Isolation and identification of α-CEHC sulfate in rat urine and an improved method for the determination of conjugated α-CEHC

Article abstract of DOI:10.1021/jf802459d

2,5,7,8-Tetramethyl-2-(2′-carboxyethyl)-6-hydroxychroman (α-CEHC), the water-soluble metabolite of α-tocopherol (α-TOH) with a shortened side chain but an intact hydroxychroman structure, has been identified in human urine and are thought to be produced in significant amount at excess intake of α-TOH. In previous studies, CEHCs in biological specimens were measured by HPLC, GC-MS or LC-MS, preceded by a hydrolysis procedure using either enzyme or methanolic HCl. In an attempt to analyze α-CEHC in rat urine accordingly, we observed that enzyme hydrolysis was relatively inefficient in releasing α-CEHC compared to high concentrations of HCl. The HCl releasable α-CEHC conjugate was isolated and chemically identified as 6-O-sulfated α-CEHC (α-CEHC sulfate). Using the synthetic α-CEHC sulfate standard, it was found that sulfatase could not hydrolyze to a significant extent. On the other hand, pretreatment with HCl at 60°C in the presence of ascorbate, followed by a one-step ether extraction, not only hydrolyzed the sulfate conjugate completely but also extracted α-CEHC with high recovery. The inclusion of ascorbate minimized the conversion of α-CEHC to α-tocopheronolactone in the HCl pretreatment. A complete procedure for the quantitative analysis of α-CEHC including HCl hydrolysis, ether extraction and reverse phase isocratic HPLC-ECD was thus established. In conclusion, α-CEHC sulfate was isolated and identified as the HCl-releasable conjugate of α-CEHC in rat urine. A rapid and sensitive method with high reproducibility for the determination of free, conjugated and total α-CEHC is then established.

Full text of DOI:10.1021/jf802459d

J. Agric. Food Chem. 2008, 56, 11105–11113 11105
Isolation and Identification of r-CEHC Sulfate in Rat
Urine and an Improved Method for the Determination
of Conjugated r-CEHC
YI-JEN LI,† SHENG-CHING LUO,† YI-JING LEE,† FU-JUNG LIN,†
CHI-CHENG CHENG,† YUNG-SHUNG WEIN,‡ YUEH-HSIUNG KUO,‡,§,|,⊥ AND
CHING-JANG HUANG*,†,#
Division of Nutritional Biochemistry, Institute of Microbiology and Biochemistry, National Taiwan
University, 1, Sec. 4, Roosevelt Rd., Taipei, 106, Taiwan, Department of Chemistry, National Taiwan
University, Taipei, Taiwan, Department of Biochemical Science and Technology, National
Taiwan University, Taipei, Taiwan, College of Pharmacy, China Medical University, Taichung, 404,
Taiwan, and Agricultural Biotechnology Research Center, Academia Sinica, Taipei, 115, Taiwan
2,5,7,8-Tetramethyl-2-(2′-carboxyethyl)-6-hydroxychroman (R-CEHC), the water-soluble metabolite
of R-tocopherol (R-TOH) with a shortened side chain but an intact hydroxychroman structure, has
been identified in human urine and are thought to be produced in significant amount at excess intake
of R-TOH. In previous studies, CEHCs in biological specimens were measured by HPLC, GC-MS
or LC-MS, preceded by a hydrolysis procedure using either enzyme or methanolic HCl. In an attempt
to analyze R-CEHC in rat urine accordingly, we observed that enzyme hydrolysis was relatively
inefficient in releasing R-CEHC compared to high concentrations of HCl. The HCl releasable R-CEHC
conjugate was isolated and chemically identified as 6-O-sulfated R-CEHC (R-CEHC sulfate). Using
the synthetic R-CEHC sulfate standard, it was found that sulfatase could not hydrolyze to a significant
extent. On the other hand, pretreatment with HCl at 60 °C in the presence of ascorbate, followed by
a one-step ether extraction, not only hydrolyzed the sulfate conjugate completely but also extract-
ed R-CEHC with high recovery. The inclusion of ascorbate minimized the conversion of R-CEHC to
R-tocopheronolactone in the HCl pretreatment. A complete procedure for the quantitative analysis of
R-CEHC including HCl hydrolysis, ether extraction and reverse phase isocratic HPLC-ECD was
thus established. In conclusion, R-CEHC sulfate was isolated and identified as the HCl-releasable
conjugate of R-CEHC in rat urine. A rapid and sensitive method with high reproducibility for the
determination of free, conjugated and total R-CEHC is then established.
KEYWORDS: Vitamin E metabolism; r-TOH; r-CEHC; sulfate conjugate; HPLC-ECD
INTRODUCTION
of R-TOH was thought to start with a radical attack on the
chroman structure resulting in the formation of R-tocopherylquino-
ne. The subsequent side chain degradation then leads to final
products, R-tocopheronic acid and the lactone derived therefrom
(the so-called Simon’s metabolites), that are excreted in urine
(3). More than a decade ago, 2,5,7,8-tetramethyl-2-(2′-carboxy-
ethyl)-6-hydroxychroman (R-CEHC), the water-soluble me-
tabolite of R-TOH with a shortened side chain but an intact
hydroxychroman structure, has been discovered in human
urine. Its chroman ring is intact, indicating that the vitamin
has not been subjected to the presumed radical scavenging
reaction in this metabolic pathway. Furthermore, this me-
tabolite is reported to be produced in significant amount at
excess intake of R-TOH (4).
Vitamin E, and in particular R-tocopherol (R-TOH) (Figure
1), is the major fat soluble antioxidant in the body, protecting
cellular membranes and other lipids against oxidative damage
caused by oxygen-derived free radicals (1). R-TOH is the most
abundant form in the body accounting for over 90% of the total
vitamin E retained, even though γ-tocopherol is generally the
most abundant form in the diet (2). For a long time, catabolism
* Corresponding author. Tel: +886-2-33662276. Fax: +886-2-
23621301. E-mail: cjjhuang@ntu.edu.tw.
† Institute of Microbiology and Biochemistry, National Taiwan
University.
‡ Department of Chemistry, National Taiwan University.
