Effects of the environment on microperoxidase-11 and on its catalytic activity in oxidation of organic sulfides to sulfoxides

Article abstract of DOI:10.1021/jo026344k

Microperoxidase-11 (MP-11, also known as heme undecapeptide of cytochrome c) was immobilized by encapsulation into sol-gel silica glass and by physisorption, chemisorption, and covalent attachment to silica gel. We then compared these species with one ano

Full text of DOI:10.1021/jo026344k

Effects of th e En vir on m en t on Micr op er oxid a se-11 a n d on Its
Ca ta lytic Activity in Oxid a tion of Or ga n ic Su lfid es to Su lfoxid es
Ekaterina N. Kadnikova† and Nenad M. Kostic´*
Department of Chemistry, Iowa State University, Ames, Iowa 50011-3111
nenad@iastate.edu
Received August 22, 2002
Microperoxidase-11 (MP-11, also known as heme undecapeptide of cytochrome c) was immobilized
by encapsulation into sol-gel silica glass and by physisorption, chemisorption, and covalent
attachment to silica gel. We then compared these species with one another and with dissolved
microperoxidase-11 as catalysts for the sulfoxidation of methyl phenyl sulfide by hydrogen peroxide.
MP-11 is prone to oligomerization in solution, both via axial ligation and via intermolecular
interactions. When the ligation oligomerization is suppressed upon immobilization, heme becomes
more accessible, and the sulfoxide yield increases 4-6 times, from 15% up to 95%. When the ligation
oligomerization of dissolved MP-11 is suppressed by protonation and acetylation of amino groups
and by addition of methanol, sodium dodecyl sulfate (SDS), or trifluoroethanol, the sulfoxide yield
increases 3-5 times (up to 76%). The oligomerization via intermolecular interactions is important
for preserving enantioselectivity in immobilized and dissolved MP-11. For MP-11 in amine-rich
and especially alcohol-rich environments, the enantioselectivity is vanishingly low, presumably
because amino and hydroxyl groups cause a conformation change in the catalyst. In other
environments, the MP-11 species are aggregated via intermolecular interactions in micellar (SDS)
solution and on the surface of the silica gel, or via axial ligation in aqueous buffer at pH 6.0. Under
these conditions, the enantioselectivity is enhanced; the enantiomeric excess (ee) becomes as high
as 46%. An understanding of the effects of the aggregation state and consequent properties on the
catalytic activity of MP-11 allowed us to control the yield and enantioselectivity of sulfoxidation
reaction.
In tr od u ction
Microperoxidases are heme peptides obtained by pro-
teolytic digestion of cytochrome c.1-3 Those designated
MP-8, MP-9, and MP-11 (Figure 1) contain the protein
segments 14-21, 14-22, and 11-21, respectively. The
heme is attached to the peptide via thioether linkages
involving the side chains of Cys 14 and Cys 17. The
imidazole group in the side chain of His 18 is one of the
axial ligands; the other one can be water or an exogenous
ligand such as cyanide, halide, or an amine.
Microperoxidases display peroxidase activity,1-5 that
is, they catalytically reduce hydrogen peroxide to water
while oxidizing a substrate to its radical cation.6-12 They
† Current address: Department of Chemistry, University of Cali-
fornia-Berkeley, Berkeley, CA 94720-1460.
(1) Adams, P. A. In Peroxidases in Chemistry and Biology; CRC
Press: Boca Raton, FL, 1991; Vol. 2, pp 171-200.
(2) Adams, P. A.; Baldwin, D. A.; Marques, H. M. In Cytochrome c:
A Multidisciplinary Approach; Scott, R. A., Mauk, A. G., Eds.;
University Science Books: Sausalito, CA, 1996; pp 632-692.
(3) Marques, H. M.; Shongwe, M. S.; Munro, O. Q.; Egan, T. J . S.
Afr. J . Chem. 1997, 50, 166-180.
F IGURE 1. Structure of monomeric microperoxidase-11.
also display oxygenase activity, akin to cytochrome P-450,
in catalyzing sulfide oxidation, amine N-demethylation,
(4) Akaza, I.; Aota, N. Talanta 1990, 37, 925-929.
(5) Wang, J .-S.; Baek, H. K.; van Wart, H. E. Biochem. Biophys. Res.
Comm. 1991, 179, 1320-1324.
(6) Dunford, H. B.; Stillman, J . S. Coord. Chem. Rev. 1976, 19, 187-
251.
(10) English, A. M.; Tsaprailis, G. Adv. Inorg. Chem. 1995, 43, 79-
125.
(7) Ortiz de Montellano, P. R. Acc. Chem. Res. 1987, 20, 289-294.
(8) Ortiz de Montellano, P. R.; Choe, Y. S.; DePhillis, G.; Catalano,
C. E. J . Biol. Chem. 1987, 262, 11641-11646.
(11) Smith, A. T.; Veitch, N. C. Curr. Opin. Chem. Biol. 1998, 2,
269-278.
(12) Moore, A. N. J .; Katz, E.; Willner, I. J . Electroanal. Chem. 1996,
417, 189-192.
(9) Dawson, J . H. Science 1988, 240, 433-439.
10.1021/jo026344k CCC: $25.00 © 2003 American Chemical Society
Published on Web 02/28/2003
2600
J . Org. Chem. 2003, 68, 2600-2608

Catalytic Activity of Microperoxidase-11
olefin oxidation, and aniline p-hydroxylation.2,13-16 The
ferryl form of the heme, created by dioxygen, H2O2, or
other oxygen-atom donors, can catalytically transfer an
oxygen atom to the substrate.7,9,17-20
an indicator of the state of the catalyst. Because the
sulfoxide yield depends on the efficiency of the oxygen-
atom transfer, and the degree of asymmetric induction
depends on the chirality of the catalyst, we can infer the
state of the MP-11 from its catalytic activity in various
media.
Proteins and other biomolecules can be immobilized
in various matrices by chemisorption, physisorption,
covalent bonding, encapsulation, cross-linking, or com-
binations of these methods.21 Much attention has recently
been devoted to the development of the sol-gel methods
for encapsulation of biomolecules.22-25 Achievements in
these studies opened many possibilities for basic and
applied research in materials science, bioanalytical chem-
istry, biocatalysis, biotechnology, and environmental
technology.24,25 In our laboratory, proteins cytochrome
c,26,27 carbonic anhydrase,28 horseradish peroxidase,29 and
lipase30 were encapsulated into sol-gel glass. Research
in our and other21,22 laboratories has shown, however,
that enzymes and even small molecules may behave
differently when free in solutions and when immobilized.
More basic research is needed before supported catalysts,
biosensors, and other composite materials can be turned
into practical chemical devices.
Spectra of MP-11 and its reactivity have been studied,
but the structure and intermolecular interactions have
not been correlated with effectiveness of MP-11 as a
catalyst for oxidation of organic substrates. In this work,
we compare catalytic properties of free (dissolved) and
variously immobilized microperoxidase-11. We chose a
simple reaction, oxidation of methyl phenyl sulfide (thio-
anisole) to sulfoxide, shown in eq 1.31-33 Our goal was
not to optimize this particular reaction, but to use it as
Exp er im en ta l P r oced u r es
Ch em icals. Reagents were obtained from commercial sources
and used without further purification. The silica gel had an
average particle size of 40 µm in the range 32-63 µm and a
surface area of 550 m2/g. Distilled water was demineralized
to an electrical resistivity greater than 17 MΩ cm.
In str u m en ts. Ultraviolet-visible (UV-vis) spectra were
recorded with a Perkin-Elmer Lambda 18 spectrophotometer.
