High-content screening identifies ganoderic acid A as a senotherapeutic to prevent cellular senescence and extend healthspan in preclinical models

Abstract

Accumulated senescent cells during the aging process are a key driver of functional decline and age-related disorders. Here, we identify ganoderic acid A (GAA) as a potent anti-senescent compound with low toxicity and favorable drug properties through high-content screening. GAA, a major natural component of Ganoderma lucidum, possesses broad-spectrum geroprotective activity across various species. In C. elegans, GAA treatment extends lifespan and healthspan as effectively as rapamycin. Administration of GAA also mitigates the accumulation of senescent cells and physiological decline in multiple organs of irradiation-stimulated premature aging mice, natural aged mice, and western diet-induced obese mice. Notably, GAA displays a capability to enhance physical function and adapts to conditional changes in metabolic demand as mice aged. Mechanistically, GAA directly binds to TCOF1 to maintain ribosome homeostasis and thereby alleviate cellular senescence. These findings suggest a feasible senotherapeutic strategy for protecting against cellular senescence and age-related pathologies.

Fig. 1. Identification of the anti-senescence natural product by high-content screening.

a Experimental scheme of the high-content screening strategy. b High-content imaging of young and natural senescent IMR-90 cells (SEN) induced by replication (n = 3 independent experiment). A merger by bright-field and fluorescence images. c Three-dimensional diagram of high-content analyses among 112 natural products with the treatment for 3 days. X axis: cell number; Y axis: SA-β-Gal; Z axis: nuclear area. Fold change (FC) was the ratio of natural products-treated groups to DMSO-treated groups. Vitamin C, flavone, and emodin had the best-improved effects on the nuclear area, cell number, and SA-β-Gal respectively. Rapamycin and fisetin were classic senotherapeutics. d Heatmap showing the senescence-related indicators in IMR-90 with the treatment of 47 candidate natural products. The indicators were displayed as SA-β-Gal, nuclear area, cell number, cell area, and cell width-length ratio. e SA-β-Gal expression in senescent HUVECs models with the treatment of 47 candidate natural products. Models were induced by replication, H₂O₂, etoposide, and the above-mixed stimulus. The natural products that decreased SA-β-Gal expression in all 4 models were put on the right side of the frame. f The relative levels of SA-β-Gal expression altered by 11 compounds in four senescent models of HUVECs. Six candidate compounds with above 30% inhibition of SA-β-Gal levels in four senescent models were labeled in black lines, while others were in red lines. g Representative images and quantification of SA-β-Gal staining after the 4-day treatment of six compounds in senescent MEF cells with different concentrations (0, 0.1, 1, 10, 100, 500, and 1000 μM). 0 μM: DMSO. (n = 3 independent experiment). h LDH release ratio in senescent MEF cells treated with different concentrations of six compounds. The asterisk indicates statistical significance between control and natural product-treated groups (n = 5 independent experiment). Comparisons are performed by Two-sided t tests or one-way ANOVA analysis. All data are expressed as the mean ± SEM. *P <  0.05, **P <  0.01, ***P <  0.001. Source data and exact P value are provided as a Source data file.

Fig. 2. GAA alleviates the senescence of various cells and the aging-related phenotype of C. elegans.

a Measurement of LDH release in natural senescent cells treated with 10 μM GAA treatment or DMSO (n = 6 independent experiments). b Representative images and quantification showing SA-β-Gal staining in young (Y), SEN (S), or GAA-treated SEN cells (G) (n = 3 independent experiments). c EdU staining and the statistics of EdU⁺ cells and d nuclear area in GAA-treated senescent cells or controls (n = 3 independent experiments). e Representative survival curve of worms treated with DMSO, rapamycin (100 μM), and GAA (10, 100, and 1000 μM). The mean and maximum lifespans were shown as the mean ± standard deviation (SD) of six independent experiments. f Lifespan analyses of worms challenged by heat stress (30 °C) and g oxidative stress (1.5% H₂O₂) with the treatment for 7 days of DMSO, rapamycin (100 μM), and GAA (1000 μM). The mean and maximum lifespans were shown as the mean ± SD of three independent experiments. h The levels of lipofuscin, ROS (DCFH-DA, 10 μM), and lipids (Nile Red, 10 μg/ml) in worms after 10 days of intervention with DMSO, rapamycin, and GAA (n = 3 independent experiments with ten nematodes for each experiment). i–k Physical function of worms after 10-day treatments (n = 3 independent experiments with 10 nematodes for each experiment); i Worms were observed the spontaneous or regular sinusoidal movement (Normal), irregular or uncoordinated movement (Sluggish), and movement only by touching (Immobile). j The swing frequency and k pharyngeal swallowing frequency were detected in worms within 1 minute. Comparisons are performed by Two-sided t tests or one-way ANOVA analysis. All data are expressed as the mean ± SEM or SD. *P <  0.05, **P <  0.01, ***P <  0.001. Source data and exact P value are provided as a Source data file.