§ Equal contribution with corresponding author.
| China Medical University.
To date, most vitamin E isoforms are shown to be catabolized
to the corresponding CEHC (4-7). These CEHCs in biological
specimens have been measured by GC-MS (4), LC-MS (8),
⊥ Academia Sinica.
# Department of Biochemical Science and Technology, National
Taiwan University.
10.1021/jf802459d CCC: $40.75  2008 American Chemical Society
Published on Web 11/08/2008

11106 J. Agric. Food Chem., Vol. 56, No. 23, 2008
Li et al.
HPLC-ECD (9, 10) or HPLC-fluorescence (11, 12) with a
prior hydrolysis and extraction process. The hydrolysis step used
either acid methylation or enzyme hydrolysis to yield either the
methyl ester (10) or free metabolite (4, 9). Acid hydrolysis in
aqueous phase has rarely been employed since Schultz et al.
(4) demonstrated that R-tocopheronolactone could be artificially
formed from R-CEHC during sample handling, while Pope et
al. (13) showed evidence of possible in vivo formation of
R-tocopheronolactone. Acid methylation, on the other hand, can
protect R-CEHC from conversion to R-tocopheronolactone (10).
There is disagreement as to whether the CEHCs are excreted
as glucuronides or sulfates even though ꢀ-glucuronidase is the
enzyme most commonly used for the hydrolysis of the
conjugates (14-17). Schultz et al. (4) could not obtain free
R-CEHC while hydrolyzing human urine with pure glucu-
ronidase (without sulfatase activity) at pH 4.5. They concluded
that R-CEHC sulfate but not glucuronide was the predominant
conjugate present in human urine. Tanabe et al. (18) observed
that when γ-CEHC was administered to rats, it was excreted
into urine as 6-O-sulfated γ-CEHC identified by using an HPLC
tandem mass spectrometry (MS-MS). On the other hand, the
procedure of Morinobu et al. (19) used ꢀ-glucuronidase without
sulfatase activity since no difference in the amount of CEHC
was detected between using the enzyme with and without
sulfatase activity. They thus suggested that the R-CEHC in
human urine is predominantly glucuronide conjugate. Using
synthetic R-CEHC glucuronide and R-CEHC sulfate standards,
Pope et al. (20) were able to detect these conjugates directly
(without prior hydrolysis) in human urine by tandem mass
spectrometry (ESI-MS-MS). Peaks with m/z of 453 and 357
were respectively designated as glucuronide and sulfate con-
jugates of R-CEHC or R-tocopheronolactone. However, they
found a comparable collision induced dissociation (CID) spec-
trum of urine extract to standard compound for R-CEHC
glucuronide, but not for R-CEHC sulfate, presumably because
of the relatively small amount. The controversy indicates that
the method for the determination of conjugated R-CEHC needs
further investigation.
Figure 1. The chemical structure of R-tocopherol, R-CEHC and its
associated compounds. (A) R-tocopherol, (B) R-CEHC, (C) R-CEHC glu-
curonide, (D) R-CEHC sulfate, (E) R-tocopheronolactone (benzoquinone)
and (F) R-CEHC-Me.
experiment was carried out under the guidelines of the care and use of
laboratory animal committee of NTU. Rats were individually housed
in metabolic cages, and urine was collected into the collection tube on
ice. All urine samples were stored in -20 °C under N2 before the
analysis.
Hydrolysis of r-CEHC Conjugates. Tested samples (rat urine or
standard compound solutions) contained 20 mg/mL of ascorbate except
for those that were tested for the protection of ascorbate. For the analysis
of free R-CEHC in the rat urine, samples were adjusted to pH 4 by
adding ascorbate and extracted by diethyl ether directly without prior
enzyme or HCl hydrolysis.
ꢀ-Glucuronidase and Sulfatase Hydrolysis. The procedures were
modified from those reported previously (7, 9). Prior to extraction, 1
mL tested samples were hydrolyzed by adding 200 µL of enzyme
solution (1000 U of ꢀ-glucuronidase or 50 U of sulfatase in 0.1 M
sodium acetate buffer, pH 4.5) and incubated for 16 or 2 h respectively
at 37 °C. After cooling on ice, the samples were extracted with diethyl
ether by mixing thoroughly and the mixtures were centrifuged at 2000
rpm for 5 min to separate the layers. An aliquot of the ether layer was
dried under vacuum, and the residue was reconstituted in mobile phase
for being analyzed using HPLC (see below). To validate the hydrolysis
procedure, samples were also incubated with various amounts of
enzyme.
HCl Hydrolysis. 0 to 12 N HCl (1 mL) was added to 1 mL tested
sample for acid hydrolysis with or without incubating in a 60 °C water
bath for 1 h under N2, extracted with diethyl ether and then analyzed
by HPLC as described above. The R-tocopheronolactone was also
extracted together with R-CEHC on this procedure.
Acid Methylation/Hydrolysis. The conjugated R-CEHC was hy-
drolyzed by acid methylation according to the method of Kiyose et al.
(10). The tested sample (1 mL, containing 20 mg ascorbate) was
lyophilized and added with 2 mL of 3 or 6 N methanolic HCl, shaking
at 60 °C for 1 h under N2. After methylation, samples were cooled on
ice, 6 mL of water was then added to each sample and extracted with
8 mL of n-hexane by shaking vigorously for 1 min. This mixture was
centrifuged at 3000 rpm for 5 min, and the upper layer was collected
and evaporated. The residue was dissolved in mobile phase which was
composed of acetonitrile (CH3CN)/H2O (43:57, v/v) containing 50 mM
ammonium acetate (pH 4.5) for HPLC analysis.
HPLC Analysis. For the analysis of R-CEHC and methyl esters of
R-CEHC (R-CEHC-Me), the detector used was a EAS coulochem II
5200A electrochemical detector. For the analysis of R-tocopherono-
lactone, a Jasco 870-UV/vis detector (870-UV) was used. The peaks
were identified by the standard compound, and concentrations of
samples were calculated according to the external standard mode.
