Biotransformation of [12C]- and [13C]-tert-amyl methyl ether and tert-amyl alcohol

Article abstract of DOI:10.1021/tx990078x

tert-Amyl methyl ether (TAME) is intended for use as a gasoline additive to increase oxygen content. Increased oxygen content in gasoline reduces tailpipe emissions of hydrocarbons and carbon monoxide from cars. Due to possible widespread use of TAME, the toxicity of TAME is under investigation. We studied the biotransformation of TAME in rats and one human volunteer after inhalation of ¹²C- or ¹³C-labeled TAME. In addition, the biotransformation of [¹³C]-tert-amyl alcohol was studied in rats after gavage. Urinary metabolites were identified by GC/MS and ¹³C NMR. Rats (two males and two females) were individually exposed to 2000 ppm [¹²C]- or [¹³C]TAME for 6 h, and urine was collected for 48 h. Free and glucuronidated 2-methyl-2,3-butanediol and a glucuronide of tert-amyl alcohol were identified by ¹³C NMR, GC/MS, and LC/MS/MS as major urinary metabolites on the basis of the relative intensities of the ¹³C NMR signals. The presence of several minor metabolites was also indicated by ¹³C NMR; they were identified as tert-amyl alcohol, 2-hydroxy-2- methylbutyric acid, and 3-hydroxy-3-methylbutyric acid. One human volunteer was exposed to an initial concentration of 27 000 ppm [¹³C]TAME by inhalation for 4 min from a 2 L gas sampling bag, and metabolites of TAME excreted in urine were analyzed by ¹³C NMR. All TAME metabolites identified in rats were also present in the human urine samples. To study tert-amyl alcohol biotransformation, male rats (n = 3) were treated with 250 mg/kg [¹³C]-tert-amyl alcohol dissolved in corn oil by garage, and urine was collected for 48 h. ¹³C NMR of the urine samples showed the presence of metabolites identical to those in the urine of [¹³C]TAME-treated rats. Our results suggest that TAME is extensively metabolized by rats and humans to tert-amyl alcohol which may be further oxidized to diols and carboxylic acids. These reactions are likely mediated by cytochrome P450-dependent oxidations.

Full text of DOI:10.1021/tx990078x

9
58
Chem. Res. Toxicol. 1999, 12, 958-964
1
2
13
Biotr a n sfor m a tion of [ C]- a n d [ C]-ter t-Am yl Meth yl
Eth er a n d ter t-Am yl Alcoh ol
†
†
‡
Alexander Amberg, Ulrike Bernauer, Dieter Scheutzow, and
,
†
Wolfgang Dekant*
Institut f u¨ r Toxikologie, Universit a¨ t W u¨ rzburg, Versbacher Strasse 9, 97078 W u¨ rzburg, Germany,
and Institut f u¨ r Organische Chemie, Universit a¨ t W u¨ rzburg, Am Hubland,
9
7070 W u¨ rzburg, Germany
Received May 11, 1999
tert-Amyl methyl ether (TAME) is intended for use as a gasoline additive to increase oxygen
content. Increased oxygen content in gasoline reduces tailpipe emissions of hydrocarbons and
carbon monoxide from cars. Due to possible widespread use of TAME, the toxicity of TAME is
under investigation. We studied the biotransformation of TAME in rats and one human
1
2
13
volunteer after inhalation of C- or C-labeled TAME. In addition, the biotransformation of
1
3
[
C]-tert-amyl alcohol was studied in rats after gavage. Urinary metabolites were identified
1
3
by GC/MS and C NMR. Rats (two males and two females) were individually exposed to 2000
1
2
13
ppm [ C]- or [ C]TAME for 6 h, and urine was collected for 48 h. Free and glucuronidated
-methyl-2,3-butanediol and a glucuronide of tert-amyl alcohol were identified by 13C NMR,
GC/MS, and LC/MS/MS as major urinary metabolites on the basis of the relative intensities
2
of the 13C NMR signals. The presence of several minor metabolites was also indicated by
NMR; they were identified as tert-amyl alcohol, 2-hydroxy-2-methylbutyric acid, and 3-hydroxy-
-methylbutyric acid. One human volunteer was exposed to an initial concentration of 27 000
13
C
3
1
3
ppm [ C]TAME by inhalation for 4 min from a 2 L gas sampling bag, and metabolites of TAME
1
3
excreted in urine were analyzed by C NMR. All TAME metabolites identified in rats were
also present in the human urine samples. To study tert-amyl alcohol biotransformation, male
1
3
rats (n ) 3) were treated with 250 mg/kg [ C]-tert-amyl alcohol dissolved in corn oil by gavage,
1
3
and urine was collected for 48 h. C NMR of the urine samples showed the presence of
1
3
metabolites identical to those in the urine of [ C]TAME-treated rats. Our results suggest that
TAME is extensively metabolized by rats and humans to tert-amyl alcohol which may be further
oxidized to diols and carboxylic acids. These reactions are likely mediated by cytochrome P450-
dependent oxidations.
In tr od u ction
Increased oxygen content in gasoline sold in certain
rodents after inhalation of high concentrations of TAME
is central nervous system depression. Subchronic inhala-
tion studies with TAME in rats have found increased
liver weights in rats exposed to 4000 ppm TAME for 4
weeks (6 h per day, 5 days per week). Significantly
reduced body weights and relative increases in adrenal,
kidney, testes, brain, and lung weights were seen in male
rats exposed to 4000 ppm TAME, but there were no
treatment-related histopathologic findings (5).
areas of the United States is required by the 1990
Amendments to the Clean Air Act in areas that failed to
meet the National Ambient Air Quality standard for
carbon monoxide or ozone. The chemicals blended with
gasoline hydrocarbons to meet the required oxygen
content are termed “oxygenates” (1). At present, the
oxygenates most often used are methyl tert-butyl ether
and methanol. However, other ethers may be used in the
The biotransformation of TAME has apparently not
been studied. tert-Amyl alcohol is a likely metabolite of
TAME formed by cleavage of the ether bond; however,
further metabolism of tert-amyl alcohol is likely as
observed with other more lipophilic alcohols. For ex-
ample, tert-butanol is extensively metabolized in rodents
and humans (6).
