A stereochemical probe of the tetrahedral intermediate in the reactions of acetyl-coenzyme A dependent acetyltransferases

Article abstract of DOI:10.1021/ja9616241

A pair of isomeric acetyl-coenzyme A (acetyl-CoA) analogs have been prepared in which the thioester group is replaced with a secondary alcohol. The two isomers differ in stereoconfiguration of the secondary alcohol and were designed to mimic the two possi

Full text of DOI:10.1021/ja9616241

9826
J. Am. Chem. Soc. 1996, 118, 9826-9830
A Stereochemical Probe of the Tetrahedral Intermediate in the
Reactions of Acetyl-Coenzyme A Dependent Acetyltransferases
Benjamin Schwartz and Dale G. Drueckhammer*
Contribution from the Department of Chemistry, Stanford UniVersity, Stanford, California 94305
ReceiVed May 15, 1996X
Abstract: A pair of isomeric acetyl-coenzyme A (acetyl-CoA) analogs have been prepared in which the thioester
group is replaced with a secondary alcohol. The two isomers differ in stereoconfiguration of the secondary alcohol
and were designed to mimic the two possible configurations of the tetrahedral intermediate or transition state in the
reactions of acetyl-CoA dependent acetyltransferases. These two isomers were tested as inhibitors of chloramphenicol
acetyltransferase and carnitine acetyltransferase, both of which have previously been predicted to form a tetrahedral
intermediate which matches the configuration of the (S)-alcohol. The (S)-isomer was the more potent inhibitor of
both enzymes, with Ki values 12-fold and 6-fold lower than the Ki values for the (R)-isomer. The (S)-isomer was
also the more potent inhibitor of phosphate acetyltransferase, acetyl-CoA synthetase, and arylamine acetyltransferase,
for which the stereochemistry of the tetrahedral intermediate was previously unknown. These results suggest a
common stereoconfiguration of the tetrahedral intermediate among acetyl-CoA dependent acetyltransferases.
The interpretation of stereochemical details of enzyme-
Scheme 1
catalyzed reactions has been useful in addressing issues of
mechanism, function, and evolution in enzyme catalysis.1
Stereochemical features of many enzyme-catalyzed reactions
can be deduced from the stereochemistry of the product, often
with the use of a substrate containing one or more stereospe-
cifically incorporated isotopic labels. Enzymes which catalyze
acyl transfer reactions are presumed to proceed via a tetrahedral
intermediate or transition state which has one of two possible
stereoconfigurations, depending on the face of attack of the acyl
group acceptor on the acyl donor.2 These include a number of
acetyl-coenzyme A (acetyl-CoA) dependent acetyltransferases,
which catalyze the transfer of the acetyl group from acetyl-
CoA 1 to an oxygen, nitrogen, or carbon nucleophile (Scheme
1).3-9 The configuration of the tetrahedral intermediate 3a or
3b cannot be deduced from product analysis, as the product 5
like the substrate 1 has an achiral sp2 carbon at the site of
reaction.
modeling analysis.10 The configuration of the tetrahedral
intermediate in the reaction of chloramphenicol acetyltransferase
has also been predicted to be that indicated by 3a, by computer
modeling of the reaction based on the known crystal structure
of the enzyme.11 Extensive work with carnitine acetyltransferase
has demonstrated the value of knowledge of the stereochemistry
of the tetrahedral intermediate in inhibitor design, and these
studies have provided further support for the predicted stereo-
chemistry of this intermediate.12 Similar observations have been
made in the design of hydroxyethylene isosteres as inhibitors
of proteases, with the configuration at a chiral secondary alcohol
center corresponding to the site of hydrolysis being important
for optimum inhibition.13,14 As part of our program in the
synthesis15 and applications16 of CoA analogs as mechanistic
tools in enzymology, we have designed a pair of secondary
alcohol analogs of acetyl-CoA 6 as stereochemical probes of
The two most well studied of the acetyl-CoA dependent
acetyltransferases are carnitine acetyltransferase3 and chloram-
phenicol acetyltransferase,4 which catalyze acetyl transfer to a
hydroxyl group of an alcohol. The tetrahedral intermediate in
the reaction of carnitine acetyltransferase has been predicted to
be that represented by 3a in Scheme 1, based on molecular
X Abstract published in AdVance ACS Abstracts, October 1, 1996.
