Genetically programmed chiral organoborane synthesis

Article abstract of DOI:10.1038/nature24996

Recent advances in enzyme engineering and design have expanded nature's catalytic repertoire to functions that are new to biology^1-3. However, only a subset of these engineered enzymes can function in living systems^4-7. Finding enzymatic pathways that form chemical bonds that are not found in biology is particularly difficult in the cellular environment, as this depends on the discovery not only of new enzyme activities, but also of reagents that are both sufficiently reactive for the desired transformation and stable in vivo. Here we report the discovery, evolution and generalization of a fully genetically encoded platform for producing chiral organoboranes in bacteria. Escherichia coli cells harbouring wild-type cytochrome c from Rhodothermus marinus⁸ (Rma cyt c) were found to form carbon-boron bonds in the presence of borane-Lewis base complexes, through carbene insertion into boron-hydrogen bonds. Directed evolution of Rma cyt c in the bacterial catalyst provided access to 16 novel chiral organoboranes. The catalyst is suitable for gram-scale biosynthesis, providing up to 15,300 turnovers, a turnover frequency of 6,100 h^-1, a 99:1 enantiomeric ratio and 100% chemoselectivity. The enantiopreference of the biocatalyst could also be tuned to provide either enantiomer of the organoborane products. Evolved in the context of whole-cell catalysts, the proteins were more active in the whole-cell system than in purified forms. This study establishes a DNA-encoded and readily engineered bacterial platform for borylation; engineering can be accomplished at a pace that rivals the development of chemical synthetic methods, with the ability to achieve turnovers that are two orders of magnitude (over 400-fold) greater than those of known chiral catalysts for the same class of transformation^9-11. This tunable method for manipulating boron in cells could expand the scope of boron chemistry in living systems.

Full text of DOI:10.1038/nature24996

LEttER
doi:10.1038/nature24996
Genetically programmed chiral organoborane
synthesis
S. B. Jennifer Kan1*, Xiongyi huang1*, Yosephine Gumulya1, Kai Chen1 & Frances h. Arnold1
Recent advances in enzyme engineering and design have expanded various non-natural reactions4,6,7,17. The resulting enzymes are fully
nature’s catalytic repertoire to functions that are new to biology1–3
.
genetically encoded and carry out their synthetic functions in their
However, only a subset of these engineered enzymes can function in bacterial expression hosts. Here, we focused on introducing boron
living systems4–7. Finding enzymatic pathways that form chemical motifs to organic molecules enantioselectively, as this would generate
bonds that are not found in biology is particularly difficult in boron-containing carbon stereocentres, which are important struc-
the cellular environment, as this depends on the discovery not tural features in functional organoboranes such as the US Food and
only of new enzyme activities, but also of reagents that are both Drug Administration (FDA)-approved chemotherapeutics Velcade and
sufficiently reactive for the desired transformation and stable Ninlaro18. They are also versatile precursors for chemical derivatization
in vivo. Here we report the discovery, evolution and generalization through stereospecific carbon–boron to carbon–carbon or carbon–
of a fully genetically encoded platform for producing chiral heteroatom bond conversion19–21
.
organoboranes in bacteria. Escherichia coli cells harbouring wild-
Although boron reagents applicable for carbon–boron bond forma-
type cytochrome c from Rhodothermus marinus8 (Rma cyt c) were tion in water are known22,23, their biocompatibility, cell permeability,
found to form carbon–boron bonds in the presence of borane–Lewis stability and reactivity in living systems, which contain an abundance of
base complexes, through carbene insertion into boron–hydrogen biomolecules, nucleic acids and metal ions, are uncertain. Nevertheless,
bonds. Directed evolution of Rma cyt c in the bacterial catalyst as boron reagents designed for in vivo chemical biology applications
provided access to 16 novel chiral organoboranes. The catalyst are precedented24–26, we reasoned that reagents suitable for biological
is suitable for gram-scale biosynthesis, providing up to 15,300 borylation could be found. We identified borane–Lewis base complexes
turnovers, a turnover frequency of 6,100h–1, a 99:1 enantiomeric as potential candidates owing to their aqueous stability and reactivity
ratio and 100% chemoselectivity. The enantiopreference of the towards carbenoid B–H insertion9–11,27 (Extended Data Fig. 2), a mech-
biocatalyst could also be tuned to provide either enantiomer of anistic pathway we thought could be adapted for use in the biological
the organoborane products. Evolved in the context of whole-cell environment owing to its orthogonality to the existing biochemistry
catalysts, the proteins were more active in the whole-cell system than of living systems.
in purified forms. This study establishes a DNA-encoded and readily
We first set out to assess whether biological organoborane
engineered bacterial platform for borylation; engineering can be production might be feasible in a bacterial cell. E. coli BL21(DE3) cells
accomplished at a pace that rivals the development of chemical harbouring wild-type cytochrome c from the Gram-negative, ther-
synthetic methods, with the ability to achieve turnovers that are two mohalophilic bacterium Rhodothermus marinus8 (Rma cyt c) were
orders of magnitude (over 400-fold) greater than those of known incubated with N-heterocyclic carbene borane28,29 (NHC-borane) 1
chiral catalysts for the same class of transformation9–11. This tunable and ethyl 2-diazopropanoate (Me-EDA) 2 in neutral buffer (M9-N
method for manipulating boron in cells could expand the scope of minimal medium, pH 7.4). After incubation at room temperature,
boron chemistry in living systems.
in vivo production of organoborane 3 was observed, with 120 turnovers
Boron-containing natural products are synthesized in the soil (calculated with respect to the concentration of Rma cyt c expressed
by the myxobacterium Sorangium cellulosum as antibiotics against in E. coli, Fig. 1a, b) and an enantiomeric ratio (e.r.) of 85:15 (R/S
Gram-positive bacteria12. In the sea, these molecules give the Jurassic isomer=6, Fig. 1c). Because the pET22b/pEC86 expression system
red alga Solenopora jurassica its distinct pink colouration13; they are translocates Rma cyt c to the E. coli periplasm for post-translational
also produced by the bioluminescent bacterium Vibrio harveyi for maturation (during which the haem cofactor is covalently ligated to the
cell–cell communication14 (Extended Data Fig. 1). To prepare boron- cyt c apoprotein)30, we assumed that borylation takes place in the peri-
containing biomolecules, living organisms produce small plasmic compartment. In the absence of Rma cyt c, E. coli yielded only
molecules that spontaneously react with boric acid available in the a trace amount of borylation product with very low stereoselectivity
environment15,16. Although this non-enzymatic method for capturing (Extended Data Table 1). Both the substrates and the organoborane
boron is sufficient for the survival of an organism, it is limited by the product were stable under these conditions. The haem cofactor
inherent affinity of a substrate towards boric acid, and lacks tunability alone could also promote the borylation reaction, although with no
and generality for synthetic biology applications. Moreover, organisms stereoselectivity. Other cytochrome c proteins, cytochromes P450, and
that produce organoboranes (compounds that contain carbon–boron globins also demonstrated carbon–boron bond-forming ability, but
bonds) are unknown.
their selectivities were unsatisfactory (Extended Data Table 1).
To improve the performance of this whole-cell catalyst, we subjected
We envisioned that enzyme-catalysed borylation could provide
living organisms with the ability to produce boron-containing products the wild-type Rma cyt c (hereafter referred to as BORWT) to site-
tailored to our needs. Such an enzyme is not known in nature, but saturation mutagenesis, sequentially targeting active-site amino
we hypothesized that existing natural proteins might be repurposed acid residues M100, V75 and M103, which are closest to the haem
and engineered to perform this task. In the past, we and others have iron in BORWT (within 7Å, Fig. 1d). Each single-site site-saturation
exploited the promiscuity of natural and engineered haem proteins for mutagenesis library was cloned using the 22c-trick method31, screened
1Division of Chemistry and Chemical Engineering, California Institute of Technology, 1200 East California Boulevard, MC 210-41, Pasadena, California 91125, USA.
*These authors contributed equally to this work.
0 0 M o n t h 2 0 1 7
| V o L 0 0 0 | n A t U R E | 1
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

