Vitamin E catabolism in women, as modulated by food and by fat, studied using 2 deuterium-labeled α-tocopherols in a 3-phase, nonrandomized crossover study

ABSTRACT

Background

Human vitamin E (α-tocopherol) catabolism is a mechanism for regulating whole-body α-tocopherol.

Objectives

To determine the roles of the intestine and liver on α-tocopherol catabolism as affected by fat or fasting, 2 deuterium-labeled (intravenous d₆- and oral d₃-) forms of α-tocopherol were used.

Methods

Healthy women received intravenous d₆-α-tocopherol and consumed d₃-α-tocopherol with a 600-kcal defined liquid meal (DLM; 40% or 0% fat, n = 10) followed by controlled meals; or the 0% fat DLM (n = 7) followed by a 12-h fast (0% fat-fast), then controlled meals ≤72 h. The order of the 3-phase crossover design was not randomized and there was no blinding. Samples were analyzed by LC/MS to determine the α-tocopherol catabolites and α-carboxyethyl hydroxychromanol (α-CEHC) in urine, feces, and plasma that were catabolized from administered oral d₃- and intravenous d₆-α-tocopherols.

Results

Urinary and plasma d₃- and d₆-α-CEHC concentrations varied differently with the interventions. Mean ± SEM cumulative urinary d₆-α-CEHC derived from the intravenous dose excreted over 72 h during the 40% fat (2.50 ± 0.37 μmol/g creatinine) and 0% fat (2.37 ± 0.37 μmol/g creatinine) interventions were similar, but a ∼50% decrease was observed during the 0% fat-fast (1.05 ± 0.39 μmol/g creatinine) intervention (compared with 0% fat, P = 0.0005). Cumulative urinary d₃-α-CEHC excretion was not significantly changed by any intervention. Total urinary and fecal excretion of catabolites accounted for <5% of each of the administered doses.

Conclusions

Differential catabolism of the intravenous d₆-α-tocopherol and oral d₃-α-tocopherol doses shows both liver and intestine have roles in α-tocopherol catabolism. During the 40% fat intervention, >90% of urinary d₃-α-CEHC excretion was estimated to be liver-derived, whereas during fasting <50% was from the liver with the remainder from the intestine, suggesting that there was increased intestinal α-tocopherol catabolism while d₃-α-tocopherol was retained in the intestine in the absence of adequate fat/food for α-tocopherol absorption.

This trial was registered at clinicaltrials.gov as NCT00862433.

Introduction

Vitamin E concentrations and forms are closely regulated by the human body (1). The major regulatory mechanism is the hepatic α-tocopherol transfer protein, which mediates the preferential α-tocopherol secretion from the liver into plasma lipoproteins for transport to extrahepatic tissues (2–4). The other mechanism is the xenobiotic catabolism of non-α-tocopherol vitamin E forms (5, 6) and “excess” α-tocopherol (3, 4). The cytochrome P450 4F2 (CYP4F2) is responsible for initiating vitamin E catabolism (6, 7). Although the tissues involved in vitamin E catabolism are not well identified, CYP4F2 is expressed in liver and kidney (8), as well as intestine (9, 10). CYP4F2 ω-hydroxylates the side chain of both tocopherols and tocotrienols to form the long-chain 13′-hydroxy (13′-OH) catabolite (6, 11). The catabolite undergoes ω-oxidation to form the 13′-carboxy (13′-COOH) catabolite, which then undergoes several rounds of β-oxidation (12) to form carboxyethyl hydroxychromanol (CEHC) (Figure 1) (11).

FIGURE 1.

Structures of α-T, intermediate catabolites (13′OH-α-T, 13′-COOH-α-T, 11′-COOH-α-T, 9′-COOH-α-T, 7′-COOH-α-T, α-CMBHC), and α-CEHC with the locations of substituent groups shown as R₁ and R₂. The locations of unlabeled (CH₃)- and deuterium-labeled (CD₃)-methyl groups are shown in the box for unlabeled (d₀), d₃-, and d₆-α-Ts and the corresponding catabolites. COOH, carboxy; OH, hydroxy; α-CEHC, 3-(6-hydroxy-2,5,7,8-tetramethylchroman-2-yl)propanoic acid; α-CMBHC, 5-(6-hydroxy-2,5,7,8-tetramethyl-chroman-2-yl)-2-methyl-pentanoic acid; α-T, α-tocopherol.

Over 25 y ago, Schultz et al. (13) suggested that α-CEHC reflects α-tocopherol adequacy. Those studies were limited by the available methods. More sensitive methodologies established that when α-tocopherol intakes in healthy people are adequate, urinary α-CEHC excretion is low, but it increases dramatically with supplemental α-tocopherol intakes (14), suggesting α-CEHC reflects α-tocopherol status. This concept was investigated in older adults, who consumed hazelnuts (∼9 mg α-tocopherol) daily for 16 wk and demonstrated limited change in plasma α-tocopherol concentrations, but showed a 33% increase in urinary α-CEHC excretion (15). By contrast, both labeled and unlabeled α-CEHC excretion amounts (16) were lower due to low α-tocopherol bioavailability in persons with metabolic syndrome compared with amounts excreted by healthy adults (17). These findings indicate that urinary α-CEHC excretion varies even with small changes in consumed α-tocopherol (15).

Recently, we reported a clinical trial in normal-weight healthy women that evaluated α-tocopherol absorption in response to 40% or 0% fat intakes consumed with the α-tocopherol dose or in response to fasting after the 0% fat meal (18). Plasma α-tocopherol fractional absorption and α-tocopherol pharmacokinetic parameters were measured in samples collected ≤72 h after intravenous administration of an emulsion containing a stable-isotope-labeled α-tocopherol [hexa-deuterium (d₆)-α-tocopherol] and oral administration of a stable-isotope-labeled α-tocopherol [tri-deuterium (d₃)-α-tocopherol]. α-Tocopherol absorption ranged on average between 55% and 74%, as evaluated using multiple methodologies, but was unaffected by dietary fat [40% compared with 0% fat in the defined liquid meal (DLM) accompanying the oral dose] or by a 12-h fast after participants consumed d₃-α-tocopherol with the 0% fat DLM. The objective of the current study was to evaluate vitamin E catabolism by using measures of labeled and unlabeled α-CEHC in plasma, urine, and feces collected during the previously reported clinical trial (18). The current report provides new and unique information concerning the contribution by the intestinal system and the liver in α-tocopherol catabolism and α-CEHC excretion by women in response to these changes in meal fat and to fasting.

Methods

Clinical research trial

The present study is a first phase of arm 1, an exploratory phase-1 clinical trial, which is embedded in a larger trial (NCT00862433) and is being conducted at the NIH Clinical Research Center (NIH CRC). This first phase of arm 1’s design, the initial vitamin E pharmacokinetic outcomes, and the power analysis have been published previously (18); highlights are described briefly herein. This study focused on the analysis of the vitamin E catabolites. The NIH National Institute of Diabetes and Digestive and Kidney Diseases/National Institute of Arthritis and Musculoskeletal and Skin Diseases Institutional Review Board (IRB) for Protection of Human Subjects approved the protocol (09-DK-0097); the Oregon State University IRB deferred to the NIH-IRB. The study was performed under an Investigational New Drug protocol (110033) issued by the FDA to ML. Sample collection for the trial described herein took place between January 2015 and August 2016.

