A Protein in Kidney Filter Cells Influences Kidney Aging and Older Bone Loss

To examine the impact of mPGES‐2 on organismal aging, we compared young (8‐week‐old) and aged (2‐year‐old) wild‐type and Ptges2‐knockout mice. No obvious phenotypic differences were detected between young wild‐type and knockout animals (Figure 1a). In contrast, aged Ptges2‐knockout mice appeared healthier than age‐matched controls, with smoother coats and improved overall appearance, and they exhibited prolonged survival (Figure 1a,b). Ptges2 deletion also improved motor performance and cognitive readouts in aged mice (Figure 1c,d). We next surveyed tissues commonly affected by aging, including kidney, bone, brain, muscle, fat, and liver. Protective effects were observed across several organs, but the most striking improvement was in the kidney, where age‐associated structural deterioration, particularly glomerular injury, was markedly attenuated (Figure 1e; Figure S1a, Appendix S1). Serum ALT, AST, BUN, and creatinine remained within physiological reference ranges and were lower in aged knockout than in age‐matched wild‐type controls (Figure S1b, Appendix S1), indicating that mPGES‐2 deficiency is well tolerated and preserves age‐related biochemical homeostasis. These data suggest that loss of mPGES‐2 improves late‐life health status, with especially prominent benefits in the kidney.

Given the pronounced renal benefit observed in aged Ptges2‐knockout mice, we next examined the role of mPGES‐2 in renal aging in greater detail. In human kidney tissue, mPGES‐2 expression increased progressively with age and showed a significant positive correlation with donor age (Figure 2a,b). A similar age‐dependent increase was observed in mouse kidneys, with prominent staining in both glomerular and tubulointerstitial compartments (Figure 2c,d). Functionally, Ptges2 deletion reduced the urinary albumin‐to‐creatinine ratio, increased glomerular filtration rate (GFR), and limited mesangial matrix expansion in aged mice (Figure S2a–d, Appendix S1). Transmission electron microscopy further showed reduced podocyte loss and improved ultrastructural organization after Ptges2 knockout (Figure 2e,f). Consistent with these findings, immunofluorescence demonstrated preservation of the podocyte markers WT‐1, synaptopodin, and nephrin in aged knockout kidneys (Figure 2g,h). Ptges2 deletion also reduced tubular cell death and renal fibrosis (Figure S2e–j, Appendix S1). At the level of cellular senescence, kidneys from aged knockout mice displayed reduced SA‐β‐gal staining and lower expression of p16, p21, p53, and senescence‐associated secretory phenotype (SASP) factors (Figure 2i–l; Figure S2k, Appendix S1). Together, these data show that mPGES‐2 contributes to structural, functional, and molecular features of renal aging.

Because global Ptges2 deletion markedly improved glomerulosclerosis and podocyte injury, we next sought to define the glomerular cell types in which Ptges2 is expressed. We therefore analyzed single‐cell transcriptomic data (GSE240374) from glomerular cells isolated from young and aged kidneys. Unsupervised clustering identified distinct glomerular cell populations, which were annotated using canonical marker genes, including Nphs1, Nphs2, and Podxl for podocytes, Pdgfrb and Acta2 for mesangial cells, Pecam1 and Kdr for endothelial cells, and Ptprc and Lyz2 for immune cells (Figure 3a,b). Among these populations, Ptges2 showed a podocyte‐enriched expression pattern and the proportion of Ptges2⁺ cells was higher in aged than in young podocytes, indicating an age‐dependent shift (Figure 3c–f). We next examined the spatial localization of mPGES‐2 by dual label immunofluorescence in structurally preserved mouse kidney sections. mPGES‐2 immunoreactivity was predominantly detected in nephrin‐positive podocytes within the glomerulus and was additionally observed in Lotus tetragonolobus lectin (LTL)‐positive proximal tubular cells, albeit at a relatively lower intensity (Figure 3g). Taken together, these findings support a podocyte enriched, but not podocyte exclusive, localization of mPGES‐2 within the kidney.

