Abstract

Background

The secretomes of thermophilic fungi have been recently identified as promising sources of enzymes for improving the biochemical conversion of lignocellulosic biomass to biofuels [1]. Readily culturable thermophilic fungi are saprophytic and found in a range of environments, such as compost, where decomposition of organic matter occurs at elevated temperatures [2]. Glycoside hydrolase (GH) enzymes from thermophilic fungi function from 60–80°C [1,3-5], whereas commercial fungal cocktails produced by mesophilic fungi perform optimally at 50°C. Performing saccharifications of pretreated biomass at elevated temperatures contributes to increased conversion rates [6] resulting in shorter incubation times and lower enzyme loadings, as well as reducing the risk of downstream contamination by competing microorganisms for fermentable sugars. Therefore, thermophilic fungi may provide new enzymatic cocktails that are optimized for the industrial biochemical conversion of biomass to sugars for fermentation to biofuels.

Prior to fermentation, lignocellulosic biomass (e.g., corn stover, switchgrass) is typically pretreated chemically to reduce its recalcitrant properties. Pretreatment increases the efficiency of subsequent enzymatic hydrolysis of plant polysaccharides. Certain ionic liquids (ILs) have shown excellent promise in generating cellulose from biomass that readily hydrolyzed by enzymatic cocktails [7]. The majority of ILs that are effective biomass solvents are also toxic to microbes, however fungi have been shown to grow in the presence of high concentrations of some ILs [8]. Recently, a strain of the thermotolerant fungus Aspergillus fumigatus was grown on switchgrass in the presence of 5% 1-butyl-3-methylimidazolium chloride, abbreviated as [C₄mim]Cl, secreting high amounts of cellulase and hemicellulase enzymes. These GHs were shown to retain residual activity in up to 20% [C₄mim]Cl [9]. In addition, increased IL tolerance has been demonstrated for thermophilic glycoside hydrolases from bacteria [10,11] and archaea [12], suggesting a correlation between thermotolerance and IL tolerance. Therefore, thermophilic fungal enzymes may express enzymes that are tolerant to residual IL present in pretreated biomass, enabling a more cost-effective IL conversion process with minimal washing required after pretreatment.

In addition, culture supernatants of thermophilic fungi have been instrumental in identifying new accessory proteins that improve cellulose hydrolysis. Increased glucose release from pretreated barley straw was observed with mixtures of Cellulclast and Novozymes 188 cocktails supplemented with thermophilic fungal crude extracts, indicating that thermophilic fungi produce enhancers for cellulose hydrolysis [13]. Biochemical and proteomic studies identified these enhancement factors as proteins that are members of GH family 61. Addition of purified GH61 proteins from thermophilic fungi Thielavia terrestris or Thermoascus aurantiacus to a commercial Trichoderma reesei enzyme cocktail gave improved cellulose hydrolysis, resulting in 2-fold lower protein loading requirements [14]. Detailed characterization of GH61 from Thermoascus aurantiacus demonstrated that these proteins are Cu-containing monoxygenases that oxidize cellulose chains, facilitating glycoside hydrolase depolymerization [15]. These discoveries demonstrate the potential for studying thermophilic fungi as sources of proteins to improve hydrolysis of lignocellulosic biomass.

Our aim is to gain an understanding of the mechanisms that thermophilic fungi use to deconstruct biomass and develop industrially relevant enzyme cocktails for saccharification of pretreated biomass. Previous studies of thermophilic fungi have indicated that Thielavia terrestris and Thermoascus aurantiacus are promising species for the production of a thermophilic enzymatic cocktail for a lignocellulose deconstruction [13]. In this work, enzymatic hydrolytic activities of supernatants from T. aurantiacus and T. terrestris grown on untreated and chemically pretreated biomass were compared. This comparison, performed on artificial biomass substrates and pretreated biomass, established T. aurantiacus as a promising system to develop thermophilic enzymatic cocktails.

