ORIGINAL ARTICLE

Double-balloon jejunal perfusion to compare absorptionof vitamin E and vitamin E acetate in healthy volunteersunder maldigestion conditionsK Nagy1, L Ramos2, M-C Courtet-Compondu1, S Braga-Lagache1, K Redeuil1, B Lobo2, F Azpiroz2,3, J-R Malagelada2,3, M Beaumont1,J Moulin1, S Acquistapache1, L Sagalowicz1, M Kussmann1,4,5,6, J Santos2,3, B Holst1 and G Williamson7

BACKGROUND/OBJECTIVES: The vitamin E derivative, a-tocopheryl acetate, is often included in formulations used in enteralnutrition. In this respect, we compared a-tocopherol and a-tocopheryl acetate absorption under ‘maldigestion’ conditions,such as occurring during enteral tube feeding, using differentially labeled RRR-[5,7-methyl-(2H6)]-a-tocopherol and RRR-[5-methyl-2H3]-a-tocopheryl acetate allowing direct comparison between free and esterified forms.SUBJECTS/METHODS: The two derivatives were given together in a single dose to six volunteers directly into the jejunumusing a double-balloon perfusion system. Perfusion lasted for 1 h, and the collected blood and effluent samples were analyzedby liquid chromatography–mass spectrometry.RESULTS: In the isolated 20-cm length of exposed jejunum, on average B6% of the two vitamin E forms were absorbed41 h based on subtraction of effluent from influent. There was substantial difference in the absolute absorbed quantity betweenindividuals, but no significant differences were observed in the absorption between the two labeled forms as assessed in theplasma. 2H3-a-tocopherol was not present in the influent, but appeared in the effluent, indicating that the acetylated form ofvitamin E is cleaved by brush border enzymes in the small intestine.CONCLUSIONS: This study shows that even in the absence of digestive enzymes and bile salts, the appropriately solubilizedacetylated form of a-tocopherol exhibits the same bioavailability as free a-tocopherol. This suggests that both forms can beabsorbed equally under maldigestion conditions such as present clinically during enteral tube feeding.

European Journal of Clinical Nutrition (2013) 67, 202–206; doi:10.1038/ejcn.2012.183; published online 5 December 2012

Keywords: vitamin E; bioavailability; jejunal segmental perfusion; pancreas

INTRODUCTIONNatural vitamin E is a generic designation for a group of eightcompounds (tocopherols and tocotrienols) synthesized byplants.1,2 The supplemented form of vitamin E is mostly itsacetate ester including the racemic mixture of eight stereoisomersaccounting for the three chiral centers at 2, 40 and 80 positions.Vitamin E is the major and most potent lipid-soluble antioxidantin vivo.3–5 It acts as a radical scavenging antioxidant in lipoproteinsand efficiently interrupts the chain propagation of lipid oxidation,thus protecting low-density lipoproteins from oxidation.6 VitaminE is also associated with lowered risk of coronary heart disease andatherosclerosis,7,8 and ischemic heart disease in cross-culturalepidemiology.9 Details on physiology, bioavailability and potencyof vitamin E are available in several excellent reviews.4,7,8,10,11

The digestion of tocopherols and tocotrienols broadly followsthat of lipids, that is, emulsification and incorporation intomixed micelles, which are transported via the unstirred waterlayer and taken up by the enterocytes, partly via active transport,especially in the proximal intestine.12 In enterocytes, tocopherolsare incorporated into chylomicrons and transported into thecirculation via the lymphatic system.7,8 Many variables influence

vitamin E absorption, and both bile acids and pancreatic juice arethought to be important for the absorption of vitamin E. Thishas been established in patients where secretion of either, or both,is severely diminished, as in cholestatic liver disease, cysticfibrosis or pancreatitis, where severe vitamin E malabsorptionoccurs.7,8 During enteral feeding of critically ill patients, nutrientsare delivered directly into the small intestine, without the addi-tion of normal digestive juices, leading to the possibility ofmalabsorption.

