2014 Volume 37 Issue 5 Pages 779-784
The interaction between angiotensin II type 1 (AT1) receptor blockers (ARBs), such as losartan potassium (LO), candesartan (CA), and telmisartan (TE), and dietary fiber was studied as to the level of free ARB in vitro. When ARB was incubated with soluble (sodium alginate, pectin, and glucomannan) or insoluble (cellulose and chitosan) dietary fiber, the levels of free LO, TE, and CA decreased. This resulted only from mixing the dietary fiber with the ARBs and differed among the types of dietary fiber, and the pH and electrolytes in the mixture. The levels of free LO and TE tended to decrease with a higher concentration of sodium chloride in pH 1.2 fluid. These results suggest that it is important to pay attention to the possible interactions between ARBs and dietary fiber.
Dietary fiber is defined as all dietary ingredients that are not digested by digestive enzymes in humans.1) Recently, many people have started to use dietary fiber supplements and to pay more attention to fiber in the diet, which can have many health benefits.2,3) Dietary fiber reduces the risk of developing health problems, such as obesity,4,5) hypertension,6–8) constipation,9,10) type 2 diabetes,9,11,12) and hyperlipidemia.9,13,14) A high intake of fiber may lower mean blood pressure in people with hypertension,7) as well as in healthy subjects.8) There are also many possible interactions between dietary fiber and drugs.15) Dietary fiber reduces the bioavailability of some minerals, nutrients, and certain drugs,9,16) although the interaction between dietary fiber and drugs remains poorly understood.
Patients with hypertension will generally be treated with various antihypertensive drugs. In recent years, numerous orally active, selective angiotensin II type 1 (AT1)-receptor blockers (ARBs) have come into clinical use.15) Losartan potassium (LO), candesartan (CA), and telmisartan (TE) are AT1-selective ARBs (Fig. 1) currently being developed for the treatment of hypertension and heart failure.
Dietary fiber adsorbs drugs, thereby altering the quantity of the free from of the drugs.2,17) Thus, it is important to understand how dietary fiber interacts with ARBs. The bioavailability of orally administered drugs depends on their absorption, which is affected by several factors, such as the pKa of a drug, the pH at the absorptive site, and the gastric emptying rate. Within the gastrointestinal tract, dietary fiber behaves as a polymer matrix with variable physicochemical properties, including susceptibility to bacterial fermentation, water-holding capacity, cation-exchange function, and adsorptive function.9) The absorption of a certain drug or substance, when administered concomitantly with guar gum and other dietary fibers, may be modified by altered gastric emptying, lower diffusion from viscous matrices than from solutions, and cholestyramine-like actions.16) The hypocholesterolemic activity of lovastatin decreases in the presence of oat bran or pectin in humans.17) This suggests that dietary fibers reduce their absorption by the gut via a possible chelation mechanism or via altered gastric emptying.16) Simultaneous administration of sodium alginate and imipramine to rats decreased both the serum concentration and the pharmacological action of imipramine, by delaying its absorption.18)
In the present study, we examined whether ARB levels are influenced by dietary fiber by measuring changes in the amounts of LO, CA, and TE when ARB was incubated with several types of soluble (pectin, sodium alginate, and glucomannan) or insoluble (cellulose and chitosan) dietary fiber (Fig. 1) in pH 1.2 and pH 6.8 fluid in vitro.
LO, TE, and CA were obtained from LKT Laboratories, Inc. (St. Paul, MN, U.S.A.). We purchased cellulose from Sigma-Aldrich (St. Louis, MO, U.S.A.), and chitosan 100 (chitosan), glucomannan (from konjac), pectin (from apples), and sodium alginate 80–120 (sodium alginate) from Wako Pure Chemical Industries, Ltd. (Osaka, Japan).
EquipmentThe Centricut mini V-50 (molecular weight (MW): 50000) ultra-filter membrane was from Kurabo Industries, Ltd. (Osaka, Japan). pH was measured using a compact pH meter (Twin pH, Horiba Ltd., Kyoto, Japan).