(1) Benner, S.; Ellington, A. D. CRC Crit. ReV. Biochem. 1988, 23, 369.
(2) For brevity, the tetrahedral species is referred to as an intermediate
hereafter, though it is unclear if it is indeed an intermediate or a transition
state. Reactions of some of the acetyltransferases are believed to involve
an acyl enzyme, the formation of which would also proceed via a tetrahedral
intermediate such as 3.5,9
(10) Gandour, R. D.; Colucci, W. J.; Fronczek, F. R. Bioorg. Chem. 1985,
13, 197.
(3) Colucci, W. J.; Gandour, R. D. Bioorg. Chem. 1988, 16, 307.
(4) Shaw, W. V.; Leslie, A. G. W. Annu. ReV. Biophys. Biophys. Chem.
1991, 20, 363.
(5) Watanabe, M.; Sofuni, T.; Nohmi, T. J. Biol. Chem. 1992, 267, 8429.
(6) Cullis, P. M.; Wolfenden, R.; Cousens, L. S.; Alberts, B. M. J. Biol.
Chem. 1982, 257, 12165.
(11) Lewendon, A.; Shaw, W. V. J. Biol. Chem. 1993, 268, 20997.
(12) Sun, G.; Savle, P. S.; Gandour, R. D.; Nica Bhaird, N.; Ramsay, R.
R.; Fronczek, F. R. J. Org. Chem. 1995, 60, 6688.
(13) Rao, B. G.; Tilton, R. F.; Singh, U. C. J. Am. Chem. Soc. 1992,
114, 4447.
(14) Rich, D. H.; Sun, C.-Q.; Prasad, J. V. N. V.; Pathiasseril, A.; Toth,
M. V.; Marshall, G. R.; Clare, M.; Mueller, R. A.; Houseman, K. J. Med.
Chem. 1991, 34, 1222.
(7) Wong, L.-J. C.; Sharpe, D. J.; Wong, S. S. Biochem. Genet. 1991,
29, 461.
(8) Coon, S. L.; Roseboom, P. H.; Baler, R.; Weller, J. L.; Namboodiri,
M. A. A.; Koonin, E. V.; Klein, D. C. Science 1995, 270, 1681.
(9) Palmer, M. A. J.; Differding, E.; Gamboni, R.; Williams, S. F.;
Peoples, O. P.; Walsh, C. T.; Sinskey, A. J.; Masamune, S. J. Biol. Chem.
1991, 266, 8369.
(15) D. P. Martin, R. T. Bibart, and D. G. Drueckhammer, J. Am. Chem.
Soc. 1994, 116, 4660.
(16) Usher, K. C.; Remington, S. J.; Martin, D. P.; Drueckhammer, D.
G.; Biochemistry 1994, 33, 7753. Schwartz, B.; Drueckhammer, D. G.;
Usher, K. C.; Remington, S. J. Biochemistry 1995, 34, 15459.
S0002-7863(96)01624-1 CCC: $12.00 © 1996 American Chemical Society

Reactions of Acetyl-CoA Dependent Acetyltransferases
J. Am. Chem. Soc., Vol. 118, No. 41, 1996 9827
Scheme 2
Scheme 4
merically enriched (S)-alcohol. The (R)-isomer was obtained
by isolation of unreacted starting material after selective
enzymatic acylation of the (S)-isomer of the racemic alcohol.
The enantiomeric purity of resolved products was determined
1
by H-NMR analysis of the corresponding MTPA esters.19,20
The absolute stereochemistry predicted from the lipase reaction
was confirmed by optical rotation19 ([R]20 ) -16.1° for the
D
(R)-isomer) and was further supported by analysis of relative
chemical shifts of the MTPA esters. In the most stable
conformation of MTPA esters, the methine proton, ester
carbonyl, and trifluoromethyl groups are aligned in one plane.
Using this model for the (R)-(+)-MTPA ester, the configuration
in which the terminal methyl group of 11 is syn to the phenyl
group ((R)-11) is expected to display upfield NMR shifts for
the methyl group relative to the isomer in which the methyl
group is syn to the methoxy group ((S)-11).20 The terminal
methyl of the (R)-(+)-MTPA ester of (R)-11 had a chemical
shift of 1.30 ppm while the methyl group from the (S)-11 ester
had a shift of 1.37 ppm, confirming predictions of absolute
stereochemistry from optical rotation.
