Biotechnology Letters 26: 223–227, 2004. © 2004 Kluwer Academic Publishers. Printed in the Netherlands.
223
Fractionation of Trichoderma reesei cellulases by hydrophobic interaction chromatography on phenyl-Sepharose Cândida T. Tomaz∗ & João A. Queiroz Department of Chemistry, University of Beira Interior, 6201-001 Covilhã, Portugal ∗ Author for correspondence (Fax: +351.275.319730; E-mail:
[email protected]) Received 17 October 2003; Revisions requested 13 November 2003; Revisions received 27 November 2003; Accepted 28 November 2003
Key words: cellulases, fractionation, hydrophobic interaction chromatography
Abstract Trichoderma reesei cellulase complex was fractionated using hydrophobic interaction chromatography with a phenyl-Sepharose column. Using a linear gradient of ammonium sulphate in the eluent buffer, a selective separation of endoglucanases was obtained at 15 ◦ C with a four-fold increase in specific activity.
Introduction Hydrophobic interaction chromatography (HIC) has been successfully employed for the fractionation and purification of proteins and other biomolecules (Queiroz et al. 2001). It takes advantage of hydrophobic interactions between non-polar regions on the surface of the proteins and hydrophobic ligands of the adsorbent to achieve separation. In HIC, protein retention can be enhanced by increasing the concentration of salting-out salts in the aqueous mobile phase. The adsorbed proteins are eluted by step-wise or gradient elution at decreasing salt concentration of the eluent (Melander & Horváth 1977). The nature and concentration of the salt thus play an important role in determining the magnitude of retention and selectivity of the HIC process (Oscarsson & Kårsnäs 1998). The pH, additives and the temperature (Hjérten et al. 1974) can also affect the retention of proteins in HIC. The type of stationary phase is the other major parameter that has to be considered to obtain success in a HIC process. The most common ligands used in HIC are linear chain alkanes and simple aromatic groups, such as phenyl. This ligand shows a mixed hydrophobic and aromatic interactions that can be used to obtain interesting results in HIC. In this work, a HIC study was undertaken to fractionate Trichoderma reesei cellulases using a phenylSepharose column. The cellulolytic enzyme system
from T. reesei consists of three different types of enzymes: cellobiohydrolases (EC 3.2.1.91), endoglucanases (EC 3.2.1.4) and a β-D-glucosidase (EC 3.2.1.21), all of which act synergistically on crystalline cellulose, during the hydrolysis of cellulose (Béguin & Aubert 1994). The evidence that some cellulases have hydrophobic sites on their surface which may bind indigo owing to hydrophobic interactions (Gusakov et al. 2000), suggested that hydrophobic supports can be used in the chromatographic fractionation of cellulases. Despite HIC not being widely used for enzyme purification, we have already reported a selective fractionation of β-glucosidase from T. reesei crude extract (Tomaz & Queiroz 1999, Tomaz et al. 2002).
Materials and methods Materials Phenyl-Sepharose 6 FF was purchased from Amersham Pharmacia Biotech. The filter paper was Whatman No.1. Hydroxyethylcellulose (medium viscosity, degree of substitution 0.8) was from Fluka. D-Cellobiose (98%, predominantly β) was obtained from Aldrich. SDS, acrylamide, N,N methylenebisacrylamide, ammonium persulphate and Coomassie Brilliant Blue G-250 were from Sigma.
224 N,N,N ,N -Tetramethylenediamine was from Aldrich. All other reagents were of analytical grade. Enzyme preparation A cellulolytic enzyme complex of Trichoderma reesei (Celluclast 1,5L) from Novo Nordisk (Denmark) was used.
Gel electrophoresis SDS-PAGE analysis was carried out in a discontinuous buffer system using a Hoefer SE 600 vertical unit (Amersham Pharmacia Biotech), according to the manufacturer’s instructions. The gel was 10% (v/v) polyacrylamide-N,N-methylenebisacrylamide (30:1) and was stained with Coomassie Brilliant Blue R-250.
Protein and enzymatic activity assays Results and discussion Protein concentration was determined by the Bradford method with BSA as standard. The filter paper activity (FPA) method was used for determination of total cellulase activity (Mandels et al. 1976). It was carried out at pH 4.8, 50 ◦ C on 50 mg filter paper strips for 1 h. The reaction was stopped by placing the tubes in a boiling water bath for 10 min and after centrifugation; the reducing sugars were measured with dinitrosalicylic acid reagent, using glucose as standard. Endoglucanase activity was measured using 1% hydroxyethylcellulose and the resulting reducing sugars were quantified by dinitrosalicylic acid method. Cellobiase (β-glucosidase) activity was determined using cellobiose as substrate. The reaction was performed at 50 ◦ C, for 30 min, using 50 mM citrate buffer, pH 4.8 and the glucose released was measured by the glucose oxidase method. Chromatographic method Chromatography was performed using an AmershamPharmacia FPLC system. Initially, the separation was performed by a gel filtration on a Sephadex G-25M gel (35 × 1.6 cm intern. diam. column), equilibrated with acetate buffer 25 mM, pH 4.8, at 39 ml h−1 . After an ultrafiltration step using an Amicon cell with a polyethersulphone membrane of 5000 Da (NMWC), the concentrated enzyme solution was fractionated by HIC, on a phenyl-Sepharose 6 FF column. The gel was packed in an isothermal column (4.8 × 1 cm intern. diam.) and equilibrated with 25 m M acetate buffer, pH 4.8, with 10% (w/v) ammonium sulphate, at 24 ml h−1 and at different temperatures (15 to 35 ◦ C). Fractions (1 ml) were collected and the protein concentration and the activity towards filter paper, hydroxyethylcellulose and cellobiose were determined.