Results

Supernatant activities on artificial substrates

Cellulase (endoglucanase, cellobiohydrolyase, beta-glucosidase) and hemicelluase (xylanase, beta-xylosidase, arabinofuranosidase) activities were measured for these supernatants using artificial substrates (Table ​(Table2).2). Comparison of the cellulase profiles indicated that higher levels of endoglucanase was present in the T. terrestris supernatant but the T. aurantiacus supernatant possesses higher levels of cellobiohydrolase and β-glucosidase activity. For hemicellulosic substates, xylanase activity was higher for T. aurantiacus but lower for β-xylosidase activity. In general, the highest activities for the artificial biomass substrates were observed on supernatants generated from growth with SG and MCC.

Table 2

Secreted cellulase and hemicellulase activities (U/mg or mU/mg) from thermophilic fungi grown on complex biomass substrates^1,2,3

Strain/Carbon source	CMC (U/mg)	pNPC (mU/mg)	pNPG (mU/mg)	Xylan (U/mg)	pNPX (mU/mg)
T. aurantiacus
Switchgrass	0.17 ± 0.03	34.4 ± 1.31	177.3 ± 2.71	0.67 ± 0.01	8.8 ± 0.81
AFEX Switchgrass	0.04 ± 0.01	25.7 ± 0.40	45.6 ± 1.71	0.10 ± 0.02	12.4 ± 1.63
IL Switchgrass	0.02 ± 0.01	9.8 ± 0.26	54.8 ± 0.43	0.35 ± 0.01	10.3 ± 1.09
Microcrystalline Cellulose	1.94 ± 0.22	12.5 ± 1.60	93.8 ± 12.86	1.90 ± 0.62	6.92 ± 0.49
T. terrestris
Switchgrass	0.79 ± 0.02	11.3 ± 0.62	26.1 ± 1.14	2.44 ± 0.38	5.7 ± 0.47
AFEX Switchgrass	0.24 ± 0.01	3.7 ± 0.07	5.7 ± 1.21	0.66 ± 0.05	0.8 ± 0.37
IL Switchgrass	0.23 ± 0.02	7.8 ± 0.25	5.7 ± 0.23	1.02 ± 0.05	3.8 ± 0.41
Microcrystalline Cellulose	1.24 ± 0.23	16.2 ± 4.38	6.9 ± 1.76	3.27 ± 1.12	4.17 ± 1.27

Open in a separate window

¹Specific activities based on protein concentrations obtained using the Bradford method.

² Three replicates were performed and the error bars represent standard deviation from the mean.

³ Activities represented by each substrate is described in the Methods section.

The eight culture supernatants were screened for glucose release on lignocellulosic biomass substrate; these tests were performed to determine the most active samples from each fungal strain. The enzymatic activities were compared to a commercially available enzyme cocktail, Cellic CTec2 (Novozymes), in the saccharification of [C₂mim][OAc]-treated switchgrass. Time course data on [C₂mim][OAc]-treated switchgrass demonstrated that T. aurantiacus enzymes performed comparably to CTec2 at 50°C, while T. terrestris enzymes released between considerably less glucose compared to the T. aurantiacus supernatant (Figure​Figure1).1). For each strain, supernatants prepared from cultures on AFEX-SG were the most active of all the samples in cellulose hydrolysis, followed by supernatants from untreated switchgrass (SG), [C₂mim][OAc]-treated switchgrass (IL-SG), and microcrystalline cellulose (MCC), respectively.

Object name is 1754-6834-5-54-1.jpg" title="Click on image to zoom" class="tileshop" src="/pmc/articles/PMC3507748/bin/1754-6834-5-54-1.jpg"/>

Open in a separate window

Figure 1

Release of glucose from 2% [C₂mim][OAc]-pretreated switchgrass. In the reactions, T. aurantiacus and T. terrestris culture supernatants were normalized to 1U CMCase activity. Glucan conversion (based on glucose only): 7 – 68%. Three replicates were performed and the error bars represent standard deviation from the mean.

