|Home | About | Journals | Submit | Contact Us | Français|
The role of hepatic xenobiotic regulatory mechanisms to modulate hepatic α-tocopherol concentrations during excess vitamin E administration remains unclear. We hypothesized that increased hepatic α-tocopherol would cause a marked xenobiotic response. Thus, we assessed cytochrome P450 oxidation systems (phase I), conjugation systems (phase II) and transporters (phase III) following daily α-tocopherol injections (100 mg/kg body wt) up to 9 days in rats. α-Tocopherol injections increased hepatic AT concentrations nearly 20-fold, along with a 10-fold increase in hepatic α-tocopherol metabolites, α-CEHC and α-CMBHC. Expressions of phase I (CYP3A2, CYP3A1, CYP2B2) and phase II (Sult2a1) protein and/or mRNA genes were variably affected by α-tocopherol injections; however, expressions of phase III transporter genes were consistently changed by α-tocopherol. Two liver efflux transporter genes, ABCB1b and ABCG2, were up-regulated following α-tocopherol injections, while OATP, a liver influx transporter, was down-regulated. Thus, an over-load of hepatic α-tocopherol increases its own metabolism, and increases genes of transporters that are postulated to lead to increased excretion of both vitamin E and its metabolites.
Hepatic vitamin E is regulated, in part, by the α-tocopherol transfer protein (α-TTP), which facilitates the secretion of α-tocopherol into plasma [1, 2]. Perhaps, more important in regard to vitamin E regulation, is the role of its metabolism [3, 4]. Vitamin Es are metabolized similarly to some xenobiotics in that they are initially hydroxylated by cytochrome P450s (CYPs), partially β-oxidized, conjugated, and excreted in urine  or bile .
The first step in vitamin E metabolism is thought to be the ω-hydroxylation of the side chain. Sontag and Parker  have investigated the CYPs involved in vitamin E metabolism; only CYP4F2 was demonstrated to be involved in the ω-oxidation of γ–tocopherol and was therefore termed tocopherol hydroxylase . They further reported “CYP4F2-mediated tocopherol-omega-hydroxylation is a central feature underlying the different biological half-lives, and therefore biopotencies, of the tocopherols and tocotrienols” . In addition, “excess” hepatic α-tocopherol leads to the up-regulation of the expression of CYP3A family proteins [3, 4, 9] and genes .
Following ω-hydroxylation and transformation to a carboxyl group, β-oxidation takes place to form the ultimate vitamin E metabolites (e.g. α-CEHC and γ-CEHC (carboxyethyl-hydroxychroman) derived from α-tocopherol and γ-tocopherol (or tocotrienols), respectively) [7, 11]. The intermediate compounds between 13′OH-tocopherols and CEHCs have been identified . β-oxidation of the hydroxylated tocopherols has been localized; the final steps in tocopherol β-oxidation likely take place in the mitochondria because these organelles were found to accumulate CEHCs .
CEHCs can be sulfated or glucuronidated [14–16]. There is controversy concerning this conjugation step. Pope et al.  reported in α-tocopherol supplemented humans that urinary α-CEHC glucuronide was present, but the sulfate conjugate could not be unambiguously characterized. In contrast, the Jiang group [17–19] has identified that sulfated-γ-CEHC is the major CEHC conjugate both in rats and in human cells in culture. Sulfated intermediates were found between 13′-OH-γ-tocopherol and γ-CEHC, suggesting that sulfation may be an important early step in intracellular trafficking to guide vitamin E metabolism. It is not clear whether the differences in the two outcomes, either sulfation or glucuronidation, relates to differences in α- and γ-tocopherol metabolism, or differences between rats and humans, or differences of measures from liver compared with urine samples. Moreover, the enzymes involved have not been identified.
