The HK-2 cell line (ShangHai FuHeng BioLogy, FH0228) was maintained in Dulbecco’s modified Eagle’s medium (HyClone Laboratories) supplemented with 10% FBS. HK-2 cells were treated with oxalate (Rhawn, China) or oxalate + fisetin (Aladdin, China) for 48 h. The culture medium of the oxalate (0.5 mM) or oxalate (0.5 mM)+ fisetin (10 μM) groups was collected (CM-2D) and replaced with fresh medium without oxalate or fisetin, and then HK-2 cells were further cultured for another 48 h; the culture medium (CM-4D) was collected for subsequent experiments in which macrophage-like M0 cells (M0-MΦs) were cultured. The THP-1 cells (iCell-h213) were cultured in 6-well plates with RPMI-1640 media supplemented with 10% FBS and 0.05 mM β-mercaptoethanol. The THP-1 cells were differentiated into M0-MΦs by treatment with 100 ng/ml PMA (Sigma) for 3 d (31). Adherent cells were further incubated with fresh medium containing LPS (100 ng/ml, Sigma) (32) and IFN-γ (50 ng/ml, Peprotech, Rocky Hill, USA) (33) for 24 h to stimulate M1 macrophage (M1-MΦ) polarization or with IL-4 (20 ng/ml, Peprotech, Rocky Hill, USA) (34) and IL-13 (20 ng/ml, Peprotech, Rocky Hill, USA) (35) for 24 h to stimulate M2 macrophage (M2-MΦ) polarization. Simultaneously, the aforementioned CM-4D from HK-2 cell culture was added. Macrophage polarization was assessed via RT–qPCR and western blot analyses. CM-4D and CM-2D were frozen at -20 °C until analysis using commercial enzyme-linked immunosorbent assay (ELISA) kits (Elabscience Biotechnology Co., Ltd., Wuhan, China) for IL-6, IL-1β, matrix metalloproteinase-3 (MMP-3) and MMP-13.

To knock down LXN or Rps3 expression, HK-2 cells were seeded overnight and incubated until they reached 60–80% confluence. HK-2 cells were transfected with LXN or Rps3 siRNA (Sigma–Aldrich) using Lipofectamine 3000 (Invitrogen, USA). The relative sequences are shown in Supplementary Table 1. During 48 h of transfection, HK-2 cells were simultaneously incubated with oxalate and then replaced with fresh culture medium for an additional 2 d of incubation; the obtained supernatant (CM-4D) was added to M0-MΦs to perform polarization experiments. For the experimental group receiving combined treatment, HK-2 cells were transfected with LXN knockdown vectors and concurrently incubated with oxalate and fisetin for 48 hours. Cellular senescence was evaluated using classic molecular markers and cell cycle regulators, including p16 and p53 (36). p16 is a critical inhibitor of CDK4/6 kinase activity that induces hypophosphorylation and sustained activation of the retinoblastoma (Rb) (37), which ultimately leads to cell cycle arrest and prevention of cell division (38). p53 is a master regulator orchestrating DNA damage responses and senescence induction (39, 40). In the group receiving combined LXN knockdown and fisetin treatment, senescence markers in HK-2 cells were assessed through SA-β-gal staining and p16 protein expression analysis.

HK-2 cells were fixed in 4% paraformaldehyde for 10 min and then stained with a Senescence β-Galactosidase Staining Kit (Abbkine, Wuhan, China). The images were observed using microscopy (Leica DMi8).

HK-2 cells were seeded into 96-well plates at 5 × 10³ cells per well. After 48 h of incubation, HK-2 cell proliferation was detected with a cell counting kit-8 (CCK8, Multi Sciences Ltd., Hangzhou, China). The OD450 absorbance was measured to evaluate cell activity.

The levels of superoxide dismutase (SOD), total antioxidant capacity (T-AOC), glutathione (GSH) and lactate dehydrogenase (LDH) were measured using commercial assay kits from Nanjing Jiancheng Bioengineering Institute of China (SOD, A001-3-2; T-AOC, A015-2–1; GSH, A006-2-1; and LDH, A020-2-2) according to the manufacturer’s instructions.

