Tritiated chiral alkanes as substrates for soluble methane monooxygenase from Methylococcus capsulatus (Bath): Probes for the mechanism of hydroxylation

Article abstract of DOI:10.1021/ja963971g

The tritiated chiral alkanes (S)-[1-²H₁, 1-³H]ethane, (R)-[1-²H₁, 1-³H]ethane, (S)-[1-²H₁, 1-³H]butane, (R)[1-²H₁, 1-³H]but

Full text of DOI:10.1021/ja963971g

1818
J. Am. Chem. Soc. 1997, 119, 1818-1827
Tritiated Chiral Alkanes as Substrates for Soluble Methane
Monooxygenase from Methylococcus capsulatus (Bath): Probes
for the Mechanism of Hydroxylation
Ann M. Valentine,† Barrie Wilkinson,‡ Katherine E. Liu,† Sonja Komar-Panicucci,†
Nigel D. Priestley,‡ Philip G. Williams,§ Hiromi Morimoto,§ Heinz G. Floss,*,‡ and
Stephen J. Lippard*,†
Contribution from the Department of Chemistry, Massachusetts Institute of Technology,
Cambridge, Massachusetts 02139, Department of Chemistry, Box 351700, UniVersity of
Washington, Seattle, Washington 98195, and National Tritium Labelling Facility and Structural
Biology DiVision, E. O. Lawrence Berkeley National Laboratory, One Cyclotron Road,
Berkeley, California 94720
ReceiVed NoVember 18, 1996X
Abstract: The tritiated chiral alkanes (S)-[1-2H1,1-3H]ethane, (R)-[1-2H1,1-3H]ethane, (S)-[1-2H1,1-3H]butane, (R)-
[1-2H1,1-3H]butane, (S)-[2-3H]butane, (R)-[2-3H]butane, and racemic [2-3H]butane were oxidized by soluble methane
monooxygenase (sMMO) from Methylococcus capsulatus (Bath), and the absolute stereochemistry of the resulting
product alcohols was determined in order to probe the mechanism of substrate hydroxylation. When purified
hydroxylase, coupling protein, and reductase components were used, the product alcohol displayed 72% retention of
stereochemistry at the labeled carbon for the ethane substrates and 77% retention for the butanes labeled at the
primary carbon. A putative alkyl radical which would yield these product distributions would have a lifetime of 100
fs, a value too short to correspond to a discrete intermediate. Intramolecular kH/kD ratios of 3.4 and 2.2 were determined
for ethane and butane, respectively. When the hydroxylations were performed with purified hydroxylase but only
a partially purified cellular extract for the coupling and reductase proteins, different product distributions were observed.
These apparently anomalous results could be explained by invoking exchange of hydrogen atoms at the R carbon of
the product alcohols. The characteristics of this exchange reaction are discussed. Hydroxylation of [2-3H]butanes
by the latter system yielded ∼90% retention of stereochemistry at the labeled carbon. The implication of these
results for the catalytic mechanism of sMMO is discussed. Together with the mechanistic information available
from a range of substrate probes, the results are best accounted for by a nonsynchronous concerted process involving
attack on the C-H bond by one or more of several pathways discussed in the text.
Introduction
ponent (H) which binds the hydrocarbon substrate and dioxy-
gen,8 a reductase (R) containing Fe2S2 and FAD cofactors which
enable it to accept electrons from NADH and transfer them to
the hydroxylase,10,13 and a regulatory or coupling component,
protein B.14,15 In addition to methane, the sMMO system can
oxidize a wide variety of substrates.16-18
The methanotrophic bacteria Methylococcus capsulatus (Bath)
and Methylosinus trichosporium OB3b rely on a methane
monooxygenase (MMO) system to convert methane into
methanol in the first step of their biosynthetic pathways.1-3 This
reaction provides the organisms with their sole source of carbon
and energy according to eq 1. MMO exists in both particulate
X-ray crystallographic studies of the resting, diferric form
of the M. capsulatus (Bath) hydroxylase,19,20 EPR and ENDOR
CH4 + O2 + NADH + H+ f CH3OH + H2O + NAD+
(8) Dalton, H.; Leak, D. J. In Gas Enzymol.; Degn, H., et al., Eds.; D.
Reidel Publishing Co.: London, 1985; pp 169-186.
(9) Nielsen, A. K.; Gerdes, K.; Degn, H.; Murrell, J. C. Microbiology
1996, 142, 1289-1296.
(10) Colby, J.; Dalton, H. Biochem. J. 1978, 171, 461-468.
(11) Fox, B. G.; Lipscomb, J. D. Biochem. Biophys. Res. Comm. 1988,
154, 165-170.
(12) Fox, B. G.; Froland, W. A.; Dege, J. E.; Lipscomb, J. D. J. Biol.
Chem. 1989, 264, 10023-10033.
(13) Colby, J.; Dalton, H. Biochem. J. 1979, 177, 903-908.
(14) Green, J.; Dalton, H. J. Biol. Chem. 1985, 260, 15795-15801.
(15) Liu, K. E.; Valentine, A. M.; Wang, D.; Huynh, B. H.; Edmondson,
D. E.; Salifoglou, A.; Lippard, S. J. J. Am. Chem. Soc. 1995, 117, 10174-
10185.
(1)
and soluble forms.4 The membrane-bound form contains copper
and is active under normal growth conditions.5-7 When grown
in media deficient in copper, methanotrophs express the soluble
MMO system.8,9 Soluble MMOs from both Methylococcus
capsulatus (Bath)10 and Methylosinus trichosporium OB3b11,12
comprise three proteins, an iron-containing hydroxylase com-
† Department of Chemistry, Massachusetts Institute of Technology.
‡ Department of Chemistry, University of Washington.
§ National Tritium Labelling Facility, Lawrence Berkeley Laboratory.
X Abstract published in AdVance ACS Abstracts, February 1, 1997.
(1) Anthony, C. The Biochemistry of Methylotrophs; Academic Press:
New York, 1982.
(16) (a) Colby, J.; Stirling, D. I.; Dalton, H. Biochem. J. 1977, 165, 395-
402. (b) Green, J.; Dalton, H. J. Biol. Chem. 1989, 264, 17698-17703.
(17) Dalton, H. In AdVances in Applied Microbiology; Academic Press:
1980; Vol. 26, pp 71-87.
(2) Liu, K. E.; Lippard, S. J. In AdVances in Inorganic Chemistry; Sykes,
A. G., Ed.; Academic Press: San Diego, 1995; Vol. 42, pp 263-289.
(3) Lipscomb, J. D. Annu. ReV. Microbiol. 1994, 48, 371-399.
(4) Stanley, S. H.; Prior, S. D.; Leak, D. J.; Dalton, H. Biotech. Lett.
1983, 5, 487-492.
(5) Nguyen, H.-H. T.; Shiemke, A. K.; Jacobs, S. J.; Hales, B. J.;
Lidstrom, M. E.; Chan, S. I. J. Biol. Chem. 1994, 269, 14995-15005.
(6) Zahn, J. A.; DiSpirito, A. A. J. Bacteriol. 1996, 178, 1018-1029.
(7) Cook, S. A.; Shiemke, A. K. J. Inorg. Biochem. 1996, 63, 273-284.
(18) Fox, B. G.; Borneman, J. G.; Wackett, L. P.; Lipscomb, J. D.
Biochemistry 1990, 29, 6419-6427.
(19) Rosenzweig, A. C.; Frederick, C. A.; Lippard, S. J.; Nordlund, P.
Nature 1993, 366, 537-543.
(20) Rosenzweig, A. C.; Lippard, S. J. Acc. Chem. Res. 1994, 27, 229-
236.
(21) Woodland, M. P.; Patil, D. S.; Cammack, R.; Dalton, H. Biochim.
Biophys. Acta 1986, 873, 237-242.
S0002-7863(96)03971-6 CCC: $14.00 © 1997 American Chemical Society

Tritiated Chiral Alkane Hydroxylation by sMMO
J. Am. Chem. Soc., Vol. 119, No. 8, 1997 1819
spectroscopy of the mixed-valent state,21-24 and other physical
methods12,22 indicate that the hydroxylase active site contains a
non-heme dinuclear iron center.25 Crystal structures of the
protein in several other forms provide clues about the nature of
species formed during turnover.26 In addition, the catalytic cycle
has been explored by using freeze-quench and stopped-flow
methodology, which allowed intermediate species to be de-
tected.15,27,28,31 The diferric resting center is reduced by the
reductase to give a diferrous center, the structure of which in
the absence of the other components is known.26 This moiety
binds dioxygen, forming a diiron(III) peroxo species, which then
converts to the active oxidizing intermediate known as com-
pound Q. Intermediates before the peroxo, and after Q, have
been postulated but not directly observed for the M. trichos-
porium OB3b enzyme.29,30 One intermediate termed T, occur-
ring after Q, has been observed only with the substrate
nitrobenzene and has been assigned as protein-bound product.31
The exact formulation of Q has not been elucidated, although
it has been suggested to be a high valent iron-oxo species.2,3
The mechanism by which Q oxidizes substrate is also not
known, and some facets of the mechanism have been postulated
by analogy to cytochrome P-450.
Figure 1. Oxidation of (S)-[1-2H1,1-3H]ethane at the C-1H bond via
a proposed substrate radical intermediate. Radical rebound before
rotation about the C-C bond in the resulting planar intermediate would
lead to products displaying retention of configuration, as shown.
Rotation before rebound would lead to inverted products.
employing a very fast radical clock substrate probe, the
calculated radical rebound rate was too fast to be rationalized
on the basis of a discrete radical intermediate.39 In the same
study, no correlation was observed between the ring opening
rate of a range of probes and the amount of ring-opened products
observed after oxidation of those probes by cytochrome P-450.
