Absorption Profile of (Poly)Phenolic Compounds after Consumption of Three Food Supplements Containing 36 Different Fruits, Vegetables, and Berries

The market of plant-based nutraceuticals and food supplements is continuously growing due to the increased consumer demand. The introduction of new products with relevant nutritional characteristics represents a new way of providing bioactive compounds and (poly)phenols to consumers, becoming a strategy to ideally guarantee the health benefits attributed to plant foodstuffs and allowing the increase of daily bioactive compound intake. A paramount step in the study of nutraceuticals is the evaluation of the bioavailability and metabolism of their putatively active components. Therefore, the aim of the present study was to investigate the absorption profile of the (poly)phenolic compounds contained in three different plant-based food supplements, made of 36 different plant matrices, which were consumed by 20 subjects in an open one-arm study design. Blood samples were collected at baseline and 1, 2, 5, and 10 h after capsule intake. Twenty quantifiable metabolites deriving from different (poly)phenolic compounds were identified. Results showed that the consumption of the three capsules allowed the effective absorption of several (poly)phenolic compounds and metabolites appearing at different times in plasma, thereby indicating different absorption profiles. The capsules thus ensured potential health-promoting molecules to be potentially available to target tissues and organs.


Introduction
Daily consumption of five portions or at least 400 g of fruits and vegetables is promoted in many national and international dietary guidelines and by public policies [1,2]. An adequate consumption 23.2 ± 3.0 (m) and 22.8 ± 3.05 (f) (kg/m 2 ) (data expressed as mean ± SD). The volunteers had to meet all the inclusion and exclusion criteria. Subjects had to be non-smokers, with a BMI between 20 and 30 kg/m 2 , not on medication, not premenopausal, following normal dietary habits (no specific diets, meals and food components.) and adhering to a wash-out period. Exclusion criteria included: not consuming more than four servings of fruits and vegetables per day, not having any type of food allergy or histamine intolerance, not displaying a high level of physical activity (defined as more than five training units/week), not having menstrual dysfunctions and not abusing alcohol.
All participants were informed about the purpose of the study via telephone calls or personal meetings. Subjects who wanted to join the study signed the informed consent form before their inclusion in the trial. Moreover, the selected subjects received a (poly)phenol-poor diet plan which they had to adhere to for 48 h before the first blood sampling. The list of permitted and forbidden foods is provided in Table S1. Relevant principles of Good Clinical Practice were followed throughout the study. The study was conducted according to the guidelines laid down in the Declaration of Helsinki, and all procedures involving human subjects were approved by the Ethics Committee of the Medical University of Graz, Austria, (EC-number: 27-507ex14/15). The trial was registered at www.clinicaltrials.gov (identifier No. NCT02587468).

Test Capsules
Capsules used for the study were provided by the Juice Plus Company/NSA LLC, Collierville, TN, USA and manufactured for Europe by Natural Alternatives International (NAI), Manno, Switzerland. The capsules contained powdered juice concentrate derived from 36 different fruits, vegetables, and berries including juice and pulp from different vegetal matrices, namely Juice PLUS+ ® Vineyard (a berry blend), Juice PLUS+ ® Fruit Blend and Juice PLUS+ ® Vegetable Blend, which were kindly supplied by the Juice PLUS+ ® company. In detail, the powder samples differed for their composition: Juice PLUS+ ® Vineyard (hereafter called "berry blend") contained 750 mg of dried powder blend of juice and pulp from grapes and berries (45.7%) including Concord grape, blueberry, cranberry, blackberry, bilberry, raspberry, redcurrant, blackcurrant, elderberry, in varying proportions, besides green tea, ginger root, grape seed, artichoke leaf powder, cocoa powder, and pomegranate powder. Juice PLUS+ ® Fruit Blend ("fruit blend") instead contained 750 mg of dried powder blend of juice and pulp (52%) of apple, orange, pineapple, cranberry, peach, acerola cherry, papaya, in varying proportions, beet root, date, and prune. Lastly, Juice PLUS+ ® Vegetable Blend ("vegetable blend") contained 750 mg of dried powder blend of juice and pulp (60%) of carrot, parsley, beet, kale, broccoli, cabbage, tomato, and spinach, in varying proportions, as well as sugar beet, garlic powder, oat, and rice bran. Moreover, the fruit and the vegetable powders were enriched with vitamins C, and folic acid) and with a natural carotenoid and tocopherol blend. The berry blend powder was enriched with vitamins C and folic acid as well as with a natural tocopherol blend.

