Oxidative Stress, Antioxidant Capabilities, and Bioavailability: Ellagic Acid or Urolithins?

Oxidative stress (OS), triggered by overproduction of reactive oxygen and nitrogen species, is the main mechanism responsible for several human diseases. The available one-target drugs often face such illnesses, by softening symptoms without eradicating the cause. Differently, natural polyphenols from fruits and vegetables possess multi-target abilities for counteracting OS, thus representing promising therapeutic alternatives and adjuvants. Although in several in vitro experiments, ellagitannins (ETs), ellagic acid (EA), and its metabolites urolithins (UROs) have shown similar great potential for the treatment of OS-mediated human diseases, only UROs have demonstrated in vivo the ability to reach tissues to a greater extent, thus appearing as the main molecules responsible for beneficial activities. Unfortunately, UROs production depends on individual metabotypes, and the consequent extreme variability limits their potentiality as novel therapeutics, as well as dietary assumption of EA, EA-enriched functional foods, and food supplements. This review focuses on the pathophysiology of OS; on EA and UROs chemical features and on the mechanisms of their antioxidant activity. A discussion on the clinical applicability of the debated UROs in place of EA and on the effectiveness of EA-enriched products is also included.


Introduction
The health effects of many fruits, fruit juices, nuts, and seeds have been associated with their high content of antioxidant polyphenols and particularly in ellagitannins (ETs) able to provide ellagic acid (EA), one of the most powerful antioxidant molecules [1][2][3]. Food chemists consider both ETs and EA as nutraceuticals, because in addition to possessing the basic nutritional values, they are gifted with several extra health benefits, therefore the dietary intake of foods containing these components often translates in relevant biological effects. For example, documented findings assert a correlation among the consumption of ET-rich foods and greater cardiovascular health [1][2][3][4] or among the consumption of fruits and vegetables and minor incidence of coronary heart disease [5]. Moreover, much empirical data led to the hypothesis that both EA and ETs might be exploited to prevent chronic and degenerative diseases such as cancer, diabetes, cardiovascular diseases, and central nervous system (CNS) disorders [6].
RONS-induced senescent cells are subject to irreversible mutations, that lead to the secretion of soluble factors (interleukins, chemokines, and growth factors), of degradative enzymes like matrix metalloproteases (MMPs), and of insoluble proteins/extracellular matrix (ECM) components, that contribute to determine inflammation, hypertension, and endothelial dysfunction, thus favoring pathological settings.
Moreover, OS is also a key player in cancer development. Consequently, to a metabolic adaptation, cancer cells produce high amounts of ROS, potentially harmful for healthy cells, but simultaneously, they are equipped with increased levels of antioxidants able to efficiently counteract OS and to defend themselves from impairments evoked by OS [27,49].
In fact, in cancer cells, several redox-sensitive transcription factors [protein p53, nuclear erythroid related factor 2 (Nrf2), nuclear factor kappa-B (NFκB)] are hyper-activated and are able to modulate the expression of both antioxidant genes and of signal transduction proteins [protein kinase C (PKC), mitogen-activated protein kinase (MAPK), serine-protein kinase (ATM), etc.].
Through to the constitutive activation of these redox-sensitive pathways, cancer cells ensure themselves the ability to proliferate and reduce death.
To induce age-related pathologies, RONS act on various pathways affecting several cellular processes as reported in Table 1. Table 1. Components influenced by reactive oxygen and nitrogen species (RONS) and effects provoked [50]. ↑ Means increase; ↓ means decrease/reduction.

RONS Origin
RONS can derive from both endogenous and exogenous sources ( Figure 2, Table 2) [50].     As a defense mechanism, by using three different kinds of NOS, i.e., epithelial NOS, neuronal NOS, and inducible NOS, cells generate NO from intracellular L-arginine, which is converted to NO • due to NADPH as an electron source. NO • and ONOO − are produced by reaction of NO with ROS, while NO • in combination with O 2 , provides ONOO • , which induces lipid peroxidation in lipoproteins [24,51,52]. Table 3 summarizes the most representative radicals and the reactive species ONOOCO 2 − produced in biological aerobic systems. Whatever their origin, RONS cause indifferently detrimental oxidative modifications of cellular macromolecules such as carbohydrates, lipids, proteins, and DNA, producing molecules, considered also markers of OS (Table 4). Table 4. Oxidative modification of cellular macromolecules: reactions involved and markers of OS produced in such way.