Participants were adult normotensive (<160/90 mm Hg), nonobese (BMI <29.9 kg/m²), nondiabetic women (aged 18–40 y) able to give informed consent (Supplemental Table 1).

We used a crossover design, but we did not randomize the order and there was no blinding (Supplemental Figure 1).

Study procedures

Participants were housed in the NIH CRC for ≤5 d during each intervention (Supplemental Figure 2). To avoid confounding results with variable vitamin C status (19), vitamin C stores were saturated in all subjects (20–22). During the ≤5-d stays, all meals were controlled, provided, and monitored by the NIH CRC Metabolic Kitchen staff. Participants were admitted the evening before dosing and consumed a low-vitamin-E dinner and snack. On dosing day, participants consumed a 600-kcal DLM at 09:45. The DLM contained 1) 40% fat, 2) 0% fat, or 3) 0% fat followed by a 12-h fast with dinner at 22:00 (18). Hereafter, these interventions are called 40% fat, 0% fat, or 0% fat-fast, respectively. Interventions included the following participants: 40% fat (n = 10), subject numbers: 1–9, 12; 0% fat (n = 10), subject numbers: 1–5, 8–11, 13; 0% fat-fast (n = 7), subject numbers 1, 2, 4, 5, 9, 10, 11. Five women participated in all 3 interventions: subject numbers: 1, 2, 4, 5, 9. Not all subjects completed all trials because some declined further participation owing to the required time commitments.

The α-tocopherol doses were 30 mg each of the oral d₃-α-tocopherol (69.3 μmol) and intravenous d₆-α-tocopherol (68.8 μmol). Each of these doses is double the current vitamin E RDA of 15 mg (23), but substantially smaller than commonly used vitamin E supplements. The dose was chosen to allow the labels to be followed for ≤30 d. No adverse effects were observed.

Deuterated-α-tocopherol was administered via the 2 routes, as follows. At 09:45 each participant consumed approximately two-thirds of the DLM. The oral dose was then administered by placing it on the tongue of the participant, who then consumed the remaining DLM. The intravenous d₆-α-tocopherol dose (5.6 mL) was drawn from a single-use vial into a syringe. Immediately upon finishing the DLM, the intravenous dose was administered via intravenous cannula by slow injection over ∼6 min. Blood sample collection from the contralateral intravenous cannula commenced with the beginning of intravenous dosing. The schedule included collection of multiple samples (Supplemental Figure 2).

Blood collection tubes (containing sodium heparin) were kept on ice for <30 min before plasma and erythrocyte isolation by centrifugation for 15 min at 500 × g at 4 °C. Plasma was divided into aliquots in cryovials, which were frozen in liquid nitrogen. Urine was collected every 4 h during the first 24 h, then every 8 h ≤96 h. Total urine volume was measured, aliquots taken, creatinine measured, and samples frozen. Because liquid intake was not limited, creatinine was used for correction of volume variability. Dietary creatinine/creatine was not controlled in the run-up to the study, but subjects were instructed to stop any dietary supplements such as creatine in the run-up.

Some participants chose not to collect feces; some subjects collected feces at 1, 2, or 3 times ≤96 h. Fecal samples collected from subjects in the 0% fat-fast intervention were insufficient for analysis.

Samples were kept frozen at −80°C until shipped on dry ice by overnight freight to the Linus Pauling Institute.

Analytical procedures: vitamin E metabolite analyses

Urine and plasma labeled and unlabeled CEHCs were extracted using a modified method of Li et al. (24), as described (25). Briefly, plasma or urine was added to a 10-mL screw-cap tube containing 0.8 mL Milli-Q water and 0.5 mL 2% ascorbic acid solution. To hydrolyze conjugates, the samples were acidified with 0.5 mL HCl and incubated for 1 h at 60°C. An internal standard (IS), trolox (Sigma), was added and CEHCs extracted with 4 mL diethyl ether; an aliquot of the ether fraction was collected and dried under nitrogen. Samples were resuspended in 1:1 (vol:vol) water:methanol and injected into the LC/MS/MS. Samples were analyzed with either an API 3000 (Applied Biosystems/MDS Sciex) with a Turbo Ion Spray source [electrospray ionization (ESI)] operated in negative mode, or a Waters XEVO TQD triple quadrupole mass spectrometer using an ESI source operated in negative mode with an Acquity UPLC H-Class Separations Module. A subset of samples was analyzed and quantified with both pieces of equipment and results were found to be not significantly different (data not shown). For the API 3000, a SymmetryShield RP-18 column (3.0 × 150 mm, 3.5 µm particle; Waters) with a Symmetry-Shield Sentry RP-18 precolumn (3.9 × 20 mm, 3.5 µm particle; Waters) was used for separation. Instrument control and data acquisition for the Xevo TQD were performed with Waters Masslynx version 4.1 TargetLynx software. The column used was an Acquity UPLC BEH C-18 column (2.1 × 50 mm, 1.7 µm particle; Waters) with an Acquity UPLC BEH C-18 VanGuard precolumn (2.1 × 5 mm, 1.7 µm particle; Waters), which was maintained at 40°C. Multiple reaction monitoring m/z ratios were obtained similarly for both instruments, as follows (parent/fragment): d₀-α-CEHC, m/z 277/163; d₃-α-CEHC, m/z 280/166; d₆-α-CEHC, m/z 283/169; d₀-γ-CEHC, m/z 263/149; and trolox, m/z 249/163. Typical retention times for HPLC separations for trolox, d₀-γ-CEHC, and α-CEHCs were 11.7, 11.8, and 12.5 min and by UPLC were 2.7, 2.8, and 2.9 min, respectively. Sample CEHC concentrations were calculated using peak area ratios of the corresponding ion to the trolox (IS) and an external standard curve of either d₀-α- or d₀-γ-CEHC (Cayman Chemical). The d₆-α-CEHC or d₃-α-CEHC excreted are reported per time point or as the cumulative d₆-α-CEHC or d₃-α-CEHC excreted, respectively calculated by summing the micromoles per gram of creatinine for each urine sample collected over 72 h. The time course of urinary d₆-α-CEHC or d₃-α-CEHC (μmol/g creatinine) or plasma (nmol/L), respectively, for each individual was used to assess the time (Tmax) of maximum concentration (Cmax), and the area under the concentration × time curves, using Prism 6 for Mac OSX (Graphpad Software).

To determine whether the 2 different routes of α-tocopherol administration differed across interventions, the roles of the liver and intestine in vitamin E catabolism were assessed. The d₃-α-tocopherol amount absorbed was estimated based on our previous findings for each participant (18), then this value was used as the denominator in the calculation of the ratio of d₃-CEHC excreted:d₃-α-tocopherol absorbed × 100, whereas the d₆-α-tocopherol amount injected was used as the denominator for the ratio of d₆-CEHC excreted:injected × 100. The difference between the 2 ratios represents the contribution of the intestine to the urinary d₃-α-CEHC excreted as the percentage of dose absorbed; the contribution of the liver equals the ratio of d₆-CEHC excreted:injected d₆-α-tocopherol.

Fecal samples were collected by participants, frozen by nursing staff, kept frozen, shipped on dry ice by overnight freight, and kept frozen until analysis. Frozen fecal samples were transferred directly from −80°C storage to tared, quart-sized, plastic buckets and weighed. Then, a homogenizing solution [1% ascorbic acid in ultrapure water (wt:vol), 10 μmol diethylenetriaminepenta-acetic acid, 12.42 μmol δ-T, and 100 nmol δ-CEHC (recovery standards)] was added to the bucket in a 2:1 ratio to sample weight. The buckets were weighed again for total solution weight and sealed shut. Samples were allowed to thaw for a maximum of 18 h in a cold room (4°C). Each bucket containing the entire collected fecal sample in homogenizing solution was mixed for 15 min in a VR-1 Digital Paint Mixer (The Cary Company). Aliquots were taken from the homogenate and frozen in a −80°C freezer for future analyses.

