Risk assessment of quercetin dihydrate and rutin in food supplements – “Other substances”

Abstract

Background “Other substances” are substances that have a nutritional and/or physiological effect and are not vitamins or minerals. Excessive intake of certain "other substances" may be associated with health risks. The Norwegian Food Safety Authority asked the Norwegian Scientific Committee for Food and Environment (VKM) to assess whether rutin (CAS number 153-18-4) and quercetin dihydrate (CAS number 6151-25-3) from Sophora Japonica (bud/flower) could pose a health risk for the Norwegian population when taken daily as oral supplements as 5 mg rutin for children from 4 years of age, 25 mg rutin for adults from 18 years of age and 500 mg quercetin dihydrate for adults from 18 years of age. Quercetin (3,3′,4′,5,7-pentahydroxyflavone) is the prototypical representative of the flavonol subclass of flavonoids. Plants contain quercetin as an aglycone or as various conjugated forms such as glycosides, including isoquercitrin (quercetin 3- O-β-D- glucopyranoside; IQ) and rutin (quercetin 3- O-β-D-rutinoside). The composition of the diverse quercetin glycosides varies between different food plants, whereas quercetin is often present in food supplements as aglycones. Rutin is composed of quercetin and rutinose, a disaccharide of rhamnose and glucose. The low oral bioavailability of quercetin and rutin, caused by their low hydrophilic solubility, can be increased by glucosyl conjugation of these molecules, i.e. by addition of various moieties such as lecithin (a mixture of various glycerophospholipids) or sugars. Quercetin Phytosome® is formulated with sunflower lecithin in a 1:1 weight ratio. Enzymatically modified quercetin glycoside (EMIQ) is an α-glycosyl isoquercitrin (α- oligoglucosyl quercetin 3- O-glucoside). EMIQ is produced through enzymatic conversion of rutin into a mixture of isoquercitrin and its α-glucosyl derivatives with 1– 10 α-glucose moieties connected. A hydroxyethylation reaction has been used to add a hydroxyethyl chain on the hydroxyl groups of rutin to form O-(β-hydroxyethyl)- rutosides (HER). As there are four hydroxyl groups available, mono-, di-, tri- and tetra- HER in various isomeric forms are present in commercial products. In the included studies, all of these chemical forms were studied, since they all are converted to quercetin aglycone in the body.Methods A broad search for review studies was conducted to identify human studies and/or animal toxicity studies on adverse health outcomes related to quercetin or rutin. However, because most reviews did not present much detail on the safety assessment in the human studies, a second systematic search for randomized controlled trials was performed in five databases. From these searches, 2526 records were obtained, which were screened at the level of title/abstract. From these, 140 randomised controlled trials were obtained in full-text and screened against the eligibility criteria. Among these, 45 studies were found to meet the inclusion criteria. Publications that in any way mentioned ‘safety’ or ‘adverse events/effects’ or ‘side-effects’ or had analysed blood or urine or any biological measure with an expressed intent to evaluate safety were considered to meet the inclusion criteria, otherwise they were excluded. Assessment of quercetin and rutin - Norwegian Scientific Committee for Food and Environment No specific adverse health outcomes related to quercetin or rutin had been identified before the literature searches. Therefore, the included papers were sorted according to how adverse effects were obtained or measured. Fifteen publications with data based on objectively measured outcomes, such as results of analyses of blood or urine, were defined as Category 1. Eight publications with only data on adverse health effects obtained or registered by subjective methods, such as self-reported adverse effects/adverse events/side-effects by participants, were defined as Category 2. Additionally, 14 publications in Category 1 included also subjectively reported adverse effects/events, i.e. altogether 23 publications reported some data on adverse effects/events or lack of such. Twenty-two publications that only mentioned briefly “no adverse effects/events/side-effects reported”, without any information on how such data were recorded or any details on the results, were defined as Category 3. The publications in Category 1 and 2 were evaluated for risk of bias using the OHAT tool and used in the risk assessment. These OHAT evaluated publications were classified as Tiers 1, 2 or 3, which represent low, moderate and high risk of bias, respectively.Results from the systematic review of adverse effects Among the Category 1 publications, twelve reported that parameters in blood or urine were analysed with the objective to investigate safety of the treatment without reporting any adverse effects. In three studies assessing treatments with 240 mg quercetin per day for 3 months, 150 mg quercetin dihydrate per day for 8 weeks and 150 mg quercetin dihydrate per day for 6 