Evaluation of double expression system for co-expression and co-immobilization of flavonoid glucosylation cascade

Chemicals

Unless otherwise stated, all chemicals and medium compounds were bought from Sigma-Aldrich (St Louis, USA), Carbosynth (Compton, Berkshire, UK), or SERVA Electrophoresis Gmb (Heidelberg, Germany). Antibiotics were purchased from Cayman Chemical Company (Ellsworth, USA). Restriction enzymes (BsaI-HFv2, BbsI-HF), T4 DNA ligase, Taq PCR Kit, Plasmid Miniprep Kit, and all the rest necessary molecular biology reagents were bought from New England Biolabs Inc. (Ipswich, USA). Biochanin A was purchased from Carbosynth (Compton, Berkshire, UK). The UHPLC grade solvents used in this study were bought from Merck KGaA (Darmstadt, Germany).

Plasmids construction

All genes, primers, plasmids, and strains used and constructed within this research are listed in Table S1. All strains harboring created plasmids were obtained by transformation of chemically competent E. coli DH5-alpha, E. coli 10-beta (for plasmid maintenance), or E. coli BL21 (DE3) (for protein production) cells. Positive clones were selected with blue-white screening using selective media supplemented with X-Gal (20 µg/mL) and appropriate antibiotic (ampicillin (100 mg/L), kanamycin (50 mg/L), or gentamicin (50 mg/L)), colony PCR procedure using Taq DNA polymerase and positive phenotype. The sequences of expression cassettes were confirmed by Macrogen Europe BV.

Flanked by BbsI restriction sites, RhaBAD and Trc empty expression cassettes (Fig. 1) including appropriate promoter, transcriptional regulator, and T7 terminator sequences were synthesized and cloned into pMA-T plasmids by Invitrogen GeneArt Gene Synthesis (Thermo Fisher Scientific, Waltham, MA, USA). Both plasmids were then digested by BbsI restriction enzyme and cassettes were inserted into pSEVA23g19g1 and pSEVA23g19g2 vectors, respectively. Plasmids (pRhaBAD_12 and pTrc_23) served as backbone vectors in the following investigations.

Fig. 1figure 1

Schematic map of the construction of: a single gene RhaBAD expression cassette; b single gene Trc expression cassette; c RhaBADTrc double expression cassette. Red lines mark restriction enzymes cleavages. Abbreviation: CDS, coding sequence; p, promoter (pJ23103 for the rhaS expression; placQ for the lacI expression); T, terminator (T500 for the rhaS expression; T1 for the lacI expression); N, N-tag

Genes encoding GmSuSy sucrose synthase and YjiC glucosyltransferase were codon-optimized for E. coli, synthesized with flanking BsaI restriction sites, and cloned into the plasmids by Invitrogen GeneArt Gene Synthesis (Thermo Fisher Scientific, Waltham, MA, USA). Both genes were cut out by BsaI-HFv2 restriction enzyme and inserted along with T7 RBS and N-His6x-tag to pRhaBAD_12 and pTrc_23 backbone vectors respectively. Plasmids (pRhaBAD-GmSuSy and pTrc-YjiC) were then digested by BbsI-HF restriction enzyme and both inserted alongside into pSEVA63g19gA vector to create a monocistronic genetic configuration with independent induction systems for both genes (pRhaBAD-GmSuSy_Trc-YjiC).

Plasmid with mCherry coding sequence flanked with BsaI restriction sites was purchased from Invitrogen GeneArt Gene Synthesis (Thermo Fisher Scientific, Waltham, MA, USA). Plasmid with GFP (green fluorescent protein) coding sequence flanked with BsaI restriction sites was kindly provided by Juan Nogales’s research group (SBG, CSIC, Madrid). Both genes were proceeded in the same manner as GmSuSy and yjiC genes, which resulted in the construction of pRhaBAD-mCherry_Trc-GFP plasmid. Firstly, mCherry and GFP sequences were cut out by BsaI-HFv2 restriction enzyme and inserted along with T7 RBS and N-His6x-tag sequences to pRhaBAD_12 and pTrc_23 backbone vectors, respectively. The expression cassettes were then digested by BbsI-HF restriction enzyme and both inserted alongside into pSEVA63g19gA vector to create a monocistronic genetic configuration with independent induction systems for both genes (pRhaBAD-mCherry_Trc-GFP).