1H nuclear magnetic resonance (NMR) spectra were recorded
at 400 MHz; the integration error was 5-10%. X-band electron
paramagnetic resonance (EPR) spectra were recorded at 120
K. Fourier transform infrared (FTIR) spectra of Nujol mulls
and diffuse-reflectance infrared Fourier transform (DRIFT)
spectra were recorded with an IR spectrometer equipped with
a TGS detector in the main compartment and a MCT detector
in the auxiliary module (AEM), which housed a Harrick
diffuse-reflectance accessory. The DRIFTS samples were held
in the Harrick microsampling cup. Diffuse-reflectance UV-
vis spectra were recorded with a spectrophotometer fitted with
a 60 mm integrating sphere accessory. Each sample was
spooned into a powder holder and analyzed in a % reflectance
mode. Circular dichroism (CD) spectra were recorded using
0.1 mm quartz cuvette. Gas chromatography-mass spectrom-
etry (GC-MS) experiments were done with a triple-quadrupole
mass spectrometer attached to a gas chromatograph. The
system was configured in the electron ionization mode. The
first quadrupole was used as a mass analyzer to scan the m/z
values from 35 to 650 at a rate of 1.2 scan/s. The second and
third quadrupoles were kept in the RF-only mode. Unit mass
resolution was achieved using perfluorokerosene as the cali-
bration and tuning reference. The GC starting temperature
was kept at 110 °C for 1 min and then raised to 170 °C at a
rate of 8 °C/min. A Chiraldex G-PN GC column (30 m × 0.32
mm) was used to separate the sulfoxide enantiomers. Lyo-
philization (freeze-drying) was done at -70 °C.
En ca p su la tion of Micr op er oxid a se-11 in Sol-Gel Silica
Gla ss. A modification of a published method34 was used.
Aqueous MP-11 (5.80 mM, 1.00 mL), aqueous sodium fluoride
(1.00 M, 0.200 mL), and aqueous poly(vinyl alcohol) (4% w/w,
0.400 mL) were combined with 0.128 mL of water. While the
mixture was vigorously stirred with a vortex shaker, tetra-
methyl orthosilicate (1.827 g, 12 mmol) was added. After ca.
10 s, when the mixture turned into a clear homogeneous
solution and warmed up, it was placed into an ice bath. The
gelation occurred ca. 10 s later, and the mixture was kept in
the ice bath for 5 min. The sealed reaction vessel was kept at
rt for 24 h and then opened. The gel was air-dried at 40 °C for
96 h. The resulting glass was ground in a mortar and shaken
with 10.0 mL of water at rt for 2 h. The solid was filtered off,
washed with water, acetone, and hexane, and dried for 12 h
at 40 °C. The final product was a powder of a light reddish-
brown color. Elemental analysis found 3.81 C, 1.51 H, and 0.10
N; the first and the last result correspond to 3.39 µmol of MP-
11 per gram of solid.
(13) Mashino, T.; Nakamura, S.; Hirobe, M. Tetrahedron Lett. 1990,
31, 3163-3166.
(14) Nakamura, S.; Mashino, T.; Hirobe, M. Tetrahedron Lett. 1992,
33, 5409-5412.
(15) Nakamura, S.; Mashino, T.; Hirobe, M. In Fifth International
Symposium on the Activation of Dioxygen and Homogeneous Catalytic
Oxidation, College Station, Texas, 1993; Barton, D. H. R., Martell, A.
E., Swyer, D. T., Eds.; Plenum: New York, 1993; p 472.
(16) Osman, A. M.; Koerts, J .; Boersma, M. G.; Boeren, S.; Veeger,
C.; Rietjens, I. M. C. M. Eur. J . Biochem. 1996, 240, 232-238.
(17) Ortiz de Montellano, P. R. Cytochrome P450: Structure, Mech-
anism, and Biochemistry, 2nd ed.; Plenum Press: New York, 1995.
(18) Lewis, D. F. V. Cytochromes P450: Structure, Function and
Mechanism; Taylor & Francis: London, 1996.
(19) Waterman, M. R.; J ohnson, E. E. Cytochrome P450; Academic
Press, Inc.: San Diego, 1996; Vols. 206 and 272.
(20) Sono, M.; Roach, M. P.; Coulter, E. D.; Dawson, J . H. Chem.
Rev. 1996, 96, 2841-2887.
(21) Bickerstaff, G. F. In Immobilization of Enzymes and Cells, 1st
ed.; Bickerstaff, G. F., Ed.; Humana Press: Totowa, NJ , 1997; Vol. 1,
pp 1-11.
(22) Avnir, D.; Braun, S.; Lev, O.; Ottolenghi, M. Chem. Mater. 1994,
6, 1605-1614.
(23) Avnir, D. Acc. Chem. Res. 1995, 28, 328-334.
(24) Livage, J . C. R. Acad. Sci., Ser. II: Mec., Phys., Astron. 1996,
322, 417-422.
(25) Dunn, B.; Zink, J . I. Chem. Mater. 1997, 9, 2280-2291.
(26) Shen, C.; Kostic´, N. M. J . Electroanal. Chem. 1997, 438, 61-
65.
(27) Shen, C.; Kostic´, N. M. J . Am. Chem. Soc. 1997, 119, 1304-
1312.
(28) Badjic´, J . D.; Kostic´, N. M. Chem. Mater. 1999, 11, 3671-3679.
(29) (a) Kadnikova, E. N.; Kostic´, N. M. J . Non-Cryst. Solids 2001,
283, 63-68. (b) Kadnikova, E. N.; Kostic´, N. M. J . Mol. Catal. B:
Enzymol. 2002, 18, 39-48.
(30) Badjic´, J . D.; Kadnikova, E. N.; Kostic´, N. M. Org. Lett. 2001,
3, 2025-2028.
(31) Drabowicz, J .; Kielbasinski, P.; Mikolajczyk, M. In The Syn-
theses of Sulphones, Sulphoxides and Cyclic Sulphides; Patai, S.,
Rappoport, Z., Eds.; Wiley: New York, 1994; pp 109-254.
(32) Mata, E. G. Phosphorus, Sulfur Silicon Relat. Elem. 1996, 117,
231-286.
(34) Reetz, M. T.; Zonta, A.; Simpelkamp, J . Biotech. Bioeng. 1996,
49, 527-534.
(33) Kagan, H. B.; Diter, P. Organosulfur Chem. 1998, 2, 1-39.
J . Org. Chem, Vol. 68, No. 7, 2003 2601

Kadnikova and Kostic´
P r ep a r a t ion of Am in op r op yl-F u n ct ion a lized Silica
Gel. Silica gel (10.0 g) was activated35,36 by refluxing with
100.0 mL of concentrated hydrochloric acid, washed with
deionized water, and dried for 6 h at 190 °C. Elemental
analysis found 0.00 C, 0.77 H, and 0.00 N. A suspension of
activated silica gel in 90.0 mL of a 10% (v/v) solution of
3-aminopropyltriethoxysilane in dry toluene was refluxed for
2 h under a nitrogen atmosphere. Aminopropyl-functionalized
silica gel was filtered off, washed with toluene and acetone,
and carefully dried for 8 h at 115 °C to afford 9.27 g of an
off-white solid. Elemental analysis found 5.24 C, 1.39 H, and
1.57 N; the first and the last result correspond to 1.12 mmol
of NH2 groups per gram of solid. Both IR and DRIFT spectra
of aminopropyl-functionalized silica gel exhibited bands at
3368 and 3298 cm-1, characteristic of an NH2 group.