Vitamin E metabolites are excreted in urine  and bile . Xenobiotic transporters are likely candidates for mediating both hepatic tocopherol and CEHC excretion. Nr1i2 (nuclear receptor subfamily 1, group 1, member 2, also known as the pregnane X receptor (PXR)) regulates a constellation of genes involved in xenobiotic detoxification [20–22]. On ligand binding, PXR binds to its response element in the promoter region of genes and induces some cytochrome P450 oxidation systems (phase I), conjugation systems (phase II) and transporters (phase III) . All of these systems are potential mediators of hepatic concentrations of vitamin E and its metabolites. Indeed, Landes et al  demonstrated “in HepG2 cells transfected with the human PXR and the chloramphenicol acetyl transferase (CAT) gene linked to two PXR responsive elements, CAT activity was most strongly induced by alpha- and gamma-tocotrienol followed by rifampicin, delta-, alpha- and gamma-tocopherol.” However, Cho et al.  showed that stimulation of PXR by the mouse PXR activator, pregnenolone 16a-carbonitrile (PCN), in wildtype compared with PXR-null mice decreased vitamin E metabolism. Moreover, they identified a new CEHC conjugate, a glucoside, using metabolomic techniques and mass spectrometry. Importantly, PCN stimulation of PXR, decreased the urinary excretion of α-CEHC glucuronide and γ-CEHC glycoside in wildtype but not PXR-null mice. Thus, the role of PXR-dependent genes in vitamin E metabolism remains unclear.
We have previously reported that two PXR-dependent xenobiotic systems, CYP3A and p-glycoprotein (multi-drug resistance gene (MDR1), an ATP-binding cassette transporter, also referred to as ABCB1) are increased in the livers from vitamin E injected rats [3, 9]. Dietary α-tocopherol in mice also up-regulates Cyp3a protein  and mRNA [25, 26]. Given that both CYP3A and ABCB1 are under the control of PXR , we hypothesized that other mechanisms to control xenobiotics might also be up-regulated by α-tocopherol. Typically, xenobiotic compounds result in the coordinate up-regulation of drug-metabolizing pathways through transcriptional induction of genes encoding phase I CYP enzymes, phase II SULT enzymes, as well as and ABC transporters, such as MDR1 . We, therefore, undertook a survey of hepatic xenobiotic regulatory mechanisms in the livers from α-tocopherol-injected rats to clarify, which mechanisms are important for vitamin E metabolism and are up-regulated by excess liver α-tocopherol. Vitamin E injections were used as these increase liver α-tocopherol rapidly and more than 10-fold in less than one week [3, 29].
The vitamin E solutions used for injection included: Vital E-300 (300 IU RRR-α-tocopherol/ml compounded with 20% ethanol and 1% benzyl alcohol) in an emulsified base (Schering-Plough Animal Health, Union, NJ, USA) or EmcellE (Stuart Products, Inc. Bedford, TX, USA), which is a clear, micellized, water-dispersible, liquid solution of RRR-α-tocopherol (500 IU/ml). Dr. Rob Stuart of Stuart Products generously provided the placebos for both the Vital E and the EmcellE formulations.
HPLC-grade methanol, hexane, ethanol, and glacial acetic acid were obtained from Fisher (Fair Lawn, NJ, USA). All other chemicals were obtained at reagent grade quality from suppliers.
The Oregon State University Animal Care and Use Committee approved animal protocols. Male Sprague-Dawley rats (Harlan, Indianapolis, IN), 275–300 g, were maintained on a 12-h light/dark cycle; housed two animals per cage; and fed Harlan Teklad Rodent Diet 8604 (Harlan Teklad, Madison, Wisconsin, USA) and water ad libitum. At least two weeks after arrival, rats received daily injections (intraperitoneal or subcutaneous, as indicated) of either vehicle (saline or placebo, as indicated) or α-tocopherol (100 mg/kg body wt, as either vital E or EmcellE, as indicated) up to 9 days. The vitamin E solutions were diluted with sterile saline, filtered (0.45 micron sterile filter, Micron Separations Inc., Westborough, MA) and the α-tocopherol concentrations determined prior to injection. A volume containing α-tocopherol equal to 100 mg/kg body weight was administered to each rat. Thus, there were no differences in the amounts of vitamin E administered as a result of the two sources of vitamin E (Vital E or EmcellE); however, there may be differences in quantity or form of the emulsifiers and/or alcohols used to solubilize the vitamin E. Volumes of the vehicle matched that of the vitamin E injections. Animals were weighed on injection days 1, 4 and 7 (or 8, as indicated). With regard to timing, we previously reported  using subcutaneous vitamin E injections in rats that both CYPs and transporters are up-regulated by 7 days.
After a 12 h fast and approximately 21 h after their last injection, rats were euthanized by intraperitoneal injection of sodium pentobarbital (65 mg/kg). Blood was collected in EDTA; plasma was obtained by centrifugation and stored at −80°C. Tissues were perfused with 0.9% saline (containing 2 U/ml heparin) using a perfusion catheter inserted into the heart. Tissues were excised and aliquots frozen in liquid N2 and stored at −80°C.
