Cellular senescence of renal tubular epithelial cells in diabetic kidney disease: emerging mechanisms and targeted interventions

Abstract

Diabetic kidney disease (DKD) is a leading cause of end-stage kidney disease (ESKD), and while current standard treatment strategies can slow disease progression, they are still unable to reverse progressive structural damage to the kidneys. Emerging evidence indicates that the senescence of renal tubular epithelial cells (RTECs) is not only a central feature of early-stage DKD but also a key pathological mechanism driving renal interstitial fibrosis and progressive kidney function decline. This review systematically elucidates the emerging molecular networks governing RTEC senescence within the DKD microenvironment. Specifically, it thoroughly explores the roles of glucolipotoxicity induced by metabolic dysregulation, persistent reactive oxygen species (ROS) stress, DNA damage, senescence-associated secretory phenotype (SASP)-mediated paracrine toxicity, abnormal epigenetic modifications, and failure of endogenous protective mechanisms such as mitophagy. Furthermore, this review provides a brief overview of current potential therapeutic strategies for delaying or eliminating senescent cells, covering research progress and clinical limitations related to sodium-glucose cotransporter 2 (SGLT2) inhibitors, metformin, mesenchymal stem cell (MSC) therapy, and potential natural products.

1. Introduction

Diabetic kidney disease (DKD), a major microvascular complication of diabetes, remains the leading global cause of chronic kidney disease (CKD) and end-stage kidney disease (ESKD) (Glastras and Pollock, 2024; Abbad et al., 2025). Which is characterized by excessive accumulation and deposition of extracellular matrix, tubular atrophy, dilatation and necrosis (Liu, 2024; De Gregorio et al., 2025; Liu L. et al., 2026) that was associated with diverse underling molecular mechanisms, such as the activation of oxidative stress and inflammation (Tang and Yiu, 2020; Rayego-Mateos et al., 2023; Wang et al., 2023a; Liu H. et al., 2025), renin-angiotensin system (Walther et al., 2021; Zhao et al., 2025) and TGF-β/Smad (Hong et al., 2025; Liu K. et al., 2025) and Wnt/β-catenin (Miao et al., 2023; Long et al., 2024; Tian et al., 2024) signaling pathway as well as the dysbiosis of gut microbiota (Stepanova, 2024; He et al., 2025; Liu et al., 2026c; Sun et al., 2026; Zhang et al., 2026) and the disorder of endogenous metabolites (Wang et al., 2023b; Corradi et al., 2024; Li et al., 2024c; Sharma et al., 2025; Song et al., 2025). Epidemiological data show that approximately 40% of patients with type 2 diabetes (T2DM) and 30% of patients with type 1 diabetes (T1DM) develop DKD, a condition that accounts for more than one-third of all cases of ESKD worldwide (Anders et al., 2018; Disease and Collaboration, 2020). Current standard treatment strategies for DKD include strict glycemic control, antihypertensive therapy, renin-angiotensin-aldosterone system (RAAS) blockers, and sodium-glucose cotransporter 2 (SGLT2) inhibitors and nonsteroidal mineralocorticoid receptor antagonists (Barrera-Chimal et al., 2022; Correa-Rotter et al., 2024). Despite these therapeutic advances, current interventions remain unable to completely halt or reverse the underlying structural kidney damage (Zhao et al., 2018; Fioretto and Pontremoli, 2022; He et al., 2024).

Historically, pathological research and clinical interventions for DKD have primarily focused on abnormal changes in glomerular hemodynamics, thickening of the basement membrane, and damage to and shedding of podocytes (Chen Y. et al., 2024; Yin et al., 2025; Zhu et al., 2026). However, a growing body of clinical evidence and basic research indicates that renal tubular epithelial cell (RTEC) injury is not only a central feature of early-stage DKD but also a key determinant of kidney function decline and long-term patient prognosis (Balakumar, 2024; Liu D. et al., 2024; Liu et al., 2025c). Under diabetic conditions, RTECs undergo profound phenotypic transitions and functional impairment. Among these changes, cellular senescence—a stress response to injury—has been established as a key pathological mechanism driving tubulointerstitial fibrosis and disease progression in DKD (Schunk et al., 2021; Chen et al., 2022) as detailed in Figure 1.

FIGURE 1.

Stimuli induce cellular senescence Figure Caption:DNA damage, paracrine propagation of senescence, metabolic dysregulation, oxidative stress, epigenetic regulation, and compromised endogenous protective mechanisms collectively contribute to the induction of cellular senescence. Senescent cells are consequently marked by sustained cell cycle arrest, genomic instability, and robust SASP secretion, all of which perpetuate tissue dysfunction and promote pathological progression. Created in BioRender. s, J. (2026) https://core.local.biorender.dev/api/short-link/ofhjdpe. Note. Abbreviations: DKD, diabetic kidney disease; IL, interleukin; RTEC, renal tubular epithelial cell; SA-β-Gal, senescence-associated β-galactosidase; SASP, senescence-associated secretory phenotype; TGF-β, transforming growth factor-β; TNF-α, tumor necrosis factor-α. p16, p38, and p53 are classical signaling molecules involved in cell cycle regulation and stress responses, and their upregulated expression is closely associated with the senescent process.

Increasing publications have suggested that cellular senescence is involved in CKD (Liu et al., 2023; Li et al., 2024b; Xiang et al., 2024; Yang et al., 2024a; Wang Y. B. et al., 2025). Cellular senescence promotes the progression of DKD by contributing to pathological processes such as the amplification of inflammation, tissue remodeling, and the exacerbation of fibrosis (Wei et al., 2024; Patel et al., 2026). The accumulation of senescent RTECs impairs the intrinsic regenerative capacity of the kidney (Lu et al., 2024b; Zhong et al., 2025). More importantly, these cells—which remain in a state of cell cycle arrest but exhibit highly active metabolism—secrete senescence-associated secretory phenotype (SASP) factors (Wang W. et al., 2026). These factors induce senescence in neighboring healthy cells through paracrine effects, ultimately driving tubular atrophy and tubulointerstitial fibrosis (Li et al., 2023c). This review aims to summarize recent findings on novel mechanisms of RTEC senescence in DKD and summarize progress regarding emerging drugs in delaying RTEC senescence, with the goal of providing a new theoretical perspective for clinical interventions in DKD.

