Hepatic enzymatic synthesis and hydrolysis of CoA esters of solvent-derived oxa acids

Article abstract of DOI:10.1002/jbt.10063

Many ethylene glycol-derived solvents are oxidized to xenobiotic alkoxyacetic acids (3-oxa acids) by hepatic enzymes. The toxicity of these ubiquitous solvents has been associated with their oxa acid metabolites. For many xenobiotic carboxylic acids, the toxicity is associated with the CoA ester of the acid. In this study, related alkoxyacetic acids were evaluated as potential substrates for acyl-CoA synthetases found in mitochondrial, peroxisomal, and microsomal fractions isolated from rat liver. Likewise, chemically synthesized oxa acyl-CoAs were used as substrates for acyl-CoA hydrolases associated with the same rat liver fractions. Activities of the xenobiotic oxygen-substituted substrates were compared with analogous physiologic aliphatic substrates by UV-vis spectrophotometric methods. All of the solvent-derived oxa acids were reasonable substrates for the acyl-CoA synthetases, although their activity was usually less than the corresponding physiologic acid. Acyl-CoA hydrolase activities were decreased compared with acyl-CoA synthetase activities for all substrates, especially for the oxa acyl-CoAs. These studies suggest that these xenobiotic carboxylic acids may be converted to reactive acyl-CoA moieties which will persist in areas of the cell proximal to lipid synthesis, β-oxidation, protein acylation, and amino acid conjugation. The interaction of these xenobiotic acyl-CoAs with those processes may be important to their toxicity and/or detoxification.

Full text of DOI:10.1002/jbt.10063

J BIOCHEM MOLECULAR TOXICOLOGY
Volume 17, Number 2, 2003
Hepatic Enzymatic Synthesis and Hydrolysis of CoA
Esters of Solvent-Derived Oxa Acids
Sree D. Panuganti, Jill M Penn, and Kathleen H. Moore
Department of Chemistry, Oakland University, Rochester, MI 48309-4477, USA; E-mail: kmoore@oakland.edu
Received 13 September 2002; revised 2 February 2003; accepted 2 February 2003
ABSTRACT: Many ethylene glycol-derived solvents
are oxidized to xenobiotic alkoxyacetic acids (3-oxa
acids) by hepatic enzymes. The toxicity of these ubiq-
uitous solvents has been associated with their oxa
acid metabolites. For many xenobiotic carboxylic acids,
the toxicity is associated with the CoA ester of the
acid. In this study, related alkoxyacetic acids were
evaluated as potential substrates for acyl-CoA syn-
thetases found in mitochondrial, peroxisomal, and mi-
crosomal fractions isolated from rat liver. Likewise,
chemically synthesized oxa acyl-CoAs were used as
substrates for acyl-CoA hydrolases associated with
the same rat liver fractions. Activities of the xeno-
biotic oxygen-substituted substrates were compared
with analogous physiologic aliphatic substrates by
UV–visspectrophotometricmethods. Allofthesolvent-
derived oxa acids were reasonable substrates for the
acyl-CoA synthetases, although their activity was usu-
ally less than the corresponding physiologic acid. Acyl-
CoA hydrolase activities were decreased compared
with acyl-CoA synthetase activities for all substrates,
especially for the oxa acyl-CoAs. These studies suggest
that these xenobiotic carboxylic acids may be converted
to reactive acyl-CoA moieties which will persist in ar-
eas of the cell proximal to lipid synthesis, ꢀ-oxidation,
protein acylation, and amino acid conjugation.
The interaction of these xenobiotic acyl-CoAs with
those processes may be important to their toxicity
INTRODUCTION
This study focuses on carboxylic acids, which are
known to arise from rapid metabolism of ethylene
glycol-based solvents. These solvents can be catego-
rized as ether, ether alcohol, or ether ester compounds
[1]. The ether alcohol solvent 2-ethoxyethanol can be
oxidized to 2-ethoxyacetic acid (3-oxapentanoic acid)
by sequential action of hepatic alcohol dehydrogenase
and aldehyde dehydrogenase [2]. Ether solvents such
as 1,2-diethoxyethane are metabolized by microsomal
O-deethylase action and subsequent alcohol oxidation
[3]. Ether esters like 2-ethoxyethyl acetate are oxidized
after ester hydrolysis [2]. Elevated levels of alkoxyacetic
acids have been detected in the urine of workers ex-
posed to relatively low levels of ethylene glycol ethers
[4,5]. Nose-only inhalation by rats of low-levels of the
ethylene glycol ether diglyme resulted in numerous
clinical and histopathological aberrations [6].
Recent studies have shown that the toxicity at-
tributed to many ethylene glycol-based solvents is
mediated by the carboxylic acid metabolite. For in-
stance, the metabolite 2-methoxyacetic acid is re-
sponsible for the multiple abnormalities (testicular,
lymphoid, hematopoietic) observed subsequent to 2-
methoxyethanol exposure [7]. Gray et al. [8] found
alkoxy-specific toxicity using a primary testicular cell
culture model with a group of solvent-derived oxa
acids. Ghanayem et al. [9] noted that the presence and
position of the ether linkage was related to the hemato-
toxicityobservedwithseveralalkoxyaceticacids. While
toxicity mechanisms are not fully elucidated, the car-
boxylic acid metabolites are clearly implicated.
ꢀ
C
and/or detoxification.
2003 Wiley Periodicals, Inc.
J Biochem Mol Toxicol 17:76–85, 2003; Published on-
line in Wiley InterScience (www.interscience.wiley.com).
DOI 10.1002/jbt.10063
KEYWORDS: Alkoxyacetic Acids; Oxa Acids; Ethy-
lene Glycol-Derived Solvents; Acyl-CoA Syn-
thetase; Acyl-CoA Hydrolases; Acyl-CoA Synthesis;
Hepatotoxicity
Many of the hepatotoxic effects of xenobiotic or
xenobiotic-derived carboxylic acids may be mediated
by the CoA ester of the xenobiotic moiety [10,11]. In
1985, Sherratt reviewed early work on the hepatic ac-
cumulation of xenobiotic acyl-CoAs and their result-
ing biochemical effects, primarily inhibition of fatty
acid ꢀ-oxidation, and other intermediary metabolic
Correspondence to: Kathleen H. Moore.