Saccharification with culture supernatants from thermophilic fungi

The saccharification experiment described above established AFEX-grown cultures of T. aurantiacus and T. terrestris as promising sources of enzymes from thermophilic fungi. [C₂mim][OAc] treated switchgrass (IL-SG) was used as a substrate, which has been shown to be an excellent substrate for commercial enzymatic cocktails [7]. Comparable amounts of glucose were released from IL-SG from T. aurantiacus enzymes and CTec2 at a range of protein loadings (Figure ​(Figure2).2). Activities from T. terrestris were consistently lower at all protein loading levels. These trends were also observed when saccharifications were performed at different temperatures (Figure ​(Figure3).3). The CTec2 cocktail and the AFEX-SG grown T. aurantiacus supernatant had the highest relative activity at 50°C; however the T. aurantiacus supernatant released almost 3-fold more glucose from IL-SG at 70°C than CTec2 and the T. terrestris supernatant. In contrast, the AFEX-SG T. terrestris supernatant had an activity maximum at 60°C, but the T. aurantiacus enzymes were 2-fold more active at the same temperature.

Object name is 1754-6834-5-54-2.jpg" title="Click on image to zoom" class="tileshop" src="/pmc/articles/PMC3507748/bin/1754-6834-5-54-2.jpg"/>

Open in a separate window

Figure 2

Glucose released from [C₂mim][OAc]-treated switchgrass at different enzyme loadings for culture supernatants from AFEX-SG grown T. aurantiacus and T. terrestris compared to equivalent enzyme loading from Cellic CTec2. The reactions were performed at 60°C with 2% (w/v) [C₂mim][OAc]-pretreated switchgrass. Three replicates were performed and the error bars represent standard deviation from the mean.

Object name is 1754-6834-5-54-3.jpg" title="Click on image to zoom" class="tileshop" src="/pmc/articles/PMC3507748/bin/1754-6834-5-54-3.jpg"/>

Open in a separate window

Figure 3

Temperature profile of the saccharification of [C₂mim][OAc]-pretreated switchgrass with T. aurantiacus/T. terrestris culture supernatants and Cellic CTec2. Enzyme loading was set at 0.06 g enzyme/g glucan and biomass loading was 2% [C₂mim][OAc]-pretreated switchgrass. Three replicates were performed and the error bars represent standard deviation from the mean.

Saccharifications of this supernatant were the tested against dilute acid and AFEX-SG, which are more recalcitrant substrates compared to IL-SG [7,17]. Comparison of saccharification of dilute acid and AFEX-SG switchgrass by the T. aurantiacus and CTec2 indicated that at high enzyme loadings, more glucose was released by the T. aurantiacus supernatant compared to the CTec2 cocktail for all three substrates (Figure ​(Figure4).4). High and low enzyme loadings were applied as suggested by CTec2 recommendations, with 1 – 6% (w/w) enzyme to cellulose representing the range of low to high enzyme concentrations.

Object name is 1754-6834-5-54-4.jpg" title="Click on image to zoom" class="tileshop" src="/pmc/articles/PMC3507748/bin/1754-6834-5-54-4.jpg"/>

Open in a separate window

Figure 4

Conversion of glucan to glucose and cellobiose using culture supernatants from AFEX-SG grown T. aurantiacus and Cellic CTec2 on various pretreated biomass substrates. Reaction conditions were: (2% w/v) biomass, 60°C, pH 5 for 72 hours with mixing. Switchgrass substrates were pretreated in three different ways: IL – ionic liquid, [C₂mim][OAc]; DA – dilute acid, H₂SO₄; AFEX-, NH₄. The saccharification reactions were performed at two different enzyme loadings, corresponding to recommended high and low enzyme loadings for Cellic Ctec2 (http://www.bioenergy.novozymes.com/). Equal amounts of protein from both T. aurantiacus supernatants and Cellic CTec2 were added to each reaction (g enzyme/g glucan): AFEX-high/low ∼ 0.065/0.016; IL-high/low ∼ 0.046/0.011; DA-high/low ∼ 0.043/0.010. Three replicates were performed and the error bars represent standard deviation from the mean.