1
future. tert-Amyl methyl ether (TAME) is also used as
an oxygenate due to relatively low production costs, the
fact that its vapor pressure is lower than those of other
ethers, and its ability to act as a high-octane gasoline
blending compound (2, 3).
Due to the potential widespread exposure of humans
to oxygenates in fuel (3, 4), studies on the toxicity of
TAME are underway. The major toxic effect seen in
In this study, we investigated the biotransformation
of TAME in rats with the aim of identifying metabolites
1
2
13
excreted in urine. Both [ C]TAME and C-labeled
TAME were used for these studies. The use of 1 C-labeled
TAME permits the detection of metabolites in urine by
3
*
To whom correspondence should be addressed: Department of
Toxicology, University of W u¨ rzburg, Versbacher Str. 9, 97078 W u¨ rzburg,
Germany. Telephone: +49(931)201 3449. Fax: +49(931)201 3446.
E-mail: dekant@toxi.uni-wuerzburg.de.
1
3
C NMR without manipulations or chromatography and
therefore may help in identifying metabolites with low
stability or volatility and products formed by the possible
reaction of TAME metabolites with endogenous chemicals
(7-9).
†
Institut f u¨ r Toxikologie, Universit a¨ t W u¨ rzburg.
Institut f u¨ r Organische Chemie, Universit a¨ t W u¨ rzburg.
Abbreviations: TAME, tert-amyl methyl ether; MS/MS, secondary
‡
1
ion mass spectrometry.
1
0.1021/tx990078x CCC: $18.00 © 1999 American Chemical Society
Published on Web 09/10/1999

Biotransformation of tert-Amyl Methyl Ether
Chem. Res. Toxicol., Vol. 12, No. 10, 1999 959
1
3
Ma ter ia ls a n d Meth od s
Sch em e 1. Syn th esis of 2-[ C]-ter t-Am yl Meth yl
Eth er
An im a ls a n d Tr ea tm en t. Male and female Fischer F344
rats were obtained from Harlan Winkelmann (Borchen, Ger-
many). Animals were kept at constant humidity and tempera-
ture (21 °C) in the animal facility of the department with a 12
h light/dark cycle. Before the experiments, the animals were
acclimated to the metabolic cages for 3 days, and control urine
was collected during this time for 12 h before the exposure.
Exp osu r e of On e Hu m a n Volu n teer to [13C]TAME by
In h a la tion . A 2 L gas sampling bag (Linde, Germany) was filled
1
3
with 99.9% oxygen, and 300 µL of [ C]TAME was added with
a microliter syringe through a septum to give a concentration
of 27 000 ppm. The volunteer inhaled the contents of the bag
for 4 min and exhaled into the bag to ensure maximum uptake
1
3
8000 series gas chromatograph. Split injection (split ratio of 1:5)
was used, and spectra were recorded from m/z 30 to 300 in 1 s
intervals.
of [ C]TAME. Urine samples were taken in 6 h intervals for
8 h.
Ch a m ber Design a n d Tr ea tm en t. The design of the
4
exposure chamber has been described in detail (10). The animals
Secondary ion mass spectrometry (MS/MS) was performed
using a Finnigan TSQ-700 triple-quadropole mass spectrometer
with electrospray ionization. Samples were introduced at a flow
rate of 0.5 mL/min in water adjusted to pH 3 by addition of
formic acid. LC/MS and LC/MS/MS spectra were obtained by
automatically switching between positive- and negative-ion
modes in alternating scans. The electrospray voltage was 4.5
kV, and the capillary temperature was 200 °C. MS/MS spectra
were recorded with a collision energy of 30 eV. To isolate the
nonvolatile metabolites of TAME, HPLC with an evaporative
light scattering detector (Sedere 55, Knauer, Germany) was used
due to the low UV absorption of the metabolites. Urine samples
were separated on a 250 mm × 4 mm steel column filled with
Partisil ODS-III using gradient elution. A linear gradient from
100% water (acidified with formic acid to pH 3) to 50% water/
acetonitrile over the course of 25 min at a flow rate of 1 mL/
min was used for separation.
were introduced into the chamber at 10 a.m., and the calculated
1
2
13
amounts of the C- or C-labeled ethers were introduced with
a microliter syringe. The concentrations of the ethers in the gas
phase of the chamber were monitored by an automatic gas
sampling valve every 10 min. The gas phase (100 µL) was
introduced into a capillary gas chromatograph (HP model 5970)
and separated with a 40 m DB1-coated fused silica column (DB1,
J &W Scientific, 40 m, 0.18 mm i.d., 0.4 µm film thickness). Ether
concentrations in the air were quantified by flame ionization
detection (FID). After the end of the 6 h exposure period, the
animals were transferred to metabolic cages and the urine was
collected on ice in 24 h intervals over the course of 48 h. Two
male and two female rats were individually exposed to each of
1
3
12
the two [ C]- and [ C]ethers.
Meta bolite An a lysis in Ur in e. Urine obtained from the
1
3
animals exposed to C-labeled ethers or control urine (720 µL)
Syn th esis of 2-[13C]-ter t-Am yl Alcoh ol. A solution of 60
mmol of ethyl bromide (Aldrich, Steinheim, Germany) in diethyl
ether was slowly added to equimolar amounts of magnesium
turnings (1.5 g) covered by 5.0 mL of diethyl ether (Scheme 1).