2. Pathological links between cellular senescence and RTECs in DKD

RTECs constitute most of the renal parenchyma and represent the most metabolically active cell population in the kidney (Qi and Yang, 2018; Cao et al., 2022; Bai et al., 2024; Wei et al., 2025). RTECs rely on ATP supplied by mitochondria to maintain high-intensity reabsorption of ions, glucose, and solutes (Li Y. et al., 2024; Deng et al., 2025; Zhou et al., 2025). Under DKD conditions, the intracellular accumulation of glucose and lipids impairs the kidneys’ normal metabolic pathways, driving metabolic reprogramming (Wang Y. et al., 2024; Li and Susztak, 2025; Cao et al., 2026). A persistently high-glucose microenvironment leads to massive glucose uptake by RTECs (Ahmad et al., 2022). This sustained hyper-reabsorption triggers a severe imbalance in energy metabolism and exacerbates intracellular hypoxia within the RTECs (Hou et al., 2024; Lu et al., 2024a).

Increasing studies indicated that metabolic reprogramming is closely related to kidney injury (Wang W. et al., 2024; Ding et al., 2025). Metabolic reprogramming in RTECs shifts cellular energy utilization from fatty acid oxidation and oxidative phosphorylation to glycolysis (Wang T. et al., 2024; Sun et al., 2025). This inefficient energy metabolism leads to insufficient ATP supply and causes unmetabolized lipids to accumulate intracellularly as lipid droplets, creating a lipotoxic environment (Cai et al., 2020; Zhao L. et al., 2023). Concurrently, lipotoxicity and advanced glycation end products (AGEs) jointly drive mitochondrial dysfunction and endoplasmic reticulum stress, triggering excessive reactive oxygen species (ROS) production and overwhelming the cellular antioxidant defense systems (Liu et al., 2025d; Wu et al., 2026). If the damage to the RTECs goes beyond the threshold for self-repair, the cells enter senescence. The most obvious feature of these senescent cells is permanent cell cycle arrest. This process depends heavily on two main signaling axes: tumor protein p53 (p53)/cyclin-dependent kinase inhibitor 1A (p21) and p16^INK4a. Both of these ultimately feed into the retinoblastoma (Rb) pathway (Gao et al., 2022). Early on during DNA damage, the ataxia-telangiectasia mutated (ATM) kinase gets activated. This kinase then phosphorylates the p53 protein to keep it stable. As a transcription factor, the activated p53 strongly drives the expression of its downstream target, p21 (Li et al., 2025). Once produced, p21 binds to cyclin E-cyclin-dependent kinase 2 (CDK2) complexes and blocks their activity, which effectively stops the cell cycle from moving forward (Eleftheriadis et al., 2022). Later on, maintaining this senescent state over a long period requires the p16^INK4a-CDK4/6-Rb pathway. What happens here is that p16^INK4a specifically inhibits CDK4/6. Because of this inhibition, the Rb protein stays in a hypophosphorylated state. In this state, Rb binds tightly to E2F transcription factors. This binding completely shuts down the transcription of genes that the cell needs to copy its DNA (Wei et al., 2024; Lu et al., 2025). Consequently, damaged or apoptotic RTECs cannot be replaced due to this permanent cell cycle exit, which directly leads to dilation of the renal tubular lumen, epithelial thinning, and tubular atrophy. This severely compromises the integrity of the renal tubular epithelial barrier and its reabsorption function, and constitutes an important histopathological basis for the progressive decline in kidney function (Al-Dabet et al., 2022). At the morphological and structural levels, senescent RTECs undergo significant changes, characterized by cellular hypertrophy and a flattened morphology (Huang et al., 2022). These changes are accompanied by nuclear structural remodeling, including a significant downregulation of lamin B1 expression and impaired nuclear membrane integrity, and chromatin remodeling into senescence-associated heterochromatin foci. Together, these changes indicate a decline in genomic stability (Lukášová et al., 2018).

The deleterious impact of senescent RTECs on kidney structure and function stems not only from their own loss of function but also from their pathological remodeling of the surrounding microenvironment, a process primarily mediated by the release of SASP (Xiong et al., 2024). The SASP is a biologically active secretome comprising various pro-inflammatory cytokines, chemokines, and profibrotic growth factors (Phillips et al., 2025). Through these secreted factors, senescent RTECs recruit monocytes and T lymphocytes to the tubulointerstitium. In the DKD microenvironment, infiltrating monocytes differentiate into macrophages and further polarize toward a profibrotic phenotype, exacerbating the deterioration of the local microenvironment (Chu et al., 2025). Furthermore, SASP mediators induce a partial epithelial-to-mesenchymal transition (pEMT) in adjacent RTECs. Key effector molecules, including the profibrotic cytokine transforming growth factor-β (TGF-β), the pro-inflammatory interleukins interleukin-6 (IL-6), interleukin-1β (IL-1β), and interleukin-8 (IL-8), tumor necrosis factor-α (TNF-α), as well as the matrix-remodeling protease matrix metalloproteinase-9 (MMP-9), have been identified as principal drivers of this process (Yang et al., 2023; Lin et al., 2024). These factors synergistically activate intracellular signaling cascades that exacerbate the diffuse progression of tubulointerstitial fibrosis. In addition, the SASP exerts a paracrine senescence effect; the ROS and inflammatory cytokines it releases can induce damage in neighboring healthy cells and drive them into a senescent state. This cycle leads to the focal accumulation of senescent cells within kidney tissue and ultimately accelerates widespread functional loss of nephrons (Wang C. et al., 2025).

