Through sequence analysis of the target gene (accession number: U97153) to the cel29B from Orpinomyces joyonii and celB from Orpinomyces PC-2, we concluded that the first 20 amino acids were the signal peptides, the catalytic domain was from M21 to P376, and the last 101 amino acids consisted of the linker and NCRPD sequences. Because the P. pastoris expression vector pPIC9K contained the α-Factor secretion signal, the PCR primers were designed to exclude the coding region of the signal peptide. The isolated celE gene was about 1.3 kb; it was sequenced and found to have a 99.8% similarity with the reported sequence [20]. There were three mutations, one was in the NCRPD, the other two were silent mutations. It was then ligated to pPIC9K plasmid to generate the expression vector pPICE. Double digestion of pPICE with EcoRI and NotI released a 1.3 kb fragment corresponding to the size of the isolated celE gene, while single digestion by SalI released the 10.5 kb fragment, the same size of the expression vector. In addition, nucleotide sequencing revealed that the genes were inserted in frame into the plasmid.

After P. pastoris transformation, thirty P. pastoris transformants with the highest resistance to G418 were selected and subjected to genomic DNA isolation. The PCR assay and nucleotides sequencing were performed, which confirmed that all the transfomants contained an integrated celE sequence in the genome DNA. Enzyme activity was then further determined, after induction in shaking flasks in BMGY and BMMY medium. With induction by 0.5% methanol for 48 h, all the recombinant strains expressed and secreted endo-β-1,4-glucanase into the culture broth. Among them, three strains exhibited relatively high production of the recombinant protein, and one of the strains, designed as P. pastoris egE, was selected for further study. The endo-β-1,4-glucanase activity reached 35.1 IU mL⁻¹ within 48 h after 0.5% methanol induction in shaking flasks. The continuous production of CelE by transgenic P.pasoris due to the expression of the celE gene was also confirmed by SDS-PAGE analysis of culture supernatants of after growth for 12–24 h in BMMY broth. The total protein profile of transgenic P.pastoris on SDS-PAGE showed an apparent single band. It was estimated to be 52 kDa from its mobility relative to standard proteins by SDS-PAGE, similar to the predicted molecular mass (51.5 kDa) (Figure 1a).

To examine the effect of carbon source on endo-β-1,4-glucanase production, we grew the highest EG producing strain P. pastoris egE on the YSM medium containing sodium carboxymethyl cellulose (CMC) as the carbon source and compared it to the control strain with empty plasmid pPIC9K. Firstly, the two strains were streaked onto YPD plates and grew for 60 h. Single colonies were then inoculated into 25 mL YPD liquid medium in 250 mL Erlenmeyer flasks, and shaken at 30 °C, 220 rpm. 10 μL of cell culture was streaked on the plates when the OD₆₀₀ was 0.8. The celE-expressing transformant grew very quickly, while the control strain hardly grew on the plate (Figure 1b). These results suggest that P. pastoris egE expressing and secreting the endo-β-1,4-glucanase CelE acquired the ability to utilize CMC as the carbon source.

Furthermore, to verify the presence of the transformed gene in the chromosome, the P. pastoris egE was cultured for multiple generations. The chromosomal DNA was then extracted from different generations of the strains as templates. PCR verification of celE gene indicated that the gene has been stably integrated into the chromosome (Figure 1c).

At 40 °C, the effect of pH on EG activity for the hydrolysis of CMC was determined. The result showed that the optimal pH of EG was at pH 6.0 (Figure 2a) and the activity at pH 5.5 and pH 7.5 was more than 80% of that at pH 6.0. At pH 6.0, EG activities were measured at various temperatures to determine the optimal temperature. The result showed that the optimal temperature for the enzyme of interest expressed in P. pastoris was 45 °C (Figure 2b). The activity at 40–55 °C was more than 80% of that at 45 °C, but decreased rapidly outside this temperature range. In addition, the pH and temperature optima of the enzyme activity were consistent with those of enzymes expressed from E. coli. Depending on these results, the endo-β-1,4-glucanase expressed in P. pastoris GS115 can be classified as a mesophilic neutral endoglucanase, which can be used as a good source for industrial applications, especially for denim biofinishing/washing.