In an attempt to measure R-CEHC in rat urine according to
reported methods (7, 9), we observed that ꢀ-glucuronidase with
sulfatase activity only released a very small amount of R-CEHC
whereas HCl treatment resulted in extremely high amount. In
the present study, therefore, we first isolated and identified
R-CEHC sulfate as the HCl releasable conjugate in rat urine.
Using the synthetic R-CEHC sulfate standard, the acid hydroly-
sis condition was optimized to achieve complete hydrolysis with
minimal conversion of R-tocopheronolactone. A rapid and
sensitive method with high reproducibility for the determination
of free, conjugated and total R-CEHC by isocratic HPLC was
thus established.
MATERIALS AND METHODS
Chemicals and Reagents. R-CEHC, R-tocopheronolactone and
R-CEHC sulfate were synthesized by a procedure similar to that reported
by Pope et al. (20). The details are described in Supporting Information
(Schemes S1-S3). ꢀ-Glucuronidase (type H-1, contains minimum
300,000 U/g ꢀ-glucuronidase activity and minimum 10,000 U/g
sulfatase activity) and sulfatase (Type H-1, contains minimum 10,000
U/g sulfatase activity) were purchased from Sigma Chemical Co. All
remaining chemicals and reagents were of the highest purity available
or HPLC-grade.
Urine Sample Preparation. Rat urine was obtained from normal
Wistar rats (purchased from the Laboratory Animal center of College
of Medicine, National Taiwan University (NTU)) fed modified AIN-
76 diets containing 50 or 500 mg/kg all-rac-R-tocopheryl acetate. The

Acid Hydrolysis of R-CEHC Sulfate
J. Agric. Food Chem., Vol. 56, No. 23, 2008 11107
Figure 2. Negative ion ESI-MS spectra of 6-O-sulfated R-CEHC. The parent ion is m/z 357. The major mass fragments were m/z 357, 277 and 243;
these fragments were assigned as shown in the figure.
r-CEHC. The R-CEHC was analyzed by HPLC-ECD using a RP18
5 µm column (Keystone BetaBasic). The mobile phase was methanol
(MeOH)/H2O (43:57, v/v) containing 50 mM sodium acetate (pH 4.5)
at a flow rate of 1 mL/min. The guard cell was set to 250 mV. Detection
and quantification of the R-CEHC was carried out with electrochemical
detection operating at an applied voltage of 150 mV, which was found
to be optimal (Figure S1 in the Supporting Information; data was
described in the Supporting Information). The retention time was 22-23
min for R-CEHC. The concentration was calculated by intrapolating
from an external R-CEHC calibration curve, and the variation was
recorded to assess the reproducibility. Stock solution of R-CEHC were
prepared in MeOH and stored at -20 °C. Working standard were
prepared and calibrated daily according to its absorbance values ε )
3230 cm-1 × (mol/L)-1 at 289 nm.
H2O and then MeOH/H2O gradient solvent system from 0/100 to 100/0
(v/v). For a rapid screening of the target fractions, collected fractions
were tracked by thin-layer chromatography (TLC) after HCl hydrolysis
and R-CEHC standard was used as the reference. The fractions
containing HCl hydrolyzable R-CEHC were combined and further
purified by a preparative HPLC using a RP-18 column (Phenomenex
Luna 5 µ C18(2), 250 mm × 10 mm) eluting with a 50% MeOH/H2O
(v/v) at a flow rate of 2 mL/min. Four fractions (fractions A-D) were
obtained and screened by TLC. The fraction D was further subjected
to the preparative chromatography again, eluted with MeOH/H2O (55:
45, v/v) containing 50 mM ammonium acetate (pH 4.5) and a total of
13 fractions (fractions D1-D13) were obtained. The fraction D9 was
then separated by HPLC with a RP-18 column eluted with MeOH/
H2O (43:57, v/v) containing 50 mM ammonium acetate (pH 4.5) at a
flow rate of 1 mL/min and detected by UV/vis detector operating at
289 nm and four fractions were collected (fractions D9-1 to D9-4).
Finally, the conjugated metabolite of R-CEHC was isolated from the
fraction D9-4 (its retention time was 7-8 min). After desalting, the
conjugated metabolite of R-CEHC was identified by NMR, MS and
IR spectroscopy using a Bruker 400 spectrometer, a Finnigan TSQ-
46C mass spectrometer and a Bio-Rad FTS-40 spectrophotometer
respectively. All spectra are shown in the Supporting Information
(Figure S6).
r-CEHC-Me. The HPLC-ECD analysis of R-CEHC-Me was
similar to that of R-CEHC described above. The retention time of
R-CEHC-Me is 13 min in this system. The concentration was calculated
by intrapolating from a calibration curve of R-CEHC-Me prepared
according to the method of Kiyose et al. (10). Pure 25, 50, 100, 200
ng of R-CEHC were methylated with methanolic HCl under N2. After
methylation, R-CEHC-Me standards were extracted from the reaction
mixture by the procedure for extracting this compound from testing
samples described above. The extracted R-CEHC-Me standards were
analyzed by HPLC-ECD to establish the calibration curves of
R-CEHC-Me.
r-Tocopheronolactone. The R-tocopheronolactone was analyzed by
HPLC-UV using a RP18 5 µm column (Merck Lichrospher), and the
mobile phase was the same as those for R-CEHC analysis. R-Toco-
pheronolactone was detected by UV absorption at 268 nm. Stock
solutions of R-tocopheronolactone were prepared in MeOH and stored
at -20 °C. Working standards were prepared daily, and the concentra-
tion was calibrated by the absorbance values at 268 nm based on ε )
22553 cm-1 × (mol/L)-1. The retention time was 23 min for R-toco-
pheronolactone. R-CEHC could also be detected by this HPLC-UV
at 289 nm, and its retention time was 21 min, despite with much lower
sensitivity.
Statistical Analysis of Data. Data are expressed as means ( SD.
The statistical significance of the content of R-CEHC in the samples
obtained from different hydrolysis methods was analyzed by oneway
ANOVA (analysis of variance) and Duncan’s multiple range test using
SAS software (SAS 9.0, Cary, NC). Results were considered to be
significant at the 95% confidence level (p < 0.05).
