A comparison of 1H8- and 2H8-toluene toxicokinetics in men

Article abstract of DOI:10.1080/004982599238830

1. To examine the bioequivalence of an isotope-labelled tracer to study toxicant disposition, we conducted 33 controlled human exposures to a mixture of 50 ppm ¹H₈-toluene and 50 ppm ²H₈-toluene for 2 h, and measured concentrations in blood and breath, and metabolite levels in urine for 100 h post-exposure. 2. A physiologically based kinetic (PBK) model found that compared with ¹H₈-toluene, ²H₈-toluene had a 6.4 ± 13% (mean ± SD) lower AUC, a 6.5 ± 13% higher systemic clearance (1.46 ± 0.27 versus 1.38 ± 0.25 l/h-kg), a 17 ± 22%, larger terminal volume of distribution (66.4 ± 14 versus 57.2 ± 10 l/kg) and a 9.7 ± 26% longer terminal half-life (38 ± 12 versus 34 ± 10 h) (p < 0.05 for all comparisons). 3. The higher ²H₈-toluene clearance may have been due to an increased rate of ring oxidation, consistent with the 17% higher observed fraction of ²H₅- versus ¹H₅-cresol metabolites in urine. 4. The larger terminal volume and half-lives for ²H₈-toluene suggested a higher adipose tissue/blood partition coefficient. 5. Observed isotope differences were small compared with interindividual differences in ¹H₈-toluene kinetics from previous studies. 6. The PBK model allowed us to ascribe observed isotope differences in solvent toxicokinetics to underlying physiologic mechanisms.

Full text of DOI:10.1080/004982599238830

xenobiotica , 1999, vol. 29, no. 1, 93±108
1
2
A comparison of H8- and H8-toluene toxicokinetics
in men
C. H. PIERCE‹*, T. A. LEWANDOWSKI‹, R. L. DILLS‹,
M. S. MORGAN‹, M. A. WESSELS‹, D. D. SHENŒ
and D. A. KALMAN‹
‹
Departments of Environmental Health and Pharmaceutics, University of
Œ
Washington, Seattle WA 98195, USA
1. To examine the bioequivalence of an isotope-labelled tracer to study toxicant
"
disposition, we conducted 33 controlled human exposures to a mixture of 50 ppm H -
)
toluene and 50 ppm H -toluene for 2 h, and measured concentrations in blood and breath,
#
)
and metabolite levels in urine for 100 h post-exposure.
"
2. A physiologically based kinetic (PBK) model found that compared with H -toluene,
)
#
H -toluene had a 6.4‰ 13 % (mean‰ SD) lower AUC, a 6.5‰ 13 % higher systemic
)
clearance (1.46‰ 0.27 versus 1.38‰ 0.25 l h-kg), a 17‰ 22 % larger terminal volume of
distribution (66.4‰ 14 versus 57.2‰ 10 l kg) and a 9.7‰ 26 % longer terminal half-life
}
}
!
(38‰ 12 versus 34‰ 10 h) (p 0.05 for all comparisons).
#
3. The higher H -toluene clearance may have been due to an increased rate of ring
)
oxidation, consistent with the 17 % higher observed fraction of H - versus H -cresol
metabolites in urine.
#
"
&
&
#
4. The larger terminal volume and half-lives for H -toluene suggested a higher
)
adipose tissue blood partition coeæcient.
}
5. Observed isotope diåerences were small compared with interindividual diåerences
"
in H -toluene kinetics from previous studies.
)
6. The PBK model allowed us to ascribe observed isotope diåerences in solvent
toxicokinetics to underlying physiologic mechanisms.
Introduction
Stable isotope-labelled compounds serve as valuable probes to understand in
vivo drug and toxicant disposition by obviating contributions from environmental
and endogenous sources (Pierce et al. 1997). Examination of the toxicokinetics and
eåects from occupational or controlled exposures to toxicants can be confounded by
"
background contributions. In the case of H -toluene, studies with controlled
)
human exposures have assumed that blood concentrations of & 100 nmol l were
}
"
well above background and solely due to administered H -toluene (Veulemans and
)
Masschelein 1978, Nise et al. 1989). Occupational studies have attributed blood
levels down to 12±110 nmol l to workplace exposure alone (Kawai et al. 1992a±c,
}
1994, Mannino et al. 1995). However, average concentrations in blood from non-
occupationally exposed groups have been measured as 3±14 nmol l, with a range of
}
up to 206 nmol l (Brugnone et al. 1989, Ashley et al. 1994, Etzel and Ashley 1994,
}
Brugnone et al. 1995, Fustinoni et al. 1995, Vrca et al. 1995). Background exposures
may therefore confound controlled and occupational exposure studies. One way to
address this concern is through the use of stable isotope-labelled probes, a technique
commonly employed in studying the in vivo disposition of drugs (Falconnet et al.
1988).
* Author for correspondence.
Xenobiotica ISSN 0049±8254 print}ISSN 1366±5928 online ’ 1999 Taylor & Francis Ltd
http:}}www.tandf.co.uk}JNLS}xen.htm http:}}www.taylorandfrancis.com}JNLS}xen.htm

94
C. H. Pierce et al.
Studies using isotopic tracers often assume identical disposition kinetics to the
native compound, yet this assumption has not often been examined, particularly for
toxicologic studies. Previous work has found that a deuterium label did not aåect
trazodone kinetics following oral dosing (Gammans et al. 1984), and that there were
#
no diåerences in H-labelled versus unlabelled lidocaine or bupivacaine kinetics
following i.v. infusion (Burm et al. 1988). Browne et al. (1982) detected no
"&
"$
diåerences in phenobarbital kinetics with the use of N and C labels. Others,
however, have conducted tracer studies assuming, but not testing, kinetic isotope
equivalence (Lof et al. 1993, Jones et al. 1994).
"
#
To assess the kinetic diåerences between H -toluene and H -toluene, we
)
administered 50 ppm of each compound as a mixture for 2 h to 26 subjects. Blood
)
"
and breath concentrations of H - and H -toluene, and urinary concentrations of
#
)
metabolites were measured in periodic samples taken for up to 100 h post-exposure.
)
By applying a previously developed physiologically based kinetic (PBK) model for
#
H -toluene (Pierce et al. 1996a), we compared the areas under the blood
)
concentration-time curves, clearances, and terminal volumes of distribution and
half-lives of the labelled and native compounds. Urinary excretion rates of the major
metabolite hippuric acid, and of the minor cresol metabolites were used to provide
an independent assessment of isotopic metabolic diåerences.
M aterials and m ethods
Chemicals
#
#
H -toluene (99.6 g-atom %), and H -o-, m-, and p-cresols (98 g-atom %, used as standards) were
)
obtained from Cambridge Isotope Laboratories (Woburn, MA, USA). Toluene (distilled in glass grade)
)
was from Burdick and Jackson (Muskegon, MI, USA) and unlabelled metabolites were obtained from
#
Aldrich (Milwaukee, WI, USA). H -hippuric acid was synthesized by reacting H -benzoyl chloride
#
&
with glycine ethyl ester (2-fold molar excess based on H -benzoic acid). The reaction mixture was
&
#
&
extracted with ether, and the ether was dried with Na SO and evaporated. The residue was recrystalized
#
%
from hexane-chloroform and had a m.p. of 53 ÊC. Mass spectrometry gave the following peaks: m z 212
}
+,
(M 10 %), 166 ([M- HCO ] 5 %), 139 ([M-CO CH H ] 17 %), 110 ([ H PhCO] 100%), and 82
+,
+,
+,
#
#
#
#
#
$
&
+,
([ H Ph] 32 %).
#
&
Subjects
Twenty-six men of ages 20±62 years of age were recruited by advertisement on campus and given a
self-administered questionnaire to screen for occupational solvent exposure. Five of these subjects
participated in two or three replicate exposures, with at least 2 weeks between exposures, for a total of 33
exposures. Smokers, those with chronic illness or on chronic medication, and those with occupational
solvent exposure were excluded from the study. Individuals with anaemia, acute respiratory infection,
rhinitis or asthma were given medical evaluations before entering the study ; all subjects were Caucasian.
Subjects provided written consent, and recruitment, exposure and sampling protocols were approved by
the Institutional Review Board of the University of Washington.
Weight (67±129 kg) and height (1.65±1.98 m) were measured, and the fraction of the body that was
adipose tissue (0.10±0.39 kg kg), was estimated by a skin fold method using Lange calipers (Durnin and
}
Womersley 1974). Alveolar ventilation rates (4.5±9.7 l min), estimated as 70 % of total ventilation, and
}
"
#
inhaled H -toluene concentrations (43±71 ppm) and H -toluene concentrations (45±68 ppm) were also
)
measured (Morgan et al. 1993). Person-speci®c H -toluene blood air partition coeæcients were
)
#
}
)
determined previously (Pierce et al. 1996a) and were used for modelling both the native and deuterated
compounds.
Exposures and sampling
Inhaled toluene concentrations were generated using a controlled screw drive which compressed a
syringe to inject solvent into the airstream. Subjects inhaled 100 ppm of an approximately equimolar
"
#
mixture of H -toluene and H -toluene for 2 h through a gated mouthpiece. A sample of inspired gas
)
from each exposure was collected for measurement of exact H - and H -toluene concentrations. The
)
"
#
)
exposure system provided real time measurement of inhaled and exhaled concentrations of total toluene,
)
respiratory ¯ow rate and exhaled CO concentrations (Morgan et al. 1993).
#

