Microbial metabolites of ingested caffeic acid are absorbed by the monocarboxylic acid transporter (MCT) in intestinal Caco-2 cell monolayers

It was previously reported that m-coumaric acid, m-hydroxyphenylpropionic acid (mHPP), and 3,4-dihydroxyphenylpropionic acid (DHPP) are major metabolites of ingested caffeic acid formed by gut microflora and would be transported by the monocarboxylic acid transporter (MCT). We have directly measured their absorption characteristics in Caco-2 cells using a coulometric detection method involving HPLC-ECD. The proton-coupled directional transport of m-coumaric acid, mHPP, and DHPP was observed, and the transport was inhibited by an MCT substrate. The permeation of m-coumaric acid and mHPP was concentration-dependent and saturable: The Michaelis constant for m-coumaric acid and mHPP was 32.5 and 12.9 mM, respectively, and the maximum velocity for m-coumaric acid and mHPP was 204.3 and 91.2 nmol (min)(-1) (mg protein)(-1), respectively. By contrast, the permeation of DHPP was nonsaturable even at 30 mM and was inversely correlated with the paracellular permeability of Caco-2 cells. Our results demonstrate that these compounds are absorbed by the MCT, although DHPP is mainly permeated across Caco-2 cells via the paracellular pathway. MCT-mediated absorption of phenolic compounds per se and their colonic metabolites would exert significant impact on human health.     

Phenolic acids are absorbed from the rat stomach with different absorption rates

The intestinal absorption characteristics of phenolic acids (PAs) have been elucidated in terms of their affinity for the monocarboxylic acid transporter (MCT). Recently, the involvement of the stomach has been implicated in the absorption of polyphenols. The present work demonstrates that the gastric absorption efficiency of each PA is apparently different between various PAs. Various PAs with different affinities for MCT were administered (2.25 mumol) to rat stomach, and then the plasma concentration of the PA was measured. The plasma concentration of ferulic acid (FA) peaked 5 min after administration in the stomach. At 5 min after administration, the plasma concentration of each PA increased in the order: gallic acid = chlorogenic acid < caffeic acid < p-coumaric acid = FA. This order matches their respective affinity for MCT in Caco-2 cells, which we have demonstrated in previous studies. These results indicated that MCT might be involved in the gastric absorption of PAs, similar to the intestinal absorption.     

Transepithelial transport of microbial metabolites of quercetin in intestinal Caco-2 cell monolayers

m-Hydroxyphenylacetic acid (mHPA), 3,4-dihydroxyphenylacetic acid (DHPA), and 4-hydroxy-3-methoxyphenylacetic acid (HMPA) are major microbial metabolites of quercetin. After administration of quercetin to human subjects, these metabolites are readily detected in blood and urine. mHPA, DHPA, and HMPA are thought to exert protective biological activity within the body due to their antioxidant properties. However, very little work has been published concerning their absorption. I have examined the absorption characteristics of the quercetin metabolites in Caco-2 cells by a coulometric detection method using HPLC-ECD. All of them exhibited nonsaturable transport in Caco-2 cells up to 30 mM, whereas HMPA and mHPA also showed proton-coupled polarized absorption. The proton-coupled directional transport of HMPA and mHPA was inhibited by the substrate of the monocarboxylic acid transporter (MCT). A considerable amount of apically loaded HMPA and mHPA was taken up and transported through to the basolateral side, while almost all of the apically loaded DHPA was retained on the apical side. Furthermore, the transepithelial flux of DHPA was inversely correlated with the paracellular permeability of Caco-2 cells, although those of HMPA and mHPA were almost constant. These results indicate that transport of DHPA was mainly via paracellular diffusion, although HMPA and mHPA were absorbed to some extent by the MCT.     

Uptake and metabolism of hydroxycinnamic acids (chlorogenic, caffeic, and ferulic acids) by HepG2 cells as a model of the human liver

Hydroxycinnamic acids are antioxidant polyphenols common in the human diet, although their potential health benefits depend on their bioavailability. To study the hepatic uptake and metabolism, human hepatoma HepG2 cells were incubated for 2 and 18 h with caffeic, ferulic, and chlorogenic acids. Moderate uptake of caffeic and ferulic acids was observed versus a low absorption of chlorogenic acid, where esterification of the caffeic acid moiety markedly reduced its absorption. Methylation was the preferential pathway for caffeic acid metabolism, along with glucuronidation and sulfation, while ferulic acid generated glucuronides as the only metabolites. Ferulic acid appeared to be more slowly taken up and metabolized by HepG2 cells than caffeic acid, with 73% and 64% of the free, nonmetabolized molecules detected in the culture medium after 18 h, respectively. In conclusion, hydroxycinnamic acids can be metabolized by the liver as suggested by the results obtained using HepG2 cells as a hepatic model system.     

