Multiple amino acid substitutions significantly improve the thermostability of feruloyl esterase A from Aspergillus niger

Bioresource Technology 117 (2012) 140–147

Contents lists available at SciVerse ScienceDirect

Bioresource Technology journal homepage: www.elsevier.com/locate/biortech

Multiple amino acid substitutions significantly improve the thermostability of feruloyl esterase A from Aspergillus niger Shuai-Bing Zhang a,b, Xiao-Qiong Pei a, Zhong-Liu Wu a,⇑ a b

Key Laboratory of Environmental and Applied Microbiology, Chengdu Institute of Biology, Chinese Academy of Sciences, P.O. Box 416, Chengdu 610041, PR China Graduate University of the Chinese Academy of Sciences, Beijing 100049, PR China

h i g h l i g h t s

g r a p h i c a l a b s t r a c t

" Half-life of thermal inactivation of

AnFaeA increases from 15 to >4000 min. " Twelve beneficial mutations lead to increased thermostability. " Release of ferulic acid from corn stalk increases to >300% using mutant AnFaeA.

a r t i c l e

i n f o

Article history: Received 10 January 2012 Received in revised form 23 March 2012 Accepted 10 April 2012 Available online 26 April 2012 Keywords: Feruloyl esterase Aspergillus niger Thermostability Pichia pastoris Directed evolution

a b s t r a c t Feruloyl esterase A from Aspergillus niger (AnFaeA) is one of the most important feruloyl esterases of industrial relevance. Previous work aided by the PoPMuSiC algorithm has identified two beneficial mutants (D93G and S187F) with thermostabilization effect. In this work, twelve additional amino acid substitutions were identified to be beneficial to the thermostability of AnFaeA after screening a random mutagenesis library constructed in Pichia pastoris. Combination of these mutations resulted in a mutant with 80% residual activity after heat treatment at 90 °C for 15 min and a half-life increasing from 15 min to >4000 min at 55 °C. The thermostabilized mutant displayed significantly enhanced performance compared to the parental AnFaeA when applied to the treatment of steam-exploded corn stalk at 60 °C together with an xylanase, demonstrating its great potential for industrial application. Ó 2012 Elsevier Ltd. All rights reserved.

1. Introduction Feruloyl esterases (FAEs) (EC 3.1.1.73) are critical enzymes involved in the complete degradation of lignocellulose, which cleave the ester bonds between hydroxycinnamic acids and arabinoxylans or certain pectins present in plant cell walls (Faulds, 2010; Wong, 2006). Four types (types A, B, C and D) of FAEs were classified based ⇑ Corresponding author. Address: Chengdu Institute of Biology, Chinese Academy of Sciences, 9 South Renmin Road, 4th Section, Chengdu, Sichuan 610041, PR China. Tel./fax: +86 28 85238385. E-mail address: [email protected] (Z.-L. Wu). 0960-8524/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biortech.2012.04.042

on their substrate utilization and primary sequence identity (Crepin et al., 2004), and a recent report proposed a new classification of 12 FAE families based on a combination of published experimental data and the interplay of descriptor-based computational analysis with pharmacophore modeling (Udatha et al., 2011). Up to now, FAEs have been widely applied to the degradation of lignocellulose, the manufacture of pulp and paper, as well as the food and pharmaceutical industries (Fazary and Ju, 2008). The feruloyl esterase A (AnFaeA) from Aspergillus niger is one of the most investigated FAEs of industrial significance, which has attracted much attention from researchers (Faulds, 2010; Hermoso et al., 2004; Juge et al., 2001; Mcauley et al., 2004; de Vries et al.,

141

S.-B. Zhang et al. / Bioresource Technology 117 (2012) 140–147

1997). Recently, we have reported the identification of thermostabilization mutations of an AnFaeA from A. niger CIB 423.1 using a rational design approach assisted with the PoPMuSiC algorithm. The resulting double mutant D93G/S187F has a 9.6-fold increase of half-life of thermal inactivation at 50 °C compared to the wildtype (Zhang and Wu, 2011). However, an economical industrial process would demand more robust catalyst to reduce the enzyme load, as well as to increase the operational flexibility. For the hydrolysis of lignocellulosic substrates, in particular, thermostable enzymes are highly desired to facilitate the application of higher reaction temperature, which would decrease the possibility of microbial contamination, promote the disorganization of the raw materials, and improve enzyme penetration. To further enhance the thermostability of AnFaeA, in the current study, we applied a directed evolution approach to identify more thermo-stabilizing mutations. This approach, which mimics natural evolution of enzymes on a laboratory timescale, has proved to be extremely valuable in improving various features of proteins, including thermostability (Bloom and Arnold, 2009; Pei et al., 2011; Zhang et al., 2010). To the best of our knowledge, no previous work has been reported on the directed evolution of FAEs. Only recently, a high-throughput screening system for FAEs has been evaluated based on a chromogenic substrate, 2-chloro-4-nitrophenyl ferulate (CNPF) with the coefficient of variance of 10.3% as determined for the single sequence library of the AnFaeAD93G/S187F mutant (Zhang et al., 2012). This screening system was adapted in this work to test a random library constructed in Pichia pastoris using the same double mutant as the parental enzyme. Multiple amino acid substitutions were identified to be beneficial for the thermostability of AnFaeA. The thermostabilized mutant AnFaeA was successfully applied to the release of ferulic acid from steam exploded corn stalk with enhanced efficiency. 2. Methods 2.1. Materials and chemicals The substrate CNPF was synthesized as previously described with >99% purity (Zhang et al., 2012). Methyl feulate (MFA) was purchased from Alfa Aesar (Tianjin, China) with 98% purity. Steam exploded corn stalk was kindly donated by Professor Hongzhang Chen at the Institute of Process Engineering, Chinese Academy of Sciences (Yang et al., 2009). Oligonucleotides were synthesized by Invitrogen Life Technologies (Shanghai, China) in PAGE-purified grade. The AnFaeA mutant, M11, containing 11 additional amino acid changes (L14F, T35I, K37I, T57I, T63I, A140T, Q121H, S163T, Q177H, V178A, and Q185R) compared with the parental double mutant D93G/S187F was de novo synthesized by Sangon (Shanghai, China). Therefore, M11 contains a total of 13 mutations compared with the native AnFaeA from A. niger CIB 423.1 (EMBL accession No. FJ430154). Restriction enzymes were purchased from New England Biolabs (Beverly, MA). DNA sequencing was carried out in Invitrogen (Shanghai, China). All other reagents were obtained from general commercial suppliers and used without further purification.

