Tracking Down a New Steroid-Hydroxylating Promiscuous Cytochrome P450: CYP154C8 from Streptomyces sp. W2233-SM

Article abstract of DOI:10.1002/cbic.201800018

CYP154C8 from Streptomyces sp. has been identified as a new cytochrome P450 with substrate flexibility towards different sets of steroids. In vitro treatment of these steroids with CYP154C8 revealed interesting product formation patterns with the same group of steroids. NMR study revealed the major product of corticosterone to be hydroxylated at the C21 position, whereas progesterone, androstenedione, testosterone, and 11-ketoprogesterone were exclusively hydroxylated at the 16α position. However, the 16α-hydroxylated product of progesterone was further hydroxylated to yield dihydroxylated products. 16-hydroxyprogesterone was hydroxylated at two positions to yield dihydroxylated products: 2α,16α-dihydroxyprogesterone and 6β,16α-dihydroxyprogesterone. To the best of our knowledge, this is the first report of generation of such products through enzymatic hydroxylation by a CYP450. In view of the importance of modified steroids as pharmaceutical components, CYP154C8 has immense potential for utilization in bioproduction of hydroxylated derivative compounds to be directly employed for pharmaceutical applications.

Full text of DOI:10.1002/cbic.201800018

Accepted Article
Title: Tracking Down a Novel Steroid Hydroxylating Promiscuous
Cytochrome P450, CYP154C8 from Streptomyces sp. W2233-
SM
Authors: Bikash Dangi, Ki-Hwa Kim, Sang-Ho Kang, and Tae-Jin Oh
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemBioChem 10.1002/cbic.201800018
Link to VoR: http://dx.doi.org/10.1002/cbic.201800018
A Journal of
www.chembiochem.org

ChemBioChem
10.1002/cbic.201800018
1
2
Tracking Down a Novel Steroid Hydroxylating Promiscuous Cytochrome P450, CYP154C8
from Streptomyces sp. W2233-SM
3
4
5
1
1
2
1,3,4
Bikash Dangi , Ki-Hwa Kim , Sang-Ho Kang , and Tae-Jin Oh
*
1
6
7
Department of Life Science and Biochemical Engineering, SunMoon University, 70 Sunmoon-ro
221, Tangjeong-myeon, Asan-si, Chungnam 31460, Republic of Korea
2
8
9
Genomics Division, National Institute of Agricultural Sciences, RDA, Jeonju, 54874, Republic
of Korea
3
1
1
0
1
Department of Pharmaceutical Engineering and Biotechnology, SunMoon University, 70
Sunmoon-ro 221, Tangjeong-myeon, Asan-si, Chungnam 31460, Republic of Korea
4
1
1
2
3
Genome-based BioIT Convergence Institute, 70 Sunmoon-ro 221, Tangjeong-myeon, Asan-si,
Chungnam 31460, Republic of Korea
1
1
1
1
4
5
6
7
Author for correspondence:
Prof. Tae-Jin Oh
1
1
2
2
2
8
9
0
1
2
Tel: +82(41)530-2677;
Fax: +82(41)530-2279;
E-mail: tjoh3782@sunmoon.ac.kr
2
3
Running title: Steroid hydroxylase CYP154C8
1
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
24
25
26
27
28
29
30
31
32
33
34
35
36
Abstract: CYP154C8 from Streptomyces sp. was identified as a novel cytochrome P450 with
substrate flexibility to different sets of steroids. The in vitro reaction of these steroids with
CYP154C8 revealed interesting product formation patterns with the same group of steroids.
Nuclear magnetic resonance study revealed the major product of corticosterone hydroxylated at
the C21 position; while progesterone, androstenedione, testosterone, and 11-ketoprogesterone
were exclusively hydroxylated at 16α position. However, the 16α-hydroxylated product of
progesterone was further hydroxylated to yield dihydroxylated products. 16-
hydroxyprogesterone was hydroxylated at two positions yielding 2α,16α-dihydroxyprogesterone
and 6β,16α-dihydroxyprogesterone. To our knowledge, this is the first report of generation of
such products in enzymatic hydroxylation by CYP450. Considering the importance of modified
steroids as pharmaceutical components, CYP154C8 has immense potential to be utilized in
bioproduction of the hydroxylated derivative compounds directly employed for pharmaceutical
applications.
3
7
Introduction
[
1]
38
39
40
41
42
43
44
45
Cytochrome P450s (CYP450s) are versatile hemoproteins present in all forms of life. They
catalyze oxyfunctionalization of a broad variety of molecules, including steroids, xenobiotics,
[2, 3]
alkanes, terpenes, antibiotics, and fatty acids.
Many physiological functions of bacterial
CYP450s remain unknown. However, their flexibility to a vast range of substrates and potential
to introduce hydroxyl groups into inactive carbon of different molecules in a regio- and stereo-
[4-8]
selective manner make them fascinating catalysts for pharmaceutical industries.
Typically,
hydroxylating compounds can bring dramatic changes to efficacy, activity, toxicity, and
solubility in parent molecules.
2
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
46
47
48
49
50
51
52
53
54
55
56
57
58
Steroids are important compounds with a wide range of therapeutic properties such as
anticancer, antiandrogenic, progestational, immunosuppressive, diuretic, anti-inflammatory, and
[9, 10]
contraceptive activities.
The physiological function of steroids depends on their structures,
especially determined by type, number, and position of their functional groups (Scheme 1). For
[11]
example, oxyfunctionalization at 11β is crucial for their anti-inflammatory activity.
Industrially, regio- and stereo-selective hydroxylation of steroids has important applications for
synthesizing highly valuable steroid hormones such as mineralocorticoids, glucocorticoids, and
sex hormones. It has been reported that 11α- and 11β-hydroxyprogesterone play an important
role in controlling blood pressure with anti-androgenic activity but with minimal estrogenic and
[7, 12–14]
progestational side effects.
Reichstein S, a derivative of progesterone and a corticosteroid
precursor with anti-inflammatory activity, is industrially hydroxylated at 11β position.
Pregnenolone is synthesized from ergosterol using recombinant Saccharomyces cerevisiae
[15]
harboring CYP450 and its redox partner adrenodoxin and adrenodoxin reductase.
59
60
61
62
63
64
65
66
67
68
A number of eukaryotic CYP450s involved in steroid synthesis are known, their
characteristic expression in membrane-bound form with low catalytic efficiency has limited their
potential uses in biotechnology applications. But bacterial CYP450s overexpressed in soluble
[16, 17]
form, making them suitable for industrial use to produce valuable compounds.
However,
only a few bacterial CYP450s capable of hydroxylating steroids have been characterized.
CYP106A2 from Bacillus megaterium ATCC 13368 and CYP106A1 from B. megaterium are
[3, 14]
known for steroid hydroxylating activity.
Similarly, CYP514C3 from S. griseus and
CYP154C5 from Nocardia francinica can oxidize steroids such as progesterone, testosterone,
[16, 18]
nandrolone, dehydroepiandrostenedione, and androstenedione at 16α position.
Recently, it
has been shown that CYP260A1 and CYP260B1 from Sporangium cellulosum and CYP109E1
3
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
[
8, 19-21]
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
from B. megaterium DSM319 can also hydroxylate steroids.
CYP154 group of enzymes
are widely present in Actinomycetes, and many of them have been identified in Streptomyces
species. Some of the CYP154 genes have been expressed and characterized. CYP154C1 has
shown an in vitro activity towards 14 and 12-membered ring macrolide YC-17 and narbomycin,
[22]
respectively. Similarly, CYP154A1 also was found to bind narbomycin, however, no catalysis
[23]
was observed with the same substrate. Interestingly, the same enzyme was shown to catalyze
unexpected intramolecular cyclization of dipentaenone to a Paterno`-Bu¨chi-like product,
[24]
without oxidation/reduction.
The crystal structure of CYP154C1 and CYP154A1 from
Streptomyces coelicolor A3(2) has been already been elucidated. A wide range of substrate
screening shows CYP154A8 from N. francinica and CYP154E1 from Thermobifida fusca can
[25]
oxidize fatty acids, primary alcohols, secondary alcohols, and terpenoids. CYP154H1 from T.
fusca oxidizes organic sulfides, different arylaliphatic sulfides, indole and indole derivatives, and
[26]
small arylaliphatic substrates like ethylbenzene, propylbenzene, and styrene.
Recently,
CYP154F1 from T. fusca has been characterized which preferably accepts small linear molecules
with a carbon chain length of 8-10 atoms. However, a mutant form of it was able to convert
[27]
larger substrates like (E)-stilbene and (+)-nootkatone.
In this study, we identified another
novel steroid hydroxylating CYP450 from Streptomyces sp. W2233-SM belonging to CYP154
family and studied its potential use in hydroxylation of diverse sets of steroids in vitro and in
vivo using Escherichia coli cells.
