Common Variants of Cytochrome P450 4F2 Exhibit Altered Vitamin E-ω-Hydroxylase Specific Activity

Abstract

Human cytochrome P450 4F2 (CYP4F2) catalyzes the ω-hydroxylation of the side chain of tocopherols (TOH) and tocotrienols (T3), the first step in their catabolism to polar metabolites excreted in urine. CYP4F2, in conjunction with α-TOH transfer protein, results in the conserved phenotype of selective retention of α-TOH. The purpose of this work was to determine the functional consequences of 2 common genetic variants in the human CYP4F2 gene on vitamin E-ω-hydroxylase specific activity using the 6 major dietary TOH and T3 as substrate. CYP4F2-mediated ω-hydroxylase specific activity was measured in microsomal preparations from insect cells that express wild-type or polymorphic variants of the human CYP4F2 protein. The W12G variant exhibited a greater enzyme specific activity (pmol product ⋅ min⁻¹ ⋅ pmol CYP4F2⁻¹) compared with wild-type enzyme for both TOH and T3, 230–275% of wild-type toward α, γ, and δ-TOH and 350% of wild-type toward α, γ, and δ-T3. In contrast, the V433M variant had lower enzyme specific activity toward TOH (42–66% of wild type) but was without a significant effect on the metabolism of T3. Because CYP4F2 is the only enzyme currently shown to metabolize vitamin E in humans, the observed substrate-dependent alterations in enzyme activity associated with these genetic variants may result in alterations in vitamin E status in individuals carrying these mutations and constitute a source of variability in vitamin E status.

Introduction

Cytochrome P450 4F2 (CYP4F2)⁵ is an ω-hydroxylase that catalyzes the first step in the only known pathway of vitamin E metabolism, the tocopherol (TOH)-ω-hydroxylase pathway (1, 2). CYP4F2 was initially characterized as catalyzing the ω-hydroxylation of leukotriene B4 as well as arachidonic acid (AA) (3, 4). The ω-hydroxylation of the terminal methyl group of the phytyl side chain of vitamin E to form the 13′-OH product is followed by the truncation of the molecule into short chain, more water-soluble carboxychromanol metabolites that are excreted in urine (5–7). Like all CYP enzymes, CYP4F2 is part of an endoplasmic reticulum membrane-bound complex that requires the cytochrome P450 (CYP) enzyme itself and reducing equivalents from NADPH via the electron donor cytochrome P450 reductase. While CYP4F2 shares its activity toward AA and leukotriene B4 with several other P450 enzymes, there is no known redundancy in TOH-ω-hydroxylase activity.

Vitamin E is the generic term for several chemical structures, including α-, γ-, and δ-TOH containing a saturated phytyl side chain and having 3, 2, and 1 methyl group on the chromanol ring, respectively. The corresponding α-, γ-, and δ-tocotrienols (T3) have 3 double bonds in the phytyl side chain. CYP4F2 exhibits profound substrate preference for non-α-TOH forms of vitamin E (8). In addition, the hepatic α-TOH transfer protein preferentially promotes the secretion of α-TOH from the liver into the bloodstream for transport to tissues (9). Together, these 2 proteins comprise an effective mechanism that results in the selective accumulation of α-TOH in tissues and the elimination of other forms of vitamin E regardless of the relative proportions of each form in the diet (8).

Plasma α-TOH concentrations and responses to supplementation in clinical trials vary widely within a healthy population but are consistent within individuals, lending support to the idea that α-TOH status is a genetically determined trait (10). Due to the apparent importance of the TOH-ω-hydroxylase pathway in vitamin E status, it has been suggested that individual single nucleotide polymorphisms (SNP) in the CYP4F2 gene may contribute to variation in vitamin E status in humans (11). Recently, 2 nonsynonymous SNP in the CYP4F2 gene have been identified: a T to G nucleotide change at mRNA position 84 leading to a tryptophan to glycine substitution at amino acid position 12 (rs3093105; W12G) and a G to A nucleotide change at mRNA position 1347 leading to a valine to methionine substitution at amino acid position 433 (rs2108622; V433M). The W12G variant has a minor allele frequency of 11 and 21% in the European American and African American populations examined, respectively, and the V433M variant has a minor allele frequency of 17 and 9%, respectively (12). The minor allele frequency for the W12G variant and the V433M variant are 6 and 26%, respectively, as determined by using a combined Asian sampling panel (Chinese and Japanese HapMap dataset) (12). The existence of the double mutant in humans has yet to be reported. Recently, the V433M polymorphism has been associated with altered warfarin dose requirements in humans (13) as well as hypertension (14) and ischemic stroke in several populations (14–16). Speculative mechanisms behind these associations relate to the role of CYP4F2 in 20-hydroxyeicosatetraenoic acid (20-HETE) production and the putative role of CYP4F2 as a phylloquinone oxidase (17).

