Discrimination of the prochiral hydrogens at the C-2 position of n-alkanes by the methane/ammonia monooxygenase family proteins

Article abstract of DOI:10.1039/c5ob00640f

The selectivity of ammonia monooxygenase from Nitrosomonas europaea (AMO-Ne) for the oxidation of C₄-C₈n-alkanes to the corresponding alcohol isomers was examined to show the ability of AMO-Ne to recognize the n-alkane orientation within the catalytic site. AMO-Ne in whole cells produces 1- and 2-alcohols from C₄-C₈n-alkanes, and the regioselectivity is dependent on the length of the carbon chain. 2-Alcohols produced from C₄-C₇n-alkanes were predominantly either the R- or S-enantiomers, while 2-octanol produced from n-octane was racemic. These results indicate that AMO-Ne can discriminate between the prochiral hydrogens at the C-2 position, with the degree of discrimination varying according to the n-alkane. Compared to the particulate methane monooxygenase (pMMO) of Methylococcus capsulatus (Bath) and that of Methylosinus trichosporium OB3b, AMO-Ne showed a distinct ability to discriminate between the orientation of n-butane and n-pentane in the catalytic site.

Full text of DOI:10.1039/c5ob00640f

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Discrimination of the prochiral hydrogens at the
C-2 position of n-alkanes by the methane/
ammonia monooxygenase family proteins†
Cite this: Org. Biomol. Chem., 2015,
3, 8261
1
Akimitsu Miyaji, Teppei Miyoshi, Ken Motokura and Toshihide Baba*
The selectivity of ammonia monooxygenase from Nitrosomonas europaea (AMO-Ne) for the oxidation of
C
4
–C
recognize the n-alkane orientation within the catalytic site. AMO-Ne in whole cells produces 1- and
-alcohols from C –C n-alkanes, and the regioselectivity is dependent on the length of the carbon
chain. 2-Alcohols produced from C –C n-alkanes were predominantly either the R- or S-enantiomers,
8
n-alkanes to the corresponding alcohol isomers was examined to show the ability of AMO-Ne to
2
4
8
4
7
while 2-octanol produced from n-octane was racemic. These results indicate that AMO-Ne can discrimi-
nate between the prochiral hydrogens at the C-2 position, with the degree of discrimination varying
according to the n-alkane. Compared to the particulate methane monooxygenase (pMMO) of Methylo-
coccus capsulatus (Bath) and that of Methylosinus trichosporium OB3b, AMO-Ne showed a distinct ability
to discriminate between the orientation of n-butane and n-pentane in the catalytic site.
Received 1st April 2015,
Accepted 25th June 2015
DOI: 10.1039/c5ob00640f
www.rsc.org/obc
able to discriminate the pro-R hydrogen at C-2 of n-butane
and n-pentane. In addition, the two pMMOs have the
The selective oxidation of n-alkanes to alcohols is accom- same substrate range and product selectivities, suggesting
1
. Introduction
3
,4
plished by several monooxygenases under ambient con- that the two pMMOs have common protein structures around
1
ditions. For instance, methane monooxygenase produces the catalytic site, which involve the recognition of n-butane
methanol selectively from methane; the most inert n-alkane. and n-pentane. However, the amino acid sequences and
A further challenge in the oxidation of n-alkanes is the the crystal structures of pMMO-Mc and pMMO-Mt show
discrimination of a specific C–H bond within the n-alkane some structural differences between the active sites,
5
molecule to selectively produce a specific regioisomer and although the shapes of overall protein are almost identical.
enantiomer of the corresponding alcohol. For an enzyme to Therefore, the amino acid residues in the active site controll-
discriminate between multiple C–H bonds in a n-alkane mole- ing the orientation of n-butane and n-pentane are possibly
cule, its active site must recognize the orientation of a conserved in the two pMMOs, and differences in the
2
n-alkane molecule in the catalytic site. Specific interactions, local structure of the active site do not affect the orientation of
hydrogen bonds and van der Waals interactions between a n-butane and n-pentane.
n-alkane and the surface of the active site play a key role in its
The location of the pMMO catalytic site is a controversial
recognition. Thus, some or all amino acid residues composing issue. Based on the crystal structure of pMMO, three metal
the active site of enzymes should affect the regio- and enantio- sites are proposed as the catalytic site of pMMO: di-copper
6
7
8
selectivity of the alcohol isomers produced from n-alkanes.
center, mono-copper center, and tri-copper center. A puta-
The active site of the copper-containing particulate tive substrate binding cavity is adjacent to each copper
3
b,9
methane monooxygenase of Methylococcus capsulatus (Bath) center.
pMMO-Mc) and Methylosinus trichosporium OB3b (pMMO-Mt) catalytic site because the recombinant protein of pMMO sub-
can recognize n-butane and n-pentane. These pMMOs produce domain containing only the di-copper center shows activity
However, the di-copper center is most likely the
(
3
6
(
R)-2-alcohols selectively from n-butane and n-pentane,
toward methane oxidation to methanol. In addition, the
indicating that the active site of these monooxygenases is chemistry of various model copper complexes and DFT calcu-
lations of the di-copper center of the pMMO active site led to
II
III
the proposal that a mixed-valent (µ-oxo) (µ-hydroxo) Cu Cu
Department of Environmental Chemistry and Engineering, Tokyo Institute of
Technology, 4259 G-1-14 Nagatsuta-cho, Midori-ku, Yokohama 226-8502, Japan.
E-mail: tbaba@chemenv.titech.ac.jp; Fax: +81-45-924-5441; Tel: +81-45-924-5480
species catalyzes the homolytic cleavage of the C–H bond in
10
methane. Furthermore, the C–H bond cleavage and C–O
bond formation proceed with retention of configuration for
†
Electronic supplementary information (ESI) available. See DOI: 10.1039/
1
1
c5ob00640f
n-alkanes.
