Journal of Biotechnology 191 (2014) 8692

Contents lists available at ScienceDirect

Journal of Biotechnology
journal homepage: www.elsevier.com/locate/jbiotec

Biooxidation of n-butane to 1-butanol by engineered P450

monooxygenase under increased pressure
Bernd A. Nebel a , Daniel Scheps a , Sumire Honda Malca a , Bettina M. Nestl a ,
Michael Breuer b , Hans-Gnter Wagner b , Boris Breitscheidel b ,
Detlef Kratz b , Bernhard Hauer a,
a
b

Institute of Technical Biochemistry, Universitaet Stuttgart, Allmandring 31, 70569 Stuttgart, Germany1
BASF SE, 67056 Ludwigshafen, Germany

a r t i c l e

i n f o

Article history:
Received 8 April 2014
Received in revised form 6 August 2014
Accepted 19 August 2014
Available online 26 August 2014
Dedicated to the 60th birthday of Prof.
Jaeger.
Keywords:
CYP153A
Fusion protein
Butane hydroxylation
1-Butanol
High pressure bioconversion

a b s t r a c t
In addition to the traditional 1-butanol production by hydroformylation of gaseous propene and by fermentation of biomass, the cytochrome P450-catalyzed direct terminal oxidation of n-butane into the
primary alcohol 1-butanol constitutes an alternative route to provide the high demand of this basic
chemical. Moreover the use of n-butane offers an unexploited ubiquitous feed stock available in large
quantities. Based on protein engineering of CYP153A from Polaromonas sp. JS666 and the improvement
of the native redox system, a highly -regioselective (>96%) fusion protein variant (CYP153AP.sp.(G254A) CPRBM3 ) for the conversion of n-butane into 1-butanol was developed. Maximum yield of 3.12 g/L butanol,
of which 2.99 g/L comprise for 1-butanol, has been obtained after 20 h reaction time. Due to the poor solubility of n-butane in an aqueous system, a high pressure reaction assembly was applied to increase the
conversion. After optimization a maximum product content of 4.35 g/L 1-butanol from a total amount
of 4.53 g/L butanol catalyzed by the self-sufcient fusion monooxygenase has been obtained at 15 bar
pressure. In comparison to the CYP153A wild type the 1-butanol concentration was enhanced vefold
using the engineered monooxygenase whole cell system by using the high-pressure reaction assembly.
2014 Elsevier B.V. All rights reserved.

1. Introduction
1-Butanol is a versatile basic chemical with a worldwide
production of more than 2.8 million tons per year and used for
various applications from components in plasticizers to solvents
for paints, coating and varnishes (Green, 2011). The major chemical process for the production of butanol is the oxo-synthesis
of propene (hydroformylation) (Weissermel and Arpe, 2003). In
recent years 1-butanol has been considered as an ideal biofuel
alternative. In this light, new fermentation processes based on
sugar, starch and cellulose are investigated as a second alternative
route. Especially the acetone-butanol-ethanol (ABE) fermentation
by Clostridium acetobutylicum represents an additional option for
butanol production (Green, 2011). Nevertheless the strict anaerobic behavior of Clostridia and their slow growth rates, as well as
the limited sustainability and restraints inherent to the microbe,

Corresponding author.
E-mail address: bernhard.hauer@uni-stuttgart.de (B. Hauer).
1
Tel.: +49 685 63193.
http://dx.doi.org/10.1016/j.jbiotec.2014.08.022
0168-1656/ 2014 Elsevier B.V. All rights reserved.

such as sporulation, are problematic for industrial fermentations

(Durre, 2008, 2011; Lutke-Eversloh and Bahl, 2011; Papoutsakis,
2008). Recently a new innovative attempt has been reported
for the production of 1-butanol in metabolically engineered
Escherichia coli (E. coli) strains comprising a set of genes involved
in the biosynthesis of metabolic pathways to produce up to 1.2 g/L
1-butanol (Atsumi et al., 2008a, 2008b; Dellomonaco et al., 2011).
Hence the selective -hydroxylation of gaseous n-butane constitutes a potential biotechnological alternative for the synthesis
of 1-butanol. The use of gaseous substrates in biocatalysis, like nbutane, which remain untapped as energy and chemical feedstock
is a relatively young research area. In literature various in vivo biotransformations to convert n-butane into 1-butanol are reported.
Several soil and marine microorganisms such as Pseudomonas
and Mycobacteria have naturally evolved metal-enzymes that
degrade alkanes into alcohols using gaseous aliphatic alkanes as
carbon and energy source (van Beilen and Funhoff, 2007). These
microbes contain oxygenase enzymes that can introduce oxygen
into the terminal, non-activated position of different aliphatic substrates. Alkane -hydroxylases are oxygenases, whose activities
depend on the chain-length of the hydrocarbon. In general, short

