Cloning, expression and characterisation of P450-

Hal1 (CYP116B62) from Halomonas sp. NCIMB

172: A self-sufficient P450 with high expression

and diverse substrate scope

Joanne L. Porter, Selina Sabatini, Jack Manning, Michele Tavanti, James L. Galman, Nicholas

J. Turner and Sabine L. Flitsch*

School of Chemistry, Manchester Institute of Biotechnology, The University of Manchester,

UK

*Corresponding author email: sabine.flitsch@manchester.ac.uk

Abstract

Cytochrome P450 monooxygenases are able to catalyse a range of synthetically challenging

reactions ranging from hydroxylation and demethylation to sulfoxidation and epoxidation. As

such they have great potential for biocatalytic applications but are underutilised due to often-

poor expression, stability and solubility in recombinant bacterial hosts. The use of self-

sufficient P450s with fused haem and reductase domains has already contributed heavily to

improving catalytic efficiency and simplifying an otherwise more complex multi-component

system of P450 and redox partners. Herein, we present a new addition to the class VII family

with the cloning, sequencing and characterisation of the self-sufficient CYP116B62 Hal1 from

Halomonas sp. NCIMB 172, the genome of which has not yet been sequenced. Hal1 exhibits

high levels of expression in a recombinant E. coli host and can be utilised from cell lysate or

used in purified form. Hal1 favours NADPH as electron donor and displays a diverse range of

activities including hydroxylation, demethylation and sulfoxidation. These properties make

Hal1 suitable for future biocatalytic applications or as a template for optimisation through

engineering.

Keywords Biocatalysis, Cytochrome P450 monooxygenase, Halomonas sp., C-H activation,

demethylation, CYP116B

Abbreviations

ALA Aminolevulinic acid

API Active pharmaceutical ingredient

BM3 CYP102A1 from Bacillus megaterium

FAD Flavin adenine dinucleotide

FMN Flavin mononucleotide

GDH Glucose dehydrogenase

Hal1 CYP116B62 from Halomonas sp. NCIMB 172

HalC CYP116B45 from Halomonas campaniensis (uncharacterised)

IMAC Immobilized metal ion affinity chromatography

IPTG β-D-1-thiogalactopyranoside

NADPH Nicotinamide adenine dinucleotide phosphate

P450 Cytochrome P450 monooxygenase

ROS Reactive oxygen species

1. Introduction

Cytochrome P450 monooxygenases (P450s or CYPs) are haem-containing enzymes that are

widespread in nature where they play a vital role in the oxyfunctionalisation of a vast variety of

physiologically important compounds including fatty acids, steroids, vitamins and xenobiotics

[1]. P450s are highly attractive biocatalysts due to their ability to catalyse the regio- and

stereoselective oxidation of a wide range of substrates, ranging from simple alkanes to highly

functionalised APIs [2, 3]. In the presence of suitable redox partners, P450s are able to shuttle

electrons supplied by NAD(P)H to enable the activation of molecular oxygen, with a single

oxygen atom introduced into the substrate and the other concomitantly reduced to water

(Scheme 1A). This initial hydroxylation can also enable subsequent reactions including alcohol

oxidation, N- and O-dealkylation (Scheme 1B), isomerisation, C-C bond cleavage, desaturation

and ring formations and expansions [4-6].

While this impressive catalytic repertoire gives P450s huge potential and desirability as green

alternatives to traditional organo- and metallocatalysts [7-9], there are key factors that continue

to limit their application in these non-physiological roles [10]. P450s are hampered by low

expression and solubility, narrow substrate scope, poor stability and expensive cofactor and

redox requirements [2, 11].

Scheme 1. (A) General P450 reaction scheme and (B) subsequent O-demethylation following

initial hydroxylation

The use of catalytically self-sufficient P450s (most commonly class VII and class VIII

enzymes) (for a detailed review of P450 systems and classification see Hannemann et. al. 2007

[12]) with the haem and reductase domains contained on a single polypeptide chain, addresses

some of these limitations by removing the necessity for additional redox partners, thereby

reducing costs associated with enzyme production and drastically simplifying system

development. The best studied of these enzymes is the fatty acid hydroxylase P450-BM3

(CYP102A1) [13], a class VIII P450 with an FMN and FAD containing reductase domain [14].

Although engineering studies have significantly expanded the substrate scope of BM3 [15-18],

the substrate scope of wild-type CYP102A enzymes is typically poor with limited promiscuity

beyond their native fatty acid substrates. Class VII enzymes such as P450-RhF [19-21] and

homologues [22-25], comprising a haem domain fused to a FMN and 2Fe-2S ferredoxin

containing reductase domain, have been shown to exhibit a more varied substrate scope [26,

27]. To combine this catalytic self-sufficiency with thermal stability, we have recently

characterised a novel panel of self-sufficient P450s from thermophilic microorganisms [28].

The panel of class VII CYP116B P450s were shown to possess significant thermostability

across the entire multi domain enzymes, with P450-TT from T. thermophilus displaying a half-

life of over 9 h at 50 °C. Although this work contributes heavily to alleviating stability issues

associated with the use of P450 biocatalysts, further work is required to make more of these

biocatalysts routinely available for an even wider range of applications.

Nature possesses a huge array of enzymes that have evolved over millions of years to adapt to

the needs of their respective host organisms. This includes organisms that have adapted to

typically unfavourable environmental factors such as extreme temperature (e.g. psychrophiles

or thermophiles), pH (e.g. acidophiles or alkaliphiles) or high salt concentrations (e.g.

halophiles). Enzymes from extremophiles, which have already evolved to function under

extreme conditions, could also be better equipped to perform under the various conditions

demanded from biocatalytic processes [29].