A simultaneous intake of fat improves vitamin E absorption as itstimulates bile flow and secretion of pancreatic enzymes(including lipase and esterases) to allow micelle formation. If avitamin E supplement is taken with a low-fat meal, the absorptionof the vitamin into the blood, and therefore its efficacy, issignificantly reduced.13 Absorption of vitamin E was minimal whentaken with only a glass of water, with cereal and semi-skimmedmilk, but significantly higher when taken with cereal and full-fatmilk or cream, and higher still when taken with a meal of toast andbutter.13 Vitamin E intake is for most people far from optimal14

and so many clinical studies investigate the bioavailabilityand metabolism of this micronutrient. Applications of stable

1Nestle Research Center, Nestec Ltd, Lausanne, Switzerland; 2Department of Gastroenterology, Institut de Recerca Vall d’Hebron, Digestive System Research Unit, HospitalUniversitari Vall d’Hebron, Universitat Autonoma de Barcelona (Department of Medicine), Barcelona, Spain; 3Centro de Investigacion Biomedica en Red de EnfermedadesHepaticas y Digestivas (CIBERehd), Madrid, Spain; 4Interdisciplinary Nanoscience Center, Aarhus University, Aarhus, Denmark; 5Proteomics and Metabonomics Core, NestleInstitute of Health Sciences, Lausanne, Switzerland; 6Ecole Polytechnique Federale Lausanne, Lausanne, Switzerland and 7School of Food Science and Nutrition, University ofLeeds, Leeds, UK. Correspondence: Professor G Williamson, School of Food Science and Nutrition, University of Leeds, Woodhouse Lane, Leeds LS2 9JT, UK.E-mail: G.Williamson@leeds.ac.ukReceived 21 June 2012; revised 11 October 2012; accepted 24 October 2012; published online 5 December 2012

European Journal of Clinical Nutrition (2013) 67, 202–206& 2013 Macmillan Publishers Limited All rights reserved 0954-3007/13

www.nature.com/ejcn

isotope-labeled tocopherols are a popular means to follow thefate of orally administered vitamin E, as these can be readilydistinguished from the endogenous vitamin E pool of the body bymass spectrometry-based methods.15–17 On the other hand, thebioavailability and health benefits of various vitamin Econstituents (free versus acetylated) under various suboptimalconditions (for example, malabsorption during tube feeding) arestill not fully understood.18,19

The aim of this study was to compare the bioavailability of freeand acetylated a-tocopherol under maldigestion conditions andto ascertain whether normal pancreatic function is necessary forits efficient absorption in healthy volunteers by applying a closedsegment intestinal perfusion technique in vivo.

MATERIALS AND METHODSSubjects, groups and centersHealthy subjects were recruited at the Vall d’Hebron Hospital, Barcelona.Inclusion criteria were X18 years, male (to eliminate gender variations),body mas index 18–25 kg/m2, baseline serum vitamin E 6–21 mg/l,triglyceride 0.45–1.85 g/l and cholesterol 1.3–2.2 g/l. Exclusion criteria werefood, latex or any other allergy related to the perfusion components, anychronic disease or major gastrointestinal surgery, bowel disorder or regularconsumption of vitamin supplements. Acetylsalicylates, nonsteroidalanti-inflammatory drugs, b2-agonists, acetaminophen, antacids, antic-holinergics, codeine or opioid derivatives were allowed for up to 3 daysbefore the perfusion, whereas any other drug or supplement wasprohibited from 2 weeks before the study; moderate alcohol or caffeineingestion was allowed up to the last 72 h before the study.

Written informed consent was obtained from each participant, includingacceptance of pre- and post-study diet and meal provided on the dayof perfusion. Subject personal information was protected by anonymouscode. Six participants were enrolled to complete the study. Clinical studyprotocol and ethical application were approved by the Ethical Committeeof the Medical Faculty, at Vall d’Hebron, Barcelona and by the NestleResearch Center Medical Unit, Study number 05.10. Nestle ResearchCenter.

