Postprandial vitamin E plasma enrichment and second meal effect in women with and without metabolic dysfunction-associated steatotic liver disease

Abstract

Background

The effect of meal fat sequence and excess body fat on post-prandial changes in vitamin E absorption is unclear.

Objectives

We investigated post-prandial changes in plasma vitamin E enrichment in response to (1) varied dietary fat sequences in healthy women and (2) excess body and liver fat in women with obesity and metabolic dysfunction-associated steatotic liver disease (MASLD).

Methods

Vitamin E (α-tocopherol, α-T) pharmacokinetics were assessed using deuterium (d)-labeled α-T administered orally (d₃-α-T) and intravenously (d₆-α-T). 13 healthy women received oral and IV α-T (30mg) with a liquid breakfast containing 40% or 0% fat, followed by 30% fat lunch and dinner at 4h and 8h, or a 12h fast (0% fat/fast). 6 women with MASLD and 10 normal weight controls received oral and IV α-T (2 mg) with similar meal sequence. Major outcomes were post-prandial plasma α-T %enrichment, Cmax, and AUCs at 4, 12, and 24h. Exploratory outcomes included the second meal effect (SME), defined as mean change in pre- vs post-2nd meal d₃-α-T %enrichment at 4h and 12h respectively.

Results

At 4h oral d₃-α-T %enrichment Cmax and AUC0–4h were not statistically different between 40% and 0%-fat groups. Following the 2nd meal at 4h, SME was 55% higher in the 0% group vs 40% group (15.8±1.1% vs 10.2±1.1%, p<0.009), resulting in greater Cmax and AUC in 0% group at 12h and 24h (p<0.04). Compared to the 0% group, the 0% fat/fast group showed reduced oral d₃-α-T %enrichment, while no differences were seen with IV d₆-α-T %enrichments. Compared with controls, the obese/MASLD cohort had 31% lower SME (0.87±0.1% vs 1.26±0.1%,p=0.047), resulting in reduced Cmax and AUC (p<0.04) at 24h.

Conclusion

The variation in the meal fat sequence (0%-to-30% vs 40%-to-30% fat) significantly enhanced SME and post-prandial d₃-α-T %enrichment, while obesity/MASLD attenuated SME and post-prandial vitamin E %enrichment.

INTRODUCTION

Vitamin E (α-tocopherol, α-T) is a fat-soluble micronutrient and its impaired absorption may result in inadequacy, even deficiency¹. Following ingestion and digestion by biliary and pancreatic enzymes, dietary fat containing vitamin E is incorporated into micelles and taken up into enterocytes by facilitated transporters^2–4. On entry into enterocytes, vitamin E and other fat-soluble dietary factors are stored temporarily in cytoplasmic lipid droplets with subsequent incorporation into chylomicrons and transfer into plasma⁴. Importantly, the mechanisms that modulate the timing and movement of vitamin E from the enterocytes to lymph, then plasma remains poorly characterized.

The roles of the meal fat and fasting on vitamin E absorption were highlighted in our previous report of healthy women who received oral and intravenous deuterium-labeled (d_n-) α-T consumed with a liquid meal at breakfast containing 0% or 40% fat, followed by controlled meals or a 12h fast⁵. While these data provided useful insights on how dietary fat influences vitamin E absorption over 72 h, the characterization of the immediate and sequential post-prandial changes, including “second meal effect” (SME) were insufficient. The SME is an important phenomenon where ingestion of a fat-containing meal several hours post-dosing induces a surge in production of intestinal triglyceride-rich lipoproteins, chylomicrons^6–11. Additionally, given that obesity is associated with reduced vitamin E bioavailability^12,13, it is unclear how the subsequent meals alter post-prandial changes in the movement of vitamin E from the enterocytes to plasma. Characterizing these post-prandial changes in healthy and chronic disease cohorts is essential for understanding different regulatory roles of the gut and liver, but also to inform dietary strategies aimed at mitigating nutrient deficiency in at-risk groups, such as individuals with fat malabsorption disorders.

The objectives of this study were to investigate effects of (1) varied and sequential dietary fat interventions, (2) fasting intervention, and (3) obesity/MASLD status on SME and post-prandial α-T enrichment in plasma. Plasma α-T % enrichments were evaluated at specific time points following meal consumption. We hypothesized that varied meal fat sequences and obesity/MASLD status would affect the kinetics and extent of vitamin E absorption. If correct, then optimizing the time of vitamin E ingestion and meal consumption could mitigate inadequate bioavailability in at-risk populations.

METHODS

Study design

This study is a secondary analysis of previously published data^5,13, focusing specifically on outcomes within the first 24 hours following intervention. This clinical trial (NCT00862433) was conducted at the NIH Clinical Research Center (CRC) and approved by the Institutional Review Board (IRB) of the National Institute of Diabetes and Digestive Kidney Diseases (NIDDK) and the National Institute of Arthritis and Musculoskeletal and Skin Diseases (NIAMS) under protocol 09-DK-0097, NCT00862433 (https://clinicaltrials.gov/study/NCT00862433).