Contract Grant Sponsor: Oakland University.
2003 Wiley Periodicals, Inc.
c
ꢀ
76

Volume 17, Number 2, 2003
CoA ESTERS OF SOLVENT-DERIVED OXA ACIDS
77
pathways[12]. Morerecently, Brass[13]notedthataccu-
mulated xenobiotic acyl-CoAs are poorly metabolized
quite frequently and may facilitate toxicity by nature of
their unusual structure or by depletion or sequestration
of available CoA. Knights and Roberts [14] have sug-
gested that there are multiple hepatic acyl-CoA syn-
thetases (ligases) that may be more or less reactive
with xenobiotic substrates than with physiologic sub-
strates. While the medium-chain acyl-CoA synthetase
of liver mitochondria has been shown to activate many
xenobiotics, including 2-arylacetic acids, that subse-
quently undergo amino acid conjugation prior to ex-
cretion [15], other xenobiotics like the 2-arylpropionic
acids seem to be activated by specific microsomal acyl-
CoA synthetases [14]. Yao et al. [16] demonstrated that
rat liver mitochondria were more versatile in the ac-
tivation of xenobiotic carboxylic acids to the CoA es-
ter than were rat heart mitochondria; in addition, only
those xenobiotics that were transformed to the CoA es-
ter were inhibitory to mitochondrial respiration. A re-
cent review [17] suggests that, especially in humans,
the hepatic formation of xenobiotic acyl-CoA moieties
will result in further involvement of the xenobiotic in
cellular processes and should not be considered mere
preparation for amino acid conjugation and subsequent
excretion.
The role of cellular acyl-CoA hydrolases or
thioesterases in the disposition of xenobiotic acids
remains uncertain. Acyl-CoA hydrolases have been
found in cytosol, mitochondria, peroxisomes, and
endoplasmic reticulum of mammalian cells [18], and
are believed to be involved in the maintenance of
physiological acyl-CoA and free CoASH pools for
both fatty acid oxidation and fatty acid synthesis [19].
Because these enzymes have some ability to hydrolyze
xenobiotics, they may have a role in the detoxification
of nonphysiological metabolites [20,21]. In a recent
detailed study on non- and partially metabolisable
xenobiotic acyl-CoAs, Garras et al. [22] suggest that
many cellular acyl-CoA hydrolases may function to
regulate acyl-CoA levels.
MATERIALS AND METHODS
Materials
Ethoxyacetic acid (3-oxapentanoic acid), 3-
methoxypropionic acid (4-oxapentanoic acid), 2-(2-
methoxyethoxy)acetic acid (3,6-dioxaheptanoic acid),
and 2-[2-(2-methoxyethoxy) ethoxy]acetic acid (3,6,9-
trioxadecanoic acid) were purchased from Aldrich
(Milwaukee, WI). Pentanoic acid, heptanoic acid,
decanoic acid, their CoA derivatives, CoA, AMP,
ATP, Tris-HCl, 5,5ꢁ-dithiobis (2-nitrobenzoic acid)
[DTNB], tetrahydrofuran, oxalyl chloride, and other
chemical and biochemical reagents were from Sigma
(St Louis, MO). Rat liver was from freshly sacrificed,
10-week-old, male, Sprague-Dawley rats. Thin layer
plates (silica gel GF uniplates) were from Analtech
(Newark, DL).
Capillary Electrophoresis Analysis
of Acyl-CoAs
Analyses of synthesized acyl-CoAs were per-
formed on an automated HP3DCE capillary elec-
trophoresis system (Hewlett-Packard) equipped with
a UV–vis multiwavelength detector and an uncoated
fused silica capillary of length 48.5 cm (detection win-
dow at 40 cm from sample injection end) and internal
diameter of 50 ꢁm. Sodium tetraborate buffer (20 mM,
pH 9.3) was used as running buffer. The capillary was
preconditioned with a first flush using 1.0 N NaOH for
2 min followed by a second flush using 20 mM sodium
tetraborate buffer (pH 9.3) for 3 min. Samples were in-
jected by the hydrodynamic method for 10 s by ap-
plying 50 mbar pressure; a potential of 30 kV positive
polarity (current = 30–50 ꢁA) was applied across the
capillary. All analyses were performed at 20◦C. Each
run was set for 15 min. Synthesized acyl-CoAs were de-
tected at 260 nm by a UV–vis detector. Migration times
of peaks and corresponding peak areas in the electro-
pherogram were used to identify and quantitate the
synthesized acyl-CoAs.
Here in we report the effectiveness of a group of
medium-chain oxa acids as substrates for hepatic acyl-
CoA synthetases associated with mitochondrial, perox-
isomal (light mitochondrial), and microsomal fractions
of rat liver. The oxa acids studied are representative of
the many xenobiotic carboxylic acids which are known
or predicted to arise from the hepatic metabolism of
ethylene glycol-derived solvents [1,2]. To obtain com-
plementary information on the reactivity of the xenobi-
otic oxa acyl-CoAs with hepatic acyl-CoA hydrolases, it
was first necessary to synthesize, purify, and quantitate
the oxa acyl-CoA substrates. This study begins to de-
fine the impact of this group of xenobiotics on normal
fatty acid metabolism.