3. Mechanisms underlying RTEC senescence

In recent years, research into the pathogenesis of DKD has made significant progress, revealing multiple molecular networks that drive the senescence of RTECs (Zhang et al., 2025a; Hu et al., 2026b) as detailed in Table 1 and Figure 2. This has provided deeper mechanistic insights into the complexity of senescence in DKD and highlights novel targets for precision interventions (Chen et al., 2018; Zhang et al., 2025b).

TABLE 1.

Mechanisms and treatments that regulate renal tubular epithelial cell senescence in diabetic kidney disease.

Experimental model	Signaling pathway	Effect	Treatment	References
C57BL/6 mice HK-2 cells	PA→RXRα↑/MR → CDKN1A (p21)↑ →Cellular Senescence	Cellular senescence Lipid accumulation Increased SASP factors	MR antagonist	(Feng et al., 2024)
db/db mice NRK-52 E cells	ALM→RXRα↑→ CDX2↑→ CFTR ↑ → β-catenin↓→ Snail↓→ EMT	Tubulointerstitial fibrosis	α-Lipoamide (ALM)	(Zhang et al., 2023)
ApoE −/− mice GEnCs	HG→RXRα↓→ circ8411↓→ miR-23a-5p↑ →ABCA1↓ →Pyroptosis	Pyroptosis	GLP-1RA	(Wu et al., 2024)
Hhip PTC-Tg mice HK-2 cells	HG→ATM/ATR↑→CDK1/CDK2↓→Cellular Senescence	Cellular senescence Glucotoxicity Increased SASP factors	Canagliflozin	(Zhao et al., 2023b)
WT C57BL/6 mice HKC-8 cells	HG→FXII ↑→uPAR→ FAK/Src↑→NOX1↑→ROS↑→ DNAdamage→↑ATM/ATR↑→p53/p21↑ →Cellular Senescence	Cellular senescence Excessive ROS Increased SASP factors	FXII blockade	(Elwakiel et al., 2024)
eNOS −/− db/db mice hGECs hPTECs	DKD→ROS↑→ASK-1↑→p38 MAPK → p53/p21→Cellular Senescence	Cellular senescence Increased SASP factors	Selective ASK-1 inhibitor	(Kajimoto et al., 2024)
Akita mice Agt PT-KO mice HK-2 cells MDCK cells	DKD→iRAS↑→Ang II↑→NOX4↑→ROS↑→p16 →Cellular Senescence	Cellular senescence Increased SASP factors	Targeted Agt knockout Losartan N-acetylcysteine	(Yang et al., 2026)
db/db mice NRK-52 E cells NRK-49 F cells	Senescent cells→Shh↑ →NRK-49 F→Tubulointerstitial fibrosis	Tubulointerstitial fibrosis	Cyclopamine	(Wang et al., 2022)
HK-2	HG+TGFβ1→Serpin E1↑ →Cellular Senescence	Cellular senescence	Serpin E1 inhibitor Dasatinib + Quercetin (D+Q)	(Chen et al., 2023)
C57BL/6 J mice Glis1 KO/Tg mice mTECs	HG→Glis1↓→KAT5↑→Histone lactylation→Cellular Senescence	Cellular senescence Increased SASP factors	Glis1 overexpression, Sodium	(Chen et al., 2025b)
db/db mice STZ-induced mice NAT10 KO mice HK-2 cells	DKD→NAT10↑→ac4C acetylation of PAPP-A mRNA→PAPP-A↑ →p53↑→Cellular Senescence	Cellular senescence	Targeted NAT10 knockout AAV Remodelin	(Hu et al., 2026b)
HK-2 cells	HG→ER-α36↓→EZH2↓→H3K27me3↓→PTEN↑ →PI3K/AKT↓→Cellular Senescence	Cellular senescence	ER-α36 overexpression	(Yu et al., 2025)
C57BL/6 mice YME1L Tg mice HK-2 cells	DKD→YME1L↓→BCL2L13/LC3↓→Mitophagy inhibition→mtROS↑ →Cellular Senescence	Cellular senescence Increased SASP factors	YME1L overexpression	(Luo et al., 2024)
C57BL/6 J mice ERRα ptKO/ptKI mice HK-2 cells	DKD→ERRα↓→NOS2↓→Citrulline↓ →ROS →Cellular Senescence	Cellular senescence Mitochondrial dysfunction Increased SASP factors	Citrulline	(Hu et al., 2026a)
C57BL/6 mice RTECs	HG→Parkin↓→GATA4↑→GAS1↑ →Cellular Senescence	Cellular senescence Increased SASP factors	Parkin overexpression	(Chen et al., 2020)

