-catenin axis in the PT’s response to chronic injury. This chapter describes the recent findings on the role of TGF-β/β-catenin axis in the PT’s response to CKD. Using genetically modified mice and biochemical and microscopy techniques, TGF-β/β-catenin axis revealed promoting mitochondrial homeostasis, regenerative Th1 immune response, G1 cell arrest, and survival. Future experimental studies should identify key downstream effectors in this axis that can be targeted to mitigate CKD progression.

1. Introduction

Fibrosis or accumulation of extracellular matrix (ECM) proteins is a common pathological feature of progressive chronic diseases. Regardless of the organ, most therapeutic strategies fail to impede the progression of fibrosis and organ loss of function [1, 2, 3]. Tubulointerstitial fibrosis (TIF), the hallmark of chronic kidney disease (CKD), affects more than 10% of humans worldwide and increases the risk of cardiovascular diseases and stroke [4, 5, 6, 7, 8]. CKD treatment represents a considerable financial burden that threatens global health care systems [9, 10, 11]. Though our understanding of mechanisms involved in TIF progression has improved in the last few decades, a curative treatment for CKD patients is still out of reach in practice. Excessive transforming growth factor-beta (TGF-β) and Wnt/β-catenin activation signaling are two common features in CKD progression [12, 13]. Several studies demonstrated the beneficial effect of systemic inhibition of TGF-β signaling in mouse models of TIF; however, genetic inhibition of TGF-β signaling in the renal epithelia and matrix-producing interstitial cells (MPICs) did not mitigate TIF progression [14, 15]. Moreover, Voelker J et al. demonstrated that the addition of TGF-β1 monoclonal antibody to renin-angiotensin system (RAS) inhibitors did not slow the progression of diabetic nephropathy, suggesting that TGF-β signaling is not exclusively the main driving force in CKD progression [3]. Recently, experimental studies in mice demonstrated that deletion of the TGF-β type II receptor (TbRII) in the proximal tubule (PT), the most metabolic renal segment, aggravates TIF in two models of CKD mimicking toxin-induced (aristolochic acid nephrotoxicity) and hypertensive (uni-nephrectomy/angiotensin 2) nephropathies, implying a possible beneficial effect of epithelial TGF-β signaling in chronic tubular injury [16, 17, 18, 19]. Wnt/β-catenin pathway is upregulated in biopsies of human diabetic kidneys and in the PT of proteinuric human patients. Several studies have shown the profibrotic effect of Wnt/β-catenin in fibroblasts/pericytes, but not in renal epithelia [20, 21, 22, 23, 24]. Recently, compelling data demonstrated the beneficial effect of constitutive active β-catenin stabilization in the PT upon chronic tubular injury [18, 25].

The PT is the most sensitive renal segment to injury partly due to its high metabolic rate, oxygen dependency, and higher exposure to toxins. Upon injury, PT cells sense their substratum integrity, undergo cell-autonomous mechanisms, and secrete growth factors/cytokines that activate renal interstitial cells to complete the post-injury recovery [26]. Unfortunately, this tubulointerstitial crosstalk usually becomes maladaptive under chronic injury, leading to TIF. Though the reparative capacity of the PT is incontestable, the mechanisms and molecular signature underlying this process are still elusive. This chapter recapitulates the most recent knowledge on how the TGF-β/β-catenin axis mediates PT adaptive response to chronic injury.

2. Renal tubulointerstitial fibrosis (TIF)

Fibrosis is a pathological condition characterized by the accumulation of extracellular matrix (ECM) proteins including collagens, fibronectin, and vimentin. TIF is the common pathological feature of late-stage renal diseases associated with progressive loss of renal function. The histopathological features of TIF are notable ECM protein accumulation in association with immune cell infiltration, tubular cell loss, myofibroblast proliferation, and rarefaction of micro-vessels. Multiple mechanisms including chronic inflammation, hypoxia, oxidative stress, cell reprogramming, and excessive growth factor activity intertwine in the pathophysiology of TIF [27]. Maladaptive growth factor (GF) activity and chronic inflammation seemingly initiate TIF, while subsequent oxidative stress constitutes a vicious microenvironment that fuels TIF progression. Peritubular capillary rarefaction and associated chronic hypoxia likely result from endothelial cell death under the biochemical/physical effect of matrix expansion and increased ROS production. Cell reprogramming is a (mal-)adaptive mechanism to the micro-environmental changes occurring in chronic diseases and cancers. To survive in this hypoxic and oxidative harmful environment, tubular cells undergo epithelial to mesenchymal transition (EMT) and accordingly, rewire their metabolic profile [28, 29, 30]. Cadherin switch or replacement of E-cadherin by N-cadherin is a key characteristic of EMT that reportedly plays an important role in the pathophysiology of TIF and can be reflective of the loss of renal parenchyma [31]. Considerable progress has been made in understanding the pathophysiology of TIF progression; however, the critical point of irreversibility in CKD progression remains a conundrum. Sustained GF/cytokine activity combined with an aberrant/excessive interstitial cell stimulation arguably constitutes a “self-enhancing loop” that promotes tubular injury and TIF. TGF-β is arguably the most potent profibrotic, though its effects depend upon the microenvironment, the dose, and the type of targeted cells. Mesenchymal activation of the Wnt/β-catenin pathway reportedly promotes fibrosis in experimental models of CKD. However, intact TGF-β/β-catenin activity is reportedly required to mediate PT adaptive response to chronic injury. TGF-β/β-catenin axis is therefore pivotal to studying post-injury PT regenerative capacity and the mechanisms leading to a maladaptive fibrotic response.

