MI65CH04-Punt
ARI
23 May 2011
17:37
V I E W
A
N
I N
C E
S
R
E
D V A
Fungal Protein Production: Design and Production of Chimeric Proteins Peter J. Punt,1 Anthony Levasseur,2 Hans Visser,3 Jan Wery,3 and Eric Record2 1 TNO Microbiology and Systems Biology, 3700 AJ, Zeist, The Netherlands; email:
[email protected] 2
INRA, UMR1163, Universit´e Aix-Marseille I, II, Case 932, 13288 Marseille Cedex 09, France
3
Annu. Rev. Microbiol. 2011. 65:57–69 The Annual Review of Microbiology is online at micro.annualreviews.org This article’s doi: 10.1146/annurev.micro.112408.134009 c 2011 by Annual Reviews. Copyright All rights reserved 0066-4227/11/1013-0057$20.00
DYADIC Nederland, 6709 PA Wageningen, The Netherlands
Keywords Aspergillus, Trichoderma, Chrysosporium, fusion-protein, purification, carbohydrate-binding module, CAZY
Abstract For more than a century, filamentous fungi have been used for the production of a wide variety of endogenous enzymes of industrial interest. More recently, with the use of genetic engineering tools developed for these organisms, this use has expanded for the production of nonnative heterologous proteins. In this review, an overview is given of examples describing the production of a special class of these proteins, namely chimeric proteins. The production of two types of chimeric proteins have been explored: (a) proteins grafted for a specific substrate-binding domain and (b) fusion proteins containing two separate enzymatic activities. Various application areas for the use of these chimeric proteins are described.
57
MI65CH04-Punt
ARI
23 May 2011
17:37
Contents INTRODUCTION . . . . . . . . . . . . . . . . . . BIFUNCTIONAL, CARBOHYDRATE-BINDING MODULE–CONTAINING FUNGAL ENZYMES . . . . . . . . . . . . . PROTEIN TARGETING VIA THE CARBOHYDRATE-BINDING MODULE . . . . . . . . . . . . . . . . . . . . . . . . DESIGN OF CHIMERIC PROTEINS: THE BACTERIAL CELLULOSOME MODEL . . . . . . . DESIGN AND PROPERTIES OF CHIMERIC FUNGAL ENZYMES BASED ON CARBOHYDRATEBINDING MODULE GRAFTING . . . . . . . . . . . . . . . . . . . . . . DESIGN AND PROPERTIES OF CHIMERIC FUNGAL ENZYMES BASED ON FUSION PROTEINS . . . . . . . . . . . . . . . . . . . . . . . SUBSTRATE BINDING OF CHIMERIC PROTEINS . . . . . . . . . . APPLICATIONS TESTS OF CHIMERIC PROTEINS . . . . . . . . . . DESIGN AND PROPERTIES OF CHIMERIC FUNGAL PROTEINS FOR ALTERNATIVE APPLICATIONS . . . . . . . . . . . . . . . . . OUTLOOK. . . . . . . . . . . . . . . . . . . . . . . . . .
58
59
60
61
62
63 64 64
65 66
INTRODUCTION In industrial biotechnology, a wide variety of products is being produced, some of these already for very long times. In particular, microbial production systems such as (lactic acid) bacteria, yeast, and filamentous fungi have a long history of use (for a review see Reference 47). The products include microbial biomass, as is the case for starter cultures in many dairy applications. In addition, organic acids such as lactic acid and citric acid are used for food preservation. Moreover, a variety of antimicrobial compounds, such as penicillins and cephalosporins, are produced by bacterial and fungal production 58
Punt et al.
systems. A well-known biotechnological application of various yeast species is the production of alcoholic beverages. In particular, filamentous fungi have been an important source of a wide array of hydrolytic enzymes to be used in food and feed applications, in textile and paper industry, and also in the chemical industry. Because fungi efficiently secrete enzymes in their culture medium, these enzymes have been amenable to industrial applications with relatively simple purification procedures. In more recent times, with the advent of genetic engineering tools for fungi, genetically modified enzymes have been progressively engineered in order to provide the physicochemical properties needed to improve the industrial processes and thus decrease the cost of the final manufactured product. These engineering tools employ site-direct mutagenesis based on the threedimensional structure of the enzyme and, more recently, so-called directed evolution, which consists of DNA-based diversity generation followed by screening for variants with optimized properties. As an alternative to directed evolution and thanks to the development of robotic laboratory stations, new enzymes can also be screened from samples representing the natural biodiversity by high-throughput screening, allowing a researcher to quickly obtain new enzyme targets (51). In the genomic era, more and more genomes are now publicly available and can be easily interrogated by using specialized databases and comparative genome mining, resulting in the identification of new enzymes. Moreover, for this purpose, so-called metagenomic approaches are being employed, allowing the discovery of new genes identified by sequencing and analysis of gene libraries prepared from genetic material of noncultured environmental samples. From all these approaches, large sets of target genes are available. To allow production of proteins encoded by any of the novel identified genes, an appropriate and efficient host becomes an obligatory tool. Because filamentous fungi are an important source of many industrial proteins and because these proteins have received Generally Regarded As Safe (GRAS)
MI65CH04-Punt
ARI
23 May 2011
17:37
notifications by the Food and Drug Administration, these eukaryotic organisms have also been selected as production host organisms for these novel proteins. In contrast to bacterial hosts, posttranslational maturation steps, such as protein glycosylation, can be performed by their eukaryotic counterparts. Compared to mammalian cell lines, such as Chinese hamster ovary cells, fungi provide faster and much cheaper fermentation conditions. In particular, various Aspergillus species, such as A. awamori, A. niger, A. oryzae, and Trichoderma reesei are now widely used in biotechnology. These fungi are nontoxic or nonpathogenic. For novel recombinant protein production, various improved fungal host strains have been developed with a particular aim of decreasing their proteolytic activities (for a review see Reference 10). In industrial biotechnology and, more specifically, in the so-called green chemistry sector, microorganisms are considered useful not only for the production of native industrial enzymes but also in industrial bioprocesses as replacements for traditional chemical processes. The products of these novel bioprocesses may include food nutrients, chemicals, polymers, detergents, pharmaceuticals, pulp and paper, textiles, and biofuels. Also, in many of these processes, (novel) proteins play an important role. In this review, we focus on the production and development of novel chimeric and bifunctional proteins in filamentous fungi, using A. niger, T. reesei, and a newly emerging and promising fungal host, Chrysosporium lucknowense, as examples (11, 22, 48, 51). Different sections are dedicated to design of chimeric proteins by fusion of different catalytic modules or by addition of nonenzymatic domains, resulting in proteins with new properties.