Study Design
In an open one-arm study design, three capsules (one berry, fruit, and one vegetable blend), were consumed by each participant on one single occasion. Before the test meal, participants were asked to follow a two-week wash-out period avoiding all food supplements and dietetic products and all kinds of drugs/medications. Additionally, participants were asked to consume a (poly)phenol-poor diet 48 h before the test day. To facilitate participant adherence to the dietary restrictions, a list of permitted and forbidden foods was supplied. The day of the test, subjects visited the lab after an overnight fast, and after the baseline blood drawing (T 0 ), they received the three capsules (one of each blend), to be consumed with 250 mL of still water. Four additional blood samples were collected, at 1, 2, 5, and 10 h after capsule intake. After the 2 h blood sampling, the subjects consumed a standardized (poly)phenol-poor snack, including white bread, cheese, ham, milk, and water ad libitum, in accordance to Pereira-Caro [30]. For each blood drawing, 5 mL of venous EDTA-blood from the elbow were Nutrients 2017, 9,194 4 of 17 collected via vein catheter, in supine position. Blood was immediately centrifuged at 2500 g for ten minutes to separate the plasma, which was frozen at −70 • C until uHPLC/MS n analyses.

Plasma Extraction
Plasma samples of all volunteers were extracted using a solid phase extraction (SPE) method as reported by Urpi-Sarda and colleagues, with some modifications [34]. Oasis ® HLB Vac cartridges (1 cc, 30 mg sorbent, 30 µm particle size) (Waters, Milford, Massachusetts, MA, USA) were conditioned with 1 mL of methanol and equilibrated with 1 mL of water. An aliquot of 800 µL of plasma was added with 16 µL of o-phosphoric acid (2.5%) and then loaded into cartridges. The cartridges were washed with 1 mL of acidified water (0.1% formic acid) and finally eluted with 1 mL of methanol containing formic acid (1.5 mol/L). The eluates were evaporated overnight to dryness by means of a rotary speed-vacuum at room temperature and reconstituted with 80 µL of methanol/acidified water (0.1% formic acid) (50:50 v/v) prior uHPLC/MS n analysis.
For UHPLC, mobile phase A was acetonitrile containing 0.2% formic acid and mobile phase B was 0.2% formic acid in water. The gradient started with 5% A, isocratic conditions were maintained for 0.5 min, and reached 95% A after 6.5 min, followed by 1 min at 95%. The starting gradient was then immediately reestablished and maintained for 4 min to re-equilibrate the column. The flow rate was 0.4 mL/min, the injection volume was 5 µL, and the column temperature was set at 40 • C.
The applied mass spectrometry (MS) method consisted in the selective determination of each target precursor ion by the acquisition of characteristic product ions in selective reaction monitoring (SRM) mode (Table 1), applying a negative ionization. To optimize the method, all the available standard compounds were infused into the MS to set the best mass parameters and to check the actual fragmentation patterns. Finally, for all the analyses, the spray voltage was set at 3 kV, the vaporizer temperature at 300 • C, and the capillary temperature operated at 270 • C. The sheath gas flow was 60 units, and auxiliary gas pressure was set to 10 units. Ultrahigh purity argon gas was used for collision-induced dissociation (CID). The S-lens values were defined for each compound based on infusion parameter optimization (Table 1). Conversely, for compounds that were not available for infusion, the S-lens values were set using the values obtained for the chemically closest available standards. Quantification was performed with calibration curves of standards, when available (Table 1). Data processing was performed using Xcalibur software (Thermo Scientific Inc., Waltham, MA, USA). All data were expressed as mean values ± SEM.