Formulation and dosage of vitamin EFor all treatments, deuterated vitamin E (RRR-[5,7-methyl-(2H6)]-a-toco-pherol; 98 atom%) and deuterated vitamin E acetate (RRR-[5-methyl-2H3]-a-tocopherol acetate; 98 atom%) obtained from Orphachem (Clermont-Ferrand, France) were used. The products were analyzed and certified forhuman consumption with a complete sample dossier and safetyevaluation. The use of deuterium-labeled vitamin E and vitamin E acetateenables differentiation from endogenous/dietary vitamin E.

The composition of the base dispersion containing the vitamin E andvitamin E acetate is shown in Table 1. Monoglyceride (Dimodan U/J, acommercial grade form of monolinolein), polyglycerol ester (0 80/D) andascorbyl palmitate were purchased from DuPont/Danisco (Grindsted,Denmark). Lysophosphatidylcholine (Sunlecithin A1) was obtained fromTayio Kagaku (Yokkaichi, Japan). Na-caseinate was obtained from Emi(Luzern, Switzerland). The preparation commenced by mixing ascorbylpalmitate, polyglycerol ester 0 80/D and Dimodan U/J. Then, deuterateda-tocopherol and deuterated a-tocopherol acetate were added tothe melted mixture to form the lipid mixture. Sunlecithin A-1(lysophosphatidylcholine) and sodium caseinate were added into bottled

water under magnetic agitation. The lipid mixture, preheated to 70 1C,was then added to the lysolecithin/sodium caseinate solution alsopreheated to 70 1C. The pre-dispersing treatment was done with anultraturrax (Kinematica, Lucerne, Switzerland) for 4 min at position 5/10then the emulsion was passed through a Rannie homogenizer (KindlerMaschinen, Zurich, Switzerland) in continuous mode for 5 min at 450 bars.The dispersions were then transferred into 100 ml Pyrex bottles undernitrogen to avoid oxidation during sterilization and storage. The cubosomedispersions were sterilized using a standard heat treatment of 30 min at121 1C in an autoclave Systec V100 (Systec, Wettenberg, Germany).

The base buffer consisted of 10.8 mM KCl, 240 mM NaCl, 4 mM Na2HPO4,20 mM glucose, 70 mM mannitol and 500 mg phenol red (nonabsorbablevolume marker) dissolved in normal distilled water and pH adjusted to7.4 with 0.5 M NaOH. The final perfusion solution was obtained by mixingthe base buffer solution and a base dispersion in a 1:1 proportion. Themixing was assisted by an ultrasonication probe and agitation to optimizehomogeneity of the product. The particles, present in the dispersion, hadan inverted bicontinuous structure of the double diamond type (spacegroup Pn-3 m). The structure was assessed by cryo-transmission electronmicroscopy and small angle X-ray scattering as reported previously.20,21

This structure corresponds to a good physical stability of the dispersionand, as shown by light scattering and optical microscopy, minimal (if any)particle aggregation between the mixing procedure and the perfusionevent. In this final perfusion solution, the concentration was 166.7 mg/lvitamin E and 250 mg/l vitamin E acetate. As the perfusion was performedat a flow rate of 5 ml/min for 60 min, a total amount of 50 mg vitamin E and75 mg vitamin E acetate was administered per treatment.

Perfusion systemAn in vivo human jejunal perfusion,22 extensively used by us23,24 wasutilized. A similar perfusion setup has also been described and evaluatedby several other groups.25–28 Briefly, it comprises a multi-channel catheterwith two occluding latex balloons (or multiple segments with moreballoons) placed at a variable distance (usually between 10 and 20 cm)apart from one another. Once the instrument is inserted, the balloons areinflated to create an isolated segment in the jejunum; a separate tubeallows proximal aspiration. A nonabsorbable dye, phenol red, can beadded as volume marker to correct for the water absorption/secretionand to ensure that there is no leakage to or from the segment. Effluent,blood and urine samples are collected to study the systemic bioavailability.As digestion is prevented, no enzymes are added and there is no biliary orpancreatic secretions, the in situ perfusion system enables the study ofabsorption processes excluding digestion and therefore resembles severemaldigestion conditions and tube feeding in clinical nutrition.

