Spiropentylacetyl-CoA, a mechanism-based inactivator of acyl-CoA dehydrogenases

Article abstract of DOI:10.1021/ja9737634

Acyl-CoA dehydrogenases (ACDs) are FAD-dependent enzymes that catalyze the conversion of an appropriate fatty acyl-CoA thioester substrate to the corresponding trans-α,β-enoyl-CoA product. Early studies have shown that the de hydrogenation stereospecific and is initiated by the abstraction of the pro-R α-H, followed-by the transfer of the pro-R β-H, as a hydride equivalent, to the bound FAD. However, recent studies of the inactivation of ACDs by a metabolic of hypoglycin A, (methylenecyclopropyl)acetyl-CoA (MCPA- CoA), led to an alternative mechanism in which the reducing equivalent is delivered from the initially formed α-anion to the bound FAD via a single electron transfer process. To further explore the observed mechanistic discrepancy, we have reexamined the inhibitory properties of a closely related MCPA-CoA analogue, spiropentylacetyl-CoA (SPA-CoA), which was previously reported as a fight-binding inhibitor for ACDs. In contrast to early results, our data showed the SPA-CoA is a mechanism-based inhibitor for pig kidney medium-chain acyl-CoA dehydrogenase (MCAD) and Megasphaera elsdenii short-chain acyl-CoA dehydrogenase (SCAD) and that the inactivation is time-dependent, active-site-directed, and irreversible. More importantly, both (R)- and (S)-SPA-CoA could effectively inactivate MCAD, and the resulting inhibitor-FAD adducts appear to have one of the three-membered rings of the spiropentyl moiety cleaved. Since the inactivation is nonstereospecific with respect to C(β)-C bond scission, the ring opening of SPA-CoA leading to enzyme inactivation is likely initiated by a spiropentylcarbinyl radical. Such a radical-induced ring fragmentation is expected to be extremely facile and may bypass the chiral discrimination normally imposed by the enzyme. Thus, these results are consistent with our early notion that MCAD is capable of mediating one-electron redox chemistry. Interestingly, it was also found that (R)-SPA-CoA is an irreversible inhibitor for while the S-epimer is only a competitive inhibitor for the same enzyme. The selective inhibition exhibited by these compounds against two closely related dehydrogenases is likely a consequence of the distinct steric and electronic demands imposed by the active sites of MCAD and SCAD. Such information is important for the design of novel class-selective inhibitors to control and/or regulate fatty acid metabolism.

Full text of DOI:10.1021/ja9737634

2008
J. Am. Chem. Soc. 1998, 120, 2008-2017
Spiropentylacetyl-CoA, A Mechanism-Based Inactivator of Acyl-CoA
Dehydrogenases
†
‡
§
|
Ding Li, Hui-qiang Zhou, Srikanth Dakoji, Injae Shin, Eugene Oh, and Hung-wen Liu*
Contribution from the Department of Chemistry, UniVersity of Minnesota, Minneapolis, Minnesota 55455
ReceiVed October 31, 1997
Abstract: Acyl-CoA dehydrogenases (ACDs) are FAD-dependent enzymes that catalyze the conversion of an
appropriate fatty acyl-CoA thioester substrate to the corresponding trans-R,â-enoyl-CoA product. Early studies
have shown that the dehydrogenation is stereospecific and is initiated by the abstraction of the pro-R R-H,
followed by the transfer of the pro-R â-H, as a hydride equivalent, to the bound FAD. However, recent
studies of the inactivation of ACDs by a metabolite of hypoglycin A, (methylenecyclopropyl)acetyl-CoA
(MCPA-CoA), led to an alternative mechanism in which the reducing equivalent is delivered from the initially
formed R-anion to the bound FAD via a single electron transfer process. To further explore the observed
mechanistic discrepancy, we have reexamined the inhibitory properties of a closely related MCPA-CoA analogue,
spiropentylacetyl-CoA (SPA-CoA), which was previously reported as a tight-binding inhibitor for ACDs. In
contrast to early results, our data showed that SPA-CoA is a mechanism-based inhibitor for pig kidney medium-
chain acyl-CoA dehydrogenase (MCAD) and Megasphaera elsdenii short-chain acyl-CoA dehydrogenase
(SCAD) and that the inactivation is time-dependent, active-site-directed, and irreversible. More importantly,
both (R)- and (S)-SPA-CoA could effectively inactivate MCAD, and the resulting inhibitor-FAD adducts
appear to have one of the three-membered rings of the spiropentyl moiety cleaved. Since the inactivation is
nonstereospecific with respect to Câ-C bond scission, the ring opening of SPA-CoA leading to enzyme
inactivation is likely initiated by a spiropentylcarbinyl radical. Such a radical-induced ring fragmentation is
expected to be extremely facile and may bypass the chiral discrimination normally imposed by the enzyme.
Thus, these results are consistent with our early notion that MCAD is capable of mediating one-electron redox
chemistry. Interestingly, it was also found that (R)-SPA-CoA is an irreversible inhibitor for SCAD, while the
S-epimer is only a competitive inhibitor for the same enzyme. The selective inhibition exhibited by these
compounds against two closely related dehydrogenases is likely a consequence of the distinct steric and electronic
demands imposed by the active sites of MCAD and SCAD. Such information is important for the design of
novel class-selective inhibitors to control and/or regulate fatty acid metabolism.
Acyl-CoA dehydrogenases (ACDs) are FAD-dependent en-
zymes that catalyze the conversion of the fatty acyl-CoA
has been shown to be pro-R specific as well.4 Since the
desaturation is effectively a two-electron redox reaction, relaying
the reducing equivalents from intermediate 2 to the active-site-
bound flavin is commonly believed to be a direct hydride
substrate (1) to the corresponding trans-R,â-enoyl-CoA product
1
(
3), with concomitant reduction of the flavin coenzyme. There
1
,3
exists six distinct subclasses of ACDs involved in fatty acid
metabolism, short-chain (SCAD), medium-chain (MCAD), long-
chain (LCAD), very long chain (VLCAD), isovaleryl-CoA
transfer process (eq 1). However, oxidation of the carbanion
(2) (a) Beinert, H. Enzymes, 2nd ed.; 1963, 7, 447. (b) Engel, P. C. In
Chemistry and Bio-chemistry of FlaVoenzymes; M u¨ ller, F., Ed.; CRC
Press: New York, 1991; p 597. (c) Schulz, H. In Fatty Acid Oxidation:
Clinical, Biochemical, and Molecular Aspects; Tanaka, K., Coates, P. M.,
Eds.; Alan R. Liss: New York, 1990; p 23. (d) Ikeda, Y.; Okamura-Ikeda,
K.; Tanaka, K. J. Biol. Chem. 1985, 260, 1311. (e) Izai, K.; Uchida, Y.;
Ori, T.; Yamamoto, S.; Hashimoto, T. J. Biol. Chem. 1992, 267, 1027.
(3) (a) Ghisla, S.; Engst, S.; Moll, M.; Bross, P.; Strauss, A.; Kim, J. J.
P. In New DeVelopments in Fatty Acid Oxidation; Coates, P. M., Tanaka,
K., Eds.; Wiley-Liss: New York, 1992; p 127. (b) Kim, J. J. P. In FlaVins
and FlaVoproteins; Curti, B., Ronchi, S., Zanetti, G., Eds.; de Gruyter: New
York, 1991; p 291. (c) Kim, J. J. P.; Wang, M.; Djordjevic, S.; Paschke, R.
In New DeVelopments in Fatty Acid Oxidation; Coates, P. M., Tanaka, K.,
Eds.; Wiley-Liss: New York, 1992; p 111. (d) Kim, J. P.; Wang, M.;
Paschke, R. Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 7523. (e) Bross, P.;
Engst, S.; Strauss, A. W.; Kelley, D. P.; Rasched, I.; Ghisla, S. J. Biol.
Chem. 1990, 265, 7116. (f) Becker, D. F.; Fuchs, J. A.; Banfield, D. K.;
Walter, F. D.; MacGillivray, R. T. A.; Stankovich, M. T. Biochemistry 1993,
32, 10736.
(
(
IVD), and short/branched chain acyl-CoA dehydrogenases
SBCAD).2 While each of these enzymes displays a preferential
substrate specificity, they all adopt a common mechanism in
which the initial step has been established to be the abstraction
of the pro-R R-H by a conserved glutamate in the active site.
This is followed by the removal of a â-H from the nascent
carbanion intermediate 2, and the stereochemistry of this step
3
†
Current address: Department of Chemistry, Texas A&M University,
College Station, TX 77843-3255.
‡
Current address: Eppley Institute for Research in Cancer, University
of Nebraska Medical Center, Omaha, NE 69198-6805.
§
Current address: Department of Chemistry, Yensei University, Seoul,
Korea.
|
Current address: Samsung Advanced Institute of Technology, 103-12,
Moonji-Dong, Yusong-Gu, Taejon, Korea 305-380.
(1) (a) Thorpe, C.; Matthews, R. G.; Williams, C. H. Biochemistry 1979,
(4) (a) Biellmann, J. F.; Hirth, C. G. FEBS Lett. 1970, 9, 55, 335. (b)
Bucklers, L.; Umani-Ronchi, A. S.; Ret e´ y, J.; Arigoni, D. Experientia 1970,
26, 931. (c) Dommes, V.; Kunau, W.-H. J. Biol. Chem. 1984, 259, 1789.