Toluene toxicokinetics in man
95
#
Figure 1. Physiologically based toxicokinetic model for inhaled H -toluene. Q
=
alveolar ventilation
alv
)
=
=
concentration in alveolar air
(litres air h) ; C
concentration in inhaled air (ppm) ; C
}
inh
(ppm) ; FP ®rst-pass metabolism ( mol h); Q
alv
=
cardiac output (litres blood h); C
=
centration in arterial blood ( mol l); C
l
con-
=
}
}
co
art =
l
=
l
concentration in venous blood ( mol l); Cl
}
}
ven
intrinsic metabolic clearance in the lung (l h); Cl
int- x
intrinsic metabolic clearance in the liver
=
}
int- h
blood ¯ow rate to tissue group i (l h); C
=
=
concentration in blood leaving tissue
group i (mg l). Subscripts (i) for tissue groups or compartments : a, adipose tissues ; r, rapidly
(l h); Q
}
}
i
vi
}
perfused tissues ; s, slowly perfused tissues ; l, liver.
One antecubital venous blood sample and one breath sample were taken prior to exposure for
"
determination of pre-exposure H -toluene levels; 16 simultaneous blood and breath samples were taken
)
after the end of exposure over the following 4 days, at sampling intervals that varied from every 15 min
immediately following exposure to every 12 h at times & 24 h after exposure. Blood samples were
collected in 5 ml Vacutainer tubes containing citrate. Exhaled breath samples were collected in 20-litre
Tedlar plastic bags pre®lled with 10 litres of dry nitrogen. Urine samples were taken prior to,
"
immediately following, and 4 and 8 h after exposure ; then with blood and breath samples at times 24 h
after exposure.
Chemical analyses
Blood samples were analysed in triplicate for H -toluene and H -toluene by a headspace method
"
#
)
(Dills et al. 1991), using gas chromatography with mass spectrometry (GC MS) detection in selected ion
)
}
mode. Low-level contamination of blood samples through contact with Vacutainer vials or catheters

96
C. H. Pierce et al.
Table 1. Values of parameters used in the physiologically based kinetic model.
Tissue group
Physiologic
parameter
Slowly perfused
Rapidly perfused
Liver
Adipose
Volume (V, litres)
0.95 BW - V
*
0.05 BW- V
liver
*
0.023 BW*
0.10±0.39 BW
(measured)
0.06±0.18 Q
(®tted to data)

55.9
adipose
‹
‹
Blood ¯ow
(Q, l h)
0.24 Q - Q
0.76 Q - Q
co
0.27 Q *
co
co
adipose
liver
co
}
Œ
Œ
Œ
K
1.54
4.64
4.64
5.97
p
l
( mol l)
( mol h)
‹
K
V
±
±
±
±
±
±
}
m
.
! ( ‹Ž
s
l
52.1 BW
}
max
.
! (%
, based on similar scaling of
=
=
=
BW body weight (kg) and Q
cardiac output (l h) 12.92 BW
}
co
cardiac output and alveolar ventilation rate (Tardifet al., 1993) and measurements of ventilation rate in
our subjects.
* From Rowland and Tozer (1989).
‹
Œ

From Tardif et al. (1993).
From Gargas et al. (1989).
From Pierce et al. (1996b).
Scaling of body weight from (Mordenti 1986).
Ž
s Maximal extrahepatic rate of metabolism was constrained to a range of 0±70 % of V
model-®tting process.
during the
max- h
"
compromised the H -toluene concentrations in pre-exposure and late post-exposure blood samples,
)
hence the H -toluene blood data were not used. The contents of breath sample bags were drawn through
"
)
two-section charcoal sorbent tubes using a calibrated personal sampling pump. The charcoal sections
"
were then analysed (US Department of Health 1985). Alveolar H - and H -toluene concentrations were
#
)
)
calculated from the exhaled breath concentrations by multiplying the levels in the collection bags by
[alveolar CO ] [bag CO ], where the alveolar CO values were measured prior to exposure, or were
}
#
assumed to be the standard value of 5.26%.
#
#
A
unsubstituted o-, m- and p-cresols was developed with a quantitation range of 0.0092±4600 mol l (Dills
sensitive gas chromatography-electron capture detection assay for phenolic deuterated and
l
}
b
et al. 1997). Cresol glucuronides and sulphates in urine were hydrolysed with -glucuronidase and
arylsulphate, extracted with tert-butyl methyl ether and benzene, and derivatized with hepta-
"
¯urorobutyric anhydride. H -Cresol levels in pre-exposure urine samples were 0.211‰ 0.32, 0.500‰ 0.66
)
l
=
and 333‰ 290 ( mol l, mean‰ SD, n 38) for the o-, m- and p-cresols, respectively. Due to the high level
}
of background p-cresol concentrations in urine from food products and amino acid metabolism, only
levels of the native and deuterated o- and m-isomers were compared.
Assays for hippuric acid were performed on duplicate 1 ml aliquots of urine. Samples were spiked
l
with 50 l 0.2 mg ml N-(3-methylbenzoyl)-alanine in ethanol (recovery standard), acidi®ed to pH 2, and
}
extracted twice with 2 ml ethyl acetate. The extracts were combined and evaporated to dryness under
vacuum. Ethyl esters of the hippuric acids were then formed by reaction at 80 ÊC for 30 min with acidi®ed
l
ethanol (100 l 1 M acetyl chloride in absolute ethanol). The reaction mixture was cooled and ethanol was
removed by evaporation under vacuum. Ethyl acetate (0.5 ml) and internal standard (50 l 2 mg ml
l
}
methyl decanoate) were added. Samples were then analysed by gas chromatography mass spectrometry
}
#
using selected ion monitoring (m z 110 for ethyl H -hippurate and m z 134 for ethyl H -hippurate).
"
}
}
&
While both H - and H -hippuric acid were measured in urine samples, relatively high contributions of
&
"
#
&
dietary and endogenous sources of H -hippuric acid were expected to preclude its use as a quantitative
&
"
&
marker of exposure (ACGIH 1991).
Physiologically based model
A previously described semiempirical PBK model (Pierce et al. 1996a) (®gure 1) was implemented
with SimuSolv software (Dow Chemical Co., Midland, MI, USA) using subject-speci®c values of body
"
weight, adipose tissue fraction, H -toluene blood air partition coeæcient, exposure concentration and
}
)
alveolar ventilation rate. Values of fractional tissue compartment volumes, fractional blood ¯ows,
tissue blood partition coeæcients and hepatic metabolic constants V
and K were taken from
}
max- h
m
literature sources (table 1). A minimum number of tissue compartments (®ve) was used to represent
toluene distribution in the body. As a small lipophilic molecule, toluene was expected to rapidly diåuse
across tissue membranes. The exchange of toluene between arterial blood and tissue groups was therefore
treated as being ¯ow-limited. Cardiac output (CO), and therefore all blood ¯ows, were scaled to (body
.
! (%
weight)
(Mordenti 1986, D’Sousa and Boxenbaum 1988, Ings 1990, Ritschel et al. 1992).