weeks, each reported a single sporadic effect in the direction of adversity but of mild severity (decrease in high density lipoprotein (HDL)-cholesterol, increased levels of tumor necrosis factor (TNF)-α and decreased glutathione (GSH) level, respectively). The first two effects were not supported by other endpoints measured in the same study and the third effect could be interpreted as not adverse. In addition, these effects had not been reported in the other included studies, which indicated that they may be chance findings. Therefore, no serious hazards were identified among these Category 1 results. Among the Category 2 publications, five studies stated that no adverse effects were reported by the participants or observed. Eight publications reported adverse event/effects with at least some detail about the observations, however, the reported effects/events were all of minor severity and were either considered not to be study drug-related, the type and numbers of reported effects/events were similar between the treatment and control groups, they occurred in only one person, were self- resolving or did not show a dose-response. Therefore, no serious hazards were identified among these Category 2 results. In addition to the systematic approach used to identify and characterize adverse effects observed in the human randomized controlled trials, additional information on absorption, distribution, metabolism and excretion (ADME), and toxic effects of the studied substances - mostly from animal studies, was included from various sources, not obtained in a systematic way.Toxicokinetics Quite a lot of data on toxicokinetics/ADME were available for the included substances, both from human pharmacokinetic studies and animal studies. The sugar moieties of the quercetin glycosides may modulate the quercetin bioavailability. These substances Assessment of quercetin and rutin - Norwegian Scientific Committee for Food and Environment are converted to quercetin aglycone, which is easier absorbed than substances such as rutin, probably by passive diffusion over the intestinal epithelium or directly via an intestinal transporter molecule. Quercetin may also be subsequently degraded by the colonic microbiota, mainly into different phenolic acids. After absorption, quercetin is extensively metabolised in enterocytes and liver, and it may be glucuronidated, sulfated and/or methylated. In the blood, primarily these quercetin conjugates are found, with only very low levels of the aglycone form. Quercetin is found in some tissues mostly as aglycone, while in other tissues, the unconjugated quercetin is present in smaller proportions. Ingested quercetin is rapidly excreted via urine and feces, and may also be metabolised and excreted via the lungs as CO2. There is interindividual variation in the quantitative ratio of the various metabolites formed and in the rate of absorption and excretion of quercetin, depending on genetic variation, individual antioxidative status and co-administration of other dietary components such as fiber or fat.Toxicological data Based on the available literature, mutagenic and genotoxic effects have been reported in some assays in vitro, but quercetin, rutin and the related substances EMIQ and IQ were not found to be genotoxic in vivo for the doses evaluated in this risk assessment. The discrepancy between in vitro mutagenicity and genotoxicity, and lack of genotoxic or carcinogenic effects in vivo, may be related to the transient nature and the instability of the quercetin quinone methide adducts, as well as various other mechanisms. In a 2-year feeding study by the US National Toxicology Program (1992), there was some evidence of carcinogenic activity of quercetin in male rats receiving up to 1900 mg/kg body weight per day of quercetin based on an increased incidence of renal tubule cell adenomas, but there was no evidence of carcinogenic activity of quercetin in female rats in the same doses. The renal tumor development may be associated with or may be a consequence of the chronic progressive nephropathy occurring only in male rats, with probably no or only little relevance for extrapolation to humans. Other long-term rat studies, two on quercetin and two on EMIQ, did not report any carcinogenic effects. The International Agency for Research on Cancer (IARC) concluded that “quercetin is not classifiable as to its carcinogenicity to humans” (Group 3). Quercetin can most likely cross the placenta since effects on the fetus have been observed after maternal exposure in mice and it is shown for several other flavonoids. Rutin may be able to bind to the estrogen receptor and exert estrogen-like effects. The available studies in mice, rats and rabbits did not find reprotoxic effects of quercetin after exposure during gestation. However, in one experiment with female mice exposed for 9 months during reproductive age, 60% reduction in number of litters was observed after exposure to 5 mg/kg body weight of quercetin for 9 months. In a case-control study, O-(β-hydroxyethyl)-rutoside (HER) treatment with oral doses of 900-1000 mg