The error prone PCR (epPCR) product was recovered and digested with NotI and EcoRI, and then ligated into pGAPZaA vector (Invitrogen, Carlsbad, CA, USA) digested with the same restriction enzymes. The ligation product was transformed into E. coli DH5a using a MicroPulser Electroporator (Bio-Rad Laboratories, Hercules, CA, USA) to maintain and propagate the library. The transformants were grown on low salt Luria–Bertani (LB) plates containing 1% (w/v) tryptone (Oxoid, Hampshire, England), 0.5% (w/v) yeast extract (Oxoid, Hampshire, England), 0.5% (w/v) sodium chloride and 2% (w/v) agar, supplemented with 25 lg zeocin/ml. Then the transformants were harvested and the plasmids were extracted with the TIANprep Mini Kit (Tiangen, Tianjin, China). The extracted plasmids were digested with NotI and EcoRI, and ligated into the shuttle vector pPIC9K (Invitrogen, Carlsbad, CA, USA) digested with the same two restriction enzymes. The ligation product was electro-transformed into E. coli DH5a and the transformants were grown in LB plates supplemented with 100 lg ampicillin/ml. The recombinant plasmids were extracted and linearized with PmeI and electrotransformed into P. pastoris KM71 (Invitrogen, Carlsbad, CA, USA). 2.3. Screening for thermostable mutants and sequence analysis The transformants were cultivated on minimal dextrose (MD) medium agar plates (2% (w/v) glucose, 1.34% (w/v) yeast nitrogen base with ammonium sulfate without amino acids (YNB), 4  105% (w/v) biotin, 2.0% agar (w/v)) for two days at 30 °C. Single colonies were picked and cultivated in 150 ll buffered glycerolcomplex medium (BMGY) (1% (w/v) yeast extract, 2% (w/v) tryptone, 1.34% (w/v) YNB, 1% (v/v) glycerol, 100 mM potassium phosphate (pH 6.0), 4  105% biotin (w/v)) on 96-well microplates with shaking at 280 rpm for 48 h at 30 °C on an INFORS Multitron incubator shaker (INFORS HT, Swissland) with the parental AnFaeA as the control. The plates were centrifuged at 1800g for 10 min and the supernatants were discarded. The expression of AnFaeA mutants were induced by adding 150 ll buffered methanol-complex medium (BMMY) (same as BMGY except using 0.5% (v/v) methanol instead of 1% (v/v) glycerol as an inducer) to the cell pellets and the cultivation was continued for 36 h. After centrifugation at 1800g for 10 min, 20 ll supernatant in each well was transferred into two duplicate 96-well plates, and one plate was pretreated at 63 °C for 30 min in a BINDER oven (BINDER, Germany). Then the reaction mixture in each well was prepared by adding 170 ll of sodium phosphate buffer (100 mM, pH 6.4) containing 2.5% (v/v) Triton X-100 and 10 ll of DMSO solution containing the substrate CNPF (20 mM) (Zhang et al., 2012). After 15 min incubation at 40 °C with shaking at 100 rpm, the release of 2-chloro-4-nitrophenol was measured at 425 nm in a Thermo Scientific Varioskan Flash microplate reader (Thermo Scientific, USA). Each of the transformants expressing thermostable mutants was cultivated in yeast extract peptone dextrose medium (YPD) containTable 1 Sequences of oligonucleotides used for site-directed mutagenesis. Target sites

Sequences of oligonucleotidesa

T57I

50 -CTCCGCGACGACATCAGCAAAGAAATTATC-30

V178A

50 -GATGCGTTCCAGGCCTCGAGCCCGG-30

K37I

50 -CATCGACTACTATCATAGGAGAGAAAATTTAC-30

G69A

50 -CCGTGGCACTGCCAGTGACACAAACC-30

C235S

50 -GAAGTACAGTGCAGTGAGGCACAGGG-30

50 -GATAATTTCTTTGCTGATGTCGTCGCGGAG-30

2.2. Construction of random mutagenesis library The plasmid pGAPZaA-faeA(D93G/S187F) encoding the AnFaeA double mutant D93G/S187F (Zhang and Wu, 2011) was used as the template. Random mutagenesis was performed with the Genemorph II Random Mutagenesis Kit (Stratagene, La Jolla, CA, USA) following the manufacturer’s protocol with primers 50 -TAGGAGGTGAATTCGCCTCCACGCAAGGCATCTC-30 (forward) and 50 TAGGAGGTGCGGCCGCTTACCAAGTACAAGCTCCGCTCG-30 (reverse).

50 -CCGGGCTCGAGGCCTGGAACGCATC-30 50 -GTAAATTTTCTCTCCTATGATAGTAGTCGATG-30 5’-CCGTGGCACTGCCAGTGACACAAACC-30 50 -CCCTGTGCCTCACTGCACTGTACTTC-30 a

The nucleotide changes are underlined.