Subcellular Fractionation
Subcellular fractionation was accomplished by dif-
ferential centrifugation [23]; specifically, rat liver was
flushed with ice-cold isotonic saline and diluted 10-
fold (wt/vol) with sucrose-phosphate buffer (0.25 M
sucrose, 10 mM potassium phosphate, 1 mM EDTA,
pH 7.4). The tissue was minced, processed for 3 s with
a Polytron PT 10 homogenizer (Brinkman) set at 30%
maximal power, and then subjected to four passes at
50% maximal power with a motor-driven teflon pestle

78
PANUGANTI, PENN, AND MOORE
Volume 17, Number 2, 2003
homogenizer. Resulting homogenate was centrifuged
in a Sorvall RC-5B instrument (DuPont Instruments,
Boston, MA) for 10 min at 750 × g to remove large cel-
lular fragments and nuclei; the resulting supernatant
was centrifuged for 10 min at 8700 × g to obtain the
mitochondrial fraction. The mitochondrial fraction was
further purified by two rounds of resuspension and re-
centrifugation. The original postmitochondrial super-
natant was centrifuged at 26,000 × g for 15 min to
pellet the peroxisomal (light mitochondrial) fraction;
the peroxisomal supernatant fraction was transferred
to ultracentrifuge tubes and centrifuged for 60 min at
100,000 × g in a Beckman L-5 Ultracentrifuge (Beckman
Instruments, Anaheim, CA). Themicrosomalpelletwas
resuspended by brief treatment with a Tissue Tearor
homogenizer.
strate and enzyme controls were run. Substrate control
contained 0.1 mM acyl-CoA and 90 mM Tris-HCl (pH
7.4) but no subcellular fraction. Enzyme control con-
tained 250 ꢁg rat liver subcellular fraction and 90 mM
Tris-HCl (pH 7.4) but no acyl-CoA substrate. Reac-
tions were terminated at intervals of 15 and 45 min
by placing the tubes in a boiling water bath for 3 min.
Denatured protein was removed by centrifugation at
16,000 × g for 3 min. Supernatants were mixed with
5,5ꢁ-dithiobis (2-nitrobenzoic acid) (0.125 mM) and the
samples were read at 412 nm on a UV–vis spectropho-
tometer (Hitachi). The actual increase in the absorbance
at 412 nm was determined by subtracting the increase in
the absorbances of enzyme and substrate controls from
that of test sample. The molar absorption coefficient of
nitrobenzoate (ε = 13,600 M−1 cm−1) was used to deter-
mine the concentration of CoA released by enzymatic
hydrolysis. Enzyme activity was calculated by dividing
the concentration of released CoA by incubation time
and amount of enzyme protein.
Protein content of resuspended fractions was de-
termined by a modified biuret procedure [24], using
bovine serum albumin as the standard. Aliquoted frac-
tions were quick-frozen and stored at −80◦C.
Acyl-CoA Synthetase Assay
RESULTS
For standard assay, duplicate 1 mL reaction mix-
tures containing 36 mM Tris-HCl (pH 8.5), 72 mM KCl,
3.6 mM MgCl2, 0.1 mM CoA, 0.4 mM ATP, 2.5 mM
carboxylic acid substrate, and 180 ꢁg rat liver protein
were incubated at 37◦C for 30 min. Reaction conditions
were chosen to inhibit competing acyl-CoA hydrolase
activity [25] and to achieve a maximal acyl-CoA syn-
thetase activity [23]. Reaction was terminated by plac-
ing tubes in a boiling water bath for 3 min to dena-
ture protein. Protein was removed by centrifugation at
16,000 × g for 3 min. Supernatants were placed on ice or
frozen pending further analysis. For UV spectrophoto-
metric determination, 0.1 mL of supernatant was mixed
with 0.9 mL deionized water (1:10 dilution). Based
on the increased absorption at 235 nm of acyl-CoA
thioesters versus free CoASH [26], absorbance mea-
surements of 1:10 dilutions of assay supernatants were
determined at 235 nm using a UV–vis spectrophotome-
ter (Shimadzu). Control absorbances (no carboxylic
acid substrate) were subtracted from sample values.
An experimentally determined molar extinction coef-
ficient (12,850 M−1cm−1) was used to convert the mea-
sured absorption to concentration and then to enzyme
activity.
Chemical Synthesis of Oxa Acyl-CoAs
Oxa acyl-CoAs are not commercially available;
therefore, they were synthesized in the laboratory for
use as substrates for rat liver acyl-CoA hydrolases.
Chemical synthesis of oxa acyl-CoAs involves the ac-
tivation of oxa acids to oxa acyl-chlorides followed by
their reaction with free CoA [26]. A reaction between
carboxylic acid (0.75 mmol) and fresh oxalyl chloride
(1 mL) was carried out at room temperature for 15 min,
followed by evaporation of excess oxalyl chloride by
a stream of nitrogen. The reaction was repeated two
more times and the final product, acyl-chloride, was
dissolved in 2 mL peroxide-free tetrahydrofuran. The
acyl-chloride was added drop wise to 0.26 mmol of CoA
dissolved in a water/tetrahydrofuran (1:2) solution, pH
8.9. The reaction was monitored by a thiol spot test, us-
ing sodium nitroprusside reagent that forms pink com-
plex upon reaction with free thiol group. When the thiol
spot test was negative, addition of acyl-chloride was
stopped and the pH was adjusted to 2–3, using Dowex
50WX2-200 cation exchange resin. After filtering out
Dowex, excess tetrahydrofuran was removed by rotary
evaporation. Unreacted carboxylic acid was removed
by repeated ether extractions. The aqueous layer con-
taining acyl-CoA was lyophilized and the powdered
product was dissolved in minimal deionized water for
characterization and quantitation.
Acyl-CoA Hydrolase Assay
For standard assay, duplicate 1 mL reaction mix-
tures containing 90 mM Tris-HCl (pH 7.4), 0.1 mM
substrate acyl-CoA, and 250 ꢁg rat liver subcellular
fraction protein were incubated at 37◦C. Parallel sub-
Synthesized acyl-CoAs were characterized and
quantitated by UV spectrum, thin layer chromatogra-
phy using freshly prepared butanol/acetic acid/water

Volume 17, Number 2, 2003
CoA ESTERS OF SOLVENT-DERIVED OXA ACIDS
79
TABLE 2. Reactivity of Carboxylic Acids with Mitochondrial
Acyl-CoA Synthetases
(5:2:3) solvent mix, and capillary electrophoresis.
