Senescence-associated hyper-activation to inflammatory stimuli in vitro

SnCs exhibit a senescence-associated hyper-activation phenotype in response to inflammatory stimulation

While it is well known that SnCs are pro-inflammatory in nature by virtue of expression of the SASP [22, 27], whether inflammatory stimulus could further significantly exacerbate their pro-inflammatory phenotype has not been studied yet. Endothelial cells being a common cell type spread throughout the body in the form of a lining layer of the blood vessels, we decided to use human umbilical vein endothelial cells (HUVEC) to examine this prospect. We also know that HUVEC are inherently responsive to various inflammatory stimuli [28]. To determine if an inflammatory stimulus could significantly exacerbate the pro-inflammatory phenotype of senescent HUVEC compared to their normal counterparts, we examined the transcriptional response of non-senescent (NC HUVEC) and ionizing radiation (IR)-induced senescent HUVEC (IR HUVEC) to lipopolysaccharide (LPS). Using the methods reported by us previously [29], senescence was induced in HUVEC by exposure to ionizing radiation or serial passaging. Induction of senescence in HUVEC by these methods were evidenced by the permanent cell cycle arrest measured by EdU staining (Supplementary Figure 1A), elevated expression of senescence-associated beta galactosidase (SA-β-Gal) activity (Supplementary Figure 1B), increased expression of CDKN2A and CDKN1A mRNA (Supplementary Figure 1C, 1D) and several SASP factors at basal conditions (Figure 1A–1F and Supplementary Figure 2).

Upon analyzing the transcriptional response of NC HUVEC and IR HUVEC to LPS stimulation, we observed that both types of cells showed a dose-dependent upregulation of mRNA expression for several cytokines and chemokines such as IL6, CCL2, CXCL1, CCL5, IL1β and TNFα (Figure 1A–1F). Moreover, at any given dose, IR HUVEC showed a significantly higher mRNA expression of these inflammatory mediators than their NC HUVEC counterparts (Figure 1A–1F).

Next, we examined the time-dependent dynamics of the mRNA expression of these inflammatory mediators upon LPS stimulation. Much like the dose-dependent response, both NC HUVEC and IR HUVEC exhibited a time-dependent response to LPS stimulation. Again, IR HUVEC expressed significantly higher levels of mRNA for the analyzed cytokines and chemokines, at most given time points, relative to NC HUVEC (Figure 1G–1L).

To verify if this exacerbated response of IR HUVEC was specific to LPS, we examined the response of NC and IR HUVEC to IL1β and TNFα, known inflammatory stimulants [28]. As observed with response to LPS, stimulation by IL1β and TNFα elicited a strong transcriptional activation of IL6, CCL2, CXCL1, CCL5, IL1β and TNFα mRNA expression in both NC HUVEC and IR HUVEC. However, stimulated IR HUVEC expressed significantly higher levels of mRNA for all tested cytokines and chemokines when compared to their NC counterparts (Figure 2A–2L). These results suggest that IR HUVEC are hyper-reactive to inflammatory stimulus, which we term senescence-associated hyper-activation.

To examine if this phenomenon was exclusive for IR HUVEC, we generated replicative senescent HUVEC (Rep-Sen HUVEC) and tested their response to LPS, IL1β and TNFα. Rep-Sen HUVEC, similar to IR HUVEC, showed exacerbated transcriptional activation of IL6, CXCL10 and CCL5 upon inflammatory stimulation (Supplementary Figure 2A–2I).

To explore whether senescence-associated hyper-activation is a general characteristic of SnCs, rather than being specific to endothelial cells, studies were extended to IR-induced senescent human adipose derived stem cells (ASCs) (Supplementary Figure 3A–3I), renal epithelial cells (RECs) (Supplementary Figure 4A–4I) and WI38 lung fibroblast (WI38) (Supplementary Figure 5A–5I). SnCs from all three cell types exhibited a higher basal level of IL6, CCL5 and CXCL10 mRNA expression as well as higher expression of these inflammatory mediators in response to LPS, IL1β and TNFα stimulation than non-SnCs with a few exceptions in which some of the cells were not very responsive to LPS stimulation. For example, non-senescent WI38 fibroblasts showed no significant change in expression upon LPS stimulation for any of the three genes analyzed, whereas senescent WI38 fibroblasts showed a significant upregulation of mRNA for CCL5, but not for IL6 and CXCL1 upon LPS stimulation.