(Poly)Phenolic Compound Absorption
The capsules consumed in the present study were previously characterized for (poly)phenolic profile and quantified for their total phenolic content [23]. Capsules were made of 36 different plant matrices, and contained a wide array of different phenolic compounds, principally ellagitannins, flavan-3-ols, flavonols, and anthocyanins in berry blend capsules, flavones, flavonols, flavanones, and anthocyanins in fruit blend capsules, and flavones and flavonols in vegetable blend capsules.
Out of the 92 monitored molecules, 20 quantifiable metabolites were identified, or tentatively identified in plasma samples. All the quantified metabolites were found as conjugated compounds with sulfate, glucuronide or glycine moieties. Three glycine-conjugated metabolites, hippuric acid, 4-hydroxyhippuric acid, and feruloylglycine, were detected. A total of five metabolites were flavonol derivatives, including conjugates of quercetin, kaempferol, myricetin, and patuletin. Three metabolites were directly linked to flavanone metabolism, namely naringenin, and hesperetin derivatives, whereas only one flavone metabolite, namely diosmetin sulfate, was detected. Among flavan-3-ol derivatives, three conjugated phenyl-γ-valerolactones were detected. Finally, five metabolites were small phenolic derivatives, including hydroxyphenylpropionic acid sulfate, ferulic acid glucuronide, pyrogallol sulfate, dihydroxybenzoic acid sulfate, and methyl-trihydroxybenzoic acid sulfate. Generally, with the exception of hippuric acid, plasma levels of the quantified metabolites did not exceed the nanomolar range. Considering those metabolites for which the origin is strictly attributable to (poly)phenolic compounds contained in the capsules, the most abundant metabolites resulted in the low molecular weight phenolics which are usually produced in the colon by microbial transformation of flavonoids.
The glycine-conjugated metabolites are ubiquitous and could originate both by endogenous precursors [18,35] and by microbial metabolism of phenolic compounds [36][37][38]. Their potential endogenous origin justifies the high concentration at baseline ( Figure S1).
Actually, the absorption curves of hippuric acid and 4-hydroxyhippuric acid did not show a notable concentration peak and their levels were basically constant during the study period. On the contrary, feruloylglycine exhibited a peak plasma concentration 2 h after consumption, indicating a stronger connection with the phenolic compounds introduced through the capsules, as only the capsules were consumed within the 2 h.
Looking to the other circulating metabolites appearing after capsule consumption, two metabolic phases could be easily distinguished. Conjugated metabolites appearing in the circulatory system within 1 or 2 h after capsule ingestion suggest an absorption in the first part of the gastro-intestinal tract. Native compounds are rapidly hydrolysed to release the aglycones, which are then conjugated by sulfotransferases (SULTs) and uridine-5 -diphosphate glucuronosyltransferases (UGTs) at both enterocyte and hepatic level before entering the systemic circulatory system [18]. Kaempferol glucuronide, quercetin glucuronide, quercetin sulfate, myricetin glucuronide, and diosmetin sulfate absorption curves are the expression of (poly)phenol metabolism in the first gastro-intestinal tract (Figure 1). A concentration peak recorded between 5 and 10 h after test meal indicates, instead, a clear interaction between the indigested (poly)phenolic fraction and the colonic microbiota. The absorption profile outlined by patuletin sulfate and hesperetin sulfate likely represents a colonocyte level absorption and a subsequent conjugation at hepatic level ( Figure 2). Actually, their plasmatic concentration reached a maximum level 5 h post capsule consumption.
Probably, naringenin and hesperetin were partially cleaved in the first gastro-intestinal tract and rapidly absorbed and metabolized at intestinal/hepatic level, resulting in a first peak approximately 1 h after capsule ingestion. However, the peak between 5 and 10 h indicates a more important role of the colonic microbiota in flavanone metabolism [30]. Similarly, ferulic acid glucuronide absorption profile showed a double phase curve ( Figure 3). Nevertheless, no ferulic acid was detected in the capsules [23], suggesting that this compound probably originated by catechol-O-methyltransferase (COMT) activity on other hydroxycinnamic acids present in the capsules [39]. A concentration peak recorded between 5 and 10 h after test meal indicates, instead, a clear interaction between the indigested (poly)phenolic fraction and the colonic microbiota. The absorption profile outlined by patuletin sulfate and hesperetin sulfate likely represents a colonocyte level absorption and a subsequent conjugation at hepatic level ( Figure 2). Actually, their plasmatic concentration reached a maximum level 5 h post capsule consumption.