Macromolecules
Reactions

Proteins
RNS with free or within polypeptide sequences L-tyrosine Nitrotyrosine (NT) [53] Fenton reaction of oxidants with L-lysine, L-arginine, L-proline, L-threonine PC [54] Proteins and lipids Michael-addition of aldehydic lipid oxidation products to L-lysine, L-cysteine, L-histidine PC [54] Proteins and lipids Complex oxidative process oxLDL [55] Proteins and carbohydrates Glyco-oxidation between L-lysine amino groups and L-arginine carbonyl groups linked to carbohydrates AGEs (N-ε-carboxymethyl-lysine pentosidine glucosepane) [ [57] In order to counteract these detrimental effects, cells have developed several repair systems able to repair or eliminate those lipids, proteins, and DNA damaged by the action of RONS. In particular, cytosolic and mitochondrial DNA repair enzymes include polymerases, glycosylases, and nucleases, while proteinases, proteases, and peptidases make up part of the proteolytic enzymes, which take care of removing damaged proteins.
In addition, biological systems have developed both physiological and biochemical mechanisms in order to minimize free radicals (FRs) production and reactive species toxicity. At physiological level, the microvascular system exerts the function of maintaining the levels of O 2 in the tissues, while at biochemical level a protective activity is exerted both by endogenous (enzymatic and non-enzymatic) and exogenous molecules, as reported in Table 5.
In this regard, GSH-Px, glutathione reductase (GR), and methionine sulfoxide reductase (MSR) act as intermediaries in the repair process of oxidative damage.

Nrf2
Regulates the expression of antioxidant proteins ARE Encodes for detoxification enzymes and cytoprotective proteins NQO1 Catalyzes the two-electron reduction of quinones and quinonoid compounds to hydroquinones MSR Carries out the enzymatic reduction of the oxidized form of methionine to methionine

EA Chemical Structure and Physical Properties
EA is a chromene-dione derivative whose chemical name is 2, 3, 7, 8-tetrahydroxy-chromeno[5 ,4,3-cde]chromene-5,10-dione, which can also be seen as a dimeric derivative of gallic acid (GA) ( Figure 3) and whose physicochemical properties are summarized in Table 6 [40,58].   EA appears as cream-colored needles (from pyridine) or yellow powder, odorless, and incompatible with strong reducing agents. EA possesses a highly thermostable structure with a melting point of about 450 °C and a boiling point of 796.5 °C [34]. Due to the weak acidic nature of its four phenolic groups (pKa1 = 5.6 at 37 °C), around neutral pH it is mainly deprotonated on positions 8 and 8′, while above pH 9.6 lactone rings open to give a carboxyl derivative [40].
EA structure includes both a hydrophilic moiety composed of four hydroxyl and two lactone groups and a lipophilic planar fragment, consisting of two hydrocarbon phenyl rings. Consequently, EA is equipped with a high degree of crystallinity, deriving from its planar and symmetrical structure and from the extensive hydrogen-bonding network, which can form within the crystals. These structural characteristics, far from being an advantage, lead the acid to be poorly soluble both in  EA appears as cream-colored needles (from pyridine) or yellow powder, odorless, and incompatible with strong reducing agents. EA possesses a highly thermostable structure with a melting point of about 450 • C and a boiling point of 796.5 • C [34]. Due to the weak acidic nature of its four phenolic groups (pKa1 = 5.6 at 37 • C), around neutral pH it is mainly deprotonated on positions 8 and 8 , while above pH 9.6 lactone rings open to give a carboxyl derivative [40].
EA structure includes both a hydrophilic moiety composed of four hydroxyl and two lactone groups and a lipophilic planar fragment, consisting of two hydrocarbon phenyl rings. Consequently, EA is equipped with a high degree of crystallinity, deriving from its planar and symmetrical structure and from the extensive hydrogen-bonding network, which can form within the crystals. These structural characteristics, far from being an advantage, lead the acid to be poorly soluble both in aqueous or in organic solvents. EA possesses an insignificant water solubility of about 9.3-9.7 µg/mL at pH 7.4 and 21 • C [8] and a very poor solubility in alcohol [33], which concretize in a very poor bioavailability and trivial absorption in gastrointestinal tract (GIT). EA water solubility increases with pH, as well as the antioxidant action [40]. EA is almost insoluble in acidic media and distilled water, while its water solubility is significantly improved by basic pH. However, in basic solutions, phenolic compounds lack stability and these molecules, under ionic form, undergo extensive transformations or are converted to quinones, as a result of oxidation. A stability study on pomegranate fruit peel extract demonstrated that EA content significantly decreases in a few weeks regardless of the pH of the solution, due to the hydrolysis of the ester group with hexahydroxydiphenic acid formation, suggesting that EA should not be stored in aqueous medium [40].
Concerning organic solvents, EA is slightly soluble in methanol, more soluble in ethanol (EtOH) and dimethyl sulfoxide (DMSO), and shows maximum solubility in N-methyl-2-pyrrolidone (NMP), confirming the effect of basic pH on EA dissolution [40].
In this regard, high concentrations of EtOH (80% or greater) could be a suggestion to solubilize EA, but such solutions are not advisable for clinical purposes. Similarly, highly diluted DMSO solutions of EA could be achievable, but DMSO is very harmful to humans. One of the most exploited vehicles for EA is polyethylene glycol (PEG) 400, as it is endowed with satisfactory biocompatibility and, at the same time, is miscible with both aqueous and organic solvents [59]. EA solubility in oils and surfactants is also provided, helpful for developing emulsifying-based techniques [40]. EA poor solubility not only prevents it from reaching cells in vivo, but also causes several difficulties in developing any EA pharmaceutical formulations [60].