Halophilic microorganisms thrive in highly saline environments (including salt lakes, mines,

and marine environments) and are found in all three domains of life: Archaea, Bacteria and

Eucarya [30]. Such highly saline environments often require their inhabitant microorganisms

to survive at elevated temperatures, alkaline pH or with low oxygen availability [31]. The

unique properties and biodiversity of salt-tolerant organisms are being exploited for (current

and potential) biotechnological applications including treatment of saline wastewaters,

production of biopolymers and biodegradation of alkanes and aromatic hydrocarbons [32, 33].

Enzymes sourced from halophilic bacteria also have huge potential but as yet have been

underutilised.

Here we present the discovery and characterisation of a new class VII P450 (Hal1) from the

halophilic bacterium Halomonas sp. NCIMB 172, the genome of which is yet to be sequenced.

Cloning, expression and characterisation of the new P450-Hal1 shows high levels of soluble

expression and wide substrate scope, both of which are important features for a P450

biocatalyst and address some of the current limitations of this technology.

2. Materials and Methods

2.1 Chemicals

All chemicals were purchased from Sigma-Aldrich unless otherwise specified. Solvents for

HPLC analysis were purchased from Honeywell and were of chromasolv grade or from Romil

and were of SpS grade purity. Competent cells were purchased from New England Biolabs and

gases were purchased from BOC.

2.2 Identification and cloning

A BLAST search was conducted using the P450-RhF (CYP116B2) amino acid sequence as the

query sequence on the UniProt website (www.uniprot.org/blast/). Genomic DNA was

purchased from DSMZ (DSM-15293) and was used as the template sequence in a typical PCR

with complementary primers. The primers were designed with suitable 5’ overhangs for In-

fusion® cloning (Clonetech) into the pET28a expression vector between the BamHI and SalI

restriction sites (Table S1). Primer synthesis and DNA sequencing was carried out by Eurofins

Genomics.

2.3 Expression and purification

Small-scale preliminary expression trials were conducted as specified in the supporting

information (Page S9). For larger scale expression a starter culture of LB medium (8 mL)

supplemented with kanamycin was inoculated with a single colony of Escherichia coli

BL21(DE3) transformed with the pET28a/Hal1 plasmid and grown overnight. The starter

culture was used to inoculate the M9 expression medium (800 mL in a 2 L baffled flask),

supplemented as outlined in the preliminary expression trials. Cultures were grown at 37 °C to

an OD600 of ~ 0.8 then cooled to the expression temperature of 20 °C and supplemented with 5-

aminolevulinic acid (5-ALA) then induced with β-D-1-thiogalactopyranoside (IPTG) (0.4

mM). Protein expression was carried out for 20 h with shaking at 200 rpm.

Protein purification was carried out by immobilized metal ion affinity chromatography (IMAC)

using Ni-agarose resin (Qiagen) in a gravity flow column, with all steps performed at < 7 °C.

Cells were harvested by centrifugation at 2831 g for 15 min (Avanti J-E centrifuge, Beckman

Coulter) then resuspended in 0.1 M KPi, 0.4 % glycerol pH 8 and lysed by ultrasonication

(Bandelin Sonopuls). Cell debris was removed by centrifugation (48384 g, 40 min) and the

clarified supernatant was incubated with the column resin for 60 min. The protein bound resin

was loaded into a gravity flow column and washed sequentially with 5 (5 CV) and 20 mM (10

CV) imidazole then P450-Hal1 was eluted with 250 mM imidazole.

The spectral absorption properties of Hal1 were determined on a CARY 50 UV/Visible spectrophotometer (Agilent Technologies). The P450 concentration was determined by CO difference spectroscopy using an extinction coefficient of 91 mM-1 cm-1 following an established protocol [34].

2.4 Buffer optimisation

Optimisation of the assay buffer was investigated by monitoring the initial rate of the Hal1

catalysed demethylation of 7-methoxycoumarin (7). Formation of 7-hydroxycoumarin was

monitored by fluorescence spectroscopy using a microtitre platereader (Tecan Infinite M200

Pro) with λex = 397 nm and λem = 466 nm. Reactions were carried out using 0.5 mM NADPH, 1

mM 7-Methoxycoumarin, 2 % (v/v) DMSO and 1 μM purified enzyme in a total reaction

volume of 200 μL with varied buffer composition. Buffer type (KPi, Borate, Tris, Tricine, BIS-

TRIS Propane, pH (7 - 10) and salt concentration (0 - 250 mM NaCl or KCl) were varied

across several experiments. Data was obtained in triplicate and is expressed as a percentage

relative to the initial rate using a starting buffer of 0.1 M KPi pH 8 and the error is the standard

deviation.

2.5 Half-life, thermal and chemical stability assays

Thermal stability assays were carried out by heating aliquots of Hal1 in the appropriate buffer

(0.1 M KPi pH 8 or 0.1 M tricine pH 8.5) at 30 °C for the half-life experiments or at a range of

temperatures for 15 min to measure residual activity with increasing temperature. The enzyme

activity toward 7 following heat treatment was measured as described above. Chemical

stability was assessed by incubation of Hal1 in the presence of 5 or 10 % (v/v) co-solvent, then

measuring residual enzyme activity in the same manner. All reactions were performed in

triplicate and the error is the standard deviation.

2.6 Cytochrome c assay

Cofactor preference was determined using a typical P450 reductase assay monitored using

cytochrome c as the external electron acceptor [35]. Reactions were performed in 0.1 M KPi

pH 8 or 0.1 M tricine pH 8.5, with 10 nm enzyme and varied cofactor concentrations (1 - 100

μM for NADPH and 0.1 - 3 mM for NADH). The initial rate of reduction of cytochrome c was

monitored at 550 nm on a Cary 300 UV-visible spectrophotometer (Varian) with temperature

control at 25 °C. Kinetic parameters were determined using an extinction coefficient of 26.4

mM-1 cm-1 and fitting the data to the Michaelis-Menten equation using SciDAVis 1.17.