reSeArCH Letter
a
N
N2
E. coli harbouring
Rma cyt c variant
N
N
N
BH2
OEt
BH3
+
Me
OEt
Me
O
M9-N buffer (pH 7.4) RT
O
1
2
3
R conꢀguration
b
d
Rma cyt c
WT (BORWT
M100D
2,490
2,300
)
0
I
1,850
II V75R/M100D (BORR*
III V75R/M100D/M103T
)
(BORR1
)
170
120
0
80
0
None
I
II
III
Haemin Haemin
+ BSA
Rma cyt c evolution
c
39
V75
M100
10
7
M103
6
0
1
1
N/A
None
I
II
III
Haemin Haemin
+ BSA
Rma cyt c evolution
e
Preincubate with
1, 2 or 3
Add remaining
f
Puriꢀed protein
Cell lysate
substrate(s)
TTNincub
TTNcontrol
6,100
BORR1
15 min
30 min
Whole cell
No incubation Add substrates
410
1 and 2
160
NHC-borane Me-EDA Product
Preincubate with:
1
2
3
30
Puriꢀed BORR1
92
53
100
Whole-cell BORR1
3
4
100
90
100
TTN retained (%)
%TTN retained =
(TTNincub / TTNcontrol) × 100%
100%
50%
BORWT
BORR1
Figure 1 | Discovery, evolution and characterization of a bacterial
catalyst for borylation. a, Reaction scheme shows a representative
conditions. d, X-ray crystal structure of wild-type Rma cyt c (PDB: 3CP5).
e, Turnover frequencies (h−1, on log scale) of BORWT and BORR1 as
whole-cell catalysts, cell lysates, or purified proteins for the production
of organoborane 3. f, Purified and whole-cell BORR1 were preincubated
with NHC-borane 1, Me-EDA 2, or organoborane 3 before they were
used as borylation catalysts to determine the inactivation effects of 1–3.
The numbers shown represent the %TTN retained after preincubation,
and are relative to a control (no incubation) of the same type of catalyst
(purified protein or whole cell). Bars and numbers above bars represent
mean values averaged over four biological replicates. Individual data
points are shown as overlays. BSA, bovine serum albumin; NHC,
N-heterocyclic carbene.
in vivo borylation reaction between NHC-borane 1 and diazo ester 2 to
yield organoborane 3. Standard substrate loading is 10 mM for both 1
and 2. The absolute configuration of biosynthesized 3 was assigned as
R by X-ray crystallography. b, c, Sequential site-saturation mutagenesis
of Rma cyt c targeting active-site amino acid residues M100, V75 and
M103 improved the turnover (b) and enantioselectivity (c) of bacterial
production of organoborane 3. Whole-cell Rma cyt c variants were
compared using E. coli cells with an optical density at 600 nm (OD600
)
of 15. Total turnover numbers (TTNs) were calculated with respect to
the concentration of Rma cyt c expressed in E. coli and represent the total
number of turnovers obtained from the catalyst under the stated reaction
as whole-cell catalysts in 96-well plates for improved borylation enan- function is readily scalable from an analytical to a millimolar scale,
tioselectivity, and the best variant was used to parent the next round of with 0.5mmol substrates, BORR1 produced organoborane 3 in 97.5:2.5
mutation and screening. With a single mutation M100D replacing the e.r. and 75% isolated yield, with a TTN of 3,000. The absolute con-
distal axial ligand, the first-generation biocatalyst exhibited a 16-fold figuration of product 3 was unambiguously assigned as R by X-ray
improvement in turnover compared with the wild type (total turn- crystallography.
over number (TTN) 1,850, Fig. 1b), with 88:12 e.r. (R/S isomer=7;
With an excellent borylating bacterium in hand, the properties and
Fig. 1c). The M100D mutation also substantially improved carbene potential of the system were assessed. We characterized the initial
transfer reactivity for Si–H insertion catalysed by Rma cyt c6. This rates of in vivo borylation and found that screening for improved
improvement in catalytic performance is probably due to the removal enantioselectivity also led to an overall rate enhancement: whole-cell
of the axial ligand from the haem iron, which opens a site primed for BORR1 is 15 times faster than BORWT, with a turnover frequency of
iron carbenoid formation and subsequent product formation32. Two 6,100 h–1. Notably, as purified protein or in cell lysate, both BORR1
subsequent rounds of mutagenesis and screening led to variant BORR1 and BORWT are orders of magnitude slower than in vivo (Fig. 1e).
(V75R M100D M103T), which exhibited a turnover of 2,490 and an e.r. When isolated BORR1 protein and whole-cell BORR1 were preincu-
of 97.5:2.5 (R/S isomer=39). This genetically programmed biological bated with Me-EDA 2 before the borylation reaction, the isolated
2
| n A t U R E | V o L 0 0 0 | 0 0 M o n t h 2 0 1 7
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

