Oxidation of Cholesterol
J. Am. Chem. Soc., Vol. 119, No. 5, 1997 857
subsequent purification steps were performed at 4 °C. Cell debris was
removed by centrifugation at 135 000g for 30 min. The supernatant
was loaded onto a column of DEAE-cellulose (DE-52, Whatman)
preequilibrated with buffer A and eluted with buffer A. Fractions
containing cholesterol oxidase were concentrated by (NH
cipitation, the pellet was redissolved in buffer A, and (NH
4
)
2
SO
SO
4
pre-
was
4
)
2
4
added to a final concentration of 1.5 M. The protein was further
purified on a butyl-Sepharose column (Pharmacia) that had been
equilibrated with buffer C and eluted with a linear gradient (0-100%)
of buffer A. Fractions were collected and analyzed by SDS-PAGE.
Fractions containing pure oxidase were pooled, concentrated, and
desalted by ultrafiltration into buffer A. Typically, 20-30 mg of pure
cholesterol oxidase was obtained per liter of culture. Protein concentra-
-
1
tions were determined by UV absorbance using ꢀ280 ) 81 924 M
-
1
cm (calculated from the molar absorptivities of tryptophan and
2
3
tyrosine ).
Purification of E361Q Cholesterol Oxidase. Cell paste of E. coli
BL21(DE3)plysS(pCO219) was purified as described above for pCO117,
the wild-type cholesterol oxidase. Typically, 20-30 mg of pure E361Q
cholesterol oxidase was obtained per liter of culture.
UV and CD Spectra of Cholesterol Oxidase. A solution of
cholesterol oxidase was prepared in buffer A. A baseline spectrum of
buffer A was subtracted from the sample spectrum. The concentration
of cholesterol oxidase was 13-19 µM for UV spectra and CD spectra
in the near UV. For CD spectra in the far UV, the protein concentration
was 54-58 µM.
Figure 1. A typical HPLC chromatogram of E361Q reaction mixture
after 3 h. HPLC conditions are given in Experimental Procedures: 3R,
cholest-4-ene-6R-hydroperoxy-3-one, k′ ) 11.5; 5R, cholest-4-ene-6R-
ol-3-one, k′ ) 11.8; 3â, cholest-4-ene-6â-hydroperoxy-3-one, k′ ) 14.3;
4
, cholest-4-ene-3,6-dione, k′ ) 15.7; 5â, cholest-4-ene-6â-ol-3-one,
k′ ) 16.5; 2, cholest-4-ene-3-one, k′ ) 23.3; 1, cholest-5-ene-3-one, k′
23.8; cholesterol, k′ ) 25.7.
)
Specific Activity Assay of Cholesterol Oxidase. Cholesterol was
added as a propan-2-ol solution (4 mM) to buffer B prewarmed to
7 °C. The final assay mixture was 1.26% propan-2-ol, 50 µM
7
.75 (s, 1, OOH); k′ ) 14.3; λmax ) 240 nm. Cholest-4-ene-6r-
1
hydroperoxy-3-one (3r): H-NMR δ 4.60 (ddd, 1, J ) 12.6, 5.4, 1.8
Hz, 6-H), 6.09 (s, 1, 4-H), 8.05 (bs , 1, OOH); k′ ) 11.5; λmax ) 240
nm. Cholest-4-ene-6r-ol-3-one (5r): m/z (EI) ) 400; H-NMR δ 4.31
3
cholesterol. The rate of formation of cholest-4-ene-3-one, 2, was
followed at 240 nm at 37 °C. The slope of the first 10% of the reaction
was determined by linear regression and converted to µmol/min‚mg
1
(
m, 1, 6-H), 6.15 (s, 1, 4-H); k′ ) 11.8; λmax ) 240 nm. Cholest-4-
1
ene-6â-ol-3-one (5â): m/z (EI) ) 400; H-NMR δ 4.33 (dd, 1, J )
-
1
-1
11
using ꢀ240 ) 12 100 M cm for 2. The rate of formation of H
2 2
O
2
4
1
.7, 2.7 Hz, 6-H), 5.80 (s, 1, 4-H); k′ ) 16.5; λmax ) 240 nm. Cholest-
was determined using a horseradish peroxidase coupled assay. The
standard assay conditions were the same as the cholest-4-ene-3-one
assay with the addition of 1.13 mM phenol, 0.87 mM 4-aminoantipyrine
1
-ene-3,6-dione (4): m/z (EI) ) 398; H-NMR δ 2.66 (dd, 1, J )
5.5, 3.6, 2- or 7-H), 6.15 (s, 1, 4-H); k′ ) 15.7; λmax ) 251 nm.