Later work employing a probe designed to discriminate between
radical and cation intermediates led to the proposal of a
nonsynchronous concerted mechanism for the P-450 reaction.40
A study of this enzyme activated by H2O2 concluded that an
Fe(III) peroxo, not a ferryl, was responsible for hydroxylation
of lauric acid.41 Theoretical calculations indicate that a more
likely scenario than radical rebound, at least for quadricyclane
oxidation, might involve HO+ insertion into a C-C bond rather
than radical rebound.42
Radical clock substrate probe studies with sMMO isolated
from M. capsulatus (Bath) afforded no evidence for the
formation of substrate radicals.43,44 Based on the known rate
constants for rearrangement of the various probes, a lower limit
of 1013 s-1 at 45 °C was estimated for a putative rebound
reaction rate constant. For enzyme isolated from M. trichos-
porium OB3b, however, evidence for a substrate radical was
detected in the form of 3-6% rearranged product alcohols.43,45
Based on the amount of rearrangement observed, a rate constant
of 6 to 9 × 1012 s-1 at 30 °C was estimated for a putative
rebound reaction for this enzyme.43 In parallel work, informa-
tion about the nature of possible intermediate substrate radicals
was obtained by allowing sMMO from M. trichosporium OB3b
to hydroxylate both (R)- and (S)-[1-2H1,1-3H]ethane.46 As
illustrated in Figure 1, if the reaction were to proceed through
a radical rebound mechanism, some inversion at the chiral
carbon atom would be expected. In the figure, intermediate Q
is depicted as a bis(oxo)iron(IV) or diferryl species, although
this formulation is only one alternative.15 Following abstraction
of a hydrogen atom from the chiral substrate, an alkyl radical
results. This substrate radical can rebound, forming the
The heme enzyme cytochrome P-450 displays similar reactiv-
ity to sMMO with the exception that it will not hydroxylate
methane.32 Cytochrome P-450 oxidations are thought to proceed
through a high valent iron oxo, or ferryl, intermediate.32-35 This
species is postulated to abstract a hydrogen atom from the
hydrocarbon substrate, resulting in a substrate radical and an
iron-coordinated hydroxyl radical. The hydroxyl radical and
substrate radical recombine in a “rebound” reaction to afford
the product alcohol, which dissociates from the active site
leaving the resting state of the enzyme. Radical clock substrate
probes which rearrange upon formation of a substrate radical
provided early support for the presence of substrate radicals in
the reaction cycle of cytochrome P-450.36-38
Recently, some investigators have questioned the validity of
the rebound model for cytochrome P-450. In one study
(22) DeWitt, J. G.; Bentsen, J. G.; Rosenzweig, A. C.; Hedman, B.;
Green, J.; Pilkington, S.; Papaefthymiou, G. C.; Dalton, H.; Hodgson, K.
O.; Lippard, S. J. J. Am. Chem. Soc. 1991, 113, 9219-9235.
(23) DeRose, V. J.; Liu, K. E.; Kurtz, D. M., Jr.; Hoffman, B. M.;
Lippard, S. J. J. Am. Chem. Soc. 1993, 115, 6440-6441.
(24) Liu, K. E.; Lippard, S. J. J. Biol. Chem. 1991, 266, 12836-12839.
(25) Feig, A. L.; Lippard, S. J. Chem. ReV. 1994, 94, 759-805.
(26) Rosenzweig, A. C.; Nordlund, P.; Takahara, P. M.; Frederick, C.
A.; Lippard, S. J. Chem. Biol. 1995, 2, 409-418.
(27) Liu, K. E.; Wang, D.; Huynh, B. H.; Edmondson, D. E.; Salifoglou,
A.; Lippard, S. J. J. Am. Chem. Soc. 1994, 116, 7465-7466.
(28) Liu, K. E.; Valentine, A. M.; Qiu, D.; Edmondson, D. E.; Appelman,
E. H.; Spiro, T. G.; Lippard, S. J. J. Am. Chem. Soc. 1995, 117, 4997-
4998.
(29) Liu, Y.; Nesheim, J. C.; Lee, S.-K.; Lipscomb, J. D. J. Biol. Chem.
1995, 270, 24662-24665.
(30) Nesheim, J. C.; Lipscomb, J. D. Biochemistry 1996, 35, 10240-
10247.
(31) Lee, S.-K.; Nesheim, J. C.; Lipscomb, J. D. J. Biol. Chem. 1993,
268, 21569-21577.
(32) Ortiz de Montellano, P. R. In Cytochrome P450: Structure,
Mechanism, and Biochemistry, 2nd ed.; Ortiz de Montellano, P. R., Ed.;
Plenum Press: New York, 1995; pp 245-303.
(39) Newcomb, M.; Le Tadic, M.-H.; Putt, D. A.; Hollenberg, P. F. J.
Am. Chem. Soc. 1995, 117, 3312-3313.
(40) Newcomb, M.; Le Tadic-Biadatti, M.-H.; Chestney, D. L.; Roberts,
E. S.; Hollenberg, P. F. J. Am. Chem. Soc. 1995, 117, 12085-12091.
(41) Pratt, J. M.; Ridd, T. I.; King, L. J. J. Chem. Soc., Chem. Commun.
1995, 2297-2298.
(33) Groves, J. T.; Han, Y.-Z. In Cytochrome P450: Structure, Mech-
anism, and Biochemistry, 2nd ed.; Ortiz de Montellano, P. R., Ed.; Plenum
Press: New York, 1995; pp 3-48.
(34) Mansuy, D.; Battioni, P. In ActiVation and Functionalization of
Alkanes; Hill, C. L., Ed.; Wiley: New York, 1989; pp 195-218.
(35) Guengerich, F. P. In Biological Oxidation Systems, Vol 1; Reddy,
C. C., Hamilton, G. A., Madyastha, K. M., Eds.; Academic Press: San
Diego, 1990; pp 51-67.
(42) Bach, R. D.; Mintcheva, I.; Esteve´z, C. M.; Schlegel, H. B. J. Am.
Chem. Soc. 1995, 117, 10121-10122.
(43) Liu, K. E.; Johnson, C. C.; Newcomb, M.; Lippard, S. J. J. Am.
Chem. Soc. 1993, 115, 939-947.
(44) Choi, S.-Y.; Eaton, P. E.; Hollenberg, P. F.; Liu, K. E.; Lippard, S.
J.; Newcomb, M.; Putt, D. A.; Upadhyaya, S. P.; Xiong, Y. J. Am. Chem
Soc. 1996, 118, 6547-6555.
(36) Ortiz de Montellano, P. R.; Stearns, R. A. J. Am. Chem. Soc. 1987,
109, 3415-3420.
(37) Bowry, V. W.; Ingold, K. U. J. Am. Chem. Soc. 1991, 113, 5699-
5707.
(38) Bowry, V. W.; Lusztyk, J.; Ingold, K. U. J. Am. Chem. Soc. 1991,
113, 5687-5698.
(45) Ruzicka, F.; Huang, D.-S.; Donnelly, M. I.; Frey, P. A. Biochemistry
1990, 29, 1696-1700.
(46) Priestley, N. D.; Floss, H. G.; Froland, W. A.; Lipscomb, J. D.;
Williams, P. G.; Morimoto, H. J. Am. Chem. Soc. 1992, 114, 7561-7562.

1820 J. Am. Chem. Soc., Vol. 119, No. 8, 1997
Valentine et al.
sMMO protein components were separated on a DEAE Sepharose fast
flow (Pharmacia LKB Biotechnology Inc.) column which was packed
(5.0 × 22 cm) and equilibrated in buffer A by using a Pharmacia FPLC
chromatographic system. All buffers were filtered and degassed prior
to use. The sample was loaded at approximately 4 mL/min by using
an in-line peristaltic pump. The loaded sample was washed at the same
flow rate with 800 mL of buffer A. An eight column volume gradient
of 0-0.4 M NaCl in buffer A was employed to elute bound protein.
The flow rate applied during the salt gradient was 3.5 mL/min. Native
polyacrylamide gel electrophoresis (PAGE) and SDS-PAGE analysis
were used to evaluate the protein fractions.
Fractions which contained protein of the correct molecular weight
according to SDS-PAGE analysis, and which reduced dichlorophenol
indophenol at a rate which was greater than 30% of the most active
fraction,51 were pooled as crude reductase protein. The crude reductase
sample was diluted with an equal volume of 20 mM Tris (pH 7.1)
with 8 mM NaTG (Tris buffer) and 50 mM NaCl and was loaded at a
flow rate of 3 mL/min on a DEAE Sepharose CL-6B column (2.6 ×
8.0 cm). The loaded sample was washed with 2.8 column volumes of
Tris buffer with 100 mM NaCl. The bound protein was eluted with a
10 column volume gradient of Tris buffer from 100-500 mM NaCl.
Fractions from DEAE Sepharose CL-6B chromatography were pooled
which had an A458nm/A340nm ratio of greater than 1.0. The pooled material
was applied to a 5′AMP Sepharose 4B affinity (Pharmacia) column
(2.5 × 11 cm). The affinity column had been regenerated and
equilibrated in Tris buffer, 50 mM NaCl.