BIFUNCTIONAL, CARBOHYDRATE-BINDING MODULE–CONTAINING FUNGAL ENZYMES Filamentous fungi are well known for their wide variety of secreted hydrolytic enzymes (9, 49),
which provide these organisms a unique capability of degrading a plethora of biomass types. Plant-biomass-derived polysaccharides are heterogeneous, and depending on the nature of the different individual monomeric building blocks and their mutual linkages, enzyme mixtures of different complexity are required to fully degrade the polysaccharide structures. The efficient hydrolysis of plant polysaccharides is a key challenge in biomass saccharification for biorefinery purposes. The extent or efficiency of degradation depends not only on the catalytic characteristics of the enzymes, but also on the interaction, i.e., the binding of key enzymes to polysaccharide fibers. Many of the plantpolysaccharide-degrading enzymes secreted by filamentous fungi can be regarded as bifunctional proteins, because they consist of a catalytic domain and a carbohydrate-binding module (CBM). CBMs were originally classified as cellulose-binding domains (16, 46). However, CBMs that bind carbohydrates other than cellulose are discovered continuously and have currently been grouped into 61 different families (http://www.cazy.org/). CBMs consist of relatively short, mostly N- or C-terminal protein domains, separated from the catalytic domain by an often glycosylated-hinge domain necessary to obtain independent folding and conformational flexibility of both domains. (For a recent review on CBMs see Reference 43.) The number and diversity of CBMs encoded by genomes of lignocellulolytic fungi vary considerably. To date, Podospora anserina is reported to contain the highest number of CBMs of all fungal genomes sequenced (14). This study uncovered 28 cellulose-binding modules belonging to CBM family 1 (CBM1). Surprisingly, T. reesei contains only 36 CBMs in total (32). The A. niger genome encodes for 40 proteins carrying CBMs of which 8 belong to CBM1 (http://www.cazy.org/). Also C. lucknowense C1 (C1) is a rich source of plant cell wall–degrading enzymes such as cellulases and hemicellulases (22). Preliminary data from the C1 genome sequence revealed an extraordinary number of CBMs (75), of which 46 belong to CBM1 (data not shown). www.annualreviews.org • Fungal Protein Production
Chimeric: consisting of two or more parts of different origin Carbohydratebinding module (CBM): protein domain binding to carbohydrate substrate CAZY: Carbohydrate Active Enzyme Database CBM1: carbohydrate-binding module family 1
59
MI65CH04-Punt
ARI
23 May 2011
17:37
For some cellulases and xylanases isolated from industrial C1 production strains, the absence or presence of CBMs has significant effects on enzymatic functionality (11, 20, 48). Intact cellobiohydrolase I (CBH1, 65 kDa) contains a C-terminal CBM that, after proteolytic processing, yields the 52-kDa CBH1 catalytic core without the CBM module and major part of the glycosylated linker (20). Both forms of CBH1 were evaluated for hydrolysis of soluble and insoluble cellulose model substrates, such as p-NP-β-D-cellobioside, carboxymethylcellulose, and Avicel. The specific activity of both enzyme forms for the soluble substrates was comparable. However, the specific activity level of the intact CBH1 against the insoluble microcrystalline cellulose Avicel was twice as high as that of the CBM-less CBH1. Furthermore, the adsorption ability of the enzymes to Avicel was notably worse for the truncated CBH1 (20). Comparable observations were made for C1 endoglucanases EG51 and EG44 (11). These proteins represent the 51-kDa CBM-intact and the 44-kDa CBM-devoid forms of the same endoglucanase (EGII). Both enzymatic forms showed similar characteristics in the hydrolysis of soluble substrates. However, in Avicel adsorption experiments, 98% of the initially added EG51 adsorbed to Avicel, whereas only 10% of the CBM-truncated EG44 adsorbed. In addition, the specific activity of EG51 against Avicel was threefold higher than that of EG44 (11). Therefore, it can be concluded that the CBM modules of these cellulases bind to the recalcitrant insoluble cellulose fibers, enabling hydrolysis by the catalytic modules more efficiently. In addition to these cellulases, C1 xylanases Xyn10A and Xyn10B contain CBMs belonging to CBM1 (48). Also for these enzymes, both forms (i.e., with and without CBM) were present in the fermentation medium. Adsorption studies with the purified enzymes revealed that more than 90% of the intact Xyn10A and Xyn10B bound to cellulose, whereas only 2%–5% of Xyn10A and Xyn10B without their CBMs adsorbed (48). 60
Punt et al.
Gusakov et al. (19) have used mixtures of natural nonprocessed C1 cellulases with and without CBMs to design artificial enzyme mixtures. It was shown that the presence of both forms in the proper ratios notably enhanced the hydrolysis of cellulosic substrates, yielding a more efficient enzyme mixture than the native C1 enzyme mixture. The mode of this synergy in the artificial mixtures was suggested to be a cellulose hydrolysis mechanism by which loosely associated CBM-devoid cellulases remove physical obstacles from the substrate that hamper the processive action of tightly bound CBM-containing cellulases (19).