Probably, naringenin and hesperetin were partially cleaved in the first gastro-intestinal tract and rapidly absorbed and metabolized at intestinal/hepatic level, resulting in a first peak approximately 1 h after capsule ingestion. However, the peak between 5 and 10 h indicates a more important role of the colonic microbiota in flavanone metabolism [30]. Similarly, ferulic acid glucuronide absorption profile showed a double phase curve ( Figure 3). Nevertheless, no ferulic acid was detected in the capsules [23], suggesting that this compound probably originated by catechol-O-methyltransferase (COMT) activity on other hydroxycinnamic acids present in the capsules [39].  Finally, most (poly)phenolic compounds passing unmodified and unabsorbed through the first gastro-intestinal tract become a suitable substrate for the locally hosted microbiota. Several modifications on native compounds have been reported to be catalyzed by microbial enzymes, resulting in the formation of low molecular weight compounds [40], which are efficiently absorbed by colonocytes before hepatic conjugation.  Finally, most (poly)phenolic compounds passing unmodified and unabsorbed through the first gastro-intestinal tract become a suitable substrate for the locally hosted microbiota. Several modifications on native compounds have been reported to be catalyzed by microbial enzymes, resulting in the formation of low molecular weight compounds [40], which are efficiently absorbed by colonocytes before hepatic conjugation. Finally, most (poly)phenolic compounds passing unmodified and unabsorbed through the first gastro-intestinal tract become a suitable substrate for the locally hosted microbiota. Several modifications on native compounds have been reported to be catalyzed by microbial enzymes, resulting in the formation of low molecular weight compounds [40], which are efficiently absorbed by colonocytes before hepatic conjugation.

Discussion
In the present study, the absorption profile of (poly)phenolic compounds derived from the ingestion of Juice PLUS+ ® berry, fruit and vegetable blend capsules has been investigated. A total of 20 circulating metabolites have been identified, or tentatively identified, and quantified. As expected, detected metabolites derived from different (poly)phenols and appeared at different times in plasma. Flavonol metabolites principally originated in the first part of the gastro-intestinal tract, as previously reported by other authors. Feliciano and colleagues [41] reported a time to reach the maximum plasma concentration (T max ) for kaempferol 3-O-glucuronide and quercetin 3-O-glucuronide of about 2 h. An absorption curve similar to those reported in the present study for quercetin 3-glucuronide and quercetin 3-sulfate was reported by Mullen and colleagues, who highlighted a concentration peak between 1 and 2 h after onion ingestion [44]. To the best of our knowledge, this is the first study in which plasma concentrations of patuletin and myricetin metabolites have been reported, suggesting that other flavonols could be effectively absorbed and hence investigated for their potential bioactivity. Moreover, a new flavone metabolite at plasmatic level has been detected, namely diosmetin sulfate. Its early peak plasma level registered at 1 h is in contrast with scientific data previously reported for other flavones, like apigenin. The T max of apigenin (measured as aglycone after enzymatic hydrolysis) has been reported to be around 7 h after consumption of parsley [45]. The few human feeding studies involving flavones available so far reported a low bioavailability for this phytochemical [18,46], which could be ascribable to the low absorption of flavones, as demonstrated in the present investigation. Conversely, several scientific data are available for naringenin and hesperetin absorption, and all studies agree about the importance of colonic microbiota activity on this class of (poly)phenols [47][48][49]. The absorption curves observed in the present work confirm that flavanone metabolism principally occurs in the large intestine. Brett and colleagues reported a plasma concentration peak of flavanone conjugates 6 h after orange consumption [50], whereas the highest naringenin and hesperetin derivative urinary excretion has been reported within 2 and 10 h after orange juice consumption [30]. Flavan-3-ols were the most representative compounds in berry blend capsules [23]. However, no catechin monomer conjugates nor dimers or oligomeric proanthocyanidins were detected in plasma. However, in vivo studies have shown that both monomers and high molecular weight flavan-3-ols are effectively degraded by the gut microbiota into hydroxyphenyl-γ-valerolactones [51][52][53]. Many phenyl-γ-valerolactone derivatives have been detected after green tea [51,54], cocoa [34,55,56], wine [57] and almond [58] consumption. Partially compensating for the absence of flavan-3-ol monomers in plasma in the present study, three phenyl-γ-valerolactones were detected and quantified.