Letter reSeArCH
N2
R1
E. coli harbouring
BORR1 (OD600 = 5)
BH2
OR2
+
R1
BH3
Me
OR2
Me
O
M9-N buffer (pH 7.4)
RT, 6 h
O
Boron reagent
Diazo ester
Organoborane
a
b
5
N
N
N
N
N
N
BH2
N
N
BH2
N
BH2
BH2
BH2
OEt
OEt
OEt
OEt
Me
OEt
12
Me
Me
Me
Me
O
O
O
O
O
4
5
6
7
2,020 TTN
99:1 e.r.
2,580 TTN
>99:1 e.r.
890 TTN
96:4 e.r.
2,160 TTN
99:1 e.r.
2,440 TTN, 96:4 e.r.
Gram-scale synthesis
2,910 TTN, 96:4 e.r., 42% isolated yield
(64% based on recovered starting material)
Cl
N
N
N
N
N
F3C
Cl
Cl
N
N
N
BH2
BH2
BH2
N
BH2
OEt
OEt
OEt
c
OEt
Me
Me
Me
Me
9
3
98:2
98:2
97:3
97:3
97:3
96:4
96:4
96:4
e.r.
O
O
O
O
8
9
10
11
1,680 TTN
97:3 e.r.
3,090 TTN
99:1 e.r.
1,310 TTN
97:3 e.r.
1,070 TTN
96:4 e.r.
15,000
12,000
9,000
6,000
3,000
d
N
N
N
N
N
N
N
N
BH2
BH2
BH2
BH2
OMe
OEt
Oi-Pr
OBn
Me
Me
Me
Me
O
O
O
O
13
3
14
15
2
4
6
8
2,760 TTN
98:2 e.r.
2,480 TTN
97:3 e.r.
1,040 TTN
96:4 e.r.
2,560 TTN
92:8 e.r.
Substrate equivalent
Figure 2 | Scope of chiral organoborane production in E. coli.
a, d, Scope of boron reagent (a) and diazo ester (d) for borylation
catalysed by E. coli harbouring BORR1. Standard substrate loading is
10 mM for both substrates. Reactions conducted in duplicate. b, Gram-
scale synthesis (8.4 mmol) of organoborane 12 catalysed by whole-cell
BORR1 (OD600 =30). The small scale preparation of 12 (2,440 TTN,
96:4 e.r.) is also reported for comparison. The absolute configuration of
biosynthesized 12 was assigned as R by X-ray crystallography.
c, Biosynthesis of organoboranes 9 (blue) and 3 (orange) catalysed by
whole-cell BORR1 (OD600 =15). One substrate equivalent (8 mM final
concentration of boron reagent and diazo ester) was added to the reaction
every 75 min. Reactions conducted in biological duplicate. Bn, benzyl; e.r.,
enantiomeric ratio.
protein retained only around 50% of its activity, whereas whole-cell at regular time intervals to E. coli expressing BORR1 (we tested the
BORR1 retained greater than 90% activity (Fig. 1f). NHC-borane sequential addition of up to eight equivalents of substrates over a period
1 and organoborane product 3 did not inactivate the enzyme. of 12h, Fig. 2c; Extended Data Table 2), organoborane 3 was produced
Me-EDA probably inactivates BORR1 through carbene transfer to with 10,400 turnovers (50% yield, 96:4 e.r.), whereas organoborane 9
the haem cofactor and/or the nucleophilic side chains of the protein, was obtained with 15,300 turnovers (73% yield, 96:4 e.r.). No substantial
a mechanism we studied previously in detail for a cytochrome P450- loss in activity or enantioselectivity was observed, demonstrating the
based carbene transferase33. The intact periplasm apparently protects potential of this bacterial catalyst for biosynthesis and incorporation
BORR1 from inactivation by Me-EDA, and carbene transfer to yield into natural or engineered metabolic pathways.
the organoborane product is generally faster than protein inactivation
Systematic modification of the diazo ester substituents from Et to
pathway(s) under those conditions. Similar observations have been Me, i-Pr or Bn revealed that the borylation ability of BORR1 is not
reported for other protein-based carbene transfer reaction systems7,34
.
limited to Me-EDA (3, 13–15, Fig. 2d). The relative insensitivity of
Analysis of colony-forming units shows that in vivo organoborane the protein to the steric bulk of the ester might indicate that, in the
production does not markedly reduce the viability of the E. coli putative iron carbenoid intermediate, this moiety is solvent-exposed
(Extended Data Fig. 3).
rather than embedded within the active site. By re-randomizing the
Next, we explored the scope of boron reagents that could function in 103 position in BORR1, a residue we thought might modulate loop
the cellular environment. Ten boron reagents were tested under turno- dynamics for improved binding of this substrate, the borylation turn-
ver-optimized conditions: although the size, solubility and lipophilicity over of 15 improved (from 2,560 to 4,200 TTN) using the triple-mutant
of these reagents varied, all were found to permeate the cell membrane V75R/M100D/M103D (BORR2, Fig. 3a). From the same site-saturation
and give the desired products with excellent selectivities and turno- library, a borylation catalyst for trifluoromethyl-substituted diazo ester
vers (Fig. 2a). Various substitutions on the NHC nitrogen are tolerated (CF3-EDA) was also discovered (V75R/M100D/M103F, BORR3).
(3–10). The reaction is chemoselective in the presence of terminal Acceptor–acceptor diazo reagents such as CF3-EDA are less reactive
olefins (5), which could function as a reaction handle suitable for towards carbenoid formation because of their electron-deficient nature
downstream biological or bio-orthogonal derivatization. Sterically and have not been used before this for enzymatic carbene-transfer
more demanding tetra- and penta-substituted NHCs are also accepted reactions. The present system tolerates this class of substrates and
(7–10). As well as imidazole-based boron reagents, triazolylidene yielded product 16 with 95:5 e.r. and 1,560 TTN.
borane and picoline borane could also be used for in vivo borylation,
To further broaden the generality of this borylation platform, we
yielding products 11 and 12 in 1,070 TTN and 2,440 TTN, respectively, re-examined the evolutionary landscape from BORWT to BORR1 to
with uniformly high selectivities (96:4 e.r.). On the gram scale, in vivo search for promiscuous mutants that might unlock new reactivities.
borylation produced 740mg of picoline organoborane 12 with 2,910 Double mutant V75P/M100D (BORP*) stood out as highly produc-
TTN, 96:4 e.r. and 42% isolated yield (64% based on recovered starting tive but poorly selective (69:31 e.r.) for Me-EDA borylation in the
material, Fig. 2b). The absolute configuration of 12 was assigned as R M100D/V75X site-saturation library. As proline-mediated helix kinks
by X-ray crystallography. When substrates were added portion-wise are known to induce structural and dynamic changes to proteins,
0 0 M o n t h 2 0 1 7
| V o L 0 0 0 | n A t U R E | 3
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