A previous study demonstrated that the V433M substitution in CYP4F2 results in decreased ω-hydroxylation of AA (12). The purpose of this study was to determine whether the amino acid substitutions resulting from these 2 nonsynonymous SNP in CYP4F2 cause altered vitamin E-ω-hydroxylase specific activity, using several major dietary forms of vitamin E as substrates.

Materials and Methods

Chemicals and materials.

TOH were obtained from Matreya, LLC. The T3 were a gift from Volker Berl (BASF). Sf9 insect cells were provided by W. Lee Kraus (Cornell University). Insect cell media was purchased from BD Biosciences. Insect cell certified fetal bovine serum albumin (BSA) was purchased from Atlanta Biologicals and antibiotic-antimycotic solution was purchased from Mediatech. β-NADPH, hemin chloride, 5-aminolevulinic acid, ferric citrate, and fraction V BSA were purchased from Sigma Chemical. The internal standard d₉-α-TOH was custom synthesized by J. Swanson (Cornell University). Pyridine and N,O-Bis-[trimethylsilyl]trifluoracetamide containing 1% trimethylchlorosilane were purchased from Pierce Chemical. Insect cell microsomes (BD Supersomes) containing human CYP4F2, cytochrome P450 reductase, and cytochrome b₅ were purchased from Gentest. Polyvinylidene difluoride membranes, Odyssey blocking buffer, and IRDye 680 conjugated goat (polyclonal) anti-rabbit IgG were purchased from Li-Cor Biosciences. Rabbit anti-human 4F2 polyclonal antibody was obtained from Fitzerald Industries. AA and 20-HETE were purchased from Cayman Chemical.

Generation of baculovirus.

Recombinant baculovirus preparations containing 4 human CYP4F2 enzyme isoforms (wild-type W12/V433, position 12 variant G12/V433, position 433 variant W12/M433, and the double variant G12/M433) were generated by Stec et al. (12) as previously described. The vector contained 2 promoters that independently drive the expression of CYP4F2 and of the required coenzyme cytochrome P450 reductase.

Preparation of microsomes containing CYP4F2 variants.

Sf9 cells were infected with recombinant baculovirus at a multiplicity of infection of 5–10. Sf9 cells were grown in insect cell media with 10% fetal bovine serum and antibiotic-antimycotic solution. At the time of infection, the media were supplemented with hemin (conjugated to BSA; 4 μmol/L), 5-aminolevulinic acid (0.5 mmol/L), and ferric citrate (100 μmol/L). The cells were incubated at 27°C for 3 d following infection. After this time, cells were collected, washed with PBS, resuspended in homogenization buffer (0.1 mol/L sodium phosphate buffer containing 10% glycerol, 1 mmol/L EDTA, 0.1 mmol/L dithiothreitol, 1 mmol/L phenylmethylsulfonyl fluoride, pH 7.5), and frozen at −80°C. The cells were then thawed, homogenized using a dounce homogenizer, and centrifuged at 10,000 × g for 20 min at 4°C. The supernatant was then subjected to high-speed centrifugation at 100,000 × g for 1 h at 4°C. The microsomal pellet was resuspended in 0.1 mol/L sodium phosphate containing 20% glycerol, 1 mmol/L EDTA, 0.1 mmol/L dithiothreitol, and 0.2 mmol/L phenylmethylsulfonyl fluoride and stored at −80°C. The total microsomal protein concentration was measured using a Bradford-based Bio-Rad assay with BSA as the standard. In addition, negative control Sf9 cells that did not receive the virus were prepared in triplicate as described above.

Measurement of CYP4F2 protein.