It was then demonstrated that phenyl-γ-valerolactones represent an intermediate step in the microbial metabolism of flavan-3-ols and that other low molecular weight compounds, such as phenylacetic, phenylpropionic, benzoic acids derivatives [59] and hippuric acid, which derive from benzoic acids [60], could be formed. Likewise, flavonols, flavones, and flavanones, could be degraded into smaller phenolics, namely phenylacetic, phenylpropionic, and benzoic acid derivatives [61]. Similarly, anthocyanins undergo an important colonic set of transformations, giving rise to low molecular weight metabolites. After glucosidic cleavage of the sugar moiety, cyanidin could be the precursor of caffeic acid, from which ferulic and isoferulic acids could be formed after COMT activity [62][63][64]. Ferulic acid may undergo further phase II metabolism, namely sulfation and glucuronidation, generating conjugated ferulic derivatives [39,65]. Feliciano and colleagues recently hypothesized that peonidin could also lead to the formation of ferulic acids [41]. Moreover, anthocyanins could be converted into small phenolics such as phenylpropionic, phenylacetic, benzoic acids, and pyrogallol [62,64,66].
Concerning pyrogallol, the high plasmatic concentration of pyrogallol sulfate between 5 and 10 h recorded in the present study is in agreement with a T max reported after cranberry consumption [41]. Gallic acid, which was present in the capsules, could have contributed to the formation of small metabolites [67], but also to other compounds like methylgallic acid [68], and, after phase II metabolism, to its 3-O-sulfate derivative [69].
It now appears clear that the extensive bioconversion of (poly)phenol compounds strictly depends on the characteristics of individual colonic microbiota, and differences in microbiota composition now allow to discriminate phenotypes associated with producers and non-producers of specific metabolites [67,70,71]. As a matter of fact, inter-individual differences in the intestinal ecology may lead to differences in bioavailability, linked to specific metabolite production and, ideally, to differences in health benefits [72]. In the present study, the absorption of phenolics and the production of their metabolites is accompanied by a considerable inter-individual variability, plausibly due to the interaction between these compounds and the gut microbiota of the host. A clear example of this large variability among participants concerns the circulating concentration of phenyl-γ-valerolactones resulting from the catabolic transformations of catechins and procyanidins, operated mainly by Clostridium coccoides and Bifidobacterium spp. [73]. By analyzing the absorption curves of each participants, five subjects resulted as abundant producers of phenyl-γ-valerolactones, whereas the remaining subjects produced only extremely small amounts of these compounds (data not shown), suggesting marked variations in the colonic microflora of the individual volunteers. The inter-individual variability can affect not only the quantity of metabolites but also the timing of their appearance. For instance, the wide bars observed in the curves of kaempferol glucuronide are ascribable to the fact that these metabolites disappeared after 5 and 10 h in almost all the participants with the only exception of two subjects, who showed a second, later peak, probably due to colonic absorption. Considering this large variability, there is an increased interest in stratifying future study participants based on their polyphenol-metabolizing phenotypes (i.e., metabotypes) [72], and the present study supports the hypothesis that this variability should be carefully considered as a confounder of in vivo studies evaluating health effects of these phytochemicals. Finally, considering all the circulating detected metabolites, excluding those compounds whose origin could be attributed to endogenous precursors, such as hippuric acid, 4-hydroxyhippuric acid, and feruloylglycine [18,35], a "global" curve could be depicted to summarize the totality of quantified phenolic metabolites at every specific time point ( Figure 5). Observing this graph, some considerations can be drawn: (i) the absorption and metabolism of (poly)phenols in the first gastro-intestinal tract (1-2 h after capsule ingestion) is low when compared to what occurs in the colon (5-10 h after capsule consumption); (ii) the plasma curves of the phenolic metabolites clearly highlighted the deep interaction between these compounds and the gut microbiota; (iii) the beneficial effects attributed to the regular consumption of these capsules does not depend on very high concentration of phenolic metabolites circulating after their consumption. In fact, plasmatic metabolites rarely exceeded nanomolar concentrations, as previously reported [18,19]. However, the effect of the regular and long time intake of these products may modify the way our organism interacts with the contained phenolics, perhaps improving its ability to absorb some of them at small intestinal level. Moreover, a modulation of the colonic microbiota in the long run seems plausible, perhaps in the direction of improved transformations, leading to increased absorption of microbial metabolites. Finally, the large variability observed in this short acute absorption study should be taken into consideration in future interventions, as not all the recruited volunteers might deal with Juice PLUS+ ® phenolics in the same way.