The loaded sample was washed with 1.3 column volumes of Tris
buffer, 95 mM NaCl. This wash eluted nonadsorbed proteins from
the column. One column volume of Tris buffer, 410 mM NaCl
followed by a three column volume rinse of Tris buffer, 50 mM NaCl
eluted nonspecifically bound protein. Reductase fractions were eluted
with Tris buffer and 50 mM NaCl containing 1.4 mM ethanol-free
NADH. SDS-PAGE analysis was performed to confirm the absence
of any contaminating protein in the purified reductase. The NADH
was removed from purified reductase by concentrating the sample to
∼10% of the original volume with an Amicon concentrating cell
equipped with an ultrafiltration PM30 membrane, which was run under
30 psi of N2 pressure. The reductase was rediluted in Tris buffer
containing 50 mM NaCl and reconcentrated. This procedure was
repeated until no NADH was detected spectrophotometrically in the
eluate. A typical yield of 20-30 mg of purified reductase was obtained
from 100 g of cell paste, and specific activities were in the range of
1200-1600 mU/mg under the conditions necessary to generate a high
product concentration (vide infra). In a standard propylene assay,
activities ranged from 3500-3800 mU/mg.22,43
unrearranged product alcohol. Alternatively, because the carbon
radical is planar, rotation about the C-C bond might occur prior
to recombination, affording a product alcohol with the opposite
configuration. The alcohol products were derivatized with a
chiral acid, and tritium nuclear magnetic resonance (NMR)
analysis of the resulting diastereomeric esters was used to
quantitate the amount of inversion that occurred during the
hydroxylation reaction. For both (R)- and (S)-chiral ethane
substrates, ≈65% of the product alcohols retained their stereo-
chemistry, whereas ≈35% were inverted.46 Taking the energy
barrier for rotation about the C-C bond to be 0.15 kcal/mol,47,48
we compute the rate constant for this process to be 4.9 × 1012
s-1 at 30 °C (eq 2). From the amount of retention of
stereochemistry in the chiral alcohol products, a value of 4.2 ×
1012 s-1 was calculated for the rebound reaction rate constant
(eq 3). This value agrees well with that estimated from the
radical clock work.43
k(rotation) ) (kT/h)(exp(-∆Gq/RT))
(2)
(3)
krebound ) krotation(% retention/% inversion - 1)
In view of the slight differences in reactivity toward radical
clock substrate probes exhibited by sMMO from the different
organisms, we were interested to determine how the M.
capsulatus (Bath) enzyme would perform in the chiral ethane
experiment.46 Accordingly, the substrates (S)-[1-2H1,1-3H]-
ethane, (R)-[1-2H1,1-3H]ethane, (S)-[1-2H1,1-3H]butane, (R)-[1-
2H1,1-3H]butane, (S)-[2-3H]butane, (R)-[2-3H]butane, and race-
mic [2-3H]butane were hydroxylated with sMMO from M.
capsulatus (Bath), the results and analysis of which are reported
here.
Experimental Section
Bacterial Growth and Protein Purification. Methylococcus cap-
sulatus (Bath) cells were grown as described previously.49 Hydroxylase
and recombinant protein B were purified as reported elsewhere, and
the iron content for hydroxylase and specific activities were in the usual
ranges.22,43 In some reactions, an unresolved mixture of protein B,
reductase, and other cellular proteins (B/R mix) from the initial DEAE
cellulose column (0.05-0.5 M NaCl fraction) was used with purified
hydroxylase to reconstitute the enzymatic system.18 In all other
experiments, purified recombinant B22,43 and purified native reductase
(vide infra) were employed.
Isolation of sMMO Reductase from Soluble Cell Extract. Protein
purification procedures were performed at 4 °C or on ice. Cell lysis
with a French pressure cell (American Instrument Co.) and the
preparation of soluble cell extract were carried out as previously
described43 except for the buffer composition, which followed a
published procedure.50 Frozen cells (∼100 g) were suspended in 200
mL of 25 mM N-morpholinopropane sulfonic acid (MOPS) buffer (pH
7.0) which contained 200 µM of Fe(NH4)2(SO4)2‚6H2O and 2 mM
cysteine (buffer A) as well as 0.01 mg/mL of DNase 1 and 2 mM
phenylmethylsulfonyl fluoride (PMSF). Buffers containing iron and
cysteine were prepared just prior to use. After the cells were lysed,
sodium thioglycolate (NaTG) was added to a final concentration of 8
mM, which was maintained in all subsequent steps of the reductase
purification. The lysed cells were centrifuged in a Beckman L-70
ultracentrifuge at 32 000 × g for 10 min. The supernatant was
centrifuged at 90 000 × g for 1 h. An additional 10 min centrifugation
(90 000 × g) of the supernatant removed residual particles, allowing
the supernatant to be filtered with a 0.2 µm syringe filter. The soluble
cell-free extract was diluted with an equal volume of buffer A, and the
Determination of Conditions for Oxidizing Labeled Alkanes.
Preliminary assays with unlabeled ethane, butane, and propylene were
carried out to determine conditions which afforded 1-2 µmol of product
alcohol. Hydroxylase was concentrated to ≈300 µM and incubated
with the B/R mixture or the purified components in varying ratios in
a 5 mL reaction flask capped with a septum. The total volume was
adjusted to 400 µL with 25 mM MOPS, pH 7.0 buffer. A syringe was
inserted through the septum, and 2 mL of the head space gas was
removed and replaced with 2 mL of the substrate gas. The mixture
was incubated for 30 s at 45 °C in a shaking incubator, after which
time 100 µL of a 0.1 M ethanol-free solution of NADH was added to
initiate the reaction. The flask was returned to the incubator, and the
reaction was allowed to proceed for 5 min. A 5 µL portion of the
solution was injected into a gas chromatograph equipped with a Porapak
Q column to quantitate the alcohol product present in the reaction
solution.43 The area of the product peak was compared to a calibration
curve of alcohol peak area versus the concentration of sample injected.
Conditions which afforded at least 1 µmol of alcohol product per 1
mL of reaction volume were subsequently employed at the National
Tritium Labelling Facility. Typically 50-80 nmol each of H and B
and 42-68 nmol of R were employed. Under these conditions, the
specific activity calculated for the hydroxylase is low, ≈40 mU/mg,
owing to product inhibition.
(47) Pacansky, J.; Dupuis, M. J. Am. Chem. Soc. 1982, 104, 415-421.
(48) Suter, H. U.; Ha, T.-K. Chem. Phys. 1991, 154, 227-236.
(49) Pilkington, S. J.; Dalton, H. In Methods In Enzymology; Academic
Press: New York, 1990; Vol. 188, pp 181-190.
(50) Fox, B. G.; Froland, W. A.; Jollie, D. R.; Lipscomb, J. D. In Methods
In Enzymology; Academic Press: New York, 1990; Vol. 188, pp 191-
202.
Reagent Synthesis. Materials and General Procedures. 1H, 2H,
and 13C NMR spectra were recorded on either a Bruker AC-200 or
AF-300 spectrometer. Chemical shifts were determined relative to the
(51) Hultquist, D. E. In Methods in Enzymology; Academic Press: New
York, 1978; Vol. 52, pp 463-473.

Tritiated Chiral Alkane Hydroxylation by sMMO
J. Am. Chem. Soc., Vol. 119, No. 8, 1997 1821
residual proton or 13C absorption of the solvent, CDCl3 or C6D6, used
as internal standard. 3H spectra were recorded at 320.13 MHz by using
an IBM-Bruker AF-300 spectrometer fitted with a tritium channel and
atmospheric pressure to yield the title compound as a colorless oil (20.4
g, 45%). 1H NMR: δH (200 MHz, CDCl3): 0.29 (3H, t, 7.3 Hz, H-4),
1.27 (2H, sext, 7.3 Hz, H-3), 1.43 (2H, t, 7.3 Hz, H-2), 3.23 (1H, bs,
OH) ppm. 2H NMR: δD (46.1 MHz, C6H6): 3.34 (s) ppm.
1
a dual H/3H 5 mm probe. Stable isotope compounds were obtained
[1-2H1]Butyraldehyde. [1-2H2]Butan-1-ol (20 g, 0.26 mol) was
heated to a gentle reflux, and a solution of K2Cr2O7 (28.3 g, 0.1 mol)
in concentrated H2SO4 (20 mL) was then added via a nonpressure
equalizing dropper funnel, at such a rate so as to maintain a constant
distillation of the product [1-2H1]butyraldehyde as a water azeotrope.
After addition of all the dichromate solution the mixture was heated at
reflux for a further 10 min (maximum stillhead temperature 89 °C).
The resulting aqueous layer was discarded, and the organic layer was
dried over MgSO4 (2 g) for 30 min and then purified by fractional
distillation. The fractions which distilled between 50 and 75 °C were
pooled and again fractionally distilled; fractions were collected with
stable stillhead temperatures of 67 °C and 70 °C. NMR analysis showed
no trace of codistilling 1-[1-2H2]butanol, but fraction 1 contained a small
quantity (<3%) of the corresponding hemiacetal. Combined yield )
6.2 g, 85.5 mmol, 33%. 1H NMR: δH (200 MHz, CDCl3): 0.88 (3H,
t, 7.4 Hz, H-4) ppm, 1.59 (2H, sext, 7.3 Hz, H-3), 2.33 (2H, t, 7.2 Hz,
H-2). 2H NMR: δD (46.1 MHz, C6H6): 9.25 (s) ppm. 13C NMR: δC
from Aldrich Chemical Co. or Cambridge Isotope Labs. (R)-O-Acetyl-
mandelic acid and (S)- and (R)-2-butanol were purchased from Aldrich
Chemical Co. and were of 99% ee. THF was dried by distillation from
Na/benzophenone ketyl, CH2Cl2 by distillation from CaH2, and pyridine
by storing over KOH pellets. All other solvents were of HPLC grade
and used without further purification. Tosyl chloride was recrystallized
from hexane and dried in vacuo overnight prior to use. Argon was
dried by passing through a bed of 3 Å molecular sieves and then
anhydrous CaCO3. Flash chromatography was performed over silica
gel 60 (230-400 mesh).
(S)-[1-2H1]Ethanol. [1-2H1]Acetaldehyde (0.56 mL, 10 mmol)
(Cambridge Isotope Labs) was added to a stirred solution of (R)-Alpine
Borane (10 mmol) in 20 mL of dry tetrahydrofuran (THF) at -78 °C
under argon. The temperature was maintained at -78 °C for 4 h, after
which time the mixture was allowed to warm to room temperature.
