Expression and characterization of Trichoderma reesei endoglucanase II in Pichia pastoris under the regulation of the GAP promoter

Trichoderma reesei is known to be one of the organisms capable for producing various types of cellulase in high concentrations. Among these cellulases, the highest catalytic efficiency of endoglucanases II (EGII, EC 3.2.1.4) are considered important for industrial application. The characterization of the EGII is necessary since it is widely used in high‐temperature reactions in the industries. In this study, the recombinant EGII protein was expressed in Pichia pastoris and it has a molecular mass of approximately 52 kDa. Recombinant EGII was purified using Ni‐NTA affinity chromatography and characterized by SDS‐PAGE and western blot analyses. The enzyme activity of recombinant EGII was measured using the Nelson Somogyi method to determine its optimum pH and temperature. The result showed that the maximum EGII expression was achieved after 72 h of culture incubation. The crude enzyme has optimum activity at pH 5.0, resulting in 16.3 U/mL and 14.6 U/mL activity at 40 °C and 50 °C, respectively. While the purified enzyme gave the specific activity of 115.7 U/mg under the optimum condition. Finally, our study demonstrated that recombinant EGII could retain the endoglucanase activity for 89% and 80% at 40 °C and 50 °C, respectively.


Introduction
Trichoderma reesei is a wellstudied fungus that is capa ble of producing large amounts of various cellulases. This fungus secretes at least six types of cellulases that con sist of two types of cellobiohydrolases and four endoglu canases (Knott et al. 2014). These enzymes are exten sively used in several industries, such as laundry deter gent, textile and pulp, paper industry, and potential for bioenergy production. Thereby, this fungus is industri ally relevant to meet the target production level of cellu lases. Nowadays, the engineering of cellulase into a high performance enzyme for biomass hydrolysis and other in dustrial applications becomes the major research priority. However, various industrial processes and conditions (par ticularly in different temperature and pH conditions) re main a challenge. The cellulase excreted from T. reesei cannot withstand a long period of exposure at high tem perature and pH during its reaction process, which leads to the leveling off of enzymatic activity (Akbarzadeh et al. 2014).
Among cellulases produced by T. reesei, endoglu canase II (EGII; EC 3.2.1.4) is predominant and showed the highest catalytic proficiency. The EGs activity of T. reesei is known to decrease about 55% when EGII was absent in the secretory complex of EGs (Qin et al. 2008; Boonvitthya et al. 2013. This evidence revealed that the presence of EGII is crucial for lignocellulosic biomass hy drolysis and other industrial applications. Several stud ies reported that EGII production and characteristic im provement were performed by improving the strain of se creting microorganisms, protein engineering, and recom bination (Ito et al. 2004; Liang et al. 2011; Charoenrat et al. 2013. Thereby, there is still ample scope for im provement, particularly to produce EGII in a heterologous expression system to facilitate protein engineering work. Several heterologous expression has been carried out to produce endoglucanases in various host microorganisms, including Escherichia coli, Yarrowia lipolytica, Saccha romyces cerevisiae, and Pichia pastoris (Nakazawa et al. 2008; Qin et al. 2008; Boonvitthya et al. 2013; Akbarzadeh et al. 2014. Yeast is commonly used for its ability to in crease protein stability since the glycosylation process has occurred. In consequence, the structural and thermal sta bility of protein may increase due to the covalent bond for mation. The covalent bond formation causes less dynamic fluctuation and reduces protein molecules' flexibility (Qin and Qu 2014). tion levels, which helps for easier purification of recombi nant protein (MacauleyPatrick et al. 2005). It has also been proven that the expression system promotes eco nomically effective production of recombinant protein as it does not need complex medium and condition (Safder et al. 2018). The expression of a foreign gene in P. pastoris includes three main steps: (a) insertion of the foreign gene into an expression vector; (b) introduction of the expres sion vector into the expression host, P. pastoris; and (c) the selection of potential strains for foreign gene expres sion (MacauleyPatrick et al. 2005). The gene encoding EGII (egII gene) from T. reesei was successfully inserted into an expression vector and introduced into P. pastoris genome. In this study, the highexpression transformant with high endoglucanase activity was selected for further steps. The expression of recombinant EGII was regulated under glyceraldehyde3phosphate dehydrogenase (GAP) constitutive promoter in the fedbatch fermentation pro cess. Recombinant EGII produced was then characterized for the determination of its optimum pH, temperature, and thermal stability.