Absorbances of synthesized sample at 232 and 260 nm
were compared with that of free CoA and commercial,
aliphatic acyl-CoAs to confirm the thioester bond for-
mation, and A260 (ε = 16,400 M−1 cm−1) was used for
quantitation.
Chain
Length
Carboxylic
Acid
Activitya
(nmol/min/mg)
Pb
Ten
Decanoic
3,6,9-Trioxadecanoic
Heptanoic
3,6-Dioxaheptanoic
Pentanoic
3-Oxapentanoic
4-Oxapentanoic
89.7 14.3
37.9 11.2
41.6 15.9
33.5 8.2
36.8 15.2
21.6 4.3
34.0 6.4
<0.01
Table 1 shows the Rf values and ratios of absorp-
tions at 232–260 nm of synthesized physiologic and oxa
acyl-CoAs. Samples on the thin layer plate were visu-
alized using short wave UV light (Ultra-violet Prod-
ucts Inc.). All synthesized samples appeared as single
spots showing their purity. Retention of oxa acyl-CoAs
on the silica gel was higher than that of corresponding
physiological acyl-CoAs supporting increased polarity
of oxa acyl-CoAs due to the presence of one or more
oxygen atoms in the hydrocarbon chain. The ratios of
absorbances at 232–260 nm for synthesized oxa acyl-
CoAs were similar to those of physiologic acyl-CoAs
confirming the formation of thioester bonds between
corresponding oxa acids and CoA. Synthesized acyl-
CoAs appeared as single, sharp peaks in capillary elec-
tropherograms (Figure 1). The migration times of syn-
thesized physiologic and oxa acyl-CoAs were similar to
those of standard (commercial) acyl-CoAs. Some of the
synthesized samples contained small amounts of free
CoA (retention time approximately 12.5 min).
Seven
Five
n.s.
n.s.
n.s.
a Means SD for n ≥ 5 duplicate experiments.
b Indicates level of significance when oxa acid activity is compared with that
of analogous saturated acid; n.s. indicates P > 0.05.
reactive as the 10-carbon substrates. Statistically, there
is no difference between the reactivity of either of the
oxapentanoic acids and pentanoic acid. Peroxisomal
acyl-CoA synthetases exhibited a pattern of reactivity
(Table 3) similar to that observed with mitochondrial
enzymes. Trioxadecanoic acid was significantly less re-
active than decanoic acid. It is also noteworthy that
dioxaheptanoic acid and both oxapentanoic acids are
activated by the peroxisomal enzymes at the same rate
as their saturated controls. Acyl-CoA synthetases asso-
ciated with the rat liver microsomes (Table 4) activated
the trioxadecanoic acid to a lesser extent (34.7%) than
did its control. There was no significant difference be-
tween microsomal activation of the dioxaheptanoic and
the two oxapentanoic acids as compared with their re-
spective controls.
Enzymatic Synthesis of Oxa Acyl-CoAs
Acyl-CoAsynthetasesintheratlivermitochondrial
fraction were able to activate all of the carboxylic acid
substrates (Table 2). Specific activity was higher with
decanoic acid than with any other substrate tested in
mitochondrial fractions. The trioxadecanoic acid was
significantly less reactive, being activated only 42% as
was the decanoic acid; the dioxaheptanoic acid was al-
most as reactive as its control. Pentanoic acid was not as
Enzymatic Hydrolysis of Oxa Acyl-CoAs
The reactivity of rat liver mitochondrial acyl-CoA
hydrolases with physiologic and oxa acyl-CoAs is sum-
marized in Table 5. Decanoyl-CoA was most effective
substrate whereas the reactivity of octanoyl-CoA and
TABLE 1. Analytical Characterization of Synthesized Acyl-
CoAs
TABLE 3. Reactivity of Carboxylic Acids with Peroxisomal
Acyl-CoA Synthetases
a
b
Sample
Rf
A232/A260
Chain
Length
Carboxylic
Acid
Activitya
(nmol/min/mg)
Pb
CoA
0.125
0.643
0.625
0.281
0.606
0.238
0.581
0.311
0.265
0.597
0.585
0.593
0.600
0.574
0.597
0.557
Standard decanoyl-CoAc
Decanoyl-CoA
Ten
Decanoic
3,6,9-Trioxadecanoic
Heptanoic
3,6-Dioxaheptanoic
Pentanoic
3-Oxapentanoic
4-Oxapentanoic
62.2 4.9
22.9 7.4
33.7 9.4
29.6 6.1
31.8 3.5
27.2 4.9
26.9 6.4
<0.01
3,6,9-Trioxadecanoyl-CoA
Octanoyl-CoA
3,6-Dioxaheptanoyl-CoA
Pentanoyl-CoA
Seven
Five
n.s.
n.s.
n.s.
3-Oxapentanoyl-CoA
a Relative mobility of product on thin layer plate in butanol/water/acetic
acid solvent mixture.
a Means SD for n ≥ 5 duplicate experiments.
b Ratio of absorbances at 232–260 nm, characteristic of thioester bond.
c Commercial acyl-CoA.
b Indicates level of significance when oxa acid activity is compared with that
of analogous saturated acid; n.s. indicates P > 0.05.

80
PANUGANTI, PENN, AND MOORE
Volume 17, Number 2, 2003
FIGURE 1. Representative capillary electropherograms of commercial and synthesized acyl-CoAs. Capillary electrophoresis conditions were
as given in Methods section. (A) 0.1 mM standard octanoyl-CoA, (B) 0.1 mM synthesized heptanoyl-CoA, and (C) 0.1 mM synthesized 3,6-
dioxaheptanoyl-CoA.