Enzyme expression screening

E. coli BL21 (DE3) harboring pRhaBAD-GmSuSy_Trc-YjiC plasmid was grown overnight in a 100-mL Erlenmeyer flask containing 30 mL of lysogeny broth (LB) supplemented with 50 mg/L gentamicin in an incubator shaker (New Brunswick Innova 44, Eppendorf, Vienna, Austria) at 37 °C and 120 rpm agitation. The next day, five baffled 250 mL shaken flask with 60 mL of terrific broth (TB) containing 50 mg/L gentamicin were prepared. Each were inoculated with 10% (v/v) of overnight preculture and incubated at 37 °C and 120 rpm agitation until an optical density at 600 nm (OD600) reached around 0.6–0.8 value. Protein expression was then induced in each culture by the addition of different concentrations of L-rhamnose and IPTG (isopropyl β-D-1-thiogalactopyranoside) (Table 1).

Table 1 Tested concentrations of the inducers for the evaluation of the enzymes expression

Expression was carried out overnight in an incubator shaker at 25 °C and 120 rpm agitation. The cultures were then harvested by centrifugation (4000 g, 30 min, 4 °C) and the pellets were resuspended in 50 mM HEPES buffer (pH 7.5), digested by lysozyme (300 μg/mL) for 1.5 h at 4 °C, and disrupted by sonication on ice using Vibra-Cell Ultrasonic Liquid Processor VCX 130 (Sonics & Materials, Inc., Newtown, USA) for 2.5 min by the following procedure—5 s pulses and 5 s pauses at 80% amplitude. The cell lysates were then recovered by centrifugation (14000 g, 30 min, 4 °C) and their catalytic activity was evaluated using enzyme assays described below. Cell lysate from strain harboring pRhaBAD-GmSuSy_YjiC plasmid (Supplementary materials) and induced by 10 mM of L-rhamnose was used as a comparison of the expression from single and double controlled expression systems.

In vivo fluorescence measurements

For the in vivo fluorescence measurements, E. coli BL21 (DE3) strain harboring pRhaBAD-mCherry_Trc-GFP plasmid was cultivated at 37 °C in the 96-well microplate in Synergy H1 microplate reader (BioTek Instruments, Vermont, USA) using 22.5 h continuous assay with 45 min intervals for shaking (15 s, 282 cpm), fluorescence (GFP: excitation 479 nm, emission 520 nm; mCherry: excitation 579 nm, emission 616 nm), and absorbance (OD600) measurements. The working volume in each well was 200 μL. Cultivation was carried out in a double concentrated M9 minimal medium (2xM9, Table S2) with enhanced buffering capacity (Azatian et al. 2020) due to pH dependence of GFP fluorescence (Doherty et al. 2010). The medium was supplemented with 50 mg/L gentamicin, 10% (v/v) of overnight preculture, 0.8% (w/w) glucose, and different concentrations of the inducers (Table 2).

Table 2 Tested concentrations of the inducers for in vivo fluorescence measurements

The uninduced culture was used as a control of the expression of fluorescent proteins. All variants were cultivated in three independent replicates. Fluorescence values were normalized against the OD600.

Enzymes overexpression

E. coli BL21 (DE3) harboring pRhaBAD-GmSuSy_Trc-YjiC or pRhaBAD-GmSuSy_YjiC plasmid was grown overnight in a 100-mL Erlenmeyer flask containing 30 mL of LB supplemented with 50 mg/L gentamicin in an incubator shaker (New Brunswick Innova 44, Eppendorf, Vienna, Austria) at 37 °C and 120 rpm agitation. The next day, 500 mL of TB containing 50 mg/L gentamicin was inoculated with 10% (v/v) of overnight preculture and incubated in a baffled 2 L shaken flask at 37 °C and 120 rpm agitation until an OD600 reached around 0.6–0.8 value. Protein expression was then induced by the addition of 10 mM of L-rhamnose and 1 mM of IPTG (pRhaBAD-GmSuSy_Trc-YjiC) or 25 mM of L-rhamnose (pRhaBAD-GmSuSy_YjiC). Expression and sonication was performed as described above in the enzyme expression screening section. The cell lysate was then recovered by centrifugation (14000 g, 30 min, 4 °C) and stored at 4 °C.