Participants

Two comparator groups of women were included in the study analysis (Supplementary Figure 1). Group A comparators included healthy normal weight adult women found to have normal physical exams, no chronic illness and liver fat <2% based on magnetic resonance spectroscopy (MRS) as previously reported^5,14. Group B comparators included healthy women as well as women with obesity and MASLD, liver fat ≥ 7%, as previously reported¹³. Exclusion criteria included tobacco and recreational drug use, alcohol abuse (>20g/d of alcohol), coagulopathy; and unwillingness to use effective contraceptive methods for the duration of the study. All women in all groups had normal liver function, normal platelet counts, and FIB-4 score less than 0.92. At time of enrollment, the cohort with MASLD were diagnosed with hepatosteatosis (HS)¹³, now referred to as MASLD based on the enrollment criteria: excess liver fat (>5% hepatic steatosis), at least one risk factor for cardiometabolic dysfunction (all participants had obesity with mean BMI of 43 kg/m², Supplementary Table 1), and no other causes of hepatic steatosis¹⁵.

Interventions

On day 0, enrolled participants were admitted to the NIH Clinical Center. On day 1, participants received a custom-made defined liquid meal (DLM) at breakfast containing 40% fat or 0% fat as previously described^5,13,14. When 2/3 of the liquid meal was consumed, oral d₃-α-T vitamin E at doses of either 30mg (group A) or 2mg (group B) was placed on the participant’s tongue, then the remainder of the liquid meal was consumed within 5 minutes (Figure 1). At that time, the intravenous d₆-α-tocopherol was administered at a dose of either 30mg (Group A) or 2mg (Group B). The end of the IV vitamin E infusion defined the beginning of the sampling study (time 0). Subsequently, group A participants received different meal interventions (Figure 1). Participants included in cohort A were part of a crossover study. There were 13 participants, but, because not all participants had data for all 3 of the interventions, there were 10 participants in the 40% fat group, 10 in the 0% fat group, and 7 in the 0% fat/fast group. In cohort B, there were 10 participants in the normal weight group and 6 in the MASLD group (Supplementary Figure 1).

Figure 1: Study design and interventions for group A comparators (1A) and group B comparators (1B).

(A) Group A study interventions (Healthy N=13) and (B) Group B study interventions (Healthy N=10, Obese/MASLD N=6).

Abbreviations: MASLD, Metabolic dysfunction-associated steatotic liver disease; IV, intravenous.

Analytical procedures

We used α-T %enrichment in plasma given this paper’s focus on the post-prandial changes 24h post-dosing. Data for plasma vitamin E values are included in Supplementary files (Supplementary Tables 2 and 3). Percent α-T enrichment: The d₃-, d₆-, d₉- (used as an internal standard) and unlabeled (d₀-α-Ts) were analyzed using high-performance liquid chromatography/mass spectrometry (LC/MS) in plasma and lipoproteins as described previously^5,13,14. Data were calculated as enrichment percentages of either d₃-α-T or d₆-α-T using the formula described below, where Total α-T is the sum of the plasma concentrations of d₀-, d₃-, d₆- α-Ts⁵:

$$d_3\text{-}\alpha \text{-T}=\frac{d_3\text{-}\alpha \text{-}T\times 100}{\mathit{\text{Total-}}\alpha \text{-T}}$$

$$d_6\text{-}\alpha \text{-T}=\frac{d_6\text{-}\alpha \text{-}T\times 100}{\mathit{\text{Total-}}\alpha \text{-T}}$$

Outcomes

Primary outcomes included differences in post-prandial α-tocopherol %enrichment curves and kinetics parameters: area under the curve (AUC) and maximal concentration (Cmax), for 0–4 hours, 0–12 hours, and 0–24 hours. Timepoints were selected based on varied dietary fat interventions (4h) and fasting (12h) post-dosing. Exploratory outcome included differences in second meal effects (SME) among comparator groups.

Statistical Methods

Area under the curve (AUC) and maximal concentration (Cmax) were calculated for each patient for 0–4 hours, 0–12 hours, and 0–24 hours using trapezoidal rule. Values obtained in the first 5 minutes were removed from these calculations for d₆-α-T %enrichment. Second meal effect (SME) was calculated for each participant by subtracting the pre-2^nd meal Cmax (0–4h) from the post-2^nd meal Cmax (0–12h).

Statistical tests were 2-sided, and p values <0.05 were considered to be statistically significant. SAS PROC MIXED was used to compare the curves in Figures 2 to 4, with participants as a repeated measure, maximum likelihood estimation, random effects for intercept and participants, and compound symmetry. AUC and Cmax were compared using SAS PROC MIXED, with a random repeated measures participant effect for cohort A (some participants in cohort A were in multiple groups), but not for cohort B. These analyses accounted for the cross-over design by including the participant ID variables.

Figure 2: Effect of varied sequential meal fat on post-prandial vitamin E plasma %enrichment and second meal effect (SME) in group A comparators of healthy women: 40%fat (N=10) vs 0% fat (N=10).