(d) Ikeda, Y.; Tanaka, K. In Fatty Acid Oxidation: Clinical, Biochemical,
and Molecular Aspects; Tanaka, K., Coates, P. M., Eds.; Alan R. Liss: New
York, 1990; p 37.
1
8, 331. (b) Ghisla, S.; Thorpe, C.; Massey, V. Biochemistry 1984, 23,
154. (c) Pohl, B.; Raichle, T.; Ghisla, S. Eur. J. Biochem. 1986, 160, 109.
3
(
d) Ghisla, S. In FlaVins and FlaVoproteins; Bray, R. C., Engel, P. C.,
Mayhew, S. G., Eds.; Walter de Gruyter & Co.: Berlin, 1984; p 385 and
references cited therein.
S0002-7863(97)03763-3 CCC: $15.00 © 1998 American Chemical Society
Published on Web 02/24/1998

Spiropentylacetyl-CoA
via a single electron transfer route involving the formation of
J. Am. Chem. Soc., Vol. 120, No. 9, 1998 2009
2
Scheme 1
a transient radical species (4) and a flavin semiquinone radical
is also possible (eq 2). The latter hypothesis has gained support
5
from recent inactivation studies described below.
Over the last two decades, a wide variety of mechanism-
based inhibitors have been reported for this important class of
enzymes.6 These inhibitors can be divided into two groups
depending on whether the enzyme itself or the FAD coenzyme
is the target for their actions. For example, the 2- and 3-alkynoyl
thioester derivatives have been shown to inactivate the acyl-
CoA dehydrogenases via trapping of the active-site base,
glutamate.6 The inactivation mechanism involves the abstrac-
tion of the R- or γ-H of the inhibitor by the active-site base,
glutamate, to generate a reactive 2,3-allenic CoA intermediate,
which is suitably poised to trap the same glutamate (Glu-367
in bacterial SCAD and Glu-376 in pig liver MCAD). On the
contrary, (methylenecyclopropyl)acetyl-CoA (MCPA-CoA, 5),
a metabolite of hypoglycin A, inactivates both MCAD and
metabolism.9 This synthetic byproduct was later found to be
spiropentylacetic acid (SPA), and its inhibitory activity was
manifested both in vivo (with SPA) and in vitro (with SPA-
CoA, 16) in a rat model. An early study had shown that SPA
itself is stable to alkaline hydrolysis with no detectable ring
1
0
opening, which led Tserng et al. to conclude that SPA-CoA
,7
9
may simply act as a tight-binding inhibitor. While this
postulation is plausible, it should be noted that SPA-CoA shares
a highly strained ring skeleton with MCPA-CoA (5), and more
bulky derivatives such as 3-indolepropionyl-CoA and trans-3-
indoleacryloyl-CoA are known substrates of acyl-CoA dehy-
1
1
drogenase. Thus, it is perplexing why SPA-CoA (16) can fit
into the active site and yet not be processed by this enzyme. In
an attempt to clarify this ambiguity, we have reexamined the
inhibitory properties of SPA-CoA against two acyl-CoA
5
,8
SCAD via covalent modification of FAD (Scheme 1), and
both enantiomers of MCPA-CoA have been shown to be
effective inhibitors.5
a,b,d
The lack of stereospecificity of bond
1
2
dehydrogenases: pig liver MCAD and a recombinant Mega-
rupture at Câ of MCPA-CoA (5) is interesting and strongly
implies that this ring-opening step leading to inactivation is
likely a spontaneous event induced by a cyclopropylcarbinyl
3
f
sphaera elsdenii SCAD. A detailed mechanistic inquiry to
address the time dependency, stereospecificity/stereoselectivity,
and the nature of this inactivation was performed with pure
enzymes. Reported herein are our results and the mechanistic
implications.
5
radical (7). The identification of 9 as the major turnover
product, when the incubation was conducted aerobically, further
sustains the intermediacy of a ring-opened radical 8 (Scheme
5
e,g
1
).
Thus, these results may be extrapolated to support the
Results and Discussions
notion that dehydrogenases are capable of mediating one-
electron oxidation-reduction.
In a recent investigation of disordered fatty acid metabolism
in MCPA-treated rats, Tserng et al. came across a synthetic
byproduct which inhibits MCAD without affecting amino acid
Preparation of Spiropentylacetyl-CoA (16). The precursor
for the synthesis of spiropentylacetyl-CoA (16, SPA-CoA) is
(methylenecyclopropyl)acetic acid (13, MCPA), which is de-
rived from cyclopropanation of 3-buten-1-ol (10) by 1,1-
13
dichloroethane in the presence of n-butyllithium. The resulting
product 11 was subjected to base-induced elimination, depro-
tection, and oxidation in sequence to give 13. Conversion of
13 to SPA (15) was achieved by the treatment with diazo-
methane and lead tetraacetate to give 14, followed by alkaline
(
5) (a) Lenn, N. D.; Shih, Y.; Stankovich, M. T.; Liu, H.-w. J. Am. Chem.
Soc. 1989, 111, 3065. (b) Lai, M.-t.; Liu, H.-w. J. Am. Chem. Soc. 1990,
12, 4034. (c) Baldwin, J. E.; Ostrander, R. L.; Simon, C. D.; Widdison,
1
W. C. J. Am. Chem. Soc. 1990, 112, 2021. (d) Lai, M.-t.; Liu, L.-d.; Liu,
H.-w. J. Am. Chem. Soc. 1991, 113, 7388. (e) Lai, M.-t.; Liu, H.-w. J. Am.
Chem. Soc. 1992, 114, 3160. (f) Baldwin, J. E.; Widdison, W. C. J. Am.
Chem. Soc. 1992, 114, 2245. (g) Lai, M.-t.; Li, D.; Oh, E.; Liu, H.-w. J.
Am. Chem. Soc. 1993, 115, 1619.
1
4
hydrolysis of the methyl ester (Scheme 2). Condensation of
SPA with isobutyl chloroformate and coupling to coenzyme
(
6) Cummings, J. G.; Thorpe, C. Biochemistry 1994, 33, 788.
5g,15
A
in aqueous THF solution (pH 8-8.5) afforded the desired
(7) (a) Fendrich, G.; Abeles, R. H. Biochemistry 1982, 21, 6685. (b)
spiropentylacetyl-CoA (16). This compound was later purified
by a HPLC Partisil-C18 column, and the overall yield from
MCPA (13) was 50%.
Powell, P. J.; Thorpe, C. Biochemistry 1988, 27, 8022. (c) Freund, K.;
Mizzer, J. P.; Dick, W.; Thorpe, C. Biochemistry 1985, 25, 5996. (d)
Lundberg, N. N.; Thorpe, C. Arch. Biochem. Biophys. 1993, 454. (e)
Frerman, F. E.; Miziorko, H. M.; Beckmann, J. D. J. Biol. Chem. 1980,
2
2
55, 11192. (f) Gomes, B.; Fendrich, G.; Abeles, R. H. Biochemistry 1981,
0, 1481.
(9) Tserng, K. Y.; Jin, S.-J.; Hoppel, C. L. Biochemistry 1991, 30, 10755.
(10) Ullman, E. F.; Fanshawe, W. J. J. Am. Chem. Soc. 1961, 83, 2379.
(11) (a) Rojas, C.; Schmidt, J.; Lee, M.-Y.; Gustafson, W. G.; McFarland,
J. T. Biochemistry 1985, 24, 2947. (b) Johnson, J. K.; Srivastava, D. K.
Biochemistry 1993, 32, 8004.
(8) (a) Osmundsen, H.; Sherratt, H. S. A. FEBS Lett. 1975, 55, 81. (b)
Kean, E. A. Biochem. Biophys. Acta 1976, 422, 81. (c) Abeles, R. H. In
Enzyme ActiVated IrreVersible Inhibitors; Seiler, N., Jung, M. J., Koch-
Weser, J., Eds.; Elsevier-North-Holland: Amsterdam, 1978; p 1. (d) Ghisla,
S.; Wenz, A.; Thorpe, C. In Enzyme Inhibitors; Brodbeck, U., Ed.; Verlag
Chim: Weinheim, Germany, 1980; p 43. (e) Wenz, A.; Thorpe, C.; Ghisla,
S. J. Biol. Chem. 1981, 256, 9809. (f) Ghisla, S.; Melde, K.; Zeller, H. D.;
Boschert, W. In Fatty Acid Oxidation; Clinical, Biochemical, and Molecular
Aspects; Tanaka, K., Coates, P. M., Eds.; Alan R. Liss, Inc.: New York,
(12) Thorpe, C. Methods Enzymol. 1981, 71, 366.
(13) Baldwin, J. E.; Parker, D. W. J. Org. Chem. 1987, 52, 1475.
(14) Zefirov, N. S.; Kozhushkov, S. I.; Ugrak, B. I.; Lukin, K. A.;
Kokoreva, O. V.; Yufit, D. S.; Struchkov, Y. T.; Zoellner, S.; Boese, R.;
de Meijere, A. J. Org. Chem. 1992, 57, 701.
(15) Dakoji, S.; Shin, I.; Becker, D. F.; Stankovich, M. T.; Liu, H.-w. J.
Am. Chem. Soc. 1996, 118, 10971.
1
990; p 185.

2010 J. Am. Chem. Soc., Vol. 120, No. 9, 1998
Li et al.
Scheme 2
lent further credence supporting that FAD of MCAD is
covalently modified by SPA-CoA. The above experiments were
also performed with SCAD, and a similar conclusion was
reached.
