Inhibition of the metalloprotease ADAM19 as a novel senomorphic strategy to ameliorate gut permeability and senescence markers by modulating senescence-associated secretory phenotype (SASP)

Increase in the expression of persistent DNA damage and β galactosidase expression in the gut of flies

Several lines of evidence have demonstrated beyond doubt that DNA damage accumulates with age [28]. Increased reactive oxygen species production and a decline in DNA repair capacity have been implicated but not completely understood and have been the focus of research [28]. To gain insight into underlying mechanisms at both molecular and cellular levels, we leveraged Drosophila melanogaster to investigate the molecular effects of exposure to ionizing radiation on intestinal permeability in flies in our previous publication [16].

The 5-day-old adult female w¹¹¹⁸ flies were exposed to ionizing radiation (100 Gy) that is known to induce DNA damage. The cells undergoing DNA double-strand breaks are known to activate the DNA damage response (DDR) [29]. This repair response can be reliably measured by quantifying the histone variant H2AX phosphorylation, producing γH2AX [29]. We performed the immunofluorescence assay in the dissected gut seven days after irradiation and observed a significant increase in cells expressing γH2AX. We also observed a concomitant increase in β-galactosidase expression by co-immunostaining the fly gut following dissection seven days after irradiation (Figure 1A, A-i and A-ii). Interestingly, we also observed a ~7-fold increase in the expression of β-galactosidase and a higher percentage of cells expressing γH2AX (42% of cells were γH2AX-positive compared to 0.4% in the control) in the gut of w¹¹¹⁸ flies in an age-dependent manner, as shown by our immunofluorescence assay in the gut of twenty-day-old flies compared to seven-day-old control flies. (Figure 1B, B-i and B-ii).

Figure 1. Increase in senescence makers in irradiated and aged flies and meltrin knockdown in the enterocytes reduces intestinal permeability and senescence makers in flies following irradiation. (A) Representative immunohistochemistry images of the gut show increased SA-β gal and γ-H2Ax positive cells in X-ray irradiated flies (100Gy) compared to control flies (0Gy). Fluorescence intensity was measured using ImageJ; Scale bar = 20 μm. (A-i). Quantification of fluorescence intensity of SA-β gal (arbitrary units, A.U.) between 0 Gy and 100 Gy. Mean fluorescence intensity was measured using ImageJ and averaged across the gut of 15–16 flies. Data represent mean ± SEM for each group. (A-ii) Quantification of γ-H2AX positive cells expressed as a percentage of total cells in both control and 100 Gy X-ray exposed groups. Data are presented as mean ± SEM from the gut of 15–16 flies. Exposure to 100 Gy of X-ray radiation significantly increased the number of γ-H2AX positive cells compared to the control group, indicating elevated DNA double-strand breaks in response to radiation. (B) Representative immunohistochemistry images of the gut show increased SA-β gal and γ-H2Ax positive cells in X-ray irradiated flies (20 days) compared to control flies (7 days). Fluorescence intensity was measured using ImageJ; Scale bar = 20 μm. (B-i) Quantification of fluorescence intensity of SA-β gal (arbitrary units, A.U.) between 7-day and 20-day-old flies. Mean fluorescence intensity was measured using ImageJ and averaged across the gut of 13–16 flies. Data represent mean ± SEM for each group. (B-ii) Quantification of γ-H2AX positive cells expressed as a percentage of total cells in both 7-day-old and 20-day-old flies. Data are presented as mean ± SEM from the gut of 13-16 flies. The gut of 20 days old flies significantly increased the number of γ-H2AX positive cells compared to the 7-day-old, indicating elevated DNA double-strand breaks in response to radiation. (C-i, C-ii) Smurf assay for assessing gut permeability was performed with 5966-GS>UAS-Meltrin RNAi flies on day ten after irradiation. Control 100Gy R (−) was maintained on 1.5 YE food without RU486, whereas meltrin expression was knocked down in ECs in 100Gy R (+). Results plotted as the mean proportion of Smurf to non-Smurf flies of 4 vials in each group with 25 flies in each vial. Error bars indicate SEM. (D) Immunohistochemistry images of gut showing an increase in SA-β gal and γ-H2Ax positive cells in non-irradiated and X-Ray irradiated in control (RU-) flies and meltrin RNAi and flies (RU+). (D-i, D-ii) Quantification of SA-β gal intensity and γ-H2Ax positive cells in irradiated-non irradiated flies in the gut of RU- and RU+ flies. (A-i, A-ii, B-i, B-ii, C, C-i, D-i and D-ii) Error bars indicate SEM. The significance level was determined by ANOVA or Student’s t-test, denoted by asterisks: ^*p < 0.05, ^**p < 0.01, ^***p < 0.001, and ^****p < 0.0001.

