Synthesis of a reaction intermediate analogue of biotin-dependent carboxylases via a selective derivatization of biotin

Article abstract of DOI:10.1021/ol990026z

An efficient and practical synthesis of 1, a unique reaction intermediate analogue of biotin-dependent carboxylases, is described. The synthesis features a selective acylation of the 1′-N of biotin. Target 1 inhibits the activity of the biotin carboxylase

Full text of DOI:10.1021/ol990026z

ORGANIC
LETTERS
1999
Vol. 1, No. 1
99-102
Synthesis of a Reaction Intermediate
Analogue of Biotin-Dependent
Carboxylases via a Selective
Derivatization of Biotin
David R. Amspacher, Carol Z. Blanchard, Frank R. Fronczek, Marcelo C. Saraiva,
Grover L. Waldrop,* and Robert M. Strongin*
Department of Chemistry and the Department of Biological Sciences,
Louisiana State UniVersity, Baton Rouge, Louisiana 70803
rob.strongin@chem.lsu.edu
Received April 9, 1999
ABSTRACT
An efficient and practical synthesis of 1, a unique reaction intermediate analogue of biotin-dependent carboxylases, is described. The synthesis
features a selective acylation of the 1′-N of biotin. Target 1 inhibits the activity of the biotin carboxylase component of acetyl CoA carboxylase.
It is the first known biotin-derived inhibitor of biotin carboxylase and should promote new kinetic and structural studies of the biotin-dependent
carboxylases.
Biotin functions as a cofactor for a group of enzymes that
catalyze carboxylation reactions, transcarboxylation reactions,
and decarboxylation reactions.1,2 Currently, we are studying
the kinetic and structural aspects of the biotin-dependent
enzyme acetyl CoA carboxylase, which catalyzes the first
step in fatty acid biosynthesis. These studies have been
hampered by the low affinity (Km ) 134 mM) that biotin
has for the enzyme.3 One way to increase the affinity of
biotin for acetyl CoA carboxylase might be to develop stable
substrate analogues that incorporate biotin. To this end, we
describe the synthesis and bioactivity of a unique first-
generation biotin-derived inhibitor (1) of acetyl CoA car-
boxylase. The synthesis of 1 should allow for a variety of
novel structural modifications of biotin.
All biotin-dependent carboxylases catalyze their respective
reactions via the two-step sequence shown in Scheme 1.2
Scheme 1
The elegant studies of Lane and co-workers showed that in
the first partial reaction biotin is carboxylated on the 1′-N.4
Since bicarbonate is the source of CO2 for all biotin-
dependent carboxylases, the carboxylation of the 1′-N of
(1) Vitamins and Coenzymes; Wagner, A. F., Folkers, K., Eds.; Wiley:
New York, 1964; pp 138-159.
(2) Good, H. G.; Barden, R. E. Annu. ReV. Biochem. 1977, 46, 385-
413.
(3) Blanchard, C. Z.; Lee, Y. M.; Frantom, P. A.; Waldrop, G. L.
Biochemistry 1999, 38, 3393-3400.
(4) Guchhait, R. B.; Polakis, S. E.; Hollis, D.; Fenselau, C.; Lane, M.
D. J. Biol. Chem. 1974, 249, 6646-6656.
10.1021/ol990026z CCC: $18.00 © 1999 American Chemical Society
Published on Web 05/17/1999

