Biology of the proximal tubule in body homeostasis and kidney disease

ABSTRACT

The proximal tubule (PT) is known as the workhorse of the kidney, for both the range and magnitude of the functions that it performs. It is not only responsible for reabsorbing most solutes and proteins filtered by glomeruli, but also for secreting non-filtered substances including drugs and uremic toxins. The PT therefore plays a pivotal role in kidney physiology and body homeostasis. Moreover, it is the major site of damage in acute kidney injury and nephrotoxicity. In this review, we will provide an introduction to the cell biology of the PT and explore how it is adapted to the execution of a myriad of different functions and how these can differ between males and females. We will then discuss how the PT regulates phosphate, glucose and acid–base balance, and the consequences of alterations in PT function for bone and cardiovascular health. Finally, we explore why the PT is vulnerable to ischemic and toxic insults, and how acute injury in the PT can lead to maladaptive repair, chronic damage and kidney fibrosis. In summary, we will demonstrate that knowledge of the basic cell biology of the PT is critical for understanding kidney disease phenotypes and their associated systemic complications, and for developing new therapeutic strategies to prevent these.

BASIC STRUCTURE AND FUNCTION OF THE PROXIMAL TUBULE

The proximal tubule (PT) is known as the workhorse of the kidney, responsible for reabsorbing the vast majority of filtered molecules, but this moniker does not adequately reflect the wide range of metabolic and transport functions of the segment. The PT plays an essential role in homeostasis, orchestrating a myriad of interconnected essential metabolic functions (Fig. 1). The PT begins at the urinary pole of Bowman's capsule and ends at the transition to the thin descending limb of the loop of Henle. Each cell is endowed with a primary cilia that features mechnosensitive calcium channels which can potentially be activated by apical shear stress or radial stretch of the tubules and instigate downstream effects such as gene transcription, protein synthesis, endocytic traffic and ion transport [1].

Figure 1:

PT cells play a key role in homeostasis. Figure created with Biorender.com.

The PT is divided anatomically into convoluted and straight parts but features three distinct, albeit overlapping sections (termed S1–3) which can be distinguished by differences in cell structure. Reabsorption predominates in the early PT region, whereas secretion becomes more prominent later and differences in cell phenotype likely represent adaptation to these different tasks. Cells in S1 display: (i) a highly developed apical brush border and large infoldings of the basolateral membrane to increase surface area for transport; (ii) a sophisticated system to reclaim and degrade filtered proteins; and (iii) a high density of mitochondria that provide energy to drive solute transport [2]. These features become less prominent in S2–3; however, cells in these regions display other features—such as a high density of smooth endoplasmic reticulum and peroxisomes—that are likely important for metabolism of secreted substances. Thus, transport functions along the PT are to some extent spatially segregated.

These differences in function are matched by differences in gene expression profiles: mRNA transcripts responsible for apical solute carriers—such as sodium-glucose co-transporter 2 (Slc5a2; SGLT2) and the sodium–hydrogen exchanger 3 (Slc9a3; NHE3)—are highly expressed in S1, whereas genes involved in excretion—such as the organic anion transporter 1 (Slc22a6; AOT1)—are more predominant in S2. However, PT segments also show substantial overlap in gene expression. For example, transcripts for the tight-junction protein claudin-2 (Cldn2) and megalin (Lrp2) can be demonstrated throughout the length of the PT [3]. Expression patterns are significantly altered in the setting injury or disease. In severe proteinuria, extension of protein uptake from S1 to S2 induces marked expansion of a transitional S1/2 hybrid population at the expense of secretory function [4].

Of interest, mice and rat kidneys demonstrate significant differences in PT gene expression on the basis of sex and PT function appears to be regulated by androgens in males but not estrogen (which plays a role in distal nephron transport). Female rats have greater density of SGLT2 but a lower abundance of other sodium co-transporters on the apical membrane. This translates to a brisker excretion of a saline load and the ability of diabetic female mice to maintain normal blood pressure on a high salt diet. During pregnancy and lactation, the kidney adapts to allow an increase in retention of salt, water and nutrients to support offspring [5]. How these observations relate to human health and disease is the subject of active research that may shed light on a range of issues including the changes in the renal response throughout the life cycle, the female “advantage” in cardiovascular disease, and potentially offer targets or rationale for therapeutic intervention.