Preparation of Mixtures of Free ARBs with Dietary FiberDietary fibers and LO were dissolved or suspended in the first (pH 1.2) (HCl/NaCl) or second (pH 6.8) (KH2PO4/NaOH) fluid, which are used for disintegration tests according to the Japanese Pharmacopoeia Sixteenth Edition, 0.05 M glycine–HCl buffer (pH 3.0; glycine/HCl), 0.1 M phosphate buffer (pH 7.0; NaH2PO4/Na2HPO4), or water. TE and CA were dissolved in a small volume of methanol or dimethylsulfoxide. TE was diluted with pH 1.2 fluid and CA was diluted with pH 6.8 fluid, because dilutions with the other solutions became turbid. In the experiment comparing LO to TE or CA, LO was dissolved in the solvent used for TE or CA. Each solution of ARB with or without dietary fiber was mixed 1 : 1 (v/v) and the pH was measured. The final concentration of each type of dietary fiber was as follows: cellulose 2.0%, chitosan 0.3%, sodium alginate 0.8%, pectin 0.2%, and glucomannan 0.5%; and that of each ARB was as follows: LO 0.11 mM, TE 0.15 mM, and CA 0.018 mM. These concentrations were based on the recommended doses for taking one dose of a supplement of each type of dietary fiber and the ARBs doses used in Japan. The mixed solutions were incubated at 37°C with shaking for 90 min. At the indicated incubation times, the mixed solutions were centrifuged in a Centricut mini V-50 (3000×g, 15 min). The solution that passed through the membrane was subjected to high-performance liquid chromatography (HPLC).
Measurements of LO, TE, and CA by HPLCThe HPLC system consisted of a controller (SLC-10AVp) equipped with variable-wavelength UV detector (SPD-10Avp), column oven (CTO-ASvp), and data processor (C-R6A; all from Shimadzu, Kyoto, Japan). The column (Inertsil ODS-2, 150×4.6 mm i.d., S 5 µm) was from GL Sciences, Inc. Japan (Tokyo, Japan). The column was maintained at 40°C (LO and TE) or 25°C (CA). The mobile phase was a mixture of H3PO4/NH4H2PO4 buffer–acetonitrile–methanol–triethylamine (60 : 30 : 10 : 0.04, v/v) for LO, 0.01 M phosphate buffer–methanol–acetonitrile (15 : 15 : 70, v/v) for TE, and acetonitrile–water–acetic acid (57 : 43 : 1, v/v) for CA. The flow rate was 1.0 mL/min (LO and TE) or 1.4 mL/min (CA). Detection was performed at 254 nm (LO and CA) or 210 nm (TE). Retention times were as follows; LO, 3.8 min; CA, 1.8 min; TE, 2.6 min. No other peak was detected under these conditions.
CalculationsThe percentage of free ARB was calculated using the following formula:
The control was ARB solution alone (no dietary fiber). In this study, free drug was defined as the amount of drug present in the solution that passed through the membrane.
Statistical AnalysisData were analyzed using the F-test for variance and Student’s or Welch’s t-test for significance using the Statcel2 software (OMS, Saitama, Japan). The p-values <0.05 were considered to indicate statistical significance.
The percentage of free LO at 0 min did not change during a 90-min incubation with several types of dietary fiber in water (Fig. 2). When incubated with cellulose and chitosan, the proportions of free LO were 85–100% during incubation for 90 min. Free LO disappeared immediately upon mixing with pectin and glucomannan. In the pH 1.2 or pH 6.8 fluid, the time course of the percentage of free LO was similar to the data in Fig. 2 (data not shown). The time courses of the percentage of free TE in the pH 1.2 fluid and CA in the pH 6.8 fluid were similar to that of LO, and the level at 0 min did not change during the 90-min incubation (data not shown).
Dietary fiber and LO were mixed in water and incubated at 37°C. Each data point represents the mean with S.D. (n=2–3).
The pH values of the mixtures of LO and dietary fibers were almost identical to those of the solutions used (Fig. 3). In water, the pH values of mixtures of LO and chitosan, sodium alginate, and pectin were 7.4, 7.5, and 3.5, respectively. Cellulose did not affect the percentage of free LO in solutions with several pH values. With chitosan, the percentage of free LO in the pH 1.2 fluid was lower than in the other solutions. Sodium alginate and pectin decreased the percentage of free LO to <20% and ca. 50%, respectively, in both the pH 6.8 fluid and pH 7.0 buffer. The percentage of free LO decreased at pH values close to 7.0, except when dissolved in water. Glucomannan decreased the percentage of free LO to <30% in all solutions used.
Dietary fibers and LO were mixed and incubated at 37°C for 30 min. Each column represents the mean with S.D. (n=3).
Figure 4 shows a comparison of the effect of dietary fibers on the level of free drug between LO and TE, and LO and CA. With sodium alginate, the percentage of free TE (80%) was significantly lower than that of LO (90%). The percentage of free CA was significantly higher than that of free LO with sodium alginate (CA: 54%, LO: 12%), pectin (CA: 86%, LO: 53%), and glucomannan (CA: 22%, LO: 0%).
