Enzymatic assembly of carbon–carbon bonds via iron-catalysed sp 3 C–H functionalization

Article abstract of DOI:10.1038/s41586-018-0808-5

Although abundant in organic molecules, carbon–hydrogen (C–H) bonds are typically considered unreactive and unavailable for chemical manipulation. Recent advances in C–H functionalization technology have begun to transform this logic, while emphasizing the importance of and?challenges associated with selective alkylation at a sp³ carbon^1,2. Here we describe iron-based catalysts for the enantio-, regio- and chemoselective intermolecular alkylation of sp³ C–H bonds through carbene C–H insertion. The catalysts, derived from a cytochrome P450 enzyme in which the native cysteine axial ligand has been substituted for serine (cytochrome P411), are fully genetically encoded and produced in bacteria, where they can be tuned by directed evolution for activity and selectivity. That these proteins activate iron, the most abundant transition metal, to perform this chemistry provides a desirable alternative to noble-metal catalysts, which have dominated the field of C–H functionalization^1,2. The laboratory-evolved enzymes functionalize diverse substrates containing benzylic, allylic or α-amino C–H bonds with high turnover and excellent selectivity. Furthermore, they have enabled the development of concise routes to several natural products. The use of the native iron-haem cofactor of these enzymes to mediate sp³ C–H alkylation?suggests that?diverse haem proteins could serve as potential catalysts for this abiological transformation, and will facilitate the development of new enzymatic C–H functionalization reactions for applications in chemistry and synthetic biology.

Full text of DOI:10.1038/s41586-018-0808-5

Letter
https://doi.org/10.1038/s41586-018-0808-5
Enzymatic assembly of carbon–carbon bonds via
iron-catalysed sp3 C–H functionalization
ruijie K. Zhang1, Kai Chen1, Xiongyi Huang1, Lena Wohlschlager1,2, Hans renata1,3 & Frances H. Arnold1*
Although abundant in organic molecules, carbon–hydrogen (C–H)
Metal carbene sp3 C–H insertion in small-molecule catalysis, in
bonds are typically considered unreactive and unavailable for particular intermolecular and stereoselective versions of this reaction,
chemical manipulation. Recent advances in C–H functionalization typically relies on transition-metal complexes based on rhodium12, irid-
technology have begun to transform this logic, while emphasizing the ium13 and other metals14–16. Artificial metalloproteins for carbene C–H
importance of and challenges associated with selective alkylation at a insertion have been created by introducing an iridium-porphyrin into
sp3 carbon1,2. Here we describe iron-based catalysts for the enantio-, variants of apo haem proteins17. Although rare, there are a few exam-
regio- and chemoselective intermolecular alkylation of sp3 C–H ples of iron carbene sp3 C–H insertion. The iron-catalysed examples
bonds through carbene C–H insertion. The catalysts, derived from a use high temperatures (for example 80°C)18, are stoichiometric19 or
cytochrome P450 enzyme in which the native cysteine axial ligand has are restricted to intramolecular reactions20, which suggests there is a
been substituted for serine (cytochrome P411), are fully genetically high activation energy barrier to the insertion of an iron carbene into a
encoded and produced in bacteria, where they can be tuned by C–H bond. However, because the protein framework of an enzyme can
directed evolution for activity and selectivity. That these proteins impart substantial rate enhancements to reactions21 and even confer
activate iron, the most abundant transition metal, to perform this activity to an otherwise unreactive cofactor22, we surmised that directed
chemistry provides a desirable alternative to noble-metal catalysts, evolution could reconfigure a haem protein to overcome the barrier for
which have dominated the field of C–H functionalization1,2. The
laboratory-evolved enzymes functionalize diverse substrates
containing benzylic, allylic or α-amino C–H bonds with high
Nature’s alkyltransferase
a
turnover and excellent selectivity. Furthermore, they have enabled
Me
the development of concise routes to several natural products. The
use of the native iron-haem cofactor of these enzymes to mediate
sp3 C–H alkylation suggests that diverse haem proteins could serve
as potential catalysts for this abiological transformation, and will
facilitate the development of new enzymatic C–H functionalization
reactions for applications in chemistry and synthetic biology.
Biological systems use a limited set of chemical strategies to form
carbon–carbon (C–C) bonds during the construction of organic mol-
ecules3. Whereas many of these approaches rely on the manipulation of
functional groups, certain enzymes—including members of the radical
S-adenosylmethionine (SAM) family—can perform alkylation of sp3
C–H bonds. This is a versatile strategy for structural diversification, as
seen by its essential role in the biosynthesis of structurally varied natu-
ral products and cofactors4–6. However, known biological machineries
for this transformation are limited to enzymes that transfer a methyl
group5,6 or conjugate a radical acceptor substrate4,7 to specific mole-
cules, with methylation as a common mode of sp3 C–alkyl installation
by radical SAM enzymes (Fig. 1a).
