Kinetic analysis of terminal and unactivated C-H bond oxyfunctionalization in fatty acid methyl esters by monooxygenase-based whole-cell biocatalysis

Article abstract of DOI:10.1002/adsc.201100440

The alkane monooxygenase AlkBGT from Pseudomonas putida GPo1 constitutes a versatile enzyme system for the ω-oxyfunctionalization of medium chain-length alkanes. In this study, recombinant Escherichia coli W3110 expressing alkBGT was investigated as whole-cell catalyst for the regioselective biooxidation of fatty acid methyl esters to terminal alcohols. The ω-functionalized products are of general economic interest, serving as building blocks for polymer synthesis. The whole-cell catalysts proved to functionalize fatty acid methyl esters with a medium length alkyl chain specifically at the ω-position. The highest specific hydroxylation activity of 104 U g_CDW^-1 was obtained with nonanoic acid methyl ester as substrate using resting cells of E. coli W3110 (pBT10). In an optimized set-up, maximal 9-hydroxynonanoic acid methyl ester yields of 95% were achieved. For this specific substrate, apparent whole-cell kinetic parameters were determined with a V_max of 204±9 U g_CDW ^-1, a substrate uptake constant (K_S) of 142±17 μM, and a specificity constant V_max/K_S of 1.4 U g _CDW^-1 μM^-1 for the formation of the terminal alcohol. The same E. coli strain carrying additional alk genes showed a different substrate selectivity. A comparison of biocatalysis with whole cells and enriched enzyme preparations showed that both substrate availability and enzyme specificity control the efficiency of the whole-cell bioconversion of the longer and more hydrophobic substrate dodecanoic acid methyl ester. The efficient coupling of redox cofactor oxidation and product formation, as determined in vitro, combined with the high in vivo activities make E. coli W3110 (pBT10) a promising biocatalyst for the preparative synthesis of terminally functionalized fatty acid methyl esters. Copyright

Full text of DOI:10.1002/adsc.201100440

FULL PAPERS
DOI: 10.1002/adsc.201100440
À
Kinetic Analysis of Terminal and Unactivated C H Bond
Oxyfunctionalization in Fatty Acid Methyl Esters by
Monooxygenase-Based Whole-Cell Biocatalysis
Manfred Schrewe,a Anders O. Magnusson,a,b Christian Willrodt,a Bruno Bꢀhler,a,*
and Andreas Schmida
a
Laboratory of Chemical Biotechnology, Department of Biochemical and Chemical Engineering, TU Dortmund
University, Emil-Figge-Str. 66, 44227 Dortmund, Germany
Fax : (+49)-231-755-7382; phone: (+49)-231-755-7384; e-mail: bruno.buehler@bci.tu-dortmund.de
Current address: DECHEMA e.V., Bioverfahrenstechnik, Theodor-Heuss-Allee 25, 60486 Frankfurt/Main, Germany
b
Received: May 27, 2011; Published online: November 29, 2011
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/adcs.201100440.
Abstract: The alkane monooxygenase AlkBGT from stant (KS) of 142Æ17 mM, and a specificity constant
Pseudomonas putida GPo1 constitutes a versatile Vmax/KS of 1.4 UgCDWÀ1 mmÀ1 for the formation of the
enzyme system for the w-oxyfunctionalization of terminal alcohol. The same E. coli strain carrying ad-
medium chain-length alkanes. In this study, recombi- ditional alk genes showed a different substrate selec-
nant Escherichia coli W3110 expressing alkBGT was tivity. A comparison of biocatalysis with whole cells
investigated as whole-cell catalyst for the regioselec- and enriched enzyme preparations showed that both
tive biooxidation of fatty acid methyl esters to termi- substrate availability and enzyme specificity control
nal alcohols. The w-functionalized products are of the efficiency of the whole-cell bioconversion of the
general economic interest, serving as building blocks longer and more hydrophobic substrate dodecanoic
for polymer synthesis. The whole-cell catalysts acid methyl ester. The efficient coupling of redox co-
proved to functionalize fatty acid methyl esters with factor oxidation and product formation, as deter-
a medium length alkyl chain specifically at the w-po- mined in vitro, combined with the high in vivo activi-
sition. TheÀh1 ighest specific hydroxylation activity of ties make E. coli W3110 (pBT10) a promising biocat-
104 UgCDW
was obtained with nonanoic acid alyst for the preparative synthesis of terminally func-
methyl ester as substrate using resting cells of E. coli tionalized fatty acid methyl esters.
W3110 (pBT10). In an optimized set-up, maximal 9-
hydroxynonanoic acid methyl ester yields of 95%
were achieved. For this specific substrate, apparent Keywords: biotransformations; coupling efficiency;
whole-cell kinetic parameters were determined with oxygenation; renewable resources; terminal func-
a Vmax of 204Æ9 UgCDWÀ1, a substrate uptake con- tionalization; whole-cell biocatalysis
Introduction
means. Multi-step processes including dehydrogena-
tion, separation, and finally oxyfunctionalization are
Fatty acid methyl esters (FAMEs) constitute an inter- necessary to obtain the desired oxygenates.[2] Direct
3
À
esting group of compounds, not only for energy stor- hydroxylation of sp C H bonds based on solid metal
age (e.g., biodiesel) but also as a renewable feedstock catalysts is a possible alternative.[3] The use of non-
for the chemical industry. FAMEs are produced by heme iron complexes was shown to allow the direct
À
the transesterification of vegetable oils or waste lipids oxidation of C H bonds under ambient conditions
yielding glycerol as by-product.[1] Via oxyfunctionali- using H2O2 as oxidant.[4] However, application of
zation of the terminal alkyl carbons, FAMEs give these methods on a preparative scale is still hindered
access to bifunctional precursors for polymer synthe- by poor selectivity, low turnover numbers, and the re-
3
À
sis. Unfortunately, the inert character of the sp C H quirement for a large substrate surplus, in spite the
bonds makes the alkyl chain of FAMEs barely acces- progress concerning selectivity and predictability
sible for specific oxyfunctionalization by chemical achieved by modifying the ligand frameworks of the
Adv. Synth. Catal. 2011, 353, 3485 – 3495
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3485

Manfred Schrewe et al.
FULL PAPERS
Figure 1. General scheme of the NADH-dependent terminal oxyfunctionalization of fatty acid methyl esters with the alkane
monooxygenase system AlkBGT (n=2–9).
catalytic complexes.[5] In general, chemical synthesis the use of whole cells is preferred over in vitro appli-
strategies and the application of metal catalysts are cations, because of the multi-component and partly
expensive and suffer from low efficiencies and poor membrane-bound nature of the enzyme complex,
selectivities.[6]
combined with the dependency on redox cofactors
As a promising alternative, biocatalysts can be ap- and a low stability of isolated enzymes.[23] In this
plied for the direct and selective oxyfunctionalization study, aiming at the conversion of renewable feed-
[7,8]
À
of inert C H bonds.