Plasma and tissue α- and γ-tocopherols were measured, as previously described [9, 30]. Liver CEHCs were extracted, following addition of an internal standard (trolox) . Extracts were hydrolyzed using enzyme (1 mg β-glucuronidase, type H-1 (has both glucuronidase and sulfatase activities) Sigma-Aldrich, St. Louis, MO, in 100 μL 10 mM potassium phosphate buffer, pH 6.8) for 30 min at 37°C, then samples were acidified, extracted (5 mL diethyl ether), and an ether aliquot dried and resuspended in 1:1 H2O:methanol containing 0.05% acetic acid and 0.05% wt:vol ascorbic acid. CEHCs and CMBHCs were detected using a high pressure liquid chromatography/mass spectrometer with an electrospray ionization source in negative single ion recording (SIR) mode, as described by us .
All the following steps were carried out on ice with centrifugation steps at 4°C. Protease inhibitor cocktail (Santa Cruz Biotechnology, Santa Cruz, CA, USA or EMD Biosciences, La Jolla, CA, USA) was added to all buffers just before use. Microsomes were isolated by ultracentrifugation, as described [3, 31] and stored at −80°C until analyzed. For membrane preparation, tissues (<500 mg) were homogenized in buffer (10 mM potassium phosphate, 150 mM KCl, 1 mM EDTA, pH 7.4), using a Potter-Elvehjem-type homogenizer and an electric drill, and then centrifuged at 14,000 × g for 20 min. The resulting homogenate was centrifuged at 100,000 × g for 90 min at 4°C. The resulting pellet was resuspended in wash buffer (100 mM potassium pyrophosphate, 1 mM EDTA, pH 7.4), then centrifuged at 100,000 × g for 90 min. The pellet was suspended in resuspension buffer (100 mM potassium phosphate, 30% glycerol, 1 mM EDTA, pH 7.3) and stored at −80°C until analyzed.
Antibodies were obtained as follows: anti-rat CYP3A2 and CYP3A1 (Millipore, Billerica, MA, USA); Sult2a1 and anti-β-actin (Santa Cruz Biotechnology, Inc, Santa Cruz, CA, USA or Merck Temecula, CA, USA), anti-human ABCB1 (MDR1, C219, Signet, Dedham, MA, USA). We previously documented that anti-human ABCB1 cross reacts with the rat protein . Western blot analyses were carried out using microsome or membrane (MDR1) preparations, as previously described .
Freshly isolated liver samples were placed in RNAlater® (Ambion, Austin, TX, USA) and stored overnight, according to manufacturer’s instructions. RNA was extracted from liver samples (~15 mg) using the RNeasy Mini Kit (Qiagen, Valencia, CA), according to manufacturer’s instructions. RNA quantity was determined by spectroscopy (NanoDrop ND-1000 UV-Vis Spectrophotometer, Thermo Scientific, Wilmington, DE). cDNA was generated from DNase treated RNA using Superscript III First-strand Synthesis System for RT-PCR (Invitrogen, Carlsbad, CA), according to manufacturer’s instructions.
As noted in Table 1, primers were either purchased from RT2 Primers (SABiosciences, formerly SuperArray, Frederick, MD, USA) or designed using Primer-BLAST (www.ncbi.nlm.nih.gov). RT-qPCR was performed using the Opticon 2 thermocycler (Bio-Rad, Hercules, CA). Detection of amplification products was done using RT2 Real-Time™ SYBR Green PCR Master Mix for RT2 primers (SABiosciences) or SYBR Green PCR Master Mix (Invitrogen, Carlsbad, CA) for self-designed primers. For housekeeping genes, beta-actin (Actb) and beta-2 microglobulin (B2M) RT2 primers were purchased (SABiosciences), while glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and hypoxanthine phosphoribosyltransferase 1(HPRT1) were self-designed. All housekeeping genes gave similar results; the housekeeping gene used for the various experiments are indicated in the legends to the figures and tables. All genes were run in triplicate; the average transcript expression for each gene of interest was determined for each rat and normalized to the average transcript expression of the respective housekeeping gene. The fold-change in transcript expression for each gene was determined in E-injected rats relative to control rats using the 2−ΔΔCt method and standard propagation of error methods as described . Many of the genes are published under multiple names (Table 1), so it should be noted that gene expression for CYP3A23/3A1 is called CYP3A1; CYP3A11 (mouse) is called CYP3A2 (cytochrome P450, family 3, subfamily a, polypeptide 2 [Rattus norvegicus]); Nr1i2 is called PXR; Nr1i3 is called CAR; and MDR1 is called ABCB1b. Primers to both Abcb1b (ATP-binding cassette, subfamily B (MDR/TAP), member 1B [Rattus norvegicus]) and Abcb1a (ATP-binding cassette, sub-family B (MDR/TAP), member 1A [Rattus norvegicus]) were tested, but only Abcb1b changed markedly in response to vitamin E injections; therefore in all but the Vital E experiment, only changes in Abcb1b mRNA were measured.