EA In Vivo Formation and Metabolism
As already reported in previous sections, except for a marginal amount (e.g., 0.7-4.7 mg/100 g of berries, fresh weight), free form of EA is produced mostly in vivo, essentially upon physiological massive ETs hydrolysis in the stomach and by gut microbiota action in the small intestine. In particular, the ETs hydrolysis leads to the production of GA and hexahydroxydiphenoic acid (HHDP), that spontaneously lactonizes to EA also known as 4,4,'5,5',6,6'-hexahydroxydiphenic acid 2,6,2'-6-dilactone [1,2,9]. Once produced, EA, except for an insignificant fraction, reaches the small and then the large intestine undamaged, where, together with the EA produced by gut microbiota from ETs, is metabolized to UROs, which, in turn, are converted to their conjugates as schematized in Figure 4.

EA Chemical Reactivity
EA easily undergoes exothermic acid-base reaction and can be effortlessly sulfonated and nitrated by the corresponding acids [8].
As to the chemical reactivity, EA can undergo three general reaction types: (a) nitrosation (electrophilic aromatic substitution, non in vivo) of the electron rich aromatic

EA Chemical Reactivity
EA easily undergoes exothermic acid-base reaction and can be effortlessly sulfonated and nitrated by the corresponding acids [8].
As to the chemical reactivity, EA can undergo three general reaction types: (a) nitrosation (electrophilic aromatic substitution, non in vivo) of the electron rich aromatic rings, by reaction with sodium nitrite, mineral acid, and pyridine [29], to produce a red quinone oxime (λ max = 538 nm). This reaction is the basis for a spectrophotometric analysis of EA ( Figure 6).  BPDE is a benzo[a]pyrene-derived carcinogen able to promote an electrophilic alkylation of genetic materials (DNA, RNA) with consequent genetic mutation of the cell. The reaction of EA with this electrophilic DNA-damaging agent has been proposed as possible mechanism for EA anticarcinogenic effects [61].
(c) in vivo other reactions: (i) EA is able to interact with several important biological macromolecules such as DNA, exercising anti-mutagenic and anti-carcinogenic activity [62][63][64].
(ii) EA can act as selective estrogen receptor modulator (SERM) with the possibility to work both as estrogenic and anti-estrogenic [65].
(iii) as nuclear hormone receptor, working as a so called "Endocrine Disruptome" with antagonist or agonist activity [66].
(iv) EA by interacting with polyphenol oxidase enzymes [29,67] can be oxidized to produce a 1,2-quinone able to develop acute cytotoxicity inside unhealthy cells, causing their death and ameliorating many not curable pathologies such as cancer [68].

The Common Sources of EA
EA is a common secondary metabolite in many medicinal plants and vegetables, where its free form is present at very low concentrations. Mainly, EA is present in glycoside forms, i.e., conjugated with a saccharide unit, such as glucose, rhamnose, arabinose, or in complex derivatives, as component of ETs. EA is a sub-fraction of ETs, in many fruits (pomegranates, persimmons, raspberries, black raspberries, wild strawberries, peaches, plums), in some nuts (walnuts, almonds), in seeds such as berry seeds, in vegetables [7,8], and in many species of medicinal plants, associated with health benefits and commonly ingested with a diet. An updated list of plants (43 species), where EA was isolated or only identified, is available in Table 7.

