Cultivation to improve in vivo solubility of overexpressed arginine deiminases in Escherichia coliand the enzyme characteristics

Construction and expression of ADI gene in E. coli

The arginine deiminase of P. putida had been studied due to its characteristics by purification from its original host [22]. Because of its high specific activity (58.8 U/mg) and in-vitro solubility, we selected the enzyme to investigate its potential in industrial conversion from L-arginine to L-citrulline. We optimized the gene sequence according to the codon frequency table of E. coli BL21(DE3) using the “codon randomization” strategy [23] and synthesized it. The artificial ADI gene (Figure 1) was constructed into different expression plasmids, including pET-30a(+), pBAD/gIIIA, pColdTF and pGEX-6P-1, for overexpression in E. coli BL21(DE3). After a normal optimization process according to the previous report [24], 0.1 mM IPTG and 0.04% L-arabinose incubated at 16°C were used to induce the rADI expression in E. coli cells. These expression plasmids, except pET-30a(+), exhibited nearly no solubility and enzymatic activity to hydrolyze L-arginine to L-citrulline. The pET30a-ADI produced overexpressed rADI proteins. However, there was only a weak corresponding band (approximately 53 kDa) in the soluble fractions (Figure 2A, lane 3), which gave the hydrolytic activity of 0.64 ± 0.04 U/ml. Most of the expressed proteins were existed in the form of inactive inclusion bodies (Figure 2A, lane 4).

Wen et al.[25] reported that truncation of a non-catalytic region in a 1,3-1,4-β-D-glucanase from Fibrobacter succinogenes improved enzymatic activity and thermotolerance. The crystal structures of ADI from Mycoplasma arginini and Pseudomonas aeruginosa showed that the enzymes shared the same catalytic triad (Cys-His-Glu) and the Cys active site [26, 27]. The catalytic region mainly located in the C-terminal region of ADI. To explore the effects of deleting the N-terminal non-catalytic fragment on solubility and activity of rADI, our construction based on pET30a-ADI, a plasmid that expressed a truncated ADI protein lacking the N-terminal 30 amino acids (T-rADI, see Figure 1). Unfortunately, instead of improving solubility of the product, the truncation resulted in complete inactivation of the enzyme (Figure 2B, lanes 2 & 4).

Co-expression with chaperone teams

Cultivation-improved solubilization of rADI

Kozai et al. [30] reported that the addition of L-arginine in the refolding solution was able to clearly increase the enzymatic activity of denatured ADI proteins. However, our experiments showed that the presence of L-arginine had no effect on the in-vitro solubilization of rADI proteins in the inclusion body form. We further tested whether L-arginine is helpful to improve the in vivo solubilization of ADI proteins in E. coli cells before the formation of inclusion bodies. E. coli cells are routinely cultivated in the LB medium, which contains peptones and yeast extracts as the carbon and nitrogen sources. We added different concentrations of L-arginine in the LB medium to cultivate the pET30a-ADI plasmid-harboring E. coli BL21(DE3) cells. Low concentrations of L-arginine (less than 1%) had no effect on the solubilization of rADI proteins, while higher concentrations (1%-5%) caused alkalization of the medium (pH10.0), in which E. coli cells developed poorly or no growth at all. If we adjusted the media containing high L-arginine concentrations to pH 7.0 or added L-arginine hydrochloride to replace L-arginine, which did not change the pH of the medium significantly, E. coli cells were able to grow normally but the solubilization of rADI proteins did not improve.

It was noted that higher concentrations of D-glucose (1% or more) supported better growth of cells, but the activities of crude extracts were prominently decreased. Higher concentrations of supplemented L-arginine (1% or more) inhibited the growth of cells because of alkalization of the medium, even with the presence of D-glucose, which caused no effect in the improvement of enzymatic activities (Table 2). For example, at the concentration of 1% L-arginine (combined with 0.5% D-glucose), the OD₆₀₀ of culture broth was less than 1.0 after 24 hrs of incubation and at the 2% concentration, there were no growth. Surprisingly, when we added different concentrations of L-arginine hydrochloride, instead of L-arginine in the LB medium with D-glucose, no improvement was observed for the solubilization of rADI proteins (Table 2). When L-arginine was combined with other sugars, such as D-galactose, D-sucrose and D-maltose, the solubilization of rADI proteins was not improved. When L-arginine was replaced with non-substrate amino acids of ADI, such as L-tryptophan, L-citrulline and L-glycine no improvement was observed for the solubilization of rADI proteins (Table 2).