Transport processes in the PT

PT reabsorption is driven by Na⁺/K⁺-ATPase pumps in the basolateral membrane that transport sodium out of PT cells, generating concentration gradients that favor influx of sodium across the apical membrane. A range of apical co-transporters capitalize on this gradient to remove other solutes from the filtrate, such as phosphate, amino acids and glucose. Water follows both via the transcellular route (utilizing constitutively active aquaporin channels) and the leaky paracellular pathway (via porous claudins), through which calcium, magnesium and potassium are also reabsorbed.

Reabsorption in the PT is influenced by signaling of G protein–coupled receptors; activation of the angiotensin II receptor and two α-adrenergic receptors, α-1A (Adra1a) and α-2B (Adra2b), augment sodium and bicarbonate reclamation whereas the PTH receptor inhibits sodium and water reabsorption [6]. The PT is also endowed with olfactory receptors which have been identified in all three segments. These receptors may play an important role in response to dietary acid load, blood pressure modulation and glucose homeostasis, and may be a future target for therapeutics [7].

Endocytosis of filtered proteins in the PT

The glomerulus filters small plasma proteins and thus determines their half-life in the circulation. A major specialized function of the PT is to reclaim and degrade these proteins, to retain important nutrients. PT cells achieve this by receptor-mediated, clathrin-dependent endocytosis, employing a receptor complex composed of two large multi-ligand receptors, megalin and cubilin, in concert with the transmembrane protein amnionless, that bind and internalize the filtered ligand–receptor complex and traffic the complex to endosomes. Receptors and ligands are then separated within acidified endosomes to allow the former to recycle to the apical membrane and the latter to proceed to cathepsin rich lysosomes for catabolism (Fig. 2A). Recent spaciotemporal studies of PT protein metabolism in vivo have revealed evidence of sequential coordinated activity between segments, whereby S1 endocytoses and degrades proteins, before releasing peptide fragments that undergo further reabsorption in downstream S2/3 segments [8].

Figure 2:

PT tubule cells are endowed with machinery to accomplish a range of tasks that contribute to homeostasis. (A) The endolysosomal system in PT cells play a key role in reabsorbing albumin and other filtered proteins. These proteins may bind either megalin or cubilin, or both. The transmembrane protein amnionless directs cubilin to the membrane. Binding leads to receptor-mediated endocytosis, catabolism of reclaimed proteins and recycling of the apical receptors. Filtered nephrotoxins such as gentamicin and myoglobulin can enter proximal tubular cells via endocytosis and cause damage. Lysine and other endocytic blockers have been shown to competitively inhibits uptake of nephrotoxins and are potential targets to reduce tubular injury. (B) Functions of the PT are typically depicted separately to focus on a particular pathway, but these occur simultaneously within cells and important interactions between functions are increasingly recognized. Glucose transport, citrate and bicarbonate reclammation, and phosphate reabsorption are shown. Action of SGLT2 promotes reabsorption by Na⁺/H⁺ exchanger (NHE3) through a common scaffolding protein. Excess glucose may affect glycolysis which is also regulated by phosphate uptake. Predicted stochiometry of NaPi-IIa features three Na⁺ ions and one divalent Pi transported for each cycle, whereas the electroneutral action of NaPi-IIc is electroneutral is predicted to transport two Na⁺ ions and one divalent Pi, each cycle. Both transporters are downregulated by retrieval and degradation (green arrows) prompted by PTH or FGF23 binding to their receptors on the basolateral membrane, the latter with Klotho as a co-factor. PTH inhbits NHE3 and induces a natriuresis. CA, carbonic anhydrase; NaDC, Na⁺ dicarboxylate cotransporter. Figure created with Biorender.com.