Volume 17, Number 2, 2003
CoA ESTERS OF SOLVENT-DERIVED OXA ACIDS
81
TABLE 4. Reactivity of Carboxylic Acids with Microsomal
TABLE 6. Reactivity of Acyl-CoAs Acids with Peroxisomal
Acyl-CoA Synthetases
Acyl-CoA Hydrolases
Chain
Length
Carboxylic
Acid
Activitya
(nmol/min/mg)
Chain
Length
Activitya
(nmol/min/mg)
Pb
Acyl-CoA
Decanoyl-CoA
3,6,9-Trioxadecanoyl-CoA
Octanoyl-CoA
3,6-Dioxaheptanoyl-CoA
Pentanoyl-CoA
3-Oxapentanoyl-CoA
4-Oxapentanoyl-CoA
Pb
Ten
Decanoic
3,6,9-Trioxadecanoic
Heptanoic
3,6-Dioxaheptanoic
Pentanoic
3-Oxapentanoic
4-Oxapentanoic
65.2 6.4
22.6 2.3
26.4 2.8
23.2 2.3
25.9 1.4
21.8 4.4
22.9 2.8
Ten
1.67 0.13
0.63 0.25
1.00 0.10
0.48 0.20
0.26 0.09
0.10 0.05
0.14 0.10
<0.01
<0.01
<0.01
Seven
Five
Eight
Seven
Five
n.s.
n.s.
n.s.
<0.01
n.s.
a Means SD for n ≥ 5 duplicate experiments.
a Means SD for n ≥ 10 duplicate experiments.
b Indicates level of significance when oxa acid activity is compared with that
of analogous saturated acid; n.s. indicates P > 0.05.
b Indicates level of significance when oxa acid activity is compared with that
of analogous saturated acid; n.s. indicates P > 0.05.
DISCUSSION
pentanoyl-CoA with these enzymes was quite similar.
While the mitochondrial enzymes were able to hy-
drolyze all of the acyl-CoA and oxa acyl-CoA sub-
strates, the rate of hydrolysis of oxa acyl-CoAs was
considerably decreased compared to that of physio-
logic acyl-CoA controls. Reactivity of peroxisomal acyl-
CoA hydrolases (Table 6) was higher with decanoyl-
CoA and octanoyl-CoA compared to pentanoyl-CoA.
Trioxadecanoyl-CoA reactivity with peroxisomal en-
zymeswas significantly lower (37%)thanthat of itscon-
trol. Dioxaheptanoyl-CoA reactivity was less than 50%
that of heptanoyl-CoA. Both oxapentanoyl-CoAs were
less hydrolyzed compared to their control by the perox-
isomal enzymes. Microsomal acyl-CoA hydrolases ex-
hibited similar type of reactivity (Table 7) as did the per-
oxisomal enzymes. Decanoyl-CoA and octanoyl-CoA
were more effective substrates than pentanoyl-CoA for
these enzymes. Trioxadecanoyl-CoA exhibited much
less reactivity with microsomal enzymes (0.47 0.16)
compared to that of decanoyl-CoA (1.58 0.12). Reac-
tivity of dioxaheptanoyl-CoA and both oxapentanoyl-
CoAs with microsomal enzymes was half that of their
respective controls.
This study shows that some medium-chain,
oxygen-substituted carboxylic acids are reasonable
substrates for hepatic acyl-CoA synthetases, especially
those found in the mitochondria and peroxisomes. In-
cluded in this group of oxa acids are two which are
known to be carboxylic acid metabolites of ethylene
glycol-based solvents. 3-Oxapentanoic acid is the major
oxidized metabolite of 2-ethoxyethanol [2]. Likewise,
3,6-dioxaheptanoic acid is derived from the ether gly-
col solvent diglyme [1,1ꢁ-oxybis(2-methoxyethane)] by
O-demethylationandsubsequentalcoholandaldehyde
oxidation [27]. We now consider our findings in light of
current knowledge of acyl-CoA synthetases, oxa acids
and their CoA derivatives, and other xenobiotic car-
boxylic acids.
As detailed in a recent review [28], liver is rich
in acyl-CoA synthetases with short-chain, medium-
chain, long-chain, and very long-chain specificities
with respect to saturated fatty acids. Mitochondria
are reported to have at least two broad specificity
medium-chain acyl-CoA synthetases that not only acti-
vate medium-chain fatty acids but also a wide range of
TABLE 5. Reactivity of Acyl-CoAs with Mitochondrial Acyl-
TABLE 7. Reactivity of Acyl-CoAs with Microsomal Acyl-
CoA Hydrolases
CoA Hydrolases
Chain
Length
Activitya
(nmol/min/mg)
Chain
Length
Activitya
(nmol/min/mg)
Acyl-CoA
Decanoyl-CoA
3,6,9-Trioxadecanoyl-CoA
Octanoyl-CoA
3,6-Dioxaheptanoyl-CoA
Pentanoyl-CoA
3-Oxapentanoyl-CoA
4-Oxapentanoyl-CoA
Pb
Acyl-CoA
Decanoyl-CoA
3,6,9-Trioxadecanoyl-CoA
Octanoyl-CoA
3,6-Dioxaheptanoyl-CoA
Pentanoyl-CoA
3-Oxapentanoyl-CoA
4-Oxapentanoyl-CoA
Pb
Ten
2.07 0.15
0.63 0.23
0.91 0.13
0.31 0.19
1.09 0.19
0.47 0.24
0.61 0.15
Ten
1.58 0.12
0.47 0.16
1.07 0.07
0.40 0.15
0.21 0.05
0.10 0.06
0.10 0.09
<0.01
<0.01
<0.01
<0.01
Eight
Seven
Five
Eight
Seven
Five
<0.01
<0.01
<0.01
n.s.
a Means SD for n ≥ 10 duplicate experiments.
a Means SD for n ≥ 10 duplicate experiments.
b Indicates level of significance when oxa acid activity is compared with that
b Indicates level of significance when oxa acid activity is compared with that
of analogous saturated acid; n.s. indicates P > 0.05.
of analogous saturated acid.

