4,5-Cyclopropanocholestan-3β-ol Substrates for Cholesterol Oxidase and Their 1H NMR Assignments

Article abstract of DOI:10.1021/jo9706537

We have assayed 4,5-cyclopropanocholestan-3-ols and 4,5-cyclopropanocholestan-3-ones and tested them as substrates and inhibitors of cholesterol oxidase. The 4,5-cyclopropanocholestan-3/3-ols (α and β) are substrates of cholesterol oxidase that are converted to their respective ketones 1000-fold more slowly than cholesterol. The induced ring-current effects of a cyclopropane ring are clearly illustrated in the ¹H NMR spectra of these sterols. These shielding effects are dramatic because of the rigidity of the steroid backbone. Assignments of the ¹H resonances of the A, B, and cyclopropyl rings of the sterols have been made using DQF-COSY and NOESY experiments. We have assigned the upfield multiplet at approximately 0.5 ppm to H_6α in both isomers. H_6α is shielded by the cyclopropyl σ bond. H_6β is deshielded by the cyclopropane ring and appears at approximately 2.0 ppm in both isomers.

Full text of DOI:10.1021/jo9706537

J . Org. Chem. 1997, 62, 5893-5897
5893
4,5-Cyclop r op a n och olesta n -3â-ol Su bstr a tes for Ch olester ol
1
Oxid a se a n d Th eir H NMR Assign m en ts
Nicole S. Sampson* and Amy E. McCann
Department of Chemistry, State University of New York, Stony Brook, New York 11794-3400
Received April 11, 1997X
We have assayed 4,5-cyclopropanocholestan-3-ols and 4,5-cyclopropanocholestan-3-ones and tested
them as substrates and inhibitors of cholesterol oxidase. The 4,5-cyclopropanocholestan-3â-ols (R
and â) are substrates of cholesterol oxidase that are converted to their respective ketones 1000-
fold more slowly than cholesterol. The induced ring-current effects of a cyclopropane ring are clearly
1
illustrated in the H NMR spectra of these sterols. These shielding effects are dramatic because
of the rigidity of the steroid backbone. Assignments of the 1H resonances of the A, B, and cyclopropyl
rings of the sterols have been made using DQF-COSY and NOESY experiments. We have assigned
the upfield multiplet at approximately 0.5 ppm to H6R in both isomers. H6R is shielded by the
cyclopropyl σ bond. H6â is deshielded by the cyclopropane ring and appears at approximately 2.0
ppm in both isomers.
We report here the assay of 4,5-cyclopropyl steroids as
substrates and inhibitors of cholesterol oxidase. These
cyclopropyl steroids were designed to test the mechanism
of cholesterol oxidase. Cholesterol oxidase (EC 1.1.3.6)
catalyzes the oxidation and isomerization of cholesterol
into cholest-4-en-3-one (Scheme 1). This enzyme is part
of the bacterial metabolic pathway for utilizing choles-
terol as its carbon source and is secreted by Gram-
positive soil bacteria, including Brevibacterium steroli-
cum (ATCC 81387) and Streptomyces sp. strain SA-COO.
The enzyme is used to assay serum cholesterol levels,
and recently, its larvicidal properties were discovered.1,2
Cholesterol oxidase is a monomer (57 kD) with a single
active site for both oxidation and isomerization and
requires one FAD per active site.3-5 The oxidation may
proceed via a radical intermediate or via direct hydride
transfer. We were interested in probing how much
conformational flexibility was available to the substrate
in the active site and envisioned that ring-opening
reactions would serve as good indicators of flexibility.