8
8
8
9
Results and Discussion
Screening of steroid modification by Streptomyces strain
Initially, Streptomyces sp. W2233-SM was checked for its biotransformation ability to diverse
9
9
0
1
steroids. Ethyl acetate extracts of whole-cell bioconversion showed multiple peaks in HPLC
4
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
92
93
94
95
96
97
98
99
chromatograms of steroid substrates used for biotransformation (Figure S1). Therefore, reaction
mixtures were further analyzed by high-resolution electrospray ionization mass spectrometry
(HR-ESI/MS) in positive ion mode. Mass spectra of these new peaks resembled exact masses of
monohydroxylated products of substrates used. The substrate progesterone with chemical
+
+
+
formula of C21
H
31
O
2
had exact mass of m/z [M+H] 315.2557 and a single product (P) had
+
+
+
+
exact mass of m/z [M+H] 331.2279, whose calculated m/z [M+H] value is 331.2268 (Figure
S1A). In case of androstenedione, two products were obtained whereas P1 was identified as
+
+
monohydroxylated product with an exact mass of m/z [M+H] 303.2019, whose calculated mass
+
1
1
1
1
1
00 for the chemical formula C19
H
27
O
3
is 303.1955 (Figure S1B). The substrate androstenedione
+
2
01
(C19
H
27
O
) had an exact mass of 287.2161. Similarly, among the three products peak of
+
+
02 nandrolone, P1 was found to be a hydroxylated product that had an exact mass of m/z [M+H]
+
03
04
291.1955 for which calculated mass for molecular formula C18
H
27
O
3
was 291.1949. Nandrolone
+
with chemical formula C18
H
27
O
2
had the exact mass of 275.2004 (Figure S1C).
1
05
Identification of strain
106 Based on 16S rRNA gene sequence, the strain showed close similarity with the genus
107 Streptomyces. A comparison of 16S rRNA gene sequences from the selected strains belonging to
108 genus Streptomyces is shown in the supplementary information (Figure S2).
1
09
Selection of CYP450 responsible for steroid hydroxylation
1
1
1
1
10
11
12
Among the 20 putative CYP450 in the genome of Streptomyces sp. W2233-SM, CYP154C8
showed the highest similarity with previously well-known steroid hydroxylases such as
CYP154C3 (74%) and CYP154C5 (66%). Phylogenetic tree analysis of CYP154C8 also
13 revealed the closest relation with CYP154C3 and CYP154C5 among the CYP154 group of
5
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
1
14 CYP450s (Figure 1). Hence, we speculated the whole-cell bioconversion of steroids in wild-type
15 strain might have been catalyzed by CYP154C8.
1
1
1
16
17
Cloning, overexpression, purification, and spectral characterization of CYP154C8 and
redox partner
1
1
1
1
1
18
CYP154C8-encoding gene was PCR amplified from the Streptomyces sp. W2233-SM using a
19
pair of primers as described in materials and methods. The sequence-confirmed gene was cloned
20 into a pET32a(+) vector under control of a T7 promoter. The redox partner, camA and camB
21
22
were cloned separately into pET28a(+) and pET32a(+), respectively. To develop the in vivo
expression system, camA and camB were also cloned into pAYAC-Duet1 vector.
1
1
1
1
1
1
1
1
1
1
23
All proteins were overexpressed and purified using recombinant E. coli BL21(DE3)
24
(Figure 2A). Theoretical molecular weight calculated for CYP154C8 was found to be ~47 kDa,
25 however, the SDS-PAGE analysis showed a band at ~70 kDa due to Trx-His-s-enterokinase
26 fusion sequence of pET32a(+) vector transcribed and translated along with the sequence of
27 CYP154C8. The purified form of CYP154C8 showed a Soret peak at 419 nm in an oxidized
28 form. Carbon monoxide-bound and dithionite-reduced form of CYP154C8 exhibited maximum
29 absorption at a wavelength of 449 nm, one of the major spectral characteristics of CYP450
30
(Figure 2B). The purity of CYP450 can be evaluated by Rz value, which is calculated as the ratio
31
of absorbance (A) at λmax of the soret band to the A at 280 nm. The purified enzyme had the Rz
[28]
32 value of 1.45, indicating high purity.
1
33
Substrate binding assay
Substrate binding to active CYP450 sites displaces the water molecule at the sixth coordination
1
34
[29]
1
35 position of heme iron, thereby causing a shift of heme iron from low spin to high spin.
6
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
1
1
1
1
1
1
1
1
1
1
1
36
CYP450 unbound with any substrate will show a maximum absorption at approximately around
420 nm, whereas substrate binding will cause a minimum absorption at 420 nm with a maximum
37
[30]
38 absorption at 390 nm. This is called type I spectral change. CYP154C8 exhibited maximum A
39 at 389 nm at high spin and at 419 nm at low spin for all substrates (Figure S3). Steroids with
40 higher hydrophobicity (group I) showed markedly lower K values, indicating tight binding.
Progesterone, androstenedione, testosterone, and nandrolone had K values of less than 0.5 µM.
However, substrates such as prednisone and cortisone had K values of 2.50 µM and 4.16 µM,
respectively, while prednisolone and hydrocortisone had K values of 12.78 µM and 15.80 µM,
respectively. 16α-hydroxyprogesterone showed the K value of 1.21 ± 0.1. Progesterone as a
value at 0.083 µM while hydrocortisone had the highest K value at
d
41
42
43
44
45
46
d
d
d
d
substrate had the lowest K
d
d
15.80 µM (Table 1).
1
1
1
1
1
1
1
1
1
1
1
47
Corticosterone did not induce a high-spin shift of heme among steroid substrates used in
48
this study. Group I steroids K
d
value appeared to be exceptionally small. These steroids had K
d
d
49
value below 0.5 µM, indicating high affinity, while group II steroids had comparatively large K
50
values. This clearly indicates with the increase in the hydrophobicity of the substrates, the
binding affinity of CYP154C8 increases. As reported previously, CYP154C3 and CYP154C5
51
52
also displayed tight binding (low K ) with steroids. Tight binding is likely to be the consequence
d
53
of steroid hydrophobicity which leads to an unfavorable entropy reduction of the water
molecules surrounding steroids in an aqueous environment of the active site. Substrate binding to
the active site releases the low-entropic water molecule from the solvation shell of steroids, thus
increasing the entropy of the system. This result in large hydrophobic interaction interfaces
54
55
56
[31]
57 which promotes the overall process of steroids binding to the active site of CYP450.
1
58
In vitro assay analysis and product identification
7
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
The in vitro reaction mixture of CYP154C8 reconstituted with PDX (redoxin) and PDR
(reductase) as redox partner, and NADH as cofactor showed a single product peak for the group I,
III, and IV steroids (Figure 3A), while group II substrates had multiple products peaks (Figure
3B). LC-MS analysis of the reaction mixtures showed the exact mass
of monohydroxylated products of the respective substrate (Table S1). However, progesterone
reaction mixture showed three product peaks (P1, P2, and P3), where P1 and P2 were shown to
have a very low conversion, and P1 was found to be a major product (Figure 4 inset I). LC-MS
analysis of the reaction mixture showed P1 and P2 as dihydroxylated products, and P3
a monohydroxylated product of progesterone (data not shown). We thought the two
dihydroxylated products might be due to subsequent hydroxylation of the 16α-
hydroxyprogesterone by CYP154C8. To confirm it, 16α-hydroxyprogesterone was purified as
described in material and method, and it was used as a substrate for in vitro reaction with
CYP154C8. Reaction mixture analysis of 16α-hydroxyprogesterone showed two product peaks
(P1 and P2) in HPLC (Figure 4 inset II); P1 and P2 were found to be a major and a minor
product of 16α-hydroxyprogesterone, respectively. The same two products were confirmed by
74 HPLC (Figure 4 inset II) and LC-MS analyses (data not shown) which were obtained in an in
75
vitro reaction with the substrate progesterone. The two dihydroxylated products, P1 and P2 of
progesterone characterized by NMR were identified as 2α,16α-dihydroxyprogesterone (major)
and 6β,16α-dihydroxyprogesterone (minor), respectively. To our knowledge, this is the first
report of such dihydroxylated products formation by any CYP450 enzyme in a two-step
oxidation of progesterone. Although, S. roseochromogenes NCIB 10984 was shown to
hydroxylate exogenous progesterone to 16α-hydroxyprogesterone and subsequently to 2β,16α-
76
77
78
79
80
[32]
81 dihydroxyprogesterone.
Some of the CYP450s like CYP17A1, cytochrome P450scc
8
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
1
1
1
1
1
1
1
1
1
1
82
(CYP11A1), cytochrome P450BioI (CYP107H1), and MycG have been known to catalyze the
[33-42]
83 multi-step oxidation of their respective substrate in
a
sequential manner.
84 The monohydroxylated products of progesterone, androstenedione, testosterone, and 11-
1
13
85
ketoprogesterone analyzed by NMR ( H-NMR, C -NMR, and 2D-NMR) showed hydroxylation
at 16α-position. 16α-hydroxydehydroepiandrosterone and its 3β-sulfate ester, 16α-
hydroxydehydroepiandrosterone sulfate, are metabolic intermediates in the biosynthesis of estriol
from dehydroepiandrosterone during pregnancy. 16α-hydroxydehydroepiandrosterone has
86
87
88
[43, 44]
89 estrogenic activity.
