The gut microbiota metabolite urolithin A, but not other relevant urolithins, induces p53-dependent cellular senescence in human colon cancer cells

Highlights

• Urolithin A exerts anti-clonogenicity and cell cycle arrest in colon cancer cells.

• Long term exposure to Urolithin A induces cellular senescence.

• Urolithin A induces cellular senescence in a p53-dependent manner.

• Phase-II metabolism limits the senescence-mediated chemoprevention.

Abstract

The promotion of senescence in cancer cells by dietary (poly)phenols gained attention as a promising chemopreventive strategy against colorectal (CRC) and other cancers. Urolithins (Uros) are ellagitannins and ellagic acid-derived gut microbiota metabolites that reach high concentrations in the human colon. They were postulated to be as potential anticancer agents in different CRC models, but their role as promoters of cellular senescence has never been comprehensively evaluated.

We evaluated long-term senescent-mediated chemoprevention of physiologically relevant doses of different Uros and representative mixtures of human urolithin metabotypes in human CRC (HCT-116, Caco-2, and HT-29) and non-tumorigenic (CCD18-Co) cell lines. Our results show that Uro-A (but not Uro-C, IsoUro-A, or Uro-B) leads to a dose-dependent anti-clonogenic effect through the increase of the senescence-associated β–galactosidase activity, rather than by reversible cell cycle arrest and(or) apoptosis which require much higher concentrations. Senescence was accompanied by an elevated p53 and p21^Cip1/Waf1 expression in HCT-116 cells (p53-wild type), but not in other CRC lines with p53 mutated or non-tumorigenic cells, which suggests that long-term senescence-mediated chemoprevention is a p53-dependent manner. Moreover, the ATP-binding cassette transporters and the phase-II metabolism of Uros limited the induction of senescence, which anticipates lower effects of conjugated Uros against systemic cancers.

Introduction

Colorectal cancer (CRC) is one of the most common malignancies (per incidence), and the second cause of cancer-related death worldwide in both sexes (Bray et al., 2018). Early diagnosis and the promotion of healthy dietary habits have been established as essential measures of primary prevention for CRC development (Shike, 1999; Magalhaes et al., 2012; Grosso et al., 2017). Epidemiological studies have described the inverse correlation between CRC incidence and a high intake of fruits and vegetables rich in phytochemicals, including phenolic compounds (Turati et al., 2015; Bingham et al., 2003). However, despite the abundant preclinical (in vitro and animal models) anticarcinogenic activity reported for different individual phenolics and phenolic-rich plant foods, the clinical evidence remains elusive so far (Núñez-Sánchez et al., 2015).

Ellagitannins (ETs) and ellagic acid (EA) are polyphenols occurring in pomegranate, walnuts, and many berries. ETs are not absorbed, and EA shows low bioavailability mainly due to its limited solubility under physiological conditions (González-Sarrías et al., 2015). Consequently, EA is metabolized by the colonic microbiota producing a family of bioavailable metabolites known as urolithins (Uros) (Cerdá et al., 2004; Tomás-Barberán et al., 2017). Remarkably, Uros reach relevant concentrations in the lumen and colonic tissues (malignant and normal) of CRC patients after the intake of an ET-rich pomegranate extract (Núñez-Sánchez et al., 2014).

The gut microbiota composition in humans differentially determines the molecular form, concentration, and bioactivity of Uros. This dissimilar metabolism is linked to the inter-individual variability of the human gut microbiota, which has led to identify three different urolithin metabotypes (UMs), i.e., metabotype A (UM-A) where only urolithin A (Uro-A) is the final urolithin produced, metabotype B (UM-B) that produces Uro-A, isourolithin A (IsoUro-A) and urolithin B (Uro-B), and metabotype 0 (UM-0) that does not produce these final Uros (Tomás-Barberán et al., 2014; Romo-Vaquero et al., 2019).

Uros have been extensively studied in CRC models, showing cell cycle arrest and apoptosis induction via modulation of genes and proteins expression, as well as signaling events associated with CRC development (González-Sarrías et al., 2009, 2014, 2016, 2017). However, Uros’ activity can vary depending on the cell type and dosage. Besides, their action has only been reported in short-term treatments and thus, we hypothesize that Uros may promote cellular senescence after long-term incubation times.

Cellular senescence has been proposed to be an anticancer mechanism since it prevents irreversible, permanent cell cycle progression on cancer cells accompanied by changes in cell morphology and activation of senescence-associated β-galactosidase (SA-β-gal) as the most common hallmarks. Besides, senescence activation is accompanied by substantial molecular changes in gene expression linked to cell growth inhibition, such as up-regulation of p53, cell cycle inhibitors, including p21^Cip1/Waf1 and p16^INK4a, as well as the hypo-phosphorylated form of retinoblastoma (Rb) (Campisi and d'Adda di Fagagna, 2007; Sikora et al., 2011; Muñoz-Espín and Serrano, 2014; Lessard et al., 2018).

To date, only one clinical trial in CRC patients reported some molecular changes in several key CRC-related genes and microRNAs in normal and cancerous colon tissues, after the intake of an ET-rich pomegranate extract (Núñez-Sánchez et al., 2015, 2017). However, the short-term intake (days) disallowed the association of molecular changes with the patients’ Uros and (or) EA levels detected in the colon tissues, which prevented possible clinical evidence of Uros against CRC.

The present study aims to investigate whether a panel of Uros and representative mixtures of UMs at physiologically relevant, but non-cytotoxic concentrations can induce senescence in human CRC and non-tumorigenic cells. Additionally, we investigated whether the cellular senescence induction was directly promoted via ATP-binding cassette (ABC) transporters, as well as whether the phase-II metabolism of Uros could limit their effect.