Fecal homogenate aliquots were prepared for extraction and analysis by LC-MS/MS, as described (18). Duplicate samples for each collection of ∼0.25 g fecal homogenate were quickly thawed, weighed, then saponified before extraction of vitamin E and 13′-OH-α-tocopherol (long-chain metabolite). For the saponification step, samples were incubated with 10 mL ethanol, 5 mL 1% ascorbic acid in ultrapure water (wt:vol), and 1.5 mL saturated KOH in a water bath at 70°C for 30 min. Upon cooling, 5 mL water (1% ascorbic acid, wt:vol), 100 μL butylated hydroxy toluene (1 mg/mL in ethanol), and an additional IS 20 μL d₉-α-tocopherol (800 μmol) were added. After hexane extraction, an aliquot was evaporated under nitrogen and the dried residue resuspended in 1 mL 1:1 (vol:vol) ethanol:methanol. An appropriate aliquot was analyzed by LC-MS/MS, as described for plasma samples. Samples were corrected for δ-tocopherol recovery, which was 60%–90%. The total amount excreted daily was calculated from the averages for each collection, then summed over the study period.

For extraction of short-chain metabolites, a methanol extraction was performed as described (26). Briefly, ∼100 mg fecal homogenate was added to 0.5 mL methanol and 20 μL trolox (45 µM, IS), mixed on a vortex for 10 s, and centrifuged 10 min at 21,000 × gat 4 °C to pellet the solid matter. The supernatant was dried down under nitrogen, reconstituted in 100 μL 1:1 (vol:vol) water:methanol, and injected using a Waters XEVO TQD triple quadrupole mass spectrometer, as described above under urine and plasma catabolite analysis. Additional multiple reaction monitoring m/z ratios were obtained as follows (parent/fragment): 5-(6-hydroxy-2,5,7,8-tetramethyl-chroman-2-yl)-2-methyl-pentanoic acid (d₀-α-CMBHC), m/z 319/163; d₃-α-CMBHC, m/z 322/166; d₆-α-CMBHC, m/z 325/169; 7′-COOH-d₀-α-tocopherol, m/z 347/163; 7′-COOH-d₃-α-tocopherol, m/z 350/166; 7′-COOH-d₆-α-tocopherol, m/z 353/169; 9′-COOH-d₀-α-tocopherol, m/z 389/163; 9′-COOH-d₃-α-tocopherol, m/z 391/166; 9′-COOH-d₆-α-tocopherol, m/z 394/169; 11′-COOH-d₀-α-tocopherol, m/z 417/163; 11′-COOH-d₃-α-tocopherol, m/z 420/166; 11′-COOH-d₆-α-tocopherol, m/z 423/169; 13′-COOH-d₀-α-tocopherol, m/z 459/163; 13′-COOH-d₃-α-tocopherol, m/z 462/166; 13′-COOH-d₆-α-tocopherol, m/z 465/169; and δ-CEHC, m/z 249/135. Analytes were measured using the IS and corrected for loss of added δ-CEHC.

Statistical design

The statistical analysis used herein was consistent with that used for the primary endpoints of the trial (18) and was conducted using Statistical Analysis System (SAS) version 9.4 (SAS Institute).

The data for Tables 1 –4 satisfy the normality requirements and were analyzed using linear mixed models in SAS PROC MIXED. For the fecal data analysis in Table 5, we found that the data were not normally distributed. None of the most common transformations (logarithmic, square root, or arcsine transformation) achieved normality of the data and, therefore, a nonparametric approach was used (i.e., Wilcoxon's matched-pairs Signed Rank test) instead of linear mixed models.

TABLE 1.

Cumulative urinary d₃- and d₆-α-CEHCs excreted after oral d₃- and i.v. d₆-α-tocopherol administration during 3 interventions¹

DLM fat	d6-α-CEHC	Vs. 0% fat, P values2	d3-α-CEHC	Vs. 0% fat, P values2	d3- vs. d6-α-CEHC, P values3	Interaction, P values3
40% fat	2.50 ± 0.37	0.6174	1.92 ± 0.36	0.2991	0.0031	0.0003
0% fat	2.37 ± 0.37		2.29 ± 0.37		0.5696
0% fat-fast	1.05 ± 0.39	0.0005	1.70 ± 0.41	0.1259	0.0171

¹The least-square mean ± SEM d₆- or d₃-α-CEHC (μmol/g creatinine) excreted by women consuming a DLM (either 40% fat, n = 10; 0% fat, n = 10; or 0% fat-fast, n = 7) was summed over 72 h, as Figure 2 shows. DLM, defined liquid meal; d₃- and d₆-α-tocopherol, RRR-α-tocopherol labeled with tri-deuterium and hexa-deuterium, respectively; 0% fat, defined liquid meal with 0% fat; 0% fat-fast, defined liquid meal with 0% fat with a 12-h fast; 40% fat, defined liquid meal with 40% fat; α-CEHC, 3-(6-hydroxy-2,5,7,8-tetramethylchroman-2-yl)propanoic acid.

²Data were analyzed as an incomplete crossover design in SAS PROC MIXED with treatment (40% fat, 0% fat, and 0% fat-fast) as a fixed effect and subject as a random effect. Using the ESTIMATE statements, the effect of DLM fat was evaluated using the contrast between 40% fat and 0% fat interventions; the effect of fasting was evaluated using the contrast between 0% fat and 0% fat-fast interventions.

³For comparisons between the i.v. d₆-α-tocopherol and oral d₃-α-tocopherol parameters, the differences between d₆-α-CEHC and d₃-α-CEHC for each parameter were calculated, then the data analyzed as a crossover design in SAS PROC MIXED with treatment (40% fat, 0% fat, 0% fat-fast) as a fixed effect and subject as a random effect; the overall interaction was analyzed as a Latin square design with double repeated measures (Kronecker product) in SAS PROC MIXED. Fixed effects were interventions (40% fat, 0% fat, 0% fat-fast), the route of α-tocopherol administration (oral, i.v.), and their interaction. The variance-covariance structure within subjects was modeled by using a Kronecker product, in which the intervention was modeled using a compound symmetry matrix and the type of treatment within treatment was modeled using an unstructured variance-covariance matrix.

TABLE 4.