Biocatalysts combine several stocks into products with added value, the use of re-
desired characteristics for the efficient conversion of combinant E. coli containing AlkBGT was investigat-
low-cost reactants to high-value products, such as high ed for the selective formation of terminal alcohols
turnover numbers, high regio- and enantioselectivity, and aldehydes from FAMEs (Figure 1). In order to
broad substrate spectrum, and environmentally identify possible implications of the whole-cell ap-
friendly reaction conditions.[9] Among the enzymes proach on substrate selectivity and availability/
catalyzing oxyfunctionalizations, oxygenases are most uptake, enriched and isolated enzyme preparations
3
À
versatile with respect to the hydroxylation of sp C H were investigated as well. Finally, coupling efficiencies
bonds,[10] and are thus considered to have a great po- of cofactor oxidation and FAME oxygenation were
tential for application in industrial processes.[8,11] Iron- determined.
containing oxygenases are the most abundant and
À
best investigated oxygenases catalyzing C H oxyfunc-
tionalizations.[10] For example, the hydroxylation of Results and Discussion
long chain fatty acids by the heme-containing cyto-
chrome P450 BM3 is well investigated.[12,13] This The w-Oxyfunctionalization of FAMEs using E. coli
enzyme features very high turnover numbers of up to W3110 Containing AlkBGT
285 sÀ1 (determined based on the NADPH oxidation
rate, frequently biased by uncoupling) but rather low FAMEs were tested as renewable substrates for ter-
regioselectivities by performing the reaction predomi- minal oxyfunctionalization by AlkBGT using resting,
nantly at three different subterminal positions.[13,14] i.e., non-growing but metabolically active, cells of E.
Absolute regioselectivity could not be achieved by coli W3110 carrying the expression plasmid pGEc47.
numerous attempts of protein engineering which, This plasmid encodes all genes of the alkane degrada-
however, resulted in variants with increased regiose- tion pathway from P. putida GPo1.[19]
lectivity.[15] In contrast, terminal oxyfunctionalization
FAMEs ranging from pentanoic acid methyl ester
of long chain alkanes and fatty acids on a commercial to dodecanoic acid methyl ester were converted by
scale is reported for Candida tropicalis.[16] Thereby, resting cells of E. coli W3110 (pGEc47) (Figure 2).
the initial hydroxylation is catalyzed by a cytochrome The alkane monooxygenase AlkBGT catalyzed the
P450 monooxygenase coupled to an NADPH oxidore- hydroxylation reaction exclusively at the w-position
ductase.[17] Similarly, the alkane hydroxylase system leading to the formation of terminal alcohols, as veri-
AlkBGT from Pseudomonas putida GPo1 is known to fied by GC-MS (see the Supporting Information). In
catalyze oxyfunctionalizations of medium chain alka- addition, small amounts of the corresponding alde-
nes and fatty acids exclusively at terminal posi- hydes were detected as products of a second oxida-
tions.[18,19] This three-component enzyme system con- tion step, making up 0–7% of the total products. Only
sists of the monooxygenase (AlkB), a rubredoxin with hexanoic acid methyl ester as substrate, were
(AlkG), and a rubredoxin reductase (AlkT).[20,21] The equal amounts of alcohol and aldehyde formed. The
latter two components are responsible for the electron highest specific hydroxylation activities were found
transfer from NADH to AlkB. AlkB is a membrane- for nonanoic and decanoic acid methyl ester conver-
bound, non-heme diiron enzyme introducing one sions with 71 and 79 UgCDWÀ1, rÀe1spectively. Hydroxy-
oxygen atom from molecular oxygen at the terminal
ACHTUNGTERNlGNUN ation activities of 31–37 UgCDW were found for the
position of alkanes. However, other aliphatic, alicy- conversion of octanoic, undecanoic, and dodecanoic
clic, and aromatic compounds are also accepted as acid methyl esters, whereas low activities of 2 and
substrates.[22] As proposed for oxygenases in general, 4 UgCDW were obtained for pentanoic and heptanoic
À1
3486
asc.wiley-vch.de
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 3485 – 3495

À
Kinetic Analysis of Terminal and Unactivated C H Bond Oxyfunctionalization
Figure 3. SDS-PAGE analysis of the isolated total mem-
Figure 2. FAME spectrum for in vivo biocatalysis with
AlkBGT. (A) Specific initial activities of resting cells of E.
coli W3110 (pGEc47) (light grey) and E. coli W3110
(pBT10) (dark grey). (B) Ratios of product amounts formed
with E. coli W3110 carrying pBT10 or pGEc47. Biomass
conc.: ~1 gCDW LÀ1; substrate conc. 0.75–1.00 mM; each bar
represents the average of at least two independent experi-
ments.
brane fractions and of the supernatant after ultracentrifuga-
tion of cell extracts from (1) E. coli W3110, (2) E. coli
W3110 (pGEc47), and (3) E. coli W3110 (pBT10). Arrow
indicates AlkB (~41 kDa). Cells were grown at 308C in the
presence of 0.5% (w/v) glucose and induced by addition of
0.025% (v/v) DCPK for 4 h.
acid methyl ester conversions, respectively. FAMEs W3110 (pGEc47) representing approximately 60 and
can therefore be considered as highly suitable sub- 20% of total membrane proteins 4 h after induction,
strates for AlkBGT-containing recombinants. The respectively. After ultracentrifugation, no AlkB was
maximum activities achieved were even higher than observed in the supernatant. This indicates that all
À1
the 30–65 UgCDW
reported for E. coli W3110 AlkB was incorporated into the cytoplasmic mem-
(pGEc47) in conversions of octane and nonane.[22,24] brane. Inclusion body formation was not observed.
In accordance with the reported values, octane and Although not visible on SDS-PAGE gels, a similar in-
nonane were converted by E. coli W3110 (pGEc47) crease in the expression levels can be assumed for
with initial activities of 64 and 59 UgCDWÀ1, respec- alkG and alkT since the regulatory system and the
tively, in the set-up used in this study.
gene order are identical in pBT10 and pGEc47.
Plasmid pGEc47 encodes, next to the alkane mono-
E. coli W3110 (pBT10) showed significantly higher
oxygenase system AlkBGT, also for an alcohol dehy- specific transformation rates (activities) for the small-
drogenase (AlkJ), an aldehyde dehydrogenase er substrates with 5 to 9 carbon atoms in the alkyl
(AlkH), and an acyl-CoA synthetase (AlkK).[25] The chain, as compared to E. coli W3110 (pGEc47)
presence of the down-stream enzymes of the alkane (Figure 2). À1 The highest specific activity of
degradation pathway in E. coli W3110 (pGEc47) 104 UgCDW was found for nonanoic acid methyl
might lead to product degradation thereby counter- ester conversion. However, the longer and more hy-
acting the desired product accumulation. E. coli spe- drophobic substrates decanoic, undecanoic, and do-
cies containing pGEc47DJ were shown to catabolize decanoic acid methyl esters were converted at higher
dodecanoic acid despite the presence of glucose.[26] rates by E. coli W3110 (pGEc47). E. coli W3110
Although alcohols and aldehydes were the only prod- (pBT10) showed specific dodecanoic acid methyl
ucts detected during conversions of FAMEs in the ester (DAME) hydroxylation rates of only 2%, as
performed short-term resting cell assays, possible uti- compared to nonanoic acid methyl ester (NAME) hy-
lization as carbon source via b-oxidation cannot be droxylation. This is in sharp contrast to the 43% ob-
excluded and the presence of the down-stream en- served with E. coli W3110 (pGEc47). Higher activities
zymes should be avoided to increase product yields.
of E. coli W3110 (pBT10) towards the shorter sub-
In order to obtain higher AlkBGT expression strates can be explained by higher AlkBGT levels in
levels, resulting in higher specific activities and to these cells, although the factor of activity increase
avoid possible product degradation, the high-copy varied between 1.3 and 4. However, catalytic activities
number plasmid pBT10 was constructed encoding for substrates with a longer alkyl chain did not corre-
only the three monooxygenase components. AlkB ex- late with AlkB expression levels.
pression levels in E. coli W3110 (pGEc47) and E. coli
The potential of the AlkBGT system for the pro-
W3110 (pBT10) were investigated by SDS-PAGE duction of terminal alcohols and the corresponding
analysis of membrane fractions (Figure 3). AlkB was acids from alkanes has been reported previously.[27]
indeed more abundant in the membrane fraction of Here, for the first time, AlkBGT is shown to be
E. coli W3110 (pBT10), as compared to E. coli highly suitable for whole-cell conversions of renewa-
Adv. Synth. Catal. 2011, 353, 3485 – 3495
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3487

Manfred Schrewe et al.