Determination of Stoichiometry of Inactivation of MCAD
and SCAD by SPA-CoA. The stoichiometry of the inactivation
between SPA-CoA and acyl-CoA dehydrogenases was deter-
3
mined using [R- H]SPA-CoA (18), which was prepared by the
reactions delineated in Scheme 3. The synthesis involved
treatment of compound 14 with lithium diisopropylamide
3
followed by quenching with HOH. The resulting product 17
was hydrolyzed in alkaline solution and then coupled with
3
coenzyme A to afford [R- H]SPA-CoA (18) in 52% overall yield
with a specific activity of 0.32 mCi/mmol. The reaction
2
conditions were first tested using H2O in the quenching step;
the isotope incorporation was found to be regioselective at the
R-C, and the extent of incorporation was g90%. Ten molar
equivalents of the labeled SPA-CoA (18) was incubated with
MCAD (40 nmol), and the inactivated enzyme was exhaustively
dialyzed as previously described. Appropriate controls without
inactivator and/or with boiled enzyme were also prepared. Our
results established a 1:1 stoichiometry between SPA-CoA and
inactivated MCAD (per subunit). It should be noted that
calculations to determine the stoichiometry of inactivation with
racemic 18 were calibrated to account for 50% loss of the
radiolabels prior to covalent modification of the enzyme. When
an analogous experiment was performed with SCAD, a similar
Studies of Activity of Spiropentylacetyl-CoA on MCAD
and SCAD. The effect of SPA-CoA (16) on the catalytic
activity of MCAD was investigated by the incubation of MCAD
with 10 mol equiv of SPA-CoA under aerobic conditions at 25
°
C. As shown in Figure 1, time-dependent loss of activity of
MCAD was noted during the incubation. It was also found
that bleaching of the flavin chromophore (λmax 450 nm) ensued
concurrently with the loss of enzyme activity (Figure 1). Since
the activity and the bleached flavin chromophore of the
inactivated enzyme remained unchanged after prolonged dialy-
sis, the inactivation is clearly irreversible and most likely
involves covalent modification of the FAD coenzyme. Similar
results were also observed when SCAD was treated with SPA-
CoA. Again, time-dependent irreversible inactivation of the
enzyme occurred with concomitant bleaching of the active-site
flavin coenzyme.
Denaturation of Inactivated Enzyme. MCAD that had been
inactivated with SPA-CoA was denatured by mixing vigorously
with four volumes of methanol at 25 °C. This treatment released
the noncovalently bound FAD from the active site of MCAD
into the buffer solution. The suspension was centrifuged, and
the protein precipitate was washed with 50 mM potassium
phosphate buffer, pH 7.6. The supernatant and the washings
were pooled, and the methanol was evaporated by a steady
stream of nitrogen. While the reduced flavin in free solution
is known to be rapidly reoxidized upon exposure to air, the
spectrum of the combined supernatant showed no indication of
any oxidized flavin chromophore. This observation provided
additional evidence substantiating the formation of covalently
modified FAD-inhibitor adducts as the cause of inactivation.
Similar results were also observed when SCAD was exposed
to SPA-CoA. Thus, an analogous mode of inactivation is
proposed for the latter case as well.
1
:1 ratio was also found between inactivated SCAD and
inhibitor. Such a stoichiometric ratio (1:1) is consistent with
an inactivation mechanism involving covalent modification of
the active site flavin coenzyme.
3
Analysis of MCAD Inactivated by [r- H]SPA-CoA. MCAD
3
was incubated with 0.5 equiv of [R- H]SPA-CoA (18) in 50
mM potassium phosphate buffer (pH 7.6) at room temperature
for 30 min. The reaction was quenched with active charcoal
5
d
(
10% solution), and the resulting suspension was mixed
vigorously and then centrifuged to precipitate the charcoal.
Interestingly, approximately 52% of the total radioactivity was
found in the supernatant. Since only substoichiometric amounts
of inhibitor are used, and the CoA derivatives and protein
molecules are readily adsorbed by charcoal, the majority of
radioactivity detected in the supernatant must result from
3
enzyme-mediated solvent exchange of [R- H]SPA-CoA. Thus,
an R-proton abstraction by the active-site base, Glu-376, leading
to its exchange with the solvent water, must be a requisite step
of SPA-CoA-mediated inactivation.
In a separate experiment, the incubation, carried out under
identical conditions, was quenched with 90% methanol instead
of charcoal. The denatured protein was precipitated, washed
free of inhibitor, and submitted for scintillation counting. Since
only about 1% of the total radioactivity was detected in the
protein precipitate, the inactivation path leading to modification
of the protein itself is insignificant. It should be pointed out
that all of these readings had been calibrated against controls
prepared in parallel with boiled enzyme. Hence, the above
findings provided direct evidence corroborating that inactivation
is likely induced by an R-proton abstraction, analogous to the
first catalytic step, and the principal target of SPA-CoA is the
flavin cofactor, not the enzyme itself.
Treatment of Reduced Flavin Adducts with Ferricenium
Hexafluorophosphate. It has been shown that ferricenium salt
+
-
(
Fc PF6 ), a strong oxidant, can be used as an alternate electron
acceptor to effectively reoxidize substrate-reduced acyl-CoA
dehydrogenases.7c,16 To assess the effect of ferricenium salt
on MCAD which has been inactivated by SPA-CoA, three
portions of FcPF6 (5 mol equiv each) were added in succession
to the sample solution and the change of the electronic spectrum
was monitored. Not surprisingly, neither MCAD reactivation
nor FAD reoxidation occurred. The inability of ferricenium
salt to reoxidize the modified flavin was also noted when MCAD
was pretreated with MCPA-CoA (5), an inhibitor which is
known to form covalent adduct(s) with FAD.5 These findings
Studies of Stereospecificity of Inactivation of MCAD and
SCAD by SPA-CoA. The mechanism emerged from the above
results is similar to that proposed for the inactivation of MCAD
and SCAD by MCPA-CoA (5) (Scheme 1). To test whether
the ring cleavage is a stereospecific process, we have prepared
,8
(
16) Lehman, T. C.; Hale, D. E.; Bhala, A.; Thorpe, C. Anal. Biochem.
1
990, 186, 280.

Spiropentylacetyl-CoA
J. Am. Chem. Soc., Vol. 120, No. 9, 1998 2011
Figure 1. Effect of SPA-CoA on the spectrum of MCAD. MCAD, 13.0 mM in 0.4 mL of 50 mM potassium phosphate buffer, pH 7.6, at 25 °C,
was incubated with 10 equiv of SPA-CoA (16) under aerobic conditions. Intermediate spectra were recorded after incubation for (a) 0 min, (b) 2
min, (c) 5 min, (d) 10 min, and (e) 30 min, respectively. The inset shows the effect of SPA-CoA on the activity of MCAD under the same
conditions.
Scheme 3
Scheme 4
(
R)- and (S)-SPA-CoA (19 and 20, respectively). The key
intermediates to these two enantiomers were (R)- and (S)-MCPA
21 and 22, respectively), and their synthesis followed a reported
(
17
procedure. Subsequent conversion of MCPA to SPA-CoA was
achieved through a series of reactions adapted from those
delineated in Scheme 2, involving methylation of MCPA as a
methyl ester, cyclopropanation of the exocyclic methylene
group, hydrolysis of the ester group, and coupling of the
resultant carboxylic acids with coenzyme A (Scheme 4). The
effect of these SPA-CoA isomers on the catalytic activity of
MCAD was analyzed by a method analogous to the successive
titration reported by Wenz et al.8e As shown in Figure 2, a
plot of the residual activity observed under aerobic conditions
versus total equivalents of SPA-CoA added led to a partition
ratio of approximately 2 for both (R)- and (S)-SPA-CoA. Since
the partition ratio defines the number of latent inhibitor
molecules converted to product relative to each turnover leading
to enzyme inactivation, the equivalent ratios strongly suggested
that the inactivation by either (R)- or (S)-SPA-CoA follows the
same chemical course. Thus, the Câ bond cleavage must be
nonstereospecific since both enantiomers of SPA-CoA are
competent irreversible inactivators for MCAD.
Furthermore, no turnover products could be detected in the
reaction mixture, and no R-H exchange was noted when (S)-
SPA-CoA was recovered from the enzyme incubation in
deuterated buffer (pD 7.2). In fact, (S)-SPA-CoA exhibited a
competitive inhibitory effect against MCPA-CoA (5) with
SCAD, as judged by the 69% reduction in the rate of inactivation
of SCAD when (S)-SPA-CoA was coincubated with 5. These
findings implicate that (S)-SPA-CoA can bind to the active site
of SCAD, albeit in a nonproductive manner. It is likely that
the steric/spatial demands of the relatively bulky spiropentyl
moiety may have obtruded the active-site base, Glu-367, of
SCAD to effectively abstract the R-H. Thus, the fact that (S)-
SPA-CoA is an irreversible inhibitor for MCAD but only a weak
competitive inhibitor for SCAD makes it a new class-selective
inhibitor for this important family of enzymes.
Analogous experiments were also performed with SCAD, and
a partition ratio of 30 was estimated for (R)-SPA-CoA (19),
which is an irreversible inactivator for SCAD. However, to
our surprise, the activity of SCAD was only slightly affected
upon treatment with excess (S)-SPA-CoA (20, 300 equiv) and
the absorbance of FAD was unchanged during incubation.