Pharmacokinetic parameters derived from the plasma d₆- and d₃-α-CEHC concentrations after oral d₃- and i.v. d₆-α-tocopherol administration during 3 interventions¹

Parameter	DLM fat	d6-α-CEHC	Vs. 0% fat, P values	d3-α-CEHC	Vs. 0% fat, P values	d3- vs. d6-α-CEHC, P values	Interaction, P values
AUC, nmol/L · h	40% fat	143 ± 19	0.1231	106 ± 20	0.5006	0.0330	0.0264
	0% fat	104 ± 19		90 ± 21		0.0830
	0% fat-fast	95 ± 23	0.7339	136 ± 24	0.0749	0.0786
Cmax, nmol/L	40% fat	6.4 ± 0.7	0.0421	4.7 ± 1.2	0.9359	0.0254	0.0088
	0% fat	4.3 ± 0.7		4.6 ± 1.2		0.6718
	0% fat-fast	3.5 ± 0.8	0.4158	7.5 ± 1.4	0.0588	0.0438
Tmax, h	40% fat	13.9 ± 1.5	0.8165	14.5 ± 1.2	0.5726	0.7813	0.0094
	0% fat	13.4 ± 1.5		13.5 ± 1.2		0.5396
	0% fat-fast	18.7 ± 1.8	0.0329	10.4 ± 1.5	0.0451	<0.0001

¹Parameters (least-square means ± SEMs) were calculated from the plasma CEHC concentrations in women as described for Table 1 and as shown in Figure 3: (D) 40% fat, n = 10; (E) 0% fat, n = 10; (F) 0% fat-fast, n = 7. Statistical analyses of the parameters derived from either d₆-α-CEHC or d₃-α-CEHC were performed as described for Table 3. Cmax, maximum concentration; DLM, defined liquid meal; d₃- and d₆-α-tocopherols, RRR-α-tocopherol labeled with tri-deuterium and hexa-deuterium, respectively; Tmax, time of maximum concentration; α-CEHC, 3-(6-hydroxy-2,5,7,8-tetramethylchroman-2-yl)propanoic acid.

TABLE 5.

α-Tocopherol and α-tocopherol catabolites, as percentage of administered vitamin E excreted in feces and urine ≤96 h, during the 40% and 0% fat interventions¹

	DLM fat
	40% Fat		0% Fat
Labeled catabolites	Oral d3-α-tocopherol	I.v. d6-α-tocopherol	Oral d3-α-tocopherol	I.v. d6-α-tocopherol
α-CEHC	0.01% (0.00%–0.01%)	0.04% (0.01%–0.06%)	0.00% (0.00%–0.01%)	0.02% (0.01%–0.03%)
α-CMBHC	0.03% (0.00%–0.05%)	0.02% (0.01%–0.06%)	0.00% (0.00%–0.01%)	0.01% (0.01%–0.01%)
7′-COOH-α-tocopherol	0.08% (0.02%–0.27%)	0.10% (0.02%–0.36%)	0.02% (0.01%–0.12%)	0.02% (0.01%–0.11%)
9′-COOH-α-tocopherol	0.01% (0.01%–0.06%)	0.04% (0.01%–0.09%)	0.01% (0.01%–0.05%)	0.05% (0.03%–0.07%)
11′-COOH-α-tocopherol	0.7% (0.1%–3.8%)	1.0% (0.3%–4.5%)	0.7% (0.1%–3.2%)	0.8% (0.1%–2.9%)
13′-COOH-α-tocopherol	0.4% (0.0%–0.5%)	0.5% (0.1%–1.0%)	0.1% (0.0%–0.4%)	0.3% (0.1%–0.5%)
13′-OH-α-tocopherol	0.1% (0.0%–0.2%)	0.1% (0.0%–0.2%)	0.3% (0.0%–0.4%)	0.1% (0.0%–0.2%)
Sum fecal α-tocopherol catabolites	1.8% (0.3%–5.0%)a	2.7% (0.4%–6.3%)a	1.2% (0.3%–4.0%)	1.3% (0.3%–3.7%)
Fecal α-tocopherol	38.0% (33.0%–56.0%)	3.4% (1.3%–4.1%)	27.6% (14.4%–33.2%)	2.1% (1.1%–2.7%)
Urinary α-CEHC	0.7% (0.4%–0.9%)b	0.9% (0.7%–1.1%)b	1.1% (0.6%–1.5%)	1.1% (0.7%–1.4%)

¹After determining that the data were not normally distributed, median (IQR) percentages for each d₃- or d₆-α-catabolite were calculated from the administered vitamin E excreted in feces over 96 h from women during the 40% fat or 0% fat interventions. The fecal catabolites were also totaled for each participant to obtain the sum of the fecal catabolites excreted by each participant. Also shown are the fecal d₃- and d₆-α-tocopherols excreted, as previously reported (18), and the urinary α-CEHC excreted, calculated as total excreted over 72 h divided by the administered dose × 100. Because the data were not normally distributed, statistical analysis was performed using Wilcoxon's matched-pairs Signed Rank test. Labeled catabolite structures are shown in Figure 1. COOH, carboxy; d₃- and d₆-α-tocopherols, RRR-α-tocopherol labeled with tri-deuterium and hexa-deuterium, respectively; OH, hydroxy; α-CEHC, 3-(6-hydroxy-2,5,7,8-tetramethylchroman-2-yl)propanoic acid, α-CMBHC, 5-(6-hydroxy-2,5,7,8-tetramethyl-chroman-2-yl)-2-methyl-pentanoic acid.

^a,bMedians (IQRs) not sharing a common superscript letter in the same row are significantly different at P < 0.005. There were no consistent trends noted between the 40% fat and 0% fat interventions for the various individual fecal catabolites. Fecal α-tocopherol contains unabsorbed d₃-α-tocopherol, so no comparisons were made between d₆- and d₃-α-tocopherol. The fecal d₆-α-tocopherol is from the i.v. dose so is likely to have been secreted from the liver in bile (27) and is included for relative comparisons of tocopherol and catabolite excretion. Only those subjects who provided fecal samples are shown: 40% fat (n = 9 fecal samples, subject numbers 1–5, 7–9, 12; n = 10 urine samples); 0% fat (n = 8 fecal samples, subject numbers 1–5, 8, 11, 13; n = 10 urine samples).

In Tables 1 –4, 2 types of comparisons are shown. For comparisons within the intravenous d₆-α-tocopherol and oral d₃-α-tocopherol parameters, data were analyzed as an incomplete crossover design in SAS PROC MIXED with treatment (40% fat, 0% fat, and 0% fat-fast interventions) as a fixed effect and subject as a random effect. Using the ESTIMATE statements, the effect of DLM fat was evaluated using the contrast between 40% fat and 0% fat interventions; the effect of fasting was evaluated using the contrast between 0% fat and 0% fat-fast interventions. For comparisons between the intravenous d₆-α-tocopherol and oral d₃-α-tocopherol parameters, the differences between d₆-α-CEHC and d₃-α-CEHC for each parameter were calculated, then the data analyzed as a crossover design in SAS PROC MIXED with treatment (40% fat, 0% fat, 0% fat-fast) as a fixed effect and subject as a random effect; the overall interaction was analyzed as a Latin square design with double repeated measures (Kronecker product) in SAS PROC MIXED. Fixed effects were interventions (40% fat, 0% fat, 0% fat-fast), the route of α-tocopherol administration (oral, intravenous), and their interaction. The variance-covariance structure within subjects was modeled by using a Kronecker product, in which the intervention was modeled using a compound symmetry matrix and the type of treatment within treatment was modeled using an unstructured variance-covariance matrix.

Data are reported as least-squares means ± SEMs, with the exception of the fecal catabolites, which are shown as medians (IQRs). All statistical tests were 2-sided and statistical significance was set as P ≤ 0.05. Because we report on an exploratory phase 1 trial and all outcomes shown in Tables 1 –5 are part of the primary outcomes (vitamin E catabolism and pharmacokinetics), no corrections for multiple-comparison adjustments were performed (28).