The Grignard reaction was initiated by the addition of traces
of iodine (12, 13). The mixture was stirred for 30 min at reflux;
a solution of 50 mmol of 2-[ C]acetone (lot no. P-7787, CIL
Cambridge Isotope Laboratories, Andover, MA) in diethyl ether
was added over the course of 30 min, and the mixture was kept
at reflux for 2 h. After cooling, hydrolysis was performed with
was introduced into separate NMR tubes (5 mm i.d.), and 80
µL of D
2
O was added. These samples were directly analyzed by
C NMR (6). Some of the urine samples were treated with
â-glucuronidase (Sigma, G7846, lot no. 72H6836) for 30 min at
1
3
3
7 °C or sulfatase (Sigma, S9751, lot no. 621-07881) for 48 h at
3
7 °C (11). Aliquots of the treated urine samples were then also
1
3
analyzed by NMR. For acid treatment, urine samples were
acidified to pH 2 with concentrated hydrochloric acid and
incubated for 1 h at 90 °C.
For gas chromatography/mass spectrometry, 0.5 mL of the
urine samples was introduced into 1.5 mL sample vials, acidified
with 2 N HCl to pH 4, and kept at 80 °C for 30 min. A volume
3
0 mL of an ice-cold, saturated NH
separated, and the aqueous layer was extracted five times with
0 mL of diethyl ether. The ether layers were combined and
dried over K CO . After evaporation of the solvent, the residue
4
Cl solution. The layers were
1
(100 µL) of the headspace from the incubations was then injected
into the gas chromatograph. To analyze less volatile metabolites,
some samples (0.5 mL) were extracted with 0.5 mL of ethyl
acetate, and 1 µL of the ethyl acetate layer was injected into
the gas chromatograph.
To identify acidic metabolites, 0.5 mL of the urine samples
was taken to dryness by lyophilization and the residues that
2
3
1
3
was distilled to yield 2-[ C]-tert-amyl alcohol (61% yield, 98%
GC/FID purity). The structures of the reaction product was
confirmed by GC/MS and 1H and C NMR.
1
13
2-[13C]-ter t-Am yl a lcoh ol: H NMR (250 MHz, D
O) δ 0.89
OH], 1.19 [6H, s,
OH], 1.52 [2H, q, J ) 8 Hz, J HC
OH]; 13C NMR (63 MHz, D
O) δ 10.7 [s,
CH C(CH OH],
OH], 75.0 [s, d with 0.5%
2
3 2 3 2
[3H, t, J ) 8 Hz, J HC ) 4 Hz, CH CH C(CH )
were obtained were treated with 500 µL of BF
3
/methanol (14%)
J
HC ) 4 Hz, CH
4 Hz, CH CH C(CH
CH CH C(CH
38.1 [d, J CC ) 39 Hz, CH
3
CH
2
C(CH
3
)
2
)
at 60 °C for 30 min. Samples were then diluted with 250 µL of
water and extracted with 1 mL of chloroform. The chloroform
layers were dried over sodium sulfate, and 1 µL of the solution
was injected into the gas chromatograph. Separation was
performed on a 30 m × 0.25 mm fused silica column coated with
DB-WAX (0.25 µm film thickness). For analysis, a temperature
gradient from 35 to 230 °C with a heating rate of 10 °C/min
was applied. Electron impact mass spectra (70 eV) were recorded
and metabolite peaks identified by comparison of the chromato-
grams of the urines from treated rats with those of untreated
controls.
3
2
3
)
2
2
3
2
3
)
2
OH], 30.0 [d, J CC ) 40 Hz, CH
CH C(CH
3
2
3 2
)
3
2
3 2
)
of a satellite doublet J CC ) 39 Hz, CH
eV) m/z (relative intensity) 74 (M - CH
100), 56 (68), 44 (33), 43 (17), 42 (19), 41 (6), 40 (12).
3
CH
2
C(CH
, 85), 60 (M - C
3
)
2
OH]; MS (70
+
+
3
2
H
5
,
Syn th esis of 2-[13C]TAME fr om 2-[13C]-ter t-Am yl Alco-
h ol. A mixture of 60 mmol of methanol and 6 mL of 10% H -
2
SO was heated to 65 °C, and then 20 mmol of 2-[13C]-tert-amyl
4
alcohol was added with a syringe (14). When the temperature
of the mixture was increased to 100 °C, an azeotropic mixture
In str u m en ta l An a lysis. 13C NMR spectra were recorded
with a Bruker 250 MHz spectrometer (Bruker AC 250) or a
Bruker 600 MHz NMR spectrometer (Avance 600). Usually,
of 2-[ C]TAME, methanol, 2-methyl-2-butene, 2-methyl-1-
13
butene, and tert-amyl alcohol were distilled off. Methanol was
removed from the reaction mixture by extraction with water.
Further purification was performed by slow distillation with dry
ice cooling on a B u¨ chi-GKR-51 Kugelrohr apparatus to yield 10%
2
000 scans were aquired for Fourier transformation (6). Gas
chromatography/mass spectrometry was performed with
Fisons MD 800 mass spectrometer coupled to a Carlo Erba GC
a
1
3
2-[ C]TAME. The GC/FID purity of the reaction product was

9
60 Chem. Res. Toxicol., Vol. 12, No. 10, 1999
98%. The structure of the reaction product was confirmed by
Amberg et al.
>
1
13
GC/MS and H and C NMR.
