Suppression of the senescence-associated secretory phenotype (SASP) in human fibroblasts using small molecule inhibitors of p38 MAP kinase and MK2

Abstract

Senescent cells show an altered secretome profile termed the senescence-associated secretory phenotype (SASP). There is an increasing body of evidence that suggests that the accumulation of SASP-positive senescent cells in humans is partially causal in the observed shift to a low-level pro-inflammatory state in aged individuals. This in turn suggests the SASP as a possible therapeutic target to ameliorate inflammatory conditions in the elderly, and thus a better understanding of the signalling pathways underlying the SASP are required. Prior studies using the early generation p38 MAPK inhibitor SB203580 indicated that p38 signalling was required for the SASP. In this study, we extend these observations using two next-generation p38 inhibitors (UR-13756 and BIRB 796) that have markedly improved selectivity and specificity compared to SB203580, to strengthen the evidence that the SASP is p38-dependent in human fibroblasts. BIRB 796 has an efficacy and toxicity profile that has allowed it to reach Phase III clinical trials, suggesting its possible use to suppress the SASP in vivo. We also demonstrate for the first time a requirement for signalling through the p38 downstream MK2 kinase in the regulation of the SASP using two MK2 inhibitors. Finally, we demonstrate that a commercially-available multiplex cytokine assay technology can be used to detect SASP components in the conditioned medium of cultured fibroblasts from both young and elderly donors. This assay is a high-throughput, multiplex microtitre-based assay system that is highly sensitive, with very low sample requirements, allowing it to be used for low-volume human biological fluids. Our initial studies using existing multiplex plates form the basis for a "SASP signature" assay that could be used as a high-throughput system in a clinical study setting. Our findings therefore provide important steps towards the study of, and intervention in, the SASP in human ageing and age-related disease.

Keywords: BIRB 796; Cellular senescence; Human ageing; Human fibroblasts; MK2.III; SB203580.

Fig. 1

Cytokine levels versus cumulative population doublings (CPD) for AG16409 and AG11081 cells. a Plot of cytokine levels versus CPD for cell strains AG16409 (top panels) and AG11081 (middle panels) and each analyte. Data is plotted as pg analyte/ml conditioned medium (normalised to 30,000 cells) versus CPD achieved. b The data from a is calculated for each PD point as a percentage of the maximum level seen in the senescent cells for each analyte. Then the average is calculated for all eight cytokines and plotted against the CPD. The data is shown as the average % ± the SE of the means of each individual analyte

Fig. 2

Bar chart showing cytokine levels in pg/ml medium for pre-senescent and senescent AG16409 and AG11081 cells. The protein levels are normalised to 30,000 cells (see Table S1 for cell CPD details)

Fig. 3

Effects of p38 and MK2 inhibitors on IL-6 expression. Levels of IL-6 in pg/ml conditioned medium (normalised to 30,000 cells) for a AG07719A cells, b AG16409 and AG08433 cells. c Data from a and b expressed as a percentage of IL-6 at senescence for each strain. d Average IL-6 levels shown as a percentage of the level seen at senescence for AG16409 cells using four biological replicates. Only one inhibitor concentration has been used. Values are mean ± SD, p values are compared to the levels seen at senescence; Student t Test. Prolif and sen refer to cells at low CPD and senescence respectively (see Table S2 for cell CPD details). The rest of the data refer to inhibitor-treated senescent cells. Key to inhibitors on x axes: SB = SB203580, UR = UR-13756, B = BIRB 796, PF = PF-3644022, MK = MK2.III; the numerical values are µM. ND is not determined