Dietary fibers and LO, TE or CA were mixed and incubated at 37°C for 30 min. TE was soluble in only the pH 1.2 fluid, and CA was soluble in only the pH 6.8 fluid, while LO was soluble in all solvents used. The effects of dietary fibers on the free drug level were compared using LO as a standard. Each column represents the mean with S.D. (n=3). * p <0.05, significantly different from LO.
In the non-sodium chloride pH 1.2 fluid, the percentage of free LO after mixing with glucomannan was significantly higher than in the pH 1.2 fluid containing 0.2% sodium chloride, whereas the other dietary fibers did not affect the percentage of free LO irrespective of the presence or absence of 0.2% sodium chloride. In the pH 1.2 fluid containing 2% sodium chloride, the percentage of free LO was significantly decreased with chitosan, sodium alginate, and pectin compared to in pH 1.2 fluid (Fig. 5A).
The percentage of free TE (Fig. 5B) was not affected by the absence of sodium chloride in the pH 1.2 fluid. In the pH 1.2 fluid containing 2% sodium chloride, the percentage of free TE decreased significantly with chitosan, sodium alginate, pectin, and glucomannan compared to in the pH 1.2 fluid.
Dietary fibers and LO or TE were mixed and incubated at 37°C for 30 min. Each column represents the mean with S.D. (n=3). * p <0.05, significantly different from the pH 1.2 fluid.
We examined the effect of dietary fibers on the levels of free ARBs by mixing them in vitro. The levels of free LO, TE, and CA were decreased by mixing with several dietary fibers. The degree of change differed among the dietary fiber types and the pH of the mixture. The levels of free ARBs were highest in the order cellulose>sodium alginate=pectin>chitosan>glucomannan in the pH 1.2 fluid, and cellulose>chitosan>pectin>sodium alginate>glucomannan in the pH 6.8 fluid. Moreover, the free LO, TE, and CA levels decreased immediately on mixing with several dietary fibers, and then did not change during the following 90-min incubation. This indicates that the coexistence of several dietary fibers with ARBs may decrease free ARB levels.
For the insoluble dietary fibers, cellulose resulted in no change in the level of free LO in the solutions examined, while chitosan decreased the free LO level in only the pH 1.2 solution. The decreases in free LO upon mixing with the soluble dietary fibers, sodium alginate and pectin, depended on the pH of the solution. The percentage of free LO decreased at pH values close to 7.0, except when dissolved in water. Glucomannan decreased the level of free ARBs in all solutions examined. The composition of the buffer made little difference at pH values ca. 7.0. However, at pH 1.2, the sodium chloride concentration influenced the free LO and TE levels. From these results, we conclude that the pH and electrolyte content of the solution may be important factors in the effect of dietary fibers on free LO levels in vitro.
Multiple linkages form between dietary fibers and surrounding molecules.9) These interactions include ionic bonds, hydrogen bonds, and weaker hydrophobic and dispersion forces, and may affect both mineral and steroid absorption. Continuous change is also observed in the structure and function of the matrix, depending on the changes in the surrounding pH and osmolality. Dietary fiber has cation- and anion-exchange properties.9) With sodium alginate and pectin, uronic acid groups affect the adsorption of minerals, such as iron and calcium. Chitosan has amino groups, which can play a role in anion exchange. LO and TE have one anionic group. Thus, chitosan would be positively charged at low pH and attracted to LO and TE. With chitosan, the level of free LO was decreased in a pH 1.2 solution, and returned to the original level when the pH of the solution was shifted to pH 6.8 (data not shown). This suggests that the interaction at low pH would be re-dissociable when the pH of the solution returned to neutral. Regarding the relationships between amino acids in the AT1 receptor and LO, the tetrazole moiety of LO interacts primarily with Lys199 and secondarily with His256 in the AT1 receptor.19) The amino group of chitosan may also affect ARB interactions.