Covalen t Attach m en t of Micr oper oxidase-11 to Am in o-
p r op yl-F u n ction a lized Silica Gel. To a solution of MP-11
(41.6 mg, 21.6 µmol) in 84.0 mL of a 0.050 M HEPES buffer
at pH 7.5 were added EDC (161.0 mg, 0.84 mmol) and
aminopropyl-functionalized silica gel (305.0 mg, 0.342 mmol
of NH2 groups). The reaction mixture was stirred at rt for 3.5
h. The modified silica gel was filtered off, washed with the
HEPES buffer and acetone, and air-dried for 30 min to afford
261.8 mg of a dark reddish-brown solid. Elemental analysis
found 6.86 C, 1.59 H, and 2.05 N; the first and the last result
correspond to 17.1 µmol of MP-11 per gram of solid.
magnesium sulfate, the solvent was removed by rotary evapo-
ration. Residue was dissolved in CD2Cl2, and the compounds
were identified by their GC retention times, mass spectra, and
1H NMR chemical shifts in comparison with authentic samples.
1
The yield of sulfoxide was measured by H NMR. We checked
the efficiency of the extraction step (i.e., mass balance) by
spiking the organic phase with a known amount of internal
standard (thioanisole) and comparing the 1H peak integrals
for thioanisole and corresponding sulfoxide before and after
spiking. Thus, we estimated the recovery of more than 97% of
organic material from the reaction mixture. The enantiomeric
excess of (S)-sulfoxide was measured by chiral GC-MS, as
described above, in the Instruments section. When the catalyst
was scarce, reactions were run at 1/10 of the stated scale, and
extraction was done directly with CD2Cl2. For recycling experi-
ments with suspended catalyst, the catalyst was filtered off,
washed with a small amount of solvent and resuspended in
the same solvent. For recycling experiments with dissolved
catalyst, the aqueous phase after methylene extraction was
quickly frozen until the next cycle. Amount of unreacted
hydrogen peroxide in the aqueous phase was estimated from
the sulfoxide yield, and sufficient hydrogen peroxide was added
in the subsequent cycle to achieve 1:1 initial ratio of thioanisole
to hydrogen peroxide in the reaction mixture.
P r ep a r a tive-Sca le Oxid a tion of Meth yl P h en yl Su l-
fid e. These experiments were done in duplicate, on a scale 10
times larger than that stated above, for both reaction and
workup. Sulfoxide was separated from unreacted sulfide by
column chromatography (hexanes/ethyl acetate, gradient from
4:1 to 0:1). After the removal of the solvent and drying in
vacuo, methyl phenyl sulfoxide was isolated as a yellowish oil.
For reaction in 50 mM sodium phosphate buffer at pH 6.0,
the sulfoxide yield was 7.2-9.1 mg (16 ( 2%). For reaction in
80:20 water/methanol, the sulfoxide yield was 18-21 mg (39
( 3%).
Assign m en t of Absolu te Con figu r a tion in th e P r e-
d om in a n t En a n tiom er of Meth yl P h en yl Su lfoxid e. An
equimolar mixture of racemic and (S)-methyl p-tolyl sulfoxide
dissolved in CD2Cl2 was analyzed by the GC-MS method. Upon
addition of 2.5 equiv of Kagan chiral shift reagent, the 1H NMR
spectrum of the mixture was taken. The gas chromatogram
showed that the retention time is longer for the (S)-enantiomer
than for the (R)-enantiomer. The 1H NMR spectrum showed
that the methyl signal of the (S)-enantiomer is shifted upfield
from that of the (R)-enantiomer.
In cu b a t ion of E n a n t iop u r e Su lfoxid e w it h Micr o-
p er oxid a se-11. (S)-Methyl p-tolyl sulfoxide (36 µmol) was
incubated with MP-11 (1.16 µmol) dissolved in 3.50 mL of
sodium phosphate buffer for 50 and 100 h. The reaction
mixture was worked up as usual, and the absolute configura-
tion of sulfoxide in both cases was found by chiral chromatog-
raphy to be S.
Ad sor p tion of Micr op er oxid a se-11 on to Silica Gel a n d
Am in op r op yl-F u n ction a lized Silica Gel. Silica gel (6.000
g) or its aminopropyl-functionalized derivative (6.417 g) was
added to a stirred solution of MP-11 (22.33 mg, 11.6 µmol for
normal “loading”; 197.84 mg, 10.28 µmol for high “loading”)
in 15.0 mL of water or, for normal loading only, in 80% (v/v)
methanol in water. The resulting suspension was lyophilized
for 36 h to afford a uniformly colored reddish-brown solid. The
normal and high loadings correspond to 1.93 and 17.1 µmol of
MP-11 per gram of solid, respectively.
N-Acetyla tion of Micr op er oxid a se-11.37 Solution of MP-
11 (50.0 mg, 26.0 µmol) in 130.0 mL of deionized water was
combined with 130.0 mL of a saturated sodium acetate
solution, and the resulting mixture was cooled to 0 °C. Acetic
anhydride (16.25 mL, 172.2 mmol) was added in five portions
at 10 min intervals. The reaction mixture was kept at 4 °C
for 15 h and lyophilized for 70 h. Residue from the first
lyophilization was dissolved in 100.0 mL of a 0.100 M aqueous
sodium hydroxide and subjected to flash chromatography on
Sephadex-10 with 0.100 M sodium hydroxide as an eluent. The
eluate was lyophilized for 72 h. Residue from the second
lyophilization was washed with four 50 mL portions of water
in an Amicon 8050 ultrafiltration cell fitted with a YC05
membrane, which has a cutoff of 500 D. Final lyophilization
for 100 h yielded 37.9 mg (73%) of N-acetylated microperoxi-
dase-11. Concentration of its aqueous solutions was deter-
mined spectrophotometrically37 using ꢀ550 ) 31.2 mM-1 cm-1
.
Gen er a l P r oced u r e for Oxid a tion of Meth yl P h en yl
Su lfid e. All experiments were done in triplicate. Micro-
peroxidase-11 in various forms (1.16 µmol) was dissolved or
suspended in 3.50 mL of a solvent (see footnotes to Table 2
for details). To the stirred solution or suspension of the catalyst
was added 36 µmol (4.27 µL) of thioanisole (methyl phenyl
sulfide) neat or dissolved in dichloromethane. Hydrogen
peroxide (240 µL of a 0.150 M aqueous solution, 36 µmol) was
added in six portions, at 6 min intervals. The reaction was
stopped 1.0 h after the first addition of hydrogen peroxide by
extracting the reaction mixture with three 1.0 mL portions of
dichloromethane; for reactions in alcoholic solutions, the
extraction solvent was a 1:1 (v/v) mixture of dichloromethane
and hexane. After the organic phase was dried over anhydrous
Resu lts
We present our findings in this section and interpret
them in the next one.
P r ep a r a tion a n d Ch a r a cter iza tion of Ma ter ia ls.
Immobilization of MP-11 on silica support was achieved
by four methods: encapsulation into sol-gel glass,34
covalent attachment by EDC to aminopropyl-functional-
ized silica gel,35,36,38,39 and adsorption by lyophilization
onto silica gel and onto aminopropyl-functionalized silica
gel. In the first two cases, loading of the catalyst was
determined by the elemental analysis. All of the methods
of immobilization afforded lightly to darkly colored red-
(35) Butterworth, A. J .; Clark, J . H.; Walton, P. H.; Barlow, S. J .
Chem. Commun. 1996, 1859-1860.