B.A. Nebel et al. / Journal of Biotechnology 191 (2014) 8692

chain alkanes (C1 C4 ) are hydroxylated by methane, propane and

butane monooxygenases (Hamamura et al., 1999). Medium-chain
alkanes (C5 C16 ) are oxidized by integral-membrane non-heme
diiron monooxygenases (alkB) or alternatively by heme ironcontaining cytochrome P450 monooxygenases (P450s or CYPs)
(van Beilen and Funhoff, 2005). Non-engineered methane and
propane monooxygenases tend to be specic for alkanes shorter
than C4 , while the soluble butane monooxygenases especially
alkane hydroxylases and cytochrome P450 monooxygenases of
the 153A subfamily (CYP153A) were recently found to naturally
oxidize n-butane (Funhoff et al., 2007; Koch et al., 2009; Scheps
et al., 2011; van Beilen and Funhoff, 2007). CYP153s are bacterial
class I P450s that operate as three-component systems, comprised
of a catalytic unit (P450 domain) and two additional redox proteins, namely an iron-sulfur electron carrier (ferredoxin) and a
FAD-containing reductase (ferredoxin reductase), which are necessary for the transfer of electrons from NAD(P)H to the P450
active site (Hlavica and Lehnerer, 2010; Munro et al., 2007; Urlacher
and Eiben, 2006; Urlacher et al., 2004). One of the rst characterized butane monooxygenases from Thauera butanivorans produced
in vitro 1-butanol with a conversion rate of 40.5 nM/min/mg
(Cooley et al., 2009; Dubbels et al., 2009; Hamamura et al., 1999).
Unlike other alkane monooxygenases, CYP153A and alkB have been
recombinantly expressed with physiological or non-physiological
redox partners in E. coli and Pseudomonas sp. Using the AlkBmonooxygenase wild type enzyme from Pseudomonas putida Gpo1
a yield of 633 M 1-butanol could be obtained. After directed evolution of this protein the yield could be increased up to 1586 M
1-butanol (Johnson and Hyman, 2006; Koch et al., 2009). A further example for the conversion of n-butane is reported by the
CYP153A6 monooxygenase from Mycobacterium sp. HXN-1500
and its mutant CYP153A6BMO1 producing 277 M and 393 M 1butanol, respectively (Funhoff et al., 2006, 2007; Koch et al., 2009;
van Beilen and Funhoff, 2007).
We recently reported -hydroxylations of medium chain
alkanes, primary alcohols and fatty acids by CYP153 enzymes
from Mycobacterium marinum (CYP153A16), Marinobacter aquaeolei VT8 and Polaromonas sp. (Honda Malca et al., 2012; Scheps
et al., 2011). CYP153A16 and CYP153A from Polaromonas sp. were
screened toward C5 C12 alkanes and C6 C12 primary alcohols
using CamA and CamB from Pseudomonas putida as electron transfer partners within a cofactor regenerating enzyme system. The
alkane substrate range of the biocatalyst oscillated between C5
and C12 , with pentane and hexane being the preferred alkane substrates. They were converted to the corresponding primary alcohols
with low activity but absolute -regioselectivity. Furthermore, in
CYP153A from Marinobacter aquaeolei VT8, a glycine at position
307, which corresponds to position 254 in Polaromonas sp., could be
reported as mutation hot spot to change substrate specicity and to
increase the activity for terminal hydroxylation reactions against noctane alkane, 1-heptanol as well as short chain fatty acids (C8 C9 )
(Mestres, 2005; Seifert et al., 2009).
Aiming to increase the efciency of electron transfer, we herein
fused CYP153A from Polaromonas sp. JS666 (CYP153AP.sp. ) to the
reductase domain of P450-BM3 from Bacillus megaterium (CPRBM3 ).
Oxygenation reactions of gaseous n-butane to 1-butanol with the
fusion protein and variants thereof were carried out in E. coli whole
cells. We also optimized our biotransformation assembly in terms
of gas ow rate and pressure. At normal pressure n-butane is poorly
soluble in water (61 mg/L at 20 C) and as a consequence, it is difcult to supply large amounts of this gaseous substrate in aqueous
medium at ambient conditions (atmospheric pressure and room
temperature) to saturate the enzyme. In order to overcome this
limitation we performed transformations using a pressurized system to increase the butane substrate concentration in the reaction
medium, which is a common strategy in chemical synthesis.

87

2. Materials and methods

All chemicals, solvents and buffer components were obtained
from Sigma-Aldrich (Schnelldorf, Germany). All gases for the experiments were obtained from Air Liquide and were at least 99.5% pure.
Pfu DNA polymerase, endonucleases, T4 DNA ligase and isopropyl
-d-thiogalactopyranoside (IPTG) were purchased from Fermentas
(St. Leon-Rot, Germany).
2.1. Strains and plasmids
Plasmids and strains used in this study are listed in Table 1.
2.2. Genetic manipulation
The wild type CYP153AP.sp. (Bpro 5301) and the corresponding
redox system with a FAD-dependent oxidoreductase (Bpro 530)
and a ferredoxin (Bpro 299) from Polaromonas sp. strain JS666
ATCC BAA-500 was introduced into the NdeI and HindIII cloning
sites of the pET-28a(+) vector. The operon was amplied by PCR
using oligonucleotides 5 -GGT CAT ATG AGA TCA TTA ATG AGT
GAA GCG ATT GTG GTA AAC AAC C-3 and 5 -AGCT AAG CTT TCA
GTG CTG GCC GAG CGG-3 . The fusion (CYP153AP.sp. -CPRBM3 ) and
mutant (CYP153AP.sp.(G254A) -CPRBM3 ) fusion chimeras were created as pET28a(+) vector constructs. The enzymatic part for the
CYP153AP.sp. -CPRBM3 constructs was created by PCR amplication
with oligonucleotide 5 -GGT GCT AGC ATG AGT GAA GCG ATT GTG
GTA AAC AAC CAA AAC G-3 and 3 -TAG CAG ACT GTT CAG TGC
TAG GTG AAG GAA TAG CGT TGA TGC GGA CGG GCA GCG ACT
CAT AGC-5 for the construct with the original linker sequence. The
CYP153A-CPRBM3 redox system was amplied by PCR with oligonucleotides 5 -CAC TGA ACA GTC TGC TAA AAA AGT ACG CAA AAA
GGC AGA AAA CGC TCA TAA TAC GCC GCT GC-3 and 3 -CAT CTC
GAG TTA CCC AGC CCA CAC GTC TTT TGC GTA TC-5 . In a following step the matching amplied products were assembled at their
described overlapping sections by PCR and ligated into pET-28a(+).
The ligated plasmid was used to transform competent E. coli DH5
cells via heat shock. Successful cloning was veried by automated
DNA-sequencing (GATC-Biotech, Konstanz, Germany).
Site-directed mutagenesis using the plasmid pET28a(+) harboring CYP153AP.sp. -CPR was mutated using the QuikChange standard
protocol. Chemical competent E. coli HMS174 (DE3) cells were
used for transformation reactions with the DpnI-treated QC-PCR
mixtures. Variant G254A was created by PCR amplication with
oligonucleotide 5 -CTG GGC AAC CTC ATT TTG CTG ATC GTC GCG
GGC AAT GAC ACG ACC CGC-3 and its complementary sequence 3 GCG GGT CGT GTC ATT GCC CGC GAC GAT CAG CAA AAT GAG GTT
GCC CAG-5 , which contains the codon GCG covering the amino
acid alanine. Competent E. coli DH5 cells were transformed with
the DpnI-treated PCR mixtures. Isolated plasmids with the desired
mutations (sequencing by GATC-Biotech, Konstanz, Germany) were
used to transform in competent E. coli HMS174 (DE3) cells.
2.3. Cultivation of biomass
The cell cultivation and protein expression for the three
CYP153A constructs ITB379, ITB413 and ITB408 was conducted
as described below. E. coli BL21(DE3) cells expressing CYP153A
variants were pre-cultured in LB medium at 37 C with shaking at
250 rpm for 24 h. 1 mM of the non-gaseous n-hexane was added
to optimize the adaption toward alkanes and to increase product
formation (Koch et al., 2009; van Beilen et al., 2006). Cultivations
of E. coli BL21(DE3) whole cells for bioconversions were carried
out in 1 L Erlenmeyer shake asks containing 200 mL terrifc broth
(TB) media supplemented with the appropriate antibiotic (1 mL/L
kanamycin solution, 50 g/mL). Cultivation of E. coli was generally