Meltrin knockdown in the gut reduces gut permeability, DNA damage, and β galactosidase expression

Our recent publication reported a GWAS analysis and identified several genes associated with increased intestinal permeability following radiation exposure. Among the candidates identified was meltrin [16]. Interestingly, the quantitative realtime PCR (qRT-PCR) indicated a significant decrease in the expression of meltrin in the gut of flies ten days following the exposure to ionizing radiation (100Gy) (Supplementary Figure 1A).

To investigate if meltrin may influence gut permeability, DNA damage, and senescence markers after irradiation, we used the Gal4-UAS system to knock down meltrin expression in enterocytes (ECs) [30]. RNAi targeting meltrin was expressed using the drug RU486 inducible (RU+), EC-specific 5966-GS driver in 5-day-old adult flies, with vehicle-treated uninduced control flies (RU-) used for comparison. Meltrin knockdown was confirmed by qRT-PCR (Supplementary Figure 1B). The Smurf assay was performed to determine the effect of meltrin as previously described [31]. Briefly, this assay involves feeding flies a blue dye, and in a few hours, if the intestine is damaged and permeable, the dye diffuses to the hemolymph, and the whole fly appears blue. Knocking down meltrin in ECs (RU+) in non-irradiated control flies showed no significant changes in gut permeability upon meltrin knockdown (Figure 1C-i). In contrast, followed by irradiation (100 Gy + RU) resulted in a reduction in intestinal permeability, as we observed in a 2-fold lower percentage of Smurf flies compared to irradiated control flies without RU486 (-RU) exposed to 100 Gy. (Figure 1C-ii).

This metalloprotease inhibitor batimastat (BB-94) can inhibit several memebers of ADAM family [32, 33]. Hence, to phenocopy the effects of its knockdown of ADAM 19 (meltrin) on intestinal permeability after irradiation, we treated w¹¹¹⁸ flies with 50 μM BB-94 in the food. We observed significantly reduced intestinal permeability in w¹¹¹⁸ female flies 14 days after irradiation (Supplementary Figure 1C). As exposure to ionizing radiation is known to cause DNA damage, we tested the effect of knocking down meltrin in ECs on the expression γH2AX and β-galactosidase. As expected, our results show a significant increase in the mean fluorescence intensity of β-galactosidase expression and percentage of cells expressing γH2AX in control flies upon exposure to 100Gy of ionizing radiation RU486 (RU-), whereas knocking down meltrin in ECs (RU+), followed by irradiation (100Gy), significantly reduced β-galactosidase expression, and the percentage of cells expressing γH2AX compared to the no RU486 (RU-) control. At the same time, non-irradiated flies did not show a change in the expression of either γH2AX or β-galactosidase in their gut with or without meltrin in ECs (Figure 1D, 1D-i, and 1D-ii). We also observed an increase in the expression of upd3 in the 7 days post irradiated fly gut, which was significantly reduced with meltrin knock down (Supplementary Figure 1D).

As we reported in our previous publication, damage caused by exposure to ionizing radiation significantly decreases the gut size in adult w¹¹¹⁸ female flies. Presumably, it is because of an increase in the apoptotic clearance of at least three days [16]. Using the same approach of dual acridine orange/ethidium bromide (AO/EB) fluorescent staining to quantify apoptosis-associated changes of cell membranes during the process of apoptosis [16] our results demonstrate a higher number of AO/EB positive cells in the gut of flies upon meltrin knockdown (RU+) on day five after irradiation compared to the control (RU-) (Supplementary Figure 1E). In contrast, apoptotic cells decreased significantly at day 10 (Supplementary Figure 1F).

