Cloning and characterization of thermophilic endoglucanase and its application in the transformation of ginsenosides

Sequence analysis of BcelFp from F. pennivorans DSM9078

The endoglucanase gene BcelFp consisted of 969 bp encoding 323 amino acids with a theoretical molecular mass of 37.89 kDa and a theoretical pI value of 5.43. The amino acid sequence of BcelFp (Genbank No. AFG35892.1) exhibited the highest similarity with GH 5 proteins from Fervidobacterium islandicum (85.7% identity, Genbank No. WP_052107242.1 ) and Fervidobacterium changbaicum (85.7% identity, Genbank No. WP_090223359.1). These proteins have not yet been characterized. The nearest characterized glycoside hydrolase (76% identity, GenBank No. WP_011994708.1) in the CAZy database was cellulase FnCel5A from Fervidobacterium nodosum. The BcelFp and FnCel5A were both from thermophilic bacteriums belonging to genus Fervidobacterium. FnCel5A was the first cellulase of the genus Fervidobacterium that had been cloned and expressed (Wang et al. 2010). The alignment of BcelFp with several characterized glycoside hydrolases from GH5 indicated that these proteins shared some conserved peptide motifs, namely NEP (residues 143–145), HYY (residues 203–205) and GEFG (residues 259–262) (Fig. 2). The Glu144 and Glu260 residues were typical catalytic residues of the GH5 enzymes, which confirmed that BcelFp belonged to GH5 family (Aspeborg et al. 2012).

Fig. 2figure 2

Multiple amino acid sequence alignment of BcelFp with several characterized glycoside hydrolases from GH5. The accession numbers of the aligned sequences are for the following organisms: ADD73709, endoglucanase FnCel5A from Fervidobacterium nodosum Rt17-B1; AAD36816, endoglucanase from Thermotoga maritima MSB8; AXU72614. endoglucanase from Clostridioides difficile. The accession numbers were indicated to the left of the amino acid sequences. Identical residues are indicated by a red background. Symbols: ↑ amino acids forming a catalytic residues

In order to gain a better understanding of the evolutionary position of BcelFp in glycoside hydrolase family 5, we constructed the phylogenetic tree using the neighbor-joining method in the MEGA4 program with bootstrap values based on 1,000 replications. The resulting consensus tree is presented in Fig. 3. BcelFp from Fervidobacterium pennivorans DSM9078 clustered within subfamily 25 and formed a separate, well-supported clade with cellulase from Acetivibrio thermocellus and endoglucanase (FnCel5A) from Fervidobacterium nodosum Rt17-B1.

Fig. 3figure 3

Phylogenetic analysis of BcelFp, and other characterized glycoside hydrolases from GH5. The units at the bottom of the tree indicate numbers of substitution events

Expression, purification and characterization of the endoglucanase BcelFp

The putative endoglucanase gene from F. pennivorans DSM9078 was cloned and expressed in Escherichia coli under the control of the IPTG-inducible promoter T7. After being induced under 16 oC for 12 h with 1 mM IPTG, the recombinant enzyme was solubly overexpressed in E.coli cells. The recombinant BcelFp was purified by His-trap affinity chromatography with a final purification of 2.3-fold and a specific activity of 297 U/mg for CMC. The expressed enzyme was determined as a single band by SDS-PAGE, with a molecular mass of approximately 38 kDa (See Additional file 1: Fig. S1), which was almost consistent with the molecular weight of 37,892 Da calculated from 323 amino acids.

The optimum pH and temperature of the purified BcelFp were determined using CMC as the substrate. The maximum activity was observed at pH 6.0, and the activity was higher than 50% of the maximum activity in the range of pH 4.0 to 6.5. While from pH 6.5, the enzyme activity decreased swiftly, representing that the enzyme was active over narrow acid pH range (Fig. 4 A). After being incubated at various pHs (pH 4.0, pH 5.0, pH 6.0 and pH 7.0) for 1 h, more than 90% of enzyme activity remained at acid pHs, while 70% of enzyme activity remained at pH 7.0, indicating that BcelFp was more stable at acid pHs (Fig. 4B). The optimal temperature for BcelFp activity was 95 oC, while it also displayed high activity between 60 and 100 oC (Fig. 4 C). Hence, the BcelFp was thermophilic enzyme.