Proteomic analysis of the T. aurantiacus secretome

Proteomic analysis of the T. aurantaicus supernatants was performed to determine abundant proteins that may be responsible for the observed saccharification activity. Analysis of the T. aurantiacus supernatants showed distinct protein band patterns for each cultivation condition on untreated and pretreated switchgrass (Figure ​(Figure5).5). The AFEX-SG supernatant had three prominent protein bands (∼50 kDa, ∼35 kDa, and ∼25 kDa), which were present at lower relative abundances in the other cultivations. Mass spectrometry analysis of the peptides from these three bands revealed that they predominantly consisted of four glycoside hydrolases: GH 7 (cellobiohydrolase), GH 5 (endoglucanase), GH10 (xylanase), and GH61 (copper-dependent polysaccharide monooxygenase) (Table ​(Table3).3). Visual comparison to T. terrestris demonstrated that there were no comparable prominent bands observed for the three cultivation conditions, which was consistent with the lower activity, therefore theT. terrestrisbands were not excised for analysis.

Object name is 1754-6834-5-54-5.jpg" title="Click on image to zoom" class="tileshop" src="/pmc/articles/PMC3507748/bin/1754-6834-5-54-5.jpg"/>

Open in a separate window

Figure 5

SDS-PAGE analysis of supernatants fromT. aurantiacus and T. terrestris on 8-16% Tricine polyacrylamide gels. Lane 1, SG-grown T. aurantiacus; lane 2, AFEX-SG grown T. aurantiacus; lane 3, IL-SG grown T. aurantiacus; lane 4, MW markers; lane 5, SG-grown T. terrestris; lane 6, AFEX-SG grown T. terrestris; lane 7, IL-SG T. terrestris.

Table 3

Mass spectrometry analysis of major protein bands isolated fromT. aurantiacusAFEX-SG secretome.

Prominent band	Description	Predicted MW	Peptides (95%)1	% Sequence coverage (95)2
∼50 kDa	Possible GH 7 Cellobiohydrolase [GenBank: "type":"entrez-protein","attrs":"text":"CAD79781","term_id":"29160309","term_text":"CAD79781"CAD79781]	48 kDa	250	72
∼35 kDa	GH 5 Endoglucanase [GenBank: "type":"entrez-protein","attrs":"text":"CAZ67882","term_id":"251730792","term_text":"CAZ67882"CAZ67882]/GH10 Xylanase [GenBank: "type":"entrez-protein","attrs":"text":"CAB65468","term_id":"6689357","term_text":"CAB65468"CAB65468]	37 kDa/36 kDa	249/176	72/65
∼25 kDa	GH 61 [PDB: 3ZUD_A]	24 kDa	60	93

Open in a separate window

¹ Number of unique peptides matching protein with 95% confidence.

² Sequence coverage of identified protein with unique proteins matching with 95% confidence.

Discussion

In this work, we have shown that culture supernatant recovered from growth of T. aurantiacus on AFEX-SG exhibited high levels of saccharification activity on IL-SG. The higher levels of protein in the supernatants from the AFEX-SG grown supernatants may be due to additional nitrogenous compounds formed during the AFEX process [18]. The T. aurantiacus activity was 2–4 fold higher than the culture supernatant from T. terrestris at comparable protein amounts. CMCase activity was not predictive of saccharification activity, as supernatants from T. aurantiacus showed up to 11.5-fold lower activity on CMC compared to supernatants from T. terrestris. Cellobiohydrolase (pNPC) and β-glucosidase (pNPG) activities correlated with overall saccharification activities, demonstrating that exo-acting cellulases are critical for efficient saccharification of cellulose to glucose [19].