FULL PAPERS
ble fatty acid methyl esters at high rates and with ab- ence of substrate and product leading to uncoupling
solute terminal regioselectivity.
of cofactor consumption and product formation.
The maximal initial NADPH oxidation rates were
3- to 4-fold higher after addition of substrates com-
pared to the background activity without substrate.
Comparable NADPH consumption rates were found
Conversion of FAMEs in vitro
Since AlkB expression levels cannot explain the low for substrate concentrations ranging from 500 to
activities of E. coli W3110 (pBT10) towards the 2500 mM NAME and 200 to 1000 mM DAME. The ad-
longer chain substrates, a component encoded on dition of lower substrate concentrations led to lower
plasmid pGEc47 obviously facilitated their conver- specific NADPH oxidation rates, indicating kinetic
sion. This can either be a protein assisting intracellu- substrate limitation. Since significant background
lar AlkB catalysis, in a similar way as reported for NADPH oxidation rates were observed without the
component B of the methane monooxygenase,[28] or a supply of FAMEs as substrate, not all reducing equiv-
factor enhancing the intracellular availability of alents might have been used for product formation.
poorly water-soluble, larger substrates. In order to get Additionally, the presence of substrates and products
more insight into the difference in substrate selectivi- might have induced uncoupling to occur at higher
ty of the two recombinant strains, in vitro biotransfor- rates.[11,30] Therefore, NADPH consumption and prod-
mations of NAME and DAME were performed using uct formation were compared over 15 min of reaction
isolated and enriched enzyme preparations. The time (Table 2). Remarkably, the coupling efficiency
enzyme preparations for AlkB and AlkG were ob- was 100% using NAME as substrate, which implies
tained from different recombinant E. coli strains sepa- that uncoupling could be avoided by addition of this
rately over-expressing the respective genes (see Ex- substrate in non-limiting amounts. In contrast, the
perimental Section). The in vitro studies were per- presence of DAME promoted uncoupling with slight-
formed by combining enriched AlkB and AlkG prep- ly increasing coupling efficiencies with increasing sub-
arations and commercially available spinach reductase strate concentrations. The average coupling efficiency
as described before.[20,22]
of 26% and the initial NADPH oxidation rate of
The background NADPH oxidation rates for in-
complete reconstitutions were low, except for AlkG
present together with the reductase, which showed in- Table 2. Product formation and coupling efficiencies in vitro.
creased background activities, probably caused by im-
Substrate csubstrate Product
NADPH
consumed
[mM][a]
Coupling
efficiency
[%]
purities of the AlkG preparation (Table 1). AlkG is
known to transfer the electrons also to acceptors
other than AlkB, for example, to cytochrome c.[29]
The reconstituted three-component enzyme system
showed significant NADPH consumption rates of
5.7 ULÀ1 without substrate addition. This implies the
occurrence of uncoupling with electron transfer to
molecular oxygen, typically resulting in the formation
of reactive oxygen species, e.g., H2O2. Such reactions
may also occur during biotransformation in the pres-
[mM]
formed
[mM][a]
NAME
DAME
500
2500
200
194
203
52
194
204
216
236
249
100
100
24
27
28
500
64
1000
70
[a]
Product formation and NADPH consumption was deter-
mined during 15 min of reaction.
Table 1. Initial NADPH consumption rates obtained for in vitro biotransformations of nonanoic acid methyl ester (NAME)
and dodecanoic acid methyl ester (DAME).
Component
NADPH consumption rate ULÀ1 [mmolminÀ1 LÀ1]
Without substrate
NAME
DAME
AlkB
AlkG
0.1
0.2
0.1
1.5
0.6
3.4
1.1[a]
n.d.[e]
2.1[a]
n.d.[e]
Reductase (Red.)
AlkB+AlkG
AlkB+Red.
AlkG + Red.
AlkB+AlkG+Red.
0.7[a]
n.d.[e]
1.4[a]
n.d.[e]
0.4[a]
n.d.[e]
4.0[a]
n.d.[e]
5.7Æ0.7 (71Æ8)[d]
14.8Æ0.1[b] (186Æ1)[d]
22.1Æ0.5[c] (276Æ6)[d]
[a]
2.5 mM substrate added.
Average for addition of 500, 1000, and 2500 mM substrate.
Average for addition of 200, 500, and 1000 mM substrate.
In brackets, rates are given as specific activities expressed in UgAlkB
n.d.: not determined.
[b]
[c]
[d]
[e]
À1
.
3488
asc.wiley-vch.de
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 3485 – 3495

À
Kinetic Analysis of Terminal and Unactivated C H Bond Oxyfunctionalization
Table 3. Comparison of specific activities obtained for nonanoic acid methyl ester (NAME) and dodecanoic acid methyl
ester (DAME) conversions in vivo and in vitro.
System
Substrate
Specific activity [UgÀ1]
Ratio activities NAME/DAME
55
E. coli W3110 (pBT10)
NAME
DAME
NAME
DAME
NAME
DAME
104Æ2[a]
1.9Æ0.4[a]
71Æ3[a]
E. coli W3110 (pGEc47)
2.3
2.6
31Æ6[a]
in vitro
186Æ1[b]
72Æ2[c]
[a]
[b]
[c]
À1
UgCDW
.
UgAlkBÀ1; based on initial NADPH oxidation activities assuming 100% coupling.
UgAlkBÀ1; based on initial NADPH oxidation activities assuming 26% coupling.
22.1 ULÀ1 (Table 1) translate into 16.4 ULÀ1 of un- the reconstituted enzyme assays correlated with the
coupled NADPH oxidation in the presence of ratio observed with E. coli W3110 (pGEc47). In con-
DAME, which is a factor of 3 higher than the uncou- trast, E. coli W3110 (pBT10) showed a much higher
pling observed in the absence of DAME. Obviously, ratio. This and the fact that the enzyme preparations
DAME addition led to a significant acceleration of were obtained from strains over-expressing alkB or
uncoupling reactions in vitro.
alkG only, indicate that pGEc47 does not encode
In an analogous in vitro set-up similarly high components supporting the AlkBGT-catalyzed con-
NADPH oxidation rates have been observed for the version of DAME but rather improves intracellular
hydroxylation of C8–C12 alkanes by AlkBGT, whereas substrate availability. As the outer membrane of
Pseudomonas cells showed decreasing product forma- Gram-negative bacteria constitutes an effective barri-
tion rates with increasing chain length (being maximal er for large and hydrophobic compounds,[33] whole-
for C8 and C9 and not detectable for C12).[22] Grant cell bioconversions of such substrates are often limit-
et al. showed that E. coli DH1 (pGEc47DJ) contain- ed by poor substrate availability in the bacterial
ing AlkBGT preferred the oxidation of octane over cell.[11,34] In cells containing pGEc47, DAME mass
dodecane. Dodecane oxidation rates constituted only transfer over the outer membrane may be facilitated
14–16% of octane oxidation rates.[26] This discrepancy by the pGEc47-encoded outer membrane protein
between in vitro NADPH oxidation and in vivo prod- AlkL, which has been hypothesized to be involved in
uct formation may, at least in part, be attributed to substrate uptake.[25]
uncoupling in the presence of long chain substrates as
A comparison of the whole-cell activities with in
was the case for FAMEs. However, mass transfer lim- vitro results obtained for FAME conversion revealed
itations also may reduce in vivo conversion rates of limitations for the larger and more hydrophobic sub-
long chain substrates such as dodecane.
strate DAME to occur on both levels, the selectivity
Van Beilen, Witholt, and co-workers reported in- of the monooxygenase AlkB and, for E. coli W3110
creased growth rates on longer chain alkanes (C12 to (pBT10), intracellular substrate availability.