O
Me
O
P
O
P
S
H
H
O
Ad
+
HO
OH
HO
OH
NH3
OR1
OR1
[4Fe-4S]
Cobalamin
Nature’s oxygenase
b
O2
OH
NAD(P)H
H
H
R2
R3
R2
R3
N
N
Fe
S
N
N
Cys
Evolved alkyltransferase
We sought to introduce a new enzymatic strategy for the alkylation
of sp3 C–H bonds. For our design, we drew inspiration from the most
widely used biological C–H functionalization transformation: C–H
oxygenation. Enzymes such as the cytochromes P450 accomplish C–H
oxygenation using a haem cofactor; their activities rely on the activation
of molecular oxygen for the controlled generation of a high-energy
iron-oxo intermediate that is capable of selective insertion into a sub-
strate C–H bond8. Analogously, we anticipated that the combination
of a haem protein and a diazo compound would generate a protein-
enclosed iron carbene species, and that this carbene could participate
in a selective C–H insertion reaction with a second substrate (Fig. 1b).
Although it has been shown that haem proteins are capable of perform-
ing carbene-transfer processes such as cyclopropanation and hetero-
atom–hydrogen bond insertions9–11, their functionalization of sp3 C–H
bonds is yet to be achieved.
R4
N2
R4
R3
H
H
R2
R3
R2
N
N
Fe
X
N
N
Fig. 1 | Enzymatic C–H functionalization systems. a, Methylation
catalysed by cobalamin-dependent radical SAM enzymes, as illustrated
by Fom3 in fosfomycin biosynthesis6. b, Oxygenation catalysed by
cytochrome P450 monooxygenase (top) and the proposed alkylation
reaction achieved under haem protein catalysis (bottom). Structural
illustrations are adapted from Protein Data Bank (PDB) ID 5UL4 (radical
SAM enzyme) and PDB 2IJ2 (cytochrome P450BM3). Ad, adenosyl; Cys,
cysteine; R, organic group; X, amino acid.
1Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, CA, USA. 2Present address: Institute of Food Technology, University of Natural Resources and Life
Sciences, Vienna, Austria. 3Present address: Department of Chemistry, The Scripps Research Institute, Jupiter, FL, USA. *e-mail: frances@cheme.caltech.edu
N A t U r e | www.nature.com/nature
© 2019 Springer Nature Limited. All rights reserved.

reSeArCH Letter
100
95
90
85
80
75
70
65
60
55
50
2,400
2,200
2,000
1,800
1,600
1,400
1,200
1,000
800
a
b
CO2Et
OMe
CO2Et
N2
H
H
2
Haem protein
OMe
MeO
MeO
1a
P450
3a
Variant
TTN
ND
13
P450BM3 wild type
P-4(A82L)
CYP119 wild type
ND
600
Globin
400
200
R. marinus NOD(Y32G)
HGG wild type
7
0
ND
ND
Mb(H64V/V68A/D122N)
Directed
evolution
Cyt c
R. marinus cyt c wild type
ND
In vitro assay
1.0-mmol scale
P411-CHF in E. coli lysate P411-CHF in E. coli cells
R. marinus cyt c
ND
ND
RT
4 °C
(V75T/M100D/M103E)
2,020 TTN
96.7:3.3 e.r.
82% yield, 1,060 TTN
98.0:2.0 e.r.
H. thermophilus cyt c
Fig. 2 | Haem-protein-catalysed sp3 C–H alkylation. a, A selected
data point is shown as a grey dot. Enantioselectivity data are represented
by green diamonds. Unless otherwise indicated, reaction conditions were
as follows: haem protein in E. coli whole cells (optical density at 600 nm,
OD600, of 30) (a) or in clarified E. coli lysate (b), 10 mM substrate 1a,
10 mM ethyl diazoacetate, 5 vol% EtOH in M9-N buffer at room
temperature (RT) under anaerobic conditions for 18 h; see Supplementary
Information for conditions of the 1.0-mmol reaction in b. Reactions
performed with lysate contain 1 mM Na2S2O4. TTN is defined as the
amount of indicated product divided by haem protein as measured by the
haemochrome assay. See Supplementary Information for the complete list
of haem proteins tested and detailed experimental procedures.