A gel filtration on Sephadex G-25 M was the first step used in the purification procedure of the crude extract of T. reesei. The fractions with cellulolytic activity (recovery of 94% FPA) were pooled and concentrated by an ultrafiltration step and then subjected to HIC on phenyl-Sepharose. Protein retention and selectivity in HIC are highly dependent upon the chemical nature of the salt used to promote binding (Oscarsson & Kårsnäs 1998). Salts useful in this regard are called antichaotropic and belong to the Hofmeister (lyotropic) series of ions for the precipitation of proteins or for their positive influence in increasing the molal surface tension of water (Melander & Horváth 1977, Melander et al. 1984). The effect of the different salts on the retention of the cellulases was performed using NaCl, which has a low ability to enhance hydrophobic interactions, and sodium sulphate and ammonium sulphate that promote a high retention of enzymes, since they are at the beginning of the lyotropic series. The hydrophobic interactions are known to increase also upon increasing the ionic strength of the medium (Oscarsson & Kårsnäs 1998). The experimental results show that even using the highest concentration of NaCl (4 M) in the mobile phase, there was not a total binding of the cellulases due to the small surface tension increment for this salt. For ammonium sulphate and sodium sulphate, the total retention of the cellulases was promoted with a lower concentration of these salts (15% and 10% w/v, respectively), since they present high increments of molal surface tension (results not shown). A selective fractionation of the cellulases in three peaks was obtained with respectively, 7% (w/v) ammonium sulphate (Tomaz et al. 2002) and 5% (w/v) sodium sulphate in the eluent buffer, using a stepwise elution. The influence of the temperature on the cellulases fractionation with phenyl-Sepharose was investigated using a linear gradient elution, from 10% to 0% (w/v) ammonium sulphate in the eluent buffer, at different
225
Fig. 1. Hydrophobic interaction chromatography on phenyl-Sepharose 6 FF at different temperatures: (A) 15 ◦ C, (B) 23 ◦ C, (C) 30 ◦ C and (D) 35 ◦ C. Buffer: 25 m M acetate, pH 4.8 containing 10% (w/v) (NH4 )2 SO4 . The elution was performed by a linear decreasing gradient of (NH4 )2 SO4 (- - -) from 10% to 0% (w/v). Sample size injected: 200 µl (3.9 mg protein).
Table 1. Cellobiase and endoglucanase specific activities assayed in each peak collected after hydrophobic interaction chromatography on phenyl-Sepharose, at 15 ◦ C (Figure 1A) and 23 ◦ C (Figure 1B), using a linear decreasing gradient of (NH4 )2 SO4 from 10% to 0% (w/v). HIC
Specific celobiase Specific endoglucanase activity (U mg−1 ) activity (U mg−1 )
Feed 15 ◦ C Peak 1 2 3 4 23 ◦ C Peak 1 2 3 4
0.09
29
– 0.12 – –
–
– 0.12 – –
– 11 95 66
− Activities not detected.
9 116 79
temperatures (15 to 35 ◦ C). In HIC, the increase of the temperature generally enhances protein retention (Hjertén et al. 1974). The chromatogram profiles obtained (Figure 1) show that the retention of T. reesei cellulases was affected in the range of the studied temperatures. At 15 ◦ C, four peaks were obtained (Figure 1A) suggesting a selective fractionation of the cellulases but, on increasing the temperature above 30 ◦ C, a progressive reduction of the peak 3 and an increase of peak 4 were observed (Figure 1C). Thus, at 35 ◦ C the chromatogram profile shows a fractionation only in three peaks (Figure 1D). This means that higher temperature promotes an increase in the retention time of some cellulases due to the enhancement of the hydrophobic interactions. The measurement of the cellulase activities (Table 1) also shows some differences between each temperature experiment, in particular on endoglucanases activity. In fact, at 35 ◦ C when the fractionation of the cellulases in three peaks was obtained (Figure 1D), the endoglucanases activity
226
Fig. 2. Ten % SDS-PAGE analysis of the main fractions obtained after hydrophobic interaction chromatography on phenyl-Sepharose 6 FF, at 23 ◦ C (Figure 1B), using a linear decreasing gradient of (NH4 )2 SO4 from 10% to 0% (w/v). Lanes: 1 – Crude extract; 2 – peak 1; 3 – peak 2; 4 – peak 3; 5 – peak 4; 6 – molecular weight markers (phosphorilase b, Mr 94 kDa, bovine serum albumin, Mr 67 kDa, ovalbumin, Mr 43 kDa, carbonic anhydrase, Mr 30 kDa, soybean trypsin inhibitor, Mr 20.1 kDa, lactoalbumin, Mr 14.4 kDa).
