Although elevated blood levels of trimethylamine N-oxide (TMAO) have been associated with atherosclerosis development in humans, the role of its gut microbiota-derived precursor, TMA, in this process has not been yet deciphered. Taking this into account, and the fact that increased intestinal fatty acid absorption contributes to atherosclerosis onset and progression, this study aimed to evaluate the effect of TMA on fatty acid absorption in a cell line that mimics human enterocytes. Caco-2 cells were treated with TMA 250 μM for 24 h. Fatty acid absorption was assessed by measuring the apical-to-basolateral transport and the intracellular levels of BODIPY-C12, a fluorescently labelled fatty acid analogue. Gene expression of the main intestinal fatty acid transporters was evaluated by real-time quantitative reverse transcription PCR. Compared to control conditions, TMA increased, in a time-dependent manner and by 20–50 %, the apical-to-basolateral transport and intracellular levels of BODIPY-C12 fatty acid in Caco-2 cells. Fatty acid transport protein 4 (FATP4) and fatty acid translocase (FAT)/CD36 gene expression were not stimulated by TMA, suggesting that TMA-induced increase in fatty acid transport may be mediated by an increase in FAT/CD36 and/or FATP4 activity and/or fatty acid passive transport. This study demonstrated that TMA increases the intestinal absorption of fatty acids. Future studies are necessary to confirm if this may constitute a novel mechanism that partially explains the existing positive association between the consumption of a diet rich in TMA sources (e.g. red meat) and the increased risk of atherosclerotic diseases.

Inflammatory bowel diseases (IBD) are a major problem for public health, with an increased incidence and impact on life quality. The effect of pre- and probiotic combination has been less studied in IBD. Using genomic and proteomic array technologies, this study examined the efficacy of a new combination of natural alternatives: prebiotics (grape pomace extract, GP) and probiotics (lactobacilli mixture, Lb mix) on inflammation and intracellular signalling routes in a cellular model of inflammation. Caco-2 cells challenged with lipopolysaccharide (LPS) for 4 h were treated with GP extract (50 μg/ml gallic acid equivalent) and Lb combination (3 × 108 colony-forming units/ml total Lb) for 24 h. The profile expressions of forty key inflammatory markers and twenty-six signalling kinases were analysed. Other markers involved in inflammation were also investigated (NF-κB/RELA, Nrf2, aryl hydrocarbon receptor, Cyp1A1, Cyp1B1); 57·5 and 60 % of investigated genes and proteins, respectively, were down-regulated by the synbiotic combination. Relevant cytokines and chemokines involved in response to microbial infection and inflammation were reduced under the level induced by LPS treatment and toward the unchallenged control. As expected, the reduction effect seems to imply mitogen-activated protein kinase and NF-κB pathway. Most of the signalling molecules activated by LPS were decreased by GP extract and Lb mix. Our study indicates that the synbiotic combination of GP extract and Lactobacillus sp. mixture exerted anti-inflammatory properties, which are able to decrease the majority of inflammatory genes, their proteins and associated signalling markers. Due to protective role of GP compounds on lactobacilli probiotic, this synbiotic combination might serve as a promising adjunctive therapy in intestinal inflammations.

Dietary oils containing large amounts of conjugated linolenic acids (CLnA) may be regarded as a source of conjugated linoleic acids (CLA), which have been suspected to bear health-promoting properties. Indeed, CLnA can be converted into CLA in mammals. The objective of the present study was to investigate the uptake of CLnA and their metabolism into CLA in Caco-2 cells, as a validated in vitro model of the intestinal barrier. Caco-2 cells were incubated for 24 h in the presence of either α-eleostearic, β-eleostearic, catalpic or punicic acid. We first observed that Caco-2 cells take these CLnA up at different rates and then convert them but with varying efficiency depending on the structure of the Δ13 double bond. Finally, the distribution of CLnA between neutral lipids (NL) and phospholipids appeared to be linked to their number of trans double bonds: the higher the number, the higher the accumulation in the NL fraction.