Exposure to ionizing radiation is known to increase intestinal permeability and reduce survival in w¹¹¹⁸ flies [16]. As ECs-specific knockdown of meltrin reduced intestinal permeability, we investigated if gut-specific knockdown of meltrin may improve survival and observed no effect on the mean or median lifespan of irradiated and non-irradiated flies (Supplementary Figure 1G).

Meltrin/ADAM19 inhibition in the gut reduces gut permeability in mice model

As mentioned earlier, meltrin is a highly conserved gene, and its mammalian homolog is ADAM19 [34, 35]. Interestingly, ADAM19 is known to be elevated in intestinal cells in inflammatory bowel disease (IBD) patients [36]. We wondered if ADAM19/meltrin may also regulate intestinal permeability in mammals, especially in response to the damage caused by genotoxic stress. To test this, we used the well-established chemotherapy model. Doxorubicin (doxo) is a well-known chemotherapy agent that induces DNA damage by inhibiting topoisomerase II [37]. However, as a chemotherapy drug, the side effect of doxo is increased intestinal permeability [38, 39]. As ADAM19 knockout mice die soon after birth due to cardiac defects [40], we used a BB-94 to pharmacologically inhibit its function, testing our hypothesis.

We treated four-month-old C57BL6 mice with or without doxo as a single intraperitoneal (i.p.) injection (doxo) (10 mg/kg) or similarly treated with vehicle controls (Ctrl) described previously [41]. BB-94 was administered with i.p. injection (5 or 20 mg/Kg, once a week for six weeks) (Figure 2A). Intestinal permeability was measured by performing ELISA to detect fecal albumin, frequently used as a proxy for intestinal permeability [42]. We performed an ELISA test on the fecal material collected from the mice six weeks after doxo treatment and detected an increase in the amount of albumin detected in the fecal matter compared to vehicle-treated control mice. However, mice treated with 5 and 20 mg/Kg BB-94 had significantly lower albumin in the fecal matter compared to doxo-treated mice (Figure 2B). There was no statistical difference between mice treated with 5 or 20 mg/Kg BB94 and the amount of albumin detected in the fecal matter collected from mice treated with either (Figure 2B).

Figure 2. ADAM19 inhibition by BB-94 treatment reduces intestinal permeability and inflammation in mice caused by doxorubicin treatment. (A) Experimental design to determine the effect of ADAM19 inhibition in a mouse model. Mice received a single dose, i.p. Dox (10 mg/Kg) or vehicle (SHAM Ctrl) at the start of the experiment. Followed by a subset of mice receiving Batimastat BB-94 doses of 5 mg/Kg and 20 mg/Kg once a week, every other week till week 5. Intestinal permeability and inflammation were measured on week six in the colon of dissected mice. (B) Albumin levels were measured using an ELISA assay in early morning fecal samples from each group. The graph presents albumin levels as mean ± SEM, expressed in mg per 100 mg/ml of fecal material, from (n = 6) across. Significant differences in albumin levels between groups (e.g., Sham vs. doxo, doxo vs. BB-94 treated mice at 5 mg/kg and 20 mg/kg, etc.) were assessed using ANOVA, with p-values indicated as ^*p < 0.05, ^**p < 0.01, ^***p < 0.001. Elevated albumin levels in feces may indicate increased intestinal permeability or gut damage, reflecting potential gastrointestinal dysfunction in doxo-treated mice, while treatment with BB-94 significantly reduced it. (C) qRT-PCR was performed with RNA prepared from dissected colon tissue of mice treated as described above. The results are presented as mean relative fold change in the gene expression of IL-6 and normalized to the housekeeping gene Actin. (D) Mean relative fold change in the gene expression of p21 and normalized to the housekeeping gene Actin. (B–D) Error bars indicate SEM. The significance level was determined by ANOVA and denoted by asterisks: ^*p < 0.05, ^**p < 0.01, ^***p < 0.001, and ^****p < 0.0001.

Doxorubicin treatment is known to increase the inflammatory response in various tissues in mice [43, 44]. We investigated the effect of BB-94 on the inflammation in the gut. Our qRT-PCR results in the colon tissue of the mice confirmed an increase in interleukin-6 (IL-6) expression upon doxorubicin treatment. However, mice that received either concentration BB-94 (5 or 20 mg/Kg) had significantly reduced expression of IL-6. (Figure 2C). Interestingly, IL-6 is the mammalian homolog of Upd3, which also increases in the expression in the gut of flies upon exposure to genotoxic stress. A similar trend was observed in TNFα expression, although it was not statistically significant (p ≤ 0.05) due to the small sample size (Supplementary Figure 2).