Toluene toxicokinetics in man
97
"
#
Figure 2. Measured concentrations of H -toluene ( 1) and H -toluene (O) in alveolar breath,
)
)
$
#
l
l
( mol m ) and H -toluene in blood (x , mol l) from 33 exposures to 50 ppm each for 2 h.
}
}
)
Because fractional blood ¯ow to the adipose tissue (Q
)
and maximal rate of extrahepatic
a/co
metabolism (V ) are diæcult to measure, may vary widely between individuals, and are important
max- x
determinants of toluene blood concentration (Pierce et al. 1996a), these parameters were ®tted to each of
"
#
the H - and H -toluene data sets. Based upon previous measurements and estimates of Q
(Astrand
a/co
)
)
and Rodahl 1986, Rowland and Tozer 1989, Leggett and Williams 1991, Ferrannini 1992, Davis and
Mapleson 1993), this model parameter was expected to vary within the bounds of 0.04±0.17. As a test of
" #
was ®t independently to the H - and H -toluene data for each exposure ; no diåerence
) )
the model, Q
a/co
was expected for this adipose blood ¯ow value. A search range for values of V
max- x
observation by Wheeler et al. (1992) that human lung microsomes contain 10.5% of the liver microsomal
was de®ned using the
activity of CYP2E1. Thummel et al. (1993) found about a 7-fold range of CYP2E1 content
(49±372 pmol nmol protein) in 12 human liver samples examined in vitro. Because there may be some
}
CYP2E1 activity in locations other than the liver and lung, and because other P450 isozymes (e.g.
CYP1A1; Kawamoto et al. 1995) may participate in toluene metabolism (Nakajima and Wang 1994) we
expected maximal extrahepatic rate of metabolism to range from 0 to 70 % of that in the liver. K and
"
m
in the liver were assumed to be the same for H - and H -toluene, based upon the relative
#
V
max- h
insensitivity of modelled blood concentrations to these parameters because of ¯ow rate-limited clearance
)
)
by the liver (Pierce et al. 1996a). Using published K , V and blood ¯ow to the liver (Rowland and
Tozer 1989, Tardif et al. 1993), hepatic metabolism was found to be much less aåected than extrahepatic
m
max- h
metabolism to small changes in V
.
max
Model goodness-of-®t
The model was ®tted to the breath ( H -toluene) or blood and breath ( H -toluene) data from each
"
#
)
to maximize the log likelihood function (LLF) (Steiner et al. 1990):
)
exposure by varying V
max- x
and Q
a/co
n
n
–
fic
#
1
1
(zi fi )
=
p
3
c3
1 logfi
LLF
n(log(2 ) 11) 1nlog
29
9n
:
:
i
i
=
=
"
"
=
where n number of measurements at each time point, zi
=
measured value of the ith data point, fi

98
C. H. Pierce et al.
=
c =
c
heteroscedasticity (weighting) parameter. was optimized
predicted value of the ith data point and
over the range 0±2 during model ®tting. Almost every ®t produced a value of 2, consistent with our
constant coeæcient of variation in assay error.
The model goodness-of-®t to each of the 33 sets of data was measured by the fraction of variation in
the data that was explained by the model. Similar to the calculation of the coeæcient of multiple
#
determination (r ), this value was computed using the following equation (Steiner et al. 1990):
n
n
–
fic
#
(zi fi )
zi
3
3
c
/
#
model error
total error
fi
i
i
=
n
=
"
"
–
–
# =
=
=
, where z
r
1
1
/(c/ )
"
f i
–
#
#
3
(zi z)
3
i
fic
=
"
i
=
"
The model was also used to determine the area under the blood concentration-time curve (AUC),
systemic clearance (Cl ), and the terminal volume of distribution (V ) and half-life (t ), using the
following equations :
s
z
/ -z
" #
¢
=
AUC
C dt
v
&
l
!
=
absorbed dose AUC
where C
=
modelled venous blood concentration ( mol l).
}
v
Cl
}
s
&
=
3
v
A
C
}
t,i
v
z
i
=
"
=
where i tissue group, A
=
l
=
amount of toluene in tissue in terminal phase ( mol) and C venous
t
blood concentration in terminal phase ( mol l)
v
l
}
C ln2
v
=
t
/
" #Õ
z
dC dt
}
v
"
were determined in the terminal phase ( 40 h post-exposure).
/ -z
" #
V
andt
z
Diåerences in 1H8-toluene and 2H8-toluene kinetics
"
To evaluate kinetic diåerences between H - and H -toluene, two con¯icting
#
)
)
concerns were addressed: the post-exposure time interval over which the model was
®tted had to be long enough to characterize adequately all phases of toluene
disposition but short enough to minimize the confounding eåect of background
exposure on ®tting the terminal phase. To this end, we visually examined the data
"
from all subjects en masse to determine when the breath concentrations of H - and
)
H -toluene began to diverge, i.e. when background exposure to H -toluene
#
"
)
signi®cantly contributed. We also repeatedly varied the model parameter
)
representing background concentration around a value of 50 ppb (Pierce et al. 1997)
"
to observe its contribution to predicted H -toluene breath levels.
)
Once the optimum time interval for ®tting the model was determined, the model
"
was optimized to the H -toluene and H -toluene data separately, for each exposure,
#
)
and Q . Values of these ®t model parameters and the kinetic
)
by varying V
max- x a/co
parameters AUC, Cl , V , and t
"
for H - and H -toluene were compared within
#
s
z
/ -z
" #
exposures using paired t-tests. A signi®cance level of p 0.05 for all tests was
) )
!
chosen.
"
As a second, model-independent method to evaluate diåerences in H - and H -
#
)
)
toluene clearances, apparent metabolic formation clearances to the o-, m- and p-
cresol metabolites were determined using the expressions (Rowland and Tozer
1989),