Compounds 6a and 6b were tested as inhibitors of chloram-
phenicol acetyltransferase, carnitine acetyltransferase, phosphate
acetyltransferase (phosphotransacetylase), acetyl-CoA syn-
thetase, and arylamine acetyltransferase. Results, along with
Km values for acetyl-CoA 1 are shown in Table 1. The primary
alcohol 9 was also tested as an inhibitor of chloramphenicol
acetyltransferase. For chloramphenicol acetyltransferase, 6a was
found to be a 12-fold better inhibitor than 6b, with 9 having a
Ki equivalent to that for 6b. The Ki for 6a at pH 8.0 was 12
µM, 1.5-fold higher than the Ki at pH 7.0. With the other four
enzymes, 6a was also the more potent inhibitor. In general,
the Ki for 6b was fairly near the Km for acetyl-CoA, with the
Ki for 6a ranging from 6-fold to about 27-fold lower than the
Km for 6b.
Scheme 3
the tetrahedral intermediate in reactions catalyzed by the acetyl-
CoA dependent acetyltransferases. We report here the synthesis
of these analogs and their inhibition of the enzymes carnitine
acetyltransferase, chloramphenicol acetyltransferase, phosphate
acetyltransferase, acetyl-CoA synthetase, and arylamine acetyl-
transferase. All five enzymes exhibit more potent inhibition
by the (S)-alcohol, suggesting a common configuration of the
tetrahedral intermediate among these enzymes.
Results
Compounds 6a and 6b were prepared by reaction of 715 with
each enantiomer of 5-amino-2-pentanol (8) as shown in Scheme
2 and as described previously for the synthesis of other CoA
analogs.15 The related primary alcohol 9 was prepared similarly
by reaction of 7 with 4-amino-1-butanol (10). The individual
isomers of 8 were prepared by catalytic transfer hydrogenation
of the corresponding nitro compounds 11 as shown in Scheme
3.17 Individual isomers of 11 were in turn prepared in >98%
ee by enzymatic resolution of racemic material18 using the lipase
from Candida rugosa (Scheme 4).19 The (S)-isomer was
obtained by isolation of the enzymatically produced (S)-acetyl
derivative 12, which was subsequently subjected to enzymatic
hydrolysis using the same enzyme to form the further enantio-
Discussion
The secondary alcohols 6a and 6b were chosen as simple
analogs of the tetrahedral intermediate represented by 3a and
3b. This choice was based largely on the utility of hydroxy-
ethylene isosteres of peptides as protease inhibitors.13,14 For
both chloramphenicol acetyltransferase and carnitine acetyl-
transferase, 6a, which matches the predicted stereochemistry
of the tetrahedral intermediate, was a better inhibitor than the
(17) Brande, E. A.; Linstead, R. P.; Wooldridge, K. R. H. J. Chem. Soc.
1954, 3586.
(18) Nakamura, K.; Kitayama, T.; Inoue, Y.; Ohno, A. Tetrahedron 1990,
46, 7471.
(20) Dale, J. A.; Mosher, H. S. J. Am. Chem. Soc. 1973, 95, 512. Ohtani,
I.; Kusumi, T.; Kashman, Y.; Kakisawa, H. J. Am. Chem. Soc. 1991, 113,
4092.
(19) Nakamura, K.; Inoue, Y.; Kitayama, T.; Ohno, A. Agric. Biol. Chem.
1990, 54, 1569.

9828 J. Am. Chem. Soc., Vol. 118, No. 41, 1996
Schwartz and Drueckhammer
Table 1. Michaelis and Inhibition Constants for Substrates and Inhibitors of Acetyltransferases
Km or Ki (mM)
chloramphenicol
acetyltransferase
carnitine
acetyltransferase
phosphate
acetyltransferase
acetyl-CoA
synthetase
arylamine
acetyltransferase
compd
1
0.03
0.008
0.1
0.04
0.01
0.06
0.4
0.02
0.35
1.1
0.18
2.1
0.6
0.03
0.8
6a
6b
9
0.1
originally predicted by Pauling,21 though values up to 3.6 kcal/
mol have been observed in methyl group enhancement of
substrate binding by certain aminoacyl-tRNA synthetases.22
While all five enzymes studied in this work exhibit selective
inhibition by the (S)-isomer of 6, the inhibition is fairly modest.