RESULTS
HPLC-ECD for r-CEHC Analysis. Our HPLC-ECD
method for R-CEHC analysis was modified from that reported
by Lodge et al. (9) in that an isocratic mobile phase was used
instead of the gradient system and the applied potential was
modified to 150 mV. Using this system, a sharp peak of
R-CEHC was obtained on the chromatograms for both the
standard compound and the extracts of urine samples. A typical
Isolation and Identification of r-CEHC Conjugate. Rat urine was
lyophilized and then extracted with MeOH for deproteinization and
desalting. The crude extract was separated by chromatography using
an open column packed with RP-18 silica gel, eluting with a CH3CN/

11108 J. Agric. Food Chem., Vol. 56, No. 23, 2008
Li et al.
1
Table 1. 13C NMR and H NMR Spectroscopic Data for R-CEHC, R-CEHC Sulfate Standard and R-CEHC Sulfate Isolated from Rat Urine (400 MHz, 100
MHz in CD3OD)
R-CEHC standard calibration curve was linear from 50 ng/mL
to 800 ng/mL (3.6-57.6 pmole, injected volume is 20 µL)
(Figure S2 in the Supporting Information; data was shown in
the Supporting Information). The method is thus applicable for
samples containing a low level of R-CEHC in biological
specimens. The detection limit was 0.45 pmol (125 pg) with a
signal-to-noise ratio (s/n) at 2.45. The coefficient of variation
(CV) of the slope of the standard calibration curve is lower
than 5%.
Identification of the HCl Releasable r-CEHC Conjugate
in the Rat Urine. To identify the HCl releasable R-CEHC
conjugate, rat urine samples were freeze-dried, extracted and
separated by flash column chromatography and preparative HPLC
(Figure S5 in the Supporting Information; the isolated flowchart
and chromatograms were shown in the Supporting Information).
The fraction D9-4 contained the target compound that can release
R-CEHC after HCl hydrolysis. As shown in Figure 2, the electron
impact mass spectrum indicated that the R-CEHC conjugate has
an exact molecular weight of 357 and its major fragments are m/z
243 and 277. The m/z 243 ion is resulted from a fragmentation at
the 3,4 carbon-carbon and oxygen-carbon bond at C2 of the
chroman ring. The fragment of m/z 277 corresponds to a loss of
80 unit which is a characteristic fragment of sulfur trioxide (SO3-).
The 1H and 13C NMR spectrum (Table 1) of the R-CEHC
conjugate shows chemical shifts indicating changes on the C6 of
the chroman ring. The fragment of m/z 357, 243, 277 in the MS
along with the similarity of the NMR chemical shifts to the
R-CEHC led to the proposed structure of the conjugated R-CEHC
in the rat urine is 6-O-sulfated R-CEHC (R-CEHC sulfate) which
best fit the formula C16H21O7S. The characteristic of sulfate (1232
and 1001 cm-1) and carboxylic acid (2700-3300 cm-1) in the
IR also support the chemical structure elucidated. This structure
was confirmed by the unambiguous synthesis of R-CEHC sulfate
from R-CEHC and trimethylamine-SO3 complex. The purity of
R-CEHC sulfate in the fraction D9-4 is about 86%. Accordingly,
the sulfate conjugate is identified as the compound in rat urine
that can release R-CEHC by HCl.
Enzyme Hydrolysis of Rat Urine Releases a Rather Small
Amount of r-CEHC Compared to HCl. While ꢀ-glucu-
ronidase was employed to hydrolyze the conjugated R-CEHC
in the rat urine samples as reported (7, 9), about 6-fold of
R-CEHC was detected as compared to the value without any
pretreatment (free R-CEHC) (2.17 ( 0.52 µM v.s. 0.33 ( 0.07
µM). Increasing the amount of enzyme could not increase the
amount of R-CEHC detected from the rat urine. In contrast,
adding HCl vastly increased the peak area at the retention time
of 22-23 min. By a calculation based on the calibration curve,
the peak area was estimated to be equivalent to 280-fold that
of the free R-CEHC (92.59 ( 1.04 µM vs 0.33 ( 0.07 µM).
To examine the chemical identity of the compound of this large
peak, a preparative HPLC was used to separate, isolate and
1
purify the compound. The identical H NMR spectrum to that
of the authentic R-CEHC standard indicated that the peak eluted
at 22-23 min after HCl pretreatment of rat urine is indeed
R-CEHC (Figure S4 in the Supporting Information; data was
described detailed in the Supporting Information).

Acid Hydrolysis of R-CEHC Sulfate
J. Agric. Food Chem., Vol. 56, No. 23, 2008 11109
Figure 3. The effect of various concentrations of HCl on the R-CEHC and R-tocopheronolactone measured from (A) rat urine, (B) R-CEHC sulfate
standard (78.6 µM) and (C) R-CEHC standard (89.9 µM) with or without ascorbate protection.
Establish the Procedure for the Analysis of r-CEHC in
Rat Urine by Acid Hydrolysis and HPLC-ECD. Various
concentrations of HCl were added to authentic R-CEHC sulfate
solution or rat urine for the determination of the amount of HCl
needed. The pH, R-CEHC and R-tocopheronolactone detected
are shown in Figure 3. A threshold HCl concentration is needed
for the release of R-CEHC from the rat urine and R-CEHC
sulfate solution. Above this threshold, the amount of released
R-CEHC increased dose-dependently with the increase of HCl
concentration. When the final concentration of HCl added
reached 3 N, the maximal amount of R-CEHC is released from
both rat urine and R-CEHC sulfate solution (Figure 3A, 3B).
As R-CEHC has been shown to be oxidized to R-tocopherono-
lactone (4), we also examined whether this can be protected by

11110 J. Agric. Food Chem., Vol. 56, No. 23, 2008
Li et al.
