Effects of Urolithin A on Mitochondrial Parameters in a Cellular Model of Early Alzheimer Disease

Abstract

(1) Background: Ellagitannins are natural products occurring in pomegranate and walnuts. They are hydrolyzed in the gut to release ellagic acid, which is further metabolized by the microflora into urolithins, such as urolithin A (UA). Accumulation of damaged mitochondria is a hallmark of aging and age-related neurodegenerative diseases. In this study, we investigated the neuroprotective activity of the metabolite UA against mitochondrial dysfunction in a cellular model of early Alzheimer disease (AD). (2) Methods: In the present study we used SH-SY5Y-APP695 cells and its corresponding controls (SH-SY5Ymock) to assess UA's effect on mitochondrial function. Using these cells we investigated mitochondrial respiration (OXPHOS), mitochondrial membrane potential (MMP), adenosine triphosphate (ATP) production, autophagy and levels of reactive oxygen species (ROS) in cells treated with UA. Furthermore, we assessed UA's effect on the expression of genes related to mitochondrial bioenergetics, mitochondrial biogenesis, and autophagy via quantitative real-time PCR (qRT-PCR). (3) Results: Treatment of SH-SY5Y-APP695 cells suggests changes to autophagy corresponding with qRT-PCR results. However, LC3B-I, LC3B-II, and p62 levels were unchanged. UA (10 µM) reduced MMP, and ATP-levels. Treatment of cells with UA (1 µM) for 24 h did not affect ROS production or levels of Aβ, but significantly increased expression of genes for mitochondrial biogenesis and OXPHOS. Mitochondrial Transcription Factor A (TFAM) expression was specifically increased in SH-SY5Y-APP695. Both cell lines showed unaltered levels of peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC1α), which is commonly associated with mitochondrial biogenesis. Results imply that biogenesis might be facilitated by estrogen-related receptor (ESRR) genes. (4) Conclusion: Urolithin A shows no effect on autophagy in SH-SY5Y-APP695 cells and its effect on mitochondrial function is limited. Instead, data suggests that UA treatment induces hormetic effects as it induces transcription of several genes related to mitochondrial biogenesis.

Keywords: Alzheimer’s; ellagitannin; hormesis; metabolite; mitochondria; neurodegeneration; polyphenol; urolithin.

Figure 1

Chemical structure of compound Urolithin A.

Figure 2

(A): Oxygen consumption of 20⁶ SY5Ymock cells and SY5Y-APP695 cells treated with ctrl or 1 µM UA. N = 9–15. Activity of OXPHOS complexes were assessed via addition of several substrates, inhibitors or uncouplers. Which substance was added in which stage of the experiment is marked with “X”. (B,C) Mitochondrial membrane potential measured of SY5Ymock cells and SY5Y-APP695 cells treated with 1 µM UA (B) or 10 µM UA (C). N = 7–11; (D,E) ATP levels of SY5Ymock and SY5Y-APP695 cells treated with 1 µM UA (D) or 10 µM (E). N = 11–13 Displayed are means ± SEM. Statistical significance was tested via two-way ANOVA followed by a false discovery rate correction according to Benjamini, Krieger, and Yekutieli. Significance is displayed as: ^ns p > 0.05, * p < 0.05, ** p < 0.01; *** p < 0.001, **** p < 0.0001. List of all statistical parameters and comparisons can be found in Supplementary Table S1.