Treatment of subjectsThe overall experimental design is shown in Figure 1. The isolated jejunalsegment was perfused at a rate of 5 ml/min using a calibrated volumetricpump (COMPAT Enteral Feeding Pump, Novartis Medical Nutrition,Minneapolis, MN, USA). Intestinal effluent was collected by gravity in ice-chilled and light-protected containers at 4 1C every 15 min, aliquoted in5 ml samples, snap frozen and stored at � 80 1C. Collection of effluentthroughout the equilibration period was every 5 min. A perfusion bufferwithout vitamin E (1:1 dilution of the base buffer with distilled water) wasused during the equilibration and wash-out period. After washing theintestinal segment with the perfusion buffer for 30 min to eliminate theintestinal content and reach equilibrium, the main perfusion (60 min) withthe final perfusion solution followed. This included a single, combineddose of 50 mg of vitamin E and 75 mg vitamin E acetate delivered with aflow rate of 5 ml/min. Then the segment was rinsed in a post-irrigation stepwith the perfusion buffer for another 30 min. The perfusate from this post-irrigation step was also collected in 15-min fractions. In parallel to theperfusion, every 30 min, 3� 2 ml of the perfusion solution, that had notpassed the catheter, was taken and frozen at � 80 1C. As a final step, tominimize absorption of distally leaked formulation after its mixing with theintestinal juice following balloon deflation, the whole gut was rinsed with1 1l of osmotic solution (Solucion Evacuante Bohm, Lab Bohm, SA, Madrid,Spain) immediately after the final jejunal perfusion and before removalof the probe.

A venous catheter was placed in the forearm after successful tubeinsertion in the jejunum. Blood samples were collected in EDTA tubes(10-ml fractions) at time 0 (B1000 hours, beginning of perfusion), and atthe following time points after beginning the perfusion: 1, 3, 5, 7, 9 and11 h (first day), 24, 26, 28, 30, 32, 34 h (second day) and 48, 56 h (third day).

Table 1. Composition of the final, ready-to-use dispersion used forthe perfusion

Component Absolute amountin 300 ml (mg)

Vitamin E (2H6-a-tocopherol) 51Vitamin E acetate (2H3-a-tocopherylacetate) 76.5Monoglyceride 1932Ascorbyl palmitate 51Polyglycerol ester 339Lysophosphatidylcholine 19.5Sodium caseinate 390

Vitamin E absorptionK Nagy et al

203

& 2013 Macmillan Publishers Limited European Journal of Clinical Nutrition (2013) 202 – 206

Samples obtained from the intervention were aliquoted and frozenimmediately. Blood samples were collected into EDTA tubes, centrifugedfor 10 min at 3000 g at 4 1C to extract the plasma, which was keptat � 80 1C.

Chemicals. High pressure liquid chromatography grade water, ethanol,1,4-dioxane, n-hexane and BHT (2,6-di-tert-butyl-4-methylphenol) wereobtained from VWR International AG (Dietikon, Switzerland). All-rac 2H9-a-tocopheryl acetate (2-methyl-5,7,8-tri(methyl-2H3)-2-(40 ,80 ,120-trimethyltri-decyl)-6-chromanol-acetate was obtained from Sigma-Aldrich ChemieGmbH (Buchs, Switzerland). All-rac 2H9-a-tocopherol (2-methyl-5,7,8-tri(methyl-2H3)-2-(40 ,80 ,120-trimethyltridecyl)-6-chromanol) was obtainedfrom Chemaphor Inc., Ottawa, Canada.