PROTEIN TARGETING VIA THE CARBOHYDRATE-BINDING MODULE The results described above show that CBMs can function as independent modules within chimeric proteins, efficiently targeting the enzymatic activity to the substrate. Their associated functions are related to the recognition of the substrate, the increase of the local concentration close to the substrate, and the modification of the interfacial properties of the substrate. These distinct properties of CBMs have been exploited in biotechnological applications where they are used as purification tags in affinity chromatography, protein targeting, cell immobilization, and diagnostics (43). Interestingly, not only well-known hydrolytic enzyme families but also several families of other, sometimes hypothetical, proteins have members containing CBMs (http://www.cazy.org/). An original CBMcontaining protein was reported in the fungus T. reesei and named swollenin (42). Swollenin also has a bimodular structure composed of an N-terminal CBM1 domain connected by a linker region to the plant expansin homologous domain. In the plant kingdom, expansins are involved in the breakdown of hydrogen bonds between cellulose microfibrils (cotton fibers, filter paper, and cell walls of Valonia algae) and other cell wall polymers without having detectable hydrolytic enzyme activity, judging from the
MI65CH04-Punt
ARI
23 May 2011
17:37
Table 1 Overview of various chimeric proteins discussed in this review Production Chimeric proteins Fusion proteins Grafted proteins
FaeA-XynB
Origin
Fungal host
Aspergillus niger
(g L−1 )
Aspergillus niger
1.4
Application Food
Reference 28
GLA-PHYT
Aspergillus niger
Aspergillus niger
>1
Animal feed
21
FaeA-XynBCBM1
Aspergillus niger
Aspergillus niger
1.5
Food
28
SWO-FaeA
Aspergillus niger/ Trichoderma reesei
Trichoderma reesei
0.025
Food
29
Man-CBM1
Aspergillus aculeatus
Aspergillus niger
0.13
Biofuel
38
Laccase-CBM1
Pycnoporus cinnabarinus
Aspergillus niger
0.01
Pulp bleaching
40
Doc-FaeA
Aspergillus niger/ Clostridium thermocellum
Aspergillus niger
0.1
Food
27
GH13-CBM20
Aspergillus niger
Aspergillus niger
NA
Purification
49
GLA-Fc
Aspergillus niger
Chrysosporium
>1
Purification
This review
production of detectable amounts of reducing sugars (12, 13, 42). The existence of this type of nonenzymatic bifunctional proteins provided the basis for our research on further chimeric protein design. Several genetic engineering approaches involving the design and production of various novel chimeric and bifunctional proteins in filamentous fungi have been explored and are described below. In Table 1 and Figure 1 an overview of the various chimeric proteins described in this review is given.
DESIGN OF CHIMERIC PROTEINS: THE BACTERIAL CELLULOSOME MODEL In a first approach to target fungal enzymes toward cellulose, the use of a bacterial cellulosome model was investigated. Bacterial cellulosomes were defined two decades ago as efficient multienzyme complexes for the degradation of cellulosic substrates (25). Cellulosomes are well-organized entities containing mixed compositions of different types of (hemi)cellulases. An enhanced synergistic action on plant cell wall degradation is promoted by the collective incorporation of enzymes into the cellulo-
some complex. This discrete multicomponent complex is mediated by cohesin-dockerin interaction, allowing the integration of enzymes and accounting for their selective incorporation into the complex. Cellulosomes were reported in anaerobic bacteria such as Clostridia and in anaerobic fungi. In the fungal kingdom, high-molecular-weight cellulosome-like
a sp
b
Cellulosome: complexes of cellulolytic enzymes consisting of catalytic subunits bound together via dockerin domains that bind to scaffold proteins, the so-called cohesins
Enzyme modules FAEA LAC MAN
sp pro
XYNB
sp
CBHB
Hinge CBM1
Chimeric enzyme modules
sp
FAEA LAC MAN
Hinge CBM1
sp
FAEA
Hinge
XYNB
sp
FAEA
Hinge XYNB Hinge CBM1
Figure 1 Schematic overview of various (a) enzyme (b) and chimeric enzyme modules described in this review. Abbreviations: FAEA, Aspergillus niger feruloyl esterase A; LAC, Pycnoporus cinnabarinus laccase; MAN, Aspergillus aculeatus mannanase; XYNB, Aspergillus niger xylanase B; CBHB, Aspergillus niger cellobiohydrolase B; CBM1, carbohydrate-binding module family 1; sp, signal peptide; pro, propeptide; hinge, O-glycosylated hinge region.
www.annualreviews.org • Fungal Protein Production
61
MI65CH04-Punt
ARI
23 May 2011
Lignin: complex polymeric structure containing monomeric aromatic building blocks Laccase: coppercontaining oxidase enzymes
62
17:37
complexes in Piromyces, Orpinomyces, and Neocallimastix have been described (2, 15, 30, 37). In this complex, a 40-amino-acid cysteine-rich, noncatalytic docking domain has been identified (45). This domain has been referred to as a potential fungal dockerin. However, this potential fungal dockerin has neither sequence similarity nor structural similarity to the bacterial counterpart (39). In addition, the putative dockerin module from Piromyces equi recognizes and binds to the oligosaccharide component of the glycosylated β-glucosidase (34). Thus, the definition of fungal cellulosome is still tentative as no cohesin module seems to be involved in the multienzymatic complex. The absence of classical fungal cohesin is a limiting factor for using fungal cellulosomes in (nano)biotechnology. Therefore, we focused on the use of a bacterial cellulosome as a template for the design of a fungal cellulosome. Our final objective was to mimic the bacterial cellulosome and produce this in a fungal cell factory (i.e., A. niger or T. reesei ) in order to produce high protein levels and to facilitate stronger synergistic effects among fungal catalytic subunits than obtained from the free-enzyme system. The first study focused on the production of a chimeric protein in A. niger composed of a fungal enzyme feruloyl esterase A (FAEA) from A. niger fused to the Cel48S dockerin from Clostridium thermocellum (28). Although for a Cterminal fusion a (truncated) chimeric protein was detected, most of the dockerin module was cleaved by proteolysis, whereas no protein was obtained for N-terminal fusion. Weak dockerin stability has already been described for natural bacterial cellulosomal enzymes (1). Consequently, both strategies were abandoned and a second approach was used to obtain a chimeric dockerin protein. In this so-called carrier approach, the Doc-FAEA gene fusion was fused downstream of the sequence of A. niger glucoamylase preceded by a fungal processing site (for review of this approach see Reference 17). The well-secreted glucoamylase from A. niger could improve the secretion efficiency of the chimeric protein by facilitating translocation and successive folding in the secretory pathway. Punt et al.