88

B.A. Nebel et al. / Journal of Biotechnology 191 (2014) 8692

Table 1
Host strains and plasmid construct used in this work.
Construct abbreviation

Strain or plasmid

Characteristics

Reference

E. coli strains
ITB379
ITB413
ITB408

BL21 (DE3)
HMS174 (DE3)
HMS174 (DE3)

pET28a(+):CYP153A-operon from Polaromonas sp.

pET28a(+):CYP153A-CPRBM3 from Polaromonas sp.
pET28a(+):CYP153A(G254A) -CPRBM3 from Polaromonas sp.

This study
This study
This study

pET28a(+)

Expression vector

Novagen, Wisconsin, USA

Plasmid

performed aerobically in orbital shakers (Multitron, Infors HT,

Bottmingen, Switzerland) at 180 rpm and 37 C.
The growth was carried out on a shaker to an OD600 of
1.11.3. Expression was induced by the addition of 0.25 mM IPTG.
The culture was supplemented with 4 g/L glycerol, 0.5 mM 5aminolevulinic acid (-ALA) and 100 mg FeSO4 in E. coli. The cells
were incubated for 24 h at 28 C and 180 rpm and harvested by a
centrifugation step at 4.000 g and 4 C for 30 min. Due to variations in the expression level of the different CYP153A variants,
23 independent cultures were prepared to assure a high enzyme
concentration. The pellets were washed with 100 mM potassium
phosphate buffer (pH 7.4). After this procedure the cells were concentrated into 100 mM potassium phosphate buffer (pH 7.5) to a
nal concentration of 50 gcww /L buffer (18.7 gcdw /L).

n-butane and 95.8 vol.% synthetic air. The n-butane gas concentration was settled at 4.2 vol.% because of safety reasons (explosion
range 1.49.4 vol.% n-butane). The total gas ow rate was adjusted
to 12.5 L/h by using mass ow units. Butane/air gas supply into
the cell slurry was guaranteed through a continuous ow rate and
the use of a sparger after mixing in a dispenser nozzle (Fig. 1). To
minimize product loss, a back ow gas collection system was used.
The collection asks were lled with 150 mL ddH2 O. After dened
time points, samples from the bioreactor ask and the collection
asks, which were installed downstream of the fermentation ask
to avoid product discharge, were taken and after a fast and tight
sealing procedure analyzed by GC/MS-headspace chromatography.
For the quantication of the formed C4 -product the amounts of 1butanol and 2-butanol in the bioconversion asks and downstream
asks were combined.

2.4. Biotransformation
After the cells were provided with 1% glycerol (v/v) and 20 mM
glucose as carbon source, the gaseous substrate was added to the
reaction mixture as illustrated in Fig. 1.
2.4.1. In vivo biotransformation of n-butane to butanol
All biotransformations were carried out with resting cells (E. coli
with CYP153A, 50 gcww /L) in 100 mM potassium phosphate buffer
pH 7.5. For bioconversions of gaseous n-butane, 100 mL of cell suspension and 15 L of antifoam 204 (SigmaAldrich) were stirred
in a 250 mL Schott-ask at room temperature. n-Butane was added
to the reaction mix with following an inlet gas ratio of 4.2 vol.%

2.4.2. In vivo biotransformation of n-butane to butanol under

increased pressure
The terminal hydroxylation of the gaseous C4-substrate was
investigated in a high pressure reactor. Into the 250 mL glass insert
100 mL cell suspension was provided and a magnetic stirring bar
was added. 10 g of cold liquid n-butane was lled in excess which
resulted in the formation of a second phase at a temperature of
5 C. Afterwards the pressure tank (Carl Roth, high-pressure autoclave II) was sealed and connected via a high pressure line to a
synthetic air gas cylinder. This setup offers the opportunity to apply
a selected pressure of 1100 bar (nal pressure 120 bar). This step
ensures also the supply of sufcient oxygen for the reaction. The

Fig. 1. Schematic illustration of the atmospheric pressure continuous gas ow assembly. A 250 mL reaction vessel was equipped with a magnetic stirrer, a gas inlet with
sparger unit and a gas outlet connected in serial with two 250 mL gas collection vessels. The two collection aks were lled with water and installed downstream to avoid
product lost (only one gas collection ask is illustrated). The inlet gas ow was controlled by a master ow and two mass ow units (n-butane and synthetic air) to adjust
the ow and gas composition. The gas mix was settled at 4.2 vol% n-butane to avoid an explosive atmosphere. The assembly was installed in a fume hood because of safety
reasons.