Scheme 2 Preparation of 1 from (+)-Biotin
biotin is accomplished by activating bicarbonate through
bonding of the ureido moiety is the major contributor to this
phosphorylation with ATP to form a carboxyphosphate
intermediate.5 The carboxyl group is then transferred from
carboxyphosphate to biotin to form carboxybiotin.
Target compound 1 (Figure 1) is formally derived from
phosphonacetic acid coupled to the 1′-N of biotin. Thus,
compound 1 should embody a stable analogue of the
naturally occurring carboxyphosphate intermediate involved
in biotin-mediated CO2 transfer.
tight binding.7,8
Given the fact that the ureido ring of biotin is of great
importance for binding to avidin and that the 1′-N is directly
involved in biotin’s role as a carboxy transfer intermediate,
it is surprising that relatively few studies have involved
functionalization of the 1′-N of biotin.9,10 The synthesis of 1
(Scheme 2) begins with protection of (+)-biotin as the
corresponding benzyl ester in 95% yield via stirring DMF
(60 mL), biotin (10.2 g, 42.6 mmol, 1 equiv), benzyl alcohol
(5.6 g, 52 mmol, 1.2 equiv), DCC (46.2 mL of a 1 M solution
in CH2Cl2, 46.2 mmol, 1.1 equiv), DMAP (0.51 g, 4.2 mmol,
0.1 equiv), and HOBt (0.64 g, 4.2 mmol, 0.1 equiv), at room
temperature for 12 h.11 Chloride 3 is obtained in 98% yield
by careful dropwise addition of Et3N (7.06 mL, 5.1 mmol,
3 equiv) and chloroacetyl chloride (1.95 mL, 25 mmol, 1.5
equiv) in three portions to a solution of biotin benzyl ester
(5.6 g, 17 mmol, 1 equiv) in CH2Cl2 at -78 °C and warming
to room temperature (12 h).12 No diacylated product is
observed under these conditions. Up to 25% diacylated
product has been observed upon acylation of biotin methyl
ester with methyl chloroformate9 or trifluoroacetic anhy-
dride.10 We obtain up to 10% diacylation with biotin methyl
1
ester and chloroacetyl chloride, as evidenced by H NMR.
(7) (a) Launer, H. F.; Fraenkel-Conrat, H. J. Biol. Chem. 1951, 193, 125-
132. (b) Green, N. M. Biochem. J. 1963, 89, 585-591. (c) Green, N. M.
AdV. Protein Chem. 1975, 29, 85-133.
Figure 1. Carboxyphosphate, biotin, and target stable reaction
intermediate analogue 1.
(8) Weber, P. C.; Ohlendorf, D. H.; Wendoloski, J. J.; Salemme, F. R.
Science 1989, 243, 85-88.
(9) Knappe, J.; Ringelmann, E.; Lynen, F. Biochem. Z. 1961, 335, 168-
176.
In the area of biotechnology, biotin has become an
essential reagent in methods used to label, detect, and purify
proteins and nucleic acids.6 These methods are based on the
remarkable affinity biotin has for the proteins avidin and
streptavidin. The dissociation constant of biotin from avidin
and streptavidin is about 10-15 M, which is one of the tightest
interactions between a protein and a ligand.7 The hydrogen
(10) Berkessel, A.; Breslow, R. Bioorg. Chem. 1986, 14, 249-261.
(11) 2: 1H NMR (DMSO-d6) δ 7.36 (s, 5H), 6.43 (s, 1H), 6.36 (s, 1H),
5.08 (s, 2H), 4.29 (m, 1H), 4.13 (m, 1H), 3.07 (m, 1H), 2.84-2.50 (m,
2H), 2.36 (t, J ) 7.2 Hz, 2H), 1.65-1.25 (m, 6H); 13C (DMSO-d6) δ 173.59,
163.56, 137.15, 129.30, 128.79, 66.18, 61.88, 60.03, 56.20, 34.15, 28.84,
25.37; IR (KBr) 3244, 2930, 1740, 1692, 1458, 1428, 1262, 1170, 737,
698 cm-1; HRMS m/z found 335.1432 (calcd for C17H23N2O3S, 335.1429
MH+).
(12) 3: 1H NMR (DMSO-d6) δ 8.11 (s, 1H), 7.36 (s, 5H), 5.08 (s, 2H),
4.81 (m, 3H), 4.19 (m, 1H), 3.20 (m, 1H), 3.03-2.83 (m, 2H), 2.37 (t, J )
7.2 Hz, 2H), 1.71-1.23 (m, 6H); 13C NMR (DMSO-d6) δ 173.56, 166.32,
156.64, 137.14, 129.14, 128.80, 66.20, 62.20, 58.84, 55.42, 44.89, 38.03,
34.10, 28.81, 25.24; IR (thin film) 3332, 3134, 3134, 2937, 1735, 1704,
1395, 1358, 1241, 792, 752, 699 cm-1; HRMS m/z found 411.1137 (calcd
for C19H23ClO4N2S, 411.1245 MH+).
(5) Knowles, J. R. Annu. ReV. Biochem. 1989, 58, 195-221.
(6) (a) Diamonds, E. P.; Christopoulos, T. K. Clin. Chem. 1991, 37, 625-
636. (b) Methods in Enzymology; Wilchek, M., Edward, E. A., Eds.;
Academic Press: New York, 1990; Vol. 184.
100
Org. Lett., Vol. 1, No. 1, 1999