Carotenoid consumption has been linked to a number of beneficial health effects, including the reduction of chronic diseases such as cancer and cardiovascular complications. However, no data are available on their action on the intestinal epithelium, being exposed to the highest concentrations of carotenoids in the human body, and where they could act preventively on intestinal inflammatory diseases such as Crohn's disease and ulcerative colitis. The objective of the present study was to investigate whether lycopene and β-carotene in micelles (M), at concentrations that could be reached via the diet (10–25 μg/ml) could aid in the reduction of TNF-α plus IL-1β-induced inflammation of Caco-2 human epithelial cells. The impact on biomarkers of inflammation, including IL-8, NO and cyclo-oxygenase-2 (through PGE-2α), and the NF-κB and mitogen-activated protein kinase (MAPK) pathways of intracellular signalling cascades were evaluated compared with controls (empty M). Furthermore, proteomic analyses were conducted from total cellular protein extracts. The results revealed that isolated carotenoids had no statistical significant anti-inflammatory effect on the biomarkers observed, or on the regulation of NF-κB and MAPK. Nevertheless, analyses of the proteome suggested that fifteen proteins were significantly (P < 0·05, expression ratio >1·3) differentially regulated following β-carotene exposure, participating mostly in metabolic activities including antioxidant mechanisms, such as glutathione S-transferase A1. Only one protein was differentially regulated by lycopene (profilin-1). To our knowledge, this is the first attempt to investigate pathways involved in the action of carotenoids on the intestinal epithelium.

Bifidobacterium spp. typical of the human intestinal microbiota are believed to influence the balance of immune responses in the intestinal mucosa. The aim of the present study was to investigate the effect of different bifidobacterial species and their mixtures in in vitro experiments with peripheral blood mononuclear cells (PBMC) and Caco-2 cells. Bifidobacterium adolescentis, B. angulatum, B. breve, B. catenulatum, B. infantis, B. longum and two combinations of these bifidobacteria simulating the species composition found in faecal samples from breast-fed (BF) and formula-fed (FF) infants were used. The levels of several cytokines were measured by direct stimulation of PBMC and by stimulation of a Caco-2/PBMC co-culture with bifidobacteria. B. catenulatum and B. breve were the strongest enhancers of interferon-γ (IFN-γ) production by direct stimulation of PBMC. B. longum was the highest inducer of IL-10 and the lowest TNF-α stimulus. In the Caco-2/PBMC system, B. breve was the highest inducer of IL-8 production by Caco-2 cells, significantly different from B. infantis, B. adolescentis and the FF mixture (P < 0·05). IFN-γ produced by PBMC stimulated with the BF mixture (containing 22 % B. breve, compared with 7 % in the FF mixture) was significantly higher compared with B. adolescentis, B. infantis and B. longum. B. adolescentis also inhibited IFN-γ production compared with the FF mixture and B. longum. The proportion of different Bifidobacterium strains seems to be an important determinant of the cytokine balance in the simulated intestinal environment studied. B. breve and the combination of the Bifidobacterium species typically found in the microbiota of BF infants have shown the most significant effects.

dl-2-Hydroxy-(4-methylthio)butanoic acid (HMTBA) is a synthetic source of dietary methionine (Met) widely used in poultry nutrition. HMTBA is transported in the intestinal epithelium by the monocarboxylate transporter 1, after which its biological utilisation relies on its conversion to l-Met. This process involves stereospecific HMTBA oxidation to 2-keto-(4-methylthio)butanoic acid (KMB) and transamination to l-Met. In the present study, we examined HMTBA conversion to l-Met, further incorporation into cellular proteins and the regulation of both processes by HMTBA supplementation in differentiated intestinal Caco-2 cells. The results showed d- and l-HMTBA oxidation in the enterocytes, this process being up-regulated by HMTBA. The data also revealed that KMB transamination is not linked to a specific amino group donor. However, the branched-chain amino acid l-leucine is the preferred amino group donor. Furthermore, transamination was not affected by HMTBA availability. The incorporation of radioactivity from HMTBA into cellular proteins was not significantly different from that of l-Met and was not affected by HMTBA supplementation. In conclusion, the results reveal the capacity of Caco-2 cells to convert HMTBA to l-Met and the up-regulation of conversion by nutritional HMTBA supplementation, thus highlighting the contribution of the intestinal epithelium in the utilisation of HMTBA as a dietary source of Met.