Results

Urinary CEHC excretion

Deuterated α-tocopherol was administered via 2 routes: the oral d₃-α-tocopherol dose was consumed with a DLM; immediately thereafter the d₆-α-tocopherol dose in an emulsion resembling chylomicrons was injected intravenously. The d₆-α-tocopherol represents 100% delivered to the circulation and delivered to the liver (18). Previously, the ratios of plasma d₃-α-tocopherol to d₆-α-tocopherol concentrations were used successfully to estimate fractional α-tocopherol absorption (18). Therefore, it was hypothesized that oral d₃-α-tocopherol would mix well with the intravenous d₆-α-tocopherol in the liver, then liver catabolism would also reflect these same ratios, and, thus, the ratios of the cumulative excreted urinary d₃-α-CEHC to d₆-α-CEHC could also be used as a measure of α-tocopherol absorption. However, this expectation was not met because the excretion of d₃- and d₆-α-CEHCs varied differently with the interventions, as described below.

Cumulative urinary d₆- and d₃-α-CEHC excretion

The pattern of the cumulative urinary d₆-α-CEHC excretion ≤72 h after administration of the intravenous d₆-α-tocopherol dose was visibly different depending on the intervention (Figure 2A). The cumulative amounts of d₆-α-CEHC excreted over 72 h—whether estimated as micromoles per gram of creatinine excreted (or nanomoles excreted, data not shown)—were not significantly different between the 40% fat and 0% fat interventions (Table 1). However, the 0% fat-fast intervention, as compared with the 0% fat, resulted in a significant ∼50% decrease in the cumulative urinary d₆-α-CEHC excreted (Table 1). Thus, during fasting less d₆-α-CEHC arising from the intravenous d₆-α-tocopherol dose was excreted in urine.

FIGURE 2.

Cumulative urinary d₃- and d₆-α-CEHCs. The oral d₃-α-tocopherol dose was consumed, immediately thereafter the i.v. d₆-α-tocopherol dose was administered during each of the interventions (40% fat, 0% fat, 0% fat-fast); the fat was administered in a defined liquid meal (DLM), given as a breakfast at 10:00. Urine samples were collected at the indicated times and the excreted d₃- and d₆-α-CEHC amounts analyzed and corrected for the grams of creatinine excreted. The values of each sample were summed ≤72 h and are reported as mean + SEM cumulative sums at each time point. The interventions are indicated as squares = 40% fat (n = 10); diamonds = 0% fat (n = 10); and circles = 0% fat-fast (n = 7). (A) Cumulative d₆-α-CEHC excreted, closed symbols; (B) cumulative d₃-α-CEHC excreted, open symbols. Table 1 shows results of statistical analyses. α-CEHC, 3-(6-hydroxy-2,5,7,8-tetramethylchroman-2-yl)propanoic acid.

The interventions had no effect on the cumulative urinary d₃-α-CEHC excreted over 72 h from the orally administered d₃-α-tocopherol (cumulative μmol/g creatinine) (Figure 2B, Table 1).

By comparing the cumulative urinary d₆- and d₃-α-CEHC during each intervention, we observed that during the 40% fat intervention significantly more d₆- than d₃-α-CEHC was excreted, during the 0% fat intervention similar amounts of d₆- and d₃-α-CEHC were excreted, and during the 0% fat-fast intervention significantly less d₆- than d₃-α-CEHC was excreted.

The d₃-α-tocopherol amount absorbed was estimated based on our previous findings for each participant (18), then this value was used as the denominator in the calculation of the ratio of d₃-CEHC excreted:absorbed. To estimate the liver contribution, the d₆-α-tocopherol amount injected was used as the denominator for the ratio of d₆-α-CEHC excreted:injected dose, whereas the difference between the 2 ratios represents the contribution of the intestine to the urinary d₃-α-CEHC excreted (Table 2). During the 40% fat intervention, the contribution by the intestine was negligible. By contrast, during the 0% fat and 0% fat-fast interventions, the intestine contributed ∼0.5%–0.6% of the d₃-α-tocopherol absorbed to the urinary d₃-α-CEHC. The liver contribution during the 40% and 0% fat trials was ∼1% of the dose; however, fasting caused a significant decrease. Thus, during the 40% fat intervention the liver significantly contributed >90% of the urinary d₃-α-CEHC excretion, whereas during the 0% fat intervention the liver contribution of ∼70% remained significantly greater than that of the intestine, whereas during the 0% fat-fast the liver and intestine both contributed ~50%.

TABLE 2.

Estimated percentage of the absorbed d₃-α-tocopherol dose that was catabolized by liver or intestine during the 3 interventions and excreted as urinary d₃-α-CEHC¹

DLM fat	n	Intestine	Vs. 0% fat, P values2	Liver	Vs. 0% fat, P values2	Intestine vs. liver, P values3	Interaction, P values3
40% fat	10	0.11% ± 0.12%	0.0050	0.89% ± 0.10%	0.3200	<0.0001	0.0002
0% fat	10	0.54% ± 0.12%		1.03% ± 0.10%		0.0040
0% fat-fast	7	0.60% ± 0.14%	0.7000	0.43% ± 0.12%	0.0010	0.1714

¹The d₆- or d₃-α-CEHC (μmol) excreted by women consuming a DLM (either 40% fat, n = 10; 0% fat, n = 10; or 0% fat-fast, n = 7) was summed over 72 h. The d₃-α-tocopherol amount absorbed was estimated based on our previous findings for each participant (18), then this value was used as the denominator in the calculation of the ratio of d₃-CEHC excreted:d₃-α-tocopherol absorbed × 100, whereas the d₆-α-tocopherol amount injected was used as the denominator for the ratio d₆-CEHC excreted:injected × 100. The difference between the 2 ratios represents the contribution of the intestine to the urinary d₃-α-CEHC excreted as the percentage of dose absorbed; the contribution of the liver equals the ratio of d₆-CEHC excreted:injected d₆-α-tocopherol. Shown are the least-square mean ± SEM d₃-α-CEHC percentages of dose catabolized by either the intestine or liver. DLM, defined liquid meal; d₃- and d₆-α-tocopherol, RRR-α-tocopherol labeled with tri-deuterium and hexa-deuterium, respectively; 0% fat, defined liquid meal with 0% fat; 0% fat-fast, defined liquid meal with 0% fat with a 12-h fast; 40% fat, defined liquid meal with 40% fat; α-CEHC, 3-(6-hydroxy-2,5,7,8-tetramethylchroman-2-yl)propanoic acid.

²After verifying that the data were normally distributed, data were analyzed as an incomplete crossover design in SAS PROC MIXED with treatment (40% fat, 0% fat, and 0% fat-fast) as a fixed effect and subject as a random effect. Using the ESTIMATE statements, the effect of DLM fat was evaluated using the contrast between 40% fat and 0% fat interventions; the effect of fasting was evaluated using the contrast between 0% fat and 0% fat-fast interventions.

³For comparisons between the liver and intestine responses, the differences for each parameter were calculated, then the data analyzed as a crossover design in SAS PROC MIXED with treatment (40% fat, 0% fat, 0% fat-fast) as a fixed effect and subject as a random effect; the overall interaction was analyzed as a Latin square design with double repeated measures (Kronecker product) in SAS PROC MIXED. Fixed effects were interventions (40% fat, 0% fat, 0% fat-fast), the liver and intestine responses, and their interaction. The variance-covariance structure within subjects was modeled by using a Kronecker product, in which the intervention was modeled using a compound symmetry matrix and the type of treatment within treatment was modeled using an unstructured variance-covariance matrix.