Reactions were performed in triplicate and the error is the standard deviation.

2.7 Substrate screening

The biotransformations depicted in Figure 5 were performed using cell lysates with a

concentration equivalent to 200 mg mL-1 wet cells (40 mg mL-1 dry cells). Expression was

performed as described above with cells resuspended in 0.1 M KPi pH 8 then lysed and used

directly with a final Hal1 concentration of 10.6 μM. Reactions were carried out using a

cofactor recycling system consisting of GDH (1 mg mL-1), D-glucose (10 mg mL-1) and

NADP+ (1 mg ml-1). Substrates were dissolved in DMSO and used at 1 mM with a final DMSO

concentration of 2 % (v/v). Reactions were performed at 20 °C, 250 rpm for 18 h then analysed

by either LC-MS (1-3, 5-8 and 10) or GC-FID (4, 9 and 11-13) (Section 1.7, Supporting

Information).

3. Results and Discussion

3.1 Discovery and cloning of the self-sufficient P450-Hal1

A blast search was performed using P450-RhF as the query sequence to identify suitable

homologues from halophilic organisms. A class VII self-sufficient P450 (HalC) (UniProt ID:

A0A060BBY3) with 56.7 % sequence identity to P450-RhF was identified from Halomonas

campaniensis (DSMZ 15293) and was selected as the target sequence for cloning from

genomic DNA. In parallel to cloning HalC, the same set of primers were used for colony PCR

of thirteen alternative Halomonas strains available to us from our in-house culture collection

(purchased from NCIMB) (Table 1).

Table 1Additional Halomonas strains screened from the NCIMB culture collection.

NCIMB culture collection no. Strain

172 Halomonas sp.

557 Halomonas aquamarina

2054 Halomonas marina

13962 Halomonas hydrothermalis

13958 Halomonas meridiana

13960 Halomonas axialensis

1550 Halomonas sp.

13959 Halomonas sulfidaeris

13957 Halomonas neptunia

1980 Halomonas aquamarina

1978 Halomonas cupida

13961 Halomonas meridiana

2116 Halomonas aquamarina

While amplification of the HalC gene was unsuccessful, several of the other screened

Halomonas strains gave amplified PCR products (Fig. S1). The amplified products were

cloned into pET28a with DNA sequencing revealing the successful amplification and sub-

cloning of a self-sufficient class VII P450 (Hal1, GenBank accession no. MG386934) from

Halomonas sp. NCIMB 172. The amino acid sequence of the newly discovered Hal1

(CYP116B62) has been deposited on the Prokaryotic Cytochrome P450 database (p450.co); a

useful resource for P450 sequence information and associated literature.

P450-Hal1 shares 96 % amino acid sequence identity to the original cloning target HalC

(sequence alignment Fig. S3, Supporting Information) and 65 to 52 % sequence identity to

other characterised members of the class VII group (Fig. 1 and Fig. S4 and S5, Supporting

Information) [22-25].

Fig. 1. Phylogenetic tree with the newly cloned Hal1 and the previously reported Class VII

P450s and the Class VIII BM3. The evolutionary analysis was performed in MEGA7 with the

sequence alignment done using ClustalW and the evolutionary history was inferred using the

neighbour-joining method. The evolutionary distances were computed using the Poisson

correction method and are in units of the number of amino acid substitutions per site.

3.2 Expression and purification of P450-Hal1

Preliminary expression trails of P450-Hal1 were conducted with differing temperatures (37 and

24 °C), media (M9 and LB with IPTG induction and auto induction LB) and expression time (4

and 20 h). SDS-PAGE analysis revealed better expression was achieved at lower temperatures

and with IPTG induction over the use of auto induction media (Fig. S6).

Following the initial trials, larger scale expression (800 mL in 2 L baffled flask) was conducted

in M9 media with IPTG induction at 20 °C for 20 h. High levels of Hal1 overexpression were

achieved giving a P450 content of 335 nmolP450 g-1cdw in corresponding cell lysate. This value is

higher than achieved under the same expression conditions for P450-AT, AX, JT, TB and TT

[28] and is in the range achieved for RhF and BM3 on a larger scale with optimised nutrient

feeding strategies [36, 37] (Table 2).

Table 2Comparison of expression levels of Hal1 with other self-sufficient P450s

P450 Conditions P450 content (nmolP450 g-1

cdw) Reference

Hal1 (CYP116B62) 2 L shake flask 335

AT (CYP116B65) 2 L shake flask 150 [28]

AX (CYP116B64) 2 L shake flask 151 [28]

JT (CYP116B63) 2 L shake flask 61 [28]

TB (CYP116B29) 2 L shake flask 144 [28]

TT (CYP116B46) 2 L shake flask 315 [28]

RhF (CYP116B2)2 L shake flask 56 [28]

1 L fermenter with oxygen and nutrient feeding strategy 316 [36]

BM3 (CYP102A1) 5 L fermenter with batch feeding strategy 483 [37]

Hal1 with a cleavable fused N-terminal polyhistidine tag was purified by immobilised metal

ion affinity chromatography (IMAC) following sonication of resuspended whole cells. SDS-

PAGE analysis showed soluble overexpression of the correct size construct (92.6 kDa) and

subsequent purification of the full-length enzyme (Fig. S7).