EA Chemical Reactivity
EA easily undergoes exothermic acid-base reaction and can be effortlessly sulfonated and nitrated by the corresponding acids [8].
As to the chemical reactivity, EA can undergo three general reaction types: (a) nitrosation (electrophilic aromatic substitution, non in vivo) of the electron rich aromatic rings, by reaction with sodium nitrite, mineral acid, and pyridine [29], to produce a red quinone oxime (λ max = 538 nm). This reaction is the basis for a spectrophotometric analysis of EA ( Figure 5).  BPDE is a benzo[a]pyrene-derived carcinogen able to promote an electrophilic alkylation of genetic materials (DNA, RNA) with consequent genetic mutation of the cell. The reaction of EA with this electrophilic DNA-damaging agent has been proposed as possible mechanism for EA anti-carcinogenic effects [61].
(c) in vivo other reactions: EA is able to interact with several important biological macromolecules such as DNA, exercising anti-mutagenic and anti-carcinogenic activity [62][63][64].
EA can act as selective estrogen receptor modulator (SERM) with the possibility to work both as estrogenic and anti-estrogenic [65]. (iii) as nuclear hormone receptor, working as a so called "Endocrine Disruptome" with antagonist or agonist activity [66].
(iv) EA by interacting with polyphenol oxidase enzymes [29,67] can be oxidized to produce a 1,2-quinone able to develop acute cytotoxicity inside unhealthy cells, causing their death and ameliorating many not curable pathologies such as cancer [68].

The Common Sources of EA
EA is a common secondary metabolite in many medicinal plants and vegetables, where its free form is present at very low concentrations. Mainly, EA is present in glycoside forms, i.e., conjugated with a saccharide unit, such as glucose, rhamnose, arabinose, or in complex derivatives, as component of ETs. EA is a sub-fraction of ETs, in many fruits (pomegranates, persimmons, raspberries, black raspberries, wild strawberries, peaches, plums), in some nuts (walnuts, almonds), in seeds such as berry seeds, in vegetables [7,8], and in many species of medicinal plants, associated with health benefits and commonly ingested with a diet. An updated list of plants (43 species), where EA was isolated or only identified, is available in Table 7. Combretaceae [107] In the following Table (Table 8) fruits, vegetables, nuts, fresh or minimally processed food and beverages are listed, in which EA was found and frequently quantified. Table 8. Contents of ellagitannins (EA) in fruits, nuts (mg/100 g of fresh weight), in beverages (mg/L), and in seeds (mg/g seed). The total EA content is estimated after acid hydrolysis so as to include ETs in the estimation [33]. Discrepancies are not unexpected as EA content values (even for the same plant material) can vary markedly depending on different extracting conditions [111,123,125], genetic diversity, climate, ripening stage soil, and storage conditions [126]. Kakadu (Australian indigenous Kakadu plum, Terminalia ferdinandiana) as well as other two Australian vegetables (i.e., Anise myrtle, Syzygium anisatum and Lemon myrtle, Backhousia citriodora) encompass a great amount of EA in the free form, thus representing an exception to the rule. Kakadu showed the highest free EA concentration [(228-14,020 mg/100 g of dried weight (DW)] [113,[125][126][127].