16α-hydroxyprogesterone and 6β-hydroxyprogesterone have shown to
90
91
inhibit 5α-reductase, and these modified steroids can be good drug target for the prostate
[45]
cancer.
1
1
1
1
1
1
1
1
2
2
2
2
2
92
However, different product formation patterns were observed when substrates (group II) had
hydroxyl or carbonyl groups at C11 and C21 positions. Corticosterone was converted into three
93
94 products (Figure 3B inset II). HPLC chromatogram showed a flat peak at the base of a major
95 product (P2) which was unusual. To further characterize this peak, LC-MS and NMR analyses
96 were carried out. Hydroxylation at the C21 position of corticosterone was found, yielding (11β)-
97 11,21,21-trihydroxypregn-4-ene-3,20-dione (21-hydroxycorticosterone) which was an unusual
98
conversion (Scheme 2). This is the first report of such product formed by any enzymatic
99 conversion reactions. Although the same product showing similar peak characteristic (a flat peak
[46]
00 at the base) in HPLC chromatogram has been reported previously,
it was a hydrocortisone
01
02
03
04
degradation product synthesized by chemical method. A similar mechanism of forming such
[47–49]
corticosteroids has also been reported previously by chemical methods.
This product was
confirmed to be the result of CYP154C8 catalysis in this study by carrying out series of control
reactions. 21-hydroxycorticosterone might have resulted due to dehydrogenation of 21-hydroxyl
9
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
2
2
2
2
05 of corticosterone by CYP154C8 giving rise keto aldehyde form. A hydrated form of the product
06
07
(21-hydroxycorticosterone) is expected to be more favorable over the keto aldehyde form in
[46]
aqueous solution, although the keto aldehyde form was never detected in our reactions with
08 corticosterone.
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
09
The ions at m/z 121 and 145 of all three products (Figure S4) indicate the presence of an
[50, 51]
10 unmodified steroidal A/B-ring.
A possible position of hydroxylation of P1 was estimated at
11
12
13
14
15
16
17
18
C16 of D-ring based on HPLC analysis, LC-MS fragmentation, and characteristics of CYP154C8.
A comparative study of P1 with hydrocortisone (a compound with same molecular weight and
chemical formula as P1) showed no similarity in HPLC retention time, while LC-MS fragments
were found to be highly similar. It showed P1 was not a C17 hydroxylated product. Comparison
of the HPLC retention pattern of P1 with other known hydroxylated products (16α hydroxylated
products of progesterone, androstenedione, and testosterone) showed strong evidence of P1
being hydroxylated at C16 position (Figure 3). This was further supported by the characteristic
of CYP154C8 which had been shown to catalyze hydroxylation at C16 position. Another product
+
+
19 P3 had the mass of m/z [M+H] 361.2007 for which the calculated mass of molecular formula
+
20
21
22
23
24
25
C
21
H
29
O
5
was 361.2010. This might be due to the presence of a double bond, a keto group, or
[19]
epoxide formation as keto groups and epoxides can form from hydroxyl groups. It is possible
that P3 might be a product obtained by oxidation of P2 at C21 (C-OH → C=O) to give keto acid.
All group II substrates were hydroxylated in multiple positions with the characteristic of peaks
similar to those of corticosterone P2. Therefore, those peaks might represent glyoxal hydrate of
respective substrates, similar to corticosterone P2 (glyoxal hydrate). Functionalization at C11
26 position of steroids might play an important role in positioning steroids at the active site of
27 CYP154C8. To determine such possibility, more steroids (group III) with the functional group
1
0
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
28
29
30
31
32
(hydroxyl or carbonyl) at C11 position were selected. However, a single product was detected in
HPLC (Figure 3A inset IV and VI) and LC-MS analysis (Table S1). Speculating that the
presence of C21 hydroxyl group might have a role in multiple product formations, steroids
lacking C11 functional group but having C21 hydroxyl group (group IV) were further selected.
A single monohydroxylated product was also detected by HPLC (Figure 3A inset V) and LC-MS
33 (Table S1). This clearly indicates that CYP154C8 catalyzes hydroxylation in more than one
34
35
36
position when the substrate has a functional group (hydroxyl or carbonyl) at C11 and C21
positions. P2 of group II steroids was a result of such structure. Hence, group II steroids having a
hydroxyl or carbonyl functional group at C11 and C21 positions were metabolized into two or
37 more products. All substrates that were hydroxylated at one position were believed to be 16α-
38
39
hydroxylated products based on HPLC peak analysis, substrate structure, and comparative
studies of CYP154C8 with already characterized similar proteins (CYP154C5 and CYP154C3).
40 Although CYP154C8 has a major characteristic of hydroxylating at C16α, the presence of
41
42
43
hydroxyl or carbonyl groups at C11 and C21 might have changed the conformation of a substrate
at active sites. It has been previously reported that CYP154C5 shows major differences in steroid
binding orientation between C17-acetyl-substituted steroids (progesterone and pregnenolone)
[31]
44 and C17-hydroxyl/carbonyl substituted steroids (androstenedione and testosterone). Therefore,
45
46
47
48
steroid orientation is highly influenced by C17 substituents. Other substrates with bulky C17
substituent groups such as lanosterol, cholesterol, and desmosterol were also screened in this
study. However, no conversion was found in vitro (data not shown). Dexamethasone, a C16-
methylated steroid, also showed no conversion. Protein multiple sequence alignment of
49 CYP154C8 and CYP154C5 shows a few differences in the active sites (Figure S5). Polar amino
50 acid Gln239 is thought to play an important role in the dynamic process of steroid binding at the
1
1
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
[
31]
251
252
253
254
255
active site of CYP154C5. It was substituted by Lys249 in CYP154C8. Other active site amino
acids Val87 and Val291 in CYP154C5 were replaced by Ala98 and Thr301 in CYP154C8,
respectively. CYP154C5 has been reported in bioconversion of the only group I like substrates; it
would be interesting to see the products of group II substrates catalyzed by it, which has not been
previously reported.
2
56 Kinetic analysis
2
2
2
2
2
57
58
59
Progesterone was found to be the most preferred substrate in in vitro conversion assay. Hence,
we selected the substrate progesterone to determine the kinetic parameters of CY154C8-
catalyzed hydroxylation. The kinetics parameters for CYP154C8 were determined by plotting
60 product formation rate vs substrate concentration. The K and kcat values were estimated to be
m
-1
61
62
28.31 ± 3.35 µM and 2.38 ± 0.078 min , respectively (Figure 5).
2
Biotransformation of steroids using recombinant E. coli
2
2
2
2
2
2
2
2
2
2
63
64
65
66
The in vivo bioconversion of steroids by E. coli BL21(DE3) cells harboring pAYAC_camAB
and pET32a_CYP154C8, and only pET32a_CYP154C8 was found to be similar. Expression
level analysis of CYP154C8, PDX, and PDR by SDS-PAGE showed a clear band of CYP154C8
in all system, however, no clear band of PDX and PDR was observed (Figure S6). This showed
67 that the intrinsic redox partner of E. coli BL21(DE3) effectively transfer reducing equivalents to
+
68 CYP154C8. In E. coli, the ferredoxin-NADP oxidoreductase transfers reducing equivalent
69
70
71
72
between NADPH and iron-sulfur clusters of various ferredoxins, and in some cases NADPH and
[52]
the FMN containing flavodoxin.
As reported previously, bovine cytochrome P450C17
hydroxylase activity is supported by NADPH-flavodoxin (ferredoxin) oxidoreductase in
[53]
combination with flavodoxin. Another cytochrome P450cin (CYP176A) is believed to have a
1
2
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
2
2
2
2
73 flavodoxin as redox partner that might utilize the endogenous E. coli flavodoxin/flavodoxin
[54]
74
75
76
reductase system. This shows endogenous flavodoxin and flavodoxin reductase might have
possibly involved in transferring reducing equivalents to CYP154C8. However, the involvement
of ferredoxin and ferredoxin reductase cannot be ignored.
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
77
78
79
80
81
82
83
In vivo bioconversion of C21 hydroxysteroids (prednisolone, hydrocortisone, prednisone,
cortisone, and 11-deoxycortisol) except corticosterone showed a major peak that had longer
retention time than the substrate peak (Figure S7 inset I). LC-MS analysis of products revealed
the exact mass of acetylated steroid of the respective substrate (Table S2). Cortisone and
prednisone were most preferred substrates for formation of such product (Figure S8). A similar
phenomenon has been previously reported where chloramphenicol acetyltransferase I (CATI)
was found to transfer an acetyl group to C21 hydroxysteroids. All the substrates were selectively
[55]
84 transformed into corresponding 21-acetoxy derivatives. while the hydroxylated products from
85
86
these substrates except 11-deoxycortisol was found to be a minor product (less than 1%).