AAV, adeno-associated virus; ABCA1, ATP-binding cassette subfamily A member 1; ac4C, N4-acetylcytidine; Agt, angiotensinogen (gene); AKT, protein kinase B; ALM, α-lipoamide; Ang II, angiotensin II; ApoE, apolipoprotein E; ASK-1, apoptosis signal-regulating kinase 1; ATM, ataxia telangiectasia mutated; ATR, ATM, and Rad3-related; BCL2L13, Bcl-2-like protein 13; CDK1, cyclin-dependent kinase 1; CDK2, cyclin-dependent kinase 2; CDKN1A, cyclin-dependent kinase inhibitor 1A (p21); CDKN2A, cyclin-dependent kinase inhibitor 2A (p16); CDX2, caudal type homeobox 2; CFTR, cystic fibrosis transmembrane conductance regulator; circ8411, circular RNA 8411; D+Q, dasatinib plus quercetin; DKD, diabetic kidney disease; DSBs, DNA double-strand breaks (referred to as DNA damage in the table); EMT, epithelial-to-mesenchymal transition; eNOS, endothelial nitric oxide synthase; ER-α36, estrogen receptor alpha 36; ERRα, estrogen-related receptor alpha; EZH2, enhancer of zeste homolog 2; FAK, focal adhesion kinase; FXII, coagulation factor XII; GAS1, growth arrest-specific 1; GATA4, GATA binding protein 4; GEnCs, glomerular endothelial cells; Glis1, GLI-similar 1; GLP-1RA, glucagon-like peptide-1 receptor agonist; H3K27me3, trimethylation of histone H3 lysine 27; HG, high glucose; hGECs, human glomerular endothelial cells; hPTECs, human proximal tubular epithelial cells; HKC-8, human kidney cell line-8; HK-2, human kidney 2; iRAS, intrarenal renin-angiotensin system; KAT5, lysine acetyltransferase 5; KO, knockout; LC3, microtubule-associated protein 1 light chain 3; MDCK, Madin-Darby canine kidney; miR-23a-5p, microRNA-23a-5p; MR, mineralocorticoid receptor; mTECs, mouse tubular epithelial cells; mtROS, mitochondrial reactive oxygen species; NAT10, N-acetyltransferase 10; NOS2, nitric oxide synthase 2; NOX1, NADPH oxidase 1; NOX4, NADPH oxidase 4; NRK-49F, normal rat kidney 49F; NRK-52E, normal rat kidney 52E; p38 MAPK, p38 mitogen-activated protein kinase; PA, palmitic acid; PAPP-A, pregnancy-associated plasma protein A; PI3K, phosphoinositide 3-kinase; PT, proximal tubule; PTEN, phosphatase and tensin homolog; ROS, reactive oxygen species; RTECs, renal tubular epithelial cells; RXRα, retinoid X receptor alpha; SASP, senescence-associated secretory phenotype; Serpin E1, serpin family E member 1; Shh, Sonic Hedgehog; Src, proto-oncogene tyrosine-protein kinase Src; STZ, streptozotocin; TGFβ1, transforming growth factor beta 1; Tg, transgenic; uPAR, urokinase-type plasminogen activator receptor; WT, wild type; YME1L, YME1-like 1 ATPase. Arrows (↑ and ↓) indicate upregulation and downregulation, respectively.

FIGURE 2.

Emerging mechanisms of renal tubular epithelial cell senescence in diabetic kidney disease. Figure Caption:This schematic diagram comprehensively illustrates the complex regulatory network underlying cellular senescence induced by the diabetic kidney disease (DKD) microenvironment, which is orchestrated by multiple parallel intracellular signaling pathways. Specifically, these upstream pathways encompass RXRa/MR, SGLT2-HHIP-ATM/ATR, FXII-mediated activation of FAK/Src leading to Rac1/NOX1-driven ROS production, the ASK1/ROS axis, Glis1-KAT5-mediated histone lactylation, NAT10-mediated ac4C acetylation of PAPP-A mRNA, the Parkin-GATA4-GAS1 axis, the ER-α36-EZH2-PTEN axis, the ERRα-NOS2 axis, YME1L-BCL2L13-mediated autophagy-associated mitochondrial disruption, as well as oxidative stress responses involving AGT-NOX4 and TGF-β-PAI-1. Collectively, these diverse signaling events converge to activate the core senescence proteins p53, p21, p16, and p38, ultimately promoting the phenotypic transition of normal cells into senescent cells characterized by a flattened morphology. Subsequently, these senescent cells secrete senescence-associated factors, including Shh, SASP, and PAI-1, which act in a paracrine manner on surrounding normal cells, thereby establishing a vicious positive feedback loop that accelerates DKD pathological progression. Created in BioRender. s, J. (2026) https://core.local.biorender.dev/api/short-link/06emrgm. Note. Abbreviations: AGT, Angiotensinogen; ac4C, N4-acetylcytidine; ASK1, Apoptosis signal-regulating kinase 1; ATM, Ataxia telangiectasia mutated; ATR, ATM and Rad3 related; BCL2L13, BCL2-like 13; DKD, Diabetic Kidney Disease; ER-α36, Estrogen receptor alpha 36; ERRα, Estrogen-related receptor alpha; EZH2, Enhancer of zeste homolog 2; FAK, Focal adhesion kinase; FXII, Coagulation factor XII; GAS1, Growth arrest specific 1; GATA4, GATA binding protein 4; Glis1, GLI-similar 1; HHIP, Hedgehog interacting protein; KAT5, Lysine acetyltransferase 5; LC3, Microtubule-associated protein 1A/1 B-light chain 3; MR, Mineralocorticoid receptor; mRNA, Messenger RNA; NAT10, N-acetyltransferase 10; NOS2, Nitric oxide synthase 2; NOX1, NADPH oxidase 1; NOX4, NADPH oxidase 4; p16, Cyclin-dependent kinase inhibitor 2A; p21, Cyclin-dependent kinase inhibitor 1A; p38, p38 mitogen-activated protein kinase; p53, Tumor protein 53; PAI-1, Plasminogen activator inhibitor-1; PAPP-A, Pregnancy-associated plasma protein-A; Parkin, E3 ubiquitin-protein ligase Parkin; PTEN, Phosphatase and tensin homolog; Rac1, Ras-related C3 botulinum toxin substrate 1; ROS, Reactive oxygen species; RXRa, Retinoid X receptor alpha; SASP, Senescence-associated secretory phenotype; SGLT2, Sodium-glucose cotransporter 2; Shh, Sonic hedgehog; Src, Proto-oncogene tyrosine-protein kinase Src; TGF-β, Transforming growth factor beta; uPAR, Urokinase-type plasminogen activator receptor; YME1L, YME1-like 1.