Strain and plasmid
Pichia pastoris SMD1168H purchased from Invitrogen (USA) was used as an expression host. Constructed plas mid pLIPITrCel5A ordered at ATUM (USA) used as an expression vector for P. pastoris transformant carrying egII gene from T. reesei was available at Research Cen ter for Biotechnology, Indonesian Institute of Sciences.
Six colonies were selected from the se lection medium and characterized by colony PCR. Specific primers, namely MFαFP1 (5' ATGAGATTCCCATCTATTTTCACCGCTGTCT 3') and TrCel5ARP (5' GAGCGGGGGATATACTTTGGAAGTAACACAA3'), were used to detect the inserted gene in the yeast genome. PCR was performed as follows: initial denaturation at 95°C for 5 min; 30 cycles of denaturation at 95°C for 30 s, annealing at 55°C for 40 s and extension at 72°C for 40 s, then final extension at 72°C for 5 min. The amplified fragments were then analyzed using 1% agarose gel electrophoresis.
The colonies were then confirmed for their expression of EGII, measured by a plate diffusion assay according to the method proposed by Ratnakomala et al. (2019). Ten microlites of each transformant cultures were added into agar wells containing 0.5% (w/v) carboxyl methylcellu lose (CMC) and incubated for 3 d at 30°C. Finally, plates were stained using 1% Congo Red solution for 15 min for color development, followed by washing the plate with 1 M sodium chloride solution to detect halo zones. The di ameter of the halo zones was measured and documented. Transformants with bigger halo zones were selected for further analysis.
Selected transformants were tested for their endoglu canase activity. The activity was measured by Nelson Somogyi (NS) method using CMC as a substrate, accord ing to the reference by Gusakov et al. (2011). About 160 µL of 6.25 mg/mL CMC in 0.1 M acetate buffer (pH 5.0) and 40 µL culture supernatant were preheated at 50°C for 5 min. Both were mixed and heated at 50°C for another 10 min. Zeropoint two milliliters of copper tartrate was added into the mixture to stop the reaction, and then the assay mixture was incubated in boiling water for 40 min. The mixture was allowed to decrease the temperature to 25°C (room temperature), then 0.2 mL of arsenomolyb date was added and incubated at room temperature for 10 min. One point four milliliters of a mixed solution and 0.4 mL of acetone was added into the assay mixture and then centrifuged at 13,000 rpm for 1 min. The optical density was measured using a spectrophotometer at λ=610 nm to estimate the quantity of reducing sugars produced in the assay mixture. One enzyme activity unit was defined as the amount of enzymeproducing one µmol of reducing sugar per min under assay conditions (Jin et al. 2011).

Culture condition of endoglucanase II expression in Pichia pastoris
The preculture of each selected colonies was prepared into 2 mL YPD medium containing 100 µg/mL ampicillin and 100 µg/mL zeocin, then incubated at 28°C for 2 d. One milliliter of preculture was added into 19 mL YPD medium containing 100 µg/mL ampicillin without zeocin in an Erlenmeyer flask and incubated at 28°C for 4 d at 250 rpm. Fedbatch culture, including sampling, was car ried out every 24 h intervals by withdrawing 4 mL of cul ture supernatant, along with the addition of 4 mL of 5× YPD medium to return the initial medium volume. The batch culture was only involving sampling every 24 h in tervals by withdrawing 4 mL culture supernatant without the addition of YPD media.

SDS-PAGE and Western Blot
Sample preparation for SDSPAGE analysis was done fol lowing the reference by Koontz (2014). One milliliter of culture from sampling was centrifuged in 12,000 rpm to separate its pellet and supernatant. The sample supernatant was added with 150 µL of 100% TCA, then vortex to ho mogenize. The sample was incubated at 4°C overnight to allow protein precipitation, then centrifuged at 12,000 rpm for 10 min to allow separation between supernatant and precipitated protein. The supernatant was removed without disrupting the pellet. The pellet was washed using 200 µL of acetone, followed by centrifugation at 12,000 for 10 min. The step of washing the pellet with acetone was repeated twice to ensure no more TCA residue was left. Pellet was allowed to dry to remove the acetone, then added with 15 µL of 5× SDSPAGE loading buffer and be ing heated for 10 min in boiling water. Samples then ana lyzed by 15% SDSPAGE gel using 90 V for 80 min to run the samples. The molecular mass was estimated from the migration distance in comparison with the prestained pro tein molecular weight marker (Thermo Scientific, USA). Western blot was performed by electrotransfer the bands from acrylamide gels onto a nitrocellulose mem brane at 90 V for 2 h. The membrane was blocked using a 10 mL blocking agent containing 1% BSA, then incubated for 1 h with gentle shaking at room temperature. The 5 µL of KPL HisDetector NickelHRP (SeraCare, USA) was added directly into the block solution, and incubation was continued for 1 h with gentle shaking. The membrane was washed three times with TBS solution containing 0.05% v/v Tween20, each time for 5 min. Finally, the detection was done by adding 5 mL KPL TMB to visualize bands of interest (Thermo Scientific, USA).