Toluene toxicokinetics in man
99
=
Cl
Rate of excretion
blood concentration
}
s-metabolite
metabolite,
metabolite,
t
t
toluene, t
toluene, t
#
for H -toluene, and
)
=
Cl
Rate of excretion
breath concentration
}
s-metabolite
"
for H - and H -toluene.
#
)
)
Use of these expressions required three assumptions : (1) the initial oxidation
step in toluene metabolism is slower than subsequent metabolic steps, which is
supported by observed reduced rates of metabolism following co-administration
with benzene, styrene, n-hexane, trichloroethylene or tetrachloroethylene in animal
studies (Snyder 1987) ; (2) toluene metabolite appearance in urine is not governed by
glomerular ®ltration rate (GFR), an assumption which was tested by comparing
Cl
values with a standard value for GFR ; and (3) toluene metabolite
s-metabolite
elimination occurs through urinary excretion, which has been determined previously
(Snyder 1987). Use of the second equation was based on a rapid equilibration
"
between blood and breath toluene levels (Pierce et al. 1996a), and equivalent H -
)
#
and H -toluene blood air partition coeæcients (Pierce et al. 1996b).
}
)
Using these expressions to evaluate isotope diåerences in cresol formation
clearances, 157 Cl
-Cl
simultaneous pairs were compared, and
s- H -o-cresol
s- H -o-cresol
# (
pairs were compared, based on the data from all 33
"
-Cl
(
172 Cl
s- H -m-cresol
s- H -m-cresol
# (
"
exposures.
(
Results
"
#
H - and H -toluene breath concentrations
)
)
The breath data in ®gure 2 indicated that after 40 h H -toluene concentrations
#
)
decreased while H -toluene levels began to plateau, an observation consistent
"
)
with expected contributions of background exposure (Pierce et al. 1997). Model
simulations suggested that a nominal 50 ppb background level of exposure was
C
"
expected to result in a H -toluene breath concentration of
$
0.5 lmol m , similar
}
)
$
to the measured 0.334 ‰ 225 and 0.764‰ 598 lmol m found in groups described as
}
non-occupationally exposed (Brugnone et al. 1989). The average observed breath
levels in the 2±40-h post-exposure period were thus 5±300 times higher than
expected background concentrations (®gure 2). We therefore considered blood and
breath concentrations in the 2±40-h period to be due to the controlled exposure, and
optimized the physiologic model to describe these data only. Similarly, the pairing
of blood and urine, and breath and urine samples to determine apparent formation
clearances to the cresols was done only over the 2±40-h period.
Optimized model parameters
"
The optimized
V
averaged 6.7‰ 4.3 lmol h-kg for H -toluene and
}
max- x
#
)
9.6‰ 5.4 lmol h-kg for H -toluene (table 2) ; and all were within the expected
}
)
physiologic bounds (0±70 % of V
, or 0±36 lmol kg-h) (Wheeler et al. 1992,
}
max- h
#
for H -toluene were signi®cantly higher than
Thummel et al. 1993). The V
!
max- x
0.05). The ®t values of Q
)
"
those for H -toluene (p
"
were 0.09‰ 0.03 for H -
a/co
)
)
#
toluene and 0.08‰ 0.03 for H -toluene; they were not signi®cantly diåerent. Two
)

100
C. H. Pierce et al.
"
Table 2. Comparison of H - and H -toluene model parameters.
#
)
)
"
#
H -toluene
Model parameter
H -toluene
)
)
Fitted
Õ %
6.7‰ 4.3 [2.33 10 ±14]
0.09‰ 0.03 [0.015±0.15]
l
( mol h-kg)
V
Q
9.6‰ 5.4 [0.34±26]**
0.08‰ 0.03 [0.03±0.14]
}
max- x
a/co
Toxicokinetic
l
AUC ( mol-h l)
Cl (l h)
12.2‰ 2.8 [7.75±19.1]
117‰ 19 [89.2±170]
4960‰ 1500 [2170±8890]
34.3‰ 10 [14.1±61.2]
11.3‰ 2.4 [6.92±19.1]**
123‰ 18 [97.8±173]*
5740‰ 1600 [2520±9270]**
37.5‰ 12 [13.4±67.6]*
}
}
S
V (l)
t
z
(h)
/ -z
" #
Goodness of ®t (% )
94‰ 2.7% [85±98 %]
90‰ 6.5% [68±98 %]**
!
!
p
0.05, **p 0.01
Values are mean‰ SD (range).
of these 66 values were below the expected physiologic range of 0.04±0.17 (Astrand
and Rodahl 1986, Rowland and Tozer 1989, Leggett and Williams 1991, Ferrannini
1992, Davis and Mapleson 1993) ; for one of these two subjects only limited data (one
#
H -toluene breath and two blood samples) were available in the terminal phase,
)
where adipose blood ¯ow is most in¯uential on blood and breath concentrations.
"
The model was able to explain 94‰ 2.7 % of the H -toluene data variability, and
)
#
90‰ 6.5 % of the H -toluene data variability in the 33 exposure sets. This signi®cant
)
0.05, paired t-test) was likely a consequence of ®tting the model to
!
diåerence (p
"
the breath data for H -toluene compared with ®tting to the blood and breath data
)
#
for H -toluene.
)
Kinetic model parameters
"
Signi®cant diåerences between H - and H -toluene were observed for all
#
)
)
#
kinetic parameters (table 2). The H -toluene AUC were on average 6.4‰ 13 %
)
"
smaller than those for H -toluene and Cl were 6.5‰ 13 % larger. In the evaluation
S
, the H -toluene values for one subject were
)
"
"
of V and t
2 SD larger than the
z
means, re¯ecting a ®nal terminal phase breath measurement about half that expected
/ -z
" #
)
from previous values. V and t
for this subject were therefore not used in the
#
z
analysis. A typical model ®t to H - and H -toluene breath concentrations in the
/ -z
" #
"
)
same subject is presented in ®gure 3.
)
#
The terminal volume of distribution (V ) was 17‰ 22 % higher for H -toluene
z
compared with H -toluene V (p 0.01 ; table 2) ; dividing by body weight gave
)
!
"
z
)
#
66.4‰ 14 versus 57.2‰ 10 l kg. Similarly, t
was 9.7‰ 26 % longer for H -
}
/ -z
" #
)
!
toluene (38‰ 12 versus 34‰ 10 h ; p 0.05).
Urinary cresol data
Both peak and 0±40 h average excretion rates (®gure 4), and peak urinary
#
concentrations (table 3) were higher for H - compared with H -o- and m-cresols.
"
(
Intra-subject paired t-tests similarly found higher levels of deuterated cresol
(
!
"
0.05). The data for H -p-cresol were of limited value due to high
metabolites (p
(
intra-individual variability, presumably due to dietary and aromatic amino acid

Toluene toxicokinetics in man
101
"
#
Figure 3. Representative measured and modelled concentrations of H -toluene ( 1, ---) and H -
toluene (O, ÐÐ) in alveolar breath from an exposure of 50 ppm each for 2 h.
)
)
"
#
Figure 4. Rates of urinary excretion of H -( 1) and H -cresols (O) from 33 exposures to 50 ppm of
(
H -toluene and H -toluene each for 2 h. Samples are plotted at mid-point collection times
(
"
#
)
relative to exposure.
)