142

S.-B. Zhang et al. / Bioresource Technology 117 (2012) 140–147

ing 1% (w/v) yeast extract, 2% (w/v) peptone and 2% (w/v) dextrose. The genomic DNA was extracted using a yeast genome DNA extraction kit (Tiangen, Tianjin, China). The encoding sequences of AnFaeA mutants were amplified by PCR with Phusion High-Fidelity DNA Polymerase (Finnzymes, Espoo, Finland) using the same primers as for epPCR. The PCR products were ligated into pGAPZaA vector for sequence analysis to confirm the nucleotide changes. 2.4. Site-directed mutagenesis Site-directed mutagenesis was performed with the Quikchange Mutagenesis Kit (Stratagene, La Jolla, CA, USA) following the manufacturer’s protocol using the plasmid pGAPZaA-faeA (D93G/ S187F) as the template with primers containing the corresponding mutations (Table 1). To construct the mutant M12, the plasmid encoding mutant M11 was applied as the template. The successful introduction of the desired mutations was confirmed by DNA sequencing at Shanghai Invitrogen Life Technologies. 2.5. Expression and purification of the parental AnFaeA and the mutants To facilitate the expression and purification of the target protein, the parental AnFaeA and the mutants Z11, M11 and M12 were constitutively expressed in P. pastoris KM71 using the pGAPZaA vector as previously described (Zhang and Wu, 2011). Briefly, each of the transformants was cultivated in YPD agar plate supplemented with 1 M sorbitol and 100 lg zeocin/ml. After incubation at 30 °C for 72 h, single colonies were picked and inoculated into 100 ml YPD medium, and incubated for 48 h at 30 °C with gyratory shaking at 230 rpm. Supernatants of the culture were filtered through a 0.45 lm MF-Millipore membrane (Millipore, Bedford, MA, USA), and purified using a Bio-scale Mini UNOsphere Q column (Bio-Rad, Hercules, CA, USA) as described previously (Zhang and Wu, 2011). The enzymes were purified with around 10-fold increased in specific activity using anion exchange columns (Supplementary data, Fig. S1).The purified enzymes were confirmed by both enzymatic activity assay and SDS–PAGE. Protein estimations were done with the BCA Protein Assay Kit (Beyotime, Shanghai, China) with bovine serum albumin as a standard (Beyotime, Shanghai, China). 2.6. Measurement of thermostability and kinetic parameters All the spectrophotometric assay of feruloyl esterase activity using the substrate CNPF was carried out on 96-well plates under the same conditions as described in Section 2.3 for the library screening unless mentioned otherwise. The assay with substrate MFA was performed in 1 ml sodium phosphate buffer (100 mM, pH 6.4) containing 1 mM MFA and 12.5 lg enzyme. After 10 min incubation at 40 °C, the reaction was immediately quenched by adding 0.4 ml of acetic acid and analyzed using HPLC. HPLC was conducted with a Shimadzu Prominence LC-20AD system connected to a PDA-detector using a Luna C18 reverse-phase column (Phenomenex, Torrance, CA, USA). A mobile phase of 80% methanol in water containing 0.5% (v/v) trifluoroacetic acid was used and absorption at 280 and 325 nm was recorded. Temperature or pH dependence of activities was measured using the substrate MFA. To measure irreversible thermal inactivation, samples of crude or purified enzymes were incubated at 55 °C. Aliquots were removed every 5 min, cooled in ice and assayed using the substrate CNPF. The thermal inactivation half-life (t1/2) of each enzyme was determined by plotting ln(A425) values versus incubation time and deduced by linear regression. For mutant Z11 containing the C235S mutation, varied incubation temperatures of 55, 70, 75, 80, 85, and 90 °C were applied.

The thermal inactivation patterns of the parental AnFaeA and mutants were tested by incubating 0.5 mg/ml purified enzyme samples at temperatures ranging from 55 to 99 °C for 15 min in a MJ Mini™ 48-Well Personal Thermal Cycler (Bio-Rad, Hercules, CA). The samples were then placed on ice and the remaining activities were assayed with the substrate CNPF or MFA. Steady-state kinetic parameters of purified parental AnFaeA and the mutants were determined with different concentrations of CNPF on a 96-well microplate for 10 min. Data were fitted to the Michaelis–Menten equation using Graph-Pad Prism v5.0 (GraphPad Software, San Diego, CA, USA) to generate estimates of Km and kcat values. 2.7. Determination of melting temperatures (Tm) of parental AnFaeA and variants The Tm values were measured using a differential scanning VPDSC microcalorimeter (Microcal Inc., Northampton, MA). The experiments were performed with 1.0 mg protein/mL in 0.1 M sodium phosphate (pH 6.4). Thermal denaturation scans were performed with freshly prepared protein solutions using a temperature scan rate of 1 °C/min. Protein samples were dialyzed for more than 20 h against 0.1 M sodium phosphate (pH 6.4). Samples were filtered through 0.22 lm pore membranes following dialysis and were degassed before measurements. 2.8. Release of ferulic acid from steam exploded corn stalk The parental AnFaeA and the most thermostabilized mutant M12 were each applied to liberate ferulic acid from steam exploded corn stalk synergistically with an purified xylanase, which was a mutant of xylanase XT6 designated as FC06T) (Zhang et al., 2010). Two-gram of steam exploded corn stalk was suspended in 50 mL sodium phosphate buffer (100 mM, pH 6.4) containing xylanase (0.2 mg/ml) and AnFaeA (0.05 mg/ml). The hydrolysis reactions were performed at 60 °C and monitored for 7 h. To measure the amount of ferulic acid released, 500 lL of the reaction solution were withdraw and quenched by the addition of 200 ll glacial acetic acid. Ten microliter of samples were analyzed on a Shimadzu Prominence LC-20AD system using a Luna C18 column (Phenomenex, Torrance, CA, USA). A mobile phase of 30% methanol in water containing 0.5% (v/v) trichloroacetic acid was used and absorption at 280 and 325 nm was recorded using a PDA detector. The control reactions were performed in the same system without the addition of AnFaeA. 2.9. Structural analysis Homology-based model of AnFaeA was constructed using the SWISS-MODEL version 8.05 (Kiefer et al., 2009) with the crystal structure of the FaeA from A. niger CBS120.49 (PDB code: 1UWC) (Mcauley et al., 2004), which shared a sequence similarity of 98% with the native AnFaeA from A. niger CIB 423.1 (EMBL accession No. FJ430154). Visualization of the modeled structure was done using the program PyMOL (Delano Scientific, Palo Alto, CA, USA). 3. Results and discussion 3.1. Screening for mutants with increased thermostability The thermostability of the native AnFaeA (EMBL accession No.: FJ430154) has been enhanced with two beneficial mutants (D93G and S187F) identified using the PoPMuSiC algorithm (Zhang and Wu, 2011). The double mutant displayed over 10-fold increase in specific activity compared with the native enzyme, which would