Western blots to quantify recombinant CYP4F2 protein were performed on the microsomes prepared from Sf9 cells. A standard curve of human CYP4F2 (4F2 Supersomes) was run on each gel. The amount of microsomal protein concentration that was loaded in each lane varied between samples in order to ensure each sample was within the standard curve that was run on each blot. Between 47 and 100 μg of microsomal protein was heated to 95°C for 5 min, electrophoresed on 12% SDS-polyacrylamide gels, and blotted onto polyvinylidene difluoride membrane. Membranes were blocked with Odyssey blocking buffer for 2 h at room temperature and then incubated with rabbit anti-human 4F2 polyclonal antibody (1:300). The membranes were then incubated with IRDye 680 conjugated goat (polyclonal) anti-rabbit IgG for 1 h at room temperature. The membranes were visualized using an Odyssey infrared imager (Li-Cor) with densito-metric analysis performed using Odyssey software. The concentration of CYP4F2 protein in the microsomal preparations prepared from each human CYP4F2 variant was derived from the standard curve on each individual blot.

Measurement of vitamin E-ω-hydroxylase activity.

A comparison of CYP4F2-dependent vitamin E-ω-hydroxylase activity among the 4 enzyme isoforms using Sf9 microsomes obtained from 3 independent transfection batches was first determined using δ-TOH as substrate. The reaction system (0.5 mL) consisted of NaP buffer (0.1 mol/L, with 0.1 mmol/L EDTA, pH = 7.4), 60 μmol/L δ-TOH added as a complex with 1% (wt:v) fraction V BSA prepared as previously described (8), and the volume of each of the 4 microsomal preparations sufficient to achieve an equivalent total microsomal protein concentration. Microsomes were preincubated with the substrate-BSA complex at 37°C for 5 min, after which NADPH was added to a final concentration of 2 mmol/L and the system incubated for 35 min at 37°C. The reaction was terminated and δ-13’-OH-TOH product extracted and derivatized as previously described (8). d₉-α-TOH was used as internal standard.

An investigation of the impact of the single and double mutations using several TOH (α-, γ-TOH) and T3 (α-, γ-, δ-T3) as substrate was then performed using microsomes from the same transfection batch. Conditions for evaluation of ω-hydroxylation of γ-TOH were similar to that of δ-TOH described above. Those used for assay of metabolism of α-TOH and T3 were similar except for substrate concentrations (α-TOH, 350 μmol/L; T3, 10 μmol/L) to account for differences in membrane uptake as characterized previously (8). In addition, several of these substrates were assayed at a lower concentration (γ-, δ-TOH, 30 μmol/L; α-T3, 4 μmol/L). The analogous 13'-ω-hydroxylated reaction products were extracted and derivatized as described above.

Measurement of AA ω-hydroxylation.

Assays to determine the AA-ω-hydroxylase specific activity were performed in NaP buffer with 42 μmol/L AA without BSA and an equal amount of microsomal protein from each of the 4 microsomal preparations from 1 transfection batch in 0.5 mL reaction volume. Reaction mixtures were preincubated for 5 min at 37°C, after which NADPH (2 mmol/L) was added and incubations continued for 35 min at 37°C. 20-HETE was extracted and derivatized after acidification as described above by using d₉-α-TOH as internal standard.

GC-MS.

A Hewlett-Packard 6890 gas chromatograph coupled to a Hewlett-Packard 5872 mass selective detector operated in selected ion mode was used for quantification of vitamin E and AA reaction products. The gas chromatograph was fitted with a Hewlett-Packard HP-1 methylsiloxane capillary column (30 m × 0.25 mm) and operated in split injection mode using helium as the carrier gas. Long-chain TOH and T3 metabolites were resolved isothermally at 280°C for 20 min. 20-HETE was resolved using a temperature program starting at 180°C, ramping to 250°C at a rate of 6°C/min, then to 280°C at a rate of 25°C/min, holding at 280°C for 13 min. Metabolite product was quantified using the d₉-α-TOH internal standard. 20-HETE values were corrected using a relative detector response factor of 50.

Calculation of enzyme specific activity.

Enzyme activity data (pmol product/min) were normalized to the microsomal CYP4F2 protein content as determined by quantitative Western blotting to yield specific activity (pmol ⋅ min⁻¹ ⋅ pmol 4F2⁻¹). Data are presented as specific activity or as percentage of wild-type specific activity.

Statistical analysis.

