Simultaneous engineering of an enzyme's entrance tunnel and active site: the case of monoamine oxidase MAO-N

An efficient directed evolution strategy for enhancing activity and manipulating stereoselectivity of a monoamine oxidase is presented.


Materials and chemicals
KOD Hot Start DNA Polymerase was obtained from Novagen. Restriction enzyme Dpn I was bought from NEB. The oligonucleotides were synthesized by Life Technologies. Plasmid preparation kit was ordered from Zymo Research, and PCR purification kit was bought from QIAGEN. DNA sequencing was conducted by GATC Biotech. All commercial chemicals were purchased from Sigma-Aldrich, Tokyo Chemical Industry (TCI) or Alfa Aesar.

Construction of the expression plasmid of WT MAO-N
The full length WT MAO-N gene was PCR amplified from the synthetic WT MAO-N gene with 6-His tag purchased from GENEWIZ, Inc (www.genewiz.com) by the following primers (forward primer: 5'-CATGCATGCCATGGGCACCAGCCGCGACGGCTATCAGTGGA -3'; reverse primer: 5'-CGCCGCGGATCCTCAGTGGTGGTGGTGGTGGTGCAGGC-3'). The recombinant WT MAO-N gene was generated by the insertion of the PCR product into pRSF-Duet-1 vector via Nco I and BamH I sites with an C-terminal 6-His tag and verified by sequencing.

The selection of mutagenesis residues
The homology model of WT MAO-N was built using the X-ray crystal structure of the D3 variant of MAO-N from Aspergillus Niger (PDB: 2VVL) as template using the modeling package in the Accelrys Discovery Studio 4.1. Then substrate was modeled into the active center. All the residues lining the active center were selected to be within 5 Å of the docked substrate. The substrate access tunnel was determined using the Caver program on clustered structures of the homology model, the minimum probe radius was 0.7 Å and the coordinate of starting point was X:-0.874, Y:-16.474 and Z: -31.969. The residues surrounding the substrate access tunnel were selected for mutagenesis.

Primer design and library creation of MAO-N
Primer design and library construction depend upon the particular amino acid chosen, and in the case of MAO-N involves twenty-three residues, which were divided into six groups ( Fig. S2-Fig. S7): 1) Amplification of the short fragments of MAO-N using mixed primers F1/R1, F2/R2, F3/R3, F4/R4 and F5/R5 for Library A, B, C, D and E respectively. Amplification of the short fragments of MAO-N using mixed primers F6/R6 and F7/R7 for Library F; 2) Over-lap PCR using the PCR products of F6/R6 and F7/R7 as template and mixed primers F6/R7; 3) Amplification of the whole plasmid pRSF-MAO-N using the PCR products of F1/R1, F2/R2, F3/R3, F4/R4, F5/R5 and over-lap PCR product of step2 as megaprimers, leading to the final variety plasmids for library generation. Other Libraries ( Figure S8-S9) were created using the same procedure as mentioned above and all the primers used are listed in Table  S4. The PCR products were digested by Dpn I and transformed into electro-competent E. coliBL21 (DE3) to create the library for screening. The transformants were plated on HiBond-C Extra membranes placed on LB agar plates containing 50 µg/ml kanamycin and 0.5 mM IPTG, respectively. The plates were incubated at 30°C for 24 hours. The resulting first round of mutant libraries were screened with the 1, 2 or 3 as substrate as described below.

Screening of the mutant libraries
The plate assay method described by Turner group was followed. Specifically, the Hi-BondC Extra membranes containing the clones were pulled from the LB agar plates and kept at -20°C for 24 hours to lyse the cells. The membranes were incubated at room temperature for 12 hours with an assay mixture containing 1 tablet of diaminobenzidine (DAB), 1 ml of potassium phosphate buffer (1 M, pH 7.0), 20 µL of screening substrate (100 mM) solution, 30 µL of horseradish peroxidase (5 mg/mL), 10 mL of 2% agarose and water up to 20 mL. Positive clones were picked and inoculated on LB agar plates (50 µg/ml kanamycin) every two hours. The selected positive clones were subjected to activity measurement (see below activity assay) and the mutations were identified by DNA sequencing and amino acid sequence verification.