After 15-18 h, the solution was cooled to 0 °C, and ethanolamine
(0.74 mL, 12.3 mmol) was added. After 30 min a white precipitate
formed. The solvent was removed by distillation, during which the
precipitate dissolved upon warming. Just before the end of the
distillation, 5 mL of benzene was added to, and distilled from, the
mixture. The benzene addition and distillation steps were repeated
twice. The final volume of the collected distillates was 32 mL. (R)-
[1-2H1]Ethanol was synthesized from (S)-Alpine Borane by using similar
procedures.
(S)-[1-2H]Ethyl Tosylate. To a solution of tosyl chloride (3.5 g,
18.4 mmol) in dry pyridine (15 mL) was added (S)-[1-2H]ethanol (2 g,
43.4 mmol). A crystalline precipitate of pyridine hydrochloride was
visible after 5 min, and the resulting suspension was stirred under argon
for 5 h at room temperature. Reaction was terminated by addition of
diethyl ether (50 mL) and washed with saturated aqueous CuSO4 (1 ×
50, 2 × 25 mL), 1 N aqueous HCl (25 mL), and saturated aqueous
NaHCO3 (25 mL), dried over MgSO4 and then filtered. The solvent
was removed in vacuo, and the title compound recovered as a white
crystalline solid by chromatography over flash silica gel eluting with
10% diethyl ether in hexane (2.42 g, 68%, mp 37 °C). 1H NMR: δH
(300 MHz, CDCl3): 1.22 (3H, t, 7.1 Hz, H-2), 2.39 (3H, s, Ar-Me),
4.06 (2H, q, 7.1 Hz, H-1), 7.30 (2H, d, 8.2 Hz, Ar), 7.73 (2H, d, 8.2
Hz, Ar) ppm. 2H NMR: (46.1 MHz, C6H6) 3.74. (R)-[1-2H]Ethanol
was used to prepare the (R)-[1-2H]ethyl tosylate.
Ethyl (2′R)-2′-Acetoxy-2′-phenylethanoate. To a solution of (R)-
O-acetylmandelic acid (830 mg, 4.3 mmol) and 4-(dimethylamino)-
pyridine (DMAP) (16 mg) in dry methylene chloride (11 mL) at -20/
-30 °C was added, dropwise over 10 min, a solution of dicyclo-
hexylcarbodiimide (DCC) (880 mg, 4.3 mmol) in dry methylene
chloride (5 mL). After 10 min a pale cream precipitate was observed.
Ethanol (500 mg 10.9 mmol) was added dropwise with the temperature
maintained between -20/-30 °C, the solution was stirred for 90 min,
and then the temperature was allowed to rise to ambient conditions
overnight.
The resulting suspension was filtered, and the solids were washed
with methylene chloride (3 × 15 mL). The mixture was washed with
brine (40 mL), saturated aqueous NaHCO3 (35 mL), the organics were
dried over MgSO4, and the solvent was removed in vacuo to leave a
yellow mass. This product was purified by chromatography over flash
silica gel eluting with ethyl acetate/hexane (10:70) to yield the title
compound as a colorless oil (955 mg, 67%). 1H NMR: δH (300 MHz,
CDCl3): 1.19 (3H, t, 7.1 Hz, H-2), 2.16 (3H, s, COCH3), 4.14 (2H, m,
H-2), 5.89 (1H, s, H-2′), 7.35 (3H, m, Ar) and 7.45 (2H, m, Ar) ppm.
[1-2H2]Butan-1-ol. To a suspension of LiAlD4 (15.4 g, 0.37 mol)
in anhydrous THF (120 mL) under argon at room temperature was
added methyl butyrate (61 g, 0.61 mol) at a rate sufficient to initiate,
and then maintain, gentle reflux over approximately 25 min. The
resulting slurry was stirred under reflux for 3 h and then overnight at
room temperature. The resulting mixture was quenched by addition
of water (10 mL) followed by diethyl ether (100 mL) and after 15 min
1 N aqueous HCl (125 mL). The organic fraction was separated, and
the aqueous was extracted with diethyl ether (4 × 50 mL). The
combined organic layers were dried over anhydrous MgSO4 for 1 h
and then filtered. The resulting solution was fractionally distilled at
13
(75.5 MHz, C6D6): 13.60, 15.55, 45.38, and 202.62 (t, JD- ) 25.9
C
Hz) ppm.
(R)-1-[1-2H1]Butanol. A solution of (S)-Alpine Borane (25 mmol)
in anhydrous THF (50 mL) was cooled to -78 °C under argon and
stirred for 1 h. [1-2H1]Butyraldehyde (2 mL, 22 mmol) was added
dropwise, and the resulting solution stirred at -78 °C for 2 h and then
at room temperature overnight. The resulting solution was cooled to
0 °C, ethanolamine (2 mL, 32.5 mmol) was added dropwise, and the
solution was stirred at 0 °C for 30 min and then at room temperature
overnight. The resulting solution was fractionally distilled to remove
all volatiles boiling below 75 °C; xylenes (30 mL) were added, and
the solution again distilled with all volatiles boiling between 75-130
°C collected (19.3 mL). (S)-1-[1-2H1]Butanol was prepared analogously
from (R)-Alpine Borane.
(R)-1-[1-2H1]Butyl Tosylate. To a solution of tosyl chloride (3.8
g, 20 mmol) in dry pyridine (20 mL) was added the (R)-butanol/xylenes
mixture (15.3 mL, max: 17.6 mmol). A crystalline precipitate of
pyridinium hydrochloride was visible after 5 min, and the resulting
lemon yellow solution was stirred under argon for 5 h at room
temperature. The reaction was terminated by addition of diethyl ether
(100 mL) and washed with saturated aqueous CuSO4 (1 × 100, 2 ×
50 mL), 1 N aqueous HCl (50 mL), and saturated aqueous NaHCO3
(50 mL), dried over anhydrous MgSO4, and then filtered. The solvent
was removed in vacuo and the (R)-1-[1-2H1]butyl tosylate recovered
as a colorless oil by chromatography over flash silica gel eluting with
7.5% diethyl ether in hexanes (2.43 g, 60%). 1H NMR: δH (200 MHz,
CDCl3): 0.87 (3H, t, 7 Hz, H-4), 1.34 (2H, sext, 8.1 Hz, H-3), 1.62
(2H, q, 7.5 Hz, H-2), 2.45 (3H, s, Me), 4.02 (1H, m, H-1), 7.34 (2H,
d, 8.3 Hz), 7.78 (2H, dd, 8.3 and 1.8 Hz) ppm. 2H NMR: δD (46.1
MHz, C6H6): 3.79 (s). 13C NMR: δC (75.5 MHz, CDCl3): 13.31,
18.50, 21.55, 30.61, 70.03 (t, 22.9 Hz), 127.79, 129.75, 133.17, and
144.59 ppm. (S)-1-[1-2H1]butyl tosylate was prepared analogously from
(S)-butanol.
(1R)-[1-2H1]Butyl (2′R)-2′-Acetoxy-2′-phenylethanoate. To a
solution of (R)-O-acetylmandelic acid (1.20 g, 6.1 mmol) and DMAP
(16 mg) in dry methylene chloride (25 mL) at -20/-30 °C was added,
dropwise over 10 min, a solution of DCC (1.40 g, 6.4 mmol) in dry
methylene chloride (10 mL) to give a pale cream precipitate. A fraction
of the (R)-1-[1-2H1]butanol/xylenes mixture (4 mL) was added dropwise
with the temperature maintained between -20/-30 °C and stirred for
90 min, and the solution was allowed to reach room temperature
overnight. The resulting cream suspension was filtered, and the filtrate
was washed with saturated aqueous NaHCO3 (2 × 50 mL) and dried
over MgSO4, and the solvent was removed in vacuo to leave a yellow
mass. The title compound was recovered by chromatography over flash
silica gel eluting with 10% ethyl acetate in hexane and analyzed by 2H
NMR to establish the optical purity of the sample; ee ) 93((3)%. 2H
NMR: δD (46.1 MHz, C6D6): 3.89 (s, D-1, major isomer), 3.82 (s,
D-1, minor isomer) ppm. The other enantiomer was prepared similarly
from (S)-1-[1-2H1]butanol and displayed 95% ee.
(()-2-Butyl Tosylate. To a solution of tosyl chloride (1.05 g, 5.5
mmol) in dry pyridine (10 mL) was added (()-2-butanol (1.20 mL, 13

1822 J. Am. Chem. Soc., Vol. 119, No. 8, 1997
Valentine et al.
mmol). A crystalline precipitate of pyridinium hydrochloride was
visible after 5 min, and the resulting solution was stirred under argon
for 5 h at room temperature. The reaction was terminated by addition
of diethyl ether (50 mL) and washed with saturated aqueous CuSO4 (1
× 50 mL, 2 × 25 mL), 1 N aqueous HCl (30 mL), and saturated
aqueous NaHCO3 (30 mL), dried over MgSO4, and then filtered. The
solvent was removed in vacuo, and the title compound was recovered
as a colorless oil by chromatography over flash silica gel eluting with
7.5% diethyl ether in hexanes (799 mg, 64%). (S)-2-Butyl tosylate
and (R)-2-butyl tosylate were prepared analogously from (S)-2-butanol
and (R)-2-butanol, respectively. 1H NMR: δH (300 MHz, CDCl3): 0.78
(3H, t, 7.9 Hz, H-4), 1.21 (3H, d, 6.3 Hz, H-1), 1.54 (2H, m, H-3),
2.40 (3H, s, Me-4′), 4.52 (1H, tq, H-2), 7.29 (2H, d, 7.9 Hz, H-2′ and
H-6′), 7.75 (2H, d, 6.6 Hz, H-3′ and H-5′) ppm.
(1 mL), and a portion (ca. 5 × 104 cpm) was analyzed for chiral purity.52
The compound 2R was treated similarly.