(36) Emne´us, J .; Gorton, L. Anal. Chim. Acta 1993, 276, 303-318.
(37) Carraway, A. D.; McCollum, M. G.; Peterson, J . Inorg. Chem.
1996, 35, 6885-6891.
(38) Lo¨etzbeyer, T.; Schuhmann, W.; Katz, E.; Falter, J .; Schmidt,
H.-L. J . Electroanal. Chem. 1994, 377, 291-294.
(39) Heleg-Shabtai, V.; Katz, E.; Willner, I. J . Am. Chem. Soc. 1997,
119, 8121-8122.
2602 J . Org. Chem., Vol. 68, No. 7, 2003

Catalytic Activity of Microperoxidase-11
F IGURE 3. Desorption by washing with phosphate buffers
of microperoxidase-11 (MP-11) immobilized in four different
ways. The catalyst encapsulated in sol-gel silica or covalently
attached to aminopropyl-functionalized silica gel remained on
the support even after prolonged washing (a). The catalyst
physisorbed on silica gel (c and e) washed off more easily than
that chemisorbed on aminopropyl-functionalized silica gel (b
and d). Buffer at pH 10.0 (d and e) is more effective than the
buffer at pH 6.0 (b and c) in MP-11 removal.
F IGURE 2. Diffuse-reflectance spectra of variously im-
mobilized microperoxidase-11 (MP-11). This catalyst MP-11
was immobilized on silica supports by encapsulation in sol-
gel silica glass (a) and by physisorption on silica gel from
aqueous (b) and alcoholic (c) solutions. The catalyst was also
immobilized on aminopropyl-functionalized silica supports by
covalent attachment (d) and by chemisorption from aqueous
solution at normal (e) and high (f) loadings. Characteristic
Soret (ca. 400 nm) and â (ca. 525 nm) bands are prominent in
all diffuse-reflectance spectra.
dish-brown solids. The EPR spectra of the immobilized
MP-11 showed a signal with g ≈ 6.0, which is charac-
teristic of high-spin iron(III) porphyrins,2,40 but no signals
with g ≈ 3.2 and g ≈ 2.1, diagnostic of low-spin iron(III)
porphyrin with imidazole and a primary amine as axial
ligands,2,40 see Figure SI 1 in Supporting Information.
Figure 2 shows diffuse-reflectance UV-vis spectra of
variously immobilized MP-11, with their characteristic
Soret (ca. 400 nm) and â (ca. 525 nm) bands. The
absorbance of encapsulated MP-11 (spectrum a) is lower
than that of surface-immobilized MP-11 (spectra b-f).
The efficiency of immobilization was tested in desorp-
tion experiments. Samples of immobilized MP-11 con-
taining the same amount of the catalyst were washed
by portions of sodium phosphate buffers at pH 6.0 and
pH 10.0. As Figure 3 shows, both encapsulated and
covalently attached MP-11 were retained on the support
even after prolonged washing, whereas chemisorbed and
physisorbed MP-11 were gradually removed. Both the
nature of the support and the pH of the buffer are
important; desorption occurred more easily from silica
gel than from aminopropyl-functionalized silica gel, and
more easily at pH 10.0 than at pH 6.0.
Cir cu la r Dich r oism Sp ectr a of Dissolved Micr o-
p er oxid a se-11. To understand the conformation of the
catalyst at reaction conditions, we recorded circular
dichroism spectra of dissolved MP-11 (Figure 4). Only
MP-11 in micellar solution (f) exhibits a strong negative
peak around 220 nm, characteristic of R-helical peptides.
In all other environments (a-e, g), MP-11 exhibits only
a small shoulder around 220 nm but a strong signal
around 203 nm, characteristic of randomly-coiled pep-
tides.
F IGURE 4. Circular dichroism (CD) spectra of dissolved
microperoxidase-11 (MP-11) and N-acetylated microperoxi-
dase-11 (N-Ac-MP-11) at the concentration of 331.4 µM used
in sulfoxidation reaction. (a) MP-11 and (b) N-Ac-MP-11 in pH
6.0 buffer; (c) MP-11 and (d) N-Ac-MP-11 in pH 3.0 buffer; (e)
MP-11 in methanol solution; (f) MP-11 in a micellar solution;
(g) MP-11 in trifluoroethanol solution. For solvent details, see
Table 2 footnotes.
with all solvents and all (blank) supports. In all solvents,
at pH 6.0, the sulfoxide yield was <1.0%; at pH 3.0, the
yield was 4 ( 1%, as a result of general acid cataly-
sis.31,41,42 The sulfoxide yield was 4 ( 1% in the presence
of undoped sol-gel silica glass, 26 ( 5% in the presence
of chromatographic quality silica gel, and 2 ( 1% when
silica gel was modified with aminopropyl groups. This
catalytic effect of silica on sulfoxidation reaction was
recently thoroughly studied by us;29 this effect should be
taken into account when investigating and designing
immobilized catalysts.
Com p a r ison of H or ser a d ish P er oxid a se a n d
Micr op er oxid a se-11 a s Su lfoxid a tion Ca ta lysts. We
Th e “Ba ck gr ou n d ” Su lfoxid a tion Rea ction . The
oxidation of thioanisole by a 2-fold molar excess of
hydrogen peroxide in the absence of MP-11 was studied
(41) Bach, R. D.; Su, M.-D.; Schlegel, H. B. J . Am. Chem. Soc. 1994,
116, 5379-5391.
(42) Adewuyi, Y. G.; Carmichael, G. R. Environ. Sci. Technol. 1986,
20, 1017-1022.
(40) Carraway, A. D.; Povlock, S. L.; Houston, M. L.; J ohnston, D.
S.; Peterson, J . J . Inorg. Biochem. 1995, 60, 267-276.
J . Org. Chem, Vol. 68, No. 7, 2003 2603

Kadnikova and Kostic´
TABLE 1. Su lfoxid a tion of Meth yl P h en yl Su lfid e by 1 Mola r Equ iv of Hyd r ogen P er oxid e Ca ta lyzed by
Micr op er oxid a se-11 (MP -11) in a 50 m M Sod iu m P h osp h a te Bu ffer a t p H 6.0, a t Am bien t Tem p er a tu r e, for 1 h
sulfoxide
yieldb
oligomerization
entry
catalyst
cyclea
eec
by axial ligationd
YESf
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
MP-11 in solution
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
15 ( 5e
12 ( 1
12 ( 3
55 ( 1
9 ( 2
36 ( 7
5 ( 3
4 ( 1
9 ( 2
MP-11 encapsulated in sol-gel silica
nd
10 ( 2
9 ( 2
6 ( 1
MP-11 covalently attached to aminopropyl-
functionalized silica gelg
92 ( 2
83 ( 2
56 ( 9
87 ( 1
95 ( 1
38 ( 4
95 ( 1
49 ( 7
46 ( 1
89 ( 2
92 ( 2
16 ( 2
93 ( 1
91 ( 1
10 ( 2
11 ( 1
11 ( 4
18 ( 2
30 ( 6
23 ( 5
28 ( 3
3 ( 1
NOh
NOh
nd
MP-11 physisorbed onto silica gel from
aqueous solutioni
MP-11 physisorbed onto silica gel from
alcoholic solutioni
3 ( 2
4 ( 6
MP-11 chemisorbed onto aminopropyl-
functionalized silica geli
17 ( 6
12 ( 5
20 ( 5
29 ( 3
11 ( 3
5 ( 1
NOh
nd
MP-11 chemisorbed onto aminopropyl-
functionalized silica gel at high loadingg
a
b
Consecutive reuse of the catalyst in three cycles; see Experimental Section for details. First cycle is in bold type. Determined by 1H
NMR spectroscopy; see Table 1 footnote e for isolated yield. c Determined by a GC-MS method. The predominant enantiomer in all cases
was (S)-PhS(O)Me. Inferred from the EPR spectra; see Figure SI 1 in Supporting Information. YES means that the g-values are consistent
d
with a low-spin iron(III) heme coordinated by histidine and a primary amine; NO means that the g-values are consistent with a high-spin
iron(III) heme coordinated by histidine and water; nd means not determined. e Isolated yield 16 ( 2%. f From ref 2: gx ) 1.46, gy ) 2.19,
gz ) 3.18. High loading for covalently attached and chemisorbed MP-11 is 17.1 µmol per gram of solid. geff ≈ 6.0. Usual loading for
physi- and chemisorbed MP-11 is 1.93 µmol per gram of solid.
g
h
i
compared the activity and enantioselectivity of these two
catalysts for the reaction in eq 1. With both catalysts,
the maximum chemical conversion is reached after 1 h.