Enzymes co-immobilization

Co-immobilization of the enzymes was performed on standard agarose based resin coated with nickel ions (Ni-agarose resin) and EziG Amber, Coral, and Opal resins. Each resin was prepared following the manufacturer protocol (see Supplementary materials) using HEPES buffer (50 mM, 50 mM KCl, 300 mM NaCl, pH 7.5). Next, E. coli cell lysate containing both enzymes and 20 mM of imidazole was added to the prepared resins. Immobilizations were carried out on ice on the orbital shaker (ELMI, Riga, Latvia) at 100 rpm agitation for 2 h. Then suspensions were centrifuged at 500 g for 5 min, supernatants were removed, and resins were washed two times with wash buffer (50 mM HEPES, 50 mM KCl, 300 mM NaCl, 5 mM imidazole, pH 7.5). Supernatants and washing solutions were collected for protein and activity measurements. The concentration of the proteins was evaluated using Bicinchoninic Acid Kit for protein determination (Sigma-Aldrich, St Louis, USA) and bovine serum albumin as a reference for the calibration curve. The enzymatic activity was determined by the standard cascade assay (described below). Solutions were also analyzed by the SDS-PAGE protocol (Laemmli 1970) using Color Prestained Protein Standard Broad Range (10–250 kDa) (New England Biolabs Inc. Ipswich, USA) as a molecular weight marker. Prepared resins were resuspended in HEPES buffer (50 mM, 50 mM KCl, pH 7.5) and stored at 4 °C.

The protein capacity of each resin was calculated as the difference between the protein amount in cell lysate before immobilization and cell lysate and wash solutions after immobilization. The enzyme activity bound to the resin (bound activity) was calculated as a difference between the cell lysate activity before (A0) and after the immobilization process (A), divided by the mass of the employed resin (g). The bound activity is expressed as U/g. Immobilization yield (%) was calculated as a ratio between ΔA (= A0-A) and A0. The observable activity of the solid catalyst was measured directly by the standard cascade assay and expressed as U/g. Recovered activity (%) is the ratio between the observable and the bound activity of the solid catalyst.

Enzyme assays

All reactions were performed in 0.8 mL total volume in 2-mL tubes at 500 rpm (soluble enzymes) or 1000 rpm (immobilized enzymes) agitation in a shaker (Eppendorf ThermoMixer C) for 15 min. The reactions were terminated by heating at 100 °C for 5 min and centrifuged (21300 g, 5 min) to remove the solid materials (precipitated proteins, or resin). The supernatants were analyzed via the described below protocols. The reaction mixtures lacking the catalyst served as a control.

YjiC

The general glucosyltransferase reaction mixture contained 0.5 mM uridine diphosphate glucose (UDP-glucose), 0.1 mM p-nitrophenol (pNP), 50 mM KCl, and 40 μL of the cell lysate or 40 μL of the suspension of co-immobilized enzymes (which corresponded to 2.8 mg of the dry resin). The reaction progress was analyzed by the measurements of the pNP absorbance at 405 nm on the Synergy H1 microplate reader (BioTek Instruments, Vermont, USA) in 200 μL working volume. The concentration of the pNP was calculated in reference to the standard curve. One unit (U) of the catalyst activity is defined as the enzyme amount releasing 1 μmol pNP β-D-glucoside/min under specific conditions.

GmSuSy

The general sucrose synthase reaction mixture contained 2 mM uridine diphosphate (UDP), 500 mM sucrose, 50 mM KCl, and 40 μL of the cell lysate or 40 μL of the suspension of co-immobilized enzymes (which corresponded to 2.8 mg of the dry resin). The reaction progress was analyzed by the detection of the fructose by the modified bicinchoninic acid (BCA) method (Waffenschmidt and Jaenicke 1987; Cerdobbel et al. 2010), which was evaluated for sucrose interference (Table S3). Reaction samples (10 μL) were added to the 190 μL of the assay solution (see Supplementary materials) and heated at 100 °C for 5 min. After cooling to room temperature, the absorbance at 560 nm was measured on the Synergy H1 microplate reader (BioTek Instruments, Vermont, USA). One unit (U) of the catalyst activity is defined as the enzyme amount releasing 1 μmol fructose/min under specific conditions.

Cascade

The general cascade reaction mixture contained 0.5 mM UDP, 500 mM sucrose, 0.1 mM pNP, 50 mM KCl, and 40 μL of the cell lysate or 40 μL of the suspension of co-immobilized enzymes (which corresponded to 2.8 mg of the dry resin). The reaction progress was analyzed in the same manner as glucosyltransferase reactions. One unit (U) of the catalyst activity is defined as the enzyme amount releasing 1 μmol pNP β-D-glucoside/min under specific conditions.

