High-level expression and molecular characterization of a recombinant prolidase from Escherichia coli NovaBlue

Expression and purification of the recombinant enzyme

Chromosomal DNA from E. coli NovaBlue was isolated by a commercial kit (Geno Plus Genomic DNA Extraction Miniprep System; Viogene Inc., Valencia, CA, USA) and used as a template for the amplification of the pepQ gene by polymerase chain reaction (PCR). PCR amplification was performed on Applied Biosystems 2400 Thermal Cycler (Thermo Fisher Scientific Inc., Waltham, MA, USA) with two sets of primers (PepQ-F: 5′-GGATCCATGGAATCACTGGCCTCGCTC and PepQ-R: 5′-AAGCTTTCACGCCAGTTTCAGATCCCG), designed on the basis of the gene sequence (Accession number: P21165) deposited in GenBank. Touchdown PCR was conducted as follows: an initial denaturation at 94 °C for 3 min, followed by 30 cycles of denaturation of 94 °C for 2 min, annealing at 53 °C for 1.5 min, and extension at 72 °C for 2 min. After a final extension at 72 °C for 2 min, the amplified fragment was digested with BamHI and HindIII, and inserted into the predigested vector pQE-30 (Qiagen, Hilden, Germany) to yield pQE-EcPepQ.

Single colony of E. coli M15 (pQE-EcPepQ) was inoculated to Luria-Bertani (LB) medium supplemented with 100 μg/mL ampicillin and 25 μg/mL kanamycin and the culture was shaken at 37 °C and 180 rpm until the optical density at 600 nm reached 0.6. High-level expression of EcPepQ in E.coli cells using a T5 promoter-based expression system in 100 mL of antibiotic-containing LB medium was initiated by adding IPTG to a final concentration of 100 μM. After incubation at 25 °C for 12 h, the culture cells were harvested by centrifugation, and were resuspended in 12.5 mL of buffer A (50 mM NaH₂PO₄, 300 mM NaCl, and 10 mM imidazole) and disrupted by sonication. The cell-free extract was subsequently applied to Ni-NTA agarose (Qiagen, Hilden, Germany) in a column, which had been equilibrated by buffer A. Afterward, the column was washed once with 30 mL of buffer A and eluted with buffer B (50 mM NaH₂PO₄, 300 mM NaCl, and 250 mM imidazole). Fractions enriched in protein and prolidase activity were pooled for gel electrophoresis and activity assay. Protein concentration was determined using the Bio-Rad protein assay kit with bovine serum albumin as the standard.

Gel electrophoresis and size-exclusion chromatography

Sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) was performed on a 12% polyacrylamide separating gel and a 5% stacking gel. For native gel electrophoresis, the experiment was carried out in a vertical slab gel apparatus with 5% stacking gel and 10% separating gel concentration. Coomassie blue R-250 dye was used to visualize protein bands on the polyacrylamide gels.

The apparent molecular mass of the native form of EcPepQ was determined by gel filtration on a Superdex 200 HiLoad 16/60 column (GE Healthcare, Chicago, IL, USA) with 25 mM Tris–HCl buffer (pH 8.0) containing 10 mM Mg(OAc)₂ and two mM DTT. A calibration curve was simultaneously established with four reference proteins, including thyroglobulin (670 kDa), aldolase (158 kDa), ovalbumin (43 kDa), and ribonuclease A (13.7 kDa). The elution volumes of reference proteins and the native enzyme were individually recorded to calculate their respective K_av values through the following equation: $$K_{\text{av}}=(V_{\text{e}}–V_{\text{o}})/(V_{\text{t}}–V_{\text{o}})$$ Where K_av, V_e, V_o, and V_t donate the distribution coefficient of a particular protein, the elution volume of a particular protein, the column void volume, and the total bed volume, respectively.