Chiral Tritiated Alkanes. The reagent LiEt3B3H was prepared as
described previously.53 The appropriate alkyl tosylate (≈0.3 mmol),
dissolved in 100 µL of THF, was added to a round bottom flask
equipped with a stopcock connected to a closed vacuum system and
containing ≈0.2 mmol of LiEt3B3H in THF. Vigorous gas evolution
occurred for 15-20 s. After a 1 h incubation period, the flask was
cooled to -78 °C. The gas was transferred to a coconut charcoal bed
(200 mg) in a test tube immersed in liquid nitrogen. The transfer was
allowed to occur for 60 s. Identical procedures were used to synthesize
(S)-[1-2H1,1-3H]ethane, (R)-[1-2H1,1-3H]ethane, (S)-[1-2H1,1-3H]butane,
(R)-1-[2H1,1-3H]butane, (S)-[2-3H]butane, (R)-[2-3H]butane, and racemic
[2-3H]butane from their corresponding alkyl tosylates.
Enzymatic Reactions. A second round bottom flask equipped with
a stopcock, sidearm, and septum was fitted to the reaction apparatus
described above and evacuated. Buffer (25 mM MOPS, pH 7.0, 100-
500 µL) was added through the septum and cooled to 77 K. Substrate
gas was transferred to the flask from the charcoal bed by opening both
stopcocks and bringing the charcoal bed to room temperature by
warming with hot air. After 60 s the stopcocks were closed, and the
buffer was brought to room temperature. The protein solution was
added through the septum. A quantity of NADH (50 or 100 µL of a
0.1 M solution) was also added. Pure O2 gas (1 mL) was added by
syringe through the septum], and the mixture was incubated at 45 ( 2
°C with constant stirring for 30 min. Cooling the flask to -78 °C
terminated the reaction. Excess substrate gas was transferred from the
reaction flask back to the charcoal bed by first cooling the bed in liquid
nitrogen and then opening the stopcocks. Transfer was allowed to
proceed for 60 s, and then the stopcocks were closed. A 10 µL aliquot
of the expected product alcohol (unlabeled) was added to the reaction
flask to act as a carrier in subsequent manipulations. This flask was
removed from the apparatus and transferred to a vacuum line where
the volatile reaction products were collected by lyophilization.
Product Analysis. To measure the radioactivity of the samples, a
1 µL aliquot of the product solution was diluted in 200 µL of methanol
before being placed in a liquid scintillation counter. NMR analysis
was carried out with a 300 MHz Bruker spectrometer at NTLF. All
(2S)-Butyl (2′R)-2′-Acetoxy-2′-phenylethanoate. To a solution of
(R)-O-acetylmandelic acid (829 mg, 4.26 mmol) and DMAP (14 mg)
in dry methylene chloride (11 mL) at -20/-30 °C (acetone/dry ice)
was added, dropwise over 10 min, a solution of DCC (880 mg, 4.27
mmol) in dry methylene chloride (4 mL). After 10 min a pale cream
precipitate was observed. (S)-2-Butanol (300 µL, 3.25 mmol) was
added dropwise with the temperature maintained between -20/-30
°C and stirred for 90 min, and then the solution was allowed to reach
room temperature over a further 90 min. The resulting suspension was
filtered, and the solids were washed with methylene chloride (3 × 15
mL). The mixture was washed with brine (40 mL) and saturated
aqueous NaHCO3 (35 mL), the combined organics were dried over
MgSO4 and the solvent was removed in vacuo to leave a yellow/white
mass. This product was purified by chromatography over flash silica
gel (45 g) eluting with ethyl acetate/hexane (5:70) to yield the title
compound as a colorless oil (678 mg, 89%). 1H NMR: δH (500 MHz,
C6D6): 0.75 (3H, t, 7.4 Hz, H-4), 0.86 (3H, d, 6.3 Hz, H-1), 1.25 (1H,
pseudo sept, 7.2 Hz, H-3R), 1.42 (1H, pseudo sept, 7.2 Hz, H-3S),
1.73 (3H, s, COCH3), 4.86 (1H, m, H-2, 6.03 (1H, s, H-2′), 7.42 (3H,
m), 7.48 (2H, m). Selective decoupling at 0.75 ppm: 1.24 (1H, dd,
13.8 and 5.5 Hz, H-3R), 1.42 (13.7 and 7.2 Hz, H-3S) ppm.
(2R)-Butyl (2′R)-2′-Acetoxy-2′-phenylethanoate. The experimen-
tal protocol for the (2S)-isomer was used, except that R-2-butanol was
employed. 1H NMR: δH (500 MHz, C6D6): 0.46 (3H, t, 7.5 Hz, H-4),
1.03 (3H, d, 6.2 Hz, H-1), 1.10 (1H, pseudo sept, 7.3 Hz, H-3S), 1.24
(1H, pseudo sept, 7.3 Hz, H-3R), 1.74 (3H, s, COCH3), 4.82 (1H, m,
H-2), 6.03 (1H, s, H-2′), 7.42 (3H, m), 7.48 (2H, m). Selective
decoupling at 0.46 ppm: 1.10 (1H, dd, 16.1 and 5.1 Hz, H-3S), 1.24
(1H, dd, 16.2 and 8.4 Hz, H-3R) ppm.
1
3H NMR spectra were H decoupled.
Derivatization.54 (R)-O-Acetylmandelic acid (47 mg, 0.24 mmol)
and DMAP (2 mg) were mixed in 2 mL of methylene chloride at -40
°C in a 100 mL round bottom flask. Over a 5 min period, a solution
of DCC (54 mg, 0.24 mmol) in 0.5 mL of methylene chloride was
added dropwise to the flask. A creamy precipitate resulted after 10
min. The enzymatic lyophilyzate was added dropwise over a 5 min
period. The resulting mixture was allowed to warm to room temper-
ature over 3-6 h and then stirred overnight. The suspension was
filtered through a silica plug (≈2 g) and washed with 35-40 mL of
methylene chloride. The filtrate was evaporated to near dryness under
a dinitrogen stream and then lyophilized to remove any methylene
chloride. The residue was suspended in benzene-d6 and filtered through
a glass wool plug directly into an NMR tube.
Oxidation of C2D6. Perdeuteroethane (Cambridge Isotope Labs,
99%) was oxidized by a reconstituted sMMO system as reported above
for unlabeled substrates except that the incubation time was 30 min.
The product alcohol was isolated from the aqueous reaction mixture
by extraction with 1 mL of methylene chloride and the extract was
dried over MgSO4. A 1 µL aliquot of distilled pyridine and 1 µL
benzoyl chloride were added. The esterification reaction was allowed
to proceed for 1 h at room temperature, and the mixture was analyzed
by gas chromatography with mass spectrometric detection (GC/MS)
on a Hewlett-Packard 5890A gas chromatograph equipped with a
Hewlett Packard 5971A mass spectrometer. An HP-1 (methyl silicone
gum) column (50 m × 0.2 mm × 0.5 µm film thickness) was employed
for the separation.
n-Butyl (2′R)-2′-Acetoxy-2′-phenylethanoate. The experimental
protocol for the (2S)-isomer was employed, except that 1-butanol was
used. 1H NMR: δH (500 MHz, C6D6): 0.61 (3H, t, 7.4 Hz, H-4), 0.99
(2H, pseudo sept, 7.4 Hz, H-3), 1.20 (2H, dquin, 6.8 and 2.0 Hz, H-2),
1.73 (3H, s, COCH3), 3.85 (1H, m, H-1S), 3.94 (1H, m, H-1R), 6.06
(1H, s, H-2′), 7.40 (3H, m) and 7.45 (2H, m) ppm.
Verification of Butyl and Ethyl Tosylate Stereochemistry. To a
mixture of (R)-O-acetylmandelic acid (42 mg, 0.2 mmol) and anhydrous
K2CO3 (28 mg, 0.2 mmol) in dry dimethylformamide (DMF) (1 mL)
stirring vigorously at room temperature under argon was added a
solution of the alkyl tosylate (0.2 mmol) in dry DMF (0.5 mL). After
30 min a fine precipitate of tosic acid had formed. After 36 h the
solvent was removed in vacuo, and the flask was pumped under high
vacuum overnight. The residual gum was washed with benzene (5 ×
3 mL), and the combined fractions were filtered through a plug of flash
silica gel (1 cm, Pasteur pipette) and eluted with benzene (10 mL).
The benzene was removed in vacuo, and the product was analyzed by
2H NMR.
(S)-[2-2H1, 3H]Acetate. A sample of the derivative product resulting
from oxidation of (S)-[2-2H1, 3H]ethane was stirred in toluene (200 µL)
and 2 N KOH (600 µL) at room temperature for 4 h. Carrier ethanol
(10 µL) was added, and the volatiles were isolated by lyophilization.
The isolated volatiles were added to a stirred solution of KMnO4 (160
mg) and K2CO3 (120 mg) in water (500 µL) at 4 °C. The resulting
mixture was brought to room temperature and stirred for 6 h. The
brown suspension was acidified to pH ≈ 1 with 6 M H2SO4, and the
volatiles were isolated by lyophilization. The recovered volatiles were
adjusted to pH 10 with 2 N KOH, and the volatiles were removed by
lyophilization. The remaining solid was dissolved in distilled water
Exchange Reactions. A 2 µL sample of CD3CD2OH or CH3CD2-
OH (Cambridge Isotope Labs, 99%) was added to 0.9 mL of B/R mix,
a reconstituted system consisting only of purified H, B, and R
components, or to buffer (25 mM MOPS, pH 7). A 100 µL portion of
0.1 M NADH was added. In some experiments, 1 µL of 0.1 M NAD+
was also added. The mixture was incubated at 45 °C with shaking for
30 min. The ethanol was isolated as described above and analyzed by
GC/MS.