Time course of urinary d₆-α-CEHC and d₃-α-CEHC excretion

The time courses of the urinary d₆-α-CEHC and d₃-α-CEHC excretion after administration of the intravenous d₆-α-tocopherol and the oral d₃-α-tocopherol doses were different depending on the intervention, with major differences visibly apparent in the first 24 h of each intervention (Figure 3A–C).

FIGURE 3.

Urinary and plasma d₃- and d₆-α-CEHC concentrations by time. Urine samples were obtained as described in the Figure 2 legend; plasma samples were obtained at the times indicated in Supplemental Figure 1. The mean ± SEM d₆-α-CEHC (closed symbol) and d₃-α-CEHC (open symbol) concentrations in (A–C) urine (μmol/g creatinine) and (D–F) plasma (nmol/L) at each time point are shown for each intervention, as indicated: (A, D) squares = 40% fat, n = 10; (B, E) diamonds = 0% fat, n = 10; and (C, F) circles = 0% fat-fast, n = 7. Tables 3 and 4 show results of statistical analyses. (G–I) Plasma α-T concentrations (μmol/L) shown here are the same as published previously (18) and use the same symbols for the interventions. α-CEHC, 3-(6-hydroxy-2,5,7,8-tetramethylchroman-2-yl)propanoic acid; α-T, α-tocopherol.

The AUCs (μmol/g creatinine × h) of the urinary d₆-α-CEHC excreted over 72 h during the 40% and 0% fat interventions were similar; however, during the 0% fat-fast intervention the urinary d₆-α-CEHC AUC was significantly less and was about half of the AUC during the 0% fat intervention (Table 3). The urinary d₃-α-CEHC AUCs were similar during all 3 interventions. During the 0% fat-fast intervention, significantly less d₆- than d₃-α-CEHC was excreted.

TABLE 3.

Pharmacokinetic parameters derived from the time course of the urinary d₃- and d₆-α-CEHCs excreted that resulted from catabolism after oral d₃- and i.v. d₆-α-tocopherol administration during 3 interventions¹

Parameter	DLM fat	d6-α-CEHC	Vs. 0% fat, P values2	d3-α-CEHC	Vs. 0% fat, P values2	d3- vs. d6-α-CEHC, P values3	Interaction, P values3
AUC,4 μmol/g creatinine × h	40% fat	13.62 ± 2.08	0.7579	10.31 ± 1.94	0.2404	0.0330	0.0264
	0% fat	13.12 ± 2.10		12.50 ± 1.96		0.0830
	0% fat-fast	6.30 ± 2.24	0.0013	9.28 ± 2.16	0.1016	0.0786
Cmax, μmol/g creatinine	40% fat	0.433 ± 0.065	0.8585	0.344 ± 0.082	0.4151	0.0721	0.0074
	0% fat	0.423 ± 0.065		0.417 ± 0.083		0.9878
	0% fat-fast	0.170 ± 0.071	0.0007	0.415 ± 0.094	0.9877	0.0338
Tmax, h	40% fat	14.9 ± 1.0	0.3472	13.2 ± 1.0	0.0182	0.6193	0.0062
	0% fat	16.2 ± 1.0		16.8 ± 1.0		0.1997
	0% fat-fast	24.6 ± 1.2	<0.0001	13.1 ± 1.2	0.0281	0.0003

¹Parameters (least-square means ± SEMs) were calculated from CEHC analyses of urine obtained from women consuming a DLM (either 40% fat, n = 10; 0% fat, n = 10; or 0% fat-fast, n = 7), as shown in Figure 3A (40% fat), B (0% fat), and C (0% fat-fast). Cmax, maximum concentration; DLM, defined liquid meal; d₃- and d₆-α-tocopherol, RRR-α-tocopherol labeled with tri-deuterium and hexa-deuterium, respectively; Tmax, time of maximum concentration; 0% fat, defined liquid meal with 0% fat; 0% fat-fast, defined liquid meal with 0% fat with a 12-h fast; 40% fat, defined liquid meal with 40% fat; α-CEHC, 3-(6-hydroxy-2,5,7,8-tetramethylchroman-2-yl)propanoic acid.

²Parameters derived from either urinary d₆-α-CEHC or from d₃-α-CEHC, respectively, were analyzed as an incomplete crossover design in SAS PROC MIXED with treatment (40% fat, 0% fat, and 0% fat-fast) as a fixed effect and subject as a random effect. Using the ESTIMATE statements, the effect of DLM fat was evaluated using the contrast between 40% fat and 0% fat interventions, whereas the effect of fasting was evaluated using the contrast between 0% fat and 0% fat-fast interventions.

³For comparisons between the i.v. d₆-α-tocopherol and oral d₃-α-tocopherol parameters, the differences between d₆-α-CEHC and d₃-α-CEHC for each parameter were calculated, then the data analyzed as a crossover design in SAS PROC MIXED with treatment (40% fat, 0% fat, 0% fat-fast) as a fixed effect and subject as a random effect; the overall interaction was analyzed as a Latin square design with double repeated measures (Kronecker product) in SAS PROC MIXED. Fixed effects were interventions (40% fat, 0% fat, 0% fat-fast), the route of α-tocopherol administration (oral, i.v.), and their interaction. The variance-covariance structure within subjects was modeled by using a Kronecker product, in which the intervention was modeled using a compound symmetry matrix and the type of treatment within treatment was modeled using an unstructured variance-covariance matrix.

⁴AUC is calculated from the urine concentrations from 0 to 72 h.

The Cmax urinary d₆- or d₃-α-CEHC excreted during the 40% and 0% fat interventions were similar (Table 3). During the 0% fat-fast intervention the urinary d₆-α-CEHC Cmax was decreased by ∼50% (compared with the 0% fat intervention); fasting did not affect the d₃-α-CEHC Cmax. Thus, the comparison between d₆- and d₃-α-CEHC shows that during fasting significantly less d₆-α-CEHC than d₃-α-CEHC was excreted.

The Tmax urinary d₆-α-CEHC excretion was similar in the 40% fat and 0% fat interventions, whereas the d₆-α-CEHC Tmax in the 0% fat-fast intervention was ∼10 h later (Table 3). Compared with the Tmax of the urinary d₃-α-CEHC excretion in the 0% fat intervention, the Tmax in the 40% fat and the 0% fat-fast interventions were significantly earlier by ∼3 h (Table 3). Thus, fasting significantly delayed the appearance of d₆-α-CEHC, but had no effect on d₃-α-CEHC excretion.

Urinary unlabeled α- and γ-CEHC excretion

The daily mean unlabeled-α-CEHC excreted in urine was unaffected by the interventions (40% fat, 1.2 ± 0.3 μmol/g creatinine, n = 10; 0% fat, 0.9 ± 0.3 μmol/g creatinine, n = 10; 0% fat-fast, 0.5 ± 0.3 μmol/g creatinine, n = 7). The daily unlabeled γ-CEHC excreted was about one-third less (P = 0.012) during the 0% fat-fast intervention than during the 0% fat, whereas the 40% fat and 0% fat interventions had no effect on unlabeled γ-CEHC excretion (40% fat, 1.2 ± 0.2 μmol/g creatinine, n = 10; 0% fat, 1.4 ± 0.2 μmol/g creatinine, n = 10; 0% fat-fast, 0.8 ± 0.2 μmol/g creatinine, n = 7).