Figure 4. Comparison of various hydrolysis procedures for the analysis of R-CEHC in rat urine. Rat urine samples containing 20 mg/mL ascorbate to
adjust pH value to near 4 were collected from rats fed the AIV-76 modified diets containing 50 mg/kg (A) or 500 mg/kg diet (B) of all-rac-R-tocopheryl
acetate. Aliquots of urine samples were hydrolyzed by sulfatase, HCl (final concentration 3 or 6 N) at room temperature, HCl at 60 °C for one hour (final
concentration 3 or 6 N). In (B) aliquots of samples were also deconjugated by methanolic HCl (final concentration 3 or 6 N) at 60 °C which produce and
can be detected as R-CEHC-Me. Values are means ( SD of 2-8 independent experiments. Values not sharing a common letter are significantly
different from one another by oneway ANOVA and Duncan’s multiple range test (p < 0.05).
adding ascorbate as an antioxidant. In the presence of ascorbate,
more R-CEHC and less R-tocopheronolactone were detected
when the R-CEHC sulfate solutions were treated with high
concentrations of HCl (Figure 3B). The protective effect of
ascorbate in the HCl pretreatment was less apparent for rat urine
(Figure 3A), but most dramatic for the R-CEHC solution
(Figure 3C). In the latter case, the recovery of R-CEHC was
near 100% and 60%, respectively, in the presence and absence
of ascorbate. Therefore, it is demonstrated that ascorbate could
protect R-CEHC in HCl hydrolysis. It is interesting to note that
adding ascorbate lowered the pH of R-CEHC solution to about
4, which might render the R-CEHC existing as the protonated
form and could be extracted almost completely without adding
HCl (Figure 3C).
monitored while R-CEHC sulfate solution was hydrolyzed with
various conditions. As shown in Table 2, acid hydrolysis with
6 N HCl at 60 °C for one hour resulted in the highest amount
of R-CEHC detected. Under the protection of ascorbate, the
production of R-tocopheronolactone was minimal and the
recovery of R-CEHC from the R-CEHC sulfate solution was
over 100% through this process of acid hydrolysis at 60 °C
(Table 3).
For comparison, the acid hydrolysis coupled with methylation
method reported by Kiyose et al. (10) was also examined. This
method which measures total CEHC after free and conjugated
CEHC in samples were converted to CEHC-Me has been
employed in many studies (21-24). As shown in Table 2 and
Figure 4A, the amount R-CEHC obtained from 6 N HCl acid
hydrolysis at 60 °C for one hour is significantly higher than the
amount of R-CEHC-Me derived from acid methylation. Notice-
To obtain a valid acid hydrolysis procedure, the released
amount of R-CEHC and R-tocopheronolactone production were

Acid Hydrolysis of R-CEHC Sulfate
J. Agric. Food Chem., Vol. 56, No. 23, 2008 11111
Table 2. The production of R-CEHC and R-Tocopheronolactone from R-CEHC Sulfate Standard by Various Hydrolysis Procedures
acid hydrolysisb
acid methylationc
3 N HCl 6 N HCl
none
sulfataseb
sulfatase + 3 N HClb
3 N HCl
6 N HCl
6 N HCl at 60 °C
R-CEHC (µM)
0.01 ( 0.00 ea,d 0.03 ( 0.01 e
58.1 ( 3.4 c
73.9
6
56.8 ( 9.4 c 63.3 ( 9.5 b
82.5 ( 7.8 a
105.0
9
43.7 ( 1.0 d 55.7 ( 0.7 c
%
0.01
4
0.04
5
72.3
23
81.3
9
55.6
2
70.9
2
Na
R-tocopheronolactone (µM) ND
ND
0.28 ( 0.15 b
0.36
3
0.65 ( 0.22 b 1.00 ( 0.15 b
3.69 ( 2.55 a
4.69
6
%
0.83
15
1.27
3
N
2
3
a Each value represents mean ( SD. ND, nondetectable. N indicates the number of independent experiments. b 78.6 µM R-CEHC sulfate solution (trimethylamine salt
of R-CEHC sulfate MW 477, 37.5 µg/mL) contains 20 mg/mL ascorbic acid (pH value around 4) were respectively hydrolyzed by sulfatase, sulfatase followed by HCl (final
concentration 3 N), HCl (final concentration 3 or 6 N) at room temperature, HCl at 60 °C for one hour (final concentration 3 or 6 N). R-CEHC released was analyzed by
HPLC-ECD using MeOH/H2O ) 43/57 (the aqueous phase contains 50 mM ammonium acetate, pH 4.5) as the mobile phase. R-Tocopheronolactone was analyzed by
HPLC-UV at 268 nm with the same mobile phase system. c For comparison, the sulfate R-CEHC solution was also deconjugated by methanolic HCl (final concentration
3 N or 6 N) at 60 °C for 1 h. The R-CEHC-Me produced was analyzed by HPLC-ECD using CH3CN/H2O ) 43/57 (the aqueous phase contains 50 mM ammonium
acetate, pH 4.5) as the mobile phase. d Values not sharing a common letter a,b,c,d,e are significantly different from one another in a horizontal row by oneway ANOVA
and Duncan’s multiple range test (p < 0.05).
Table 3. The Recovery Test of the R-CEHC Analysis by Spiking the R-CEHC or R-CEHC Sulfate Standard before Acid Hydrolysis
3 N HCl
6 N HCl at 60 °C
samplesb
rat urine
spike 1c
R-CEHC sulfate (41.9 µM)
R-CEHC sulfate (78.6 µM)
nonspike
spike 2c
nonspike
spike 1
nonspike
spike 1
R-CEHC
µM
129.9 ( 9.7a
169.9 ( 5.8
160.0 ( 4.5
21.6 ( 3.4
51.6
55.4 ( 0.0
75.9 ( 3.2
96.6
113.2 ( 5.8
%
recoveryd
%
111.0
95.7
94.0
103.7
a Each value represents mean ( SD of 2 to 5 independent experiments. b Rat urine was collected from rats fed an AIN-76 modified diet containing 500 mg/kg diet of
all-rac-R-tocopheryl acetate. R-CEHC sulfate standard is dissolved in 60 mM Na2HCO3 solution to a final concentration of 41.9 or 78.6 µM. Both contained 20 mg/mL
ascorbic acid and were hydrolyzed with HCl (final concentration 3 N) at room temperature or HCl at 60 °C for 1 h (final concentration 6 N). R-CEHC released was extracted
by diethyl ether and analyzed by HPLC-ECD using MeOH /H2O ) 43/57 (aqueous phase contains 50 mM sodium acetate, pH 4.5) as the mobile phase. c Spike 1 and
2 respectively indicate the samples were added with 35.97 nmol of R-CEHC (R-CEHC MW 278, 10 µg) or 31.44 nmol of R-CEHC sulfate (trimethylamine salt of R-CEHC
sulfate MW 477, 15 µg). d Recovery (%) ) [(spike - non-spike)/35.97 or 31.44] × 100.
ably, sulfatase released very minimal amount of R-CEHC from
R-CEHC sulfate solution and rat urine samples, unless high
concentration of HCl was added after enzyme reaction before
the extraction procedure (Table 2, Figure 4B).