3.1. Metabolic dysregulation driving senescence

RTEC senescence is closely associated with the metabolic dysregulation in DKD. Feng et al. demonstrated that in the high-fat diet/streptozotocin (HFD/STZ)-induced DKD model, the expression of retinoid X receptor alpha (RXRα) was significantly elevated in RTECs. In this context, RXRα directly binds to the mineralocorticoid receptor (MR) to synergistically activate the transcription of downstream senescence-related genes, such as p21, thereby driving cellular senescence. Furthermore, the RXRα/MR complex can also disrupt renal glucose and lipid metabolism by upregulating phosphoenolpyruvate carboxykinase 1 (PCK1) and pyruvate dehydrogenase kinase 4 (PDK4), further exacerbating the senescence cycle (Feng et al., 2024). However, recent studies indicate that α-lipoamide can delay DKD fibrosis by upregulating RXRα and activating the RXRα/caudal type homeobox 2 (CDX2)/cystic fibrosis transmembrane conductance regulator (CFTR)/β-catenin signaling pathway (Zhang et al., 2023). Additionally, GLP-1 receptor agonists (GLP-1RA) can also effectively reduce cholesterol accumulation and pyroptosis by targeting the RXRα/circ8411/miR-23a-5p/ATP-binding cassette subfamily A member 1 (ABCA1) axis (Wu et al., 2024). These findings suggest that while RXRα, in concert with MR, promotes senescence and fibrosis, it can also exert pleiotropic renoprotective effects under specific pharmacological interventions. Zhao et al. found that overexpression of Hedgehog-interacting proteins (HHIPs) in renal proximal tubular cells (RPTCs) enhances SGLT2 protein expression and glucose uptake. This excessive influx triggers cellular glucotoxicity, activates ATM/ATM- and Rad3-related (ATR)-mediated cell cycle arrest, and induces cellular senescence—a pathogenic cascade that can be effectively abrogated by the SGLT2 inhibitor canagliflozin (Zhao X. P. et al., 2023). Both the RXRα/MR and HHIP pathways are activated by DKD-induced metabolic dysregulation. By reprogramming intracellular metabolic pathways, they exacerbate glucolipotoxicity, ultimately culminating in RTEC senescence and tubulointerstitial fibrosis.

3.2. ROS and DNA damage driving senescence

Persistent oxidative stress and DNA damage are pivotal drivers of cellular senescence in RTECs (Small et al., 2012; Wang et al., 2019; Kim et al., 2026). Coagulation factor XII (FXII) is a serine protease precursor synthesized in the liver and secreted into plasma (Pham et al., 2012; Maas and Renné, 2018); it plays a key role in contact activation within the intrinsic coagulation pathway (Göbel et al., 2018; Kluge et al., 2022; Shamanaev et al., 2023). Previous studies have found that FXII levels are abnormally elevated in DKD, suggesting a possible association with inflammatory mechanisms (Xiao et al., 2023). Mechanistic studies by Elwakiel et al. have demonstrated that under pathological conditions, the proenzyme form of FXII binds to the urokinase-type plasminogen activator receptor (uPAR) on the RTEC surface and, by recruiting integrin β1 as a co-receptor, activates the focal adhesion kinase (FAK)/proto-oncogene tyrosine-protein kinase Src (Src)/Ras-related C3 botulinum toxin substrate 1 (Rac1)/NADPH oxidase 1 (NOX1) pathway, generating large amounts of ROS, which leads to persistent DNA double-strand breaks (DSBs) and subsequently triggers a p21-dependent senescence program (Elwakiel et al., 2024). Acting as a critical intracellular sensor for ROS, the apoptosis signal-regulating kinase 1 (ASK-1) pathway is deeply implicated in renal fibrosis via the upstream regulation of p38 mitogen-activated protein kinase (p38 MAPK) and c-Jun N-terminal kinase (JNK) (Tesch et al., 2016). In the DKD milieu, ROS accumulation leads to the hyperactivation of ASK-1 and its downstream p38 MAPK pathway (Liles et al., 2018). Research by Kajimoto et al. found that sustained activation of the p38 MAPK pathway is closely associated with the activation of the p53/p21 pathway and drives the expression of SASP, jointly promoting the entry of RTECs into a state of senescence (Kajimoto et al., 2024). Yang et al. found that high expression of angiotensinogen (AGT) specific to the proximal tubule significantly upregulates NADPH oxidase 4 (NOX4) levels (Yang et al., 2026). NOX4 is widely localized on the membranes of mitochondria and the endoplasmic reticulum; its pathological activation leads to the production of large amounts of ROS, triggering severe mitochondrial oxidative stress (Jia et al., 2024; Lai et al., 2024; Dai et al., 2026). Experimental animals exhibited disruption of mitochondrial cristae structure accompanied by oxidative DNA damage, ultimately leading to p16^INK4a-dependent senescence and apoptosis in RTECs (Yang et al., 2026). Collectively, pathogenic mediators such as FXII and AGT induce profound ROS generation and subsequent DSBs, forcing RTECs into senescence. Concurrently, stress sensors like ASK1 amplify these oxidative signals to solidify the senescent phenotype. Thus, dismantling this ROS-senescence axis presents a therapeutic avenue for DKD intervention.

3.3. Paracrine signaling driving senescence

Paracrine signaling is a pivotal mechanism driving the propagation of senescence within the tubulointerstitium (Zhao et al., 2019; Hickson et al., 2021; Wang et al., 2024b). Through the secretion of SASP factors, senescent RTECs induce paracrine effects, precipitating premature senescence in adjacent, otherwise healthy cells (Yang et al., 2024b). Research by Wang et al. revealed that senescent RTECs can activate adjacent fibroblasts by secreting the Sonic Hedgehog (Shh) protein, promoting their proliferation and activation, thereby accelerating tubulointerstitial fibrosis (Wang et al., 2022). The TGF-β/Smads pathway is a canonical signaling cascade involved in cellular senescence. TGF-β activates Smad proteins to upregulate p21 and promote nuclear translocation of p16^INK4a, thereby inducing cell cycle arrest in glomerular endothelial cells (Ueda et al., 2021). In RTECs, TGF-β1 induces upregulation of Serpin E1 (PAI-1). PAI-1 secreted by senescent tubular cells serves as a critical effector molecule mediating paracrine senescence, and its specific inhibitor, TM5441, effectively abrogates this paracrine effect (Chen et al., 2023). Furthermore, Smad3 downstream of TGF-β1 also contributes to DKD renal injury by suppressing autophagy in RTECs (Yang et al., 2021). However, Chen et al. found that complete blockade of interleukin-33 (IL-33) signaling paradoxically aggravated kidney injury in DKD, while partial blockade effectively ameliorated renal damage and cellular senescence (Chen L. et al., 2024). This suggests that, within a certain physiological range, some SASP factors may exert protective effects in the kidney. Nevertheless, mounting evidence indicates that their net effect in the DKD microenvironment is predominantly detrimental.