subset of haem proteins that were tested for promiscuous C–H alkylation
activity. Structural illustrations are of the following superfamily members
with the haem cofactor shown as red sticks: cytochrome P450BM3 (top,
PDB 2IJ2), sperm whale myoglobin (middle, PDB 1A6K) and R. marinus
cytochrome c (bottom, PDB 3CP5). cyt c, cytochrome c; HGG, Hell’s Gate
globin; H. thermophilus, Hydrogenobacter thermophilus; Mb, sperm whale
myoglobin; ND, not detected. b, Directed evolution of a cytochrome P411
for enantioselective C–H alkylation (reaction shown in a). Bars represent
mean TTN values averaged over four reactions (performed from two
independent cell cultures, each used for duplicate reactions); each TTN
the iron carbene C–H insertion reaction and acquire this new function product 3a with 60 TTN. Unlike the native monooxygenase activity,
(Fig. 1b). the C–H alkylation process does not require reducing equivalents
In initial studies, we tested a panel of 78 haem proteins that included from the FAD and FMN domains of these enzymes. Surmising that
variants of cytochromes P450, cytochromes c and globin homologues. these domains may not be needed for the C–H alkylation reaction,
The haem proteins in whole Escherichia coli cells were combined with we performed systematic truncations of P411-gen6 to determine the
p-methoxybenzyl methyl ether (1a) and ethyl diazoacetate (2) at room minimally sufficient domain(s) for retaining catalytic activity. Notably,
temperature under anaerobic conditions; the resulting reactions were removal of the FAD domain, which contains 37% of the amino acids in
analysed for the formation of C–H alkylation product 3a (Fig. 2a; the full-length protein, created an enzyme with higher C–H alkylation
see Supplementary Information for the complete list of haem proteins activity: P411∆FAD-gen6 delivers 3a with 100 TTN, a 1.7-fold increase
tested). Haem proteins from two superfamilies were found to show in TTN compared with P411-gen6 (Supplementary Fig. 2). This indi-
low levels of this promiscuous activity, establishing the possibility of cates that the FAD domain may have (negative) allosteric effects on the
creating C–H alkylation enzymes with very different protein architec- C–H alkylation activity. Further studies with these truncated enzymes
tures. Among the proteins tested were variants of cytochrome P450BM3 revealed that they could be used in whole E. coli cells, in clarified
from Bacillus megaterium in which the axial cysteine ligand is substi- E. coli cell lysate and as purified proteins (Supplementary Table 3). Eight
tuted for serine, known as cytochrome P411s31. We found that one of additional rounds of mutagenesis and screening yielded P411-CHF
these variants, P-4(A82L)22, which differs from the wild type by 18 (P411∆FAD C–H functionalization enzyme; for the full list of changes,
mutations, provided 3a with a total turnover number (TTN) of 13. In see Supplementary Information).
addition, nitric oxide dioxygenase from Rhodothermus marinus with a
P411-CHF displays a 140-fold improvement in activity over
tyrosine-to-glycine mutation at position 32 (R. marinus NOD(Y32G)) P-4(A82L) and delivers 3a with excellent stereoselectivity (2,020
catalysed the reaction with 7 TTN. A second alkane substrate, 4-ethy- TTN, 96.7: 3.3 enantiomeric ratio (e.r.) using clarified E. coli lysate).
lanisole (1i), was also accepted by the nascent C–H alkylation enzymes, Subsequent studies showed that the stereoselectivity could be improved
albeit with lower turnover numbers (Supplementary Table 2). The haem by conducting the reaction at lower temperature (for example, 4°C)
cofactor alone (iron protoporphyrin IX) or in the presence of bovine without substantial change to TTN (Supplementary Table 4). Enzymatic
serum albumin was inactive (Supplementary Tables 1 and 2).
C–H alkylation can be performed on a millimole scale: using 1.0 mmol
With P411 P-4(A82L) as the starting template, sequential rounds of substrate 1a, E. coli harbouring P411-CHF at 4°C furnished 3a in 82%
site-saturation mutagenesis and screening in whole E. coli cells were isolated yield, 1,060 TTN, and 98.0: 2.0 e.r. (Fig. 2b). Preliminary mech-
performed to identify increasingly active and enantioselective biocat- anistic investigations were pursued to investigate the nature of the C–H
alysts for C–H alkylation. Amino acid residues chosen for mutagenesis insertion step. Independent initial rates measured for reactions with
included those that line the active site pocket, reside on loops and other substrate 1a or deuterated substrate 1a-d2 revealed a normal kinetic
flexible regions of the protein, or possess a nucleophilic side chain23. isotope effect of 5.1 for C–H alkylation catalysed by P411-CHF, sug-
Improved variants were subsequently evaluated in reactions using gesting that C–H insertion is rate-determining (Supplementary Fig. 5).
clarified E. coli lysate with p-methoxybenzyl methyl ether (1a) and
Using E. coli harbouring P411-CHF, we assayed a range of ben-
4-ethylanisole (1i) (Fig. 2b and Supplementary Fig. 1). Five rounds of zylic substrates for coupling with ethyl diazoacetate (Fig. 3). Both
mutagenesis and screening yielded variant P411-gen6, which furnished electron-rich and electron-deficient functionalities on the aromatic
N A t U r e | www.nature.com/nature
© 2019 Springer Nature Limited. All rights reserved.