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem., 2015, 13, 8261–8270 | 8261

View Article Online
Paper
Organic & Biomolecular Chemistry
For determination of the amino acid residue(s) involved in
substrate recognition, site-directed mutagenesis is the most
effective technique. However, this genetic engineering pro-
2
. Results and discussion
2.1 n-Butane and n-pentane oxidation by AMO-Ne
cedure cannot be applied to the study of pMMO, because the Both 1-butanol and 2-butanol were produced from the oxi-
heterologous gene expression system of pMMO is not estab- dation of n-butane by AMO-Ne in whole cells, as reported in
lished. Alternatively, by comparing the protein structures and ref. 14a. As shown in Fig. 1, butanols were produced at a con-
enzyme functions among the family proteins, the key amino stant rate for about 15 min after the reaction started, and
acid residue(s) involved in the molecular recognition can be the production rate decreased gradually. Butyraldehyde and
predicted.
2-butanone, which can be produced by further oxidation of
In this study, we focused on the ammonia monooxygenase butanols, were not observed in the reaction. We further con-
of Nitrosomonas europaea (AMO-Ne) as a family protein of firmed that no butanols consumption was observed when the
pMMO-Mc and pMMO-Mt. AMO-Ne plays a role in the physio- whole cells of AM-Ne were incubated with butanols. These
1
2
logical oxidation of ammonia to hydroxylamine, which is results indicate that the whole cells of AMO-Ne specifically
different physiological role to pMMO. Crystal structure of show activity toward n-butane oxidation to butanols.
AMO-Ne has not been revealed. However, the high amino acid
In the initial stage of n-butane oxidation, 82% of the
sequence homology between AMO-Ne and the two pMMOs product was 2-butanol; the distribution of butanol regio-
1
3
(
approximately 40% identity and 65% similarity) suggests isomers was almost constant during the reaction, as shown
that these enzymes are evolutionarily related each other, in Fig. 1. These results indicate that AMO-Ne in whole cells
and indicates that the tertiary and quaternary protein struc- produces 2-butanol selectively from n-butane. In other words,
tures of these enzymes have high similarity. In addition, n-butane oxidation proceeds predominantly when the C-2
the amino acid residues composing the metal binding site position of n-butane is located close to the catalytic site of
in the two pMMOs are well conserved in AMO-Ne, which AMO-Ne. Further, we investigated the selectivity for 2-butanol
suggests that AMO-Ne has the same catalytic site as the two enantiomers from n-butane. Of the 2-butanol produced, 77%
pMMOs. In fact, AMO-Ne oxidizes some n-alkanes including was the S-enantiomer, as shown in Table 1; the enantio-
methane as non-growth substrates to the corresponding selectivity was also almost constant during the reaction. Thus,
1
4a
alcohols.
AMO-Ne shows some selectivity for (S)-2-butanol in n-butane
On the other hand, the substrate binding site of AMO-Ne oxidation. The selectivity indicates that the pro-S hydrogen at
may be larger than that of the two pMMOs based on the larger the C-2 position of n-butane is preferably oriented toward the
1
4
substrate range of AMO-Ne. In addition, differences in the catalytic site within AMO-Ne.
structure of the substrate binding site of AMO-Ne compared to
For the n-pentane oxidation by AMO-Ne, both 1-pentanol
the two pMMOs is suggested by the difference in regio- and 2-pentanol were produced (Fig. S1(A) in ESI†) as reported
1
4b
selectivity of AMO-Ne for alcohols from alkanes,
and its in ref. 14a; 3-pentanol and other oxidation products such as
1
5
enantioselectivity for epoxides from alkenes. The difference ketone and aldehyde were not observed. The selective pro-
in the local structure of these enzymes may also affect their
ability to discriminate between the prochiral C–H bonds at the
C-2 position of n-alkanes, which enables the enzyme to selec-
tively generate a specific enantiomer of 2-alcohols. Further,
AMO-Ne may be proper enzyme for predicting the key amino
acid residue(s) involved in the molecular recognition of
n-alkane in pMMO/AMO family proteins through the com-
parison of their protein structures.
We report here the influence of the protein structure of
copper-containing monooxygenases on their ability to recog-
nize n-alkanes. We investigated the selectivity for 2-alcohol
4 8
enantiomers produced by the oxidation of C –C n-alkanes
using the whole cells of AMO-Ne. In addition, we constructed
a homology protein structure model of the substrate binding
cavity in AMO-Ne based on the protein structure of pMMO-Mc.
According to the product selectivity and the spatial structure of
the substrate binding cavity, the orientation of the n-alkanes in
the substrate binding cavity of AMO-Ne was estimated. Further-
more, comparing the n-alkane orientations and the spatial
structure of the substrate binding cavity, we discussed the
amino acid residue(s) affecting the regio- and enantioselecti-
vities of n-alkane oxidation to the corresponding alcohol by
pMMO/AMO family proteins.
Fig. 1 Butanol production in n-butane oxidation by AMO-Ne. ○, ●:
-butanol; □, ■: 1-butanol. Arrows indicate axis of data. The reaction
was carried out at 30 °C in 50 mM phosphate buffer (pH 7.0) containing
.2 mM duroquinol, and 4.5 g-wet cells per L whole cells. The initial
concentration of n-butane (dissolved) was 230 µM.
