Epigallocatechin Gallate Effectively Affects Senescence and Anti-SASP via SIRT3 in 3T3-L1 Preadipocytes in Comparison with Other Bioactive Substances

Abstract

Aim: We investigated different bioactive compounds including epigallocatechin gallate (EGCG), anthocyanidin, resveratrol, phloretin, spermidine, butyrate, and β-hydroxybutyrate with regard to their effect on SIRT3 via NRF2 and modulation of the proinflammatory senescence-associated secretory phenotype (SASP) in senescence induced 3T3-L1 preadipocytes.

Methods: For induction of senescence, 3T3-L1 preadipocytes were incubated with bromodeoxyuridine (BrdU) for 8 days. Cell cycle inhibition was observed, and β-galactosidase activity was measured. After BrdU treatment, cells were treated with different bioactive compounds in various concentrations for 96 h. ELISA was used for determining proinflammatory cytokine IL6 in SASP cells.

Results: CDKN1a increased significantly after BrdU incubation compared to untreated control (p < 0.01). All secondary plant ingredients used for treatment, but not anthocyanidin 50 μM, decrease CDKN1a expression (p < 0.05), whereas most endogenous substances did not attenuate CDKN1a. IL6 secretion positively correlated with CDKN1a (p < 0.01), whereas EGCG could diminish both, IL6 and CDKN1a with the strongest effect (p < 0.01). Although NRF2 positively correlated with SIRT3 activation (p < 0.05), only resveratrol (p < 0.01) and anthocyanidin (p < 0.05) could activate NRF2 significantly. Solely anthocyanidin 50 μM (p < 0.05) and 100 μM (p < 0.01) and EGCG 50 μM (p < 0.01) could increase SIRT3 expression. Activation of SIRT3 with EGCG correlated with lowered IL6 secretion significantly (p < 0.05) but not with anthocyanidin.

Conclusion: Accumulation of senescent cells in adipose tissue plays an important role in obesity and age-related diseases. SIRT3, located in the mitochondria, can regulate ROS via different pathways. Thus, targeting SIRT3 activating compounds such as EGCG may delay senescence of cells and senescence induced inflammatory processes.

Figure 1

The experimental design outlined per day. Timeline for cell plating and senescence induction and incubation with different substances are addressed in this figure.

Figure 2

Beta galactosidase activity staining (blue) in 3T3-L1 preadipocytes. Untreated confluent 3T3-L1 preadipocytes show minor amount of senescence cells (a). After BrdU treatment for 8 days, cells show a typical senescence phenotype and increased beta galactosidase activity, which is indicated by blue staining (b). After induction of senescence with BrdU for 8 days, following a treatment with roxithromycin 100 μM for 96 h beta gallactosidase activity decreased resulting in significant less blue stained cells (c). EGCG 100 μM treatment for 96 h after 8days of BrdU treatment could decrease β-gal activity but not in the extend as roxithromycin (d).

Figure 3

Percentage of β-galactosidase activity of all substances and concentrations in response to BrdU control. Beta-gal activity was significantly enhanced in BrdU cells compared to untreated cells (p < 0.01). Anthocyanidin, EGCG, resveratrol, EGCG-resveratrol-spermidine mix, BHB, and roxithromycin could change senescence phenotype and diminish beta-gal activity significantly (^∗p < 0.05; ^∗∗p < 0.01). The results were expressed as mean ± SD. Statistical significance between compounds and concentrations to the control was determined by one-way ANOVA with Dunnett's post hoc test.

Figure 4

RQ value of CDKN1a mRNA expression. Cells incubated with BrdU showed an increase in gene expression, thus cell cycle arrest. Subsequent treatment with different substances showed a decrease of CDKN1a expression with secondary plant ingredients (a), but for endogenous substances this result could only be reached for 4 mM BHB (b). Statistical significance was defined ^∗p < 0.05; ^∗∗p < 0.01; ^∗∗∗∗p < 0.001. The results were expressed as mean ± SD. Statistical significance between compounds and concentrations to the control was determined by one-way ANOVA with Dunnett's post hoc test.

Figure 5

IL6 levels pg/ml measured with ELISA in cells treated with BrdU and different substances, using the highest concentration. Compared to untreated cells IL6 secretion increased significantly (p = 0.002). Regarding BrdU control, most secondary plant ingredients besides resveratrol could decrease IL6 secretion significantly. Statistical significance was defined ^∗p < 0.05; ^∗∗p < 0.002. The results were expressed as mean ± SD. Statistical significance between compounds and concentrations to the control was determined by one-way ANOVA with Dunnett's post hoc test (a). Pearson's correlation analysis showed a significant correlation of CDKN1a expression with IL6 secretion including all substances (^∗∗p < 0.002) (b).

Figure 6

Pearson's correlation between CDKN1a and NRF2. Cell cycle arrest thus increase in CDKN1a positively correlates with NRF2 expression (p < 0.01) (a). Not all secondary plant ingredients are NRF2 activators. Relative quantification values of NRF2 comparing all treatments regarding BrdU control (b+c). Anthocyanidin 50 μM, resveratrol 15 μM, and both roxithromycin concentrations increased NRF2 antioxidative defense pathway significantly (b). EGCG, resveratrol, spermidine mix 30 μM, both phloretin concentrations, but also endogenous substances, like spermidine and butyrate diminished NRF2 assuming lower ROS (b+c). Statistical significance was defined ^∗p < 0.05; ^∗∗p < 0.01; ^∗∗∗p < 0.005; ^∗∗∗∗p < 0.001. The results were expressed as mean ± SD. Statistical significance between compounds and concentrations to the control was determined by one-way ANOVA with Dunnett's post hoc test.

Figure 7

Positive Pearson's correlation between NRF2 and SIRT3 expression of all substances (a). Relative quantification values of NRF2 comparing all treatments regarding BrdU control (b+c). Significant activation of SIRT3 can be seen with anthocyanidin and EGCG both at a concentration of 50 μM. EGCG 100 μM increased SIRT3 expression but not significant (b). Both concentrations of roxithromycin increased SIRT3 expression. The other secondary pant ingredients did either not influence SIRT3 expression or like the endogenous substances ameliorate SIRT3 (c). Statistical significance was defined ^∗p < 0.05; ^∗∗p < 0.01; ^∗∗∗p < 0.005; ^∗∗∗∗p < 0.001. The results were expressed as mean ± SD. Statistical significance between compounds and concentrations to the control was determined by one-way ANOVA with Dunnett's post hoc test.

Figure 8

Pearson's correlation IL6 levels pg/ml and SIRT3 gene expression RQ regarding treatment with secondary plant ingredients. SIRT3 activation diminished IL6 secretion in senescent cells, thus ameliorating inflammation regarding secondary plant ingredients but not endogenous substances (p < 0.05) (a). Positive Pearson's correlation SIRT3 and NRF2 of EGCG in a dose dependent manner (0 μM, 50 μM, and 100 μM) (p < 0.05) (b).