The aim of this study was to investigate the effects of unhydrolysed/intact casein and eight different sodium casein hydrolysates (a–h) on the viability and growth of human cancer cell lines. Both human Jurkat T cells and Caco-2 cells were incubated with increasing concentrations of the test compounds (0·5–10% v/v) for 24 h. Cell viability was assessed using the MTT, lactate dehydrogenase (LDH) release and Trypan Blue assays. Cell growth was monitored using the MTT, Trypan Blue and Bromodeoxyuridine (BrdU) proliferation assays. Casein hydrolysates b, c and f had an inhibitory effect on the viability and growth of both cell lines. The casein hydrolysates did not negatively affect the membrane integrity of both Jurkat and Caco-2 cells. In Jurkat cells hydrolysates a and h had an inhibitory effect on DNA synthesis after 24 h, while in Caco-2 cells DNA synthesis was not affected. In conclusion, we found that the different casein hydrolysates had cell-specific effects which target particular functions within the cell. Overall, casein hydrolysates had no effect on membrane integrity while they had varied effects on mitochondrial activity and DNA synthesis in the different cell lines.

Transcriptomic studies are useful for elucidating molecular mechanisms through which changes in nutrient availability produce pleiotropic effects on whole-body and tissue physiology. To further the knowledge of gene-regulatory effects of Zn on tissues important for adult and fetal Zn nutrition, we analysed the responses of human intestinal Caco-2 and placental JAR cells to changes in Zn supply. Analysis of oligonucleotide microarrays demonstrated that, despite the analogous roles of the two tissues in nutrient transfer, different genes respond to changes in Zn availability in intestinal cells compared with placental cells. A number of Fe- and Cu-related genes were identified as targets for regulation by Zn, revealing potential mechanisms underlying reported dietary interactions between Zn and other metals. We established that there are fundamental differences in Zn-regulated transcriptional control in Caco-2 compared with JAR cells. We demonstrated that Zn-induced transcriptional activation of the metallothionein 2A promoter occurs over different, and physiologically relevant, concentration ranges in Caco-2 and JAR cells, indicating that these cell lines sense changes in the extracellular Zn concentration over different ranges. Also, we established that mRNA levels of the Zn-responsive metal response element binding transcription factor (MTF)-1, and its homologue MTF-2, are regulated by Zn in Caco-2 but not JAR cells, which may in part underlie differential gene responses to Zn in intestinal and placental cells. The present study identified a number of novel molecular targets that may underlie symptoms associated with deficient or excessive Zn supply and highlighted the necessity of taking account of cell- and tissue-specific processes when investigating Zn-regulated gene expression in mammals.

The effect of processing (homogenization, lyophilization, acid-extraction) meat products on iron uptake from meat combined with uncooked iron-fortified cereal was evaluated using an in vitro digestion/Caco-2 cell model. Beef was cooked, blended to create smaller meat particles, and combined with electrolytic iron-fortified infant rice cereal. Chicken liver was cooked and blended, lyophilized, or acid-extracted, and combined with FeSO4-fortified wheat flour. In the beef–cereal combination, Caco-2 cell iron uptake, assessed by measuring the ferritin formed by cells, was greater when the beef was blended for the greatest amount of time (360 s) compared with 30 s (P < 0·05). Smaller liver particles (blended for 360 s or lyophilized) significantly enhanced iron uptake compared to liver blended for 60 s (P < 0·001) in the liver–flour combination. Compared to liver blended for 60 s, acid-extraction of liver significantly enhanced iron uptake (P = 0·03) in the liver–flour combination. Homogenization of beef and homogenization, lyophilization, or acid-extraction of chicken liver increases the enhancing effect of meat products on iron absorption in iron-fortified cereals.