In all experiments, the (S)-enantiomer of PhS(O)Me
predominated. Whereas the enzyme horseradish peroxi-
dase afforded 5 ( 1% of PhS(O)Me with an enantiomeric
excess of 91 ( 3%,29 the catalyst MP-11 afforded a
somewhat higher yield but a lower enantioselectivity
(Table 1, entry 1).
When the same MP-11 catalyst was used repeatedly
(entries 1-3), the yields remained constant (and low), but
the enantioselectivity vanished. Concomitant bleaching
of the Soret band was a symptom of the oxidative
degradation of the heme.10,37,43-47 Because MP-11 achieved
better yields than horseradish peroxidase, we chose it for
further studies of the various means of immobilization
and their effects on the catalytic activity.
Reten tion of Su lfoxid e Con figu r a tion . Incubation
of enantiopure (S)-TolS(O)Me with MP-11 under sulfoxi-
dation reaction conditions, but in the absence of hydrogen
peroxide, yielded only enantiopure (S)-TolS(O)Me, even
after 100 h. Thus, MP-11 does not racemize chiral
sulfoxide, once the latter is formed.
Su lfoxid a tion Ca ta lyzed by Im m obilized Micr o-
p er oxid a se-11. Variously immobilized MP-11 turned out
to be more efficient as a catalyst than free MP-11 in
solution. The yield of sulfoxide increased ca. 3-fold when
MP-11 was encapsulated into sol-gel silica (entry 4 in
Table 1). Covalently attached, physisorbed, and chemi-
sorbed MP-11 (entries 7, 10, 16, and 19) proved to be even
more effective, yielding ca. 6 times more sulfoxide than
free MP-11 did. The enantioselectivity induced by the
immobilized catalyst was lower than (entries 4, 7, and
16) or similar to (entries 10 and 19) that induced by the
free catalyst. The catalytic effect of the matrix, which we
reported for encapsulated horseradish peroxidase,29 did
not noticeably lower the enantioselectivity of MP-11
physisorbed onto silica gel from aqueous solution, because
the smaller size of MP-11 and higher loadings on the
surface of the silica gel resulted in a better surface
coverage.
With the exception of the encapsulated species (entries
4-6), immobilized MP-11 fared well in the recycling
experiments (entries 7-18): yields in the first and second
cycles were almost always high. The enantioselectivity
generally stayed constant in the successive catalyst cycles
for most of the supporting matrices (entries 4-12 and
16-18).
To explain the increase in the reactivity of MP-11 upon
immobilization, we hypothesized that this catalyst on
various supports is aggregated less than in solution or
not at all. To verify this hypothesis, we prepared mono-
meric MP-11 in solution and studied its catalytic ef-
fectiveness.
(43) Clore, G. M.; Hollaway, M. R.; Orengo, C.; Peterson, J .; Wilson,
M. T. Inorg. Chim. Acta 1981, 56, 143-148.
(44) Spector, A.; Ma, W.; Wang, R.-R.; Kleiman, N. J . Exp. Eye Res.
1997, 65, 457-470.
Su lfoxid a t ion Ca t a lyzed b y Mon om er ic Micr o-
p er oxid a se-11 in Solu tion . There are several ways to
preclude oligomerization of MP-11:2 by separating the
molecules,40,48-50 changing the solvent,5,40,49-51 or prevent-
ing the amino groups in one MP-11 molecule from
coordinating to the iron atom in another.2,3,5,37,52-57 Re-
(45) Cantoni, L.; Gibbs, A. H.; Matteis, F. D. Int. J . Biochem. 1981,
13, 823-830.
(46) Schaefer, W. H.; Harris, T. M.; Guengerich, F. P. Biochem. 1985,
24, 3254-3263.
(47) Korzekwa, K.; Osawa, Y. Proc. Natl. Acad. Sci. U.S.A. 1991,
88, 7081-7085.
2604 J . Org. Chem., Vol. 68, No. 7, 2003

Catalytic Activity of Microperoxidase-11
TABLE 2. Su lfoxid a tion of Th ioa n isole by 1 Mola r Equ iv of Hyd r ogen P er oxid e Ca ta lyzed by Dissolved
Micr op er oxid a se-11 (MP -11) a t Am bien t Tem p er a tu r e for 1 h a t p H 6.0, Un less Oth er w ise Noted
sulfoxide
yieldb
oligomerization
entry
catalyst
MP-11 at pH 6.0
cyclea
eec
by axial ligationd
YESf
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
1
2
3
1
2
3
1
1
1
2
3
1
2
3
1
2
3
15 ( 5e
12 ( 1
12 ( 3
56 ( 3
41 ( 2
27 ( 1
48 ( 4
58 ( 3
41 ( 8k
33 ( 3
37 ( 4
76 ( 3
38 ( 1
9 ( 3
36 ( 7
5 ( 3
4 ( 0
25 ( 3
10 ( 1
4 ( 3
N-acetylated MP-11 at pH 6.0
NOg
N-protonated MP-11 at pH 3.0h
N-acetylated MP-11 at pH 3.0h
MP-11 in alcoholic solutionj
12 ( 2
6 ( 3
6 ( 1
1 ( 1
NOi
nd
NOl
0 ( 1
MP-11 in micellar solutionm
MP-11 in TFE solutionn
34 ( 5
35 ( 3
46 ( 5
3 ( 3
6 ( 3
NOl
nd
71 ( 5
38 ( 1
9 ( 3
4 ( 1
a
b
Consecutive reuse of the catalyst; see Experimental Section for details. First (or the only) cycle is in bold type. Determined by 1H
NMR spectroscopy; see Table 2 footnotes e and k for isolated yields. c Determined by a GC-MS method. The predominant enantiomer in
all cases was (S)-PhS(O)Me. Inferred from the EPR spectra; see Figure SI 1 in Supporting Information. YES means that the g-values
d
are consistent with a low-spin iron(III) heme coordinated by histidine and a primary amine; NO means that the g-values are consistent
with a high-spin iron(III) heme coordinated by histidine and water; nd means not determined. e Isolated yield 16 ( 2%. f From ref 2: gx
g
h
) 1.46, gy ) 2.19, gz ) 3.18. From ref 37: gx ) gy ) 5.9, gz ) 2.0. Potassium hydrogen phthalate (50 mM) was used as a buffer at pH
3.0. Because of the significant absorbance of phthalate in the UV region, sodium citrate (50 mM) was used as a pH 3.0 buffer for CD
measurements. From ref 2: geff ) 5.70, g| ) 2.00. Methanol/water (80:20, v/v). Isolated yield 39 ( 3%. From ref 40: g ≈ 6.0. SDS
eff
i
j
k
l
m
n
(5%, w/w) in 0.1 M Me4NBr. TFE/buffer (50/50, v/v).
sults obtained by all of these methods are shown in Table
2.
is ca. 5 times higher than when these molecules are
allowed to aggregate (entry 1). In the subsequent cycles
(entries 13 and 14), the yield drops, but the enantio-
selectivity remains approximately the same. A short-
coming of this method is that the micellar components
(SDS and Me4NBr) are partially leached into the organic
phase during extraction, as 1H NMR spectra of the
organic phase showed.