In line with the single‐cell transcriptomic data showing a podocyte‐enriched expression pattern of Ptges2 in aged glomeruli, and considering the crucial role of podocytes in maintaining the glomerular barrier and filtration function (Nagata 2016), we generated podocyte‐specific Ptges2‐knockout mice to further test whether podocyte mPGES‐2 is functionally required for renal aging. In aged mice, podocyte‐specific Ptges2 deletion significantly lowered the urinary albumin‐to‐creatinine ratio and increased glomerular filtration rate (Figure 4a,b). Glomerulosclerosis, podocyte damage, and renal fibrosis were also attenuated (Figure 4c–g; Figure S3a,b, Appendix S1). Importantly, podocyte‐specific Ptges2 knockout also reduced renal aging, as demonstrated by reduced SA β galactosidase (β‐Gal) positivity and lower expression of aging biomarkers (Figure 4h–l). These results indicate that podocyte mPGES‐2 is a major determinant of aging‐related kidney injury.

Given that mPGES‐2 was also detected in renal tubular cells and that global Ptges2 deletion ameliorated tubular injury, including tubular dilation, brush border loss, and apoptosis, we next asked whether tubular mPGES‐2 makes an independent contribution to renal aging. To answer it, we generated tubule‐specific Ptges2‐knockout mice. Tubule‐specific deletion produced only partial improvement in renal aging and associated kidney injury (Figure S4a–j, Appendix S1), and the magnitude of protection was clearly less than that observed in podocyte‐specific knockout mice, which largely phenocopied the global knockout. These findings suggest that tubular mPGES‐2 contributes to renal aging, but to a lesser extent than podocyte mPGES‐2.

mPGES‐2 was initially identified as a PGE₂ synthase, but it was later recognized as a dual‐function enzyme that preferentially generates MDA rather than PGE₂ under most in vivo conditions (Yamada and Takusagawa 2007). Consistent with this view, several studies, including ours, have shown that Ptges2 manipulation does not measurably alter renal PGE₂ synthesis in a number of pathological settings (Jania et al. 2009; Yamada and Takusagawa 2007; Zhong et al. 2024; Zhong, Quan, et al. 2023). However, whether mPGES‐2 contributes to renal aging through PGE₂‐related signaling remains unclear. To address this, we measured the levels of its two metabolites, PGE₂ and MDA. Unexpectedly, global Ptges2 knockout did not affect MDA content but significantly reduced PGE₂ levels in aged mice (Figure 5a,b). Likewise, podocyte‐specific Ptges2 silencing reduced renal PGE₂ content (Figure 5c). To exclude a compensatory contribution from other PGE₂ synthases, we measured mPGES‐1 and cPGES and found neither was affected by Ptges2 knockout (Figure S5a, Appendix S1).

Next, we examined the expression of four prostaglandin E₂ receptors (EP), the downstream targets of PGE₂, and found that all were upregulated with aging (Figure S5b, Appendix S1). However, Ptger1 (EP1) was the only receptor whose expression was significantly downregulated by Ptges2 knockout in vivo (Figure 5d), and the only receptor selectively induced by D‐galactose in MPC5 podocytes (Figure S5c, Appendix S1), suggesting that EP1 may be the preferentially inducible PGE₂ receptor in podocytes under senescence‐inducing conditions.

Functionally, exogenous PGE₂ reduced podocyte viability and aggravated cell senescence in MPC5 cells (Figure S5d–g, Appendix S1), supporting a pro‐senescent role of PGE₂ in podocytes. To assess whether this response is cell‐type biased within the kidney, we performed parallel experiments in HK‐2 tubular epithelial cells. PGE₂ exposure in HK‐2 tubular epithelial cells induced early changes in p16, p21, and p53 expression and partially reduced cell viability, but did not produce a robust SA‐β‐gal‐positive senescence phenotype within the experimental time frame (Figure S5h–l, Appendix S1). This cell type‐biased susceptibility, consistent with the limited regenerative capacity and intrinsic vulnerability of terminally differentiated podocytes, reinforces why podocyte mPGES‐2 is the rate limiting intrarenal node of the aged kidney.