Purification of recombinant endoglucanase II
The cellfree medium culture was harvested using cen trifugation after 72 h of incubation and purified manually using NiNTA sepharose affinity chromatography column (Thermo Fisher Scientific, Massachusetts, U.S.). One milliliter of column material was washed with a fivebed volume of 20% ethanol, then washed with aquadest with the same volume. The column was then equilibrated with a fivebed volume of 50 mM PBS buffer (pH 7.4). A three bed volume of crude EGII sample was added and incu bated at 4°C for 2 h. The sample was allowed to flow through the column and collected in an Eppendorf tube, labeled as a flowthrough fraction. The column was then being washed with a threebed volume of 50 mM PBS pH 7.4. Finally, a threebed volume of an elution buffer con taining 250 mM of imidazole was added into the column. Each 250 µL was collected as one fraction in a 1.5 mL centrifuge tube. Then proteins from collected fractions were then analyzed using 15% SDSPAGE gel to identify which fraction contained purified EGII. The protein con centration was estimated using a Bovine Serum Albumin (BSA) standard curve following Carter's method (2013). The electropherogram of SDSPAGE containing EGII pro teins and a series of BSA with known concentration were subjected to ImageJ 1.53e software (Rasband, 19972018). By calculating the area under the curve (AUC) of the BSA protein band, the linear regression equation was made. The AUC of EGII was analyzed, and the value was entered into the equation to get the concentration of recombinant EGII.

Characterization of recombinant endoglucanase II
Characterization was done by following the NS method ac cording to the reference by Gusakov et al. (2011) as men tioned in section 2.2. The optimum pH and temperature for enzymatic activity assay were established using the stan dard procedure, according to the reference by Bajaj et al. (2009). To determine pH's effect on enzyme activity, var ious buffers including citrate, acetate, phosphate, tris, bi carbonate, and carbonate buffers with pH 3, 5, 6.8, 8, 9, and 10 respectively were used. To determine the optimum temperature of recombinant endoglucanase, the tempera ture used was varied (40°C, 50°C, 60°C and 70°C) under standard assay conditions. One percent of the CMC solu tion was used as a substrate to determine optimum pH and temperature. To determine the thermal stability of recom binant EGII, equal quantities of purified EGII were pre incubated for 15, 30, 45, and 60 min at optimum temper ature and pH, which already predetermined. After pre incubation, enzyme activities were measured using stan dard assay conditions. Enzyme activity can be calculated by using the formula as the following: Whereas enzymespecific activity can be calculated by using the following formula:

Selection of recombinant clones
A study conducted by (Sivashanmugam et al. 2009) sug gested that longtime storage of glycerol stock utilized for preculture preparation often leads to lowyield protein ex pression, although the colony used was known to previ ously produce a high yield of the target protein. Therefore, colony selection should be reconducted. In this study, we did colony selection from glycerol stock using colony PCR followed by enzyme activity and expression screening.
The recombinant clones from glycerol stock were spread on YPD plates containing zeocin (500 µg/mL) and ampicillin (100 µg/mL). Six random recombinant clones were selected for screening after incubation at 28°C for 36 h. Colony PCR was performed using specific primers to detect the presence of the egII gene in transformant colonies. pLIPITrCel5A plasmid carrying egII gene was used as a template for positive control while P. pastoris SMD1168H nontransformed colony was used for nega tive control. The amplified products were visualized on an agarose gel as seen in Fig. S1, and were approximately 625 bp in length. The result ensures that all selected trans formants were inserted with egII gene.    Colony selection was then continued with a qualitative and quantitative enzymatic assay, namely plate diffusion and NS assay. In plate diffusion assay, clone selection was based on the ratio between well and halo zone diameter. Whereas in NS assay, clone selection was based on the ac tivity of EGII to breakdown CMC and produce reducing sugars, by which the reducing sugars will react with the NS reagent. CMC was chosen as the substrate as it exhibits an amorph structure suitable for the hydrolysis mechanism by EGII (Biswas, 2014). This study showed that EGII was successfully expressed in all selected clones, as indicated by the halo zone formation. Clones 14 have a higher halo zone diameter, among others (Table 1 and Figure 1). Re combinant clones were then selected by determining their enzyme activities using the NS method (Table 1 and Fig  ure 2). From those analyses, clones 13 showed higher endoglucanase activity compared to other colonies.