By using this strategy, the production of an intact recombinant Doc-FAEA protein was obtained and the production yield was estimated to be 100 mg of chimeric protein per liter in nonoptimized cultures. This strategy enabled us to produce the first fungal enzyme suitable to be incorporated in vitro into a bacterial cellulosome. However, limited proteolytic stability of this type of bacterial dockerin-containing proteins was still a serious problem, as exemplified by the results obtained with a fusion protein containing a bacterial dockerin from Clostridium cellulolyticum. In this case, the recombinant protein was produced completely cleaved from the dockerin (A. Levasseur, personal communication). Therefore, in subsequent approaches we focused exclusively on the use of fungal protein domains for chimeric protein design.
DESIGN AND PROPERTIES OF CHIMERIC FUNGAL ENZYMES BASED ON CARBOHYDRATEBINDING MODULE GRAFTING Considering different sectors of application, the grafting of fungal CBMs to (multimodular) chimeric enzymes could be of interest by improving the efficiency of the bioprocess and therefore decreasing the required charge of the considered enzymes. The primary goal is to decrease the price of the overall bioprocess and/or decrease the environmental burden by using lower amounts of polluting chemical agents. In the development of new bioprocesses, enzymes have become attractive tools in relation to reaching both economical and ecological goals. Many bioprocesses, in the biofuel, food, and pulp and paper sectors, rely on a mixture of proteins with cellulase or hemicellulase activity. These enzymes are produced in various fungal hosts, including T. reesei, A. niger, and C. lucknowense. In several of these processes, lignin-degrading enzymes have also been considered but not yet widely applied. One of the enzymes considered for lignin degradation is laccase, a copper-containing oxidoreductase
MI65CH04-Punt
ARI
23 May 2011
17:37
commonly found in plants and fungi (5). Based on their lignin-degrading capacity, fungal laccases are studied for pulp bleaching (24, 44). In our research, we considered the production of a chimeric laccase from P. cinnabarinus grafted to the CBM1 domain of CbhB of A. niger via its natural linker. The lac-CBM chimeric enzyme would be a new kind of enzyme that could link to cellulose and act on lignin. A. niger was selected as a potential host to produce this 100kDa protein on the basis of previous success in producing good yields of active laccase proteins from various origins (31, 36, 41). Indeed, the production of the chimeric lac-CBM1 was achieved in culture flasks of A. niger D15 (40). However, a minor part of the protein seems to be truncated at the linker position. As observed for the dockerin fusions, chimeric protein was produced only if the CBM1 was placed at the C terminus. The produced enzyme was demonstrated to have the biochemical properties of the native laccase, except that the optimal temperature was higher, 85◦ C instead of 65◦ C, and the thermal stability was markedly decreased. Similar results are obtained for a mannanase-CBM that was designed for the biofuel sector. Mannanases are major enzymes involved in hemicellulose degradation and may act synergistically with cellulases to enhance plant biomass degradation. Chimeric mannanase-CBM1 production was achieved in A. niger at a higher yield (130 mg liter−1 ) by combining the mannanase from Aspergillus aculeatus and the CBM of A. niger CbhB (38). All the kinetic and physicochemical properties, except for the thermal stability, were close to those of the native protein. For instance, half-life times of the chimeric enzymes were seven times higher at 65◦ C due to the fusion of the linker CBM part. Araki et al. (4) suggested the CBM performs a protecting role on the connected enzyme. In the same way, the deletion of the CBM from the native xylanase XYN10B leads to a decrease in thermostability and optimal temperature (3). However, additional glycosylation as present in the hinge region between the catalytic and
the substrate-binding domain could affect thermostability of the chimeric protein (6).
DESIGN AND PROPERTIES OF CHIMERIC FUNGAL ENZYMES BASED ON FUSION PROTEINS In addition to grafting substrate-binding domains, designing fusion proteins consisting of two individual proteins was also considered. Potential candidate enzymes selected as a partner for protein engineering were feruloyl esterase A and xylanase B. Feruloyl esterases hydrolyze ester bonds that link ferulic acid to plant cell wall polysaccharides and thus constitute a strategic group of enzymes with a potentially broad range of applications in the food, pharmaceutical, paper pulp, and biofuels industries. Ferulic acid (4-hydroxy-3-methoxy-cinnamic acid), one of the most abundant hydroxycinnamic acids in the plant world, is an attractive phenolic compound. Ferulic acid can be used as an antioxidant (18) and photoprotectant (33), or it can be transformed by microbial conversion into “natural” vanillin, an expensive flavoring agent (26, 35). In our research, synergistic effects were investigated in the case of different bifunctional feruloyl esterase–xylanase fusion proteins
Hemicellulose: carbohydrate structure containing a cellulose backbone consisting of glucose monomers and many different other sugar monomers, such as xylose
Glucoamylase
Hinge region
CH2
CH2 Fc-domain
C H 3 C H3
Protein A binding
Figure 2 Schematic representation of the structure of the glucoamylase-Fc protein. www.annualreviews.org • Fungal Protein Production
63
MI65CH04-Punt
ARI
23 May 2011
17:37
(27). In this study also, CBM1 with a flexible linker region was grafted to the fusion proteins, generating a fully active trifunctional chimeric enzyme. Similarly, fully active fusion proteins consisting of glucoamylase and phytase have been generated (21). In contrast, production of a chimeric fusion protein, consisting of swollenin and feruloyl esterase, led to a protein with a lower than expected Vmax value (29), likely because the nonflexible hinge region was engineered in a swollenin-feruloyl esterase fusion protein.