1
3
1
2
-[ C]TAME: H NMR (250 MHz, D
Hz, J HC ) 4 Hz, CH CH C(CH OCH
Hz, CH CH C(CH OCH ], 1.55 [2H, q, J ) 8 Hz, J HC ) 4 Hz,
CH CH C(CH OCH ], 3.20 [3H, s, J HC ) 4 Hz, CH
]; 13C NMR (63 MHz, D
O) δ 10.4 [s, CH
], 26.5 [d, J CC ) 36 Hz, CH CH C(CH OCH
CH C(CH OCH ], 51.2 [s, CH
2
O) δ 0.85 [3H, t, J ) 8
3
2
3
)
2
3
], 1.17 [6H, s, J HC ) 4
3
2
3
)
2
3
3
2
3
)
2
3
3
CH
CH
2
C(CH
C(CH )
3
)
3 2
2
-
-
OCH
OCH
J
3
3
2
3
2
3
2
3
)
2
3
], 33.9 [d,
C(CH
CC ) 38 Hz, CH
3
2
3
)
2
3
3
CH
2
3
)
2
-
OCH ], 80.1 [s, d with 0.5% of a satellite doublet J CC ) 40 Hz,
CH
3
3
+
CH
- CH
2
C(CH
3
)
2
OCH
3
]; MS (70 eV) m/z (relative intensity) 88
+
(M
3
, 75), 74 (M - C
2
H
5
, 100), 72 (58), 56 (88), 46 (15),
4
4 (79), 42 (30), 41 (12), 40 (19).
Syn th esis of 2-Meth yl-2,3-bu ta n ed iol. 2-Methyl-2,3-bu-
tanediol was prepared by reduction of 3-hydroxy-3-methyl-2-
butanone (13). LiAlH (30 mmol, 1,120 mg) was dissolved in 15
mL of diethyl ether followed by slow addition of 12 mmol of
-hydroxy-3-methyl-2-butanone in 10 mL of diethyl ether at
F igu r e 1. 13C NMR spectrum (151 MHz, 2000 scans) of a 24 h
urine sample from a male Fisher F344 rat exposed to 2000 ppm
4
1
3
3
2-[ C]TAME for 6 h. The following structural assignments for
the NMR signals were made: δ 72.8, 3-hydroxy-3-methylbutyric
acid; δ 74.9, tert-amyl alcohol; δ 76.4, 2-methyl-2,3-butanediol;
δ 76.7, glucuronide of 2-methyl-2,3-butanediol; δ 79.8, 2-hy-
droxy-2-methylbutyric acid; and δ 83.1, glucuronide of tert-amyl
alcohol.
room temperature. The mixture was then stirred for 2 h at
reflux. After the mixture had been cooled with ice, 2.5 mL of
water and then 2 mL of a 20% solution of sodium hydroxide
were added. After extraction with diethyl ether, the organic
2 3
layers were combined and dried over K CO . After removal of
the solvent, 2-methyl-2,3-butanediol was isolated by fractional
distillation under reduced pressure.
standard procedures (Scheme 1); the purity of TAME and
tert-amyl alcohol used for the exposures was >98% as
checked by GC/MS.
1
2
-Meth yl-2,3-bu ta n ed iol: H NMR (250 MHz, D
3H, d, J ) 6 Hz, CH CH(OH)C(CH OH], 1.17 [9H, s, CH
CH(OH)C(CH OH], 3.63 [1H, q, J ) 6 Hz, CH
OH]; C NMR (63 MHz, D O) δ 19.3 [s, CH
OH], 25.9 [s, CH CH(OH)C(CH OH], 27.0 [s, CH
OH], 76.4 [s, CH CH(OH)C(CH OH], 76.7 [s, CH
OH)C(CH
, 9), 71 (24), 59 (100), 45 (22), 43 (91), 41 (27).
2
O) δ 1.15
[
3
3
)
2
3
-
-
-
3
)
2
3
CH(OH)C(CH
CH(OH)C(CH
3
3
)
2
Biotr a n sfor m a tion of TAME. Two male and two
1
3
2
3
)
2
female rats were individually exposed by inhalation to
3
3
)
2
3
CH(OH)C-
12
13
2
000 ppm (initial concentration) [ C]- or [ C]TAME.
(
(
CH
3
)
2
3
3
)
2
3
CH-
-
Monitoring of the exposure chamber air for TAME
indicated a continuous decrease in the air concentrations
of TAME due to uptake of the ethers by the rats and
biotransformation to less volatile metabolites (data not
shown). At the end of the exposure period of 6 h, the
TAME concentrations in the chamber were less than 300
ppm, indicating extensive uptake and biotransformation
of TAME by the rats. Analysis of the chamber air by gas
chromatography did not reveal the formation of volatile
and exhaled metabolites of TAME that could be detected
by the flame ionization detector. At the end of the
exposure period, the animals were transferred to meta-
bolic cages and urine samples were collected in 24 h
+
3
)
2
OH]; MS (70 eV) m/z (relative intensity) 89 (M
CH
3
Syn th esis of 2-Meth yl-1,2-bu ta n ed iol. 2-Methyl-1,2-bu-
tanediol was prepared by reduction of 2-hydroxy-2-methylbu-
tyric acid (Aldrich). LiAlH
1
2
4
(25 mmol, 960 mg) was dissolved in
2 mL of diethyl ether followed by slow addition of 12 mmol of
-hydroxy-2-methylbutyric acid in 10 mL of diethyl ether at
room temperature. The mixture was then stirred for 2 h at
reflux. After the mixture had been cooled with ice, 2.5 mL of
water followed by 2 mL of a 20% solution of sodium hydroxide
was added. After extraction with diethyl ether, the organic
layers were combined and dried over K CO . After removal of
2 3
the solvent, 2-methyl-1,2-butanediol was isolated by fractional
distillation under reduced pressure.
1
3
1
intervals for 48 h and analyzed by GC/MS and C NMR
spectroscopy to identify metabolites.