Genetic defects to this endolysosomal system (ELS) in the PT—such as Dent disease, Lowe syndrome and cystinosis—produce low molecular weight proteinuria and progressive kidney disease [9]. The ELS also represents an important entry route for nephrotoxins, including aminoglycosides, radiocontrast agents, myoglobin (rhabdomyolysis) and immunoglobulin light chains (myeloma), and is thus an important potential target for pharmacological blockade and prevention or amelioration of endocytosis-mediated injury. This strategy has been successfully demonstrated with pretreatment of infused α2-macroglobulin receptor-associated protein that binds megalin to prevent aminoglycoside nephrotoxicity in rats [10] and with dietary supplementation of lysine which competitively inhibits uptake of disulfide-rich proteins in a mouse model of nephropathic cystinosis [11]. In addition, increased lysine intake decreased hypertension and kidney injury in a rat model of salt-sensitive hypertension. In these animals and in a pilot human physiology study, the benefits of lysine on PT cell function appeared to extend beyond the acute effect on receptor-mediated endocytosis and impact cell metabolism and deplete carbon metabolites [12].

The PT also releases extracellular vesicles containing PT cell apical proteins into the lumen, which can be delivered to distal segments. These extracellular vesicles may play an important role both in normal physiology by promoting intra-nephron communication and also may contribute to inflammatory and fibrotic response to injury [13].

Metabolism in the PT

The high transport activity of the PT necessitates a large supply of ATP, provided by densely packed and elongated mitochondria. PT mitochondria can utilize a range of substrates from the filtrate or peritubular capillaries including metabolites such as lactate and pyruvate which can be metabolized via the tricarboxylic acid cycle (TCA), and fatty acids which serve as a particularly important fuel by either β-oxidation or by providing TCA substrate [14, 15]. Amino acids absorbed from the filtrate or generated from protein breakdown are also used for PT metabolism; in particular, glutamine can be reclaimed in the filtrate and returned to the circulation or metabolized in response to systemic acidosis (see below). This flexibility is governed by intracellular and extracellular metabolic sensors, not just by substrate abundance [15].

Conversely, PT cells lack significant anaerobic glycolytic capacity which explains their extreme vulnerability to aerobic insults, including ischemia and mitochondrial toxins.

ROLE OF THE PROXIMAL TUBULE IN BODY HOMEOSTASIS AND THERAPEUTIC TARGETS

Phosphate homeostasis and potential therapeutic interventions

Phosphate reabsorption in the PT is key to maintain whole-body phosphate homeostasis (Fig. 2B). Apical phosphate transport is mediated by NaPi-IIa (also known as NPT2a or SLC34A1), NaPi-IIc (SLC34A3) and PiT2 (SLC20A2), but the basolateral efflux pathway is unknown. Phosphate reabsorption is regulated by parathyroid hormone (PTH) and fibroblast growth factor-23 (FGF23) which both cause internalization and lysosomal degradation of NaPi-IIa and NaPi-IIc. Accordingly, diseases with unregulated secretion of PTH or FGF23 lead to phosphaturia and hypophosphatemia—examples include primary hyperparathyroidism and tumor-induced osteomalacia. Phosphate reabsorption through NaPi-IIa also regulates FGF23 secretion through a kidney–bone feedback loop that depends on glycolysis and secretion of glycerol-3-phosphate [16].

In CKD, both calcium and phosphate homeostasis are disturbed due to impaired 1,25-vitamin D production and phosphate retention which results in a secondary increase in PTH and FGF23. This compensatory response comes at the expense of collateral damage to bone, heart and blood vessels, a constellation referred to as CKD–mineral bone disorder (CKD-MBD).