Based on these observations, we further tested the involvement of the PGE₂/EP1 axis by genetic and pharmacological approaches. Ptges2 knockdown increased cell viability and attenuated podocyte senescence, whereas these protective effects were reversed by exogenous PGE₂ administration (Figure S6a–e, Appendix S1). In contrast, Ptges2 overexpression inhibited cell growth and accelerated cell senescence, both of which were partially blocked by SC51089, an EP1 receptor antagonist (Figure S6f–j, Appendix S1). Collectively, these findings support the notion that mPGES‐2 promotes podocyte senescence and accelerates renal aging, at least in part, through the PGE₂/EP1 signaling pathway.

As shown in Figure S1a, Ptges2 knockout improved bone architecture in aged mice, prompting us to investigate the role of mPGES‐2 in age‐related bone loss. Because osteoporosis frequently coexists with renal aging and chronic kidney injury (Caldiroli et al. 2025), we considered the possibility that mPGES‐2 might influence skeletal aging in parallel with its renal effects. Micro‐computed tomography showed that aging‐induced osteoporosis was markedly ameliorated in Ptges2‐knockout mice, as reflected by improved trabecular microarchitecture and increased bone mineral density (BMD; Figure 5e,f).

Bone homeostasis is maintained by the balance between osteoblast‐mediated bone formation and osteoclast‐mediated bone resorption (Zhu et al. 2020). Consistent with improved skeletal integrity, Ptges2 knockout reduced cartilage degeneration and favored osteoblast formation, as shown by safranin O‐fast green, toluidine blue, tartrate‐resistant acid phosphatase (TRAP), and alkaline phosphatase (ALP) staining (Figure S7a–d, Appendix S1). Accordingly, osteoblast‐associated markers increased, whereas osteoclast‐associated markers decreased in aged Ptges2‐knockout mice (Figure 5g). These data indicate that loss of mPGES‐2 preserves bone homeostasis during aging.

We next asked whether podocyte‐specific Ptges2 deletion exerts a similar skeletal effect. Indeed, mice lacking Ptges2 in podocytes showed improved motor activity and grip strength (Figure 6a). In parallel, osteoporosis and imbalanced mineral metabolism were substantially alleviated in podocyte‐specific Ptges2‐knockout mice (Figure 6b–d), as observed in global Ptges2 knockout mice. Cartilage degeneration was reduced, osteoclast‐related markers declined, and osteoblast‐related markers increased (Figure 6e–g). Thus, podocyte mPGES‐2 contributes not only to renal aging but also to age‐related deterioration of bone.

To determine whether this skeletal phenotype reflected a direct action of PGE₂ within bone, we measured bone PGE₂ content and found no significant change following Ptges2 knockout (Figure S8a, Appendix S1), arguing against a dominant bone autonomous PGE₂ mechanism. Given the established kidney‐bone axis (Massy and Drueke 2020; Quarles 2008), we next examined renal endocrine mediators relevant to bone health. Calcitriol [1,25(OH)₂D₃] is the active form of vitamin D and is normally produced in the kidneys (Selamet et al. 2018; Yoon et al. 2023), where adequate levels are necessary for proper bone mineralization and for balanced bone remodeling through regulation of osteoblast and osteoclast activity (Parfitt et al. 1984; Tang et al. 2019). Calcitriol can also influence α‐klotho, a kidney derived hormone tightly linked to systemic aging (Kurosu et al. 2005; Lau et al. 2012; Xu and Sun 2015).

In the present study, we found that Ptges2 knockout increased calcitriol levels in aged mice (Figure S8b, Appendix S1). Since calcitriol is produced by CYP27B1 in the kidneys, the key activating enzyme involved in its synthesis, we observed the mRNA levels of Cyp27b1 and found that it was upregulated in Ptges2 knockout mice (Figure S8c, Appendix S1). Renal and circulating α‐klotho levels were likewise increased in aged Ptges2‐knockout mice (Figure S8d, Appendix S1). Importantly, podocyte‐specific Ptges2 deletion also enhanced calcitriol production, accompanied by increased CYP27B1 expression (Figure 6h–j), and increased renal α‐klotho abundance (Figure 6k). These data suggest that podocyte mPGES‐2 influences renal endocrine function relevant to bone homeostasis.

