Rearrangement-free hydroxylation of methylcubanes by a cytochrome P450: The case for dynamical coupling of C-H abstraction and rebound

Article abstract of DOI:10.1021/jacs.9b08064

The highly strained cubylmethyl radical undergoes one of the fastest radical rearrangements known (reported k = 2.9 × 10¹⁰ s^-1 at 25 °C) through scission of two bonds of the cube. The rearrangement has previously been used as a mechanistic probe to detect radical-based pathways in enzyme-catalyzed C-H oxidations. This paper reports the discovery of highly selective cytochrome P450-catalyzed methylcubane oxidations which notionally proceed via cubylmethyl radical intermediates yet are remarkably free of rearrangement. The bacterial cytochrome P450 CYP101B1 from Novosphingobium aromaticivorans DSM 12444 is found to hydroxylate the methyl group of a range of methylcubane substrates containing a regio-directing carbonyl functionality at C-4. Unlike other reported P450-catalyzed methylcubane oxidations, the designed methylcubanes are hydroxylated with high efficiency and selectivity, giving cubylmethanols in yields of up to 93%. The lack of cubane core ring-opening implies that the cubylmethyl radicals formed during these CYP101B1-catalyzed hydroxylations must have very short lifetimes, of just a few picoseconds, which are too short for them to manifest the side reactivity characteristic of a fully equilibrated P450 intermediate. We propose that the apparent ultrafast radical rebound can be explained by a mechanism in which C-H abstraction and C-O bond formation are merged into a dynamically coupled process, effectively bypassing a discrete radical intermediate. Related dynamical phenomena can be proposed to predict how P450s may achieve various other modes of reactivity by controlling the formation and fate of radical intermediates. In principle, dynamical ideas and two-state reactivity are each individually able to explain apparent ultrashort radical lifetimes in P450 catalysis, but they are best considered together.

Full text of DOI:10.1021/jacs.9b08064

Subscriber access provided by Karolinska Institutet, University Library
Article
Rearrangement-Free Hydroxylation of Methylcubanes by a Cytochrome
P450: The Case for Dynamical Coupling of C–H Abstraction and Rebound
Md Raihan Sarkar, Sevan D. Houston, Gregory Paul Savage, Craig M.
Williams, Elizabeth H. Krenske, Stephen G. Bell, and James J. De Voss
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.9b08064 • Publication Date (Web): 19 Nov 2019
Downloaded from pubs.acs.org on November 19, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 15
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Rearrangement-Free Hydroxylation of Methylcubanes by a
Cytochrome P450: The Case for Dynamical Coupling of C–H
Abstraction and Rebound
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Md. Raihan Sarkar,†,§,^ Sevan D. Houston,‡,§ G. Paul Savage,¶ Craig M. Williams,*,‡ Elizabeth H.
Krenske,*,‡ Stephen G. Bell*,† and James J. De Voss*,‡
‡
†Department of Chemistry, University of Adelaide, Adelaide, SA 5005, Australia; School of Chemistry and Molecular
¶
Biosciences, University of Queensland, Brisbane, QLD 4072, Australia; CSIRO Manufacturing, Ian Wark Laboratory,
Melbourne, VIC 3168, Australia.
ABSTRACT: The highly strained cubylmethyl radical undergoes one of the fastest radical rearrangements known (reported k = 2.9
× 1010 s–1 at 25 °C) through scission of two bonds of the cube. The rearrangement has previously been used as a mechanistic probe to
detect radical-based pathways in enzyme-catalyzed C–H oxidations. This paper reports the discovery of highly selective cytochrome
P450-catalyzed methylcubane oxidations which notionally proceed via cubylmethyl radical intermediates yet are remarkably free of
rearrangement. The bacterial cytochrome P450 CYP101B1 from Novosphingobium aromaticivorans DSM 12444 is found to
hydroxylate the methyl group of a range of methylcubane substrates containing a regio-directing carbonyl functionality at C-4. Unlike
other reported P450-catalyzed methylcubane oxidations, the designed methylcubanes are hydroxylated with high efficiency and
selectivity, giving cubylmethanols in yields of up to 93%. The lack of cube ring opening implies that the cubylmethyl radicals formed
during these CYP101B1-catalyzed hydroxylations must have very short lifetimes, of just a few picoseconds, which are too short for
them to manifest the side reactivity characteristic of a fully equilibrated P450 intermediate. We propose that the apparent ultrafast
radical rebound can be explained by a mechanism in which C–H abstraction and C–O bond formation are merged into a dynamically
coupled process, effectively bypassing a discrete radical intermediate. Related dynamical phenomena can be proposed to predict how
P450s may achieve various other modes of reactivity by controlling the formation and fate of radical intermediates. In principle,
dynamical ideas and two-state reactivity are each individually able to explain apparent ultrashort radical lifetimes in P450 catalysis,
but they are best considered together.
incipient cubylmethyl cation 3 is implicated in the acid-
catalyzed rearrangement of cubylmethanol 2 to homocubanol 5
INTRODUCTION
The ability of cytochrome P450s to carry out a range of
difficult chemical transformations has driven intense activity
across many disciplines of chemistry seeking to uncover,
understand, and engineer new realms of P450 reactivity.1
Studies of new P450–substrate pairings shed important new
light on the details of these enzymes’ complex catalytic
mechanisms.2 Chemical probes, in particular strained
hydrocarbon substrates, have played an important role in
determining the involvement of radical intermediates and the
(via 4); in this rearrangement, a 1,2-carbon–carbon bond
migration leads to the expansion of one edge of the cube.6a In
contrast, the cubylmethyl radical 7, generated by photolysis of
precursor 6, rearranges via scission of two or more of the cube
C–C bonds to give ladderene (9) and bis-cyclobutenyl (10)
radicals.7, The first two β-scission steps in the rearrangement
8
of 7 are among the fastest radical rearrangements known, with
rate constants of 2.9 × 1010 s–1 and >1.5 × 1011 s–1 at 25 °C.8b
Scheme 1. Distinct reactivities of methylcubane derivatives
in cationic versus radical manifolds
,
timing of the bond forming and breaking processes.3 4
This paper reports novel mechanistic insights which emanate
from unexpected behavior observed in P450-catalyzed
oxidations of methylcubanes. Cubane itself is a highly strained
but metabolically stable hydrocarbon5 which has recently been
introduced as a benzene bioisostere in drugs and agrichemicals.6
These applications are possible because the strong C–H bonds
of cubane (similar in strength to a benzene C–H bond)5a,6c,7 give
it a low susceptibility to P450-mediated oxidative drug
metabolism. Conversely, derivatives of methylcubane (1) are
susceptible to enzymatic degradation, including by P450s.
Putative intermediates in different enzyme-catalyzed oxidations
of 1 include – formally at least – the cubylmethyl cation and
radical, both of which are known to undergo rapid skeletal
rearrangements when generated chemically (Scheme 1). An
1
(a) Cationic rearrangement
“
”
fast
OH
OH
2
3
4
5
(b) Radical rearrangement
S
k =
2.9 x 1010 s–1
O
hn
N
O
7
6
8
k >
1.5 x 1011 s–1
slower
9
10
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 15
Scheme 2. Reported products of oxidations of methylcubane (1) by various MMO and cytochrome P450 enzymes
1
2
3
4
OH
OH
OH
HO
OH
HO
1
2
11
12
13
14
5
5
6
7
8
MMO M. cap.b,d
MMO M. tri.e
CYP2B1b,d
100%
40%
22%
54–78%
42–71%
0%
42%c
nr
0%
0%
0%
0%
nr
<1%
6–14%
0% total methylcubanols
18% total methylcubanols
38%
29%
13%
9–17%
1–13%
CYP2B1b,f
5–17%
10–20%
6–19%
7–20%
CYP2E1b,f
9
aM. cap: Methylococcus capsulatus (Bath), M. tri.: Methylosinus trichosporium OB3b, CYP: cytochrome P450. bProducts were derivatized
as acetates. cLater reassigned10a as 5. dRef. 9. eRef. 10b. fRef. 10a. nr: not reported. The total yields of products in these examples were low:
<25 μmoles total product or <1% yield reported for the CYP and M. cap. MMO reactions; ca. 5% yield for the M. tri. MMO reaction.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
These fast rearrangements, and their distinctive products,
have enabled methylcubane to be used as a mechanistic probe
substrate to distinguish between cationic and radical pathways
rebounds by recombining with the Fe(OH) group to give the
product Fe–alcohol complex. A key piece of evidence
supporting the involvement of radical intermediates in these
reactions is the detection of rearranged products of definite
radical-based origin; for example, the oxidation of norcarane
involves a cyclopropylcarbinyl radical ring opening, I → II.