EA Antioxidant Power: The Proposed Mechanisms of Action
A large variety of natural antioxidants are present in food and phenolic compounds encompass more than 8000 molecules, sharing the structural feature of presenting a phenol moiety.
EA hydroxyl groups and the lactone systems are home to hydrogen bonds, but can also act as electron acceptors and hydrogen donors. Consequently, EA is endowed with the capacity to accept electrons from different substrates and with the possibility to participate in antioxidant redox reactions, thus resulting a very efficient FRs scavenger [8].
Antioxidants can be considered chemical protectors classified on the basis of their mode of action [128]. Primary antioxidants (Type I, or chain breaking) are chemical species able to prevent oxidation by acting as free-radical scavengers, while secondary antioxidants (Type II, or preventive) act through indirect pathways and by retarding the oxidation process. In this case, antioxidants work through metal chelation, decomposition of hydroperoxides to non-radical species, by repairing of primary antioxidants with hydrogen or electron donation, by deactivating of singlet oxygen, impounding of triplet oxygen, and by absorbing UV radiation [128]. Since EA exerts its protective antioxidant effects through both primary and secondary ways of action, it can be classified as a multiple-function antioxidant.
Concerning the primary way of action, the reactions between EA and FRs are second order reactions and depend both on the concentration of EA and FRs, and on factors influencing their chemical structures such as the medium pH, polarity, and the reaction conditions. In general, the antioxidant capacity of EA is strongly influenced by the reaction medium and, in particular, it is reduced in solvents able to form hydrogen bonds with EA and it is improved in solvents favoring EA ionization to anion phenoxide [129]. In general, the antiradical properties of different natural and synthetic primary antioxidants with OH groups, derive from their capacity to donate hydrogen atom to a FR. Due to this transfer, non-radical species or a new radical, more stable and less reactive than the previous ones, form and possess the ability to exert antioxidant effects by several mechanisms.
EA can exert antioxidant effects mainly through three of the above-mentioned reaction mechanisms. The first is based on the SET reactions, while the second and third are based on the HAT and SPLHAT reactions, respectively. Although the result is the same, i.e., the inactivation of FRs to neutral, cationic, or anionic species, the kinetics and secondary reactions involved in the processes are different. The SET reactions involve the transmission from EA of single electrons with formation of a more stable EA • and charged species. While, the HAT and SPLHAT reactions involve the inactivation of FRs through the donation of a hydrogen atom with formation of EA • and uncharged species. In this regard, EA can react with RONS through HAT reaction, breaking the chain leading to the generation of new reactive toxic species, thanks to its different hydroxyl groups, which are good hydrogen donors [134][135][136][137][138]. In particular, the reaction mechanism of EA with ROO • consists of a transfer of the hydrogen cation from the EA hydroxyls to the radical species, forming a transition state of an H-O bond with one electron. On the other hand, the hydroxyl groups can interact with the π-electrons of the benzene ring providing molecules endowed with the ability to generate free long-living radicals stabilized by delocalization, able to interfere and modify radical-mediated oxidation processes, by SET reactions. In vitro tests have shown that the type and polarity of solvents strongly influence the antioxidant capacity and the mechanism of action of EA.
The alcohols may act either as acceptors of hydrogen bonds, thus reducing EA antiradical effects by hydrogen atom transfer (HAT) reactions, or favoring the ionization of the EA to anion phenoxides, which can react rapidly with the peroxyl radicals, through an electron transfer, thus improving EA radical scavenging activity by SET reactions [129].
Since EA is also a Type II antioxidant, it exerts its effects against FRs thanks to its ability in inhibiting the endogenous production of oxidants and in particular of radical hydroxyl ( • OH), which is the most reactive and electrophilic of the oxygen-based radicals.
• OH is the factor main responsible for tissue and DNA damage and therefore, its inhibition is of primary importance for reducing OS generated from the metal-catalyzed Fenton reaction and the HWR (Table 3), according to Equations (1)-(4), respectively, involving the reduced forms of Fe and Cu.
Phenolic structures as that of EA, for the presence of the hydrophobic benzenoid rings and for the capability of the phenolic hydroxyl groups to form hydrogen-bonding interactions, could also interact with enzymes involved in radical generation, such as various cytochrome P450 isoforms, lipoxygenases, cyclooxygenase, and xanthine oxidase, thus inhibiting RONS over production [134]. Additionally, EA can work synergistically with other endogenous and exogenous antioxidants, such as ascorbic acid, β-carotene, and β-tocopherol, thus increasing their effectiveness and can regulate intracellular glutathione levels [134].
Regardless, it is necessary to remember that due to some of their hydroxyl groups, phenolic antioxidants including EA, under certain conditions, including high dosage, high concentrations of transition metal ions, alkali pH, and the presence of oxygen molecules, can act as pro-oxidants [141].
These groups may sometimes induce significant DNA damage in the presence of Cu (II) or may create ROS through the reduction of Cu (II) → Cu. The pro-oxidant activity, peculiar of small polyphenols, can trigger apoptosis in cancer cells [142]. In contrast, large molecular-weight phenols, such as ETs, have little or no pro-oxidant properties [143].

Structure and Chemistry
The term urolithins (UROs) refers to a very large family of metabolites produced from free EA not absorbed in the stomach and to a lesser extent by the mammalian gut-microbiota. UROs are dibenzopyran-6-one derivatives with different hydroxyl substitutions, which can be considered a combination of benzocoumarin and iso-benzocoumarin, including mainly urolithin A (URO A), urolithin B (URO B), and their isomeric forms Iso URO A and Iso URO B (Figure 4, Section 3.2).
The chemical and physical computed properties of URO A and URO B are reported in Table 10.