Tritiated Chiral Alkane Hydroxylation by sMMO
J. Am. Chem. Soc., Vol. 119, No. 8, 1997 1823
Results
Synthesis of Chiral Tritiated Alkanes. The method for
preparing chiral tritiated alkanes applied known and unambigu-
ous chemistry developed for the synthesis of chiral methyl
tritiated acetates in high yield, activity, and enantiomeric purity.
[1-2H1]Acetaldehyde was reduced to (S)- or (R)-[1-2H1]ethanol
by (R)- or (S)-Alpine Borane, respectively. The ethanols were
isolated by distillation and converted to their tosylates by
treatment with tosyl chloride in pyridine. An aliquot was also
converted to the (R)-O-acetylmandelic acid ester derivative in
the presence of DCC-DMAP. 2H NMR analysis of the products
demonstrated that the alcohols had been synthesized in 88 (
4% ee. Experiments with standard materials were carried out
to assure that the C-1 center in the alcohol does not racemize
when this reaction is carried out below -10 °C.55 Treatment
of the ethyl tosylates with (R)-O-acetylmandelic acid and K2CO3
in DMF converted them to their mandelate derivatives with
inversion at C-1. 2H NMR experiments again demonstrated that
the ethanols had been prepared in 88 ( 4% ee and confirmed
the stereochemical fidelity of the derivatization procedure. For
the final generation of chiral tritiated ethane, the tosylate group
was converted to tritide by using the reagent LiEt3B3H at
maximum specific activity.53 This reaction proceeds via a direct
SN2 displacement, affording ethane of unambiguous stereo-
chemistry.
Figure 2. Possible reaction products from the oxidation of (S)-[1-
2H1,1-3H]ethane. Products 1 and 2 arise from reaction at the C-H bond,
products 3 and 4 from reaction at the C-D bond, and product 5 from
reaction at the unlabeled carbon. Alcohols arising from hydroxylation
of the C-T bond are not detected by 3H-NMR spectroscopy. The ratio
of the sum of products 1 and 2 to the sum of products 3 and 4 yields
the intramolecular kH/kD isotope effect. After derivatization, the products
displaying retention of stereochemistry (1 and 3) can be distinguished
from those exhibiting inversion (2 and 4), and the total percent retention
can be determined.
Primary tritiated butanes were generated in a similar manner.
The [1-2H1]butyraldehyde was prepared by LiAl2H4 (>98 atom
% 2H) treatment of methyl butyrate to produce [1-2H2]butanol,
which was converted to the aldehyde by treatment with K2Cr2O7
in H2SO4. This method takes advantage of the primary KIE
for this process, yielding product with enriched deuterium
content. No [1-1H2]butanol species were formed, as verified
by 13C NMR spectroscopy. Butanes carrying tritium in the
secondary position were prepared initially from (S)- and (R)-
2-butanol of >99% ee.
Analysis of Lyophilyzates. For each alkane enantiomer,
hydroxylation can take place either at the hydrogen atom or
deuterium atom of the chiral carbon, at the tritium atom, or at
an unlabeled carbon. Products that arise from hydroxylation
of the C-T bond are not detected, because 3H NMR spectros-
copy is the analytical method. Intramolecular kinetic isotope
effects, kH/kD, were calculated from the relative amounts of C-H
3
versus C-D hydroxylation products. The H NMR data also
afforded information about the distribution of hydroxylation at
each carbon atom of the substrate. These ratios of products,
C2/C1 and C4/C1, for ethane and butane, respectively, were
also determined. In alkanes labeled at the primary position,
the tritiated carbon is designated as C1. In the butanes labeled
at a secondary carbon, that carbon is referred to as C2.
The possible alcohol products of the reaction of (S)-[1-2H1,1-
3H]ethane with sMMO are depicted in Figure 2, and the
observed 3H NMR spectra are given in Figure 3. The 3H NMR
spectrum of the reaction products in Figure 3A contains four
major resonances at δ ) 1.19 (s), δ ) 3.65 (s), δ ) 3.67 (s),
and δ ) 4.84 (s) ppm. The resonance at δ ) 4.84 ppm was
Figure 3. 3H NMR analyses of the oxidation products of (S)-[1-2H1,1-
3H]ethane. (A) Spectrum of the underivatized alcohols. The far upfield
peak corresponds to oxidation at the unlabeled carbon (product 5 from
Figure 2). The narrower downfield peak at ∼3.65 ppm (inset) arises
from alcohols 3 and 4 from Figure 2. The upfield peak, split by the
integer-spin deuterium atom, originates from alcohols 1 and 2. The far
downfield peak is assigned to tritiated water. (B) Spectrum of the
mandelate derivatives. Resonances 1-4 (inset) correspond to the
mandelates of the like-numbered products in Figure 2. The large upfield
peak results from the mandelate of product 5.
3
assigned to HHO, and the singlet at δ ) 1.19 ppm to the
product of oxidation occurring at the unlabeled carbon atom.
The inset in Figure 3 shows an expansion of the two signals at
δ ) 3.65 and 3.67 ppm. The narrower, more downfield
resonance is attributed to the products of oxidation at the C-D
bond (RCHTOH products, where R ) CH3), whereas the
broader, more upfield signal is assigned as arising from
oxidation at the C-H bond (RCDTOH products). The splitting
of the latter signal is due to coupling to the deuterium atom
present in the molecule. A kinetic isotope effect can be
calculated from the ratio of alcohol products for hydroxylation
(52) Floss, H. G.; Tsai, M. D. AdV. Enzymol. 1979, 50, 243-302.
(53) Andres, H.; Morimoto, H.; Williams, P. G. J. Chem. Soc., Chem.
Commun. 1990, 627-628.

1824 J. Am. Chem. Soc., Vol. 119, No. 8, 1997
Valentine et al.
1.47 ppm correspond to the diastereomeric products formed from
hydroxylation at the unlabeled secondary carbon. The farthest
upfield resonance is due to hydroxylation of the primary carbon
more distant from the tritium label.
Analysis of Derivatives. Conversion of the ethanol products
to their ethyl mandelates54 can afford five 3H NMR resonances,
two sets arising from diastereomers and one from another
product. These signals are shown in Figure 3B. The products
arising from stereochemical inversion of (S)-ethane are labeled
2m and 4m, those with retention are labeled 1m and 3m. The
peaks correspond to mandelates of the alcohols with the same
number of designations in Figure 2. The loss of stereochemistry
was calculated by adding the areas of peaks 2m and 4m and
dividing by the combined areas of all four peaks. The value
determined was corrected for the %ee of the substrate, 88 (
2% for ethanes, 95% for primary labeled butanes, and 99% for
butanes labeled at the 2° carbon.52 The corrected percent
inversion is given by eq 4, where x is the true fraction which
has undergone inversion, e is the percent of the major substrate
enantiomer, f is the observed percent of products displaying
retention, and ee is the percent enantiomeric excess of the
substrate (e minus the percent of the minor substrate enanti-
omer).
x ) (e - f )/ee
(4)
3
The H NMR spectrum of the mandelate derivatives of the
hydroxylation products of (S)-[1-2H1,1-3H]butane are given in
Figure 4B (whole spectrum) and inset (expanded version). The
3
mandelates of the diastereomeric esters have H NMR signals
at δ ) 3.8-3.9 which were analyzed analogously to the ethane
products above. The 3H spectrum of the mandelate derivatives
of alcohols formed with (S)-[2-3H]butane is given in Figure 5B.
The signal at δ ) 4.82 ppm comes from the mandelate of (R)-
2-butanol, which indicates retention of stereochemistry. The δ
) 4.86 ppm peak is indicative of inversion to form (S)-2-butanol.
Table 1 summarizes the results for ethane hydroxylations, Table
2 reports results for oxidations of (R) and (S)-[1-2H1,1-3H]-
butane, and Table 3 lists the findings for butane substrates
labeled at the C2 position.
Figure 4. 3H NMR analyses of the oxidation products of (S)-[1-3H,1-
2H]butane. (A) Spectrum of the alcohol products. Tritiated water appears
far downfield. The signal at δ ) 3.7 ppm corresponds to the primary
alcohols CH3CH2CH2CHT(OH) and CH3CH2CH2CDT(OH). Other
signals are identified in the text. (B) The spectra of corresponding
mandelate derivatives. The inset is an expansion of the region of the
diastereomeric esters of the products arising from oxidation at the
labeled carbon.
Oxidation by Purified Components. Hydroxylations of
chiral ethane took place with an intramolecular kH/kD ratio of
3.4 ( 0.4 and a C2/C1 ratio, reflecting relative hydroxylation
rates at the unlabeled versus the labeled carbon atom, of 3.3:1.
The ethanol products displayed 72% retention of stereochem-
istry. For butanes labeled at the primary position, smaller
intramolecular isotope effects were observed, together with a
slightly higher ratio of hydroxylation at the unlabeled primary
carbon to the one bearing the tritium atom. The amount of
retention of stereochemistry was slightly larger for these alcohols
than for the ethanes. The differences between results from
oxidations of the (R)- and (S)-enantiomers of each substrate were
within experimental error.
Oxidation by Hydroxylase and B/R Mix. When H and a
mixture of B and R which contained other cellular proteins were
used to hydroxylate the chiral alkanes, very low or even inverse
(<1) intramolecular kinetic isotope effects were observed. This
phenomenon was determined to occur not because the protein
preferentially oxidized the C-D over the C-H bond but because
the product distributions were disturbed by a second enzymatic
process which was acting on, and even interconverting, the
product alcohols. Several factors were diagnostic of this
secondary process. In multiple trials, the results were more
variable than when purified proteins were used. A sign that
the product distributions from H + B/R mix suffered from a
process that masked their true values was that the amount of
of the C-H bond versus products hydroxylated at the C-D
bond. The sum of the areas of these two peaks can be compared
to the area at δ ) 1.19 ppm to obtain information about the
relative hydroxylation rates at the two different carbon atoms.