2
-Meth yl-1,2-bu ta n ed iol: H NMR (250 MHz, D
3H, t, J ) 8 Hz, CH CH C(CH )(OH)CH OH], 1.13 [3H, s, CH
CH C(CH )(OH)CH OH], 1.52 [2H, q, J ) 8 Hz, CH CH C(CH
OH)CH OH], 3.43 [2H, s, CH )(OH)CH
OH]; 13C
NMR (63 MHz, D )(OH)CH OH], 24.5
s, CH CH C(CH CH C(CH )(OH)CH
OH], 70.8 [s, CH OH], 76.4 [s, CH
CH )(OH)CH
CH
2
O) δ 0.89
[
3
2
3
2
3
-
2
3
2
3
2
3
)-
A typical NMR spectrum of a urine sample from a male
(
2
3
3
CH
CH
2
C(CH
C(CH
3
2
13
1
rat exposed to [ C]TAME is shown in Figure 1. The H-
decoupled 13C NMR spectrum exhibited several reso-
nances which were also present in the spectra of urine
from control animals (data not presented). The structures
of these endogenous compounds were assigned to urea,
creatinine, hippurate, and glucose by comparison with
2
O) δ 9.9 [s, CH
2
3
2
[
3
2
3
)(OH)CH
CH
2
OH], 32.8 [s, CH
3
2
3
2
-
3
2
C(CH
3
)(OH)CH
2
3
CH
2
C-
+
(
-
(
3
2
OH]; MS (70 eV) m/z (relative intensity) 89 (M
3
, 41), 75 (89), 73 (98), 71 (54), 58 (42), 57 (94), 55 (83), 53
30), 45 (73), 42 (100), 41 (79).
1
3
Ch em ica ls. 2-Hydroxy-2-methylbutyric acid and 3-hydroxy-
literature data (15) and reference spectra (6). The
C
1
3
3
-methyl-2-butanone were purchased from Aldrich, and 3-hy-
atom in [ C]TAME metabolites was expected to give
resonances in the range of chemical shifts between 70
and 90 ppm. In control urine, only one resonance was
droxy-3-methylbutyric acid and 2-methyl-2,4-butanediol were
obtained from Tokyo Kasei Kogyo Ltd. (Tokyo, J apan). All other
chemicals were obtained from commercial suppliers at the
highest purity available.
13
observed in that region of the C NMR spectra (δ 78.3).
1
3
Urine from animals exposed to [ C]TAME exhibited
several signals in the region between 70 and 90 ppm,
indicative of three major and at least three minor
metabolites. The structures of metabolites were eluci-
dated by a combination of 13C NMR spectroscopy, GC/
MS, and LC/MS/MS.
Resu lts
Syn th esis of 13C-La beled ter t-Am yl Alcoh ol a n d
TAME. The 13C-labeled compounds were synthesized for
identification of TAME and tert-amyl alcohol metabolites
1
4
without the need for C-label and detection of nonvolatile
metabolites. The use of 13C-labeled compounds permits
metabolite identification and may provide information
about the relative quantities of metabolites that are
present. The 13C-labeled compounds were synthesized by
The NMR signal at δ 72.8 was identical in chemical
13
shift to C-2 (which carries the C-label) of 3-hydroxy-3-
methylbutyric acid. The presence of this compound in the
1
2
13
urine of both [ C]- and [ C]TAME-treated rats was also
confirmed by GC/MS. Gas chromatographic separation

Biotransformation of tert-Amyl Methyl Ether
Chem. Res. Toxicol., Vol. 12, No. 10, 1999 961
F igu r e 2. Mass spectrum of 2-methyl-2,3-butanediol present
in the urine (collected for 24 h after the end of exposure) of a
1
3
male Fischer F344 rat exposed to 2-[ C]TAME (2000 ppm) for
1
2
F igu r e 3. Electrospray mass spectrum (A) and further frag-
6
h (A) and to [ C]TAME (2000 ppm) for 6 h (B).
1
3
mentation (m/z 326) (B) of the isolated metabolite with a
C
resonance at δ 76.7 at a collision energy of 30 mV. Spectra were
recorded after direct loop injection of the metabolite isolated
by preparative HPLC.
of the urine from those animals showed peaks not present
in urine from control animals. The mass spectra of one
1
2
of these peaks from [ C]TAME-treated animals were
identical to those of authentic 3-hydroxy-3-methylbutyric
acid (data not shown). In the mass spectra of this peak
1
3
obtained by separation of urine from [ C]TAME-treated
animals, several fragments were shifted by 1 mass unit,
suggesting the presence of 13C in the molecule (data not
presented). These observations confirm 3-hydroxy-3-
methylbutyric acid as a minor urinary metabolite of
TAME in rats.
The minor signal in the NMR spectrum of urine from
C]TAME-treated animals at δ 74.9 was identical in
[13
chemical shift to that of C-2 in tert-amyl alcohol. This
compound was also identified as a minor TAME metabo-
lite by GC/MS (data not shown).
The signal at δ 76.4 ppm represents a major TAME
metabolite in the rat. The C-2 atom of synthetic 2-methyl-
2
,3-butanediol also exhibited a resonance at δ 76.4.
Moreover, mass spectra recorded from a peak present in
1
2
the urine of [ C]TAME-treated rats, but not in the urine
of control animals, were identical to that of 2-methyl-
2
,3-butanediol (Figure 2). Again, in GC/MS of urine
1
3
samples from [ C]TAME-treated rats, several fragments
of the compound were shifted by 1 mass unit due to the
high abundance of 13C in the molecule (Figure 2). These
observations confirm 2-methyl-2,3-butanediol as a major
metabolite of TAME in rat urine.
F igu r e 4. COSY-NMR spectrum of the isolated metabolite with
13C resonance at δ 76.7. For recording NMR spectra, the
a
metabolite was isolated by preparative HPLC and dissolved in
O.
