Due to its potentially beneficial impact on human health, the polyphenol quercetin has come into the focus of medicinal interest. However, data on the bioavailability of quercetin after oral intake are scarce and contradictory. Previous investigations indicate that the disposition of quercetin may depend on the sugar moiety of the glycoside or the plant matrix. To determine the influence of the sugar moiety or matrix on the absorption of quercetin, two isolated quercetin glycosides and two plant extracts were administered to 12 healthy volunteers in a four-way crossover study. Each subject received an onion supplement or quercetin-4'-O-glucoside (both equivalent to 100 mg quercetin), as well as quercetin-3-O-rutinoside and buckwheat tea (both equivalent to 200 mg quercetin). Samples were analyzed by HPLC with a 12-channel coulometric array detector. In human plasma, only quercetin glucuronides, but no free quercetin, could be detected. There was no significant difference in the bioavailability and pharmacokinetic parameters between the onion supplement and quercetin-4'-O-glucoside. Peak plasma concentrations were 2.3 +/- 1.5 microg x mL(-1) and 2.1 +/- 1.6 microg x mL(-1) (mean +/- SD) and were reached after 0.7 +/- 0.2 hours and 0.7 +/- 0.3 hours, respectively. After administration of buckwheat tea and rutin, however, peak plasma levels were--despite the higher dose-only 0.6 +/- 0.7 microg x mL(-1) and 0.3 +/- 0.3 microg x mL(-1), respectively. Peak concentrations were reached 4.3 +/- 1.8 hours after administration of buckwheat tea and 7.0 +/- 2.9 hours after ingestion of rutin. The terminal elimination half-life was about 11 hours for all treatments. Thus, the disposition of quercetin in humans primarily depends on the sugar moiety. To a minor extent, the plant matrix influences both the rate and extent of absorption in the case of buckwheat tea administration compared with the isolated compound. The site of absorption seems to be different for quercetin-4'-O-glucoside and quercetin-3-O-rutinoside. The significance of specific carriers on the absorption of quercetin glycosides, as well as specific intestinal beta-glucosidases, needs to be further evaluated.

Aim: The two bacterial species, Eubacterium ramulus and Enterococcus casseli¯avus, which had previously been isolated from human faeces using the¯avonoid quercetin-3-glucoside as the growth substrate, were tested for their ability to utilize this compound in vivo. Methods and Results: Germ-free rats were associated with Eu. ramulus and subsequently with Ent. casseli¯avus and vice versa. Identi®cation and enumeration of the bacterial cell counts in faeces and intestinal contents were performed by whole cell¯uorescence in situ hybridization. Eubacterium ramulus and Ent. casseli¯avus occurred in caecal and colonic contents at cell counts of up to 10 10 g À1 dry weight. In the jejunum, only Ent. casseli¯avus was found (10 9 g À1 dry weight). Upon oral administration of 32 mmol quercetin-3-glucoside, quercetin was detected in the faeces and urine of germ-free rats (2Á2 Â 10 À1 À8Á1 Â 10 À1 mmol 24-h À1 faeces collection and 1Á0 Â 10 À2 À2Á8 Â 10 À1 mmol 24-h À1 urine collection, respectively) and of rats monoassociated with Ent. casseli¯avus (7Á9 Â 10 À1 À2Á7 mmol 24-h À1 faeces and 1Á0 Â 10 À1 À5Á9 Â 10 À1 mmol 24-h À1 urine, respectively). In contrast, the faeces and urine of rats associated with Eu. ramulus contained 3,4-dihydroxyphenylacetic acid (4Á7 Â 10 À2 À3Á6 mmol 24-h À1 faeces and 2Á4 Â 10 À2 À1Á0 mmol 24-h À1 urine, respectively) but only low, or undetectable, concentrations of faecal quercetin (up to 9Á3 Â 10 À2 mmol 24-h À1 faeces; detection limit 2Á5 Â 10 À2 mmol). Urinary quercetin concentrations varied markedly from undetectable amounts up to 1Á0 mmol 24-h À1 urine (detection limit 1Á0 Â 10 À2 mmol). Isorhamnetin was found in the urine of all animals independent of their bacterial status. There were no signi®cant differences between the groups (2Á0 Â 10 À2 À2Á8 Â 10 À1 mmol 24-h À1 urine). In complete intestinal tissues of animals, associated with both species, quercetin-3-glucoside and its metabolites were detected by a more sensitive and selective method at concentrations that were two to three orders of magnitude lower than in faeces or urine. Conclusions: These results indicate that Eu. ramulus may be a key organism for the bacterial transformation of¯avonoids in the gut.

No abstract

A human study was performed to investigate the influence of different dietary flavonoids on the faecal population of the flavonoid-degrading bacterium Eubacterium ramulus. Twenty-eight healthy subjects, divided into five groups, consumed for 6 days a flavonoid-free diet. On day 4 of this intervention period the study participants ingested a single dose of quercetin (14 mg kg(-1) body mass (bm)), rutin (pure or as buckwheat leaves, 28 mg kg(-1) bm) or a placebo. During the first 3 days of the intervention period the total faecal flora decreased by 61-88% and the population of E. ramulus by 40-75%. The oral intake of the flavonoids resulted in a dramatic increase in the faecal E. ramulus population. The relative proportion of E. ramulus rose from minimally 0.2% (on day 1) to maximally 6.9% of the total flora on day 8. The faecal concentrations of degradational products were highly variable but reflected the bacterial breakdown of the flavonoids.

Purified flavonoid extract from illuminated parsley (Petroselinum crispum (Mill.) Nym.) cell culture was administered by gavage to Wistar rats. The dose corresponded to 6.9 mg flavonoids on aglycone base/kg body mass. Segments of the gastrointestinal wall from stomach to colon, their luminal contents, and liver and kidneys were collected at time intervals between 1 and 12 h and investigated by HPLC of the respective extracts for flavonoids. The spreading of the flavonoids was accompanied by partial deglycosylation that began already in the stomach where at first quercetin and later apigenin, chrysoeriol and isorhamnetin aglycones were detected. We got evidence of flavonoid absorption by the stomach that does not require the liberation of aglycones. Due to obvious differences in metabolization and absorption rates the composition and the content of flavonoids changes in the gastrointestinal segments and their contents with time. Flavonoids could be detected neither within the gastrointestinal lumen after 12 h nor in the kidneys at any time. But traces of flavonoids were found in the livers at 1.5 and 12 h.