SUBSTRATE BINDING OF CHIMERIC PROTEINS For the various chimeric protein produced, the binding capacity of the CBM was evaluated, showing protein-specific differences. On the S
M
E1
E2
E3
E4
E5
one hand, mannanase-CBM showed a higher capacity of binding for the microcrystalline cellulose compared to a mechanical softwood pulp (38). On the other hand, laccase-CBM revealed a very low to no binding to Avicel, and then an increased level of binding between MN300 cellulose and a softwood kraft pulp, corresponding to a delignification of approximately 90% (40). Concerning the trifunctional CBM1containing feruloyl esterase–xylanase enzyme, this chimeric protein possesses affinity for Avicel, confirming that CBM conserved its function in the chimeric enzyme (27). These results reveal that the heterogeneity of the lignocellulosic substrates, and especially the interaction between cellulose and the hemicellulosic/lignin networks, will complicate straightforward interpretation of substrate-binding experiments. To compare various results, simple biomimetic substrates (8) should be used to test chimeric enzymes and better understand the mechanisms of CBM-mediated substrate binding involved in lignocellulosic degradation.
APPLICATIONS TESTS OF CHIMERIC PROTEINS
Figure 3 SDS-PAGE of protein A purification of the glucoamylase-Fc protein. Abbreviations: S, culture supernatant of the recombinant C1 host strain; M, protein size marker; E1–E5, subsequent elutions of the protein A–bound protein fraction; SDS-PAGE, sodium dodecyl sulfate polyacrylamide gel electrophoresis. The arrow indicates the position of the glucoamylase-Fc protein. 64
Punt et al.
As described above, in several cases chimeric proteins were successfully produced. Subsequently, experiments were performed in relation to the biotechnological use of these proteins. In consideration of pulp-bleaching applications, compared with bleaching results obtained with the native laccase, the chimeric lac-CBM enzyme reached the same level of delignification with a fourfold-lower dosage of protein in combination with saving 50% of the ClO2 , which is used in elemental chlorine-free bleaching to minimize the production of toxic organochlorine compounds (40). Similarly, the mannanase-CBM was tested in combination with the cellulase/xylanase cocktail and a β-glycosidase from T. reesei and A. niger for hemicellulose saccharification. Glucose release increased to 13% (38). Also, the action of T. reesei swollenin was evaluated in combination with a plant cell-wall-degrading enzyme, i.e., FAEA from
MI65CH04-Punt
ARI
23 May 2011
17:37
A. niger (29). The efficiency of the resulting chimeric protein (SWOI-FAEA) was tested for ferulic acid release using destarched, but not heat-treated, wheat bran as a model substrate and showed that SWOI-FAEA was more efficient than FAEA alone. By using heat treatment (at 130◦ C for 10 min), the positive effect of the chimeric SWOI-FAEA was not observed. On the one hand, the CBM1 of SWOI may increase the local concentration of the FAEA enzyme in close proximity of the substrate. On the other hand, the particular mobility of the expansin module of SWOI may disrupt the lignocellulose structure and facilitate the lateral diffusion of the chimeric protein along the surface of the cellulose microfibrils (12, 13). Also, the bifunctional feruloyl esterase–xylanase enzyme had a synergistic effect on natural substrates compared with the free enzymes (27), suggesting that the physical proximity of each enzymatic partner to each other and the targeted substrate leads to a positive effect.
DESIGN AND PROPERTIES OF CHIMERIC FUNGAL PROTEINS FOR ALTERNATIVE APPLICATIONS In the various examples described above, the purpose of designing chimeric proteins was to generate novel proteins with improved catalytic capabilities. However, other applications of chimeric proteins can also be considered. One other important alternative application is the use of chimeric proteins for purification purposes. In this approach, a specific substratebinding domain fused to a protein of interest may be used to purify the chimeric protein from the culture supernatant of the host organism. In particular, in filamentous fungal hosts, this approach is useful, as in most cases a mixture of different secreted proteins is produced in the background of the protein of interest. Recently, an example of this approach was published in relation to the purification and analysis of a specific amylase-like glycosyl hydrolase family 13 protein in A. niger (50), by producing a GH13-
CBM20 fusion protein and purifying the protein based on selective binding to starch. The advantage of this approach to purification of secreted fungal proteins over the use of other purification tags such as His-tags is that the CBM-based tags show mostly good proteolytic stability. On the basis of research conducted for the production of pharmaceutical proteins such as monoclonal antibodies, specific proteinbinding modules, in particular the protein A–binding domain of the heavy-chain antibody chain, have also been considered as purification tags. This so-called Fc domain has been fused
Lignocellulose: carbohydrate structure containing a cellulose backbone consisting and lignin-based cross-linking molecules CBM20: carbohydrate-binding module family 20
CONTROLLED FED-BATCH FERMENTATION PROTOCOL For production of sufficient amounts of chimeric proteins for application studies, simple shakeflask cultivation experiments are often not appropriate, particularly because these relatively uncontrolled conditions may result in unwanted proteolysis. Therefore, a small-scale fed-batch fermentation setup was developed for 2–5 liter fermentors, allowing up to 1 gram per 1 liter yields of recombinant protein. The following setpoints are used for A. niger: pH 4.5, T = 30◦ C, Airflow 1.2 liter min−1 . The pH was maintained using 8 M KOH and 2 M H3 PO4 . The set points for the fermentation were as follows. Dissolved oxygen was maintained at 20% by controlling agitation speed. Initial agitation speed was 400 rpm. Agitation may be increased up to 800 rpm. If dissolved oxygen drops below 20%, air sparging is replaced for O2 sparging. After glucose depletion (usually 24 h after start of the batch culture), the feed was started. Preculture: 2x 100 ml preculture in potato dextrose broth, inoculated with 1∗ 106 spores ml−1 and incubated at 30◦ C at 125 rpm until sufficient growth (24–48 h), were used to inoculate the fermentor. Batch culture: Medium composition: glucose (30 g liter−1 ), NaNO3 (10 g liter−1 ), MgSO4 .7H2 O (0.8 g liter−1 ), KH2 PO4 (2 g liter−1 ), CaCl2 .2H2 O (0.1 g liter−1 ), yeast extract (5 g liter−1 ), tryptone (5 g liter−1 ), Aspergillus metal trace elements (10 ml liter−1 ) (7), antifoam (struktol J673 10%; 1 ml liter−1 ). Feed culture: glucose (200 g liter−1 ), KH2 PO4 (10 g liter−1 ), NaNO3 (30 g liter−1 ), yeast extract (10 g liter−1 ), tryptone (10 g liter−1 ). The feed rate used was 5 g h−1 . Cultures are terminated 48 to 56 h after start of the feed.