102
C. H. Pierce et al.
Table 3. Urinary cresol sample concentrations and formation clearance rates.
"
#
H -toluene
Metabolite
Urinary concentration
H -toluene
)
)
mean‰ SD [range]
l
o-Cresol ( mol l)
}
Background
Peak
0.21‰ 0.32
1.03‰ 0.70 [0.24±3.72]
±
3.86‰ 2.1 [0.65±10.8]
l
m-Cresol ( mol l)
}
Background
Peak
0.50‰ 0.66
1.15‰ 1.0 [0.081±5.14]
±
6.37‰ 2.9 [1.15±16.1]
Form ation clearance rate
(CL
)
mean‰ SD, [n]
s-metabolite
o-Cresol (litres blood h)
}
±
0.55‰ 0.41 [137]
10.6‰ 8.7 [160]
0.95‰ 0.66 [151]
18.8‰ 15[173]
24.5‰ 15 [153]
474‰ 320 [178]
(litres breath h)
}
4.80‰ 6.3 [175]
±
9.83‰ 21 [176]
±
±
m-Cresol (litres blood h)
}
(litres breath h)
}
p-Cresol (litres blood h)
}
(litres breath h)
}
!
All H - versus H -toluene diåerences were signi®cant (p 0.05).
"
#
)
)
catabolism sources of these metabolites (®gure 4). Neither the apparent formation
clearance of toluene to o- and m-cresol (0.55 and 0.95 l h ; table 3) or to p-cresol
}
(24.5 l h, suggesting tubular secretion) was governed by an assumed glomerular
}
®ltration rate of 7.5 l h.
}
"
#
The combined use of breath concentrations of H - and H -toluene paired with
)
urinary excretion rates of H - and H -o- and m-cresol metabolites revealed
)
"
#
(
that were 3.59‰ (3.92 times those of Cl
(p
0.05), and
!
Cl
Cl
s- H -o-cresol
# (
s- H -o-cresol
" (
!
that were 4.71‰ 3.87 times those of Cl
(p 0.05). The
s- H -m-cresol
" (
s- H -m-cresol
# (
average formation clearances to the labelled metabolites were about twice as high as
to unlabelled metabolites (table 3). Using the paired values, these diåerences
amounted to an increase of 0.40 litres blood h to o-cresol and 0.75 litres blood h to
}
}
#
m-cresol for H -toluene compared with H -toluene. Using the observed
"
)
of 24.5 l h and assuming that H -toluene is cleared 4.15
)
#
Cl
}
s- H -p-cresol
# (
)
"
([3.59 14.71] 2) times more rapidly than H -toluene to p-cresol, an 18.7 l h
}
}
)
increase would be expected. Thus the net increase in H - versus H -toluene Cl
#
"
s
)
)
based on cresol formation was 19.9 l h (0.40 10.75 118.7), a 17 % increase over the
}
"
H -toluene Cl of 117 l h.
}
s
)
Discussion
Stable isotope labels have been used in a wide variety of studies, including the
evaluation of zinc, iron, selenium and copper absorption (Couzy et al. 1993, Boza et
al. 1995) ; the measurement of gastrointestinal transit time (Heine et al. 1995) ; the
determination of steroid pharmacokinetics (Kasuya 1994) ; the evaluation of
transdermal nitroglycerin absorption (Kochak et al. 1992) ; and the toxicokinetics of
lead in the rat (Smith et al. 1992). While less of a concern for higher mass isotopes,
most of these studies have assumed, but not tested, a kinetic equivalence between
labelled and unlabelled substances. Jones et al. (Jones et al. 1994) suggested the use
#
of H -toluene as a toxicokinetic tracer without consideration of possible isotope
)
eåects, and Lof et al. (Lof et al. 1993) conducted human exposures to H -toluene
#
)
assuming kinetic equivalence with the native toxicant. This is the ®rst study of

Toluene toxicokinetics in man
103
"
#
Figure 5. Rates of urinary excretion of H -hippuric acid ( 1) and H -hippuric acid (O) from 33
&
exposures to 50 ppm of H -toluene and H -toluene each for 2 h. Samples are plotted as mid-
&
"
#
)
point collection times relative to exposure.
)
which we are aware, aside from our preliminary observations (Morgan et al. 1993),
to compare the kinetics of a stable isotope-labelled and unlabelled toxicant in
humans.
#
Model ®tting to the data indicated a 6.5 % higher systemic clearance of H -
)
toluene compared with H -toluene, a small diåerence compared with the 2-fold
"
)
interindividual variability observed. The estimated increase in metabolic clearance
based on formation of cresol metabolites was 17 % ; the diåerence in the two
#
estimates may have been due to a decrease in formation clearance of H - relative to
)
H -toluene, to hippuric acid. While we were not able quantitatively to evaluate this
"
)
diåerence due to endogenous and dietary sources of urinary H -hippuric acid, an
"
&
examination of excretion rates supports this explanation. In ®gure 5, the increase
C
400 lmol h (200±600).
#
from background to peak levels of H -hippuric acid excretion is
150 lmol h
}
&
(0±150), whereas the increase of H -hippuric acid is
C
"
}
&
Dividing by each subject’s body weight, our clearances were 1.46‰ 0.27 l h-kg
}
#
"
for H -toluene and 1.38‰ 0.25 l h-kg for H -toluene, both very similar to 15
}
)
)
previous studies which found an average clearance of 1.54‰ 0.44 l h-kg (range
}
"
0.9±2.3 l h-kg) in groups of men experimentally exposed to H -toluene (Wallen et
}
)
al. 1984, 1985, Wallen 1986, Hjelm et al. 1988, Lof et al. 1990, Hjelm et al. 1994).
In these previous studies, there was an average intersubject coeæcient of variation
of 38 %, with a range of 10±50 %. This degree of imprecision suggests that isotopic
diåerences in human toluene clearance are small compared with other sources of
variability.

104
C. H. Pierce et al.
Deuteration has been shown to reduce the rate of alkyl C-H bond oxidation by
#
P450 enzymes, primarily due to the increased isotopic mass and a more stable C- H
bond (Langenhove 1986). White and McCarthy (1986) found a deuterium isotope
eåect of 2.6 in the hydrogen abstraction from toluene using an in vitro rabbit liver
system. A pronounced decrease in P450-mediated formation of carbon monoxide
#
"
following inhalation of
H
versus H -dichloromethane in mouse has also been
-
# #
observed (Andersen et al. 1994). Van Beerendonk et al. (1995) have reported that a
completely deuterated form of the ¯ame retardant tris (2,3-dibromopropyl)
"
phosphate showed a
3-fold decrease in cytochrome P450 metabolism in the
isolated perfused rat liver ; there was no signi®cant change in the P450 independent
glutathione S-transferase detoxi®cation pathway. Similar in vitro results were seen
with deuterated ethylbenzene, where deuteration of the alkyl side chain resulted in
an approximate 55 % decrease in formation of phenyl ethanol (McMahon and
Sullivan 1966). The apparent partial shift from hippuric acid to cresol metabolites
observed in this study for the deuterated compound compared with the proteated
analogue is consistent with these observations, given that methyl group oxidation is
expected to be the rate-limiting step in hippuric acid formation.
Cytochrome P450-mediated ring oxidation occurs by addition of oxygen through
an arene oxide mechanism, where C-H bond cleavage is no longer the rate-limiting
step (Langenhove 1986). Based on deuteration experiments with a number of
aromatic compounds, Tomaszewski et al. (1975) suggested that oxidation of the
aromatic ring would not be aåected by deuteration, since this occurs by direct
addition of oxygen to the C-C bond rather than by breakage of the C-H bond, as
would occur during oxidation of a methyl side chain. A study in humans using
labelled and unlabelled carbamazepine (Browne et al. 1994), where ring oxidation is
the rate-limiting step in metabolism, revealed no pharmacokinetic diåerences. In
#
our study, the increased H - versus H -(o- and m-) cresol production rates were
"
&
indicative of a higher fractional rate of ring oxidation.
&
There is some previous evidence of increased overall rate of toluene metabolism
following deuteration. Previous research (Ling and Hanzlik 1989) determined that
stepwise deuteration of the toluene methyl group in vitro led to a stepwise increase
in overall oxidation with a decrease in the yield of the benzyl alcohol metabolite
relative to cresol. Complete substitution of the methyl group led to a 48 % increase
in overall oxidation. This suggested that benzyl alcohol release is much slower than
#
cresol release from the enzyme-product complex. Thus, the reduced rate of H -
)
versus H -toluene binding to the enzyme for benzyl alcohol formation shifts more
"
)
substrate to the cresol pathway, resulting in a net increased rate of toluene
metabolism. Our observations of increased clearance and increased fraction of cresol
#
for H -
metabolites are consistent with this explanation - the average value of V
max- x
toluene was 82 % higher, the AUC 6.4 % lower, and the systemic clearance 6.5 %
)
higher than the non-deuterated counterpart. While our modelling approach assumed
that diåerences in extrahepatic metabolism (V
) accounted for the observed
max- x
isotopic diåerences in clearance, shifts in hepatic V
and K , which because of
max- h
m
¯ow-limited hepatic clearance have smaller eåects on total clearance, may have also
contributed.
A potential minor contributor to the observed diåerences in clearance is an
isotopic shift in the blood air partition coeæcient (K ), which could result in a
}
b/a
changed rate of exhalation. Recent analysis of pre-exposure blood samples from 29
experiments using a vial equilibration technique (Pierce et al. 1996b) in our