Kinetic Studies of Inhibition of MCAD and SCAD by (R)-
and (S)-SPA-CoA. The competence of each enantiomer of
SPA-CoA to inactivate MCAD and SCAD was further studied
by kinetic analysis. As mentioned above, incubation of SCAD
with (R)-SPA-CoA (19) resulted in time-dependent loss of
(17) (a) Kabat, M. M.; Wicha, J. Tetrahedron Lett. 1991, 32, 531. (b)
Lai, M.-t.; Oh, E.; Shih, Y.; Liu, H.-w. J. Org. Chem. 1992, 57, 2471. (c)
Lai, M.-t.; Oh, E.; Liu, H.-w. Bioorg. Med. Chem. Lett. 1992, 2, 1423.

2012 J. Am. Chem. Soc., Vol. 120, No. 9, 1998
Li et al.
Figure 4. Effect of (R)- and (S)-SPA-CoA on the spectrum of MCAD
monitored at 450 nm. MCAD, 13.0 mM in 0.4 mL of 50 mM potassium
phosphate buffer, pH 7.6, was incubated under aerobic conditions at
Figure 2. Effect of SPA-CoA on the catalytic activity of MCAD. A
series of samples containing MCAD (13.0 mM) and appropriate
amounts of SPA-CoA in 0.4 mL of 50 mM potassium phosphate buffer
2
5 °C with 20 equiv of inhibitor: (A) (R)-SPA-CoA (19) and (B) (S)-
SPA-CoA (20). A is the absorbance at a given time during the
incubation, and A is the absorbance at the end of the inactivation. The
slope of the semilogarithmic plot of the changes of the absorbance (A
) versus the incubation time gave the pseudo-first-order rate
constants.
t
f
(pH 7.6) were prepared. The incubation was carried out aerobically at
f
25 °C for 1 h, and the residual activity of each sample was then assayed.
-
A
t
These figures show the percentage of residual activity versus the ratio
of SPA-CoA to enzyme: (A) (R)-SPA-CoA (19) and (B) (S)-SPA-
CoA (20).
spectrophotometrically at 450 nm. Shown in Figure 4, At
represents the absorbance at a given time during the incubation,
while the absorbance at the end of the inactivation is designated
as Af. The slope of a semilogarithmic plot of the changes of
the absorbance (Af - At) versus the incubation time gave the
-1
observed inactivation rate constants (kobs) of 0.15 and 0.41 min
for (R)- and (S)-SPA-CoA (0.26 mM), respectively, under the
1
8
assay conditions.
Although the partition ratios for the
inactivation of MCAD are the same for both isomers, the
inactivation caused by (S)-SPA-CoA (20) is about 2.7 times
faster than the observed rate for the R-enantiomer (19). Given
the steric bulk of the spiropentyl group, it is conceivable that
the alignment of the CR-H bond toward the active-site base is
different for the R- and S-isomers. Such an aberrance in bond
orientation may have hampered the abstraction of the CR proton
by the active site base and contributed to the disparity of the
inactivation rate. It is noteworthy that the rate of inactivation
of MCAD by SPA-CoA is about 2.5-7 times slower than that
Figure 3. Effect of (R)-SPA-CoA on the catalytic activity of SCAD.
The enzyme (9.0 mM) was incubated aerobically with (R)-SPA-CoA
(19) in 0.4 mL of 50 mM potassium phosphate buffer, pH 7.6, at 25
1
7c
°C. The observed inactivation rates (kobs) at different inhibitor concen-
by MCPA-CoA.
Thus, MCPA-CoA is a more effective
trations (16.0, 31.9, 114, 570, and 1140 mM) were recorded and plotted
against inhibitor concentration. The inset shows the double-reciprocal
plot of kobs versus (R)-SPA-CoA concentration.
inactivator than SPA-CoA.
Characterization of the Covalent Adduct(s) between
MCAD and SPA-CoA. While deprotonation of an R-H and
formation of a covalent adduct with flavin as the initial and
final episodes of SPA-CoA-mediated inactivation are established
by the above experiments, the detailed reaction course and the
nature of the reactive intermediate(s) involved remain elusive.
Since both SPA-CoA and MCPA-CoA possess a highly strained
enzyme activity. The loss of activity follows saturation kinetics
and also exhibits dependence on inhibitor concentration. The
apparent dissociation constant (KI) and the inactivation rate
constant (kinact) were deduced from the double-reciprocal plots
of first-order rate constants (kobs) versus values of inhibitor
concentration. As shown in Figure 3, the KI and kinact were
(18) Attempts to determine kinact and K for the inactivation of MCAD
I
-
1
determined to be 25.5 mM and 0.11 min , respectively, from
this analysis. Since bleaching of the FAD chromophore
accompanied time-dependent loss of enzymic activity, the extent
of inhibition of MCAD by (R)- and (S)-SPA-CoA was followed
by 19 or 20 based on a similar approach shown in Figure 3 proved to be
problematic. The inactivation follows first-order kinetics only at high
inhibitor concentration but deviates from the first-order kinetics at low
inhibitor concentration. The amount of inhibitor (260 mM) used in the
incubations shown in Figure 4 was at the saturation conditions.

Spiropentylacetyl-CoA
J. Am. Chem. Soc., Vol. 120, No. 9, 1998 2013
Scheme 5
cyclic skeleton, it is conceivable that the inactivation by SPA-
CoA, similar to that caused by MCPA-CoA, also involves
formation of a ring-opened reactive intermediate which couples
with FAD. Thus, structural characterization of the resultant
inhibitor-FAD adduct(s) may provide significant insights into
the inactivation mechanism. To facilitate the structural assign-
ment, we have chemically synthesized the doubly labeled (C4′
1
3
C- H detected HMQC spectrum (500 MHz, 2H
1
Figure 5.
2
O)
showing the resonances of the covalent adducts between MCAD and
2′,4′- C ]spiropentylacetyl-CoA (23). Number of transients ) 64;
2
temperature ) 25 °C; accumulation time ) 10 h.
13
[
13
and C2′, 23) and the singly labeled (C4′, 24) [ C]SPA-CoA.
As shown in Scheme 5, cyclopropanation of allenic alcohol 25
13
using 2 equiv of [ C]CH2I2 and 3 equiv of trimethylaluminum
fragmentation of one or both rings of the spiropentyl moiety to
give ring-opened intermediates (31-33), which can then co-
valently modify the FAD coenzyme (route A and/or B).
However, the fact that the ring skeleton of methyl spiropentyl-
acetate, a model for SPA-CoA, is stable under basic conditions10
infers that the relief of ring strains by the R-anion-induced ring
scission is insufficient to compensate for the required activation
gave 26 in good yield.19 The [2′,4′- C2]spiropentyl alcohol
13
26 was oxidized to the corresponding acid (27), which was later
coupled to coenzyme A. The crude product was purified by
HPLC to give 23 in 29% overall yield. The monolabeled SPA-
CoA (24) was obtained by starting from methylenecyclopropyl
13
alcohol (28), which was treated with 1 equiv of [ C]CH2I2 and
equiv of trimethylaluminum to give 29 in excellent yield. The
2
9
energy. Furthermore, the lack of resonance stabilization of the
spiropentyl alcohol 29 was then oxidized with Jones reagent,
and the resultant acid was coupled to coenzyme A. Purification
by HPLC gave 24 in 60% overall yield.
ring-opened products (31-33) also renders the ring cleavage
thermodynamically less favored. Interestingly, our finding that
both (R)- and (S)-SPA-CoA are effective inhibitors for MCAD
clearly demonstrated that Câ bond cleavage leading to ring
fragmentation is nonstereospecific. Since in normal catalysis,
the Câ-H bond rupture step is pro-R-specific (eqs 1 and 2),4
the lack of stereospecificity found in this inactivation study
strongly suggests that the ring opening of SPA-CoA is likely a
spontaneous event, which is not directly enzyme-catalyzed.
Hence, analogous to the inactivation mechanism proposed for
MCPA-CoA, a one-electron oxidation of the anion intermediate
30 to give a transient R-spiropentyl radical 34 is an appealing
alternative (Scheme 6, route C).
The incubation was conducted anaerobically with MCAD
13
(
0.47 mmol) using 5 equiv of [ C]SPA-CoA in 50 mM
potassium phosphate buffer (pH 7.6) at 25 °C, and the progress
of inactivation was monitored at A450. After 1 h, most of the
unreacted SPA-CoA was removed by ultrafiltration. The
a,b
1
3
1
resulting mixture was concentrated and subjected to C- H-
detected HMQC (heteronuclear multiple quantum coherence)
experiments. The spectra obtained from these incubations all
consisted of an array of resonances within the range of 1-3.2
1
ppm in the H dimension with cross-peaks between 15 and 45
1
3
ppm in the C dimension (Figure 5). Evidently, more than
one 13C-enriched FAD adduct was generated in the incubation.
In view of the great ring strain associated with spiropentane
(63.5 kcal/mol), the spiropentylcarbinyl (SPC) radical (34),
once formed, should undergo ring opening more rapidly than
the cyclopropylcarbinyl (CPC) radical, which has a ring strain
of 27.6 kcal/mol and fragments at a rate of 10 s at 25 °C.
It has been shown that incorporation of the CPC moiety into a
1
13
20
Since the chemical shifts of these new H and C signals are
distinct from those of an intact spiropentane structure, ring
cleavage of SPA-CoA must have occurred during inactivation.
Furthermore, the absence of any signal downfield from 3.2 and
2
0
8
-1
21
1
13
4
5 ppm in the H and C dimensions, respectively, clearly
2
2
indicated that none of the covalent adducts contain an olefinic
carbon as part of the appended chain to FAD.
more strained system, and/or the addition of multiple methyl
(
20) Benson, S. W.; Cruickshank, F. R.; Golden, D. M.; Haugen, G. R.;
O’Neal, H. E.; Rodgers, A. S.; Shaw, R.; Walsh, R. Chem. ReV. 1969, 69,
79.