Plasma d₆-α- and d₃-α-CEHC concentrations during each intervention

The mechanism of CEHC export from the liver is unknown, but CEHCs are found in plasma and excreted in urine. Therefore, it was not unexpected that the patterns of the plasma d₆-α- and d₃-α-CEHC concentrations (Figure 3D–F) were similar to the time course of the urinary d₆-α- and d₃-α-CEHC excreted (Figure 3A–C). Remarkably, plasma d₆-α- and d₃-α-CEHC concentrations, especially during the 0% fat-fast intervention, did not follow the plasma d₆-α- and d₃-α-tocopherol concentrations (Figure 3G–I) published previously (18).

The 3 interventions—40% fat, 0% fat, and 0% fat-fast—had no significant effects on the plasma d₆-α-CEHC AUC or on the d₃-α-CEHC AUC (Table 4). During the 40% fat intervention, the plasma d₆-α-CEHC AUC was significantly greater than the d₃-α-CEHC AUC.

The d₆-α-CEHC Cmax were significantly greater in the 40% fat intervention than those in the 0% fat intervention, but the d₆-α-CEHC Cmax were similar during the 0% and 0% fat-fast interventions. The d₃-α-CEHC Cmax were similar during all interventions. Notably, during the 40% fat intervention the d₆-α-CEHC Cmax were significantly greater than the d₃-α-CEHC Cmax, they were similar during the 0% fat trial, but during the 0% fat-fast, the Cmax for d₆-α-CEHC were significantly less than those of d₃-α-CEHC (Table 4).

The plasma d₆- and d₃-α-CEHC Tmax were similar in the 40% fat and 0% fat interventions, whereas during the 0% fat-fast intervention the d₆-α-CEHC Tmax was delayed by ∼5 h and the d₃-α-CEHC Tmax appeared ∼3 h earlier (Table 4). Thus, during the 0% fat-fast intervention the peak plasma d₆-α-CEHC was delayed significantly by ∼8 h relative to the d₃-α-CEHC Tmax (Table 4).

Fecal d₆- and d₃-α-catabolites and d₆- and d₃-α-tocopherols excreted during each intervention

The expectation based on rat studies (27, 29) was that excess α-tocopherol would be excreted from the liver via the bile and into the feces. Our previous report (18) noted that relatively little of the intravenous d₆-α-tocopherol was excreted in feces. We therefore evaluated whether any of the catabolites found between 13′ COOH-α-tocopherol and α-CEHC were present in the feces of the study participants in the 40% and 0% fat interventions (Table 5). Insufficient fecal material was collected during the 0% fat-fast intervention for analysis.

During the 40% fat intervention, significantly greater amounts of total fecal d₆-α-tocopherol catabolites than of d₃-α-tocopherol catabolites were excreted (Table 5). During the 0% fat intervention, no differences were observed. Notably, in both interventions, fecal d₆-α-tocopherol, fecal d₆-α-tocopherol catabolites, and urinary d₆-α-CEHC represented in total only ∼5%–7% of the intravenously administered dose. Even less of the orally administered dose was excreted as fecal d₃-α-tocopherol catabolites and urinary d₃-α-CEHC. It should be noted that the fecal d₃-α-tocopherol also included nonabsorbed d₃-α-tocopherol. Thus, if calculated per absorbed dose, then the fecal d₃-α-tocopherol catabolites would approximately equal the d₆-α-tocopherol catabolites during the 40% intervention.

Discussion

To our knowledge, this is the first time that vitamin E catabolism has been studied in people using an intravenous, deuterium-labeled α-tocopherol that was administered as an oil:water emulsion to mimic chylomicrons. Based on the observed rapid d₆-α-tocopherol clearance observed during the first 20 min after intravenous injection (18), the d₆-α-tocopherol and d₆-α-CEHC reflect liver α-tocopherol trafficking and catabolism, respectively, whereas the d₃-α-tocopherol and d₃-α-CEHC reflect both intestinal and liver modulation of these respective actions. For convenience, it is assumed that the extrahepatic tissues have similar negligible effects on catabolism, an assumption which may be incorrect if the kidney is a significant contributor (30). Nonetheless, the intravenous d₆-α-tocopherol emulsion allows evaluation of catabolism of vitamin E delivered to the liver without impact of the intestine, whereas the oral dose measures catabolism in both the intestine and the liver. We observed that during the 40% fat intervention >90% of the catabolism to urinary α-CEHC took place in the liver, whereas during fasting the intestine and liver played equal roles in α-tocopherol catabolism. Catabolism by the intestine is likely because it expresses CYP4F2 (9, 31–33), the enzyme that initiates vitamin E catabolism (6, 11).

The 3 interventions had no differential effects ≤72 h on the plasma d₆-α-tocopherol concentrations arising from the intravenous dose (Figure 3G–I), as we reported (18). Because the d₆-α-tocopherol in the emulsion was cleared from the plasma with an estimated half-life of ∼3–4 min, virtually 100% of a relatively large bolus (30 mg or 68.8 μmol d₆-α-tocopherol) was delivered to the liver in <1 h (18). Thus, it was anticipated that this large α-tocopherol amount might stimulate liver α-tocopherol catabolism, but the interventions that were expected to alter intestinal α-tocopherol catabolism instead affected the liver. Specifically, the 0% fat-fast intervention caused an ∼50% decrease in the cumulative urinary d₆-α-CEHC excretion, whereas the switch from 40% to 0% fat had no effect (Figure 2A, Table 1). The decreased d₆-α-tocopherol catabolism during the 12-h fast suggests that increased α-tocopherol secretion from the liver into plasma occurred as a result of the reduced delivery of meal fat and α-tocopherol to the liver. However, increased d₆-α-tocopherol secretion—evaluated as increased plasma d₆-α-tocopherol—was not observed in paired comparisons of the 0% and 0% fat-fast interventions in the 7 women who completed both interventions (data not shown). However, plasma α-tocopherol concentrations are reported in micromoles per liter, whereas α-CEHC are present at nanomoles per liter; thus, small changes in α-tocopherol catabolism may not be detected by evaluating plasma α-tocopherol concentrations. An alternative explanation is that α-tocopherol catabolism was downregulated by fasting. Given that 1) CYP4F2 initiates vitamin E catabolism, 2) the CYP4F2 gene is regulated by sterol regulatory element-binding protein (SREBP) (34, 35), and 3) fasting decreases SREBP-1c (36), it is hypothetically possible that hepatic α-tocopherol catabolism was decreased by fasting. This suggestion is supported by the observations during the 0% fat-fast intervention that the unlabeled α-CEHC excreted was on average nearly half of that during the 0% fat intervention (although the differences did not reach statistical significance) and the unlabeled γ-CEHC excreted was significantly decreased by one-third. However, the increased role of the intestine during the 0% fat and 0% fat-fast interventions suggests that the lack of fat during the oral d₃-α-tocopherol administration increased intestinal catabolism, whereas fasting decreased liver α-tocopherol catabolism (Table 2).