These rearrangements, with known (chemically determined)
rate constants, have been used as clocks to quantify the rate
constants for P450 rebound processes.4
in enzyme-catalyzed C–H oxidation reactions.8d, , The
9 10
oxidations of methylcubane by methane monooxygenase
(MMO) and mammalian cytochrome P450 enzymes have been
studied in detail. Key examples are shown in Scheme 2. The
percentages listed in Scheme 2 represent product distributions
rather than yields. The oxidation of methylcubane by the MMO
from Methylococcus capsulatus (Bath) yielded solely
cubylmethanol 2, derivatized as the acetate for analysis.9 In
contrast, the oxidation of methylcubane by the MMO from
k =
2 x 108 s–1
P450
Methylosinus trichosporium OB3b gave
a mixture of
OH
cubylmethanol 2 (40%), methylcubanols 11–13 (18%), and a
“radical rearrangement product” assigned as ladderene 14
(42%).10b A subsequent report10a suggested that the “radical
rearrangement product” was in fact not ladderene 14 but
homocubanol 5, which would have likely been derived via a
cationic pathway.
I
II
This paper reports the discovery of the first high-yielding
cytochrome P450-catalyzed methylcubane oxidations. These
reactions ostensibly proceed via cubylmethyl radical
intermediates yet remarkably show no evidence for any skeletal
rearrangement. The reactions involve the bacterial cytochrome
P450 CYP101B1 from Novosphingobium aromaticivorans
DSM 12444, which is found to selectively hydroxylate the
methyl group of a series of methylcubane substrates containing
a directing group at the C-4 position relative to methyl. The
unexpectedly selective hydroxylations appear to indicate an
ultrafast rebound process. Rationalization of these results leads
to the proposal that a full understanding of P450 catalysis must
take into account not only the well-known contributions of
different spin states, but also the dynamics of the bond forming
and breaking events. In this picture, the lifetime of the radical
“intermediate” depends on the extent of dynamical coupling of
the C–H abstraction and C–O bond-forming processes, which
may be so closely separated in time that the intermediate is
bypassed entirely, or nearly so.
In the cytochrome P450-catalyzed reactions, mixtures of
products were obtained, comprising cubylmethanol 2,
methylcubanols 11–13, and sometimes homocubanol 5.10 Most
mammalian P450 systems used in these studies are known to
have spacious substrate binding pockets that would allow the
substrate to bind in multiple different orientations, explaining
the low regioselectivities. Likewise, the contrasting
regioselectivities displayed by the two MMOs in Scheme 2 can
be attributed to the larger substrate binding pocket of the M.
trichosporium enzyme as compared with the M. capsulatus
MMO.10b The contrasting results of the MMO and P450
methylcubane oxidations can be further rationalized in terms of
the different reactive intermediates (non-heme versus heme
iron) and degrees of substrate specificity of each system (small-
substrate specific MMO versus promiscuous P450).9,10
In all of these reported P450 methylcubane oxidations, the
turnover rates and yields of oxidized products were very low
(<1%). This, together with the volatilities of the reactant and
products, meant that the product mixtures could not easily be
quantified with accuracy. Hence, a definitive conclusion about
the mechanism(s) of methylcubane oxidation could not be
reached. This remains an unresolved problem, especially for
P450-catalyzed oxidations which appear to yield both radical-
and cation-derived products.
The prevailing mechanistic understanding of P450-catalyzed
hydroxylations is centered on a radical rebound mechanism.3 In
the classic rebound picture, a hydrogen atom is abstracted from
the substrate by an iron-oxo oxidant (Cpd I) to give an
intermediate carbon-centered radical.3a–d The radical then
ACS Paragon Plus Environment

Page 3 of 15
Journal of the American Chemical Society
group in the case of 20). Substrates 15–19 were produced from
RESULTS AND DISCUSSION
1
2
3
4
5
6
7
cubane-1,4-dicarboxylic acid, while 20 was synthesized from
the corresponding monomethyl ester (see Schemes S2 and S3
of the Supporting Information).
Substrate design and synthesis. Our previous work
demonstrated that C–H oxidations catalyzed by the N.
aromaticivorans CYP101B1 were more efficient and selective
when the substrate incorporated a carbonyl substituent.11 For
example, α-ionone, isobornyl acetate, cyclooctyl acetate, and 2-
adamantyl acetate were each oxidized at a C−H bond distal to
Enzyme assays and identification of products. The results
of in vitro assays of substrate binding (via heme iron spin state
shift assays) and catalytic turnover of 15–20 by the N.
aromaticivorans CYP101B1 enzyme are shown in Table 1,
Figure 1, and Figures S1–S3. Substrates 17–20 induced a 50%
shift of CYP101B1 to the high spin form, indicating a good fit
between these substrates and the active site. Substrate 15
induced a smaller shift, whereas 16 did not alter the spin state
appreciably. The ethylcubane derivative 20 produced a more
extensive shift of the CYP101B1 spin state compared with
methyl analogue 18.
the
carbonyl-containing
substituent.
Although
a
crystallographic structure of CYP101B1 is not yet available,
this regio-control presumably involves a specific interaction
between the carbonyl functionality and one or more hydrogen
bond donors in the CYP101B1 active site. In the present work,
the six methylcubane substrates 15–20 (Table 1) were designed
and synthesized, each possessing
functionality at the C-4 position of the cubane ring system to
promote oxidation at the methyl group (or substituted methyl
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a carbonyl-containing
Table 1. Substrate binding, turnover and coupling efficiency data for the CYP101B1 catalyzed oxidation of methylcubane
derivatives.a
CYP101B1
Substrate
%HS
heme
NADH oxidation rate
Product formation rate
Coupling (%)
30
251 ± 10
< 10
< 5
(15)
<5
60
50
85
70
39b
nd
nd
< 2.5
66
(16)
374 ± 10
446 ± 5
468 ± 7
690 ± 28
< 10
(17)
(18)
296 ± 29
189 ± 35
298 ± 32
41
(19)
43
(20)
Turnover activities were measured using a ArR:Arx:CYP101B1 concentration ratio of 1:10:1 (0.5 μM CYP enzyme, 50 mM Tris, pH 7.4).
(ArR and Arx are the protein redox partners required for CYP101B1 activity.11d–g) Rates are reported as the mean  S.D. (n  3 unless
otherwise noted) and are given in nmol.nmol-CYP–1.min–1. The coupling is the percentage of consumed NADH utilized for the formation of
products. nd: not determined due to lack of product formation. bMeasured by a single experiment.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 15
Little or no product was detected in enzyme turnover assays 18–20. The homocubanol 25 could in principle be formed either
1
2
3
4
5
6
7
8
of 15, 16, or of 17. The latter result was unexpected, given the
significant spin state change induced by binding of 17 to
CYP101B1. By contrast, the oxidations of the better matched
substrates 18–20 each gave significant quantities of a single
by an enzyme-catalyzed oxidation of 18 following a cationic
pathway, or by an acid-catalyzed rearrangement of the alcohol
21 post enzymatic hydroxylation of 18. GC co-elution
experiments with these product standards led to the assignment
of the major products of the oxidations of 18 and 20 as
cubylalkanols 21 and 23, respectively (Scheme 3b).12 The
oxidation product of substrate 19 was assigned as
cubylmethanol 22 by NMR and MS analysis following a larger
scale CYP101B1-catalyzed oxidation (vide infra). In the
oxidation of 18, no homocubanol 25 was detected, either by
GC-MS or by GC based comparison with the authentic material.
Likewise, no products with MS fragmentation patterns
analogous to that of homocubanol 25 were observed in the
CYP101B1-catalyzed oxidations of 19 or 20.
major product, corresponding to
a
regioselectively
monooxygenated species, as detected by GC and GC-MS
analysis. The rates of product formation varied from 189 to 298
nmol.nmol-CYP–1.min–1 The coupling efficiencies (i.e. the
.
proportion of consumed NADH reducing equivalents utilized
for the formation of oxygenated products) were good, ranging
from 41 to 66%. These rates and efficiencies are comparable to
those previously observed for CYP101B1 with carbonyl-
substituted adamantanes and norisoprenoids, which displayed
rates of 166–1350 nmol.nmol-CYP–1.min–1 and coupling
efficiencies of 48–99%.11d–f
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 3 (a) Possible products and (b) assigned products of
the CYP101B1-catalyzed oxidations of 18–20
The CYP101B1-catalyzed oxidations of 18 and 19 were
carried out on a larger scale using excess NADH in order to
enable NMR analysis of the metabolite (Figures S2 and S4).
GC-MS analysis of the extracted crude products showed that in
each case, a single major product was present, identical to the
one detected in the smaller scale experiments, and no detectable
substrate remained. Minor products (<10%) were now also
detectable, which were identified (using MS analysis and co-
elution experiments) as originating from ester hydrolysis,
transesterification, and further oxidation reactions involving the
substrates and/or their oxidized products (Figures S2–S4).
Figure 1 GC-MS analysis of the turnovers of (a) 18 and (b) 19 by
CYP101B1. The turnovers are shown in black, a product control
(21, see see Scheme 3b) is shown in blue, and a control turnover
where no NADH was added is shown in red. Peaks labeled with
asterisks represent impurities or products of non-enzymatic
hydrolysis/transesterification (see SI for details).