C16) for recombinant Pseudomonas strains with
NAME was found to be a preferred substrate for
mutant alkB genes.[31] It was concluded that the ex- terminal oxyfunctionalization by AlkBGT both in
change of the bulky tryptophan residue with the terms of coupling and whole-cell activity. To the best
smaller amino acids cysteine or serine (W55C and of our knowledge, 104 UgCDWÀ1, as determined here,
W55S) enlarged the substrate binding pocket and is the highest whole-cell activity reported for
thus enabled larger substrate molecules to access the AlkBGT to date.
enzyme with reduced steric hindrance. Since uncou-
pling is described for substrates not fitting well into
the active site of an enzyme,[32] the described muta- Characterization of E. coli W3110 (pBT10) as
tions might have led to an improved coupling efficien- Catalyst for NAME Oxyfunctionalization
cy, higher alkane conversion rates, and subsequently
to higher growth rates.
To determine apparent whole-cell kinetics, resting cell
In order to compare the obtained in vitro and in assays with varying NAME and 9-hydroxynonanoic
vivo results, the initial in vitro NAME and DAME hy- acid methyl ester (HNAME) concentrations were per-
droxylation rates were calculated from the NADPH formed. Alcohol conversion was investigated because
oxidation rates assuming 100 and 26% coupling, re- low aldehyde concentrations have been found in ini-
spectively (Table 3). Interestingly, the ratio of initial tial experiments. The conversions of both substrates
NAME and DAME hydroxylation rates observed in followed Michaelis–Menten-like kinetics (Figure 4).
Adv. Synth. Catal. 2011, 353, 3485 – 3495
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3489

Manfred Schrewe et al.
FULL PAPERS
Figure 4. Apparent Michaelis–Menten-like kinetics for
whole-cell conversions of (A) nonanoic acid methyl ester
(NAME) (dashed line represents the NAME solubility of
133 mM in water) and (B) 9-hydroxynonanoic acid methyl
ester (HNAME) catalyzed by E. coli W3110 (pBT10). Initial
activities determined within 5 min of biotransformation are
given applying a biomass concentration of 0.08 gCDW lÀ1.
Even with NAME concentrations exceeding the solu-
bility limit of 133 mM in water, the kinetics followed
Michaelis–Menten-like behaviour (Figure 4, A).
Considering apparent kinetic parameters (Table 4),
the initial hydroxylation step is preferred over alcohol
oxidation. Thereby, NAME hydroxylation catalyzed
by E. coli W3110 (pBT10) shows an efficiency which
is comparable to styrene epoxidation by recombinant
E. coli containing the styrene monooxygenase StyAB
(Vmax =107 UgCDWÀ1; KS =64 mm)[35] and pseudocu-
mene hydroxylation by E. coli containing the xylene
monooxygenase XylMA (Vmax =351 UgCDWÀ1; KS =
202 mm),[36] two of the most efficient biocatalytic
whole-cell oxyfunctionalization reactions reported.
Figure 5. Biotransformation of 1.0, 2.5, 5.1, and 9.8 mM of
NAME with resting E. coli W3110 (pBT10). Dashed line
represents the initial substrate concentration. A biomass
concentration of 1.2 gCDW lÀ1 was applied.
The maximum specific activities reached in the
three biotransformations with initial substrate concen-
trations of 2.5, 5.1,Àa1nd 9.8 mM were on the expected
level of 100 UgCDW as seen in previous experiments,
whereas the addition of 1 mM substrate led to sub-
Optimization of Nonanoic Acid Methyl Ester
Oxyfunctionalization
In order to optimize product yields, NAME was strate limitation (Table 5).
added in concentrations varying from 1 to 10 mM to
After the first 10 min of the biotransformations
resting E. coli W3110 (pBT10) and the time course of with 5.1 and 9.8 mM NAME, approximately 1 mM of
the biotransformation was followed. In all cases, the HNAME was produced (Figure 5, C and D). Using
terminal alcohol was the main product accumulating, the Michaelis–Menten equation and the determined
with up to 0.3 mM of the aldehyde as minor by-prod- apparent kinetic parametersÀ(1Table 4), specific alcohol
uct (Figure 5, A–D). The addition of 5.1 mM substrate oxidation rates of 39 UgCDW would be expected with
appeared to be optimal. 4.8 mM substrate were con- 1 mM of the alcohol present. However, the maximum
verted within 2 h of reaction time resulting in alcohol alcohol oxidation activities observed in these experi-
and aldehyde concentrations of 4.5 and 0.2 mM, re- ments were only 8 UgCDWÀ1. This indicates that the al-
spectively. The product concentrations correlated to cohol lost the competition for the catalyst against
yields based on consumed substrate of 95 and 4% for NAME, which is in accordance with its significantly
alcohol and aldehyde, respectively (Table 5).
higher substrate uptake constant as compared to
NAME (Table 4). Similar results have been published
Table 4. Apparent kinetic parameters for the whole-cell conversions of nonanoic acid methyl ester (NAME) and 9-hydroxy-
nonanoic acid methyl ester (HNAME) with resting E. coli W3110 (pBT10).
Vmax/KS [UgCDWÀ1 mMÀ1]
À1
Substrate
Apparent Vmax [UgCDW
]
Apparent KS [mM]
NAME
HNAME
204Æ9
71Æ3
142Æ17
823Æ70
1.4
0.1
3490
asc.wiley-vch.de
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 3485 – 3495

À
Kinetic Analysis of Terminal and Unactivated C H Bond Oxyfunctionalization
Table 5. Maximum specific activities and yields obtained in bioconversions with different initial NAME concentrations.
Time Initial sub-
[min] strate conc.
[mM]
Remaining
substrate [%] tion rate [UgCDW
Max. NAME oxiÀd1a- Max. HNAME oÀx1ida- Y alco- Y alde-
Y oxidation
]
tion rate [UgCDW
]
hol
hyde
products [%][a]
[%][a]
[%][a]
12
30
120
120
1.0
2.5
5.1
9.8
13
9
7
83
8
5
8
8
69
76
95
92
13
7
4
82
83
99
94
100
103
96
45
2
[a]
Yields were calculated based on converted substrate.
recently for octane conversions by recombinant E.
coli expressing alkBGT,[26] and for xylene monooxyge-
nase containing E. coli converting pseudocumene and
toluene.[36] Alcohol oxidation was slow as long as the
initial substrate was present in excess and the alcohol
concentration was below a critical limit.
During biotransformations with initial substrate
concentrations of 1 and 2.5 mM, alcohol and aldehyde
concentrations decreased once NAME was almost de-
pleted (Figure 5, A and B). This and the decrease in
total substrate and product recovery indicate product
degradation. If product degradation led to the forma-
tion of molecules carrying a carboxyl group (either by
ester hydrolysis or by formation of nonanedioic acid
monomethyl ester), these molecules could enter the
b-oxidation cycle. Product degradation can be caused
by host intrinsic enzymes as reported by Meyer et al.
for m-nitrotoluene and its oxidized derivatives,[37] or
by the overoxidation of the products catalyzed by
AlkBGT. Overoxidation is a feature often observed
for monooxygenases.[11] For xylene monooxygenase
XylM from P. putida mt-2 with a sequence identity of
25% to AlkB,[38] the overoxidation of toluene to ben-
zoic acid was reported.[39] Here, the occurrence of
Figure 6. Initial volumetric and specific NAME oxyfunction-
alization activities of resting E. coli W3110 (pBT10) as a
function of biomass concentration. The dashed line repre-
sents the expected specific activity according to the kinetics
determined for NAME conversion (Figure 4 A) using its sol-
ubility limit of 133 mM in water as substrate concentration.