reSeArCH Letter
a
M99
T101
M100
BORR2
M103D
[B]
[B]
Site-saturation mutagenesis
15
OBn
OEt
4,200 TTN
92:8 e.r.
Me
Me
F3C
M103
O
O
BORR*
BORR1
M103T
3
2,490 TTN
98:2 e.r.
T98
V75R
Y71
BORR3
M103F
[B]
16
OEt
1,560 TTN
95:5 e.r.
BORWT
M100D
M89
V75
O
BORP1
[B]
17
340 TTN
94:6 e.r.
OEt
Ph
M99Y T101A M103F
BORP*
V75P
O
BORP2
(R)-18
1,010 TTN
96:4 e.r.
[B]
Y71C M89C M99C
Ph
Ph
CF3
N
[B] =
BORG1
N
H2B
BORG*
V75G
[B]
(S)-18
1,120 TTN
90:10 e.r.
M89F T98V M99L
T101L M103F
CF3
b
pin =
O
O
20
OH
78% yield
94:6 e.r.
N
O3
N
Ph
Ph
CF3
BH3
NaB
Bpin
N
N
Pinacol, HCl
MeCN, 40 °C
E. coli harbouring BORP2
BH2
CF3
(R)-18
Ph
CF3
19
M9-N buffer (pH 7.4) RT
Ph
+
1. Li
2. NaB
21
OH
N2
CH
90% yield
38% yield
(over 2 steps)
96:4 e.r.
1,100 TTN, 96:4 e.r.
70% isolated yield based on
recovered starting material
O2
Br
Ph
CF3
CF3
3
Figure 3 | Expanding the generality and utility of biological borylation.
a, The generality of in vivo borylation was expanded through directed
evolution to accommodate bulky substrates (15, 17) and less reactive
acceptor–acceptor diazo reagents (16), to move beyond diazo ester-based
substrates ((R)-18, (S)-18), and to provide either enantiomer of the
organoborane products ((R)-18, (S)-18). Reactions conducted in biological
quadruplicate. Solid arrows represent site-saturation mutagenesis studies.
BORP* was discovered in the M100D V75X site-saturation mutagenesis
library for Me-EDA borylation. Amino acid residues targeted during
directed evolution are depicted in the X-ray crystal structure of wild-type
Rma cyt c (PDB: 3CP5). The absolute configuration of biosynthesized
(R)-18 was assigned by X-ray crystallography. b, Derivatization of
biocatalytic product. Organoborane (R)-18 was biosynthesized with
E. coli harbouring BORP2 (OD600 =30) on a 1.3-mmol scale in 40% isolated
yield (70% based on recovered starting material) for derivatization studies.
Conversion to pinacol borane 19 was achieved with retention of the
stereogenic carbon centre (stereoselectivity determined after derivatization
to alcohol 20). The yield reported for 19 was determined by 19F NMR.
We demonstrated the stereospecific transformation of 19 to alcohol 20 and
Matteson homologation–oxidation product 21.
we asked whether the V75P mutation might provide access to a unique BORG* was further tuned through mutations M89F, T98V, M99L,
reaction space. Ethyl 2-diazophenylacetate (Ph-EDA) is a bulky donor– T101L and M103F (BORG1, Extended Data Table 3) to yield organob-
acceptor diazo reagent inactive towards BORWT, but when added to orane (S)-18 with 90:10 e.r. and 1,120 TTN.
E. coli harbouring BORP* with NHC-borane 1, Ph-EDA was trans-
Chiral α-trifluoromethylated organoboranes are useful synthetic
formed to organoborane 17 in 100 TTN and 75:25 e.r. (Fig. 3a). By building blocks that combine the unique properties of fluorinated
accumulating three additional loop mutations though directed evo- motifs with the versatile synthetic applications of organoboranes35.
lution (M99Y, T101A and M103F, Extended Data Table 3), BORP* However, methods for their asymmetric preparation are rare11,36
.
evolved into a synthetically useful catalyst (BORP1) for the borylation Our ability to biosynthesize both enantiomers of these molecules may
of Ph-EDA, supporting 340 turnovers with an e.r. of 94:6.
have applications in pharmaceutical and agrochemical synthesis. For
BORP* also allows us to move beyond diazo ester-based sub- example, product (R)-18 was converted to pinacol boronate 19 with
strates and apply bacterial production to a different class of chiral retention of the stereogenic carbon centre (Fig. 3b). Through well-
organoboranes: although inactive towards BORWT, CF3-substituted established stereospecific transformations19–21, pinacol boronates can
(diazomethyl)benzene (CF3-DMB) reacted with NHC-borane 1 in be diversified into a broad array of chiral compounds. We demonstrated
the presence of BORP* to yield organoborane (R)-18 in vivo with 74 the transformation of 19 to alcohol 20, a motif found in compounds
turnovers and modest selectivity (79:21 e.r.). We enhanced this through useful for the treatment of cancer37 and neurodegenerative diseases38,
three cysteine mutations at Y71, M89 and M99 (BORP2, Extended Data and the Matteson homologation–oxidation product 21, both of which
Table 3) to produce organoborane (R)-18 in 96:4 e.r. and 1,010 TTN. were obtained with good stereocontrol.
Through X-ray crystallography, the absolute configuration of (R)-18
In conclusion, we present a platform for biological borylation,
which can be tuned and configured through DNA manipulation.
was unambiguously assigned.
Finally, we asked whether the stereochemical preference of biological Microorganisms are powerful alternatives to chemical methods for
borylation could be switched. Towards this end, examination of the producing pharmaceuticals, agrochemicals, materials and fuels. They
M100D V75X site-saturation library for CF3-DMB borylation led are available by fermentation on a large scale and at low cost, and their
us to identify a variant (V75G M100D; BORG*) having an inverted genetically encoded synthetic abilities can be systematically modified
stereochemical preference to BORP* in the carbon–boron bond- and optimized. Borylation chemistry can now be added to the vast
forming step (31:69 e.r. for R/S isomer; 340 TTN). The selectivity of synthetic repertoire of biology.
4
| n A t U R E | V o L 0 0 0 | 0 0 M o n t h 2 0 1 7
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