The Ki for 6a ranges from 4-fold to 20-fold lower than the Km
for acetyl-CoA. A number of factors could contribute to this
modest inhibition relative to the inhibition of proteases by the
related peptide isosteres.10,11 Steric interactions between the
hydroxyl group or other nucleophilic atom of the nucleophilic
substrate and the hydrogen atom at the alcohol carbon may have
a small negative effect on binding, even with the configuration-
ally preferred inhibitor, though a similar effect might be seen
in the related protease inhibitors. Furthermore, efficient rec-
ognition of a tetrahedral intermediate or mimic vs a planar
trigonal species requires recognition of three corners of the
tetrahedron. In the peptide isosteres, the hydroxyl group and
the two peptide moieties attached to the tetrahedral alcohol
center may all serve as good recognition elements. While the
hydroxyl and CoA moieties of the acetyl-CoA analogs may
serve as good recognition elements, the small and unfunction-
alized methyl group may be a poor third recognition element
so that greatly enhanced affinity relative to acetyl-CoA is not
observed. Perhaps more importantly, less enzymatic stabiliza-
tion of the tetrahedral intermediate may be necessary in acyl
transfer reactions involving CoA thioesters than in protease-
catalyzed amide hydrolysis. This differential stabilization is
predicted on the basis of lower ground state energies of amide
bonds vs thioester bonds due to the significant π orbital
delocalization found in amides but similar energies for the
tetrahedral intermediates. The protease-catalyzed reactions thus
involve a greater activation energy, requiring these enzymes to
provide more stabilization of the tetrahedral species. Indeed,
site-directed mutagenesis studies of the serine protease subtilisin
predict approximately 10 kcal/mol stabilization provided by the
oxyanion hole of these enzymes,23,24 compared to only 3 kcal/
mol predicted by similar studies of chloramphenicol acetyl-
transferase.25,26 Weaker binding of the tetrahedral intermediate
by the acetyltransferases relative to the proteases could translate
to weaker binding of an analog of this intermediate. However,
despite the modest inhibition of acetyltransferases by 6a and
6b, the stereodiscrimination of the two isomers 6a and 6b is
clear and provides a prediction of the stereochemistry of the
tetrahedral intermediate.
Figure 1.
epimer 6b. 6a was also the better inhibitor of the phosphate
and arylamine acetyltransferases and of acetyl-CoA synthetase,
which in the reverse direction catalyzes the transfer of an acetyl
group from acetyl-CoA to AMP. For these three enzymes, the
stereoconfiguration of the tetrahedral intermediate had not been
previously predicted. The results reported here suggest that
these three enzymes also proceed through a tetrahedral inter-
mediate which matches the stereochemistry of the (S)-alcohol
6a.
The expected basis for selective inhibition by the isomer of
6 which matches the configuration of the tetrahedral intermediate
is illustrated in Figure 1. In the ternary complex of enzyme,
nucleophilic substrate (represented by ROH, as in the chloram-
phenicol and carnitine acetyltransferases), and 6a, the hydroxyl
group of 6a is expected to bind in the oxyanion binding site of
the enzyme. The absence of increased inhibition of chloram-
phenicol acetyltransferase at higher pH supports inhibition by
the neutral form of the alcohol rather than the alkoxide. The
potential steric interaction between the hydrogen atom on the
chiral carbon of the inhibitor and the nucleophilic oxygen is
probably fairly small. Binding of 6b in the active site as shown
in Figure 1 would result in more severe steric interaction
between the hydroxyl oxygens of 6b and the nucleophilic
substrate and would not benefit from interactions of the hydroxyl
oxygen of 6b with the oxyanion binding site. Alternatively,
binding as illustrated by 6b′ would allow the hydroxyl group
to bind in the oxyanion binding site, with the hydrogen and
methyl groups reversed relative to that shown for the structure
of the complex with 6a. This binding mode would also be
expected to result in unfavorable steric interactions, in this case
between the hydroxyl oxygen of the nucleophile and the methyl
group of the 6b, and would result in a loss of favorable binding
interactions for the methyl group. The fact that the isomer of
6 which matches the predicted stereochemistry of the tetrahedral
intermediate for chloramphenicol acetyltransferase and carnitine
acetyltransferase is the better inhibitor for both enzymes supports
the above analysis.