As a supplement study, we have expressed both the truncated (lacking the NCRPD) and full-length CelE in P. pastoris GS115 and found little difference between the enzymes in terms of activity and optimal pH and temperature. The NCRPD has been found in two groups of proteins with different functions. The NCRPD found in aerobic bacteria/fungi binds cellulose or other carbohydrates, and is usually named CBM (carbohydrate binding motif). In anaerobic fungi, the NCRPD functions as a dockerin domain or docking domian [4,36,37]. It is generally believed that NCRPD is not required for the catalytic activity of the enzyme. Instead, the domains in the anaerobic fungi are always responsible for the assembly of the multiple scaffoldins to be formed as the cellulosomes. This established role of NCRPD is consistant with the finding that the domain has little effect on the catalytic activity or properties of expressed CelE. However, we did find a slight enhanced stability of the enzyme produced from P. pastoris when the full-length sequence was used. Therefore the full-length enzyme was subsequently used for all the experiments.

We also compared the expressed CelE with the Thermoascus aurantiacus Cel5A, a GH5 member whose crystal structure has been previously determined. A previously reported study has shown that eight different residues are completely conserved among the GH5 enzymes [38]. These include the catalytic proton donor (Glu133 in Cel5A), the neiboring Asn132 and the nucleophile Glu 240. Five other conserved residues are His 93 and Trp273 (implicated in substrate binding), His 198 and Tyr 200 (in interaction with the nucleophile Glu240) and Arg49 (which forms H-bonds with conserved Asn132, His198 and Glu240). As shown in Figure 3, these eight conserved residues are conserved in the CelE protein.

Recombinant P. pastoris strain egE was cultivated in BCG medium in shaking flasks. Meanwhile, different methanol addition concentrations were discussed, and the endo-β-1,4-glucanase activity in supernatants was assayed every 12 h after methanol induction (Figure 4). We noticed that the enzyme production had barely changed when cells were induced by 0.5% methanol in BMMY medium and BCG medium, reaching a level of 57.3 IU mL⁻¹ and 54.4 IU mL⁻¹ respectively, after 96 h induction, demonstrating that the BCG medium was also appropriate for enzyme production. It should be emphasized here that the BMMY medium is a defined medium in common use, while the BCG medium is a complex medium designed in this study, with corn steep liquor and glucose as the main components. Like BMMY medium, BCG medium contains 100 mM potassium phosphate buffer. The enzyme production was increased significantly more rapidly in BMMY medium than in BCG medium in the first 12–24 h, while the EG accumulation in the broth became closer later. In this sense, the BCG medium could replace the BMMY medium. It is possible that the glucose in BCG medium may prime the cells at the early stage for increased production of the heterologous enzyme at later stages. This was consistent with determination of cell number: about 1.5 times more cells were grown in BCG medium than in BMMY medium at 24 h. Although glucose would repress the synthesis of methanol metabolizing enzymes [39], it had been already exhausted in the first 24–30 h induction time, having little impact on enzyme expression. Besides, the minimal media FBSH has also been tried for induction but was found to have lower enzyme production (39.5 IU mL⁻¹) after induction by 0.5% methanol for 96 h.

Methanol accumulation inside the broth has previously been shown to negatively influence the production level [40]. In the study of added methanol concentrations, we found that at 1.0%, the enzyme activity rapidly rose after methanol supplementation and reached its peak (72.5 IU mL⁻¹) when inducted for 96 h, which were 33.3% and 11.2% higher than the 0.5% (54.4 IU mL⁻¹) and 1.5% (65.2 IU mL⁻¹), respectively. The control strain (transformed by plasmid pPIC9K) has no endo-β-1,4-glucanase production at all (data not shown).

According to previous studies, a lower methanol concentration has resulted in a lower cell growth rate and, consequently, a lower yield of recombinant proteins [28,39,41]. However, it is also important to note that the induction of AOX1 promoter may decline at high methanol concentrations due to toxic products of alcohol oxidation having inhibitory effects directly on the AOX enzyme and cell growth in general [42,43]. A substantial higher laccase activity was obtained from recombinant P. pastoris at 0.5% than at 1.0% methanol concentration [44].