Heterologous expression of Endoglucanase II by recombinant clones in YPD media
The expression profile of recombinant protein produced by six recombinant clones in YPD media was analyzed through 15% SDSPAGE gel after 96 h of incubation, in cluding sampling every 24 h intervals ( Supplementary Fig  ure 2). Clone 2 was chosen for further expression anal ysis as it showed the thickest band of interest at a size of around 50 kDa with lesser nontarget bands. More over, clone 2 also showed a high enzymatic activity, which refers to plate diffusion and NS assay (Figure 1 and Fig  ure 2). Clone 2 was then used for recombinant EGII pro duction through fedbatch and batch fermentation (Fig  ure 3). The production of recombinant EGII through fed batch and batch fermentation was done to compare both methods and know which method is less timeconsuming and produce a higher level of the target protein. Feeding was included in fedbatch fermentation by adding a new 5×YPD medium right after sampling, whereas batch fer mentation only included sampling. The maximum num ber of cells achieved through fedbatch fermentation was 3.76x10 7 cells/mL after 72 h of incubation. Meanwhile, the maximum number of cells achieved through batch fer mentation was only 2.71x10 7 cells/mL after 48 h of in cubation. The protein expression was done through fed batch fermentation (Figure 4a) showed a thicker band of interest in each 24 h intervals compared to batch fermen tation (data not shown). More protein bands were present in the protein expression profile from batch fermentation with less than 50 kDa in size. MacauleyPatrick et al. (2005) suggest that the secreted recombinant proteins can be proteolytically degraded in the culture medium. This might happen due to cellbound proteases, extracellular proteases, and/or intracellular proteases from lysed cells.
The problems due to proteolysis can be foreseen in the recombinant protein production: (a) reduction of product yield when product is degraded, and (b) reducing biolog ical activity when the product is truncated. In this study, protease production may occur due to the insufficiency of nutrients that lead to cell lysis. In comparison, fedbatch fermentation supplied enough nutrients added during pro duction periods, which promoted optimal growth of cul ture and higher protein expression (Hadiyanto et al. 2013). Figure 4a showed that P. pastoris expressed the highest concentration of recombinant EGII after 72 h of incuba tion. Thereby, fedbatch fermentation for 72 h of incuba tion was used for further production of recombinant EGII. The expression profile of recombinant EGII by clone 2 every 24 h intervals was visualized through SDSPAGE analysis (Figure 4a). The maximum expression of recom binant EGII was reached after 72 h with 3.49 mg/L pro tein secreted from 20 mL of fermentation culture. The growth of transformants reached stationary phase after 72 h; thereafter, it entered the death phase. The estimated molecular mass of recombinant EGII was 52 kDa. EGII is a glycoprotein with 397 amino acids and has a molecu lar weight of 40 kDa without glycosylation (Garvey et al. 2014). However, its molecular weight may increases up to 48 kDa with native glycosylation done by T. reesei (Ak barzadeh et al. 2014). Sun et al. (2018) reported that the expression system by P. pastoris might increase up to 4 kDa of molecular weight due to glycosylation. In this study, the SDSPAGE analysis of recombinant EGII pro tein expressed in P. pastoris showed a thick band with a slightly larger molecular mass than that of native EGII from T. reesei. However, there was not enough evidence to show that the increase in molecular mass is due to gly cosylation. Treatment of recombinant EGII protein with endoglycosidase H may be required as supporting data to signify if glycosylation takes place.

