Glucuronidation And Sulfation Of Flavonoids And The Effects Of Flavonoids On P-glycoprotein

Background and Purpose:Flavonoids are important natural organic compounds that widely distributed in the plant kingdom. Many foods such as fruits, vegetables and medicinal herbs contain flavonoids. Flavonoids have a wide range of pharmacological effects, including protecting cardiovascular, digestive and respiratory systems, anti-inflammatory, anticarcinogenic, estrogen-like effect, analgesic effect, and impacts on enzymes. Despite claims that flavonoids are beneficial to human health, it is well known that their bioavailabilities are poor (generally less than5%), which restricted the full use of flavonoids.Studies have shown that the poor bioavailability of flavonoids is mainly due to the wide range of phase II metabolism in vivo. The phase II metabolites of flavonoids were subsequently rapidly excreted by the efflux transporters in the organism cells, resulting in the minimal amount of flavonoids that can be used. The key work to improve the bioavailability of flavonoids may includes studies on the absorption, distribution, metabolism and excretion (ADME) of flavonoids, especially those on the phase Ⅱ mechanism of flavonoids and the interaction between flavonoids and their responsible efflux transporters. Glucuronidation and sulfation are the main two metabolic pathways of phase Ⅱ metabolism of flavonoids, which are catalyzed by UDP-Glucuronosyltransferases (Ugts) and sulfotransferase (Suits), respectively. The past researches mainly focused on the glucuronidation of flavonoids but paid less attention to the sulfation generally due to the extreme difficulties in detecting the sulfates of flavonoids in biological fluids.In the present study, several flavonoids were selected to investigate the glucuronidation and sulfation of flavonoids. A group of flavonoids sharing similar chemical structures were used to find out the relationship such as competing or collaborative between glucuronidation and sulfation in FVB mouse intestinal S9. A group of mono-hydroxyl and di-hydroxyl flavones were used to find out the species difference between human and mouse in human liver S9and C57mouse liver S9. Meanwhile, considering that the drug efflux transporters play an essential role in the metabolism of flavonoids, we chose P-glycoprotein (P-gp), which is the most widely distributed and have most substrates in vivo to investigate the effects of flavonoids on the expression and activity of P-gp in LS174T human colon cancer cells.Methods:1. Seven flavonoids with similar chemical structures,5-hydroxyflavone (5-HF),6-hydroxyflavone (6-HF),7-hydroxyflavone (7-HF),5,6-dihydroxyflavone (5,6-diHF),6,7-dihydroxyflavone (6,7-diHF),5,7-dihydroxyflavone (5,7-diHF), and5,6,7-trihydroxyflavone (5,6,7-THF) were selected to investigate the relationshiPbetween Ugts and Sults which catalyzing the phase II metabolism of flavonoids, as well as whether these two kinds of enzymes show specificity on the chemical structure.A FVB mouse intestinal perfusion model was used along with three small intestine S9fraction systems:(A) sulfation only,(B) glucuronidation only, or (C) simultaneous sulfation and glucuronidation or Sult-Ugt co-reaction. The mouse intestinal perfusion model referenced the rat in situ perfusion model which is already used maturely. The rat perfusion model is conducted on fully anesthetized animal and corresponds well to the human jejunal perfusion model. For flavonoids phase II metabolism, the rat intestinal perfusion model can only provide the glucuronidated metabolites but little sulfated metabolites, which makes the model difficult to simulate the sulfation of flavonoids in human gut. Thus, mouse intestinal perfusion model was established. To examine the excretion rates of flavonoids glucuronides and sulfates form small intestine and colon, we could1) compare the metabolic characteristics of flavonoids with different hydroxyl position, for example, the selectivity and specificity of Ugts and Sults on the hydroxyls;2) compare the excretion rates of flavonoids glucuronides and sulfates form small intestine and colon, and find out the different distribution of Ugts/Sults between small intestine and colon; and3) compare the excretion rates of flavonoids glucuronides and sulfates form small intestine and colon, and find out the different distribution of efflux transporters between small intestine and colon.Three small intestine S9fraction systems were created.1) Sulfation single reaction system. In this reaction system, Sults, which catalyze the sulfation of flavonoids, were provided by small intestine S9fractions. PAPS was the sulfonic acid grouPdonor. The reaction condition was held by the KPI buffer, and the whole reaction was conducted under a37℃water bath. The reaction was stopped by adding cold I.S. solution after a certain time. Samples were processed and analyzed by UPLC. By examining the amounts of flavonoid metabolites generated in the reaction system, the reaction rates were calculated. The blank control reaction was carried out accompanying.2) Glucuronidation single reaction system. According to the sulfation single reaction system, change the PAPS to UDPGA, and change the reaction time and enzyme concentration on basis of the glucuronidation rate, keePother conditions remain unchanged and calculate the reaction rate of glucuronidation.3) Sult-Ugt co-reaction system.Reference to the above two reaction systems, put PAPS and UDPGA together to the reaction system, adjust the reaction time and enzyme concentration in order to achieve the requirements of the determination and keePother conditions unchanged. The sulfation and glucuronidation rates were calculated respectively. These reaction rates in combine system were compared with the rates in single reaction systems, and the similarities and differences of sulfation and glucuronidation in single and combined reaction systems were examined.The sulfation and glucuronidation rates in the S9reaction systems were fitted to the sulfates and glucuronides excretion rates in the perfusion model to analyze the consistence of the two models.2. Four mono-hydroxyl flavones,3-hydroxyflavone (3-HF),5-hydroxyflavone (5-HF),7-hydroxyflavone (7-HF),4’-hydroxyflavone (4’-HF), and three di-hydroxyl flavones,3,5-dihydroxylflavone (3,5-diHF),3,4’-dihydroxylflavone (3,4’-diHF),3,7-dihydroxylflavone (3,7-diHF), were selected to investigate the glucuronidation and sulfation characteristics of flavonoids in human liver S9by using the human liver S9incubation model. Besides, C57mouse liver S9incubation model was conducted to find the species difference between human and mouse. C57mouse intestinal perfusion model and C57mouse intestine S9incubation model were used to investigate the glucuronidation and sulfation of flavonoids in vivo and in vitro. The metabolites excretion rates in vivo and the metabolites formation rates in vitro were compared to analyze the consistence of the two models.3. Two flavonoids with different hydroxyl groups were selected to investigate their effects on the expression and activity of P-glycoprotein. LS174T cells were treated with5,7-diHF and5,6,7-THF for6days. After the treatment, total protein was collected and the expression level of P-gPwas analyzed by Western blot. Data was analyzed by Quantity One software and the statistic analysis was finished by SPSS13.0.Results:1. To investigate the formation rates of glucuronides and sulfates of the seven selected flavonoids, FVB mouse intestinal S9fractions were used. Both the glucuronidation and sulfation reactions were conducted at three substrate concentrations (10,20and40μM). Rates at20μM were used to illustrate the pattern.In mouse intestinal S9fraction, both Ugts and Sults were the most active against7-OH groups over other hydroxyl groups. All selected flavonoids containing7-OH grouPwere only glucuronidated and sulfated at7-OH even if other hydroxyl groups were presented. The glucuronidation rate at7-OH group was statistically faster (P<0.05) than reaction rates at other groups in both the single glucuronidation system and the coreaction system. As to flavonoids without7-OH in structure, such as5,6-diHF,5-OH and6-OH are likely to occur the binding reaction without significant difference. For the sulfation reactions in vitro, the fastest formation rate of sulfate at20μM was found in6,7-diHF to7-O-S, and this was also the fastest reaction in the perfused small intestine. For the glucuronidation reaction, the fastest formation rate was found in5,6,7-THF to7-O-G.The formation rates of glucuronides and sulfates of flavonoids in the simultaneous reaction system were also investigated. The rank order of glucuronidation rates was similar between the single reaction and the co-reaction system, while the sulfation rates showed statistical differences (P<0.05) when the two reactions were conducted simultaneously. The sulfation rates in co-reaction system were significantly higher than those in sulfation single reaction system. In the mouse intestinal perfusion model, the excretion rates of both sulfates and glucuronides of flavonoids in small intestine were significantly higher (P<0.05) than those in colon, indicating that the amount of Ugts and Sults, as well as efflux transporters in small intestine and colon are different. In the small intestine, the excretion rates of glucuronides of flavonoids were significantly higher (P<0.05) than those of sulfates except for6,7-diHF. While in colon, this pattern was reversed, implying that sulfation may be a more important metabolism than glucuronidation in the colon of mice.We plotted rates of glucuronidation and sulfation by mouse intestinal S9fractions against excretion rates of glucuronides and sulfates from the perfused intestine. The results indicated that the S9fractions could not accurately model glucuronidation in vivo, while the sulfation of flavonoids in the S9fractions still revealed useful information about sulfation in vivo (as the slope of correlation curve was0.6767). The reason for the inconsistence in glucuronidation may due to the efflux transporters in the intestine of whole animal.2. To investigate the characteristics of glucuronidation and sulfation of3-HF,5-HF,7-HF,4’-HF,3,5-diHF,3,4’-diHF and3,7-diHF, human liver S9and C57mouse liver and intestine S9were used. Both the glucuronidation and sulfation reactions were conducted at three substrate concentration (5,10and40μM). Rates at10μM were used to illustrate the pattern.The results showed that glucuronidation and sulfation occurred more easily on7-OH and4’-OH, while3-OH could not be sulfated at all. The comparison between reaction rates in human liver S9and mouse liver S9showed that there were statistical differences (P<0.05) between glucuronidation/sulfation rates in human liver S9and C57mouse liver S9. Sults were found more active in human liver S9than in mouse liver S9. As to Ugts, the differences were not consistent. For example, the glucuronidation rates of3,7-diHF in human liver S9was7-40times of those in C57mouse liver S9, while for4’-HF, the glucuronidation rates in human liver S9were lower than those in mouse liver S9. This indicates that species difference may vary depending on the hydroxyl groups in flavonoids.The regional metabolism of flavonoids in the mouse intestinal perfusion model was then compared with the glucuronidation and sulfation rates in the in vitro reaction systems. The correlation factor between sulfates excretion from the intestinal perfusion model and sulfates production in the in vitro S9fraction was0.9834, and a weak correlation was found between intestinal flavonoid glucuronides excretion and in vitro glucuronidation (with the correlation factor as0.7988).3. The Western blot analysis showed that after treatments by5,7-diHF and5,6,7,-THF for six days, the P-gp expression levels in LS174T cells were statistically different from that in normal cells (P<0.05).5,7-diHF at10μM (t=4.805, P=0.041) and20μM (t=6.315, P=0.024) could significantly inhibit the expression of P-gp, while5,6,7,-THF at50μM could up-regulate the P-gp expression significantly (t=-4.473, P=0.047).Conclusions:In conclusion, minor structural changes can significantly impact the disposition of flavonoids in intestine via the sulfation and glucuronidation pathways. Ugts and Sults from mouse intestinal S9fraction showed higher catalytic rates toward7-OH groups than other hydroxyl groups. The rates of glucuronidation and sulfation in human liver S9were statistically different from those in C57mouse liver S9. Activity of Sults in human liver S9was significantly higher than that in C57mouse liver S9. Flavonoids could impact the P-gp protein expression on human colon cancer cells LS174T cells to a certain extent.