143

S.-B. Zhang et al. / Bioresource Technology 117 (2012) 140–147

result in higher sensitivity during high-through screening (Zhang et al., 2012). Therefore, in this work, the double mutant, AnFaeA(D93G/S187F) was used as the parental enzyme to further improve the thermostability of AnFaeA. The epPCR library was successfully constructed in P. pastoris, which has been becoming a popular host expressing heterologous proteins (Daly and Hearn, 2005), and started being applied in directed evolution very recently (Fernández et al., 2010; Hartner et al., 2008; Liu et al., 2008). The shuttle vector pPIC9K was applied for the library construction because it displayed higher transformation efficiency than the vector pGAPZaA, which facilitated the high-throughput screening. Although multiple insertion events could occur spontaneously at about 1–10% of the single insertion events, the screening process, which involves the measurement of residual activities, would not be affected. The resulted random mutagenesis library containing around 40,000 colonies was first constructed in E. coli and then transferred into P. pastoris KM71 for protein expression and screening. A low inducer concentration (0.5% methanol) was applied to avoid significant change of protein expression on 96-well plates. Ten randomly picked colonies were sequenced to estimate the diversity of the library, which revealed an error rate of 1.9 nucleotide changes/kb. A total of 10,000 mutants were then screened for feruloyl esterase activity toward CNPF using a high-throughput screening system for FAEs developed in our lab (Zhang et al., 2012), and approximately 55% of them were active. The activities with or without heat treatment were tested for each mutant to address the concern that improved thermostability might negatively impact the catalytic activity. Significantly increased enzymatic activity (>30%) without heat treatment was not observed for any individual mutants, which ruled out the possibility of multi-copy transformants affecting protein expression under the current cultivation conditions. After heat treatment at 63 °C for 30 min on 96-well plates, the parental AnFaeA lost around 70–80% activity, and eleven mutants displayed significantly increased residual activity of over 40% without obvious loss of activity, and the best mutant Z11 containing the C235S substitution retained around 90% activity (Fig. 1). Further measurement of their half-lives of thermal inactivation (t1/2) at 55 °C using the crude enzymes showed that the t1/2 values for most of the mutants were ranging from 8.7 to 16.9 min, over 1.8-fold increase compared to that of the parental AnFaeA (Table 2, entries 2–11). However, the best mutant Z11 displayed an unexpected quick drop of activity with less than 50% activity remaining after 10min treatment. On the other hand, with prolonged incubation at 55 °C, further inactivation of Z11 was very slow (Fig. 2). Elevated temperatures were then applied to test the thermal inactivation of Z11. To our surprise, the mutant retained more activity after treatment at temperatures ranging from 70 to 85 °C than at

Table 2 Half-lives of thermal inactivation for the parental AnFaeA and mutantsa. Entry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Enzyme Parental Z1 Z2 Z3 Z4 Z5 Z6 Z7 Z8 Z9 Z10 Z11 – – – –

b

Mutations

t1/2 (min)

Relative t1/2

– Q177H T63I A140T T57I/V178A T35I S163T K37I/G69A Q185R L14F Q121H C235S T57I V178A K37I G69A

4.7 ± 0.1 9.4 ± 0.1 11.5 ± 0.6 11.1 ± 0.2 10.9 ± 0. 8 15.8 ± 1.4 16.9 ± 1.3 11.6 ± 0.6 11.6 ± 1.0 9.4 ± 0.3 8.7 ± 0.5 15.0 ± 0.6 6.8 ± 0.2 7.5 ± 0.3 11.7 ± 0.5 4.9 ± 0.1

1 2.0 2.5 2.4 2.3 3.4 3.6 2.5 2.5 2.0 1.8 3.2 1.4 1.6 2.5 1

a All the data were acquired after heat treatment at 55 °C except Entry 12, for which the treatment was at 90 °C. Crude enzymes were applied for all experiments. b The parental AnFaeA is a mutant AnFaeA from Aspergillus niger CIB 423.1 containing the D93G/S187F mutation.

Fig. 2. Thermal inactivation of mutant Z11 (C235S). Residual activities were measured after treating crude enzyme samples at 55 (e, in dotted line), 70 (j), 75 (h), 80 (d), 85 (s) or 90 () °C, and samples were taken at intervals and assayed. Each point presented is the mean ± S.D. (range) of duplicate assays.

55 °C, and showed over 50% residual activity after 5-min treatment at 90 °C (Fig. 2). Its t1/2 was estimated to be 15 min at 90 °C (Table 2, entry 12). Sequence analysis of the eleven mutants revealed a total number of 13 mutations (Table 2). The effect of each single mutation in mutants T57I/V178A and K37I/G69A on the half-life of thermal inactivation was analyzed, and the results showed that the T57I, V178A, K37I mutations were beneficial to the thermostability of AnFaeA with 1.5–3-fold improvement in t1/2 (Table 2, entries 13– 15). Collectively, twelve beneficial amino acid substitutions were identified from screening one random library.