Recent studies have shown two areas of sequence homology
between an arylalkylamine acetyltransferase and several other
For further support of the significance of interactions with
the methyl group in inhibitor binding, the primary alcohol 9
was tested as an inhibitor of chloramphenicol acetyltransferase.
The Ki 12-fold higher than the Ki for 6a but near the Ki for 6b
indicates that binding of the methyl group by the enzyme
contributes about 1.5 kcal/mol to the binding of 6a, which
matches the stereochemistry of the tetrahedral intermediate,
while the methyl group has little effect on binding by the
mismatched isomer 6b. The additional binding imparted by
the methyl group of 6a is above the value of 0.9 kcal/mol
(21) Pauling, L. Festschr. Prof. Dr. Arthur Stoll Siebzigsten Geburstag
1958, 597.
(22) Fersht, A. R., Shindler, J. S.; Tsui, W.-C. Biochemistry 1980, 19,
5520.
(23) Braxton, S.; Wells, J. A. J. Biol. Chem. 1991, 266, 11797.
(24) Values of 13-14 kcal/mol have been predicted on the basis of
computational studies (Daggett, V.; Schroder, S.; Kollman, P. J. Am. Chem.
Soc. 1991, 113, 8926).
(25) Lewendon, A.; Murray, I. A.; Shaw, W. V.; Gibbs, M. R.; Leslie,
A. G. W. Biochemistry 1990, 29, 2075.
(26) Lewendon, A.; Shaw, W. V. J. Biol. Chem. 1993, 268, 20997.

Reactions of Acetyl-CoA Dependent Acetyltransferases
J. Am. Chem. Soc., Vol. 118, No. 41, 1996 9829
(13.3 mg, 0.10 mmol). The reaction was stirred until the formation of
precipitate ceased, after which 3,3′-diamino-N-methyldipropylamine
(32.2 uL, 0.20 mmol) was added to quench. After 5 min, the solution
was diluted with ether (5 mL) and washed with cold, dilute hydrochloric
acid (2 × 5 mL), cold, saturated sodium bicarbonate (5 mL), and
saturated brine (5 mL). The ether was dried over magnesium sulfate,
filtered, and concentrated in Vacuo to yield a clear oil (40 mg, 0.11
mmol). 1H-NMR for the (+)-MTPA ester of (S)-5-nitro-2-pentanol
(CDCL3): δ 1.37, d, 3H, JH-H ) 6.4 Hz; δ 1.6-1.8, m, 2H; δ 2.0-
2.1, m, 2H; δ 3.52, d, 3H, JH-H ) 1.2 Hz; δ 4.36, t, 2H, JH-H ) 6.8
Hz; δ 5.1-5.3, m, 1H; δ 7.3-7.6, m, 5H. 1H-NMR for the (+)-MTPA
ester of (R)-5-nitro-2-pentanol (CDCL3): δ 1.30, d, 3H, JH-H ) 6.4
Hz; δ 1.6-1.8, m, 2H; δ 2.0-2.1, m, 2H; δ 3.52, d, 3H, JH-H ) 1.2
Hz; δ 4.36, t, 2H, JH-H ) 6.8 Hz; δ 5.1-5.3, m, 1H; δ 7.3-7.6, m,
5H.
acetyl-CoA dependent acetyltransferases. This suggests an
evolutionary relationship among these enzymes, though this
homology is not shared by many other acetyl-CoA dependent
acetyltransferases. Evolutionarily related enzymes would be
expected to proceed through a tetrahedral intermediate of the
same stereoconfiguration. The results of this work support a
possible evolutionary relationship among acetyltransferases, as
the five enzymes studied all show inhibition by the same isomer
of 6. Further studies of additional acetyltransferases using this
pair of probes will permit further investigation of whether the
stereochemistry of the tetrahedral intermediate is conserved
among all of this class of enzymes or whether other acetyl-
transferases will be found for which the tetrahedral intermediate
has the opposite stereochemistry.
(R)-5-Amino-2-pentanol. To a solution of (R)-5-nitro-2-pentanol
(0.60 g, 4.5 mmol) in cyclohexene (9.0 mL) was added palladium (10%
on carbon, 30 mg). After heating overnight at 85 °C, the reaction was
filtered through Celite and concentrated in Vacuo to yield a clear oil
(0.37 g, 3.6 mmol). The amine was used without further purification.