Activity assays

Prolidase activity was measured by determining the release of proline from Ala-Pro by the recombinant enzyme. The amount of proline in the reaction mixture was estimated by the colorimetric method reported elsewhere (Ghosh et al., 1998). At the beginning of the enzyme reaction, EcPepQ together with 25 mM Tris–HCl buffer (pH 8.0) and other reaction components (one mM Ala-Pro and 0.1 mM Mn²⁺) were warmed up separately at 60 °C for a short period of time (∼2 min). The catalytic action was initiated by adding 0.1 mL of a suitable dilution of enzyme to the reaction components and sufficient distilled water to bring the final volume up to 0.5 mL. The reaction mixture was incubated at 60 °C for 10 min and then stopped the catalytic action by adding 0.5 mL of acetic acid (3.5 N) and 0.5 mL of ninhydrin reagent (3%, w/v; 3.0 g ninhydrin in a mixture of 60 mL of glacial acid and 40 mL of phosphoric acid). Afterward, the resultant solution was heated at 100 °C for 10 min. The amount of substrate hydrolyzed was calculated from the increase in absorbance at 515 nm. One unit of EcPepQ activity is defined as the amount of enzymes that catalyze the release of one μmole of proline from Ala-Pro per minute at 60 °C and pH 8.0.

Effect of different temperatures (4–80 °C) on the prolidase activity was evaluated with 25 mM Tris–HCl buffer (pH 8.0). The thermal stability of EcPepQ was studied at pH 8.0 and in the temperature range of 4–80 °C for 10 min, and the residual activity was immediately measured under the standard assay conditions. The pH optimum of EcPepQ was determined at 60 °C using 25 mM phosphate-citrate buffer (pH 2.8–5.0), 25 mM sodium acetate buffer (pH 3.6–5.9), 25 mM sodium citrate buffer (pH 4.0–6.0), 25 mM sodium dihydrogen phosphate buffer (pH 6.0–8.0), 25 mM Hepes-NaOH buffer (pH 6.8–8.2), 25 mM Tris–HCl buffer (pH 7.0–9.0), and 25 mM glycine-NaOH buffer (pH 8.6–10.6). The pH stability of EcPepQ was evaluated at 30 °C for 30 min by incubating the enzyme sample (10 μg/mL) at different buffer systems, and the residual activity was then measured under the standard assay conditions. Stimulation of EcPepQ by metal cofactor was determined through assaying the enzymatic activity in presence of 1.0 mM metal ions.

Kinetic parameters of EcPepQ for the hydrolysis of Ala-Pro were determined using 45 μg/mL of the purified enzyme. Substrate concentrations ranging from 0.2 to 14.2 mM were applied to the reaction mixture. The assays were performed at 60 °C for 10 min in Eppendorf tubes in a total volume of 0.5 mL. To determine the kinetic constants of EcPepQ, a Lineweaver–Burk plot was created with data points derived from double-reciprocal transformation.

Circular dichroism spectroscopy

The circular dichroism (CD) spectra of enzyme samples were recorded in the far-UV region (190–260 nm) on the JASCO J-815 spectropolarimeter (JASCO Corporation, Tokyo, Japan) with the light path of one mm. Prior to spectral analysis, the desalted enzyme sample at a final concentration of 0.2 mg/mL was individually incubated at the indicated conditions or treated with different types of organic co-solvents (5–70%, v/v) for 30 min. The spectra were recorded at room temperature (except for the experiment of heat-induced denaturation) using an external circulating water bath. Each spectrum was an average of at least 10 scans and the control signals were taken and subtracted from the respective test to minimize the chances of any false signal generation due to the salts, chaotropic agents or solvents. The data were expressed as mean residue molar ellipticity (deg·cm²·/dmol) based on a residue number of 443 and a mean residue weight (MRW) of 50,176 Da. Mean residue molar ellipticity can be calculated as follows: [θ]_mrw = MRW × θ_obs/10 × c × l, where θ_obs is the observed ellipticity in degree at a given wavelength, c is the protein concentration in mg/mL, and l is the pathlength in cm. The spectra were also quantitatively analyzed by the DICHROWEB server to estimate the content of secondary structure elements (Lobley, Whitmore & Wallace, 2002).