Transepithelial transport of alpha-lipoic acid across human intestinal Caco-2 cell monolayers

Alpha-lipoic acid (LA) is used in dietary supplements or food with antioxidative functions. The mechanism for the intestinal absorption of alpha-lipoic acid was investigated in this study by using human intestinal Caco-2 cell monolayers. LA was rapidly transported across the Caco-2 cell monolayers, this transport being energy-dependent, suggesting transporter-mediated transport to be the mechanism involved. The LA transport was strongly dependent on the pH value, being accelerated in the acidic pH range. Furthermore, such monocarboxylic acids as benzoic acid and medium-chain fatty acids significantly inhibited LA transport, suggesting that a proton-linked monocarboxylic acid transporter (MCT) was involved in the intestinal transport of LA. The conversion of LA to the more antioxidative dihydrolipoic acid was also apparent during the transport process.     

Absorption of chlorogenic acid and caffeic acid in rats after oral administration

Absorption of orally administered chlorogenic acid (5-caffeoylquinic acid) and caffeic acid in rats was studied to obtain plasma pharmacokinetic profiles of their metabolites. Rats were administered 700 micromol/kg body weight of chlorogenic or caffeic acid, and blood was collected from the tail for 6 h after administration. Ingested caffeic acid was absorbed from the alimentary tract and was present in the rat blood circulation in the form of various metabolites. On the other hand, only traces of metabolites, supposedly caffeic and ferulic acids conjugates, were detected in rat plasma for 6 h after chlorogenic acid administration. Chlorogenic acid and small amounts of caffeic acid were found in the small intestine for 6 h after chlorogenic acid administration. These results suggest that chlorogenic acid is not well absorbed from the digestive tract, unlike caffeic acid, and subject to almost no structural changes to the easily absorbed forms.     

Tea polyphenols inhibit the transport of dietary phenolic acids mediated by the monocarboxylic acid transporter (MCT) in intestinal Caco-2 cell monolayers

It was previously reported that a fluorescent marker dye, fluorescein, is transported via the monocarboxylic acid transporter (MCT). Fluorescein transport was competitively inhibited by MCT substrates such as ferulic and salicylic acids. Tea polyphenols, in particular, epigallocatechin gallate (EGCg) and epicatechin gallate (ECg), inhibited the transport of fluorescein. Tea polyphenols also inhibited the transport of salicylic and ferulic acids, suggesting tea polyphenols might be substrates of MCT. However, the transepithelial flux of tea polyphenols was much lower than that of the MCT substrates and was inversely correlated with the paracellular permeability of Caco-2 cell monolayers. These findings suggest that tea polyphenols are not substrates but inhibitors of MCT. Furthermore, the transepithelial transport of these polyphenols is mainly via paracellular diffusion. However, directional transport of ECg and EGCg from the basolateral to the apical side was observed, indicating that the behavior of tea polyphenols in the intestinal epithelium is complex.     

Pharmacokinetic study of caffeic and rosmarinic acids in rats after oral administration

p-Coumaric and ferulic acid are actively taken up by monocarboxylic acid transporter (MCT), whereas gallic acid, caffeic acid (CA), and rosmarinic acid (RA) are absorbed by paracellular diffusion in human intestinal Caco-2 cells, although CA has low affinity for MCT. We previously demonstrated that p-coumaric acid has a much higher absorption efficiency than gallic acid in rats, owing to the MCT-mediated absorption of p-coumaric acid in vivo (J. Agric. Food Chem. 2004, 52, 2527-2532). Here, absorption of orally administered CA and RA in rats has been studied to investigate their intestinal absorption characteristics and pharmacokinetics in vivo and to compare the results with those of p-coumaric and gallic acids obtained under identical conditions. Rats were given 100 micromol/kg body weight of CA and RA, and blood was collected from the portal vein and abdominal artery after administration. CA, RA, and their metabolites were quantified by a coulometric detection method using HPLC-ECD. The serum concentration of intact CA and RA in the portal vein peaked at 10 min after administration, with a C(max) of 11.24 micromol/L for CA and 1.36 micromol/L for RA. The area under the curve (AUC) for intact CA and RA in the portal vein was calculated from the serum concentration-time profile to be 585.0 and 60.4 micromol min L-1, respectively. The absorption efficiency of CA was about 9.7-fold higher than that of RA. Overall, the absorption efficiency of these compounds in vivo increases in the order: gallic acid = RA < CA < p-coumaric acid, which is in good agreement with results obtained in Caco-2 cells in vitro.     