(A-B) Plasma labeled α-T % enrichment for the first 24h following administration of oral d₃-α-T (A) with either 40% fat breakfast (blue circle) or 0% fat breakfast (red circle). (B) Plasma α-T % enrichment following administration of IV d₆-α-T with either 40% fat breakfast (blue circle) or 0% fat breakfast (red circle). C) Difference in AUC means for oral d₃-α-T (open circle) and intravenous d₆-α-T (black circle), between the 40% fat and 0% fat group at 4h, 12h and 24h post-dosing. D) Second meal effect (SME) based on change in pre-2^nd meal d₃-α-T %enrichment Cmax at 4h vs post-2^nd meal Cmax at 12h in the 40% fat group (blue circle) and 0% fat group (reed circle). E) Graphical representation of the change in SME characterized by the difference between Cmax at 0–12h and Cmax (Table 1) in the 40% fat (blue circle) and 0% fat (red circle) group. n=10 per group, with 7 participated in both 40% and 0% interventions (see supplementary figure 1). Statistical significance is shown for each time interval, with p-values representing the interaction between time and dietary fat using repeated measures regression analysis.

Figure 4: Effect of health status on post-prandial vitamin E plasma %enrichment: Healthy women (N=10) vs women with MASLD (N=6).

(A-B) Plasma labeled α-T % enrichment for the first 24h following administration of oral d₃-α-T (A) and IV d₆-α-T (B) with identical dietary fat contents in healthy normal weight controls (dark blue circle) and cohort with metabolic dysfunction-associated steatotic liver disease (MASLD) (orange circle). C) Difference in AUC means for oral d₃-α-T (white circle) and intravenous d₆-α-T (black circle), between the healthy controls and cohort with MASLD at 4h, 12h and 24h post-dosing. D) Second meal effect based on change in pre-2^nd meal d₃-α-T %enrichment Cmax at 4h vs post-2^nd meal Cmax at 12h in the healthy cohort (dark blue) and cohort with MASLD (orange). E) Graphical representation of the change in SME characterized by the difference between Cmax at 0–12h and Cmax (Table 1) in the MASLD (orange circle) and healthy cohort (dark blue circle). N=10 for the healthy controls and n=6 for the cohort with MASLD. Statistical significance is shown for each time interval, with p-values representing the interaction between time and dietary fat using repeated measures regression analysis

RESULTS

Effect of varied sequential meal fat on post-prandial vitamin E plasma enrichment in normal weight healthy women.

To assess the effect of fat on differences in plasma d₃-α-T %enrichment, we compared outcomes in group A participants that received either 40% or 0% fat with breakfast at time 0 h, followed by identical 30% meal fats at lunch and dinner (Figures 1A and 2A). At 4 h, differences between the 40% and 0% fat groups with regards to d₃-α-T %enrichment Cmax (3.8±0.7% vs 1.9±0.7%, p=0.076) or AUC_0–4h (6.7±1.4 vs 2.3±1.3, p=0.062) was not statistically significant (Table 1). Following lunch, with 30% fat, there was an exponential increase in d₃-α-T %enrichment in both groups, consistent with a second meal effect (SME, Figure 2A). Following this SME, the Cmax became comparatively higher in the 0% fat vs 40% fat group at 12h (17.5±1.0% vs 13.9±1.0%, p=0.003) and 24h (18.0±0.9% vs 14.8±0.9%, p=0.008, Table 1). Similarly, the AUC became comparatively higher in the 0% group at 12h (106.4±7.7 vs 94.7±7.7, p=0.036) and 24h (302.1±14.8 vs 249.5±14.8, p<0.001), Table 1. Consistent with these dynamic changes in post-prandial d₃-α-T %enrichment, repeated measures regression showed significant differences between 40% vs 0% d₃-α-T %enrichment at 4, 12 and 24 h (p<0.001 for each comparison Figure 2A).

Table 1:

C_max and AUC means (SEM) of d₃-α-T %enrichment in group A comparators: 40% fat versus 0% fat

	AUC (%.h)				Cmax (%)
	40% fat (n=10)	0% fat (n=10)	0% – 40%	pvalue	40% fat (n=10)	0% fat (n=10)	0% – 40%	pvalue
0–4 hours	6.7 (1.4)	2.3 (1.3)	−4.4 (1.9)	0.062	3.8 (0.7)	1.9 (0.7)	−1.91 (0.89)	0.076
0–12 hours	94.7 (7.7)	106.4 (7.7)	11.7 (4.3)	0.036	13.9 (1.0)	17.5 (1.0)	3.67 (0.77)	0.003
0–24 hours	249.5 (14.8)	302.1 (14.8)	52.6 (8.3)	<0.001	14.8 (0.9)	18.0 (0.9)	3.24 (0.83)	0.008
SME* (Difference (0–12 – 0–4))					10.2 (1.1)	15.8 (1.1)	5.51 (1.44)	0.009

Table 1 presents data from Figure 2A. Participants received 30mg of Oral and IV α-T with breakfast (0h) containing either 40% or 0% fat, followed by 30% fat lunch (4h) and dinner (8h) at 4h. Comparison between 40% fat and 0% fat interventions of d₃-α-T %enrichment Cmax (%) and AUC (%.h). Cmax and AUC are calculated separately for 0–4, 0–12, and 0–24 hours and the Cmax difference between 0–4 hours and 0–12 hours are calculated for each participant. Means, SEM, and p values are from a repeated measures mixed-effects linear regression model. Thirteen participants are included, 7 have data for both of the comparison groups, 6 have data for only one.