Toluene toxicokinetics in man
105
#
laboratory suggested that the K for H -toluene is about 3 % lower than that for
b/a
H -toluene (p 0.05). However, using average anthropometric and model-®t
)
!
"
)
values, this diåerence would result in only about a 0.5 % diåerence in Cl .
s
#
"
H -toluene demonstrated a 17‰ 22 % larger V compared with H -toluene, a
small diåerence compared with the 4-fold interindividual variability seen (table 2).
z
)
)
#
"
While the t
for H -toluene was 9.6‰ 26 % longer than for H -toluene, this
/ -z
" #
diåerence was much smaller that the 4-fold interindividual range. The t
)
)
of
/ -z
" #
#
"
38‰ 12 h for H - and 34‰ 10 h for H -toluene were within the wide range of values
found in previous studies : 10 h (Monster et al. 1993), 10.2±15.7 h (Hjelm et al.
)
)
1994), 12±65 h (Carlsson and Ljungquist 1982), 15.8 h (Droz and Guillemin 1986),
C
19.2±22.1 h (Brugnone et al. 1983), 21.4 h (Matsumoto et al. 1992),
72 h (Nise
and érbñk 1988), 77 h (Cohr and Stokholm 1979), and 90 and 118‰ 86 h (Nise et al.
1989). As with the clearances, this range suggests that the observed kinetic
diåerences in t
due to isotopic labelling are small compared with other sources of
/ -z
" #
variation.
Due to its high lipophilicity, & 96 % of toluene is expected to be in adipose tissue
C
in the terminal phase of disposition ( 30 h) (Pierce et al. 1996a). The rate-limiting
step in terminal phase elimination is release from adipose tissue, the rate of which is
mediated by the volume of tissue, tissue blood ¯ow and the tissue blood partition
}
coeæcient. The ®t fractional blood ¯ows to adipose tissue were not diåerent between
"
#
H - and H -toluene data sets. Nor was there a diåerence in tissue volume, since the
)
isomers were co-administered to each subject. The observed diåerences in terminal
)
#
volume and half-life may thus be due to an increased H -toluene adipose
)
tissue blood coeæcient. However, we have not found a signi®cant diåerence in the
}
#
human adipose tissue air partition coeæcient for H - versus
"
H
toluene
}
)
)
(1017‰ 162 versus 962‰ 298, mean‰ SD), nor in the blood air partition coeæcient
}
(18.1‰ 1.40 versus 18.3‰ 1.39 ; Pierce et al. 1996b). In addition, previous reports
have found decreased lipophilicity with deuteration, assessed using octanol water
}
partition coeæcients or reverse-phase HPLC, for a range of aliphatic and aromatic
hydrocarbons (Falconnet et al. 1988). Thus, the reasons for the larger terminal
#
volume of distribution and longer half-life for H - compared with H -toluene are
"
)
)
not currently understood.
C
than H -toluene ; an example of this diåerence is presented in ®gure 3. H -toluene
#
Overall, our results indicated that H -toluene was cleared
6 % more rapidly
)
"
#
)
)
also exhibited a 17 % larger terminal volume of distribution and a 10 % longer
terminal half-life. These small distinctions are unlikely to be signi®cant in the
practical evaluation of toluene toxicokinetics, given the large intersubject and
interstudy diåerences previously reported. The stable isotope label was therefore a
valuable tool to elucidate toluene kinetics.
This investigation extended the use of physiologically based modelling to test the
use of stable isotope-labelled tracers to represent the unlabelled compound in vivo.
Such an approach allowed us to ascribe observed isotope diåerences in solvent
toxicokinetics to underlying physiologic mechanisms.
Acknow ledgem ents
Valuable discussion was provided by Ken Kunze. A portion of this work was
conducted through the Clinical Research Facility of the University of Washington,