1H-NMR (D2O): δ 1.15, d, 3H, JH-H ) 3.0 Hz; δ 1.4-1.6, m, 4H; δ
2.6-2.8, t, 2H; δ 3.7-3.9, m, 1H.
(S)-5-Amino-2-pentanol. To a solution of (S)-5-nitro-2-pentanol
(0.28 g, 2.1 mmol) in cyclohexene (5.0 mL) was added palladium (10%
on carbon, 20 mg). After heating overnight at 85 °C, the reaction was
filtered through Celite and concentrated in Vacuo to yield a clear oil
(0.17 g, 1.7 mmol). The amine was used without further purification.
1H-NMR (D2O): δ 1.15, d, 3H, JH-H ) 3.0 Hz; δ 1.4-1.6, m, 4H; δ
2.6-2.8, t, 2H; δ 3.7-3.9, m, 1H.
Compounds 6a and 6b should serve as general probes for
predicting the stereochemistry of the tetrahedral intermediate
in reactions catalyzed by acetyl-CoA dependent acetyltrans-
ferases. This information may be useful in guiding the design
of inhibitors of this class of enzymes. These probes may also
serve as new nonhydrolyzable acyl-CoA analogs for use in
studies of enzyme-ligand complexes by X-ray crystallogra-
phy.16 Such studies may provide additional insights into the
means of stabilization of the tetrahedral intermediate in these
enzymes and how it relates to stabilization of the analogous
intermediates in reactions of the protease and esterase enzymes.
These and related studies are currently underway.
(R)- and (S)-(2-Hydroxypropyl)dethio-Coenzyme A. (R)-5-Amino-
2-pentanol (0.35 g, 3.4 mmol) was dissolved in water (1.5 mL), and
the pH was adjusted to 10.0 with 4 M HCl. A 40 mg sample of
adenosine 5′-(trihydrogen diphosphate) 3′-(dihydrogen phosphate) 5′-
[(R)-3-hydroxy-4-[[3-(propylthio)-3-oxopropyl]amino]-2,2-dimethyl-4-
oxobutyl] ester15 was added, and the pH was again adjusted to 10.0.
The reaction was followed by analytical reversed-phase HPLC using
conditions previously descibed.15 The product eluted with a retention
time of 12.2 min, and the side product due to hydrolysis of the
thiopropyl ester eluted at 10.3 min. After the complete disappearance
of the thioester starting material (retention time 17.8 min) was observed,
the reaction was acidified with 4 M HCl to pH 5.0 and purified by
preparative reversed-phase HPLC, and residual phosphates were
removed using a SPICE cartridge. Similar conditions were used to
prepare the (S)- and (R/S)-(2-hydroxypropyl)dethio-coenzyme A. 1H-
NMR (D2O): δ 0.67, s, 3H; δ 0.79, s, 3H; δ 1.06, d, 3H, JH-H ) 6.0
Hz; δ 1.35-1.45, m, 2H; δ 1.5-1.7, m, 2H; δ 2.41, t, 2H, JH-H ) 6.8
Hz; δ 2.93, t, 2H, JH-H ) 7.6 Hz; δ 3.32, t, 2H, JH-H ) 6.8 Hz; δ
3.4-3.5, m, 0.5H; δ 3.6-3.7, m, 1.5H; δ 3.89, s, 1H; δ 4.1-4.2, m,
2H; δ 6.03, d, 1H, JH-H ) 5.2 Hz; δ 8.21, s, 1H; δ 8.44, s, 1H.
(2-Hydroxyethyl)dethio-Coenzyme A. 4-Amino-1-butanol (0.3 g,
4.0 mmol, purchased from Aldrich) was dissolved in water (1.0 mL),
and the pH was adjusted to 10.0 with 4 M HCl. A 15 mg sample of
adenosine 5′-(trihydrogen diphosphate) 3′-(dihydrogen phosphate) 5′-
[(R)-3-hydroxy-4-[[3-(propylthio)-3-oxopropyl]amino]-2,2-dimethyl-4-
oxobutyl] ester was added, and the pH was again adjusted to 10.0.