B.A. Nebel et al. / Journal of Biotechnology 191 (2014) 8692

compression step at the beginning of each reaction was made as

slowly as possible to avoid physical stress. For each time point a
new reaction setup was necessary because the used high pressure
assembly did not allow continuous sampling.
2.5. Analysis
2.5.1. P450 activity assay
Concentrations of the P450 enzymes were determined by the
carbon monoxide (CO) differential spectral assay, based on the
formation of the characteristic FeII -CO complex at 448 nm. The
cells were disrupted by sonication on ice (4 2 min, 2 min intervals). Enzymes in cell-free extracts were reduced by the addition
of 10 mM sodium dithionite from a freshly prepared 1 M stock
solution, and the carbon monoxide complex was formed by slow
bubbling with CO gas for approximately 30 s. The concentrations
were calculated using the absorbance difference at A450 and A490
(Ultrospec 3100pro spectrophotometer, Amersham Biosciences)
and an extinction coefcient of 91 mM1 cm1 (Omura and Sato,
1964).
2.5.2. Headspace GC/MS analysis
To overcome time consuming extraction procedures a GC/MSheadspace method for 1-butanol products analysis was established.
Samples were analyzed by using a GC/MS QP-2010 instrument (Shimadzu, Japan) equipped with a FS-Supreme-5-column
(30 m 0.25 mm 0.25 m) in combination with a CombiPal Sampler operated in headspace mode and a 2.5 mL heated headspace
syringe. Helium as carrier gas (ow rate 0.69 mL/min) was applied
with a split-ratio of 15:1. Electron impact (EI) ionization was used
and the mass range from 20 to 200 m/z was detected. Both detector
interface and injector temperature were set at 250 C. One milliliter
of the biotransformation mix was lled into a 20 mL headspace
vial. To overcome detector saturation, the biotransformation culture was diluted with the appropriate buffer medium. 100 L of the
internal standard solution (10 mM 1-hexanol) was added and the
headspace vials were sealed tightly. Temperature program: start
40 C, hold 5 min, 5 C/min to 85 C, hold 1 min, 60 C/min to 300 C.
For quantication of 1-butanol, the system was calibrated with the
internal standard 1-hexanol. Standard samples with different concentrations (0.012 mM of 1-butanol and 2-butanol) in 100 mM
potassium phosphate buffer were measured via GC/MS.
3. Results and discussion
3.1. Engineering of biocatalysts
With the goal of evaluating an alternative route to produce
1-butanol, we started the engineering of the catalyst based on

89

the knowledge from directed evolution strategies by the group

of F. H. Arnold on CYP153A6BMO1 (Koch et al., 2009). Following
these results of the CYP153A6BMO1 and its ability to -oxidize
alkanes, cycloalkenes and alicyclic compounds (Funhoff et al.,
2006), by using a homology-based approach we were able to
identify CYP153AP.sp. possessing a valine at position 95, which is
equivalent to alanine residue 97 in CYP153A6BMO1 . The increased
activity of CYP153A6BMO1 toward n-butane gas has been attributed
to a single amino acid substitution Ala97Val (Koch et al., 2009). With
regard to the high amino acid sequence similarity (85%) between
CYP153AP.sp. and CYP153A6 and the preference of CYP153AP.sp. for
alkanes over primary alcohol and fatty acids (Honda Malca et al.,
2012), the biocatalyst CYP153AP.sp. was therefore considered to be
promising starting candidate for the hydroxylation of n-butane to
1-butanol.
In biocatalytic reactions with oxygenases whole cells are preferred over isolated enzymes primarily because cells are capable
of regenerating NAD(P)H. Cells may also confer higher oxygenase stability by providing a protected compartment or a better
organization of its components. Native CYP153A enzymes depend
on a ferredoxin reductase and a ferredoxin to obtain reducing
equivalents from NAD(P)H. Low expression levels of the separate
redox proteins or inability to interact in equimolar ratios with
the CYP might impair whole cell biotransformations using these
enzymes. Although most P450 and electron-transferring proteins
are independent proteins encoded by different genes, there are
self-sufcient monooxygenases that contain both P450 and P450
reductase in a single polypeptide chain (Munro et al., 2007). The
electron transfer from the redox complex to the catalytic domain
is mostly slow and one of the rate-limiting aspects in many CYP
systems (Guengerich, 2002). The improvement of electron delivery
to the heme-domain is important for the efcient conversion of
substrate to product by cytochrome P450 enzymes (OReilly et al.,
2011; Robin et al., 2009). A typical example of such a self-sufcient
P450 monooxygenase is the long-chain fatty acid hydroxylase
P450-BM3 from Bacillus megaterium (Davis et al., 1996; Gustafsson
et al., 2004; Narhi and Fulco, 1986, 1987). This monooxygenase consists of the P450 (CYP102A1) and P450 reductase domain and shows
extremely high molecular turnover of >1500 min1 (Munro et al.,
1997, 2002). To address the problems of low catalytic efciency,
inefcient electron transfer and complicated inner cellular arrangement of three single proteins, a functional self-sufcient CYP153A
fusion complex was created and further engineered (Robin et al.,
2009; Scheps et al., 2013).
We designed and characterized a fusion construct of the hemedomain from CYP153AP.sp. from Polaromonas sp. (CYP153AP.sp. CPRBM3 , ITB413) containing a C-terminally-linked CPR reductase
domain of P450-BM3 (Fig. 2). Furthermore, a previous study has
shown that a single point mutation of glycine at position 307