Letter reSeArCH
Online Content Methods, along with any additional Extended Data display items and
Source Data, are available in the online version of the paper; references unique to
these sections appear only in the online paper.
24. Chang, M. C. Y., Pralle, A., Isacoff, E. Y. & Chang, C. J. A selective, cell-permeable
optical probe for hydrogen peroxide in living cells. J. Am. Chem. Soc. 126,
15392–15393 (2004).
25. Halo, T. L., Appelbaum, J., Hobert, E. M., Balkin, D. M. & Schepartz, A. Selective
recognition of protein tetraserine motifs with a cell-permeable, pro-fluorescent
bis-boronic acid. J. Am. Chem. Soc. 131, 438–439 (2009).
26. Kim, J. & Bertozzi, C. R. A bioorthogonal reaction of N-oxide and boron
reagents. Angew. Chem. Int. Ed. 54, 15777–15781 (2015).
27. Li, X. & Curran, D. P. Insertion of reactive rhodium carbenes into boron–
hydrogen bonds of stable N-heterocyclic carbene boranes. J. Am. Chem. Soc.
135, 12076–12081 (2013).
28. Curran, D. P. et al. Synthesis and reactions of N-heterocyclic carbene boranes.
Angew. Chem. Int. Ed. 50, 10294–10317 (2011).
29. Würtemberger-Pietsch, S., Radius, U. & Marder, T. B. 25 years of N-heterocyclic
carbenes: activation of both main-group element–element bonds and NHCs
themselves. Dalton Trans. 45, 5880–5895 (2016).
30. Arslan, E., Schulz, H., Zufferey, R., Künzler, P & Thöny-Meyer, L. Overproduction
of the Bradyrhizobium japonicum c-type cytochrome subunits of the cbb3
oxidase in Escherichia coli. Biochem. Biophys. Res. Commun. 251, 744–747
(1998).
31. Kille, S. et al. Reducing codon redundancy and screening effort of
combinatorial protein libraries created by saturation mutagenesis. ACS Synth.
Biol. 2, 83–92 (2013).
32. Mara, M. W. et al. Metalloprotein entatic control of ligand–metal bonds
quantified by ultrafast X-ray spectroscopy. Science 356, 1276–1280 (2017).
33. Renata, H. et al. Identification of mechanism-based inactivation in P450-
catalyzed cyclopropanation facilitates engineering of improved enzymes.
J. Am. Chem. Soc. 138, 12527–12533 (2016).
34. Hernandez, K. E. et al. Highly stereoselective biocatalytic synthesis of key
cyclopropane intermediate to ticagrelor. ACS Catal. 6, 7810–7813 (2016).
35. Argintaru, O. A., Ryu, D., Aron, I. & Molander, G. A. Synthesis and applications of
α-trifluoromethylated alkylboron compounds. Angew. Chem. Int. Ed. 52,
13656–13660 (2013).
36. Jiang, Q., Guo, T. & Yu, Z. Copper-catalyzed asymmetric borylation:
Construction of a stereogenic carbon center bearing both CF3 and
organoboron functional groups. J. Org. Chem. 82, 1951–1960 (2017).
37. Kanouni, T., Stafford, J. A., Veal, J. M. & Wallace, M. B. Histone demethylase
inhibitors. WO 2014/151106 A1 (2014).
38. Scopes, D. Pyrrolo [3,2-E] [1,2,4] triazolo [1,5-A] pyrimidines derivatives as
inhibitors of microglia activation. US patent 2012/0289523 A1 (2012).
received 22 July; accepted 2 November 2017.
Published online 29 November 2017.
1. Renata, H., Wang, Z. J. & Arnold, F. H. Expanding the enzyme universe:
accessing non-natural reactions by mechanism-guided directed evolution.
Angew. Chem. Int. Ed. 54, 3351–3367 (2015).
2. Hyster, T. K. & Ward, T. R. Genetic optimization of metalloenzymes: enhancing
enzymes for non-natural reactions. Angew. Chem. Int. Ed. 55, 7344–7357
(2016).
3. Hammer, S. C., Knight, A. M. & Arnold, F. H. Design and evolution of enzymes
for non-natural chemistry. Curr. Opin. Green Sustainable Chem. 7, 23–30
(2017)
4. Coelho, P. S. et al. A serine-substituted P450 catalyzes highly efficient carbene
transfer to olefins in vivo. Nat. Chem. Biol. 9, 485–487 (2013).
5. Jeschek, M. et al. Directed evolution of artificial metalloenzymes for in vivo
metathesis. Nature 537, 661–665 (2016).
6. 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).
7. Tinoco, A., Steck, V., Tyagi, V. & Fasan, R. Highly diastereo- and enantioselective
synthesis of trifluoromethyl-substituted cyclopropanes via myoglobin-
catalyzed transfer of trifluoromethylcarbene. J. Am. Chem. Soc. 139,
5293–5296 (2017).
8. Stelter, M. et al. A novel type of monoheme cytochrome c: biochemical and
structural characterization at 1.23 A resolution of Rhodothermus marinus
cytochrome c. Biochemistry 47, 11953–11963 (2008).
9. Cheng, Q.-Q., Zhu, S.-F., Zhang, Y.-Z., Xie, X.-L. & Zhou, Q.-L. Copper-catalyzed
B–H bond insertion reaction: a highly efficient and enantioselective C–B
bond-forming reaction with amine–borane and phosphine–borane adducts.
J. Am. Chem. Soc. 135, 14094–14097 (2013).
10. Chen, D., Zhang, X., Qi, W.-Y., Xu, B. & Xu, M.-H. Rhodium(i)-catalyzed
asymmetric carbene insertion into B–H bonds: highly enantioselective access
to functionalized organoboranes. J. Am. Chem. Soc. 137, 5268–5271 (2015).
11. Hyde, S. et al. Copper-catalyzed insertion into heteroatom–hydrogen bonds
with trifluorodiazoalkanes. Angew. Chem. Int. Ed. 55, 3785–3789 (2016).
12. Irschik, H., Schummer, D., Gerth, K., Höfle, G. & Reichenbach, H. The tartrolons,
new boron-containing antibiotics from a myxobacterium, Sorangium
cellulosum. J. Antibiot. 48, 26–30 (1995).
Supplementary Information is available in the online version of the paper.
13. Wolkenstein, K., Sun, H., Falk, H. & Griesinger, C. Structure and absolute
configuration of Jurassic polyketide-derived spiroborate pigments obtained
from microgram quantities. J. Am. Chem. Soc. 137, 13460–13463 (2015).
14. Chen, X. et al. Structural identification of a bacterial quorum-sensing signal
containing boron. Nature 415, 545–549 (2002).
15. Elshahawi, S. I. et al. Boronated tartrolon antibiotic produced by symbiotic
cellulose-degrading bacteria in shipworm gills. Proc. Natl Acad. Sci. USA 110,
E295–E304 (2013).
16. Dembitsky, V. M., Al Quntar, A. A. & Srebnik, M. Natural and synthetic small
boron-containing molecules as potential inhibitors of bacterial and fungal
quorum sensing. Chem. Rev. 111, 209–237 (2011).
17. 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).
18. Das, B. C. et al. Boron chemicals in diagnosis and therapeutics. Future Med.
Chem. 5, 653–676 (2013).
19. Miyaura, N. & Suzuki, A. Palladium-catalyzed cross-coupling reactions of
organoboron compounds. Chem. Rev. 95, 2457–2483 (1995).
20. Leonori, D. & Aggarwal, V. K. Lithiation–borylation methodology and its
application in synthesis. Acc. Chem. Res. 47, 3174–3183 (2014).
21. Leonori, D. & Aggarwal, V. K. Stereospecific couplings of secondary and tertiary
boronic esters. Angew. Chem. Int. Ed. 54, 1082–1096 (2015).
22. Tehfe, M. A. et al. A water-compatible NHC-borane: Photopolymerizations in
water and rate constants for elementary radical reactions. ACS Macro Lett. 1,
92–95 (2012).
Acknowledgements This work was supported in part by the National Science
Foundation, Office of Chemical, Bioengineering, Environmental and Transport
Systems SusChEM Initiative (grant CBET-1403077), the Gordon and Betty
Moore Foundation through grant GBMF2809 to the Caltech Programmable
Molecular Technology Initiative, and the Jacobs Institute for Molecular
Engineering for Medicine at Caltech. X.H. is supported by a Ruth L. Kirschstein
National Institutes of Health Postdoctoral Fellowship (F32GM125231). We
thank O. F. Brandenberg, S. Brinkmann-Chen, T. Hashimoto, R. D. Lewis, and
D. K. Romney for discussions and/or comments on the manuscript, and
N. W. Goldberg and A. Zutshi for experimental assistance. We are grateful to
S. Virgil, N. Torian, M. K. Takase and L. Henling for analytical support, and
H. Gray for providing the pEC86 plasmid.
Author Contributions S.B.J.K. and X.H. designed the research with guidance
from F.H.A. S.B.J.K., X.H., Y.G. and K.C. performed the experiments and analysed
the data. S.B.J.K., X.H. and F.H.A. wrote the manuscript with input from all
authors.
Author Information Reprints and permissions information is available at
www.nature.com/reprints. The authors declare competing financial interests:
details are available in the online version of the paper. Readers are welcome to
comment on the online version of the paper. Publisher’s note: Springer Nature
remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations. Correspondence and requests for materials should be
addressed to F.H.A. (frances@cheme.caltech.edu).
23. Handa, S., Wang, Y., Gallou, F. & Lipshutz, B. H. Sustainable Fe–ppm Pd
nanoparticle catalysis of Suzuki-Miyaura cross-couplings in water. Science
349, 1087–1091 (2015).
reviewer Information Nature thanks M. Fischbach and the other anonymous
reviewer(s) for their contribution to the peer review of this work.
0 0 M o n t h 2 0 1 7
| V o L 0 0 0 | n A t U R E | 5
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