3.4. Epigenetic regulation accelerating senescence

A major clinical challenge in halting the progression of DKD is “metabolic memory,” in which kidney damage persists even after blood glucose levels have been brought under control (Kato and Natarajan, 2019; Chen Z. et al., 2024). Research evidence suggests that metabolic memory is primarily mediated by epigenetic mechanisms, including histone modifications, DNA methylation, and non-coding RNA dysregulation (Kato and Natarajan, 2019).

Chen et al. demonstrated that in patients with DKD and in an STZ-induced DKD mouse model, the transcription factor GLI-similar 1 (Glis1) was significantly downregulated in RTECs. Glis1 can directly bind to the lysine acetyltransferase 5 (KAT5), interfering with its interaction with histones, thereby reducing the level of lactylation and delaying cellular senescence. Overexpression of Glis1 can alleviate high-glucose-induced senescence of RTECs and renal fibrosis (Chen et al., 2025b). Hu et al. revealed that N-acetyltransferase 10 (NAT10) is significantly upregulated in diabetic renal tubules, where it enhances the mRNA stability of pregnancy-associated plasma protein A (PAPP-A) via N4-acetylcytidine (ac4C) acetylation. The subsequent upregulation of PAPP-A protein interacts with p53, activating the p53/p21 signaling cascade to precipitate RTEC senescence (Hu et al., 2026b). In addition, the protective role of estrogen receptor α36 (ER-α36) in DKD, finding that ER-α36 expression is downregulated under high-glucose conditions, leading to suppressed enhancer of zeste homolog 2 (EZH2) expression, which in turn reduces the level of trimethylation of histone H3 lysine 27 (H3K27me3) modification in the phosphatase and tensin homolog (PTEN) promoter region, relieve the transcriptional repression of PTEN, upregulates PTEN, and inhibits the phosphoinositide 3-kinase (PI3K)/protein kinase B (AKT) pathway, ultimately promoting senescence and apoptosis in RTECs (Yu et al., 2025). The above studies indicate that various epigenetic modifications—including histone lactylation, RNA acetylation, and histone methylation—play important regulatory roles in the senescence of RTECs in DKD, providing a new perspective for gaining a deeper understanding of the pathogenesis of DKD and exploring therapeutic strategies targeting epigenetic regulation.

3.5. Impairment of endogenous protective mechanisms promoting senescence

Intrinsic protective mechanisms can regulate the anti-senescence network in multiple ways. In terms of maintaining mitochondrial quality control, the YME1-like 1 ATPase (YME1L), a mitochondrial inner membrane metalloproteinase, binds to its interacting protein BCL2 like 13 (BCL2L13) and promotes its phosphorylation. This post-translational modification enhances the interaction between BCL2L13 and microtubule-associated protein 1 light chain 3 (LC3), thereby driving the mitophagic clearance of damaged mitochondria. Consequently, this pathway mitigates mitochondrial dysfunction and excessive ROS production, effectively alleviating RTEC senescence (Luo et al., 2024). The estrogen-related receptor alpha (ERRα) acts at the transcriptional level to regulate citrulline metabolism by activating nitric oxide synthase 2 (NOS2); this metabolic axis prevents cellular senescence by maintaining mitochondrial function and redox balance (Hu et al., 2026a). Furthermore, the E3 ubiquitin ligase Parkin can target GATA binding protein 4 (GATA4), a key transcription factor involved in senescence, and by promoting its ubiquitination and degradation, effectively suppresses cell cycle arrest, inflammation, and the secretion of fibrogenic factors driven by the GATA4/growth arrest-specific gene 1 (GAS1) signaling axis (Chen et al., 2020).

4. Current anti-senescence strategies and limitations

4.1. Clinical and preclinical therapies

Senomorphics are a class of drugs that do not eliminate senescent cells but rather suppress or modulate their detrimental functions, and include agents such as SGLT2 inhibitors and metformin (Scisciola et al., 2023). Dapagliflozin prevents high-glucose-induced senescence of RTECs by blocking the C1q/TNF-related protein 1 (CTRP1)/SGLT2 axis and the ROS–DNA damage–p53/p21 signaling pathway (Eleftheriadis et al., 2022; Gu et al., 2026). Empagliflozin attenuates senescence in RTECs by upregulating sine oculis homeobox homolog 1 (Six1) and wingless-type MMTV integration site family member 4 (Wnt4), thereby suppressing nuclear factor κB (NF-κB) and its downstream inflammatory cytokines (Chen et al., 2025a). Metformin downregulates the cell cycle-related protein E2F transcription factor 1 (E2F1) in RTECs under hyperglycemic conditions, reduces the phosphorylation levels of ATM and checkpoint kinase 2 (Chk2), and consequently attenuates DNA damage and the senescent phenotype (Liang et al., 2022). Metformin upregulates the RNA-binding protein muscleblind-like protein 1 (MBNL1) and prolongs the half-life of miR-130a-3p; both of these molecules specifically bind to and inhibit the transcriptional activator signal transducer and activator of transcription 3 (STAT3), thereby significantly attenuating the accumulation of senescence-associated protein p21 and substantially reducing the proportion of senescence-associated β-galactosidase (SA-β-gal)-positive cells (Jiang et al., 2020). However, senomorphics still lack senescence-specific molecular targets, which limits their capacity to achieve precise, tissue-targeted senolysis.