Cells were incubated with 0.5% Triton X-100 (MERCK) for 20 min and washed with PBS. Nonspecific binding was blocked by incubation with normal horse serum (Gibco) at 37 °C for 30 min. The cells were then incubated with anti-p16 antibody (Abcam, ab108349) or anti-Rps3 antibody (Abcam, ab128995) at 4 °C overnight, followed by incubation with an Alexa Fluor 488-conjugated goat anti-rabbit IgG secondary antibody (AmyJet Scientific Inc., Wuhan, China) in the dark at 37 °C for 1 h. Coverslips with stained cells were further stained with DAPI (Abcam). Positive cells were visualized and imaged using a fluorescence microscope (Leica DMi8).

Cellular RNA extraction, reverse transcription, and RT–qPCR were performed as described previously (41). The transcript levels were normalized to those of GAPDH. Gene expression was calculated using the 2-ΔΔCt method. The specific primers used are shown in Supplementary Table 1.

Protein preparation and western blot analysis were performed as described previously (42, 43). The primary antibodies used were as follows: anti-VEGF rabbit antibody (Abcam, ab52917), anti-MMP13 rabbit antibody (Abbkine, Wuhan, China, ABP51805), anti-IL-1β rabbit antibody (Abcam, ab254360), anti-MCP1 rabbit antibody (Abcam, ab214819), anti-LXN rabbit antibody (Abcam, ab154744), anti-Rps3 rabbit antibody (Abcam, ab128995), anti-p16 rabbit antibody (Abcam, ab108349), anti-p53 rabbit antibody (Abcam, ab32389), anti-TNF-α rabbit antibody (Abbkine, Wuhan, China, ABM0127), anti-iNOS rabbit antibody (Abcam, ab283655), anti-Arg-1 rabbit antibody (Abcam, ab96183) and anti-CD163 rabbit antibody (Abcam, ab182422). β-actin or GAPDH was used for normalization.

All experimental procedures were performed following the rules of the National Institutes of Health Guide for the Care and Use of Laboratory Animals, and approval was obtained from the Ethics Committee on the Use of Live Animals of Harbin Medical University. Male Sprague–Dawley rats (180–220 g, 6–8 weeks old) were randomly divided into six groups: the control group, oxalate group, oxalate+fisetin group, oxalate+AAV-Null group and oxalate+AAV-shLXN group,oxalate + fisetin + AAV-shLXN group. Normal control rats were given an equal volume of PBS. The rats received intraperitoneal injection of 60 mg/kg glyoxylic acid (Sigma–Aldrich) five times a week for four weeks (oxalate group) (44) to establish the kidney stone rat model. Fisetin, a classic senolytic, was intragastrically administered at 50 mg/kg to kidney stone model rats five days a week for four weeks (45)from the first day of glyoxylate injection. To knockdown LXN expression in vivo, rats were transduced with an AAV serotype 9 vector encoding a green fluorescent protein reporter together with either short hairpin RNAs (shRNAs) targeting LXN in the kidney (AAV-shLXN) or an empty vector (AAV-null) (Shanghai GeneChem Co., Ltd. Shanghai, China). The rats were injected with either AAV-shLXN or AAV-Null via the tail vein, followed by an additional glyoxylate injection (oxalate+AAV-shLXN group). The rats treated with glyoxylate and AAV-Null vector injection (oxalate + AAV-Null group) were used as the virus transfection control group. All animal kidney tissues were obtained on Day 30 after the first injection of glyoxylate.

Rat kidneys were embedded in paraffin, and the samples were sectioned. Hematoxylin and eosin (HE) staining, Von Kossa staining and immunohistochemical analysis were performed as described previously (46–49).Immunostaining was performed with specific antibodies against the target proteins LXN (Bioss, bs-1971R), Rps3 (Absin, abs116130), p53 (Bioss, bs-2090R), p16 (Bioss, bs-0740R), iNOS (Bioss, bs-0162R) and CD163 (Bioss, bs-2527R). Images were captured using a microscope (Leica DMi8). Kidney injury scores were evaluated using a scale ranging from 0 to 4 as described before (50). Briefly, 0 indicating normal; 1 representing less than 25%; 2 indicating 25-50%; 3 representing 50-75%; and 4 indicating greater than or equal to 75%. Quantitative evaluation of Von Kossa staining and immunohistochemical analysis were performed using Image J software as described previously (51).