3.2. Combination of beneficial amino acid substitutions

Fig. 1. Residual activities of thermostable AnFaeA mutants identified from epPCR library.

Since beneficial amino acid substitutions commonly show a synergistic or additive effect on the thermostability of proteins (Giver et al., 1998; McLachlan et al., 2008; Zhang et al., 2010), the combination of the twelve beneficial amino acid substitutions

144

S.-B. Zhang et al. / Bioresource Technology 117 (2012) 140–147

Fig. 3. Thermal inactivation of mutants M11 (d) and M12 (s). Residual activities were measured after treating purified enzyme samples at 55 °C, and samples were taken at intervals and assayed. Each point presented is the mean ± S.D. (range) of duplicate assays.

identified in this work were carried out to generate the mutant M12 for enhanced thermostability. Another combinational mutant M11 was also constructed with all beneficial amino acid substitutions except C235S, owing to the exceptional characteristic of this mutation. Both M11 and M12 were functionally expressed in P. pastoris, and the enzymes were purified (Supplementary data, Fig. S1) and were subjected to the measurement of half-lives of thermal inactivation at 55 °C. Both mutant enzymes displayed significantly enhanced thermostability as shown in Fig. 3 with M12 being the best. After 72-h heat-treated at 55 °C, they both retained over 90% activity (t1/2 >4000 min), while the parental AnFaeA had a t1/ 2 of 15 min when purified enzyme was applied. Based on the crystal structure of the AnFaeA from A. niger CBS120.49 (PDB code: 1UWC) (Mcauley et al., 2004) which shares a 96.6% similarity with our parental AnFaeA, four amino acid residues at positions 37, 57, 121 and 177 are located on the surface of the protein and eight amino acid residues at positions 14, 35, 63, 140, 163, 178, 185 and 235 exist in the inner of protein. Due to diversity of A. niger strains (de Vries et al., 1997), one of the identified substitutions, T35I, exists in the sequence of AnFaeA from A. niger CBS120.49. The intramolecular interactions of parental AnFaeA and mutant M12 (PDB-P and PDB-M1 of the Electronic Annex) associated with those amino acid substitutions were analyzed using the PIC server (http://pic.mbu.iisc.ernet.in/index.html) (Tina et al., 2007), which is programed to reveal canonical interactions such as hydrogen bonds, hydrophobic interactions, and electrostatic interactions that is well known as the dominant structural factors responsible for protein thermostability. Stabilization effect was found for some of those amino acid substitutions. The substitution of L14F would add a new putative hydrophobic interaction with Val15 within 5 Å, and add five new putative aromatic–aromatic interactions within 4–7 Å. K37I and T63I would add 1 and 8 new putative hydrophobic interactions with other residues within 5 Å, respectively. Q121H would add seven new putative hydrogen bonds, and have one new ionic interaction with Asp 124 within 6 Å. Q185R and A140T would each add one new putative hydrogen bond, although A140T would reduce putative hydrophobic interactions with Leu141 and Leu155 within 5 Å. In addition, four of the substitutions, L14F, T63I, A140T, Q185R could result in reduced local structural entropy of AnFaeA, which is considered having a linear relationship with protein thermostability and may lead to the improved thermostability of AnFaeA (Chan et al., 2004).

Fig. 4. Thermal inactivation of parental AnFaeA (s) and mutants Z11 (d), M11 (h) and M12 (j) assayed with substrate CNPF (a) or MFA (b). Purified enzymes were treated at elevated temperatures for 15 min before measuring residual activities. Each point presented is the mean ± S.D. (range) of duplicate assays.

On the other hand, there are several substitutions would lead to the reduction of those intramolecular interactions. The V178A substitution would reduce three putative hydrogen bonds, and three putative hydrophobic interactions within 5 Å. The T57I substitution would reduce the putative hydrogen bond with Asn55. The S163T and Q177H substitutions would reduce two and four putative hydrogen bonds, respectively. The C235S substitution would reduce the number of putative hydrogen bonds six to five.

3.3. Characterization of the parental AnFaeA and mutants The thermal inactivation patterns of parental AnFaeA and mutants Z11, M11, M12 were measured using either CNPF (Fig. 4a) or MFA (Fig. 4b) as the substrate. Those two substrates returned similar thermo-inactivation tendency for all the enzymes. M12 appeared as the most thermo-resistant mutant, which retained around 80% activity when treated at temperatures from 70 to

Table 3 Kinetic parameters and melting temperatures of the parental AnFaeA and mutants. AnFaeAa

Km (mM)

kcat (min1)

kcat/Km (mM1 min1)

Tm (°C)

Parental Z11 M11 M12

3.65 ± 0.37 6.34 ± 0.59 3.96 ± 0.28 4.08 ± 0.45

62.8 ± 4.0 53.4 ± 3.6 72.4 ± 3.4 47.6 ± 3.4

17.2 ± 2.1 8.4 ± 1.0 18.3 ± 1.5 11.7 ± 1.5

58.4 52.2 69.4 65.6

a Additional amino acid substitutions in those mutants compared with the parental AnFaeA (D93G/S187F) are as follows: Z11 (C235S), M11(L14F, T35I, K37I, T57I, T63I, A140T, Q121H, S163T, Q177H, V178A, Q185R) and M12 (L14F, T35I, K37I, T57I, T63I, A140T, Q121H, S163T, Q177H, V178A, Q185R, and C235S).

S.-B. Zhang et al. / Bioresource Technology 117 (2012) 140–147

Fig. 5. Temperature dependence of activities of parental AnFaeA (o) and mutants Z11 (d), M11 (h) and M12 (j). Activities were normalized as percentages of the activities at optimal temperature. Each point presented is the mean ± S.D. (range) of duplicate assays using the substrate MFA.