To investigate the underlying mechanism, we collected conditioned medium from D galactose treated Ptges2 overexpressing MPC5 podocytes (OE) and applied it to renal tubular cells. Conditioned medium from OE podocytes contained significantly higher PGE₂ than control medium (Figure S9a, Appendix S1) and reduced CYP27B1 and α‐klotho expression in HK2 cells at both mRNA and protein levels (Figure S9b,c, Appendix S1), supporting a paracrine glomerulotubular interaction in which podocyte derived PGE₂ downregulates the tubular endocrine program. Together, these findings support an indirect kidney to bone mechanism in which elevated podocyte mPGES‐2 contributes to age related skeletal decline by impairing the renal endocrine support required for bone homeostasis.

To assess the translational potential of targeting mPGES‐2 in late life, we treated 20‐month‐old mice with SZ0232, a selective mPGES‐2 inhibitor identified by our group (Zhong et al. 2022), at 0.5 mg/kg intraperitoneally every 2 days for 2 months. SZ0232 improved aging‐associated injury in several organs, with the most prominent protection observed in the kidney (Figure 7a; Figure S10, Appendix S1), closely paralleling the phenotype of Ptges2‐knockout mice. SZ0232 reduced glomerulosclerosis, improved podocyte integrity, diminished renal fibrosis, and suppressed tubular apoptosis (Figure 7b–d; Figure S11a–d, Appendix S1). It also attenuated renal senescence, as shown by reduced SA‐β‐gal staining and lower expression of aging‐related proteins (Figure 7e–g). Consistent with the genetic models, SZ0232 lowered renal PGE₂ levels (Figure 7h). In bone, SZ0232 improved trabecular microarchitecture, increased bone mineral density and bone mineral content, and rebalanced bone remodeling (Figure 7i–l; Figure S11e,f, Appendix S1), changes that were accompanied by increased renal calcitriol content (Figure 7m). These results support mPGES‐2 inhibition as a potentially actionable strategy for the treatment of renal aging and its skeletal complication.

The sources of key reagents and materials are detailed in Table(Appendix). S1 S1

All animals were maintained in a specific pathogen‐free animal facility at an ambient temperature of 22°C and 50% humidity under a 12/12 h light/dark cycle, with free access to water and food. All experimental protocols were performed in strict accordance with the recommendations of the Guide for the Care and Use of Laboratory Animals of the China Association for Laboratory Animal Science and were approved by the Animal Ethics Committee of Xuzhou Medical University (202301T006). C57BL/6J Ptges2^+/− and Ptges2^flox/+ mice were generated by GemPharmatech Co. Ltd. (Nanjing, China) using CRISPR/Cas9 technology. Nphs‐2 Cre mice were gifted by Professor Zhanjun Jia of Nanjing Medical University. Ksp‐Cre transgenic mice were gifted by Professor Baoxue Yang of Peking University. Ptges2^−/− mice were obtained by self‐crossing C57BL/6J Ptges2^+/− mice. Podocyte‐specific Ptges2 knockout mice (Ptges2^f/f;Nphs2‐Cre) were generated by crossbreeding Nphs‐2 Cre mice with Ptges2^flox/+ mice. Tubule‐specific Ptges2 knockout mice (Ptges2^f/f;Ksp‐Cre) were generated by crossbreeding Ksp‐Cre transgenic mice with Ptges2^flox/+ mice. Littermate Ptges2^f/f mice served as controls. Ptges2 knockout was confirmed by tail genotyping and western blotting as previously reported (Zhong et al. 2024).