Attempted Inhibition of 6-Oxygenation Activity. The specific
(
Aldrich, Milwaukee, WI), and 10 U of horseradish peroxidase (Sigma,
activities of E361Q and wild-type cholesterol oxidase were measured
in buffer B with cholesterol (50 µM, 1.6% propan-2-ol) in the presence
of NaCN (1 mM), dithiothreitol (1 mM), or tris(carboethoxy)phosphine
St. Louis, MO). The formation of quinonimine at 510 nm was followed
as a function of time. The slope of the first 10% of the reaction was
-
1
determined and converted to µmol/min‚mg using ꢀ510 ) 5780 M
(3 mM) by following the increase in absorbance at 240 nm at 37 °C.
-1
min . This ꢀ was calculated by calibration of solutions using the above
referenced ꢀ240 of 2.
The tris(carboethoxy)phosphine reaction mixture was analyzed by
HPLC (Vide supra). Catalase (21 U) and E361Q cholesterol oxidase
The individual specific activities for the formation of 2, 3, 4, and 5
were measured under the same conditions as the A240 assay using [26-
C]cholesterol. The quantity of each product formed was analyzed
(16 µg) were added to a 1 mL solution of cholesterol (140 µM, 3.3%
propan-2-ol) in buffer B. The products formed were analyzed by HPLC
14
(Vide supra).
by HPLC (Vide infra).
Anaerobic Titration of E361Q Cholesterol Oxidase. A quartz
Time Course of Cholesterol Oxidase Reaction. Cholesterol or
cholest-5-ene-3-one, 1, was added as a propan-2-ol solution (3 mM)
to buffer B prewarmed to 37 °C, to a final concentration of 140 µM.
Reactions were initiated by the addition of cholesterol oxidase (6 nM
wild-type or 294 nM E361Q). Reaction mixtures were incubated under
foil to protect them from ambient light in a screw-capped vial with a
silicone seal at 37 °C. Aliquots (1 mL) were removed at various time
points and analyzed by HPLC (Vide infra).
HPLC Analysis. Samples were analyzed with a Model 680 gradient
controller, three M510 solvent pumps, and a Model 490 multiwave-
length detector (Waters Corp., Milford, MA) or a Model PDA-1
photodiode array detector (Rainin Instrument, Woburn, MA). The
following conditions were used: stationary phase, Microsorb-MV®
C-18 column (Rainin Instrument Corp., Woburn, MA, 5 µm, 10 Å,
24
cuvette, modified as described by Williams et al., was used for
anaerobic titrations. A 32 µM solution of E361Q cholesterol oxidase
was prepared in buffer B. Protein solutions were deoxygenated by
lyophilization and resuspended in degassed water under an Ar
atmosphere. Residual O
titration with cholesterol. The solution was judged to be O
titration of the FAD spectrum with 0.5 equiv of cholesterol showed a
0% reduction in absorption and remained constant, i.e., the FADH
formed was not reoxidized. This titration served to generate 0.5 equiv
of 1 in situ in the absence of O . Thus, oxidized E361Q cholesterol
oxidase was incubated with 1 in the absence of O . Flavin spectra
2
(<5 µM) was removed from the solution by
2
free when
5
2
2
2
were recorded every 20 min. After 1 h, the FAD was completely
reduced by the addition of a second 0.5 equiv of cholesterol. Thus,
reduced E361Q was incubated with 1. Spectra were recorded every
4
.6 × 250 mm); gradient elution at 1.25 mL/min; solvent A, CH
solvent B, propan-2-ol; solvent C, CH CN/H O (1:1, v:v); detection at
12 and 240 nm and by scintillation counting of fractions. A 25-min
3
CN;
2
2 2
0 min over 1 h. The FADH was reoxidized by O to FAD.
3
2
Initiation of 1 Oxidation by E361Q Cholesterol Oxidase.
A
2
solution of E361Q cholesterol oxidase (877 nM) in buffer B was
ultrafiltered (NMWCO ) 30 000) to remove cholesterol oxidase and
the filtrate added to a solution of 1 (50 µM, 1.2% propan-2-ol) in buffer
B. The rate of product formation was followed by UV assay at 240
nm. A portion of the filtrate was tested for residual cholesterol oxidase
activity by measuring the specific activity with cholesterol as substrate
in the absence of 1. In a second experiment, E361Q cholesterol oxidase
isocratic elution with 80% A and 20% C followed by 10-min linear
gradient to 85% A and 15% B followed by 25-min isocratic elution at
the same conditions yielded the separation shown in Figure 1. Samples
were injected directly from assay solutions. Product ratios were
determined by liquid scintillation counting of collected fractions and,
when necessary, integration of peak area as detected at 212 nm. Isolated
peaks were characterized as described below.
Product Characterization. Cholest-4-ene-6â-hydroperoxy-3-one
3â): H-NMR δ 4.42 (dd, 1, J ) 3.6, 2.4 Hz, 6-H), 5.87 (s, 1, 4-H),
(
24) Williams, C. H., Jr.; Arscott, L. D.; Matthews, R. G.; Thorpe, C.;
1
(
Wilkinson, K. D. Meth. Enzymol. 1979, 62, 185-197.
Sampson, Nicole S.