We checked the importance of maintaining the second-
ary structure of microperoxidase-11. A solution of 2,2,2-
trifluoroethanol (TFE) stabilizes the R-helices in peptides
and proteins.58,59 Yields and their recycling trend for MP-
11 in TFE-containing solutions (entries 15-17) are the
same as for MP-11 in micellar solution (entries 12-14),
but the enantioselectivity is as low as that obtained in
methanol solution (entries 9-11).
When primary amino groups of MP-11 are acetylated
(entry 4) or protonated (entry 7), the yield of sulfoxide
increases 2-3 times over that achieved with oligomeric
MP-1151 (entry 1). Acetylation does not markedly affect
the enantioselectivity (entries 4 vs 1); protonation lowers
it (entries 7 vs 1 and 8 vs 4). In successive cycles, the
yield gradually decreases, while the enantioselectivity
vanishes (entries 4-6).
When MP-11 is made monomeric in alcoholic solution
(entry 9), the yield of sulfoxide again increases 2-3 times
in comparison with oligomeric MP-11 (entry 1). Enan-
tiomeric excesses are very low in the first cycles and
nonexistent in subsequent ones, but the yield stays
approximately constant in successive cycles (entries
9-11).
When molecules of MP-11 are separated by entrapment
Discu ssion
in the detergent micelles (entry 12), the yield of sulfoxide
Ad sor p tion of Micr op er oxid a se-11 on Silica a n d
Mod ified Silica Su r fa ces. The strength of MP-11
adsorption depends on the nature of the support and the
pH of the buffer. The overall charge, estimated on the
basis of the known pKa values for conversions between
different states of MP-11,2 is -2 at pH 6.0 and -3 at pH
10.0. At both pH values, the surface of silica gel is
negatively charged owing to deprotonation of the hy-
droxyl groups. The surface of aminopropyl-functionalized
silica gel is positively charged at pH 6.0 and approxi-
mately neutral at pH 10.0. The greater retention at pH
6.0 and on the aminopropyl-modified surface, evident in
Figure 2, is caused by electrostatic interactions. The
(48) Mazumdar, S.; Medhi, O. K.; Mitra, S. Inorg. Chem. 1991, 30,
700-705.
(49) Othman, S.; Lirzin, A. L.; Desbois, A. Biochemistry 1993, 32,
9781-9791.
(50) Das, D. K.; Medhi, O. K. J . Chem. Soc., Dalton Trans. 1998,
1693-1698.
(51) Huang, Y.-P.; Kassner, R. J . J . Am. Chem. Soc. 1981, 103,
4927-4932.
(52) Wang, J . S.; van Wart, H. E. J . Phys. Chem. 1989, 93, 7925-
7931.
(53) Munro, O. Q.; Marques, H. M. Inorg. Chem. 1996, 35, 3752-
3767.
(54) Munro, O. Q.; Marques, H. M. Inorg. Chem. 1996, 35, 3768-
3779.
(55) Fan, B.; Fontenot, D. L.; Larsen, R. W.; Simpson, M. C.;
Shelnutt, J . A.; Falcon, R.; Martinez, L.; Niu, S.; Zhang, S.; Niemczyk,
T.; Ondrias, M. R. Inorg. Chem. 1997, 36, 3839-3846.
(56) Carraway, A. D.; Miller, G. T.; Pearce, L. L.; Peterson, J . Inorg.
Chem. 1998, 37, 4654-4661.
(58) Benson, D. R.; Hart, B. R.; Zhu, X.; Doughty, M. B. J . Am. Chem.
Soc. 1995, 117, 8502-8510.
(59) Nelson, J . W.; Kallenbach, N. R. Proteins: Struct., Funct., Genet.
1986, 1, 211-217.
(57) Marques, H. M.; Perry, C. B. J . Inorg. Biochem. 1999, 75, 281-
291.
J . Org. Chem, Vol. 68, No. 7, 2003 2605

Kadnikova and Kostic´
repulsion between MP-11 and the silica gel surface is
stronger at higher pH, which facilitates the removal of
microperoxidase-11. The effect of pH is less pronounced
for aminopropyl-functionalized silica gel, since its surface
is neutral or even positively charged.
aggregated via axial ligation of primary amino groups
(at the N-terminus and in lysine side chains) in one
molecule of MP-11 to the iron atom in another. At
physiological pH, the major species are pentamers and
hexamers.2 The presence (spectrum a) or absence (spectra
b-g) of oligomerization, caused by axial ligation, is not
reflected in the positions and intensity of CD peaks.
As shown in Tables 1 and 2, MP-11 catalyst was tested
and recycled under a range of conditions. Comparisons
of the results for the first cycle under various conditions
allow us to judge the reactivity and selectivity of the
catalyst. Comparisons of the results for consecutive cycles
under the same conditions reveal the recyclability of the
catalyst.
Effect of Im m obiliza tion on th e Oxygen a se Activ-
ity of Micr op er oxid a se-11. The yields in Table 1 show
that the oxygen-atom transfer from the ferryl form of MP-
11 to the substrate becomes more efficient when MP-11
is immobilized. Breakup of the aforementioned oligomers
upon immobilization presumably makes more heme
groups available for catalysis, and the yield increases.
We tested this hypothesis in two ways. First, EPR
spectra of our immobilized MP-11 contain signals char-
acteristic of the monomer37,40 and lack signals charac-
teristic of the oligomers.2,40 Second, since several methods
for deoligomerization of microperoxidases in solution are
known, we applied them in this study of the reaction in
eq 1.
Protonation or acetylation prevents primary amino
groups (at the N-terminus and in lysine side chains) in
one molecule of MP-11 from coordinating to the iron atom
in another.2,3,5,37,52-57 The sulfoxidation yields obtained
upon blocking the amino groups in these two ways are
similar (Table 2, entries 4 and 7), but lower than the
yields obtained with some of the immobilized catalysts
(Table 1, entries 7, 10, 16, and 19).
Microperoxidases are deoligomerized in alcoholic
solution.5,40,49-51,65,66 With the consequent greater expo-
sure of the heme groups, the sulfoxide yield increases,
as entries 9 and 1 in Table 2 show.
Diffu se-Reflecta n ce UV-vis Sp ectr a of Im m obi-
lized Micr op er oxid a se-11. Since our materials are
opaque, colored powders, transmittance spectroscopy
could not be done. We chose the diffuse-reflectance
method to record at the UV-vis spectra of immobilized
MP-11. Since the process of diffuse reflectance is non-
linear,60 the bandwidth observed in such spectra depends
on the particle size and on the nature of the material.