We also assessed differences in d₆-α-T %enrichment between the 40% and 0% comparators for the IV administered d₆-α-T, which bypasses intestinal absorption and more specifically evaluates the role of the liver⁵. The d₆-α-T %enrichment Cmax and AUCs were not significantly different between groups at any of the time points assessed (Table 2 and Figure 2B). To compare the divergent patterns of the oral d₃-α-T vs IV d₆-α-T, we plotted the mean AUC differences in 40% fat group vs 0% fat group (Figure 2C and Tables 1 and 2). In contrast to the dynamic changes in the enrichment following oral d₃-α-T %enrichment, a relatively flat line was observed with d₆-α-T % enrichment following IV administration in either fat group (Figure 2C).

Table 2:

C_max and AUC means (SEM) of d₆-α-T %enrichment in group A comparators: 40% fat versus 0% fat

	AUC (%.h)				Cmax (%)
	40% fat (n=10)	0% fat (n=10)	0% – 40%	pvalue	40% fat (n=10)	0% fat (n=10)	0% – 40%	pvalue
0–4 hours	58.4 (4.1)	64.4 (4.1)	6.0 (5.7)	0.33	21.8 (1.2)	23.5 (1.2)	1.66 (1.42)	0.29
0–12 hours	237.7 (11.2)	248.0 (11.2)	10.3 (13.2)	0.46	23.4 (1.2)	24.9 (1.2)	1.59 (1.11)	0.20
0–24 hours	483.9 (19.9)	495.0 (19.9)	11.2 (22.3)	0.64	23.4 (1.2)	24.9 (1.2)	1.59 (1.11)	0.20
SME* (Difference (0–12 – 0–4))					2.1 (0.5)	1.4 (0.5)	−0.69 (0.88)	0.47

Table 2 presents data from Figure 2B. Participants received 30mg of Oral and IV α-T with breakfast (0h) containing either 40% or 0% fat, followed by 30% fat lunch (4h) and dinner (8h). Comparison between 40% fat and 0% fat interventions of d₆-α-T %enrichment Cmax (%) and AUC (%.h). Cmax and AUC are calculated separately for 0–4, 0–12, and 0–24 hours and Cmax difference between 0–4 hours and 0–12 hours are calculated for each participant. Means, SEM, and p values are from a repeated measures mixed-effects linear regression model. Thirteen participants are included, 7 have data for both of the comparison groups, 6 have data for only one.

We quantified the SME by comparing the mean change in pre-2^nd meal (4h) vs post-2^nd meal d₃-α-tocopherol Cmax at 12h (Figure 2D, Table 1). Compared with the 40% group, the 0% group had a 55% higher mean change in oral d₃-α-T % enrichment Cmax (10.2 ± 1.1% vs 15.8 ± 1.1%, p=0.009, Figure 2E, Table 1). In contrast, no significant differences were observed with IV d₆-α-T % enrichment (p=0.47, Table 2).

Effect of fasting on vitamin E plasma enrichment in normal weight women.

To assess the effect of fasting, we compared differences in plasma d3-α-T %enrichment in the 0%fat group vs 0% fat/fast group at defined times post-dosing (Figures 1A and Figure 3). Repeated measures regression showed that at 4h, d₃-α-T %enrichment values were similar in both groups (p=0.98, Figure 3A), and no differences observed with either Cmax (1.9±0.5% vs 1.6±0.5%, p=0.70, Table 3) or AUC (2.33±0.67 vs 1.92±0.78, p=0.66, Table 3), which was not surprising given that there were no differences in treatments up to 4h.

Figure 3: Effect of fasting on vitamin E plasma %enrichment in group A healthy comparators of healthy women: 0%fat (N=10) vs 0%fat/fast (N=7).

(A-B) Plasma labeled α-T % enrichment for the first 24h following simultaneous administration of oral d₃-α-T (A) and IV d₆-α-T (B) with 0% fat breakfast followed by standard meals with 30% fat (red circle) or 12h fast (green circles). C) Difference in AUC means for oral d₃-α-T (white circle) and intravenous d₆-α-T (black circle), between the 0% fat and 0% fat/fast groups at 4h, 12h and 24h post-dosing. For A and B n=10 for the 0% fat group, and n=7 for the 0%fat-fast group (with 7 participant participated in both interventions (see supplementary figure 1). Statistical significance is shown for each time interval, with p-values representing the interaction between time and fasting intervention using repeated measures regression analysis.

Table 3:

C_max and AUC means (SEM) of d₃-α-T %enrichment in group A comparators: 0% fat versus 0% fat then fast.