A substrate concentration of 2.5 mM was applied. * Only
1 mM of substrate was added (data from kinetic assay).
AlkBGT-catalyzed overoxidation seemed to be the pected considering the kinetics for NAME conversion
dominant effect since no further degradation was ob- (Figure 4, Table 4). ForÀ1biomass concentrations be-
served as long as the NAME concentration was not tween 0.5 and 1.3 gCDW L , activities were found to be
limiting (Figure 5, C and D).
in this range. The application of lower biomass con-
À1
The obtained maximalÀ1hydroxylation activities of centrations (0.08 and 0.16 g
L ) resulted in signifi-
CDW
approximately 100 UgCDW at biomass concentrations cantly higher specific activities. Direct substrate
À1
of 1.2 g
L
(Table 5) were significantly lower than uptake from organic phase droplets (undissolved sub-
CDW
the activities observed during kinetic analyses. Apply- strate) could be a possible explanation for these high
À1
ing very low biomass concentrations of 0.08 g
L , conversion rates. Such direct uptake is considered to
CDW
À1
maximal specific activities of 175 UgCDW were deter- be one of the mechanisms how hydrocarbon degrad-
mined (Figure 4). In order to investigate the influence ers get access to poorly water soluble substrates.[40]
of the biomass concentration on the specific activity, Bacterial cells collide with and/or attach to these
biotransformations with varying cell concentrations droplets and thereby face far higher substrate concen-
were perfomed. Volumetric activities correlated line- trations as present in the aqueous surrounding. The
À1
arly with biomass concentrations up to 1.3 g
L
addition of 2.5 mM NAME and mixing on an orbital
CDW
reaching a plateau with higher biomass concentrations shaker is expected to result in the formation of sub-
(Figure 6). Mass transfer limitations of either sub- strate droplets in the aqueous phase. Whereas such
strate or oxygen probably caused this effect.
direct uptake may have occurred in experiments with
The corresponding specific activities observed can very low biomass concentrations, direct access to a
be divided into three different stages (Figure 6). For substrate droplet may decrease with increasing bio-
an aqueous substrate concentration corresponding to mass conÀce1 ntrations. At cell concentrations above
À1
the solubility limit of 133 mM, 100 UgCDW can be ex- 1.3 gCDW L in the set-up used, oxygen mass transfer
Adv. Synth. Catal. 2011, 353, 3485 – 3495
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3491

Manfred Schrewe et al.
FULL PAPERS
over the gas/liquid phase boundary and/or substrate coli W3110 (pEGc47) expressing all genes of the
mass transfer over the organic/aqueous phase boun- alkane degradation pathway of P. putida GPo1. This
dary may have become limiting, resulting in decreased points to the presence of a specific substrate uptake
specific activities. However, such mass transfer can system in the latter strain. Next to substrate availabili-
significantly be improved by the use of standard bio- ty, substrate selectivity and coupling efficiency of the
reactor configurations, e.g., stirred tank reactors, AlkBGT enzyme complex are engineering targets to
making the use of high cell densities feasible for bio- improve the w-oxyfunctionalization of longer chain
catalytic NAME hydroxylation.
FAMEs.
Conclusions
Experimental Section
This study showed the high potential of the whole-cell Chemical, Bacterial Strains, and Media
biocatalyst E. coli W3110 (pBT10) expressing the
All chemicals used for this study were purchased from
alkane monooxygenase genes alkBGT for the specific
w-oxyfunctionalization of medium chain length
FAMEs as renewable feedstocks. It also demonstrated
the control of catalyst efficiency by various parame-
ters other than enzyme characteristics. The exclusive
formation of terminally functionalized products from
unactivated alkyl carbons and the high hydroxylation
rates achieved with AlkBGT in vivo, ranging among
the highest oxyfunctionalization rates reported for
whole-cell biocatalysis, underline the versatility of this
enzyme system. The efficient coupling of cofactor
consumption and product formation (100%), the high
alcohol yield (95%), and the high conversion rate
(104 UgCDWÀ1) achieved with the substrate nonanoic
acid methyl ester offer an excellent starting point for
the preparative synthesis of terminally functionalized
FAMEs in a biotransformation process. Beside sub-
strate selectivity of AlkBGT, substrate availability
and uncoupling appeared to be critical for the conver-
sion of longer and more hydrophobic substrates, of
which the intracellular availability was higher in E.
Sigma–Aldrich (Steinheim, Germany) or Carl Roth (Karls-
ruhe, Germany) with the highest purity available except for
9-hydroxynonanoic acid methyl ester (purity of ꢀ95%, TCI
Europe N.V., Zwijndrecht, Belgium) and 12-hydroxydodeca-
noic acid methyl ester (purity of ꢀ95%, Evonik Degussa
GmbH, Marl, Germany). All strains and plasmids used in
this study are listed in Table 6. For cloning purposes, E. coli
DH5a was used. For expression and biotransformation ex-
periments E. coli W3110 was transformed with the different
plasmids. Strains were cultivated in M9[41] or modified M9
(M9*)[42] medium supplemented with 1 mLLÀ1 USFe trace el-
ement solution[43] and 0.5% (w/v) glucose or in LB
medium.[41] The antibiotics (50 mgLÀ1 kanamycin or
12.5 mgLÀ1 tetracycline) were added to all media if appro-
priate. Solid media contained 15 gLÀ1 agar.
DNA Manipulation Techniques
Standard techniques were used for restriction analysis, clon-
ing, and agarose gel electrophoresis.[41] Phusion High-Fideli-
ty DNA Polymerase obtained from Finnzymes Oy (Espoo,
Finland) was used for all PCR reactions. All restriction en-
donucleases were purchased from Fermentas GmbH (St.
Table 6. Bacterial strains and plasmids used in this study.
Strain or Plasmid
Characteristics
Ref.
Strain
E. coli DH5a
[44]
supE44 DlacU169 (F80 lacZDM15) hsdR17 recA1 endA1 gyrA96
thi-1 relA1
[45]
E. coli W3110
Plasmid
pCOM10
FÀ, lÀ1, rph-1, IN(rrnD-rrnE)1
[46]
alkane responsive broad-host-range vector; PalkB; alkS
pSMART-HCKan
blunt-end subcloning vector, Kmr
Lucigen Corporation
(Middleton, WI, USA)
[19]
pGEc47
contains genes necessary for growth on alkanes (alkBFGHJKL and
alkST) in broad-host-range vector pLAFR1, Tcr
pSMART-HCKan containing alkBFG
pSMART-HCKan containing alkT
pCOM10 containing alkBFG
pSMART-BFG
pSMART-T
pBG10
this study
this study
this study
this study
pBT10
contains genes for alkane monooxygenase (alkBFG and alkST) in
broad-host range vector pCOM10, Kmr
pAMB
pAMG
contains gene for the oxygenase component alkB in
broad-host range vector pCOM10, Kmr
this study
this study
contains gene for the rubredoxin alkG in broad-host
range vector pCOM10, Kmr
3492
asc.wiley-vch.de
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 3485 – 3495

À
Kinetic Analysis of Terminal and Unactivated C H Bond Oxyfunctionalization
Leon-Rot, Germany). All primers were purchased from Eu-
rofins MWG (Ebersberg, Germany). Successful cloning was
confirmed by sequencing by Eurofins MWG.