106
C. H. Pierce et al.
supported by NIH grant RR-37. Principal ®nancial support was provided by the
Superfund Basic Research program, NIEHS ES 04696.
References
merican onference of overnmental ndustrial
ygienists
A
C
G
I
H
(ACGIH), 1991, Documentation of
the Threshold Limit Values and Biological Exposure Indices, 6th edn (Cincinnati : ACGIH).
ndersen lewell ahle earhart
A
, M., C
, H. III, M
, D. and G
, J., 1994, Gas uptake studies of deuterium
isotope eåects on dichloromethane metabolism in female B6C3F1 mice in vivo. Toxicology and
Applied Pharmacology, 128, 158±165.
shley
onin
ardinali
ccraw ooten
, J. and W , J., 1994, Blood concentrations of
volatile organic compounds in a nonoccupationally exposed US population and in groups with
A
, D., B
, M., C
, F., M
suspected exposure. Clinical Chemistry, 40, 1401±1404.
strand
odahl
, K., 1986, Textbook of Work Physiology: Physiological Bases of Exercise, 3rd
A
B
B
, P.-O. and R
edn (New York : McGraw-Hill).
ox agles ilson
oza
airweather
ait
, S., 1995, The validity of extrinsic
, J., F , T., E
stable isotopic labeling for mineral absorption studies in rats. Journal of Nutrition, 125, 1611±1616.
rowne angenhove ostello iemann reenblatt
, J., W
, P. and F
-T
, T., L
, A. V., C
, C., B
, K. and G
, D., 1982, Pharmaco-
kinetic equivalence of stable-isotope-labeled and unlabeled drugs. Journal of Clinical Phar-
macology, 22, 309±315.
rowne
zabo
avoudi
osephs
, H. and J , E., 1994, Determination of pharmacokinetic drug
interactions with carbamazepine and phenytoin using stable isotope labeling and simple high-
B
B
, T., S
, G., D
performance liquid chromatography ultraviolet detection technique. Neurology, 44, 2410±2411.
erpelloni
}
rugnone
obbi
yyad
iuliari
erbellini
, M. and P , L., 1995, Blood
, F., G
, M., A
, K., G
, C., C
toluene as a biological index of environmental toluene exposure in the ` normal’ population and in
occupationally exposed workers immediately after exposure and 16 hours later. International
Archives of Occupational and Environmental Health, 66, 421±425.
rugnone
erbellini
postoli
ocatelli
ariotto
, M. and M , P., 1983, Decline of blood
B
B
, F., P
, L., A
, P., L
and alveolar toluene concentration following two accidental human poisonings. International
Archives of Occupational and Environmental Health, 53, 157±165.
rugnone
erbellini
accini
asini
aranelli
omeo
obbi
, M. and
, A., 1989, Breath and blood levels of benzene, toluene, cumene and styrene in non-
, F., P
, L., F
, G. B., P
, F., M
, G., R
, L., G
edde
Z
occupational exposure. International Archives of Occupational and Environmental Health, 61,
303±11.
urm
e
, A., D -B
oer
an leef
, A., V -K
ermeulen
, J. and V , N., 1988, Pharmacokinetics of lidocaine and
B
bupivacaine and stable isotope labelled analogues: a study in healthy volunteers. Biopharmaceutics
and Drug Disposition, 9, 85±95.
arlsson
jungquist
, E., 1982, Exposure to toluene : concentration in subcutaneous adipose
C
C
C
, A. and L
tissue. Scandinavian Journal of Work Environment and Health, 8, 56±62.
tokholm
ohr
, K. and S
Environment and Health, 5, 71±90.
astenmayer ansourian
, J., 1979, Toluene. A toxicologic review. Scandinavian Journal of Work
ouzy
uinchard
unoz ox irren
-B , R. and D , H., 1993,
, F., K
, P., M
, R., G
, S., M
Zinc absorption in health elderly humans and the eåect of diet. American Journal of Clinical
Nutrition, 58, 690±694.
ousa
oxenbaum
, H., 1988, Physiological pharmacokinetic models: some aspects of theory,
D’S
, R. and B
practice and potential. Toxicology and Industrial Health, 4, 151±171.
apleson
, W., 1993, A physiological model for the distribution of injected agents, with
avis
D
D
, N. and M
special reference to pethidine. British Journal of Anaesthesia, 70, 248±258.
ellamy alman
, D., 1997, Quantitation of o-, m- and p-cresol and deuterated
ills
, R., B
, G. and K
analogs in human urine by gas chromatography with electron capture detection. Journal of
Chromatography B, 703, 105±113.
ills
roz
ent
heckoway
alman
, D., 1991, Quanti®cation of volatile solvents in blood
D
D
D
, R., K
by static headspace analysis. Talanta, 38, 365±374.
uillemin
, S., C
, H. and K
, P. and G
Occupational Medicine, 28, 593±602.
urnin omersley
, M., 1986, Occupational exposure monitoring using breath analysis. Journal of
, J. and W
, J., 1974, Body fat assessed from total body density and its estimation from
skin fold thicknessÐmeasurements on 481 men and women aged from 16 to 72 years. British
Journal of Nutrition, 32, 77±97.
tzel
shley
, D., 1994, Volatile organic compounds in the blood of persons in Kuwait during
E
F
, R. and A
the oil ®res. International Archives of Occupational and Environmental Health, 66, 125±129.
alconnet ayar echalany herrah enchekroun arrupt esta
, B.
, J., T
razier
Labelled Compounds. Third International Symposium. Innsbruck, Austria (Elsevier), pp.
, N. E., B
, A., C
, Y., B
, Y., C
, P., T
and B
, J., 1988, Isotope eåects on lipophilicity. In Synthesis and Applications of Isotopically
355±360.

Toluene toxicokinetics in man
107
errannini
F
F
, E., 1992, The haemodynamics of obesity : a theoretical analysis. Journal of Hypertension, 10,
1417±1423.
ustinoni uratti
iampiccolo
olombi
, A., 1995, Biological and environmental
, S., B
, M., G
, R. and C
monitoring of exposure to airborne benzene and other aromatic hydrocarbons in Milan traæc
wardens. Toxicology Letters, 77, 387±392.
ammans
ackenthun
formulations using stable isotope methodology. British Journal of Clinical Pharmacology, 18,
ussell
, A. and R , J., 1984, Comparative bioavailability of trazadone
G
G
H
H
H
, R., M
431±437.
argas
urgess
oisard
ason
ndersen
, G. and A , M., 1989, Partition coeæcients of low-
molecular-weight volatile chemicals in various liquids and tissues. Toxicology and Applied
, M., B
, R., V
, D., C
Pharmacology, 98, 87±99.
erthold
"$
, P., 1995, A novel stable isotope breath test : C-labeled glycosyl
eine
jelm
lein
, W., B
, H. and K
ureides used an noninvasive markers of intestinal transit time. American Journal of Gastro-
enterology, 90, 93±98.
of
, E., L , A., S
ato
olmsjo
undmark
ortro m
, B. and N , A., 1994, Dietary and ethanol
, A., C
, A., L
induced alterations of the toxicokinetics of toluene in humans. Occupational and Environmental
Medicine, 51, 487±491.
jelm
,
aslund
alle n
, P. and W , M., 1988, In¯uence of cigarette smoking on the toxicokinetics of
, E., N
toluene in humans. Journal of Toxicology and Environmental Health, 25, 155±163.
R., 1990, Interspecies scaling and comparisons in drug development and toxicokinetics.
Xenobiotica, 20, 1201±1231.
ngs
I
ones
unlap
ospe r
, S. M. J , 1994, Stable-isotope dilution GC-MS for determination of
J
, A., D
toluene in submilliliter volumes of whole blood. Journal of Analytical Toxicology, 18, 251±254.
asuya
, M. and G
K
K
, Y., 1994, Stable isotope methodology in the pharmacokinetic study of steroids. Nippon Rinsho,
52, 635±641.
izunuma
awai
awai
asugi
origuchi
keda
, S. and I , M., 1994, Toluene in blood as a
, T., M
, K., Y
, T., H
marker of choice for low-level exposure to toluene. International Archives of Occupational and
Environmental Health, 66, 309±315.
asugi
izunuma
origuchi
guchi
chida
wami
keda
, O. and I ,
K
, T., Y
, T., M
, K., H
, S., I
, H., U
, Y., I
M., 1992a, Comparative evaluation of urinalysis and blood analysis as means of detecting exposure
to organic solvents at low concentrations. International Archives of Occupational and Environmental
Health, 64, 223±234.
awai
awai
asugi
izunuma
origuchi
keda
, M., 1992b, Urinalysis versus blood
K
K
, T., Y
analysis, as a tool for biological monitoring of solvent exposure. Toxicology Letters, 63, 333±343.
asugi izunuma origuchi orioka iyashita chida
, T., M
, K., H
, S. and I
, T., Y
, T., M
, K., H
, S., M
, I., M
, K., U
, Y. and
keda
I
, M., 1992c, Monitoring of workers exposed to a mixture of toluene, styrene and methanol
vapours by means of diåusive air sampling, blood analysis and urinalysis. International Archives
of Occupational and Environmental Health, 63, 429±435.
awamoto
oga
urata
atsuda
odama
, S. and K , Y., 1995, Eåects of ALDH2,
CYP1A1 and CYP2E1 genetic polymorphisms and smoking and drinking habits on toluene
K
K
, T., K
, M., M
, K., M
metabolism in humans. Toxicology and Applied Pharmacology, 133, 295±304.
eal
ochak
erner
ambol
, M. and S , N., 1992, Variational analysis of the transdermal
, G., B
, B., L
delivery rate from two prototypical ethanol-water nitroglycerin TTS devices and Transderm-
Nitro 10 in the normal population. Journal of Pharmacokinetics and Biopharmaceutics, 20,
443±459.
angenhove
L
L
L
, A. V., 1986, Isotope eåects: de®nitions and consequences for pharmacologic studies.
Journal of Clinical Pharmacology, 26, 383±389.
eggett
illiams
, L., 1991, Suggested reference values for regional blood volumes in humans.
, R. and W
Health Physics, 60, 139±154.
anzlik
ing
, K. and H
rate-limiting step in cytochrome P-450 catalysis. Biochemical and Biophysical Research
Communications, 160, 844±849.
jelm olmsjo
, R., 1989, Deuterium isotope eåects on toluene metabolism. Product release as
a
of
undmark
ordstro m
ato
S , A., 1993,
L
L
, A., H
, E. W., C
, A., L
, B.-O., N
, A. and
#
Toxicokinetics of toluene and urinary excretion of hippuric acid after human exposure to H -
toluene. British Journal of Industrial Medicine, 50, 55±59.
)
of
allen
jelm
, E., 1990, In¯uence of paracetamol and acetylsalicylic acid on the
, A., W
toxicokinetics of toluene. Pharmacology and Toxicology, 66, 138±141.
annino chreiber ldous shley oolenaar
, M. and H
lmaguer
, D., 1995,
M
, D., S
, J., A
, K., A
, D., M
, R. and A
Human exposure to volatile organic compounds : a comparison of organic vapor monitoring badge
levels with blood levels. International Archives of Occupational and Environmental Health, 67,
59±64.
atsumoto
oga
ata
adoya
higematsu
, A., 1992, The changes of gasoline
M
, T., K
compounds in blood in a case of gasoline intoxication. Clinical Toxicology, 30, 653±662.
ahon ullivan
, H., 1966, Microsomal hydroxylation of ethlybenzene. Stereospeci®city
, M., S
, T., K
, T. and S
c
M M
, R. and S
and the eåect of phenobarbital induction. Life Sciences, 5, 921±926.