Letter reSeArCH
CO2Et
R
CO2Et
a
H
H
E. coli harbouring P411-CHF
+
Ar
R
M9-N buffer, RT
N2
Ar
CO2Et
OMe
CO2Et
OMe
CO2Et
CO2Et
OMe
OMe
F3C
MeO
Me
Br
3a, 2,150 TTN
3b, 840 TTN
3c, 410 TTN
96.5:3.5 e.r.
3d, 640 TTN
96.7:3.3 e.r.
92.9:7.1 e.r.
98.6:1.4 e.r.
CO2Et
OMe
CO2Et
OMe
CO2Et
CO2Et
Me
Me
O
O
Me
Si
H
3h, 930 TTNa
3e, 390 TTN
3f, 900 TTN
3g, 1,210 TTN
86.4:13.6 e.r.
89.0:11.0 e.r.
69.0:31.0 e.r.
97.9:2.1 e.r.
CO2Et
Me
CO2Et
Me
CO2Et
Me
CO2Et
Me
Me
Me
MeO
MeO
Me
3i, 530 TTN
3j, 50 TTN
3k, 340 TTN
3l, 140 TTN
97.9:2.1 e.r.
>99:1 e.r.
96.1:3.9 e.r.
97.4:2.6 e.r.
CO2Et
H
H
b
H
H
E. coli harbouring
P411 variant
CO2Et
or
M9-N buffer, RT
MeO
MeO
MeO
1m
3m (C–H insertion)
3m′ (Cyclopropanation)
+
P411-CHF
P-I263F
Enzyme-controlled
reactivity
CO2Et
N2
3m, 100 TTN
>99:1 e.r.
3m′, 280 TTN
59:41 d.r.b
>25:1
>49:1
2
Fig. 3 | Substrate scope for benzylic C–H alkylation with P411-CHF.
a, Experiments were performed using E. coli expressing cytochrome
P411-CHF (OD600 = 30) with 10 mM substrate 1a–1l and 10 mM ethyl
diazoacetate at room temperature under anaerobic conditions for 18 h;
reported TTNs are the average of four reactions (performed from
two independent cell cultures, each used for duplicate reactions). See
Supplementary Fig. 12 for the full list of alkane substrates. aSi–H insertion
product 3hʹ is also observed (Supplementary Fig. 7). b, Reaction selectivity
for carbene C–H insertion or cyclopropanation can be controlled by the
protein scaffold. Experiments were performed as in a using the indicated
P411 variant. bDiastereomeric ratio (d.r.) is given as cis:trans; e.r. was not
determined.
ring are well-tolerated (3a–3e, 3h); cyclic substrates are also suitable that C–H alkylation product 3m dominated, with selectivity > 25:1
coupling partners (3f, 3g). The functionalization of alkyl benzenes at (Fig. 3b, Supplementary Fig. 6). By contrast, a related full-length P411
secondary benzylic sp3 C–H bonds was found to be successful (3i–3l). variant P-I263F, containing 13 mutations in the haem domain relative
Notably, in the biotransformation of substrate 1l, which contains both to P411-CHF, catalysed only the formation of cyclopropane product
tertiary and secondary benzylic C–H bonds, P411-CHF preferentially 3mʹ. Additionally, despite the established reactivity of silanes with iron
functionalizes the secondary position despite its higher C–H bond dis- carbenes10, P411-CHF delivered C–H alkylation product 3h when sub-
sociation energy. The carbene intermediate derived from ethyl diazoac- strate 1h was used in the reaction (Si–H insertion product 3hʹ was
etate belongs to the acceptor-only class. Compared to the more widely also observed, but its formation may not be catalysed by P411-CHF,
used donor/acceptor carbenes, acceptor-only intermediates are more Supplementary Fig. 7). Reaction with P-I263F, by contrast, provided
electrophilic, and as a result selective reactions with this carbene class only the Si–H insertion product. These examples demonstrate an
are still a major challenge for small-molecule catalysts13,16. Our results exceptional feature of macromolecular enzymes: different products
show that P411-CHF can control this highly reactive intermediate to can be obtained simply by changing the amino acid sequence of the
furnish the desired sp3 C–H alkylation products, and does so with high protein catalyst.
enantioselectivity.