Surprisingly, no acetylated product of corticosterone was detected which is also a C21
87 hydroxysteroid. The in vitro reaction with acetyl-CoA did not show any acetylated products for
88
89
90
91
such substrates. The plasmid vector pAYACDuet-1 used for the cloning of camA and camB gene
as redox partner contains chloramphenicol resistance gene. The activation of the
chloramphenicol acetyltransferase in E. coli to detoxify the antibiotic chloramphenicol might
[56]
have involved in the acetylation of C21 steroids. To confirm about it, the pCDFDuet-1 vector
92 which does not contain chloramphenicol resistance gene was selected for the expression of camA
93
94
95
and camB in E. coli cells. The In vivo bioconversion of such steroids harboring
pET32a_cyp154C8 and pCDF_camAB recombinant plasmids showed no possible acetylated
product peak in HPLC chromatogram (Figure S7 inset II). This clearly indicates that C21
1
3
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
2
2
2
2
3
3
3
3
3
3
3
96
97
98
99
00
01
02
hydroxysteroids were acetylated by chloramphenicol acetyltransferase. The hydroxylated
product yield of C21 hydroxysteroids with the pCDFDuet-1 vector system was not found to have
improved, however, the bioconversion of progesterone and androstenedione was increased,
comparatively (Figure 6). Progesterone was most favored substrate with overall bioconversion of
~93% that showed a major conversion (~90%) into a monohydroxylated product (16α-
hydroxyprogesterone) and a low conversion (~3%) into two dihydroxylated products (2α,16α-
dihydroxyprogesterone and 6β,16α-dihydroxyprogesterone ). The least favored substrate in vivo
03 were group II substrates which had a very low bioconversion into hydroxylated products. The
04 bioconversion of corticosterone was estimated to be 7.2 ± 2.5 %. The group I substrates
05 (progesterone, androstenedione, testosterone, and nandrolone) along with 11-ketoprogesterone
06
07
were the most preferred substrates by an in vivo bioconversion (Figure S9).
3
Conclusions
3
3
3
3
3
3
3
3
3
08
09
10
11
12
13
14
15
We have investigated a new steroid hydroxylating CYP450 by genome mining. CYP154C8 was
found to exhibits oxyfunctionalization of broad steroid substrates. The regio- and stereo-specific
hydroxylation at C16α position with the steroids lacking functional group at C11 and C21
position, and stereo- and regio-selective hydroxylation of corticosterone at C21 position bearing
functional group at both positions is an interesting mechanism of functioning. Along with this,
CYP154C8 can subsequently hydroxylate 16α-hydroxyprogesterone to give rise two
dihydroxylated products. We believe CYP154C8 can be a suitable model, and biocatalyst for
tailoring CYP450 activities to generate steroids with therapeutic values and other high-value
16 metabolites.
3
17 Experimental Section
1
4
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
3
3
3
3
3
3
3
3
18 Chemicals and reagents: All steroid substrates and sodium formate were purchased from Tokyo
19
20
21
Chemical Industry Co., Ltd. (Korea). Isopropyl-1-thio-β- -galactopyranoside (IPTG), 1,4-
D
dithiothreitol (DTT), and kanamycin were obtained from Duchefa Bohemie (Korea). Ampicillin
(Amp), chloramphenicol (Cm), α-aminolevulinic acid (ALA), nicotinamide adenine dinucleotide
22 (NADH), cytochrome C, catalase, and formate dehydrogenase were purchased from Sigma-
23
24
25
Aldrich (Korea). Restriction enzymes were obtained from Takara Clontech (Korea). T4 DNA
ligase, DNA polymerase, and dNTPs were from Takara Bio (Japan). All other chemicals were
high-grade products obtained from commercially available sources.
3
3
3
3
3
3
3
3
3
3
3
3
3
3
26
27
28
29
30
31
32
33
34
35
36
37
Sequence accession numbers and bioinformatics analysis: Gene responsible for encoding
CYP450 was found based heme-binding domain signature, FxxGx(H/R)xCxG, as query
sequence from Streptomyces sp. W2233-SM genome (unpublished). The nucleotide sequence of
CYP154C8 and 16S ribosomal RNA gene sequence of the strain has been deposited into the
GenBank under the accession number MF398962 and MG198705, respectively. Microbial
identification was carried out based on 16S ribosomal gene sequence. The name of CYP450 was
assigned by Dr. David Nelson (http://drnelson.utmem.edu/CytochromeP450.html). The
[57]
evolutionary history was inferred using the Neighbor-Joining method.
The optimal tree with
the sum of branch length = 6.82869639 is shown. The percentage of replicate trees in which the
associated taxa clustered together in the bootstrap test (500 replicates) are shown below the
[58]
branches.
The tree is drawn to scale, with branch lengths in the same units as those of the
evolutionary distances used to infer the phylogenetic tree. The evolutionary distances were
[59]
38 computed using the Poisson correction method and are in the units of the number of amino
39 acid substitutions per site. The analysis involved 25 amino acid sequences. All positions
1
5
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
3
3
40
41
containing gaps and missing data were eliminated. There were a total of 289 positions in the final
[60]
dataset. Evolutionary analyses were conducted in MEGA6.
3
3
3
3
3
3
3
3
3
3
3
3
3
42
43
44
45
Strains, media, and conditions: Wild-type Streptomyces sp. W2233-SM isolated from Korean
TM
-1
soil was cultivated in Bacto Tryptic Soy Broth (TSB; 17.0 g L pancreatic digest of casein, 3.0
-1
-1
-1
-1
g L papaic digest of soybean, 5 g L sodium chloride, 2.5 g L dextrose, and 2.5 g L
dipotassium phosphate, media was adjusted to pH 7.3) at 28°C with shaking (200 rpm). After
46 120 h of culture, the cells were collected by centrifugation followed by washing twice with Tris
47 buffer (pH 7.4). The washed cells were resuspended in potassium phosphate buffer (pH 7.4) and
48
stored in -20°C until use. E. coli XL1 Blue (Stratagene, La Jolla, CA, USA), pMD20-T (Takara),
49 pET28a(+), pET32a(+), and pAYAC-Duet1 (Novagen, Germany) were used for sub-cloning and
50
51
52
DNA manipulation. E. coli BL21(DE3) (Stratagene) was used for recombinant protein
expression and whole-cell biotransformation. All E. coli strains were grown at 37°C in Luria-
Bertani (LB) media or plates supplemented with ampicillin (100 µg/ml), streptomycin (50
53 µg/ml), and chloramphenicol (34 µg/ml) when required. X-gal and IPTG were used for colony
54
55
screening and protein expression.
3
3
3
3
3
3
3
3
Whole-cell biotransformation by wild-type strain: Streptomyces sp. W2233-SM were
56 screened for biotransformation capability of steroids. The bioconversion assay was performed in
57
58
59
60
potassium phosphate buffer (pH 7.4) with substrates (progesterone, androstenedione, and
testosterone) at concentration of 1 mM at 28°C (200 rpm). After 24 h of bioconversions, the
reaction mixture was extracted with ethyl acetate. The ethyl acetate fractions were collected,
dried, and reconstituted in 500 ml of acetonitrile. High-performance liquid chromatography
61 photo-diode array and liquid chromatography-mass spectrometry (LC-MS) analysis were used
62 for further characterization.
1
6
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
63
64
65
66
67
68
Cloning and overexpression of CYP154C8 and redox partner: The CYP154C8 encoding
sequence (1,266 bp, 421 amino acids, accession number MF398962) was amplified from the
genomic DNA of Streptomyces sp. W2233-SM. The PCR primers used for amplification of the
gene were designed as 5’ - GAA TTC ATG AAC GGT CAG TCA GCG A - 3’ (EcoRI) as
forward and 5’ - AAG CTT TCA GCT GCC GTG GAG CA - 3’ reverse primer (HindIII). The
letters underlined indicate the restriction site for an endonuclease in bracket, respectively. The
69 PCR product obtained was cloned into pMD20-T vector using E. coli XL1-Blue and the
70
71
72
73
74
75
76
nucleotide sequence was confirmed by automated sequencing (Macrogen, Korea). Further, the
gene product was ligated into pET32a(+) vector to create pET32aCYP154C8 construct. The
resulting construct encoding N-terminal His6-tag protein under the control of a T7 promoter was
transformed into chemically competent E. coli Bl21 (DE3) and plated on LB agar containing 100
µg/ml ampicillin. A single colony from the plate was selected and grown overnight at 37°C. 1 ml
of overnight grown seed culture was added to 100 ml of LB-medium supplemented with 100
µg/ml ampicillin and incubated at orbital shaker (180 rpm) at 37°C until cell density was about
77 0.6-0.8 at OD600nm. Cultures were induced with 0.5 mM of IPTG, and supplemented with 1 mM
78 of 5-aminolevulinic acid hydrochloride (5-ALA) and 0.5 mM FeCl to support heme synthesis.
79 The cells were incubated for 48 h at 20°C. The cell pellets were harvested by centrifugation
80 (3,500 rpm) for 20 min at 4°C and washed twice with 50 mM Tris-HCl buffer (pH 7.4)
81 containing 10% glycerol, 100 mM NaCl, and 1 mM DTT. The in vitro reconstituted system was
3
[61]
82
83
84
85
supported by redox partners PDX (camA) and PDR (camB).