1
The H and 13C NMR data of the crude extract from the
oxidation of 18 agreed with the data of the authentic 21 product
standard. The 1H NMR spectrum showed a pair of characteristic
signals in the range 3.9–4.2 ppm, each integrating to 3H, as
expected for the intact 1,4-disubstituted cubane framework. A
2H singlet was present at 3.71 ppm, corresponding to the
hydroxymethyl CH2 group. If any rearranged products (namely
ladderenes or bis-cyclobutenes, cf. Scheme 1) had been formed
in these oxidations, their exo-methylene protons would have
been expected to resonate at approximately 4.5–4.7 ppm and
Identification of the products of the CYP101B1-catalyzed
oxidations of 18–20 was achieved by comparison with authentic
samples of the plausible oxidized derivatives 21 and 23–25
(Scheme 3a). The cubylalkanols 21–24 correspond to site-
selective hydroxylations of the methyl or ethyl C–H bonds of
ACS Paragon Plus Environment

Page 5 of 15
Journal of the American Chemical Society
their cyclobutene proton signals would have appeared in the
Several experiments were performed in an effort to
1
1
2
3
4
5
6
7
8
range 5.8–6.6 ppm. No such signals were present in the H
chemically generate the cubylmethyl radical 28 (derived from
substrate 18) and study its reactivity (Scheme 4 and S6–S8).
The expected rearranged products of 28, after trapping by a
hydrogen atom donor, are 29–31. In two key experiments,
shown in Scheme 4b, a Barton–McCombie deoxygenation of
imidazole thiocarbamate ester 32 (Bu3SnH/AIBN) and a Barton
decarboxylation of N-hydroxypyridine-2-thione ester 34 (h)
were performed. Both reactions led to complex mixtures of
products. Alkene(s) that may have been derived from a
rearrangement of radical 28 comprised <5% of the products
from both reactions observable by 1H NMR spectroscopy. The
major detectable products obtained from the reaction of 32 were
carbamate 33 and the alcohol 21 (40% combined), both of
which were likely formed through heterolytic pathways. In the
attempted Barton decarboxylation reaction of 34, a typical14
thioether byproduct 35 was formed in 12% yield, suggesting
that the cubylmethyl radical intermediate 28 was indeed
generated. By way of contrast, a Barton decarboxylation of the
homologous acid 36 (Scheme 4c) gave ethylcubane derivative
20 in 91% yield, indicating that unlike the cubylmethyl radical,
it is possible to efficiently introduce and trap a radical at a
primary carbon β to the cube through chemical means.
1
NMR spectra of the crude extracts. H NMR analysis of crude
22 from 19 revealed analogous signals to those observed for 21
with the addition of a 2H singlet at 2.68 ppm for the methylene
group adjacent to the ester moiety.
These larger scale experiments allowed the yield of 21 from
the oxidation of 18 to be calculated. Calibration of the levels of
substrate and product present after the oxidation revealed that
CYP101B1 had converted 93 ± 10% of methylcubane 18 into
cubylmethanol 21. This is much higher than the yields reported
previously for P450-catalyzed oxidations of parent
methylcubane (<1%, Scheme 2). It further indicates that 21
itself is not a good substrate for CYP101B1.13
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
There was no evidence for hydroxylation of any of the cubane
C−H bonds of 18–20 (Figure S4). To determine if CYP101B1
could indeed hydroxylate a cubane C–H bond, assays were
performed on methyl cubane-1-carboxylate (26). This substrate
proved capable of binding to the enzyme (20% HS) and
initiating the catalytic cycle (as measured by NADH oxidation)
but it furnished only trace amounts of an apparently
monooxygenated metabolite (Figures S2, S3).
CO2Me
Scheme 4. (a) Possible products formed from cubylmethyl
radical 28 in the presence of a hydrogen atom donor; (b)
Synthetic reactions designed to generate and trap
cubylmethyl radical 28 and its rearrangement products; (c)
Generation and trapping of a cubylethyl radical from 36
26
Chemical characterization of cubylmethyl radical
rearrangements. The initially formed intermediates in the
hydroxylations of substrates 18–20 would be expected to be the
corresponding 4-substituted cubylmethyl radicals. Given the
rapid skeletal rearrangement of the parent cubylmethyl radical
7 (k = 2.9  1010 s–1 at 25 °C, Scheme 1), it is very surprising to
find no rearranged products in the oxidations of 18–20.
In order to determine if the cubylmethyl radicals derived from
the 4-substituted methylcubane derivatives 18–20 are somehow
less reactive toward rearrangement than the parent cubylmethyl
radical, the effects of the substituents on the first -scission step
were computed with quantum chemistry (Figure 2).
Calculations with the high-accuracy CBS-QB3 method
predicted small barriers (G‡) of 3.2 and 3.0 kcal/mol for the -
scission of the 4-methyl and 4-ester substituted cubylmethyl
radicals 27 and 28, respectively, equivalent to the calculated
G‡ value for  itself (3.2 kcal/mol) and in very good agreement
with the measured rate constant for 7 (calc. k = 2.8 × 1010 s–1,
expt.
k
=
2.9
×
1010 s–1). These -scissions are
thermodynamically favored by >20 kcal/mol. The second -
scission (not shown), which is even faster, is downhill by a
further 17 kcal/mol. All three radicals’ rearrangements are
therefore predicted to be extremely fast and effectively
irreversible reactions. Carboxyl or carboxymethyl substituents
at the 4-position are not anticipated to slow the rearrangement
appreciably.
Figure 2. Barriers and energetics for the first -scission step in the
rearrangements of cubylmethyl radicals, calculated with CBS-QB3
(kcal/mol).
In general, the products of methylcubane rearrangements are
highly susceptible to further reactions such as polymerization.8a
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 15
This reactivity is likely to be compounded by the electron-
kcal/mol. Returning to the non-rearranged radical faces a
sizable kinetic barrier.
1
2
3
4
5
6
7
8
withdrawing groups present in 29–31. The absence of unraveled
cubanes among the major products of the synthetic reactions
depicted in Scheme 4b does not necessarily indicate that the
cubylmethyl radical 28 was not formed; it more likely reflects
the instability of the unraveled cubane toward further
degradation. Of course, similar fragility would be expected for
any unraveled cubane derivatives produced in the enzyme-
catalyzed oxidations of 18–20. However, if ladderene or
bicyclobutenyl derivatives had indeed been formed in the
enzyme-catalyzed processes but had quickly degraded, then
significant losses of material would have been observed. In
reality, the yields were high; e.g. 18 gave near quantitative
conversion to cubylmethanol 21. This sets the oxidations of
methylcubanes 18–20 apart from all other previously studied
P450-catalyzed methylcubane oxidations (Scheme 2). Their
high yields mean that the formation of rearranged products can
be ruled out with a level of confidence that was not possible in
the previous (very low yielding) methylcubane oxidations.
On this basis, it is concluded that the rearrangement rate
constants for the radicals derived from 18–20 in the active site
of CYP101B1 can be taken to be equivalent to the rate constant
measured for the parent cubylmethyl radical 7 in solution, i.e. 3
× 1010 s–1. The observed near-quantitative hydroxylation of 18–
20 implies that the rebound of the cubylmethyl radicals must
out-compete rearrangement by at least 1–2 orders of magnitude,
from which it follows that the rebound rate constants are  3 ×
1011–1012 s–1.
Typical reported rate constants3 for radical rebound in P450-
catalyzed hydroxylations are 2–3 × 1010 s–1. Therefore,
substrates 18–20 are the first examples of methylcubanes whose
apparent rebound rates can definitively be categorized as
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ultrafast.
Several
other
ultrafast
probes
(e.g.
methylphenylcyclopropyl type probes) have even faster
apparent rebound rate constants, up to 1.2 × 1013 s–1.3g,10a,23d The
large variations observed for rebound rates, spanning a factor of
103, has led to much speculation about the mechanism and the
nature of the radical “intermediate”. Rebound rate constants of
3 × 1011 – 1.2 × 1013 s–1 imply radical lifetimes on the order of
0.08–3 picoseconds. These lifetimes, especially those in the
sub-picosecond regime, are very short, almost bordering on the
lifetimes of transition states (TSs), which are typically <60
Mechanistic implications. The use of radical clock probes
to characterize enzyme mechanisms and to quantify rebound
rate constants involves several caveats. These are briefly
examined here as they relate to the mechanism of CYP101B1-
catalyzed hydroxylation of 18–20.
The first assumption is that the enzyme does indeed abstract
a hydrogen atom from the substrate to generate a carbon-
centered radical. The prevailing view of P450 catalysis is very
much in favor of radical formation21 and certainly, in the case
of 18–20, the absence of homocubanol products represents
strong evidence against a cationic pathway. Density functional
theory (DFT) calculations (see the Supporting Information)
using a heme–methylthiolate model of Cpd I suggest that a
P450 should have no special difficulty in abstracting a hydrogen
atom from methylcubane to form a cubylmethyl radical. The
energy profile for hydrogen atom abstraction by the model Cpd
I is similar to those predicted by DFT for a range of other C–H
bonds.22–24
,
femtoseconds.19 20 As a group, the radicals derived from
ultrafast P450 probe substrates seem to represent a special class
of “intermediate” which, unlike normal P450 radical
intermediates, do not survive long enough to equilibrate within
the active site.