Enzymatic C–H alkylation is not limited to the functionalization of
Enzymes can exhibit excellent reaction selectivity arising from benzylic C–H bonds. Structurally dissimilar molecules containing ally-
their ability to form multiple interactions with substrates and inter- lic or propargylic C–H bonds are excellent substrates for this chemistry
mediates throughout a reaction cycle. We proposed that the protein (Fig. 4a). In contrast to 1a–1m, which contain a rigid benzene ring,
scaffold could be tuned to create complementary enzymes that can compounds 4a–4c and 4e feature flexible linear alkyl chains. Their suc-
access different reaction outcomes available to a substrate. When P411- cessful enantioselective alkylation suggests that the enzyme active site
CHF was challenged with 4-allylanisole (1m)—a substrate that can can accommodate substrate conformational flexibility while enforcing
undergo both C–H alkylation and cyclopropanation—we observed a favoured substrate orientation relative to the carbene intermediate.
N A t U r e | www.nature.com/nature
© 2019 Springer Nature Limited. All rights reserved.

reSeArCH Letter
O
O
H
H
R3
R1
R3
E. coli harbouring P411-CHF
R4
R4
+
R1
R2
M9-N buffer, RT
N2
R2
a
Allylic and propargylic substrates
CO2Et
Me
Me
Br
Me
CO2Et
OMe
CO2Et
OMe
CO2Et
CO2Et
Me
OMe
OMe
MeO
5d, 70 TTNa
97.0:3.0 e.r.
5e, 190 TTNb
99.0:1.0 e.r.
5a, 3,750 TTN
93.6:6.4 e.r.
5b, 950 TTN
94.7:5.3 e.r.
5c, 2,760 TTN
96.3:3.7 e.r.
CO2Et
N2
HO2C
Me
Me
2
Me
Me
Decarboxylative
alkenylation
2 stepsc
CO2Et
OMe
CO2H
OMe
E. coli harbouring P411-CHF
86% yield
471 mg at 2.4-mmol scale
Quantitative
ref. 25
OMe
OMe
4a
(+)-5a, 2,810 TTN
(+)-6
(+)-Lyngbic acid
94.7:5.3 e.r.
b
Alkyl amines
Me
OMe
Cl
Me
CO2Et
CO2Et
N
N
N
N
N
Me
Me
CO2Et
CO2Et
CO2Et
8a, 2,330 TTN
8b, 2,230 TTN
8c, 2,030 TTN
82.8:17.2 e.r.
8d, 2,150 TTN
83.7:16.3 e.r.
8e, 500 TTN
90.3:9.7 e.r.
E. coli harbouring P411-CHF
See below
CO2Et
OMe
42% yield
N
N
2
49 mg at 0.5-mmol scale
Me
Me
N
OMe
(–)-8f, 1,050 TTNd
(–)-Cuspareine
Me
1
73.0:27.0 e.r.
7f
+
3 stepse
50% yield overall
CO2Et
E. coli harbouring P411-gen5
N2
CO2Et
OMe
OMe
2
N
N
85% yield
Me
Me
(+)-Cuspareine
594 mg at 3.0-mmol scale
(+)-8f, 1,310 TTN
91.1:8.9 e.r.
c
Alternative diazos
Me
O
Me
Me
O
O
O
Me
OMe
OMe
N
OBn
N
N
N
O
Me
Me
Me
Me
MeO
10b, 280 TTNf
10d, 150 TTN
10a, 360 TTN
10c, 500 TTNf
78.0:22.0 e.r.
71.0:29.0 e.r.
Fig. 4 | Application of P411 enzymes for sp3 C–H alkylation. a, Allylic
and propargylic C–H alkylation. Unless otherwise indicated, experiments
were performed using E. coli expressing cytochrome P411-CHF with
10 mM substrate 4a–4e and 10 mM ethyl diazoacetate; reported TTNs
are the average of four reactions (performed from two independent cell
cultures, each used for duplicate reactions). aTTN was calculated on the
basis of isolated yield from a reaction performed on a 0.25-mmol scale.
bThe cyclopropene product was also observed (Supplementary Fig. 8).
cHydrogenation, followed by hydrolysis. b, Enzymatic alkylation of
substrates containing α-amino C–H bonds. Unless otherwise indicated,
reactions were performed on a 0.5-mmol scale using E. coli expressing
cytochrome P411-CHF with substrates 7a–7f and ethyl diazoacetate;
TTNs were calculated on the basis of isolated yields of products shown.
dIsolated in 9:1 r.r. for 8f:8fʹ determined by 1H nuclear magnetic resonance
analysis; TTN is reported for the sum of the regioisomers. eReduction,
halogen exchange and Suzuki–Miyaura cross-coupling. c, Enzymatic C–H
alkylation with alternative diazo reagents. Unless otherwise indicated,
reactions were performed on a 0.5-mmol scale using E. coli expressing
cytochrome P411-CHF with coupling partner 1a or 7a and diazo
compounds 9a–9d; TTNs were calculated on the basis of isolated yields of
products shown. fVariant P411-IY(T327I) was used. See Supplementary
Information for the complete list of substrates (Supplementary Figs. 12,
13), information about enzyme variants, and full experimental details.