The recovery of R-CEHC by the procedure established was
further validated by spiking rat urine samples or R-CEHC sulfate
solution with R-CEHC or R-CEHC sulfate standards. The
recovery ranged from 96 to 111% (Table 3).
that sulfate conjugate might be the dominant conjugate form of
R-CEHC in rat urine. This is also in agreement with the
observation on γ-CEHC of Tanabe et al. (18).
Surprisingly, sulfatase has very low efficiency in hydrolyzing
R-CEHC sulfate. Chiku et al. (5) isolated the metabolite of
δ-tocopherol from the urine of rats given radioactive δ-toco-
pherol by TLC after hydrolysis of conjugates with sulfatase in
various conditions. By treatment with sulfatase in acetate buffer,
radioactivity moved far from the origin; but in a buffer
containing phosphate ion, radioactivity remained at the origin.
Their results indicated that the phosphate ion is a sulfatase
inhibitor and the urinary metabolite of δ-tocopherol was excreted
as conjugates of sulfate. Based on this report, we have tried to
improve the condition of enzyme reaction by excluding the
phosphate ion in the reaction mixture of sulfatase and the
R-CEHC sulfate standard. The release of free R-CEHC, how-
ever, was still very low. In contrast, the sulfatase could hy-
drolyze its optimal substrate p-nitrocatechol sulfate to produce
p-nitrocatechol and free sulfate efficiently in acetate buffer (data
not shown). The sulfatase used in our study is thus indeed active.
Therefore, the inefficiency of sulfatase hydrolysis of R-CEHC
sulfate is speculated to be attributed to the number of methyl
groups substituted in the chroman ring which has exerted a
steric hindrance that might interfere in the catalytic function
of the sulfatase enzyme. δ-CEHC sulfate, in contrast, with
only one methyl group at the meta-position to the sulfate
group on the chroman ring, was shown to be hydrolyzable
with sulfatase. It is not known whether such interference also
DISCUSSION
All forms of tocopherols and tocotrienols (R, ꢀ, γ and δ) are
metabolized to CEHCs (25, 26). However, there is disagreement
in the detection of conjugated forms of CEHCs in urine
samples (4, 18-20). Despite the disagreement, hydrolyzed with
ꢀ-glucuronidase is still the most widely employed process before
the extraction for analyzing CEHCs in specimens from humans
(14), rats (15) or mice (16, 17). However, we observed that
pretreatment of rat urine with ꢀ-glucuronidase (with sulfatase
activity) or sulfatase released only very small amount of
R-CEHC (Table 2) in contrast to a vast amount released by
HCl hydrolysis. We then used the release of R-CEHC by HCl
to track the original chemical species and found that R-CEHC
sulfate is the HCl releasable conjugate of R-CEHC in the rat
urine. It is known that ester linkage of a sulfate conjugate could
be hydrolyzed by acid catalysis, especially with high concentra-
tion of HCl. This is in agreement with our observation that HCl
dose-dependently released R-CEHC from rat urine and R-CEHC
sulfate solution. Together with the result that ꢀ-glucuronidase
only released very limited amount of R-CEHC, it is suggested

11112 J. Agric. Food Chem., Vol. 56, No. 23, 2008
Li et al.
existed in the hydrolysis of R-CEHC glucuronide by the
metabolites are excreted in urine or bile (26, 27). Since we found
that HCl can release such a large amount of R-CEHC from rat
urine and the HCl releasable form was identified to be R-CEHC
sulfate, it is conceivable that R-CEHC sulfate is the major and
dominate form of R-CEHC in rat urine. Indeed, Jiang et al. (12)
also identified sulfated long-chain carboxychromanols (sulfated
9′-, 11′-, and 13′-carboxychromanol) as novel vitamin E meta-
bolites from the human A549 cells or rats treated with γ- or
δ-tocopherol and they further provided evidence that sulfation
may occur parallel with ꢀ-oxidation. Therefore, sulfation may
play an important role in the vitamin E catabolism and merit
further investigation.
In conclusion, we observed that HCl can release a very large
amount of R-CEHC from rat urine. This acid releasable
conjugate was isolated, purified and identified to be R-CEHC
sulfate. A relatively simple and rapid method for the measure-
ment of R-CEHC was then developed. In this procedure, samples
are added with ascorbate as the antioxidant, acid hydrolyzed
withheat,extractedbydiethyletherandanalyzedbyHPLC-ECD.
The method is sensitive and reproducible and can be used for
further study of the vitamin E catabolism to R-CEHC.
glucuronidase enzyme.
Using our synthetic R-CEHC sulfate, we found that acid
hydrolysis with 6 N HCl at 60 °C for one hour almost
completely hydrolyzed the conjugate. Schultz et al. (4) showed
an almost complete conversion of R-CEHC to R-tocopherono-
lactone after bubbling oxygen to a solution of 70 µM R-CEHC
in 0.1 M HCl for 24 h at room temperature, suggesting that
urinary R-tocopheronolactone found in the early report might
be artificially produced through the chemical isolation process.
However, Pope et al. (13) has been able to detect R-tocopher-
onolactone, in human urine and speculated it as an indicator of
in vivo oxidative stress. Irrespective the origin of the R-toco-
pheronolactone in the urine, our procedure of including a high
concentration of ascorbate as an antioxidant could keep the
production of R-tocopheronolactone at a minimal level through
the acid hydrolysis with heat.