82
PANUGANTI, PENN, AND MOORE
Volume 17, Number 2, 2003
xenobiotic carboxylic acids for subsequent glycine con-
jugation [29,30]. Mitochondria, peroxisomes, and mi-
crosomes appear to have an identical long-chain acyl-
CoA synthetase [31]. While most active with fatty acids
of 10–18 carbons, the long-chain enzyme also functions
on medium and very long-chain substrates [32]. Very
long-chain acyl-CoA synthetase activities have been as-
sociated with peroxisomes and microsomes; however,
substrate-specificity is not well-established [28]. Lay-
ered on to the work with physiologic substrates is a
large body of work focused on acyl-CoA synthetase
activation of xenobiotic carboxylic acids [17]. There
is general agreement that the mitochondrial medium-
chain enzyme can activate some aromatic xenobiotics;
there is conflicting evidence concerning the existence
of one or more additional xenobiotic-specific acyl-CoA
synthetases [17]. In addition, the long-chain acyl-CoA
synthetase, in its microsomal and peroxisomal lo-
cales, has been associated with activation of carboxy-
lates with known hypolipidemic and/or peroxisome-
proliferating properties [33,34].
In light of the above information, the subcellular
preparations used in this study would be expected to
be reactive with the medium-chain (5–10 carbon) phys-
iologic substrates used as controls. In all three fractions,
the decanoic acid was the most effective substrate. The
shorter acids, heptanoic and pentanoic, were activated
most effectively in the mitochondria, perhaps because
of the medium-chain specific enzyme [29]. Heptanoic
and pentanoic acid were activated poorly by the micro-
somal enzymes (presumably the long-chain acyl-CoA
synthetase); peroxisomal enzymes had intermediate
activity, suggesting different long-chain acyl-CoA
specificity, additional synthetases, and/or some mito-
chondrial contamination. When considering the acti-
vation of the oxa acids, a somewhat different pattern
emerges. Overall, it appears that the oxa acids will be
activated most effectively by the mitochondrial and
peroxisomal acyl-CoA synthetases, with relatively little
activation by the microsomal enzymes. Thus the per-
oxisomal and mitochondrial acyl-CoA pools have the
potential for significant oxa acyl-CoA accumulation.
Like acyl-CoA synthetases, acyl-CoA hydrolases
are ubiquitous enzymes found in cytosol, peroxisomes,
endoplasmic reticulum, and mitochondria of mam-
malian cells [18,35,36]. Although these enzymes are
widely distributed, their physiological functions are
not completely understood. An obvious function of
these enzymes is the maintenance of the physiologi-
cal acyl-CoA concentrations and free CoASH pools in
the tissue for fatty acid metabolism [19]. It has been
hypothesized that these enzymes also have the ability
to hydrolyze xenobiotics and, thus, are involved in the
detoxification of nonphysiological metabolites [20,21].
Acyl-CoA hydrolase reaction is a potential mechanism
to oppose an accumulation of detrimental amounts of
acyl-CoAs and subsequent sequestration of CoA. Se-
questration of CoA in the form of acyl-CoAs has a pro-
found inhibitory effect on normal fatty acid metabolism
[37,38]. Thus, acyl-CoA hydrolases play a very impor-
tant role in the regulation of fatty acid metabolism.
As reported in previous studies [36], hydrolase
assay results in this study suggest the presence of
medium- and long-chain acyl-CoA hydrolases in the
mitochondrial fraction and long-chain acyl-CoA hy-
drolases in peroxisomal and microsomal fractions. Just
as decanoic acid was the most effective substrate for
acyl-CoA synthetases in all three fractions, decanoyl-
CoA was the most effective substrate for all hydro-
lase populations. Octanoyl-CoA was hydrolyzed by
all three fractions to the same extent. Pentanoyl-CoA
was hydrolyzed more by mitochondrial enzymes, but
poorly hydrolyzed by peroxisomal and microsomal en-
zymes. The rate of hydrolysis of oxa acyl-CoAs ranged
from 29 to 55% that of physiologic acyl-CoAs.
When the relative activities of synthetase and hy-
drolase were compared, the enzymatic synthesis of
all acyl-CoAs was considerably higher than their hy-
drolysis by respective acyl-CoA hydrolases in all three
fractions. The ratio of activity of synthetase to that
of hydrolase (Figure 2) was considerably higher for
trioxadecanoyl-CoA (60.1) when compared to that of
decanoyl-CoA (43.3) in the mitochondrial fraction, but
the ratios are almost the same in peroxisomal and mi-
crosomal fractions. Dioxaheptanoyl-CoA has shown
considerablyhighersynthetasetohydrolaseratioscom-
pared to its control in all three fractions; a similar
pattern was observed for both oxapentanoyl-CoAs in
all three fractions. The higher synthetase to hydrolase
ratios observed with oxa acyl substrates suggest the po-
tential for significant accumulation of oxa acyl-CoAs
and subsequent CoA sequestration.
Although little work has been done with the
medium-chain length oxa acids which are the focus
of this paper, it is useful to review what is known
about related longer-chain oxa acids. Aarsland and
Berge [34] determined that 3-oxaheptadecanoic acid
and 4-oxaoctadecanoic acid were converted to CoA
esters at rates 25 and 5% that of palmitic acid; the
peroxisome-proliferating effect of the 3-oxa compound
was significantly less than its 3-thia analogue. A series
of studies focused on the use of oxa and thia medium
chain acyl-CoAs as probes for the reaction mechanisms
of enzymes involved in fatty acid ꢀ-oxidation have
shown that 3-oxa or 3-thia substitutions would make
the fatty acid incapable of ꢀ-oxidation [39]. Specifi-
cally, 3-oxaoctanoyl-CoA inhibited medium-chain acyl-
CoA dehydrogenase; in contrast, 4-oxaoctanoyl-CoA is
a weak substrate for the enzyme [39]. A feeding study
with long chain 3-oxa fatty acid resulted in inhibited

Volume 17, Number 2, 2003
CoA ESTERS OF SOLVENT-DERIVED OXA ACIDS
83
FIGURE 2. Comparison of ratios of acyl-CoA synthetase to acyl-CoA hydrolase activities for physiological and oxa acyl-CoAs in various
subcellular fractions.