N A t U r e | www.nature.com/nature
© 2019 Springer Nature Limited. All rights reserved.

Letter reSeArCH
To demonstrate the utility of this biotransformation, we applied the haem protein diversity can be leveraged to generate families of C–H
methodology to the formal synthesis of lyngbic acid (Fig. 4a). Marine alkylation enzymes that emulate the scope and selectivity of nature’s
cyanobacteria incorporate this versatile biomolecule into members of C–H oxygenation catalysts.
the malyngamide family of natural products; likewise, total-synthesis
approaches to malyngamides typically access lyngbic acid as a strategic
Data availability
intermediate en route to the target molecules24. Using E. coli harbouring
All relevant data are provided in Supplementary Information. Any additional infor-
P411-CHF, intermediate 5a was produced on a 2.4-mmol scale in 86%
isolated yield, 2,810 TTN, and 94.7:5.3 e.r. Subsequent hydrogenation
and hydrolysis provided (R)-(+)-6 in quantitative yield, which can be
transformed to (R)-(+)-lyngbic acid by decarboxylative alkenylation25.
As part of our investigation into the substrate scope of the reaction,
we challenged P411-CHF with alkyl amine compounds. Compounds
of this type are typically challenging substrates for C–H functionaliza-
tion methods because the amine functionality may coordinate to and
inhibit the catalyst or undergo undesirable side reactions (for exam-
ple, ylide formation and its associated rearrangements)26. Using 7a or
7b, substrates that have both benzylic C–H bonds and α-amino C–H
mation is available from the corresponding author on request.
Received: 18 July 2018;Accepted: 1 November 2018;
Published online 19 December 2018.
1. Hartwig, J. F. & Larsen, M. A. Undirected, homogeneous C–H bond
functionalization: challenges and opportunities. ACS Cent. Sci. 2, 281–292
(2016).
2. Saint-Denis, T. G., Zhu, R.-Y., Chen, G., Wu, Q.-F. & Yu, J.-Q. Enantioselective
C(sp3)‒H bond activation by chiral transition metal catalysts. Science 359,
eaao4798 (2018).
3. Frey, P. A. & Hegeman, A. D. Enzymatic Reaction Mechanisms Ch. 14 (Oxford
Univ. Press, New York, 2007).
bonds, P411-CHF delivered the corresponding β-amino ester prod- 4. Yokoyama, K. & Lilla, E. A. C–C bond forming radical SAM enzymes involved in
the construction of carbon skeletons of cofactors and natural products. Nat.
Prod. Rep. 35, 660–694 (2018).
5. Bauerle, M. R., Schwalm, E. L. & Booker, S. J. Mechanistic diversity of radical
S-adenosylmethionine (SAM)-dependent methylation. J. Biol. Chem. 290,
3995–4002 (2015).
6. McLaughlin, M. I. & van der Donk, W. A. Stereospecifc radical-mediated
B12-dependent methyl transfer by the fosfomycin biosynthesis enzyme Fom3.
Biochemistry 57, 4967–4971 (2018).
7. Shisler, K. A. & Broderick, J. B. Glycyl radical activating enzymes: structure,
mechanism, and substrate interactions. Arch. Biochem. Biophys. 546, 64–71
(2014).
8. Poulos, T. L. Heme enzyme structure and function. Chem. Rev. 114, 3919–3962
(2014).
9. Coelho, P. S., Brustad, E. M., Kannan, A. & Arnold, F. H. Olefn cyclopropanation
via carbene transfer catalyzed by engineered cytochrome P450 enzymes.
Science 339, 307–310 (2013).
10. Kan, S. B. J., Lewis, R. D., Chen, K. & Arnold, F. H. Directed evolution of
cytochrome c for carbon–silicon bond formation: bringing silicon to life. Science
354, 1048–1051 (2016).
11. Brandenberg, O. F., Fasan, R. & Arnold, F. H. Exploiting and engineering
hemoproteins for abiological carbene and nitrene transfer reactions. Curr. Opin.
Biotechnol. 47, 102–111 (2017).
12. Liao, K. et al. Site-selective and stereoselective functionalization of non-
activated tertiary C–H bonds. Nature 551, 609–613 (2017).
13. Weldy, N. M. et al. Iridium(iii)-bis(imidazolinyl)phenyl catalysts for
enantioselective C–H functionalization with ethyl diazoacetate. Chem. Sci. 7,
3142–3146 (2016).