Statistical analysis was performed using IBM SPSS 27.0. Data were presented as the mean ± Standard Deviation (SD). Two-tailed Student’s t-tests and one way ANOVA followed by Tukey’s test. were used for normally distributed data when appropriate. A non-parametric test was used for non-normally distributed data. P < 0.05 was considered statistically significant.

Fisetin, a classic senolytic agent, was employed in this study to treat cells for evaluating the impact of anti- senescence effect on aberrant macrophage polarization and subsequent stone formation. To determine the effects of oxalate with or without fisetin on cell viability, HK-2 cells were treated with oxalate (0, 0.5 or 1.0 mM) and varying concentrations of fisetin (0, 5, 10, 20 or 40 μM) for 48 h. HK-2 cell viability was significantly lower at a dose of 1.0 mM oxalate compared with that of the vehicle control. Incubating HK-2 cells with 0.5 mM oxalate and various concentrations of fisetin (5 and 10 μM) had no significant effect on cell viability. Incubating HK-2 cells with 40 μM fisetin or 0.5 mM oxalate+ 20 μM fisetin significantly decreased cell viability compared with that of the control group (Figure 1A). Since substantial viability drops demonstrated the cytotoxic potential of oxalate and fisetin, the maximum non-cytotoxic doses (just before the significant viability decrease) from agent’s concentration gradient were selected for subsequent cellular senescence experiments. Substantial numbers of senescent cells were evaluated via senescence-associated β-galactosidase (SA-β-gal) activity, and the results revealed that the frequency of SA-β-gal(+) senescent cells increased significantly after 0.5 mM oxalate treatment, and that 10 μM fisetin significantly decreased the proportion of SA-β-gal (+) senescent cells (Figure 1B). Typical images are shown in Figure 1C. Therefore, we treated HK-2 cells with 0.5 mM oxalate for 48 h to establish a model of RTEC injury and 10 μM fisetin was administered simultaneously to evaluate the effect of anti-senescence in the subsequent experiments. In addition to SA-β-gal activity detection, the expression of typical markers of cellular senescence (e.g. p16 and p53) were also determined via immunofluorescence staining and western blotting. Immunofluorescence revealed that p16 staining was more intense in HK-2 cells incubated with oxalate than in control HK-2 cells, and 10 μM fisetin inhibited p16 staining intensity under oxalate induction (Figure 1D). Western blot analysis revealed that oxalate significantly increased the levels of p16 and p53, and fisetin reversed this effect of oxalate on p16 and p53 expression (Figure 1E).