Activity assay
The enzyme specific activities were assayed using a SPECTRAMAX M2e (MD, USA) at 30°C. Initial rates of the reaction were measured via the absorbance of a dye (ε = 29.4 mmol L -1 cm -1 ) at 510 nm , which was produced by the action of horseradish peroxidase with the liberated hydrogen peroxide from the oxidation of the amine by MAO-N or variants, 4-aminoantipyrine, and 2,4,6-tribromo-3hydroxybenzoicacid. The assay mixture (0.2 mL total volume) contained 174 μL of phosphate buffer (50 mM, pH 7.4), 2 μL of a 2, 4, 6-tribromo-3-hydroxybenzoic acid stock solution (20 mg/mL in DMSO), 2 μL of 4-aminoantipyrine stock solution (15 mg/mL in H2O), 2 μL of an amine stock solution (0.5 M in DMSO), and 2 μL of a horseradish peroxidase stock solution (5 mg/mL). The reaction was started by the addition of 20 g of enzyme in 20 l of phosphate buffer (50 mM, pH 7.4). One enzyme unit (U) was defined as the amount of enzyme that produced 1 μmol of hydrogen peroxide per min. The activity assays were performed in triplicate with the supernatant extract of vector PRSF-duet induced expression as control experiments.

Expression and purification of WT MAO-N and positive mutants
The WT and positive mutants were inoculated in 5 mL LB containing 50 µg/mL kanamycin and cultured overnight at 37 °C with shaking. The overnight cultures were scaled up to 800 mL TB containing 50 µg/mL kanamycin and induced by 0.5% lactose at 28°C for 20-22 hours. Then the cultures were harvested by centrifugation at 6,000  g and resuspended in a PBS buffer (20 mM, pH 7.4) containing 500 mM NaCl, 20 mM imidazole. The cell pellets were disrupted by sonication and the cell debris was removed by centrifugation at 15,000  g for 60 min. The soluble protein samples were loaded onto a nickel affinity column (GE Healthcare) and washed with 20~500 mM imidazole solution containing 500 mM NaCl and 20 mM PBS buffer (pH 7.4). The purified proteins were desalted and concentrated with centrifugation filtration devices. The protein concentrations were determined by Bradford method.

Determination of kinetic parameters
The kinetic parameters were obtained by measuring the initial velocities of the enzymatic reaction and curve-fitting according to the Michaelis-Menten equation. The activity assay was performed in a mixture containing a varying concentration of 1 (0.25-10 mM) and 2 (0.5-10 mM). All experiments were conducted in triplicate.

Preparative scale deracemization of substrates using recombinant cells of MAO-N mutants
Deracemization of substrate 1, 2, 4 and 5 were carried out as follows: Cell pellet from E. coli cultures (5 g) containing mutant LG-I-D11 or LG-J-B4 was resuspended in 98 mL of phosphate buffer (100 mM, pH 7.4). Substrate 1(147 mg, 1.0 mmol) or 2 (104 mg, 0.5 mmol) or 4 (161 mg, 1mmol) or 5 (175 mg, 1mmol) in 2 mL of DMSO and borane-ammonia complex (124 mg, 4 mmol) were added and mixed. The mixture was shaken at 200 rpm and 30°C on an orbital shaker and the reaction was monitored by chiral HPLC analysis. When deracemization was finished, the pH of the reaction mixture was carefully adjusted to 11 with 5 M NaOH solution. The suspension was extracted three times with 100 mL of ethyl acetate and the phase separation was facilitated by centrifugation (6000 g, 15 min). The combined organic layer was dried over anhydrous sodium sulfate and filtered. Removal of the solvent and purification by preparative thin layer chromatography gave the product.
Deracemization of 1 (147 mg) gave 108 mg product (73% isolated yield), which was identified as S isomer by comparison of the retention time on a HPLC with that of the authentic sample. The ee value (>99%) was determined by chiral HPLC analysis performed on an Agilent 1200 using Chiracel AD-H column (4.6 mm × 250 mm, DAICEL CHIRAL TECHNOLOGIES CO.LTD). A mixture of hexane, ethanol and hexane containing 0.5% diethylamine (88:2:10) was used as eluent at 1.0 mL/min of flow rate and the column temperature was controlled at 30°C. [1] The retention times for (S)-and (R)-1 were 10 Deracemization of 2 (104 mg) gave 90 mg product (86% isolated yield), which was identified as S isomer by comparison of the retention time on a HPLC with that of the authentic sample. The ee value (93.4%) was determined by chiral HPLC analysis performed on an Agilent 1200 using Chiracel OD-H column (4.6 mm × 250 mm, DAICEL CHIRAL TECHNOLOGIES CO.LTD). A mixture of hexane and isopropanol (97:3) was used as eluent at 1.0 mL/min of flow rate and the column temperature was controlled at 40°C. [3] The retention times for (S)-and (R)-2 were 8.597 and 13.172 min, respectively.  Deracemization of 4 (161 mg) gave 129 mg product (80% isolated yield), which was identified as S isomer by comparison of the retention time on a HPLC with that of the authentic sample. The ee value (>99%) was determined by chiral HPLC analysis performed on an Agilent 1200 using Chiracel OD-H column (4.6 mm × 250 mm, DAICEL CHIRAL TECHNOLOGIES CO.LTD). A mixture of hexane/isopropanol/diethylamine (99:1:0.1) was used as eluent at 0.5 mL/min of flow rate and the column temperature was controlled at 40°C. [4] The retention times for (S)-and (R)-4 were 18.465 and 21.592 min, respectively. Deracemization of 5 (175 mg) gave 142 mg product (81% isolated yield), which was identified as S isomer by comparison of the retention time on a HPLC with that of the authentic sample. The ee value (>99%) was determined by chiral HPLC analysis performed on an Agilent 1200 using Chiracel OD-H column (4.6 mm × 250 mm, DAICEL CHIRAL TECHNOLOGIES CO.LTD). A mixture of hexane/isopropanol/N,N-diethylamine (99:1:0.1) was used as eluent at 0.5 mL/min of flow rate and the column temperature was controlled at 40°C. [4] The retention times for (S)-and (R)-5 were 12.111 and 12.969 min, respectively.