Senolytics are a class of drugs that selectively induce apoptosis in senescent cells, with the core strategy being the elimination of pre-existing senescent cells (Robbins et al., 2021). Dasatinib plus Quercetin (D+Q) is currently the most extensively validated senolytic combination in DKD research, with multiple studies demonstrating its renoprotective effects. D+Q treatment not only reduces the abundance of p16^INK4a- and p21-positive cells in the kidney, but also effectively suppresses proinflammatory macrophage infiltration and senescence-associated inflammation, while restoring the expression of the endogenous anti-aging proteins α-Klotho and Sirtuin-1 (SIRT1) (Bian et al., 2026). Alvespimycin is a potent inhibitor of heat shock protein 90 (HSP90). In mouse models of diabetes combined with acute kidney injury, short-term administration of Alvespimycin was shown to partially clear senescent RTECs, but exerted only limited improvement in renal function (Tesch et al., 2024). B-cell lymphoma 2 (Bcl-2) family inhibitors, such as navitoclax (ABT-263), attenuate sustained RTEC senescence induced by Repeated Low-Dose Cisplatin through triggering apoptotic pathways (Li S. et al., 2023). However, their efficacy in DKD remains unclear, and they carry a risk of thrombocytopenia.

Mesenchymal stem cells (MSCs) are a population of stromal cells endowed with self-renewal capacity and multipotent differentiation potential, which can be isolated from various tissues including umbilical cord, adipose tissue, and bone marrow (Han et al., 2025). They exert their therapeutic effects predominantly through the release of paracrine factors. The secretome of these cells comprises both soluble factors, such as growth factors, cytokines, chemokines, and hormones, and insoluble factors encapsulated within extracellular vesicles (EVs) (Nikfarjam et al., 2020; Trigo et al., 2025). Recent studies have revealed that MSCs and their EVs possess considerable therapeutic promise for DKD, as evidenced in both in vitro and in vivo models (Ni et al., 2023; Li Q. et al., 2024; Wang et al., 2024a; Zhang et al., 2024). Liao et al. found that MSC-derived exosomes significantly suppressed the transcription of the cell cycle negative regulators p16^INK4a and cyclin-dependent kinase inhibitor 2D (p19^INK4d), while partially restoring the expression level of lamin B1. Moreover, the mRNA levels of the SASP key inflammatory factors IL-6 and C-C motif chemokine ligand 7 (CCL7) were also markedly downregulated, collectively alleviating the senescence phenotype of RTECs (Liao et al., 2021). Ko et al. demonstrated that SMUP-Cells suppressed the expression of Gremlin 2 (GREM2) in RTECs, thereby blocking Smad3 signaling and concurrently inhibiting both epithelial-to-mesenchymal transition and cellular senescence in RTECs (Ko et al., 2025). Konari et al. demonstrated that intravenously administered bone marrow-derived mesenchymal stem cells (BM-MSCs) could transfer their mitochondria to renal proximal tubular epithelial cells (PTECs) in streptozotocin-induced diabetic animals. This mitochondrial transfer attenuated PTECs injury by upregulating the expression of superoxide dismutase 2 (SOD2), the anti-apoptotic protein Bcl-2, downregulating the pro-apoptotic protein Bcl-2-associated X protein (Bax), and suppressing the production of ROS (Konari et al., 2019). In addition, the two studies primarily focusing on glomerular cells also observed the therapeutic effects of mesenchymal stem cells on RTECs. Human umbilical cord-derived MSCs (hucMSCs) activated AMP-activated protein kinase (AMPK)/mammalian target of rapamycin (mTOR) pathway-dependent autophagy through paracrine action, alleviated high glucose-induced podocyte senescence, and concurrently mitigated pathological damage to RTECs (Li et al., 2023b). Placenta-derived MSCs (P-MSCs) upregulated SIRT1/forkhead box O1 (FOXO1) pathway-mediated autophagy, significantly reduced blood glucose, blood lipids, and the urinary albumin-to-creatinine ratio in DKD rats, and ameliorated both tubular injury and glomerular mesangial matrix deposition (Liu H. et al., 2024).

Proteolysis-targeting chimeras (PROTACs) represent a novel class of small-molecule drugs composed of a ligand that recognizes the protein of interest (POI), a linker, and a ligand that engages an E3 ubiquitin ligase (He et al., 2022). PROTACs recruit specific E3 ligases via the ligand moiety to induce ubiquitination of the target protein, which is subsequently degraded through the ubiquitin-proteasome pathway (Békés et al., 2022). As diverse pathogenic proteins can be eliminated by recruiting distinct E3 ligases using tailored ligands, PROTAC technology holds considerable promise for precision drug discovery and development. For instance, P1705434, a Von Hippel-Lindau (VHL)-recruiting PROTAC, has been shown to hijack the VHL E3 ligase to promote ubiquitin-proteasome degradation of Smad3 while concomitantly attenuating VHL-mediated degradation of HIF-2α, thereby correcting the overactivation of TGF-β/Smad3 signaling and the imbalance between hypoxia-inducible factor-1α (HIF-1α) and hypoxia-inducible factor-2α (HIF-2α). This dual modulation effectively ameliorated tubular cell injury and fibrosis in both cisplatin-induced and folic acid-induced nephropathy models (Ruan et al., 2026). Additionally, the bromodomain-containing protein 4 (BRD4)-targeting PROTAC degrader AT1 has been reported to improve fibrosis in DKD by degrading BRD4 and downregulating the expression of TGF-β and IL-6 (Yang et al., 2025).