Purification and confirmation of recombinant endoglucanase II
The culture supernatant was purified manually using Ni NTA sepharose resin. The column was eluted with an iso cratic buffer containing 250 mM imidazole to release the recombinant protein bound to the resin. All fractions col lected were analyzed using SDSPAGE. Among 12 elu tion fractions collected, the elution fraction 3 was known to have the thickest band of interest, indicating the high est concentration of purified protein (data not shown). All samples collected from sampling for each 24 h and elu tion fraction 3 were analyzed using SDSPAGE followed by Western blot to confirm the recombinant EGII protein (Figure 4b). The result exhibited a single band of 52 kDa for each lane, which corresponds to a theoretical molec ular mass of EGII recombinant proteins expressed in P. pastoris (Bai et al. 2016). The concentration of purified recombinant EGII was then quantified using a BSA stan dard curve followed by ImageJ analysis, resulting in 0.21 mg/mL of purified EGII.

Characterization of recombinant endoglucanase II
The endoglucanase activity of recombinant EGII was as sayed at different pH (3.0, 5.0, 6.8, 8.0, 9.0, 10.0). The results showed that the recombinant EGII exhibited op timum activity at pH 5.9 ( Figure 5). Activity above pH 8.0 was negligible as it did not show enzymatic activity. The native endoglucanase by T. reesei also showed the same optimum pH (Li et al. 2013). The study conducted by Boonvitthya et al. (2013) revealed that the production of crude EGII in P. pastoris was around 10 U/ml under a controlled condition with the optimum pH at 5 to 6 and the temperature of 4060°C. Endoglucanase activity measurement was using vari ous temperatures ranging from 30°C to 70°C. The opti mum temperature was found to be 40°C to 50°C, as shown in Figure 6. Native endoglucanase also showed a similar range of optimum temperature, which was at 45°C to 55°C (Kamal et al. 2017). The effect of temperature optimum on enzyme stability is shown in Figure 6. Preincubation was done at 40°C and 50°C with different time intervals up to 60 min. About 89% of endoglucanase activity was retained at 40°C after 60 min, and more than 80% of en doglucanase activity was retained after 60 min at 50°C (Figure 7). Based on the literature, native EGII secreted from T. reesei retained 60% of its enzymatic activity at 50   after 40 min of incubation (Kamal et al. 2017). This ev idence indicated the thermal stability of our recombinant EGII is slightly improved compared to native EGII derived from T. reesei as it can retain 80% endoglucanase activity after incubation at 50°C for 60 min.
The result of the thermal stability assay obtained was slightly lower than the study conducted by Samanta et al. (2012), which showed that recombinant EGII could retain relative endoglucanase activity up to 90% at 50°C. How ever, this study reported that purified EGII protein showed 24.3 U/mL of enzyme activity after preincubation at 50°C in pH 5.0, which gave the specific activity of 115.7 U/mg. The specific activity of our recombinant EGII is higher than the result reported by using the E. coli expres sion system (Nakazawa et al. 2008).
Nglycosylation, a posttranslational modification generally occurred in fungi, has important roles in en zyme stability. A study conducted by Han et al. (2020) demonstrated that the addition of Nglycosylation at par ticular amino acid sites has successfully increased the ther mal stability of recombinant endoglucanase in P. pas toris. Optimized stabilization is associated with entropy which mostly dependent on glycosylation sites position (ShentalBechor and Levy 2008). Generally, glycans at tached to flexible regions within random coils would re strict the conformational space and promote entropic re duction to increase conformational stability at high tem peratures (Dotsenko et al. 2016; Adney et al. 2009). This additional targeted Nglycosylation was deserved to be adopted in our future study to improve the thermal stability of our recombinant EGII.
Optimization of fermentation is also required to im prove the expression of P. pastoris by exploring various carbon and nitrogen sources to enhance microbial growth and protein expression. Further study to maintain a spe cific growth rate is also highly suggested by implement ing a short carbonstarving period through an exponential feeding strategy.

Conclusions
Recombinant EGII proteins were successfully expressed in the P. pastoris expression system with an optimum in cubation time of 72 h using fedbatch fermentation. Char acterization of recombinant EGII using SDSPAGE and Western blot analyses showed that the EGII protein has a molecular weight of approximately 52 kDa. The enzy matic assay demonstrated that the crude enzyme has op timum activity at pH 5.0, which results in 16.3 and 14.6 U/mL activity at a temperature of 40°C and 50°C, re spectively. The specific activity of recombinant EGII re sulted in 115.7 U/mg in its optimum pH and temperature. Moreover, recombinant EGII can maintain 89% of its en doglucanase activity at 40°C and more than 80% at 50°C for 60 min, indicating the improvement in thermal stability compared to native EGII derived from T. reesei.