3. The kidney and proximal tubule (PT)

The kidney ensures body fluid and electrolytes homeostasis and secretes two hormones, renin and erythropoietin, respectively, regulating blood pressure and erythrocyte regeneration. Renal anatomy comprises two macroscopic parts: The cortex and the medulla. Its structural and functional unit, the nephron, includes the glomerulus and the tubular system. Under normal physiology, the kidney receives 20% of cardiac output (5 L/min), from which only 10% of the oxygen delivery is extracted. This inefficient use of oxygen in comparison to other metabolic organs such as the brain (34%) and the heart (65%) is due to the heterogeneous vascularization of the cortex and medulla. The bulk of renal blood flow (90%) is directed to the cortex (30 to 50 mmHg), resulting in a relatively low oxygen tension in the medulla (10 to 20 mmHg), creating a cortico-medullary metabolic switch [32, 33, 34]. The proximal tubule (PT), the most abundant segment dwelling in the renal cortex, relies on high cortical perfusion to achieve 70% water and sodium reabsorption and maintain its normal physiology [35, 36]. In end-stage renal disease (ESRD), the accumulation of extracellular matrix (ECM) proteins dramatically decreases cortical oxygen diffusion and imposes a metabolic adaptation in the PT to overcome chronic hypoxia and oxidative stress [37]. In addition to the cortico-medullary oxygen supply gradient, while distal renal epithelia express E-cadherin to ensure cell–cell adhesion, the PT relies on N-cadherin. N-cadherin is mostly expressed in mesenchymal cells and in high oxygen-dependent organs including the heart and the brain. Whether this E-to-N-cadherin switch plays a role in PT metabolism and survival in chronic injury is unknown. N-cadherin expression is targeted by TGF-β signaling, and its specific expression in the PT implies its possible role in PT oxidative metabolism. The PT is the most sensitive renal segment to injury partly due to its high metabolic rate and oxygen dependency but also its higher exposure to diverse toxins. PT energy supply relies on their abundant mitochondrial network, representing up to 40% of the PT cell volume, and the utilization of fatty acid beta-oxidation to couple oxygen consumption to ATP production [38, 39, 40, 41]. The PT is divided into three parts: the convoluted tubules (S1 and S2) dwelling in the cortex and the parse recta (S3) located at the cortico-medullary junction and in the outer medulla (Figure 1). S3 PT cells are the most vulnerable to ischemia/toxins-induced acute kidney injury (AKI), and depending on the injury severity, injured cells dedifferentiate and proliferate or undergo necrosis [42, 43]. Mitochondria are crucial for PT cell survival by hosting and orchestrating intrinsic apoptosis. Injured PT is not only a victim but can become pivotal in the initiation of TIF progression through the paracrine effects of their secreted GFs and cytokines. Indeed, in acute kidney injury (AKI), injured PT cells secrete GFs and chemokines that promote epithelial recovery and reparative activation of interstitial cells, leading to post-injury recovery. However, in chronic injury, this tubulointerstitial crosstalk becomes maladaptive, leading to TIF [26, 44, 45, 46, 47]. Thus, promoting adaptive PT cell response to chronic injury represents an important therapeutic strategy to counteract TIF progression. Among several pathways, TGF-β signaling has been reported to induce N-cadherin expression, a terminal differentiation marker of the PT cells [48], and Wnt/β-catenin is crucial in kidney development and β-catenin participates in the cellular structure integrity by linking cadherins to the actin cytoskeleton [49, 50].

Figure 1.

Representative microscopy images of PTs (S3) isolated from tdTomato reporter mice carrying Cre-recombinase expression under the control of gamma-glutamyl transpeptidase (γGT-Cre; WT; tdTomato). From Dr. Nlandu Khodo Stellor.

4. TGF-β and β-catenin axis in tubule-interstitial fibrosis

4.1 TGF-β signaling

TGF-β is the most potent fibrotic factor in CKD. Three mammalian isoforms (TGF-β1, 2, and 3), encoded by distinct genes, have been identified. TGF-β1 is the predominant and best-characterized isoform and will be the focus of this chapter [51, 52, 53]. TGF-βs function through two serine/threonine kinase receptors, the type II (TbRII) and type I (TbRI or ALK5) TGF-β receptors. TβRII binds the active ligand and phosphorylates ALK5, which subsequently activates downstream Smad2/3. Active Smad2/3 oligomerizes with Smad4 and translocates to the nucleus, where their interactions with transcriptional co-factors lead to the expression of TGF-β target genes. TGF-β also activates Smad-independent pathways including MAP kinase [54, 55].