Enzyme characterization

The effect of the pH on enzyme activity was performed at 30 °C in the pH range 5.5–9.5 using the following buffers: 50 mM sodium citrate (pH 5.5, 6.5), 50 mM HEPES (pH 7.5, 8.5), and 50 mM glycine–NaOH (pH 9.5). The dependence of the enzyme activity on the temperature was checked at 20, 30, 40, and 50 °C at 50 mM HEPES buffer (pH 7.5). The enzyme activity at various Mg2+ ions concentrations (0.1, 1, 5, 10, and 50 mM) was evaluated at 40 °C and 50 mM HEPES buffer (pH 7.5). Enzyme stability was assessed at different pH values (6.5, 7, 7.5, 8, 8.5) and different temperatures (30, 35, 40, 45, and 50 °C) by incubation for 6 and 24 h in 50 mM HEPES buffer (50 mM KCl, 10 mM MgCl2, 500 mM sucrose). The residual cascade activity was checked using a standard cascade assay at 40 °C and was calculated as the percentage of the activity at the beginning of the incubation. In the all above-described characterizations, cell lysates and enzymes co-immobilized on Ni-agarose resin were used. Each test was performed in triplicates.

Reusability of the co-immobilized cascade

Reactions were performed in 50 mM HEPES buffer (pH 7.5) containing 30 mg of solid catalyst, 0.5 mM UDP, 500 mM sucrose, 0.1 mM pNP, 50 mM KCl, and 10 mM MgCl2 in 0.6 mL total volume in sealed spin columns (Sigma-Aldrich, St Louis, USA) inserted into 2-mL tubes at 1000 rpm agitation in a shaker (Eppendorf ThermoMixer C) at 40 °C for 10 min. At the end of each cycle of reaction, columns were open, reaction mixtures were centrifuged (1000 g, 2 min) and fresh reaction mixtures were loaded to resealed spin columns for the consecutive cycle of catalysis. Filtrates were analyzed in the same manner as in the glucosyltransferase assay. Each carrier was reused 12 times in triplicates. The residual cascade activity was calculated as the percentage of the activity at the first cycle of reaction, separately for each resin.

Fed-batch conversion of Biochanin A

The semi-preparative reaction was performed using co-immobilized on EziG Opal resin enzymes, in a total volume of 40 mL in 50-mL Eppendorf falcon in an incubator shaker (New Brunswick Innova 44, Eppendorf, Vienna, Austria, 30 °C, 120 rpm) for 48 h. Reaction was carried out at 50 mM HEPES buffer (pH 7.5) containing 0.5 mM UDP, 500 mM sucrose, 50 mM KCl, 10 mM MgCl2, 10% (v/v) DMSO (dimethyl sulfoxide), 150 mg of solid carrier, and 68.75 mg Biochanin A in total. Biochanin A dissolved in DMSO (64.5 mM stock) was stepwise added to the reaction at 0, 2, 4, 6, 8, and 24 h. Reaction samples (20 μL) were withdrawn hourly for the first 8 h and then after 24, 25, 26, and 48 h of the catalysis, extracted with 100 μL of ethyl acetate, and centrifuged (21300 g, 5 min). Thirty microliters of organic fractions was diluted in 170 μL of methanol and analyzed using UHPLC performed on Dionex Ultimate 3000 UHPLC + instrument (Thermo Fisher Scientific, Waltham, MA, USA) equipped with a DGP-3600A dual pump liquid control module, a TCC-3200 thermostated column compartment, a WPS-3000 autosampler, a diode array detector (DAD), and an analytical C-18 Acclaim RSLC Polar Advantage II (2.2 μm, 2.1 × 100 mm, Thermo Fisher Scientific) column, thermostated at 40 °C with 0.1% formic acid solution in water (A) and 0.1% formic acid solution in acetonitrile (B) as mobile phase in gradient elution program (0–3 min: 15–98% B; 3–4.2 min: 98% B; 4.2–4.4 min: 98–15% B; 4.4–6 min: 15% B) at 0.7 mL/min flow. The system control and data acquisition were done using Chromeleon 6.80 software (Dionex, Sunnyvale, USA). The detection was carried out at 280 nm, and the identification of the substrate and product was based on the retention time and UV spectrum of authentic standard compounds and product structure was also confirmed by NMR spectra analysis (Supplementary materials). The total turnover number (TTN) was calculated as the ratio of moles of generated product (Sissotrin) divided by the moles of UDP added to the reaction.

Statistical analysis

The means, standard deviations of the mean, and Pearson’s correlation coefficients (PCC) were calculated from triplicate experiments using Statistica software (version 13.3).

Accession numbers

Synthetic genes used in the study were deposited in NCBI Genbank database under accession numbers: OP381218 for Sucrose synthase and OP381219 for YjiC glycosyltransferase.

留言 (0)

沒有登入
gif