HER per day for 3-5 weeks during the second and/or third month of pregnancy was found to be associated with a higher risk of certain congenital abnormalities. Similarly, malformation of the limbs the offspring (syndactyly) was found Assessment of quercetin and rutin - Norwegian Scientific Committee for Food and Environment in mice after exposure to approximately 67 mg/kg body weight of quercetin for about two weeks during gestation. The only available information on allergenicity, sensitization and irritation was that EMIQ was not a skin sensitizer or irritant in mice.Uncertainty Among the publications in Category 1, ten were evaluated as having low risk of bias and five with moderate risk of bias, and among the publications in Category 2, six were evaluated as having low risk of bias and two with high risk of bias. The main objective in most of these randomized controlled trials was not to examine adverse effects, but beneficial effects. Heterogeneity or mechanisms of action could not be evaluated for most publications due to the lack of reported adverse effects. Furthermore, the publications included were heterogeneous both in relation to the outcomes examined and study duration. In addition to the administered dose(s), the actual exposure to quercetin or rutin is determined by their purity and stability. Information on purity or stability was rarely stated in the available publications, which contribute to the uncertainty of the doses actually causing the reported effects or the lack of affects. To be able to use the included studies in the risk assessment, the given doses of quercetin- and rutin-related substances were recalculated to the corresponding dose of the common substance quercetin aglycone, into which all the related substances are metabolized. However, mostly only one pharmacokinetic study was available per modified substance and, therefore, there is some uncertainty regarding the general applicability of this information, affecting our calculations of quercetin and rutin exposure. Information from human studies with other designs than randomized controlled trials was not systematically included in this risk assessment.Conclusions Based on a systematic review of randomized controlled trials examining effects of quercetin or rutin, which resulted in the inclusion and evaluation of 23 publications with adult participants, VKM considers that exposure to the three requested doses (500 mg quercetin dihydrate, 5 and 25 mg rutin) taken daily for at least up to 3 months in adults does not pose a health risk. Two of the included publications found no adverse effects after administration for up to 6-10 months. No acute toxicity of a single or short-term (5-7 days) exposure was indicated by the results. No specific treatment-related and dose-dependent adverse effects could be identified from the included studies which reported a few outcomes in a potentially adverse direction among the parameters measured in blood or urine, and adverse effects/events/side-effects reported by the participants. By expert judgement, the weight of evidence for absence of adverse effects related to quercetin or rutin in the 23 included randomized controlled trials is judged to be “moderate”. VKM was also requested to consider if 5 mg rutin per day could pose a health risk for children from 4 years of age. None of the included studies investigated exposure Assessment of quercetin and rutin - Norwegian Scientific Committee for Food and Environment specifically in children. None of the included studies compared susceptibility to adverse effects in adults and children. Based on the results for adults and supporting evidence from one excluded study with higher daily doses (approximately 40-70 mg rutin plus 100-150 mg quercetin) for 6.5 months, VKM concludes that 5 mg rutin per day up to 6.5 months will not cause adverse effects in children other than possibly transient irritability. Some data indicated that O-(β-hydroxyethyl)-rutoside (HER) and quercetin may induce teratogenic effects in offspring, shown in humans and mice, respectively. Regarding these teratogenic effects, they were observed at similar doses in humans, but at a higher dose in mice, compared with the dose of quercetin dihydrate (recalculated to quercetin aglycone) VKM was requested to evaluate. Because of the lack of sufficient data on pregnant women and their fetuses, and the lack of data on breast-feeding women and their infants, as well as on children in general, it is not known whether these groups may potentially be more susceptible to these substances than adults. Some data indicate that persons with chronic nephropathy or estrogen-dependent cancer may be vulnerable to adverse effects of quercetin.Data gaps There were few publications having evaluation of adverse effects of quercetin and rutin as the main objective. Furthermore, many of the included studies were small and of short duration, even some with single dose administration. Very little data were found on effects of quercetin and rutin on children and pregnant women, and no data on adolescents and breastfeeding women.