Thermal unfolding of EcPepQ samples (∼0.15 mg/mL) in 25 mM Tris–HCl buffer (pH 8.0) was determined by monitoring the temperature-dependent changes of the molar ellipticity at 222 nm. In this experiment, EcPepQ samples were heated at scan rates of 0.5, 1, and 2 °C/min. The alterations in molar ellipticity at 222 nm were further fitted with a two-state model (Pace, 1990) to acquire the midpoint (T_m) of the unfolding curve. To explore the possible refolding of denatured EcPepQ, the temperature of thermoelectrically controlled cell holder was reduced by 0.5 °C/min and measurements were taken once every minute.

Fluorescence spectroscopy

Fluorescence analysis of EcPepQ samples was performed at 25 °C in a JASCO FP-6500 fluorescence spectrophotometer with an excitation wavelength of 295 nm. All fluorescence spectra were corrected for the contribution of 25 mM Tris–HCl buffer (pH 8.0), chaotropic agent and organic co-solvents. Before fluorescence spectroscopy, the purified EcPepQ was individually mixed with an appropriate amount of buffer and various amounts of guanidine hydrochloride (GdnHCl) or organic co-solvents to produce solutions with a protein concentration of approximately 0.15 mg/mL. Fluorescence spectra were then recorded for these samples after 30 min of equilibration at room temperature. The emission profiles of EcPepQ samples were recorded from 305 to 500 nm at a scanning rate of 240 nm/min. The maximal peak of fluorescence spectra and the changes in fluorescence intensity were brought together to calculate the average emission wavelength (AEW) (λ) according to Eq. (1) (Royer, Mann & Matthews, 1995). (1) $$\langle \lambda \rangle =\frac{{\displaystyle {\sum }_{i={\lambda }_1}^{{\lambda }_N}\left(F_i\cdot {\lambda }_i\right)}}{{\displaystyle {\sum }_{i={\lambda }_1}^{{\lambda }_N}\left(F_i\right)}}$$ in which F_i is the fluorescence intensity at the specific emission wavelength (λ_i).

To calculate the transition point and ΔG_N-U of GdnHCl-treated EcPepQ_, the unfolding data were further fitted with Eq. (2) (Pace, 1990). (2) $$y_{obs}=\frac{y_N+y_U•e^{-\left(\frac{\Delta G_{{(H_{2}O)}_{N\to U}}-m_{N\to U}[GdnHCl]}{RT}\right)}}{1+e^{-\left(\frac{\Delta G_{(H_{2}O)N\to U}-m_{N\to U}[GdnHCl]}{RT}\right)}}$$ where y_obs represents the observed biophysical signal, y_N and y_U are the calculated signals of the native and unfolded states, respectively, [GdnHCl] is the concentrations of the chaotropic agent, ΔG_N-U is the free energy change for the N↔U process, and the m_N-U represents the sensitivity to denaturant concentration.