We also assayed the effects of the combined addition of 0.5% L-arginine and 0.5% D-glucose on the expression of another ADI, which was from Listeria welshimeri serovar 6b str. SLCC5334 (NC_008555). The total activity of the enzyme was 0.1187 ± 0.04 U/ml in the presence of 0.5% L-arginine and 0.5% D-glucose, which was 7.2-fold higher than it expressed alone (0.0165 ± 0.01 U/ml) and 2.9-fold higher than it co-expressed with pGro7 (0.0412 ± 0.02 U/ml). The increase rate was similar to that for the rADI from Pseudomonas putida. These results suggested that the combined addition of L-arginine and D-glucose was applicable in solubilization of different ADI proteins.

Enzymatic and conversion characteristics of rADI

After purification using Ni-NTA affinity chromatography, the recombinant proteins showed as a single band with a size of 53 kDa on SDS-PAGE gel (Figure 3A). We assayed enzymatic characteristics of the purified rADI proteins. The rADI, which contained some additional terminal fragments such as a His-tag and some restriction sites, showed similar characteristics as the original ADI obtained from Pseudomonas putida[22]. For example, the stability of proteins was observed between pH 7.2 and pH 9.0 after 12 hrs incubation (Figure 3B). The optimal pH was 6.0 for enzymatic activity, but the activity was sharply decreased with the variation of pH value (above 7.2 or below 4.5) (Figure 3C). The enzymatic activity of rADI was mostly retained after incubation at 30°C or 40°C for 90 min, clearly decreased after incubation at 50°C for 90 min (Figure 3D). Cu²⁺ (10 mM), SDS (1 mM, 10 mM) and DTT (1 mM, 10 mM) strongly inhibited the activity of rADI, Cr³⁺ (10 mM), Co²⁺ (10 mM) had a slight inhibition for rADI, while other metal ions had no remarkable effect observed on rADI (Table 3). Interestingly, the suitable temperature for the enzymatic activity of rADI was between 30°C and 50°C and the optimal temperature was 40°C (Figure 3E). The optimal temperature was 50°C for the original ADI measured in Pseudomonas putida[22]. At the optimal 40°C, the specific activity of rADI reached 76 ± 0.03 U/mg, which was also higher than 58.8 U/mg of the original at 50°C.We performed conversion experiments using the sodium phosphate buffer (500 mM, pH 6.0) containing 10% L-arginine. The L-citrulline products increased linearly at 30°C or 40°C. After 22 hrs of incubation, 20 g of L-arginine substrates were almost converted into L-citrulline by the crude enzyme from 25 ml cultures. The addition of the batch substrate or not, had no impact on the conversion (Figure 4). The above enzymatic characteristics suggested that rADI was a potential candidate for industrial conversion of L-arginine to L-citrulline.

Strains and cultivation conditions

Construction of recombinant plasmids

The ADI gene studied in this study is taken from P. putida (GenBank accession no. P41142). We optimized the gene sequence according to the codon frequency table of E. coli BL21(DE3) using the “codon randomization” strategy [23]. The artificial gene and its corresponding amino acid sequences (Figure 1) were deposited in GenBank under the accession no. KJ411883. After synthesized at Biosune (Shanghai, China), the gene was amplified using a specific primer pair (Forward: CATGCCATGGCTATGAGCGCCGAAAAACAGAAGTATG; Reverse: CCCAAGCTTGATAATCGATAGGGTCGCGAACAATG) to generate the digestion sites of endonucleases NcoI and HindIII (italics). After purification, the amplified fragment was digested using the restriction enzymes and ligated into plasmid pET-30a(+), named pET30a-ADI, which was confirmed by sequencing. The construction of recombinant plasmids, transformation and protein expression were performed according to the protocol described by Sambrook and Russell [36].