Previous explanations for unexpected ratios of
rearranged to unrearranged products. Several theories have
previously been advanced to account for the paradox of ultrafast
rebound in P450-catalyzed oxidations.21 The most widely
accepted and versatile picture is the two-state model of
–
reactivity developed by Shaik and coworkers.22 24 In this
picture, the reactive oxidant Cpd I exists in two almost
equienergetic electronic states – a doublet (low spin, LS) and a
quartet (high spin, HS). After hydrogen atom abstraction has
taken place and the necessary reorientation of radical and
Fe(OH) group with respect to each other has occurred, the
energy profile for rebound differs depending on the spin state.
On the HS potential energy surface a barrier of approximately
4 kcal/mol or more is found, but on the LS surface the barrier is
smaller or even zero. This endows the HS intermediate with a
longer lifetime than the LS intermediate, potentially making the
HS intermediate the chief origin of any rearranged products and
the LS intermediate the chief origin of unrearranged products.
Apparent ultrafast rebound could be explained by this model if
the flux through a barrierless LS rebound pathway is much
greater than the flux through the HS pathway. Various factors
have been proposed to influence the LS vs HS partitioning,
including the donor ability of the radical, the geometry and
strength of the bond to the proximal ligand, and specific
A second assumption is that the rate constant for the ring
opening of the radical clock probe within the enzyme active site
is the same as in bulk solution. Conceivably, an enzyme might
impede the ring opening simply by physically constraining the
radical within the active site. However, this appears unlikely for
18–20, since CYP101B1 is able to specifically bind and
efficiently oxidize larger substrates such as 2-adamantyl
isobutyrate and cyclododecyl acetate, suggesting a sterically
undemanding active site.11f,g Newcomb et al.23c explored similar
ideas with a range of ultrafast probes; their findings with probes
that underwent little molecular motion upon rearrangement, and
with probes that underwent isotopically induced metabolic
branching, led to the conclusion that the rearrangements were
not slowed in the enzyme.15 Furthermore, calculations suggest
that the weak interaction of the radical with the (HO)Fe moiety
in the active site does not have a significant effect on the
rearrangement barrier for either cubylmethyl16 or
cyclopropylmethyl-type17 radicals.
contributions from the protein environment.1f,
25
One further assumption normally made is that the
rearrangement of the probe radical in the active site is
irreversible.18 For our cubylmethyl radicals, it seems very likely
that the rearrangement is irreversible within the active site. DFT
calculations (see SI) predict that the first -scission of the
cubylmethyl radical in the presence of a model Cpd II is
energetically downhill by 23 kcal/mol and the second -scission
(which is even faster than the first) is downhill by a further 17
DFT calculations (see SI) show that the two-state energy
profile for methylcubane hydroxylation is fairly typical for a
P450-catalyzed hydroxylation, with no unusual spin-state
related effects. After the intermediate cubylmethyl radical has
formed, its rebound onto FeOH to give the hydroxylated
product has a small potential energy barrier of 3–4 kcal/mol in
either spin state. The rearrangement of the cubylmethyl radical
has a similar barrier, 3–4 kcal/mol, on both surfaces.
ACS Paragon Plus Environment

Page 7 of 15
Journal of the American Chemical Society
Intinsically, therefore, the radical has comparable propensities substrate binding lead to the prediction that the C–H abstraction
1
2
3
4
5
6
7
8
toward rebound or rearrangement in either spin state. The exact
spin-state preference for CYP101B1 cannot easily be
determined. Whilst a mechanism in which one spin state
dominates cannot be ruled out, there is no obvious reason why
this should be assumed to be the case. Could an alternative
picture equally explain the apparent ultrafast rebound?
and rebound phases of the hydroxylation can be brought close
together in time to form a process that is “dynamically
concerted”, or nearly so.26
A
reaction consisting of two bond-forming/breaking
processes is defined as dynamically concerted if the time gap
between the first and second bond forming/breaking events is
less than 60 fs (this being the lifetime for a transition state with
a zero energy barrier).26 If the time gap is longer than 60 fs, the
reaction is defined as dynamically stepwise. As a paradigm for
understanding reactivity, this type of dynamical effect has been
rapidly gaining prominence, with relevance to both small-
Groves3g,h has drawn attention to the importance of cage
effects in determining the ratios of rearranged to unrearranged
products obtained from radical clock probes. Apart from the
rebound or rearrangement of the incipient caged radical
[R•…(HO)Fe], an alternative pathway is cage escape, where R•
and (HO)Fe become separated by a solvent molecule or active
site side chain. The escaped R• eventually rearranges before
recombining with (HO)Fe. For radicals that rearrange relatively
slowly, greater-than-expected ratios of rearranged to
unrearranged products can be produced if cage escape competes
with rebound. This behavior has been observed, for example,
for a series of bicyclic cyclopropylcarbinyl-type probes.3i The
hydroxylations of 18–20 display the exact opposite effect (not
as much rearrangement as expected), suggesting that in these
reactions the reverse situation is in play: namely, R• is kept
close to Fe(OH) by complementary enzyme–substrate binding.
Given that any slowing of the rearrangement due to steric
constraints or due to the R•…(HO)Fe interaction in the active
site is not supported by our evidence (discussed above), it seems
likely that the key to the lack of rearrangement is a specific
enzyme–substrate binding interaction, presumably involving
the carbonyl functionality of 18–20 and one or more hydrogen
bond donors in CYP101B1.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
molecule transformations and enzyme-catalyzed processes.20,
27
To illustrate how dynamical effects may lead to ultrafast
P450-catalyzed hydroxylation, Figure 3a shows the potential
energy surface for methylcubane oxidation by a heme–
methylthiolate model of Cpd I, calculated with DFT.16 This
surface represents a reaction occurring in a solvent continuum
in the HS state. The medium has a polarity similar to that of the
interior of an enzyme, but has no explicitly modeled solvent
molecules or functional groups around the substrate and Fe
complex. The energy surface is projected onto two coordinates
representing the C–H distance and the COH angle. Together
these two coordinates differentiate the hydrogen abstraction and
rebound processes, as shown by the arrows in Figure 3a.
Starting from the hydrogen abstraction TS at the lower left of
the diagram, the minimum energy path (arrow 1) leads downhill
onto a broad plain where the radical intermediate resides.
Before rebound can occur, the radical and OH group must
reorient so that the singly occupied orbital of the radical can
overlap with oxygen to form the C–O bond (arrow 2). A large
change of about 60° in the COH angle occurs during this
process. Movement along the reorientational coordinate leads
to a small energy barrier at the rebound TS and then a steep
downhill pathway to the alcohol product (arrow 3).
Dynamical coupling of hydrogen abstraction and radical
rebound. We propose that a more complete picture of P450
catalysis can be gained by considering not only two-state
reactivity but also the way in which the enzyme controls the
reaction trajectory. Some simple considerations about enzyme–
Figure 3. (a) Potential energy surface (high spin) for the hydroxylation of methylcubane by a model Cpd I, calculated with B3LYP-D3(BJ)/6-
31G(d)-LANL2DZ in implicit (CPCM) diethyl ether. (b) Hypothetical effect of merging the hydrogen atom abstraction and rebound steps
into a dynamically concerted process as a result of complementary enzyme–substrate binding. The surface in (b) was generated by raising
the energies of the points in the upper portion of (a) to simulate holding R• in proximity to Fe(OH).
An important aspect of this energy surface is that after the
hydrogen atom abstraction has taken place, some time will
elapse while the energy of the intermediate becomes
redistributed into the appropriate nuclear modes to enable the
radical and Fe(OH) to align appropriately for C–O bond
formation. The most important modes involve the repositioning
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 15
of R• with respect to Fe(OH) and the torsion around the Fe–OH reactants or underwent desaturation). A complete lack of side-
1
2
3
4
5
6
7
8
bond. During this energy redistribution phase, the radical would
roam easily around the shallow energy well. This period of time
is when the radical is most likely to undergo side reactions such
as skeletal rearrangement.
reactivity would likely require both the LS and HS reactions to
feature dynamically coupled hydrogen abstraction and rebound.