3.3 UV-visible absorption properties of P450-Hal1

The UV-visible spectral properties were determined for the newly sequenced Hal1 and were

characteristic of a typical P450 haemoprotein. (Fig. S8). The spectrum of oxidised Hal1

displayed a main Soret band with absorption maximum at 420 nm and the α and β bands

positioned at 569 and 536 nm. Reduction with sodium dithionite caused the main Soret band to

weaken in intensity and shift to 417 nm. Subsequent addition of CO to the reduced sample

caused the expected shift in absorption, in this case to 452 nm, upon formation of the Fe(II)-

CO complex. Only a slight shoulder was observed at ~ 429 nm representing the inactive P420

form.

3.4 Effect of buffer, pH and salt on the activity of P450-Hal1

The influence of pH, buffer type and salt on the activity of Hal1 was investigated using 7-

methoxycoumarin (7) as substrate. A pH profile for Hal1 was generated using three different

buffers (phosphate buffer pH 7 – 8, TRIS pH 7.5 – 9 and borate buffer pH 8 - 10) across a

combined pH range of 7 – 10 (Fig. 2A). The pH optimum was found to be 8.5, with the

enzyme showing little activity at less than pH 7.5 or at higher than pH 9. The use of different

buffers at pH 8.5 influenced the initial rate of activity with TRIS giving the lowest rate

(although still double the activity obtained in 0.1 M KPi pH 8) (Fig. 2B). Of the buffers tested,

Tricine and BIS-TRIS Propane at a buffering strength of 0.1 M, gave the best results with

increases in activity of over 3-fold when compared to 0.1 M KPi pH 8.

Fig. 2. Activity of P450-Hal1 monitoring the effect of (A) pH and buffer (at strength 50 mM)

and (B) buffer type and strength at pH 8.5. Activity was measured towards 7-methoxycoumarin

and expressed as a percentage in comparison to the activity displayed by the enzyme in the

original buffer (0.1 M KPi pH 8).

The effect of salt on the activity of Hal1 was assessed given that its native host, Halomonas sp.

NCIMB 172, was isolated from skate skin in the North Sea [38]. Enzyme activity was

measured in the presence of low to moderate salt concentrations (0 – 250 mM NaCl or KCl)

(Fig. S9). The addition of very low concentrations (< 50 mM) of sodium chloride or potassium

chloride salt had little effect on the activity of Hal1, however overall enzyme activity decreased

with further increasing salt concentration. The same trend was observed with both salts (NaCl

and KCl) and with either potassium phosphate pH 8 or Tricine pH 8.5 used as the assay buffer.

3.5 Nucleotide cofactor preference

Steady-state kinetic studies were conducted to determine the nucleotide cofactor preference of

P450-Hal1. The Hal1 catalysed reduction of cytochrome c, used as an external electron

acceptor, was monitored in 0.1 M KPi pH 8 using either NADH or NADPH as the electron

donor (Table 3). Our experiments revealed that P450-Hal1 is bifunctional accepting both

NAD(P)H cofactors. However, a clear preference was observed toward NADPH with a kcat/KM

value of 4.9 ± 0.9 x 106 compared to 2.4 ± 0.2 x 104 M-1 s-1 when using NADH as electron

donor. This preference toward NADPH is consistent with other self-sufficient class VII (RhF

homologues with FMN-FeS redox system) and class VIII (BM3 homologues with FMN-FAD

redox system) enzymes all of which favour NADPH.

Kinetic parameters for cytochrome c reduction by Hal1 were also calculated with NADPH

using 0.1 M tricine as the reaction buffer. The alternative reaction buffer gave a 3-fold increase

in catalytic rate with a kcat value of 69 ± 3 compared to 22 ± 1 s-1 obtained using 0.1 M KPi pH

8. However, a simultaneous increase in KM gave a lower overall kcat/Km value of 3.1 ± 0.6 x 106

M-1 s-1. The higher electron transport rate in tricine buffer is consistent with the results of the

pH screen, which was assessed based on initial product formation rates (Fig. 2A).

Table 3Kinetic properties of Hal1 using NADPH and NADH as reductants and cytochrome c as

electron acceptor in 0.1 M KPi buffer pH 8 and using NADPH in 0.1 M tricine pH 8.5.

0.1 M KPi pH 8 0.1 M Tricine pH 8.5

NADPH NADHCofactor Preference

Ratio kcat/KM

(NADPH/NADH)NADPH

KM (μM) 4.7 ± 0.8 700 ± 16199

23 ± 5

kcat (s-1) 22 ± 1 17 ± 2 69 ± 3

kcat/KM (M-1 s-1) 4.9 ± 0.9 x 106 2.4 ± 0.2 x 104 3.1 ± 0.6 x 106

3.6 Stability of P450-Hal1

The stability of Hal1 was assessed primarily by measurement of residual enzyme activity over

time with incubation at either 4 or 30 °C with a comparison between 0.1 M KPi pH 8 and 0.1

M tricine pH 8.5 (Fig. 3). With potassium phosphate buffer and incubation at 4 °C Hal1

retained activity over the analysis period, while at 30 °C a loss in activity of approximately 30

% was observed over the same 8 h analysis period. In contrast to the relative stability displayed

in phosphate buffer, with the use of tricine buffer at pH 8.5 Hal1 displayed a half-life of 1 h

with incubation at 30 °C and 3 h with incubation at 4 °C. Additionally, residual activity was

monitored following 15 min incubation at increasing temperatures from 25 to 45 °C (Fig. S10,

Supporting Information). Residual activity curves for Hal1 were obtained under four buffer

conditions: 0.1 M KPi pH 8 without and with 50 mM NaCl (Fig. S10A, B) and 0.1 M tricine

pH 8.5 without and with 50 mM NaCl (Fig. S10C, D). The extracted T50 values were

approximately 34 °C and were within error regardless of buffer and the presence of salt (Table

S3).