TGF-β signaling plays a pivotal role in normal (organ development, tissue homeostasis, and injury response) and pathological conditions (cancers and organ fibrosis) by mediating cell differentiation, migration, survival, and death. TGF-β signaling is required for kidney morphogenesis through inhibition of HGF-induced branching morphogenesis and promotion of tubule elongation [56, 57]. TGF-β displays multiple and opposing effects depending on the microenvironment and the targeted cell type. TGF-β is a tumor suppressor through induction of growth arrest and cell death; though it promotes malignancy by inducing epithelial to mesenchymal transition (EMT), particularly E-to-N-cadherin switch [58]. TGF-β1 and TbRII knockout mice die around weaning age due to severe inflammation in most organ systems [59, 60]. Genetic or systemic/pharmacological inhibition of TGF-β mitigates TIF in mouse models of CKD; whereas kidney-specific Smad 2 knockout exacerbates TGF-β/Smad3 mediated fibrosis [61, 62, 63]. However, compelling data revealed the beneficial effect of the epithelial TGF-β signaling in experimental models of TIF, though the underlying subcellular mechanisms remained unclear in the last decades. Recent studies demonstrated that deletion of TbRII promotes PT cell survival and reduces TIF progression in two models of CKD by promoting cell survival and β-catenin activation, leading to reduced tubular injury (atrophy, dilatation, and flattening) and fibrosis (increased collagen accumulation) (Figure 2) [18]. Among multiple mechanisms in the pathophysiology of CKD, excessive TGF-β signaling has been reported to mediate mitochondria dysfunction through disruption of the electron transport chain (ETC) and suppression of antioxidant factors. However, TGF-β enhances mitochondrial function (oxygen consumption rate, mitochondrial membrane potential) and glycolysis in primary podocytes through activation of the mTOR pathway [64, 65]. Smad 2, a TGF-β’s downstream effector, promotes mitochondrial fusion by interacting with mitofusin 2, a pro-fusion mitochondrial shaping protein. TGF-β induces autophagy/mitophagy, a protective mechanism whereby damaged organelles (mitochondria) are eliminated from cells and mediates mitochondrial elongation in ARPE-19 cells through the down-regulation of Opa3 [66, 67, 68]. TGF-β has been shown to induce the expression of glucose 6-phosphate dehydrogenase (G6pd), an enzyme that commits glucose breakdown to the pentose phosphate pathway. A recent study reported that TGF-β signaling mediates PT adaptive response to chronic injury by modulating mitochondrial homeostasis and Th1 immune response [69].

Figure 2.

H&E (structure), Sirius red (fibrosis) images showing deleterious (increased injury and fibrosis) effects of abrogating TGF-β signaling in mice lacking Tgfbr2 (TbRII) in the PT (γGT; Tgfbr2^fl/fl) upon aristolochic acid (AA)-induced CKD (left). Graphs showing increased PT injury (KIM-1) and decreased renal function as reflected by increased blood urea nitrogen (BUN) in AA-injured γGT;Tgfbr2^fl/fl mice (from Nlandu Khodo et al. JASN 2017).

4.2 Wnt/β-catenin signaling

β-catenin is a multifunctional protein that functions as a structural anchor linking cell–cell adhesion proteins, notably cadherins, to the actin cytoskeleton at the adherens junctions. Interestingly, β-catenin is the signaling molecule of the Wnt/β-catenin pathway, which is crucial in organ development and in multiple pathogenic states including organ fibrosis and cancers. The binding of Wnts to the LRP/Frizzled receptor complex rescues β-catenin from phosphorylation by the destruction complex (Axin2/APC/GSK3β) and escapes the proteasomal degradation. This allows β-catenin cytosolic stabilization, nuclear translocation (activation), and interaction with its canonical co-transcriptional factors TCF/LEF to activate the transcription of pro-survival genes including c-myc, cyclin D1, and VEGF [70, 71]. β-catenin can also interact with other co-transcription factors, including FoxOs, to mediate non-canonical signaling [72]. Wnt/β-catenin signaling mediates branching morphogenesis during kidney development. Wnt/β-catenin signaling is silenced in adult kidneys except in the inner medulla and is enhanced in both the cortex and medulla upon AKI to promote epithelial self-renewal potentially through reduction of apoptosis and/or altered proliferation [73, 74, 75]. Moreover, β-catenin and K-ras synergy leads to neoplasms, whereas its expression in the metanephric mesenchyme leads to renal dysplasia [76]. Wnt pathway is upregulated in biopsies of human diabetic kidneys and in PT of proteinuric human patients [22]. Several studies have shown the profibrotic effect of Wnt/β-catenin in fibroblasts/pericytes but not in renal epithelia. A study reported a maladaptive effect of Wnt signaling in the PT, but the authors overexpressed a Wnt ligand (Wnt9a), which likely had distant and maladaptive effects on interstitial/stromal cells [21]. Recent studies demonstrated the beneficial role of constitutive active β-catenin in PT under chronic tubular injury by reducing tubular injury and fibrosis (Figure 3) [25].