Superoxide dismutase (SOD), an antioxidant enzyme, catalyzes the decomposition of superoxide radicals into oxygen and hydrogen peroxide, thereby reducing oxidative stress damage. Total antioxidant capacity (T-AOC) can be used to evaluate the overall antioxidant level of all antioxidant substances (including enzymatic and non-enzymatic antioxidants) in samples such as serum. Glutathione (GSH) is a critical non-enzymatic antioxidant within cells that directly scavenges free radicals and participates in maintaining redox homeostasis. Lactate dehydrogenase (LDH) is a key enzyme in glycolysis that catalyzes the interconversion between pyruvate and lactate. It is commonly used as a biomarker for cellular damage or death, as it is released into the bloodstream upon rupture of the cell membrane. Subsequently, the levels of the aforementioned oxidative and antioxidative indicators, including SOD, T-AOC, GSH and LDH, were measured to evaluate injury to RTECs. The levels of SOD, T-AOC and GSH were significantly lower in the oxalate group than in the control group, whereas the LDH level was higher. Compared with those of the oxalate group, the oxalate + fisetin group presented elevated SOD, T-AOC and GSH levels and reduced LDH levels (Figure 2A). Next, we used a model of oxalate-induced senescent HK-2 cells as described above to determine whether HK-2 cell senescence paralleled the increased release of proinflammatory factors of the senescence-associated secretory phenotype (SASP). We collected mRNA and protein from control cells and senescent HK-2 cells treated with oxalate with or without fisetin in 2-d cultures. The mRNA and protein expression of the proinflammatory SASP-associated factors VEGF, MMP13, MCP-1 and IL-1β were upregulated by oxalate, and this upregulation was partially reversed by fisetin (Figures 2B, C). The changes in the concentrations of the SASP-associated factors IL-1β, MMP13, IL-6 and MMP3 in the culture medium from 2-day-old cultures (CM-2D) were similar to the MMP13 and IL-1β expression in control cells (Figure 2D). On the third day, the supernatants of HK-2 cells induced with oxalate with or without fisetin were replaced with fresh culture medium, and 2 days later, the culture medium (CM-4D) was obtained for subsequent experiments on macrophage polarization to exclude the effects of oxalate or fisetin on the biological behavior of macrophages. IL-1β, MMP13, IL-6 and MMP3 levels of CM-4D from HK-2 cell cultures after oxalate induction were significantly upregulated, and this upregulation was also significantly decreased in the CM-4D from oxalate+fisetin-treated HK-2 cell cultures (Supplementary Figure 1).

To study whether senescence of RTECs can influence the polarization of M0-MΦs into either inflammatory macrophages (M1-MΦs) or anti-inflammatory macrophages (M2-MΦs), THP-1 cells were induced to differentiate into macrophages via PMA treatment. Based on IFNγ and LPS induction toward M1-MΦs or IL-4 and IL-13 induction toward M2-MΦs for another 24 h, we assayed macrophage polarization in the culture medium (CM-4D) from HK-2 cells treated with oxalate or oxalate+fisetin. In current macrophage research, TNF-α (52) and iNOS (53) are well-established biomarkers for identifying M1 macrophages, whereas Arg-1 (54) and CD163 (55) are widely recognized as markers of M2 macrophages (56). Therefore, in accordance with conventional macrophage classification schemes (57, 58), we employed TNF-α and iNOS for the identification of M1 macrophages and used Arg-1 and CD163 to characterize M2 macrophages in this study. As shown in Supplementary Figure 2, compared with those in native macrophages, the mRNA expression of TNF-α and iNOS was increased in M1 macrophages, and the expression of Arg-1 and CD163 mRNA was increased in M2 macrophages, confirming the success of macrophage induction. As shown in Figures 3A, B, CM-4D from oxalate-induced HK-2 cells (CM-4D-oxalate) significantly increased TNF-α and iNOS mRNA/protein expression in M1-MΦs, and this upregulation was significantly reversed by CM-4D from oxalate+fisetin-treated HK-2 cells (CM-4D-oxalate+fisetin). However, there was no significant change in Arg-1 or CD163 mRNA/protein expression in M2-MΦs in response to CM-4D from oxalate +/- fisetin-treated HK-2 cells (Figures 3C, D). These results indicated that CM-4D oxalate induced the polarization of M0-MΦs toward M1-MΦs but not toward M2-MΦs.