Characteristic Soret (ca. 400 nm) and â (ca. 525 nm)
bands are present in all spectra of immobilized MP-11
(Figure 2). The majority of our materials (spectra a-c
and e-f) are silica particles with MP-11 layered on the
surface via covalent attachment, chemisorption or physi-
sorption. The spectra of these materials have similar
intensities; the highest intensity is observed for the
material with the highest loading of MP-11 on the surface
(spectrum a). In another type of material, MP-11 encap-
sulated in sol-gel silica glass, microperoxidase-11 mol-
ecules are distributed through the volume of each par-
ticle. Therefore, a smaller number of these molecule are
located closer to the particle surface, where they can be
detected by diffuse-reflectance spectroscopy. Thus, a
lower overall intensity is observed for encapsulated MP-
11 (spectrum d).
Con for m a tion of Dissolved MP -11. In the molecule
of MP-11, undecapeptide 11-21 of cytochrome c is
tethered to the heme via three residues: Cys 14, Cys 17,
and His 18; see Figure 1. Therefore, the pentapeptide
segment 14-18 is confomationally rigid; NMR63,64 and
resonance Raman63 spectra consistently indicate one turn
of the R-helix. This R-helical motif is stabilized by
intramolecular hydrogen bonds,62 which are especially
favored by the hydrophobic environment in the micelle.63
The rest of the peptide chain is very flexible. Circular
dichroism spectra of MP-11 in aqueous buffer at pH 9.064
or in 90% TFE solution at pH 7.065 show that, at these
conditions, the peptide moiety in MP-11 is essentially a
random coil. Calculation of the helical content from the
relative intensities of the CD peak at ca. 203 (random
coil) and the small shoulder at ca. 220 nm (R-helix)
indicates that only a small fraction of the undecapeptide
is R-helical in aqueous buffer.64
Because MP-11 is monomeric in micellar solution,40,48-50
the yield of sulfoxidation achieved by the catalyst under
these conditions is high (entry 12 in Table 2). Since the
concentration of trifluoroethanol (TFE) was relatively
high,58,59 we had to consider the possible effects of the
alcohol group and the TFE hydrophobicity; these effects
are expected to favor the monomeric form of MP-11. The
catalysts in TFE solution (entries 15-17) and in micellar
solution (entries 12-14) afford similarly high yields of
sulfoxide.
En a n tioselectivity of th e Su lfoxid a tion Rea ction
Ca ta lyzed by Va r iou sly Im m obilized a n d Dissolved
MP -11. The enantioselectivity of the reaction is induced
by the chiral environment around the active site in the
catalyst. If the heme is exposed but the peptide confor-
mation is appreciably changed, the yield may be rela-
tively high, but the enantioselectivity may be relatively
low. Indeed, Tables 1 and 2 show that the yield of
sulfoxide and the enantioselectivity in its formation are
not necessarily correlated.
We recorded the CD spectra of MP-11 and its N-
acetylated derivative in all solvents and at the same
concentration that we used in the sulfoxidation reactions;
see Figure 4. Only in a micellar solution (spectrum f) is
the peptide in MP-11 significantly R-helical. In all other
solvents (spectra a-e and g), the peptide moiety in MP-
11 and N-Ac-MP-11 is randomly coiled. This catalyst is
(60) Hapke, B. In Theory of Reflectance and Emittance Spectroscopy;
Cambridge University Press: New York, 1993; pp 284-234.
(61) Mondelli, R.; Scaglioni, L.; Mazzini, S.; Bolis, G.; Ranghino, G.
Magn. Reson. Chem. 2000, 38, 229-240.
(62) Low, D. W.; Gray, H. B.; Duus, J . O. J . Am. Chem. Soc. 1997,
119, 1-5.
We correlated the enantioselectivity with the concen-
tration of MP-11 on the support surface and with the
nature of the support. Using the atomic coordinates for
the heme and the amino acid residues 11-21 of cyto-
(63) Ma, J .-G.; Vanderkooi, J . V.; Zhang, J .; J ia, S.-l.; Shelnut, J . A.
Biochem. 1999, 38, 2787-2795.
(64) Urry, D. W. J . Biol. Chem. 1967, 242, 4441-4448.
(65) Santucci, R.; Picciau, A.; Antonini, G.; Campanella, L. Biochim.
Biophys. Acta 1995, 1250, 183-188.
2606 J . Org. Chem., Vol. 68, No. 7, 2003

Catalytic Activity of Microperoxidase-11
chrome c in the RASMOL program,67 we estimated the
cross-sectional “footprint” of MP-11 to be 130-140 Å2.
Knowing the loading of MP-11 per gram of silica gel,
whose surface area is 550 m2/g, we estimated the ratio
of the surface area and the MP-11 “footprint”. This ratio
was 26-38 for MP-11 that was covalently attached or
chemisorbed at high loading and 490-530 for MP-11
physisorbed or chemisorbed at normal loading. In all
cases, MP-11 molecules can be sufficiently far apart not
to interact with one another. Indeed, the EPR spectra
showed that axial ligation does not occur in any of these
four samples, containing immobilized MP-11. However,
intermolecular interactions of other types (i.e., hydro-
phobic forces or interpeptide hydrogen bonding) are still
conceivable. The average number of surface aminopropyl
groups per footprint of MP-11 molecule can be estimated
from the known concentration of these groups on the
surface, which is 1.12 mmol/g. This number is zero for
physisorption on silica gel and ca. 2 for chemisorption
on the aminopropyl-functionalized silica gel or covalent
attachment to aminopropyl-functionalized silica gel.
Consider enantioselectivities for the following pairs of
entries in Table 1: 16 vs 19 (first cycles) and 17 vs 20
(second cycles). Evidently, an increase in the MP-11
concentration on the aminopropyl-modified surface (en-
tries 19 and 20) does not affect the enantioselectivity of
sulfoxidation; the average number of surface aminopropyl
groups per adsorbed MP-11 molecule is the same. Com-
paring enantioselectivities for the catalysts with the same
MP-11 loading on different supports (aminopropyl-func-
tionalized silica gel in entries 7-9, 16-18, and 19-21
vs silica gel in entries 10-12), we conclude that at the
same loading of MP-11 on the surface, the presence of
aminopropyl groups lowers the enantioselectivity in
sulfoxidation reaction. On the other hand, our desorption
studies (see above) showed that the presence of the
surface aminopropyl groups strengthens the interactions
between the support and microperoxidase-11. We pre-
sume that these stronger interactions between the sur-
face and the catalyst cause a conformation change in the
catalyst, which lowers the enantioselectivity.
The absence of enantioselectivity of MP-11 dissolved
in methanol (Table 2, entries 9-11) suggests that the
conformation of the catalyst differs from that in water.
To further test the effect of an alcoholic environment on
the enantioselectivity, MP-11 was physisorbed onto silica
gel not only from aqueous solution (Table 1, entries 10-
12) but also from methanol solution (entries 13-15).
Interestingly, MP-11 physisorbed onto silica gel from
methanol solution delivers the same low levels of enan-
tioselectivity as MP-11 in methanol (Table 2, entries
9-11) and not the higher levels observed for MP-11
physisorbed onto silica gel from aqueous solution. Thus,
these experiments show that both the amine-rich and
alcohol-rich environments lead to a greatly diminished
chiral induction by microperoxidase-11. It is now plau-
sible to conclude that the low enantioselectivity of
encapsulated MP-11 (Table 1, entries 4-6) is caused by
methanol (byproduct of hydrolysis of tetramethoxysilane)
in the silica sol prior to entrapment of the catalyst.
The low enantioselectivity observed for MP-11 in
trifluoroethanol solution (entries 15-17) mirrors that in
methanol solution (entries 9-11), again a consequence
of the alcohol functionality (as shown above for metha-
nol). Circular dichroism measurements by us (Figure 4,
spectrum g) and others65 consistently show a vanishingly
low R-helicity of MP-11 dissolved in trifluoroethanol
solution.