Cumulatively, this data suggests that SnCs exhibit a senescence-associated hyper-activation phenotype upon being stimulated with a prominent inflammatory stimulant.

SnCs secrete high levels of inflammatory cytokines and chemokines

To investigate whether the increased mRNA levels for the multiple inflammatory mediators in SnCs translate into an elevated secretion of the corresponding factors, we analyzed the conditioned media from NC HUVEC and IR HUVEC with or without LPS stimulation. From the data presented in Figure 3 and Supplementary Table 1, it is evident that IR HUVEC secreted significantly higher levels of multiple cytokines and chemokines relative to NC HUVEC under basal condition. More importantly, when stimulated with LPS, IR HUVEC produced much greater levels of these factors than LPS-stimulated NC HUVEC (Figure 3 and Supplementary Table 1), confirming that SnCs indeed have a senescence-associated hyper-activation phenotype.

Senescence-associated hyper-activation is not associated with increased expression of the Toll-like receptor4 (TLR4)

To investigate whether senescence-associated hyper-activation is due to an increased expression of surface receptors for the inflammatory stimulants, we quantified the expression of TLR4 (a primary receptor for LPS [30]), IL1 receptor 1 (IL1R1), IL1R2 and TNF receptor 1 (TNFR1) in NC and IR HUVECs using western blotting (Figure 4). We found that both cells expressed similar levels of TLR4 (Figure 4B), IL1R1 (Figure 4C) and IL1R2 (Figure 4D), while IR HUVECs expressed a slightly higher level of TNFR1 than NC HUVECs (Figure 4E). This finding suggests that the senescence-associated hyper-activation in responses to LPS and IL1β stimulation is not due to an increased expression of surface receptors but possibly due to the changes in their downstream pathways. However, increased expression of TNFR1 may partially contribute to the senescence-associated hyper-activation to TNFα.

Senescence-associated hyper-activation is mediated by the activation of p38

We hypothesized that senescence-associated hyper-activation is attributable to the hyper-activity of some of the intracellular signaling pathways responsible for the SASP phenotype. To test this hypothesis, we examined if the p38 pathway becomes hyper-activated in SnCs upon inflammatory stimulation. Our results not only corroborated previous demonstrations of SnCs having an elevated basal p38 activation [26], but also showed that IR HUVEC exhibited a significantly higher activation of p38 upon IPS stimulation compared to NC HUVEC (Figure 5A, 5B).

To determine whether senescence-associated hyper-activation effect was dependent on the activation of p38, NC HUVEC and IR HUVEC were pre-treated with losmapimod, a specific p38 inhibitor [31], prior to LPS stimulation. Our analysis showed that the upregulation of IL6, TNFα, CCL5, and IL1β mRNA expressions in IR HUVEC with or with LIP stimulation were abrogated or significantly reduced by the losmapimod pretreatment, demonstrating that activation of p38 mediates not only the expression of SASP at the basal conditions but also that of senescence-associated hyper-activation in response to LPS stimulation (Figure 5C–5F).