The CDK inhibitor p21 regulates p53-mediated growth arrest following DNA damage [45]. Interestingly, our qRT-PCR in doxorubicin-treated mice showed a trend of increased cell cycle checkpoint maker p21 expression in the colon, which was significantly reduced by both doses (5 mg/kg and 10 mg/kg) of BB-94 (Figure 2D).

ADAM19 modulates cellular response to genotoxic stress

In response to genotoxic stressors like doxorubicin or irradiation, mammalian cells that accumulate persistent DNA damage can permanently escape the cell cycle by a phenomenon described as cellular senescence [46]. Senescent cells are also known to secrete a complex mix of pro-inflammatory cytokines, chemokines, growth factors, and matrix-modifying proteins collectively called the SASP [47].

ADAM19 inhibition reduced intestinal permeability and inflammation by reducing the senescence phenotype and proinflammatory upd3 and β-galactosidase in flies and IL-6 and p21 expression in mice colon. We then reasoned if ADAM19 may regulate senescence phenotype. To this end, we tested the effect of ADAM19 knockdown or inhibition in human lung fibroblasts IMR-90 with doxorubicin (250 nMol, a well-established model of cellular senescence. We observed a two-fold increase in the expression of ADAM19 in doxo-treated senescence cells. Using shRNA, we observed a significant decrease in its expression (Supplementary Figure 3A).

The qRT-PCR indicates a significant increase in the expression of pro-inflammatory SASP IL-1α, IL-6, and IL-8, whereas ADAM19 knockdown or its inhibition by BB94 significantly reduced the expression of these SASP factors, with no effect of control shRNA (Figure 3A and Supplementary Figure 3B, 3C).

Figure 3. ADAM19 knockdown and pharmacological inhibition reduced senescence in human fibroblast cells caused by irradiation. (A) The mean fold change in the expression levels of senescence-associated secretory phenotype (SASP) cytokines IL-6, IL-8, and IL-1α was measured by qRT-PCRIMR-90 cells ten days after exposure to 100 Gy X-ray. The expression levels were quantified relative to untreated control samples and normalized to the housekeeping gene actin (ACTB) to control for differences in input RNA and reverse transcription efficiency. Data are presented as mean ± SEM from three independent experiments. The results represent the relative fold change in gene expression compared to control. Increased expression of SASP cytokines, such as IL-6, IL-8, and IL-1α, indicates a senescence response and may reflect the inflammatory environment of irradiation-induced cellular damage, which was significantly reduced in cells with ADAM19 knockdown or inhibition following treatment with BB-94. (B-i) IMR-90 cells were subjected to SA β-Gal staining ten days after irradiation (10Gy), following 10 µM BB-94 treatment. (B-ii) IMR-90 cells were subjected to SA β-Gal staining ten days after irradiation (10Gy), following transduction with either ADAM19 or scrambled shRNA. The percentage of SA β-Gal positive cells was quantified and presented as mean ± SEM, with three independent experiments with two technical replicates per group. (C) MCF-7 cells were treated with conditioned media collected from IMR-90 cells ten days after irradiation with or without ADAM19 shRNA or with or without Batimastat (10 µM). Results plotted as mean % EdU positive cells. (A–C) Error bars indicate SEM. The significance level was determined by ANOVA and denoted by asterisks: ^*p < 0.05, ^**p < 0.01, ^***p < 0.001, and ^****p < 0.0001. In (A, B-i, B-ii and C) “NS” represent non senescence and ‘Sen’ represent senescence.

The lysosomal β-galactosidase activity in senescent cells is well documented and often quantified to measure senescence. We observed an increase 6-fold in the percentage of β-galactosidase-expressing cells in senescent cells (Sen) ten days after doxorubicin treatment. As expected, we observed a significant increase in the expression of SA β-gal in the proportion of cells treated with doxo (10 days after doxo treatment) compared to non-senescence (NS) control cells. While knocking down ADAM19 significantly decreased the cell activity of SA β-gal (from 64% to about 25%). Control shRNA did not affect the β-galactosidase-expressing cells (Figure 3B-i). Similarly, BB-94 treatment (10 mM) significantly lowered the percentage of cells expressing β-galactosidase compared to vehicle-treated Sen cells (from about 70% to about 30%) (Figure 3B-ii).