These indicators could then be used in our ongoing
investigation of the steroid and membrane binding
properties of cholesterol oxidase. We reasoned that if
there was a radical intermediate in the oxidation, a
cyclopropyl substrate such as 1 might undergo ring
opening. The ring-opened radical could potentially in-
activate the enzyme. Alternatively, if the cyclopropyl
steroids 1a or 1b were converted to the ketones 2a or
2b, the ketones might undergo nucleophilic ring opening
in the active site.6 The active site base for isomerization,
glutamate-361,7,8 is positioned over the â-face of the
steroid, and inspection of the X-ray structure indicates
that it is close enough to form a covalent bond. Thus,
we undertook the synthesis of steroids 1-2 and their
assay with cholesterol oxidase. In addition, spectral
characterization of steroids 1a and 1b revealed a large
perturbation of 1H chemical shifts at C6 due to the
cyclopropyl ring. The results of our enzyme assays of 1a ,
1b, 2a , and 2b and the spectral assignments of 1a and
1b are presented.
Discu ssion
We synthesized steroids 1a and 1b using a modifica-
tion of the procedures of Dauben and co-workers.9 Use
of sonication in the cyclopropanation reactions resulted
in an increase of 20% in product yield compared to
previously used methods. Cholest-4-en-3-one was re-
duced with LiAlH4 to yield cholest-4-en-3-ol. The 3R- and
3â-epimers were separated by silica gel chromatography.
The cholest-4-en-3-ols were cyclopropanated in separate
reactions using CH2I2 and zinc with sonication to activate
the metal to yield 1b and the epimer of 1a . The
3-hydroxy moiety was epimerized via Collins oxidation
to form the 3-ketocyclopropyl steroid 2a , followed by a
reduction with LiAlH4 to provide 1a . Ketone 2b was
prepared in an analogous manner (Scheme 2).
We envisioned that, upon incubation with cholesterol
oxidase, the cyclopropyl steroids 1a ,b would behave as
substrates and/or as irreversible inhibitors. We per-
formed, therefore, two types of assays with the recom-
binant form of the Streptomyces cholesterol oxidase.10 In
the first assay, the rate of production of H2O2 was
followed using horseradish peroxidase, and thus the rate
of oxidation of the 3-hydroxy moiety was measured.
Alcohol 1a was oxidized at 0.1% of the rate of cholesterol
oxidation; 1b was oxidized at 0.2% of the rate for
cholesterol with 50 µM substrate. The limited solubility
of the steroids precluded measurement of kcat and Km.
For every equivalent of alcohol consumed, 1 equiv of H2O2
X Abstract published in Advance ACS Abstracts, August 1, 1997.
(1) Purcell, J . P.; Greenplate, J . T.; J ennings, M. G.; Ryerse, J . S.;
Pershing, J . C.; Sims, S. R.; Prinsen, M. J .; Corbin, D. R.; Tran, M.;
Sammons, R. D.; Stonard, R. J . Biochem. Biophys. Res. Commun. 1993,
196, 1406-13.
(2) Greenplate, J . T.; Duck, N. B.; Pershing, J . C.; Purcell, J . P.
Entomol. Exp. Appl. 1995, 74, 253-258.
(3) Uwajima, T.; Yagi, H.; Terada, O. Agric. Biol. Chem. 1974, 38,
1149-1156.
(7) Kass, I. J .; Sampson, N. S. Biochem. Biophys. Res. Commun.
(4) Vrielink, A.; Lloyd, L. F.; Blow, D. M. J . Mol. Biol. 1991, 219,
533-554.
1995, 206, 688-693.
(8) Sampson, N. S.; Kass, I. J . J . Am. Chem. Soc. 1997, 119, 855-
(5) Li, J .; Vrielink, A.; Brick, P.; Blow, D. M. Biochemistry 1993,
32, 11507-11515.
862.
(9) Dauben, W. G.; Laug, P.; Berezin, G. H. J . Org. Chem. 1966, 31,
(6) Liu, H.-w.; Walsh, C. T. In The Chemistry of the Cyclopropyl
Group; Rappoport, Z., Ed.; Wiley-Interscience: New York, 1987; Vol.
Part 2, p 969.
3869-3871.
(10) Nomura, N.; Choi, K.-P.; Murooka, Y. J . Ferm. Bioeng. 1995,
79, 410-416.