90 °C, and over 60% activity even after heat-treated at 99 °C for 15 min (Fig. 4). The mutant M11, lack of the C235S mutation compared with M12, quickly lost activity after treatment at temperatures higher than 65 °C (Fig. 4), indicating significant impact of the C235S substitution on thermal stability. The melting temperatures (Tm) of M12 and M11 increased as well with upward shifts of 7.2 and 11.0 °C, respectively, compared to that of parental AnFaeA (Table 3). At the same time, the C235S mutation caused unexpected downward shifts of the Tm values of Z11 and M12 of 6.2 and 3.8 °C compared with those of the parental AnFaeA and M11, respectively (Table 3). The C235S mutation also resulted in a decrease in temperature optima for the mutant Z11 (Fig. 5). And surprisingly, the M12 mutant had a temperature optima of only 60 °C, same as the M11 mutant, which was lacking of the C235S mutation compared with M12, and appeared much less thermostable than M12 as shown in Fig. 4. The pH optima of the enzymes was not affected by the mutations, and all the mutants Z11, M11 and M12 showed a similar pH/activity profile as the parental AnFaeA with the pH optima at pH 4.6 (data not shown). Cys235 is located in the inner of the protein, and is the only free cysteine among a total of seven Cys residues within the sequence of AnFaeA. Other Cys residues form three disulfide bonds (Cys91– Cys94, Cys29–Cys258 and Cys227–Cys234) and are critical for structure stability (Hermoso et al., 2004). The replacement of the free cysteine could presumably eliminate thiol oxidation and avoid the formation of incorrect disulfide bonds, resulting in reduced irreversible denaturation of the protein. The thermostabilization effect of replacing cysteine residue with serine or alanine has been reported in several cases (Amaki et al., 1994; Heinzelman et al., 2009), and one of them showed that the C6S mutation in a superoxide dismutase (SOD) could cause increased resistance to thermal inactivation but decreased melting temperature at the same time (McRee et al., 1990). This phenomenon closely resembles our results for the C235S mutation, and might be caused by the reversible unfolding of the protein, since the both the SOD and the AnFaeA were stored on ice after heat treatment and the activities were assayed at a decreased temperature, which was standard procedure for thermostability measurement, but could allow refolding to occur after thermal inactivation when the thermal inactivation was caused by noncovalent conformational changes rather than covalent changes. However, the temperature dependence of activity for those SOD mutants was not reported. Our results indicated that when the enzymes were assayed at a constant temperature without heating

145

and cooling process, the C235S mutation would lose the thermostabilization effect (Fig. 5), which further supported the hypothesis that the C235S mutation might increase the probability of correct refolding of the AnFaeA during the cooling process after heat treatment, which contribute significantly to its apparently high thermostabilization effect during standard assays. Another significant effect of the C235S mutation was that it led to an unusual thermal inactivation pattern for the mutant Z11 with the highest residual activity of >60% achieved after heat treatment at around 70 °C, while at lower temperature of 55 °C, its residual activity was <40%, even lower than the parental enzyme (Fig. 4a and b). Assays with either CNPF or MFA as the substrate returned similar patterns, and it was also in accordance with the phenomenon observed during the measurement of t1/2 using crude enzymes when heat treatment at 55 °C resulted in faster loss of activity than at 70 °C (Fig. 2). A similar unusual thermal inactivation pattern has been reported for a-chymotrypsin as the ‘‘zig-zag’’ temperature dependence of the rate constant of irreversible thermoinactivation (Levitsky et al., 1994a; Levitsky et al., 1994b). A possible explanation for this phenomenon is that the thermal inactivation of the enzyme at low temperature and high temperature are subjected to different mechanisms and the ‘high-temperature’ denatured form is more stable against irreversible thermoinactivation than the ‘‘low-temperature’’ form (Levitsky et al., 1994a). In our case, the mutant Z11 treated at higher (70–85 °C) and lower temperatures (55 °C) might undergo different conformational changes to two forms of unfolded intermediates, and the one obtained at higher temperature might be more readily to refold to the native form. Sophisticated studies on irreversible and reversible thermoinactivation of those enzymes might help to elucidate the real mechanism in the future. Functionally known FAEs all contain the same Cys residue as AnFaeA, while putative FAEs or lipase superfamily members contain varied residues such as Asn, Ile or Ser at the corresponding positions. Two well-studied lipases with Ser residue instead of Cys are from Rhizomucor miehei (RmL) (Boer et al., 1988) and Rhizopus delemar (RdL) (Haas et al., 1991). They share 30.24% and 24.74% sequence identity with AnFaeA, and same as AnFaeA, they both contain three disulfide bonds, while on the position that corresponds to the free Cys235 in AnFaeA, there is the Ser residue. However, with t1/2 values of 35 min (55 °C) and <5 min (50 °C), respectively, the thermostability of neither RmL nor RdL is higher than the parental AnFaeA, indicating that Ser is not the only factor in stability, and the Cys to Ser substitution in native protein does not necessarily indicate native thermostability. And there is no report on any unusual thermal inactivation patterns for those lipases. The steady-state kinetic parameters for the parental AnFaeA and mutants were determined at 40 °C with CNPF as the substrates using purified enzymes (Table 3). All the enzymes tested retained 100% activity under these conditions to avoid the interference of thermostability. The results demonstrated that the C235S mutation had a slightly negative impact on the catalytic efficiency of the enzymes judged by the increased apparent Michaelis constants (Km) and decreased turnover frequencies (kcat) of mutants Z11 and M12 compare with those of the parental AnFaeA and M11, respectively (Table 3). Despite of that, the kcat of M12 remained 76% of that of the parental AnFaeA (Table 3), and its excellent resistance toward irreversible thermal inactivation would facilitate the application in biomass degradation at higher reaction temperature. 3.4. Release of ferulic acid from steam exploded corn stalk Corn stalk is a typical natural biomass. FAEs play a key role in enhancing the accessibility of others related enzymes and subsequent hydrolysis of hemicellulose fibers by removing the ferulic