The reaction was followed by analytical reversed-phase HPLC. The
product eluted with a retention time of 11.4 min, and the side product
due to hydrolysis of the thiopropyl ester eluted at 10.2 min. After the
complete disappearance of the thioester starting material (retention time
17.8 min) was observed, the reaction was acidified with 4 M HCl to
pH 5.0 and purified as (2-hydroxypropyl)dethio-coenzyme A. 1H-NMR
(D2O): δ 0.56, s, 3H; δ 0.68, s, 3H; δ 1.2-1.4, m, 4H; δ 2.25, t, 2H,
JH-H ) 6.4 Hz; δ 2.93, t, 2H, JH-H ) 6.0 Hz; δ 3.26, t, 2H, JH-H ) 6.0
Hz; δ 3.35-3.45, m, 1.5H; δ 3.6-3.7, m, 0.5H; δ 3.79, s, 1H; δ 4.0-
4.1, m, 2H; δ 6.02, d, 1H, JH-H ) 6.0 Hz; δ 8.18, s, 1H; δ 8.41, s, 1H.
Inhibition Studies of Chloramphenicol Acetyltransferase. Inhibi-
tion of CAT by the acetyl-CoA analogs was determined as previously
described15 in 0.8 mL of Tris buffer (0.1 M, pH 7.0) with 0.01 unit of
chloramphenicol acetyltransferase, 0.1 mM chloramphenicol, 0.2 mM
DTNB, 0.01-0.08 mM acetyl-CoA, inhibitor, and water to bring the
final volume to 1.0 mL per assay. The reactions were monitored at
412 nm using ꢀ412 ) 13.6 × 103 M-1 cm-1 for DTNB. Ki values were
Experimental Section
Materials. Chloramphenicol acetyltransferase from Escherichia coli,
carnitine acetyltransferase from pigeon breast muscle, phosphate
acetyltransferase (phosphotransacetylase) from Bacillus stearothermo-
philus, acetyl-CoA synthetase from Bakers yeast, and arylamine
acetyltransferase from pigeon liver were all obtained from Sigma and
used without further purification.
(R)-5-Nitro-2-pentanol. To a solution of racemic 5-nitro-2-pentanol
(6.0 g, 45.0 mmol, prepared as described previously18) suspended in
hexanes (800 mL) was added isopropenyl acetate (11.0 mL, 250 mmol)
and lipase (10.0 g from C. rugosa). The heterogeneous reaction was
stirred vigorously with aliquots removed once an hour for analysis by
1H-NMR. After 30% conversion was observed, the reaction was filtered
and concentrated in Vacuo to remove hexanes and excess isopropenyl
acetate. The acetate and alcohol were separated on silica gel (800 mL)
with 1:1 hexane-ethyl acetate (Rf ) 0.60 for acetate, 0.27 for alcohol;
visualized with KMnO4). The acetate was used for preparation of the
(S)-alcohol (see below). The isolated alcohol (3.9 g, 29.3 mmol) was
resuspended in hexanes (500 mL), and isopropenyl acetate (8.6 mL,
194 mmol) and lipase (11.3 g) were added. The reaction was monitored
as before, and was filtered after 54% conversion to acetate. The acetate
and alcohol were separated as before to yield the (R)-nitro alcohol as
a clear oil (2.1 g, 15.8 mmol). 1H-NMR for 5-nitro-2-pentanol
(CDCL3): δ 1.22, d, 2H, JH-H ) 6.6 Hz; δ 1.4-1.6, m, 2H; δ 1.9-
2.0, br s, 1H; δ 2.0-2.3, m, 2H; δ 3.8-4.0, m, 1H; δ 4.45, t, 2H, JH-H
) 7 Hz. 1H-NMR for acetate (CDCL3): δ 1.25, d, 3H, JH-H ) 6.4
Hz; δ 1.6-1.7, m, 2H; δ 2.05, s, 3H; δ 2.0-2.1, m, 2H; δ 4.42, t, 2H,
JH-H ) 7.0 Hz; δ 4.9-5.0, m, 1H.
(S)-5-Nitro-2-Pentanol. The acetate isolated after 30% enzymatic
conversion of the racemic alcohol (1.73 g, 9.9 mmol) was dissolved in
phosphate buffer (10 mM, 100 mL, pH 7.5). Lipase (500 mg) was
added, and the pH was maintained by the addition of sodium hydroxide
(0.5 M) with a pH Stat. After 50% hydrolysis (as measured by the
addition of 10 mL of base), the reaction was extracted with methylene
chloride (3 × 25 mL) and the combined organics were washed with
water (2 × 25 mL). The methylene chloride layer was dried over
magnesium sulfate, filtered, and concentrated in Vacuo to yield the (S)-
nitro alcohol as a clear oil (0.40 g, 2.9 mmol).