Here, 20‐month‐old mice fed a standard laboratory chow diet were administered either SZ0232 or saline control. SZ0232 was injected intraperitoneally at a dose of 0.5 mg/kg once every 2 days for 2 months. SZ0232 was dissolved in a saline solution at 0.1 mg/mL. At the end of the experimental period, blood and urine samples were collected, and the mice were sacrificed to harvest the kidney, liver, brain, heart, muscle, bone, and fat tissues for further analysis.

HEK293T cells were purchased from ATCC and cultured in DMEM (C11995500BT, Gibco) supplemented with 10% heat‐inactivated fetal bovine serum (FBS) (A5256701, Thermo Fisher Scientific) and 1X penicillin–streptomycin solution (C0222, Beyotime Biotechnology). The conditionally immortalized murine podocyte cell line (MPC5) was purchased from BNCC (Beina Biotechnology Co. Ltd.). MPC5 cells were cultured in RPMI 1640 medium (C11875500BT, Thermo Fisher Scientific) supplemented with 10% FBS and 100 U/mL penicillin plus 0.1 mg/mL streptomycin. All cells were grown at 37°C in a humid atmosphere containing 5% CO₂.

D‐galactose was used to induce cell senescence. Cells were exposed to D‐galactose (G0750‐25G, Sigma‐Aldrich↗) at a dose of 15 mg/mL for 60 h. 16,16‐dimethyl Prostaglandin E₂ (14,750, Cayman Chemical) was added to cells for 36 h. SC51089 (HY‐108563, MedChemExpress) was added to the medium for 60 h at a final concentration of 10 μM. After the experiments, cells were collected for further analysis. In addition, MPC5 podocytes stably overexpressing Ptges2 (OE) or carrying the empty vector (NC) were cultured overnight and switched to serum free medium containing D‐galactose (15 mg/mL) for 24 h. Conditioned medium was clarified by centrifugation, and PGE₂ concentration was measured by ELISA (Cayman 514010). HK2 cells at 70%–80% confluence were then incubated with NC‐CM or OE‐CM (1:1 with fresh DMEM) for 48 h. Cyp27b1 and α‐klotho mRNA were quantified by qRT‐PCR; CYP27B1 and α‐klotho protein by western blotting.

The mouse Ptges2 gene was cloned into the expression vector pCDH‐CMV2‐puro (Jiangsu Laisen Co. Ltd.). The shRNA sequence targeting mouse Ptges2 (5′‐CGTGAGAAGGACTGAGATCAA‐3′) was designed and inserted into the lentiviral expression vector pLKO.1‐puro (Shanghai GenePharma Co. Ltd.). To produce lentiviral particles, HEK293T cells were transiently transfected with the recombinant plasmids along with the packaging plasmids psPAX2 and pMD2.G for 48 h. The viral supernatant was then collected, filtered, and used to infect MPC5 cells for 48 h. Infected cells were subsequently selected with puromycin dihydrochloride (ab141453; Abcam) for an additional 72 h.

Chips of human kidney tissues were provided by ZK Biotechnology Co. Ltd. The organ donors are men and women aged from 16 to 82 year‐old with informed consent. In all cases, informed written consent was collected by the company where the organs were recruited. For this human experiment, the study was approved by the Human Ethics Committee of ZK Biotechnology Co. Ltd. The detailed information of donors was listed in Tablein Appendix. S2 S1

Motor coordination and capability were evaluated using a balance‐beam task. Prior to beginning the experiment, mice were acclimated to the testing area for at least 60 min, starting from a wider beam to the narrow testing beam, in which mice were placed at the start and allowed to traverse the beam to reach the endpoint, repeating four consecutive trials. After training, the mice were allowed to walk across a narrow beam, and the event of falling from the beam was recorded. The test was performed 3 times and the overall score for each mouse was the average of the three trial scores. Forelimb grip strength (N‐F/kg) is determined using a Grip Strength Meter (Columbus Instruments, Columbus, OH). The mice were placed on the grid, keeping the torso horizontal, and allowing only their forepaws to attach to the grid before measurements were taken. The mouse was gently pulled back by its tail, ensuring that the mouse gripped the top portion of the grid and the torso remained horizontal, and the maximal grip strength value of the mouse was recorded and displayed on the screen. This procedure was performed 10 times and the mean value was calculated to obtain forelimb grip strength measurements.