Figure 3.

H&E (structure) and Sirius red (fibrosis) images showing beneficial effects (decreased injury and fibrosis) in mice carrying constitutive stabilization of β-catenin in the PT (γGT;Ctnnb1^exo3fl/fl) upon AA-induced CKD (from Nlandu Khodo).

4.3 TGF-β/β-catenin axis and proximal tubule response to chronic injury

Sustained GFs/cytokines activity combined with aberrant/excessive interstitial cells stimulation arguably constitutes a “self-enhancing loop” that promotes tubular injury and TIF. Aberrant mesenchymal TGF-β and Wnt/β-catenin signaling arguably mediate TIF progression, though their effects depend on the type of targeted cells and the microenvironment. Experimental studies demonstrated that genetic inhibition of TGF-β signaling in the PT worsens TIF, and stabilization of β-catenin in the PT mitigates TIF under chronic injury, implying the pivotal role of synergistic β-catenin/TGF-β axis in PT adaptive response to chronic injury and survival [18, 25].

The TGF-β/β-catenin axis promotes cell differentiation, which is a prerequisite in post-injury survival prior to proliferation. Several studies illustrated the mechanisms whereby the TGF-β pathway activates Wnt/β-catenin signaling, though additional evidence must be provided to clarify how TGF-β regulates β-catenin activation in PT cells (Figures 4–6). According to the literature, TGF-β may regulate β-catenin activation at different levels of β-catenin signaling. TGF-β has been reported to inhibit DKK1, an inhibitor of the LRP5 receptor, suggesting a possible regulatory effect of TGF-β at the level of the receptor activation. TGF-β has been reported to promote Ser9 phosphorylation of GSK3β, leading to its inactivation and, subsequently, to β-catenin activation; finally, several studies have reported a direct interaction of pSMAD3 and with β-catenin that prevents β-catenin degradation and promotes its nuclear accumulation [77, 78, 79]. β-Catenin is one of the components of the intercellular junctions (aderens junctions), where it directly interacts with the intracellular fragments of cadherins. TGF-β represses E-cadherin expression while increasing N-cadherin expression [80, 81]. The TGF-β’s repressive effect on E-cadherin expression involves the activation of zinc-finger transcription factors including Snail1, Slug, and Twist1. Recent studies in human PT cells (HK-2) reported the role of histone deacetylase (Hdac), particularly Hdac8, as an important regulator of TGF-β repressive effect on E-cadherin expression. Moreover, TGF-β may post-transcriptionally regulate cadherin expression through the induction of proteinases including Sheddases (ADAM 10 and TACE) and other metalloproteases [82]. This repressive effect of TGF-β signaling on E-cadherin may increase β-catenin cytosolic pool and promote its activation. Decreased E-cadherin expression has been reported to increase β-catenin cytosolic pool and promote β-catenin signaling [83]. TGF-β signaling induces N-cadherin expression in PT cells, and the putative expression of N-cadherin in the PT implies an intrinsic role of TGF-β signaling in the maintenance of PT cell terminal differentiation and functional homeostasis.

Figure 4.

Western blot showing increased β-catenin nuclear accumulation (activation) upon treatment of PT cells isolated from Tgfbr2^fl/fl mice (TβRII^flox/flox) with TGF-β. α-tubulin and H3-histone are used as cytosolic and nuclear fraction markers, respectively (left). qPCR showing basal decrease of canonical β-catenin transcriptional activity as reflected by decreased Axin2 in PT cells lacking TbRII (TβRII^−/−) (from Nlandu Khodo et al. JASN 2017).

Figure 5.

Western blot and quantification showing decreased β-catenin nuclear accumulation (activation) in mice lacking Tgfbr2 (TbRII) in the PT (γGT;Tgfbr2^fl/fl) 3 weeks after AA-induced injury. α-tubulin and H3-histone are used as cytosolic and nuclear fraction markers, respectively (from Nlandu Khodo et al. JASN 2017).

Figure 6.

TGF-β receptor signaling is proposed to facilitate successful recovery from cellular damage by interacting with the Wnt-β-catenin pathway. Genetic ablation of the TGF-β receptor in proximal tubule inhibits normal nuclear localization of β-catenin, resulting in increased death, cell cycle arrest, and expression of fibrotic factors (from Basile DP and Mehrotra P, JASN 2017).