Enzyme expression and purification

In order to overproduce EcPepQ, the PCR-amplified DNA fragment encoding an approximately 57 kDa protein was digested with BamHI and HindIII, and cloned into the expression vector pQE-30 to yield pQE-EcPepQ (Fig. 1A). Such construction allows the expressed EcPepQ bearing 10 additional amino acid residues at its N-terminus, which facilitates the single-step purification of the recombinant protein by metal-affinity chromatography. Initially, enzyme production by E. coli M15 (pQE-EcPepQ) was evaluated at 28 °C in a five mL medium containing essential antibiotics and 100 μM IPTG for a period of 24 h to select an appropriate time for further optimization. With this information, we then investigated effects of temperature and inducer concentrations on the production of active enzyme by E. coli M15 (pQE-EcPepQ). In these experiments, one mL of bacterial culture was harvested and disrupted by sonication, and the cell-free extract was subsequently analyzed for its specific activity toward Ala-Pro. Results demonstrated that the production of functional EcPepQ achieved a maximum upon a 12 h induction of E. coli M15 (pQE-EcPepQ) with 100 μM IPTG. One major band corresponding to a molecular mass of about 57 kDa was clearly observed in the cell-free extracts of IPTG-induced recombinant cells when the cultivation temperatures were set at above 16 °C (Fig. 1B). It is noteworthy that less amount of the recombinant enzyme were produced by E. coli M15 (pQE-EcPepQ) cultivated at 4 °C. Apparently, the optimal IPTG concentration for the production of active EcPepQ was 100 μM (Fig. 1C). Based on these observations, the best conditions for the high-level production of active EcPepQ by the recombinant cells were the cultivation of E. coli M15 (pQE-EcPepQ) cells at 25 °C, an inducer concentration of 100 μM, and a growth period of 12 h. Under the aforementioned conditions, the specific activity of the cell-free extract from E. coli M15 (pQE-EcPepQ) reached 36.9 U/mg.

The recombinant enzyme in the cell-free extract of E. coli M15 (pQE-EcPepQ) cells was further purified to near homogeneity by an affinity procedure that uses Ni-NTA resin. SDS–PAGE analysis of the pooled fractions exhibited a predominant band with a molecular mass of approximately 57 kDa (Fig. 2A), a finding which is in close agreement with the value deduced from the gene sequence of EcPepQ. The His₆-tagged enzyme could be purified nearly16-fold with a yield of 68.4% by single-step affinity chromatography. The native state of EcPepQ was also determined by fast performance liquid chromatography (FPLC) gel filtration (Fig. 2B). The experimental result showed that EcPepQ was eluted from the HiLoad 16/600 Superdex@200 PG column just after aldolase (158 kDa), in a fraction corresponding to a calculated molecular mass of 114.4 kDa (Fig. 2B). Owing to gel electrophoresis analysis of the purified enzyme suggested only one type of subunit (Figs. 2A and 2C), the native state of EcPepQ is most likely to be composed of two identical 57 kDa subunits.

Biochemical characterization of EcPepQ

The effect of temperature on EcPepQ activity was investigated at pH 8.0 over a temperature range from 4 to 80 °C. As shown in Fig. 3A, the maximum activity for EcPepQ was at 60 °C. The pH effect was also studied at 60 °C in the range of 2.7–11.0 (Fig. 3B). The maximum activity was observed at pH 8.0, whereas the enzyme was sensitive to pH shift with more than 80% of activity attenuation at pH 5.0 and 10.0. The thermostability of EcPepQ was evaluated by incubating the enzyme at temperatures between 4 °C and 80 °C for 10 min. As shown in Fig. 3C, the enzyme displayed remarkable stability with more than 90% of the initial activity preserved at temperatures below 30 °C. However, there was a considerable decrease in its activity at temperatures greater than 40 °C. In addition, the effect of pH on the stability of EcPepQ was investigated by incubating the enzyme at 30 °C and different pH’s for 1 h. The experimental results showed that it was stable in the pH value of 8.0, with 65–80% of the full activity at a pH range of 7.0–9.0 (Fig. 3D). However, the enzyme was almost inactive at pH values beyond 5.0 and 10.0.

The effects of several metal ions on EcPepQ activity at a final concentration of one mM are shown in Fig. 4A. It is noteworthy that the addition of Mn²⁺ ion into the reaction mixture greatly stimulated EcPepQ activity, increasing it by approximately 43-fold as compared with the control. To verify the optimal concentration of Mn²⁺ ion, EcPepQ activity was assayed under different concentrations of this metal ion. The results clearly indicated that Mn²⁺ ion at a final concentration of 0.1 mM had the greatest stimulation effect on the prolidase activity (Fig. 4B). However, the presence of Mg²⁺ and Fe²⁺ ions in the reaction mixture appeared to slightly enhance the enzymatic activity. Partial inhibition (∼1% inhibition) was observed in the presence of Zn²⁺ ion, and the strongest inhibitory effect was found in Cu²⁺ and Ni²⁺ ions.