M9* precultures of recombinant E. coli W3110 carrying
pAMB or pAMG were used to inoculate a bioreactor (KLF
2000, Bioengineering AG, Wald, Switzerland) with 1.5 L of
M9 medium containing 3% (w/v) glucose. The dissolved
oxygen tension (DOT) was determined with an autoclavable
amperometric probe (Mettler–Toledo GmbH, Schwerzen-
bach, Switzerland) and the pH (pH probe: Mettler–Toledo
GmbH, Schwerzenbach, Switzerland) was automatically
kept at 7.2 by adding 30% phosphoric acid and 25%
NH4OH. Temperature, agitation frequency, and aeration
rate were maintained at 308C, 1.5 LminÀ1, and 1500 rpm, re-
spectively. After batch phase, 0.4% (v/v) USFe trace solution
were added to the cells and an exponential feed [50% (w/v)
glucose, 13.7 gLÀ1 magnesium sulfate] was started for a glu-
cose limited exponential growth rate m=0.2 hÀ1. After 3–4 h
of exponential growth, alk-gene expression was induced by
addition of 0.025% (v/v) DCPK and after another 4 h of
fed-batch cultivation, the cells were harvested by centrifuga-
tion, washed with 10 mM TRIS-HCl (pH 8.0) for AlkB en-
richment or 50 mM TRIS-HCl (pH 7.5) for AlkG enrich-
ment, and stored as pellet at À208C or immediately used for
protein enrichment.
A two-step cloning strategy was followed to obtain an
alkBGT expression vector with a gene organization closely
related to the original operons in P. putida GPo1. The
alkBFG genes were amplified by PCR from plasmid
pGEc47[19] using the primers 5’-AAGGGAATTCCATATGCTT
GAGAAACACAGAGTTC-3’ containing an NdeI restriction site
(underlined) and 5’-AAAATTCGCGTCGACAAGCGCTGAATGGG
TATCGG-3’ containing a SalI restriction site (underlined).
The PCR product was cloned into the blunt-end cloning
vector pSMART-HCKan (Lucigen Corporation, Middleton,
WI, USA) giving pSMART-BFG. After restriction with
NdeI and SalI, the fragment obtained was ligated into the
alkane responsive broad-host-range expression vector
pCOM10[46] cut with the same enzymes resulting in the for-
mation of plasmid pBG10 containing alkBFG under control
of the PalkB promoter.
The alkT gene was amplified from plasmid pGEc47 by
PCR using the primers 5’-TGAGACAGTGGCTGTTAGAG-3’
and 5’-TAATAACCGCTCGAGAACGCTTACCGCCAACACAG-3’
containing an XhoI restriction site (underlined). The PCR
product was cloned into pSMART-HCKan giving pSMART-
T. After restriction with PacI and XhoI, the fragment ob-
tained was ligated into pBG10 cut with the same restriction
enzymes. The resulting plasmid pBT10 contained alkT under
the control of PalkS downstream of the alkS gene.
Protein Isolation and Enrichment
For isolation of total membrane fractions from alkBGT ex-
pressing strains, cell pellets were resuspended in a 55 mM
sodium-potassium phosphate buffer (pH 7.5) and disrupted
by two passages through a French press (5.5 MPa, SLM-
Aminco, Rochester, NY, USA). Cell debris was removed by
centrifugation at 12,000 g and 48C for 15 min. The superna-
tant was centrifuged at 200,000 g and 48C for 2 h. The re-
sulting pellet containing the membrane fraction was resus-
pended in the same buffer and analyzed by SDS-PAGE.
For AlkB enrichment, spheroplasting was performed as
described previously,[47] with the modification that no EDTA
was added as it is known to impair AlkB activity.[48] The re-
sulting pellet was stored at À208C. For further purification,
the pellet was resuspended in 10 mM TRIS-HCl (pH 8.0) to
a concentration corresponding to 25 gCDW LÀ1. Bacterial
DNA was degraded by adding 5 mM magnesium sulphate
and 0.25 UmLÀ1 DNase, and incubation on ice for 20 min.
Then, the solution was slowly passed through a French
press. Cell debris was removed by centrifugation at 12,000 g
and 48C for 30 min. The supernatant was centrifuged at
250,000 g and 48C for 90 min. The pellet from the ultracen-
trifugation was directly used or stored at À208C until
needed. AlkB was enriched by density gradient centrifuga-
tion. A discontinuous sucrose gradient was prepared ranging
from 30–55% (w/v) sucrose, with steps of 5% sucrose. Each
density fraction had a volume of 1.76 mL except the 30%
fraction, which had a volume of 0.88 mL. The pellet ob-
tained by ultracentrifugation was dissolved in 25% sucrose
and 0.8 mL of this solution was loaded on top of the gradi-
ent. Thus, the total volume in each centrifugation tube was
10.5 mL. After centrifugation at 170,000 g and 48C for 22 h,
the solution was fractionized by carefully pipetting the su-
crose gradient from the top. After analyzing the different
fractions by SDS-PAGE, the AlkB containing fractions were
pooled and stored at À208C. The yield was 15% and purity
was estimated by SDS-PAGE to be over 80%.
Plasmids for separate alkB and alkG expression were con-
structed to facilitate the enrichment of single enzyme com-
ponents for in vitro studies. Both genes were amplified from
pBT10 using for alkB the primers 5’-GAGAATTCCATATGCTT
GAGA-3’ containing a NdeI restriction site (underlined) and
5’-GCTGGAAGATCTACGATGCTACCGCAGAG-3’ containing a
BglII restriction site (underlined) and, for alkG, the primers
5’-GGGAATTCCATATGGCTAGCTATAAATGCCC-3’ containing
a NdeI restriction site (underlined) and 5’-TTACGCGGATCCT
CACTTTTCCTCGTAGAGCA-3’ containing a BamHI restric-
tion site (underlined). The PCR products were cut with the
respective restriction enzymes and separately ligated into
pCOM10[46] cut with NdeI and BamHI resulting in pAMB
and pAMG, carrying the alkB and alkG genes, respectively,
under control of the PalkB promoter.
Cultivation and Gene Expression
Cultivation was generally performed in orbital shakers
(Multitron, Infors HT, Bottmingen, Switzerland) at 200 rpm
and 378C for LB cultures or 308C for minimal medium cul-
tures. Precultures of 5 mL LB medium were inoculated with
a single colony of the recombinant E. coli strain and grown
for 10 h. Subsequently, 500 mL of the precultures were used
to inoculate 50 mL M9* medium. These cultures were incu-
bated overnight and used either to inoculate a bioreactor
(see below) or 100 mL M9* shake flask cultures to an initial
biomass concentration of 0.03 gCDW LÀ1. The culturesÀ1were
incubated until a biomass concentration of 0.08 gCDW L was
reached. Then, alkBGT expression was induced by the addi-
tion of 0.025% (v/v) dicyclopropyl ketone (DCPK) and the
cells were incubated for another 4 h. Subsequently, the
AlkBGT-containing cells were harvested by centrifugation
(4595 g, 10 min, 48C).
Adv. Synth. Catal. 2011, 353, 3485 – 3495
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3493

Manfred Schrewe et al.
FULL PAPERS
For AlkG enrichment, cells were resuspended in 50 mM
TRIS-HCl (pH 7.5) to a density of 40 gCDW LÀ1 and disrupted
by 5 sequential French press passages. After removal of cell
debris by centrifugation at 4595 g and 48C for 30 min, the
supernatant was incubated at 708C for 15 min and then cen-
trifuged at 12,000 g and 48C for 10 min to remove heat-de-
natured proteins. An ice-cold and saturated ammonium sul-
phate solution was added dropwise to the supernatant under
slow stirring in an ice-bath. The fraction between 30% and
60% saturation was collected and stored at À208C until
used. The AlkG preparation had a purity of 50% as deter-
mined by SDS-PAGE.