Statistical analysis was performed using Prism version 4.0 (Graphpad Software, San Diego, CA) or JMP Statistical software (SAS, Cary, NC). Data were log-transformed to equalize variances, as necessary, and were analyzed by one- or two-way ANOVA, as appropriate. Post hoc tests were performed using Tukey’s Multiple Comparisons when overall group effects were found to be significant, e.g. P-value < 0.05. Data are expressed as mean ± SD.
Previously, subcutaneous injections of Vital E were used to rapidly increase liver α-tocopherol concentrations [3, 9], but these injections caused some inflammation and skin thickening; therefore, an alternative injection route, intraperitoneal injection of Vital E, was investigated.
On day 1, the body weights of the rats were 354±27 g. By day 8, the body weights of the Vital E injected group had decreased to 303±23 g, while the control group decreased to 346±22 g (n=5/group). Although there was a significant interaction of the injections with time (p<0.0001), the effect of the vitamin E injections on body weights was not statistically significant.
Vital E intraperitoneal injections increased (p <0.0001) plasma, liver and extra-hepatic tissue α-tocopherol concentrations. Plasma α-tocopherol concentrations were 75 ± 21 μmol/L compared with 12 ± 2 in saline injected animals. Notably, in response to the intraperitoneal vitamin E injections, liver α-tocopherol concentrations increased nearly 20-fold, while the lung, kidney and adrenal gland only increased 2.5 to 3-fold (Figure 1). Because the vital E contains α-tocopherol from plant sources, γ-tocopherol is present, which also increased liver γ-tocopherol concentrations, as compared with the saline injected animals (Table 3). These data are consistent with the regulation of plasma and tissue concentrations by the liver and confirm the findings previously reported using subcutaneously injected vital E [3, 9].
As previously noted using subcutaneous injections , Vital E vitamin E intraperitoneal injections increased hepatic membrane protein, ABCB1 (P<0.0005, Figure 2A). In contrast, microsomal CYP3A2 protein was unchanged (Figure 2B) and a phase II enzyme, SULT2A1 protein (Figure 2C), decreased (P=0.0119) in the vital E injected relative to the saline injected controls.
To evaluate whether increased liver α-tocopherol concentrations up-regulated genes involved in xenobiotic metabolism, RT-qPCR was performed. Of particular interest was to determine whether genes for nuclear receptors were modulated. Therefore, mRNA levels were determined for PXR, CAR and PPARγ. Although both PXR and CAR have been identified as regulators of CYP3A  and SULTs [28, 34] and α-tocopherol as a PPARγ ligand , only PXR mRNA from the vitamin E treated rats was increased ~1.5-fold relative to the controls (p=0.04, Table 2).
A constellation of genes was evaluated by RT-qPCR, including representatives of phase I (cytochrome P450 enzymes), phase II (conjugation enzymes) and phase III (transporters) (Table 2). In vital E injected animals, CYP3A1 mRNA doubled (P<0.02), ABCB1b mRNA increased 3.6-fold (P<0.01), while the mRNA levels of CYP2B2, CYP3A2, GSTM3, SULT2A1, SULT2B1, and ABCB1a were not altered significantly by the vital E relative to saline injections (Table 2).
Previously saline was injected into the control animals because an appropriate vehicle control could not be obtained. We, therefore, switched to a different vitamin E solution, EmcellE, because the manufacturer provided a placebo formulation (and was subsequently able to provide a vital E placebo). To compare this new vitamin E solution to our previous experiments, pilot trial 1 was carried out.
On day 1, the body weights of the rats were 326±11 g. By day 7, the body weights of the EmcellE vitamin E injected group had decreased to 301±3 g, while the placebo group weighed 346±19 g (n=3/group). Although there was a significant interaction of the injection types over time on body weights (p=0.005), the effect of the vitamin E injection compared with placebo on body weights did not reach statistical significance at day 7. Similarly, to Vital E, there was thickening of skin and some inflammation at the injection sites.