Multiple tissues isolated from the above mouse models were fixed in 4% paraformaldehyde overnight. Tissues were embedded in paraffin and cross‐sectioned for histological analysis. As for Hematoxylin–eosin (H&E) and PAS staining, the paraffin‐embedded sections were de‐paraffinized, dehydrated with xylene and ethanol, and successively stained with H&E or PAS (Crowley et al. 2009; Fu et al. 2020). Masson's trichrome staining was performed using a kit (G1340, Solarbio), according to the manufacturer's instructions. Sirius red staining was conducted by Servicebio Technology Co. Ltd. As for immunohistochemical analysis, the prepared slides were stained with the indicated primary antibodies. As for transmission electron microscopy (TEM) analysis, fresh kidney cortex samples were collected and immediately fixed in 2.5% glutaraldehyde. The samples were then performed for TEM by Servicebio Technology Co. Ltd. Images were systematically captured using an Olympus BX43F microscope.

Alkaline phosphatase (ALP) and tartrate‐resistant acid phosphatase (TRAP) are marker enzymes for osteoblasts and osteoclasts, respectively, and these enzymes are markers that highlight the presence of osteoblasts and osteoclasts (Sun et al. 2020; Zeng et al. 2016). Briefly, bone tissue was fixed with 4% paraformaldehyde, decalcified, and embedded in paraffin. Paraffin sections were prepared and deparaffinized for staining with ALP (D001‐2‐2, Nanjing Jiancheng Bioengineering Institute) and TRAP according to the instructions of the indicated commercial kits (G1050‐50T, Servicebio Technology Co. Ltd.). Images were captured using an Olympus BX43F microscope.

Safranin O‐Fast Green, and toluidine blue staining were used to detect cartilage, mucin, and mast cell granules in the bone (Hu and Olsen 2016). For Safranin O‐Fast Green staining (G1371, Solarbio), deparaffinized sections were stained with Weigert's iron hematoxylin working solution for 5 min and washed gently in distilled water for 10 min. Sections were stained with a fast green solution for 5 min and rinsed quickly with a 1% acetic acid solution for 10–15 s. Images were captured using an Olympus BX43F microscope. For toluidine blue staining (G2543, Solarbio), deparaffinized sections were stained with a toluidine blue working solution for 2–3 min and rinsed gently with three changes of distilled water for 3 min. Next, the slides were dehydrated with 95% absolute alcohol, followed by two changes of xylene and coverslips.

Micro‐computed tomography (μCT) was used to evaluate bone microstructure and mineralization. The femurs were collected, and soft tissues were removed. The femurs were then placed with gauze in a sample holder and scanned using CT (VNC‐102) at the Shanghai Aiyou Biotechnology Center. The images were captured using Cruiser and analyzed using Avatar software. Trabecular morphometry was characterized by measuring bone mineral density (BMD), bone volume fraction (BVF), bone surface area to bone volume ratio (BS/BV), bone volume fraction (BV/TV; %), trabecular thickness (Tb.Th), trabecular number (Tb.N), and trabecular spacing/separation (Tb.Sp). All μCT parameters were derived using the manufacturer's protocols.

Frozen tissue sections were prepared for immunofluorescence staining. Frozen sections were thawed and washed with PBS. After that, the sections were incubated with 0.3% Triton X‐100 for 1 h and blocked with 10% goat serum for 1 h at room temperature. The sections were then incubated with the indicated primary antibody for 12 h at 4°C. The primary antibodies used were as follows: Wilms tumor protein (ab89901, Abcam), nephrin (AF3159, R&D systems), synaptopodin (sc‐515842, Santa Cruz), lotus tetragonolobus lectin (L32480, Thermo Fisher Scientific), and mPGES‐2 (10881‐1‐AP, proteintech). Detection was performed using the following secondary antibodies: Donkey Anti‐Mouse IgG (H+L) Dylight488 (E032211, Earthox), Goat Anti‐Rabbit IgG (H+L) Dylight488 (E032220, Earthox), Goat Anti‐Mouse IgG (H+L) Dylight594 (E032410, Earthox), and Goat Anti‐Rabbit IgG (H+L) Dylight594 (E032420, Earthox). Nuclei were counterstained with DAPI (C1002, Beyotime Biotechnology). Confocal images were captured using a Leica STELLARIS 5 microscope with Leica Application Suite X (LAS X) imaging software. For image quantification, podocyte markers were calculated specifically in glomerular areas. Images were quantified using the Image J software. The detailed information of donors was listed in Table S2 in Appendix S1.