2
1
8262 | Org. Biomol. Chem., 2015, 13, 8261–8270
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
Table 1 Comparison of the selectivity for 2-alcohol enantiomers in n-butane and n-pentane oxidation among AMO-Ne, pMMO-Mt, and
pMMO-Mc
Distribution of 2-alcohol
enantiomers (%)
a
Amount of 2-alcohol
c
Biocatalyst
(μmol g-wet per cells)
R-Form
S-Form
ee (%)
b
n-Butane oxidation
AMO-Ne
14 ± 2
6.8 ± 1.5
7.9 ± 1.4
23 ± 1
76 ± 1
75 ± 1
77 ± 1
24 ± 1
25 ± 1
54 (S)
52 (R)
50 (R)
pMMO-Mt
pMMO-Mc
n-Pentane oxidation
AMO-Ne
pMMO-Mt
b
3.1 ± 0.9
3.4 ± 1.2
8.1 ± 1.5
38 ± 1
86 ± 1
86 ± 1
62 ± 1
14 ± 1
14 ± 1
24 (S)
72 (R)
72 (R)
pMMO-Mc
a
b
The amount of 2-alcohol produced over a reaction time of 10 min. The reactions were carried out at least three times separately at 30 °C in
5
0 mM phosphate buffer (pH 7.0) containing 1.2 mM duroquinol, 4.5 g-wet cells per L whole cells, and n-butane or n-pentane. Initial
concentration of n-butane and n-pentane dissolved in the reaction mixture was 230 μM and 92 μM, respectively. The data are shown as the
c
average of separate reactions. Defined as |R − S|/(R + S) × 100, where R and S are the amount of each enantiomer. Major enantiomer is shown in
parentheses.
duction of 1- and 2-pentanols indicates that the oxidation The enantioselectivity of AMO-Ne for 2-alcohols from n-butane
reaction is selectively proceeded at C-1 and C-2 position of and n-pentane is clearly distinct from that of pMMO-Mc and
n-pentane. The major product of the n-pentane oxidation was pMMO-Mt. AMO-Ne predominantly produces (S)-2-alcohols
1
-pentanol; the selectivity for 1-pentanol was 74%, and the from n-butane and n-pentane, while the two pMMOs produce
selectivity was constant during the reaction. This regio- (R)-2-alcohols selectively. The inversion of the prochiral hydro-
selectivity indicates that n-pentane oxidation proceeds pre- gen orientation at the C-2 position of n-butane and n-pentane
dominantly when the C-1 position of n-pentane is located is probably due to the structural difference in the substrate
close to the catalytic site of AMO-Ne. The difference in the oxi- binding site between AMO-Ne and the two pMMOs.
dation position for n-pentane versus n-butane suggests that the
2
.2 Comparison of amino acid residues composing putative
binding state of n-pentane in the AMO-Ne active site is
different. However, 62% of the 2-pentanol produced was the
substrate binding cavity
S-enantiomer, which is consistent with the n-butane oxidation. In pMMO, putative cavities where the substrate could bind are
9
Thus, AMO-Ne also has the ability to recognize the pro-S located adjacent to the metal binding centers (Fig. 2). Com-
hydrogen at the C-2 position of n-pentane, although the parison of the amino acid residues composing the surface of
selectivity is lower than that of n-butane.
the cavity of the two pMMOs and AMO-Ne (Fig. 3) shows that
To compare the ability of AMO-Ne to discriminate between three amino acid residues of AMO-Ne (His 141 and Ala 156 of
the prochiral hydrogens at the C-2 position of n-butane and AmoB, and Met 56 of AmoC) adjacent to the di-copper center
n-pentane with those of pMMO-Mc and pMMO-Mt, we sub- are substituted for Trp, Ile, and Leu, respectively, in the two
mitted n-butane and n-pentane for oxidation with the whole pMMOs. On the other hand, all of the amino acid residues
cells of pMMO-Mc and pMMO-Mt. The amount of 2-alcohol composing the cavities adjacent to the mono-copper center
produced in the reactions over 10 min and the distributions and the tri-copper center are completely conserved for the
of enantiomers in the 2-alcohol are summarized in Table 1. three enzymes. Amino acid residues proximal to the substrate
The distributions were almost consistent with results of the binding site tend to affect the ability of substrate binding
3
16
previous works by us and other group.
more strongly than those distal from the site. Therefore, the
The amount of 2-alcohol produced was dependent on the distinct product selectivities for AMO-Ne and the pMMOs
kind of bacterial strains as shown in Table 1. In this study, we suggest that one or more of the amino acid residues compos-
used whole cells for the reactions due to the instability of iso- ing the substrate binding site of AMO-Ne should be different
lated enzymes in vitro. Possibly, morphology of bacterial cells from those of the two pMMOs, and that the amino acid resi-
and composition of bacterial membranes affect the access of dues likely affect the enzyme’s ability to control the orientation
n-butane and n-pentane to the enzymes in cells, which are not of n-butane and n-pentane. Thus, according to the results of
the properties of enzyme itself. Therefore, differences in the the amino acid alignment, the cavity adjacent to the di-copper
amount of 2-alcohol between the bacteria shown in Table 1 center in the three enzymes is most likely the substrate
cannot be considered simply as differences in the ability binding site. The di-copper center was proposed as the cata-
between the enzymes.
lytic site by other group based on methane oxidation activity
On the other hand, the distribution of enantiomers in the of the recombinant pMMO subdomain containing only the
6
reactions can be considered the recognition ability of enzymes. di-copper center.
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem., 2015, 13, 8261–8270 | 8263

View Article Online
Paper
Organic & Biomolecular Chemistry
between AMO-Ne and the two pMMOs are shown by white
letters on a black background.
2.3 Spatial structure of the putative substrate binding cavity
To evaluate the influence of the different amino acid residues
composing the cavity’s surface described in section 2.2 on the
orientation of n-butane and n-pentane in the substrate binding
site, protein structures of AMO-Ne and pMMO are required.
However, the crystal structure of AMO-Ne has not been
revealed. Therefore, we constructed a homology model of the
cavity adjacent to the di-nuclear copper site of AMO-Ne (Fig. 4(B))
based on the pMMO-Mc protein structure (PDB ID: 3RGB)
(
Fig. 4(A)). The model was constructed by replacing the three
amino acid residues (Trp136 and Ile151 of PmoB, and Ile78 of
pMMO-Mc) with His, Ala, and Met, respectively.