They were expressed as His-
tagged proteins in E. coli BL21(DE3) using plasmid constructs pET28a(+) and pET32a(+) as
[62]
described previously. For the in vivo system, the camA-harboring plasmid pET28a_PDX was
digested with NdeI and XhoI restriction enzymes to obtain camA. The camB-carrying plasmid
1
7
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
3
3
3
3
86
pET32a_PDR was digested with EcoRV and HindIII to obtain camB. These genes were
subsequently ligated into respective sites of vector pAYAC-Duet to create plasmid
pAYAC_camAB. Similarly, camA and camB genes previously cloned into pCDF-duet1 vector
87
88
[63]
89 was also used separately for in vivo conversion of steroid substrates.
3
3
3
3
3
3
3
3
3
3
90
CYP154C8 and redox partner purification: Cell pellets were harvested after protein
91 overexpression and lysed by ultra-sonication followed by centrifugation at 15,000 rpm for 40
2
+
92
93 using TALON His-tag; soluble protein extracts were mixed and agitated with TALON His-tag
94 resin on ice for 60 min. Protein-bound resins were pre-equilibrated with two-column volumes of
min at 4°C. Soluble fractions of these proteins were purified with Ni affinity chromatography
95 equilibrium buffer (potassium phosphate, pH 7.4). Bound proteins were then eluted with elution
96 buffer (potassium phosphate, pH 7.4, 10% glycerol, and 100 mM NaCl) containing 20 mM, 100
97
98
99
mM, and 250 mM imidazole. Fractions containing concerned proteins were pooled and
concentrated by ultra-filtration using Amicon centrifugal filters (Millipore) with molecular
weight cut-off (MWCO) of 30 kDa for CYP154C8 and PDR and MWCO of 10 kDa for PDX.
4
4
4
4
4
4
4
4
4
00 Determination of enzyme concentrations: CYP154C8 concentration was determined based on
[
64]
01
02
03
04
05
CO A-difference spectra as described previously. Proteins (2 ml) were diluted with potassium
phosphate buffer (50 mM, pH 7.4) and separated into two cuvettes. A pinch of sodium dithionite
was added into both cuvettes. One cuvette of samples was bubbled with CO. Both samples were
then scanned using a Biochrome Libra S35PC UV/Visible Spectrophotometer (England) between
wavelength of 400 and 500 nm repeatedly until the reading at 449 nm was lower than the highest
-1
-1
06 reading. CYP154C8 amount was calculated based on ε449 - 489 = 91 mM cm . PDR concentration
07 was determined as the average of concentrations calculated from wavelengths 378, 454, and 480
-1
-1
08
nm using extinction coefficients (ε) of 9.7, 10.0, and 8.5 mM cm , respectively, and PDX
1
8
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
4
4
4
4
09
concentration was also determined as the average of concentrations calculated from wavelengths
-1
-1
[65]
10 415 and 454 nm using extinction coefficients of 11.1 and 10.4 mM cm , respectively. PDX
11
12
and PDR activities were determined by monitoring cytochrome C reduction at 550 nm using
-1
-1 [66]
extinction coefficient of 19.1 mM cm .
4
4
4
4
4
4
4
4
4
4
4
4
13
Substrate binding assay: CYP154C8 dissociation constants (K ) were determined by titrating
d
14
substrate concentrations until they were saturated. CYP154C8 enzyme was prepared at
15 concentration of 1 µM in 50 mM phosphate buffer (pH 7.4). Steroid substrates were prepared in
16 DMSO. A values of all samples were measured at wavelength of 350–500 nm using Biochrome
17 Libra S35PC spectrophotometer. K of different steroids were obtained by fitting plots of
resulting A differences (Abs389 nm - Abs419 nm) against substrate concentrations using tight binding
d
18
19
[67]
equation as described previously. Graphpad prism 6 software was used to fit titration data
[
68]
20 points to the following quadratic equation : Aobs=Amax{([S]+[E
t
]+K
]))^0.5}/2[E], where Aobs was the A difference at any ligand concentration, Amax was the
A difference at ligand saturation, S was substrate concentration, E was enzyme concentration,
and K was dissociation constant for enzyme-ligand complex. The K values were calculated as
D
)-(([S]+[E
]+K
t D
)^2-
21
22
23
(4[S][E
t
t
d
d
24 average of three different experiments.
25 In vitro assay: The reaction mixture contained 3 µM CYP154C8, 6 µM PDR , 24 µM PDX , 1 U
4
4
4
4
4
4
4
26
27
28
29
formate dehydrogenase, 100 µg/ml catalase, 1 mM MgCl , 150 mM sodium formate, and 500
2
µM substrate in 50 mM potassium phosphate buffer (pH 7.4; 10% glycerol and 100 mM NaCl).
The reaction was initiated by adding 250 µM of NADH to make a final mixture volume of 500
µL. The reaction mixture was incubated at 30°C with shaking (200 rpm) for 2 h. All the steroid
30 substrates were prepared in DMSO at stock concentration of 20 mM. Reactions were then
31 extracted twice with 500 µl of ethyl acetate. Organic phases were collected and dried for analysis.
1
9
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
4
4
4
4
4
4
4
4
4
4
4
32
33
34
Kinetic analysis: NADH consumption rates were spectrophotometrically measured at 25°C in
phosphate buffer (50 mM, pH 7.4). The reaction mixture contained 1 µM CYP154C8, 1 µM
PDR, 18 µM PDX, 0.5 mM substrate, and 250 µM NADH. NADH consumption by enzymes
-1
-1
35 was measured at wavelength of 340 nm (ɛ= 6.22 mM cm ) with a UV-Vis scanning photometer
36 (UV-1601PC, Shimadzu, Tokyo, Japan). NADH consumption rates with different substrate
37 concentrations (10–180 µM) were measured to determine K . All reaction mixtures were
extracted with ethyl acetate after all NADH was consumed. Product formation was calculated
m
38
39
40
41
42
using the area of the product formed from HPLC. Coupling efficiency was calculated using
amounts of product formed at different substrate concentrations. Coupling efficiency (%) was
determined as the amount of product formed (µM)/total NADH consumed (µM) × 100. K
m
and
kcat values were calculated by plotting product formation rate vs. concentration.
4
4
4
4
4
4
4
4
4
4
4
4
43
Whole-cell bioconversion by E. coli BL21(DE3) and product purification for analysis:
Frozen E. coli cells stored at -20°C overexpressing CYP154C8, PDX, and PDR were
44
45 resuspended in potassium phosphate buffer (pH 7.4) to make the desired final OD600 value of 50
.
46
47
48
49
50
51
52
53
Bioconversion were carried out in 1 ml and 100 ml scale for analytical and preparative purpose,
respectively, at 30°C for 24 h and shaking 200 rpm with the addition of glucose (0.5 mg/ml). All
the steroids stocks were prepared in DMSO at the concentration of 100 mM. From these
prepared stocks, respective amounts of steroid were added to reaction mixtures to achieve final
steroid concentration of 1 mM. After 24 h of incubation, the bioconversions and the control were
extracted with ethyl acetate. 1 ml of bioconversions was extracted twice with 1 ml of ethyl
acetate while 100 ml of bioconversions was extracted with 300 ml of ethyl acetate. The
supernatant was dried under pressure and dissolved in HPLC-grade methanol. The samples were
54 filtered through 0.45 µm pore sized PTFE filter, and subjected to preparative HPLC (Shimadzu)
2
0
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
4
4
4
4
4
4
4
4
55
56
57
58
with a C18 column (Mightysil RP-18 GP, 150 × 4.6 mm, 5 µm, Kanto Chemical, Japan). To
estimate the amount of product formed, λmax (absorption maximum) of products and substrates
were identified. The peak area of each product and substrate were calculated by using their λmax
and conversion percentage were calculated. The hydroxylated product of corticosterone for NMR
59 analysis was prepared similar to an in vitro conversion assay described previously. A total
60
61
62
volume of 100 ml reaction mixture was distributed into 0.5 ml and incubated at 30°C and 200
rpm for 2 h. Further, the samples for purification of the hydroxylated product were prepared as
described above.
4
4
4
4
4
4
4
4
63
64
65
66
Analytical methods: Samples were injected into UHPLC and separated with Mightysil Reverse
phase C18 GP column (4.6 × 250 mm, 5 µm). Water (A) and acetonitrile (B) were used as mobile
phases for separation. The reaction mixtures were analyzed with a gradient system of B at 15%
for 0–10 min, 50% for 10–20 min, 70% for 20–25 min, and 15% for 25–40 min at a flow rate of
67 1 mL/min. Substrates and their products were detected by UV-A at 242 and 245 nm. Reaction
68 mixtures were analyzed with SYNAPT G2-S/ACUITY UPLC liquid chromatography
69 quadrupole time-of-light/electrospray ionization mass spectrometry (Waters, Milford, MA, USA)
70
in positive ion mode.