Figure 3

(A) Citrate synthase activity of SY5Ymock and SY5Y-APP695 samples. Cells were incubated with either 1 µM UA or its DMSO control (ctrl). Enzyme activity waws adjusted to protein content of the samples. N = 7–13. (B) ROS level measured in form of DCFDA/H2DCFDA fluorescence in SY5Ymock and SY5Y-APP695 cells. N = 8; (C,D) Mitochondrial membrane potential of SY5Ymock and SY5Y-APP695 cells treated with 1 µM UA (C) or 10 µM UA (D) whose complex I activity was inhibited via addition of 25 µM rotenone. Dashed line corresponds to MMP of uninhibited SY5Ymock/APP695 control cells. N = 7–11; (E,F) ATP levels of SY5Ymock and SY5Y-APP695 cells treated with 1 µM UA (E) or 10 µM (F) whose complex I activity was inhibited via addition of 25 µM rotenone. The dashed line shows the basal ATP level of SY5Ymock/-APP695 cells not treated with 20 µM rotenone. Dashed line corresponds to MMP of uninhibited SY5Ymock/APP695 control cells. N = 7–14; Displayed are means ± SEM. Statistical significance was tested via one-way ANOVA followed by a false discovery rate correction according to Benjamini, Krieger and Yekutieli. Significance is displayed as: ^ns p > 0.05, * p < 0.05, *** p < 0.001 and **** p < 0.0001. List of all statistical parameters and comparisons can be found in Supplementary Table S1.

Figure 4

Relative normalized mRNA expression of relevant genes relevant to OXPHOS and mitochondrial biogenesis. Genes related to OXPHOS: (A) citrate synthases (CS), (B) complex I (NDUFV1), (C) complex IV (COX5D), and (D) complex V (ATP5D). Genes related to mitochondrial biogenesis: (E) Sirtuin 1 (SIRT1), (F) cAMP response element-binding protein ranscription factor (CREB), (G) Peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC1α), (H) nuclear respiratory factor 1 (NRF1), (I) mitochondrial transcription factor A (TFAM), (J) GA Binding Protein Transcription Factor Subunit Alpha (GABPα/NRF2), (K) estrogen-related receptor alpha (ESRRα), (L) estrogen-related receptor gamma (ESRRγ). Displayed are data of SY5Ymock and SY5Y-APP695 cells which were measured seperately from each other. Data of both cell lines is adjusted to aged SY5Ymock ctrl group = 100%; N = 8–10; Displayed are means ± SEM; Statistical significance was tested via two-way ANOVA followed by a false discovery rate correction (FDRC) according to Benjamini, Krieger, and Yekutieli. Results were normalized to the mRNA expression levels of three housekeeping genes (beta-actine (ACTβ), Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and Phosphoglycerate Kinase 1 (PGK1)) according to the MIQE guidelines. Significance is displayed as: ^ns p > 0.05, * p < 0.05, ** p < 0.01, *** p < 0.001 and **** p < 0.0001. List of all statistical parameters and comparisons can be found in Supplementary Table S1.

Figure 5

(A) Expression of microtubule-associated proteins 1A/1B light chain 3 (MAP1LC3) gene related to autophagy; (B) Expression of Sequestosome 1 (SQSTM1) gene related to autophagy. Displayed are data of SY5Ymock (left side) and SY5Y-APP695 cells (right side) which were measured separately from each other. Data of both cell lines is adjusted to SY5Ymock ctrl group = 100%; N = 8–10; Displayed are means ± SEM. Results were normalized to the mRNA expression levels of three housekeeping genes (beta-actine (ACTβ), Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and Phosphoglycerate Kinase 1 (PGK1)) according to the MIQE guidelines. (C) Fluorescence of marker dye binding to autophagosomes of SY5Ymock and SY5Y-APP695 cells. Displayed are means ± SEM. N = 10. (D) Western blots of LC3B-I, LC3B-II, and p62 in SY5Ymock and SY5Y-APP695 treated with 1 µM UA or ctrl. β-Actine was used for housekeeping and results have been adjusted to it (Figure 2E–G). Gels were loaded with 15 µg protein. (E,F) Western blot results from LC3B-I and LC3B-II adjusted to β-actine. Displayed are means ± SEM. N = 9. (G) Western blot results from p62 adjusted to β-actine. Displayed are means ± SEM. N = 8. Cells were incubated with either 1 µM UA or its DMSO control (ctrl). Statistical significance was tested via two-way ANOVA followed by a false discovery rate correction according to Benjamini, Krieger, and Yekutieli. Significance is displayed as: ^ns p > 0.05, * p < 0.05, ** p < 0.01 and **** p < 0.0001. List of all statistical parameters and comparisons can be found in Supplementary Table S1.