Fig. 3. GAA treatment does not cause abnormal proliferation or damage in middle-aged mice and tumor-bearing mice.

a Experimental procedures for 2-month-old C57BL/6J mice fed with a GAA-added diet (120 mg/kg) for 3 months. ND: normal diet. b Tandem mass spectrum of GAA and GAA contents detected in the heart, kidney, liver, lung, and serum of mice at the end of the experiment (n = 3 or 4 mice). c Biochemical analysis to measure organ function, including heart (CKMB), d kidney (creatinine), and e liver (ALT) (n = 10 mice). f Immunohistochemistry images and quantification of the marker for cell proliferation (Ki67) in the liver (n = 3 mice). g EdU⁺ cells and h LDH levels in a variety of young cells with the presence or absence of 10 μM GAA (n = 3 or 6 independent experiments). i Experimental scheme for 2-month-old C57BL/6J mice subcutaneously inoculated with Hep1-6 cells (5 × 10⁶) and subsequent intraperitoneal administered with GAA (15 mg/kg·bw) or vehicle once every 3 days for 18 days. j Gross morphology and measures of tumor volume and tumor weight values of mice at the study endpoint (n = 9 mice). k Representative images and quantification of Ki67 expression (n = 3 mice). l Organ index (organ weight relative to body weight) in normal mice and tumor-bearing mice (n = 9 mice). m Representative images of crystal violet staining in cancer cells after GAA (10 μM) treatment for 48 h. The colonies of the cells were manipulated as black (n = 3 independent experiments). n Representative images of scratch assay in cancer cells after GAA (10 μM) treatment for 48 h (n = 3 independent experiments). The red line marked the outline after cell migration. Comparisons are performed by Two-sided t tests or one-way ANOVA analysis. All data are expressed as the mean ± SEM. *P <  0.05, **P <  0.01, ***P <  0.001. Source data and exact P value are provided as a Source data file.

Fig. 4. The treatment of GAA or DQ alleviates aging-related disorders of mice exposed to IR.

a Experimental scheme. Mice were subjected to a low dose (2.8 Gy, 1 Gy/min) of whole-body IR after 5 days of treatment with GAA (15 mg/kg·bw) or vehicle by gavage, followed by the oral administration of GAA, DQ (5 + 50 mg/kg·bw), or vehicle. b Body weight monitoring of IR mice. c Representative images of H&E staining in spleen and testis at the experiment endpoint (n = 4 mice). d Gross organs stained with SA-β-Gal. e Immunohistochemical image and quantification of γH2AX expression in heart, liver, kidney, and lung (n = 4 mice). f ELISA to test the levels of serum klotho (aging-related protein) and g IL-6 (a cytokine of SASP) (n = 10 mice). h Representative images of pathological detection in the heart, liver, kidney, and lung (n = 4 mice). i Biochemical analysis of CKMB, j creatinine, k AST, and l ALT (n = 8 or 10 mice). m Behavioral performance of physical function including forelimb grip and n rotarod testing (n = 8 or 10 mice). Comparisons are performed by Two-sided t tests or one-way ANOVA analysis. All data are expressed as the mean ± SEM. *P <  0.05, **P <  0.01, ***P <  0.001. Source data and exact P value are provided as a Source data file.