The purified hydroxylated products were dissolved in dimethyl-sulfoxide (DMSO-d
subjected to NMR analyses on 800 MHz using Varain Unity INOVA spectrometer (Varian, Palo
4
4
4
4
4
4
71
72
73
74
75
6
) and
1
13
Alto, CA, USA). One dimensional NMR ( H-NMR and C-NMR) was performed followed by
two-dimensional NMR, heteronuclear multiple bond correlation, correlation spectroscopy,
rotating-frame overhauser effect spectroscopy (ROESY), and heteronuclear single quantum
76 coherence spectroscopy (HSQC) to elucidate exact structures when appropriate.
2
1
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
1
4
77
16α-Hydroxyprogesterone. H NMR (800 MHz, DMSO-d6) data for 16α-hydroxyprogesterone were: δ 5.64
4
4
4
4
4
4
4
4
78
79
80
81
82
83
84
85
(s, 1H), 4.73 (d, J = 4.8 Hz, 1H), 4.53 (dt, J = 7.5, 5.3 Hz, 1H), 2.44 (d, J = 6.5 Hz, 1H), 2.43 – 2.38 (m, 1H), 2.41 –
2.36 (m, 1H), 2.25 (dt, J = 13.0, 2.4 Hz, 1H), 2.17 (dt, J = 16.9, 3.8Hz, 1H), 2.10 (s, 2H), 1.97 (ddd, J = 13.4, 5.1, 3.1
Hz, 1H), 1.90 (dt, J = 12.5, 3.4 Hz, 1H), 1.75 (ddt, J = 12.3, 5.4, 2.7 Hz, 1H), 1.65 – 1.58 (m, 2H), 1.54 (dq, J = 13.0,
3.6 Hz, 1H), 1.51 (dd, J = 10.8, 3.4 Hz, 1H), 1.50 – 1.46 (m, 1H), 1.49 – 1.45 (m, 2H), 1.35 (qd, J = 13.1, 4.0 Hz,
1
3
1H), 1.14 (s, 3H), 1.01 – 0.95 (m, 2H), and 0.58 (s, 3H). C NMR (201 MHz) data for 16α-hydroxyprogesterone
were: δ 208.55 (C20), 198.52 (C3), 171.30 (C5), 123.66 (C4), 73.36 (C17), 70.89 (C16) 53.41 (C9), 53.44 (C14),
44.75 (C13), 38.63 (C10), 38.39 (C12), 35.46 (C1), 35.95 (C15), 34.89 (C8), 34.05 (C2), 32.39 (C6), 32.19 (C21),
32.01 (C6), 20.66 (C11), 17.32 (C19), and 14.67 (C18).
1
4
86
16α-Hydroxyandrostenedione. H NMR (800 MHz, DMSO-d6) data for 16α-hydroxyandrostenedione were:
4
4
4
4
4
4
4
4
87
88
89
90
91
92
93
94
δ 5.64 (s, 1H), 4.73 (d, J = 4.8 Hz, 1H), 4.53 (dt, J = 7.5, 5.3 Hz, 1H), 2.44 (d, J = 6.5 Hz, 1H), 2.43 – 2.38 (m, 1H),
2.41 – 2.36 (m, 1H), 2.25 (dt, J = 13.0, 2.4 Hz, 1H), 2.17 (dt, J = 16.9, 3.8 Hz, 1H), 2.10 (s, 2H), 1.97 (ddd, J = 13.4,
5.1, 3.1 Hz, 1H), 1.90 (dt, J = 12.5, 3.4 Hz, 1H), 1.75 (ddt, J = 12.3, 5.4, 2.7 Hz, 1H), 1.65 – 1.58 (m, 2H), 1.54 (dq,
J = 13.0, 3.6 Hz, 1H), 1.51 (dd, J = 10.8, 3.4 Hz, 1H), 1.50 – 1.46 (m, 1H), 1.49 – 1.45 (m, 2H), 1.35 (qd, J = 13.1,
13
4.0 Hz, 1H), 1.14 (s, 3H), 1.01 – 0.95 (m, 2H), and 0.58 (s, 3H). C NMR (201 MHz) data for 16α-
hydroxyandrostenedione were: δ 208.55 (C20), 198.52 (C3), 171.30 (C5), 123.66 (C4), 73.36 (C17), 70.89 (C16),
53.41 (C9), 53.44 (C14), 44.75 (C13), 38.63 (C10), 38.39 (C12), 35.46 (C1), 35.95 (C15), 34.89 (C8), 34.05 (C2),
32.39 (C6), 32.19 (C21), 32.01 (C6), 20.66 (C11), 17.32 (C19), and 14.67 (C18).
1
4
95
16α-Hydroxytestosterone. H NMR (900 MHz, DMSO-d6) data for 16α-hydroxytestosterone were: δ 5.63 (d,
4
4
4
4
5
5
5
96
97
98
99
00
01
02
J = 1.8 Hz, 1H), 4.67 (d, J = 4.8 Hz, 1H), 4.61 (d, J = 5.1 Hz, 1H), 3.82 (dt, J = 9.2, 3.8 Hz, 1H), 3.21 (t, J = 4.8 Hz,
1H), 2.39 (ddd, J = 16.5, 14.8, 5.1 Hz, 1H), 2.38 (tdd, J = 14.3, 5.4, 2.0 Hz, 2H), 2.24 (ddd, J = 14.6, 4.2, 2.4 Hz,
1H), 2.16 (dt, J = 16.7, 3.8 Hz, 1H), 1.96 (ddd, J = 13.3, 5.1, 3.1 Hz, 1H), 1.74 (ddt, J = 12.0, 5.7, 2.8 Hz, 1H), 1.70
(dt, J = 12.6, 3.4 Hz, 1H), 1.64 (dt, J = 13.0, 9.3 Hz, 1H), 1.60 (dt, J = 14.1, 4.4 Hz, 1H), 1.53 – 1.47 (m, 2H), 1.35
(ddd, J = 13.2, 7.5, 2.1 Hz, 1H), 1.34 (qd, J = 13.1, 4.2 Hz, 1H), 1.20 (ddd, J = 12.5, 10.9, 7.5 Hz, 1H), 1.14 (s, 3H),
13
1.04 (dt, J = 12.9, 4.2 Hz, 1H), 0.96 – 0.89 (m, 1H), 0.93 – 0.88 (m, 1H), and 0.68 (s, 3H). C NMR (226 MHz)
data for 16α-hydroxytestosterone were: δ 198.58 (C3), 171.59 (C5), 123.60 (C4), 88.97 (C17), 77.04 (C16), 53.83
2
2
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
5
5
03
04
(C9), 48.22 (C14), 43.36 (C13), 38.72 (C10), 36.82 (C12), 35.49 (C1), 34.73 (C15), 34.07 (C2), 32.44 (C6), 31.76
(C7), 20.31 (C11), 17.37 (C19), and 12.82 (C18).
1
5
05
(11β)-11,21,21-Trihydroxypregn-4-ene-3,20-dione. H NMR (900 MHz, DMSO-d6) data for (11β)-
5
5
5
5
5
5
5
5
5
5
06
07
08
09
10
11
12
13
14
15
11,21,21-trihydroxypregn-4-ene-3,20-dione were: δ 5.57 (d, J = 1.8 Hz, 1H), 4.89 (s, 1H), 4.19 (q, J = 3.3 Hz, 1H),
2.86 (t, J = 9.2 Hz, 1H), 2.45 (tdd, J = 14.0, 5.6, 1.9 Hz, 1H), 2.39 (ddd, J = 16.4, 13.7, 5.0 Hz, 1H), 2.22 – 2.16 (m,
2H), 2.13 – 2.07 (m, 1H), 2.08 (dd, J = 13.5, 2.5 Hz, 1H), 2.03 (dddd, J = 13.8, 11.7, 9.0, 2.9 Hz, 1H), 1.92 (ddt, J =
17.2, 7.0, 3.6 Hz, 1H), 1.88 (ddd, J = 22.0, 10.9, 3.9 Hz, 1H), 1.82 – 1.76 (m, 1H), 1.68 (dddd, J = 12.2, 9.8, 7.1, 2.9
Hz, 1H), 1.60 (dddd, J = 12.4, 9.3, 6.7, 3.7 Hz, 1H), 1.48 (dd, J = 13.6, 3.4 Hz, 1H), 1.37 (s, 3H), 1.31 – 1.23 (m,
13
1H), 1.12 (ddd, J = 12.6, 10.5, 7.1 Hz, 1H), 1.03 – 0.96 (m, 1H), 0.95 – 0.92 (m, 1H), and 0.79 (s, 3H). C NMR
(226 MHz) data for (11β)-11,21,21-trihydroxypregn-4-ene-3,20-dione were: δ 208.70 (C20), 198.58 (C3), 172.83
(C5), 121.94 (C4), 90.68 (C21), 66.70 (C11), 57.48 (C14), 57.18 (C17), 55.86 (C9), 47.34 (C12), 44.15 (C13), 39.31
(C10), 34.46 (C1), 33.93 (C2), 32.96 (C7), 31.81 (C6), 31.61 (C8), 24.67 (C15), 23.30 (C16), 20.85 (C19), and
16.31 (C18).