Sodium dodecyl sulphate-polyacrylamide gel electropho-
resis (SDS-PAGE) was performed by the method of
Laemmli.[49] Gels were stained with Coomassie blue. A mo-
lecular weight standard (Fermentas SM0671) was used as
reference.
Analytical Methods
FAMEs and corresponding products were quantified by GC
(Trace GC UltraTM, ThermoFisher Scientific, Waltham,
USA) equipped with
a
FactorFour-VF-5 ms capillary
column, containing 5% phenyl- and 95% dimethylpolysilox-
ane (30 m, ID 0.25 mm, DF 0.25 mm, Varian, Palo Alto, Cali-
fornia, USA) with molecular nitrogen as carrier gas. Detec-
tion was accomplished by an FID. For pentanoic and hexa-
noic acid methyl esters the temperature profile was (i) 408C
for 3 min, (ii) 40 to 1708C at 158CminÀ1, (iii) 170 to 3008C
at 1008CminÀ1, and (iv) 3008C for 3 min. For all other
FAMEs the temperature profile was (i) 80 to 2808C at
158CminÀ1; (ii) 280 to 3008C at 1008CminÀ1, and (iii)
2.5 min at 3008C. Products were quantified based on stan-
dard calibration curves for available standards. Concentra-
tions of non-available compounds were estimated based on
their carbon content, comparison with the available stand-
ards (9-hydroxynonanoic and 12-hydroxydodecanoic acid
methyl ester), as well as from substrate depletion.
The products of FAME conversions were identified by
gas chromatography mass spectrometry (GC-MS) using a
Varian 1200 Quadrupole MS coupled to a Varian CP-3800
Gas Chromatograph using helium as carrier. The same
column and temperature profiles as used for GC-FID analy-
sis were applied. The oxyfunctionalized products were iden-
tified by their characteristic MS fragmentation patterns.
Biotransformations
The AlkBGT activity of resting cells was determined using
cells resuspended in 1 mL of 50 mM potassium phosphate
buffer (pH 7.4) containing 1% (w/v) glucose at a determined
cell density between 0.08 and 2.7 gCDW LÀ1. The reaction mix-
tures were incubated for 5 min at 308C in an orbital shaker
(300 minÀ1) in order to adapt to the reaction conditions. All
assays contained 2.5% (v/v) ethanol after addition of the
substrate stock solution. The reactions were started by addi-
tion of substrate solution and stopped after different time
intervals by addition of 1 mL ice-cold diethyl ether contain-
ing dodecane (0.2 mM) as internal standard for gas chroma-
tography (GC) analysis. Extraction was performed by vor-
texing the sample for 1 min. Subsequently, the samples were
centrifuged (4595 g, 48C, 5 min) for improved phase separa-
tion, the ether phase was dried over anhydrous Na2SO4, and
Acknowledgements
This study was financially supported by the Federal Ministry
of Education and Research (BMBF), grant number 0315205.
analyzed by GC. One unit (U) of whole-cell activity was de- References
fined as 1 mmol of product being produced per minute. Spe-
cific activities (UgCDWÀ1) were calculated per gram cell dry
weight (CDW).
[1] a) D. Y. C. Leung, X. Wu, M. K. H. Leung, Appl.
Energy 2010, 87, 1083–1095; b) A. Y. Tremblay, P. G.
Cao, M. A. Dube, Energy Fuels 2008, 22, 2748–2755.
[2] J. A. Labinger, J. E. Bercaw, Nature 2002, 417, 507–514.
[3] R. Burch, M. J. Hayes, J. Mol. Catal. A: Chem. 1995,
100, 13–33.
[4] a) F. Gozzo, J. Mol. Catal. A: Chem. 2001, 171, 1–22;
b) A. A. Fokin, P. R. Schreiner, Adv. Synth. Catal. 2003,
345, 1035–1052.
[5] a) M. S. Chen, M. C. White, Science 2007, 318, 783–787;
b) M. S. Chen, M. C. White, Science 2010, 327, 566–571.
[6] a) R. A. Periana, G. Bhalla, W. J. Tenn, K. J. H. Young,
X. Y. Liu, O. Mironov, C. J. Jones, V. R. Ziatdinov, J.
Mol. Catal. A: Chem. 2004, 220, 7–25; b) E. E. Wolf,
Methane conversion by oxidative processes, Van Nos-
trand Reinhold, New York, 1992.
[7] a) W. A. Duetz, J. B. van Beilen, B. Witholt, Curr.
Opin. Biotechnol. 2001, 12, 419–425; b) J. B. van Beilen,
E. G. Funhoff, Curr. Opin. Biotechnol. 2005, 16, 308–
314; c) S. F. Ye, F. Neese, Curr. Opin. Chem. Biol. 2009,
13, 89–98.
In vitro activity assays with a reconstituted enzyme system
with enriched AlkB and AlkG enzyme preparations were
performed essentially as described earlier.[22] The reaction
volume was 1 mL and contained 80 mg AlkB, 400 mg AlkG,
80 mU spinnach ferredoxin reductase, 300 mm NADPH, 0.2
to 2.5 mM substrate, 1% acetone, and 50 mM TRIS-HCl
(pH 7.5). All components except for the substrate and ace-
tone were mixed in a cuvette and tempered at 308C. The re-
action was started by the addition of nonanoic acid methyl
ester or dodecanoic acid methyl ester dissolved in acetone
followed by short mixing. The consumption of NADPH was
followed by measuring the decrease in absorption at
340 nm. After 15 min, the reaction was stopped by extract-
ing the remaining substrate and products formed with an
equal volume of ethyl acetate containing 0.5 mM decane as
internal standard for GC analysis. The product concentra-
tion was determined by GC-analysis. AlkB was applied as
the rate-limiting component. The extraction efficiency for
nonanoic and dodecanoic acid methyl ester and the corre-
sponding products was determined with known amounts of
the compounds added to the assay solution containing all
the components.
[8] B. Bꢀhler, A. Schmid, J. Biotechnol. 2004, 113, 183–210.
[9] S. M. Resnick, K. Lee, D. T. Gibson, J. Ind. Microbiol.
Biotechnol. 1996, 17, 438–457.
3494
asc.wiley-vch.de
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 3485 – 3495

À
Kinetic Analysis of Terminal and Unactivated C H Bond Oxyfunctionalization
[10] B. Bꢀhler, A. Schmid, Selective enzymatic hydroxyla-
[27] a) B. Witholt, M. J. de Smet, J. Kingma, J. B. van Bei-
À
tions, in: Handbook of C H Transformations, (Ed.: G.
Dyker), Vol. 2. Wiley-VCH, Weinheim, 2005, pp 516–
529.
len, M. Kok, R. G. Lageveen, G. Eggink, Trends Bio-
technol. 1990, 8, 46–52; b) O. Favre-Bulle, T. Schouten,
J. Kingma, B. Witholt, Bio-Technology 1991, 9, 367–
371; c) M. G. Wubbolts, O. Favre-Bulle, B. Witholt, Bio-
technol. Bioeng. 1996, 52, 301–308; d) A. Bosetti, J. B.
Van Beilen, H. Preusting, R. G. Lageveen, B. Witholt,
Enzyme Microb. Technol. 1992, 14, 702–708.
[28] a) B. G. Fox, Y. Liu, J. E. Dege, J. D. Lipscomb, J. Biol.
Chem. 1991, 266, 540–550; b) W. A. Froland, K. K. An-
dersson, S. K. Lee, Y. Liu, J. D. Lipscomb, J. Biol.
Chem. 1992, 267, 17588–17597.
[11] J. B. van Beilen, W. A. Duetz, A. Schmid, B. Witholt,
Trends Biotechnol. 2003, 21, 170–177.
