Reducing Inflammatory Aging Signals with Urolithin A: A New Approach to Slow Cell Aging

Aging is associated with increased systemic sterile inflammation (inflammaging), which promotes several age‐associated diseases (Franceschi and Campisi 2014). Key drivers of inflammaging include senescence‐associated secretory phenotype (SASP) factors released by senescent cells (Olivieri et al. 2018). While the exact components of SASP vary between different senescent cells and tissues, core SASP factors include pro‐inflammatory chemokines, matrix‐degrading enzymes, and several damage‐associated molecular pattern (DAMP) molecules (Coppe et al. 2010). Pharmacological inhibition of SASP using small molecules known as senomorphics has been proposed as a potential intervention for age‐associated diseases (Kim and Kim 2019; Lagoumtzi and Chondrogianni 2021). However, such treatments include several flavonoid inhibitors of the p38 MAPK/NF‐κB pathway (Freund et al. 2011; Zhang et al. 2023), free radical scavengers, and Janus kinase (JAK) pathway inhibitors (Niedernhofer and Robbins 2018) that are nonselective and broadly inhibit pathways also involved in homeostatic immune responses to various physiological challenges, thus limiting their systemic therapeutic application (Zhang et al. 2017).

Here we present data indicating that the gut metabolite Urolithin A (UA) acts as a senomorphic compound. Digestive tract bacteria naturally produce UA through the metabolism of ellagitannins and ellagic acid, which are abundant in berries, nuts, and pomegranates (D'Amico et al. 2021). UA has been reported to be a potent anti‐inflammatory agent, alleviating several age‐related conditions in vivo (D'Amico et al. 2021; Ishimoto et al. 2011; Larrosa et al. 2010). Preclinical studies have also shown its protective role against aging and age‐related conditions affecting the muscles, brain, joints, and other organs (D'Amico et al. 2021). In a recent clinical trial, UA supplementation improved muscular endurance in older adults (Liu et al. 2022).

To test the effect of UA on senescent cells, we induced senescence in the human fetal lung fibroblast IMR‐90 line via treatment with 300 nM doxorubicin (S‐dox). We observed robust senescence induction as measured by senescence‐associated beta‐galactosidase (SA‐β‐gal) staining (93%) 10 days following doxorubicin treatment compared to non‐senescent controls (< 2%) (Figure 1A,B). We confirmed this in two additional fibroblast cell lines (Figure S1A,B). In addition to chemotherapy‐induced senescence, we also examined the effects of UA in the context of replicative senescence (RS) (Campisi 1997). Both doxorubicin‐treated S‐dox and RS cells displayed significantly elevated expression of the cell cycle checkpoint inhibitors p16^INK4A and p21^CIP. Treatment with UA had no significant effect on the expression of these markers (Figure 1C,D). Previously, UA pre‐treatment has been shown to reduce the percentage of SA‐β‐gal expressing cells and p53 and p21 expression in senescent Organ of Corti 1 (HEI‐OC1) cells and cochlear explants (Cho et al. 2022) or mesenchymal stem cells derived from nucleus pulposus (Cho et al. 2022; Shi et al. 2021). In another study, treatment with UA did not affect the expression of SA‐β‐gal in human skin fibroblasts undergoing replicative senescence (Liu et al. 2019). While our results show no significant effect on the expression of p16^INK4A and p21^CIP, albeit with a trend towards lower expression, it may be due to the diversity of senescence phenotypes. However, UA did not reduce cell viability in either the proliferating IMR‐90 or the S‐dox in the tested concentration range (Figure S1C). It is important to note that while UA did not have any senolytic effect, senescence is heterogeneous and in other cell types, or modes of senescence induction, UA may induce cytotoxicity. In particular, UA has been shown to stimulate autophagy (Leng et al. 2013; Xavier et al. 2013). In senescent cells with extensive mitochondrial dysfunction and dysregulation, this can lead to autophagy‐induced cell death.