reSeArCH Letter
MethOdS
from the column by running a gradient from 20mM to 500mM imidazole over 10
column volumes. The purity of the collected protein fractions was analysed using
Detailed experimental methods are available in the Supplementary Information.
Data reporting. No statistical methods were used to predetermine sample size. SDS–PAGE. Pure fractions were pooled and concentrated using a 10kDa molecular
The experiments were not randomized and the investigators were not blinded to weight cut-off centrifugal filter and buffer-exchanged with 0.1M phosphate buffer
allocation during experiments and outcome assessment.
(pH=8.0). The purified protein was flash-frozen on dry ice and stored at −20°C.
Materials. Plasmid pET22b(+) was used as a cloning vector, and cloning was per- P450 and globin concentrations were determined in triplicate using published
formed using Gibson assembly39. The cytochrome c maturation plasmid pEC8630 extinction coefficients and the haemochrome assay described below.
was used as part of a two-plasmid system to express prokaryotic cytochrome Haemochrome assay. A solution of sodium dithionite (10mg ml−1) was prepared
c proteins. Cells were grown using Luria-Bertani medium or HyperBroth in M9-N buffer. Separately, a solution of 1M NaOH (0.4ml) was mixed with
(AthenaES) with 100 μg ml−1 ampicillin and 20 μg ml−1 chloramphenicol pyridine (1ml), followed by centrifugation (10,000g, 30s) to separate the excess
(LBamp/chlor or HBamp/chlor). Cells without the pEC86 plasmid were grown with aqueous layer to give a pyridine–NaOH solution. To a cuvette containing 700μl
100 μg ml−1 ampicillin (LBamp or HBamp). Electrocompetent E. coli cells were protein solution (purified protein or heat-treated lysate) in M9-N buffer, 50μl of
prepared following a published protocol40. T5 exonuclease, Phusion polymerase, dithionite solution and 250μl pyridine–NaOH solution were added. The cuvette
and Taq ligase were purchased from New England Biolabs (NEB). M9-N minimal was sealed with Parafilm, and the UV-Vis spectrum was recorded immediately.
medium (abbreviated as M9-N buffer; pH 7.4) was used as a buffering system Cytochrome c concentration was determined using ε550–535 =22.1mM−1 cm−1
for whole cells, lysates and purified proteins unless otherwise specified. M9-N (ref. 41). Protein concentrations determined by the haemochrome assay were in
buffer was used without a carbon source; it contains 47.7mM Na2HPO4, 22.0mM agreement with those determined by the bicinchoninic acid assay (Thermo Fisher)
KH2PO4, 8.6mM NaCl, 2.0mM MgSO4 and 0.1mM CaCl2.
using BSA for standard curve preparation.
Plasmid construction. All variants described in this paper were cloned and Mutagenesis library construction. Cytochrome c site-saturation mutagenesis
expressed using the pET22b(+) vector (Novagen). The gene encoding Rma cyt c libraries were generated using a modified version of the 22-codon site-saturation
(UniProt ID B3FQS5) was obtained as a single gBlock (IDT), codon-optimized method31. For each site-saturation library, oligonucleotides were ordered such that
for E. coli, and cloned using Gibson assembly39 into pET22b(+) (Novagen) the coding strand contained the degenerate codon NDT, VHG or TGG. The reverse
between restriction sites NdeI and XhoI in frame with an N-terminal pelB complements of these primers were also ordered. The three forward primers were
leader sequence (to ensure periplasmic localization and proper maturation; mixed together in a 12:9:1 ratio, (NDT:VHG:TGG) and the three reverse primers
MKYLLPTAAAGLLLLAAQPAMA) and a C-terminal 6×His-tag. This plasmid were mixed similarly. Two PCRs were performed, pairing the mixture of forward
was co-transformed with the cytochrome c maturation plasmid pEC86 into E. cloni primers with a pET22b(+) internal reverse primer, and the mixture of reverse
EXPRESS BL21(DE3) cells (Lucigen).
primers with a pET22b(+) internal forward primer. The two PCR products were
Cytochrome c expression and purification. Purified cytochrome c proteins gel-purified, ligated together using Gibson assembly39, and transformed into
were prepared as follows. One litre HBamp/chlor in a 4 l flask was inoculated E. cloni EXPRESS BL21(DE3) cells.
with an overnight culture (20ml, LBamp/chlor) of recombinant E. cloni EXPRESS Mutagenesis library screening in whole cells. Single colonies were picked with
BL21(DE3) cells containing a pET22b(+) plasmid encoding the cytochrome c toothpicks off of LBamp/chlor agar plates, and grown in deep-well (2ml) 96-well
variant, and the pEC86 plasmid. The culture was shaken at 37°C and 200rpm plates containing LBamp/chlor (400μl) at 37°C, 250rpm shaking, and 80% relative
(no humidity control) until the OD600 was 0.7 (approximately 3h). The culture humidity overnight. After 16h, 30μl aliquots of these overnight cultures were
was placed on ice for 30min, and isopropyl β-D-1-thiogalactopyranoside (IPTG) transferred to deep-well 96-well plates containing HBamp/chlor (1 ml) using a
and 5-aminolevulinic acid (ALA) were added to final concentrations of 20μM 12-channel EDP3-Plus 5–50μl pipette (Rainin). Glycerol stocks of the libraries
and 200μM, respectively. The incubator temperature was reduced to 20°C, and were prepared by mixing cells in LBamp/chlor (100μl) with 50% v/v glycerol (100μl).
the culture was shaken for 22h at 200rpm. Cells were collected by centrifugation Glycerol stocks were stored at −78°C in 96-well microplates. Growth plates were
(4000g, 15min, 4°C), and the cell pellet was stored at −20°C until further use (at shaken for 3h at 37°C at 250rpm, and 80% relative humidity. The plates were
least 24h). The cell pellet was resuspended in buffer containing 100mM NaCl, then placed on ice for 30min. Cultures were induced by adding 10μl of a solution,
20mM imidazole, and 20mM Tris-HCl buffer (pH 7.5 at 25°C) and cells were prepared in sterile deionized water, containing 2mM IPTG and 20mM ALA. The
lysed by sonication (2min, 2s on, 2s off, 40% duty cycle; Qsonica Q500 sonicator). incubator temperature was reduced to 20°C, and the induced cultures were shaken
Cell debris was removed by centrifugation for 20min (5,000g, 4°C). Supernatant for 20h (250rpm, no humidity control). Cells were pelleted (4,000g, 5min, 4°C),
was sterile-filtered through a 0.45μm cellulose acetate filter and purified using a resuspended in 380μl M9-N buffer, and the plates containing the cell suspensions
1ml Ni-NTA column (HisTrap HP, GE Healthcare) using an AKTA purifier FPLC were transferred to an anaerobic chamber. To deep-well plates of cell suspensions
system (GE Healthcare). The cytochrome c protein was eluted from the column by were added NHC-borane substrate (10μl per well, 400mM in MeCN) and diazo
running a gradient from 20 to 500mM imidazole over 10 column volumes. The reagent (10 μl per well, 400mM in MeCN). The plates were sealed with aluminium
purity of the collected cytochrome c fractions was analysed using SDS–PAGE. Pure sealing tape, removed from the anaerobic chamber, and shaken at 500rpm for
fractions were pooled and concentrated using a 3kDa molecular weight cut-off 6h (24h for reactions with Ph-EDA or CF3-DMB because of their lower aqueous
centrifugal filter and dialysed overnight into 0.05M phosphate buffer (pH=7.5) solubility). After quenching with hexanes:ethyl acetate (4:6 v/v, 0.6ml), internal
using 3kDa molecular weight cut-off dialysis tubing. The dialysed protein was standard was added (20μl of 20mM 1,2,3-trimethoxybenzene in toluene). The
concentrated again, flash-frozen on dry ice, and stored at −20°C. The concentra- plates were then sealed with sealing mats and shaken vigorously to thoroughly mix
tion of cytochrome c was determined in triplicate using the haemochrome assay the organic and aqueous layers. The plates were centrifuged (4,000g, 5min) and
described below.
the organic layer (200μl) was transferred to autosampler vials with vial inserts for
Cytochrome P450 and globin expression and purification. Purified P450s and gas chromatography–mass spectrometry (GC–MS) or chiral high performance
globins were prepared differently from the cytochrome c proteins, and described liquid chromatography (HPLC) or supercritical fluid chromatography (SFC)
as follows. One litre HBamp in a 4l flask was inoculated with an overnight culture analysis. Hits from library screening were confirmed by small-scale biocatalytic
(20ml, LBamp) of recombinant E. cloni EXPRESS BL21(DE3) cells containing a reactions.
pET22b(+) plasmid encoding the P450 or globin variant. The culture was shaken Cell lysate preparation. Cell lysates were prepared as follows: E. coli cells express-
at 37°C and 200rpm (no humidity control) until the OD600 was 0.7 (approximately ing Rma cyt c variant were pelleted (4,000g, 5min, 4°C), resuspended in M9-N
3h). The culture was placed on ice for 30min, and IPTG and ALA were added to buffer and adjusted to the appropriate OD600. Cells were lysed by sonication (2min,
final concentrations of 0.5mM and 1mM, respectively. The incubator temperature 1s on, 2s off, 40% duty cycle; Qsonica Q500 sonicator), aliquoted into 2ml micro-
was reduced to 20°C, and the culture was shaken for 20h at 200rpm. Cells were centrifuge tubes, and the cell debris was removed by centrifugation for 10min
collected by centrifugation (4°C, 15min, 4,000g), and the cell pellet was stored at (14,000g, 4°C). The supernatant was sterile-filtered through a 0.45μm cellulose
−20°C until further use (at least 24h). The cell pellet was resuspended in buffer acetate filter, and the concentration of cytochrome c protein lysate was determined
containing 100mM NaCl, 20mM imidazole, and 20mM Tris-HCl buffer (pH 7.5 using the haemochrome assay. Using this protocol, the protein concentrations we
at 25°C). Haemin (30mg ml−1, 0.1M NaOH; Frontier Scientific) was added to the typically observed for OD600 =15 lysates were in the 8–15μM range for wild-type
resuspended cells such that 1mg of haemin was added for every 1g of cell pellet. Rma cyt c and 1–10μM for other Rma cyt c variants.
Cells were lysed by sonication (2min, 1s on, 2s off, 40% duty cycle; Qsonica Q500 Small-scale whole-cell bioconversion. In an anaerobic chamber, NHC-borane
sonicator). Cell debris was removed by centrifugation for 20min (27,000g, 4°C). (10μl, 400mM in MeCN) and diazo reagent (10μl, 400mM in MeCN) were added
Supernatant was sterile-filtered through a 0.45μm cellulose acetate filter, and puri- to E. coli harbouring Rma cyt c variant (380μl, adjusted to the appropriate OD600
)
fied using a 1ml Ni-NTA column (HisTrap HP, GE Healthcare) using an AKTA in a 2 ml crimp vial. The vial was crimp-sealed, removed from the anaerobic
purifier FPLC system (GE healthcare). The P450 and globin proteins were eluted chamber, and shaken at 500rpm at room temperature for 6h (24h for reactions
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