Gangliosides are glycosphingolipids found in cell membranes and human milk with important roles in cell proliferation, differentiation, growth, adhesion, migration, signalling and apoptosis. Similar changes in ganglioside composition occur during embryonic development, lactation and cancer cell differentiation. It is not known, however, whether ganglioside compositional changes that occur in differentiating colon cancer cells reflect changes that occur during intestinal development. The Caco-2 cell line is commonly used to study physiological and pathophysiological processes in the small intestine and colon. Therefore, to examine this question, undifferentiated and differentiated Caco-2 cells were grown and total lipid was extracted from cell supernatant fractions using the Folch method. The upper aqueous phase containing gangliosides was collected and purified. Total gangliosides were measured as ganglioside-bound N-acetyl neuraminic acid, while individual ganglioside content was quantified via a colorimetric assay for sialic acid and scanning densitometry. The total ganglioside content of differentiated Caco-2 cells was 2·5 times higher compared with undifferentiated cells. Differentiated Caco-2 cells had significantly more (N-acetylneuraminyl) 2-galactosylglucosyl ceramide (GD3) and polar gangliosides, and a lower N-acetylneuraminylgalactosylglucosylceramide (GM3):GD3 ratio than undifferentiated cells. The present study demonstrates that the total ganglioside content and individual ganglioside composition of differentiated Caco-2 cells are similar to those of human colostrum and neonatal rat intestine. Differentiated Caco-2 cells may therefore be an alternative model for studying physiological and pathological processes in the small intestine and colon, and may help to elucidate possible functions for specific gangliosides in development and differentiation. Further research using more sensitive techniques of ganglioside analysis is needed to confirm these findings.

Colonic fermentation products, SCFA, have various effects on colonic functions. Here, we found that physiological concentrations of SCFA immediately promote epithelial barrier function in the large intestine. Solutions of mixed and individual SCFA were applied to the caecal walls mounted on Ussing-type chambers. Transepithelial electrical resistance (TER) increased rapidly and reached a peak 35 % higher than that in the control specimen within 10 min post application of the SCFA mixture (80 acetate, 40 propionate, 20 butyrate (mmol/l)). The Lucifer yellow permeability, a paracellular transport marker, was dose-dependently reduced by the mixed SCFA, acetate and propionate solutions. Inhibition of monocarboxylate transporter-1 did not influence the increase in TER with acetate; however, lowering the pH (from 7·5 to 5·5) clearly enhanced the effect of acetate. Non-metabolizable, bromo and chloro derivatives of SCFA also increased TER. These results suggest that passive diffusion of SCFA is dominant and the metabolism of SCFA is not required for the promotive effect of SCFA on barrier function. We also observed that individual SCFA dose-dependently increased TER in T84 and Caco-2 cells, which indicates that SCFA directly stimulate epithelial cells. Depletion of membrane cholesterol and inhibitors of phosphatidylinositol-3 kinase and Gq protein attenuated the acetate-mediated promotive effect. Finally, we found that the mucosal application of the SCFA mixture dose-dependently suppressed [3H] mannitol transport from the caecal lumen to the mesenteric blood in the anaesthetized rats. We conclude that physiological concentrations of SCFA immediately enhance barrier function of the colonic epithelium through cholesterol-rich microdomain in the plasma membrane.

Recent studies support beneficial effects of polyphenols in various chronic inflammatory diseases, for example, the inflammatory bowel diseases. Inhibition of NF-κB activation by polyphenols could explain part of their anti-inflammatory properties, but few data are available on the intestine. The purpose of the present study was thus to investigate the effects of some polyphenols on NF-κB activation using human intestinal Caco-2 cells. Effects of standard polyphenols (50 μmol/l) were studied on different cellular events associated with NF-κB activation: (i) NF-κB activity using cells transiently transfected with a NF-κB–luciferase construct and stimulated by inflammatory agents (IL-1β, TNF-α or lipopolysaccharides (LPS)); (ii) phosphorylation of the inhibitor of κB (IκB-α) analysed by Western blot; (iii) secretion of IL-8 quantified by ELISA assay. Results showed that chrysin and ellagic acid inhibited NF-κB activity, whereas genistein and resveratrol increased it. These effects were independent of the nature of the inducer, indicating that polyphenols may modulate NF-κB activation by acting on a common event to the cytokine- and LPS-mediated cascades. Chrysin strongly reduced (2·5-fold) IL-1β-induced IκB-α phosphorylation, whereas ellagic acid increased it (1·7-fold). Ellagic acid, genistein and epigallocatechin gallate reduced (4- to 8-fold) IL-1β-induced IL-8 secretion, while resveratrol promoted (1·7-fold) the secretion. Chrysin also diminished IL-8 secretion by 1·6-fold (but P>0·05). The data indicate that polyphenols can modulate the NF-κB activation pathway in the intestine. Chrysin could block NF-κB activation via the inhibition of IκB-α phosphorylation. The other molecular targets of the active polyphenols are still to be identified.