In the cavity of pMMO-Mc, there are two pockets around
the bottom indicated as Pckt-1 and Pckt-2 in Fig. 4(A). The
spatial volume of Pckt-2 is larger than that of Pckt-1. The sub-
stitution of Ile for Ala makes the volume of Pckt-1 larger
because of the difference in the van der Waals volumes of Ile
3
17
and Ala (0.165 and 0.090 nm, respectively). Furthermore,
the substitution of Trp for His makes the entrance of the cavity
wider. These structural changes are consistent with the differ-
ence in the substrate range between pMMO-Mc and AMO-Ne;
AMO-Ne can oxidize larger substrates, such as aromatics, than
Fig. 2 Three metal binding sites in the protein structure of pMMO-Mc.
(
1
4
A) Di-nuclear copper site; (B) mono-nuclear copper site; (C) tri-nuclear pMMO-Mc. In conclusion, one or more of the three amino
copper site.
acid residues adjacent to the di-copper site may affect the sub-
strate range and the selectivity for n-butane and n-pentane oxi-
dation to alcohol isomers.
Alternatively, the amino acid residues at locations other
than the surface of the putative cavity in pMMO/AMO affect
catalytic selectivity. Once a method for site-directed muta-
genesis of pMMO or AMO is established, the three amino acid
residues mentioned above will be promising targets for chan-
ging the substrate scope, and the regio- and enantioselectivity
of n-alkane oxidation to alcohols. In addition, construction of
the model of substrate binding site in pMMO/AMO family pro-
teins by the theoretical calculation such as QM/MM method
can expect the regio- and enantioselectivity.
Copper ions are shown as blue balls, amino acid residues
differing between pMMO-Mc and AMO-Ne are shown as red
sticks, and the Connolly surface of the protein is shown in
gray. The right side of the cavity opens up to the outside of the
protein molecule.
2.4 Orientation of n-butane and n-pentane in the putative
Fig. 3 Comparison of the amino acid residues composing the cavities
adjacent to the three metal binding centers in pMMO-Mc, pMMO-Mt, substrate binding site of AMO-Ne and pMMO-Mc
and AMO-Ne.
Based on the protein structure of the putative substrate
binding cavity shown in Fig. 4, the proposed orientation of
n-alkanes relative to the catalytic site is shown in Fig. 5. In the
case of pMMO-Mc, (R)-2-alcohols are produced selectively from
The three different colored ribbons represent the peptides n-butane and n-pentane. In n-alkane oxidation by pMMO-Mc,
composing pMMO-Mc: PmoA, PmoB, and PmoC are blue, proton abstraction and hydroxyl group insertion occurs with
1
1
purple, and pale purple, respectively.
concerted mechanism. According to this mechanism, the
The amino acid residues composing the protein cavity are enantioselectivity indicates that the pro-R hydrogen at the C-2
indicated as bold letters, and the residues that are different position of the n-alkanes is predominantly located close to the
8264 | Org. Biomol. Chem., 2015, 13, 8261–8270
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 4 Protein surface structure of the cavity adjacent to the di-nuclear copper center of pMMO-Mc and AMO-Ne. (A) Protein structure of
pMMO-Mc (PDB: 3RGB). (B) Homology model structure of AMO-Ne.
Fig. 5 Proposed orientation of n-butane and n-pentane relative to the di-nuclear copper center in the putative substrate binding cavity in
S R
pMMO-Mc (A) and AMO-Ne (B). R: Ethyl or n-propyl group; Me: methyl group; H and H : pro-S and pro-R hydrogens, respectively.
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem., 2015, 13, 8261–8270 | 8265

View Article Online
Paper
Organic & Biomolecular Chemistry
catalytic site in the substrate binding site of pMMO-Mc. This These volumes may be optimum for packing the alkyl chains
orientation can be explained by the model shown in Fig. 5(A)- within Pckt-1; fitting the inner surface of Pckt-1 to the alkyl
①
, where the pro-S hydrogen is directed toward Pckt-1, the chains makes the conformers shown in Fig. 5(B)-② and -③
methyl group is toward Pckt-2, and the alkyl group, such as stable and favorable.
ethyl and n-propyl, is toward the entrance of the cavity. Alterna-
In the oxidation of n-butane and n-pentane to alcohols with
tively, if the pro-S hydrogen is oriented toward the catalytic AMO-Ne, the selectivity for 1-alcohols was higher than that of
site, then the methyl group is directed toward Pckt-2 and the pMMO-Mc. This result indicates that the C-1 position of the
alkyl group is directed toward Pckt-1 as shown in Fig. 5(A)-②, n-alkane molecule is preferably positioned close to the cata-
and/or the alkyl group is directed toward Pckt-2 and the lytic site of AMO-Ne unlike that of pMMO-Mc. The larger space
methyl group is directed toward Pckt-1 as shown in Fig. 5(A)- around the entrance of the cavity in AMO-Ne than that in
③
. However, (S)-2-pentanol was a minor product in n-pentane pMMO-Mc may cause the difference in the n-butane and
oxidation to 2-pentanol. Probably, the volume of Pckt-1 is not n-pentane orientation within the substrate binding site. There-
large enough to accommodate methyl, ethyl and n-propyl fore, n-butane and n-pentane may bind around the entrance
3
groups, whose molecular volumes are about 0.027 nm , prior to diffusing into the cavity, resulting in orientation of the
3
3
18
0
.044 nm , and 0.061 nm , respectively, and Pckt-2 is not C-1 position of the n-alkanes close to the catalytic site.
large enough for ethyl and n-propyl groups. Thus, these orien-
tations are minor conformers for the binding of n-butane and
n-pentane at the catalytic site in pMMO-Mc.