4.3.1 TGF-β/β-catenin axis and mitochondria

Mitochondria are cylindrical organelles (0.5 to 1 mm of diameter) delimited by a double membrane, outer (OMM) and inner (IMM) membranes, which form two compartments: inter-membrane space (IMS) and mitochondrial matrix. The OMM is less specialized and quasi-permeable to small molecules compared to the IMM, which is selectively permeable and forms a series of folds (cristae) projecting into the matrix. The matrix is the site of the tricarboxylic acid cycle (TCA), which performs the oxidation of Acetyl-CoA from multiple sources (hexoses, fatty and amino acids) to generate NADH, further converted to adenosine-5′-triphosphate (ATP), through the IMM electron transport chain (ETC). The ETC is composed of four multi-protein complexes that receive and transfer electrons and create a proton (H+) gradient to generate ATP using F0F1 ATPase (complex V) enzymatic activity. The NADH coenzyme Q reductase or NADH dehydrogenase (complex I) accepts and transfers 2 electrons from NADH to coenzyme Q , which transfers them to cytochrome bc1 or coenzyme QH2-cytochrome c reductase (complex III). Through cytochrome c, complex III transfers electrons to cytochrome AA3 (complex IV), which reduces oxygen in two molecules of water.

Mitochondria are abundant and elongated in highly energetic and healthier cells, whereas they fragment under cellular stress [84, 85]. They possess their own genome of 16.6 K bp, coding for 37 genes (13 polypeptides of the ETC enzymes, 22 tRNA, and 2 rRNA); however, most mitochondrial proteins (1100) involved in mDNA replication, structure, and function come from the nuclear genome, implying a considerable role of nuclear genome derived factors in the maintenance of mitochondria homeostasis.

The mechanism whereby TGF-β signaling affects ETC protein expression and function remains unclear for a long time. Mitochondria are composed of proteins coded by their own genome and nuclear genome, and mitochondrial protein sorting and localization are tightly regulated. Nuclear genome-coded protein import into mitochondria is mediated by different translocases and porins. The Tomm complex and Vdac (in the OMM) are involved in protein import into the IMS while Timm23/Pam3 (in the IMM) complex facilitates their transport to the mitochondrial matrix. The sorting and assembly machinery (SAM) and Timm22 are involved in protein insertion into the OMM and IMM, respectively [86, 87]. Mitochondria are the cornerstone of PT cell metabolism and survival, respectively, through fatty acid oxidation (FAO) and modulation of response to apoptosis/injury. They represent an important proportion of PT cell volume while cells of the inner medulla, displaying parsimonious oxygen consumption, have lower mitochondrial density. Given its high dependency on mitochondrial oxygen metabolism, the PT necessitates a powerful antioxidant buffering system to neutralize mitochondrial metabolism-derived ROS and maintain its functional integrity.

The quality control of mitochondria is regulated at different levels: morphology, density, and function. The morphology is mainly regulated by mitochondrial shaping proteins (MSP), which mainly include pro-fission (dynamin-related protein Drp1 and its downstream Fis1), pro-fusion (Mfn1/2 and the dynamin family GTPase Opa1) proteins. Several studies have reported the pivotal role of Pgc-1α in mitochondrial biogenesis, while Pink1/Parkin and BNIP3/Nix have been associated with cell disposal of damaged mitochondria through mitophagy [88, 89]. Mitochondrial dysfunction has been implicated in a broad range of inherited and acquired renal diseases, including tubular defects (Fanconi and Bartter-like syndromes), cystic disease, AKI, glomerular diseases (FSGS), and CKD. Several data have reported disruption of mitochondrial respiration in TIF, including inactivation of complex IV in CKD patients [90]. The uremic toxins impair ETC function and cause cell dedifferentiation. Moreover, the reduction of mitochondrial copy number and increased oxidative stress in skeletal muscle is associated with CKD in humans and mice [91].

TGF-β signaling is necessary to maintain and promote homeostatic OXPHOS, and excessive TGF-β signaling has detrimental effects on oxygen consumption rate (OCR) and ATP production. PT cells lacking TbRII show decreased basal OCR and slightly increased extracellular acidification rate (ECAR) in line with decreased mitochondrial coupling efficiency, suggesting that TGF-β signaling inhibition directly impacts the PT cell’s capacity to efficiently produce ATP via OXPHOS. The absence of TGF-β signaling in mouse PT cells induces a metabolic switch from OXPHOS to increased lactate production consistent with anaerobic glycolysis preference and decreased expression of the polymerase γ. Moreover, the ubiquinone metabolism and complex I are the most affected pathway in PT cells lacking the TbRII, implying that deletion of TbRII induces mitochondrial dysfunction and metabolic switch, partly through impaired expression and function of complex I subunits. Mitochondria homeostasis involves its quality control, which is mainly mediated by their renewal through biogenesis and mitophagy. Pgc1α, the master regulator of mitochondrial biogenesis, has been shown to protect against CKD, though it is regulated in a dose-dependent manner by TGF-β. PT cells lacking the TbRII have decreased Pink1 protein level, an important protein in mitophagy, suggesting that TGF-β signaling regulates mitochondrial homeostasis by promoting mitophagy and expression of complex I subunits [69].