Formation Pathway and Metabolism of UROs
From a chemical point of view, UROs form through the opening and decarboxylation of one of EA lactone rings and the sequential and progressive removal of hydroxyls from different positions (dehydroxylase activities). The gut microbiota metabolic activity is responsible for these in vivo transformations. In humans, the complete metabolism of EA leads to the production of two UROs subtype, i.e., URO A and B and of their structural isomers Iso URO A and Iso URO B (Figure 4, Section 3.2).
In particular, after the opening of one of the EA lactone rings, catalyzed by a lactonase to give Luteic Acid (LA), a first decarboxylation occurs thanks to a decarboxylase which provides the first metabolite urolithin M-5 (pentahydroxy-urolithin). From this, in the small intestine by removal of one hydroxyl group from different positions tetrahydroxy-urolithin isomers (URO D, URO M-6) are produced, subsequently trihydroxy-urolithins (URO C, URO M-7) are synthetized thanks to the removal of a second hydroxyl and finally dihydrox-urolithins (URO A and Iso URO A) are obtained after the removal of a third one. In addition, monohydroxy-urolithins (URO B and Iso URO B) are also detected in the large intestine, particularly in those cases in which Iso URO A is produced. Further degradation of UROs to remove the second lactone ring has not been reported so far [43] although it should not be discarded [38].
Thanks to their chemical structure, UROs can be easily absorbed, circulate in plasma, reach tissues, including the central nervous system [9], exert their activities and then, they are incorporated into enterohepatic circulation or are conjugated with methyl, glucuronic acid, or sulfate and are eliminated with feces and urine. In this regard, results obtained in animals treated with ETs-rich diet (food and beverages) lead to the identification and quantification of the metabolites in plasma, urine, feces, and that enter tissues. The prevalent metabolites detected in plasma and urine correspond to URO-A, Iso URO-A, and URO-B, mainly conjugated with glucuronic acid [43]. The available data regarding UROs concentrations in plasma, urine, and different tissues are sometimes very different and affected by an overestimation, depending on the analytical methods adopted. Regardless, the glucuronide conjugated of URO A, B, and Iso URO A has been detected in plasma in the range of 0.045-35 µM and in the urine up to 100 µM, even if concentrations of 5330 and 6185 µM have been also reported for URO A and URO B, respectively.
Studies for evaluating UROs concentration in the tissues were mainly performed on rats and pigs and showed a strong dependence on the tissue target and a very large range of 100-1050 ng/g with decreasing accumulation from prostate, intestine, liver, kidney, to lung [43].

Influence of Individual Metabotype on UROs Production
As described for isoflavones, whose metabolism varies among individuals which may be either O-desmethylangolensin (ODMA) 'producers' or 'non-producers' [46], the EA metabolism and consequently the type and quantity of UROs produced depends on the gut microbiota composition that in turn depends on the individual state of health and age, on the environmental and life conditions, and on human metabotype (UM) [44,47].
Three different UMs exist: UM 0 (no UROs producers), UM A (producers of URO A), and UM B (producers of URO B and/or Iso URO A). UM dictates gut microbiota composition and, therefore, UROs production. In addition, due to the influence of age, health status, and obesity, UM can change also across the lifespan, consequently UROs type production and the amount of UROs produced for a single individual can change across the lifespan and with age [9,47]. In this regard, in vitro studies concerning metabolism of EA by Gordonibacter urolithinfaciens and G. pamelaea (Gordonibacter genera belonging to Coriobacteriaceae family) showed the production of tri-hydroxyl derivative URO C only [45]. Monitoring this metabolism by HPLC-MS, it was observed that such bacteria are able to produce only species of UROs not completely transformed to the final metabolites URO-A, Iso URO-A, and URO-B. In particular, the analysis showed the sequential production of URO-M5, URO-M6 up to URO-C from which, also after longer incubation periods, no further hydroxyl was removed by these microbes [43]. These findings demonstrated that the type of UROs produced depends on the type of bacteria which form the gut microbiota and that for the complete transformation of EA into URO-A and URO-B, other bacteria, still unknown, are necessary.
Regardless, it is also possible that Gordonibacter species manage the complete catabolism of EA to final UROs only at physiological conditions found in vivo, which might be critical for their functioning. Differently, other genera from Coriobacteriaceae family tested by Selma et al. in 2014, were not able to produce any kind of UROs [45].
In a study on healthy human volunteers subjected to an acute consumption of 800 mg of pomegranate extract, it was reported that URO A and B were detected after 8 and 24 h in, altogether, three of the 11 subjects, whereas URO A-glucuronide was detected in six of the 11 subjects. Hydroxyl-URO A was found in three subjects at several time points from 2 to 24 h and URO A-glucuronide was found over a period of 2-24 h in six of the subjects, stating that UROs type production varies from individual to individual and that they can circulate in plasma up to 24 h after the intake [144].
The dependence of EA metabolism and types of UROs produced by the human gut microbiota composition and UM was confirmed also by studies on ex vivo cultures [145].
In order to establish the impact of aging on the distribution of UMs and the potential correlation with obesity, lifestyle, and health status, a study was performed on a large Caucasian cohort (n = 839), aged from 5 to 90 years. The findings from this study confirmed, for the first time, that aging is the main factor, followed by health status and weight, that determinates the gut microbiota composition and the type of UROs produced, with potential consequences for human health [47].