S0022-3263(97)00653-1 CCC: $14.00 © 1997 American Chemical Society

5894 J . Org. Chem., Vol. 62, No. 17, 1997
Sampson and McCann
Sch em e 1
Ta ble 1. 1H Ch em ica l Sh ifts (Er r or 0.01 p p m )
Sch em e 2
proton
1a
1b
H1R
H1â
H2R
H2â
H3R
H4R
H4â
H6R
H6â
H7R
H7â
H19
H28
H29
0.86
1.27
1.84
1.40
4.05
naa
0.82
0.48
1.99
1.01
1.54
1.15
0.29
0.092
0.99
1.29
1.62
1.20
4.33
1.14
na
0.54
2.00
0.89
1.54
0.94
0.045
0.67
a
Not applicable.
was produced. The product(s) were isolated from an
incubation mixture of cholesterol oxidase and 1a or 1b
and characterized. We determined that 1a was converted
to 2a and 1b is converted to 2b by mass spectrometry
and TLC comparison with authentic material. These
products are consistent with the alcohol:H2O2 stoichiom-
etry observed. Thus, the products were ring-intact
cyclopropanocholestanones.
This lack of reaction could be because of improper
alignnment of glutamate-361 with respect to the cyclo-
propyl ring or insufficient electrophilicity of the ketone.
Upon characterization of the alcohol 1b, we discovered
that the spectral assignments of the protons in the A and
B steroid rings in the literature were incomplete.9,12
Protons attached to cyclopropyl rings characteristically
appear at high field. In the 1H NMR spectrum of 1b,
there are three high-field resonances, but only two
correspond to cyclopropyl protons. J oska and Fajkos
assigned the third resonance to H6, but did not designate
whether it was H6R or H6â, and they did not report J
coupling to support their assignment.12 We have used
DQF-COSY and NOESY experiments to assign the
cyclopropyl and A- and B-ring proton resonances of the
steroids. The peak assignments and J couplings are
summarized in Tables 1 and 2. The DQF-COSY experi-
ment was used to identify the cyclopropyl peaks at 0.045,
0.667, and 1.14 ppm. Their assignment was based on
following the proton couplings in 1b from H3R, which is
distinctive with its downfield resonance at 4.33 ppm, to
H4, and from H4 to H28 and H29 (Figure 1). The third
upfield resonance at 0.54 ppm, however, was not coupled
to any of the other cyclopropyl hydrogens. This peak was
coupled to a downfield resonance at 2.00 ppm, with a
large 2J of 13.6 Hz, indicating that they are geminal
protons. We used a NOESY experiment to assign these
resonances. There is an NOE cross-peak between H19
and the resonance at 2.00 ppm and another between H4R
and the resonance at 0.54 ppm (Figure 2). We assigned
In a second assay, we incubated the cyclopropanoster-
oids with cholesterol oxidase and catalase and periodi-
cally removed aliquots. Catalase was included because
it is known that extended incubation of cholesterol
oxidase with H2O2 leads to inactivation of the enzyme.11
The aliquots were assayed for remaining enzymatic
activity by the addition of cholesterol. Incubation of
cholesterol oxidase with 1a , 1b, 2a , or 2b did not result
in irreversible inactivation of cholesterol oxidase.
Our assay results are consistent with either mecha-
nism. The oxidation may proceed via a hydride transfer
without formation of radical intermediate. Alternatively,
the radical intermediate may not partition into a ring-
opening pathway and simply undergoes a second 1 e-
transfer step. Even if ring opening does occur, the
conformational constraints of the active site might pre-
vent geometric rearrangement and favor rapid reversion
to the cyclopropyl ring, resulting in no apparent parti-
tioning into the ring-opening pathway. The question of
whether oxidation proceeds via hydride transfer or radi-
cal intermediate remains open. Furthermore, we do not
see nucleophilic ring opening in the presence of 2a or 2b.
(11) Loubathugel, C.; Tritsch, D.; Biellmann, J . F. C. R. Acad. Sci.,
Ser. III 1994, 317, 299-303.
(12) J oska, J .; Fajkos Collect. Czech. Chem. Commun. 1980, 45,
1850-1859.