Fucose is present in foods, and it is a major component of human mucin glycoproteins and glycolipids. l-Fucose can also be found at the terminal position of many cell-surface oligosaccharide ligands that mediate cell-recognition and adhesion-signalling pathways. Mucin fucose can be released through the hydrolytic activity of pathogens and indigenous bacteria, leading to the release of free fucose into the intestinal lumen. The immunomodulating effects of free fucose on intestinal epithelial cells (enterocyte-like Caco-2) were investigated. It was found that the presence of l-fucose up regulated genes and secretion of their encoded proteins that are involved in both the innate and adaptive immune responses, possibly via the toll-like receptor-2 signalling pathway. These include TNFSF5, TNFSF7, TNF-α, IL12, IL17 and IL18.Besides modulating immune reactions in differentiated Caco-2 cells, fucose induced a set of cytokine genes that are involved in the development and proliferation of immune cells. These include the bone morphogenetic proteins (BMP) BMP2, BMP4, IL5, thrombopoietin and erythropoietin. In addition, the up regulated gene expression of fibroblast growth factor-2 may help to promote epithelial cell restitution in conjunction with the enhanced expression of transforming growth factor-β mRNA. Since the exogenous fucose was not metabolised by the differentiated Caco-2 cells as a carbon source, the reactions elicited were suggested to be a result of the direct interaction of fucose and differentiated Caco-2 cells. The presence of free fucose may signal the invasion of mucin-hydrolysing microbial cells and breakage of the mucosal barrier. The intestinal epithelial cells respond by up regulation and secretion of cytokines, pre-empting the actual invasion of pathogens.

Invasion in and translocation across enterocytes are major events during Campylobacter jejuni-induced enteritis in humans. C. jejuniin vitro infection of cell monolayers typically results in loss of tight junction integrity, which could contribute to translocation. In the present study, we wanted to investigate whether butyrate is able to confer protection to Caco-2 cells against C. jejuni invasion, thus reducing paracellular permeability and limiting C. jejuni translocation. Protection of Caco-2 cells against C. jejuni invasion was assessed using a gentamicin protection assay. Transwell systems were used to investigate the impact of butyrate on translocation of C. jejuni across a Caco-2 monolayer and its effect on transepithelial resistance during infection. Butyrate protected Caco-2 cells against C. jejuni invasion in a concentration-dependent manner. Differentiated Caco-2 cells were less susceptible to C. jejuni invasion than 3-d-old undifferentiated cells and higher concentrations of butyrate and longer incubation times were needed to become refractive for invasion. C. jejuni translocation over Caco-2 monolayers was reduced when monolayers were treated with butyrate and this was accompanied by an enhanced drop in transepithelial resistance. The present study showed that butyrate is able to protect Caco-2 cells from two major virulence mechanisms of C. jejuni, namely invasion and translocation, but not from a decline in transepithelial resistance.

β-Cryptoxanthin (β-CX), a provitaminic carotenoid of potential interest for health, is found principally in Citrus fruit in both free and esterified forms. Little is known about the intestinal absorption of β-CX especially with regard to the esterified forms. The aim of this study was to evaluate the absorption of free and esterified β-CX using simulated digestion coupled with the Caco-2 model. Bioaccessibility was investigated by measuring the transfer of carotenoids from different citrus juices into micelles using an in vitro digestion system. Then, carotenoid uptake was evaluated by adding carotenoid-rich micelles (from the in vitro digestion) or synthetic micelles (made from synthetic lipids and carotenoids purified from citrus juice) to human intestinal cells (Caco-2 TC7 clone). Our results showed that β-cryptoxanthin esters (β-CXE) were partially hydrolysed during the in vitro digestion. The bioaccessibility of free β-CX measured was significantly higher (40 (sd 1·05) %) than that of β-carotene (30 (sd 1·9) %) and β-CXE (16 (sd 1·5) %). In the same way, the incorporation of free β-CX (27 (sd 1·01) %) into synthetic micelles exceeded (P < 0·05) that of β-carotene (10 (sd 0·7) %) and β-CXE (8·8 (sd 0·4) %). In the case of micelles from in vitro digestion, the uptake of β-carotene, free β-CX and β-CXE forms by Caco-2 cells was 14·3 (sd 1·8), 3·9 (sd 1·3), and 0·7 (sd 0·08) % respectively. These results showed a preferential uptake by Caco-2 cells of β-carotene and free β-CX compared with the two esters of β-CX.