2
.5
6 8
C –C n-alkane oxidation by AMO-Ne
Previous reports have indicated that AMO-Ne can oxidize
As shown in Fig. 4, the protein structure of the putative sub- larger substrates in molecular size than n-pentane
3
13
strate binding cavity in AMO-Ne is locally different from that (0.096 nm ). Thus, we submitted C –C n-alkanes to the oxi-
6
9
in pMMO-Mc. The change from Ile to Ala in Pckt-1 increases dation conditions using AMO-Ne whole cells to investigate the
3
the volume of Pckt-1 by about 0.075 nm , which is estimated dependence of the recognition ability for the pro-R hydrogen
to be the change in the volume of the amino acid residue sub- at the C-2 position of n-alkanes on the carbon chain length.
stituted. Therefore, Pckt-1 in AMO-Ne is large enough to Alcohol isomers were produced from C –C n-alkanes using
6
8
accommodate methyl, ethyl, and n-propyl groups, as shown in whole cells, but not from n-nonane. This results suggests that
Fig. 5(B)-② and -③. In this case, the pro-S hydrogen can point the cavity size of the substrate binding site in AMO-Ne can
3
toward the di-copper site, which produces (S)-2-alcohols selec- accommodate n-octane (0.147 nm ), but not n-nonane
3
18
tively when the oxidation reaction proceeds at the C-2 position (0.164 nm ).
of n-butane and n-pentane. In fact, the S-enantiomer was the
Oxidation of C –C n-alkanes afforded only 1- and 2-alco-
6
8
major product when the oxidation reaction of n-butane and hols (Fig. S1(B)–(D) in ESI†). As shown in Table 2, the distri-
n-pentane proceeded at C-2 position, as described in section bution of 1- and 2-alcohols is dependent on the n-alkane, as
2
.1. This result indicates that the orientations shown in Fig. 5 reported previously.14a 1-Hexanol was the major product from
(B)-② and -③ are preferable conformers for the binding of n-hexane, and its distribution in the product was 74%.
n-butane and n-pentane at the catalytic site in AMO-Ne than that However, for n-heptane the amount of 1-heptanol and 2-hepta-
in Fig. 5(B)-①. This may be caused by the optimum packing of nol produced was almost the same. This result indicates that
methyl, ethyl, and n-propyl groups in Pckt-1. The molecular the position of the n-alkane molecule relative to the catalytic
volumes of methyl, ethyl and n-propyl groups are calculated to site is strongly influenced by the chain length of the n-alkane.
3
3
3
18
be about 0.027 nm , 0.044 nm and 0.061 nm , respectively.
In the case of n-alkane oxidation with manganese porphyrin,
a
Table 2 n-Alkane oxidation by AMO-Ne
Distribution
of 2-alcohol
enantiomers (%)
Distribution of alcohol
regioisomers (%)
c
Amount of alcohols
b
d
Substrate
(μmol g-wet per cells)
1-Alcohol
2-Alcohol
R
S
ee (%)
n-C
n-C
n-C
n-C
n-C
H
4 10
H
5 12
H
6 14
H
7 16
H
8 18
16.5 ± 2.1
12.7 ± 2.2
13.6 ± 2.6
6.4 ± 1.8
1.8 ± 1.2
18 ± 1
74 ± 1
73 ± 1
48 ± 2
97 ± 1
82 ± 1
26 ± 1
27 ± 1
52 ± 2
5 ± 1
23 ± 1
38 ± 1
72 ± 1
83 ± 1
48 ± 1
77 ± 1
62 ± 1
28 ± 1
17 ± 1
52 ± 1
54 (S)
24 (S)
44 (R)
66 (R)
4 (S)
a
The reactions were carried out at least three times separately at 30 °C in 50 mM phosphate buffer (pH 7.0) containing 1.2 mM duroquinol,
b
4
.5 g-wet cells per L whole cells, and n-alkane. The data are shown as the average of three separate reactions. Initial concentration of substrate
c
dissolved in the reaction mixture was: [n-C
4
H
10] = 230 μM; [n-C
5
H
12] = 92 μM; [n-C
6
H
14] = 60 μM; [n-C
7
H
16] = 38 μM; [n-C
8
H
18] = 31 μM. The
d
amount of 2-alcohol produced over a reaction time of 10 min. Defined as |R − S|/(R + S) × 100, where R and S are the amount of each
enantiomer. Major enantiomer is shown in parentheses.
8266 | Org. Biomol. Chem., 2015, 13, 8261–8270
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
the position of oxidation in the n-alkane can be changed by
1
9
introducing bulky groups around the Mn center. Therefore,
it is reasonable to assume that the steric hindrance caused by
the amino acid residues around the catalytic site affect the
position of the n-alkane relative to the catalytic site.
For the 2-alcohols generated from C
enantiomer was predominant; however, essentially racemic
-octanol was produced from n-octane (Table 2). These distri-
4 7
–C n-alkanes, one
2
butions indicate that the ability of AMO-Ne to discriminate
between the prochiral hydrogens at the C-2 position varies
according to the chain length of the n-alkane. The enantio-
selectivity for (R)-2-alcohols increased as the chain length of
the n-alkanes increased from four to seven. This increment is
consistent with the increment in the enantioselectivities of the
two pMMOs for (R)-2-alcohols when the n-alkane chain length
increased from four to five, as previously discussed (see
Table 1). The increment in the enantioselectivity for (R)-2-
alcohol means that the recognition ability for C–H bond orien-
tation at the C-2 position of the n-alkane gradually changed for
increasing molecular size. The cavity volume of the substrate
binding site in AMO-Ne may determine the C–H bond orien-
tation of n-alkanes relative to the catalytic site.