Recently, several new approaches have targeted the PT to improve CKD-MBD. For example, inhibiting salt-inducible kinases downstream of the PTH receptor stimulated 1,25-vitamin D production in a mouse model of CKD-MBD [17]. Another study showed that a small-molecule selectively and dose-dependently inhibited NaPi-IIa and lowered plasma phosphate through a phosphaturic effect in a rat CKD model [18]. Proximal tubular phosphate handling may also have clinical relevance beyond CKD-MBD. Two recent studies showed that a lower tubular maximum phosphate reabsorption capacity was associated with worse outcomes in patients with heart failure [19] and polycystic kidney disease [20], independent of estimated glomerular filtration rate. Furthermore, significant proteinuria has been shown to impair tubular phosphate excretion by inhibiting FGF23-mediated degradation of NaPi-IIa; this resistance to FGF23 may contribute to cardiovascular disease observed in individuals with advanced CKD and proteinuria [21].

Acid–base balance and inhibition of bicarbonate reclamation

The typical Western diet constitutes an acid load that is buffered by bicarbonate and then excreted by the kidney. The PT contributes through three processes: (i) proton secretion in the tubular lumen via NHE3, (ii) synthesis and regulation of urinary buffers that carry protons including ammonium and titratable acid, and (iii) reabsorption of the base equivalent citrate (Fig. 2B) [22]. PT cells respond to chronic acidosis by increasing apical and basolateral uptake of glutamine and coordinated upregulation of genes that encode the requisite transporters and catabolic enzymes which convert glutamine to α-ketoglutarate to malate, then ultimately to glucose. In the process, two ammonium and two bicarbonate molecules are generated, the former transported to the filtrate and the latter to the blood via the basolateral Na⁺/3HCO₃⁻. Thus, metabolism of glutamine to generate ammonium is linked to gluconeogenesis. Reabsorption of filtered citrate via the sodium dicarboxylate cotransporter (SLC13A2; NaDC1) can also be upregulated and provide substrate for the TCA cycle and three bicarbonate for each citrate metabolized.

CKD impairs the capacity of the PT to excrete the daily acid load and, therefore, deteriorating kidney function leads to progressive acid retention and metabolic acidosis. In early CKD, acid retention does yet not decrease plasma bicarbonate, but may be detectable as lower urinary citrate excretion [23]. Hyperkalemia, common in CKD, also decreases PT ammonia generation and therefore contributes to hyperkalemia-induced or type 4 renal tubular acidosis [24]. The development of metabolic acidosis in CKD is associated with a more rapid decline in kidney function; a possible mechanistic link is the observation that metabolic acidosis can cause PT injury by shifting mitochondrial metabolism towards oxidized NAD required for ammoniagenesis. This metabolic switch decreased fatty acid oxidation leading to lipid accumulation and cellular injury with an uptake defect [25]. Acidosis-induced proximal tubular injury was preventable by the NAD precursor nicotinamide and treatable with bicarbonate. Clinically, sodium bicarbonate can be given to treat metabolic acidosis but the evidence that it attenuates CKD progression is mixed. Instead, strategies to limit dietary acid load are increasingly emphasized [26].

Bicarbonate reclamation can be inhibited with the sulfonamide diuretic, acetazolamide (ACTZ). As with other diuretics, ACTZ is protein bound and delivered to the PT lumen primarily by organic anion transport. ACTZ is unique because it does not inhibit a specific transporter but rather, inhibits the enzyme carbonic anhydrase (CA): CA IV embedded in the apical brush border and the cytosolic form CA II, closely linked to both the apical antiporter NHE3 [27] and the basolateral electrogenic cotransporter Na⁺-HCO₃⁻ (NBCe1), and facilitates their action. The close association between intracellular CAII and transport proteins is known as a metabolon complex that links metabolic enzymes to structural elements. Inhibition of CA limits sodium and bicarbonate reclamation and leads bicarbonaturia and metabolic acidosis. Further, inhibition of CA appears to cause alkalinization of PT cells and inhibit ammoniagenesis thus limiting normal renal adaptation to the resultant metabolic acidosis [28]. ACTZ has only a modest diuretic effect because of the ability of the more distal nephron to reabsorb sodium, but its unique action on PT cells translates into important indications: to ameliorate acute mountain sickness, to diminish metabolic alkalosis from loop and thiazide diuretics, and to facilitate decongestion in congestive heart failure.