Fig. 2. Establishing an optimized self-sufcient fusion complex comprising mutation G254A for increased activity (CYP153AP.sp.(G254A) -CPRBM3 ). Construction of a functional
self-sufcient fusion protein with a CPRBM3 redox protein from the CYP153AP.sp. natural multiple protein complex. Abbreviations: CPR: cytochrome P450 reductase (FAD/FMNcontaining reductase domain) from Bacillus megaterium, FdR and Fdx: natural FAD-containing oxidoreductase and [2Fe-2S] ferredoxin of CYP153A enzyme from Polaromonas
sp.

90

B.A. Nebel et al. / Journal of Biotechnology 191 (2014) 8692

Fig. 3. Production and product release of 1-butanol in the reaction vessel by the CYP153AP.sp. natural multiple protein complex (ITB379) into reaction vessel (dotted line)
and accumulation of 1-butanol in the gas collection ask #1 and #2 (dashed line and hashed line) during process. The 1-butanol production (mM butanol formation) was
determined by GC-HS/MS. The reactions were performed in the presence of 100 mL cell suspension (50 gcww /L, 15 L antifoam 204, 100 mM potassium phosphate buffer
pH 7.5) over 24 h reaction time. The biotransformation was done in a continuous reactor system assembled with two washing asks to avoid product discharge (total ow
12.5 L/h, 4.2 vol% n-butane and 95.8 vol% synthetic air).

in CYP153A from Marinobacter aquaeolei, which corresponds to

position 254 in Polaromonas sp. caused not only higher in vitro
-hydroxylation activity but also improved regioselectivity for
the terminal position (Chen et al., 2012; Honda Malca et al.,
2012). To reinforce the lessons learned, we created the mutant
CYP153AP.sp.(G254A) -CPRBM3 (ITB408) by leveraging the knowledge
gained by the fusion enzyme CYP153AP.sp. -CPRBM3 (ITB413). P450
CO difference spectral analyses showed that cell extracts of all
CYP153AP.sp. constructs (ITB379, ITB413 and ITB408; 0.2 gcww /mL)
yielded in soluble and active enzymes between 2.8 and 3.1 M. This
indicates that all cytochrome P450 monooxygenase systems were
functionally expressed in similar yields.
Furthermore, previous efforts to engineer CYP153 genes for
alkane hydroxylation resulted in a host that has been adapted
through prolonged cultivation on alkanes to obtain signicant
growth on these substrates without any mutations occurring in the
CYP153 genes themselves (Koch et al., 2009; van Beilen et al., 2006).
Based on the results of Koch et al. and van Beilen et al. n-hexane was
used during the growth phase potentially enabling enrichment for
activity on alkanes and adaption of the production host to substrate
and product as small as butane and butanol.
3.2. In vivo biotransformation
In order to exploit the ability of the E. coli host system to produce 1-butanol with the heterologously expressed CYP153AP.sp.
enzymes including ITB379, ITB413 and ITB408, bioconversions at
atmospheric pressure and continuous gas ow for 20 h were performed. Total butanol concentration was enhanced by improving
the biotransformation assembly through the increase of the inlet
gas ow rate and aeration. The total gas ow was adjusted at
12.5 L/h. The maximum product concentration was observed at
12.5 L/h and a butaneair ratio of 4.2 vol.% to 95.8 vol.%.
With implementation of gas collection asks into the reaction
assembly a product release out of the reaction vessel could be determined. Fig. 3 shows the 1-butanol release out of the reaction vessel
at a total air ow of 12 L/h at room temperature (dotted line). Over a
time scale of 24 h the 1-butanol concentration in the reaction vessel
decreased and a signicant accumulation of the compound in collection ask one and two could be observed (dashed line and hashed
line). Based on these results and to overcome product lost all experiments were performed with two gas collection asks. Furthermore
the quantication the product concentration in the bioconversion
asks and downstream asks were combined.

A further positive aspect of the continuous product removal out

of the reaction vessel is the enhanced 1-butanol production by the
prevention of cell damage and cell death due to an accumulation
of polar products in the cell membrane (Isken and de Bont, 1998;
Steen et al., 2008). Also the addition of a glycerol/glucose mixture,
reported to have a benecial effect on cell function and nicotinamide cofactor regeneration, was investigated (Gudiminchi et al.,
2012). Due to the fact that glycerol is known to be a driving force
for cofactor regeneration in whole cell-mediated redox biocatalysts (Blank et al., 2008), media containing either 0.050.3% glucose,
0.52% glycerol or a mixture of glucose/glycerol were tested. In the
absence of glycerol or glucose, butanol concentrations less than
0.5 mM were detected. A mixture of 20 mM glucose and 1% glycerol
was determined to be the most efcient carbon source concentration for butanol production (data not shown).
Under the improved conditions we observed that the optimized
mutant monooxygenase enzyme system produced a maximum
of 42 mM butanol (3.12 g butanol/L using 50 gcww resting cells)
under atmospheric pressure after 20 h (Table 2). Due to the high
-regioselectivity of 96% of the mutant, a product titer of 40 mM
1-butanol and 2 mM 2-butanol has been obtained. In comparison to the wild type system, the concentration of butanol using
the mutant enzyme system could by increased vefold. All threeenzyme systems were tested under the same reaction conditions
at ve different time points (Fig. 4). On the basis of the described
stepwise improvement of the wild type a signicant increase of
butanol production by the self-sufcient fusion complex and a
further optimized product concentration by the mutated multiple
domain enzyme system could be achieved.
Moreover, the engineered monooxygenase systems are very
selective and products do not suffer from overoxidation. No
oxidation to butanal or butanoic acid and further reaction to 1,4butanediol was detected.
3.3. Pressurized in vivo biotransformation
The solubility of n-butane gas in water or aqueous media is
rather low (61 mg/L at 20 C) and thus, constitutes a critical parameter for a biocatalytic process. In an attempt to enhance substrate
availability, we performed additional in vivo biotransformations
using the wild type under different pressure conditions (120 bar).
The best product concentrations were obtained at a pressure of
15 bar, experiments carried out at more than 15 bar caused a
decrease in total butanol production (Fig. 5).