Sample analysis. Sample preparation, liquid chromatography andmass spectrometric analysis of the samples were performed as reportedelsewhere.29

Statistical evaluation. Blood and effluent data were standardized by themedian influent value. Vitamin E and vitamin E acetate area under thecurve, in blood and in effluent, were calculated on the standardized valuesand were corrected for baseline. Area under the curve values werecalculated according to the trapezoidal rule. Taking into account thedistribution, Wilcoxon signed-rank tests were performed. Statisticalanalyses were done with Statistical Analysis Systems software (version9.1; SAS Institute, Cary, NC, USA).

RESULTSLuminal disappearance of vitamin ESimultaneous perfusion of differentially labeled RRR-[5,7-methyl-(2H6)]-a-tocopherol and RRR-[5-methyl-2H3]-a-tocopherol acetateinto the jejunum was performed, and influent and effluentsamples were analyzed by liquid chromatography–mass spectro-metry. Mean data on all six subjects are shown in Figure 2. Asexpected, the analyzed vitamin E concentrations in the influentsamples are close to the expected perfused concentration.The samples were also analyzed for 2H3-a-tocopherol. Althoughfree 2H3-a-tocopherol is not present in the perfusion solution(influent), only its acetylated form, 2H3-a-tocopherol appeared inthe effluent (Figure 3) demonstrating that the acetylated form ofvitamin E is cleaved by wall-bound enzymes in the lumen of thesmall intestine.

Plasma appearance of perfused vitamin EBoth 2H3- and 2H6-a-tocopherol appeared in the plasma as aconsequence of perfusion (Figure 4). Appearance of labeledtocopherols originating from the perfusion started at 2–5 h andreached Cmax at B10 h. This is in accordance with the findingsof Cheeseman et al.30 who found similar kinetics of free andacetylated a-tocopherol but in the presence of pancreas and bile

Figure 1. Flowchart showing the design of the perfusion experiment.

0

100

200

300

400

500

600

-20

Co

nce

ntr

atio

n [

mic

rom

ole

s/L

]

Time [min]

A

BC

D

100806040200

Figure 2. Mean concentrations of 2H6-a-tocopherol (B and D) and of 2H3-a-tocopherylacetate (A and C) during perfusion (n¼ 6) for influent(A and B) and effluent (C and D).

Vitamin E absorptionK Nagy et al

204

European Journal of Clinical Nutrition (2013) 202 – 206 & 2013 Macmillan Publishers Limited

salts (that is, normal oral administration). From the Tmax onwards,both forms of tocopherol diminished slowly to 30–50% of Cmax at56 h. This suggests that elimination/metabolism half-time ofa-tocopherol from plasma is around 2 days, consistent withthe literature on normally fed subjects16–19,31 and on 14C-labeleda-tocopherol.32 These suggest that the applied self-assemblysystem, as described in the Materials and methods section,enables the absorption of vitamin E even in the absence of bilesalts and pancreatic enzymes. The concentration of unlabeleda-tocopherol was in the range of 20–30 mmol/l for the six subjects.The averaged absolute levels of deuterated vitamin E were 1–2%of the endogenous, nonlabeled form. The latter appeared stablefor all subjects along the duration of the study despite therestriction in the diet. Data from all volunteers are combinedand shown in Table 2. Statistical analysis shows that there is nodifference between the absorption of the two forms of tocopherol(2H6-2H3-a-tocopherol estimated difference¼ � 0.049; lower95% confidence interval � 0.144, upper 95% confidenceinterval¼ 0.045, P-value¼ 0.156). Although Brisson et al.33 havereported that in Caco-2 cells the absorption of free and acetylatedvitamin E was similar, this study is the first to show this in humansin the absence of bile salts and pancreatic secretions.