Glucose reabsorption and SGLT2 inhibition

Filtered glucose is reabsorbed by the sodium-glucose transporters SGLT2 (SLC5A2) in S1–2 and SGLT1 (SLC5A1) in S3, and returned to the interstitial space via glucose transporters 1 and 2 (GLUT1 and GLUT2). Pharmacological inhibition of SGLT2 by gliflozins (SGLT2 inhibitors, SGLT2i) improves kidney and cardiovascular outcomes in patients with diabetic kidney disease, CKD and heart failure. The reasons for these beneficial effects are still incompletely understood. Early hypotheses focused on classic understanding of glomerular hemodynamics and a decrease in hydraulic pressure prompted by tubuloglomerular feedback and later, the SGLT2i's were dubbed the beta blockers of the kidneys, “resting” the PT from metabolic work. Yet, emerging data have shed new light into the complex metabolic program of the PT cells which is far more interconnected than previously appreciated or taught. One example is the ability of SGLT2i to induce natriuresis, lower blood pressure and normalize extracellular volume. This effect is not just from inhibition of SGLT2 and the ensuing osmotic diuresis but likely mediated by an interaction between SGLT2 and NHE3 through MAP17, a scaffolding protein that forms a complex with SGLT2 [29], or a direct, SGLT2-independent effect on NHE3 of some SGLT2i [30]. Downregulation of other transporters including urate and organic anion transporters has also been observed [31]. Another possibility is that SGLT2i's reverse negative effects of sodium-glucose hyperreabsorption on cell metabolism. SGLT2 is upregulated in diabetes and heart failure and has dramatic effects on PT cells leading to tubular growth, senescence and changes in energy metabolism [32]. Further insight comes from study of the rare disorder Fanconi–Bickel syndrome in which GLUT2 mutations prevent efflux of glucose leading to glycogen accumulation [33]. Inhibition of glucose influx by SGLT2i prevented glycogen accumulation and partially reversed PT dysfunction. Although expression of SGLT2 is restricted to the PT, SGLT2i also have beneficial off-target effects in distal tubular cells [34] and cardiomyocytes [35], and even the gut microbiome, leading to a decrease in formation of uremic toxins [31]. This latter effect may have particular impact on the PT since PT cells play an important role in excretion of these anions and this new insight into the “interactome” between gut microbiome and the PT is an intriguing segway into study of the gut–kidney axis [31].

TUBULAR SECRETION AND NEPHROTOXICITY

The tubular epithelium of the PT plays the primary role in renal excretion of many medications, metabolites, xenobiotics, environmental and uremic toxins, which are poorly filtered often due to albumin binding [36]. These molecules enter PT cells from the blood by an array of multi-specific transporters from the solute carrier transport (SLC) and ATP-binding cassette families. These transporters facilitate secretion first by influx on the basolateral side, movement across the cell and efflux at the apical membrane. The tandem action of these transporters can rapidly remove a range of substrates from the circulation, including molecules that are protein bound [37]. The relative nonspecific nature of these families of transporters facilitates excretion of a remarkable number of substances including antibiotics, antiviral agents, diuretics, bile acids and products of the gut microbiome. Substances may be excreted unchanged or after metabolic processing to increase solubility before secretion. Thus, the PT represents an important site for drug metabolism, and explains why the PT is the most frequent site of nephrotoxicity [38]. In addition, decreased clearance of endogenous secretory substances has been associated with progression of CKD, perhaps as a marker of tubular interstitial disease [39].