Fig. 5. GAA treatment extends the healthspan of aged mice.

a Experimental design for 16-month-old male C57BL/6J mice fed with a GAA-added diet (120 mg/kg, n = 40 mice) and vehicle (n = 46 mice) for 6 months. b Survival curve of mice from 16 months to 22 months of age. c Life expectancy estimation in GAA-treated aged mice and their counterparts. Days of life gained in aged mice with GAA treatment compared to the controls. d The mean frailty index of mice treated with GAA and their counterparts. Mice exhibited various aging disorders and a higher risk of multiple morbidities when they approached the end of life (higher percentage of lifespan). The lifespan of the mice was divided into five quantiles (Q), and the frailty score of mice in each quantile of lifespan was calculated. Hence, this graph of the area under the curve may indicate whether the intervention can delay the onset of morbidity. e Monitoring of the frailty index for each mouse every two months. Mixed models to analyze repeated measurements of longitudinal data. Each dot represented the total score of one animal. f Gross organs and tissues stained with SA-β-Gal. g Immunoblots and densitometric analysis of the protein levels of senescent markers, including p53, p21, and p16 (n = 3 mice). h Assessment of serum klotho and i IL-6 levels by ELISA (n = 10 mice). j–l Indicators related to organ function (CKMB, creatinine, and ALT) (n = 10 mice). m, n Representative images in pathological analysis of heart, kidney, liver, and lung (n = 3 mice). Young: Young+vehicle; Old: Old+vehicle. Comparisons are performed by Two-sided t tests or one-way ANOVA analysis. All data are expressed as the mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001. Source data and exact P value are provided as a Source data file.

Fig. 6. GAA treatment improves physical, metabolic, and brain-related dysfunction in aged mice.

a Measurement of time in rotarod, b running distance on the treadmill, and c grip strength of experimental mice (n = 8 mice). d Body composition analysis in lean mass of mice by nuclear magnetic resonance (n = 12 mice). e Representative microCT images of bone microarchitecture at the femur. Quantification of microCT-derived bone volume fraction (BV/TV), trabecular number (Tb.N), trabecular thickness (Tb.Th), and trabecular separation (Tb.Sp) (n = 6 mice). f–h The path map during the behavioral experiment. Comparative statistics including the f times entered into and time spent in the open arms in elevated plus maze testing, g the distance and the frequencies of access to the central zone in open field testing, h the time interacting with the novel objects in novel object recognition testing (n = 10 or 12 mice). i–k Biochemical index of lipid metabolism, including TG, FFA, TC, HDL-C, and LDL-C (n = 10 or 15 mice). l The weight distribution of the aged mice in each group at the end of the experiment. m Glucose tolerance test and area under the curve of mice (n = 8 mice). n Insulin levels and the corresponding o insulin resistance index (n = 10 mice). Comparisons are performed by Two-sided t tests or one-way ANOVA analysis. All data are expressed as the mean ± SEM. *P <  0.05, **P <  0.01, ***P <  0.001. Source data and exact P value are provided as a Source data file.

Fig. 7. GAA treatment improves aging-related phenotype in obese mice.

a Schematic design for 2-month-old C57BL/6J mice fed with a western diet or normal diet for 16 weeks, and then given a treatment with GAA for 12 weeks. b Mean body weight monitoring in obese and control mice and the statistics of final body weight (n = 10 mice). c Representative images and quantification of tissues stained with SA-β-Gal, including heart, kidney, liver, and lung. The statistical data were expressed as the ratio of positive stained area (n = 3 mice). d The circulating klotho and e IL-6 levels were determined by ELISA (n = 7–10 mice). f Measurements of organ function of the heart (CKMB), g kidney (creatinine), and h, i liver (ALT and AST) (n = 8 or 10 mice). j Masson staining to assess the fibrosis in organs (n = 4 mice). k Physical function testing in the time spent on rotarod and l forelimb grip (n = 10 mice). m Representative microCT images and quantification of bone volume fraction, trabecular number, trabecular thickness, and trabecular separation (n = 5 mice). n Sectional images of H&E staining and area quantification in epididymal white adipose tissue (eWAT) and inguinal white adipose tissue (iWAT). Oil red staining and quantification of the liver (n = 4 mice). o Glucose tolerance test analysis and quantification by area under the curve (n = 8 mice). p Assessment of insulin levels and q insulin resistance index (n = 8 or 10 mice). Comparisons are performed by Two-sided t tests or one-way ANOVA analysis. *P <  0.05, **P <  0.01, ***P <  0.001. Source data and exact P value are provided as a Source data file.