The expression of Lamin B1 is known to be diminished in senescent cells (Supplementary Figure 3D) [48]. However, ADAM19 knockdown or inhibition did not restore. Also, the loss of HMGB1 from the nucleus of the senescent cells is also used to characterize senescence, and we did not observe any improvement in nuclear localization of HMGB1 in senescent cells with ADAM19 knockdown (Supplementary Figure 3E) [49].

The SASP, secreted by senescent cells, is often attributed to causing systemic inflammation [50]. Chronic exposure to SASP is known to drive paracrine senescence [51]. However, paradoxically, exposure to proinflammatory SASP can create a pro-carcinogenic microenvironment and the development of aging-associated cancers [52]. Previously, cancer cell lines cultured with SASP collected from senescent cells were shown to proliferate at a faster rate. Our results agree with the published literature, as shown by the EdU staining of MCF7 cells grown with culture media (CM) collected from senescent cells proliferated two times faster than CM collected from non-senescent cells. However, the proliferation of MCF7 cells was significantly reduced when grown in CM media collected from senescent cells with ADAM19 knockdown or inhibition as measured by EdU staining performed two days after treatment with the CM collected from senescent or non-senescent IMR-90 cells treated with or without ADAM19 knockdown or inhibition with Batimastat (10 µM), (Figure 3C).

ADAM19 in modulating the SASP secretion from senescent cells

ADAM19 is known to be a shedder and is primarily shown to be involved in ectodomain shedding [20]. For a comprehensive understanding of the effect of ADAM19 on the SASP composition, we performed proteomic analysis using data-independent acquisitions, (DIA) [53, 54] of the CM (conditioned media) collected from senescent or non-senescent IMR-90 cells following ADAM19 knockdown or inhibition with or without Batimastat (Figure 4A). We quantified a total of 472 proteins (Supplementary Table 1). We determined the overlap in proteins that contain the meltrin/ADAM19 cleavage site and fold changes having corrected q-values of 0.05 or lower in comparisons between ADAM19 knockdown vs. control (a19 vs. ctrl: 82 proteins) and ADAM19 irradiated vs. control irradiated (a19-IR vs. IR: 30 proteins) (Figure 4B-i, B-ii). Using the hyperfunction in R to calculate the probability of observing an overlap of 12 proteins between these two groups, we found that this overlap was unlikely to have occurred by chance, given the total number of proteins quantified (p-value of 3.68e-4). To identify a common senescence-specific ADAM19 binding motif among the overlapping proteins with significant fold changes, we performed motif analysis (https://myhits.sib.swiss/), which found a significant signal for a cysteine-rich region (N-score = 9.186, E-value = 0.014). Further, we utilized https://string-db.org/ to identify known connections and perform gene ontology of the 12 overlapping proteins. These proteins are strong interactors with each other, as shown in the protein interaction network (Figure 4C and Supplementary Table 1). We reported the enrichment p-value for the number of interactions and ontology terms with an FDR of 5% or lower. Thus, these 12 SASP proteins are significantly regulated by ADAM19 and potentially targeted through a common cysteine-rich binding motif. In summary, in this experiment, we have identified ADAM19-altered secretory factors in senescence cells. Based on the sequence similarity of ADAM19’s cleavage site using motif analysis, we identified 12 SASP proteins as potentially directly targeted by ADAM19 knockdown.

Figure 4. ADAM19 knockdown or inhibition alters secretory factors from senescent cells. (A) Heat map showing changes in the secretome composition from senescent IMR-90 cells with or without ADAM19 shRNA. (B-i) Venn diagram shows altered protein between ADAM 19 vs. IR and IR only groups. (B-ii) Venn diagram shows altered protein with cleavage motif in groups between ADAM 19 vs. IR and ADAM19 vs. control. (C) Protein network 12 commonly altered protein in groups between ADAM 19 vs. IR and IR.