How might the enzyme enact dynamically coupled hydrogen
abstraction and rebound? There are likely many contributing
factors, but two seem especially important: firstly, controlling
the redistribution of the excess energy into the relevant
reorientational modes after hydrogen atom abstraction (i.e.,
motion of R• relative to Fe(OH) or torsion around the Fe–OH
bond), and secondly, setting up a specific approach trajectory
towards the hydrogen abstraction transition state. As is evident
in Figure 3b, trajectories that approach the hydrogen atom
abstraction transition state from a more oblique angle (e.g.
dashed line) are most likely to proceed entirely downhill
thereafter, bypassing the intermediate. This would entail
passing through transition structures that differ from the ideal
linear C–H–O alignment.31
In general, reactions occurring on energy surfaces with
shallow intermediates following a large initial energy barrier (of
which the reaction in Figure 3a is an example) are influenced
by non-statistical dynamics. Dynamically concerted pathways
enable a reaction to avoid competing fates for the intermediate
since the bond forming/breaking events are brought close
together in time, within approximately the period of a molecular
vibration.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
A dynamically concerted (or nearly so) reaction within the
confines of the CYP101B1 active site provides a way to explain
the apparent ultrafast rebound of 18–20 even without invoking
any preference for LS vs HS pathways. In the model reaction
(Figure 3a), once the hydrogen atom abstraction TS has been
passed, continued motion along the C–H abstraction coordinate
has a negligible potential energy barrier. However, within an
enzyme this type of motion would be restricted, either through
general steric confinement or through specific enzyme–
substrate interactions in the active site. Under these
circumstances, the flat region at the top left of the energy
surface is replaced by a wall, preventing R• from moving away
from Fe(OH). This scenario is represented schematically in
Figure 3b, where the energies of points in the upper portion of
the diagram from Figure 3a have been raised to simulate the
effect of these active site constraints. The reactive trajectories
on this surface are different from those in the model reaction:
they will be concentrated in a channel where R• remains in close
proximity to Fe(OH). The hydrogen atom abstraction and
rebound steps are now more tightly coupled, shortening the
lifetime of the intermediate [R•…(HO)Fe] (arrow 1 in Figure
3b). This minimizes the opportunity for the radical to undergo
side reactions (rearrangement in this case). The apparent
ultrashort radical lifetime is no longer paradoxical, since the
“cubylmethyl radical” formed during the reaction is chemically
different from the one formed in the solution-phase kinetic
studies in which the rearrangement rate was measured; it does
not have the opportunity to equilibrate.28
The high level of regioselectivity observed in the oxidations
of 18–20 highlights the importance of the carbonyl functionality
engaging in specific enzyme–substrate interactions to control
both the reaction mechanism and the site selectivity. The
oxidations of methylcubane by other P450s (Scheme 2) are
much less selective but they still produce some cubylmethanol
(2), potentially also by a dynamically coupled mechanism. The
formation of rearranged products cannot be ruled out in those
examples due to low recovery of materials but it appears likely
that in such low-yielding and unselective reactions, stepwise
pathways would also be expected, leading to some
rearrangement.
In general, a dynamics-centric picture of P450 reactivity
predicts that the species formed by hydrogen atom abstraction
from the substrate may appear to behave as either a typical
intermediate or more similar to a transition state, depending on
how tightly the enzyme controls the reactive trajectory.27 The
observed lifetime of the “intermediate” would be the average of
the lifetimes of all trajectories which individually could range
from stepwise toward dynamically concerted and thus explain
situations where mixtures of rearranged and unrearranged
products are formed from ultrafast probe substrates.32 The idea
of dynamically bypassing a radical intermediate in P450-
catalyzed hydroxylations was previously noted by Shaik et al.,
who suggested that it would represent an alternative to the two-
state reactivity paradigm.25 Functionally, a reaction in which
both dynamically concerted and dynamically stepwise
trajectories are possible would be equivalent to a reaction in
which a discrete intermediate favors different products on the
LS and HS surfaces. While two-state reactivity and dynamics
can each independently account for ultrafast rebound, it seems
more likely that the dynamical concepts are a complement to,
rather than a replacement for, the two-state reactivity paradigm.
Both are important.
The energy surfaces in Figure 3 depict the HS state. We
utilized the HS state for these calculations because HS is the
state most likely to follow a stepwise mechanism in the
conventional two-state reactivity paradigm.16 HS is therefore
the state in which dynamical effects would need to be strongest
if they were to be responsible for bringing about an apparently
concerted hydroxylation. However, the same general principles
about the reaction trajectory apply more or less equally to both
the HS and the LS states. Our calculations show that the two
states have similar general energetic features in methylcubane
oxidation. Even though the LS mechanism of a P450
hydroxylation is generally regarded as effectively concerted in
the sense that C–O bond formation often has a negligible energy
barrier, it still has the same reorientation phase in between
hydrogen abstraction and rebound and is therefore still capable
of experiencing a time delay between those two processes,
during which rearrangement could occur.29 This idea is
supported by recent molecular dynamics simulations30 on the
oxidation of propane, which showed that, out of 101 trajectories
of the model LS Cpd II run for 3.6 ps at 300 K in the gas phase,
46 trajectories gave the alcohol while one trajectory led to the
[R•…(HO)Fe] complex (the other trajectories reverted to
In addition to explaining the variation observed in rebound rates
with differing cyclopropyl radical clocks and why ultrafast
probes such as methylphenylcyclopropanes33 and their
contrained analogues23c appear to undergo less rearrangement
that expected, dynamical pictures can also be proposed to
explain a range of other aspects of cytochrome P450 reactivity.
For example, different degrees of dynamical concertedness
could explain why the mechanisms of oxidations catalyzed by
a certain P450 vary depending on the nature of the enzyme–
substrate pairing. In the oxidations of cyclopropyl-containing
fatty acids 37 (Scheme 5a) by P450BM3, long-chain (C16)
substrates underwent ring opening solely by a radical
ACS Paragon Plus Environment

Page 9 of 15
Journal of the American Chemical Society
mechanism whereas shorter-chain (C12) substrates (which were in preference from tightly coupled hydrogen abstraction +
1
2
3
4
a poorer fit in the active site) reacted through both radical and
cationic pathways.34 For the C12 substrates, the cation-derived
products are believed to result from electron transfer to the
heme Fe(OH) complex from the intermediate radical. A change
rebound for C16 substrates to less tightly coupled pathways for
C12 substrates would explain the formation of both cation- and
radical-derived products from the C12 substrates.
5
6
7
8
Scheme 5. Some aspects of P450 reactivity that may be explained by dynamical effects. (a) Substrate-dependent pathway
variations involving radical vs cation formation; (b) variations in rebound rates in the oxidations of diastereomeric substrates
proceeding via a common radical.
9
(a) Cytochrome P450BM3-catalyzed oxidations of cyclopropyl fatty acids
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
R1
HO2Cn(H2C)
R
P450BM3
P450BM3
OH
OH
OH
HO2Cn(H2C)
37
R
R1
n = 7, R = CH3
n = 7, R = C3H7
n = 9, R = C3H7
n = 7, R = CH3
only
OH
R
R
R1
radical pathway:
hydroxylation,
ring-opening
cationic pathway:
cis-trans isomerization,
dehydrogenation
(b) Cytochrome P450cam-catalyzed oxidations of a- and b-thujone
P450cam
R1 R2
OH
hydrogen
abstraction
O
O
O
O
rebound
a-thujone:
OH
kreb = 0.2 x 1010 s–1
b-thujone:
a-thujone 38:
40
kreb = 2.8 x 1010 s–1
R1 = H, R2 = Me
b-thujone 39:
same “intermediate”,
different rebound rate constants?
R1 = Me, R2 = H
great deal of scope for the generality of these dynamical effects
to be tested computationally by means of molecular dynamics
trajectory calculations and experimentally by careful design and
selection of probe/P450 pairs.
Related dynamical ideas could explain the unexpected
variations in rebound rates observed for certain probe substrates
whose structures differ in subtle stereochemical ways, e.g. the
oxidations of -thujone (38) and -thujone (39) by P450cam
(Scheme 5b).35 These oxidations ostensibly proceed via the
same radical intermediate (40), yet their rebound rates differ by
an order of magnitude. Differences in the geometry of the
P450–substrate interaction at the outset of the hydrogen
abstraction would be expected to define different reactive
trajectories characterized by distinct radical lifetimes, thereby
accounting for the apparent difference in rebound rates. Two-
state reactivity is less able to explain this behavior.
In sum, substrates bound with high specificity and held in an
optimal orientation relative to Cpd I could lead to a
hydroxylation mechanism that approaches dynamical
concertedness. Non-optimally bound substrates may instead
follow a trajectory that results in a slower rebound leading to a
stepwise reaction with a longer-lived radical intermediate.
Multiple orientations of the substrate could lead to a mixture of
both of these pathways, leading to experimental results
characteristic of either or both concerted and non-concerted
reactions. Irrespective of the specific assumptions on which our
analysis of the CYP101B1/methylcubane system is based, the
relevance of dynamical concepts to enzyme catalysis is
expected to be wide ranging.
Conversely, in other enzyme–substrate pairs, a tightly
controlled stepwise trajectory could be strategically used by
Nature to favor the formation of a discrete radical intermediate.
This would be a way to divert the radical toward alternative
fates prior to (or instead of) rebound. An example would be
P450-catalyzed dehydrogenation reactions, which are thought
to involve single-electron oxidation or hydrogen atom
CONCLUSIONS
abstraction of the initially formed radical.34, In this case,
36
We have demonstrated that a cytochrome P450 efficiently
and selectively hydroxylates the methyl group of methylcubane
derivatives designed to bind specifically to the active site of the
enzyme with the methyl group proximal to the reactive iron-oxo
moiety. Hydroxylation solely at the methyl group of a
methylcubane is very difficult to achieve using chemical
steering the course of the reaction through a stepwise pathway
would ensure a longer lifetime for the intermediate, giving it the
time to transfer an electron or hydrogen atom to the heme Fe–
OH group. These examples illustrate just a few of the possible
impacts of dynamical control over P450 reactivity. There is a
ACS Paragon Plus Environment

Journal of the American Chemical Society
methodologies, due to the fast skeletal rearrangements of the Present Address
Page 10 of 15
1
2
3
4
5
6
7
8
^ Department of Pharmaceutical Technology, University of Dhaka,
Dhaka, Bangladesh.
cubylmethyl radical (or cation) intermediate, yet the
CYP101B1 enzyme from N. aromaticivorans delivers
cubylmethanols in high yield without any evidence for
rearrangement of the cubane core. These unexpectedly efficient
hydroxylations are proposed to reflect a mechanistic paradigm
in which the enzyme controls the degree of dynamical
concertedness of the reaction. Bringing the C–H abstraction and
C–O bond formation steps close together in time effectively
bypasses the cubylmethyl radical intermediate, such that the
radical loses the opportunity to undergo otherwise fast side
reactions. This type of dynamical picture of P450 reactivity
represents a second possible explanation for apparent ultrafast
rebound rates, alongside the well tested two-state reactivity
paradigm. Both models may well be equally important. A
variety of examples of P450 catalysis are presented that
potentially can be explained in dynamical terms, which could in
future be probed by molecular dynamics trajectory calculations.