1
5
16
11-keto,16α-Hydroxyprogesterone. H NMR (900 MHz, DMSO-d6) δ 5.65 (d, J = 1.8 Hz, 1H), 4.97 (d, J
5
5
5
5
5
5
5
5
17
18
19
20
21
22
23
24
= 4.8 Hz, 1H), 4.58 – 4.54 (m, 1H), 2.75 (d, J =12.2 Hz, 1H), 2.67 (d, J = 6.7 Hz, 1H), 2.56 (ddd, J = 13.6, 5.1, 3.3
Hz, 1H), 2.46 – 2.40 (m, 2H), 2.45 – 2.38 (m, 2H),2.35 (d, J = 12.1 Hz, 1H), 2.27 (ddd, J = 14.7, 4.1, 2.3 Hz, 1H),
2.22 (d, J = 11.2 Hz, 1H), 2.18 – 2.12 (m, 1H), 2.13(dt, J = 16.8, 3.7 Hz, 1H), 2.10 (s, 3H), 1.88 – 1.83 (m, 2H), 1.73
(td, J = 13.0, 8.7 Hz, 1H), 1.64 (td, J = 14.2, 4.4 Hz,1H), 1.60 (ddd, J = 13.4, 7.7, 1.7 Hz, 1H), 1.32 (s, 3H), 1.21 (qd,
1
3
J = 13.2, 4.3 Hz, 1H), 0.49 (s, 2H). C NMR (226 MHz, DMSO-d6) data for 11-keto,16α-hydroxyprogesterone
were: δ 209.31 (C11), 208.16 (C20), 198.75 (C3), 169.56 (C5), 124.15 (C4), 71.43 (C17), 71.12 (C16), 61.79 (C9),
56.26 (C12), 51.73 (C14), 47.24 (C13), 38.19 (C10), 36.47 (C8), 35.46 (C15), 34.39 (C1), 33.76 (C2), 32.04 (C7),
31.97 (C6), 31.95 (C21), 17.18 (C19), 15.59 (C18).
1
5
25
2α,16α-Dihydroxyprogesterone. H NMR (700 MHz, DMSO-d6) δ 5.63 (d, J = 1.3 Hz, 1H), 5.03 (d, J =4.4
5
5
5
5
26
27
28
29
Hz, 1H), 4.71 (d, J = 4.9 Hz, 1H), 4.52 (dddd, J = 7.9, 6.3, 4.8, 1.0Hz, 1H), 4.14 (ddd, J = 13.7, 5.3, 4.7 Hz, 1H),
3.32 (s, 1H), 2.42 (d, J =6.5 Hz, 1H), 2.35 (td, J = 13.9, 4.5 Hz, 1H), 2.27 (dt, J = 14.3, 3.8 Hz,1H), 2.11 (dd, J =
12.7, 5.5 Hz, 1H), 2.09 (s, 3H), 1.92 – 1.87 (m, 1H),1.74 (ddt, J = 12.8, 5.7, 2.4 Hz, 2H), 1.60 (td, J = 15.1, 8.8 Hz,
1H), 1.52– 1.45 (m, 2H), 1.47 (dd, J = 12.9, 5.1 Hz, 3H), 1.47 – 1.43 (m, 1H), 1.39– 1.31 (m, 1H), 1.21 (s, 3H), 1.01
2
3
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
1
3
5
5
5
5
30
31
32
33
– 0.94 (m, 2H), 0.56 (s, 3H). C NMR (176 MHz, DMSO-d6) data for 2α,16α-dihydroxyprogesterone δ 208.11
(C20), 199.53 (C3), 170.39 (C5), 121.10 (C4), 72.90 (C17), 70.44 (C16), 68.80 (C2), 53.48 (C9), 52.89 (C14), 44.34
(C13), 44.30 (C1), 40.08 (C10), 37.87 (C12), 35.45 (C15), 33.93 (C8), 31.76 (C21), 31.55 (C7) , 31.49 (C6), 20.09
(C11), 17.55 (C19), 14.22 (C18).
1
5
34
6β,16α-Dihydroxyprogesterone. H NMR (700 MHz, DMSO-d6) δ 5.67 (s, 0H), 5.11 (d, J = 2.7 Hz, 0H),
5
5
5
5
5
5
5
5
35
36
37
38
39
40
41
42
4.73 (dd, J = 4.8, 1.4 Hz, 1H), 4.54 (dt, J = 6.8, 3.4 Hz, 1H), 4.16 (dd, J = 5.4, 2.9 Hz, 0H), 2.49 – 2.41 (m, 1H), 2.44
(d, J = 6.5 Hz, 1H), 2.20 (dt, J = 17.8, 3.1 Hz, 1H), 2.09(s, 1H), 1.94 (ddd, J = 12.9, 4.7, 2.6 Hz, 1H), 1.90 (s, 1H),
1.86 – 1.80 (m, 0H), 1.77 (dt, J = 13.6, 2.8 Hz, 0H), 1.65 – 1.57 (m, 1H), 1.57 (d, J = 8.7 Hz, 0H), 1.55– 1.50 (m,
1H), 1.50 – 1.43 (m, 1H), 1.38 (td, J = 12.9, 3.9 Hz, 0H), 1.27 (s, 1H),1.16 – 1.13 (m, 0H), 0.97 – 0.92 (m, 1H), 0.58
13
(s, 1H). C NMR (176 MHz, DMSO-d6) data for 6β,16α-dihydroxyprogesterone δ 208.17 (C20), 199.31 (C3),
168.94 (C5), 125.23 (C4), 73.01 (C17), 71.01 (C6), 70.47 (C16), 52.96 (C14), 52.90 (C9), 44.41 (C13), 38.66 (C7),
37.96 (C12), 37.62 (C10), 36.52 (C1), 35.58 (C15), 33.90 (C2), 31.78 (C21), 28.96 (C8), 20.21 (C11), 18.95 (C19),
14.25 (C18).
543
544
545
546
547
548
549
550
551
552
553
Acknowledgements
This research was supported by a grant (NRF-2016R1D1A3B03933814) of the Basic Science
Research Program through the National Research Foundation of Korea (NRF) funded by the
Ministry of Education, Science and Technology, Republic of Korea. In addition, this work was
carried out with the support of “Cooperative Research Program for Agriculture Science &
Technology Development (Project title: National Agricultural Genome Program, Project No.
PJ01045705)” Rural Development Administration, Republic of Korea. We would like to thank
Division of Magnetic Resonance, Korea Basic Science Institute, Ochang, Chungbuk, Korea for
NMR analyses.
2
4
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
5
5
54
55
Keywords: Cytochrome P450, CYP154C8, electron transport partner, steroid hydroxylation,
whole-cell biotransformation.
5
5
56
57
References
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
[1]
[2]
[3]
[4]
[5]
[6]
D. R. Nelson, Hum. Genomics 2009, 4, 59–65.
L. M. Podust, D. H. Sherman, Nat. Prod. Rep. 2012, 29, 1251–1266.
R. Bernhardt, V. B. Urlacher, Appl. Microbiol. Biotechnol. 2014, 98, 6185–6203.
S. B. Mahato, A. Mukherjee, J. Steroid Biochem. 1984, 21, 341–342.
K. E. Smith, S. Latif, D. N. Kirk, J. Steroid Biochem. 1990, 35, 115–120.
H. Agematsu, N. Matsumoto, Y. Fujii, H. Kabumoto, S. Doi, K. Machida, J. Ishikawa, A.
Arisawa, Biosci. Biotechnol. Biochem. 2006, 70, 307–311.
[7]
[8]
T. Sakaki, Biol. Pharm. Bull. 2012, 35, 844–849.
I. K. Jóźwik, F. M. Kiss, Ł. Gricman, A. Abdulmughni, E. Brill, J. Zapp, J. Pleiss, R.
Bernhardt, A. W. Thunnissen, FEBS J. 2016, 283, 4128–4148.
V. Luu-The, J. Steroid Biochem. Mol. Biol. 2013, 137, 176–182.
[9]
[10] W. L. Miller, Mol. Cell Endocrinol. 2013, 371, 5–14.
[11] M. Bureik, R. Bernhardt, Wiley-VCH Verlag GmbH & Co. KGaA. 2007, Weinheim,
Germany.
[12] R. Bernhardt, J. Biotechnol. 2006, 124, 128–145.
[13] K. T. Nguyen, C. Virus, N. Günnewich, F. Hannemann, R. Bernhardt, ChemBioChem
2012, 13, 1161–1166.
[14] G. Y. Lee, D. H. Kim, D. Kim, T. Ahn, C. H. Yun, Arch. Pharm. Res. 2015, 38, 98–107.
2
5
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
76
77
78
79
80
[15] V. B. Urlacher, S. Lutz-Wahl, R. D. Schmid, Appl. Microbiol. Biotechnol. 2004, 64, 317–
325.