B.A. Nebel et al. / Journal of Biotechnology 191 (2014) 8692

91

Table 2
Product formation after bioconversion of n-butane in CYP153A resting cells.

Total butanol titer [mM/50 gcww /20 h]*

Final 1-butanol titer [g/L]**
Initial product rate [mM/h]**
Specic activity [mg/h/M P450]
Space time yield [g/L/h]

ITB379

ITB413

ITB408

ITB379 at 15 bar

ITB408 at 15 bar

10.2 1.8
0.72 0.13
1.4
1.37
0.04 0.007

27.9 5.2
1.98 0.38
4.3
4.22
0.10 0.019

42.2 6.3
2.99 0.47
6.0
5.88
0.15 0.023

19.8 2.2
1.41 0.16
2.9
2.84
0.06 0.008

61.2 5.4
4.35 0.40
8.2
8.04
0.23 0.020

No oxidation to butanal or butanoic acid and further reaction to 1,4-butanediol was detected by GCHS/MS.
The total butanol (1-butanol and 2-butanol) and nal 1-butanol titer was determined after 20 h reaction time directly by GCHS/MS and the initial product rate was
calculated within the rst four hours of the hydroxylation of n-butane using wild type ITB379, fusion construct ITB413 and mutant variant ITB408.
**

product was measurable) without further oxidation and reaction

of the butanol product.
This tendency reects that the increased solubility of the substrate n-butane at higher pressure is an important factor to increase
the product concentration. The lower butanol concentration at
higher-pressure conditions (>15 bar) might be explained by deactivation of the enzyme system due to physical stress of pressurizing
the system. Further experiments using even more rened techniques for examination will be necessary to take a closer look at
the deactivation of the biocatalytic system.

70

total butanol [mM]

60
50
40
30
20
10
0
4

12

16

4. Conclusion

me [h]

Fig. 4. Comparison of total butanol titer with different CYP153AP.sp . constructs (ITB379, ITB413 and ITB408) at atmospheric pressure and the mutant
CYP153AP.sp.(G254A)- CPRBM3 (ITB408) under 15 bar pressure over 20 h reaction time.
The total butanol production (mM butanol formation) was determined by GCHS/MS. The whole-cell bioconversions were carried out in the presence of 100 mL
cell suspension (50 gcww /L, 15 L antifoam 204, 100 mM potassium phosphate buffer
pH 7.5). The biotransformation was done in a continuous reactor system (total
ow 12.5 L/h, 4.2 vol% n-butane and 95.8 vol% synthetic air) and a high pressure
reactor assembly. The values represent means standard deviation of triplicate
experiments.

total buttanol [mM]

20

15

10

5
0

10
pressure [bar]

15

20

Fig. 5. Pressure-dependency of butane oxidation with CYP153AP.sp. wild type

(ITB379). The total butanol concentration (mM butanol formation) was determined
by GC-HS/MS. The reactions were performed at the indicated pressure in the presence of 100 mL cell suspension (50 gcww /L, 100 mM potassium phosphate buffer pH
7.5) and 10 g of cold liquid n-butane performed in a high pressure reactor system.
The pressure was adjusted between 1 and 20 bar, by connecting the high pressure
unit to a synthetic air gas cylinder. For each pressure point a new reaction setup was
necessary because due to the fact that the used high pressure assembly did not allow
continuous sampling. The values represent means standard deviation of triplicate
experiments.

Moreover, the bioconversion of n-butane catalyzed by mutant

was increased from 42.2 mM (1 bar) to 61.2 mM (15 bar) after 20 h
reaction time using 50 gcww resting cells. This results in a maximum
concentration of 4.35 g/L 1-butanol at 15 bar which reects a 96%
selectivity of the overall process using the monooxygenase mutant
enzymes (Table 2). These results represent a raise in initial product
rate by a factor of 2.8 in a time course of 14 h (linear increase of

To conclude, we have shown that the utilization of enzyme engineering strategies as well as the implementation of a high pressure
reaction system resulted in improvements of the overall whole
cell activity and nal 1-butanol concentration by the combination
of a self-sufcient fusion complex and a domain-based directed
evolution strategy. In comparison to the CYP153AP.sp. monooxygenase wild type and the previous published CYP153A6-BMO1
system by Koch et al., an increase of activity of the production of
1-butanol, from 0.72 and 0.86 g/L, respectively, to 2.99 g/L has been
obtained by the fusion protein and its mutant. Moreover, investigations in the use of a pressure-reactor system up to 20 bar have
been carried out in order to overcome the substrate limitation of
n-butane in an aqueous system. These process engineering strategies led to an additional increase of the enzymatic regioselective
production of butanol up to 4.53 g/L which results in a yield of
4.35 g/L 1-butanol at 15 bar. The high regioselectivity of >96% of
the optimized CYP153AP.sp.(G254A) -CPRBM3 fusion protein (ITB408)
could be maintained during the entire experimental process. To
run an economical feasible process a nal 1-butanol concentration
more than 50 g/L as well as a space time yield higher than 3 g/L/h
are required. To reach the goals our preliminary studies indicate
that there is the opportunity to dene various options for improving the performance of the system. The activity of the enzyme
could be increased by applying different reductase units and the
improvement of the linker system (OReilly et al., 2011; Robin
et al., 2009). The optimization of the C-source during biotransformation, the increase of the cell stability and the implementation of
a continuous pressure system are topics waiting for more detailed
studies.
In summary the nal 1-butanol concentration could be
increased vefold by using the optimized self-fusion construct and
the implantation of the high pressure reactor system in comparison
to the wild type system at atmospheric pressure.