Three independent transfection batches of the 4 microsomal enzyme isoform preparations were each assayed in triplicate for the initial comparison of specific activities using δ-TOH as substrate. Each of the 3 batches was analyzed independently using 1-way ANOVA. When significant main effects of enzyme isoform were detected, post hoc multiple comparisons were performed using a Tukey correction. The comparison of specific activities of the 4 enzyme isoforms using various TOH and T3 substrates was performed in triplicate using the same transfection batch. To assess the effect of enzyme isoform on CYP4F2 specific activity, Student’s t test was used to compare the specific activity of each variant enzyme to that of the wild-type enzyme for each substrate individually. In addition, for each of the 3 variants, the mean percent of the wild-type value for each TOH substrate was compared with the mean percent of the wild-type value of the other 2 TOH substrates using 1-way ANOVA followed by multiple comparisons with a Tukey correction. Similarly, for each of the 3 variants, the mean percent of the wild-type value for each T3 substrate was compared with the mean percent of the wild-type value for the other 2 T3 substrates using 1-way ANOVA followed by multiple comparisons with a Tukey’s correction. Differences in AA-ω-hydroxylase specific activity among the 4 enzyme isoforms were determined using 1-way ANOVA followed by multiple comparisons with a Tukey’s correction. Values in the text are means ± SD. All statistical analyses were performed using JMP version 8 (SAS Institute). If needed, responses were log-transformed to meet the assumptions of the statistical tests. All tests were 2-sided and P < 0.05 was considered significant.

Results

Expression of CYP4F2 variants in Sf9 microsomes.

A 53-kDA protein that comigrated with human CYP4F2 standard was observed in each of the 4 CYP4F2 microsomal preparations [W12/V433 (wild type), G12/V433, W12/M433, G12/M433]. A representative blot from 1 of the transfection batch is shown in Fig. 1A. Negative control microsomal preparations did not show a band at 53 kDA (Fig. 1B).

FIGURE 1.

CYP4F2 expression of 4 enzyme isoforms in Sf9 microsomal preparations. Lanes 1–4: CYP4F2 standard of 0.05, 0.10, 0.15, and 0.20 pmol, respectively. (A) Lanes 5–8, CYP4F2 isoforms: WT, W12G, V433M, W12G/V433M, respectively. (B) Lanes 9 and 10 (from a separate blot): CYP4F2 standard of 0.15 pmol and negative control microsomes, respectively. Western blotting using antibodies against CYP4F2 detected a 53-kDa protein that comigrated with CYP4F2 standard. Blots are representative of 3 independent transfection batches.

δ-TOH ω-hydroxylase specific activity of wild-type and variants of CYP4F2.

All assayed microsomal preparations except the negative controls showed production of δ-13’-OH-TOH as determined by electron impact MS. Analysis of log-transformed data from 1 transfection batch indicated the mean δ-13’-OH-TOH production of wild-type enzyme was significantly different from the 3 variants after normalization to microsomal CYP4F2 protein content (Fig. 2). The W12G substitution resulted in significantly higher δ-13’-OH-TOH production (275% of wild-type enzyme; P < 0.0001) compared with wild type. In contrast, the amino acid substitution at position 433 resulted in decreased δ-13’-OH-TOH production (66% of wild-type enzyme; P < 0.02). The effect of the double mutation was the same as the single mutation at position 433, exhibiting diminished specific activity (60% of wild-type enzyme; P < 0.007). Microsomal preparations from all 3 transfection batches had similar differences in specific activity due to the mutations, except that in 1 batch the activity of the double mutant did not significantly differ from that of the wild type (data not shown). The relative enzyme specific activities using 30 μmol/L δ-TOH were consistent with the results observed using 60 μmol/L δ-TOH. The W12G variant exhibited 145% of wild-type enzyme specific activity, whereas the V433M variant and double variant had 37 and 66% of wild-type specific activity, respectively, at the lower concentration.

FIGURE 2.

δ-TOH-ω-hydroxylase specific activity of CYP4F2 isoforms. δ-13-OH-TOH formation was determined in microsomal incubations containing 60 μmol/L δ-TOH-BSA complex. Data are means ± SD, n = 3. Means without a common letter differ, P < 0.05.

Comparative metabolism of various TOH and T3 by CYP4F2 variants.

The W12G enzyme variant had greater specific activity toward all 3 TOH tested (255 ± 22% of wild type; P < 0.0001) (Fig. 3A). In sharp contrast, the V433M enzyme variant had lower ω-hydroxylase activity toward all 3 TOH (57 ± 14% of wild type; P < 0.03) (Fig. 3B). Among the TOH substrates, α-TOH was the most negatively affected by the V433M substitution but was the least affected by the gain-of-function effect of the W12G substitution. The W12G/V433M double mutant had reduced specific activity toward all 3 TOH, similar to what was observed for the V433M variant (48 ± 14% of wild type; P < 0.002) (Fig. 3C). The relative specific activities of the enzyme isoforms obtained using 30 μmol/L γ-TOH were consistent with those observed using 60 μmol/L γ-TOH. The W12G variant resulted in 162% of wild-type specific activity, whereas the V433M variant and the double variant had 47 and 68% of the wild-type specific activity, respectively, at the lower concentration.