Energy minimization studies (Sybil 6.1) show that a benzyl
moiety blocks the 3′-N more effectively than does a methyl.
Compound 3 (6.9 g, 17 mmol, 1 equiv) undergoes an
Arbuzov reaction at 100 °C with P(OEt)3 (20 mL, neat) to
afford the phosphonate ester 4 in 89% yield.13 The structure
of 4 is confirmed by X-ray crystallography.14 Phosphonate
ester hydrolysis of 4 (4.75 g, 9.3 mmol, 1 equiv) promoted
by TMSBr (3.7 mL, 28 mmol, 3 equiv) affords 3.0 g (72%)
of phosphonic acid.15 Subsequent saponification of 5 (0.51
g, 1.1 mmol, 1 equiv) with LiOH (0.21 g, 4.9 mmol, 4.5
equiv) affords the water-soluble target 1 (59%) for inhibition
studies.16
If 1 is in fact a reaction intermediate analogue of biotin-
dependent carboxylases, then it should act as an inhibitor.
To test this hypothesis, the effect of 1 on the activity of biotin
carboxylase from Escherichia coli was examined. Biotin
carboxylase is one component of E. coli acetyl CoA
carboxylase and catalyzes the first partial reaction shown in
Scheme 1.17 We chose it as a model for biotin-dependent
carboxylases because the gene for the enzyme has been
cloned and overexpressed.18,19 In addition, the crystal struc-
ture of biotin carboxylase has been determined, which is the
first and only three-dimensional model of a biotin-dependent
carboxylase.20
regression analysis yielded a slope inhibition constant of 8.4
( 1.1 mM. While this is a modest degree of inhibition,
compared to the Km for biotin in biotin carboxylase (134
mM), simply by positioning a phosphonacetyl moiety on the
1′-N of biotin, the affinity of biotin for biotin carboxylase
increases dramatically. Moreover, a previous study found that
none of the current biotin derivatives inhibited biotin
carboxylase;21 therefore, 1 represents the first biotin-derived
inhibitor of biotin carboxylase.
In conclusion, we have developed an efficient route toward
1, a reaction intermediate analogue of a biotin-dependent
carboxylase, based on the reaction intermediate carboxy-
phosphate. Compounds 1 and 3 embody a versatile new
biotin-based substrate and electrophilic coupling template,
respectively, that should promote ready access to new
families of active biotin materials. Currently we are preparing
and testing a variety of congeners of 1 as inhibitors of biotin
carboxylase. These compounds could serve as leads for novel
antihyperlipidemic drugs and highly specific biodegradable
herbicides, on the basis of biotin carboxylase’s role in fatty
acid biosynthesis.
Acknowledgment. We thank Dr. W. Dale Treleaven for
assistance in obtaining NMR spectra. This research was
supported by a grant from the NIH (Grant No. GM51261)
to G.L.W. The latter stages of this work were supported by
the Petroleum Research Fund of the American Chemical
Society (Grant No. 32234-AC4) to G.L.W. Support is also
acknowledged from Louisiana State University. Mass spec-
trometry was provided by the Washington University Mass
Spectrometry Resource, an NIH Research Resource (Grant
Compound 1 does inhibit the activity of biotin carboxylase.
Shown in Figure 2 are progress curves where the activity of
biotin carboxylase is measured in the absence and presence
of increasing amounts of 1. As the concentration of 1
increases, the initial velocity of biotin carboxylase decreases.