(21) (a) Griller, D.; Ingold, K. U. Acc. Chem. Res. 1980, 13, 317. (b)
Nonhebel, D. C. Chem. Soc. ReV. 1993, 347. (c) Newcomb, M. Tetrahedron
993, 49, 1151.
22) (a) Newcomb, M.; Manek, M. B.; Glenn, A. G. J. Am. Chem. Soc.
1991, 113, 949. (b) Bowry, V. W.; Lusztyk, J.; Ingold, K. U. J. Am. Chem.
Soc. 1991, 113, 5687. (c) Eaton, P. E.; Yip, Y. C. J. Am. Chem. Soc. 1991,
13, 7692. (d) Choi, S.-Y.; Eaton, P. E.; Newcomb, M.; Yip, Y. C. J. Am.
Mechanisms of Inactivation of MCAD and SCAD by SPA-
CoA. On the basis of the inactivation mechanism of MCAD
by MCPA-CoA and the experimental results presented herein,
likely pathways for irreversible inactivation of acyl-CoA de-
hydrogenase by SPA-CoA (16) can now be proposed. As shown
in Scheme 6, one possibility is that the reaction advances via
CR-H deprotonation, followed by an anion (30)-induced ring
2
1
(
1
(19) Russo, J. M.; Price, W. A. J. Org. Chem. 1993, 58, 3589.
Chem. Soc. 1992, 114, 6326.

2014 J. Am. Chem. Soc., Vol. 120, No. 9, 1998
Li et al.
Scheme 6
groups22b or phenyl groups23 to stabilize the product from a CPC
ring opening, could accelerate the radical rearrangement rate
constants by 1-3 orders of magnitude. In fact, a recent study
of the ring opening of (2-methylenecyclopropyl)carbinyl radical
inactivation. On the other hand, electron rebound from the
flavin semiquinone to reduce these inhibitor radicals prior to
their covalent coupling with the oxidized flavin coenzyme is
also feasible. It should be pointed out that there are at least
two common alkylation sites on FAD, the N5 and C4a on the
9
-1
(
°
analogue of 7), led to a rate constant of (3-4) × 10 s at 5
9
-1 24
C, which at 25 °C was estimated to be (6-7) × 10 s .
A
isoalloxazine ring, both of which are receptive to covalent bond
formation with the putative reactive intermediate(s).8
d,e,28
Thus,
proportionate rate acceleration for SPC radical is expected on
the premise that the ring strains of methylenecyclopropyl system
(
a blending of these possibilities furnishes a series of structural
candidates for the inhibitor-FAD adduct(s) (Scheme 7).29
Limited by the availability and affordability of the labeled
2
5
5) and spiropentyl skeleton (16) are at least comparable.
Studies by Newcomb et al. on secondary ethoxycarbonyl-
substituted radicals showed that these radicals fragment as fast
or faster than their alkyl radical counterparts due to transition
state polarization effects.26 Since 34 is a secondary ester-
substituted radical, its ring-opening rate may also be further
enhanced by transition-state polarization effects. Thus, the
nonstereospecificity of ring scission observed for SPA-CoA is
consistent with a radical mechanism that may be considered to
be spontaneous and consequently may bypass the chiral
discrimination normally imposed by the enzyme.
1
3
precursors, only two C-labeled SPA-CoA were synthesized
see Scheme 5), and the HMQC NMR spectrum of the modified
(
enzyme is shown in Figure 5. The complexity of the spectrum
indicates the presence of more than one species which are likely
formed by joining FAD and the ring-opened intermediates in
1
3
various combinations (Scheme 7). While the C peak at 42
ppm may be ascribed to a N-linked carbon shown in 38, an
unambiguous identification of this species and the assignment
of other signals in Figure 5 must await more spectral analysis
and comparison with appropriate model compounds. However,
the most valuable information revealed in Figure 5 is the lack
of any resonance in the downfield region. Since none of the
As depicted in Scheme 6, rupture of the spiropentyl system
(34) could generate either a tertiary cyclopropyl (35) or a
primary cyclopropylcarbinyl radical (36) (routes D and E,
respectively). Consecutive ring opening of 36 could afford an
acyclic radical intermediate 37 (route F).27 The one-electron-
reduced flavin semiquinone could be intercepted by any and/or
all of these transient radical species leading to irreversible,
covalent modification of the cofactor and ultimately enzyme
2
labeled carbons become sp hybridized in the adducts, secondary
ring opening appears to be unlikely. Thus, compounds 39 and
4
0 may be eliminated as possible structures.
(27) Cyclopropylcarbinyl radicals (such as 34) readily undergo â-scission
because the SOMO can assume an eclipsed conformation with respect to a
â,γ bond. On the contrary, cyclopropyl radicals (such as 35) cannot attain
such a conformation without the development of great strain, and therefore,
â-scission is hampered due to a much higher activation energy (Kennedy,
A. J.; Walton, J. C.; Ingold, K. U. J. Chem. Soc., Perkin Trans. 2 1982,
751).
(
23) (a) Castellino, A. G.; Bruice, T. C. J. Am. Chem. Soc. 1988, 110,
512. (b) Newcomb, M.; Manek, M. B. J. Am. Chem. Soc. 1990, 112, 9662.
c) Newcomb, M.; Johnson, C. C.; Manek, M. B.; Varick, T. R. J. Am.
Chem. Soc. 1992, 114, 10915.
24) Horner, J. H.; Johnson, C. C.; Lai, M.-t.; Liu, H.-w.; Martin-Esker,
A. A.; Newcomb. M.; Oh, E. Bioorg. Med. Chem. Lett. 1994, 4, 2693.
25) The ring strain of methylenecyclopropane has been estimated to be
7
(
(28) (a) Olson, S. T.; Massey, V.; Ghisla, S.; Whitfield, C. D.
Biochemistry 1979, 18, 4724. (b) Ghisla, S.; Olson, S. T.; Massey, V.;
Lhoste, J. M. Biochemistry 1979, 18, 4733. (c) Ghisla, S.; Ogata, H.; Massey,
V.; Schonbrunn, A.; Abeles, R. H.; Walsh, C. T. Biochemistry 1976, 15,
1791. (d) Schonbrunn, A.; Abeles, R. H.; Walsh, C. T.; Ghisla, S.; Ogata,
H.; Massey, V. Biochemistry 1976, 15, 1798. (e) Zeller, H. D.; Ghisla, S.
In FlaVins and FlaVoproteins; Edmondson, D. E., McCormick, D. B., Eds.;
Walter de Gruyter: Berlin, 1987; p 161.
(29) The structures listed in Scheme 7 for the inhibitor-enzyme adducts
are by no means a complete set of all possible variations. For example, a
fully oxidized C-6-substituted flavin has also been proposed as the structure
of a minor component isolated from the incubation of MCPA-CoA (5) with
MCAD (Zeller, H. D.; Ghisla, S. In FlaVins and FlaVoproteins; Edmondson,
D. E., McCormick, D. B., Eds.; Walter de Gruyter: Berlin, 1987; p 161).
(
(
2
0
4
0.8 kcal/mol. It should be noted that the apparently greater ring strain
associated with the spiropentyl system is misleading since it contains one
more cyclopropyl ring than the methylenecyclopropane case. Interestingly,
recent calculations showed that the major source of the “strain” resulting
from the introduction of a trigonal center into cyclopropane is not the
increase in angle strain but the loss of a very strong cyclopropane C-H
bond (Johnson, W. T. G.; Borden, W. T. J. Am. Chem. Soc. 1997, 119,
5
930).
(26) Newcomb, M.; Horner, J. H.; Emanuel, C. J. J. Am. Chem. Soc.
1
997, 119, 7147.

Spiropentylacetyl-CoA
J. Am. Chem. Soc., Vol. 120, No. 9, 1998 2015
Scheme 7
inhibitor and the (S)-SPA-CoA (20) was only a competitive
inhibitor of SCAD. Recently, cyclobutylacetyl-CoA (41, CBA-
CoA) has been shown to be a substrate for MCAD, but an
inhibitor for SCAD by forming a covalent adduct with its flavin
coenzyme. Another related example is (E)-2,3-epoxybutyryl-
CoA (42), which is an irreversible inactivator for SCAD but a
31
competitive inhibitor for MCAD.15 The selectivity exhibited
by these compounds on inhibition against two closely related
dehydrogenases is likely a consequence of the distinct steric
and electronic demands imposed by the active sites of MCAD
and SCAD. Further investigation may allow us to better define
the specific substrate/inhibitor binding characteristics of these
enzymes. Such information is important for the design of novel
class-selective inhibitors to control and/or regulate fatty acid
metabolism.
The inference that no secondary ring cleavage occurs is in
agreement with a radical mechanism, since the rate of the ring
cleavage of 36 is expected to be significantly retarded comparing
to that of the first ring opening due to the decrease of ring strain.