The Michaelis–Menten kinetic constants, K_m and V_max, for the purified EcPepQ were determined by using varying concentrations of Ala-Pro. Prolidase activity was measured under standard assay conditions as described earlier and the obtained results were plotted as a graph of enzyme activity (U/mL) against concentration of substrate [μM], which yields a hyperbolic curve with K_m and V_max values. From the graph, K_m and V_max values of EcPepQ were determined to be 8.8 ± 1.1 mM and 434.8 μM/min, respectively. A k_cat value of 926.5 ± 2.0 s⁻¹ was further obtained through the Lineweaver–Burk plot.

Spectroscopic studies

Figure 5 presents the far-UV CD spectra of EcPepQ samples. The CD spectrum of the recombinant enzyme displays two strong peaks of negative ellipticity at 208 and 222 nm, indicative of a substantial α-helical content with lesser amounts of β-sheet and random coil (54% α-helix, 9% β-sheet and 21% random coil). The representative peaks for α-helix were significantly diminished at temperatures above 60 °C (Fig. 5A). The thermal unfolding of EcPepQ was initiated at 50 °C and the CD signal at 222 nm was completely disappeared at the denaturation temperature of 75 °C (Fig. 5B). Also, it is evident that the thermal denaturation of EcPepQ followed a two-state process with a well-defined T_m value of 64.2 ± 0.2 °C. To further explore whether the unfolding process was reversible or not, thermal denaturation of the enzyme sample was determined by monitoring the ellipticity at 222 nm at three constant heating rates. After the thermal denaturation went to completion, the enzyme samples were reversely cooled down to 20 °C using the same scan speed. Figure 5B shows the transition curves obtained with EcPepQ solutions at heating rates of 0.5, 1, and 2 °C/min. It is clear that there were no significant differences between these heating rates with respect to the transition temperatures. Besides, it can be seen that the native secondary structure of the enzyme was not recovered immediately after the unfolded protein was cooled down from 90 to 20 °C (Fig. 5B). This observation indicates that the thermal denaturation of EcPepQ is highly irreversible even at the very early stages of the unfolding process.

To probe the change in the tertiary structure of EcPepQ as a function of temperature, the enzyme conformations were also analyzed by fluorescence spectroscopy. Figure 5C shows the intrinsic fluorescence spectra of EcPepQ at different temperatures. Clearly, the fluorescence intensity of EcPepQ was decreased by more than 16% when the enzyme samples were analyzed at temperatures above 60 °C. The tryptophan emission fluorescence spectrum for EcPepQ was maximized at a wavelength of around 333.6 nm. It could be seen that the λ_max was individually displayed 1.2, 3.8, and 4.2 nm blue shift when the temperatures were set at 60, 70, and 80 °C (Fig. 5C). Together with the aforementioned CD data, it can be concluded that a significant change in the EcPepQ structure has occurred as a consequence of temperature elevation.

GdnHCl-induced denaturation of EcPepQ

The function of proteins depends on their ability to acquire a unique three-dimensional structure. It is generally known that GdnHCl acts as a classical denaturant to bring about unfolding of proteins by disrupting intramolecular interactions mediated by non-covalent forces. As shown in Fig. 6A, EcPepQ treated with GdnHCl at concentrations of less than 0.5 M retained >80% of the prolidase activity. An increase in concentration up to 1.0 M resulted in 7% of the activity remaining, whereas the enzyme was completely inactivated after treatment with 1.2 M GdnHCl. These results suggest that GdnHCl concentrations below 0.5 M only cause a relatively small change in the molecular structure of EcPepQ and its catalytic activity can tend to be recovered upon removal of the denaturant. Giving the fact that EcPepQ contains a total of seventeen tryptophanyl residues, fluorescence spectroscopy is very suitable for its conformational study. Unfolding of this enzyme at different concentrations of GdnHCl was accordingly performed and the obtained data were shown in Fig. 6B. The AEW that reported on the changes in both fluorescence wavelength and fluorescence intensity was used to calculate the thermodynamic parameter of the unfolding process. As shown in Fig. 6B, EcPepQ started to unfold at 1.2 M denaturant and exhibited a [GdnHCl]_0.5,N-U value of 1.98 M, which corresponds to a free energy change (ΔG_N-U) of 6.18 kcal/mol.