4,5-Cyclopropanocholestan-3â-ol Substrates
J . Org. Chem., Vol. 62, No. 17, 1997 5895
Ta ble 2. 1H, 1H Cou p lin g Con sta n ts (Er r or 0.2 Hz)
Deter m in ed fr om 1D 1H Sp ectr a a n d DQF -COSY
Exp er im en t
J (Hz)
1a
1b
2J
1R, 1â
2R, 2â
6R, 6â
7R, 7â
28, 29
1R, 2R
1R, 2â
1â, 2R
1â, 2â
2R, 3R
2â, 3R
3R, 4R
3R, 4â
4R, 28
4R, 29
4â, 28
4â, 29
-14.2
-14.7a
-13.5
nmb
-4.9
4.9
nm
3.7
nm
7.3
-14.5
-15.0
-13.6
-12.3
-4.6
14.5
3.4
3J
3.8
nm
7.2
7.4
noc
nad
no
7.2
na
9.3
4.7
na
na
na
na
4.9
9.8
a
b
Not accurate due to non-first-order splitting. Cross-peak
observed, but coupling constant not measurable. c No cross-peak
observed. Not applicable.
d
H6R to the upfield resonance at 0.54 ppm and H6â to the
downfield resonance at 2.00 ppm on the basis of these
NOE cross-peaks and geometric constraints. The protons
resonances for 1a were assigned in an analogous fashion,
and the relevant portions of their DQF-COSY and
NOESY spectra are shown in Figures 1 and 2.
The conformations of the A- and B-rings of 1a and 1b
were determined using distance constraints from the
NOESY experiment and approximate dihedral angles
calculated from the J couplings in the DQF-COSY and
1D-1H spectra (Scheme 3). The A-ring of 1a is a pseudo-
chair that is trans-fused to the B-ring. The A-ring of 1b
is a pseudo-chair that is cis-fused to the B-ring. These
conformations corresponded to the lowest energy confor-
mations calculated in the absence of NMR constraints.
The number of A/B/cyclopropyl-ring minimum energy
conformers is limited to two in each case because of the
rigid framework of the fused cyclopropyl steroid. It is
this same rigidity that makes the shielding effect of the
cyclopropyl ring on the H6’s so pronounced.
In sterol 1b, H6R is positioned over the cyclopropane
ring (the C4-C5-C6-H6R dihedral is -25.6°) and is
shifted upfield. H6â is positioned outside the ring (the
C4-C5-C6-H6â dihedral is -141.5°) and is shifted down-
field from the position of a typical cyclohexane methylene
at 1.4 ppm. The same is true for sterol 1a (the C4-C5-
C6-H6R dihedral is +27.1° and the C4-C5-C6-H6â dihe-
dral is -88.7°). These shifts may be due to an induced
ring current in the cyclopropyl ring.13 Using the shield-
ing contours generated by Poulter and co-workers13 and
cyclopropane-H6 distances from our minimizations, we
calculated the expected shift difference due to the cyclo-
propane for H6R and H6â. The calculated chemical shifts
agree with those observed. Similar long range upfield
and downfield shifts have been reported for other cyclo-
propyl steroids.14-16
F igu r e 1. 1H NMR DQF-COSY spectra of (a) 1a (600 MHz)
and (b) 1b (500 MHz) at 25 °C. Cross-peaks between H3, H4,
H6, H28, and H29 are labeled with their assignments.
oxidase X-ray crystal structure,4,5 one might expect that
1b would not fit in the enzyme active site because of the
conformation imposed by the cis-fused A/B ring system.
In fact, Slotte reported that 5â-cholestan-3â-ol is not a
substrate for cholesterol oxidase.17 However, the H2O2
assay used in his work was not continuous, and a very
slow rate such as that for 5â-cholestan-3â-ol or 1b would
not be observed. Without more detailed kinetic assays,
we cannot say exactly why 1a and 1b are oxidized so
In summary, it is remarkable that 1b is a substrate
for cholesterol oxidase. After inspection of the cholesterol
(13) Poulter, C. D.; Boikess, R. S.; Brauman, J . I.; Winstein, S. J .
Am. Chem. Soc. 1972, 94, 2291-2296.