Fig. 3. Half-life of Hal1 in 0.1 M tricine pH 8.5 at either 30 or 4 °C. Residual activity of Hal1

in 0.1 M KPi at 30 °C is also shown (grey circles) whilst at 4 °C in the same buffer the activity

of the enzyme was retained over the time period.

The comparable thermal stability of Hal1 in either phosphate or tricine buffer compared with

the respective half-life measurements, where the half-life of Hal1 tricine buffer is significantly

reduced, suggests that this effect is not caused by structural instability. A possible explanation

could be negative interference of tricine in the substrate-binding site or during electron

transport. Either scenario would increasingly hinder catalysis over the course of the reaction, as

observed in the half-life measurements. As such for subsequent biotransformations involving

Hal1, which were conducted over an 18 h period and are detailed below (Section 3.7),

potassium phosphate was the preferred buffer.

As well as thermal stability, the chemical stability of Hal1 was also assessed. Residual enzyme

activity was measured following an incubation period in the presence of various co-solvents at

a concentration of 5 or 10 % (v/v) with phosphate buffer (Fig. 4). In the presence of 5 %

methanol, isopropanol, ethylene glycol, acetone and dimethyl sulfoxide (DMSO) Hal1 retained

greater than 90 % activity and approximately 75 % activity in the presence of ethanol and

dimethylformamide (DMF). An increase in co-solvent concentration to 10 % resulted in lower

residual activity except in the case of ethylene glycol and DMSO, the latter of which displayed

increased activity (> 125 %) with both 5 and 10 % co-solvent. Tetrahydrofuran was the least

compatible co-solvent for use in Hal1 reactions with no significant levels of activity detected.

Fig. 4. Effect of co-solvent on the stability of Hal1. Residual activity toward 7-

methoxycoumarin was measured following 30 min incubation in the presence of either 5 or 10

% (v/v) co-solvent in 0.1 M KPi pH 8.5.

Solvent stability measurements were also performed in tricine buffer with the same co-solvents

at 5 or 10 % (v/v) concentrations. These results (Fig. S11, Supporting Information) show

similar trends with DMSO being the preferred co-solvent. The activity of Hal1 in the presence

of these water miscible organic solvents is impressive, especially in comparison to that of the

prototype class VIII P450-BM3. For example, wild-type BM3 retains approximately 70 %

(Hal1 > 125 %) of its original activity with the addition of DMSO and just 20 % (Hal1 > 50 %)

in acetone and 10 % (Hal1 = 40 %) in the presence of ethanol each at a concentration of 10 %

(v/v) [39].

3.7 Substrate profile of P450-Hal1

The substrate scope of P450-Hal1 was investigated using soluble cell lysate (equivalent to the

lysate from 40 mg mL-1 dry cell weight) containing a final P450 concentration of 10.6 μM.

Reactions were conducted for 20 h with NADP+ and using a cofactor recycling system

comprising of D-glucose and glucose dehydrogenase (GDH) to generate the more costly

NADPH cofactor (Fig. 5 and Table S4, Supporting Information). Figure 5 shows the thirteen

substrates evaluated and the corresponding conversions catalysed by Hal1. The main product

could be identified by comparison with an authentic standard, or in the case of

hydroxydecanoic acid from the unique product fragmentation obtained through GC-MS

analysis of the trimethylsilyl derivative (Fig. 5, grey panels). For several substrates there were

also unidentified products, the number and approximate distribution of these have been

indicated (Fig. 5, dark panels) to provide a fingerprint of the products produced by Hal1.

P450-Hal1 catalysed the biotransformation of several methyoxy- compounds to give the O-

demethylated products with the best conversions achieved for 4-methoxyacetophenone (1) (83

%) > 2-hydroxy-4-methoxyacetophenone (2) (71 %) > 4-methoxybenzophenone (5) (40 %) >

4-methoxybenzonitrile (6) (36 %) > 7-methoxycoumarin (7) (35 %) > 5-methoxytetralone (8)

(25 %). Hydroxylation activity was observed for decanoic acid (4) (54 %) and the widely used

anti-inflammatory drug diclofenac (3) with hydroxylation at the 5’- position to give 5’-

hydroxydiclofenac (one of the common human metabolites) with a conversion of 62 %. In

addition sulfoxidation activity was demonstrated with methyl phenyl sulphide (9) to give the

sulphoxide product and trace levels of activity were observed for fluorene (10). No detectable

levels of activity were observed with styrene (11), ethylbenzene (12) or α-isophorone (13) as

substrate.

Fatty acids are the physiological substrates of class VIII CYP102A P450s (e.g. BM3), this is

the first report of activity toward these substrates by a class VII CYP116B enzyme. BM3

preferentially hydroxylates at the sub-terminal ω-1, ω-2 or ω-3 positions depending on

substrate, with decanoic acid (4) the major product is the ω-1 hydroxyacid. Here the Hal1

catalysed hydroxylation of 4 gives the ω-2 hydroxyacid (8-hydroxydecanoic acid) as the major

product (Fig. S17 and S27, Supporting Information).

Fig. 5. Activity profile of Hal1 showing the conversions (%) for various substrates. The panel

to the right of the structures indicates product profiles with identified products highlighted on

the structure and by the grey panel in the profile box where dark panels indicate unidentified

products.