General procedure for the deracemization
Deracemization of substrate 6,7,8,9,10,11 and 12 were carried out as follows: Cell pellet from E. coli cultures (25 mg) containing mutant LG-I-D11 or LG-J-B4 was resuspended in 0.5 mL of phosphate buffer (100 mM, pH 7.4) containing borane-ammonia complex (80 mmol/L). Substrate in DMSO (5 μL, 1 mol/L) were added and mixed. The mixture was shaken at 200 rpm and 30°C on an orbital shaker for 24 h. HPLC samples were prepared as follows: aqueous NaOH-solution (50 μL, 5 M) was added to the reaction mixture in the Eppendorf tube, followed by 0.8 mL of methyl tert-butyl ether. After vigorous mixing, the sample was centrifuged at 13000 x g for 1 minute. The organic phase was separated, dried over sodium sulfate and analyzed by chiral HPLC.
The configuration of 6, 7, 8, 9, 10 and 12 was determinate by comparison with literature HPLC retention time. The configuration of 11 was assigned by analogy with the other compounds.

Modelling and molecular dynamics simulation
The 3D structural models of WT MAO-N ,LG-F-B6, LG-J-B4 and LG-I-D11 were constructed based on Xray crystal structures of the mutation Aspergillus niger MAO-N (PDB ID: 2VVL and 2VVM) using Schrodinger2015-3 [7] Energy minimization of the constructed model was done using AMBER16 [8] . Active site pocket volumes were calculated in Accelrys Discovery Studio 4.1 by searching for cavities. AMBER16 was used to carry out molecular dynamics simulation of final model using ff14SB.redq force field. The MD trajectories were further analyzed, along with the binding energy, to identify relevant binding poses and structures of entrance and exit channels that were sampled in the simulation. All atom molecular dynamics simulations have been performed using AMBER16 molecular dynamics package [8] . The bonded and non-bonded description of the interactions between the various atoms has been described using the AMBER16 force fields which include the ff14SB.redq force field parameters. The ANTECHAMBER module and GAFF2 with AM1-BCC charges [9] are used to obtain force field parameters for ligands. Initially, we perform a series of energy minimization steps to eliminate any bad contacts in the initially built structures. During the minimization, protein and FAD are restrained with harmonic force constants 200 kcal/mol. The minimization step involves 2500 steps steepest descent followed by 2500 steps of conjugate gradient method. After the energy minimization, the system is slowly heated up to 300 K in 100 ps MD using 1 fs integration time step, while restraining the solute with 20 kcal/mol harmonic force constant. After this, we perform 200 ps NPT equilibration of the structures with no harmonic restraints. Finally, 50 ns NPT production simulations are performed at 300 K and 1 atm pressure with 2 fs integration time step. We have implemented periodic boundary condition across the system using a TIP3P water box [10] . We use Particle Mesh Ewald (PME) techniques integrated with AMBER package to account for the long range part of the electrostatic interactions [11] . During the dynamics, all the bonds involving hydrogen are restrained using the SHAKE algorithm [12] . Langevin thermostat with collision frequency of 1 ps-1 is used to maintain the constant temperature while the pressure is controlled by anisotropic Monte-Carlo barostat [13] . The accelerated GPU version of PMEMD [14] was performed on Nvidia K40 series cards. We have employed CPPTRAJ [15] functionality of AMBERTOOLs [8] to perform various analyses on the equilibrium MD simulation trajectories. The images and graphics of the structures shown here were generated using the software packages VMD [16] , Grace (version 5.1.25) [17] and PyMOL [18] .               Table S1. Results of screening libraries of MAO-N for amine 3 using code NDT.