(14) Bhacca, N. S.; Wolff, M. E.; Ho, W. Tetrahedron Lett. 1968, 52,
5427-30.
(15) Templeton, J . F.; Paslat, V. G.; Wie, C. W. Can. J . Chem. 1978,
56, 2058-64.
(16) Marat, K.; Templeton, J . F.; Ling, Y.; Lin, W.; Gupta, R. K.
Magn. Reson. Chem. 1995, 33, 529-40.
(17) Slotte, J . P. J . Steroid Biochem. Mol. Biol. 1992, 42, 521-526.

5896 J . Org. Chem., Vol. 62, No. 17, 1997
Sampson and McCann
WI, and solvents, of reagent or HPLC grade, were supplied
by Fisher Scientific, Pittsburgh, PA, unless otherwise specified.
Water for assays and chromatography was distilled, followed
by passage through a Barnstead NANOpure filtration system
to give a resistivity better than 18 MΩ.
All reactions were carried out in oven-dried glassware under
a dry N2 or Ar atmosphere. Solvents were dried and distilled
from Na/benzophenone ketyl. Thin layer chromatography was
performed with precoated silica plates (silica gel 60 F254
,
Merck). Flash chromatography was performed according to
the established protocol18 with silica gel 60 (Mallinckrodt) at
an average height of 20 cm. A Fisher ultrasonic bath (Model
FS9H) was used as a source of ultrasound. Melting points
were determined on Thomas Hoover capillary melting point
apparatus (Philadelphia, PA) and are uncorrected.
NMR Sp ectr oscop y. All spectra were acquired in CDCl3
1
and were externally referenced to TMS in CDCl3. 1H, H-1H
1
DQF-COSY,19,20 and H-1H NOESY21 spectra were performed
on Varian INOVA 500 and INOVA 600 spectrophotometers
(500 and 600 MHz) using the parameters supplied by the
manufacturer. 13C spectra and DEPT experiments were
recorded on a Varian GEMINI 2300 spectrophotometer (75.45
MHz). 1H-1H DQF-COSY spectra were performed with a 90°
pulse width and 3 s relaxation delay. The NOESY spectra
were performed with a 1 s relaxation delay and 1100 ms
mixing times. Both 1H-1H DQF-COSY and NOESY spectra
were acquired with a homospoil-90°-homospoil sequence
preceding d1. Datasets were transformed using a shifted sine
bell with digital resolutions of 0.6 Hz or better. 1H NMR data
are reported in the following manner: chemical shift in ppm
(multiplicity, integration, coupling constant in hertz). Only
resolved resonances are reported. 13C NMR data are reported
in the following manner: chemical shift in ppm (multiplicity).
En er gy Min im iza tion of Str u ctu r es. Steroids 1a and
1b were built and energy minimized using the Builder and
Energy modules of InsightII v 95.0 by Biosym, Inc., San Diego,
CA. The cvff parameter set was used for all calculations.
Different starting conformers were minimized using molecular
mechanics to determine their relative energies.
Ch olest-4-en -3r-ol a n d Ch olest-4-en -3â-ol. The reduc-
tion of cholest-4-en-3-one (1.00 g, 2.6 mmol) was performed
according to the procedure of Dauben.9 The mixture was
purified by silica gel chromatography (EtOAc-hexanes 1:13
v/v) to give 95 mg (9%) of cholest-4-en-3R-ol as a colorless oil:
[1H NMR δ 5.25 (d, 1, J ) 1.6), 4.13 (ddd, 1, J ) 5.9, 6.9, 1.6)]
and 873 mg (86%) of cholest-4-en-3â-ol as a white solid: mp
126.0-126.5 °C (lit.9 mp 131-132 °C); 1H NMR δ 5.43 (d, 1, J
) 2.4), 4.05 (bm, 1).