	Cmax (%)				AUC (%.h)
	0% fat (n=10)	0% fat/ fast (n=7)	0% fat/ fast – 0% fat	p value	0% fat (n=10)	0% fat/ fast (n=7)	0% fat/ fast – 0% fat	p value
0–4 hours	1.9 (0.5)	1.6 (0.5)	−0.22 (0.55)	0.70	2.3 (0.7)	1.9 (0.8)	−0.4 (0.9)	0.66
0–12 hours	17.6 (0.9)	7.7 (1.1)	−9.97 (1.23)	<0.001	107.4 (6.5)	41.0 (7.5)	−66.4 (7.8)	<0.001
0–24 hours	17.9 (1.0)	15.5 (1.0)	−2.45 (0.63)	0.008	301.7 (15.4)	206.1 (16.8)	−95.6 (13.0)	<0.001

Table 3 presents data from Figure 3A. Participants received 30mg of Oral and IV α-T with breakfast (0h) containing 0% fat, followed by 30% fat lunch (4h) and dinner (8h) (0% group) or by a 12h fast (0% fat/fast group). Comparison between 0% fat and 0% fat/fast interventions of d₃-α-T %enrichment Cmax (%) and AUC (%.h). Cmax and AUC are calculated separately for 0–4, 0–12, and 0–24 hours are calculated for each participant. Means, SEM, and p values are from a repeated measures mixed-effects linear regression model. Thirteen participants are included, 7 have data for both of the comparison groups, 6 have data for only one.

At 4h, the 0% fat group received lunch with 30% meal fat, while the 0% fat/fast did not receive any meals until 12h post-dosing (Figure 3A). In contrast to the increase in d₃-α-T %enrichment following this 2^nd meal in the 0% fat group (SME), an attenuated increase was observed in the 0% fat/fast group. Consequently, there was significant difference in repeated measures regression at 0–12h (p<0.001, Figure 3B), with the 0% fat/fast group showing comparatively reduced Cmax (17.6±0.9% vs 7.7±1.1%, p<0.001, Table 3) and AUC (107.4±6.5 vs 41.0±7.5, p<0.001, Table 3).

At 12h, the 0% fat/fast group received their 2^nd meal (containing 30% fat), with subsequent increase in d₃-α-T % enrichment but not enough to obviate differences with 0%fat group at 0–24h (p<0.001, Figure 3A). Similarly, by 24h, the 0% fat/fast group had comparatively lower reduced Cmax (15.5±1.0% vs 17.9±1.0%, p=0.008, Table 3) and AUC (206.1±16.8 vs 301.7±15.4, p<0.001, Table 3).

To determine the liver’s response to the fasting intervention, we assessed differences in the intravenously administered d₆-α-T across similar time points (Figure 3B). Using repeated measures regression, there were no differences in d₆-α-T %enrichment over 4h (p=0.96), with similar d₆-α-T %enrichment Cmax (23.6±1.2% vs 23.4±1.3%, p=0.88, Table 4) and AUCs (64.5±4.0 vs 66.2±4.4, p=0.61, Table 4).

Table 4:

C_max and AUC means (SEM) of d₆-α-T %enrichment in group A comparators: 0% fat versus 0% fat then fast.

	Cmax (%)				AUC (%.h)
	0% fat (n=10)	0% fat/ fast (n=7)	0% fat/ fast – 0% fat	p value	0% fat (n=10)	0% fat/ fast (n=7)	0% fat/ fast – 0% fat	p value
0–4 hours	23.6 (1.2)	23.4 (1.3)	−0.20 (1.21)	0.88	64.5 (4.0)	66.2 (4.4)	1.71 (3.19)	0.61
0–12 hours	25.0 (1.3)	25.5 (1.4)	0.50 (0.80)	0.55	248.8 (11.7)	265.2 (12.6)	16.4 (8.5)	0.10
0–24 hours	25.0 (1.3)	25.5 (1.4)	0.50 (0.80)	0.55	496.2 (20.9)	515.5 (22.2)	19.3 (14.3)	0.23

Table 4 presents data from Figure 3B. Participants received 30mg of Oral and IV α-T with breakfast (0h) containing 0% fat, followed by 30% fat lunch (4h) and dinner (8hr) (0% group) or by a 12h fast (0% fat/fast group). Comparison between 0% fat and 0% fat/fast interventions of d₆-α-T %enrichment Cmax (%) and AUC (%.h). Cmax and AUC are calculated separately for 0–4, 0–12, and 0–24 hours are calculated for each participant. Means, SEM, and p values are from a repeated measures mixed-effects linear regression model. Thirteen participants are included, 7 have data for both of the comparison groups, 6 each have data for only one.

However, after 4h, there were significant differences by repeated measures regression, with comparatively increased d₆-α-T %enrichment in the 0% fast/fat group at 0–12h and 0–24h (p<0.001 for both, Figure 3B). At 0–12h, the 0% fat/fast intervention showed statistically non-significant differences in Cmax (25.0±1.3% vs 25.5±1.4%, p=0.55, Table 4) and AUC (248.8±11.7 vs 265.2±12.6, p=0.10, Table 4). Similar observations were found at 24h with Cmax (25.0±1.3% vs 25.5±1.4%, p=0.55, Table 4) and AUC (496.25±20.9 vs 515.5±22.2, p=0.23, Table 4). Across 0–24h, the comparative decrease in oral d3-α-T AUCs corresponded with increase in d₆-α-T AUCs (Figure 3C).