Marine oil-derived n-3 fatty acids have been shown to stimulate intestinal Ca absorption in animal studies, but the effects of such fatty acids on Ca absorption in human subjects are relatively unknown. In particular, n-3 fatty acids may be of therapeutic value for some Crohn's disease patients who experience Ca malabsorption. Therefore, the aim of the present study was to investigate the effect of 20 : 5n-3 and 22 : 6n-3 on transepithelial Ca transport across monolayers of healthy Caco-2 cells as well as of TNF-α-treated Caco-2 cells (an in vitro model of Crohn's disease). Caco-2 cells were seeded onto permeable filter supports and allowed to differentiate into monolayers, which were treated with 80 μm-20 : 5n-3, 80 μm-22 : 6n-3, or 40 μm-20 : 5n-3+40 μm-22 : 6n-3 for 6 or 8 d, with or without co-treatment with TNF-α (10 ng/ml) (n 11–15 monolayers per treatment). On day 16, transepithelial and transcellular transport of 45Ca and fluorescein transport (a marker of paracellular diffusion) were measured. Treatment of healthy and inflamed Caco-2 cells with 20 : 5n-3, 22 : 6n-3 and both fatty acids combined for 8 d significantly (P < 0·005–0·01) increased total transepithelial Ca transport compared with that in control, effects which were mediated by an enhanced rate of transcellular Ca transport. The effects of n-3 fatty acids on Ca absorption after 6 d were less clear-cut. In conclusion, the present in vitro findings highlight the need to investigate the effect of marine oil-based n-3 fatty acids on Ca absorption in vivo in studies of healthy human subjects as well as of Crohn's disease patients.

Experimental evidence suggests that most herbs and spices possess a wide range of biological and pharmacological activities that may protect tissues against O2-induced damage. The objectives of the present study were: first, to determine the effects of plant extracts on the viability, membrane integrity, antioxidant status and DNA integrity of Caco-2 cells and second, to investigate the cytoprotective and genoprotective effects of these plant extracts against oxidative stress in Caco-2 cells. The plant extracts examined were rosemary (Rosmarinus officinalis L.), oregano (Origanum vulgare L.), sage (Salvia officinalis L.) and echinacea (Echinacea purpurea L.). Cell membrane integrity was assessed by the lactate dehydrogenase release assay. Viability was determined by the neutral red uptake assay (NRUA) and the concentration of compound that resulted in 50 % cell death (IC50) was calculated. Antioxidant status of the cells was assessed by measuring GSH content, catalase activity and superoxide dismutase activity. To examine their cytoprotective and genoprotective effects, Caco-2 cells were pre-treated with each plant extract for 24 h followed by exposure to H2O2. DNA damage was assessed by the comet assay and cell injury was determined by the NRUA. Rosemary was the most toxic (IC50 123 μg/ml) and echinacea the least toxic (IC50 1421 μg/ml). Sage was the only plant extract to affect the antioxidant status of the cells by increasing GSH content. Sage, oregano and rosemary protected against H2O2-induced DNA damage (olive tail moment and percentage tail DNA), whereas protection against H2O2-induced cytotoxicity was afforded by sage only.

An in vitro digestion/Caco-2 cell model was applied to explore the impact of lactic acid (LA) fermentation by Lactobacillus pentosus FSC1 and Leuconostoc mesenteroides FSC2 on the Fe bioavailability of carrot juice. The redox state of Fe in fermented carrot juice was also assessed as a crucial factor for absorption. LA fermentation was shown to improve mineral solubility to different extents at simulated physiological conditions: Mn (2-fold); Fe (1·5–1·7-fold); Zn (1·2-fold); Cu (1-fold). Soluble Fe2+ was increased about 16-fold by LA fermentation, and about one third of the Fe2+ remained soluble after in vitro digestion (about 4–5-fold higher than in fresh juice). Data on cell-line studies showed a 4-fold increase in the efficiency of Fe uptake, but not in transepithelial transfer by Caco2 cells, as a result of fermentation. The increases in Fe2+ level and the efficiency of cellular Fe uptake were strain-dependent. To sum up the effect on both Fe solubility and cellular uptake efficiency, the amount of cellularly absorbed Fe from Ln. mesenteroides FSC2-fermented juice was about 20 % higher than that from L. pentosus FSC1-fermented juice (22·7 v. 19·2 μg/l juice per mg protein). To conclude, LA fermentation enhanced Fe absorption by Caco-2 cells from carrot juice because of increases in not only Fe solubility after digestion, but also the efficiency of cellular Fe uptake. The fermentation-improved efficiency of Fe uptake was possibly due to the increased level of soluble Fe2+ rather than a being a strain-specific event.