Characterization of metabolites of hydroxycinnamates in the in vitro model of human small intestinal epithelium caco-2 cells

Hydroxycinnamic acids are antioxidant phenolic compounds which are widespread in plant foods, contribute significantly to total polyphenol intakes, and are absorbed by humans. The extent of their putative health benefit in vivo depends largely on their bioavailability. However, the mechanisms of absorption and metabolism of these phenolic compounds have not been described. In this study, we used the in vitro Caco-2 model of human small intestinal epithelium to investigate the metabolism of the major dietary hydroxycinnamates (ferulate, sinapate, p-coumarate, and caffeate) and of diferulates. The appearance of metabolites in the medium versus time was monitored, and the various conjugates and derivatives produced were identified by HPLC-DAD, LC/MS, and enzyme treatment with beta-glucuronidase or sulfatase. Enterocyte-like differentiated Caco-2 cells have extra- and intracellular esterases able to de-esterify hydroxycinnamate and diferulate esters. In addition, intracellular UDP-glucuronosyltransferases and sulfotransferases existing in Caco-2 cells are able to form the sulfate and the glucuronide conjugates of methyl ferulate, methyl sinapate, methyl caffeate, and methyl p-coumarate. However, only the sulfate conjugates of the free acids, ferulic acid, sinapic acid, and p-coumaric acid, were detected after 24 h. The O-methylated derivatives, ferulic and isoferulic acid, were the only metabolites detected following incubation of Caco-2 cells with caffeic acid. These results show that the in vitro model system differentiated Caco-2 cells have the capacity to metabolize dietary hydroxycinnamates, including various phase I (de-esterification) and phase II (glucuronidation, sulfation, and O-methylation) reactions, and suggests that the human small intestinal epithelium plays a role in the metabolism and bioavailability of these phenolic compounds.     

Absorption of phenolic acids in humans after coffee consumption

Despite extensive literature describing the biological effects of polyphenols, little is known about their absorption from diet, one major unresolved point consisting of the absorption of the bound forms of polyphenols. In this view, in the present work we studied the absorption in humans of phenolic acids from coffee, a common beverage particularly rich in bound phenolic acids, such as caffeic acid, ferulic acid, and p-coumaric acid. Coffee brew was analyzed for free and total (free + bound) phenolic acids. Chlorogenic acid (5'-caffeoylquinic acid), a bound form of caffeic acid, was present in coffee at high levels, while free phenolic acids were undetectable. After alkaline hydrolysis, which released bound phenolic acids, ferulic acid, p-coumaric acid, and high levels of caffeic acid were detected. Plasma samples were collected before and 1 and 2 h after coffee administration and analyzed for free and total phenolic acid content. Two different procedures were applied to release bound phenolic acids in plasma: beta-glucuronidase treatment and alkaline hydrolysis. Coffee administration resulted in increased total plasma caffeic acid concentration, with an absorption peak at 1 h. Caffeic acid was the only phenolic acid found in plasma samples after coffee administration, while chlorogenic acid was undetectable. Most of caffeic acid was present in plasma in bound form, mainly in the glucuronate/sulfate forms. Due to the absence of free caffeic acid in coffee, plasma caffeic acid is likely to be derived from hydrolysis of chlorogenic acid in the gastrointestinal tract.     

