Fig 2: Deficiency of progranulin leads to ganglioside accumulation in mouse and human brain tissues.a Ganglioside degradation pathway in the lysosome. The names of glycosyl hydrolases (green), activator proteins (purple), and associated metabolic diseases (red) are indicated in the scheme. b Quantification of mono-sialyated and di-sialyated ganglioside species isolated from Grn+/+ (gray) (n = 6), Grn+/R493X (blue) (n = 4), and GrnR493X/R493X (purple) (n = 4) mouse brains. c Quantification of mono-sialyated and di-sialyated ganglioside species isolated from the frontal lobes of control (pink), FTD-TDP43-A (sporadic-non-GRN) (green), and FTD-TDP43-A (GRN) (blue) human brains. Box plots display mean ± the minimum and maximum number in the data set of control (n = 3), FTD-TDP43-A (sporadic-non-GRN) (n = 6) or FTD-TDP43-A (GRN) (n = 12). Box plots display mean ± the minimum and maximum number. One-way ANOVA, followed by multigroup comparison (Dunn’s) test, was performed. *p < 0.05, **p < 0.01. G is for ganglioside; M/D/T are for monosialic, disialic, or trisialic; and the number refers to the order of discovery. GM2-AP GM2 ganglioside activator protein, SAP-B/C/D saposin-B/C/D, GLB1 galactosidase beta 1, HEXA beta-hexosaminidase subunit alpha, NEU3/4 neuraminidase 3/4, GALC galactosylceramidase, GCase glucosylceramidase beta, ASAH1 N-acylsphingosine amidohydrolase 1, SAP-C/D saposin-C/D.

Fig 3: Galactose-modified duocarmycin prodrugs preferentially target senescent cells with high SA-ß-galactosidase activity. (a) Representative pictures of fluorescent SA-ß-galactosidase staining in IMR90 ER:RAS cells treated with prodrug A or vehicle. Scale bar, 100 µm. (b) Single-cell intensities value for DDAO galactoside in a representative well of a 96-well plate seeded with IMR90 ER:RAS cells treated with DMSO or 4OHT. Grey dotted line indicates quantification cut-off. Cells were considered positives for SA-ß-gal when their cell intensity was > 165. (c) Quantification of SA-ß-galactosidase activity in IMR90 ER:RAS cells treated with prodrug A or vehicle. Statistical significance was calculated using two-way ANOVA. (d-g) The senolytic properties of prodrug A depend on the lysosomal ß-galactosidase (GLB1). (d) Experimental set-up. (e) Representative pictures of cytochemical SA-ß-Gal staining in IMR90 ER:RAS infected with different shRNAs against GLB1 or an empty vector and (f) quantification (n = 3). Statistical significance was calculated using one-way ANOVA. (g) Quantification of cell survival of senescent and control IMR90 ER:RAS infected with different shRNAs targeting GLB1 or an empty vector and treated with ABT-263 or prodrug A for 3 days (n = 3). Statistical significance was calculated using two-tailed, Student's t test. All error bars represent mean ± SD; n represents independent experiments; ns, not significant; *p < .05; **p < .01; ***p < .001, ****p < .0001

Fig 4: Establishment of natural aging animal model. Wild-type C57 mice were divided into young and old groups. (A) Photograph of mice. (B) Body weight (g). (C) Expression of GLB1, p53, p21, and p16 detected through western blotting; GAPDH: internal reference. (D) Relative protein expression levels of GLB1, p53, p21, and p16. N = 6; ** P < 0.01 and *** P < 0.001.