www.annualreviews.org • Fungal Protein Production
65
MI65CH04-Punt
ARI
23 May 2011
Fed-batch: fermentation process based on feeding of a growth limiting nutrient substrate to a culture
17:37
to several nonantibody-based pharmaceuticals, R , which is a tumor necrosis such as Enbrel factor–blocking protein, allowing efficient purification of the bioactive protein (23). To demonstrate the usefulness of Fc domains as a purification tag for proteins produced in fungal hosts, we designed a chimeric fusion protein consisting of the A. niger glucoamylase catalytic domain, a glycosylated hinge region, and the Fc domain (Figure 2). Expression of this protein in Chrysosporium lucknowense and growth of the recombinant strains under suitable fed-batch fermentation conditions allowed efficient production of the glucoamylase Fc fusion protein. Applying culture supernatant samples containing a wide variety of native hydrolases to commercially available protein A–based purification columns allowed the efficient purification of the specific Fc-containing fusion protein (Figure 3). Because binding of the Fc domain to protein A relies on dimerization of the Fc domain, this result demonstrates that the recombinant Fc domain is fully functional in dimerization.
OUTLOOK On the basis of the various results described, using chimeric proteins to develop improved fungal enzyme-based industrial bioprocesses provides promising options. Currently, only a limited number of possibilities have been explored. Functional characterization has been carried out for few carbohydrate-binding domains (in particular those interacting with starch and cellulose). Many more CBM domains are expected to be identified and characterized by ongoing genome mining (49). In relation to bifunctional enzymes, our research will provide more detailed information on enzyme structure and synergy, allowing better design of chimeric proteins and taking into account protein flexibility and aspects of glycosylation. Both specific CBM domains and commonly used Fc domains can be used efficiently for protein purification in filamentous fungal hosts, allowing these hosts to also be considered for the production of pharmaceutical proteins, such as full-length antibodies.
SUMMARY POINTS 1. Filamentous fungi such as Aspergillus, Trichoderma, and Chrysosporium are suitable hosts for chimeric protein production. 2. Chimeric proteins are susceptible to proteolytic processing. 3. Chimeric proteins grafted to substrate-binding domains show increased substrate interaction. 4. Chimeric fusion proteins show synergistic enzymatic activities. 5. Substrate-binding domains grafted to chimeric proteins allow efficient protein purification.
DISCLOSURE STATEMENT The authors are not aware of any affiliations, memberships, funding, or financial holdings that might be perceived as affecting the objectivity of this review.
LITERATURE CITED 1. Adams JJ, Pal G, Jia Z, Smith SP. 2006. Mechanism of bacterial cell-surface attachment revealed by the structure of cellulosomal type II cohesin-dockerin complex. Proc. Natl. Acad. Sci. USA 103:305–10 66
Punt et al.