Secretory clearance is subject to competitive inhibition. This is most obvious with medications that compete with creatinine for transport via organic cation transporters such as integrase inhibitors or the “pharmaco-enhancer” cobicistat. Yet the secretory pathway has also been implicated in tubular toxicity attributed to excessive tubular drug exposure or mismatch in the rate of influx by basolateral transporters compared with the rate of efflux at the apical membrane as with cisplatin, aristolochic acid and tenofovir disoproxil fumarate (TDF). With TDF, mitochondria are known to be a target of toxicity based on clinical findings of urinary solute wasting, along with histological studies that demonstrated dysmorphic mitochondria. A recent study demonstrated that an active metabolite of TDF inhibits V complex (ATPase synthase) activity and expression, functionally crippling the PT mitochondria [38]. Finally, given the diversity of substrates for the organic solute transport system, this arrangement may also play a role in remote sensing and signaling to promote interorgan communication [40].

PROXIMAL TUBULE CELL INJURY

The biological nature of the PT makes it particularly vulnerable to injury. The high reabsorption rate and associated expression of various transporters leads to uptake of toxins in PT cells at much higher rates than other cell types. Similarly, the abundance of ATPase transporters, which remain active during injury, causes ATP to be depleted in low-oxygen conditions faster than other segments, sensitizing the PT to ischemia.

Whether induced by ischemia, toxins or other insults, PT injury at the cellular level falls into two broad categories: sublethal injury or cell death (Fig. 3). During acute kidney injury (AKI), many PT cells show signs of cell stress or injury, as indicated by the upregulation of injury markers (such as kidney injury molecule 1, KIM-1), loss of brush border and loss of apical/basolateral polarity [41]. This sublethal injury reduces PT function but can go largely undetected. If the injury is severe enough, a portion of these PT cells undergo programmed cell death (apoptosis, necroptosis or ferroptosis) or necrosis. The predominant form of cell death that causes AKI is a matter of great debate, but it appears that the type of cell death is injury and context dependent. Regardless of the form of cell death, the loss of PT cells is strongly associated with kidney functional decline. This is because, unlike sublethal injury, loss of PT cells creates denuded basement membrane which allows for back leak of the glomerular filtrate and prevents concentration of urine [42]. Moreover, recent detailed intravital imaging studies have revealed that release of material from necrotic cell death (necroptosis, ferroptosis or necrosis) can propagate injury along the PT [43, 44].

Figure 3:

Adaptive vs maladaptive repair of PT cells: following injury to the PT, some cells die while most are sublethally injured. Cells that die leave behind denuded basement membrane which triggers surviving cells to dedifferentiate and change shape to cover the denuded area. The cells then divide to repopulate the tubule. Sublethally injured cells can undergo adaptive or maladaptive repair. Generally, sublethally injured cells undergo some level of dedifferentiation, lose the brush boarder, upregulate glycolysis and secrete proreparative cytokines, such as transforming growth factor β (TGFβ). Once recovered, the cells can redifferentiate and return to baseline function. In the case of severe injury, some PT cells are unable to repair themselves or divide to repair the tubule. Instead, they continually attempt repair by secreting cytokines and attempt cell division only to have the cell cycle arrested. This maladaptive repair state continually stresses the cell, leading to increased reliance on glycolysis and cytokine secretion. If the cells remain in this maladaptive repair state, it will cause premature aging and senescence. Senescence greatly amplifies the profibrotic response and activation of myofibroblasts to lay down extracellular matrix, eventually causing kidney fibrosis.

PT response to injury: adaptive vs maladaptive repair

An episode of AKI, particularly severe AKI, can predispose to CKD through AKI-to-CKD transition. This process largely depends on how the PT heals. Loss of neighboring cells due to cell death triggers surviving PT cells to divide and replace the damaged epithelium, resulting in a complete functional recovery in most cases [41]. To achieve this remarkable recovery, the dogma regarding recovery in AKI is that surviving PT cells would undergo dedifferentiation to reversibly revert to an earlier developmental state and then change morphology, enter the cell cycle and divide to cover areas of exposed basement membrane. Once the tubule is repopulated, the PT can redifferentiate to their original state to achieve a full adaptive repair [41].