3.8 Investigation of catalytic parameters

Further experiments were conducted using 4-methoxyacetophenone (1) (the substrate for which

the highest conversion was achieved in the initial assessment of substrate scope, Fig 5) to

determine the electron coupling efficiency and product formation rate. Inefficient coupling of

electrons from NADPH to product formation, in this case 4-hydroxyacetophenone, can result

in the formation of reactive oxygen species (ROS). As such, this uncoupling is an important

parameter in P450 catalysed reactions since it can influence the reaction rate and affect the

stability of the entire construct. The coupling efficiency for P450 catalysed reactions is

dependent on both enzyme and substrate. For example, P450-BM3 and homologues have some

of the highest reported coupling efficiencies (>95 %) with their native fatty acid substrates [40-

42]. However, with non-physiological substrates, efficiencies are significantly reduced and as

low as 1% [43-46]. Here the coupling efficiency of the Hal1 catalysed demethylation of 4-

methoxyacetophenone is 15 ± 2 % (11 ± 1 % in 0.1 M tricine buffer, pH 8), which is not

unexpected considering this is not the natural substrate for this enzyme.

Experiments were also conducted to determine product formation rates and assess performance

at reduced biocatalyst concentrations. The product formation rate of the Hal1 catalysed

demethylation of 4-methoxyacetophenone is 1.48 ± 0.02 nmol nmolP450-1 min-1 (determined

using 10 μM enzyme and 1 mM substrate concentration) (Fig S12, Supporting Information).

The total substrate conversion decreased substantially with decreasing enzyme concentration

and in all cases the reaction did not appear to proceed at any significant rate after 4 h (Fig. S13,

Supporting Information). Stability tests already conducted with Hal1 showed that in KPi buffer

pH 8, the enzyme retained > 80 % original activity after a 4 h period (Fig. 3). As such, it is

likely that stability is not the sole contributing factor to the reduced effective reaction times.

Based on the reduced coupling efficiency for this non-physiological substrate, it is probable

that decoupling plays a role in the reduced stability of the process both directly through

formation of ROS (e.g. H2O2) and indirectly through generation of high concentrations of

gluconic acid (a GDH byproduct of NADPH cofactor recycling resulting from spontaneous

hydrolysis of gluconolactone) [47]. This reduced operational stability is consistent with

previous observations of BM3 containing systems [14, 16]. It is postulated that implementation

of an alternative cofactor recycling system or addition of catalase to the reaction medium could

aid in extending the useful reaction time. The presence of catalase to scavenge hydrogen

peroxide has previously been demonstrated to significantly extend enzyme stability especially

when using purified enzyme [48, 49]. However this is not always the case and it is possible that

ROS cause immediate damage to the enzyme and as such, subsequent removal could have

limited effect [50].

4. Conclusion

P450s are synthetically useful enzymes but have yet to reach their potential due to practical

considerations concerning expression, solubility, stability and toxicity. Increasing diversity in

the pool of characterised sequences can only aid in the future utilisation of these P450

biocatalysts. Here we have cloned, sequenced, expressed and characterised P450-Hal1 from

Halomonas sp. NCIMB 172, the genome of which has not yet been sequenced. Hal1 shows

excellent overexpression in a recombinant E. coli host, especially in comparison with other

class VII enzymes. We have shown that Hal1 and can be utilised from lysate or readily purified

by IMAC. The significant effect of buffer, pH and salt on Hal1 activity shows that these factors

need to be carefully evaluated when screening new P450s for biocatalytic purposes.

P450-Hal1 can accept both NADH and NADPH and is capable of catalysing the

biotransformation of a variety of diverse substrates primarily producing the corresponding

demethylated products. In addition to demethylation activity, Hal1 also catalyses the

sulfoxidation of methyl phenyl sulphide and hydroxylation of the anti-inflammatory drug

diclofenac. Interestingly, Hal1 is the first reported class VII P450 with hydroxylation activity

towards decanoic acid. Furthermore Hal1 displays different regioselectivity than that of the

well-characterised fatty acid hydroxylase P450-BM3. Hal1 presents a robust system with high

levels of expression and diverse substrate scope, making it a valuable addition to the P450

biocatalyst toolbox.

Acknowledgements

The authors gratefully acknowledge funding received from the European Union’s Seventh

Framework Programme for research, technological development and demonstration under

grant agreement number 613849 (BIOOX) and from the European Union’s Horizon 2020

Programme for research and innovation actions H2020-LEIT BIO-2014-1 under grant

agreement number 635734 (ROBOX). The authors also gratefully acknowledge funding from

CoEBio3 Phase III.

Any statement made herein reflects only the views of the authors. The European Union is not

liable for any use that may be made of the information contained herein.

Supplementary Information

Supporting data associated with this article can be found online at: xxx

References

[1] D.R. Nelson, A world of cytochrome P450s, Philos. Trans. R. Soc., B 368 (2013) 20120430.

[2] R. Bernhardt, V.B. Urlacher, Cytochromes P450 as promising catalysts for biotechnological application: Chances and limitations, Appl. Microbiol. Biotechnol. 98 (2014) 6185-6203.

[3] V.B. Urlacher, M. Girhard, Cytochrome P450 monooxygenases: An update on perspectives for synthetic application, Trends Biotechnol. 30 (2012) 26-36.

[4] F.P. Guengerich, A.W. Munro, Unusual cytochrome P450 enzymes and reactions, J. Biol. Chem. 288 (2013) 17065-17073.

[5] F.P. Guengerich, Common and Uncommon Cytochrome P450 Reactions Related to Metabolism and Chemical Toxicity, Chem. Res. Toxicol. 14 (2001) 611-650.

[6] J.A. McIntosh, C.C. Farwell, F.H. Arnold, Expanding P450 catalytic reaction space through evolution and engineering, Curr. Opin. Chem. Biol. 19 (2014) 126-134.

[7] H.M. Girvan, A.W. Munro, Applications of microbial cytochrome P450 enzymes in biotechnology and synthetic biology, Curr. Opin. Chem. Biol. 31 (2016).