Overall, this work adds a further nuance to the complex
dynamical picture of P450 catalysis, which more broadly
involves coordinated motions of enzyme and substrate
occurring during substrate binding, oxidation, and product
release.
ORCID
Md. Raihan Sarkar: 0000-0003-4807-5071
Sevan Houston: 0000-0002-9309-7037
Paul Savage: 0000-0001-7805-8630
Craig Williams: 0000-0002-3834-7398
Elizabeth Krenske: 0000-0003-1911-0501
Stephen Bell: 0000-0002-7457-9727
James De Voss: 0000-0002-2659-5140
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Author Contributions
§These authors contributed equally.
Notes
CSIRO (G. P. S.) and UQ (C. M. W.) have a formal relationship
with Boron Molecular Pty Ltd who supply cubane intermediates.
ACKNOWLEDGMENT
We thank the University of Adelaide and the University of
Queensland (UQ) for financial support. Md.R.S. thanks the
University of Adelaide for an International PhD scholarship.
S.D.H. gratefully acknowledges the Northcote Trust and Britain-
Australia Society for their award of the Northcote Graduate
Scholarship. The financial support of the Australian Research
Council (FT140100355 to S.G.B, FT110100851 to C.M.W.,
DP180103047 to E.H.K.) is gratefully acknowledged. Computer
resources were provided by the National Facility of the Australian
National Computational Infrastructure through the National Merit
Allocation Scheme and by the University of Queensland Research
Computing Centre. Prof. P. V. Bernhardt and Mr. T. Vanden Berg
are thanked for collecting and solving, respectively, the X-ray
crystallographic structure of 1,4-diiodocubane.
ASSOCIATED CONTENT
Supporting Information
Complete experimental details, GC-MS, NMR, and MS data, and
computational methods and data. The Supporting Information is
available free of charge on the ACS Publications website.
AUTHOR INFORMATION
Corresponding Author
* stephen.bell@adelaide.edu.au; c.williams3@uq.edu.au;
e.krenske@uq.edu.au; j.devoss@uq.edu.au.
ACS Paragon Plus Environment

Page 11 of 15
Journal of the American Chemical Society
1
2
REFERENCES
D’Accolti, L.; Fusco, C.; Gandolfi, R.; Eaton, P. E.; Curci, R.
Oxyfunctionalization of non-natural targets by dioxiranes. 6. On the
selective hydroxylation of cubane. Org. Lett. 2009, 11, 3574. (d)
Fokin, A. A.; Lauenstein, O.; Gunchenko, P. A.; Schreiner, P. R.
Halogenation of cubane under phase-transfer conditions: single and
double C–H-bond substitution with conservation of the cage
structure. J. Am. Chem. Soc. 2001, 123, 1842.
1. (a) Ahalawat, N.; Mondal, J. Mapping the substrate recognition
pathway in cytochrome P450. J. Am. Chem. Soc. 2018, 140, 17743.
(b) Chen, K.; Zhang, S.-Q.; Brandenberg, O. F.; Hong, X.; Arnold, F.
H. Alternate heme ligation steers activity and selectivity in
engineered cytochrome P450-catalyzed carbene-transfer reactions. J.
Am. Chem. Soc. 2018, 140, 16402. (c) Guengerich, F. P. Mechanisms
of cytochrome P450-catalyzed oxidations. ACS Catal. 2018, 8,
10964. (d) Lewis, R. D.; Garcia-Borràs, M.; Chalkley, M. J.; Buller,
A. R.; Houk, K. N.; Kan, S. B. J.; Arnold, F. H. Catalytic iron-carbene
intermediate revealed in a cytochrome c carbene transferase. Proc.
Natl. Acad. Sci U. S. A. 2018, 115, 7308. (e) Zaragoza, J. P. T.; Yosca,
T. H.; Siegler, M. A.; Moënne-Loccoz, P.; Green, M. T.; Goldberg,
D. P. Direct observation of oxygen rebound with an iron-hydroxide
complex. J. Am. Chem. Soc. 2017, 139, 13640. (f) Ramanan, R.;
Dubey, K. D.; Wang, B.; Mandal, D.; Shaik, S. Emergence of
3
4
5
6
7
8
9
6. (a) Eaton, P. E. Cubanes: starting materials for the chemistry of the
1990s and the new century. Angew. Chem. Int. Ed. 1992, 31, 1421.
(b) Wlochal, J.; Davies, R. D. M.; Burton, J. Cubanes in medicinal
chemistry: synthesis of functionalized building blocks. Org. Lett.
2014, 16, 4094. (c) Chalmers, B. A.; Xing, H.; Houston, S.; Clark,
C.; Ghassabian, S.; Kuo, A.; Cao, B.; Reitsma, A.; Murray, C.-E. P.;
Stok, J. E.; Boyle, G. M.; Pierce, C. J.; Littler, S. W.; Winkler, D. A.;
Bernhardt, P. V.; Pasay, C.; De Voss, J. J.; McCarthy, J.; Parsons, P.
G.; Walter, G. H.; Smith, M. T.; Cooper, H. M.; Nilsson, S. K.;
Tsanaktsidis, J.; Savage, G. P.; Williams, C. M. Validating Eaton’s
hypothesis: cubane as a benzene bioisostere. Angew. Chem. Int. Ed.
2016, 55, 3580. (d) Wlochal, J.; Davies, R. D. M.; Burton, J.
Synthesis of novel amino acids containing cubane. Synlett 2016, 27,
919. (e) Auberson, Y. P.; Brocklehurst, C.; Furegati, M.; Fessard, T.
C.; Koch, G.; Decker, A.; La Vecchia, L.; Briard, E. Improving
nonspecific binding and solubility: bicycloalkyl groups and cubanes
as para-phenyl bioisosteres. ChemMedChem 2017, 12, 590. (f) Xing,
H.; Houston, S. D.; Chen, X.; Ghassabian, S.; Fahrenhorst-Jones, T.;
Kuo, A.; Murray, C.-E. P.; Conn, K.-A.; Jaeschke, K. N.; Jin, D.-Y.;
Pasay, C.; Bernhardt, P. V.; Burns, J. M.; Tsanaktsidis, J.; Savage, G.
P.; Boyle, G. M.; De Voss, J. J.; McCarthy, J.; Walter, G. H.; Burne,
T. H. J.; Smith, M. T.; Tie, J.-K.; Williams, C. M. Cyclooctatetraene:
a bioactive cubane paradigm complement. Chem. Eur. J. 2019, 25,
2729. (g) Houston, S. D.; Chalmers, B. A.; Savage, G. P.; Williams,
C. M. Enantioselective synthesis of (R)-2-cubylglycine including
unprecedented rhodium mediated C–H insertion of cubane. Org.
Biomol. Chem. 2019, 17, 1067. (h) Reekie, T. A.; Williams, C. M.;
Rendina, L. M.; Kassiou, M. Cubanes in medicinal chemistry. J. Med.
Chem. 2019, 62, 1078. (i) Mykhailiuk, P. K. Saturated bioisosteres of
benzene: where to go next? Org. Biomol. Chem. 2019, 17, 2839. (j)
Houston, S. D.; Fahrenhorst-Jones, T.; Xing, H.; Chalmers, B. A.;
Sykes, M. L.; Stok, J. E.; Farfan Soto, C.; Burns, J. M.; Bernhardt, P.
V.; De Voss, J. J.; Boyle, G. M.; Smith, M. T.; Tsanakstidis, J.;
Savage, G. P.; Avery, V. M.; Williams, C. M. The cubane paradigm
in bioactive molecule discovery: further scope, limitations and the
cyclooctatetraene complement. Org. Biomol. Chem. 2019, 17, 6790.
(k) Xing, H.; Houston, S. D.; Chen, X.; Jin, D.-Y.; Savage, G. P.; Tie,
J.-K.; Williams, C. M. Determining the necessity of phenyl ring -
character in warfarin. Bioorg. Med. Chem. Lett. 2019, 29, 1954.
7. Zhang, Q. L.; Chen, B. Z. DFT studies on the cascade
rearrangement reactions of the cubylcarbinyl radical. J. Phys. Org.
Chem. 2011, 24, 147.