Effect of sequential meal fat on post-prandial vitamin E plasma enrichment in healthy women and obese women with MASLD

To determine how excess body fat influences post-prandial vitamin E trafficking, we investigated differences in a separate cohort of normal weight healthy women and women with obesity/MASLD (group B comparators) who received 2mg each of oral d₃-α-T and intravenous d₆-α-T, and identical meal fat over 24h (Figure 1B and Figure 4). Using repeated measures regression, no differences were observed in d₃-α-T %enrichment at 0–4h (p=0.70, Figure 4A), with similar Cmax (0.48±1.10% vs 0.33±0.13%, p=0.37, Table 5) and AUC (0.62±0.13 vs 0.41±0.17, p=0.34, Table 5).

Table 5:

C_max and AUC means (SEM) of d₃-α-T %enrichment in group B comparators: healthy controls versus MASLD.

	AUC (%.h)				Cmax (%)
	Healthy (n=10)	MASLD (n=6)	MASLD - Healthy	pvalue	Healthy (n=10)	MASLD (n=6)	MASLD – Healthy	pvalue
0–4 hours	0.62 (0.13)	0.41 (0.17)	−0.21 (0.21)	0.34	0.48 (0.10)	0.33 (0.13)	−0.15 (0.16)	0.37
0–12 hours	11.97 (1.11)	8.50 (1.43)	−3.47 (1.81)	0.076	1.75 (1.14)	1.20 (0.18)	−0.54 (0.22)	0.030
0–24 hours	30.22 (2.36)	21.35 (3.04)	−8.87 (3.85)	0.037	1.76 (0.14)	1.20 (0.18)	−0.55 (0.22)	0.027
SME* (Difference (0–12 – 0–4))					1.26 (0.11)	0.87 (0.14)	−0.39 (0.18)	0.047

Table 5 presents data from Figure 4A. Participants received 2mg of Oral and IV α-T with breakfast (0h) containing 40% fat, followed by 30% fat lunch (4h) and dinner (8h). Table shows comparison between healthy controls and MASLD of d₃-α-T %enrichment Cmax (%) and AUC (%.h). Cmax and AUC are calculated separately for 0–4, 0–12, and 0–24 hours are calculated for each participant. Means, SEM, and p values are from a repeated measures mixed-effects linear regression model.

Following the 2^nd meal at 4h, there was an excursion in d₃-α-T %enrichment in both groups consistent with a second meal effect (SME, Figure 4A). At 0–12h, the MASLD cohort had significantly reduced Cmax (1.75±0.14% vs 1.20±0.18%, p=0.03, Table 5) but reduction in AUC was not statistically significant (11.97±1.11 vs 8.50±1.43, p=0.076, Table 5). At 0–24h the MASLD cohort had significantly reduced Cmax (1.76 ± 0.14% vs 1.20 ± 0.18%, p=0.027, Table 5) and reduced AUC (30.22±2.36 vs 21.35±3.04%, p=0.037, Table 5).

We assessed the role of the liver by comparing the intravenously administered d₆-α-T in both groups (Figure 4B). Using repeated measures regression, we found significant differences in d₆-α-T %enrichment at 4h, 12h and 24h (p<0.001 for all, Figure 4B). The cohort with MASLD had significantly reduced Cmax at 4h (2.15±0.120% vs 1.52±0.16%, p=0.007), 12h (2.37±0.12% vs 1.65±0.16%, p=0.003) and 24h (2.37±0.12% vs 1.65±0.1%, p=0.003, Table 6). Similarly, the MASLD cohort had significantly reduced AUCs at 4h (5.57±0.33 vs 4.02±0.42 p=0.011), 12h (23.70±1.27 vs 16.73±1.68 p=0.005 and 24h (46.39±2.44 vs 32.79±3.15 p=0.004, Table 6).

Table 6:

C_max and AUC means (SEM) of d₆-α-T %enrichment in group B comparators: healthy controls versus MASLD.

	AUC (%.h)				Cmax (%)
	Healthy (n=10)	MASLD (n=6)	MASLD - Healthy	pvalue	Healthy (n=10)	MASLD (n=6)	MASLD – Healthy	pvalue
0–4 hours	5.57 (0.33)	4.01 (0.42)	−1.56 (0.53)	0.011	2.15 (0.12)	1.52 (0.16)	−0.63 (0.20)	0.007
0–12 hours	23.70 (1.27)	16.73 (1.64)	−6.97 (2.07)	0.005	2.37 (0.12)	1.65 (0.16)	−0.72 (0.20)	0.003
0–24 hours	46.39 (2.44)	32.79 (3.15)	−13.59 (3.99)	0.004	2.37 (0.12)	1.65 (0.16)	−0.72 (0.20)	0.003
*SME (Difference (0–12 – 0–4))					0.22 (0.029)	0.13 (0.038)	−0.09 (0.05)	0.080

Table 6 presents data from Figure 4B. Participants received 2mg of Oral and IV α-T with breakfast (0h) containing 40% fat, followed by 30% fat lunch (4h) and dinner (8h). Comparison between healthy controls and MASLD of d₆-α-T %enrichment Cmax and AUC. Cmax (%) and AUC (%.h) are calculated separately for 0–4, 0–12, and 0–24 hours and the Cmax difference between 0–4 hours and 0–12 hours are calculated for each participant. Means, SEM, and p values are from a repeated measures mixed-effects linear regression model.