T. aurantiacus supernatant saccharification activities were comparable to the commercial enzymatic mixture Cellic CTec2 at 50°C at similar protein mass loadings and retained higher levels of activity up to 70°C. The temperature stability of the T. aurantiacus enzyme mixture is consistent with assays performed on purified endoglucanase and cellobiohydrolase enzymes from T. aurantiacus strains [4]. Cellic CTec2 has been shown to produce oxidized sugar products gluconic acid and cellobionic acid, suggesting the presence of GH61, a recently discovered Cu-containing cellulose monoxygenase expressed by many fungi, in the enzyme mixtures [20]. GH 61 from T. aurantiacus has been shown to reduce enzyme loadings 2-fold with Celluclast/N188 enzymatic cocktails from Novozymes and to be part of a cellulose oxidoreductase enzymatic system with cellobiose dehydrogenase that is synergistic with the well-studied hydrolytic system [13,15]. The presence of a prominent band in the corresponding to a GH61 protein at ∼25 kDA in SDS-PAGE analysis of the T. aurantiacus supernatants suggests that it may contribute to the high observed saccharification activity of T aurantiacus supernatants. The observed dominance of GH7, GH5, GH61 and GH10 in the AFEX-SG grown T. aurantiacus supernatants is similar to the GH profile of an optimal mixture of enzymes for glucose release from pretreated corn stover that was designed through computational modeling and high-throughput assays, a method referred to as GENPLANT [21]. The similarity between the GENPLANT-designed cocktail and the T. aurantiacus supernatants suggests that T. aurantiacus possesses very favorable properties for efficient cellulose hydrolysis.

Using chemical/UV mutagenesis and plate-based screeing, strains of Trichoderma reesei have been developed that secrete large amounts of cellulases and these strains are the basis for industrial production of cellulases for biomass deconstruction [22]. T. aurantiacus is an attractive candidate for a similar program of mutagenesis and screening, because it already secretes large amounts of cellulase enzymes, as demonstrated in the work described here, and it has a readily observed sexual state that could be used to select for specific mutations to enhance secretion of cellulases and remove deleterious mutations that accumulate during random mutagenesis and screening [23]. Therefore, T. aurantiacus has the potential to be a useful fungal species to understand cellulase secretion and biomass deconstruction under thermophilic conditions.

In conclusion, cultivation of T. aurantiacus and T. terrestris on biomass substrates revealed that T. aurantiacus grown on AFEX-pretreated switchgrass afforded high levels of cellulase enzymes in culture supernatants that were used to saccharify pretreated switchgrass at higher temperature and lower protein loadings than Cellic CTec2. Proteomic analysis of the supernatant demonstrated that the high activity could be attributed to the presence of high levels of proteins from the GH5, GH7, GH10 and GH61 families, which have been shown to be key enzymes in efficient cellulose hydrolysis. This work demonstrates that T. aurantiacus is a promising platform for the production of thermophilic enzymes for biomass deconstruction [1].

Methods

Abbreviations

AFEX: Ammonia Fiber Expansion pretreatment; BCA: Bicinchronic Acid; bG: beta-Glucosidase; [C2mim][OAc]: 1-ethyl, 3-methyl imidazolium acetate; CBH: Cellobiohydrolase; CMC: Carboxymethyl Cellulose; IL: Iionic Liquid pretreatment (in this work, 1-ethyl-3-methylimidazolium acetate); MCC: Microcrystalline Cellulose; SG: Switchgrass.

Competing interests

The authors declare that they have no competing interests.

Author’s contribution

SDM and SWS designed the experiments. SDM performed all experiments, analyzed data, and wrote the manuscript. TB and CP setup and performed proteomic sample analysis. SWS, BAS and PAD reviewed and edited the manuscript. All authors read and approved the final manuscript.

References