Intestinal absorption of p-coumaric and gallic acids in rats after oral administration

Ferulic acid (FA) and p-coumaric acid (CA) are absorbed by the monocarboxylic acid transporter (MCT) in Caco-2 cells, although gallic acid (GA) is not. Therefore, the MCT is selective for certain phenolic acids. Absorption of orally administered CA and GA in rats was studied to obtain serum pharmacokinetic profiles and to investigate their intestinal absorption characteristics in vivo. Rats were administered 100 micromol/kg body weight of CA and GA, and blood was collected from the portal vein and abdominal artery after administration. CA, GA, and their metabolites were quantified with a highly selective and sensitive coulometric detection method using high-performance liquid chromatography-electrochemical detection. Ingested CA was rapidly absorbed in the gastrointestinal tract in an intact form. The serum concentration of intact CA in the portal vein peaked 10 min after dosing (C(max) was 165.7 micromol/L). In contrast, GA was slowly absorbed, with a t(max) for intact GA of 60 min and a C(max) of 0.71 micromol/L. The area under the curve for intact CA and GA was calculated from the serum concentration profile in the portal vein to be 2991.3 and 42.6 micromol min L(-)(1), respectively. The relative bioavailability of CA against GA was about 70. This is the first demonstration that absorption efficiency of CA is much higher than that of GA in vivo. The absorption characteristics of CA are clearly different from those of GA. These findings are in good agreement with the results obtained in vitro using a Caco-2 cell system.     

Profiling the chlorogenic acids and other caffeic acid derivatives of herbal chrysanthemum by LC-MSn

Four samples of herbal chrysanthemum have been profiled qualitatively by LC-MS5 to identify their component chlorogenic acids and partially characterize other caffeic acid derivatives. The chlorogenic acids were minor components, and the four samples varied markedly in profile. Three p-coumaroylquinic acids, three feruloylquinic acids, four caffeoylquinic acids, six dicaffeoylquinic acids, and two tricaffeoylquinic acids were detected, 13 for the first time from this source. Partial characterization of minor components suggested the presence of five caffeoyl-hexose esters and caffeic acid-4-beta-d-glucose that have not previously been reported from this source, and eight caffeoylquinic acid glycosides and 16 dicaffeoylquinic acid glycosides that have not previously been reported in nature. Succinic acid-containing chlorogenic acids and chlorogenic acids based on epi-quinic acid, previously reported in Chrysanthemum spp., were not detected in these samples.     

Absorption and bioavailability of artepillin C in rats after oral administration

Artepillin C (AC), an active ingredient of Brazilian propolis, permeates intact across Caco-2 cells by transcellular passive diffusion. The permeation of AC across Caco-2 cells is as efficient as that of phenolic acids and the microbial metabolites of poorly absorbed polyphenols, which are actively absorbed by the monocarboxylic acid transporter (MCT) (Biochim. Biophys. Acta 2005, 1713, 138-144). Here, the absorption of orally administered AC in rats has been studied to evaluate its pharmacokinetics and bioavailability in vivo in comparison with those of p-coumaric acid (CA), a substrate of MCT. Rats were given 100 micromol/kg of body weight of AC or CA, and blood was subsequently collected from the portal vein and abdominal artery. AC, CA, and their metabolites were quantified by coulometric detection using HPLC-ECD. The serum concentration of intact AC and CA in the portal vein peaked at 5-10 min after administration, with a C(max) of 19.7 micromol/L for AC and 74.8 micromol/L for CA. The area under the curve (AUC) for intact AC and CA in the portal vein was calculated from the serum concentration as 182.6 and 3057.3 micromol.min.L(-1), respectively. The absorption efficiency of CA was about 17-fold higher than that of AC. Furthermore, the bioavailability of CA was about 278-fold higher than that of AC, and the ratio of AUC in the abdominal artery to AUC in the portal vein was 0.04 and 0.70, for AC and CA, respectively. Thus, AC is likely to be more susceptible to hepatic elimination than is CA. The bioactive compound of AC in vivo should be investigated further.     

LC/ES-MS detection of hydroxycinnamates in human plasma and urine

Hydroxycinnamates are components of many fruits and vegetables, being present in particularly high concentrations in prunes. An abundance of phenolic compounds in the diet has been associated with reduced heart disease mortality. However, little is known about the absorption and metabolism of these metabolites after normal foods are consumed. An LC--electrospray--MS method was developed to measure the concentration of caffeic acid in human plasma and urine, but it can also be applied to ferulic acid and chlorogenic acid. The limit of detection was found to be 10.0 nmol/L for caffeic acid and 12.5 nmol/L for ferulic and chlorogenic acids. The method was tested on samples of plasma and urine collected from volunteers who consumed a single dose of 100 g of prunes and increased levels were observed, demonstrating that the method is capable of detecting changes in hydroxycinnamate levels induced by dietary consumption.     