8. (a) Eaton, P. E.; Yip, Y. C. The preparation and fate of
cubylcarbinyl radicals. J. Am. Chem. Soc. 1991, 113, 7692. (b) Choi,
S. Y.; Eaton, P. E.; Newcomb, M.; Yip, Y. C. Picosecond radical
kinetics. Bond cleavage of the cubylcarbinyl radical. J. Am. Chem.
Soc. 1992, 114, 6326. (c) Della, E. W.; Head, N. J.; Mallon, P.;
Walton, J. C. Homolytic reactions of cubanes. Generation and
characterization of cubyl and cubylcarbinyl radicals. J. Am. Chem.
Soc. 1992, 114, 10730. (d) Newcomb, M. Competition methods and
scales for alkyl radical reaction kinetics. Tetrahedron 1993, 49, 1151.
9. 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. S.
Regiochemical variations in reactions of methylcubane with tert-
butoxyl radicals, cytochrome P-450 enzymes, and a methane
monooxygenase system. J. Am. Chem. Soc. 1996, 118, 6547.
10. (a) Newcomb, M.; Shen, R.; Choi, S. Y.; Toy, P. H.; Hollenberg,
P. F.; Vaz, A. D. N.; Coon, M. J. Cytochrome P450-catalyzed
hydroxylation of mechanistic probes that distinguish between
radicals and cations. Evidence for cationic but not for radical
intermediates. J. Am. Chem. Soc. 2000, 122, 2677. (b) Jin, Y.;
Lipscomb, J. D. Probing the mechanism of C–H activation: oxidation
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
function
in
P450-proteins:
A
combined
quantum
mechanical/molecular mechanical and molecular dynamics study of
the reactive species in the H2O2-dependent cytochrome P450SP and
its regio- and enantioselective hydroxylation of fatty acids. J. Am.
Chem. Soc. 2016, 138, 6786.
2. See, for example: (a) Alkhalaf, L. M.; Barry, S. M.; Rea, D.; Gallo,
A.; Griffiths, D.; Lewandowski, J. R.; Fulop, V.; Challis, G. L.
Binding of distinct substrate conformations enables hydroxylation of
remote sites in thaxtomin D by cytochrome P450 TxtC. J. Am. Chem.
Soc. 2019, 141, 216. (b) Tsutsumi, H.; Katsuyama, Y.; Izumikawa,
M.; Takagi, M.; Fujie, M.; Satoh, N.; Shin-ya, K.; Ohnishi, Y.
Unprecedented cyclization catalyzed by a cytochrome P450 in
benzastatin biosynthesis. J. Am. Chem. Soc. 2018, 140, 6631. (c)
Nagel, R.; Peters, R. J. Diverging mechanisms: cytochrome-P450-
catalyzed demethylation and -lactone formation in bacterial
gibberellin biosynthesis. Angew. Chem. Int. Ed. 2018, 57, 6082. (d)
Dang, T.-T. T.; Franke, J.; Tatsis, E.; O’Connor, S. E. Dual catalytic
activity of a cytochrome P450 controls bifurcation at a metabolic
branch point of alkaloid biosynthesis in Rauwolfia serpentina.
Angew. Chem. Int. Ed. 2017, 56, 9440. (e) ) Grant, J. L.; Mitchell, M.
E.; Makris, T. M. Catalytic strategy for carbon−carbon bond scission
by the cytochrome P450 OleT. Proc. Natl. Acad. Sci. U. S. A. 2016,
113, 10049.
3. (a) Ortiz de Montellano, P. R. Substrate Oxidation by Cytochrome
P450 Enzymes. In Cytochrome P450: Structure, Mechanism, and
Biochemistry, 4th edition.; Ortiz de Montellano, P. R., Eds.; Springer:
Switzerland, 2015, Vol. 1, Ch 4. (b) Ortiz de Montellano, P. R.
Hydrocarbon hydroxylation by cytochrome P450 enzymes. Chem.
Rev. 2010, 110, 932. (c) Denisov, I. G.; Makris, T. M.; Sligar, S. G.;
Schlichting, I. Structure and chemistry of cytochrome P450. Chem.
Rev. 2005, 105, 2253. (d) Poulos, T. L. Heme enzyme structure and
function. Chem. Rev. 2014, 114, 3919. (e) Rittle, J.; Green, M. T.
Cytochrome P450 compound I: capture, characterization, and C-H
bond activation kinetics. Science 2010, 330, 933. (f) Groves, J. T.;
McClusky, G. A. Aliphatic hydroxylation via oxygen rebound.
Oxygen transfer catalyzed by iron. J. Am. Chem. Soc. 1976, 98, 859.
(g) Huang, X.; Groves, J. T. Beyond ferryl-mediated hydroxylation:
40 years of the rebound mechanism and C–H activation. J. Biol.
Inorg. Chem. 2017, 22, 185. (h) Cooper, H. L. R.; Groves, J. T.
Molecular probes of the mechanism of cytochrome P450. Oxygen
traps a substrate radical intermediate. Arch. Biochem. Biophys. 2011,
507, 111. (i) Austin, R. N.; Luddy, K.; Erickson, K.; Pender-Cudlip,
M.; Bertrand, E.; Deng, D.; Buzdygon, R. S.; van Beilen, J. B.;
Groves, J. T. Cage escape competes with geminate recombination
during alkane hydroxylation by the diiron oxygenase AlkB. Angew.
Chem. Int. Ed. 2008, 47, 5232.
4. Auclair, K.; Hu, Z.; Little, D. M.; Ortiz de Montellano, P. R.;
Groves, J. T. Revisiting the mechanism of P450 enzymes with the
radical clocks norcarane and spiro[2,5]octane. J. Am. Chem. Soc.
2002, 124, 6020.
5. (a) Biegasiewicz, K. F.; Griffiths, J. R.; Savage, G. P.;
Tsanaktsidis, J.; Priefer, R. Cubane: 50 years later. Chem. Rev. 2015,
115, 6719. (b) Doedens, R. J.; Eaton, P. E.; Fleischer, E. B. The bent
bonds of cubane. Eur. J. Org. Chem. 2017, 2627. (c) Annese, C.;
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 15
22. Shaik, S.; Cohen, S.; Wang, Y.; Chen, H.; Kumar, D.; Thiel, W.
of methylcubane by soluble methane monooxygenase from
Methylosinus trichosporium OB3b. Biochemistry 1999, 38, 6178.
11. (a) Bell, S. G.; Wong, L. L. P450 enzymes from the bacterium
Novosphingobium aromaticivorans. Biochem. Biophys. Res.
Commun. 2007, 360, 666. (b) Bell, S. G.; Dale, A.; Rees, N. H.;
Wong, L. L. A cytochrome P450 class I electron transfer system from
Novosphingobium aromaticivorans. Appl. Microbiol. Biotechnol.
2010, 86, 163. (c) Yang, W.; Bell, S. G.; Wang, H.; Zhou, W.;
Hoskins, N.; Dale, A.; Bartlam, M.; Wong, L. L.; Rao, Z. Molecular
characterization of a class I P450 electron transfer system from
Novosphingobium aromaticivorans DSM12444. J. Biol. Chem. 2010,
285, 27372. (d) Hall, E. A.; Bell, S. G. The efficient and selective
biocatalytic oxidation of norisoprenoid and aromatic substrates by
CYP101B1 from Novosphingobium aromaticivorans DSM12444.
RSC Adv. 2015, 5, 5762. (e) Hall, E. A.; Sarkar, M. R.; Lee, J. H. Z.;
Munday S. D.; Bell, S. G. Improving the monooxygenase activity and
the regio- and stereoselectivity of terpenoid hydroxylation using ester
directing groups. ACS Catal. 2016, 6, 6306. (f) Sarkar, M. R.; Hall,
E. A.; Dasgupta, S.; Bell, S. G. The use of directing groups enables
the selective and efficient biocatalytic oxidation of unactivated
adamantyl C-H bonds. ChemistrySelect 2016, 1, 6700. (g) Sarkar, M.
R.; Dasgupta, S.; Pyke, S. M.; Bell, S. G. Selective biocatalytic
hydroxylation of unactivated methylene C–H bonds in cyclic alkyl
substrates. Chem. Commun. 2019, 55, 5029.
P450 enzymes: their structure, reactivity, and selectivity—modeled
by QM/MM calculations. Chem. Rev. 2010, 110, 949.
1
2
3
4
5
6
7
8
23. (a) Ogliaro, F.; de Visser, S. P.; Cohen, S.; Sharma, P. K.; Shaik,
S. Searching for the second oxidant in the catalytic cycle of
cytochrome P450: a theoretical investigation of the iron(III)-
hydroperoxo species and its epoxidation pathways. J. Am. Chem. Soc.
2002, 124, 2806. (b) Meunier, B.; de Visser, S. P.; Shaik, S.