4.3.2 TGF-β/β-catenin axis and inflammation

TGF-β is a multi-faceted cytokine that mediates pro-and anti-inflammatory responses depending on the microenvironment and the targeted cell types. However, global TbRII or TGF-β1 knockout causes lethal inflammatory disorders in mice, notably by regulating lymphocyte homeostasis [59, 60], suggesting its intrinsic anti-inflammatory function. Mitochondrial damage-associated molecular patterns (DAMPS) reportedly lead to the activation of Cgas/Sting/IFNγ axis in renal fibrosis [92, 93]. Though macrophages and dendritic cells initiate inflammation in response to tubular injury, T cells are involved in the whole evolution of injury [94]. Abrogation of TGF-β signaling in the PT leads to increased IFNγ and TNFα+ CD4+ cells in kidneys, whereas CD8+ cells are augmented but not significantly different between genotypes. Moreover, the percentage of the reno-protective Foxp3+ (T reg) CD4+ cells out of CD45+ cells is decreased in the kidneys of mice lacking TbRII in the PT cells. Cgas/Sting proteins, which are responsive to DAMPS and induction of IFNγ, are reportedly increased in the absence of TbRII in the PT, suggesting that TbRII deletion-induced mitochondrial dysfunction worsens Th1 inflammatory response in acute to chronic PT injury models in mice [69]. Consistently, several studies reported the anti-inflammatory effect of Wnt/β-catenin activation by inhibiting TNFα and IFNγ pathways [95, 96, 97]. Taken together, PT specific TGF-β/β-catenin axis activation ameliorates response to chronic injury, partly, by promoting expression of epithelial factors that mediate renal recovery (“PRLs”) which can act autonomously at the level of PT metabolism (mitochondria) and inflammatory cell–cell communication (Figure 7).

Figure 7.

Illustration depicting the mechanisms whereby abrogation of TGF-β signaling worsens chronic kidney disease (CKD)/TIF (Nlandu Khodo S et al. JASN 2017; Kayhan M et al. Nat. Comm 2023).

4.3.3 TGF-β/β-catenin axis and cell cycle

The cell cycle is a ubiquitous and highly coordinated process involved in organ development and tissue homeostasis whose dysfunctions lead to several pathological states including CKD [98]. It is characterized by the phase of nuclear division (mitosis) and the interlude between two mitoses (interphase). The mitosis (M phase) is divided into four stages (prophase, metaphase, anaphase, and telophase) characterized by chromosomal dynamics and segregation into two daughter nuclei. The interphase is composed of the G1, S, and G2 phases. The S phase is characterized by DNA replication while G1 and G2 phases are cell cycle gaps during which the cell is prepared for DNA synthesis and mitosis, respectively.

The cell cycle arrest is an important mechanism during kidney injury response whose dysfunction participates in the pathogenesis of CKD. While G0/G1 cell cycle arrest is considered as a mechanism contributing to the maintenance of organ homeostasis, G2/M cell cycle arrest has been associated with the maladaptive repair process following kidney injury and increased susceptibility to cell death. The G1 phase of the cell cycle is an intermediate stable state during which the two daughter cells that exit the mitosis must sense and integrate the microenvironmental stimuli to enter a resting stage (G0) or to, commit to a new cycle or die.

The cell cycle integrity is intimately regulated by cyclin-dependent kinases (CDKs), a family of serine/threonine protein kinases activated at specific points of the cell cycle. CDK activation necessitates interaction with proteins called cyclins, whose expressions fluctuate throughout the cell cycle, ensuring periodic and specific activation of CDKs during the cell cycle. Several studies identified specific CDK/Cyclin interactions during the progression of the cell cycle, notably in G1 (CDK2, CDK4, CDK6 and Cyclins type D/E), S (CDK2 and cyclin type A), G2 and M (CDK1 and Cyclins type A/B). The G1/S phase transition is characterized by the retinoblastoma protein (Rb) hyperphosphorylation and activation of E2F transcription factors, and these two events constitute the restriction point (R point) or the commitment to a new cell cycle. In addition to CDK/Cyclins, the cell cycle is also controlled by two families of Cyclin-dependent kinase inhibitors (CKIs), the INK4 (p15, p16, p18, and p19) and the Cip/Kip (p21, p27, and p57). CKIs counteract the cell cycle progression by specifically binding to CKDs or CKD/Cyclin complexes. The INK4 family specifically inactivates G1 CDK (CDK4 and CDK6) while the Cip/Kip family inhibits G1 CDK/Cyclins complexes and M CDK/Cyclin complex (CDK1/Cyclin type B).