4â,5â-Meth a n o-5r-ch olesta n -3â-ol (1b). A mixture of Zn
powder (2.03 g, 31.0 mmol) and dry DME (30 mL) maintained
under a static atmosphere of Ar was sonicated for 20 min at
60 °C. Alcohol cholest-4-en-3â-ol (600 mg, 1.55 mmol) and a
small crystal of I2 were added to the solution, followed by
dropwise addition of CH2I2 (936 µL, 11.6 mmol) over 5 min
with continuous sonication and heating. After the DME
reached reflux, sonication and heating were continued for
another 4 h. The mixture was then diluted with Et2O (40 mL),
followed by slow addition of saturated aqueous NH4Cl (30 mL).
The mixture was filtered through a pad of silica. The pad was
washed three times with Et2O. The combined organic extracts
were washed with 5% (w/v) aqueous Na2S2O3, H2O and brine,
dried (MgSO4), and concentrated by rotary evaporation. The
yellow residue was immediately chromatographed on silica gel
(EtOAc-toluene 1:9 v/v) to afford 517 mg of 1b as a clear
colorless oil (83%). Subsequent recrystallization from acetone
produced white crystals: mp 87.0-88.5 °C (lit.9 mp 94.0-95.0
F igu r e 2. 1H NMR NOESY spectra of (a) 1a (600 MHz) and
(b) 1b (500 MHz) with 1100 ms mixing time at 25 °C. Cross-
peaks between H4, H6, H19, H28, and H29 are labeled with their
assignments.
slowly by cholesterol oxidase. Their turnover does indi-
cate, however, that the active site can accommodate
bulky substrates and that the protein is flexible. The
rigid frameworks of 1a and 1b illustrate the induced ring-
current effects of a cyclopropane on neighboring protons.
H6R and H6â are a clear case of induced ring current
effects as opposed to shielding by σ bonds, because we
see both downfield and upfield shifts, depending on the
position of the proton relative to the ring.
1
°C); H NMR see Tables 1 and 2; 13C NMR δ 63.87 (d), 56.54
(d), 56.29 (d), 46.01 (d), 42.34 (s), 40.07 (t), 39.47 (t), 36.12 (t),
(18) Still, W. C.; Kahn, M.; Mitra, A. J . Org. Chem. 1978, 43, 2923-
2924.
(19) Rance, M.; Sørensen, O. W.; Bodenhausen, G.; Wagner, G.;
Ernst, R. R.; Wu¨thrich, K. Biochem. Biophys. Res. Commun. 1983, 117,
479-485.
Exp er im en ta l Meth od s
Gen er a l P r oced u r es. Cholesterol and horseradish per-
oxidase were from Sigma Chemical Co., St. Louis, MO.
pCO117 was a generous gift from Y. Murooka.10 All other
chemicals were supplied by Aldrich Chemical Co., Milwaukee,
(20) Piatini, U.; Sørenson, O. W.; Ernst, R. R. J . Am. Chem. Soc.
1982, 104, 6800-6801.
(21) States, D. J .; Haberkorn, R. A.; Ruben, D. J . J . Magn. Reson.
1982, 48, 286-292.

4,5-Cyclopropanocholestan-3â-ol Substrates
J . Org. Chem., Vol. 62, No. 17, 1997 5897
Sch em e 3
35.76 (d), 35.59 (d), 34.29 (t), 32.96 (s), 30.70 (t), 30.43 (s), 28.25
(t), 27.98 (d), 27.46 (d), 26.36 (t), 26.35 (t), 24.27 (t), 23.79 (t),
22.79 (q), 22.53 (q), 21.54 (t), 21.36 (q), 18.63 (q), 11.88 (q),
10.55 (t). Anal. Calcd for C28H48O: C, 83.93; H, 12.07.
Found: C, 83.93; H, 12.07.
Escherichia coli.8 The assay conditions were 1.13 mM phenol,
0.87 mM 4-aminoantipyrine, and 10 U of horseradish peroxi-
dase in 50 mM sodium phosphate, pH 7.0 buffer, with 0.025%
(w/v) Triton X-100 at 37 °C, 2.1 nM cholesterol oxidase.