This prediction is also consistent with the slow ring-opening
Experimental Section
rate observed by Bowry et al. on the â-(methylcyclopropyl)-
General. GC-MS analysis was performed on an HP 5890A gas-
liquid chromatograph and a VG 7070E-HF spectrometer. Ultraviolet-
visible spectroscopy was recorded on a Shimadzu UV-160, or a Hewlett-
Packard 8452A spectrophotometer. Radioactivity was measured by
liquid scintillation counting on a Beckman LS 3801 counter using
Ecoscint A biodegradable scintillation solution from National Diag-
nostics (Manville, NJ). Analytical GC was performed on an HP 5890
instrument, using a cross-linked phenylmethyl silicone gum fused silica
capillary column (0.53 mm × 30 m). High-performance liquid
chromatography (HPLC) analysis and/or purification were conducted
with either a Hewlett-Packard 1090A instrument equipped with an
carbinyl radical, which is a related mimic of 36.2
2b
More
importantly, a rapid electron rebound from flavin semiquinone
may result in quenching of radical 36 to generate the corre-
sponding anion 31 and thus prevent further ring cleavage.
Summary and Conclusion
In summary, we have extended our inquiry to understand the
mechanistic intricacies of an important class of enzymes, acyl-
CoA dehydrogenases, by conducting detailed mechanistic
investigations with an unnatural inhibitor, SPA-CoA (16). In
contrast to an early report, this compound was found to be a
mechanism-based inactivator for pig kidney MCAD and a
recombinant SCAD originally derived from M. elsdenii. Using
tritium-labeled SPA-CoA (18) in the incubation revealed the
formation of an irreversible and covalent linkage between SPA-
CoA and the FAD cofactor of MCAD and SCAD. The tritium
washout results provided direct evidence that CR proton abstrac-
tion is a requisite step of inactivation. On the basis of the
1
13
HP3392 integrator or a Beckman 110B instrument. H NMR and
C
NMR spectra were recorded on an IBM NR/200 or NR/300 or Varian
U-300 or U-500 spectrometer. Chemical shifts are reported on the δ
scale relative to the internal standard (tetramethylsilane or appropriate
solvent peaks) with coupling constants given in hertz. NMR assign-
ments labeled with an asterisk (*) may be interchangeable. Flash
column chromatography was performed on columns of various
diameters with J. T. Baker (230-400 mesh) silica gel by elution with
the solvents reported. Analytical thin-layer chromatography (TLC) was
carried out on Merck silica gel 60 G-254 plates (25 mm). TLC spots
were visualized either with UV light or by dipping the plates into the
HMQC analysis with 1 C-labeled SPA-CoA samples (23 and
3
2
4), the single-ring-opened radicals (36 and perhaps 35) are
4 2 4
staining solutions of KMnO (1%) or vanillin/ethanol/H SO (1:98:1)
postulated as the reactive intermediates responsible for FAD
adducts (Schemes 6 and 7).30 SPA-CoA in enantiomerically
pure forms (19 and 20) were synthesized, and both isomers were
shown to be competent inactivators for MCAD with nearly
identical partition ratios (kcat/kinact ≈ 2). Clearly, the inactivation
is nonstereospecific with respect to the ring cleavage. Since
the Câ-H cleavage of the trans dehydrogenation step in normal
catalysis is pro-R specific, the lack of stereospecificity of Câ-C
bond rupture for SPA-CoA strongly suggests that this ring-
opening step leading to enzyme inactivation is likely a spon-
taneous event induced by a spiropentylcarbinyl radical (34).
These results provided further evidence supporting that MCAD
is capable of mediating one-electron redox chemistry. Similar
studies were also performed with SCAD. Interestingly, the (R)-
SPA-CoA (19) was found to be a mechanism-based irreversible
or phosphomolybdic acid (7% EtOH solution) and then heating. The
drying reagent used after routine workup was anhydrous magnesium
sulfate. Solvents, unless otherwise specified, were of analytical reagent
grade or the highest quality commercially available. For anhydrous
reactions, the solvents were pretreated prior to distillation as follows:
tetrahydrofuran (THF) was dried over sodium and benzophenone;
methylene chloride, dimethyl sulfoxide (DMSO), and dimethylform-
amide (DMF) were dried over calcium hydride; pyridine and triethyl-
amine were dried over KOH. It should be noted that the tritium-labeled
compounds reported in this paper were not submitted for NMR and/or
mass spectroscopic analysis due to the possibility of radioactive
contamination. However, satisfactory analytical results were obtained
with unlabeled and/or deuterium-labeled analogues.
Enzymes. The short-chain acyl-CoA dehydrogenase (SCAD) was
purified from an Escherichia coli strain BL21(DE3)pLys which carries
3f
and overexpresses the M. elsdenii SCAD gene (pWTSCADT7-7). The
medium-chain acyl-CoA dehydrogenase (MCAD), from pig kidney,
was purified according to the procedure of Thorpe.12 Concentration
of the holoenzyme was determined spectrophotometrically on the basis
(
30) Ab initio calculations were carried out to compute the energies of
primary cyclopropylcarbinyl and tertiary cyclopropyl radicals using an
unrestricted Hartree-Fock level of theory with 6-31G* basis set (UHF/6-
/G*). The results suggested that the energy difference is about 5 kcal/
mol, favoring the primary cyclopropylcarbinyl radical. These energy
differences reflect the bond dissociation energies (BDEs) of primary alkyl
C-H versus the tertiary cyclopropyl C-H bonds (Borden, W. T. Personal
communication).
-
1
-1
3
of a molar absorptivity of 13.9 mM cm at 446 nm for oxidized
-
1
-1
SCAD and 15.4 mM cm at 446 nm for oxidized MCAD.
(31) Shin, I.; Li, D.; Becker, D. F.; Stankovich, M. T.; Liu, H.-w. J.
Am. Chem. Soc. 1994, 116, 8843.

2016 J. Am. Chem. Soc., Vol. 120, No. 9, 1998
Li et al.
transferred into a NMR tube, and subjected to 13C- H-detected HMQC
1
Enzyme Assay. The enzyme activity was determined by a published
procedure using phenazine methosulfate (PMS) or phenazine ethosulfate
PES) as an external electron carrier to mediate the transfer of reducing
equivalents from substrate (octanoyl-CoA for MCAD and butyryl-CoA
analysis on a Varian 500 spectrometer.
(
2-[2-(1-Chloro-1-methylcycloprop-2-yl)ethoxy]tetrahydropyran
(11). To a chilled solution (0 °C) of 3,4-dihydro-2H-pyran (23.5 g,
279 mmol) in anhydrous ether (150 mL) were added p-toluenesulfonic
acid (0.1 g) and 3-buten-1-ol (10; 5 g, 69 mmol). The resulting mixture
was stirred at room temperature for 5 h and was then quenched by the
addition of concentrated ammonium hydroxide (1 mL) and methanol
3
f,12
for SCAD) to 2,6-dichlorophenolindophenol (DCPIP).
Determination of Partition Ratio. A series of sample solutions
each containing 13.0 mM MCAD (or 9.0 mM SCAD) and an
appropriate amount of 19 or 20 were prepared in 400 mL of 50 mM
(
10 mL). The solvent was evaporated in vacuo, and ether was added
potassium phosphate buffer, pH 7.6, to give [I]
00. These samples were stirred gently at 25 °C for 1 h. Subsequently,
0 0
/[E] ratios of 1.0-
to the residue. The precipitated ammonium p-toluenesulfate was
filtered, the filtrate was concentrated, and the crude product was purified
by flash column chromatography (0-5% ethyl acetate in hexanes) to
1
an aliquot of 50 mL was taken from each sample and was added to the
assay cocktail. The activity was measured immediately. The control
experiments were carried out in the same manner without an inactivator.
The partition ratio was deduced by plotting the residual activity versus
give 2-(3-butenyloxy)tetrahydropyran in nearly quantitative yield (10.8
1
g): H NMR (CDCl
3
) δ 5.83 (1H, m, dCH-C), 5.08 (1H, dd, J )
), 5.02 (1H, m, dCH ), 4.59 (1H, m, 2-H of THP),
.79 (2H, m, CH O)*, 3.46 (2H, m, CH O of THP)*, 2.34 (2H, m,
CHdCH ) δ
), 1.80-1.50 (6H, m, Hs of THP); 13C NMR (CDCl
35.3 (dCH), 116.3 (CH d), 98.8 (C-2 of THP), 66.8 (side chain
CH O), 62.3 (CH O of THP), 34.2 (CH CHdCH ), 30.7, 25.5, 19.6
Cs of THP).
1
7.2, 1.8, dCH
2
2
[
I]
0
/[E]
0
.
3
2
2
Kinetic Analysis of Inactivation. In a typical inactivation experi-
CH
2
2
3
ment, an appropriate amount of inactivator (16, 19, or 20) was added
to the enzyme solution (MCAD, 13.0 mM; SCAD, 9.0 mM) in 400
mL of 50 mM potassium phosphate buffer, pH 7.6, at 25 °C. At various
time intervals, aliquots (50 mL) were removed and diluted with assay
cocktail (0.7 mL) and the residual enzyme activity was determined.
For the “protection” experiment, the enzyme (SCAD 14.4 mM; MCAD
1
2
2
2
2
2
(
A total of 10.8 g (69 mmol) of 2-(3-butenyloxy)tetrahydropyran and
.4 g (85 mmol) of 1,1-dichloroethane were mixed in 50 mL of
anhydrous ether. The solution was cooled to between -30 and -40
C and maintained at that temperature throughout the 3.5-h addition
of 48 mL of 1.6 M n-butyllithium in hexane. After the addition was
complete, the cooling bath was removed and the solution was stirred
for 12 h at 20 °C. The reaction was quenched by the addition of 20
mL of water, and all the salts present in the flask were dissolved within
8
1
3.0 mM) was co-incubated with MCPA-CoA (5, 27.0 mM) and
°
compound 16 (46.6 mM) at 25 °C. The rate of flavin chromophore
bleaching was monitored at 450 nm. Meanwhile, the residual enzyme
activity was also determined over time as described above.