Maillard products arise from condensation reactions between amino acids or proteins with reducing sugars during food processing. As ubiquitous components of human food, these early or advanced glycation products may be subject to intestinal absorption. The present study was performed to investigate the intestinal uptake of Maillard products and to determine whether they are substrates for peptide and amino acid transporters expressed at the apical membrane of Caco-2 cells. At a concentration of 10mm, Nɛ-(carboxymethyl)-L-lysine, Nα-hippuryl-Nɛ-(1-deoxy-d-fructosyl)-l-lysine,Nα-hippuryl-Nɛ-(carboxymethyl)-L-lysine and Nɛ-(1-deoxy-d-fructosyl)-l-lysine inhibited the [14C]glycylsarcosine uptake mediated by the H+–peptide co-transporter PEPT1 by 13 to 45%.For Nɛ-(1-deoxy-d-fructosyl)-l-lysine, an inhibitory constant of 8·7mm was determined, reflecting a low affinity to PEPT1 in comparison with natural dipeptides. Uptake of l-[3H]lysine was weakly affected by Nɛ-(carboxymethyl)-L-lysine, Nα-hippuryl-l-lysine and Nα-hippuryl-Nɛ-(carboxymethyl)-L-lysine but strongly inhibited by Nɛ-(1-deoxy-d-fructosyl)-l-lysine (81%). None of the Maillard products was able to inhibit the uptake of l-[3H]leucine by more than 15%. We also studied the transepithelial flux of Maillard productsacross Caco-2 cell monolayers cultured on permeable filters. The flux rates of Maillard products ranged from 0·01 to 0·3%/cm2 per h and were shown to be muchlower than those of carrier substrates such as glycylsarcosine, l-proline and the space marker 14C]mannitol. We conclude that the Maillard products investigated in the present study are neither transported by PEPT1 nor by carriers for neutral amino acids. The low transepithelial flux measured for these compounds most probably occurs by simple diffusion.

The Fe solubility test is a commonly used, easy and relatively cheap in vitro tool for predicting Fe bioavailability in food matrices. However, the outcome of a recent field trial comparing the effect on Fe status of Tanzanian infants of processed V. unprocessed complementary foods (CF), with otherwise the same composition, challenged the validity of this test for predicting Fe bioavailability. In the solubility test, significant more soluble Fe was observed in processed compared with unprocessed foods (mean 18·8 (sem 0·21) V. 4·8 (sem 0·23) %; V<0·001). However, in the field trial, no significant difference in Fe status was seen between processed and unprocessed CF groups after 6 months' follow-up. Therefore, twenty-four samples of these CF (twelve processed and twelve unprocessed batches) were analysed in triplicate for Fe availability using an in vitro digestion–Caco-2 cell culture method and results were compared with solubility results. Significantly more soluble Fe was presented to Caco-2 cells in the processed compared with unprocessed samples (mean 11·5 (sem 1·16) V. 8·5 (sem 2·54) % p=0·028), but proportionally less Fe was taken up by the cells (mean 3·0 (sem 0·40)p. 11·7 (sem 2·22) %; p=0·007). As a net result, absolute Fe uptake was lower (not significantly) in processed compared with unprocessed CF (mean 1·3 (sem 0·16) V. 3·4 (sem 0·83) nmol/mg cell protein; p=0·052). These data clearly demonstrate that the Fe solubility test was not a good indicator of Fe bioavailability in these particular food matrices. In contrast, the results of an in vitro Caco-2 model supported the effects observed in vitro.