Senescence-associated hyper-activation is also mediated by the activation of the NF-κB pathway

Based on the study by Freund et al., showing that p38-MAPK acts upstream of the NF-κB pathway to induce SASP, we hypothesized that NF-κB activation also mediates senescence-associated hyper-activation. To test this hypothesis, we analyzed NF-κB p65 levels in the cytoplasm and nucleus of NC HUVEC and IR HUVEC at baseline and upon stimulation with LPS, IL1β and TNFα, by immunocytochemistry (Supplementary Figure 6) and by immunoblotting (Figure 6A, 6B). Both results showed that IR HUVEC exhibited a significantly higher baseline level of nuclear NF-κB, as well as a significantly higher nuclear translocation of NF-κB upon LPS stimulation than NC HUVEC (Figure 6A, 6B and Supplementary Figure 6). Similarly, IR HUVEC showed a greater translocation of NF-κB p65 into the nucleus upon stimulation with IL1β and TNFα than NC HUVEC (Supplementary Figure 6). Collectively, these results suggest that IR HUVEC respond to inflammatory stimulation by hyper-activation of the NF-κB pathway.

To validate the dependence of senescence-associated hyper-activation on NF-κB activity, NC HUVEC and IR HUVEC were treated with BMS-345541, a potent inhibitor of NF-κB activation [32], prior to LPS stimulation. Our analysis revealed that the upregulation of IL6, TNFα, CCL5 and IL1β mRNA expressions in IR HUVEC with or without LPS stimulation were abrogated or significantly reduced by the pretreatment with BMS-345541 (Figure 6C–6F), confirming that the NF-κB pathway plays an important role in the induction of both SASP and senescence-associated hyper-activation in IR HUVEC.

Cell culture

Human umbilical vein endothelial cells (HUVEC, Cat. No. PCS-100-010), human WI38 fibroblasts (WI38, Cat. No. CCL-75) and human renal epithelial cells (REC, Cat. No. PCS-400-012) were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). Adipose stem/stromal cells (ASC) were previously isolated from adipose tissue (female donors) obtained during elective liposuction procedure (Traktuev et al., 2008). HUVEC were cultured in EBM-2 (Cat. No. CC-3156, Lonza, Basel, Switzerland) media supplemented with EGM-2 SingleQuots (Cat. No. CC-4176, Lonza, Basel, Switzerland). WI38 cells were cultured in DMEM medium (Cat. No. 12430054, Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 10% fetal bovine serum (FBS, Cat. No. 89510-188, VWR, Radnor, PA, USA). REC were cultured in REBM medium (CC-3191, Lonza, Basel, Switzerland) supplemented with REGM SingleQuots (CC-4127, Lonza, Basel, Switzerland). ASC were cultured in EBM-2 (Cat. No. CC-3156, Lonza, Basel, Switzerland) media supplemented with EGM-2MV SingleQuots (Cat. No. CC-4147, Lonza, Basel, Switzerland). To prevent contamination all culture media were supplemented with 100 U/mL penicillin and 100 μg/mL streptomycin (Cat. No. 15140122, Thermo Fisher Scientific, Waltham, MA, USA). Cells were cultured in a humidified incubator at 37° C with 5% CO₂.

Senescence induction

Cells of early passages (HUVEC < 10 passages; WI-38 < 25 passages; REC < 10 passages; and ASC < 4 passages) were considered to be non-senescent cells (NC). Two methods of senescence induction - replicative exhaustion and ionizing radiation, were used as previously described [29]. In short, to generate replicative senescent cells (SnCs), HUVEC were passaged until they stopped replicating further. To generate IR-induced SnCs, all cell types were exposed to 20 Gy of X-rays, followed by a culturing for 10 days. The induction of cellular senescence was validated by analyzing the expression of CDKN2A and CDKN1A mRNA, positive of SA-β-Gal, and lack of cell proliferation. SA-β-Gal was tested using SA-β-Gal staining kit (Cat. No. 9860, Cell Signaling Technologies, Danvers, MA, USA), and cell proliferation was assessed using EdU cell proliferation kit (Cat. No. C10337, Invitrogen, Carlsbad, CA, USA), following the manufacturers’ instructions.

Inflammatory stimulation

Various concentrations of LPS (Cat. No. L4391, Sigma-Aldrich, St. Louis, MO, USA), IL1β (Cat. No. 200-01B, Peprotech, Rocky Hill, NJ, USA), and TNFα (Cat. No. 300-01A, Peprotech, Rocky Hill, NJ, USA) were used to stimulate cells for different durations of time at the indicated concentrations and times presented in each figure legend.