DISCUSSIONWe compared the amount of 2H3-a-tocopherol appearing inplasma as a result of perfusing 2H3-a-tocopherol acetate to

published values after oral consumption of vitamin E. Taking Cmax

as a basis for comparison, the absorbed amount of 2H3-a-tocopherol originating from 2H3-a-tocopherol acetate (averagedfor all subjects) was 0.402 mM or 0.174 mg/l in our study. Weobserved substantial interindividual variation even in such aperfusion system where differences between individuals such astransit time and digestive secretions are eliminated. This compareswith previously published values where an intake of 75 mgvitamin E acetate resulted in a plasma Cmax of 4 mM after oral intakewith standard breakfast containing 30–40 g fat.18,19,31 Based onthis, vitamin E absorption via our perfusion model is B10%compared with absorption reported in the literature via normal,oral intake conditions. These values suggest efficient absorption inthe closed segmental in vivo perfusion setup, as only 20 cm of thegut is available 41 h for absorption, whereas via oral intake thewhole gut length is available for absorption during the entiretransit time.

We also compared the amount of absorbed 2H3-a-tocopherolappearing in plasma as a result of perfusing 2H3-a-tocopherolacetate with the initial absolute amount of 2H3-a-tocopherol acetateused during the perfusion. Assuming that the absorbed Cmax

0.402mM2H3-a-tocopherol is distributed exclusively in the plasma

fluid (B4 l), this corresponds to 1.609mmol absolute amount of2H3-a-tocopherol. Considering the fact that the perfusion included157. 9mmol (75 mg) absolute amount of 2H3-a-tocopherol acetate,this corresponds to 1% absorbed amount from the whole perfusedvolume.

By applying the segmental in vivo perfusion model to healthyvolunteers, we show that even with the exclusion of digestiveenzymes and bile salts, the appropriately solubilized acetylatedform of vitamin E can be extensively cleaved and ultimatelyabsorbed in the small intestine. As the observed bioavailability ofa-tocopherylacetate is similar to that of free a-tocopherol, theacetylated form can be used to deliver vitamin E under thesemaldigestion conditions. We propose that either brush borderenzymes, enzyme released after cellular damage or enzymes onthe exterior of shed cells are responsible for the extensivecleavage of a-tocopherylacetate.

CONFLICT OF INTERESTKornel Nagy, Marie-Claude Courtet-Compondu, Sophie Braga-Lagache, KarineRedeuil, Maurice Beaumont, Julie Moulin, Simon Acquistapache, Laurent Sagalowiczand Birgit Holst are employees of Nestle (Nestec Ltd). Gary Williamson was previouslya part time employee of Nestle (Nestec Ltd) and Martin Kussmann was previously anemployee of Nestle (Nestec Ltd). Remaining authors declare no conflict of interest.

ACKNOWLEDGEMENTSWe thank Rosemarie Jenni, Corinne Appolonia-Nouzille, Veronique Clement and DrsLaurent Fay, Michael Affolter, Corinne Magliola, Serge Rezzi and Martin Leser for theconstructive consultations and for their valuable help. We also thank Milagros Gallart,Monserrat Casellas, Purificacion Rodrıguez for their help in preparing perfusionbuffers and assisting on experimental procedures. This work was funded by NestecLtd, Switzerland.

0

10

20

30

40

50

-20Co

nce

ntr

atio

n [

mic

rom

ole

s/L

]

Time [min]

Subject 1Subject 2Subject 3Subject 4Subject 5Subject 6

100806040200

Figure 3. Concentration of 2H3-a-tocopherol (n¼ 6) after subtractionof background in the effluent resulting from hydrolysis of perfused2H3-a-tocopherylacetate. Note that the maximum value in the caseof subject 4 was 157 mM.

0

20

40

60

80

100

0

Pla

sma

con

cen

trat

ion

(nan

om

ole

s/L

)

Time after treatment [h]605040302010

Figure 4. Plasma concentration of stable isotope-labeled a-toco-pherols derived from the perfusion for one subject. 2H3-a-tocopherol (K) is derived from perfused 2H3-a-tocopherylacetateand 2H6-a-tocopherol (&) from perfused 2H6-a-tocopherol. The errorbars represent analytical triplicates. In the case of this subject, theunlabeled a-tocopherol was 28 (±2.5)mM along the time courseof the study.