Acknowledgements
BASF SE, Germany is acknowledged for nancial support. Edgar
Magin, BASF SE for technical assistance in setting up the high pressure reactor.

92

B.A. Nebel et al. / Journal of Biotechnology 191 (2014) 8692

References
Atsumi, S., Cann, A.F., Connor, M.R., Shen, C.R., Smith, K.M., Brynildsen, M.P., Chou,
K.J., Hanai, T., Liao, J.C., 2008a. Metabolic engineering of Escherichia coli for 1butanol production. Metab. Eng. 10, 305311.
Atsumi, S., Hanai, T., Liao, J.C., 2008b. Non-fermentative pathways for synthesis of
branched-chain higher alcohols as biofuels. Nature 451, 8689.
Blank, L.M., Ebert, B.E., Buhler, B., Schmid, A., 2008. Metabolic capacity estimation of
Escherichia coli as a platform for redox biocatalysis: constraint-based modeling
and experimental verication. Biotechnol. Bioeng. 100, 10501065.
Chen, M.M.Y., Snow, C.D., Vizcarra, C.L., Mayo, S.L., Arnold, F.H., 2012. Comparison of random mutagenesis and semi-rational designed libraries for improved
cytochrome P450 BM3-catalyzed hydroxylation of small alkanes. Protein Eng.
Des. Sel. 25, 171178.
Cooley, R.B., Dubbels, B.L., Sayavedra-Soto, L.A., Bottomley, P.J., Arp, D.J.,
2009. Kinetic characterization of the soluble butane monooxygenase from
Thauera butanivorans, formerly Pseudomonas butanovora. Microbiology 155,
20862096.
Davis, S.C., Sui, Z., Peterson, J.A., Ortiz de Montellano, P.R., 1996. Oxidation of omegaoxo fatty acids by cytochrome P450BM-3 (CYP102). Arch. Biochem. Biophys. 328,
3542.
Dellomonaco, C., Clomburg, J.M., Miller, E.N., Gonzalez, R., 2011. Engineered reversal
of the beta-oxidation cycle for the synthesis of fuels and chemicals. Nature 476,
355359.
Dubbels, B.L., Sayavedra-Soto, L.A., Bottomley, P.J., Arp, D.J., 2009. Thauera butanivorans sp. nov., a C2C9 alkane-oxidizing bacterium previously referred to as
Pseudomonas butanovora. Int. J. Syst. Evol. Microbiol. 59, 15761578.
Durre, P., 2008. Fermentative butanol production: bulk chemical and biofuel. Ann.
N.Y. Acad. Sci. 1125, 353362.
Durre, P., 2011. Fermentative production of butanol the academic perspective.
Curr. Opin. Biotechnol. 22, 331336.
Funhoff, E.G., Bauer, U., Garcia-Rubio, I., Witholt, B., van Beilen, J.B., 2006. CYP153A6,
a soluble P450 oxygenase catalyzing terminal-alkane hydroxylation. J. Bacteriol.
188, 52205227.
Funhoff, E.G., Salzmann, J., Bauer, U., Witholt, B., van Beilen, J.B., 2007. Hydroxylation and epoxidation reactions catalyzed by CYP153 enzymes. Enzyme Microb.
Technol. 40, 806812.
Green, E.M., 2011. Fermentative production of butanol the industrial perspective.
Curr. Opin. Biotechnol. 22, 337343.
Gudiminchi, R.K., Randall, C., Opperman, D.J., Olaofe, O.A., Harrison, S.T., Albertyn,
J., Smit, M.S., 2012. Whole-cell hydroxylation of n-octane by Escherichia coli
strains expressing the CYP153A6 operon. Appl. Microbiol. Biotechnol. 96 (6),
15071516.
Guengerich, F.P., 2002. Rate-limiting steps in cytochrome P 450 catalysis. Biol. Chem.
383, 15531564.
Gustafsson, M.C.U., Roitel, O., Marshall, K.R., Noble, M.A., Chapman, S.K., Pessegueiro,
A., Fulco, A.J., Cheesman, M.R., von Wachenfeldt, C., Munro, A.W., 2004. Expression, purication, and characterization of Bacillus subtilis cytochromes P450
CYP102A2 and CYP102A3: avocytochrome homologues of P450 BM3 from
Bacillus megaterium. Biochemistry 43, 54745487.
Hamamura, N., Storfa, R.T., Semprini, L., Arp, D.J., 1999. Diversity in butane
monooxygenases among butane-grown bacteria. Appl. Environ. Microbiol. 65,
45864593.
Hlavica, P., Lehnerer, M., 2010. Oxidative biotransformation of fatty acids by
cytochromes P450: predicted key structural elements orchestrating substrate
specicity, regioselectivity and catalytic efciency. Curr. Drug Metab. 11,
85104.
Honda Malca, S., Scheps, D., Kuhnel, L., Venegas-Venegas, E., Seifert, A., Nestl, B.M.,
Hauer, B., 2012. Bacterial CYP153A monooxygenases for the synthesis of omegahydroxylated fatty acids. Chem. Commun. 42, 51155117.
Isken, S., de Bont, J.A., 1998. Bacteria tolerant to organic solvents. Extremophiles 2,
229238.