Not All Living Species Produce UROs
The metabolism, which transforms EA in UROs, does not occur in all living species, due to the incapability to perform such metabolic pathway characterizing the microbiota of some species. Regardless, the UROs production from ETs has been reported for different animals (Table 11).  [44].
Concerning UROs production in rats, ETs from strawberries with different degrees of polymerization, showed different metabolisms and, in particular, ETs with low degree of polymerization were metabolized to URO A and considerable amounts of nasutins, while ETs with high degrees of polymerization were converted, probably by the action of Eubacterium only to nausitins [43,147].
Regardless, UROs are metabolites present in humans [38,44,148] pigs [38], in beavers, mice, sheep, and cows [146]. Additionally, in castoreum and in pigs fed oak acorns the removal of EA phenolic hydroxyls without opening the lactone ring gives rise to a relevant number of nasutin metabolites [146].

Mechanisms of Antioxidant Effects of UROs
Among UROs, URO A, Iso URO A, and URO B are the most common UROs found in humans and animals. Differently from EA, that is both a type I and a type II antioxidant able to exert antioxidant effects both by SET reactions and HAT ones, UROs have been identified as potent antioxidants only by the oxygen radical absorbance capacity (ORAC) chemical in vitro assay [149]. ORAC measures antioxidant inhibition of oxidation induced by peroxyl radicals and therefore reflects classical radical chain-breaking antioxidant activity by HAT mechanism [150].
In effect, unconjugated UROs do not have other functional groups in addition to two (URO A) or even one (URO B) phenolic hydroxyl groups, that preferentially act as hydrogen donors, favoring HAT reactions and the electron-withdrawing carboxyl group, being part of a lactone ring, cannot promote the SET mechanism. All this evidence suggests that the antioxidant activity of UROs is mediated exclusively by the HAT mechanisms [149].
Although UROs play a minimal role as a direct radical scavenger, their cytoprotective action has to be attributed mainly to the improvement of the cellular antioxidant defenses and to their activity as oxidases inhibitors [150].

EA or UROs: An Open Debate
The question about who really is responsible between EA and UROs for the health benefits deriving from the intake of ETs-rich foods, has not been clarified at all, and it is still much debated and under investigation so far. Regardless, the hypothesis that UROs are the actual bioactive compounds, firstly launched by Cerda et al. in 2005 [152], it is now the most consolidated and trustworthy one. This assumption finds justifications in the fact that practically only UROs and their phase II conjugates, are adsorbed and are able to circulate in the blood and to reach the different target tissues, where they may interact with the cell machinery and trigger different molecular and cell responses. Differently, the rather non-sense blood concentrations achieved by EA are not enough for justifying benefits coming from ETs-rich plants intake.
On the other hand, the knowledge of the distribution of UROs in human tissues is still too limited and incomplete to elect UROs as promising molecules for clinical applications devoted to prevent and/or treat OS and ageing-related disorders, due to the complexity of performing such investigations [38]. The current knowledge concerning the in vivo concentrations of UROs and their conjugates induce to reflect whether the in vitro reported UROs biological effects may be of real relevance in vivo [43].
In order to clarify which fluids and/or biological tissues are that ones where UROs, their conjugates or nusatins can be found after the intake of ETs rich foods, ETs, or EA, a complete study was developed, whose results are listed in Table 12 [43]. In the table the acronym URO was left out. Table 12. Animal and human biological fluids and tissues where UROs and/or their conjugates and/or nusatins were detected in several experimental types and conditions and after intake of different ET-rich foods [43].

Hazardous Implications Related to UROs Exposure
Results obtained from a randomized clinical trial showed that the inter-individual variability in the improvement of cardiovascular risk biomarkers, in overweight-obese individuals consuming pomegranate, depends on different UMs and type of UROs produced.
In particular, a high cardiovascular risk was associated to the UMB that dictates for a gut microbiota composition that determinates a higher production URO B and Iso URO A, rather than URO A or no production [47]. In fact, in this regard, the positive effects of UROs are obscured by a negative side effect Iso URO A and URO B [47].
In recent years, the dual redox behavior of natural polyphenols has increasingly captured the interest of researchers and the need to investigate their pro-oxidant capabilities in addition to the antioxidant effects has become urgent [149]. In this regard, UROs have received contradictory reports on their antioxidant capacity, and their pro-oxidant properties have been recently studied.
The redox properties of URO A and URO B, have been investigated by using more than one assay method including ORAC assay, three cell-based assays, copper-initiated pro-oxidant activity (CIPA) assay, and cyclic voltammetry, and the findings unveiled that UROs, although strong antioxidants in the ORAC assay, are mostly pro-oxidants in cell-based assays [149].
Citing Rahal et al., 2014 [153]: "Pro-oxidant refers to any endobiotic or xenobiotic that induces oxidative stress either by generation of ROS or by inhibiting antioxidant systems". As reported, the pro-oxidant activity of natural polyphenols depends on their dosage and on the presence of proper amounts of metal ions [154], but while small polyphenols can exhibit considerable pro-oxidant activity, large molecular-weight phenols, such as ETs, have little or no pro-oxidant properties [144].
In this context, the tendency of small polyphenols to exert pro-oxidants effects is considerable in the presence of high concentrations of transition metal ions, such as Cu 2+ or Fe 2+ . A proposed mechanism suggests that, firstly, UROs reduce Cu 2+ to Cu + and, subsequently, Cu + is re-oxidized in a Fenton-like reaction by the action of H 2 O 2 or O 2 , leading to the production of oxygen radicals [149]. Considering that in living cells, a small amount of hydrogen peroxide is produced as a result of cellular metabolism, and that transition metal ions are available, the proposed scenario is without a doubt feasible.