Letter reSeArCH
with Ph-EDA or CF3-DMB). At the end of the reaction, the crimp vial was opened of reagents. Each equivalent is 2.5μl solution of NHC-BH3 substrate in MeCN
and the reaction was quenched with hexanes:ethyl acetate (4:6 v/v, 0.6ml), followed (2M) and 2.5μl Me-EDA solution in MeCN (2M). The time interval between each
by the addition of internal standard (20μl of 20mM 1,2,3-trimethoxybenzene in equivalent was 75min. All four group sets were shaken at 480rpm in the anaerobic
toluene). The reaction mixture was transferred to a microcentrifuge tube, vortexed chamber until the completion of the addition and reaction for the last group set.
(10s, 3 times), then centrifuged (14,000g, 5min) to completely separate the organic The vials were then removed from the anaerobic chamber and quenched with 1ml
and aqueous layers (the vortex–centrifugation step was repeated if complete of ethyl acetate. The reaction mixture was transferred to a microcentrifuge tube,
phase separation was not achieved). The organic layer (200μl) was removed for vortexed (10s, 3 times), then centrifuged (14,000g, 5min) to completely separate
GC–MS and chiral SFC or HPLC analysis. All biocatalytic reactions reported were the organic and aqueous layers (the vortex–centrifugation step was repeated if
performed in n=2–4 technical replicates from at least two biological replicates. complete phase separation was not achieved). The organic layer was removed.
The TTNs reported are calculated with respect to Rma cyt c expressed in E. coli Another 1ml of ethyl acetate was added for a second round of extraction and
and represent the total number of turnovers obtained from the catalyst under the the organic solutions of two rounds of extraction were combined. The combined
stated reaction conditions. For reactions using OD600 =15 E. coli cells, the catalyst organic extracts were diluted (1.5-, 2-, and 5-fold dilution for experiments using
loadings were 0.0001–0.0015mol% of enzymes with respect to the limiting reagent 4, 6, and 8 equivalents of reagents, respectively). Internal standard (10μl of 20mM
in the reaction. The gborylation product/gdry cell weight ratios ranged from approximately 1,2,3-trimethoxybenzene in toluene) was added to 300μl of the diluted extract
0.05 (wild-type) to approximately 2 (engineered variant).
before the sample was subjected to GC–MS and chiral HPLC analysis to determine
Cell viability assay. The colony forming units (cfu) of whole-cell reactions the yield, TTN, and e.r.
(+ borylation) and controls without borylation reagents (− borylation) were Data availability. Data supporting the findings of this study are available within the
determined with biological replicates according to the following procedures. Six paper and its Supplementary Information, or are available from the corresponding
2ml screw cap vials containing 380μl suspension of E. coli harbouring BORR1 author upon reasonable request. Crystallographic coordinates and structure factors
(OD600 =15) were transferred to an anaerobic chamber. To three of these vials were have been deposited with the Cambridge Crystallographic Data Centre (https://
added NHC-borane 1 (10μl, 400mM in MeCN) and Me-EDA 2 (10μl, 400mM in www.ccdc.cam.ac.uk/) under reference numbers 1572198, 1572200 and 1572201
MeCN). These vials were capped and shaken at 500rpm in the anaerobic chamber for organoboranes 3, 18 and 12, respectively.
(+ borylation). The remaining three vials were capped and shaken in the absence
of reagents 1 and 2 (− borylation). After 2.5h, all six vials were removed from
the anaerobic chamber. Aliquots of cell suspension were removed from the vials
and subjected to serial dilution to obtain stock solutions of 106, 107, and 108-fold
dilution. 50μl of each stock solution was plated on LBamp/chlor agar plates and incu-
bated at 37°C overnight. The cfu of the cell suspensions were calculated based
on the colony counts of the 107-dilution plate. The cfu for each vial are shown in
Extended Data Fig. 2.
39. Gibson, D. G. et al. Enzymatic assembly of DNA molecules up to several
hundred kilobases. Nat. Methods 6, 343–345 (2009).
40. Sambrook, J., Frisch, E. & Maniatis, T. Molecular Cloning: A Laboratory Manual
(Cold Spring Harbor Laboratory Press, 1989).
41. Berry, E. A. & Trumpower, B. L. Simultaneous determination of hemes a, b,
and c from pyridine hemochrome spectra. Anal. Biochem. 161, 1–15 (1987).
42. Yang, J.-M. et al. Catalytic B−H bond insertion reactions using alkynes as
carbene precursors. J. Am. Chem. Soc. 139, 3784–3789 (2017).
43. Allen, T. H., Kawamoto, T., Gardner, S., Geib, S. J. & Curran, D. P. N-heterocyclic
carbene boryl iodides catalyze insertion reactions of N-heterocyclic carbene
boranes and diazoesters. Org. Lett. 19, 3680–3683 (2017).
44. Wang, Z. J. et al. Improved cyclopropanation activity of histidine-ligated
cytochrome P450 enables the enantioselective formal synthesis of
levomilnacipran. Angew. Chem. Int. Ed. 53, 6810–6813 (2014).
Biosynthesis of organoboranes 9 and 3 via serial substrate addition. Twelve 2ml
screw cap vials containing 400μl suspension of cells harbouring Rma cyt c BORR1
(OD600 =15) and 100μl of glucose (250mM) were transferred to an anaerobic
chamber. The 12 vials were grouped into 4 group sets to determine the yield, TTN
and e.r. for reactions involving the stepwise addition of 2, 4, 6 or 8 equivalents
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

reSeArCH Letter
Extended Data Figure 1 | Examples of boron-containing natural products.
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

