18. B. von Dawans, C. Kirschbaum, M. Heinrichs,
Psychoneuroendocrinology 36, 514–522 (2011).
19. Materials and methods are available as supplementary
materials on Science Online.
20. C. A. Rowland, Psychol. Bull. 140, 1432–1463
21. S. K. Carpenter, J. Exp. Psychol. Learn. Mem. Cogn. 35,
22. S. K. Carpenter, J. Exp. Psychol. Learn. Mem. Cogn. 37,
23. S. K. Carpenter, E. L. DeLosh, Mem. Cognit. 34, 268–276
24. M. A. Pyc, K. A. Rawson, Science 330, 335 (2010).
25. E. A. Wing, E. J. Marsh, R. Cabeza, Neuropsychologia 51,
26. G. S. Everly, J. M. Lating, A Clinical Guide to the Treatment of
the Human Stress Response (Springer Science+Business
Media, ed. 3, 2013).
27. E. J. Hermans et al., Science 334, 1151–1153
A full description of the materials, methods, and statistical
analyses described in this manuscript is available in the
supplementary materials. The data described in this paper are
available from the first author upon request. This research was
funded by the U.S. Army Natick Soldier Research, Development
and Engineering Center.
Materials and Methods
Tables S1 to S3
6 July 2016; accepted 24 October 2016
Directed evolution of cytochrome c
for carbon–silicon bond formation:
Bringing silicon to life
S. B. Jennifer Kan, Russell D. Lewis, Kai Chen, Frances H. Arnold*
Enzymes that catalyze carbon–silicon bond formation are unknown in nature, despite the
natural abundance of both elements. Such enzymes would expand the catalytic repertoire
of biology, enabling living systems to access chemical space previously only open to
synthetic chemistry. We have discovered that heme proteins catalyze the formation of
organosilicon compounds under physiological conditions via carbene insertion into silicon–
hydrogen bonds. The reaction proceeds both in vitro and in vivo, accommodating a broad
range of substrates with high chemo- and enantioselectivity. Using directed evolution, we
enhanced the catalytic function of cytochrome c from Rhodothermus marinus to achieve
more than 15-fold higher turnover than state-of-the-art synthetic catalysts. This carbon–
silicon bond-forming biocatalyst offers an environmentally friendly and highly efficient route
to producing enantiopure organosilicon molecules.
Silicon constitutes almost 30% of the mass of Earth’s crust, yet no life form is known to have the ability to forge carbon–silicon bonds (1). Despite the absence of organo- silicon compounds in the biological world,
synthetic chemistry has enabled us to appreciate
the distinctive and desirable properties that have
led to their broad applications in chemistry and
material science (2, 3). As a biocompatible carbon isostere, silicon can also be used to optimize
and repurpose the pharmaceutical properties of
bioactive molecules (4, 5).
The natural supply of silicon may be abundant, but sustainable methods for synthesizing
organosilicon compounds are not (6–8). Carbon–
silicon bond-forming methods that introduce
silicon motifs to organic molecules enantioselectively rely on multistep synthetic campaigns
to prepare and optimize chiral reagents or catalysts; precious metals are also sometimes needed
to achieve the desired activity (9–19). Synthetic
methodologies such as carbene insertion into
silanes can be rendered enantioselective using
chiral transition metal complexes based on rhodium (11, 12), iridium (13), and copper (14, 15).
These catalysts can provide optically pure products, but not without limitations: They require
halogenated solvents and sometimes low temperatures to function optimally and have limited
turnovers (<100) (16).
Because of their ability to accelerate chemical
transformations with exquisite specificity and
selectivity, enzymes are increasingly sought-after
complements to, or even replacements for, chemical synthesis methods (17, 18). Biocatalysts that
are fully genetically encoded and assembled inside
of cells are readily tunable with molecular biology
techniques. They can be produced at low cost
from renewable resources in microbial systems
and perform catalysis under mild conditions.
Although nature does not use enzymes to form
carbon–silicon bonds, the protein machineries
of living systems are often “promiscuous”—that
is, capable of catalyzing reactions distinct from
their biological functions. Evolution, natural or
in the laboratory, can use these promiscuous
functions to generate catalytic novelty (19–21).
For example, heme proteins can catalyze a variety of non-natural carbene-transfer reactions
in aqueous media, including N–H and S–H insertions, which can be greatly enhanced and made
exquisitely selective by directed evolution (22–24).
We hypothesized that heme proteins might
also catalyze carbene insertion into silicon–
hydrogen bonds. Because iron is not known to
catalyze this transformation (25), we first examined whether free heme could function as a
catalyst in aqueous media. Initial experiments
showed that the reaction between phenyldimethylsilane and ethyl 2-diazopropanoate (
Me-EDA) in neutral buffer (M9-N minimal medium,
pH 7.4) at room temperature gave racemic organosilicon product 3 at very low levels, a total turnover number (TTN) of 4 (Fig. 1A). No product
formation was observed in the absence of heme,
and the organosilicon product was stable under
the reaction conditions.
We next investigated whether heme proteins
could catalyze the same carbon–silicon bond-forming reaction. Screening a panel of cytochrome
P450 and myoglobin variants, we observed product formation with more turnovers compared to
the hemin and hemin with bovine serum albumin
(BSA) controls, but with negligible enantioinduc-tion (table S4). Cytochrome c from Rhodothermus
marinus (Rma cyt c), a Gram-negative, thermo-halophilic bacterium from submarine hot springs
in Iceland (26), catalyzed the reaction with 97%
enantiomeric excess (ee), indicating that the reaction took place in an environment where the
protein exerted excellent stereocontrol. Bacterial
cytochromes c are well-studied, functionally conserved electron-transfer proteins that are not
known to have any catalytic function in living
systems (27). Other bacterial and eukaryotic cytochrome c proteins also catalyzed the reaction,
but with lower selectivities. We thus chose Rma
cyt c as the platform for evolving a carbon–silicon
The crystal structure of wild-type Rma cyt c
[Protein Data Bank (PDB) ID: 3CP5; (26)] reveals
that the heme prosthetic group resides in a hydrophobic pocket, with the iron axially coordinated
to a proximal His (H49) and a distal Met (M100),
the latter of which is located on a loop (Fig. 1, B
and C). The distal Met, common in cytochrome c
proteins, is coordinatively labile (28, 29). We hypothesized that M100 must be displaced upon
iron-carbenoid formation, and that mutation
of this amino acid could facilitate formation of
this adventitious “active site” and yield an improved carbon–silicon bond-forming biocatalyst.
Therefore, a variant library made by site-saturation
mutagenesis of M100 was cloned and recombinantly expressed in Escherichia coli. After protein
expression, the bacterial cells were heat-treated
(75°C for 10 min) before screening in the
presence of phenyldimethylsilane (10 mM),
Division of Chemistry and Chemical Engineering, California
Institute of Technology, Pasadena, CA 91125, USA.
*Corresponding author. Email: email@example.com