Authors Opinions, Future Perspectives and Conclusions
Hippocrates, over 2500 years ago, coined the phrase "Let food be the medicine and medicine be the food" and in the past, folk medicine, that made use of spices, plants, herbs, fruits, and seeds for the treatment of several diseases, gave it a considerable applicative importance.
Some foods are more than fuel for the organism, and could contain nutrients as ETs and EA, that proved to possess activities essential to preclude diseases and to own high potentials for being effectively used as a sort of medicine to prevent and/or enhance illnesses.
Nowadays, an extensive literature, mainly based on in vitro tests, reports the several EA health properties, based on its multi-target antioxidant effects, able to counteract OS and aging-related diseases. As a consequence, a worldwide market concerning the production of plants concentrated extracts, EA-enriched foods, EA-based functional foods or food supplements, rapidly has been developed, with the claim to provide tools to increase the daily intake of EA and with a global conspicuous involvement of large capitals. In this regard, the actual potential, effectiveness, and safety of EA-enriched products should be discussed and a more wide-ranging debate is needed.
Since EA is practically not soluble and only an insignificant fraction of the EA contained in these products will actually be absorbed in GIT, essentially, all EA ingested will be metabolized by gut microbiota to UROs, which, easily absorbed, will circulate in blood and reach cells and tissues. Ironically, by the intake of EA-enriched products, in place of improving EA in vivo concentration that of UROs is increased, both in plasma and in tissues.
Nevertheless, depending on the age of individuals, their health conditions, and the composition of their gut microbiota, the types of UROs produced and their concentration can vary and the individual variability in the responses to UROs exposure is unpredictable and may lead to heterogeneous comebacks that could promote also adverse effects, such as cardiovascular disorders. Furthermore, by exerting both antioxidant and pro-oxidant activities, high concentration of UROs may cause DNA damage and apoptosis, leading to the development of OS-related pathologies such as cancer. Consequently, EA-enriched products' commercialization should be subject to more careful control and regulation, and after their intake, URO concentrations and tissues distributions, should be monitored and kept under continuous control, individual by individual.
An uncontrolled large production and distribution of EA-enriched foods and food supplements represent an outstanding problem, which could have heavy negative repercussions on human health rather than positive.
Regardless, further detailed and scrupulous toxicity evaluations are necessary both for EA and UROs, before their therapeutic application in humans could be possible. Clinical investigations need to verify the interesting preclinical findings and to validate if the very good results observed in animals are confirmed in humans.
It is relevant to understand also the inter-individual variability of the population in response to EA and ETs-containing foods, for a better organization of a healthy diet and for establishing the most suitable dosages of polyphenolic nutraceuticals and foods rich in ETs and EA for each person.
As far as the authors' opinion is concerned, it seems anyway conceivable to propose that EA remains the robust compound worthy of further extensive investigations.
In relation to its chemical structure and antioxidant mechanisms, EA might be either a safe nutraceutical, an innovative therapeutic, or a template molecule for the development of novel drugs able to fight RONS for promoting human health.
Finally, although attractive, the hypothesis that declares UROs as the actual substances responsible of beneficial effects coming from the ingestion of EA-rich foods is far from being clearly demonstrated, because studies investigating the in vivo EA effects are limited or even missing.
There is no evidence for attributing with certainty also to EA an in vivo activity, but it is equally incorrect to affirm the opposite and attribute in vivo activity to UROs only.
Author Contributions: Conceptualization, methodology, investigation, data curation, writing-original draft preparation, S.A.; writing-review and editing, visualization, S.A., B.M. and G.Z. All authors have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.