To assess comparative trends with oral d₃-α-T vs IV d₆-α-T, we plotted the mean AUC differences in both groups (Figure 4C) and found a progressive increase in mean differences over time, with steeper changes in IV d₆-α-T vs oral d₃-α-T (Figure 4C). To quantify SME, we compared the mean change in pre-2^nd meal (4h) vs post-2^nd meal (12h) Cmax (Figure 4D, Table 5). Compared with the normal weight controls, the MASLD cohort had 31% lower mean change in oral d₃-α-T % enrichment Cmax (1.26±0.11% vs 0.87±0.14%, p=0.047, Figure 4E, Table 5), while differences in IV d₆-α-T % enrichment were not significant (p=0.08, Table 6).

DISCUSSION

In this study, we assessed the effect of sequential meal fat on post-prandial changes in plasma vitamin E (α-tocopherol) %enrichment over 24h, building on our previous studies that used oral (d₃-α-T) and intravenous (d₆-α-T) deuterium labeled α-tocopherol to evaluate roles of the gut and liver in the vitamin E absorptive process^5,13. To assess effect of varied dietary fat sequences, we compared healthy women who received a breakfast-lunch sequence containing 40%−30% (40% fat group) vs 0%−30% (0% fat group, Figure 1A). In both groups, the biggest post-prandial increase in oral d₃-α-T %enrichment was observed following the 2^nd meal at 4h, with Cmax occurring at approximately 12h post-dosing, consistent with the second meal effect SME^6–9. Our findings showed 55% higher SME in the 0%−30% vs 40%−30% meal fat sequence (p=0.009, Table 1), with SME defined as change in pre- vs post-2^nd meal Cmax. Consistent with this increased SME, the d₃-α-T %enrichment Cmax and AUCs were significantly higher in the 0% vs 40% comparators at 12h (p=0.003 and p=0.036 respectively) and 24h (p=0.008 and p=0.001 respectively, Table 1).

In contrast to previous depictions of SME based on fat-containing meals^6,7, findings here show that an initial meal containing minimal dietary fat followed by comparatively higher fat content significantly augments SME and post-prandial vitamin E enrichment in normal weight cohorts. These data illustrate the multi-stage absorptive process of vitamin E in the gut, with orally ingested d₃-α-T taken up by the gut, stored in cytoplasmic vacuoles and released from the gut as chylomicrons before take-up by the liver^3,4. While precise mechanisms are unclear, the robust SME in the 0% group (relative to the 40% groups) is likely explained by enhanced release of the previously ingested d₃-α-T, perhaps due to the higher threshold of dietary fat in the second meal.

In contrast with the robust SME changes observed with oral d₃-α-T in the 0% vs 40% comparators, no differences were observed with the IV d₆-α-T Cmax and AUCs at either 12h (p=0.2 and p=0.46 respectively) or 24h (p=0.2 and p=0.64 respectively, Table 2). However, in the 0% group, the notable drop in IV d₆-α-T %enrichment at 4h (Figure 2B), likely reflects a dilution effect in d₆-α-T %enrichment following the SME and the resulting increase in available pool of d₃-α-T. Given that the intravenously administered d₆-α-T bypasses intestinal absorption and specifically evaluates the role of the liver⁵, these findings indicate that SME is primarily influenced by gut enterocytes and not the liver. Like other lipids, vitamin E depends on micellization in the intestinal lumen for subsequent enterocyte uptake followed by incorporation into chylomicrons and their secretion into lymph^12,16,17. Our findings indicate that while the enterocyte’s capacity to store vitamin E is independent of meal fat at time of vitamin E ingestion, the post-prandial release of this stored vitamin E is modulated by variation in the sequence of oral dietary fat ingestion (0%−30% vs 40%−30%). The enterocyte plays a central role in nutrient sensing, fat metabolism, production and regulation of enteroendocrine hormones, as well as temporary storage as cytoplasmic vacuoles and release into chylomicrons^9,18–21. While precise mechanisms are unknown, findings here coupled with previous studies linking SME to insulin resistance⁶, raises the possibility that variation in dietary fat sequences potentially induce altered nutrient sensing, neurohumoral and enteroendocrine responses regulated by the gut enterocytes²¹.

While an initial meal containing minimal dietary fat followed by comparatively higher fat content significantly augmented SME and post-prandial vitamin E enrichment in normal weight cohorts, it is unclear whether similar dietary intervention would improve vitamin E absorption and bioavailability in chronic diseases associated with impaired absorption of vitamin E. Given the complex and distinct pathological changes and mechanisms in chronic diseases, dedicated dietary studies are needed in cohorts with chronic diseases, specifically targeting aberrant post-prandial changes in vitamin E absorption.