The EmcellE, like Vital E, injections dramatically increased hepatic liver α- and γ-tocopherol concentrations (Table 3).
EmcellE subcutaneous injections decreased CYP3A1 protein levels (0.2±0.1) five-fold compared with the placebo (1.0 ± 0.2, p=0.005, relative to β-actin, with the mean of the placebo group set at 1). With regard to gene changes (Table 4), EmcellE injections decreased CYP3A1 mRNA to one third relative to the placebo (p=0.012), while ABCB1b mRNA increased nearly 5-fold (p=0.019). No changes in PXR were noted with EmcellE injections (Table 4). These outcomes were unexpected and largely different from our previous studies with Vital E. The reason for the different outcomes of these two experiments with different injectable forms of vitamin E was then sought.
Given that previously we had compared the vitamin E injected to saline injected animals, a second pilot trial was carried out using both EmcellE placebo and saline injected animals (pilot trial 2). On day 1, the body weights of the rats were 320±7 g. By day 7, the body weights of the EmcellE vitamin E injected group had decreased to 303±11 g, while the placebo group weighed 350±9 g and the saline injected group weighed 343±10 g (interaction of injections over time p<0.0001, n=4/group). The saline and placebo groups were not significantly different from each other, while the effect of the EmcellE injections was to decrease body weights relative to either the saline and placebo injections on day 7 (p<0.01 for each comparison).
With regard to CYP3A2 protein levels, the saline and the placebo groups were not significantly different from each other, while the EmcellE group showed a marked decrease in CYP3A2 protein relative to either the saline or placebo groups (main effect, p=0.0013, p<0.05 for each comparison). Similarly, the CYP3A1 mRNA in the EmcellE group was about one third of that in the saline or in the placebo groups (Figure 3B). The SULT2A1 mRNA and PXR mRNA were not significantly different between groups, although both were somewhat lower in the EmcellE injected group (Figure 3A and C). In marked contrast, ABCB1 mRNA was 10-times higher in the EmcellE injected group (Figure 3D).
To clarify the source of the discrepancy in the responses of the xenobiotic pathways to the various injections, 5 groups of rats (EmcellE, EmcellE placebo, Vital E, Vital E placebo, and saline) were injected subcutaneously with the indicated solution daily for 7 days. By day 7, the body weights of the two vitamin E injected groups decreased somewhat, while the placebo and saline groups did not (Figure 4, interaction of injections over time p<0.0001, n=4/group), but none of the paired comparisons reached statistical significance.
The vitamin E injections (EmcellE and Vital E) dramatically and similarly increased hepatic liver α- and γ-tocopherol concentrations relative to the placebo and saline conditions (Table 3).
With regard to CYP3A2 protein levels, the two vitamin E injected groups (EmcellE and vital E) showed a marked decrease in CYP3A2 protein relative to both placebo groups (main effect, p=0.0013, Figure 5D); the two placebo groups were not significantly different from the each other.
With regard to xenobiotic gene expression, a striking pattern was observed; the vitamin E injected animals had similarly low mRNA expression compared with the placebo groups and the saline injected animals, which had higher expression for the following genes: CYP3A1, CYP3A2, CYP2B2 and SULT2A1 (Figure 5A, B, C, and E). The PXR mRNA was significantly different between groups (main, effect, P=0.0028) with the lowest values in the vitamin E injected groups.
With regard to transporters, we examined multiple transporters because we had previously found that Vital E injections increased ABCB1 protein . Both vitamin E injections increased ABCB1b mRNA, with the EmcellE injection tripling the gene expression relative to the saline injected rats (Figure 6). It should be noted that in all of the experiments reported herein in rats injected with either vital E (Table 1) or EmcellE (Table 4), an increase in ABCB1b gene expression was observed.