Steroids were added as a propan-2-ol solution to a final
concentration of 50 µM. The final assay mixture was never
more than 1.6% propan-2-ol. The activities of the cyclopropyl
steroids 1a ,b were determined by following the rate of forma-
tion of H2O2 and thus, indirectly, 3â-alcohol oxidation. The
formation of quinone imine at 510 nm was followed as a
function of time.
An a lysis of P r od u cts of th e Rea ction Ca ta lyzed by
Ch olester ol Oxid a se w ith 1a a n d 1b a s Su bstr a tes. The
product of enzymatic turnover was isolated from an activity
assay mixture that did not contain horseradish peroxidase,
phenol, or 4-aminoantipyrine. The reaction product was
isolated as previously described.7 The product from the
enzymatic reaction of 1a co-spotted on TLC (1:9 EtOAc:
toluene) with 2a and had m/ z 398 (EI, 70 eV). Upon
treatment of the enzymatic product with LiAlH4 (as described
for the preparation of 1a ), the major product co-spotted with
1a . The product from the enzymatic reaction of 1b co-spotted
on TLC (1:9 EtOAc:toluene) with 2b and had m/ z 398 (EI, 70
eV). Upon treatment of the enzymatic product with LiAlH4
(as described for the preparation of 1a ), the major product co-
spotted with 1b.
4r,5r-Meth a n o-5â-ch olesta n -3-on e (2a ). Treatment of
cholest-4-en-3R-ol (263 mg, 0.7 mmol) in DME (6 mL) with Zn
(891 mg, 13.6 mmol), a small crystal of I2, and CH2I2 (406 µL,
5.04 mmol), under identical conditions as those described for
1b, yielded 213 mg of 4R,5R-cyclopropano-5â-cholestan-3R-ol
(78%) as a white solid: mp 102.5-103.0 °C (lit.9 mp 105-106
°C). Careful addition of CrO3 (300 mg, 3.0 mmol) to a solution
of dry pyridine (485 µL, 6.0 mmol) in CH2Cl2 (5 mL) resulted
in a deep red solution that was stirred for 15 min at rt. To
this mixture was added a solution of 4R,5R-cyclopropano-5â-
cholestan-3R-ol (213 mg, 0.53 mmol) in CH2Cl2 (5 mL), which
upon addition of 4R,5R-cyclopropano-5â-cholestan-3R-ol changed
from deep red to black. The formation of a black precipitate
was observed. After 2 h of continuous stirring, the solution
was decanted from the residue. The residue was washed with
Et2O (3×), and the combined organic extracts were washed
with 5% NaOH (3×), 5% HCl, 5% NaHCO3, and brine, dried
(MgSO4), and evaporated to dryness to afford 206 mg of 2a
(97%) as a white solid: mp 134.0-134.5 °C (lit.9 mp 136-137
1
°C); H NMR δ 2.19 (dd, 1, J ) 9.75, 5.25), 2.09 (ddd, 1, J )
13.5, 3.9, 3.6), 1.97 (ddd, 1, J ) 12.3, 3.0, 3.0), 1.80 (bm, 1);
13C NMR δ 209.71 (s), 56.27 (d), 56.01 (d), 51.99 (d), 42.59 (s),
39.88 (t), 38.47 (t), 36.11 (t), 35.90 (d), 35.76 (d), 34.93 (d), 34.43
(s), 32.73 (t), 32.32 (t), 30.76 (t), 30.71 (t), 28.20 (t), 27.98 (d),
24.20 (t), 23.81 (t), 22.80 (q), 22.54 (q), 21.67 (t), 18.62 (q), 18.62
(t), 17.80 (q), 12.01 (q); mass spectrum (EI, 70 eV) m/ z 398.
Anal. Calcd for C28H46O: C, 84.36; H, 11.63. Found: C, 84.15;
H, 11.69.
Assa y for Ir r ever sible In h ibition . Steroids 1a , 1b, 2a ,
and 2b were incubated separately with 90 nM cholesterol
oxidase in 50 mM sodium phosphate, pH 7.0 buffer, with
0.025% (w/v) Triton X-100 at 37 °C. At various time intervals,
the amount of active cholesterol oxidase remaining was
measured by removing an aliquot (20 µL), diluting to 1 mL
with assay buffer, adding 50 µM cholesterol, and following the
appearance of cholest-4-en-3-one at 240 nm (ꢀ240 ) 12 100 M-1
cm-1 22). The incubations were followed for 5 days.