Fly culture and stocks

Flies were reared on a standard laboratory diet (cornmeal 8.6%, yeast 1.6%, sucrose 5%, agar 0.46%, mixture of propionic acid and orthophosphoric acid 1%). Eclosed adults were transferred to standard food after 14 days of egg laying. For Gene-Switch Gal4 drivers, 100 μM of RU486 (dissolved in 95% ethanol) was used (the media was then referred to as ‘+RU486’). The control diet contained the same volume of 95% ethanol and was referred to as ‘-RU486’. The following fly strains were obtained from Bloomington Drosophila stock center w¹¹¹⁸, meltrin RNAi (58331), and 5961-GS, a kind gift from Dr. Heinrich Jasper.

Radiation exposure

Adult, female, 5-day-old flies were exposed to X-rays at 320 kV and 10 mA for 10 min to achieve 100 Gy and maintained on either +RU486 and -RU486 diets.

Smurf gut permeability assay

We performed the Smurf assay as previously described with slight modifications [31]. Briefly, we introduced 25 flies into an empty vial that held a 2.0 cm × 4.0 cm piece of filter paper. To serve as their food source, we applied 300 µl of a 2.5% blue dye solution (FD&C #1) in a 5% sucrose solution to saturate the paper. After a 24-hour incubation at 25°C, we counted the number of flies displaying visible blue dye outside their intestines to identify as “Smurf” flies.

Immunohistochemistry

Immunohistochemistry was performed post-14 days of X-ray irradiation. Briefly, flies were dissected for gut in PBS and immediately fixed with 4% PFA for 30 minutes. Samples were washed for 10 minutes three times with PBS (Phosphate buffer saline) and then incubated with PT (PBS, 0.3% triton X-100) for 30 minutes. Samples were washed for 10 minutes three times with PBS, and blocking was performed with 5% BSA in PT for 1 hour at room temperature. Samples were incubated with primary antibody overnight at 4°C, washed for 10 minutes three times with PT, and incubated with secondary antibody for 2 hours at room temperature and post-incubation wash three times with PT. Nuclei were stained using DAPI. Samples were mounted with Fluoromount G (Southern Biotech, USA), and images of midgut were captured in a Zeiss 780 confocal microscope. The following antibodies were used in this study: anti-GLB1 (abcam#203749: 1/200), anti-mouse γ-H2Av (DSHB: 1/200), anti-rabbit Alexa fluor 488 (Cell Signaling Technology, USA: 1/500) and anti-mouse Alexa fluor 555 (Cell Signaling Technology: 1/500). These images were analyzed using Image J by counting γ-H2AX positive cells and total intensity of β galactoside within an ROI of the 11583-unit area and normalized with a total number of nuclei within this area. The nucleus of more than two foci was marked as γ-H2AX positive cells.

Bioinformatics analysis

Pathway analysis of Senescence-Associated Secretory Phenotype (SASP) proteins identified through proteomics with and without ADAM19 knockdown (KD) was performed using tools from https://reactome.org/ [78]. Significant proteins with an ADAM19 cleavage site identified through proteomics in each group, ADAM19 vs. control and ADAM19 with IR vs. control with IR, were compared to lists of identified SASP proteins [13, 14]. The SASP proteins of each group were then coded as 1, and non-SASP proteins as 0, and the groups were separately inputted into the Pathway Browser analysis tool, reactome.org, without including interactors.

Cell culture

IMR-90 fetal lung fibroblasts (ATCC, USA: Cat# CCL-186) cultured in IMR-90 cells were cultured in complete media containing Dulbecco’s modified Eagle’s medium (DMEM) (Corning, USA; Cat# 10-013-CV) supplemented with 10% fetal bovine serum (FBS) (Millipore Sigma, USA; Cat# F4135) and 1× penicillin-streptomycin (Corning; Cat# 30-001-CI). The cells were a 5% CO₂ and 3% O₂ incubator.

Knockdown by ADAM19 using lentiviral shRNA

We selected MISSION shRNA Plasmid (SHCLNG-NM_023038, Sigma, USA) to stable knockdown ADAM19. The viral packaging and introduction of lentivirus into host cells were done according to the Addgene pLKO.1 protocol (Addgene, USA). The selection of cells stably expressed ADAM19 shRNA, and the control-shRNA started 72 h post-transfection. The growth medium was replaced with a fresh selection medium containing 10 µg/mL of puromycin. The puromycin-containing medium was refreshed every other day for four days. QRT-PCR was performed to determine the expression of ADAM19.