D
2
produced a prominent secondary ion at m/z 150 (glucu-
ronic acid) and several minor fragments indicative of the
consecutive loss of water and/or ketene. COSY-NMR
spectra (Figure 4) of the isolated peak exhibited a set of
resonances typical for glucuronic acids and their deriva-
tives (16) (δ 3-5) and the presence of the 2-methyl-2,3-
The most abundant signal in the 13C NMR spectrum
1
3
from urine of [ C]TAME-treated rats at δ 76.7 was
identified as a glucuronide of 2-methyl-2,3-butanediol on
the basis of the following observations. (1) Electrospray
MS/MS of the metabolite isolated from the urine of rats
exposed to [ C]TAME by HPLC showed signals at m/z
3
1
3
13
butanediol moiety containing C (δ 1-1.2, 3.6). More-
1
2
26 and 342 (Figure 3). (2) The metabolite from [ C]-
over, the presumed glucuronide was slowly cleaved by
glucuronidase to give 2-methyl-2,3-butanediol. Taken
together, the data conclusively identify the major urinary
metabolite of TAME as a glucuronide of 2-methyl-2,3-
butanediol.
TAME had signals at m/z 325 and 341, indicating the
addition of two sodium ions or one sodium and one
potassium ion to the molecular ion of 2-methyl-2,3-
butanediol glucuronide (Figure 3). The ion at m/z 326

9
62 Chem. Res. Toxicol., Vol. 12, No. 10, 1999
Amberg et al.
F igu r e 7. 13C NMR spectrum (151 MHz, 32 000 scans) of a 6
h urine sample from a male human volunteer exposed to 27 000
ppm TAME (in a 2 L gas sampling bag) for 4 min. The following
structural assignments for the NMR signals were made: δ 72.8,
3
2
-hydroxy-3-methylbutyric acid; δ 74.9, tert-amyl alcohol; δ 76.4,
-methyl-2,3-butanediol; δ 76.7, glucuronide of 2-methyl-2,3-
butanediol; δ 79.8, 2-hydroxy-2-methylbutyric acid; and δ 83.1,
glucuronide of tert-amyl alcohol.
Moreover, the NMR spectra also were consistent with the
presence of the glucuronic acid moiety (δ 3-5) and the
tert-amyl alcohol moiety in the molecule.
F igu r e 5. Electrospray mass spectrum (A) and further frag-
mentation (m/z 310) (B) of the isolated metabolite with a 13
resonance at δ 83.1 at a collision energy of 30 mV. Spectra were
recorded after direct loop injection of the metabolite isolated
by preparative HPLC.
Of the other minor metabolites that were present, the
C
compound giving the signal at δ 79.8 was identified as
13
2
-hydroxy-2-methylbutyric acid by comparison of the
C
NMR spectra of a synthetic reference and by GC/MS
data not presented).
No major differences in the intensity of the 13C NMR
(
signals of the individual metabolites were observed when
comparing the NMR spectra of urines from male and
1
3
female rats exposed to [ C]TAME. Moreover, the kinetics
of excretion of the metabolites with the urine was similar
in male and female rats.
Identical metabolites were also present in rat urine
1
3
samples collected between 24 and 48 h after [ C]TAME
exposure. Relative concentrations of these metabolites,
however, were different in the samples between 24 and
4
8 h compared to those in the 0-24 h sampling period,
indicating differences in excretion kinetics of the indi-
vidual metabolites (data not shown).
To confirm that human metabolism of TAME is identi-
cal to TAME metabolism in rats, one human volunteer
1
3
was exposed to [ C]TAME. To be able to use small
amounts of the expensive 13C-labeled material, [ C]-
TAME (300 µL of liquid) was added to a 2 L gas sampling
bag filled with 2 L of pure oxygen. The volunteer inhaled
the TAME-containing oxygen for 4 min and exhaled into
the sampling bag. This procedure ensured that most of
the TAME present in the sampling bag was taken up by
the volunteer. No major discomfort or symptoms of
toxicity were associated with the procedure. Urine of the
volunteer was collected in 6 h fractions for 48 h and
13
F igu r e 6. COSY-NMR spectrum of the isolated metabolite with
1
3
a
C resonance at δ 83.1. For recording NMR spectra, the
metabolite was isolated by preparative HPLC and dissolved in
D
2
O.
The third intense signal (δ 83.1) in the 13C NMR
analyzed by C NMR. The presence of identical metabo-
13
1
3
spectra of urine samples from [ C]TAME-treated rats
also slowly disappeared after acid (or glucuronidase)
treatment of the urine. This observation suggests that
the compound also represents a glucuronide. This as-
sumption is supported by electrospray mass spectra
lites of TAME observed in rats was indicated by the
recorded 13C NMR spectra (Figure 7). In contrast to the
1
3
case in rats, free [ C]-tert-amyl alcohol was present in
1
3
significant amounts in human urine after [ C]TAME
inhalation and the glucuronide of 2-methyl-2,3-butane-
diol was only a minor excretion product. All other
detected metabolites were present in the human urine
samples at concentrations similar to those seen in rat
(Figure 5) and COSY-NMR spectra (Figure 6) of the
isolated compound. The mass spectra that were recorded
were very similar to those of the glucuronide of 2-methyl-
1
3
2
,3-butanediol with a difference of m/z 16 in major
urine (based on relative signal intensities in C NMR).
1
3
fragments suggesting a glucuronide of tert-amyl alcohol.
Metabolites of [ C]TAME were also detected in the

Biotransformation of tert-Amyl Methyl Ether
Chem. Res. Toxicol., Vol. 12, No. 10, 1999 963
Sch em e 2. Biotr a n sfor m a tion of TAME in Ra tsa
F igu r e 8. 13C NMR spectrum (63 MHz, 2000 scans) of a 24 h
urine sample from a male Fischer F344 rat treated orally with
1
3
2
50 mg/kg 2-[ C]-tert-amyl alcohol. The following structural
assignments for the NMR signals were made: δ 72.8, 3-hydroxy-
3
2
-methylbutyric acid; δ 74.9, tert-amyl alcohol; δ 76.4, 2-methyl-
,3-butanediol; δ 76.7, glucuronide of 2-methyl-2,3-butanediol;
δ 79.8, 2-hydroxy-2-methylbutyric acid; and δ 83.1, glucuronide
of tert-amyl alcohol.
human urine sample collected up to 48 h after TAME
inhalation.