5-caffeoylquinic acid and caffeic acid down-regulate the oxidative stress- and TNF-alpha-induced secretion of interleukin-8 from Caco-2 cells

Although chlorogenic acid (CHA) easily reaches a millimolar level in the gastrointestinal tract because of its high concentration in coffee and fruits, its effects on intestinal epithelial cells have been little reported. We investigated in this study the down-regulative effects of 5-caffeoylquinic acid (CQA), the predominant isomer of CHA, on the H(2)O(2-) or TNF-alpha-induced secretion of interleukin (IL)-8, a central pro-inflammatory chemokine involved in the pathogenesis of inflammatory bowel diseases, in human intestinal epithelial Caco-2 cells. After the cells had been pre- and simultaneously treated with CQA, the oversecretion of IL-8 and overexpression of its mRNA induced by H(2)O(2) were significantly suppressed in a dose-dependent manner in the range of 0.25-2.00 mmol/L. We further found that a metabolite of CQA, caffeic acid (CA), but not quinic acid, significantly inhibited the H(2)O(2)-induced IL-8 secretion and its mRNA expression in the same dose-dependent manner. Both CQA and CA suppressed the TNF-alpha-induced IL-8 secretion as well. Caffeic acid at 2.00 mmol/l was able to absolutely block the H(2)O(2)- or TNF-alpha-induced oversecretion of IL-8 in Caco-2 cells. However, CQA and CA did not suppress the TNF-alpha-induced increase in the IL-8 mRNA expression, indicating that the suppressive mechanisms are different between TNF-alpha-induced and H(2)O(2)-induced IL-8 production models. These results suggest that the habit of drinking coffee and/or eating fruits with a high CHA content may be beneficial to humans in preventing the genesis of inflammatory bowel diseases.     

In vitro and in vivo stability of caffeic acid phenethyl ester, a bioactive compound of propolis

The in vitro biochemical stability of caffeic acid phenethyl ester in rat and human plasma was investigated and compared with the stability of other caffeic acid esters (chlorogenic acid and rosmarinic acid). The incubation of the compounds in rat plasma for up to 6 h showed that caffeic acid phenethyl ester, but not the other compounds, was hydrolyzed, whereas human plasma did not affect the stability of all the assayed compounds. The products in rat plasma were caffeic acid and an unknown compound, which was identified by mass spectrometry as caffeic acid ethyl ester, produced by transesterification in the presence of ethanol used as vehicle for standard compounds. Specific inhibitors of different plasma esterases allowed the identification of a carboxylesterase as the enzyme involved in the metabolism of caffeic acid phenethyl ester. The oral administration in rats of caffeic acid phenethyl ester in the presence of both ethanol and 2-(2-ethoxyethoxy)ethanol gave rise to a dramatic increase of caffeic acid, as well as low levels of caffeic acid phenethyl ester, caffeic acid ethyl ester, and caffeic acid 2-(2-ethoxyethoxy)ethyl ester, in urine collected within 24 h after treatment. These results suggest that caffeic acid phenethyl ester is hydrolyzed also in vivo to caffeic acid as the major metabolite and that its biological activities should be more properly assayed and compared with those of caffeic acid, its bioactive hydrolysis product. Moreover, alcohols should be carefully used in vivo as solvents for caffeic acid phenethyl ester, since they can give rise to new bioactive caffeic acid esters.     

Identification and antioxidant activity of novel chlorogenic acid derivatives from bamboo (Phyllostachys edulis)