14. Wang, Y., Wen, X., Cui, X. & Zhang, X. P. Enantioselective radical cyclization for
construction of 5-membered ring structures by metalloradical C‒H alkylation.
J. Am. Chem. Soc. 140, 4792–4796 (2018).
15. Caballero, A. et al. Silver-catalyzed C–C bond formation between methane and
ethyl diazoacetate in supercritical CO2. Science 332, 835–838 (2011).
16. Wu, W.-T., Yang, Z.-P. & You, S.-L. in Asymmetric Functionalization of C–H Bonds
(ed. You, S.-L.) Ch. 1 (Royal Society of Chemistry, Cambridge, 2015).
17. Key, H. M., Dydio, P., Clark, D. S. & Hartwig, J. F. Abiological catalysis by artifcial
haem proteins containing noble metals in place of iron. Nature 534, 534–537
(2016).
uct with high efficiency (8a and 8b, Fig. 4b). Notably, benzylic C–H
insertion was either not observed (with 7a, Supplementary Fig. 9) or
was considerably suppressed (with 7b, Supplementary Fig. 10), despite
the typically lower bond dissociation energies of benzylic C–H bonds
compared to α-amino C–H bonds. Additionally, N-aryl pyrrolidines
(7c–7e) were found to be excellent substrates and were selectively alky-
lated at the α-amino sp3 position. Using P411-CHF, the sp3 C–H alkyla-
tion of 7c outcompetes a Friedel–Crafts type reaction on the aryl ring,
which is a favourable process with other carbene-transfer systems27,28
.
Furthermore, alkylation product 8d offers a conceivable strategy for
the synthesis of β-homoproline, a motif that has been investigated for
medicinal chemistry applications29.
Given that P411-CHF alkylates both primary and secondary
α-amino C–H bonds, we investigated whether the enzyme could
be selective for one of these positions. Using N-methyl tetrahydro-
quinoline 7f as the alkane substrate, P411-CHF afforded β-amino
ester products with 1,050 TTN and a 9:1 ratio of regioisomers (C2:C1,
and 73.0:27.0 e.r. for (−)-8f) (Fig. 4b). The tetrahydroquinoline
ring is a prevalent structural motif in natural products and bioactive
molecules30, and its selective functionalization could provide a con-
cise strategy for the synthesis of alkaloids. To improve the selectivity
for the alkylation of 7f, we tested variants along the evolutionary
lineage from P-4(A82L) to P411-CHF. We found that, compared with
P411-CHF, P411-gen5 showed even better regioselectivity and the
opposite stereochemical preference for C–C bond formation. In a
reaction on 3.0-mmol scale, E. coli harbouring P411-gen5 delivered
(+)-8f in 85% yield with excellent selectivity (1,310 TTN, >50:1
regiomeric ratio (r.r.), 91.1:8.9 e.r.). In only a few steps, the enzy-
matic product was successfully transformed to the alkaloid (R)-(+)-
cuspareine30 (Fig. 4b).
Finally, we explored the introduction of different alkyl groups. Using
different diazo reagents, enzymatic C–H alkylation can diversify one
alkane substrate, such as 7a, to several products (10a–10c in Fig. 4c
and Supplementary Fig. 11). The diazo substrate scope extends beyond
ester-based reagents: Weinreb amide diazo compound 9c and diazoke-
tone 9d were found to participate in enzymatic C–H alkylation to fur-
nish products 10c and 10d, respectively. Additional substitution at the
α-position of the carbene, however, is generally not well-tolerated by
P411-CHF and the current related enzymes. With the exception of 10b,
reactions using disubstituted carbene reagents did not yield appreciable
amounts of desired products (Supplementary Fig. 11).
18. Mbuvi, H. M. & Woo, L. K. Catalytic C–H insertions using iron(iii) porphyrin
complexes. Organometallics 27, 637–645 (2008).
19. Li, Y., Huang, J.-S., Zhou, Z.-Y., Che, C.-M. & You, X.-Z. Remarkably stable iron
porphyrins bearing nonheteroatom-stabilized carbene or (alkoxycarbonyl)
carbenes: isolation, X-ray crystal structures, and carbon atom transfer reactions
with hydrocarbons. J. Am. Chem. Soc. 124, 13185–13193 (2002).
20. Grifn, J. R., Wendell, C. I., Garwin, J. A. & White, M. C. Catalytic C(sp3)–H
alkylation via an iron carbene intermediate. J. Am. Chem. Soc. 139, 13624–
13627 (2017).
21. Herschlag, D. & Natarajan, A. Fundamental challenges in mechanistic
enzymology: progress toward understanding the rate enhancements of
enzymes. Biochemistry 52, 2050–2067 (2013).