Further confirming senescence induction, most doxorubicin‐treated IMR‐90 cells had two or more γH2AX foci, indicating activation of the DNA‐damage response (DDR) triggered by the double‐strand DNA breaks (Hernandez‐Segura et al. 2018; Kinner et al. 2008). The number of individual foci has been used as an indication of the severity of the DNA damage (Noubissi et al. 2021). Even though treatment with UA did not reduce the proportion of cells with the activated DDR (Figure 1E,F), we observed fewer foci in UA‐treated senescent cells. However, it did not reach statistical significance in either the NS or the S‐dox cells (Figure S1D). The loss in the nuclear localization of high mobility group box 1 (HMGB1), a member of the highly conserved nonhistone DNA‐binding high‐mobility group protein family, is a well‐established senescence marker (Davalos et al. 2013). Its secretion from senescent cells has been reported as part of DAMP signaling, a key driver of the senescent phenotype (Wiley and Campisi 2021). Notably, treatment with UA significantly reduced the percentage of cells that lost HMGB1 from the nucleus in both the doxorubicin‐treated and RS cells (Figure 1E,G).

Treatment with UA also resulted in a significant reduction in the expression of prototypical SASP factors, including interleukin 6 (il6), interleukin 8 (il8), and interleukin 1‐alpha (il1α), in both models of senescence induction, with no effect on the NS cells (Figure 1H–J, Figure S1E,F). Reduced gene expression was followed by a significant decrease in the secretion of IL‐6 and IL‐8 after UA treatment, as measured by ELISA in both doxorubicin‐treated and replicative senescent cells (Figure 2A,B).

The primary physiological function of SASP from senescent cells is to recruit immune cells, facilitating tissue repair, and regeneration during wound healing. The pro‐inflammatory chemokines C‐C motif ligand 2 (CCL2) and C‐C motif chemokine ligand 5 (CCL5) can recruit and activate macrophages and NK cells, promoting local inflammation (Budamagunta et al. 2021; Luciano‐Mateo et al. 2020). As expected, expression of both chemokines was significantly upregulated in senescent cells. Treatment with UA mildly reduced expression, specifically in doxorubicin‐treated cells, suggesting that UA has a distinct mechanism of action dependent on the type of senescence induction (Figure 2C,D).

We have previously demonstrated that, in addition to their pro‐inflammatory function, SASP factors induce senescence in neighboring cells (Admasu et al. 2023). As UA reduced the levels of IL‐6 and IL‐8 in the conditioned media (CM), we tested if its treatment abrogates paracrine senescence by culturing proliferating IMR‐90 in the presence of media collected from either control or UA‐treated senescent cells. Our results showed significantly fewer SA‐β‐gal‐positive cells when treated with CM from UA‐treated senescent cells compared to CM from vehicle‐treated senescent cells, suggesting that it acts as an inhibitor of paracrine senescence (Figure 2E). While UA might have a broader impact on other SASP factors, its effect on the expression of IL‐6 and IL‐8 at least partially explains the ability of UA to reduce paracrine senescence, as both these cytokines are causally linked to paracrine senescence (Acosta et al. 2013; Kojima et al. 2013; Ortiz‐Montero et al. 2017; Li et al. 2020).

Finally, we examined the possible mechanisms associated with UA‐mediated regulation of cytokines and chemokines. Cellular senescence is associated with the accumulation of cytosolic DNA released from structurally impaired nuclei and mitochondria. Through an evolutionarily conserved mechanism of pathogen detection, misplaced DNA fragments are recognized by intracellular DNA sensors, including cyclic GMP‐AMP Synthase (cGAS) and the cGAS effector protein Stimulator of Interferon Genes (STING) (Sun et al. 2013). cGAS binding to double‐stranded DNA (dsDNA) results in cyclic guanosine monophosphate–adenosine monophosphate (cGMP) buildup as a secondary messenger that binds to and activates STING. STING, in turn, induces the translocation of transcription factors nuclear factor‐kappa B (NF‐κB) and IFN regulatory factor 3 (IRF3) into the nucleus (Sun et al. 2013). This promotes the expression of pro‐inflammatory SASP factors, which is further fueled by the direct exchange of cGAMP between neighboring cells through cGAMP transporters and gap junctions, thereby creating a more widespread inflammation (Loo et al. 2020; Aguado et al. 2021; Chen et al. 2016). The cGAS‐STING pathway has been proposed as a potential target for pharmacological intervention in disease treatments and controlling inflammation and immunity (Chin et al. 2023). We found that doxorubicin‐treated senescent cells indeed showed evidence of cytosolic DNA accumulation as demonstrated by induction of γH2AX foci (Figure S1G) and dsDNA‐specific pico green (PG) staining present outside of the nuclear boundaries (Figure 2F, Figure S1H). STING phosphorylation at the Ser366 residue was previously shown to be required for STING‐dependent activation of IRF3 downstream (Liu et al. 2015). Our results show that pSTING is significantly more abundant in senescent cells, reaching its highest levels shortly after treatment with doxorubicin, before returning to baseline by Day 7 (Figure S1I,J), while UA treatment downregulates pSTING in senescent cells (Figure 2G). Furthermore, we found that the cytosolic DNA foci were significantly reduced following treatment with UA, a finding consistent across all three cells we tested (Figure 2F, Figure S1K,L). These data suggest that UA may reduce the leakage of dsDNA, which aligns with previous reports demonstrating that UA enhances cytosolic DNA clearance in cancer cell lines (Madsen et al. 2024). Finally, to confirm that UA acts by regulating the cGAS‐STING signaling pathway, we tested a combination of UA treatment with a STING inhibitor 9‐nitro‐oleic acid (9‐NO), which inhibits STING palmitoylation at cysteines 88 and 91 (C88/91) that are crucial for STING activation (Hansen et al. 2018). The 9‐NO treatment downregulated expression of the SASP factors (il6, il8, and il1α). Co‐treatment with UA did not have any additional effect on this inhibition, suggesting that UA and STING regulate target gene expression through the same pathway (Figure 2H–J).