Activation of TGF-β, known for its role in organ development and tissue homeostasis, has been reported to regulate both G2/M and G1 cell cycle arrest in dose and context-dependent manner. This effect may be inherent to the pleiotropic nature of TGF-β signaling or likely mediated by some of its downstream effectors, such as Forkhead box O 1 (FoxO1) transcription factor, known to induce G1 arrest and mediate cytoprotection under oxidative stress. Depending on the dose, TGF-β-induces cell cycle arrest at G0/G1 or G2/M phases [99, 100, 101]. TGF-β-induced cell cycle arrest at G0/G1 is likely the mechanism by which this growth factor promotes survival in chronic kidney injury. Indeed, though renal epithelia are relatively quiescent in homeostasis, renal injury stimulates cell cycle entry. Most compelling data demonstrate that epithelial cells in either G0 or G1 have improved survival in response to chronic stress (oxidative stress, hypoxia, toxins, …) than cells at later stages in the cell cycle (S, G2/M), which are more vulnerable to apoptosis [102, 103]. TGF-β stimulates G1 arrest in epithelial cells, likely through multiple pathways that target CDK, which, in conjunction with cyclins, promote G1 progression to S [104, 105, 106]. Two families of CDK inhibitors that play an important role in regulating TGF-β-dependent G1 arrest are INK4 and the Cip/Kip family [107, 108, 109, 110]. G1 arrest may also be beneficial by decreasing the number of injured epithelia that progress to G2 arrest. Other authors have shown that injured renal epithelia that are arrested in G2/M acquire a secretory phenotype and produce increased amounts of profibrotic GFs in chronic renal injury [111]. Palbociclib, recently FDA-approved for breast cancer, induces G1 arrest by inhibiting CDK4/6, a mechanism like that of the INK4 family of CDK inhibitors. This pharmacologic inhibitor protected renal function in a murine model of acute kidney injury and CKD [112]. TGF-β signaling in renal epithelia increases transcription of INK4 proteins (e.g., p15INK4B) but not that of Cip/Kip family members p21 or p27. Consistent with an inhibition of CDK4/6, TGF-β signaling increases the number of PT cells in G0/G1 and reduces G2/M arrest. By inducing G1 arrest, TGF-β signaling may augment epithelial cell survival in the hypoxic CKD microenvironment and decrease TIF by limiting the number of epithelial cells that progress to G2/M arrest [113]. FoxO1, a forkhead transcription factor and established regulator of G1 arrest, is regulated by TGF-β signaling in mouse PT cells [114]. TGF-β and FoxO1 promote G1 arrest in a variety of ways, but they both target similar CDK inhibitors of the INK4 and Cip/Kip families. In fact, FoxO has been reported to interact with Smads and promote transcription of the p15INK4B gene, a well-described target of TGF-β [115, 116]. TGF-β may augment FoxO1-mediated G1 arrest through both increased FoxO1 transcription and through interactions with Smads to promote the expression of CDK inhibitors [115, 117, 118]. Treatment of PT cells with BIO, a β-catenin activator, decreases aristolochic acid toxicity-induced G2/M arrest and promotes G0/G1 arrest in the absence of TGF-β signaling in accordance with the synergistic interaction between TGF-β and β-catenin in cell cycle regulation in PT cells [18].

5. Discussion

CKD is a silent-onset disease mainly caused by diabetes and hypertension. Regardless of the initial insult, tubulointerstitial fibrosis (TIF) is the hallmark of late-stage CKD that directly correlates with the loss of renal function. Though our understanding of the pathophysiology of CKD has tremendously improved in the last decades, there is no curative treatment for CKD patients. Alternative replacement therapies include dialysis and renal transplantation. Renal injury can occur at both the glomerular and the tubulointerstitial (lesions in the renal tubules/proximal tubules and the peritubular capillaries) compartments, leading to elevated level of proteins in the urine (proteinuria) and the decline of the glomerular filtration rate (GFR). The treatments are adjusted according to the disease etiology and target the amelioration of specific clinical parameters, notably proteinuria and GFR. Several mechanisms intertwine in the pathophysiology of CKD, including chronic hypoxia, inflammation, oxidative stress, and cell reprogramming, and have been targeted to mitigate CKD progression in animals and patients [119, 120, 121]. The activation of the renin-angiotensin-aldosterone (RAA) system in response to injury-induced systemic and intrarenal hemodynamic changes is a common pathophysiological feature that fuels CKD progression. Thus, RAAS inhibitors and voltage-dependent calcium channel antagonists have been the main therapeutic strategies to slow CKD progression, and specific families of these inhibitors/antagonists are adjusted according to the patient’s history [122]. Renal anemia is an important feature of CKD progression, which is traditionally treated by recombinant erythropoietin (Epo) supplementation. The elucidation of the molecular principles of oxygen-sensing which control Epo production, notably the HIF-PHD-VHL interplay, in the last few years (Nobel Prize 2019) has led to the development of novel oral PHD inhibitors as a potential alternative to the expensive Epo supplementation to treat anemic CKD patients, though a better understanding of the renal oxygen-signaling pathway becomes even more crucial [123]. Lately, sodium-glucose transport protein 2 (SGLT2) inhibitors have emerged as an efficient strategy to treat CKD patients while providing cardiovascular benefits [124, 125]. TGF-β is a pleiotropic factor and arguably the most potent profibrotic factor in CKD progression. TGF-β and Wnt/β-catenin axis is emerging as an important pathway that couples renal epithelial homeostasis and oxygen metabolism. TGF-β signaling interferes with most of the therapeutic targets in CKD treatment including the RAA system and HIF pathway [126, 127, 128, 129]. Upon chronic injury, growth factors including TGF-β and Wnt/β-catenin orchestrate tubulointerstitial interactions, which lead to stromal/interstitial cell activation and fibrosis. Although several studies have demonstrated the harmful effects of systemic TGF-β activation, targeted TGF-β signaling inhibition has not revealed beneficial in clinical studies. Previous study demonstrated that pan-mesenchymal inhibition of TGF-β signaling does not mitigate CKD progression [15]. Surprisingly, a recent study has shown that deletion of Tgfbr2 in Pdgfrβ positive cells prevents the decline of Epo production without altering fibrosis [130]. However, by targeting a broad population of stromal cells, it is difficult to sort out the intrinsic effect of TGF-β signaling on renal erythropoietic cells from the effect of cellular interaction within the subpopulation. Abrogation of TGF-β signaling in the proximal tubule (PT), the most vulnerable renal segment to injury, worsens CKD in mice [18, 69]. TGF-β signaling reportedly induces G1 cell cycle arrest; pharmacological induction of G1/S cell cycle arrest using selective inhibitor of CDK4/6 revealed protection in mouse models of CKD [113]. As TGF-β signaling, Wnt/β-catenin signaling displays multiple and opposing effects in renal fibrosis. While mesenchymal activation of Wnt/β-catenin promotes renal fibrosis, its proximal tubular activation mitigates CKD progression partly by activating cytoprotective/antioxidant pathways [25]. Despite its beneficial effect in PT response to injury, long-term activation of Wnt/β-catenin leads to neoplasms and increase the risk of developing renal tumors in case of activating K-RAS mutation [76].