Based on our observations that the ABCB1b transporter was consistently increased by the elevated vitamin E status, a survey of transporter genes was carried out. Most of the transporter genes tested did not vary greatly; ABCC1, ABCC2, and ABCC3 were increased significantly by the EmcellE injections, but not vital E, making it less likely that this is a vitamin E-dependent effect (Figure 7). However, genes for two transporters, ABCG2 (also known as BCRP, breast cancer receptor protein) and organic anion transporting polypeptide 2 (OATP), were markedly altered by both vitamin E injectables (Figure 8). ABCG2 mRNA increased by more than 7-fold in the livers from either EmcellE or Vital E injected rats compared with the saline injected animals (interaction p=0.0004, paired comparisons, p<0.05). In contrast, OATP mRNA decreased to about one third in the livers from either EmcellE or Vital E injected rats compared with the saline injected animals (interaction p<0.0001, paired comparisons, p<0.05).
Vitamin E metabolites were also measured in the livers of these rats (Figure 9). Both the concentrations of α-CEHC and its immediate precursor, α-CMBHC, increased more than 10-fold in the livers from rats injected either with EmcellE or Vital E compared with the saline or placebo injected animals. In contrast, the γ-CEHC concentrations decreased with α-tocopherol injections. Both α-tocopherol and its metabolites (α-CEHC and α-CMBHC) were highly correlated with both ABCG2 and ABCB1b mRNA expression, while they were negatively correlated with OATP mRNA expression (Table 5). The only paired comparison that did not reach statistical significance was the paired comparison of ABCB1b with α-CMBHC.
The purpose of this study was to characterize the xenobiotic pathways involved in vitamin E metabolism based on the premise that α-tocopherol triggers modulation of CYP enzymes and xenobiotic transporters through PXR. Previously, we reported that CYP3A and MDR1 proteins were up-regulated in rats subcutaneously injected with vital E compared with those injected with saline [3, 9].
The findings concerning the effect of excess hepatic α-tocopherol on ABC transporters were consistent across all experiments. In our previous studies [3, 9] and the ones reported herein, ABCB1b (MDR1 or p-glycoprotein), both the protein and the mRNA, increased with increased hepatic α-tocopherol. It, however, is not clear as to whether α-tocopherol, or its metabolite α-CEHC (or even γ-tocopherol, or γ-CEHC) were the stimulatory molecule(s) because all were well correlated with the gene changes observed (Table 5).
We also did a survey of various hepatic transporter genes and found that ABCG2 was up-regulated and OATP was down-regulated by vitamin E injections. ABCG2 and ABCB1 proteins are both localized to the liver canaliculus, where they are thought to be involved in efflux of xenobiotics from the liver into the bile . ABCG2 expression is not limited to the liver, but “has been increasingly recognized for its important role in the absorption, elimination, and tissue distribution of drugs and xenobiotics” . Both ABCG2 and ABCB1 transport lipophillic molecules, but ABCG2 also has been reported to transport sulfates and glucuronides , so may be important for trafficking CEHCs from the liver. In contrast, OATP is a transporter involved in influx to the liver from the blood stream [36, 38, 39]. Thus, a decrease in OATP expression might limit the hepatic re-uptake of CEHCs. Further experimentation is needed to identify the roles of these transporters in regulating liver α-tocopherol concentrations. Moreover, the regulation of the transporters, themselves, is not well understood. PXR is not the only nuclear receptor implicated in regulation of ABC-transporters; other nuclear receptors relevant for their expression include the liver X receptor, farnesoid X receptor, and peroxisome proliferator-activated receptors α and γ .
Although we confirm the up-regulation of ABCB1 protein and gene expression associated with increased hepatic α-tocopherol, we do not consistently find up-regulation of CYP3A protein or gene expression. It is generally accepted that PXR activates CYP3A gene expression . Given that α-tocopherol had been demonstrated to bind to PXR  and that in mice dietary α-tocopherol could up-regulate mouse CYP3a11 genes [4, 25], it seemed plausible that α-tocopherol would increase hepatic CYP3A in rats. However, our previous studies lacked an appropriate placebo; saline was used as the vehicle. For the studies described herein, we have obtained placebo solutions that contain all of the components, with the exception of vitamin E. When gene expression is compared between the placebo- and the saline only-injected animals only Sult2A1 and PXR expression were the same for saline and both placebos (Figure 5); the two placebos had similar effects, except on CYP2B2 expression, which was less for the Vital E placebo. CYP3A2 protein was higher in the livers from rats injected with Vital E placebo compared with saline-injected, suggesting that there may be a component in the emulsifier that stimulates CYP3A2 protein expression.
The differences between the experiments previously published and the ones presented herein are difficult to reconcile. Despite our best efforts to mimic the previous conditions, there may be a component that was in the vital E solution that is no longer present. Other variables could include the amount of sodium pentobarbital administered, the degree of fasting, or there may be some additional differences in animals or diets, or some other unknown factors.