22. Prier, C. K., Zhang, R. K., Buller, A. R., Brinkmann-Chen, S. & Arnold, F. H.
Enantioselective, intermolecular benzylic C–H amination catalysed by an
engineered iron-haem enzyme. Nat. Chem. 9, 629–634 (2017).
23. Renata, H. et al. Identifcation of mechanism-based inactivation in P450-
catalyzed cyclopropanation facilitates engineering of improved enzymes. J. Am.
Chem. Soc. 138, 12527–12533 (2016).
24. Zhang, J.-T. et al. Total synthesis of malyngamides K, L, and 5″-epi-C and
absolute confguration of malyngamide L. J. Org. Chem. 76, 3946–3959
(2011).
This study demonstrates that a cytochrome P450 can acquire the
ability to construct C–C bonds from sp3 C–H bonds, and that the activ-
ity and selectivity of the reaction can be greatly enhanced using directed
evolution. Nature provides a huge collection of possible alternative
starting points for expanding the scope of this reaction even further and
for achieving other selectivities. The cytochrome P450 superfamily can
25. Edwards, J. T. et al. Decarboxylative alkenylation. Nature 545, 213–218 (2017).
26. He, J., Hamann, L. G., Davies, H. M. L. & Beckwith, R. E. J. Late-stage C–H
functionalization of complex alkaloids and drug molecules via intermolecular
rhodium-carbenoid insertion. Nat. Commun. 6, 5943 (2015).
access an immense set of organic molecules for its native oxygenation 27. Yang, J.-M., Cai, Y., Zhu, S.-F. & Zhou, Q.-L. Iron-catalyzed arylation of α-aryl-α-
chemistry; we foresee that P411-derived enzymes and other natural
diazoesters. Org. Biomol. Chem. 14, 5516–5519 (2016).
N A t U r e | www.nature.com/nature
© 2019 Springer Nature Limited. All rights reserved.

reSeArCH Letter
28. Xu, B., Li, M.-L., Zuo, X.-D., Zhu, S.-F. & Zhou, Q.-L. Catalytic asymmetric arylation
of α-aryl-α-diazoacetates with aniline derivatives. J. Am. Chem. Soc. 137,
8700–8703 (2015).
29. Cardillo, G., Gentilucci, L., Qasem, A. R., Sgarzi, F. & Spampinato, S.
Endomorphin-1 analogues containing β-proline are μ-opioid receptor agonists
and display enhanced enzymatic hydrolysis resistance. J. Med. Chem. 45,
2571–2578 (2002).
30. Sridharan, V., Suryavanshi, P. A. & Menéndez, J. C. Advances in the chemistry of
tetrahydroquinolines. Chem. Rev. 111, 7157–7259 (2011).
31. Coelho, P. S. et al. A serine-substituted P450 catalyzes highly efcient
carbene transfer to olefns in vivo. Nat. Chem. Biol. 9, 485–487
(2013).
analytical support; and B. Stoltz for use of polarimeter and gas chromatography
equipment.
Reviewer information Nature thanks B. de Bruin, N. Turner and T. Ward for their
contribution to the peer review of this work.
Author contributions R.K.Z. designed the overall research with F.H.A. providing
guidance. R.K.Z. and H.R. designed and conducted the initial screening of haem
proteins; R.K.Z. and L.W. performed the directed evolution experiments. R.K.Z.,
K.C. and X.H. designed and performed the substrate scope studies. R.K.Z. and
F.H.A. wrote the manuscript with input from all authors.
Competing interests A provisional patent application has been filed through the
California Institute of Technology based on the results presented here.
Acknowledgements This work was supported by the National Science
Foundation (NSF), Division of Molecular and Cellular Biosciences (grant MCB-
1513007). R.K.Z. acknowledges support from the NSF Graduate Research
Fellowship (grant DGE-1144469) and the Donna and Benjamin M. Rosen
Bioengineering Center. X.H. is supported by a Ruth L. Kirschstein National
Institutes of Health Postdoctoral Fellowship (grant F32GM125231). L.W.
received support from the Austrian Marshall Plan Foundation. We thank
A. Z. Zhou for experimental assistance; N. W. Goldberg, S. C. Hammer, K. E.
Hernandez, Z. Jia, A. M. Knight, G. Kubik, R. D. Lewis, C. K. Prier, D. K. Romney
and J. Zhang for discussions; S. Virgil, M. Shahgholi and D. VanderVelde for
Additional information
Supplementary information is available for this paper at https://doi.org/
10.1038/s41586-018-0808-5.
Reprints and permissions information is available at http://www.nature.com/
reprints.
Correspondence and requests for materials should be addressed to F.H.A.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.
N A t U r e | www.nature.com/nature
© 2019 Springer Nature Limited. All rights reserved.