[8] K.J. McLean, M. Hans, B. Meijrink, W.B. van Scheppingen, A. Vollebregt, K.L. Tee, J.-M. van der Laan, D. Leys, A.W. Munro, M.A. van den Berg, Single-step fermentative production of the cholesterol-lowering drug pravastatin via reprogramming of Penicillium chrysogenum, Proc. Nat. Acad. Sci. U. S. A. 112 (2015) 2847-52.

[9] R. Fasan, Tuning P450 Enzymes as Oxidation Catalysts, ACS Catalysis 2 (2012) 647-666.

[10] M.T. Lundemo, J.M. Woodley, Guidelines for development and implementation of biocatalytic P450 processes, Appl. Microbiol. Biotechnol. 99 (2015).

[11] E. O'Reilly, V. Kohler, S.L. Flitsch, N.J. Turner, Cytochromes P450 as useful biocatalysts: Addressing the limitations, Chem. Commun. 47 (2011) 2490-2501.

[12] F. Hannemann, A. Bichet, K.M. Ewen, R. Bernhardt, Cytochrome P450 systems—biological variations of electron transport chains, Biochim. Biophys. Acta, Gen. Subj. 1770 (2007) 330-344.

[13] L.O. Narhi, A.J. Fulco, Characterization of a catalytically self-sufficient 119,000-dalton cytochrome P-450 monooxygenase induced by barbiturates in Bacillus megaterium, J. Biol. Chem. 261 (1986) 7160-7169.

[14] 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, P450 BM3: the very model of a modern flavocytochrome, Trends Biochem. Sci. 27 (2002) 250-257.

[15] S.D. Munday, O. Shoji, Y. Watanabe, L.-L. Wong, S.G. Bell, Improved oxidation of aromatic and aliphatic hydrocarbons using rate enhancing variants of P450BM3 in combination with decoy molecules, Chem. Commun. 52 (2016) 1036-1039.

[16] R. Fasan, M.M. Chen, N.C. Crook, F.H. Arnold, Engineered Alkane-Hydroxylating Cytochrome P450BM3 Exhibiting Nativelike Catalytic Properties, Angew. Chem. Int. Ed. 46 (2007) 8414-8418.

[17] J.-b. Wang, A. Ilie, M.T. Reetz, Chemo- and Stereoselective Cytochrome P450-BM3-Catalyzed Sulfoxidation of 1-Thiochroman-4-ones Enabled by Directed Evolution, Adv. Synth. Catal. 359 (2017) 2056-2060.

[18] S. Schulz, M. Girhard, S.K. Gassmeyer, V.D. Jaeger, D. Schwarze, A. Vogel, V.B. Urlacher, Selective enzymatic synthesis of the grapefruit flavor (+)-nootkatone, ChemCatChem 7 (2015) 601-604.

[19] G.A. Roberts, G. Grogan, A. Greter, S.L. Flitsch, N.J. Turner, Identification of a new class of cytochrome P450 from a Rhodococcus sp, J. Bacteriol. 184 (2002) 3898-3908.

[20] G.A. Roberts, A. Celik, D.J. Hunter, T.W. Ost, J.H. White, S.K. Chapman, N.J. Turner, S.L. Flitsch, A self-sufficient cytochrome p450 with a primary structural organization that includes a flavin domain and a [2Fe-2S] redox center, J. Biol. Chem. 278 (2003) 48914-20.

[21] D.J. Hunter, G.A. Roberts, T.W. Ost, J.H. White, S. Muller, N.J. Turner, S.L. Flitsch, S.K. Chapman, Analysis of the domain properties of the novel cytochrome P450 RhF, FEBS Lett. 579 (2005) 2215-20.

[22] L. Liu, R.D. Schmid, V.B. Urlacher, Engineering cytochrome P450 monooxygenase CYP 116B3 for high dealkylation activity, Biotechnol. Lett 32 (2010) 841-5.

[23] A.J. Warman, J.W. Robinson, D. Luciakova, A.D. Lawrence, K.R. Marshall, M.J. Warren, M.R. Cheesman, S.E. Rigby, A.W. Munro, K.J. McLean, Characterization of Cupriavidus metallidurans CYP116B1--a thiocarbamate herbicide oxygenating P450-phthalate dioxygenase reductase fusion protein, FEBS J. 279 (2012) 1675-93.

[24] Y.C. Yin, H.L. Yu, Z.J. Luan, R.J. Li, P.F. Ouyang, J. Liu, J.H. Xu, Unusually broad substrate profile of self-sufficient cytochrome P450 monooxygenase CYP116B4 from Labrenzia aggregata, Chembiochem 15 (2014) 2443-9.

[25] D. Minerdi, S.J. Sadeghi, G. Di Nardo, F. Rua, S. Castrignano, P. Allegra, G. Gilardi, CYP116B5: a new class VII catalytically self-sufficient cytochrome P450 from Acinetobacter radioresistens that enables growth on alkanes, Mol. Microbiol. 95 (2015) 539-54.

[26] E. O'Reilly, M. Corbett, S. Hussain, P.P. Kelly, D. Richardson, S.L. Flitsch, N.J. Turner, Substrate promiscuity of cytochrome P450 RhF, Catal. Sci. Technol. 3 (2013) 1490-1492.

[27] A. Celik, G.A. Roberts, J.H. White, S.K. Chapman, N.J. Turner, S.L. Flitsch, Probing the substrate specificity of the catalytically self-sufficient cytochrome P450 RhF from a Rhodococcus sp, Chem. Commun. (2006) 4492-4.

[28] M. Tavanti, J.L. Porter, S. Sabatini, N.J. Turner, S. Flitsch, A panel of new thermostable CYP116B self-sufficient cytochrome P450 monooxygenases catalysing C-H activation with diverse substrate scope, ChemCatChem (2017) DOI:10.1002/cctc.201701510.