Frozen tissue sections were prepared for β‐Gal staining according to the manufacturer's instructions (G1580, Solarbio). Briefly, the prepared slides were fixed and incubated with the SA‐β‐Gal staining solution in a humidified chamber and protected from light at 37°C for 4 h to overnight. When the color reaction was complete, rinsing in 1X PBS and mounted with coverslips. Images were quantified using the ImageJ software. β‐Gal positive areas were quantified using the Image J software.

Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay relies on the enzyme terminal deoxynucleotide transferase (TdT), which attaches deoxynucleotides to the 3′‐hydroxyl terminus of DNA breaks. This assay is widely used to detect apoptosis and cell death. For the TUNEL assay, frozen tissues were fixed with 4% paraformaldehyde for 4–24 h at 4°C. After fixation, the tissues were cut into thin slices of 10 μm or less. The prepared slices were then permeabilized, which allowed reagents, such as the TdT enzyme, to penetrate the cell nucleus according to a commercial kit (MA0223‐1, MeilunBio). To quantify apoptotic cells, the number of TUNEL‐positive (green) cells was counted and divided by the total number of cells in each field using the ImageJ software.

Blood samples were collected from the orbital vein and centrifuged to collect serum. Serum blood urea nitrogen (BUN) and creatinine levels were measured using commercial kits (C013‐2‐1, Nanjing Jiancheng Bioengineering Institute; C011‐2‐1, Nanjing Jiancheng Bioengineering Institute) according to the manufacturer's instructions.

Serum and kidney samples were prepared, and the content of α‐Klotho was detected according to the procedure of a commercial kit (SEH757Mu, Cloud‐Clone Corp.).

GFR was measured transcutaneously using a MediBeacon transdermal GFR monitor (MediBeacon Inc., Mannheim, Germany). Briefly, the flank region of each mouse was depilated one day prior to measurement. On the day of the experiment, mice were anesthetized with isoflurane and the sensor was affixed to the depilated skin. A 5‐min baseline signal was acquired before a single bolus of FITC‐sinistrin (150 mg/kg body weight) was administered via tail vein injection. Mice were then individually housed and transcutaneous fluorescence was recorded continuously for 90 min. GFR values (μL/min per 100 g body weight) were calculated from the fluorescence decay curve using MPD Studio software (MediBeacon Inc.).

Urine samples were collected using metabolic cages in which mice were provided with food and water ad libitum. The collected urine samples were centrifuged to remove debris. The supernatant protein and creatinine concentrations in the urine samples were measured using a Mouse Urine Albumin Assay Kit (EIA06044m; Wuhan Xinqidi Biotech Co. Ltd.) and a Urine Creatinine Kit (C011‐2‐1; Nanjing Jiancheng Bioengineering Institute), respectively, according to the manufacturer's instructions. Urinary albumin levels are reported as the ratio of urinary albumin to creatinine.

Kidney cortical samples were homogenized in phosphate‐buffered saline (PBS) containing 1 mM ethylenediaminetetraacetic acid and 10 μM indomethacin. Samples were centrifuged at 10,000 × g for 15 min at 4°C to collect supernatants. The levels of PGE₂ in the supernatants were measured using an ELISA kit (514010, Cayman Chemical) according to the manufacturer's instructions.