Senescence induction

IMR-90 cells were treated with 300 nM of doxorubicin Hydrochloride (Millipore Sigma, USA; Cat# 504042) in DMEM complete media for 24 h, and cells were maintained in DMEM complete media by changing fresh media every other day for additional eight days, doxorubicin treated and non-senescent control cells were treated with low serum DMEM (0.2% FBS DMEM media) for one day before harvesting cells for various assays.

Senescence-associated ß-galactosidase staining

Senescent-associated ß-galactosidase activity in IMR-90 cells was assessed using the Senescence Detection Kit (BioVision, USA; Cat# K320), following the manufacturer’s instructions. The percentage of SA-ß-gal-positive cells was counted manually. Four fields were quantified per well (n = 3).

Quantitative real time PCR

Cell pellets of senescent IMR-90 fibroblasts were collected nine days after doxorubicin treatment of IMR-90 cells. The total RNA was isolated using Quick-RNA MiniPrep (Zymo Research, USA; Cat# R1055), following the manufacturer’s protocol. cDNA was synthesized using the High-Capacity cDNA RT kit (Life Technologies, USA). Transcripts were analyzed using a Roche LightCycler 480 II in 384-well plates and the UPL probe system. Bioline SensiFast Probe No-ROX was used as a master mix. Each gene’s mean cycle threshold (Ct) was normalized to tubulin levels in the same sample (delta Ct). Unpaired two-sample t-tests determined differences in mean delta Ct values between treatment groups. The delta-delta calculated the fold change. Ct method (fold = 2ddCt).

Mouse model

C56BL/6 mice were purchased from The Jackson Laboratory and housed under controlled temperature (22–24°C), humidity (40–60%), and a 12-hour light-dark cycle at the Buck Institute. All animal procedures were approved by the Buck Institute Institutional Animal Care and Use Committee (IACUC). The six-month-old male mice received a single intraperitoneal (i.p.) doxorubicin injection (10 mg/kg) to induce widespread senescence. A sub-set of mice received BB-94 i.p. injection (5 mg/Kg or 20 mg/Kg) once a week till the completion (6 weeks) starting 48 hours after Dox injection. After six weeks, the animals were euthanized by CO₂ inhalation followed by cervical dislocation, and tissues were harvested and flash-frozen in liquid nitrogen.

Albumin ELISA

The albumin concentration in freshly collected fecal pellets from mice was determined by a standard ELISA test using the manufacturer’s instructions (Bethyl Laboratories, USA #E99-134). Samples were collated from single-housed mice in cages without bedding for several hours to collect fecal pellets from individual mice. The fecal pellets were re-suspended in 1xPBS for ELISA.

Collection of secreted soluble proteins for mass spectrometry

Proteins secreted into a serum-free medium over a 24-hr period were collected. The conditioned medium was centrifuged at 10,000 × g at 4°C for 30 minutes to remove debris. The supernatant containing soluble secreted was saved and processed for mass spectrometry.

Proteomic sample preparation

Chemicals

Acetonitrile (#AH015) and water (#AH365) were from Burdick and Jackson. Iodoacetamide (IAA, #I1149), dithiothreitol (DTT, #D9779), formic acid (FA, #94318-50 ML-F), and triethylammonium bicarbonate buffer 1.0 M, pH 8.5 (#T7408) were from Sigma Aldrich, urea (#29700) was from Thermo Fisher Scientific, sequencing grade trypsin (#V5113) was from Promega and HLB Oasis SPE cartridges (#186003908) were from Waters.

Protein concentration and quantification

Samples were concentrated using Amicon Ultra-15 Centrifugal Filter Units with a three-kDa cutoff (Millipore Sigma #UFC900324) as per the manufacturer’s instructions and transferred into 8 M urea/50 mM triethylammonium bicarbonate buffer at pH 8. Protein quantitation was performed using a BCA Protein Assay Kit (Pierce #23225).