Fig. 8. GAA alleviates cellular senescence partially mediated by improving ribosome dysfunction.

a KEGG and b GSEA analyses of differential proteins (P < 0.05, fold change >1.5) in the kidney and heart of natural aging model, kidney in IR-induced premature aging model, and natural senescent HUVECs (n = 5 independent experiments). c Heatmaps depicting all detectable proteins of ribosome pathway in the above aging-related models. *GAA treatment significantly regulated the protein compared to the counterparts, with P < 0.05. d Biological processes in GO analysis for differential mRNAs between senescent cells and non-senescent cells. Data were downloaded from Gene Expression Omnibus (GEO) database. e Representative images of the O‐propargyl puromycin (OPP, 10 μM) incorporation experiment in young and senescent HUVECs with GAA treatment or not (n = 3 independent experiments). Fluorescence intensity of OPP to measure ribosomal translation function. f Heatmap of detectable proteins related to ribosome pathway in the kidney of middle-aged mice treated with GAA or vehicle. g Fluorescence intensity of OPP in HUVECs when exposed to CX-5461 (CX, 125 nM) or cycloheximide (CHX, 125 nM) for 5 days and intervened with GAA ion (10 μM) (n = 3 independent experiments). h, i Representative images and quantification of senescence-related markers, including the levels of LDH release, SA-β-Gal, EdU, γH2AX, and nuclear area (n = 3 or 6 independent experiments). j SA-β-Gal staining of cells exposed to CX and CHX with the treatment of GAA. Cells included IMR-90, BMSCs, HK-2, H9C2, and C2C12, and the exposure and intervention time were 3 days, 3 days, 10 days, 6 days, and 5 days, respectively (n = 3 independent experiments). k Fluorescence intensity of OPP in HUVECs after silencing the RPL7 and RPL32 by siRNA and treated by GAA (10 μM) (n = 3 independent experiments). l, m Representative images and quantification of senescence-related markers, including the levels of LDH release, SA-β-Gal, EdU, γH2AX, and nuclear area (n = 3 or 6 independent experiments). Comparisons are performed by Two-sided t tests or one-way ANOVA analysis. All data are expressed as the mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001. Source data and exact P value are provided as a Source data file.

Fig. 9. Identification of TCOF1 as a direct binding protein for GAA.

a Schematic steps for identifying GAA-binding proteins using microarrays fabricated with recombinant human proteins. b Magnified image of HuProt^TM and Bio-Celastrol binding to GAA spot on the protein array. c Computational docking and molecular simulation. d GAA promoted TCOF1 resistance to different temperature gradients by cellular thermal shift assays (CETSA) analysis in senescent HUVECs (n = 3 independent experiments). e Fluorescence intensity of OPP in HUVECs after silencing the TCOF1 by siRNA and treated by GAA (10 μM) (n = 3 independent experiments). f, g Representative images and quantification of senescence-related markers, including the levels of LDH release, SA-β-Gal, EdU, γH2AX, and nuclear area (n = 3 or 6 independent experiments). h The expression level of TCOF1 analyzed by proteomics (n = 6 independent experiments). i, j Phosphorylation detection of TCOF1 in HUVECs (Phos-assay) and aging kidney (LC-MS/MS analysis) (n = 3 independent experiments or mice). k GAA promoted TCOF1 resistance to proteases and phosphatase by drug affinity responsive target stability (DARTS) analysis in senescent HUVECs (n = 3 independent experiments). l Mechanism diagram. GAA interacts with TCOF1, preventing its dephosphorylation induced by phosphatase. This interaction stabilizes pTCOF1, which in turn supports RPs production. Accordingly, GAA ensures proper ribosome biogenesis and subsequent ribosomal function, thereby preventing cellular senescence potentially by alleviating the DNA damage response (DDR), modulating the p53/p21 pathway, and ZAKα kinase activity. Dashed lines indicate previous reports. Comparisons are performed by Two-sided t tests or one-way ANOVA analysis. All data are expressed as the mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001. Source data and exact P value are provided as a Source data file.