FIGURE 3.

Alterations in ω-hydroxylation of TOHs by CYP4F2 isoforms W12G (A), V433M (B), and W12G/V433M (C). Enzyme specific activity (pmol product ⋅ min⁻¹ ⋅ pmol CYP4F2⁻¹) for each of the 3 variants is expressed as percent of the wild-type activity value. Dotted line represents 100% of wild type. Data are means ± SD, n = 3. *Different from wild type, P < 0.03. Within each graph, means without a common letter differ, P < 0.05.

As was the case with TOH substrates, the W12G variant had much higher specific activities toward all 3 T3 substrates (353 ± 2% of wild type; P < 0.003) (Fig. 4A). In contrast, the V433M substitution had little or no effect on specific activity toward the T3, with a marginally significant difference from the wild type evident with δ-T3 as substrate (P = 0.046) (Fig. 4B). Again, the results of the double mutant were similar to those obtained with the V433M variant (Fig. 4C). In contrast to the case with TOH, none of the 3 variants significantly differed in specific activity among the 3 T3 substrates. Additionally, the relative specific activities of the enzyme isoforms obtained using 4 μmol/L α-T3 were consistent with those observed using 10 μmol/L α-T3. The W12G variant had 203% of the wild-type specific activity and the V433M variant and the double variant had 84 and 106% of the wild-type specific activity at the lower concentration.

FIGURE 4.

Alterations in ω-hydroxylation of T3s by CYP4F2 isoforms W12G (A), V433M (B), and W12G/V433M (C). Enzyme specific activity (pmol product ⋅ min⁻¹ ⋅ pmol CYP4F2⁻¹) for each variant is expressed as percent of the wild-type activity value. Dotted line represents 100% of wild type. Data are means ± SD for each T3 evaluated, n = 3. *Different from wild type, P < 0.05.

ω-Hydroxylation of AA by CYP4F2 variants.

In light of the finding of a significant gain-of-function effect of the W12G variant on vitamin E metabolism, the 4 enzyme isoforms were reevaluated for AA ω-hydroxylation. The specific activity of the W12G variant was greater than that of wild-type enzyme (190% of wild-type enzyme; P < 0.0009) (Fig. 5). The V433M variant had a severe loss of function (20% of wild-type enzyme; P < 0.0009), an effect similar to the loss of function observed with the double mutant (22% of wild-type enzyme; P < 0.0009). Thus, the effects of the 3 variants on ω-hydroxylation of AA mirrored the effect of the ω-hydroxylation of the 3 TOH.

FIGURE 5.

Metabolism of AA by CYP4F2 isoforms. 20-HETE formation was determined in incubations containing 42 μmol/L AA. Data are means ± SD, n = 3. Arachidonate ω-hydroxylase specific activity was compared among the 4 enzyme isoforms. Means without a common letter differ, P < 0.05.

Discussion

CYP4F2 catalyzes the initial step in the only known enzyme-mediated pathway of vitamin E metabolism, the vitamin E-ω-hydroxylase pathway (2). The product of the reaction, the hydroxyl derivative of 1 of the 2 terminal methyl groups of the hydrophobic side chain, then undergoes further metabolism to polar carboxychromanols that are excreted in the urine. No other major human liver CYP enzymes have this enzyme activity (1) and thus, unlike many other CYP-catalyzed reactions, redundancy of function appears lacking. CYP4F2 exhibits pronounced substrate specificity among the various naturally occurring forms of vitamin E, such that those forms with incomplete methylation of the phenol ring or with unsaturated side chains are metabolized more rapidly (8). Consequently CYP4F2 appears to play a unique role in the postabsorptive catabolism of vitamin E that contributes, along with the hepatic α-TOH transfer protein, to the widely expressed phenotype of selective tissue deposition of α-TOH over other forms of vitamin E. Here we tested the hypothesis that 2 common mutations in the human CYP4F2 gene would alter the activity of the enzyme toward vitamin E substrates. Consistent with this hypothesis, the W12G variant had significantly greater ω-hydroxylase specific activity toward the 3 TOH tested and all 3 T3. In sharp contrast, the V433M variant had significantly lower specific activity toward the TOH but was without significant effect with the T3 substrates. Two TOH and one T3 were also tested at lower concentrations and the relative specific activities of the 4 enzyme isoforms were consistent across both substrate concentrations. All concentrations tested were either near or substantially below the apparent Michaelis-Menten constant (K_m) for TOH-ω-hydroxylase activity in human liver microsomes (8).