4â,5â-Meth a n o-5r-ch olesta n -3-on e (2b). Treatment of
1b (100 mg, 0.25 mmol) in CH2Cl2 (1 mL) with CrO3‚pyridine
complex (1:2) under identical conditions as those described for
2a gave 91 mg of 2b (91%) as a white solid: mp 82.0-84.0 °C
(lit.9 mp 89.0-89.5 °C); 1H NMR δ 2.09 (m, 3), 1.96 (dd, 1, J )
12.7, 3.42), 1.81 (bm, 1), 1.62 (m, 2); 13C NMR δ 210.11(s), 56.17
(d), 56.12 (d), 45.86 (d), 42.39 (s), 39.75 (t), 39.46 (t), 36.81 (s),
36.08 (t), 35.76 (d), 35.50 (d), 35.08 (s), 34.28 (s), 32.47 (t), 32.28
(t), 30.10 (t), 28.18 (t), 27.96 (d), 27.60 (t), 24.16 (t), 23.81 (t),
22.77 (q), 22.52 (q), 21.65 (t), 20.63 (q), 18.59 (q), 17.83 (t),
11.91 (q); mass spectrum (EI, 70 eV) m/ z 398. Anal. Calcd
for C28H46O: C, 84.36; H, 11.63. Found: C, 84.39; H, 11.67.
4r,5r-Meth a n o-5â-ch olesta n -3â-ol (1a ). The reduction
of 2a (148 mg, 0.371 mmol) was performed using the procedure
of Dauben.9 Recrystallization from MeOH and acetone pro-
duced 143 mg of 1a (96%) as colorless plates: mp 126.0-127.0
°C (lit.12 mp 131-132 °C); 1H NMR see Tables 1 and 2; 13C
NMR δ 6.87 (d), 56.29 (d), 56.26 (d), 52.04 (d), 42.68 (s), 40.00
(t), 39.47 (t), 36.13 (t), 36.06 (d), 35.77 (d), 33.25 (s), 32.93 (t),
30.82 (s), 30.14 (t), 29.37 (t), 28.21 (t), 27.97 (d), 27.97 (t), 26.69
(d), 24.14 (t), 23.81 (t), 22.78 (q), 22.52 (q), 21.15 (t), 19.20 (q),
18.62 (q), 13.08 (t), 12.11 (q). Anal. Calcd for C28H48O: C,
83.93; H, 12.07. Found: C, 83.78; H, 12.04.
Ack n ow led gm en t. We thank J ohn Sinclair for
initiating the synthesis of 1 and Prof. Y. Murooka for
kindly providing the pCO117 clone of cholesterol oxi-
dase. The 70-VSE mass spectrometer in the Mass
Spectrometry Laboratory, School of Chemical Sciences,
University of Illinois, was purchased in part with a
grant from the Division of Research Resources, National
Institutes of Health. The NMR Spectroscopy facility at
SUNY Stony Brook is supported by a grant from the
NSF (CHE 9413510). Funding for this work was pro-
vided by grants from NSF (MCB9405394, N.S.) and NIH
(HL53306, N.S.), a Camille and Henry Dreyfus New
Faculty Award (N.S.), and a DOE/GAANN fellowship
(A.McC.).
Su p p or tin g In for m a tion Ava ila ble: Reaction scheme
and 1H NMR, DQF-COSY, and NOESY spectra of all com-
pounds (11 pages). This material is contained in libraries on
microfiche, immediately follows this article in the microfilm
version of the journal, and can be ordered from the ACS. See
any current masthead page for ordering information.
Activity Assa y w ith 1a a n d 1b. Recombinant Strepto-
myces cholesterol oxidase was purified as described from
(22) Smith, A. G.; Brooks, C. J . W. Biochem. J . 1977, 167, 121-
129.
J O9706537