Even when epithelial cell recovery appears complete, a population of incompletely repaired and dedifferentiated PT cells can remain and contribute to fibrosis and CKD progression [41]. These cells continue to attempt repair and secrete cytokines that are pro-reparative but also profibrotic. They may attempt cell division, only to be halted in cell cycle arrest (Figs 3 and 4). The resulting profibrotic but antiproliferative state has been characterized as maladaptive repair (the term we will use) or failed repair [45]. Studies using single-cell RNA sequencing (RNA-seq) have confirmed the presence of dedifferentiated maladaptive repair cells over a year after AKI in mice and found that maladaptive repair is associated with kidney fibrosis [46]. In particular, RNA-seq studies have identified a plethora of biomarkers or biomarker combinations associated with maladaptive repair [47]. A defining characteristic for adaptive vs maladatpive PT repair appears to be the regulation of a transcription factor called SOX9. Adaptive repair requires activation and subsequent deactivation of SOX9; whereas, persistent activation of SOX9 appears to activate maladaptive repair through WNT activity, activation of adjacent fibroblasts and spaciotemporally contribute to fibrosis [48]. If maladaptively repaired cells do not exit this state, stress caused by mixed signals triggers premature aging and growth arrest in the form of senescence (Fig. 3). Senescence greatly amplifies secretion of pro-inflammatory and profibrotic cytokines, accelerating progression of fibrosis and CKD [49].

Figure 4:

Maladaptive repair: images of maladaptively repaired cells expressing injury and profibrotic proteins weeks after kidney function recovery. Mice were injected with aristolochic acid to induce AKI-to-CKD transition. Forty-two days after injury, kidneys were harvested and stained for the PT injury marker KIM-1, the profibrotic cytokine transforming growth factor β (TGFβ) and the myofibroblast marker α-smooth muscle actin (αSMA). The image shows PT cells with unresolved injury (KIM-1 positive) 42 days after injury have upregulated profibrotic cytokines (TGFβ) leading to the activation of aSMA positive myofibroblasts. (Unpublished data from the Brooks lab.) Scale bar = 10 µm.

Although the PT's ability to enter the cell cycle is essential for organ recovery, cell division on top of injury can lead to unfavorable outcomes (Fig. 5). As above, PT cells are particularly sensitive to drops in ATP levels. Recently, it was shown that the energy demands of cell division can lead to PT ATP depletion and repeated cycles can lead to replicative stress and cell death [50]. This is supported by the finding that pharmacologically arresting PT cells in the G1/S transition can prevent cell death and subsequent injury [51]. In contrast, cell cycle arrest in the G2/M transition has been widely reported to be associated with maladaptive repair [41]. Alternative cell cycle programs may also play a role in repair [52]. It has been reported that a population of PT cells become polyploid cells by replicating their genome without cell division [53]. These polyploid tubular cells can contribute to the functional recovery of the kidney but may ultimately contribute to maladaptive repair. Thus, PT cell division is a double-edged sword that promotes acute function at the cost of long-term consequences, especially in the case of severe injury requiring cells to undergo multiple rounds of division.

Figure 5:

Cell cycle programs and AKI-to-CKD transition: repair following injury to normal or “resting” PT cells may lead to normal progression through the mitotic cell cycle (center gray circle) but alternative programs are increasingly recognized as contributing to kidney injury or repair. Arrest at G1/S may yield cytoprotection whereas arrest at G2/M, promoted by cyclin G1 and CDK5, can contribute to maladaptive repair and senescence. Endoreplication (center) promoted by the YAP-1 can lead to mono- or multi-nuclear polyploid cells. Repeated mitotic cycles (right) can lead to ATP depletion and cell death. Furthermore cell cycle related proteins, such as cyclin G1 and CDK5, can directly regulate adaptive/maladaptive repair.