One known and two novel antioxidant compounds have been isolated from bamboo (Phyllostachys edulis). The butanol-soluble extract of the bamboo leaves was found to have a significant antioxidant activity, as measured by scavenging the stable 1,1-diphenyl-2-picrylhydrazyl (DPPH) free radical and the superoxide anion radical (O(2)(-)) in the xanthine/xanthine oxidase assay system. Antioxidant activity-directed fractionation of the extract led to the isolation and characterization of three structural isomeric chlorogenic acid derivatives: 3-O-(3'-methylcaffeoyl)quinic acid (1), 5-O-caffeoyl-4-methylquinic acid (2), and 3-O-caffeoyl-1-methylquinic acid (3). Compounds 2 and 3 were isolated and characterized for the first time from the natural products. In the DPPH scavenging assay as well as in the iron-induced rat microsomal lipid peroxidation system, compounds 2 (IC(50) = 8.8 and 19.2 microM) and 3 (IC(50) = 6.9 and 14.6 microM) showed approximately 2-4 times higher antioxidant activity than did chlorogenic acid (IC(50) = 12.3 and 28.3 microM) and other related hydroxycinnamates such as caffeic acid (IC(50) =13.7 and 25.5 microM) and ferulic acid (IC(50) = 36.5 and 56.9 microM). Among the three compounds, compound 1 yielded the weakest antioxidant activity, and the DPPH scavenging and lipid peroxidation inhibitory activity (IC(50) = 16.0 and 29.8 microM) was lower than those of chlorogenic and caffeic acids. All three compounds exhibited both superoxide scavenging activities and inhibitory effects on xanthine oxidase. Their superoxide anion (O(2)(-)) scavenging activities (IC(50) = 1, 4.3 microM; 2, 2.8 microM; and 3, 1.2 microM) were markedly stronger than those of ascorbic acid (IC(50) = 56.0 microM), alpha-tocopherol (IC(50) > 100 microM), and other test compounds, although their inhibition effects on xanthine oxidase may contribute to the potent scavenging activity. alpha-Tocopherol exerted a significant inhibitory effect (65.5% of the control) on superoxide generation in 12-O-tetradecanoylphorbol-13-acetate-induced human promyelocytic leukemia HL-60 cells, and compound 3 showed moderate activity (36.0%). On the other hand, other compounds including 1, 2, chlorogenic acid, and other antioxidants were weakly active (24.8-10.1%) in the suppression of superoxide generation.     

苜蓿植株及根际土壤中主要酚酸和香豆素物质含量测定

通过高效液相色谱(HPLC)法测定不同试验地间18个苜蓿(Medicago sativa L.)品种植株及根际土壤中主要酚酸和香豆素物质的含量,分析其在不同品种苜蓿植株中的分布特征及不同试验地间差异。经过ASE 350型加速溶剂萃取仪萃取苜蓿植株及根际土壤中酚酸类和香豆素物质,萃取液存放于2oC冰箱,并用0.45μm有机滤膜过滤后通过HPLC测定自毒物质含量。结果显示,18个苜蓿品种中香豆素、阿魏酸、绿原酸、咖啡酸的含量存在差异,其中香豆素、绿原酸含量显著高于阿魏酸、咖啡酸含量。各苜蓿品种间单一自毒物质含量差异显著(P0.05),香豆素、阿魏酸、绿原酸和咖啡酸的总含量差异明显。武威试验地酚酸和香豆素物质平均含量比会宁试验地低4.01%。研究表明:不同苜蓿品种间香豆素、阿魏酸、绿原酸、咖啡酸物质总含量差异显著;苜蓿植株中香豆素、阿魏酸、绿原酸、咖啡酸物质总含量与根际土壤中含量差异极显著。     

Caffeic acid derivatives in the roots of yacon (Smallanthus sonchifolius)

Five caffeic acid derivatives were found in the roots of yacon, Smallanthus sonchifolius (Poepp. and Endl.) H. Robinson, Asteraceae, as the major water-soluble phenolic compounds. The structures of these compounds were determined by analysis of spectroscopic data. Two of these were chlorogenic acid (3-caffeoylquinic acid) and 3,5-dicaffeoylquinic acid, common phenolic compounds in plants of the family Asteraceae. Three were esters of caffeic acid with the hydroxy groups of aldaric acid, derived from hexose. The structure of the aldaric moiety was determined by hydrolysis and comparison of NMR spectra with those of standard aldaric acids. The compounds were novel caffeic acid esters of altraric acid: 2,4- or 3,5-dicaffeoylaltraric acid, 2,5-dicaffeoylaltraric acid, and 2,3,5- or 2,4,5-tricaffeoylaltraric acid.     

Combined impact of pH and organic acids on iron uptake by Caco-2 cells

Previous studies have shown that organic acids have an impact on both Fe(II) and Fe(III) uptake in Caco-2 cell. However, to what extent this effect is correlated with the anion of organic acids per se, or with the resulting decrease in pH, has not yet been clarified. Therefore, we studied the effect of five organic acids (tartaric, succinic, citric, oxalic, and propionic acid) on the absorption of Fe(II) and Fe(III) in Caco-2 cells and compared this with sample solutions without organic acids but set to equivalent pH by HCl. The results showed that the mechanisms behind the enhancing effect of organic acids differed for the two forms of iron. For ferric iron the organic acids promoted uptake both by chelation and by lowering the pH, whereas for ferrous iron the promoting effect was caused only by the lowered pH.