A number of reports employed the method of Kiyose et al.
(10) to detect CEHC by methylation coupled to acid hydrolysis
using methanolic HCl. CEHC-Me is considered to be more
stable that can prevent the artificial production of tocopherono-
lactone. This method was also conducted in this study for a
comparison to our procedure of acid hydrolysis with heat. Using
the R-CEHC sulfate solution, it was observed that acid methy-
lation resulted in the detection of about 70% of R-CEHC, in
contrast to about 105% by our procedure. Similarly, the amount
of R-CEHC in the rat urine sample detected by our procedure
of acid hydrolysis with heat was significantly higher than the
acid methylation process (Figure 4A).
ABBREVIATIONS USED
R-TOH, R-tocopherol; R-CEHC, 2,5,7,8-tetramethyl-2-(2′-
carboxyethyl)-6-hydroxychroman; R-CEHC sulfate, 6-O-sulfated
R-CEHC; R-CEHC-Me, R-CEHC methyl ester; HPLC-ECD,
high performance liquid chromatography-electrochemical
detector.
Our HPLC-ECD method for the measurement of R-CEHC
was modified from Lodge et al. (9) by changing the voltage in
the detection of ECD and by using mobile phase without
gradient. Furthermore, an external calibration curve was used
instead of the internal standard. In addition to measure the
amount of total R-CEHC by prior acid hydrolysis with heat,
free form of R-CEHC can also be determined by acidifying the
urine samples to pH 4 with ascorbate prior to extraction. Using
ascorbic acid to acidify to a pH of 4 can adequately protonate
R-CEHC for complete extraction but avoid the hydrolysis of
the conjugated form by adding HCl. Therefore, the amount of
R-CEHC extracted under pH 4 can be defined as the amount
of free R-CEHC in the samples. On the other hand, the amount
of R-CEHC extracted after 6 N HCl at 60 °C for one hour in
the presence of ascorbate can be defined as the amount of total
R-CEHC in the samples, which includes free and all conjugat-
ed R-CEHC. Stahl et al. (7) reported that about 35% of total
R-CEHC was present as glucuronide conjugate in human serum
but γ-CEHC are all present as the free form. Lodge et al. (9)
indicated that, for both R- and γ-CEHC, the free form typically
comprises from 5 to 25% of the total metabolites in human urine.
In our study, over 99.6% of R-CEHC in the rat urine existed in
the conjugated forms. Based on our result (Figure 3), it is
probable that, even without enzyme hydrolysis, a significant
proportion of the sulfate conjugate has been hydrolyzed while
acidified with HCl to lower pH before extraction.
Supporting Information Available: The procedure of chemi-
cal synthesis of standard compound, chromatography and
standard calibration curve of R-CEHC standard, and isolation
procedure and spectra of the vitamin E metabolite in rat urine
are available free of charge via the Internet at http://
pubs.acs.org.
LITERATURE CITED
(1) Brigelius-Flohe, R.; Traber, M. G. Vitamin E: function and
metabolism. FASEB J. 1999, 13 (10), 1145–55.
(2) Jiang, Q.; Christen, S.; Shigenaga, M. K.; Ames, B. N. Gamma-
tocopherol, the major form of vitamin E in the US diet, deserves
more attention. Am. J. Clin. Nutr. 2001, 74 (6), 714–22.
(3) Eisengart, A.; Milhorat, A. T.; Simon, E. J.; Sundheim, L. The
metabolism of vitamin E. II. Purification and characterization of
urinary metabolites of alpha-tocopherol. J. Biol. Chem. 1956, 221
(2), 807–17.
(4) Schultz, M.; Leist, M.; Petrzika, M.; Gassmann, B.; Brigelius-
Flohe, R. Novel urinary metabolite of alpha-tocopherol, 2,5,7,8-
tetramethyl-2(2′-carboxyethyl)-6-hydroxychroman, as an indicator
of an adequate vitamin E supply? Am. J. Clin. Nutr. 1995, 62
(6), 1527S–1534S.
(5) Chiku, S.; Hamamura, K.; Nakamura, T. Novel urinary metabolite
of d-delta-tocopherol in rats. J. Lipid Res. 1984, 25 (1), 40–8.
(6) Swanson, J. E.; Ben, R. N.; Burton, G. W.; Parker, R. S. Urinary
excretion of 2,7, 8-trimethyl-2-(beta-carboxyethyl)-6-hydroxy-
chroman is a major route of elimination of gamma-tocopherol in
humans. J. Lipid Res. 1999, 40 (4), 665–71.
(7) Stahl, W.; Graf, P.; Brigelius-Flohe, R.; Wechter, W.; Sies, H.
Quantification of the alpha- and gamma-tocopherol metabolites
2,5,7, 8-tetramethyl-2-(2′-carboxyethyl)-6-hydroxychroman and
2,7, 8-trimethyl-2-(2′-carboxyethyl)-6-hydroxychroman in human
serum. Anal. Biochem. 1999, 275 (2), 254–9.
The pathway in which vitamin E is catabolized to CEHC and
its conjugate is speculated to be mediated by phase I and phase
II enzymes that are responsible for the elimination of drugs,
xenobiotics as well as endogeneous compounds with low
solubility. The phase I enzyme reaction might be cytochrome
P-450 mediated ω-hydroxylation which is followed by stepwise
ꢀ-oxidation of the phytyl side chain. The phase II enzymes,
such as UDP glucuronosyl transferases (UGTs) and sulfotrans-
ferase, then catalyze the conjugation reaction before the final
(8) Himmelfarb, J.; Kane, J.; McMonagle, E.; Zaltas, E.; Bobzin, S.;
Boddupalli, S.; Phinney, S.; Miller, G. Alpha and gamma
tocopherol metabolism in healthy subjects and patients with end-
stage renal disease. Kidney Int. 2003, 64 (3), 978–91.

Acid Hydrolysis of R-CEHC Sulfate
J. Agric. Food Chem., Vol. 56, No. 23, 2008 11113
(9) Lodge, J. K.; Traber, M. G.; Elsner, A.; Brigelius-Flohe, R. A
rapid method for the extraction and determination of vitamin E
metabolites in human urine. J. Lipid Res. 2000, 41 (1), 148–54.