Fig. 4figure 4

(A) Effect of pH on enzyme activity. (B) Effect of pH on enzyme stability. The activities were determined by assays with CMC as substrate following incubation of the enzyme at pH 4 (■), pH5(●), pH 6 (▶) and pH 7 (◀) for the indicated times. (C) Effect of temperature on enzyme activity. (D) Effect of temperature on enzyme stability. The activities were determined by assays with CMC as substrate following incubation of the enzyme at 85 °C (▲), 90 °C (●), and 95 °C (■) for the indicated times. (E) Differential Scanning Calorimetry (DSC) analysis of BcelFp. The enzyme was concentrated to 1.5 mg/ml in 50 mM PBS buffer (pH 6.0). The equilibrated enzyme was scanned from 35 to 120 oC at a rate of 2.0 K/min. The enzyme scan was corrected using a buffer–buffer baseline

Fig. 5figure 5

HPLC analysis of ginsenoside Rb1, Rb2, Rc and Rd during biotransformation process by using BcelFp. Ginsenoside standards were indicated on the peaks. Numbers were used to indicate the product peaks

Fig. 6figure 6

Biotransformation pathways of ginsenoside Rb1, Rb2, Rc and Rd by using BcelFp.

The effect of temperature on enzyme stability was investigated using CMC as the substrate. As seen in Fig. 4D, the enzyme was incubated for various lengths of time at 85, 90 and 95 oC at pH 6.0 and then the residual activities were measured. The enzyme was very stable at high temperature, and its half-lives at 85, 90 and 95 oC were 60, 25, and 10 h, respectively. When DSC analysis was performed to determine the thermal transition mid-point (Tm) of BcelFp, it was 96 oC (Fig. 4E).

The effects of metal ions, and EDTA on BcelFp activity were also investigated (Table 1). Na+ and Mn2+ were able to significantly increase enzyme activity under the high concentration of 10 mM. Moreover, Fe3+, Co2+, Zn2+ were able to inhibit enzyme activity under concentrations of 5 mM and 10 mM. No significant effect was observed in the presence of Ca2+, NH4+, Ba2+, and K+ under the high concentration of 10 mM. The chelating agent EDTA did not inhibit the activity, which indicated that BcelFp was not metalloprotein.

Table 1 Effects of metal ions and EDTA on the enzyme activity of BcelFp Substrate specificity of the endoglucanase BcelFp

The substrate specificity of BcelFp was investigated using substrates including CMC, barley β-glucan, RAC, Avicel, Laminarin, soluble starch, pustulan, and pNPG. CMC is typical substrate for determining endoglucanase activity, while Avicel is used for measuring exoglucanase activity (Percival Zhang et al. 2006). The pNPG is a synthetic compound which serves as an optimal substrate for β-glucosidase. As seen in Table 2, BcelFp displayed high activity towards CMC, but undetectable activity on Avicel and pNPG, therefore it was an endoglucanase. Moreover, BcelFp was able to hydrolyze RAC, structure of which was similar to CMC. In addition, selectivity of BcelFp on β-1,6-, β-1,3-, ɑ-1,4-, ɑ-1,6- linkages in glycosides were also tested. Barley β-glucan is composed of a mixed-linked β-1,3/1,4-glucans. Laminarin contains primarily β-1,3 and a portion of β-1,6-glycosidic bonds. Soluble starch consists of ɑ-1,4/1,6-glucans. Pustulan is a linear β-1,6-linked glucans. BcelFp exhibited highest activity on barley β-glucan, while no activity on laminarin, soluble starch, and pustulan, which further indicating that BcelFp can randomly hydrolyze the β-1,4-glucopyranosyl linkage, but cannot act on β-1,6-, β-1,3-, ɑ-1,4-, ɑ-1,6- linkages in glycosides.