By the proposed n-alkane orientation in AMO-Ne shown in
Fig. 5(B)-②, the increment in the enantioselectivity for (R)-2-
alcohol can be explained. The volume of Pckt-1 (Fig. 5(B)) is
3
approximately 0.075 nm . This volume is enough for accom-
3
modating a propyl group (∼0.061 nm ), which is the respective
sidechain for oxidation of n-pentane at the C-2 position as
shown in Fig. 6(a). However, the volume is not large enough to
3
accommodate a butyl group (∼0.078 nm ), which is the
respective sidechain for oxidation at the C-2 position of
n-hexane. Because of steric hindrance in Pckt-1, the pro-R
hydrogen at the C-2 position should be oriented (Fig. 5(B)-①)
as the major binding mode when the chain length of the
n-alkane is six (Fig. 6(b)) and seven. Therefore, (R)-2-alcohol is
the major enantiomer in n-hexane and n-heptane oxidations.
On the other hand, based on the proposed orientation
shown in Fig. 5(B)-③, the alkyl group (R) is directed toward
the entrance of the substrate binding cavity. The space around
the entrance has enough space for accommodating pentyl
group. According to the selectivity for 2-alcohol enantiomers,
when the carbon number of the n-alkane increases, the orien-
tation of n-alkane shown in Fig. 5(B)-③ should change to that
shown in Fig. 5(B)-①. When the carbon number of the
n-alkane increases from four to seven, n-alkane may not be
able to enter deep enough into the cavity for interact methyl
group with Pckt-1 as shown in Fig. 6(c). In this case, the
methyl group can interact with the surface of Pckt-2 as shown
in Fig. 6(b), resulting that pro-R hydrogen is directed toward
the active site.
Fig. 6 Putative orientation of n-pentane (a) and n-hexane (b)(c) in the
substrate binding cavity of AMO-Ne.
group, inversion of configuration at C-2 position of n-alkane
The hypothesis of n-alkane orientation within AMO-Ne may occur to some extent. Alternatively, pro-R hydrogen is
cavity described above is proposed by assuming that n-alkane abstracted from C-2 position of n-alkane but the hydroxyl
oxidation to alcohol occurs with concerted mechanism pro- group may be inserted from the opposite face radical inter-
posed as the reaction mechanism of pMMO-Mc. Therefore, mediate. Considering these inversions, n-alkane orientations
after the homolytic cleavage between pro-S hydrogen and in the substrate binding cavity are more complicated than
carbon of n-alkane and prior to the insertion of hydroxyl those shown in Fig. 5.
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem., 2015, 13, 8261–8270 | 8267

View Article Online
Paper
Organic & Biomolecular Chemistry
a
Table 3 n-Butane and n-pentane oxidation by AMO in whole cells of N. europaea with various electron donor
Distribution of
2-alcohol
enantiomers (%)
Distribution of alcohol
regioisomers (%)
b
c
d
Substrate
Amount of alcohols (μmol g-wet per cells)
1-Alcohol
2-Alcohol
R
S
ee (%)
n-Butane oxidation
Duroquinol
16.5 ± 2.4
16.9 ± 3.2
13.9 ± 2.2
18 ± 1
20 ± 1
21 ± 1
82 ± 1
80 ± 1
79 ± 1
23 ± 1
23 ± 1
24 ± 1
77 ± 1
77 ± 1
76 ± 1
54 (S)
54 (S)
52 (S)
4
NH Cl
N H
2 4
n-Pentane oxidation
Duroquinol
12.7 ± 3.2
11.7 ± 3.0
8.4 ± 2.1
76 ± 1
76 ± 1
71 ± 1
24 ± 1
24 ± 1
29 ± 1
38 ± 1
39 ± 1
45 ± 1
62 ± 1
61 ± 1
55 ± 1
24 (S)
22 (S)
10 (S)
4
NH Cl
N H
2 4
a
The reactions were carried out at least three times separately at 30 °C in 50 mM phosphate buffer (pH 7.0) containing 1.2 mM duroquinol,
b
4
.5 g-wet cells per L whole cells, and n-alkane. The data are shown as the average of three separate reactions. Initial concentration of substrate
c
dissolved in the reaction mixture was: [n-C
4
H
10] = 230 μM; [n-C
5
H
12] = 92 μM; [n-C
6
H
14] = 60 μM; [n-C
7
H
16] = 38 μM; [n-C
8
H
18] = 31 μM. The
d
amount of 2-alcohol produced over a reaction time of 10 min. Defined as |R − S|/(R + S) × 100, where R and S are the amount of each
enantiomer. Major enantiomer is shown in parentheses.
2
.6 Influence of electron donor on selectivity
n-alkane oxidized, to discriminate between the C-1 and C-2
positions, and to discriminate between the prochiral hydro-
gens at the C-2 position of n-butane and n-pentane. We predict
that at least one of the three amino acid residues at the di-
copper site affects the discriminating ability of the pMMO/
AMO family proteins. Mutagenesis of the amino acid residues
has the potential to invert or increase the enantioselectivity for
the oxidation of n-alkanes to 2-alcohols. In addition, substrate
analogue can also be applied for tuning the regio- and enantio-
selectivity of for pMMO/AMO family proteins.
In the reactions described above, duroquinol was used as the
electron donor for AMO-Ne. On the other hand, ammonium
ions (NH
electron donors for AMO-Ne.
size of the substrates for AMO-Ne, duroquinol is too large to
enter into the substrate binding site of AMO-Ne. In contrast,
NH
respectively, thus they can be accommodated in the substrate
binding site of AMO-Ne. In the case of cytochrome P450 BM-3,
co-occupation of substrate or its analogue alters product distri-
+
4
) and hydrazine (N
2
H
2
) can be used as alternative
According to the molecular
1
4b
+
4
and N
2
H
4
are similar in size to methane and ethylene,
2
0
butions. These electron donors may be able to co-exist with
n-alkane molecules in the substrate binding cavity, thus 4. Experimental section
affecting the orientation of the n-alkane.