Kidney cortical samples were homogenized on ice in 300 μL of the malondialdehyde (MDA) lysis buffer containing BHT. The samples were centrifuged at 13,000 × g for 10 min to remove insoluble material. Thereafter, 200 μL of the supernatant from each homogenized sample was placed in a microcentrifuge tube. The MDA concentration was measured using an MDA assay kit (A003‐4‐1, Nanjing Jiancheng Bioengineering Institute) according to the manufacturer's instructions.

Calcitriol1,25‐Dihydroxycholecalciferol (1,25‐dihydroxyvitamin D₃) is the active form of vitamin D normally present in the kidneys. The detection of calcitriol is based on competitive enzyme‐linked immunosorbent assay technology, in which TMB is converted into blue under the catalysis of peroxidase and then converted into the final yellow color under the action of acid. The intensity of the yellow color was inversely proportional to the amount of calcitriol bound to the plate. Briefly, the kidney tissue was homogenized in PBS to obtain the supernatant. The kit components and samples were equilibrated at room temperature before use. Supernatants were processed and analyzed using a commercial kit (E‐HS30655Mo, Enovabio).

RNA samples were isolated using TRIzol reagents (15596026CN, Thermo Fisher Scientific). cDNA was generated via reverse transcription using the HyperScript III RT SuperMix for qPCR with gDNA Remover (R202‐02, Enzyartisan). The specific primers for target genes were designed using the Primer3 software (available at http://frodo.wi.mit.edu/primer3/↗) and were listed in the Table S3 of Appendix S1. qRT‐PCR amplification was performed using 2 × S6 Universal SYBR qPCR Mix (Q204‐01, Enzyartisan). Levels of the housekeeping gene beta‐actin were used as an internal control for the normalization of RNA quantity and quality differences among the samples. The qRT‐PCR data were acquired using Roche LightCycler 480 II/96 with the Light Cycler 480 SW 1.5 software. The fold changes in gene expression were normalized to beta‐actin expression.

Tissues and cell pellets were collected and resuspended in RIPA lysis buffer and homogenized to lysates. The lysates were incubated on ice for 30 min and centrifuged at 4°C for 15 min at 12,000 g. Protein levels of soluble supernatant were estimated using a BCA Protein Assay Kit (23227, Thermo Fisher Scientific) according to the manufacturer's protocols. Equal amounts of proteins were applied to a 10 or 12.5% sodium dodecyl sulfate‐polyacrylamide gel for electrophoresis to separate proteins, after which the proteins were electrophoretically transferred onto nitrocellulose membranes (Millipore). The membranes were blocked using 3% bovine serum albumin (A8020, Solarbio) for 1 h at room temperature and then incubated with indicated antibodies overnight at 4°C. In this study, the following antibodies were used: p16 (SC‐1661, Santa Cruz), p21 (SC‐6246, Santa Cruz), p53 (SC‐126, Santa Cruz), mPGES‐2 (10881‐1‐AP, Proteintech). The secondary antibodies: HRP‐labeled Goat Anti‐Mouse IgG (H+L) (A0216, Beyotime) and HRP‐labeled Goat Anti‐Rabbit IgG (H+L) (A0208, Beyotime) were used. Protein bands were detected with an ECL Plus substrate (044–500 mL, Beijing Fluorescence Biotechnology Co. Ltd). Band intensities were analyzed using the ImageJ software.

Statistical analyses were performed using GraphPad Prism version 8 (GraphPad Software Inc., La Jolla, CA, USA). All samples were randomly allocated to the treatment groups. All statistical measurements are presented as the mean ± standard error of the mean (SEM). Unpaired Student's t‐test (two‐tailed) was used to assess the statistical significance of an observed parameter between experimental groups. To compare more than two groups, data were analyzed using one‐way ANOVA, followed by Tukey's post hoc test. *p Values < 0.05 were considered statistically significant. Details regarding other materials and methods are provided in the Appendix S1.

Single cell transcriptomic data analyzed in this study are publicly available under GEO accession GSE240374. The data that support the findings of this study are available from the corresponding author upon reasonable request.