Analysis of the production of various inflammatory cytokines and chemokines

NC HUVEC and IR HUVEC were cultured with medium alone or with 30 ng/mL LPS for 24 hours. The conditioned media from the cultures was then harvested and subjected to analysis using Human Cytokine Array/Chemokine Array 71-Plex Panel by the Eve Technologies Corporation (Calgary, AB, Canada). Values are listed in Supplementary Table 1 (Supplementary Table 1).

Quantitative polymerase chain reaction (qPCR)

RNA isolated from cells using RNeasy mini kit (Cat. No. 74106, Qiagen, Hilden, Germany) was converted into cDNA using a high capacity cDNA reverse transcription kit (Cat. No. 4368813, Applied Biosystems, Foster City, CA, USA). Gene expression was then quantified using gene specific primers (Supplementary Table 2) and fast SYBR green master-mix (Cat. No. 4385617, Applied Biosystems. Foster City, CA, USA) as per the manufacturer’s instructions. The expression of GAPDH was used for normalization. Level of gene expression in untreated NC cells was used as baseline and fold change in gene expression was defined based on ΔΔCT method.

Pathway inhibitory studies

NC HUVEC and IR HUVEC were pre-treated with 1μM of losmapimod (Cat. No. HY10402, Med Chem Express, New Jersey, USA) or BMS-345541 (Cat. No. HY-10519, Med Chem Express, New Jersey, USA) for 2 hours before the cells were stimulated with 30 ng/mL LPS for 3 hours before being harvested for RNA isolation for qPCR.

Whole protein isolation and sample preparation for immunoblot analysis

Control and treated cells were washed with 1X PBS before RIPA buffer (Cat. No. BP-115, Boston Bioproducts, Ashland, MA, USA) was added to the dish. The resultant lysate was then frozen at -80° C for later use. These protein lysates were then thawed and quantified using Pierce BCA protein assay kit (Cat. No. 23225, Thermo Fisher Scientific, Waltham, MA, USA) and normalized before adding 4X Laemmli loading buffer (Cat. No. BP-110R, Boston Bioproducts, Ashland, MA, USA). Samples were then heated to 95° C for 10 minutes before using for western blotting as we previously reported [29].

Cell fractionation and protein isolation

Cells were trypsinised and harvested as pellets after various treatments. They were lysed and then fractionated to obtain cytoplasmic and nuclear proteins for immunoblot analysis using the NE-PER nuclear and cytoplasmic extraction kit (Cat. No. 78833, Thermo Fisher Scientific, Waltham, MA, USA) following the manufacturer’s instructions.

Western blotting

Western blotting was done as previously described [29]. In short, 10 μL of each prepared protein sample was used to run western blot using 15-well 4-20% Mini-PROTEAN TGX Precast Protein Gels (Cat. No. 4561096, Bio-Rad Laboratories, Hercules, CA, USA). The samples were run at 200V for 45 minutes before transferring onto a PVDF membrane using the Trans-Blot turbo transfer system (Bio-Rad Laboratories, Hercules, CA, USA). The PVDF membranes were blocked with 5% milk for 1 hour at room temperature before incubating with primary antibody overnight at 4° C. The membranes were then washed and incubated with horse radish peroxidase (HRP)-conjugated secondary antibody for 1 hour at room temperature. Protein-antibody complexes were revealed with HRP substrate on ChemiDoc imaging system (Bio-Rad Laboratories, Hercules, CA, USA). Antibodies used are listed in Supplementary Table 3 (Supplementary Table 3).

Statistical analysis

Data are expressed as mean ± SEM, unless mentioned otherwise. When comparing more than two groups, data sets were analyzed using repeated measures analysis of variance (ANOVA) on Graphpad Prism (San Diego, CA, USA). Post hoc comparisons were performed between group means using Sidak multiple comparisons test. To compare two groups, Student's t test was used. p < 0.05 was considered significant.