Given the current findings and the pleiotropic effects (cell type, microenvironment, and dose dependency) of TGF-β and Wnt/β-catenin axis, it will be difficult to systemically target these pathways as therapeutics strategies to treat CKD patients due to potential on and off-target side effects. Nanoparticle drug delivery strategies (DDS) can be used to specifically target TGF-β and Wnt/β-catenin activation in specific cell types, notably in PT cells, and overcome the deleterious effect of systemic activation of TGF-β and Wnt/β-catenin axis in clinical studies. Though the proposed approach appears promising, future studies should aim to identify downstream effectors whereby the TGF-β and Wnt/β-catenin axis mediate PT adaptive response to chronic renal injury.

6. Conclusion

Chronic kidney disease (CKD) is a growing health problem affecting 10% of the global population, and this number will likely increase given the huge burden of diabetes and hypertension, the two leading causes of CKD. Glomerular and tubulointerstitial injuries both lead to CKD through progressive tubulointerstitial fibrosis (TIF), and chronic tubular injury is an important component in the pathophysiology of TIF progression. Understanding how tubular epithelia respond to chronic injury is therefore crucial for the development of effective therapy to halt CKD progression. TGF-β is uncontestable, the master profibrotic factor involved in the pathophysiology of CKD, and Wnt/β-catenin activation in mesenchymal cells promotes renal fibrosis. However, these two pathways mediate many different responses, some of which may be beneficial in renal chronic injury.

This chapter summarizes the most recent findings describing the mechanisms whereby the TGF-β/β-catenin axis mediates proximal tubule (PT) adaptive response to chronic renal injury. TGF-β/β-catenin axis mediates PT adaptive response to chronic injury by promoting mitochondrial homeostasis, adaptive Th1 inflammatory response, and promotes G0/G1 cell cycle arrest upon chronic renal injury. Several mechanisms can be targeted to hamper CKD progression including complex I expression and function, oxidative stress, glycolysis, mitophagy, Th1 immune response, and cell cycle arrest. Targeting inhibition of the TGF-β/β-catenin axis failed to mitigate CKD in clinical practice. Although promoting tubular TGF-β/β-catenin axis signaling revealed benefits upon chronic injury in mice, it is not safe to pharmacologically target direct TGF-β/β-catenin activation as a therapeutic strategy against CKD due to its pleiotropic effects. However, identification of post-injury key regenerative effectors in this axis may help to pinpoint specific anti-fibrotic factors that can be targeted to mitigate CKD progression in humans.

Acknowledgments

I acknowledge Professor Leslie Gewin for directing the project on the role of TGF-β signaling and β-catenin in the proximal tubule response to chronic injury at Vanderbilt University Medical Center and for her tremendous mentoring contribution to my career; and Professor Roland Wenger for hosting my research group in the Institute of Physiology, Faculties of Medicine and Sciences, University of Zurich.

This work was supported by the Swiss National Science Foundation (AMBIZIONE grant PZ00P3_179916 and the NCCR “Kidney.CH” junior grant to S.N.K.).

Conflict of interest

The author has no conflicts of interest.