MI65CH04-Punt
ARI
23 May 2011
17:37
2. Ali BR, Zhou L, Graves FM, Freedman RB, Black GW, et al. 1995. Cellulases and hemicellulases of the anaerobic fungus Piromyces constitute a multiprotein cellulose-binding complex and are encoded by multigene families. FEMS Microbiol. Lett. 125:15–21 3. Araki R, Ali MK, Sakka M, Kimura T, Sakka K, Ohmiya K. 2004. Essential role of the family-22 carbohydrate-binding modules for β-1,3-1,4-glucanase activity of Clostridium stercorarium Xyn10B. FEBS Lett. 561:155–58 4. Araki R, Karita S, Tanaka A, Kimura T, Sakka K. 2006. Effect of family 22 carbohydrate-binding module on thermostability of Xyn10B catalytic module from Clostridium stercorarium. Biosci. Biotechnol. Biochem. 70:3039–41 5. Baldrian P. 2006. Fungal laccases—occurrence and properties. FEMS Microbiol. Rev. 30:215–42 6. Benoit I, Asther M, Sulzenbacher G, Record E, Marmuse L, et al. 2006. Respective importance of protein folding and glycosylation in the thermal stability of recombinant feruloyl esterase A. FEBS Lett. 580:5815– 21 7. Bennett J, Lasure L. 1991. Growth media. In More Gene Manipulations in Fungi, ed. J Bennett, L Lasure, pp. 441–58. San Diego, Academic 8. Boukari I, Putaux JL, Cathala B, Barakat A, Saake B, et al. 2009. In vitro model assemblies to study the impact of lignin-carbohydrate interactions on the enzymatic conversion of xylan. Biomacromolecules 10:2489–98 9. Braaksma M, Martens-Uzunova ES, Punt PJ, Schaap PJ. 2010. An inventory of the Aspergillus niger secretome by combining in silico predictions with shotgun proteomics data. BMC Genomics 11:584 10. Braaksma M, Punt PJ. 2008. Aspergillus as a cell factory for protein production: controlling protease activity in fungal production. In The Aspergilli: Genomics, Medical Aspects, Biotechnology, and Research Methods, ed. GH Goldman, SA Osmani, pp. 441–55. Boca Raton, FL: CRC Press 11. Bukhtojarov FE, Ustinov BB, Salanovich TN, Antonov AI, Gusakov AV, et al. 2004. Cellulase complex of the fungus Chrysosporium lucknowense: isolation and characterization of endoglucanases and cellobiohydrolases. Biochemistry 69:542–51 12. Cosgrove DJ. 2000. Expansive growth of plant cell walls. Plant Physiol. Biochem. 38:109–24 13. Cosgrove DJ. 2000. Loosening of plant cell wall by expansin. Nature 407:321–26 14. Espagne E, Lespinet O, Malagnac F, Da Silva C, Jaillon O, et al. 2008. The genome sequence of the model ascomycete fungus Podospora anserina. Genome Biol. 9:R77 15. Fujino Y, Ogata K, Nagamine T, Ushida K. 1998. Cloning, sequencing, and expression of an endoglucanase gene from the rumen anaerobic fungus Neocallimastix frontalis MCH3. Biosci. Biotechnol. Biochem. 62:1795–98 16. Gilkes NR, Warren RAJ, Miller RJ, Kilburn DG. 1988. Precise excision of the cellulose binding domains from two Cellulomonas fimi cellulases by a homologous protease and the effect on catalysis. J. Biol. Chem. 263:10401–7 17. Gouka RJ, Punt PJ, Van Den Hondel CA. 1997. Glucoamylase gene fusions alleviate limitations for protein production in Aspergillus awamori at the transcriptional and (post) translational levels. Appl. Environ. Microbiol. 63:488–97 18. Graf E. 1992. Antioxidant potential of ferulic acid. Free Radic. Biol. Med. 13:435–48 19. Gusakov AV, Salanovich TN, Antonov AI, Ustinov BB, Okunev ON, et al. 2007. Design of highly efficient cellulase mixtures for enzymatic hydrolysis of cellulose. Biotechnol. Bioeng. 97:1028–38 20. Gusakov AV, Sinitsyn AP, Salanovich TN, Bukhtojarov FE, Markov AV, et al. 2005. Purification, cloning and characterization of two forms of thermostable and highly active cellobiohydrolase I (Cel7A) produced by the industrial strain of Chrysosporium lucknowense. Enzyme Microbiol. Technol. 36:57–69 21. Heerikhuisen M, van den Hondel CAMJJ, Punt PJ. 2005 Aspergillus sojae. In Production of Recombinant Proteins. Novel Microbial and Eukaryotic Expression Systems, ed. G Gellissen, pp. 191–214. Weinheim: Wiley-VCH Verlag 22. Hinz SWA, Pouvreau L, Joosten R, Bartels J, Jonathan MC, et al. 2009. Hemicellulase production in Chrysosporium lucknowense C1. J. Cereal Sci. 50:318–23 23. Huang C. 2009. Receptor-Fc fusion therapeutics, traps, and MIMETIBODYTM technology. Curr. Opin. Biotechnol. 20:692–99 24. Ibarra D, Romero J, Mart´ınez MJ, Mart´ınez AT, Camarero S. 2006. Exploring the enzymatic parameters for optimal delignification of eucalypt pulp by laccase mediator. Enzyme Microbiol. Technol. 39:1319–27 www.annualreviews.org • Fungal Protein Production
10. Provides an overview of one of the major bottlenecks in protein production in filamentous fungi.
17. Provides an overview of one of the most successful approaches for heterologous protein production in filamentous fungi.