Compound 1 was found to exhibit linear competitive
inhibition with respect to the substrate ATP. Fitting the data
to the equation for linear competitive inhibition by nonlinear
(13) 4: 1H NMR (CDCl3) δ 7.28 (s, 5H), 5.04 (s, 2H), 4.85 (t, 1H),
4.09 (q, J ) 6.9 Hz, 4H), 3.78 (m, 2H), 3.09 (m, 1H), 2.94 (m, 1H), 2.29
(t, J ) 7.2 Hz, 2H), 1.74-1.36 (m, 6H), 1.28 (t, J ) 6.9 Hz, 6H); 31P
NMR (DMSO-d6) δ 21.3, (s); 13C NMR (DMSO-d6) δ 173.56, 165.02,
157.51, 156.60, 137.14, 129.30, 128.79, 66.20, 62.57, 62.11, 60.62, 57.68,
55.46, 38.29, 34.64, 34.108, 32.92, 28.81, 28.58, 25.24; IR (thin film) 3328,
3262, 2981, 2935, 2867, 1737, 1678, 1396, 1351, 1251, 1149, 1025, 753,
699 cm-1; HRMS m/z found 513.1828 (calcd for C23H33N2O7PS, 513.1824
MH+).
(14) Crystal data for 4: C23H33N2O7PS, monoclinic space group P21, a
) 10.546(1) Å, b ) 12.550(1) Å, c ) 10.674(1) Å, â ) 112.15(1)°, V )
1308.4(5) Å3, Z ) 2, colorless, T ) 299 K, R ) 0.096, GOF ) 4.753.
Displacement parameters of the benzyl ester and phosphonate groups are
large; low-temperature data collection is planned.
(15) 5: 1H NMR (DMSO-d6) δ 7.96 (s, 1H), 7.36 (s, 5H), 5.1 (s, 2H),
4.77 (t, 1H), 4.1 (m, 1H), 3.5 (m, 2H), 3.2 (m, 2H), 3.07-2.80 (m, 2H),
2.37 (t, J ) 7.2 Hz, 2H), 1.69-1.36 (m, 6H); 31P NMR (DMSO-d6) δ 15.5
(s); 13C NMR (DMSO-d6) δ 173.57, 166.40, 157.48, 156.61, 129.31, 128.81.
66.20, 62.10, 57.64, 55.40, 38.47, 35.44, 34.11, 28.82, 28.60, 25.25; IR
(KBr) 3434, 2935, 2856, 1737, 1682, 1399, 1356, 1233, 1184, 1149, 753,
698 cm-1; HRMS m/z found 457.1194 (calcd for C19H25N2O7PS, 457.1198
MH+).
(16) 1: 1H NMR (D2O) δ 4.50 (m, 1H), 4.34 (m, 1H), 3.25 (m, 1H),
2.92-2.64 (m, 2H), 2.43-2.36 (m, 2H), 2.06 (t, J ) 7.2 Hz, 2H). 1.69-
1.25 (m, 6H); 31P NMR (D2O) δ 14.9 (s); 13C NMR (D2O) δ 157.49, 62.33,
60.59, 55.69, 41.38, 40.05, 39.86, 37.65, 28.59, 28.00, 25.00; IR (KBr)
3344, 3196, 2922, 2851, 1702, 1661, 1567, 1431, 1138, 1074, 869 cm-1
;
MS m/z found 394.4 (calcd for C12H17LiN2NaO7PS, 394.0552 MH+).
Figure 2. Progress curves for biotin carboxylase activity with
increasing amounts of 1. The activity of biotin carboxylase was
measured by following the production of ADP using pyruvate kinase
and lactate dehydrogenase. The oxidation of NADH by lactate
dehydrogenase at 340 nm is measured with respect to time. The
substrate concentrations were held constant at 9 mM for bicarbonate,
0.1 mM for ATP, and 100 mM for biotin.
(17) Lane, M. D.; Moss, J.; Polakis, S. E. Curr. Top. Cell Regul. 1974,
8, 139-195.
(18) Li, S. J.; Cronan, J. E., Jr. J. Biol. Chem. 1992, 267, 855-863.
(19) Kondo, H.; Shiratsuchi, K.; Yoshimoto, T.; Masuda, T.; Kitazono,
A.; Tsuru, D.; Anai, M.; Sekiguchi, M.; & Tanabe, T. Proc. Natl. Acad.
Sci. USA 1991, 88, 9730-9733.
(20) Waldrop, G. L.; Rayment, I.; Holden, H. M. Biochemistry 1994,
33, 10249-10256.
(21) Tipton, P. A.; Cleland, W. W. Biochemistry 1988, 27, 4317-4325.
Org. Lett., Vol. 1, No. 1, 1999
101

No. P41RR0954), and the Nebraska Center for Mass
Spectrometry.
including an X-ray crystallographic file in CIF format. This
material is available free of charge via the Internet at
http://pubs.acs.org.
Supporting Information Available: Figures giving NMR
spectra, text giving procedures for the preparation of
compounds 1-5, and X-ray structural information for 4,
OL990026Z
102
Org. Lett., Vol. 1, No. 1, 1999