Co-expression with the chaperone teams

The plasmids pGro7, pG-KJE8, pTf16 and pKJE7 encode different chaperone teams (Table 4). The pET30a-ADI was co-expressed with these four plasmids respectively. E. coli BL21(DE3) cells harboring plasmid pET30a-ADI alone or together with a chaperone-encoding plasmid were cultivated in 5 ml of LB medium containing 40 μg/ml kanamycin and 20 μg/ml chloramphenicol for the selection of recombinants. After overnight shaking at 37°C, the culture was inoculated (1%) into fresh LB medium containing antibiotics and 0.04% L-arabinose (pG-KJE8 was induced by 5 μg/ml tetracycline and 0.04% L-arabinose). The culture was shaken at 37°C to induce the expression of chaperones. When the OD₆₀₀ of the culture reached 0.6-0.8, isopropyl-β-D-thiogalactopyranoside (IPTG) was added to a final concentration of 0.1 mM to induce the expression of rADI. After additional 24 hrs of incubation at 16°C, the cells were harvested by centrifugation at 8,000 × g and 4°C for 10 min. The cell pellets were washed, resuspended in Tris-NaCl buffer (50 mM Tris, 150 mM NaCl, pH 8.0) and broken by sonication on ice. After centrifugation at 12,000 × g and 4°C for 30 min, the separated supernatant and cell debris were assayed enzymatic activities and in sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE, 10% gel concentration).

Cultivation-improved solubilization of rADI

The E. coli BL21(DE3) cells harboring the combined plasmids of pET30a-ADI and pGro7 were cultivated as described above. To assay the effects of cultivation on the solubility of rADI, different saccharides (D-glucose, D-sucrose, D-galactose or D-maltose) were added to a final concentration of 0.5% in the LB medium containing the antibiotics and L-arabinose. Together with the addition of IPTG, different amino acids (L-arginine, L-tryptophan, L-citrulline and L-glycine) were added to a final concentration of 0.5% when the culture OD₆₀₀ reached 0.6-0.8. The cultures were incubated for additional 24 hrs at 16°C. E. coli BL21(DE3) harboring pET-30a(+) was used as negative control.

Purification of rADI

After harvested, cells were broken by sonication, and the supernatants were applied to the Ni-NTA affinity chromatography (GE Healthcare, Buckinghamshire, U.K.) to purify rADI. The protein concentration was determined using the Bradford method (Bio-Rad, Munich, Germany). Different fractions were collected and assayed enzymatic activities and in SDS-PAGE (10%).

Activity assays of rADI

The rADI activity was determined using a modified method of diacetyl monoxime thiosemicarbazide [17, 37]. 40 mM L-arginine was dissolved in 500 mM sodium phosphate buffer (pH 6.0). After the addition of rADI, the mixture was incubated at 40°C for 10 min. Then, acid-ferric solution (containing 15% H₂SO₄, 60% H₃PO₄ and 0.005% FeCl₃) and DAM-TSC solution (containing 1% diacetyl monoxime and 0.06% thiosemicarbazide) were added to stop the reaction and the mixture was heated at 100°C for 10 min for coloration. The amount of L-citrulline was evaluated by measuring the optical density at 530 nm. The protein concentration was determined using Bradford-assay with bovine serum albumin as standard. One unit of the ADI activity was defined as the amount of enzymes to produce one μmol L-citrulline per minute under the assay conditions. All the assays were performed in triplicate. The total enzymatic activity was defined as units per milliliter of culture and the specific enzymatic activity was shown as units per milligram of purified enzyme.

Conversion rate of crude enzyme

The crude enzymes were prepared from sonication-broken cells using the above-described methods. 20 ml crude extracts from 100 ml culture broth were divided into four equal parts. The conversion of L-arginine to L-citrulline was assayed in 200 ml reaction buffer (500 mM sodium phosphate buffer, pH 6.0) at 30°C or 40°C. 20 g L-arginine was added at the beginning or batch-supplemented from 5 g to 20 g at different time points.