108
Toluene toxicokinetics in man
onster
ezi c
, A., K ’ , S., G
evel
e
olff
W , F. A., 1993, Evaluation of biological
M
, I. V. D. and D
monitoring parameters for occupational exposure to toluene. International Archives of Oc-
cupational and Environmental Health, 65, S159±162.
ordenti
, J., 1986, Man versus beast: pharmacokinetic scaling in mammals. Journal of Pharmaceutical
M
M
Sciences, 75, 1028±1040.
ills
organ
alman
, R. and K , D., 1993, Evaluation of stable isotope-labeled probes in the study
, M., D
of solvent pharmacokinetics in human subjects. International Archives of Occupational and
Environmental Health, 65 (1 Suppl.), S139±142.
akajima
ang
, R., 1994, Induction of cytochrome P450 by toluene. International Journal of
N
N
N
, T. and W
Biochemistry, 26, 1333±1340.
ttewell kerfving
ise
ise
rbák
, P., 1989, Elimination of toluene from venous blood
, G., A
and adipose tissue after occupational exposure. British Journal of Industrial Medicine, 46, 407±411.
rbák
, R., S
, S. and é
, G. and é
International Archives of Occupational and Environmental Health, 60, 31±35.
ills ewandowski organ essels hen
, P., 1988, Toluene in venous blood during and after work in rotogravure printing.
ierce
ierce
ierce
alman
, D., 1997,
P
P
P
R
R
, C., D
, R., L
, T., M
, M., W
, M., S
, D. and K
Estimation of background exposure to toluene using a physiologically-based kinetic model.
Journal of Occupational Health, 39, 130±137.
ills
diåerences in H -toluene toxicokinetics assessed by a semi-empirical physiologically-based
organ
othstein
hen
alman
, D. and K , D., 1996a, Interindividual
, C., D
, R., M
#
, M., N
, G., S
)
model. Toxicology and Applied Pharmacology, 139, 49±61.
ills
ilvey
alman
, G. and K , D., 1996b, Partition coeæcients between human blood or
, C., D
., S
adipose tissue and air for aromatic solvents. Scandinavian Journal of Work Environment and
Health, 22, 112±118.
itschel
achharajani
interspecies scaling of pharmacokinetic parameters. Comparative Biochemical Physiology, 103C,
ohnson
ussain
, R. and H , A., 1992, The allometric approach for
, W., V
, N., J
249±253.
owland
ozer
, T. N., 1989, Clinical Pharmacokinetics : Concepts and Applications, 2nd edn
, M and T
(Philadelphia: Lea & Febiger).
sterloh iemayer
mith
legal
, A., 1992, Stable isotope labeling of lead
S
S
S
T
, D., O
, J., N
, S. and
F
compartments in rats with ultralow lead concentrations. Environmental Research, 57, 190±207.
nyder
, R. (ed.), 1987, Ethel Browning’s Toxicity and Metabolism of Industrial Solvents, vol. 1.
Hydrocarbons, 2nd edn (Amsterdam : Elsevier).
teiner
ey c roskey
, E., R , T. and M C , P., 1990, Reference GuideÐSimuSolv Modeling and Simulation
Software, vol. 2. (Midland : Dow Chemical Co.).
ardif apare rishnan rodeur
, J., 1993, Physiologically based modeling of the
, R., L
, S., K
, K. and B
toxicokinetic interaction between toluene and m-xylene in the rat. Toxicology and Applied
Pharmacology, 120, 266±273.
hummel
harasch
en¯urane de¯uorination catalyzed by cytochrome P-450 2E1. Drug Metabolism and Disposition,
odoll
unze
, T. and K , K., 1993, Human liver microsomal
T
, K. E., K
, E. D., P
21, 350±357.
omaszewski
erina
aly
, D. and D , W., 1975, Deuterium isotope eåects during formation of phenols
by hepatic monoxygenases. Evidence for an alternative to the arene oxide pathway. Biochemistry,
T
, J., J
4, 2024±2030.
epartment of ealth
ducation and elfare
W , 1985, NIOSH Manual of Analytical Methods.
US D
H
, E
NIOSH 84±100 (Cincinnati : NIOSH), pp. 1501-1±7.
an eerendonk earson eijer ulder
elson
eerman
, J., 1995,
V
B
, G., P
, P., M
, D., M
, G., N
, S. and M
Deuterium isotope eåect on the metabolism of the ¯ame retardant tris(2,3-dibromopropyl)
phosphate in the isolated perfused rat liver. Fundamental and Applied Toxicology, 28, 111±117.
eulemans
asschelein
, R., 1978, Experimental human exposure to toluene. International
V
V
, H. and M
Archives of Occupational and Environmental Health, 42, 105±117.
ozicevic aracic uchs rpic ajic
rca
, A., B
alinar
, M., 1995, Visual
, D., K
, V., F
, R., P
-M
, D. and M
evoked potentials in individuals exposed to long-term low concentrations of toluene. Archives of
Toxicology, 69, 337±340.
allen
W
W
W
W
, M., 1986, Toxicokinetics of toluene in occupationally exposed workers. Scandinavian Journal
of Work Environment and Health, 12, 588±593.
allen olm ordqvist
, M., H
and elimination. British Journal of Industrial Medicine, 42, 111±116.
allen aslund ordqvist
, M., 1984, The eåects of ethanol on the kinetics of toluene in
, S. and N
, M., 1985, Coexposure to toluene and p-xylene in man : uptake
, M., N
man. Toxicology and Applied Pharmacology, 76, 414±419.
heeler righton uenthner
, T., 1992, Detection of human lung cytochromes P450 that
, P. and N
, C., W
, S. and G
are immunochemically related to cytochrome P450IIE1 and cytochrome P450IIIA. Biochemical
Pharmacology, 44, 183±186.
hite
c arthy
, R. and M C , M., 1986, Active site mechanics of liver microsomal cytochrome P-450.
Archives of Biochemistry and Biophysics, 246, 19±32.
W