Letter reSeArCH
Extended Data Figure 2 | Summary of known catalytic systems for
metal–carbenoid insertion reactions of boranes. a, Rh2(esp)2-catalysed
borylation of diazo esters with NHC-boranes27. b, Cu(MeCN)4PF6-
catalysed borylation of diazo esters with phosphine-borane9.
c, [Rh(C2H4)2Cl]2-catalysed borylation of diazo esters with
amine-borane10. d, Cu(MeCN)4PF6-catalysed borylation of
CF3-substituted (diazomethyl)benzene with phosphine-borane11
.
e, Rh2(R-BTPCP)4-catalysed borylation using alkynes as carbene
precursors42. f, I2-catalysed borylation of diazo esters with
NHC-boranes43
.
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

reSeArCH Letter
Extended Data Figure 3 | Effect of biological borylation on E. coli cell
viability. Cell viability assay was performed in biological triplicate, see
Methods section for experimental protocol.
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

Letter reSeArCH
extended data table 1 | Preliminary borylation experiments
with haem and haem proteins using NhC-borane (1) and
Me-edA (2) as substrates
N/A, not applicable; NR, no product was detected; n.d., not determined. Experiments with
cytochromes c, globin, or cytochromes P450 were performed using E. coli harbouring the
corresponding protein (OD600 =15). Reactions were performed in biological triplicate. TTNs
reported represent mean values averaged over three biological replicates, and the errors are
one standard deviation. Within the detection limit of the instrument, variability in e.r. was not
observed. Unreacted starting materials were observed at the end of all reactions and no attempt
was made to optimize these reactions. Experiments with haemin were performed using 100μ
M haemin, 10mM NHC-borane 1, 10mM Me-EDA 2, 10mM Na2S2O4. Experiments with haemin
and BSA were performed using 100μM haemin in the presence of BSA (0.75mg ml−1) instead.
Experiments to determine E. coli cell background reaction were performed with E. cloni EXPRESS
BL21(DE3) cells containing a pET22b(+) plasmid encoding halohydrin dehalogenase from
Agrobacterium tumefaciens (UniProt ID Q93D82) instead of Rma cyt c. A. tumefaciens halohydrin
dehalogenase is inactive towards NHC-borane 1 and Me-EDA 2. P411 CIS4 and BM3 Hstar44 are
previously reported engineered BM3 P450 variants.
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

reSeArCH Letter
extended data table 2 | Biosynthesis of organoboranes 3 and 9 via
serial substrate addition
Experiments performed in biological duplicate. See Methods for experimental protocol.
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

Letter reSeArCH
extended data table 3 | directed evolution of whole-cell Rma
cyt c for improved enantioselectivity in the biosynthesis of
organoboranes 17, (R)-18 and (S)-18
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

Corresponding author(s): Frances H. Arnold
Initial submission Revised version
Final submission
Life Sciences Reporting Summary
Nature Research wishes to improve the reproducibility of the work that we publish. This form is intended for publication with all accepted life
science papers and provides structure for consistency and transparency in reporting. Every life science submission will use this form; some list
items might not apply to an individual manuscript, but all fields must be completed for clarity.
For further information on the points included in this form, see Reporting Life Sciences Research. For further information on Nature Research
policies, including our data availability policy, see Authors & Referees and the Editorial Policy Checklist.
` Experimental design
1. Sample size
Describe how sample size was determined.
Biocatalytic experiments were carried out in replicates (typically repeated two to
four times) to ensure independent experiments are reproducible. This information
is available in all relevant figure legends. No statistical methods were used to
predetermine the sample size.
2. Data exclusions
Describe any data exclusions.
3. Replication
No data was excluded from analysis.
Describe whether the experimental findings were
reliably reproduced.
All experimental findings were reliably reproduced using biological replicates.
4. Randomization
Describe how samples/organisms/participants were
allocated into experimental groups.
N.A.
N.A.
5. Blinding
Describe whether the investigators were blinded to
group allocation during data collection and/or analysis.
Note: all studies involving animals and/or human research participants must disclose whether blinding and randomization were used.
6. Statistical parameters
For all figures and tables that use statistical methods, confirm that the following items are present in relevant figure legends (or in the
Methods section if additional space is needed).
n/a Confirmed
The exact sample size (n) for each experimental group/condition, given as a discrete number and unit of measurement (animals, litters, cultures, etc.)
A description of how samples were collected, noting whether measurements were taken from distinct samples or whether the same
sample was measured repeatedly
A statement indicating how many times each experiment was replicated
The statistical test(s) used and whether they are one- or two-sided (note: only common tests should be described solely by name; more
complex techniques should be described in the Methods section)
A description of any assumptions or corrections, such as an adjustment for multiple comparisons
The test results (e.g. P values) given as exact values whenever possible and with confidence intervals noted
A clear description of statistics including central tendency (e.g. median, mean) and variation (e.g. standard deviation, interquartile range)
Clearly defined error bars
See the web collection on statistics for biologists for further resources and guidance.
1

` Software
Policy information about availability of computer code
7. Software
Describe the software used to analyze the data in this
study.
Data analysis: Microsoft excel
HPLC: ChemStation for LC systems Rev. B.02.01-SR1, Agilent Technologies
GCMS software: GCMS solution Version 2.72, Shimadzu Corporation
SFC: ChemStation for LC 3D systems Rev. B.03.01, Agilent Technologies
Chiral GC: GC ChemStation Rev. B.04.02, Agilent Technologies
X-ray crystallography: The structure was solved by direct methods using SHELXS
and refined against F2 on all data by full-matrix least squares with SHELXL-2014
For manuscripts utilizing custom algorithms or software that are central to the paper but not yet described in the published literature, software must be made
available to editors and reviewers upon request. We strongly encourage code deposition in a community repository (e.g. GitHub). Nature Methods guidance for
providing algorithms and software for publication provides further information on this topic.
` Materials and reagents
Policy information about availability of materials
8. Materials availability
Indicate whether there are restrictions on availability of
unique materials or if these materials are only available
for distribution by a for-profit company.
No unique materials were used.
9. Antibodies
Describe the antibodies used and how they were validated N/A
for use in the system under study (i.e. assay and species).
10. Eukaryotic cell lines
a. State the source of each eukaryotic cell line used.
N/A
N/A
N/A
b. Describe the method of cell line authentication used.
c. Report whether the cell lines were tested for
mycoplasma contamination.
d. If any of the cell lines used are listed in the database
of commonly misidentified cell lines maintained by
ICLAC, provide a scientific rationale for their use.
N/A
` Animals and human research participants
Policy information about studies involving animals; when reporting animal research, follow the ARRIVE guidelines
11. Description of research animals
Provide details on animals and/or animal-derived
materials used in the study.
N/A
Policy information about studies involving human research participants
12. Description of human research participants
Describe the covariate-relevant population
N/A
characteristics of the human research participants.
2