Plasma d₃-α-CEHC concentrations (Figure 3D–F) during the first 12 h after dosing also showed that the interventions had a major impact on the orally administered d₃-α-tocopherol catabolism to d₃-α-CEHC. During the 0% fat-fast intervention, both the plasma (Figure 3F, Table 4) and urine (Figure 3F, Table 3) d₃-α-CEHC reached a maximum at ∼10–13 h, which is before the plasma d₃-α-tocopherol peak at ∼20 h (Figure 3I) (18); these findings support the conclusion that the intestine catabolizes α-tocopherol, while it is retained in the intestine during fasting. The comparisons between the 0% fat and 0% fat-fast interventions show that eating a meal 4 h after dosing prevented the increases in plasma d₃-α-CEHC observed during fasting (Figure 3E, F). The plasma d₃-α-CEHC concentrations (Figure 3E, Table 4, Tmax) suggest that initially d₃-α-tocopherol catabolism occurs in the intestine, when there is insufficient fat during the 0% fat intervention to facilitate chylomicron formation. Four hours later after the meal, d₃-α-tocopherol absorption from the intestine is observed (18). Thus, the data consistently suggest that in the absence of adequate fat in the intestine for vitamin E absorption, there is increased intestinal α-tocopherol catabolism, while simultaneously in the liver there is decreased α-tocopherol catabolism. Importantly, the fecal long-chain (11′-COOH- and 13′-COOH-α-tocopherol) catabolites derived from both the oral and the intravenous doses suggest that both the liver and the intestine are sources of these catabolites, but further studies are needed to define the role of the intestine in vitamin E catabolism.

The consumption of various different kinds and amounts of dietary fats may have a large impact on vitamin E catabolism because CYP4F2 is not solely involved in vitamin E catabolism, but plays an important role in fatty acid oxidation (37). Among CYP4F2 functions are 20-hydroxyeicosatetraenoic acid production (38, 39), and catabolism of leukotrienes (40), long-chain (16- to 26-carbon) SFAs, unsaturated fatty acids, and branched-chain fatty acids (37). CYP4F2 is also a vitamin K oxidase that initiates vitamin K catabolism (41, 42). In addition, CYP4F2 protein concentrations in human liver vary widely (8) and its regulation is not well understood. Although CEHC is a product of xenobiotic metabolism (43), the potential functions of these vitamin E catabolites have been actively studied because they retain their antioxidant activity and are more water-soluble than the parent compounds (44). Health benefits (45) and vitamin E's anti-inflammatory activities have been attributed to the activity of its catabolites (46), especially the 13′-COOH-α-tocopherol catabolite (47) because it inhibits 5-lipoxygenase—a key enzyme in the biosynthesis of leukotrienes from arachidonic acid (20:4n–6) (48). Recently, a metabolomics analysis of urine from persons with bladder cancer compared with urine from healthy controls showed that α-CEHC was increased in persons with cancer (49). The authors suggested that vitamin E catabolism may be upregulated by inflammation secondary to tumors (49).

This study has limitations including a small sample size, lack of randomization or blinding, and compliance issues leading to imbalance with attendant potential for baseline and residual confounding. The lack of randomization to sequence was based on the underlying assumption that the sequence effect was negligible. This assumption was justified because the labeled α-CEHC concentrations in the first urine sample (4 h collection) were less than one-twentieth the area counts of the next sample. Thus, baseline samples were very low, consistent with background isotope amounts, irrespective of prior dose. In addition, the half-life previously calculated from the plasma α-tocopherol concentrations was 30–40 h; thus, the minimum of 2 mo between repeat studies in participants was very long relative to the half-life. Another limitation is the lack of fecal sample collection during the fasting intervention. Finally, CYP4F2 activity or gene polymorphisms in liver or intestinal biopsy specimens were not measured, nor was biliary catabolite excretion directly measured.

In summary, this study shows that both liver and intestine have roles in vitamin E catabolism (Figure 4). The impact of fasting on vitamin E catabolism has not previously been appreciated and thus this aspect of previous studies evaluating its catabolism should be taken into account. Finally, and perhaps most remarkably in this trial, the α-tocopherol dose reaching the liver within a 24-h period was ∼45 mg (30 mg injected, ∼15 mg absorbed). This dose was 3 times the RDA (23), yet over 72 h only ∼2% was excreted in urine and ∼5% in feces (Table 5). Further, both the plasma and urine α-CEHC concentrations suggested that the major portion of the catabolism occurred in the first 24 h, with the apparent peak in α-CEHC from the labeled doses near 12–18 h. These data suggest that the delivery of α-tocopherol in triglyceride-rich lipoproteins to the liver accounts for the major pathway that delivers vitamin E to a catabolic pathway. Labeled α-CEHC appears to leave the plasma and be excreted in urine in a linear manner, whereas the unlabeled α-CEHC was excreted at a relatively constant daily rate of 1.2 ± 0.3 μmol/g creatinine during the 40% fat intervention, a concentration similar to our previous report (14). The interesting role of vitamin E catabolites as anti-inflammatory agents suggests that they may have unappreciated functions in human health (45, 46, 48).

FIGURE 4.

Disposition of α-T and its catabolites. During consumption of a meal containing 40% fat, a portion of the dietary α-T is absorbed, whereas the remainder is excreted in feces. The absorbed α-T travels in chylomicrons in the lymph through the circulation to the liver, where it is preferentially trafficked back into the plasma for α-T delivery to tissues. Excess α-T can be catabolized in the liver to long-chain carboxy-catabolites (see Figure 1 for structures); both α-T and long chain-α-T-catabolites were found excreted in feces (see Table 5). Urinary α-CEHC excreted during the 40% fat intervention was found to be derived largely from nonintestinal (e.g., liver and possibly kidney) sources (see Table 2). The 0% fat intervention increased the proportion of α-T catabolized in the intestine and excreted as urinary α-CEHC. The 0% fat-fast intervention, similarly to the 0% fat intervention, increased the proportion of α-T catabolized in the intestine, but decreased the proportion catabolized in the liver and then excreted as urinary α-CEHC (see Table 2). The size of the circles gives a general impression of the amounts of α-T (solid circle), long chain-α-T-catabolites (circle with horizontal lines), and α-CEHC (circle with diagonal lines). α-CEHC, 3-(6-hydroxy-2,5,7,8-tetramethylchroman-2-yl)propanoic acid; α-T, α-tocopherol.

Notes

Supported by NIH National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) grant DK081761 and Office of Dietary Supplements, ODIR, NIH (to MGT); Intramural Research Programs, NIDDK, NIH; Metabolic Unit Staff, NIDDK Clinical Core Staff, and Clinical Center Nutrition and Nursing Department Staff, NIH; NIDDK grant DK053213-11 (to ML). The intravenous deuterated-α-tocopherol and the intravenous emulsion were provided as a gift by Manfred Eggersdorfer, DSM Nutritional Products AG, Kaiseraugst, Switzerland. DSM also provided a gift in support of the purchase of a mass spectrometer for the Traber lab. The funders had no input into study outcomes.

Supplemental Table 1 and Supplemental Figures 1 and 2 are available from the “Supplementary data” link in the online posting of the article and from the same link in the online table of contents at https://academic.oup.com/ajcn/.

Data Availability: Data described herein are made publicly and freely available without restriction as an online supplement to this work.

Abbreviations used: CEHC, carboxyethyl hydroxychromanol; Cmax, maximum concentration; α-CMBHC, 5-(6-hydroxy-2,5,7,8-tetramethyl-chroman-2-yl)-2-methyl-pentanoic acid; COOH, carboxy; CYP4F2, cytochrome P450 4F2; DLM, defined liquid meal; ESI, electrospray ionization; IRB, Institutional Review Board; IS, internal standard; NIH CRC, NIH Clinical Research Center; OH, hydroxy; SAS, Statistical Analysis System; SREBP, sterol regulatory element-binding protein; Tmax, time of maximum concentration; 0% fat, defined liquid meal with 0% fat; 0% fat-fast, defined liquid meal with 0% fat with a 12-h fast; 40% fat, defined liquid meal with 40% fat.