We have monitored the cells number by OD₆₀₀ every 12 h as well. The OD₆₀₀ for 0.5%, 1.0%, 1.5% methanol concentration were 18.2, 25.7 and 27.1 at 84 h, respectively. The OD₆₀₀ in BMMY medium was 19.8 with 0.5% methanol induction. Thus, low methanol concentration (0.5%) was not adequate for enzyme production. Although a higher concentration (1.5%) was beneficial for cell growth, the EG activity was affected negatively. Induction of the AOX1 promoter by higher methanol concentration could accelerate the cell growth as well as the rate of the protein synthesis, allowing less time for the peptide chains to fold properly [44]. Moreover, the toxic products from rapid metabolism may also be a negative factor.

The GH5 endo-β-1,4-glucanase EGII from T.reesei was of good quality in denim finishing and other applications [45,46]. In previous research, Qiao et al. [47] has expressed the EGII in P. pastoris GS115 using pPIC9K. The EG is a typical acidic cellulase and the optima pH of 5.0. The specific activity of the enzyme was less than 60% than at pH 5. The highest EG activity of recombinant P. pastoris/EGII in shaking flasks was only 8.73 IU mL⁻¹. Liu et al. [48] have also expressed the EGIV from T.reesei in P. pastoris. The EG activities of cultivation supernatant of P. pastoris EGIV was 2.4 IU mL⁻¹. Obviously, these EG activities were much lower than those from our study. Besides, the specificity activity of the crude enzyme from the 96 h fermentation broth (1.0% methanol concentration) was also determined to be 24.7 IU mg⁻¹. This was superior to other studies, e.g., Thongekkaew et al. [49] expressed the EG from Cryptococcus sp. S-2 in Pichia pastoris, and the specific activity of crude enzyme was 4.93 IU mg⁻¹. In another study, the specific activity was 7.7 IU mg⁻¹ with the EG from Volvariella volvacea [50]. The anaerobic fungi produce EGs with high specific activity which would greatly enhance its catalytic ability, indicating the great potential of recombinant CelE enzyme produced by P. pastoris. It is of substantial value for industrial application, such as the textile bio-finishing or paper deinking, which are preferable for the function of EGs. In addition, to the best of our knowledge, this is the first report of functional expression of neutral GH5 endo-β-1,4-glucanase gene from anaerobic fungi in Pichia pastoris. Further characterization of the enzyme, including substrate specificity and identification of critical amino acid residues, would be of great interest, and we intend to pursue these studies in the future as our long-term research direction.

Anaerobic fungus, Orpinomyces PC-2, was described by Borneman et al. (1989). The fungus was grown anaerobically at 39 °C as previously described [51]. Pichia pastoris strain GS115 and plasmid pPIC9K were purchased from Invitrogen Corporation (San Diego, CA, USA). The plasmid pPIC9K was used for expression in P. pastoris. Expression of inserts was controlled by the methanol-inducible alcohol oxidase gene (AOX1) promoter. Plasmids were introduced into E. coli DH5α to prepare enough amount (about 10 μg) of plasmids required for transformation of P. pastoris.

The E. coli DH5α was cultivated in LB medium. P. pastoris GS115 was cultured in YPD medium for propagation. The minimal media FBSH [52], BMGY and BMMY liquid medium were used for P. pastoris recombinants growth, heterologous enzyme expression and screening, as described in the manual of methods for the easy-select P. pastoris expression system (www.invitrogen.com/content/sfs/manuals/easyselecte_man.pdf). YSM medium was used to evaluate CMC utilization by recombinant yeast. BCG medium was used for enzyme expression in shaking flasks.

CMC was purchased from Sigma Chemical Co., Ltd (St Louis, USA). Enzymes (Taq DNA polymerase and T4 DNA ligase) were purchased from Sangon (Shanghai, China). Protein markers and restriction endonucleases were purchased from Takara (Dalian, China).

The Orpinomyces PC-2 was cultured [51] and isolated were done by the methods of Rozman and Komel DNA manipulation [53]. The celE gene was amplified from the genomic DNA by PCR according to its sequence reported [51]. N-terminal analysis by SignalP program (http://www.cbs.dtu.dk/services/SignalP/) suggested that the first 20 amino acid residues encoded the signal peptide, which was deleted by PCR method. Primer sequences were: E1: 5′-CGCGAATTCATGAGAGAAATATTCCATCCAAAG-3′, E2: 5′-GCGGCCGCATAAATACCACACCAGTTACCAT-3′.