Tritium Washout Study. The tritium-labeled inactivator (18, 20
nmol, 0.32 mCi/mmol) was incubated with enzyme (SCAD or MCAD,
0.0 nmol) in 500 mL of 50 mM potassium phosphate buffer (pH 7.6)
1
5-20 min. The organic layer was collected and concentrated. The
residual light yellow oil was purified by flash column chromatography
4
at 25 °C. A 100-mL aliquot of the reaction mixture was removed at
time zero and counted for radioactivity. After a period of 60 min, the
incubation was quenched by addition of activated charcoal (10%
solution), and the resulting suspension was mixed vigorously on a vortex
mixer for 1 min. The suspension was centrifuged, and the supernatant
was removed and analyzed for radioactivity by a scintillation counter.
These readings were calibrated against controls prepared in parallel
with boiled enzyme (100 °C, 10 min).
1
(
5% ethyl acetate in hexanes) to give 11 in 61% yield (9.2 g):
NMR (CDCl ) δ 4.60 (1H, m, 2-H of THP), 3.84 (2H, m, CH O), 3.50
O of THP), 1.92-1.53 (10H, m), 1.59 (3H, s, CH ), 0.92
1H, m); 13C NMR (CDCl
) δ 99.1 (C-2 of THP), 67.1 (CH O), 62.4
CH O of THP), 45.6, 31.2, 30.0, 25.7, 23.0, 22.9, 22.0, 19.8.
-[2-(Methylenecycloprop-2-yl)ethoxy]tetrahydropyran (12). Com-
H
3
2
(
(
(
2H, m, CH
2
3
3
2
2
2
pound 11 (9.2 g, 42 mmol) was added dropwise over a period of 30
min to an anhydrous dimethyl sulfoxide solution (15 mL) containing
Determination of Reversibility of Inactivation and Stoichiometry
of Tritium Incorporation. An appropriate amount of enzyme (SCAD
or MCAD) was incubated with SPA-CoA (16) and/or the labeled SPA-
CoA (18) (10 equiv) for an extended period until no further inactivation
ensued. The incubation mixture was dialyzed against 50 mM potassium
phosphate buffer, pH 7.6, at 4 °C for 2 days with 10 buffer changes.
The residual enzyme activity of each of the incubation mixtures was
determined before and after dialysis. Further, aliquots (50 mL) of the
5
.1 g (46 mmol) of potassium tert-butoxide. After stirring at 70 °C
for 7 h, the reaction was cooled slowly to ambient temperature and
then poured into ice water. The aqueous solution was extracted with
ether (3 × 50 mL), and the pooled ether extracts were washed with
brine, dried, filtered, and concentrated in vacuo. The crude residue
was purified by flash column chromatography (5% ethyl acetate in
hexanes) to give 12 in 83% yield (6.3 g): 1H NMR (CDCl
5
3
m), 1.24 (1H, m), 0.78 (1H, m); C NMR (CDCl
9
1
3
) δ 5.43,
), 4.60 (1H, t, J ) 3.2, 2-H of THP),
O), 3.48 (2H, m, CH O of THP), 1.84-1.51 (9H,
) δ 103.6 (dCH ),
O of THP), 30.2, 25.0,
.41 (1H each, d, J ) 1.2, dCH
.80 (2H, m, CH
2
“
labeled-incubation” mix (with 18) were removed at various time
2
2
intervals and submitted for scintillation counting. The readings were
calibrated against a control prepared in parallel with boiled enzyme.
In addition, a portion of the inactivated enzyme, devoid of unbound
inactivator 18, was treated with 90% MeOH to denature the protein.
After centrifugation, the radioactivity in the supernatant and in the
precipitate was determined by a scintillation counter.
13
3
2
7.8 (C-2 of THP), 69.5 (CH
9.1, 15.0, 8.3.
2 2
O), 61.7 (CH
(Methylenecyclopropyl)acetic Acid (13). Compound 12 (6.3 g,
3
4.6 mmol) and p-toluenesulfonic acid monohydrate (1.9 g, 0.01 mol)
were dissolved in 200 mL of methanol. This mixture was stirred at
room temperature overnight. Methanol was removed in vacuo, and
the residue was diluted with water. The aqueous layer was extracted
with ether, and the combined ether layers were dried, filtered, and
Treatment of Reduced Flavin Adducts with Ferricenium Hexa-
fluorophosphate (FcPF
according to a literature procedure. In a typical experiment, enzyme
MCAD, 5 nmol) was incubated aerobically with SPA-CoA (16, 25
nmol) until no further inactivation (monitored by FAD bleaching and
enzyme activity) occurred. At which time, FcPF (15 equiv) was added
6 6
). The ferricenium salt (FcPF ) was prepared
7
d
(
concentrated to give a colorless oil in 60% yield (1 g):
(CDCl ) δ 5.42 (1H, s, dCH ), 5.37 (1H, s, dCH ), 3.72 (2H, t, J )
1
H NMR
3
2
2
6
6.3, CH OH), 1.68-1.46 (3H, m, CH CH OH and 1-H), 1.26 (1H, m,
2
2
2
in three portions; the reoxidation of the flavin chromophore was
monitored at 450 nm and the remaining enzyme activity before and
after ferricenium treatment was also determined. A control experiment
was run in parallel in which the SPA-CoA was substituted by the natural
substrate, octanoyl-CoA (10 nmol), and/or MCPA-CoA (5, 25 nmol).
3-H), 0.78 (1H, m, 3-H); 13C NMR (CDCl ) δ 136.1 (dCH ), 103.0
3
2
(CH dC), 62.5 (CH OH), 36.0 (CH CH OH), 12.6 (C-1), 9.2 (C-3).
2
2
2
2
(
Methylenecyclopropyl)ethanol (1 g, 10 mmol) obtained above was
dissolved in 30 mL of acetone. Jones reagent was added dropwise at
°C until the color of the reaction mixture remained red brown. The
0
Two-Dimensional 1H{ C} HMQC Analysis. The 13C-labeled
inhibitors 23 and 24 (5 mol equiv) were separately incubated with
MCAD (20 mg, 0.47 mmol) in 50 mM potassium phosphate buffer
13
resulting solution was stirred for 1 h. The excess oxidizing reagent
was quenched with 2-propanol. The reaction solution was then diluted
with water followed by repeated extraction with ether. The combined
ether layers were extracted with 10% NaOH solution (3 × 50 mL),
and the combined aqueous phases were acidified to pH < 2 with 6 N
HCl. The resulting solution was re-extracted with three 50-mL portions
of ether, and the pooled organic phases were dried and concentrated in
(pH 7.6) at room temperature. The reaction was incubated for 1 h to
ensure complete inactivation and then ultrafiltered through an Amicon
YM-10 filter to remove unreacted SPA-CoA and any possible turnover
products. The resulting mixture was concentrated to 500 mL,

Spiropentylacetyl-CoA
J. Am. Chem. Soc., Vol. 120, No. 9, 1998 2017
vacuo to give 13 as a colorless oil in 70% yield (0.8 g): 1H NMR
g, 140 mmol), diisopropylamine (29 g, 286 mmol), paraformaldehyde
(11 g, 360 mmol), and 275 mL of dry THF. The mixture was stirred
vigorously as the copper(I) iodide (14 g, 73 mmol) was added in small
portions. The reaction mixture was heated at a gentle reflux with
stirring overnight. The resulting brown suspension was cooled to room
temperature, filtered through a Celite plug and concentrated under
reduced pressure. The dark brown residue was diluted with water and
acidified with 3 N HCl. The aqueous layer was extracted with ether,
and the combined ethereal layers were washed with water and brine,
dried, filtered, and concentrated in vacuo to give a yellow-orange oil.
The crude product was purified by flash chromatography (5% ethyl
(
1
(
1
CDCl
.70 (1H, m, 1-H), 1.37 (1H, m, 3-H), 0.88 (1H, m, 3-H); 13C NMR
CDCl ) δ 179.1 (CdO), 134.1 (C-2), 104.4 (dCH ), 37.7 (CH CO H),
0.9 (C-1), 9.5 (C-3).
Methyl Spiropentylacetate (14). To a solution of 13 (200 mg, 1.8
3
) δ 5.50, 5.41 (1H each, s, dCH
2
), 2.37 (2H, d, J ) 7.2, R-H),
3
2
2
2
mmol) in ether (10 mL) was added a solution of diazomethane (prepared
from 2.5 g of diazald in 25 mL of ether) dropwise at 0 °C. The reaction
was stirred for 10 min, followed by the addition of an equal amount of
diazomethane in ether and 20 mg of Pd(OAc) . After the mixture was
2
stirred for an additional 2 h, the resulting suspension was filtered
through a Celite plug, concentrated, and purified by flash chromatog-
ether in pentane) to give the title compound 25 in 56% yield (6.7 g):
1
raphy (2% ethyl ether in pentane) to give product 14 in 72% yield
H NMR (CDCl
3
) δ 5.09 (1H, m, CH
2
dCdCH-), 4.68 (2H, m,
OH), 2.23 (2H, m,
) δ 86.2
OH), 31.3
1
(
181 mg): H NMR (CDCl
3
) δ 3.66 (3H, s, OCH
3
), 2.38, 2.25 (1H
CH
CH
2
dCdCH-), 3.45 (2H, td, J ) 7.0, 7.0, CH
2
13
each, dd, J ) 16.1, 6,8, side chain CH
2
), 1.38 (1H, m), 0.87 (1H, m),
2
CH
dCdCH-), 74.9 (CH
CH OH).