146

S.-B. Zhang et al. / Bioresource Technology 117 (2012) 140–147

X.-Q.P.) of the Chinese Academy of Sciences, and the Province Science Foundation of Sichuan, China (2010SZ0128). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.biortech.2012.04. 042. References

Fig. 6. Liberation of ferulic acid from steam exploded corn stalk. The reactions were catalyzed with parental AnFaeA (d) or mutant M12 (j) at 60 °C in the presence of an xylanase. Each point represents the average value of three independent measurements.

acid side chains and crosslinks (Wong, 2006). The reactions involving AnFaeA are generally performed at 50 °C for a variety of industrial applications such as the pulp bleaching industry (Record et al., 2003), the degradation of feruloylated oligosaccharides from sugar-beet pulp and wheat bran (Ralet et al., 1994), the enzymatic saccharification of wheat straw (Tabka et al., 2006). To determine the effect of enhanced thermostability to the hydrolysis of steam exploded corn stalk, the reactions were performed at an elevated temperature of 60 °C. An xylanase XT6 mutant FC06T with a t1/2 of 182 min at 75 °C (Zhang et al., 2010) was applied synergistically to cleave the xylan main chain to expose the feluloyl ester bond embedded in the corn stover, which could have a significant effect on the action of FAEs (Koseki et al., 2009). The efficiency of the parental AnFaeA and mutant M12 were compared using equal quantity of proteins. The parental AnFaeA appeared completely lost activity in 30 min. The release of ferulic acid catalyzed with the mutant M12 was around 2-fold of that achieved with the parental AnFaeA after 15-min reactions, and continued rising throughout the data collection period of 7 h to over three times as much as that released by the parental AnFaeA (Fig. 6).

4. Conclusions Multiple thermostabilization mutations were identified from screening a random mutagenesis library and AnFaeA mutants with significantly enhanced stability were constructed, which would pave the way for applying elevated reaction temperatures to increase process flexibility in the hydrolysis of lignocellulosic substrates using AnFaeA in combination with other lignocellulose degrading enzymes with compatible thermostability. The results have not only demonstrated the advantage of thermostabilized mutants of AnFaeA for the release of ferulic acid from steam exploded corn stalk, but also indicated great potential of those thermostable FAEs in a lot more industrial applications to possibly increase process efficiency and economics.

Acknowledgements We thank Professor Hongzhang Chen at the Institute of Process Engineering, Chinese Academy of Sciences, for providing the steam exploded corn stalk. This work was supported by the Program of 100 Distinguished Young Scientists, the Knowledge Innovation Program (KSCX1-YW-11B2), and the Open Fund of Key Laboratory of Environmental and Applied Microbiology (Y1C5101106 to

Amaki, Y., Nakano, H., Yamane, T., 1994. Role of cysteine residues in esterase from Bacillus stearothermophilus and increasing its thermostability by the replacement of cysteines. Appl. Microbiol. Biotechnol. 40, 664–668. Bloom, J.D., Arnold, F.H., 2009. In the light of directed evolution: pathways of adaptive protein evolution. Proc. Natl. Acad. Sci. USA 106, 9995–10000. Boer, E., Huge-Jensen, B., Christensen, M., Thim, L., Fiil, N.P., 1988. Rhizomucor miehei triglyceride lipase is synthesized as a precursor. Lipids 23, 701–706. Chan, C.-H., Liang, H.-K., Hsiao, N.-W., Ko, M.-T., Lyu, P.-C., Hwang, J.-K., 2004. Relationship between local structural entropy and protein thermostability. Proteins: Struct. Funct. Bioinform 57, 684–691. Crepin, V.F., Faulds, C.B., Connerton, I.F., 2004. Functional classification of the microbial feruloyl esterases. Appl. Microbiol. Biotechnol. 63, 647–652. Daly, R., Hearn, M.T.W., 2005. Expression of heterologous proteins in Pichia pastoris: a useful experimental tool in protein engineering and production. J. Mol. Recognit. 18, 119–138. Faulds, C.B., 2010. What can feruloyl esterases do for us? Phytochem. Rev. 9, 121– 132. Fazary, A.E., Ju, Y.-H., 2008. The large-scale use of feruloyl esterases in industry. Biotechnol. Mol. Biol. Rev. 3, 095–110. Fernández, L., Jiao, N., Soni, P., Gumulya, Y., de Oliveira, L.G., Reetz, M.T., 2010. An efficient method for mutant library creation in Pichia pastoris useful in directed evolution. Biocatalysis Biotransform. 28, 122–129. Giver, L., Gershenson, A., Freskgard, P.-O., Arnold, F.H., 1998. Directed evolution of a thermostable esterase. Proc. Natl. Acad. Sci. USA 95, 12809–12813. Haas, M.J., Allen, J., Berka, T.R., 1991. Cloning, expression and characterization of a cdna encoding a lipase from Rhizopus delemar. Gene 109, 107–113. Hartner, F.S., Ruth, C., Langenegger, D., Johnson, S.N., Hyka, P., Lin-Cereghino, G.P., Lin-Cereghino, J., Kovar, K., Cregg, J.M., Glieder, A., 2008. Promoter library designed for fine-tuned gene expression in Pichia pastoris. Nucleic Acids Res. 36, 1–15. Heinzelman, P., Snow, C.D., Smith, M.A., Yu, X., Kannan, A., Boulware, K., Villalobos, A., Govindarajan, S., Minshull, J., Arnold, F.H., 2009. SCHEMA recombination of a fungal cellulase uncovers a single mutation that contributes markedly to stability. J. Biol. Chem. 284, 26229–26233. Hermoso, J.A., Sanz-Aparicio, J., Molina, R., Juge, N., González, R., Faulds, C.B., 2004. The crystal structure of feruloyl esterase a from Aspergillus niger suggests evolutive functional convergence in feruloyl esterase family. J. Mol. Biol. 338, 495–506. Juge, N., Williamson, G., Puigserver, A., Cummings, N.J., Connerton, I.F., Faulds, C.B., 2001. High-level production of recombinant Aspergillus niger cinnamoyl esterase (FAEA) in the methylotrophic yeast Pichia pastoris. FEMS Yeast Res. 1, 127–132. Kiefer, F., Arnold, K., Künzli, M., Bordoli, L., Schwede, T., 2009. The swiss-model repository and associated resources. Nucleic Acids Res. 37, D387–D392. Koseki, T., Hori, A., Seki, S., Murayama, T., Shiono, Y., 2009. Characterization of two distinct feruloyl esterases, AoFaeB and AoFaeC, from Aspergillus oryzae. Appl. Microbiol. Biotechnol. 83, 689–696. Levitsky, V.Y., Melik-Nubarov, N.S., Siksnis, V.A., Grinberg, V.Y., Burova, T.V., Levashov, A.V., Mozhaev, V.V., 1994a. Reversible conformational transition gives rise to ‘zig-zag’ temperature dependence of the rate constant of irreversible thermoinactivation of enzymes. Eur. J. Biochem. 219, 219–230. Levitsky, V.Y., Panova, A.A., Mozhaev, V.V., 1994b. Correlation of high-temperature stability of a-chymotrypsin with ‘salting-in’ properties of solution. Eur. J. Biochem. 219, 231–236. Liu, Z., Pscheidt, B., Avi, M., Gaisberger, R., Hartner, F.S., Schuster, C., Skranc, W., Gruber, K., Glieder, A., 2008. Laboratory evolved biocatalysts for stereoselective syntheses of substituted benzaldehyde cyanohydrins. Chem. Bio. Chem. 9, 58– 61. Mcauley, K.E., Svendsen, A., Patkar, S.A., Wilson, K.S., 2004. Structure of a feruloyl esterase from Aspergillus niger. Acta Crystallogr. D. Biol. Crystallogr. D60, 878– 887. McLachlan, M.J., Johannes, T.W., Zhao, H., 2008. Further improvement of phosphite dehydrogenase thermostability by saturation mutagenesis. Biotechnol. Bioeng. 99, 268–274. McRee, D.E., Redford, S.M., Getzoff, E.D., Lepock, J.R., Hallewell, R.A., Tainer, J.A., 1990. Changes in crystallographic structure and thermostability of a Cu, Zn superoxide dismutase mutant resulting from the removal of a buried cysteine. J. Biol. Chem. 265, 14234–14241. Pei, X.-Q., Yi, Z.-L., Tang, C.-G., Wu, Z.-L., 2011. Three amino acid changes contribute markedly to the thermostability of b-glucosidase bglc from Thermobifida fusca. Bioresour. Technol. 102, 3337–3342.