It now appears clear that the extensive bioconversion of (poly)phenol compounds strictly depends on the characteristics of individual colonic microbiota, and differences in microbiota composition now allow to discriminate phenotypes associated with producers and non-producers of specific metabolites [67,70,71]. As a matter of fact, inter-individual differences in the intestinal ecology may lead to differences in bioavailability, linked to specific metabolite production and, ideally, to differences in health benefits [72]. In the present study, the absorption of phenolics and the production of their metabolites is accompanied by a considerable inter-individual variability, plausibly due to the interaction between these compounds and the gut microbiota of the host. A clear example of this large variability among participants concerns the circulating concentration of phenyl-γ-valerolactones resulting from the catabolic transformations of catechins and procyanidins, operated mainly by Clostridium coccoides and Bifidobacterium spp. [73]. By analyzing the absorption curves of each participants, five subjects resulted as abundant producers of phenyl-γ-valerolactones, whereas the remaining subjects produced only extremely small amounts of these compounds (data not shown), suggesting marked variations in the colonic microflora of the individual volunteers. The inter-individual variability can affect not only the quantity of metabolites but also the timing of their appearance. For instance, the wide bars observed in the curves of kaempferol glucuronide are ascribable to the fact that these metabolites disappeared after 5 and 10 h in almost all the participants with the only exception of two subjects, who showed a second, later peak, probably due to colonic absorption. Considering this large variability, there is an increased interest in stratifying future study participants based on their polyphenol-metabolizing phenotypes (i.e., metabotypes) [72], and the present study supports the hypothesis that this variability should be carefully considered as a confounder of in vivo studies evaluating health effects of these phytochemicals. Finally, considering all the circulating detected metabolites, excluding those compounds whose origin could be attributed to endogenous precursors, such as hippuric acid, 4-hydroxyhippuric acid, and feruloylglycine [18,35], a "global" curve could be depicted to summarize the totality of quantified phenolic metabolites at every specific time point ( Figure 5). Observing this graph, some considerations can be drawn: (i) the absorption and metabolism of (poly)phenols in the first gastro-intestinal tract (1-2 h after capsule ingestion) is low when compared to what occurs in the colon (5-10 h after capsule consumption); (ii) the plasma curves of the phenolic metabolites clearly highlighted the deep interaction between these compounds and the gut microbiota; (iii) the beneficial effects attributed to the regular consumption of these capsules does not depend on very high concentration of phenolic metabolites circulating after their consumption. In fact, plasmatic metabolites rarely exceeded nanomolar concentrations, as previously reported [18,19]. However, the effect of the regular and long time intake of these products may modify the way our organism interacts with the contained phenolics, perhaps improving its ability to absorb some of them at small intestinal level. Moreover, a modulation of the colonic microbiota in the long run seems plausible, perhaps in the direction of improved transformations, leading to increased absorption of microbial metabolites. Finally, the large variability observed in this short acute absorption study should be taken into consideration in future interventions, as not all the recruited volunteers might deal with Juice PLUS+ ® phenolics in the same way.

Conclusions
In conclusion, the consumption of Juice PLUS+ ® Vineyard, Fruit, Vegetable blend, containing 36 different fruits, vegetables, and berries, allowed the effective absorption of several (poly)phenolic compounds. The capsules therefore ensure potential health-promoting molecules to be potentially available to target tissues and organs, presumably becoming responsible for the previously observed health effects.