This value may reflect the relative reactivities of the C-H versus
C-D bonds under the assumption that, once bound to the active
site, ethane is free to rotate and present either of its carbon atoms
to the active hydroxylating species. Reaction at the tritium atom
3
will not be revealed in the H NMR spectrum but is expected
to be much slower. These experiments were carried out under
Vmax conditions, in which a saturating amount of substrate was
added to the enzyme, conditions which could lead to an
artificially low estimate of intermolecular isotope effects.
3
The H NMR spectrum of the alcohol products formed by
hydroxylation of (S)-[1-2H1,1-3H]butane with sMMO is given
in Figure 4A. The downfield singlet again corresponds to
tritiated water. The peaks at δ ) 3.6 ppm arise from
hydroxylation at the labeled carbon atom. The far upfield peaks
correspond to hydroxylation at C2, C3, and C4. Oxidation of
(S)-[2-3H]butane by H + B/R mix leads to the product spectrum
in Figure 5A. Tritiated water appears far downfield. The signal
at δ ) 3.8 ppm arises from hydroxylation at the labeled carbon
(CH3CT(OH)CH2CH3). The inset in Figure 5A shows the
remaining alcohol products. The resonance at δ ) 1.54 ppm
arises from hydroxylation at the primary carbon next to the
tritium label (CH2(OH)CHTCH2CH3). The two signals at δ )
(54) Parker, D. J. Chem. Soc., Perkin Trans. 2 1983, 83-88.

Tritiated Chiral Alkane Hydroxylation by sMMO
J. Am. Chem. Soc., Vol. 119, No. 8, 1997 1825
Table 3. Results of Oxidizing (R)-, (S)-, and Racemic
[2-3H]Butane with a Mixture of H, B, and R Containing Other
Cellular Proteins
substrate
(R)-[2-3H]butane
(S)-[2-3H]butane
racemate
% (S) alcohol
% (R) alcohol
C3/C2a
86.5 (retention)
13.5 (inversion)
2.6
5.4
4.3 (inversion)
95.7 (retention)
2.2
4.6
42.8
57.2
2.3
C1/C2a
4.8
3
a Calculated from the product alcohol H NMR spectrum.
Retention of stereochemistry was primarily seen with both
(R)-[2-3H]butane and (S)-[2-3H]butane (Table 3). Products
obtained from a racemic mixture of (R)-[2-3H]butane and (S)-
[2-3H]butane showed only a slight preference for hydroxylation
of one enantiomer over the other (Table 3). The product ratios
were 1:1.3.
Oxidations of C2D6 and Exchange Reactions. In order for
an exchange reaction of hydrogen atoms at the R carbon atom
to explain the anomalous results obtained with H + B/R mix,
the process would have to be stereoselective and reversible,
properties characteristic of an enzymatic reaction. The pos-
sibility that such an exchange could occur was tested by
oxidizing C2D6 both with H + B/R mix and with purified
component proteins. When the latter were used, less than 2%
of the ethanol was detected by GC/MS to be CD3CDHOH, the
rest being CD3CD2OH. Oxidation with H + B/R mix resulted
in ∼12-13% CD3CDHOH in the product mixture. The
exchange reaction is thus independent of the hydroxylation
reaction. Incubation of CD3CD2OH in a solution containing
cellular extract, NAD+, and NADH resulted in as much as 25%
exchange of deuterons for protons at the R carbon. Much less
exchange (∼5%) was observed when NAD+ was not added,
and the presence or absence of hydroxylase did not affect the
amount of exchange.
Discussion
Figure 5. 3H NMR analyses of the oxidation products of (S)-[2-3H]-
butane. (A) Spectrum of the alcohol products. Tritiated water appears
far downfield. The product of hydroxylation at the labeled carbon
appears at δ ) 3.8 ppm. The inset is an expansion of the upfield region,
as described in the text. (B) The spectra of corresponding mandelate
derivatives. The inset is an expansion of the region of the esters of the
products arising from oxidation at the labeled carbon. The peak at δ )
4.86 ppm corresponds to the mandelate of (S)-2-butanol, signifying
inversion of stereochemistry of the starting alkane. The signal at δ )
4.82 ppm arises from the (R)-2-butanol which results from retention
of stereochemistry.
The present results indicate that hydroxylation of both chiral
ethane and butane substrates by sMMO from M. capsulatus
(Bath) affords products which retain their stereochemistry to a
large degree (71-96%). These data, together with other results
in the literature, can be explained by a simple unifying
mechanism for the hydroxylation step, as described in the
following analysis.
Differences between H + B/R Mix and Purified Protein
Components. A comparison of the results summarized in
Tables 1, 2, S1, and S2 clearly indicates the disparity between
oxidations performed using purified sMMO component proteins
versus partially purified cellular extracts. Control oxidations
of C2D6 by these two systems as well as the C-H/C-D
exchange reactions observed with C2D5OH support the hypoth-
esis that a cellular protein other than sMMO hydroxylase
catalyzes exchange of the hydrogen atoms at the R carbon of
the product alcohol. One likely such enzyme is an alcohol
dehydrogenase (ADH), which catalyzes the reaction given in
eq 5. Further discussion of this process is supplied as
Supporting Information.
Table 1. Results of Oxidizing (R)- and (S)-[1-2H1,1-3H]Ethane
with Purified H, B, and R
substrate
(R)-ethane
(S)-ethane
a
kH/kD
3.8
3.3
72
3.0
3.4
71
C2/C1a
% retention
3
a Calculated from the product alcohol H NMR spectra.
Table 2. Results of Oxidizing (R)- and (S)-[1-2H1,1-3H]Butane
with Purified H, B, and R
NAD+ + RCH2OH h H+ + NADH + RCHO (5)
substrate
(R)-butane
(S)-butane
a
kH/kD
2.5
3.8
76
1.8
3.3
78
C4/C1a
Alkanes Are Not Highly Constricted in the Active Site
Prior to Hydroxylation. The reaction products obtained with
racemic [2-3H]butane indicate clearly that the active site does
not rigidly constrain the substrate, since there was only a slight
discrimination between the two enantiomers of this probe. The
exchange reaction described above would not affect secondary
alcohols. An extremely crowded active site would most likely
% retention
a Calculated from the product alcohol H NMR spectra.
3
retention of stereochemistry in products arising from oxidation
of the C-H bond was not the same as the amount of retention
in products from oxidation of the C-D bond. Details are
available as Supporting Information.

1826 J. Am. Chem. Soc., Vol. 119, No. 8, 1997
Valentine et al.
readily distinguish between an ethyl and a methyl substituent
at the reacting carbon atom.
observed and at a higher temperature, where the putative ethyl
radical would rotate even more rapidly. The hydroxylase
proteins from the two organisms are very similar in amino acid
sequence and structure, the main differences between the two
systems apparently being the rates of formation and decay of
intermediates in the reaction cycle, M. trichosporium OB3b
being the faster of the two.15,31 Our work with the radical clock
substrate probes implied a putative rebound rate constant >1013
s-1 for sMMO from M. capsulatus (Bath) and 3-6 × 1012 s-1
for sMMO from M. trichosporium OB3b,43 and the results have
been interpreted in terms of a nonsynchronous radical-like
mechanism,44 which is further elaborated below. Because some
of the experiments with the M. trichosporium OB3b enzyme
were carried out with identical batches of substrates reported
here, it is unlikely that the different results are due to variations
in the substrates or analytical procedures. When the copper-
containing particulate MMO was studied by using the same
methodology, hydroxylation was found to proceed with com-
plete retention of configuration at the chiral atom.59
An extremely large intermolecular isotope effect (50-100)
has recently been reported for the decay of the oxidizing
intermediate Q in the presence of CD4 versus CH4 for hydroxy-
lase from M. trichosporium OB3b.30 This very large isotope
effect was accompanied by intermolecular isotope effects in
product distributions of ∼20 and intramolecular isotope effects
of 4-13. The results were interpreted as signifying that,
following hydrogen atom abstraction as depicted in Figure 1,
the methyl radical reabstracts the protium or deuterium from
the iron-bound hydroxyl group, regenerating a different form
of the original high-valent intermediate. If the abstraction but
not the radical recombination step displays a large isotope effect,
this pathway might account for the discrepancy between kinetic
isotope effects for reactivity of the intermediate Q toward these
substrates versus the resulting product alcohols. No evidence
was presented that such a complicated process exists when
ethane or butane is used as substrate, and the implications of
such a process on the chiral ethane results were not addressed.
In the absence of a fourth isotope of hydrogen, chiral ethanes
are the closest analogs to methane for which absolute stereo-
chemistry can be determined.
Kinetic Isotope Effects. When chiral alkanes were hydroxy-
lated by the sMMO system reconstituted from the purified
components, intramolecular isotope effects in the range of 3-4
were determined, values similar to those obtained with other
substrates.43 Intramolecular kinetic isotope effects of ≈4 with
these chiral substrates are similarly observed with sMMO from
M. trichosporium OB3b.46 In the radical clock experiments with
M. capsulatus (Bath), an intramolecular isotope effect of ≈5
was interpreted as indicating that substantial C-H bond
breakage is involved in the hydroxylation step.43 With cyto-
chrome P-450 hydroxylations, kH/kD values of 7-10 have been
reported.32 Such an isotope effect is expected to be g1, the
precise value depending on the linearity of the O‚‚‚H-C unit in
the transition state56 and on whether the transition state is early
or late.57
The C2:C1 hydroxylation ratio depends upon the magnitude
of the intramolecular kinetic isotope effect at C1, referred to as
kH/kD, assuming that the hydroxylating species has equal access
to hydrogen atoms on both the C1 and C2 positions of chiral
ethane. In other words, ethane rotation would occur in the active
site, rendering both ends of the molecule accessible to the active
hydroxylating species. The ratio of C2/C1 products would thus
reflect the relative ability to hydroxylate C-H versus C-D
bonds, statistically corrected for the numbers of such bonds.