Both rat CYP3A1 and 3A2 are members of the CYP3A superfamily and are ~70% identical to human CYP3A4 . In the experiments presented herein, high liver α-tocopherol concentrations were associated with decreased CYP3A protein and mRNA expression; both CYP3A1 and 3A2 were tested to insure that we did not overlook their up-regulation. However, our findings are consistent with reports that α-tocopherol metabolism decreased PXR-mediated responses in mice . Moreover, in humans CYP3A does not appear to increase. The role of PXR in regulating CYP3A has led to the expectations that vitamin E supplements could dysregulate pharmaceutical drug metabolism [43, 44]. Three studies to test this hypothesis in humans have not demonstrated decreased drug efficacy. When Leonard et al  gave vitamin E supplements to hypercholesterolemic patients, the supplements did not alter efficacy of either simvastatin or lovastatin, two drugs metabolized via a CYP3A-dependent mechanism. Werba et al.  approached this problem by studying the effect of simvastatin (20 mg/day) or pravastatin 40 (mg/day) on α- and γ-tocopherol concentrations in hypercholesterolemic humans. Although both drugs decreased circulating cholesterol and α-tocopherol concentrations, only simvastatin raised γ-tocopherol concentrations. These findings suggest that simvastatin and γ-tocopherol competed for the same sites for metabolism, while pravastatin, which is not dependent on CYP3A for metabolism, did not. Clarke et al  hypothesized that α-tocopherol supplements would induce CYP3A4 in humans and thus decrease the plasma concentration of the CYP3A4 substrate, midazolam. Although they found increases in both plasma α-tocopherol concentrations and CEHC excretion, no changes in midazolam concentrations were detected. Taken together, these findings suggest that stimulation of PXR- or CYP3A-dependent metabolism tends to decrease, rather than increase, vitamin E metabolism. These outcomes are contrary to the expectation that vitamin E functioning as a PXR-ligand would increase PXR-responsive events. However, these reports are not consistent with the finding that α-tocopherol increased PXR-dependent increases in CYP3A and lead to increased vitamin E metabolism, as shown in a cell culture model using D-galactosamine-treated human hepatocytes . Thus, we are left with the quandary that there are no clear outcomes as to how and whether α-tocopherol alters CYP3A or PXR. Certainly, the findings from the experiments reported herein show that if PXR gene expression is decreased CYP3A expression is also decreased or unchanged (Table 6); however, these changes were not dependent on the liver α-tocopherol concentrations, suggesting that PXR was dependent upon some other factor.
We anticipated that rat hepatic SULT2A expression would also be PXR-dependent; however, this was not the case (Table 6). The transcriptional control of rodent SULT2A is not solely dependent upon regulation by PXR. Motifs required for transcriptional activation by PXR and other nuclear receptors, including constitutive androstane receptor, farnesoid X receptor and vitamin D receptor have been reported to be located in the 5′-flanking regions of rodent SULT2A genes [49, 50]. Thus, the lack of SULT2A regulation by α-tocopherol is not surprising, given the variety of possible stimulatory molecules.
In summary, our studies have demonstrated that an over-load of hepatic α-tocopherol increases its own metabolism, and increases transporters that are postulated to lead to increased excretion of both vitamin E and its metabolites. Based on the work of Parker’s group [7, 8, 51] and the studies reported herein, it is clear that CYP4F2 is the only cytochrome P450 that initiates tocopherol metabolism. However, CYP4F2 also functions as a predominant leukotriene B4 and arachidonate omega-hydroxylase , as well as being involved in vitamin K metabolism ; thus is not specific only for vitamin E. Additionally, we have found that CYP4F does not change with hepatic α-tocopherol status . Thus, it would appear that vitamin E metabolism is largely dependent on the affinity of CYP4F2 for its substrates, as demonstrated by Sontag and Parker . The studies described herein emphasize that ABC-transporters, ABCB1 and ABCG2 are up-regulated in association with elevated hepatic α-tocopherol, suggesting they are involved in the disposition of either or both α-tocopherol and α-CEHC. Further studies examining the activity of these transporters are needed to confirm these observations.
Neha Patel provided excellent technical assistance. We acknowledge funding from NIH: National Institutes of Diabetes and Digestive and Kidney Disease and Office of Dietary Supplements (DK067930).
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.