Effects of co-solvents on the catalytic activity and molecular structure of EcPepQ

Concentration effects of 12 different organic co-solvents on the enzymatic activity of EcPepQ were investigated with the widely used substrate, Ala-Pro. As shown in Fig. 7, the enzyme exhibited distinct sensitivity to these organic co-solvents. It is noteworthy that the catalytic capability of EcPepQ was well preserved in most of the organic co-solvents tested up to concentrations of 15% (v/v). Isopropanol and tetrahydrofuran (THF) clearly inactivated the enzyme at a concentration of 30% (v/v). Conversely, formamide, methanol, ethylene glycol, and glycerol were found to be the most compatible co-solvents to EcPepQ (Fig. 7). Among the remaining co-solvents, the least level of disability to the functionality of EcPepQ was dimethyl sulfoxide (DMSO), followed by 1,4-dioxane, acetonitrile, acetone and ethanol. It is also important to mention that the enzymatic activity of EcPepQ was primarily decreased as a consequence of an increase in co-solvent concentrations (Fig. 7). Some literatures have shown that DMSO at low concentrations (<10%, v/v) actually promotes the stabilization of some proteins (Huang, Dong & Caughey, 1995; Arakawa, Kita & Timasheff, 2007; Batista et al., 2013). Although the protective mechanism of DMSO remains obscure, protein-solvent preferential interactions might be appropriate to interpret its beneficial effect on the conformational stability of proteins in aqueous solutions (Timasheff, 2002; Arakawa, Kita & Timasheff, 2007).

To explore the relationship between the functional inactivation and structural disruption of EcPepQ in the water/organic co-solvent mixtures, CD and fluorescence studies were carried out with our laboratory facilities. The far-UV CD spectra of EcPepQ were essentially determined at the co-solvent concentrations that resulted in ≥90% decreases in the prolidase activity, and compared with that measured under solvent-free condition (EcPepQ in 20 mM Tris–HCl buffer, pH 8.0). Considering the presence of DMSO, 1,4-dioxane, and formamide had strong negative interference on the ellipticity signals of EcPepQ, these three co-solvents were excluded from the analysis of CD spectra. As the data presented above, the far-UV CD spectrum of the solvent-free enzyme sample displayed two strong peaks of negative ellipticity at 208 and 222 nm (Fig. 8A). The spectrometric characteristic of EcPepQ was significantly diminished in the presence of water-miscible organic co-solvents, especially in the presence of 70% acetonitrile, 70% methanol, and 60% ethanol. These data clearly indicates that high concentrations of some organic co-solvents lead to profound alterations in the far-UV spectra, reflecting a substantial loss of the secondary structure of EcPepQ.

Fluorescence emission spectra were also determined upon the excitation of the enzyme samples at 295 nm to monitor changes to the microenvironments of tyrosine and tryptophan residues (Royer, 2006). The fluorescence spectrum of solvent-free enzyme sample exhibited an emission maxima at 330.8 nm (Fig. 8B). Organic co-solvent concentrations that impaired the prolidase activity of EcPepQ to less than 10% generally resulted in either a red shift or a blue shift in the maximum wavelength and a subtle increase in the emission intensity as well. Based on these facts, we speculate that at least some of the fluorescent residues of EcPepQ undergo a relevant change in their local environments (Royer, 2006). Such results further enhance the idea that significant changes in the molecular structure of EcPepQ have occurred upon treatment with high concentrations of organic co-solvents, in good agreement with the experimental results obtained from the analysis of CD spectra.