(+)-r-Methoxy-r-(trifluoromethyl)phenylacetate (MTPA Ester)
of (R)- and (S)-5-Nitro-2-pentanol. To a solution of (-)-MTPA
chloride (26 uL, 0.14 mmol) in dry pyridine (300 uL) and carbon
tetrachloride (300 uL) was added either the (R)- or (S)-5-nitro-2-pentanol

9830 J. Am. Chem. Soc., Vol. 118, No. 41, 1996
Schwartz and Drueckhammer
calculated from double-reciprocal plots of 1/V vs 1/[acetyl-CoA] at four
concentrations of the inhibitors.
5.4 × 103 M-1 cm-1 for the thioester. Concentrations of AMP,
magnesium chloride, and pyrophosphate were limited due to the
formation of precipitates in the assay. Vigorous mixing before the
addition of acetyl-CoA ensured solubilization of all components. Ki
values were calculated from double-reciprocal plots of 1/V vs 1/[acetyl-
CoA] at four concentrations of the inhibitors.
Inhibition Studies of Carnitine Acetyltransferase. Inhibition was
determined as previously described27 in 0.8 mL of potassium phosphate
buffer (25 mM, pH 7.5) with 0.026 unit of carnitine acetyl transferase,
2.0 mM L-carnitine, 0.2 mM DTNB, 0.01-0.1 mM acetyl-CoA,
inhibitor, and water to bring the total volume to 1.0 mL per assay. The
reactions were monitored at 412 nm using ꢀ412 ) 13.6 × 103 M-1 cm-1
for DTNB. Ki values were calculated from double-reciprocal plots of
1/V vs 1/[acetyl-CoA] at four concentrations of the inhibitors.
Inhibition Studies of Phosphotransacetylase. Inhibition was
determined in 0.8 mL of Tris buffer (0.1 M, pH 7.5) with 0.17 unit of
phosphotransacetylase, 20 mM potassium phosphate, 20 mM am-
monium chloride, 0.035-0.14 mM acetyl-CoA, inhibitor, and water
to bring the total volume to 1.0 mL per assay. The reactions were
monitored at 240 nm using ꢀ240 ) 5.4 × 103 M-1 cm-1 for the thioester.
Ki values were calculated from double-reciprocal plots of 1/V vs
1/[acetyl-CoA] at four concentrations of the inhibitors.
Inhibition Studies of Arylamine Acetyltransferase. Inhibition was
determined using a modified procedure of that previously described28
in 0.8 mL of potassium phosphate buffer (0.0625 M, pH 7.8) with 0.04
unit of arylamine acetyltransferase, 1 mM EDTA, 0.1 mM aniline, 1
mM DTNB, 0.112-1.12 mM acetyl-CoA, inhibitor, and water to bring
the total volume to 1.0 mL per assay. The reactions were monitored
at 412 nm using ꢀ412 ) 13.6 × 103 M-1 cm-1 for DTNB. The large
amount of enzyme used per assay and the replacement of p-nitroaniline
with aniline were necessary due to the failure to detect activity using
the previously described procedure.28 Ki values were calculated from
double-reciprocal plots of 1/V vs 1/[acetyl-CoA] at three concentrations
of the inhibitors.
Inhibition Studies of Acetyl-CoA Synthetase. Inhibition was
determined in 0.8 mL of Tris buffer (0.05 M, pH 8.0) with 0.007 unit
of acetyl-CoA synthetase, 1 mM adenosine 5′-monophosphate (AMP),
1 mM sodium pyrophosphate, 1 mM magnesium chloride, 0.058-0.58
mM acetyl-CoA, inhibitor, and water to bring the total volume to 1.0
Acknowledgment. This work was supported by National
Science Foundation Grant MCB9321371.
JA9616241
mL per assay. The reactions were monitored at 240 nm using ꢀ240
)
(27) Ramsay, R. R.; Tubbs, P. K. FEBS Lett. 1975, 45, 21.
(28) Brooks, S. P. J.; Storey, K. B. Anal. Biochem. 1993, 212, 452.