84
PANUGANTI, PENN, AND MOORE
Volume 17, Number 2, 2003
ethers and their acetates. Int Arch Occup Environ Health
1993;64:495–498.
mitochondrial ꢀ-oxidation, induced acyl-CoA oxidase
levels, and fatty liver development; conversely, 4-oxa
fatty acid stimulated mitochondrial ꢀ-oxidation and
decreased fatty liver [40]. With our finding of 3-oxa and
4-oxa medium chain acyl-CoA formation via mitochon-
drial and peroxisomal acyl-CoA synthetases, there is
a clear potential for interaction of these moieties with
both ꢀ-oxidation systems.
5. Sohnlein B, Letzel S, Weltle D, Rudiger HW, Angerer J.
Occupational chronic exposure to organic solvents. Ex-
aminations concerning the evaluation of a limit value for
2-ethoxyethanol, 2-ethoxyethyl acetate and the genotoxic
effects of these glycol ethers. Int Arch Occup Environ
Health 1993;64:479–484.
6. Valentine R, O’Neill AJ, Lee KP, Kennedy GL Jr. Sub-
chronic inhalation toxicity of diglyme. Food Chem Toxi-
col 1998;37:75–86.
7. Miller RR, Ayres JA, Calhoun LL, Young JT, McKenna
MJ. Comparative short-term inhalation toxicity of ethy-
lene glycol monomethyl ether and propylene glycol
monomethyl ether in rats and mice. Toxicol Appl Phar-
macol 1981;61:368–377.
8. Gray TJB, Moss EJ, Creasy DM, Gangolli SD. Studies on
thetoxicityofsomeglycolethersandalkoxyaceticacidsin
primary testicular cell cultures. Toxicol Appl Pharmacol
1985;79:490–501.
9. Ghanayem BI, Burka LT, Matthews HB. Structure-activity
relationships for the in vitro hematotoxicity of N-
alkoxyacetic acids, the toxic metabolites of glycol ethers.
Chem-Biol Interact 1989;70:339–352.
10. Dodds PF. Incorporation of xenobiotic carboxylic acids
into lipids. Life Sci 1991;49:629–649.
11. Aarsland A, Berge RK, Bremer J, Aarsaether N.
The hypolipidemic peroxisome proliferating drug, bis-
(carboxymethylthio)-1.10 decane, a dicarboxylic metabo-
lite of tiadenol is activated to an acylcoenzyme A
thioester. Biochim Biophys Acta 1990;1033:176–183.
12. Sherratt HAS. Acyl-CoA esters of xenobiotic carboxylic
acids as biochemically active intermediates. Biochem Soc
Trans 1985;13:856–858.
13. Brass EP. Overview of CoA metabolism and its role in
cellular toxicity. Chem-Biol Interact 1994;90:203–214.
14. Knights KM, Roberts BJ. Xenobiotic acyl-CoA formation:
Evidence of kinetically distinct hepatic microsomal long-
chain fatty acid and nafenopin-CoA ligases. Chem-Biol
Interact 1994;90:215–223.
In conclusion, this study demonstrates that a re-
lated group of xenobiotic carboxylic acids can be con-
verted to xenobiotic acyl-CoA moieties by the action
of one or more hepatic acyl-CoA synthetases. We have
also shown that these acyl-CoAs can be hydrolyzed to
some extent by hepatic acyl-CoA hydrolases. Whether
the decreased reactivities of acyl-CoA synthetases and
hydrolases with oxa acyl substrates or the altered syn-
thetase to hydrolase ratios are significant is not yet clear.
The persistence of these oxa acyl-CoA species will de-
pend on a number of other cellular enzymes involved
in fatty acid metabolism, such as carnitine acyltrans-
ferases, acyl-CoA dehydrogenase, monoacyl and di-
acyl glycerol acyltransferases, N-acyltransferase, and
UDP-glucuronyltransferases. Theimpactofthexenobi-
otic acyl-CoA moieties on metabolism may be multiple
[10,13,17] and awaits further study. In addition, a re-
cent study suggests that xenobiotic carboxylates, which
are endogenously converted to acyl-CoAs, may mod-
ulate transcription factor action and subsequent gene
expression [41].
ACKNOWLEDGMENT
Early technical assistance by Amy Komendera and
Xavier Tato is gratefully acknowledged. This project
was supported by Oakland University’s HHMI Un-
dergraduate Program in Biological Communication
and by the Oakland University Research Excellence
Fund.
15. Hutt AJ, Caldwell J. Amino acid conjugation. In: Mulder
GJ, editor. Conjugation Reactions in Drug Metabolism.
London: Taylor and Francis; 1990; pp. 273–305.
16. Yao K-W, Mao L-F, Luo MJ, Schulz H. The relationship be-
tween mitochondrial activation and toxicity of some sub-
stituted carboxylic acids. Chem-Biol Interact 1994;90:225–
234.
17. Knights KM. Role of hepatic fatty acid:CoA ligases in
the metabolism of xenobiotic carboxylic acids. Clin Exper
Pharmacol Physiol 1998;25:776–782.
REFERENCES
18. Alexson SEH, Osmundsen H, Berge RK. The presence of
acyl-CoA hydrolase in rat brown adipose tissue peroxi-
somes. Biochem J 1989;262:41–46.
19. Lee FJS, Lin LW, Smith JA. A glucose-repressible gene
encodes, acetyl-CoAhydrolasefromSaccharomycescere-
visiae. J Biol Chem 1990;265(13):7413–7418.
20. Alexson SEH, Mentlein R, Wersledt C, Hellman U.
Isolation and characterization of microsomal acyl-CoA
thioesterase. Eur J Biochem 1993;214:719–727.