4.1 Culture of bacteria
To examine the influence of the electron donors on
Whole cells of N. europaea (ATCC19718) were used for the
n-butane, n-pentane, n-hexane, n-heptane, and n-octane
oxidation reactions. The cells were grown on a 3 L scale in
enantioselectivity, we performed n-butane and n-pentane oxi-
dation by AMO-Ne with NH Cl or N H . As shown in Table 3,
4 2 2
almost no effect by the two electron donors was observed on
the regio- and enantioselectivity for the oxidation of n-butane
2 2
to butanol isomers. In contrast, N H affected the selectivity
2
1
mineral salt medium as described previously. During culture
growth, the pH of the medium was adjusted to 7.8 by the
addition of 5% (w/v) Na CO . The cells were grown to logarith-
of the n-pentane oxidation in favor of (R)-2-pentanol. The
increased selectivity for (R)-2-alcohols was previously observed
for longer chain lengths of n-alkanes (see Table 2). In con-
clusion, N H may co-exist with n-pentane in the substrate
2
3
mic phase, and then harvested by centrifugation (12 000g for
5 min). The cells were washed with 50 mM phosphate buffer
1
containing 2 mM MgCl , pH 7.7, followed by centrifugation at
2
2
2
+
12 000g for 15 min. The collected cells were suspended in
binding cavity and affect its orientation, while NH
be too small to affect the orientation.
4
may
5
0 mM phosphate buffer containing 2 mM MgCl
2
, pH 7.0,
frozen in liquid nitrogen, and stored at −80 °C.
Whole cells of M. capsulatus (Bath) and M. trichosporium
OB3b were also used for n-butane and n-pentane oxidation
reactions. These bacteria were prepared as reported
3
. Conclusions
3
b
previously.
In conclusion, we have shown that the substrate binding site
of AMO-Ne can discriminate between the prochiral hydrogens
at the C-2 position of n-alkanes. The orientation of the
4.2 n-Alkane oxidation to alcohol
n-alkane relative to the catalytic site is dependent on the n-Alkane oxidation reactions were carried out using a 50 mL
n-alkane. Furthermore, AMO-Ne differs from pMMO-Mc and 3-neck flask in a water bath at 30 °C under atmospheric
pMMO-Mt in its ability to discriminate between the size of the pressure. The flask contained 13.5 mL of 50 mM phosphate
8268 | Org. Biomol. Chem., 2015, 13, 8261–8270
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
buffer (pH 7.0) containing 1 mM duroquinol. For the oxidation
of n-butane, the flask was put under vacuum, and then filled
Acknowledgements
with n-butane and air (provided by the balloons). For the oxi- This work was supported by JSPS KAKENHI Grant Number
dation of liquid n-alkanes, the n-alkane was added using a 23760739, 25630363.
glass syringe to the buffer solution in the flask under atmos-
pheric pressure. To initiate the reaction, bacterial cells
(
(
1.5 mL) suspended in 50 mM phosphate buffer (pH 7.0)
ca. 0.5 mg dry cells per mL) were added to the reactor using a
References
glass syringe. Aliquots of the reaction mixture were withdrawn
and immediately placed on ice to quench the reaction. The ali-
quots were centrifuged briefly to pellet the bacterial cells, then
products were extracted from the supernatant using dichloro-
methane and analyzed by gas chromatography. The amount of
product was determined using a TC-WAX column (30 m ×
1 J. A. Labinger, J. Mol. Catal. A: Chem., 2004, 220,
27–35.
2 K. Godula and D. Sames, Science, 2006, 312, 67–72.
3 (a) S. J. Elliot, M. Zhu, L. Tso, H.-H. T. Nguyen, J. H.-K. Yip
and S. I. Chan, J. Am. Chem. Soc., 1997, 119, 9949–9955;
(b) A. Miyaji, T. Miyoshi, K. Motokura and T. Baba, Biotech-
nol. Lett., 2011, 33, 2241–2246.
0
.25 mm i.d., 0.25 µm film thickness, GL Science) and a
flame ionization detector. The injector and detector were
kept at 250 °C. The flow rate of carrier gas (He) was 34 cm
4 K. J. Burrows, A. Cornish, D. Scott and I. J. Higgins, J. Gen.
Microbiol., 1984, 130, 3327–3333.
−
1
min . Temperature programs for separating each alcohol
enantiomers were as follows: 40 °C for 5 min, 10 °C min−
to 230 °C.
5 (a) L. L. Lieberman and A. C. Rosenzweig, Nature, 2005,
434, 177–182; (b) A. S. Hakemian, K. C. Kondapalli,
J. Telser, B. M. Hzoffman, T. L. Stemmler and
A. C. Rosenzweig, Biochemistry, 2008, 47, 6793–6801.
6 (a) R. Balasubramanian, S. M. Smith, S. Rawat,
L. A. Yatsunyk, T. L. Stemmier and A. C. Rosenzweig,
Nature, 2010, 465, 115–119; (b) R. A. Himes, K. Barnese and
K. D. Karlin, Angew. Chem., Int. Ed., 2010, 49, 6714–6716;
(c) S. M. Smith, R. Balasubramanian and A. C. Rosenzweig,
Methods Enzymol., 2011, 495, 195–210; (d) M. A. Culpepper,
G. E. Cutsail 3rd, B. M. Hoffman and A. C. Rosenzweig,
J. Am. Chem. Soc., 2012, 134, 7640–7643.
7 (a) M. Martinho, D. W. Choi, A. A. DiSpirito,
W. E. Antholine, J. D. Semrau and E. Muenck, J. Am. Chem.
Soc., 2007, 125, 15783–15785; (b) E. F. Liew, D. Tong,
N. V. Coleman and A. J. Holmes, Microbiology, 2014, 160,
1267–1277.