was found only in the third peak fraction. On the other hand, at lower temperatures (Figures 1A and 1B), the endoglucanases activity appears in peak 3 and peak 4, corresponding to different forms of endoglucanases. In this case, the decrease of the temperature promotes a diminution in the hydrophobic interactions and a slight decrease in the retention time of the less hydrophobic endoglucanases that are found in peak 3. The specific activity of the endoglucanases was increased compared to injected fractions. At 15 ◦ C (Figure 1A), a four-fold and a three-fold increase in specific activity were detected, respectively, in peak 3 and 4 (Table 1). At 23 ◦ C (Figure 1B), a three-fold increase for peak 3 and a two-fold increase in specific activity for peak 4 were found (Table 1). The SDS-PAGE analysis (Figure 2) showed a selective separation of the different isoforms of endoglucanases (Mr 48–55 kDa) in peaks 3 and 4, despite the presence of some contaminants mainly in lane 5. In fact, the endoglucanases appear to be more hydrophobic than cellobiohydrolases and βglucosidase, since they are the last to be eluted in the HIC experiments, as previously observed (Tomaz &
Queiroz 1999). Also, Gusakov et al. (2000) showed a certain correlation between the washing performance of cellulases and the overall percentage of the surface hydrophobic residues. They found that endoglucanases show a better denim-washing performance than cellobiohydrolases due to the high quantity of these hydrophobic residues that may promote a better indigo binding. In all experiments no type of cellulolytic activity was found in peak 1 and β-glucosidase activity was detected only in peak 2 with a slight enhancement in specific activity compared with injected fractions (Table 1), as obtained previously (Tomaz et al. 2002). The total cellulolytic activity was decreased in all peaks probably due to the fractionation of the different type of the cellulases that prevents enzyme synergism. If the cellulases are separated, the action between them is not possible and the total cellulolytic activity decreases. Although total cellulolytic activity does not provide information on the levels of individual enzymes or on their role in the hydrolysis of the cellulose, this method can be useful to give information about variation of this process due to the presence or absence of different type of cellulases. In conclusion, the results obtained suggest that the purification of cellulases can be achieved using HIC on phenyl-Sepharose since the correct choice of the ionic strength of the mobile phase and temperature were done. In fact, using an appropriate temperature, salt and salt concentration, a selective fractionation of β-glucosidase and endoglucanases were obtained with a remarkable increase in the specific activity. Despite the role of temperature in HIC is not simple, this parameter can be manipulated to achieve the correct separation of the cellulases.
References Béguin P, Aubert J-P (1994) The biological degradation of cellulose. FEMS Microbiol. Rev. 13: 25–58. Gusakov AV, Sinitsyn AP, Berlin AG, Markov AV, Ankudimova NV (2000) Surface hydrophobic amino acid residues in cellulose molecules as a structural factor responsible for their high denim-washing performance. Enzyme Microb. Technol. 27: 664–671. Hjertén S, Rosengren J, Påhlman S (1974) Hydrophobic interaction chromatography. The synthesis and the use of some alkyl and aryl derivatives of agarose. J. Chromatogr. 101: 281–288. Mandels M, Andreotti R, Roche C (1976) Measurement of saccharifying cellulose. Biotechnol. Bioeng. Symp. 6: 21–33. Melander W, Horváth C (1977) Salt effects on hydrophobic interaction in precipitation and chromatography of proteins: an
227 interpretation of lyotropic series. Arch. Biochem. Biophys. 183: 200–215. Melander W, Corradini D, Horváth C (1984) Salt mediated retention of proteins in hydrophobic interaction chromatography. Application of solvophobic theory. J. Chromatogr. 317: 67–85. Oscarsson S, Kårsnäs P (1998) Salt-promoted adsorption of proteins onto amphiphilic agarose-based adsorbents. II. Effects of salt and salt concentration. J. Chromatogr. A 803: 83–93.
Queiroz JA, Tomaz CT, Cabral JMS (2001) Hydrophobic interaction chromatography of proteins. J. Biotechnol. 87: 143–159. Tomaz CT, Queiroz JA (1999) Studies on the chromatographic fractionation of Trichoderma reesei cellulases by hydrophobic interaction. J. Chromatogr. A 865: 123–128. Tomaz CT, Duarte D, Queiroz JA (2002) Comparative study for cellulases fractionation on some hydrophobic interaction chromatography adsorbents. J. Chromatogr. A 944: 211–216.