2
OH), 1.66 (1H, br s, OH); C NMR (CDCl
3
3
0
1
.79-0.72 (3H, m), 0.64 (1H, m), 0.55 (1H, m); 1 C NMR (CDCl
73.8 (CdO), 51.5 (OCH ), 37.5 (CH CO), 14.5, 13.3, 12.1, 6.1, 3.5.
Spiropentylacetic Acid (15). To a solution of methyl spiropentyl-
3
) δ
(CH
(CH
[2′,4′- C
2 2
25 (48 mg, 0.57 mmol) and [ C]CH I (0.4 g, 1.43 mmol) in 5 mL of
2
2
dCdCH-), 61.6 (CH
2
3
2
2
2
13
2
](2-Hydroxyethyl)spiropentane (26). To a solution of
13
acetate (180 mg, 1.3 mmol) in a mixture of 20 mL of ethanol and
water (v/v ) 4:1) was added sodium hydroxide (52 mg, 1.3 mmol).
After refluxing for 10 min, the solution was acidified with 2 N HCl.
The mixture was then extracted with methylene chloride, and the
combined organic extracts were dried over magnesium sulfate. Sub-
methylene chloride was added dropwise trimethylaluminum (1.45 mL,
2.85 mmol, 3.0 M in hexanes) over 10 min at 0 °C under nitrogen.
The first equivalent should be added very slowly due to the exothermic
formation of aluminum alkoxide. After the clear reaction mixture was
stirred at room temperature for 16 h, it was cooled to 0 °C. The reaction
was quenched with 5 mL of 1 N NaOH, and the resulting mixture was
stirred vigorously for an additional 1 h. The organic layer was
separated, and the aqueous layer was extracted with methylene chloride
(2 × 5 mL). The combined organic layers were washed with water,
dried, filtered, and concentrated in vacuo. The crude product was
purified by flash column chromatography (10:1 to 4:1 hexanes/ethyl
sequent filtration and concentration gave spiropentylacetic acid in 91%
1
yield (149 mg): H NMR (CDCl
2
3
) δ 10.62 (1H, br s, COOH), 2.45,
CO), 1.38 (1H, m), 1.05 (1H,
.30 (1H each, dd, J ) 16.4, 6.8, CH
dd, J ) 7.7, 4.4), 0.87-0.66 (4H, m), 0.58 (1H, t, J ) 4.4); C NMR
CDCl ) δ 180.2 (CdO), 37.9 (CH CO), 14.8, 13.3, 12.3, 6.4, 3.8.
Spiropentylacetyl-CoA (16). Compound 16 was prepared from 15
in 76% yield (43 mg) according to a well-documented mixed anhydride
2
13
(
3
2
1
2
1
procedure: H NMR ( H
.22 (1H, d, J ) 6.1, ribose anomeric H), 4.91 (1H, buried under HOH
peak), 4.66, 4.31 (1H each, s, ribose H's), 4.32 (2H, s, ribose CH O),
.09 (1H, s), 3.91 (1H, dd, J ) 9.7, 4.7), 3.62 (1H, dd, J ) 9.7, 4.6),
.51 (2H, t, J ) 6.2), 3.38 (2H, t, J ) 6.2), 3.03 (2H, t, J ) 6.2), 2.71,
CO of SPA), 2.50 (1H, dd, J ) 16.4, 7.6,
CO of SPA), 2.49 (2H, t, J ) 6.4), 1.37 (1H, m), 1.05 (1H, dd, J
2
O) δ 8.59, 8.29 (1H each, s, adenine H’s),
acetate) to give the desired product 26 in 85% yield (54 mg): H NMR
2
6
(CDCl
3
) δ 3.67 (2H, t, J ) 6.9, CH
2
OH), 1.75-0.63 (7H, m), 1.10
2
(1H, m), 0.47 (1H, m). Since the product is a mixture of diastereomers,
13
4
3
C NMR of 26 displays three signals at δ 12.05, 6.12, and 3.67 in a
13
ratio of about 2:1:1. C NMR (CDCl
62.9 (CH OH), 35.8 (CH CH
[2′,4′-13C ]Spiropentylacetic Acid (27). Compound 26 (54 mg, 0.48
3
) of the unlabeled species: δ
2
OH), 22.3, 14.1, 12.1, 6.1, 3.7.
(
1H, dd, J ) 16.4, 6.5, CH
2
2
2
CH
2
2
)
(
7.7, 4.2), 0.96, 0.82 (3H each, s, Me’s), 0.92-0.80 (2H, m), 0.76
mmol) was oxidized by Jones reagent (see the synthesis of 13 for
procedure) to give the corresponding acid 27 in 71% yield (0.43 mg):
1H, m), 0.70 (1H, m), 0.62 (1H, t, J ) 4.2); high-resolution FAB-MS
+
13
calcd for C28
H
45
3
N
O
7 17
3
P S (M + 1) 876.1806, found 876.1844.
C NMR (CDCl ) δ 12.1, 6.13, 3.50 (mixture of diastereomers).
3
Methyl [r- H]Spiropentylacetate (17). To a chilled (-78 °C)
lithium diisopropylamide solution (3.2 mmol, prepared from 1.3 mL
of n-butyllithium (2.5 M in hexane) and 0.32 g of diisopropylamine in
THF) was added dropwise a solution of methyl spiropentylacetate (14,
13
[
2′,4′- C
2
]Spiropentylacetyl-CoA (23). Acid 27 (20 mg, 0.16
mmol) was coupled with coenzyme A (50 mg, 65.1 mmol) to give the
13
final product 23 in 85% yield (48 mg): C NMR (D
2
O) δ 13.9, 7.71,
5
.43 (mixture of diastereomers); high-resolution FAB-MS calcd for
3
06 mg, 2.14 mmol in 50 mL of THF). The resulting mixture was
13
+
C
2
C
26
H
45
N
7
17 3
O P S (M + H) 878.1873, found 878.1858.
gradually warmed to 0 °C and stirred for 2 h. The reaction was cooled
13
[4′- C](2-Hydroxyethyl)spiropentane (29). Synthesis of com-
3
2
again to -78 °C and slowly mixed with 20 mL of H O (S.A. 1.8
pound 29 was achieved by a procedure similar to that used to make
mCi/mmol). The reaction mixture was stirred at 0 °C for 1 h and
quenched with ice-cold saturated ammonium chloride. After being
stirred for another 10 min, the mixture was extracted with ether and
the combined organic extracts were dried, filtered, and concentrated
in vacuo. The crude product was purified by flash chromatography
compound 26. In this experiment, 2.1 equiv of trimethylaluminum (0.54
13
mL, 2.0 M solution in hexanes) and 1.1 equiv of [ C]CH
2
I
2
(0.15 g,
0
5
1
.56 mmol) were used for the cyclopropanation of alcohol 28 (50 mg,
1 mmol, an intermediate generated during the synthesis of 13 from
13
2). The desired product was obtained in 82% yield (47 mg):
) δ 6.17, 3.67 (mixture of diastereomers).
4′- C]Spiropentylacetyl-CoA (24). Compound 24 was prepared
in 73% combined yield (44 mg) from the corresponding acid which in
C
(
2% ethyl ether in pentane) to afford product 17 in 95% yield (0.28 g).
NMR (CDCl
3
2
2
In a separate experiment, the reaction was quenched with H O and
13
[
the extent of isotope incorporation at R-C was estimated to be >90%
on the basis of NMR integration.
13
2
turn was derived from 29. C NMR (D O) of the intermediate acid:
3
[
r- H]Spiropentylacetyl-CoA (18). By adapting a similar procedure
13
δ 6.14, 3.50. C NMR of 24: δ 5.19, 2.89 (mixture of diastereomers).
used for the synthesis of spiropentylacetic acid (15) from methyl
spiropentylacetate (14), the labeled analogue was obtained from methyl
FAB-MS calcd for 13
C C
1
H
N
O
P
S (M + H) : 877.2. Found: 877.2
+
27
45
7
17
3
3
3
[
R- H]spiropentylacetate (17) in 76% (0.13 g). The [R- H]spiropen-
Acknowledgment. We thank Dr. Vikram Roongta for
providing technical assistance in obtaining the HMQC spectra
and Professor Marian Stankovich for the generous gift of E.
coli (pWTSCADT7-7)-BL21(DE3)pLys strain. We owe a special
debt of gratitude to Professor Weston Borden for sharing with
us his computation results. This work was supported by a
National Institutes of Health grant (GM40541). I.S. is a recipient
of the Stanwood Johnston Memorial Fellowship from the
University of Minnesota.
tylacetic acid was later coupled with coenzyme A to give the title
compound in 72% yield (41 mg), with a specific activity of 0.32 mCi/
mmol.
(R)- and (S)-Spiropentylacetyl-CoA (19 and 20). (R)- and (S)-
Spiropentylacetyl-CoA (19 and 20) were obtained from the correspond-
ing (methylenecyclopropyl)acetic acids (21 and 22, respectively) by
following the procedure used for the synthesis of spiropentylacetyl-
CoA (16) from (methylenecyclopropyl)acetic acid (13). Preparations
1
7
of 21 and 22 have already been reported.
,4-Pentadien-1-ol (25). To a round-bottom flask equipped with a
condenser fitted with a CaCl drying tube were added 3-butyn-1-ol (10
3
2
JA9737634