(10) Kiyose, C.; Saito, H.; Kaneko, K.; Hamamura, K.; Tomioka, M.;
Ueda, T.; Igarashi, O. Alpha-tocopherol affects the urinary and
biliary excretion of 2,7,8-trimethyl-2 (2′-carboxyethyl)-6-hydroxy-
chroman, gamma-tocopherol metabolite, in rats. Lipids 2001, 36
(5), 467–72.
(11) Hattori, A.; Fukushima, T.; Imai, K. Occurrence and determination
of a natriuretic hormone, 2,7,8-trimethyl-2-(beta-carboxyethyl)-
6-hydroxy chroman, in rat plasma, urine, and bile. Anal. Biochem.
2000, 281 (2), 209–15.
(12) Jiang, Q.; Freiser, H.; Wood, K. V.; Yin, X. Identification and
quantitation of novel vitamin E metabolites, sulfated long-chain
carboxychromanols, in human A549 cells and in rats. J. Lipid
Res. 2007, 48 (5), 1221–30.
(13) Pope, S. A.; Clayton, P. T.; Muller, D. P. A new method for the
analysis of urinary vitamin E metabolites and the tentative
identification of a novel group of compounds. Arch. Biochem.
Biophys. 2000, 381 (1), 8–15.
(14) Yoshikawa, S.; Morinobu, T.; Hamamura, K.; Hirahara, F.;
Iwamoto, T.; Tamai, H. The effect of [gamma]-tocopherol
administration on [alpha]-tocopherol levels and metabolism in
humans. Eur. J. Clin. Nutr. 2005, 59, 900–5.
(15) Leonard, S. W.; Gumpricht, E.; Devereaux, M. W.; Sokol, R. J.;
Traber, M. G. Quantitation of rat liver vitamin E metabolites by
LC-MS during high-dose vitamin E administration. J. Lipid Res.
2005, 46, 1068–75.
(16) Kluth, D.; Landes, N.; Pfluger, P.; Muller-Schmehl, K.; Weiss,
K.; Bumke-Vogt, C.; Ristow, M.; Brigelius-Flohe, R. Modulation
of Cyp3a11 mRNA expression by alpha-tocopherol but not
gamma-tocotrienol in mice. Free Radical Biol. Med. 2005, 38
(4), 507–14.
(17) Traber, M. G.; Siddens, L. K.; Leonard, S. W.; Schock, B.; Gohil,
K.; Krueger, S. K.; Cross, C. E.; Williams, D. E. alpha-Tocopherol
modulates Cyp3a expression, increases gamma-CEHC production,
and limits tissue gamma-tocopherol accumulation in mice fed high
gamma-tocopherol diets. Free Radical Biol. Med. 2005, 38 (6),
773–85.
(19) Morinobu, T.; Yoshikawa, S.; Hamamura, K.; Tamai, H. Measure-
ment of vitamin E metabolites by high-performance liquid
chromatography during high-dose administration of alpha-toco-
pherol. Eur. J. Clin. Nutr. 2003, 57 (3), 410–4.
(20) Pope, S. A.; Burtin, G. E.; Clayton, P. T.; Madge, D. J.; Muller,
D. P. Synthesis and analysis of conjugates of the major vitamin
E metabolite, alpha-CEHC. Free Radical Biol. Med. 2002, 33 (6),
807–17.
(21) Ikeda, S.; Tohyama, T.; Yamashita, K. Dietary sesame seed and
its lignans inhibit 2,7,8-trimethyl- 2(2′-carboxyethyl)-6-hydroxy-
chroman excretion into urine of rats fed gamma-tocopherol. J.
Nutr. 2002, 132 (5), 961–6.
(22) Saito, H.; Kiyose, C.; Yoshimura, H.; Ueda, T.; Kondo, K.;
Igarashi, O. Gamma-tocotrienol, a vitamin E homolog, is a
natriuretic hormone precursor. J. Lipid Res. 2003, 44 (8), 1530–
5.
(23) Uto, H.; Kiyose, C.; Saito, H.; Ueda, T.; Nakamijra, T.; Igarashi,
O.; Kondo, K. Gamma-tocopherol enhances sodium excretion as
a natriuretic hormone precursor. J. Nutr. Sci. Vitaminol. (Tokyo)
2004, 50 (4), 277–82.
(24) Uchida, T.; Ichikawa, T.; Abe, C.; Yamashita, K.; Ikeda, S. Dietary
sesame seed decreases urinary excretion of alpha- and gamma-
tocopherol metabolites in rats. J. Nutr. Sci. Vitaminol. (Tokyo)
2007, 53 (4), 372–6.
(25) Lodge, J. K.; Ridlington, J.; Leonard, S.; Vaule, H.; Traber, M. G.
Alpha- and gamma-tocotrienols are metabolized to carboxyethyl-
hydroxychroman derivatives and excreted in human urine. Lipids
2001, 36 (1), 43–8.
(26) Birringer, M.; Pfluger, P.; Kluth, D.; Landes, N.; Brigelius-Flohe,
R. Identities and differences in the metabolism of tocotrienols
and tocopherols in HepG2 cells. J. Nutr. 2002, 132 (10), 3113–8.
(27) Sontag, T. J.; Parker, R. S. Cytochrome P450 omega-hydroxylase
pathway of tocopherol catabolism. Novel mechanism of regulation
of vitamin E status. J. Biol. Chem. 2002, 277 (28), 25290–6.
Received for review August 7, 2008. Revised manuscript received
September 30, 2008. Accepted September 30, 2008. The study was
financially supported by a grant (NSC 90-2320-B-002-102) from the
National Science Council, Taiwan.
(18) Tanabe, M.; Fukushima, T.; Usui, N.; Aoyama, N.; Tsunoda, M.;
Imai, K. Intravenous administration of 2,7,8-trimethyl-2-(beta-
carboxyethyl)-6-hydroxy chroman (gamma-CEHC) to rats and
determination of its plasma concentration and urinary sodium
excretion. Biomed. Chromatogr. 2004, 18 (9), 727–34.
JF802459D