Digestion

Aliquots of each sample containing 25–100 µg protein were brought to equal volumes with 50 mM triethylammonium bicarbonate buffer at pH 8. The mixtures were reduced with 20 mM DTT (37°C for 1 hour), then alkylated with 40 mM iodoacetamide (30 minutes at RT in the dark). Samples were diluted 10-fold with 50 mM triethylammonium bicarbonate buffer at pH eight and incubated overnight at 37°C with sequencing grade trypsin (Promega) at a 1:50 enzyme: substrate ratio (wt/wt).

Desalting

Peptide supernatants were collected and desalted with Oasis HLB 30 mg Sorbent Cartridges (Waters, USA; #186003908), concentrated, and re-suspended in a solution containing mass spectrometric ‘Hyper Reaction Monitoring’ retention time peptide standards (HRM, Biognosys, USA; #Kit-3003) and 0.2% formic acid in water.

Mass spectrometry analysis

Samples were analyzed by reverse-phase HPLC-ESI-MS/MS using the Eksigent Ultra Plus nano-LC 2D HPLC system (USA) combined with a cHiPLC system directly connected to an orthogonal quadrupole time-of-flight SCIEX TripleTOF 6600 or a TripleTOF 5600 mass spectrometer (SCIEX, USA). Typically, mass resolution in precursor scans was ~45,000 (TripleTOF 6600), while fragment ion resolution was ~15,000 in ‘high sensitivity’ product ion scan mode. After injection, peptide mixtures were transferred onto a C18 pre-column chip (200 µm × 6 mm ChromXP C18-CL chip, 3 µm, 300 Å, SCIEX) and washed at 2 µl/min for 10 min with the loading solvent (H₂O/0.1% formic acid) for desalting. Peptides were transferred to the 75 µm × 15 cm ChromXP C18-CL chip, 3 µm, 300 Å, (SCIEX), and eluted at 300 nL/min with a three hours gradient using aqueous and acetonitrile solvent buffers.

All samples were analyzed using data-independent acquisitions (DIA), specifically variable window DIA acquisitions [53, 79, 80]. In these DIA acquisitions, windows of variable width (5 to 90 m/z) are passed in incremental steps over the full mass range (m/z 400–1250). The cycle time of 3.2 sec includes a 250 msec precursor ion scan followed by 45 msec accumulation time for each of the 64 DIA segments. The variable windows were determined according to the complexity of the typical MS1 ion current observed within a certain m/z range using a SCIEX ‘variable window calculator’ algorithm (more narrow windows were chosen in ‘busy’ m/z ranges, wide windows in m/z ranges with few eluting precursor ions). DIA tandem mass spectra produce complex MS/MS spectra, a composite of all the analytes within each selected Q1 m/z window. All collected data were processed in Spectronaut using a pan-human library that provides quantitative DIA assays for ~10,000 human proteins.

Processing, quantification, and statistical analysis of MS data

DIA acquisitions were quantitatively processed using the proprietary Spectronaut v12 (12.020491.3.1543) software from Biognosys. A pan-human spectral library was used for Spectronaut processing of the DIA data. Quantitative DIA MS2 data analysis was based on extracted ion chromatograms (XICs) of 6–10 of the most abundant fragment ions in the identified spectra. Relative quantification was performed by comparing different conditions (senescent versus control) to assess fold changes. To account for variation in cell number between experimental groups and biological replicates, quantitative analysis of each sample was normalized to cell number by applying a correction factor in the Spectronaut settings. There were three replicates for each experimental condition: X-irradiated senescent fibroblasts, mock-irradiated non-senescent fibroblasts, ADAM19 shRNA-treated senescent fibroblasts, BB94-treated senescent fibroblasts, ADAM19 shRNA-treated non-senescent fibroblasts. Significance was assessed using FDR-corrected q-values < 0.05.

Data visualization

Heatmaps were visualized in R using the heatmap2 function of the ‘gplots’ package [81].

Data availability

All raw files are uploaded to the Center for Computational Mass Spectrometry, MassIVE, and can be downloaded using the following link https://massive.ucsd.edu/ProteoSAFe/dataset.jsp?task=a8341a1bcba442e7ae151e15919ac92b (MassIVE ID number: MSV000085224). Data uploads include the protein identification and quantification details, spectral library, and FASTA file used for mass spectrometric analysis. MassIVE: MSV000085224 (https://massive.ucsd.edu/ProteoSAFe/dataset.jsp?task=a8341a1bcba442e7ae151e15919ac92b).