From the observed kH/kD ratios for the chiral ethanes, 3.8 and
3.0 (Table 1), we compute a C2:C1 ratios of 2.38 and 2.25,
respectively, smaller than the observed values of 3.3 and 3.4.
This difference might reflect secondary kinetic isotope effects
or some other factor not yet identified.
Calculation of an Apparent Rebound Rate Constant. The
use of many calibrated radical probes has led to several varying
values for the rate of the consensus oxygen rebound mechanism
in C-H bond functionalization processes. For chiral ethane
probes, the known barrier to rotation for an ethyl radical can
be used to calibrate the probe at different temperatures. These
data can be applied to the ratio of retention and inversion to
yield an apparent rebound rate constant. No racemization occurs
in the derivatization of the alcohols when the reaction takes
place below -10 °C.55
Mechanistic Implications. We now inquire what, given the
foregoing analysis, the experimental results would indicate if
the hydroxylation step were to occur with complete retention,
with racemization, or with complete inversion of stereochem-
istry. None of the manipulations, including the derivatization
procedures, scrambles the stereochemistry.55 In hydroxylations
of chiral ethane by pMMO,59 the products were treated by the
same methodology and 100% retention of configuration was
observed, providing evidence against scrambling during the
workup. Total inversion of stereochemistry would suggest a
concerted reaction mechanism in which the C-O bond was
formed on the face opposite to the C-H bond being broken,
inverting the remaining three C-X bonds (where X ) C or
H). Total retention would indicate that the hydroxylation
proceeds in a concerted manner with formal oxygen atom
insertion into a C-H bond or occurs in two successive steps:
C-O or C-metal bond formation at the alkyl carbon atom
followed by C-H bond cleavage or reductive elimination.
Racemization at the chiral carbon atom would be indicative of
a process involving a radical which had sufficient time to rotate
freely before the relatively slow recombination. Branched
pathways would complicate any of these interpretations.
Equation 2 yields a calculated value of krotation ) 1 × 1013
s-1 at 45 °C for the ethyl radical. Therefore, when we include
the data for our % retention/% inversion in eq 3, we calculate
that any radical capture process must occur at approximately
krebound ) 1 × 1013 s-1. This value indicates that the process is
occurring on a time scale approaching that of a single bond
vibration. Such an interpretation appears impossible, however,
if it corresponds to the lifetime of a discrete intermediate species.
We can calculate the maximum rate for any process of this type,
in which ∆Gq ) 0, i.e., the decomposition of a transition state,
to be 6.3 × 1012 s-1 at 318 K. This result leads us to conclude
either that our approach to calculating the rebound rate is not a
good approximation as believed or that the present consensus
oxygen rebound mechanism for C-H bond functionalization
is insufficient to account for hydroxylation of ethane by sMMO.
Comparisons with Literature. Very similar stereochemical
results were obtained when chiral alkane was converted to
product alcohols with sMMO from M. trichosporium OB3b.46
The products consistently had ≈2:1 retention versus inversion
of stereochemistry observed for both substrate enantiomers and
for both ethane and butane substrates.46,58 With sMMO from
M. capsulatus (Bath), greater retention of configuration was
(58) Froland, W. A.; Wilkinson, B.; Priestley, N. D.; Morimoto, H.;
Williams, P. G.; Lipscomb, J. D.; Floss, H. G. Manuscript in preparation.
(59) Wilkinson, B.; Zhu, M.; Priestley, N. D.; Nguyen, H.-H. T.;
Morimoto, H.; Williams, P. G.; Chan, S. I.; Floss, H. G. J. Am. Chem. Soc.
1996, 118, 921-922.
(55) Priestley, N.; Wilkinson, B.; Floss, H. G. Unpublished results.
(56) O’Ferrall, R. A. M. J. Chem. Soc. (B) 1970, 785-790.
(57) Ritchie, C. D. In Physical Organic Chemistry: The Fundamental
Concepts; Marcel Dekker: New York, 1990; pp 289-307.

Tritiated Chiral Alkane Hydroxylation by sMMO
J. Am. Chem. Soc., Vol. 119, No. 8, 1997 1827
substrate probes were consistent with either no substrate radical
formation or an extremely fast rebound step having a rate
constant on the order of 1013 s-1, too large really for formation
of a discrete radical intermediate.
One proposal for the mechanism by which the oxidizing
intermediate in sMMO reacts with substrate is depicted in the
upper pathway of Figure 6. This scheme should result in
complete retention of stereochemistry at the chiral carbon and
is therefore not supported by the current results as the only
mechanism at work.
Another explanation, perhaps the most satisfactory one, for
the observed product distributions is that the reaction proceeds
through a nonsynchronous concerted mechanism,40,44 as pre-
sented in the lower part of Figure 6. In this representation,
intermediate Q is depicted as a diiron(IV) dioxo species,
although other formulations are possible. The hydrocarbon
substrate is doubly activated by two iron-bound oxo groups,
either one of which could evolve into the hydroxyl group of
the product alcohol. In this scenario, O-H bond formation,
C-H bond breaking, and C-O bond formation occur in a
concerted but not totally synchronous manner. For the pathway
shown on the upper arrow, preferential C-O bond formation
at the oxygen atom that takes the hydrogen (1+2) would lead
to predominant retention of stereochemistry at the labeled
carbon. As shown on the bottom arrow, C-O bond formation
at the other oxygen atom would invert the stereochemistry
(1+3). Formation of a pentavalent carbon species65 and
pseudorotation could also cause inversion [bottom arrow, 1 +
(2 or 3) + 4]. This mechanism is consistent with the observed
kinetic isotope effects and explains the lack of ring-opened
products in the radical clock studies, since a free substrate radical
is never formed. It also explains the greater retention of
stereochemistry for the C2 chiral butane hydroxylation; steric
bulk limits backside attack and inversion of stereochemistry,
and it would also retard pseudorotation. Oxidation of primary
carbons preferentially over secondary over tertiary ones, the
opposite trend to that expected for a radical reaction, is also
explained by this mechanism, because a primary carbon atom
would have the least steric restrictions to formation of such a
highly ordered transition state. In the absence of direct
observation of substrate intermediates, diagnostic substrates such
as the tritiated chiral alkanes described here offer the most
information thus far about the characteristics of the transition
state.
Figure 6. (Top) One alternative for the oxidation of alkane by a high
valent iron-oxo intermediate without formation of a free substrate
radical. This mechanism features oxygen atom insertion into an iron-
bound alkane. In this scheme, protonation of one oxygen atom to form
water occurs before substrate activation. It is unclear how inverted
products could occur via this mechanism. (Bottom) A nonsynchronous
concerted mechanism for oxidation of an alkane by sMMO. In one
manifestation of this mechanism (upper arrow), the alkane molecule is
doubly activated by one iron and its bound oxygen atom, leading to
retention. Inversion can occur (bottom arrow) either by backside attack
from the other oxygen atom (1 + 3) or by pseudorotation of a
pentavalent carbon species [1 + (2 or 3) + 4]. This mechanism currently
has no precedence in non-heme iron model chemistry, in which primary
alkane hydroxylation by dioxygen does not occur. Instead, nonproduc-
tive decomposition pathways result.63,64
Given the above analysis, it is necessary to delineate a
mechanism for sMMO hydroxylation of small alkanes in such
a manner as to produce predominant, but not complete, retention
of stereochemistry at the oxidized center. It is possible that
there exists some radical or possibly carbocation character in
the substrate during the hydroxylation step which inverts to a
limited degree. This scenario might be interpreted in several
ways. There could be parallel reaction pathways, one involving
a substrate radical (as in Figure 1) and one, or more, not (as in
Figure 6). The degree to which a radical versus nonradical
mechanism is followed may be determined by how snugly the
substrate fits into the active site. Parallel mechanistic pathways
have similarly been invoked in porphyrin model systems to
explain the stereochemical consequences of hydroxylation
chemistry depending on steric interactions of the substrate with
the catalyst.60-62 Alternatively, abstraction of a hydrogen atom
to form a substrate radical might occur in every reaction. The
rate constant for the rebound reaction may be so large that only
a small quantity of the substrate radical has sufficient lifetime
for rotation about the C-C bond to occur before recombination
with the bound hydroxyl group. Results with radical clock
Acknowledgment. This work was supported by grants from
the National Institute of General Medical Science (GM32134
to S.J.L. and GM32333 to H.G.F.) and Shell (S.J.L.). The NTLF
was supported by the Biomedical Research Technology Pro-
gram, National Center for Research Resources, U.S. National
Institutes of Health under Grant P41 RR 01237, through Dept.
of Energy Contract DE-AC03-76SF00098 with the University
of California. We thank Professors M. Newcomb, S. N. Brown,
and M. Lieberman for helpful discussions.
(60) Groves, J. T.; Viski, P. J. Org. Chem. 1990, 55, 3628-3634.
(61) Groves, J. T.; Viski, P. J. Am. Chem. Soc. 1989, 111, 8537-8538.
(62) Groves, J. T.; Stern, M. K. J. Am. Chem. Soc. 1987, 109, 3812-
3814.
Supporting Information Available: A full discussion of
the oxidation of [1-2H1,1-3H]ethane and [1-2H1,1-3H]butane by
B/R mix (Tables S1 and S2) and the exchange process (Figure
S1) (8 pages). See any current masthead page for ordering and
Internet access instructions.
(63) Feig, A. L.; Becker, M.; Schindler, S.; Eldik, R. v.; Lippard, S. J.
Inorg. Chem. 1996, 35, 2590-2601.
(64) Feig, A. L.; Masschelein, A.; Bakac, A.; Lippard, S. J. J. Am. Chem.
Soc. 1997, 119, 334-342.
(65) Shestakov, A. F.; Shilov, A. E. J. Mol. Catal. A 1996, 105, 1-7.
JA963971G