N-Acetylation of MP-11 does not change the enantio-
selectivity of sulfoxidation reaction (Table 2, entry 4 vs
entry 1). A decrease in enantioselectivity as the pH is
lowered from 6.0 to 3.0 (entries 1 vs 7 and 4 vs 8) is
attributable to the enhanced “background” reaction,
which does not involve microperoxidase-11.
The incipient R-helical motif of residues 14-18 is
stabilized by intramolecular hydrogen bonds,62 which in
turn are favored by the hydrophobic environment within
the micelle.63 The good enantioselectivity in entries 12-
14 may be due to this R-helicity. Whereas in simple
solution the enantioselectivity is nearly lost in the
repeated cycles, in the micelle, it is completely preserved.
As spectrum f in Figure 4 shows, the micellar solution is
the only one in which MP-11 exhibits CD peaks charac-
teristic of a significant R-helical content. The enantio-
selectivity of the catalyst indeed seems to arise from its
conformation.
However, in monomeric MP-11 molecule (Figure 1),
most of the chiral groups are located on the side of heme
opposite from that participating in oxygen transfer.
Although MP-11 in micellar solutions is not oligomerized
via axial ligation,40,48-50 it may still be aggregated via
intermolecular interactions (i.e., hydrophobic forces or
interpeptide hydrogen bonding). MP-11 physisorbed onto
silica gel from aqueous solution is not oligomerized via
axial ligation (Table 1); the relatively higher enantio-
selectivity of this catalyst (Table 1, entries 10-12) can
also be induced by noncovalent aggregation. In these
aggregates, a chiral environment around the exposed
heme side (the active site) will make the catalyst enan-
tioselective.
To test the aggregation explanation, we studied the
effect of the concentration of N-acetylated MP-11 on the
enantioselectivity of this catalyst. In N-Ac-MP-11, oligo-
merization via axial ligation is now impossible, since all
primary amino groups are blocked. Solutions of N-acetyl-
MP-11 obey Beer’s law at concentrations as high as 20-
60 µM;37,57 above this limit, “noncovalent aggregation of
heme units” was suggested.2,57 We examined the concen-
tration range from ca. 331 µM (our usual reaction
conditions) to 16.6 µM (5% of this value). The increase
in sulfoxide yield upon this 20-fold dilution (Table 3,
entry 4 vs entry 1) can be a consequence of partial
blocking of the access to the heme groups in the non-
covalent aggregates. Importantly, the data in Table 3
show that the enantioselectivity is diminished upon
dilution, reaching practically zero at the catalyst con-
centration of 33.1 µM. Evidently, in noncovalent ag-
gregates, which exist at higher concentrations of N-acetyl-
MP-11 (entries 1 and 2), a greater chirality of the active
site environment is reflected in higher enantioselectivi-
ties. Breakup of these aggregates upon dilution is re-
flected in very low enantioselectivities (entries 3 and 4).
These results again show that the catalytic activity of
MP-11 depends on its aggregation state and consequent
(66) Aron, J .; Baldwin, D. A.; Marques, H. M.; Pratt, J . M.; Adams,
P. A. J . Inorg. Biochem. 1986, 27, 227-243.
(67) Sayle, R. RasWin Molecular Graphics 2.6-ucb.; Glaxo Research
and Development: Greenford, Middlesex, UK, 1995.
J . Org. Chem, Vol. 68, No. 7, 2003 2607

Kadnikova and Kostic´
TABLE 3. Su lfoxid a tion of Th ioa n isole by 1 Mola r
Equ iv of Hyd r ogen P er oxid e Ca ta lyzed by N-Acetyla ted
Micr op er oxid a se-11 (N-Ac-MP -11) a t Am bien t
Tem p er a tu r e a n d p H 6.0 for 1 h , w ith a
silica glass or physisorption onto silica gel from metha-
nolic solution), the detrimental effect of alcoholic environ-
ment remains even after the immobilized MP-11 is
resuspended in the aqueous buffer for the reaction.
In the absence of effects caused by these functional
groups, aggregation of small chiral monomers of MP-11
or its N-acetylated derivative actually preserves the high
levels of enantioselectivity. This aggregation can occur
via axial ligation (MP-11 in aqueous buffer) or via
noncovalent interactions (MP-11 physisorbed on silica gel
from aqueous solution, MP-11 in micellar solution, and
N-acetylated MP-11 in aqueous buffer). Using the most
selective catalysts such as MP-11 physisorbed on silica
gel from aqueous solution and MP-11 in micellar solution
allows an enantioselectivity of ca. 40% to be achieved and
sustained in repeated uses of the same catalyst.
Some of our catalysts, such as MP-11 covalently
attached to aminopropyl-functionalized silica, are active
and robust, but are less enantioselective than MP-11 in
aqueous solution. Other catalysts, such as MP-11 in
micellar solution, are both active and enantioselective,
but cannot be recycled efficiently. A compromise among
these desirable qualities was achieved by adjusting the
oligomerization state of the catalyst and the reaction
medium.
N-Ac-MP -11:P h SMe Mola r Ra tio of 1:31a
[N-Ac-MP-11],
µM
sulfoxide
yieldb
entry
dilution
eec
1
2
3
4
331
110
33.1
16.6
none
1:3
1:10
1:20
56 ( 3
74 ( 4
93 ( 1
92 ( 7
25 ( 3
15 ( 7
4 ( 3
5 ( 2
a
Only the first cycle is reported; see Experimental Section for
b
details. Each experiment was run in triplicate. Determined by
1H NMR spectroscopy. c Determined by a GC-MS method. The
predominant enantiomer in all cases was (S)-PhS(O)Me.
properties. An understanding of this dependence allowed
us to control the yield and enantioselectivity of oxidation
of organic sulfides by hydrogen peroxide.
Con clu sion
The product yield in the oxygenase-like reactions of
MP-11 increases when the heme is made more accessible
by prevention of oligomerization of the catalyst caused
by axial ligation. This can be achieved by chemisorption
and physisorption of MP-11 onto silica gel or amino-
propyl-functionalized silica gel. Methods for similar
prevention of oligomerization by axial ligation (“mono-
merization”) in microperoxidases, which were previ-
ously used in spectroscopic studies,2,3,37,40,48,50,52,53,56,57 are
successfully applied in this study for controlling the
oxygenase-like function of microperoxidase.
The effects of the medium and of the oligomerization
state on the enantioselectivity of the catalyst are subtler
and more intricate than their effects on the reaction yield.
The enantioselectivity is minimal when the hydrogen-
bonding groups present on the support (NH2) or in the
medium (OH) interact with the MP-11 molecule and
presumably alter its conformation. If during the prepara-
tion of the catalyst alcohol is present in large concentra-
tions (as is the case during encapsulation into sol-gel
Ack n ow led gm en t. We thank Dr. Kamel A. Harrata
for help with optimization of the chiral GC-MS condi-
tions, Stephen Veysey for the elemental analyses, David
P. Klein for help with the DRIFTS experiments, and
Dr. R. David Scott for help with the EPR spectroscopy
setup and data collection. We are also grateful to Dr.
Frank G. Padera of Perkin-Elmer Instruments for
recording the diffuse-reflectance UV-vis spectra. This
research was supported by the National Science Foun-
dation.
Su p p or tin g In for m a tion Ava ila ble: EPR spectra of
variously immobilized microperoxidase-11. This material is
available free of charge via the Internet at http://pubs.acs.org.
J O026344K
2608 J . Org. Chem., Vol. 68, No. 7, 2003