Biotr a n sfor m a tion of ter t-Am yl Alcoh ol in Ra ts.
Studies on the metabolism of tert-amyl alcohol were
included in this study to confirm structures of metabolites
“downstream” from the formation of tert-amyl alcohol and
to determine if any of the minor metabolites represent
products formed by oxidation of TAME at other sites in
the molecule leaving the methyl ether function intact.
a
Metabolites found in urine are underlined: (1) TAME, (2) tert-
1
3
amyl alcohol, (3) 2-methyl-2,4-butanediol, (4) 3-hydroxy-3-meth-
ylbutyric acid, (5) 2-methyl-2,3-butanediol, (6) glucuronide of
-methyl-2,3-butanediol, (7) 2-methyl-1,2-butanediol, (8) 2-hy-
droxy-2-methylbutyric acid, and (9) glucuronide of tert-amyl
alcohol.
Moreover, [ C]-tert-amyl alcohol was available from the
1
3
synthesis of [ C]TAME. Male rats (n ) 3) were treated
2
1
3
with 250 mg/kg [ C]TAME (dissolved in corn oil) by
gavage. Urine was collected in 24 h intervals for 48 h
1
3
and analyzed by C NMR.
The 13C NMR spectra of the urine samples (Figure 8)
The structures of the TAME metabolites that have
been delineated suggest a complex biotransformation of
TAME (Scheme 2). The first and major step in the
biotransformation of TAME is oxidation of the methyl
group to give an unstable hemiacetal which decomposes
to tert-amyl alcohol. The low concentrations of tert-amyl
alcohol recovered in the urine of rats exposed to both
TAME and tert-amyl alcohol suggest extensive further
metabolism of this alcohol by conjugation and by further
oxidation like that for tert-butanol (6, 17, 18). Glucu-
ronidation of tert-amyl alcohol seems to be a major step
in tert-amyl acohol biotransformation, resulting in the
excretion of a glucuronide. In addition, tert-amyl alcohol
is oxidized to several diols in reactions likely involving
cytochrome P450-catalyzed CH-based oxidation. The
major pathway of tert-amyl alcohol oxidation occurs at
the C-3 atom to give 2-methyl-2,3-butanediol, which,
including the glucuronide formed, is the major product
of TAME and tert-amyl alcohol biotransformation ex-
creted in the urine of rats exposed to TAME. Oxidation
of the carbon atom at the 3-position to the alcohol moiety
seems to be a minor process resulting in 2-methyl-2,4-
butanediol as an intermediate, which is further oxidized
to 3-hydroxy-3-methylbutyric acid.
1
3
collected after treatment of rats with [ C]-tert-amyl
alcohol were very similar to those obtained from rats
1
3
treated with [ C]TAME, suggesting that all the metabo-
1
3
lites formed from [ C]-tert-amyl alcohol are identical to
1
3
those formed from [ C]TAME. On the basis of relative
signal intensities in 13C NMR, tert-amyl alcohol glucu-
ronide and 2-methyl-2,3-butanediol and its glucuronide
1
3
are also major metabolites of [ C]-tert-amyl alcohol
excreted in urine; free tert-amyl alcohol, 2-hydroxy-2-
methylbutyric acid, and 3-hydroxy-3-methylbutyric acid
were identified by 13C NMR as minor metabolites. In the
urine samples taken 48 h after exposure, only 2-methyl-
2
,3-butanediol and its glucuronide were present, sug-
gesting rapid elimination of the tert-amyl alcohol glucu-
ronide.
Discu ssion
The biotransformation of TAME and the presumed
major metabolite tert-amyl alcohol was studied in rats.
In addition, to confirm an identical metabolism, biotrans-
1
3
formation of [ C]TAME was also studied in one human
volunteer after TAME inhalation. The aim of the study
was to identify metabolites formed and excreted to
prepare for a study on the kinetics of excretion of TAME
and its metabolites in rats and humans. The use of 13C-
labeled ethers permitted the detection and identification
of metabolites by 13C NMR and identification of metabo-
lites by mass spectrometry due to differences of 1 mass
unit in selected fragments of 13C-containing metabolites.
Oxidation of the methyl group next to the alcoholic
function in tert-amyl alcohol also represents a minor
pathway which results in the intermediate formation of
2-methyl-1,2-butanediol which may be further oxidized
to give 2-hydroxy-2-methylbutyric acid. In humans,
biotransformation of TAME after inhalation seems to be
qualitatively similar to that of rats; however, both the

9
64 Chem. Res. Toxicol., Vol. 12, No. 10, 1999
Amberg et al.
2
4 and 48 h urine samples exhibited differences in the
Refer en ces
relative signal intensities for the individual metabolites,
suggesting possible quantitative differences in routes of
TAME biotransformation between humans and rats. In
rats, on the basis of metabolite structures and semiquan-
titative comparison of signal intensities, glucuronidation
or further oxidation of tert-amyl alcohol is the almost
exclusive pathways of TAME biotransformation. In the
human volunteer, conjugation of the tert-amyl alcohol
that is formed or oxidation to the 2,3-diol seems to be
less efficient than the mechanism of rats, and metabolites
formed by oxidations in other parts of the molecule are
also excreted in larger amounts. These differences may
be due to differences in substrate specificity of P450s and
P450 profiles between rats and humans.