[16] P. Bracco, D. B. Janssen, A. Schallmey, Microb. Cell Fact. 2013, 12, 95.
[17] F. M. Kiss, D. Schmitz, J. Zapp, T. K. Dier, D. A. Volmer, R. Bernhardt, Appl. Microbiol.
Biotechnol. 2015, 99, 8495–8514.
81 [18] T. Makino, Y. Katsuyama, T. Otomatsu, N. Misawa, Y. Ohnishi, Appl. Environ.
82
83
84
85
86
87
88
89
90
91
92
Microbiol. 2014, 80, 1371–1379.
[19] Y. Khatri, Y. Carius, M. Ringle, C. R. Lancaster, R. Bernhardt, FEBS Lett. 2016, 590,
4638–4648.
[20] Y. Khatri, M. Ringle, M. Lisurek, J. P. von Kries, J. Zapp, R. Bernhardt, ChemBioChem
2016, 17, 90–101.
[21] S. G. Salamanca-Pinzon, Y. Khatri, Y. Carius, L. Keller, R. Müller, C. R. Lancaster, R.
Bernhardt, FEBS Lett. 2016, 590, 1838–1851.
[22] L. M. Podust, Y. Kim, M. Arase, B. A. Neely, B. J. Beck, H. Bach, D. H. Sherman, D. C.
Lamb, S. L. Kelly, M. R. Waterman, J. Biol. Chem. 2003, 278, 12214–12221.
[23] L. M. Podust, H. Bach, Y. Kim, D. C. Lamb, M. Arase, D. H. Sherman, S. L. Kelly, M. R.
Waterman, Protein Sci. 2004, 13, 255–268.
93 [24] Q. Cheng, D. C. Lamb, S. L. Kelly, L. Lei, F. P. Guengerich, J. Am. Chem. Soc. 2010,
94
95
132, 15173–15175.
[25] B. C. Von, P. Le-Huu, V. B. Urlacher, ChemBioChem 2013, 4, 2189–2198.
96 [26] A. Schallmey, G. den Besten, I. G. Teune, R. F. Kembaren, D. B. Janssen, Appl.
97
98
Microbiol. Biotechnol. 2011, 9, 1475–1485.
[27] A. Rühlmann, G. Groth, V. B. Urlacher, ChemBioChem 2018, 19, 1–9.
2
6
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
5
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
99
00
01
02
03
04
05
06
[28] Z. Su, J. H. Horner, M. Newcomb, Chemistry 2012, doi:10.1002/chem.201202254.
[29] J. Schenkman, S. Sligar, D. Cinti, Parmacol. Ther. 1981, 12, 43–71.
[30] T. T. B. Ly, Y. Khatri, J. Zapp, M. C. Hutter, R. Bernhardt, Appl. Microbiol. Biotechnol.
2012, 95, 123–133.
[31] K. Herzog, P. Bracco, A. Onoda, T. Hayashi, K. Hoffmann, A. Schallmey, Acta
Crystallogr. D Biol. Crystallography. 2014, 70, 2875–2889.
[32] J. R. Berrie, R. A. Williams, K. E. Smith, J. Steroid Biochem. Mol. Biol. 1999, 71, 153–
165.
07 [33] B. C. Chung, J. Picado-Leonard, M. Haniu, M. Bienkowski, P. F. Hall, J. E. Shively, W.
08
09
L. Miller, Proc. Natl. Acad. Sci. 1987, 84, 407–411.
[34] F. K. Yoshimoto, R. Auchus, J. Steroid Biochem. Mol. Biol. 2015, 151, 52–65.
10 [35] P. Swart, A. C. Swart, M. R. Waterman, R. W. Estabrook, J. I. Mason, J. Clin.
11
12
13
14
15
16
17
18
19
20
21
Endocrinol. Metab. 1993, 77, 98–102.
[36] F. K. Yoshimoto, E. Gonzalez, R. J. Auchus, F. P. Guengerich, J. Biol. Chem. 2016, 291,
17143–17164.
[37] R. Davydov, N. Strushkevich, D. Smil, A. Yantsevich, A. Gilep, S. Usanov, B. M.
Hoffman, Biochemistry 2015, 54, 7089–7097.
[38] R. C. Tuckey, Placenta. 2005, 26, 273–281.
[39] R. C. Tuckey, W. Li, J. K. Zjawiony, M. A. Zmijewski, M. N. Nguyen, T. Sweatman, A.
Slominski, FEBS J. 2008, 275, 2585–2596.
[40] M. J. Cryle, J. J. De Voss, Chem. Commun (Camb). 2004, 1, 86–87.
[41] Y. Anzai, S. Li, M. R. Chaulagain, K. Kinoshita, F. Kato, J. Montgomery, D. H. Sherman,
Chem. Biol. 2008, 15, 950-959.
2
7
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
22
23
24
25
26
27
28
29
30
31
32
33
[42] W. Zhang, Y. Liu, J. Yan, S. Cao, F. Bai, Y. Yang, S. Huang, L. Yao, Y. Anzai, F. Kato,
L. M. Podust, D. H. Sherman, S. Li, J. Am. Chem. Soc. 2014, 136, 3640–3646.
[43] U. Raju, H. L. Bradlow, M. Levitz, Ann. N. Y. Acad. Sci. 1990, 586, 83–87.
[44] J. Marcos, W. Y. Craig, G. E. Palomaki, E. M. Kloza, J. E. Haddow, M. Roberson, L. A.
Bradley, C. H. Shackleton, Prenat. Diagn. 2009, 29, 771–780.
[45] A. Yamada, M. Yamada, Y. Fujita, T. Nishigami, K. Nakasho, K. Uematsu,
J.
Biol.
Chem. 2001, 276, 4604–4610.
[46] F. Zhang, J. Zhou, Y. Shi, P. Tavlarakis, K. Karaisz, J. Pharm. Biomed. Anal. 2016, 128,
333–341.
[47] M. Li, B. Chen, S. Monteiro, A. M. Rustum, Tetrahedron Lett. 2009, 50, 4575–4581.
[48] M. Li, B. Chen, M. Lin, T. M. Chan, X. Fu, A. Rustum, Tetrahedron Lett. 2007, 48,
3901–3905.
34 [49] J. Hansen, H. Bundgaard, Int. J. Pharmaceut. 1980, 6, 307–319.
35 [50] M. Thevis, W. Schänzer, J. Am. Soc. Mass Spectrom. 2005, 16, 1660–1669.
36 [51] H. Li, L. P. Sheng, B. Wang, Z. L. Yang, S. Y. Liu, J. Chromatogr. Sci. 2015, 53, 285–
37
294.
38 [52] C. M. Jenkins, M. R. Waterman, Biochemistry 1998, 37, 6106–6113.
39 [53] C. M. Jenkins, M. R. Waterman, J. Biol. Chem. 1994, 269, 27401–27408.
40
41
42
43
[54] D. B. Hawkes, G. W. Adams, A. L. Burlingame, P. R. Ortiz de Montellano, J. J. De Voss,
J. Biol. Chem. 2002, 277, 27725–27732.
[55] A. Mosa, M. C. Hutter, J. Zapp, R. Bernhardt, F. Hannemann, ChemBioChem 2015, 16,
1670–1679.
2
8
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800018
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
44
45
46
47
48
49
50
[56] W. V. Shaw, L. C. Packman, B. D. Burleigh, A. Dell, H. R. Morris, B. S. Hartley, Nature
1979, 282, 870–872.
[57] N. Saitou, M. Nei, Mol. Biol. Evol. 1987, 4, 406–425.
[58] J. Felsenstein, Evolution 1985, 39, 783–791.
[59] E. Zuckerkandl, L. Pauling, 1965, pp. 97-166. Academic Press, New York.
[60] K. Tamura, G. Stecher, D. Peterson, A. Filipski, S. Kumar, Mol. Biol. Evol. 2013, 30,
2725–2729.
51 [61] J. A. Peterson, M. C. Lorence, B. Amarneh. J. Biol. Chem. 1990, 265, 6066–6073.
52
53
54
55
56
57
[62] S. Bhattarai, K. Liou, T. J. Oh, Arch. Biochem. Biophys. 2013, 539, 63–69.
[63] N. P. Niraula, S. Bhattarai, N. R. Lee, J. K. Sohng, T. J. Oh, J. Microbiol. Biotechnol.
2012, 22, 1059–1065.
[64] T. Omura, R. Sato, J. Biol. Chem. 1964, 239, 2370–2378.
[65] M. M. Purdy, L. S. Koo, P. R. Ortiz de Montellano, J. P. Klinman, Biochemistry 2004, 43,
271–281.
58 [66] P. W. Jr Roome, J. C. Philley, J. A. Peterson, J. Biol. Chem. 1983, 258, 2593–2598.
59 [67] J. W. Williams, J. F. Morrison, Methods Enzymol. 1979, 63, 437–467.
60 [68] J. B. Johnston, H. Ouellet, P. R. Ortiz de Montellano, J. Biol. Chem. 2010, 285, 36352–
6
6
61
36360.
6
62
2
9
This article is protected by copyright. All rights reserved.