Amplified PCR products were resolved by 1% agarose gel electrophoresis and ligated into the pMD18-T simple vector. The cloned DNA fragments were verified through DNA sequencing. Gel slices containing the expected size band was excised and extracted with Takara Agarose Gel DNA Purification Kit Ver. 2.0 (Dalian, China). The celE gene was inserted into the EcoRI and NotI sites of the vector pPIC9K to generate the pPICE construct.

Pichia pastoris strain GS115 was grown in 500 mL of yeast extract-peptone-dextrose (YPD) medium for 18 h and prepared for transformation. The plasmid pPICE were linearized and used to transform P. pastoris by electroporation (0.75 kV, 25 uF, 200 Ω, SCIENTZ-2B Genepulser; Ningbo, China). Transformants screening of multiple inserts was according to the Invitrogen pPIC9K Expression Manual and antibiotic G418 was used in five concentrations: 1.0, 2.0, 3.0, 4.0 and 5.0 mg/mL. To induce expression and screen the high EG producing recombinants, strains with highest resistence to G418 were selected and taken the procedure respectively: Single clone was cultured in 5 mL YPD medium at 30 °C overnight with shaking at 220 rpm. The cells were then incubated in 250 mL shaking flask containing 25 mL BMGY medium at 30 °C for 16–18 h. Harvest the cells by centrifuging at 1500–3000 g for 5 min at room temperature. To induce expression, decant the supernatant and resuspend cell pellet in 25 mL BMMY medium. The following procedures were according with that of Invitrogen Pichia Expression Kit (Version F). The EG activity was assayed and compared at 48 h. Three replicate experiments were performed for every experiment. From the transformants that exhibited the highest enzymatic activities, one was selected for fermentation and characterization.

After centrifugation, the supernatant samples were loaded on 12% SDS-PAGE gels. After electrophoresis, the gels were stained for 40 min in 30% methanol, 10% acetic acid and 0.1% Coomassie brilliant blue and then destained for 12 h in 30% methanol and 10% acetic acid. For the molecular mass determination the protein molecular weight marker from Takara (Dalian, China) was used.

The endo-β-1,4-glucanase activity of the supernatant samples against CMC substrates was estimated by incubating 1.0 mL of 1.0% CMC-Na solution in 50 mM citric acid buffer (pH 6.0) with 0.5 mL of the diluted crude enzyme solution (about 10–400 fold varying with the enzyme activity, also diluted in 50 mM citric acid buffer) at 45 °C for 15 min. The reaction was terminated by addition of 2 mL of 3,5-dinitrosalicyclic acid reagent and brown color was developed after holding all of the assay tubes in boiling water for 5 min. The contents of the tubes were diluted to 10 ml with distilled water and the absorbance was measured at 540 nm [54]. One unit of Endo-β-1,4-glucanase activity was defined as the quantity required to release 1 μmol of glucose per minute (IU mL⁻¹).

The endo-β-1,4-glucanase produced from transgenic P. pastoris strains were measured in different conditions. The optimum pH of the expressed endoglucanase was determined at 45 °C at every half point between 4.0 and 9.0 and adjusted by 50 mM citrate buffer solution (pH 4.0–6.5), 50mM sodium phosphate buffer (pH 6.5–8.0) and 50 mM borate saline buffer (pH 8.0–9.0). The endo-β-1,4-glucanase activities were measured at 5 °C intervals between 30 and 70 °C to determine the optimum temperature. The temperature was controlled in a water bath shaker.

Pichia pastoris GS115 transformed with the pPICE plasmid which had highest endo-β-1,4-glucanase activity was inoculated on YPD solid medium at 30 °C for 2 days. Single colony was then inoculated into 25 mL YPD liquid medium in 250 mL Erlenmeyer flasks, and shaken at 30 °C, 220 rpm for 24 h. Then the seed culture were harvested and transferred into 25 mL BMMY/FBSH/BCG medium in the 250 mL shake flasks. Three replicated experiments were performed for every experiment. The inoculum ratio was 10% (v/v), and the flasks were shaken at 30 °C, 220 rpm. Methanol was supplemented every 24 h to a final concentration of 0.5%, 1.0% or 1.5% (v/v).