Mechanism of oxidation reactions catalyzed by cytochrome P450
enzymes. Chem. Rev. 2004, 104, 3947. (c) Newcomb, M.; Le Tadic,
M.-H.; Putt, D. A.; Hollenberg, P. F. An incredibly fast apparent
oxygen rebound rate constant for hydrocarbon hydroxylation by
cytochrome P-450 enzymes. J. Am. Chem. Soc. 1995, 117, 3312. (d)
Newcomb, M.; Le Tadic-Biadatti, M.-H.; Chestney, D. H.; Roberts,
E. S.; Hollenberg, P. F. J. Am. Chem. Soc. 1995, 117, 12085. (e)
Newcomb, M.; Toy, P. H. Hypersensitive radical probes and the
mechanisms of cytochrome P450-catalyzed hydroxylation reactions.
Acc. Chem. Res. 2000, 33, 449.
24. (a) Jin, S.; Makris, T. M.; Bryson, T. A.; Sligar, S. G.; Dawson,
J. H. Epoxidation of olefins by hydroperoxo–ferric cytochrome P450.
J. Am. Chem. Soc. 2003, 125, 3406. (b) Vaz, A. D.; McGinnity, D.
F.; Coon, M. J. Epoxidation of olefins by cytochrome P450: evidence
from site-specific mutagenesis for hydroperoxo-iron as an
electrophilic oxidant. Proc. Natl. Acad. Sci. U. S. A. 1998, 95, 3555.
(c) Vaz, A. D.; Pernecky, S. J.; Raner, G. M.; Coon, M. J. Peroxo-
iron and oxenoid-iron species as alternative oxygenating agents in
cytochrome P450-catalyzed reactions: switching by threonine-302 to
alanine mutagenesis of cytochrome P450 2B4. Proc. Natl. Acad. Sci.
U. S. A. 1996, 93, 4644. (d) de Visser, S. P.; Ogliaro, F.; Harris, N.;
Shaik, S. Multi-state epoxidation of ethene by cytochrome P450: a
quantum chemical study. J. Am. Chem. Soc. 2001, 123, 3037.
25. Shaik, S.; Cohen, S.; de Visser, S. P.; Sharma, P. K.; Kumar, D.;
Kozuch, S.; Ogliaro, F.; Danovich, D. The “rebound controversy”: an
overview and theoretical modeling of the rebound step in C–H
hydroxylation by cytochrome P450. Eur. J. Inorg. Chem. 2004, 207.
26. Yang, Z.; Houk, K. N. The dynamics of chemical reactions:
atomistic visualizations of organic reactions, and homage to van ’t
Hoff. Chem. Eur. J. 2018, 24, 3916.
27. For an example of enzymatic control over reaction trajectories
revealed by molecular dynamics trajectory simulations, see: Yang,
Z.; Yang, S.; Yu, P.; Li, Y.; Doubleday, C.; Park, J.; Patel, A.; Jeon,
B.; Russell, W. K.; Liu, H.; Russell, D. H.; Houk, K. N. Influence of
water and enzyme SpnF on the dynamics and energetics of the
ambimodal [6+4]/[4+2] cycloaddition. Proc. Natl. Acad. Sci. U. S. A.
2018, 115, E848.
28. In solution, the energy surface for hydroxylation would be
expected to lie somewhere in between the two extremes depicted in
Figures 3a and 3b. The solvent cage would impose some local
constraints, hampering the dissociation of the incipient radical
[R•…(HO)Fe], but perhaps not as tightly as an enzyme active site.
This scenario is relevant to C–H activation reactions catalyzed by
synthetic mimics of cytochrome P450s.
29. Computations on the LS and HS energy surfaces are generally
found to differ significantly only after the reorientation phase.23
30. (a) Elenewski, J. E.; Hackett, J. C. Ab initio dynamics of the
cytochrome P450 hydroxylation reaction. J. Chem. Phys. 2015, 142,
064307. For earlier MD simulations on a related hydroxylation
reaction, see: (b) Yoshizawa, K.; Kamachi, T.; Shiota, Y. A
theoretical study of the dynamic behavior of alkane hydroxylation by
a compound I model of cytochrome P450. J. Am. Chem. Soc. 2001,
123, 9806. For related MD simulations on MMO-catalyzed
hydroxylation, see: Guallar, V.; Gherman, B. F.; Miller, W. H.;
Lippard, S. J.; Friesner, R. A. Dynamics of alkane hydroxylation at
the non-heme diiron center in methane monooxygenase. J. Am.
Chem. Soc. 2002, 124, 3377.
31. These ideas bear some relationship to an earlier model of
concerted O insertion proposed by Newcomb and co-workers23d
involving a side-on approach of the ferryl oxygen to the C–H bond.
32. Individual trajectories for a given reaction can display a broad
range of time gaps ranging from dynamically concerted to
dynamically stepwise. For example, the time gap for a di-π-methane
rearrangement was shown to vary from 13 to 1160 fs depending on
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
12. The MS fragmentation patterns of 21 and 23 indicated that the
cubane skeleton remained intact (Figure S3).
13. Newcomb et al. previously reported a similar finding, showing
that cubylmethanol 2 was not a substrate for CYP2B4.10a
14. Newcomb, M.; Kaplan, J. Rate constants and Arrhenius function
for scavenging of octyl radical by an N-hydroxypyridine-2-thione
ester. Tetrahedron Lett. 1987, 28, 1615.
15. The structures of the cubylmethyl radical 7 and its ring-opened
products 8 and 9, computed as part of the CBS-QB3 calculations
(shown below), suggest that these two -scissions produce only a
relatively small change in the volume of space occupied by the
radical.
16. See the Supporting Information for details.
17. The possibility that the rate of rearrangement is affected by the
weak interaction of R• with (HO)Fe has been considered by several
groups. Calculations on rearrangements of cyclopropylmethyl type
radicals suggested that the R•…(HO)Fe interaction did not have a
significant effect on the rearrangement barrier (at least in model
systems lacking the steric contraints of an enzyme active site). See:
(a) Kumar, D.; de Visser, S. P.; Sharma, P. K.; Cohen, S.; Shaik, S.
Radical clock substrates, their C–H hydroxylation mechanism by
cytochrome P450, and other reactivity patterns: what does theory
reveal about the clocks’ behavior? J. Am. Chem. Soc. 2004, 126,
1907. (b) Jäger, C.; Hennemann, M.; Clark, T. The effect of a
complexed lithium cation on a norcarane-based radical clock. Chem.
Eur. J. 2009, 15, 2425.
18. Frey, P. A. Radicals in enzymatic reactions. Curr. Opin. Chem.
Biol. 1997, 1, 347.
19. Schramm, V. Transition states and transition state analogue
interactions with enzymes. Acc. Chem. Res. 2015, 48, 1032.
20. Patel, A.; Chen, Z.; Yang, Z.; Gutiérrez, O.; Liu, H.-w.; Houk, K.
N.; Singleton, D. A. Dynamically complex [6+4] and [4+2]
cycloadditions in the biosynthesis of spinosyn A. J. Am. Chem. Soc.
2016, 138, 3631.
21. An early theory, proposed by Newcomb et al.,10a posited that there
are two different oxidants – an ironoxo species and an iron-
hydroperoxo species – which insert O and OH+, respectively, into the
C–H bond in a concerted fashion. This idea was suggested to explain
how P450s could furnish both radical- and cation-derived products.
However, computational studies by Shaik, Thiel, and coworkers22
predict that the iron-hydroperoxo oxidant has insufficient reactivity
to effect C–H hydroxylation.
ACS Paragon Plus Environment

Page 13 of 15
Journal of the American Chemical Society
35. He, X.; Ortiz de Montellano, P. R. Radical rebound mechanism
which vibrational modes were thermally activated in the TS. See:
in cytochrome P-450-catalyzed hydroxylation of the multifaceted
radical clocks - and -thujone. J. Biol Chem. 2004, 279, 39479.
36. (a) Bell, S. G.; Zhou, R.; Yang, W.; Tan, A. B.; Gentleman, A. S.;
Wong, L. L.; Zhou, W. Investigation of the substrate range of
CYP199A4: modification of the partition between hydroxylation and
desaturation activities by substrate and protein engineering. Chem.
Eur. J. 2012, 18, 16677; (b) Wong, S. H.; Bell, S. G.; De Voss, J. J.
P450 catalysed dehydrogenation. Pure. Appl. Chem. 2017, 89, 841.
1
2
3
4
5
6
7
8
Jiménez-Osés, G; Liu, P.; Matute, R. A.; Houk, K. N. Competition
between concerted and stepwise dynamics in the triplet di--methane
rearrangement. Angew. Chem. Int. Ed. 2014, 53, 8664.
33. Atkinson, J. K.; Ingold, K. U. Cytochrome P450 hydroxylation of
hydrocarbons: variation in the rate of oxygen rebound using
cyclopropyl radical clocks including two new ultrafast probes.
Biochemistry 1993, 32, 9209.
34. Cryle, M. J.; Hayes, P. Y.; De Voss, J. J. Enzyme–substrate
complementarity governs access to a cationic reaction manifold in the
P450BM3-catalysed oxidation of cyclopropyl fatty acids. Chem. Eur.
J. 2012, 18, 15994.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 15
TOC graphic
1
cytochrome
P450
2
3
4
OH
MeO2C
MeO2C
(CYP101B1)
5
6
7
8
9
MeO2C
“ultrafast rebound”
no radical rearrangement
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
14

Page 15 of 15
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
15