The structures of the metabolites that have been
elucidated and the delineated mechanisms of their
formation indicate that the formation of reactive inter-
mediates in TAME metabolism is unlikely. Thus, the
chronic toxicity of TAME should be low based on predic-
tions made from the metabolism structures. Indeed, the
available data for the toxicity of TAME support this
prediction. Rats exposed for 4 weeks to concentrations
of up to 4000 ppm TAME (6 h per day, 5 days per week)
showed only increased relative liver weights without
treatment-related histopathological findings. The major
toxic effects were severe CNS depression which is not
likely to be caused by metabolites (5).
(
1) Costantini, M. G. (1993) Health effects of oxygenated fuels.
Environ. Health Perspect. 6, 151-160.
(2) Vainiotalo, S., Pekari, K., and Aitio, A. (1998) Exposure to methyl
tert-butyl ether and tert-amyl methyl ether from gasoline during
tank lorry loading and its measurement using biological monitor-
ing. Int. Arch. Occup. Environ. Health 71, 391-396.
(
3) Health Effects Institute (1996) The potential health effects of
oxygenates added to gasoline, Health Effects Institute, Cam-
bridge, MA.
(
4) Vainiotalo, S., Peltonen, Y., Ruonakangas, A., and Pfaffli, P. (1999)
Customer exposure to MTBE, TAME, C6 alkyl methyl ethers, and
benzene during gasoline refueling. Environ. Health Perspect. 107,
133-140.
(
5) White, R. D., Daughtrey, W. C., and Wells, M. S. (1995) Health
effects of inhaled tertiary amyl methyl ether and ethyl tertiary
butyl ether. Toxicol. Lett. 82/83, 719-724.
(
6) Bernauer, U., Amberg, A., Scheutzow, D., and Dekant, W. (1998)
1
2
13
Biotransformation of C- and 2- C-labeled methyl tert-butyl
ether, ethyl tert-butyl ether, and tert-butyl alcohol in rats:
1
3
Identification of metabolites in urine by C nuclear magnetic
resonance and gas chromatography/mass spectrometry. Chem.
Res. Toxicol. 11, 651-658.
(7) Fennell, T. R., and Sumner, S. C. J . (1994) Identification of
1
3
metabolites of carcinogens by C NMR spectroscopy. Drug Metab.
Rev. 26, 469-481.
(
8) Nauhaus, S. K., Fennell, T. R., Ashgarian, B., Bond, J . A., and
Sumner, S. C. J . (1996) Characterization of urinary metabolites
from Sprague-Dawley rats and B6C3F1 mice exposed to [1,2,3,4-
1
3
C]butadiene. Chem. Res. Toxicol. 9, 764-773.
(
9) Shockcor, J . P., Unger, S. E., Wilson, I. D., Foxall, P. J ., Nicholson,
J . K., and Lindon, J . C. (1996) Combined HPLC, NMR spectros-
copy, and ion-trap mass spectrometry with application to the
detection and characterization of xenobiotic and endogenous
metabolites in human urine. Anal. Chem. 68, 4431-4435.
In summary, the results of this study show that TAME
is extensively metabolized in rats. The structures of the
metabolites excreted in urine suggest a major role for
cytochrome P450 enzymes in TAME biotransformation.
Moreover, the results obtained demonstrate the feasibil-
ity of using 13C-labeled chemicals and C NMR for
metabolism studies.
(
10) Loizou, G. D., Urban, G., Dekant, W., and Anders, M. W. (1994)
Gas-uptake pharmacokinetics of 2,2-dichloro-1,1,1-trifluoroethane
(HCFC-123). Drug Metab. Dispos. 22, 511-517.
(11) Henderson, R. F., Sabourin, P. J ., Bechtold, W. E., Steinberg, B.,
and Chang, I. Y. (1993) Disposition of inhaled isobutene in F344/N
rats. Toxicol. Appl. Pharmacol. 123, 50-61.
13
(
12) Kropf, H. (1975) Herstellung von Alkoholen durch Aufbaureak-
tionen. In Houben-Weyl. Handbuch der organischen Chemie, pp
1
021-1022, Verlag Chemie, Weinheim, Germany.
(
(
(
13) H u¨ nig, S., M a¨ rkl, G., and Sauer, J . (1979) Integriertes organisches
Ack n ow led gm en t. The research described in this
article was conducted under a contract from the Health
Effects Institute (HEI, Research Agreement 96-3), an
organization jointly funded by the United States Envi-
ronmental Protection Agency (EPA) (Assistance Agree-
ment X-816285) and certain motor vehicle and engine
manufacturers. The contents of this article do not neces-
sarily reflect the views of HEI, or its sponsors, nor do
they necesssarily reflect the views and policies of EPA
or motor vehicle and engine manufactures. Parts of this
work were also supported by the Biomed Program of the
European Union (Contract BMH4-CT96-0184). We thank
Dr. Zafir Alam (Thermoquest Ltd., Hamil Hampstead,
U.K.) for recording LC/MS/MS spectra.
Praktikum, Verlag Chemie, Weinheim, Germany.
14) Meerwein, H. (1975) A¨ ther. In Houben-Weyl. Handbuch der
organischen Chemie, p 14, Verlag Chemie, Weinheim, Germany.
15) Nicholson, J . K., and Wilson, I. D. (1987) High-resolution nuclear
magnetic resonance spectroscopy of biological samples as an aid
to drug development. Prog. Drug Res. 31, 427-479.
(16) Aldrich (1998) Aldrich Library of FT-NMR, 1(1) 817B, Adrich
Chemical Co., Steinheim, Germany.
(
17) Cederbaum, A. I., and Cohen, G. (1980) Oxidative demethylation
of tert-butyl alcohol by rat liver microsomes. Biochem. Biophys.
Res. Commun. 97, 730-736.
(18) J ohanson, G., Nihlen, A., and Lof, A. (1995) Toxicokinetics and
acute effects of MTBE and ETBE in male volunteers. Toxicol. Lett.
8
3, 713-718.
TX990078X