S.-B. Zhang et al. / Bioresource Technology 117 (2012) 140–147 Ralet, M.-C., Faulds, C.B., Williamson, G., Thibault, J.-F., 1994. Degradation of feruloylated oligosaccharides from sugar-beet pulp and wheat bran by ferulic acid esterases from Aspergihs niger. Carbohydr. Res. 263, 257–269. Record, E., Asther, M., Sigoillot, C., Pagès, S., Punt, P.J., Delattre, M., Haon, M., van den Hondel, C.A.M.J.J., Sigoillot, J.-C., Lesage-Meessen, L., Asther, M., 2003. Overproduction of the Aspergillus niger feruloyl esterase for pulp bleaching application. Appl. Microbiol. Biotechnol. 62, 349–355. Tabka, M.G., Herpoël-Gimbert, I., Monod, F., Asther, M., Sigoillot, J.C., 2006. Enzymatic saccharification of wheat straw for bioethanol production by a combined cellulase xylanase and feruloyl esterase treatment. Enzyme Microb. Technol. 39, 897–902. Tina, K.G., Bhadra, R., Srinivasan, N., 2007. PIC: protein interactions calculator. Nucleic Acids Res. 35, W473–W476. Udatha, D.B.R.K.G., Kouskoumvekaki, I., Olsson, L., Panagiotou, G., 2011. The interplay of descriptor-based computational analysis with pharmacophore modeling builds the basis for a novel classification scheme for feruloyl esterases. Biotechnol. Adv. 29.

147

Vries, R.P., Michelsen, B., Poulsen, C.H., Kroon, P.A., van den Heuvel, R.H.H., Faulds, C.B., Williamson, G., van den Hombergh, J.P.T.W., Visser, J., 1997. The faea genes from Aspergillus niger and Aspergillus tubingensis encode ferulic acid esterases involved in degradation of complex cell wall polysaccharides. Appl. Environ. Microbiol., 4638–4644. Wong, D.W.S., 2006. Feruloyl esterase. Appl. Biochem. Biotechnol. 133, 87–112. Yang, S., Ding, W., Chen, H., 2009. Enzymatic hydrolysis of corn stalk in a hollow fiber ultrafiltration membrane reactor. Biomass Bioenergy 33, 332–336. Zhang, S.-B., Ma, X.-F., Pei, X.-Q., Liu, J.-Y., Shao, H.-W., Wu, Z.-L., 2012. A practical high-throughput screening system for feruloyl esterases: substrate design and evaluation. J. Mol. Catal. B: Enzym. 74, 36–40. Zhang, S.-B., Wu, Z.-L., 2011. Identification of amino acid residues responsible for increased thermostability of feruloyl esterase a from Aspergillus niger using the popmusic algorithm. Bioresour. Technol. 102, 2093–2096. Zhang, Z.-G., Yi, Z.-L., Pei, X.-Q., Wu, Z.-L., 2010. Improving the thermostability of Geobacillus stearothermophilus xylanase XT6 by directed evolution and sitedirected mutagenesis. Bioresour. Technol. 101, 9272–9278.