Table 2. Baseline corrected and standardized AUC values ofdeuterated vitamin E after perfusion in a 0–56-h time period

N MeanAUC (%influent)

s.d. s.e.m. Minimum Maximum

2H3-a-tocopherol

6 1.93 2.32 0.95 0.42 6.33

2H6-a-tocopherol

6 1.84 2.2 0.9 0.4 5.99

Abbreviation: AUC, area under the curve.

Vitamin E absorptionK Nagy et al

205

& 2013 Macmillan Publishers Limited European Journal of Clinical Nutrition (2013) 202 – 206

REFERENCES1 Ball G. Bioavailability and analysis of vitamins in foods. Vitamin E. Chapman and

Hall: London 195–239, 1998.2 Ruperez FJ, Martin D, Herrera E, Barbas C. Chromatographic analysis of

alpha-tocopherol and related compounds in various matrices. J Chromatogr A2001; 935: 45–69.

3 Burton GW, Joyce A, Ingold KU. Is vitamin E the only lipid-soluble, chain-breakingantioxidant in human blood plasma and erythrocyte membranes? Arch BiochemBiophys 1983; 221: 281–290.

4 Hoppe PP, Krennrich G. Bioavailability and potency of natural-sourceand all-racemic alpha-tocopherol in the human: a dispute. Eur J Nutr 2000; 39:183–193.

5 Ingold KU, Webb AC, Witter D, Burton GW, Metcalfe TA, Muller DP. Vitamin Eremains the major lipid-soluble, chain-breaking antioxidant in human plasmaeven in individuals suffering severe vitamin E deficiency. Arch Biochem Biophys1987; 259: 224–225.

6 Morrissey PA, Sheehy PJ. Optimal nutrition: vitamin E. Proc Nutr Soc 1999; 58:459–468.

7 Bramley PM, Elmadfa I, Kafatos A, Kelly FJ, Manios Y, Roxborough HE et al. VitaminE. J Sci Food Agr 2000; 80: 913–938.

8 Traber MG, Sies H. Vitamin E in humans: demand and delivery. Ann Rev Nutr 1996;16: 321–347.

9 Gey KF, Puska P, Jordan P, Moser UK. Inverse correlation between plasma vitaminE and mortality from ischemic heart disease in cross-cultural epidemiology.Am J Clin Nutr 1991; 53: 326S–334S.

10 Brigelius-Flohe R, Traber MG. Vitamin E: function and metabolism. FASEB J 1999;13: 1145–1155.

11 FAO/WHO (Food and agriculture organization of the united nations/WorldHealth Organization). Human Vitamin and Mineral Requirements 2002121–131.

12 Muller DP, Harries JT, Lloyd JK. The relative importance of the factors involved inthe absorption of vitamin E in children. Gut 1974; 15: 966–971.

13 Jeanes YM, Hall WL, Ellard S, Lee E, Lodge JK. The absorption of vitamin E isinfluenced by the amount of fat in a meal and the food matrix. Br J Nutr 2004; 92:575–579.

14 Bruno RS, Leonard SW, Park S, Zhao YY, Traber MG. Human vitamin E require-ments assessed with the use of apples fortified with deuterium-labeledalpha-tocopheryl acetate. Am J Clin Nutr 2006; 83: 299–304.

15 Roxborough HE, Burton GW, Kelly FJ. Inter- and intra-individual variation inplasma and red blood cell vitamin E after supplementation. Free Radic Res 2000;33: 437–445.

16 Traber MG, Rader D, Acuff RV, Ramakrishnan R, Brewer HB, Kayden HJ. Vitamin Edose-response studies in humans with use of deuterated RRR-alpha-tocopherol.Am J Clin Nutr 1998; 68: 847–853.

17 Traber MG, Ramakrishnan R, Kayden HJ. Human plasma vitamin E kineticsdemonstrate rapid recycling of plasma RRR-alpha-tocopherol. Proc Natl Acad SciUSA 1994; 91: 10005–10008.