[12] a) S. N. Daff, S. K. Chapman, K. L. Turner, R. A. Holt,
S. Govindaraj, T. L. Poulos, A. W. Munro, Biochemistry
1997, 36, 13816–13823; b) A. W. Munro, D. G. Leys,
K. J. McLean, K. R. Marshall, T. W. B. Ost, S. Daff,
C. S. Miles, S. K. Chapman, D. A. Lysek, C. C. Moser,
C. C. Page, P. L. Dutton, Trends Biochem. Sci. 2002, 27,
250–257; c) Y. Miura, A. J. Fulco, Biochim. Biophys.
Acta 1975, 388, 305–317.
[29] A. Perry, L. Y. Lian, N. S. Scrutton, Biochem. J. 2001,
354, 89–98.
[13] M. A. Noble, C. S. Miles, S. K. Chapman, D. A. Lysek,
A. C. Mackay, G. A. Reid, R. P. Hanzlik, A. W. Munro,
Biochem. J. 1999, 339, 371–379.
[30] B. Bꢀhler, J. B. Park, L. M. Blank, A. Schmid, Appl.
Environ. Microbiol. 2008, 74, 1436–1446.
[31] J. B. van Beilen, T. H. M. Smits, F. F. Roos, T. Brunner,
S. B. Balada, M. Rothlisberger, B. Witholt, J. Bacteriol.
2005, 187, 85–91.
[32] K. Lee, J. Bacteriol. 1999, 181, 2719–2725.
[33] a) L. Leive, Ann. N. Y. Acad. Sci. 1974, 235, 109–129;
b) H. Nikaido, M. Vaara, Microbiol. Rev. 1985, 49, 1–
32.
[34] S. D. Doig, H. Simpson, V. Alphand, R. Furstoss, J. M.
Woodley, Enzyme Microb. Technol. 2003, 32, 347–355.
[35] J. B. Park, B. Bꢀhler, T. Habicher, B. Hauer, S. Panke,
B. Witholt, A. Schmid, Biotechnol. Bioeng. 2006, 95,
501–512.
[36] B. Bꢀhler, B. Witholt, B. Hauer, A. Schmid, Appl. En-
viron. Microbiol. 2002, 68, 560–568.
[37] D. Meyer, B. Witholt, A. Schmid, Appl. Environ. Mi-
crobiol. 2005, 71, 6624–6632.
[38] M. Suzuki, T. Hayakawa, J. P. Shaw, M. Rekik, S. Har-
ayama, J. Bacteriol. 1991, 173, 1690–1695.
[14] a) P. P. Ho, A. J. Fulco, Biochim. Biophys. Acta 1976,
431, 249–256; b) L. O. Narhi, A. J. Fulco, J. Biol. Chem.
1986, 261, 7160–7169; c) N. Shirane, Z. H. Sui, J. A. Pe-
terson, P. R. O. Demontellano, Biochemistry 1993, 32,
13732–13741; d) A. J. Warman, O. Roitel, R. Neeli,
H. M. Girvan, H. E. Seward, S. A. Murray, K. J.
McLean, M. G. Joyce, H. Toogood, R. A. Holt, D.
Leys, N. S. Scrutton, A. W. Munro, Biochem. Soc.
Trans. 2005, 33, 747–753.
[15] a) C. F. Oliver, S. Modi, M. J. Sutcliffe, W. U. Primrose,
L. Y. Lian, G. C. K. Roberts, Biochemistry 1997, 36,
1567–1572; b) P. Meinhold, M. W. Peters, A. Hartwick,
A. R. Hernandez, F. H. Arnold, Adv. Synth. Catal.
2006, 348, 763–772; c) J. C. Lewis, F. H. Arnold, Chimia
2009, 63, 309–312; d) A. B. Carmichael, L. L. Wong,
Eur. J. Biochem. 2001, 268, 3117–3125.
[16] a) S. Picataggio, T. Rohrer, K. Deanda, D. Lanning, R.
Reynolds, J. Mielenz, L. D. Eirich, Bio-Technology
1992, 10, 894–898; b) W. H. Lu, J. E. Ness, W. C. Xie,
X. Y. Zhang, J. Minshull, R. A. Gross, J. Am. Chem.
Soc. 2010, 132, 15451–15455.
[17] a) J. M. Lebeault, E. T. Lode, M. J. Coon, Biochem.
Biophys. Res. Commun. 1971, 42, 413–419; b) D. San-
glard, O. Kꢂppeli, A. Fiechter, J. Bacteriol. 1984, 157,
297–302.
[39] B. Bꢀhler, A. Schmid, B. Hauer, B. Witholt, J. Biol.
Chem. 2000, 275, 10085–10092.
[40] a) P. Goswami, H. D. Singh, Biotechnol. Bioeng. 1991,
37, 1–11; b) M. Rosenberg, E. Rosenberg, J. Bacteriol.
1981, 148, 51–57.
[41] J. Sambrook, D. W. Russell, Molecular cloning – A lab-
oratory manual, Cold Spring Harbor Laboratory Press,
Cold Spring Harbor (NY), 2001.
[18] M. Kusunose, M. J. Coon, E. Kusunose, J. Biol. Chem.
1964, 239, 1374–1380.
[19] G. Eggink, R. G. Lageveen, B. Altenburg, B. Witholt, J.
[42] S. Panke, V. de Lorenzo, A. Kaiser, B. Witholt, M. G.
Wubbolts, Appl. Environ. Microbiol. 1999, 65, 5619–
5623.
Biol. Chem. 1987, 262, 17712–17718.
[20] E. J. McKenna, M. J. Coon, J. Biol. Chem. 1970, 245,
3882–3889.
[43] B. Bꢀhler, I. Bollhalder, B. Hauer, B. Witholt, A.
Schmid, Biotechnol. Bioeng. 2003, 81, 683–694.
[44] D. Hanahan, J. Mol. Biol. 1983, 166, 557–580.
[45] B. J. Bachmann, Derivatisations and genotypes of some
mutant derivatives of Escherichia coli K-12, in: Escheri-
chia coli and Salmonella typhimurium – Cellular and
molecualar biology, (Ed.: C. F. Neidhardt), American
Society for Microbiology, Washington, DC, Vol. 2,
1987, pp 1190–1219.
[21] a) S. Benson, M. Fennewald, J. Shapiro, C. Huettner, J.
Bacteriol. 1977, 132, 614–621; b) J. A. Peterson, D.
Basu, M. J. Coon, J. Biol. Chem. 1966, 241, 5162–5164.
[22] J. B. van Beilen, J. Kingma, B. Witholt, Enzyme
Microb. Technol. 1994, 16, 904–911.
[23] A. Schmid, J. S. Dordick, B. Hauer, A. Kiener, M.
Wubbolts, B. Witholt, Nature 2001, 409, 258–268.
[24] O. Favre-Bulle, B. Witholt, Enzyme Microb. Technol.
1992, 14, 931–937.
[46] T. H. M. Smits, M. A. Seeger, B. Witholt, J. B. van Bei-
len, Plasmid 2001, 46, 16–24.
[47] H. van Heerikhuizen, E. Kwak, E. F. J. van Bruggen, B.
Witholt, Biochim. Biophys. Acta 1975, 413, 177–191.
[48] M. Nieboer, J. Kingma, B. Witholt, Mol. Microbiol.
1993, 8, 1039–1051.
[25] J. B. van Beilen, G. Eggink, H. Enequist, R. Bos, B.
Witholt, Mol. Microbiol. 1992, 6, 3121–3136.
[26] C. Grant, J. M. Woodley, F. Baganz, Enzyme Microb.
Technol. 2011, 48, 480–486.
[49] U. K. Laemmli, Nature 1970, 227, 680–685.
Adv. Synth. Catal. 2011, 353, 3485 – 3495
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3495