Compared with those in healthy controls, the expression of four senescence-associated genes, including LXN, GLS, PTGS1 and CFB, in nephrolithiasis patients (59) was found to be significantly upregulated in HK-2 cells treated with oxalate with or without fisetin (10 μM). RT–qPCR revealed that the LXN, GLS, PTGS1 and CFB genes were expressed at distinctly higher levels in oxalate-induced HK-2 cells than in control cells and were further significantly downregulated by fisetin. The ratios of the relative expression of the above genes compared with those of the control group are shown in Figure 4A, and the relative LXN mRNA levels are shown in Figure 4B. LXN, the gene expression of which varied most significantly after oxalate + fisetin intervention, was selected for further study. The protein expression of LXN after oxalate treatment with or without fisetin was confirmed to have a similar effect on gene expression (Figure 4C). To further evaluate the regulatory effect of LXN on cellular senescence, experiments with LXN knockdown in HK-2 cells were carefully performed, and the alterations in biological behaviors related to cellular senescence were evaluated. The knockdown efficiency of LXN was determined by RT–qPCR and western blot analyses. At 48 h after transfection, 75% knockdown efficiency was achieved, as shown by both gene and protein expression (Supplementary Figures 3A, B). The results of senescence-associated β-galactosidase (SA-β-gal) activity experiments showed that the increased proportion of SA-β-gal (+) cells induced by oxalate was significantly decreased by LXN knockdown (Figure 4D). In addition, LXN knockdown significantly decreased the protein expression of p16 and SASP-associated factors (VEGF, MMP13, MCP-1 and IL-1β), which were dramatically increased in oxalate-induced HK-2 cells (Figures 4E, F). Consistent with changes in the protein expression of SASP-associated factors in HK-2 cells, the concentrations of SASP factors (i.e., IL-1β, MMP13, IL-6 and MMP3) in the supernatant of HK-2 cell cultures demonstrated similar trends after oxalate treatment with or without LXN knockdown intervention (Figure 4G). Interestingly, the group treated with LXN knockdown combined with fisetin showed a significant reduction in both the proportion of SA-β-gal (+) cells and the protein expression of p16 compared to the groups treated with LXN knockdown or fisetin alone (Supplementary Figure 4). The combined treatment of fisetin and LXN knockdown reduced the levels of senescence markers to an extent that was nearly additive but not fully linear, suggesting the effect of LXN occurs through the fisetin–senescence pathway.

To observe the influence of delaying senescent HK-2 cell accumulation induced by LXN knockdown on macrophage polarization, HK-2 cells subjected to LXN knockdown were simultaneously incubated with oxalate, which was replaced with fresh culture medium for an additional 2 d, after which the supernatants were obtained (CM-4D) and added to PMA + IFNγ + LPS- or PMA + IL-4 + IL-13-stimulated THP-1 cells. As shown in Figures 4H, I, CM-4D from oxalate-induced scramble siRNA-transfected HK-2 cells (CM-4D-oxalate+Scr siRNA) significantly increased TNF-α and iNOS mRNA/protein expression in M1-MΦs, and this upregulation was significantly reversed by CM-4D from oxalate-treated HK-2 cells after LXN knockdown (CM-4D-oxalate+si-LXN). Furthermore, there was no significant change in Arg-1 or CD163 mRNA/protein expression in M2-MΦs treated with CM-4D from HK-2 cells exposed to oxalate+Scr siRNA or oxalate+si-LXN compared with those in the control group (Figures 4J, K). These results suggest that M1-MΦ polarization induced by culture medium from HK-2 cells under oxalate+LXN-knockdown (CM-4D-oxalate+si-LXN) intervention was significantly lower than that induced by CM-4D-oxalate+Scr siRNA. Overall, we concluded that LXN knockdown effectively attenuated the oxalate-induced senescent response and subsequently attenuated proinflammatory M1 macrophage polarization.

Because LXN has been reported to perform its biological function by binding to the protein ribosomal protein 3 (Rps3) (60) and Rps3 has an important extraribosomal role in the DNA damage response and in the regulation of p53 degradation (61), we detected the expression of Rps3 and p53 in the next step to investigate the mechanism by which LXN inhibits the senescence of HK-2 cells. Interestingly, less Rps3 was localized to the nucleus in HK-2 cells under oxalate induction than in control cells, and the decreased nuclear translocation of Rps3 was significantly increased by fisetin (10 μM) treatment (Figure 5A). This finding was further confirmed by the lack of significant changes in the mRNA and protein expression of Rps3 in the presence of oxalate with or without fisetin treatment (Figures 5B, C). To evaluate the relationships among LXN, Rps3 and p53, LXN or Rps3 knockdown was performed. The knockdown efficiency of Rps3 is shown in Supplementary Figures 3C, D. As demonstrated in Figures 5D, E, knockdown of LXN had little effect on the expression levels of Rps3 mRNA, p53 mRNA and Rps3 protein, whereas p53 protein expression was significantly downregulated. When Rps3 was knocked down, the expression levels of LXN mRNA, p53 mRNA and LXN protein did not significantly change, whereas p53 protein expression was significantly upregulated, suggesting that p53 is downstream of LXN/Rps3 and that LXN exerts its effects in combination with Rps3 to affect the nuclear translocation of Rps3; these findings are further supported by the result that nuclear translocation of Rps3 was obviously elevated after LXN silencing, as shown by immunofluorescence experiments (Figure 5F).