In response to the fasting intervention in the 0%fat/fast group, there was continued d₃-α-T %enrichment in a linear pattern despite the absence of meals or dietary fat ingestion during the period of fast (Figure 3A), raising the possibility of a coordinated compensatory process by the brain-gut axis to maintain vitamin E homeostasis in response to fasting^22,23. When the 0%fat/fast group received the 2^nd meal after a 12h fasting period, there was an attenuated SME in d₃-α-T %enrichment, indicating timing of the 2^nd meal influences the magnitude of SME and the subsequent bioavailability.

To investigate how post-prandial vitamin E physiology and SME are altered in the presence of excess body/liver fat, we assessed outcomes in a separate cohort of healthy women and women with MASLD (group B comparators) who received 2mg doses of oral and IV labeled α-T and identical meal fat as previously reported (Figure 1B)¹³. Findings showed the cohort with MASLD had 55% lower SME (p=0.047, Figure 4E) and significantly reduced d₃-α-T %enrichment Cmax and AUCs at 24h (p=0.027 and p=0.037 respectively, Figure 4A, Table 5). Given that the cohort with MASLD had an average liver fat of 12% (vs <2% in healthy cohort, Supplementary Table 1), these data suggest that post-prandial d₃-α-T %enrichment is attenuated by excess liver fat, consistent with previous studies describing reduced oral bioavailability in the obesity related syndromes including metabolic syndrome and MASLD^12,13.

To further explore effect of liver fat, we compared IV d₆-α-T parameters (Figure 4B) and found significantly reduced Cmax and AUCs in the MASLD group across all timepoints (Figure 4, Table 6). In contrast with oral d₃-α-T %enrichment in which discernable differences emerged post-2^nd meal (after 4h), differences in IV d₆-α-T %enrichment were evident at 4h, indicating reduced enrichment parameters are due to underlying liver pathology (excess liver fat) rather than post-prandial SME. Additionally, when oral d₃-α-T % enrichment plots were adjusted for IV d₆-α-T (d₃/d₆-α-T ratios), differences observed with oral d₃-α-T (Figure 4A) were no longer present (Supplementary Figure 2A), in contrast with the healthy comparator groups that retained similar patterns as the oral d₃-α-T % enrichment plots (Figures 2A & 3A, Supplementary Figures 2B & 2C). Taken together, findings suggest that excess liver fat attenuates post-prandial vitamin E enrichment, consistent with previous analysis showing reduced vitamin E kinetics as a function of body and liver fat, as well as the vitamin E sequestration hypotheses—the concept of sequestration, whereby excess liver fat decreases vitamin E release into plasma¹³.

To integrate the SME concepts with previous descriptions of α-T enrichment in lipoproteins in the healthy comparator groups⁵ and the MASLD/healthy¹³, we compared changes in α-T enrichment in plasma and lipoproteins in healthy comparators (supplementary Figure 3) and MASLD/healthy comparator (Supplementary Figure 4). Across all comparator groups, there was a consistency between changes in α-T % enrichment in plasma and triglyceride rich lipoproteins (TRL), including comparatively reduced d₃-α-T %enrichment in the cohort with MASLD relative to healthy controls. These data suggest the attenuated SME in MASLD is primarily a function of disease status (excess body and liver fat), rather than an inherent mechanistic difference between triglycerides and vitamin E, such that the same oral dose of d₃-α-T is being enriched into a larger pool of lipids/lipoproteins in cohorts with obesity and MASLD.

This study had several strengths and weaknesses. Strengths included use of dual oral and intravenous stable isotope that expands on our understanding of the complex multi-organ processes that regulate nutritional physiology and offers a glimpse of how this process may be disrupted in chronic diseases such as obesity and MASLD. An additional strength is the use of varied meal fat sequences to investigate physiological and pathophysiological changes in nutrient biology. Study limitations include the female-only cohorts that precluded analysis of sex differences. Similarly, the study did not include obese cohorts without MASLD to better delineate the role of excess body vs liver fat in pathophysiological vitamin E changes. Additionally, studies in the obese/MASLD cohort did not include varied dietary fat sequences and fasting interventions, thus it is unclear how observations in these normal weight healthy women would vary in cohorts with obesity or MASLD.

In conclusion, the findings here show that in healthy women, a low-to-high meal fat sequence augments post-prandial oral vitamin E enrichment and SME, and a compensatory increase in vitamin E release by the liver in response to fasting interventions. Additionally, in women with MASLD, SME and post-prandial vitamin E enrichment are attenuated. In addition to expanding our understanding of the physiological and pathophysiological changes in the post-prandial phase of the vitamin E absorption, these data provide a foundation for future studies aimed at exploring how adjustments to dietary fat intake can be utilized in augmenting vitamin E bioavailability in cohorts with chronic metabolic diseases. Additionally, to clarify mechanisms, future studies would investigate enteroendocrine changes in response to controlled nutritional interventions, weight management programs, and anti-obesity medications.

Abbreviations

α-T

RRR-α-tocopherol

d₃-α-T

d₃-RRR-α-tocopherol

d₆-α-T

d₆-RRR-α-tocopherol

AUC

Area under the curve

Cmax

Maximum concentration

MASLD

Metabolic dysfunction-associated steatotic liver disease

SME

Second meal effect