[29] J.A. Littlechild, Enzymes from Extreme Environments and Their Industrial Applications, Frontiers in bioeng. and biotechnol. 3 (2015) 161-161.

[30] Y. Ma, E.A. Galinski, W.D. Grant, A. Oren, A. Ventosa, Halophiles 2010: life in saline environments, Appl. Environ. Microbiol. 76 (2010) 6971-81.

[31] A. Ventosa, J.J. Nieto, Biotechnological applications and potentialities of halophilic microorganisms, World J. Microbiol. Biotechnol. 11 (1995) 85-94.

[32] B.Z. Fathepure, Recent studies in microbial degradation of petroleum hydrocarbons in hypersaline environments, Frontiers in Microbiol. 5 (2014) 173.

[33] E. Acikgoz, B. Ozcan, Phenol biodegradation by halophilic archaea, Int. Biodeterior. Biodegrad. 107 (2016) 140-146.

[34] T. Omura, R. Sato, The Carbon Monoxide-binding pigment of Liver Microsomes, J. Biol. Chem. 239 (1964) 2370-2378.

[35] F.P. Guengerich, M.V. Martin, C.D. Sohl, Q. Cheng, Measurement of cytochrome P450 and NADPH–cytochrome P450 reductase, Nat. protoc. 4 (2009) 1245-1251.

[36] J.M. Klenk, B.A. Nebel, J.L. Porter, J.K. Kulig, S.A. Hussain, S.M. Richter, M. Tavanti, N.J. Turner, M.A. Hayes, B. Hauer, S.L. Flitsch, The self-sufficient P450 RhF expressed in a whole cell system selectively catalyses the 5-hydroxylation of diclofenac, Biotechnol. J. 12 (2017) 1600520-1600520.

[37] S. Pflug, S.M. Richter, V.B. Urlacher, Development of a fed-batch process for the production of the cytochrome P450 monooxygenase CYP102A1 from Bacillus megaterium in E. coli, J. Biotechnol. 129 (2007) 481-8.

[38] J. Liston, The occurance and distribution of bacterial types on flatfish, J. Gen. Microbiol. 16 (1957) 205-216.

[39] T.S. Wong, F.H. Arnold, U. Schwaneberg, Laboratory evolution of cytochrome p450 BM-3 monooxygenase for organic cosolvents, Biotechnol. Bioeng. 85 (2004) 351-8.

[40] J.L. Porter, J. Manning, S. Sabatini, M. Tavanti, N.J. Turner, S.L. Flitsch, Characterisation of CYP102A25 from B. marmarensis and CYP102A26 from P. halophilus: P450 homologues of BM3 with preference towards hydroxylation of medium chain fatty acids, ChemBioChem n/a-n/a.

[41] M.A. Noble, C.S. Miles, S.K. Chapman, D.A. Lysek, A.C. MacKay, G.A. Reid, R.P. Hanzlik, A.W. Munro, Roles of key active-site residues in flavocytochrome P450 BM3, Biochem. J 339 (1999) 371-379.

[42] M. Dietrich, S. Eiben, C. Asta, T.A. Do, J. Pleiss, V.B. Urlacher, Cloning, expression and characterisation of CYP102A7, a self-sufficient P450 monooxygenase from Bacillus licheniformis, Appl. Microbiol. Biotechnol. 79 (2008) 931-940.

[43] Q.-S. Li, J. Ogawa, R.D. Schmid, S. Shimizu, Engineering Cytochrome P450 BM-3 for Oxidation of Polycyclic Aromatic Hydrocarbons, Applied and Environmental Microbiology 67 (2001) 5735-5739.

[44] S. Kille, F.E. Zilly, J.P. Acevedo, M.T. Reetz, Regio- and stereoselectivity of P450-catalysed hydroxylation of steroids controlled by laboratory evolution, Nature Chemistry 3 (2011) 738.

[45] D. Degregorio, S. D'Avino, S. Castrignanò, G. Di Nardo, S.J. Sadeghi, G. Catucci, G. Gilardi, Human Cytochrome P450 3A4 as a Biocatalyst: Effects of the Engineered Linker in Modulation of Coupling Efficiency in 3A4-BMR Chimeras, Frontiers in Pharmacology 8 (2017) 121.

[46] M.W. Peters, P. Meinhold, A. Glieder, F.H. Arnold, Regio- and Enantioselective Alkane Hydroxylation with Engineered Cytochromes P450 BM-3, J. Am. Chem. Soc. 125 (2003) 13442-13450.

[47] S.Q. Pham, P. Gao, Z. Li, Engineering of recombinant E. coli cells co-expressing P450pyrTM monooxygenase and glucose dehydrogenase for highly regio- and stereoselective hydroxylation of alicycles with cofactor recycling, Biotechnol. Bioeng. 110 (2013) 363-373.

[48] M. Bordeaux, D. de Girval, R. Rullaud, M. Subileau, E. Dubreucq, J. Drone, High-cell-density cultivation of recombinant Escherichia coli, purification and characterization of a self-sufficient biosynthetic octane omega-hydroxylase, Appl. Microbiol. Biotechnol. 98 (2014) 6275-83.

[49] S.C. Maurer, K. Kühnel, L.A. Kaysser, S. Eiben, R.D. Schmid, V.B. Urlacher, Catalytic Hydroxylation in Biphasic Systems using CYP102A1 Mutants, Adv. Synth. Catal. 347 (2005) 1090-1098.

[50] A. Pennec, C.L. Jacobs, D.J. Opperman, M.S. Smit, Revisiting Cytochrome P450-Mediated Oxyfunctionalization of Linear and Cyclic Alkanes, Adv. Synth. Catal. 357 (2015) 118-130.