18 Lodge JK, Traber MG, Elsner A, Brigelius-Flohe R. A rapid method for the extrac-tion and determination of vitamin E metabolites in human urine. J Lipid Res 2000;41: 148–154.

19 Proteggente AR, Turner R, Majewicz J, Rimbach G, Minihane AM, Kramer K et al.Noncompetitive plasma biokinetics of deuterium-labeled natural and syntheticalpha-tocopherol in healthy men with an apoE4 genotype. J Nutr 2007; 135:1063–1069.

20 Sagalowicz L, Michel M, Adrian M, Frossard P, Rouvet M, Watzke HJ et al. Crys-tallography of dispersed liquid crystalline phases studied by cryo-transmissionelectron microscopy. J Microsc 2006; 221: 110–121.

21 Yaghmur A, de CL, Sagalowicz L, Leser ME, Glatter O. Control of the internalstructure of MLO-based isasomes by the addition of diglycerol monooleate andsoybean phosphatidylcholine. Langmuir 2006; 22: 9919–9927.

22 Lennernas H, Ahrenstedt O, Hallgren R, Knutson L, Ryde M, Paalzow LK. Regionaljejunal perfusion, a new in vivo approach to study oral drug absorption in man.Pharm Res 1992; 9: 1243–1251.

23 Santos J, Bayarri C, Saperas E, Nogueiras C, Antolin M, Mourelle M et al.Characterisation of immune mediator release during the immediate response tosegmental mucosal challenge in the jejunum of patients with food allergy.Gut 1999; 45: 553–558.

24 Santos J, Saperas E, Nogueiras C, Mourelle M, Antolin M, Cadahia A et al.Release of mast cell mediators into the jejunum by cold pain stress in humans.Gastroenterology 1998; 114: 640–648.

25 Lennernas H. Human jejunal effective permeability and its correlation withpreclinical drug absorption models. J Pharm Pharmacol 1997; 49: 627–638.

26 Lennernas H. Human intestinal permeability. J Pharm Sci 1998; 87: 403–410.27 Lennernas H, Nylander S, Ungell AL. Jejunal permeability: a comparison between

the ussing chamber technique and the single-pass perfusion in humans. PharmRes 1997; 14: 667–671.

28 Takamatsu N, Kim ON, Welage LS, Idkaidek NM, Hayashi Y, Barnett J et al. Humanjejunal permeability of two polar drugs: cimetidine and ranitidine. Pharm Res2001; 18: 742–744.

29 Nagy K, Courtet-Compondu MC, Holst B, Kussmann M. Comprehensive analysisof vitamin E constituents in human plasma by liquid chromatography-massspectrometry. Anal Chem 2007; 79: 7087–7096.

30 Cheeseman KH, Holley AE, Kelly FJ, Wasil M, Hughes L, Burton G. Biokinetics inhumans of RRR-alpha-tocopherol: the free phenol, acetate ester, and succinateester forms of vitamin E. Free Radic Biol Med 1995; 19: 591–598.

31 Leonard SW, Good CK, Gugger ET, Traber MG. Vitamin E bioavailability fromfortified breakfast cereal is greater than that from encapsulated supplements. AmJ Clin Nutr 2004; 79: 86–92.

32 Chuang JC, Matel HD, Nambiar KP, Kim SH, Fadel JG, Holstege DM et al. Quan-titation of [5-14CH3]-(2R, 40R, 80R)-alpha-tocopherol in humans. J Nutr 2011; 141:1482–1488.

33 Brisson L, Castan S, Fontbonne H, Nicoletti C, Puigserver A, Ajandouze H.Alpha-tocopheryl acetate is absorbed and hydrolyzed by Caco-2 cells compara-tive studies with alpha-tocopherol. Chem Phys Lipids 2008; 154: 33–37.

Vitamin E absorptionK Nagy et al

206

European Journal of Clinical Nutrition (2013) 202 – 206 & 2013 Macmillan Publishers Limited