Although the altered cell cycle response and/or maladaptive repair was originally considered a corruption of the normal repair process, recent evidence suggests that it is controlled by specific cellular pathways including the protein p21 which can lead to cell cycle arrest, p53 which favors transcription of pro-fibrotic genes, and Yes-associated protein (YAP-1) which contributes to the polyploid response [53, 54]. Lineage tracing experiments revealed that maladaptively repaired PT cells have specific transcriptional profiles that maintain the maladaptive state [55]. Furthermore, recent evidence suggests that cell cycle regulators, such as cyclin G1 and cyclin dependent kinase 5 (CDK5), play unique roles outside of the cell cycle and directly activate profibrotic and maladaptive repair pathways (Fig. 5) [45]. These data suggest that maladaptive repair is not a passive response to injury but a specific active pathway leading to CKD, making it an attractive therapeutic target.

Targeting AKI-to-CKD transition

Recently, we identified cell cycle related genes that regulate PT maladaptive repair, cyclin G1 and CDK5. Neither is required for adaptive dedifferentiation and repair; instead, deletion of either specifically blocks the long-term dedifferentiation and promoted adaptive repair of the PT without fibrosis or CKD [45]. Another therapeutic target relates to PT metabolism: healthy PT cells are largely reliant on mitochondrial metabolism and fatty acid substrates, with limited glycolytic capacity. When PT cells undergo dedifferentiation, mitochondrial metabolism is reduced and cells depend on glycolysis [56]. This glycolytic shift is also necessary for cells to become senescent [49]. Importantly, the PT glycolytic shift is dependent, at least in part, on glucose reabsorption from the filtrate as SGLT2 inhibition has been shown to reduce PT glycolysis, maladaptive repair and senescence in diabetic kidney disease [57, 58]. This may further explain some of the protective effects of SGLT2 inhibition in CKD and point to additional strategies to prevent AKI-to-CKD transition.

To date, targeting senescence cells directly has yielded mixed results. Depletion of cells positive for senescence marker p16 or by disrupting pro-senescence transcription factors only demonstrated a marginal effect on fibrosis and CKD [49]. Likewise, deletion of INK4a worsened fibrosis [49]. On the other hand, depleting transcription factor high mobility group box-1 (HMGB1) attenuated the transition to CKD in a model of persistent kidney hypoxia [59]. Furthermore, deletion of interleukin-22, globally or its receptor from the tubule, ameliorates drug induced PT injury and is a potential target for therapy [60]. In addition, viral vectors targeting p16 have been shown to reduce senescence and fibrosis [49]. Anti-senescence therapies using senolytic drugs have shown promise in pre-clinical and clinical studies [61, 62]. Interestingly, one-time depletion of senescent cells is ineffective whereas longer-term senolytic therapy is more effective suggests senescent cells may be continuously formed in the setting of chronic injury. Thus, targeting a terminal cell state may not translate to an effective therapeutic. Therapeutics targeting the underlying causes of maladaptive repair and senescence, such as mitochondrial and metabolic dysfunction, may prove more effective at preventing CKD.

In summary, the PT is a complex and heterogeneous epithelium that sits at the nexus of body homeostasis, and alterations in PT function explain many of the clinical features of kidney diseases. Importantly, modern research techniques are uncovering the molecular secrets of the PT and identifying potential treatment strategies to preserve kidney function in CKD and prevent its systemic complications.

FUNDING

A.M.H. was supported by The Swiss National Centre for Competence in Research (NCCR) Kidney Control of Homeostasis and a Swiss National Science Foundation project grant (310030_184688). C.R.B. is supported by the National Institute of Diabetes and Digestive and Kidney Diseases grant DK121101. E.J.H. is supported by grants from the Dutch Kidney Foundation (CP18.05) and the European Research Council (ERC-COG 101125504).

AUTHORS’ CONTRIBUTIONS

All four authors made substantial contributions to the conception and design of the work. All authors drafted a segment of the manuscript and worked to revise it critically.

DATA AVAILABILITY STATEMENT

No new data were generated or analysed in support of this research.

CONFLICT OF INTEREST STATEMENT

C.R.B. reports having ownership interest in DermYoung LLC, which is unrelated to the current work.