In aged mice, UA supplementation improved muscle strength, reduced inflammation, and promoted autophagic and mitochondrial health (Luan et al. 2021). UA treatment has also been shown to increase self‐renewal and prolong the survival of T memory stem cells (TSCM), thereby preventing T cell exhaustion and promoting effective anti‐tumor T cell responses, further supporting its use as an anti‐aging therapeutic (Luan et al. 2021). Senescent cells are known to have dysfunctional mitochondria, which results in oxidative stress. Over time, damaged mitochondria can release mitochondrial DNA (mtDNA) into the cytosol. Cytosolic mtDNA can act as a DAMP molecule, triggering inflammation through pathways such as the cGAS‐STING and the secretion of the SASP and DAMP. UA is known to activate mitophagy, so UA treatment may reduce cytosolic DNA by activating mitophagy in senescent cells.

Additionally, while cGAS‐STING activation primarily serves as an inflammatory response, it can also act as a feedback mechanism to promote mitochondrial quality control by activating mitophagy. Studies have demonstrated that cGAS‐STING activation can upregulate mitophagy‐related genes or enhance mitophagy activity through phosphorylation of autophagy adaptors by TBK1. Upregulation of mitophagy may reduce the accumulation of cytosolic DNA, inflammation, oxidative stress, and mitochondrial dysfunction (Jiménez‐Loygorri et al. 2024). UA has been previously shown to regulate mitophagy, accelerating mitochondrial turnover (Ryu et al. 2016; Toney et al. 2019; D'Amico et al. 2021). To test whether UA restores the mitochondrial integrity and function in senescent cells, we measured the proportion of mitochondria with a higher membrane potential, indicative of a functional electron‐transport chain and ATP production. However, UA did not significantly alter the proportion of active mitochondria in senescent fibroblasts, indicating a different mechanism at play (Figure S1M). It is likely that UA‐mediated regulation of mitophagy is context‐dependent as some of the insults triggering senescence, such as ultraviolet (UVA) radiation, are known to have a more severe effect, disrupting autophagy more generally (Xie et al. 2022; Ahsan et al. 2019). Taken together, these results suggest that UA may mitigate senescence‐associated damage and the subsequent release of DNA into the cytosol by distinct routes in context‐dependent ways, which may rely in part on the preservation of structural integrity of the nuclei and the mitochondria, as well as on the stimulation of autophagy.

In conclusion, these data demonstrate that UA's senomorphic effect is mediated by reducing cytosolic DNA and cGAS‐STING‐mediated inflammatory response, thus targeting the root cause of the condition, DAMP signaling.

Amit Sharma, Manish Chamoli: conceptualization. Amit Sharma: supervision, funding acquisition. Anna Barkovskaya, Ashley Brauning, Aditi Thambala, Kriti Bhardwaj, Anand Rane: data collection. Anna Barkovskaya, Ashley Brauning, Aditi Thambala: formal analysis. Anna Barkovskaya, Ashley Brauning: writing – original draft preparation; Julie K. Andersen, Manish Chamoli, Amit Sharma: writing – review and editing.

M.C. and J.K.A. are co‐founders of Symbiont Bio and declares no conflicts of interest.