1
The 2-alcohol enantiomers were identified using a method
2
2
previously reported. The dichloromethane solutions of the
-alcohol products previously extracted (400 µL) were dried
over MgSO and derivatized with trifluoroacetate (22 µL) to the
2
4
corresponding esters. Enantioselectivities were determined by
using a Chiraldex G-TA column (20 m × 0.25 mm i.d., 0.12 µm
film thickness, Sigma-Aldrich) and
a flame ionization
detector. The injector and detector were kept at 250 °C. The
−
1
flow rate of carrier gas (He) was 34 cm min . Temperature
programs for separating each alcohol enantiomers were
as follows: 2-butanol and 2-pentanol: 30 °C for 2 min, 2 °C
−
1
min to 170 °C; 2-hexanol and 2-heptanol: 30 °C for 2 min,
1
5
°C min− to 170 °C; 2-octanol: constant at 60 °C.
The representative chromatograms of the enantiomers with
their retention time in the reaction mixtures were shown in
ESI (Fig. S2†). Retention times of each enantiomers are shown
in the chromatograms. The tetention times were identified by
using authentic standard prepared by the reaction with tri-
8 S. I. Chan and S. S. Yu, Acc. Chem. Res., 2008, 41, 969–
979.
9 R. Balasubramanian and A. C. Rosenzweig, Acc. Chem. Res.,
2007, 40, 573–580.
fluoroacetate with commercially available racemic 2-alcohols 10 (a) Y. Shiota and K. Yoshizawa, Inorg. Chem., 2009, 48, 838–
and either enantiomer of 2-alcohols. Racemic 2-alcohols
were purchased from Kanto Chemical Co. (S)-2-butanol,
845; (b) K. Yoshizawa, in Copper-oxygen chemistry, ed.
K. D. Karlin and S. Itoh, John Wiley & Sons, Inc., Hoboken,
NJ, USA, 2011, p. 197; (c) R. A. Himes and K. D. Karlin,
Curr. Opin. Chem. Biol., 2009, 13, 119–131.
(
R)-2-hexanol, and (R)-2-octanol were obtained from Tokyo
Chemical Industry Co., (R)-2-hexanol was from Wako
Pure Chemical Industries, and (S)-2-heptanol was from Sig- 11 (a) B. Wilkinson, M. Zhu, N. D. Priestley, H.-H. T. Nguyen,
ma-Aldrich. Based on the area of each peaks due to
the 2-alcohol enantiomers, distribution of enantiomers was
calculated.
H. Morimoto, P. G. Williams, S. I. Chan and H. G. Floss,
J. Am. Chem. Soc., 1996, 118, 921–922; (b) S. S.-F. Yu,
L.-Y. Wu, K. H.-C. Chen, W.-I. Luo, D.-S. Huang and
S. I. Chan, J. Biol. Chem., 2003, 278, 40658–40669.
1
1
2 C. Bédard and R. Knowles, Microbiol. Rev., 1989, 53, 68–84.
3 A. J. Holmes, A. Costello, M. E. Lidstrom and J. C. Murrell,
FEMS Microbiol. Lett., 1995, 132, 203–208.
4
.3 Alignment of amino acid sequences
The amino acid sequences of the enzymes were obtained from 14 (a) M. R. Hyman, I. B. Murton and D. J. Arp, Appl. Environ.
the NCBI protein database, and multiple alignments of these
Microbiol., 1988, 54, 3187–3190; (b) M. R. Hyman and
P. M. Wood, Biochem. J., 1983, 212, 31–37; (c) W. K. Keener
and D. J. Arp, Appl. Environ. Microbiol., 1994, 60, 1914–
1920.
2
3
sequences were performed using ClustalW. A protein struc-
ture model of AMO-Ne was constructed and visualized using
MolFeat ver. 4.6 (FiatLux Co., Tokyo, Japan).
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem., 2015, 13, 8261–8270 | 8269

View Article Online
Paper
Organic & Biomolecular Chemistry
1
1
1
5 C. A. G. M. Weijers, C. G. van Ginkel and J. A. M. de Bont, 21 S. A. Ensign, M. R. Hyman and D. J. Arp, J. Bacteriol., 1993,
Enzyme Microb. Technol., 1988, 10, 214–218. 175, 1971–1980.
6 S. Park, K. Morley, G. P. Horsman, M. Holmquist, 22 (a) J. H. Mommers, S. M. de Wildeman, W. A. Koolen and
K. Hult and R. J. Kazlauskas, Chem. Biol., 2005, 12, 45–54.
7 Y. Harpaz, M. Gerstein and C. Chothia, Structure, 1994, 2,
A. L. Duchateau, J. Chromatogr. A, 2008, 1182, 215–218;
(b) S. Peter, S. M. Kinne, X. Wang, R. Ullrich, G. Kayser,
J. T. Groves and M. Hofrichter, FEBS J., 2011, 278, 3667–
3675.
6
41–649.
1
1
8 A. Bondi, J. Phys. Chem., 1964, 68, 441–451.
9 B. R. Cook, T. J. Reinert and K. S. Suslick, J. Am. Chem. Soc., 23 M. A. Larkin, G. Blackshields, N. P. Brown, R. Chenna,
1
986, 108, 7281–7286.
P. A. McGettigan, H. McWilliam, F. Valentin, I. M. Wallace,
A. Wilm, R. Lopez, J. D. Thompson, T. J. Gibson and
D. G. Higgins, Bioinformatics, 2007, 23, 2947–2948.
2
0 D. A. Rock, B. N. Perkins, J. Wahlstrom and J. P. Jones,
Arch. Biochem. Biophys., 2003, 416, 9–16.
8270 | Org. Biomol. Chem., 2015, 13, 8261–8270
This journal is © The Royal Society of Chemistry 2015