67
MI65CH04-Punt
ARI
23 May 2011
25. Describes a unique cellulolytic multiprotein complex, the cellulosome.
42. Discovery of a completely new noncatalytic cellulose active protein, swollenin.
43. Describes carbohydrate binding modules and their applications.
68
17:37
25. Lamed R, Setter E, Kenig R, Bayer EA. 1983. The cellulosome: a discrete cell surface organelle of Clostridium thermocellum which exhibits separate antigenic, cellulose-binding and various cellulolytic activities. Biotechnol. Bioeng. Symp. 13:163–81 26. Lesage-Meessen L, Delattre M, Haon M, Thibault JF, Ceccaldi BC, et al. 1996. A two-step bioconversion process for vanillin production from ferulic acid combining Aspergillus niger and Pycnoporus cinnabarinus. J. Biotechnol. 50:107–13 27. Levasseur A, Navarro D, Punt PJ, Bela¨ıch JP, Asther M, Record E. 2005. Construction of engineered bifunctional enzymes and their overproduction in Aspergillus niger for improved enzymatic tools to degrade agricultural by-products. Appl. Environ. Microbiol. 71:8132–40 28. Levasseur A, Pag`es S, Fierobe HP, Navarro D, Punt P, et al. 2004. Design and production in Aspergillus niger of a chimeric protein associating a fungal feryloyl esterase and a clostridial dockerin domain. Appl. Environ. Microbiol. 70:6984–91 29. Levasseur A, Saloheimo M, Navarro D, Andberg M, Monot F, et al. 2006. Production of a chimeric enzyme tool associating the T. reesei swollenin with the Aspergillus niger feruloyl esterase A for release of ferulic acid. Appl. Microbiol. Biotechnol. 73:872–80 30. Li X, Chen H, Ljungdahl L. 1997. Two cellulases, CelA and CelC, from the polycentric anaerobic fungus Orpinomyces strain PC-2 contain N-terminal docking domains for a cellulase—hemicellulase complex. Appl. Environ. Microbiol. 63:4721–28 31. Madzak C, Otterbein L, Chamkha M, Moukha S, Asther M, et al. 2005. Heterologous production of a laccase from the basidiomycete Pycnoporus cinnabarinus in the dimorphic yeast Yarrowia lipolytica. FEMS Yeast Res. 5:635–46 32. Martinez D, Berka RM, Henrissat B, Saloheimo M, Arvas M, et al. 2008. Genome sequencing and analysis of the biomass-degrading fungus Trichoderma reesei (syn Hypocrea jecorina). Nat. Biotechnol. 26:553–60 33. Mathew S, Abraham TE. 2004. Ferulic acid: an antioxidant found naturally in plant cell walls and feruloyl esterases involved in its release and their applications. Crit. Rev. Biotechnol. 24:59–83 34. Nagy T, Tunnicliffe RB, Higgins LD, Walters C, Gilbert HJ, Williamson MP. 2007. Characterization of a double dockerin from the cellulosome of the anaerobic fungus Piromyces equi. J. Mol. Biol. 373:612–22 35. Oddou J, Stentelaire C, Lesage-Meesen L, Asther M, Ceccaldi BC. 1999. Improvement of ferulic acid bioconversion into vanillin by use of high-density cultures of Pycnoporus cinnabarinus. Appl. Microbiol. Biotechnol. 53:1–6 36. Otterbein L, Record E, Longhi S, Asther M, Moukha S. 2000. Molecular cloning of the cDNA encoding laccase from Pycnoporus cinnabarinus I-937 and expression in Pichia pastoris. Eur. J. Biochem. 267:1619–25 37. Peer A, Smith SP, Bayer EA, Lamed R, Borovok I. 2009. Noncellulosomal cohesin- and dockerin-like modules in the three domains of life. FEMS Microbiol. Lett. 291:1–16 38. Pham TA, Berrin JG, Record E, To Kim A, Sigoillot JC. 2010. Hydrolysis of softwood by Aspergillus mannanase: role of a carbohydrate binding module. J. Biotechnol. 148:163–70 39. Raghothama S, Eberhardt RY, Simpson P Wigelsworth D, White P, et al. 2001. Characterization of a cellulosome dockerin domain from the anaerobic fungus Piromyces equi. Nat. Struct. Biol. 8:775–78 40. Ravalason H, Herpo¨el-Gimbert I, Record E, Bertaud F, Grisel S, et al. 2009. Fusion of a family 1 carbohydrate binding module to the Pycnoporus cinnabarinus laccase for efficient softwood kraft pulp biobleaching. J. Biotechnol. 142:220–26 41. Record E, Punt PJ, Chamkha M, Labat M, Van Den Hondel CAMJJ, Asther M. 2002. Expression of the Pycnoporus cinnabarinus laccase gene in Aspergillus niger and characterization of the recombinant enzyme. Eur. J. Biochem. 269:602–9 42. Saloheimo M, Paloheimo M, Hakola S, Pere J, Swanson B, et al. 2002. Swollenin, a Trichoderma reesei protein with sequence similarity to the plant expansins, exhibits disruption activity on cellulosic materials. Eur. J. Biochem. 269:4202–11 43. Shoseyov O, Shani Z, Levy I. 2006. Carbohydrate binding modules: biochemical properties and novel applications. Microbiol. Mol. Biol. Rev. 70:283–95 44. Sigoillot C, Record E, Belle V, Robert JL, Levasseur A, et al. 2004. Natural and recombinant fungal laccases for paper pulp bleaching. Appl. Microbiol. Biotechnol. 64:346–52 45. Steenbakkers PJ, Li XL, Ximenes EA, Arts JG, Chen H, et al. 2001. Noncatalytic docking domains of cellulosomes of anaerobic fungi. J. Bacteriol. 183:5325–33 Punt et al.
MI65CH04-Punt
ARI
23 May 2011
17:37
46. Tomme P, Van Tilbeurgh H, Pettersson G, Van Damme J, Vandekerckhove J, et al. 1988. Studies of the cellulolytic system of Trichoderma reesei QM 9414. Analysis of domain function in two cellobiohydrolases by limited proteolysis. Eur. J. Biochem. 170:575–81 47. Tramper J, Zhu Y. 2011. Modern Biotechnology: Panacea or New Pandora’s Box? Wageningen: Wageningen Acad. Publ. 48. Ustinov BB, Gusakov AV, Antonov AI, Sinitsyn AP. 1988. Comparison of properties and mode of action of six secreted xylanases from Chrysosporium lucknowense. Enzyme Microbiol. Technol. 43:56–65 49. Van de Kaaij RM, Ram AFJ, Schaap P, Punt PJ. 2010. Genomic approaches for identification of the biopolymer degrading enzyme network of Aspergillus niger. In The Mycota, Vol. 10, Pt. 2: Industrial Applications, ed. M. Hofrichter pp. 407–24. Heidelberg: Springer-Verlag GmbH 50. Van der Kaaij RM, Yuan XL, Franken A, Ram AF, Punt PJ. 2007. Characterization of two novel, putatively cell wall-associated and GPI-anchored, alpha-glucanotransferase enzymes of Aspergillus niger. Eukaryot. Cell 6:1178–88 51. Verdoes JC, Punt PJ, Burlingame R, Bartels J, Van Dijk R. 2007. A dedicated vector for efficient library construction and high throughput screening in the hyphal fungus Chrysosporium lucknowense. Ind. Biotechnol. 3:48–57
RELATED RESOURCES Carbohydrate-Active enZYme Database: http://www.cazy.org/ Fungal Genome Initiative: http://www.broadinstitute.org/annotation/fungi/fgi/ Fungal Genomics Program, Joint Genome Institute, Department of Energy: http:// genome.jgi-psf.org/programs/fungi/index.jsf
www.annualreviews.org • Fungal Protein Production
69