The experimental procedures for animal studies are schematically depicted in Figure 6A. Blood Urea Nitrogen (BUN) is a common diagnostic readout, with higher levels indicating impaired renal function and reduced glomerular filtration rate. Serum creatinine (Scr) is also a widely used and clinically essential biomarker for evaluating renal function. Higher Scr levels typically reflect a greater degree of kidney impairment. To investigate the in vivo role of cellular senescence in kidney stone formation as well as renal impairment, the levels of Scr and BUN were measured. Compared with those in the normal group, the Scr and BUN levels in the oxalate group were notably increased on Day 30, and the Scr and BUN levels in the oxalate+fisetin group were markedly lower than those in the oxalate group, indicating the amelioration of renal impairment after fisetin administration (Figures 6B, C). Renal histological analyses showed that fisetin attenuated oxalate-induced injury to RTECs, as demonstrated by lower-grade dilatation of the tubules and decreased levels of mononuclear cell infiltrates in the interstitium of rats in the oxalate+fisetin group compared with that in the oxalate group (Figure 6D). Von Kossa staining of isolated kidney tissues revealed similar differences in crystal formation (Figure 6E), suggesting that fisetin has a protective role in ameliorating the formation of kidney stones. Furthermore, we examined the expression levels of the LXN/Rps3/p53 pathway in kidney tissue by immunohistochemistry staining. Our results showed that, compared with that under normal conditions, LXN and p53 protein expression was significantly increased in the oxalate group, whereas fisetin administration significantly decreased LXN and p53 protein expression. Compared with that in the oxalate group, the total Rps3 protein staining intensity was not significantly different, and less Rps3 protein staining was observed in the cell nuclei of the oxalate+fisetin group (Figure 6F). Immunohistochemical staining of the polarization biomarkers iNOS (M1-MΦs) and CD163 (M2-MΦs) was also performed to evaluate macrophage polarization, and immunohistochemical staining revealed that iNOS was significantly increased in the kidney tissue of the oxalate group. This increase in iNOS staining was significantly decreased in the oxalate+fisetin-treated group. However, there was no significant difference in CD163 staining among the control, oxalate and oxalate+fisetin groups (Supplementary Figure 5).

To study the effect of the LXN gene on cellular senescence in nephrolithiasis, AAV-shLXN or AAV-null vectors (as controls) were injected via the tail vein into stone model rats on the first day. Scr/BUN determinations, H&E analysis and Von Kossa staining showed that, compared with that of the stone model oxalate group, the oxalate+AAV-Null group had a similar effect on crystal deposition (data not shown). Interestingly, H&E, Von Kossa staining and immunohistochemical analysis revealed that, compared with those of the fisetin group, AAV-shLXN had a similar effect on crystal deposition, LXN/Rps3/p53 and iNOS/CD163 protein staining; more importantly, the protective effect of AAV-shLXN was even greater than that of fisetin (Figures 6D–F, Supplementary Figure 5). More importantly, the combined treatment with fisetin and LXN knockdown led to a more pronounced reduction in Scr and BUN levels, tubular dilation, mononuclear cell infiltration, crystal deposition, and expression levels of LXN, p53, and iNOS in renal tissue—as assessed by immunohistochemical staining—compared to each treatment alone (Supplementary Figures 5, 6). Taken together, our in vivo experimental results suggest that senescence-targeting therapy effectively inhibits cellular senescence through the LXN/Rps3/p53 pathway, thus reducing macrophage polarization toward the M1 phenotype and ultimately exerting a protective effect against kidney stone formation. LXN gene silencing might be a potential therapeutic method for treating nephrolithiasis.