Table 2 Enzyme specificity for BcelFp on various substrates

The substrate specificity of BcelFp on PPD- and PPT-type ginsenosides was also measured. As seen in Table 3, the sugar moieties including glucose, arabinopyranose, and arabinofuranose linked to C3 and C20 position in PPD-type ginsenosides. And the sugar moieties including glucose, xylose, and rhamnose linked to C6 and C20 position of PPT-type ginsenosides. The relative activity of BcelFp for the PPD-type ginsenosides was in the order Rd > Rb1 > Rb2 > Rc. However, BcelFp showed undetectable activity on GypXVII, CO, CMc1, F2, and Rg3. In addition, for PPT-type ginsenosides as substrates, the enzyme exhibited no activity on R1, Re, Rg1, Rg2 and Rh1, indicating that BcelFp didn’t hydrolyze sugar moieties linked to the C6 and C20 position of PPT-type ginsenosides.

Table 3 Substrate specificity of BcelFp on PPD- and PPT-type ginsenosides Biotransformation of PPD-type ginsenosides by the endoglucanase BcelFp

For the verification of the biotransformation pathways of the four PPD-type ginsenosides (Rb1, Rb2, Rc, and Rd) by using BcelFp, the HPLC analyses were performed. As shown in Fig. 5, ginsenoside Rb1 Rb2, Rc, and Rd had decomposed dramatically after 1 h of reaction. With longer reaction time, still only one product was detected for each substrate (data not shown). Peak 1 and 4 were identified to be GypXVII and F2 by comparison of their retention time with standards, respectively. As lack of standards for peak 2 and 3, their structures were further analyzed by HPLC-MS. As seen in Additional file 1: Fig. S2, The molecular weight of peak 2 was calculated to be 916 u based on its [M-H]− ion at m/z 915. The difference between 1078 u (Rb2) and 916 u (peak 2) is 162 u, which corresponds to a glucose moiety. So, BcelFp catalyzed the release of one glucose residue from Rb2 to generate peak 2. As Rb2 has an outer glucose linked to C3 position and an outer arabinopyranose linked to C20 position, the deglycosylation should occur at C3 position of Rb2 to generate CO. Similarly, peak 3 was identified to be CMc1 (See Additional file 1: Fig. S3). Combination with the 13 C NMR analysis, product 1–4 were finally confirmed to be GypXVII, CO, CMc1 and F2 (See Additional file 1: Table. S1), respectively. Based on the structural analysis, GypXVII, CO, CMc1, and F2 were generated by hydrolysis of the outer glucose linked to the C3 position of ginsenoside Rb1, Rb2, Rc and Rd, respectively. The biotransformation pathways of ginsenoside Rb1, Rb2 Rc and Rd catalyzed by BcelFp were presented in Fig. 6. In addition, GypXVII, CO, CMc1, and F2 were not further hydrolyzed. Hence, BcelFp didn’t hydrolyze sugar moieties linked to C20 position and the inner glucose residue at C3 position of PPD-type ginsenosides.

As seen in Table 4, the Michaelise-Menten constants (Km), Maximum Velocity (Vmax), and catalytic efficiencies (kcat/Km) for Rb1, Rb2, Rc, and Rd were listed. The order of the kcat/Km values of BcelFp for ginsenosides (Rd > Rb1 > Rb2 > Rc) was the same as that observed for relative activity. However, the Vmax values and the substrate affinity of the enzyme followed the orders Rb1 > Rc > Rb2 > Rd, and Rd > Rb1 > Rb2 > Rc, respectively. BcelFp had higher catalytic efficiency for Rd rather than Rb1, Rb2 and Rc. The kcat/Km value of BcelFp for ginsenoside Rd was 27.91 mM-1s-1, which was higher than that of β-glucosidase from Gordona terrae (2.21 mM-1s-1) (Shin et al. 2015), and β-glycosidase from Paenibacillus mucilaginosus (1.78 mM-1s-1) (Cui et al. 2014).

Table 4 Kinetic parameters for BcelFp with PPD-type ginsenosides

留言 (0)

沒有登入
gif