Despite their promise, clinical translation of MSCs, MSC-EVs, and PROTACs is hampered by major limitations. MSC therapies face immunogenicity, thrombosis, and tumorigenicity risks (Mei et al., 2024). MSC-EVs lack renal tropism, show only marginal kidney accumulation under pathological conditions, and have poor targeting efficiency, with long-term safety and efficacy unverified. PROTACs suffer from off-target effects, E3-ligase-dependent non-specific degradation, and poor pharmacokinetics. Moreover, clinical evidence for these approaches in DKD remains absent.

4.2. Natural products with anti-senescence potential

Increasing clinical evidence has highlighted that traditional Chinese medicines (TCM) were demonstrated to be as a key therapy for treating CKD patients (Mao et al., 2020; Dong et al., 2024; Acquah-Mills et al., 2025; Chen Y. et al., 2025; Wu et al., 2025; Wang K. E. et al., 2026). Accumulating studies have suggested that CKD, including DKD, could be improved by TCM prescriptions (Ruan et al., 2024; Amarasiri et al., 2025; Yuan et al., 2025; Liu et al., 2026b; Liu et al., 2026d), medicinal plant extracts (Liu P. Y. et al., 2024; Gao et al., 2025; Jiang et al., 2025; Miao et al., 2025) and natural compounds (Miao et al., 2022; Ajibade et al., 2024; Li et al., 2024d; Pan et al., 2024; Wang Y. N. et al., 2024; Ben-Azu et al., 2025; Miao and Vaziri, 2025). Many natural products, including TCM, also show some anti-senescence potential (Zhang et al., 2022; Teh et al., 2024; Wang Y. et al., 2025; Yan et al., 2025a; Yan et al., 2025b; Qiao et al., 2026). Quercetin, a natural flavonoid, can, in addition to its combined use with dasatinib, attenuate RTEC senescence by activating SIRT1, which subsequently initiates PINK1/Parkin-mediated mitophagy (Liu et al., 2020; Wang and Wu, 2025). Resveratrol is a multifunctional phenolic compound found in various natural plants. Studies have shown that resveratrol can reduce serum AGE levels and reverse the p53/p21-and C/EBP homologous protein (CHOP)-mediated cellular senescence pathways activated by AGEs (Lan et al., 2023). Resveratrol can also inhibit renal tubular apoptosis, alleviate oxidative stress, and improve renal fibrosis by activating SIRT1 and downregulating the NOX4/ROS pathway (Uddin et al., 2021). Quercetin and resveratrol can also reduce methylglyoxal (MG) levels in DKD cells and upregulate the expression of the anti-senescence marker protein senescence marker protein 30 (SMP30), thereby reversing high-glucose-induced cellular senescence; when used in combination, the two compounds exhibit a superior synergistic effect in reducing MG levels and increasing SMP30 expression (Abharzanjani and Hemmati, 2021). However, quercetin alone has limited ability to clear senescent cells and must rely on the synergistic effect of dasatinib; the use of dasatinib is associated with rare renal adverse reactions (Adegbite et al., 2023). Moreover, natural polyphenols like resveratrol are plagued by poor aqueous solubility and extensive first-pass metabolism, resulting in suboptimal systemic bioavailability. Overcoming these pharmacokinetic limitations will necessitate structural optimization or the development of advanced drug delivery systems.

5. Conclusion and perspectives

RTEC senescence plays a fundamental role in the pathogenesis of DKD, serving as a core driver of tubulointerstitial fibrosis and progressive renal dysfunction. In recent years, although significant progress has been made in elucidating the mechanisms of senescence—including metabolic dysregulation, ROS and DNA damage, paracrine effects, epigenetic regulation, and the failure of endogenous protective mechanisms—clinical targeted interventions for senescent RTECs still face significant challenges. Although existing treatments such as SGLT2 inhibitors and metformin can indirectly improve metabolism, they lack direct molecular targets, making it difficult to precisely eliminate senescent cells. MSCs and their EVs still require further evaluation regarding their ability to achieve strict renal tropism and long-term safety. Meanwhile, D+Q carries a potential risk of nephrotoxicity, while quercetin and resveratrol alone suffer from low bioavailability.

In the future, in-depth research into the dynamic crosstalk and synergistic pathogenic mechanisms between RTEC senescence and stress events in the renal tubular microenvironment will be key to broadening the scope of senotherapeutic targets. At the same time, there is an urgent need to develop a new generation of anti-senescence drug delivery systems that combine multi-target regulation capabilities with low toxicity and tissue selectivity. By developing targeted carriers capable of specifically recognizing surface markers on senescent RTECs, it is expected that the local accumulation efficiency and bioavailability of anti-senescence agents in the renal tubules will be significantly enhanced, thereby maximizing therapeutic efficacy while minimizing the risk of systemic off-target effects. Finally, establishing standardized preclinical-to-clinical translational pipelines—including the generation of humanized mouse models or organoid systems that recapitulate DKD-associated senescence, the identification of reliable senescence biomarkers for patient stratification, the design of rational combination regimens, and the conduct of large-scale clinical trials—will be critical to steadily advancing anti-senescence strategies for DKD from bench to bedside.

Funding Statement

The author(s) declared that financial support was received for this work and/or its publication. This study was supported by the National Natural Science Foundation of China (Nos. 82274489), Key Research and Development Plan Project of Heilongjiang Province (Nos. 2025ZX05A05).

Author contributions

JSh: Conceptualization, Methodology, Software, Visualization, Writing – original draft, Writing – review and editing. JSu: Methodology, Writing – original draft, Writing – review and editing. JG: Investigation, Writing – original draft, Writing – review and editing. JL: Data curation, Writing – review and editing. YD: Formal Analysis, Writing – review and editing. SH: Validation, Writing – review and editing. CW: Supervision, Writing – review and editing. LW: Funding acquisition, Writing – review and editing. PL: Project administration, Resources, Supervision, Writing – review and editing.

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.