21. Mukherjee JJ, Jay FT, Choy PC. Purification, characteriza-
tion and modulation of a microsomal carboxyl esterase
in rat liver for the hydrolysis of acyl-CoA. Biochem J
1993;295:81–86.
1. Riddick JA, Bunger WB, Sakano TK. Organic solvents.
In: Weissberger A, editor. Techniques of Chemistry. New
York: Wiley; 1986. Vol 2, pp. 1–21.
2. EHC 2-Methoxyethanol, 2-Ethoxyethanol, and Their
Acetates. WHO; Geneva, Switzerland: 1990. Vol. 115,
pp. 11–74.
3. Richards DE, Begley KB, DeBord DG, Cheever KL,
Weigel W, Tirmenstein MA, Savage RE. Comparative
metabolism of bis(2-methoxyethyl)ether in isolated rat
hepatocytes and in the intact rat: Effects of ethanol on in
vitro metabolism. Arch Toxicol 1993;67:531–537.
4. Sakai T, Araki T, Masuyama Y. Determination of uri-
nary alkoxyacetic acids by a rapid and simple method
for biological monitoring of workers exposed to glycol
22. Garras A, Elholm M, Sleboda J, Froyland L, Osmundsen
H, Berge RK. On the effects of thia fatty acid analogues

Volume 17, Number 2, 2003
CoA ESTERS OF SOLVENT-DERIVED OXA ACIDS
85
on hydrolases involved in the degradation of metabolis-
able and non-metabolisable acyl-CoA esters. Xenobiotica
1997;27(8)781–799.
rat liver. Eur J Biochem 1979;98:165–172.
33. Amigo L, McElroy MC, Morales MN, Bronfman M. Sub-
cellular distribution and characteristics of cipofibroyl-
CoA synthetase in rat liver. Its possible identity with
long-chain acyl-CoA synthetase. Biochem J 1992;284:283–
287.
23. DeWitt CM, Moore KH. Capillary electrophoretic assay
of xenobiotic acyl-CoA formation. Am Lab 1995;27:28C–
28I.
24. Gornall AB, Bardawill CS, David MM. Determination of
serum proteins by means of the biuret reaction. J Biol
Chem 1949;177:751–766.
25. Hosaka K, Mishira M, Kamiryo T, Numa S. Long-chain
acyl-CoA synthetase I and II from Candida lipolytica. In:
Lowenstein JM, editor. Methods in Enzymology. San
Diego: Academic Press; 1981; Vol. 71, pp. 325–333.
26. Becker CM, Harris RA. Influence of valproic acid on hep-
atic carbohydrate and lipid metabolism. Archiv Biochem
Biophys 1983;223:381–391.
27. Cheever KL, Richards DE, Weigel W, Lal JB, Dinsmore
AM, Daniel FB. Metabolism of bis (2-methoxyethyl) ether
in the adult rat: Evaluation of the principal metabolite as a
testicular toxicant. Toxicol Appl Pharmacol 1988;94:150–
159.
34. Aarsland A, Berge RK. Peroxisome proliferating sulphur-
and oxy-substituted fatty acid analogues are activated to
acyl CoA thioesters. Biochem Pharmacol 1991;41:53–61.
35. Moore KH, Dandurand DM, Kiechle FL. Fasting induced
alterations in mitochondrial palmitoyl-CoA metabolism
may inhibit adipocyte pyruvate dehydrogenase activity.
Int J Biochem 1992;24(5):809–814.
36. Yamada J, Furihata T, Tamura H, Watanabe T, Suga T.
Long chain acyl-CoA hydrolase from rat brain cytosol;
purification, characteraization and immunohistochemi-
cal localization. Arch Biochem Biophys 1996;326(1):106–
114.
37. Alexson SEH, Svensson LT, Nedergaard J. NADH sen-
sitive propionyl-CoA hydrolase in brown adipose tis-
sue mitochondria of the rat. Biochim Biophys Acta
1989;1005:13–19.
38. Brass EP, Ruff LJ. Rat hepatic CoA is redistributed in re-
sponse to mitochondrial acyl-CoA accumulation. J Nutr
1992;122(11):2094–2100.
39. Lau S-M, Brantley RK, Thorpe C. The reductive half-
reaction in acyl-CoA dehydrogenase from pig kidney:
Studies with thiaoctanoyl-CoA and oxaoctanoyl-CoA
analogues. Biochem 1988;27:5089–5095.
40. Skorve J, Asiedu D, Solbakken M, Gjestdal J, Songstad
J, Berge RK. Comparative effects of oxygen and sulfur-
substituted fatty acids on serum lipids and mitochon-
drial and peroxisomal fatty acid oxidation in rat. Biochem
Pharmacol 1992;43:815–822.
28. Watkins PA. Fatty acid activation. Prog Lipid Res
1997;36:55–83.
29. Kasuya F, Igarashi K, Fukui M. Inhibition of
a
medium chain acyl-CoA synthetase involved in glycine
conjugation by carboxylic acids. Biochem Pharmacol
1996;52:1643–1646.
30. Vessey DA, Kelley M, Wassen RS. Characterization of
the CoA ligases of human liver mitochondria catalyz-
ing the activation of short- and medium-chain fatty acids
and xenobiotic carboxylic acids. Biochim et Biophys Acta
1999;1428:455–462.
31. Miyazawa S, Hashimoto T, Yokota S. Indentity of long-
chain acyl-CoA synthetase of microsomes, mitochondria,
and peroxisomes in rat liver. J Biochem 1985;98:723–733.
32. Tanaka T, Hosaka K, Hoshimaru M, Numa S. Purification
and properties of long-chain acyl-CoA synthetase from
41. Hertz R, Magenheim J, Berman I, Bar-Tana J. Fatty acyl-
CoA thioesters are ligands of hepatic nuclear factor-4ꢂ.
Nature 1998;392:512–516.