The percentages of free drug between LO and TE and between LO and CA with dietary fiber were slightly different. The percentage of free TE in the pH 1.2 fluid was lower than LO only with sodium alginate. The percentage of free drug was higher with CA than LO with sodium alginate, pectin, and glucomannan in the pH 6.8 fluid. LO may be more susceptible to reduced efficacy due to dietary fiber binding than CA, because the decrease in the free LO level was greater than that of CA. The differences in the structure of CA and LO could influence the level of free drug. The behavior of TE and/or CA toward dietary fiber also seems to differ slightly. LO and TE have one anionic group, while CA has two. The differences in some pharmacological activities among ARBs may depend on their chemical structures.20)
In mixtures with sodium alginate or glucomannan, the decrease in the level of free LO in the pH 6.8 fluid, which is similar to the pH in the small intestine, was greater than in the pH 1.2 fluid, which is similar to the pH in the stomach. LO and CA cilexetil are absorbed mainly in the jejunum,21) intestine and small intestine,22) respectively. Dietary fiber may affect the absorption of these drugs in the intestine. In the present study, free LO and TE tended to decrease with a higher concentration of sodium chloride in the pH 1.2 fluid. Hence, this may also occur in gastric fluid, which contains a high concentration of sodium salt. Foods with high levels of sodium salt could also influence the interaction between drugs and dietary fiber. In a previous study, a high-salt diet increased systolic blood pressure (SBP) in spontaneously hypertensive rats (SHRs) compared to normal controls. In addition, administration of LO by gavage reduced SBP in SHRs fed a normal-salt diet but not in SHRs fed a high-salt diet.23) We consider that these results have been caused by the fiber in the LO administered to the high-salt group, although the type of dietary fiber used was not reported.
There are many reports of the effects of food on the bioavailability of ARBs. Nakashima et al.24) reported that the area under concentration curves (AUCs) of LO and its metabolite decreased by 80% and 86%, respectively, in the fed state, and a decreased Cmax and prolonged Tmax were observed. In another study, the administration of a tablet of TE with food reduced the bioavailability of a 40-mg dose of TE by 6%; thus, TE may be taken with or without food.25) Irie et al.26) reported that with the administration of a capsule of TE in the fed state, Cmax and AUC were decreased and Tmax was prolonged. The rate of absorption was decreased by the intake of food. Although the pharmacokinetics differed during fasting, the effect on the decrease in blood pressure was unaffected. They concluded that it was unnecessary to consider the fed state for TE. For the CA cilexetil, for which there are data on a combination tablet with amlodipine, an effect of the fed state on its pharmacological activities was found; the Cmax and AUC increased with intake of food.15) We estimate that dietary fiber intake in a normal-salt diet would not influence the effects of ARBs. The consumption of dietary fiber in a high-salt diet, or of supplements containing large amounts of dietary fiber taken with ARBs, would be expected to have effects on the levels of free ARBs and to decrease the hypotensive effects of ARBs.
Dietary fibers are polysaccharides. The chemical structures of cellulose, chitosan, and sodium alginate comprise straight chains, while pectin and glucomannan have branched chains. The solubility and the rate of hydration of polysaccharides are related to the chemical structure, as the main chain is difficult to hydrate because water must first permeate the side chain.27) In water, the decrease in the free level of LO for pectin and glucomannan was larger than for cellulose, chitosan and sodium alginate, which have straight-chain chemical structures. The ability of LO solution permeating among dietary fiber may be related to the free LO level. We found that free ARB levels decreased particularly with soluble dietary fibers. Water-holding capacity, which is related to water solubility, could be an important factor in the decrease in free ARBs. High viscosity is one characteristic of soluble dietary fibers. The viscosities of 0.8% sodium alginate and 0.2% pectin at 20°C in water were 29.3 and 1.1 mm2/s, respectively (data not shown). The decrease in the free LO level in water was less marked with sodium alginate than with pectin. These data indicate that the effect of dietary fibers on the free ARB level is more sensitive to the pH and electrolyte concentration of the solution than its viscosity in vitro at the concentrations used. Glucomannan is a hydrophilic molecule that forms viscous solutions. In vivo, viscous solutions have been reported to delay gastric emptying and the absorption of minerals and nutrients from the small intestinal lumen.9) Dietary fiber in the gastrointestinal tract acts as a polymer matrix with variable physicochemical properties, such as water-holding capacity, cation-exchange, and adsorption.9) To determine the effects of dietary fibers on the free ARB level, it is important to consider the physical and chemical properties of the mixture in vitro, as well as the pharmacokinetics in vivo.
In conclusion, interactions between ARBs and dietary fibers occur, leading to decreased free ARB at least in vitro. Although the mechanism underlying these findings is unclear, it is important to pay attention to possible interactions between ARBs and dietary fibers. For example, Watanabe et al.18) reported that simultaneous administration of sodium alginate and imipramine in rats decreased the pharmacological effect of imipramine, which is in agreement with the results of another in vitro study.2) We are planning an in vivo study to examine the influence of dietary fibers and foods high in sodium salt on the pharmacokinetics of ARBs.