Our previous characterization of wild-type CYP4F2 activity toward a wide variety of naturally occurring and synthetic vitamin E substrates indicated that the T3 were in general metabolized more readily than the TOH (8). The directionality of the effects of the 2 common SNP investigated here, in comparison to the wild-type enzyme, indicates that the T3-over-TOH substrate preference is retained, if not accentuated, in those variants. The biological rationale for the observed differences in catabolism of the different forms of vitamin E, even among the polymorphic forms of CYP4F2, remains an enigma. An accumulation of T3 or TOH that are subject to higher rates of catabolism has been reported to result in beneficial effects in animal models (18–20) and cytotoxic effects in cell culture models (21).

In an investigation of the functional effects of the W12G and V433M variants of CYP4F2, Stec et al. (12) reported a loss of AA ω-hydroxylase activity by the V433M variant and a trend toward higher specific activity accompanying the W12G substitution. Both as a positive control for our findings with vitamin E and to reevaluate the W12G variant in light of the substantial gain-of-function with both TOH and T3, we compared AA ω-hydroxylase specific activity among the 4 enzyme isoforms. Under our experimental conditions, findings with the V433M variant and double mutant mirrored those previously reported and the W12G variant exhibited a significant 2-fold enhancement of specific activity, similar to what was observed with the vitamin E substrates. The physiological importance of the latter finding is difficult to evaluate due to the redundancy in arachidonate 20-hydroxylase activity among multiple CYP450 enzymes and the potential for compensatory degradation of 20-HETE.

The means by which these 2 amino acid substitutions act to affect enzyme function is unclear. The 3-dimensional structure of CYP4F2 has not been determined or modeled; thus, the amino acid residues that participate in substrate binding, channeling, and catalysis have not been identified. Sequence heterogeneity in the amino acid 67–114 region of CYP4F3A and CYP4F3B correlates with the differences in substrate specificity of these two 4F isoforms, suggesting this region may play a critical role in determining substrate specificity of other CYP4F enzymes (22), but neither of the current substitutions lies within this putative substrate recognition domain. Other possibilities include alterations in protein structure that affect how the enzyme orients in the membrane, which may in turn influence the access by membrane-bound substrates like vitamin E. Different forms of vitamin E may reside at different depths in biological membranes (23).

The impact of these 2 variations in CYP4F2 structure on vitamin E status in humans is at present unknown. Both of the naturally occurring variants are relatively common in European American- and African American-derived sampling panels as well as in the Chinese and Japanese HapMap dataset (12). In vivo investigations indicate that decreased CYP4F2 activity results in altered vitamin E status. Yamashita et al. (24) reported that dietary sesamin, a sesame seed lignan later found to be an inhibitor of CYP4F2 (25), resulted in elevated concentrations of α- and γ-TOH in rat serum. Additionally, short-term administration of moderate amounts of sesame seeds to humans resulted in elevations in serum γ-TOH levels (26). These findings raise the distinct possibility that the CYP4F2 variants investigated here may contribute to variation in vitamin E status. Plasma α-TOH levels and responses to vitamin E supplementation are consistent over time within individuals yet are highly variable between individuals, leading to the suggestion that α-TOH status is influenced by genetic factors (10). Individuals carrying 1 or 2 of the V433M variant alleles may have increased plasma and tissue levels of TOH and display a more pronounced response to TOH supplementation than individuals carrying wild-type alleles. In contrast, individuals carrying the W12G allele may have lower plasma and tissue concentrations of TOH and little or no change in response to TOH or T3 supplementation. Underlying genetic variability in vitamin E status and response to supplementation may contribute to inconsistent and null findings in both epidemiological and randomized controlled trials regarding vitamin E status and its relationship to disease risk. These possibilities are currently under investigation.

In conclusion, 2 common nonsynonymous SNP in the CYP4F2 gene cause significant alterations in vitamin E-ω-hydroxylase specific activity in a variant-dependent manner. The W12G variant had increased enzyme activity toward both TOH and T3, whereas the V433M variant exhibited reduced enzyme activity toward TOH but not T3. The effect of these SNP on vitamin E status and the response to vitamin E supplementation in humans has important clinical implications and should be investigated.