Johnson, E.L., Hyman, M.R., 2006. Propane and n-butane oxidation by Pseudomonas
putida GPo1. Appl. Environ. Microbiol. 72, 950952.
Koch, D.J., Chen, M.M., van Beilen, J.B., Arnold, F.H., 2009. In vivo evolution of butane
oxidation by terminal alkane hydroxylases AlkB and CYP153A6. Appl. Environ.
Microbiol. 75, 337344.
Lutke-Eversloh, T., Bahl, H., 2011. Metabolic engineering of Clostridium acetobutylicum: recent advances to improve butanol production. Curr. Opin.
Biotechnol. 22 (5), 634667.
Mestres, J., 2005. Structure conservation in cytochromes P450. Proteins 58, 596609.
Munro, A.W., Daff, S.N., Turner, K.L., Chapman, S.K., 1997. Redox characterisation of
avocytochrome P-450 BM3 from Bacillus megaterium. Biochem. Soc. Trans. 25,
S628.
Munro, A.W., Leys, D.G., McLean, K.J., Marshall, K.R., Ost, T.W., Daff, S., Miles, C.S.,
Chapman, S.K., Lysek, D.A., Moser, C.C., Page, C.C., Dutton, P.L., 2002. P450 BM3:
the very model of a modern avocytochrome. Trends Biochem. Sci. 27, 250257.
Munro, A.W., Girvan, H.M., McLean, K.J., 2007. Cytochrome P450 redox partner
fusion enzymes. Biochim. Biophys. Acta 1770, 345359.
Narhi, L.O., Fulco, A.J., 1986. Characterization of a catalytically self-sufcient
119,000-dalton cytochrome P-450 monooxygenase induced by barbiturates in
Bacillus megaterium. J. Biol. Chem. 261, 71607169.
Narhi, L.O., Fulco, A.J., 1987. Identication and characterization of two functional
domains in cytochrome P-450BM-3, a catalytically self-sufcient monooxygenase induced by barbiturates in Bacillus megaterium. J. Biol. Chem. 262,
66836690.
Omura, T., Sato, R., 1964. The carbon monoxide-binding pigment of liver microsomes. II. Solubilization, purication, and properties. J. Biol. Chem. 239,
23792385.
OReilly, E., Kohler, V., Flitsch, S.L., Turner, N.J., 2011. Cytochromes P 450 as useful
biocatalysts: addressing the limitations. Chem. Commun. 47, 24902501.
Papoutsakis, E.T., 2008. Engineering solventogenic clostridia. Curr. Opin. Biotechnol.
19, 420429.
Robin, A., Roberts, G.A., Kisch, J., Sabbadin, F., Grogan, G., Bruce, N., Turner, N.J.,
Flitsch, S.L., 2009. Engineering and improvement of the efciency of a chimeric
[P450cam-RhFRed reductase domain] enzyme. Chem. Commun., 24782480.
Scheps, D., Malca, S.H., Hoffmann, H., Nestl, B.M., Hauer, B., 2011. Regioselective omega-hydroxylation of medium-chain n-alkanes and primary alcohols
by CYP153 enzymes from Mycobacterium marinum and Polaromonas sp. strain
JS666. Org. Biomol. Chem. 9, 67276733.
Scheps, D., Honda Malca, S., Richter, S.M., Marisch, K., Nestl, B.M., Hauer, B., 2013.
Synthesis of -hydroxy dodecanoic acid based on an engineered CYP153A fusion
construct. Microbial. Biotechnol. 6, 694707.
Seifert, A., Vomund, S., Grohmann, K., Kriening, S., Urlacher, V.B., Laschat, S., Pleiss,
J., 2009. Rational design of a minimal and highly enriched CYP102A1 mutant
library with improved regio-, stereo- and chemoselectivity. ChemBioChem 10,
853861.
Steen, E.J., Chan, R., Prasad, N., Myers, S., Petzold, C.J., Redding, A., Ouellet, M.,
Keasling, J.D., 2008. Metabolic engineering of Saccharomyces cerevisiae for the
production of n-butanol. Microb. Cell Fact. 7, 36.
Urlacher, V.B., Eiben, S., 2006. Cytochrome P450 monooxygenases: perspectives for
synthetic application. Trends Biotechnol. 24, 324330.
Urlacher, V.B., Lutz-Wahl, S., Schmid, R.D., 2004. Microbial P450 enzymes in biotechnology. Appl. Microbiol. Biotechnol. 64, 317325.
van Beilen, J.B., Funhoff, E.G., 2005. Expanding the alkane oxygenase toolbox: new
enzymes and applications. Curr. Opin. Biotechnol. 16, 308314.
van Beilen, J.B., Funhoff, E.G., 2007. Alkane hydroxylases involved in microbial alkane
degradation. Appl. Microbiol. Biotechnol. 74, 1321.
van Beilen, J.B., Funhoff, E.G., van, L.A., Just, A., Kaysser, L., Bouza, M., Holtackers,
R., Rothlisberger, M., Li, Z., Witholt, B., 2006. Cytochrome P450 alkane hydroxylases of the CYP153 family are common in alkane-degrading eubacteria lacking
integral membrane alkane hydroxylases. Appl. Environ. Microbiol. 72, 5965.
Weissermel, K., Arpe, H.P., 2003. Industrial Organic Chemistry. Wiley-VCH, Weinheim.