Biological pretreatment of wheat straw: Effect of fungal culturing on enzymatic hydrolysis of carbohydrate polymers

The aim of the study was comparative analysis of degradation of wheat straw lignin by white-rot fungi and its implications on the eciency of enzymatic hydrolysis of holocellulose. Cyclocybe cylindracea, Ganoderma resinaceum, Irpex lacteus, Pleurotus ostreatus and Trametes versicolor were the species studied. Peroxidases were predominantly responsible for lignin degradation even though high laccase activities were detected, except in the case of Irpex lacteus where laccase activity was not detected. Studied fungal species showed various ability to degrade lignin in wheat straw which further affected release of reducing sugars during enzymatic saccharication. The highest rate of lignin degradation was noticed in sample pretreated with Irpex lacteus (50.9 ± 4.1%). Among all tested species only Ganoderma resinaceum was suitable lignin degrader with the 2-fold higher hydrolysis yield (51.1 ± 4.7%) than in the control, and could have signicant biotechnological application due to lower cellulose loss. A key mechanism of carbohydrate component convertibility enhancement was lignin removal in the biomass. Long time consumption, the low sugar yields and unpredictable fungal response still remain the challenge of the fungal pretreatment process. assess the white-rot fungal delignication mechanisms crucial to enhance eciency of enzymatic saccharication of wheat straw after solid-state culturing. (taking into account all data from the beginning to the end of the experiment) is conrmed by one-way ANOVA (P < 0.01). According to performed Tukey post-hoc test, signicant differences (P < 0.01) are observed between I. lacteus and all other studied species except T. versicolor, meaning that in term of dry matter weight loss I. lacteus can be distinguished from other species. 5.4% 34.1 ± 2.5%) selective in delignication among all studied fungal species. One-way ANOVA was used to assess the difference between all fungal species considering rates of degradation, separately for lignin, hemicellulose and cellulose. Signicant difference was conrmed between fungal species toward all three polymers (P < 0.01). Tukey post-hoc test was also performed and for lignin, signicant differences (P < 0.01) were observed between P. ostreatus and all studied fungal species, except C. cylindracea, as well as between C. cylindracea and I. lacteus. Considering hemicelluloses, signicant differences are observed between C. cylindracea and I. lacteus, C. cylindracea and T. versicolor, I. lacteus and P. ostreatus, P. ostreatus and T. versicolor. For cellulose, I. lacteus stands signicant difference between this and all other tested species. this study exhibited various capacities to produce ligninolytic enzymes and contribution of individual enzymes in the enzymatic pool was evident. High peroxidase activity in G. resinaceum, P. ostreatus and T. versicolor was expected and is in line with results reported by other authors 12,18,25 . If not absent, laccase activity in all cases was higher at the beginning of cultivation expressing the trend of decrease in activity level after day 10 or day 15, which was also observed in other researches 26 . This can be highly related to intensive degradation of small molecular weight compounds by laccase in the initial phase of fungal growth 27 . Comparing all studied species it can be observed that I. lacteus was the only species without laccase activity. MnP has been considered the main ligninolytic enzyme in I. lacteus which is in agreement with previous studies 18 . loss is expected during pretreatment with white rot fungi the relative carbohydrate enrichment of the biomass, i.e. by comparing the ratio carbohydrates:lignin in pretreated wheat straw samples was relatively lower. Previous research also showed that the decrease in wheat straw lignin as the result of fungal culturing produced a low enhancement of carbohydrate component by increasing glucan ratio in the biomass 28 . Pretreatment of lignocellulosic biomass with white rot fungi such as I. lacteus and Ceriporiopsis subvermispora have been previously studied 41,42,43 . As in the occasion of G. resinaceum of our study, an ecient enzymatic conversion of pretreated lignocellulosic material was achieved, but the increased conversion came at the expense of considerable loss of holocellulose during pretreatment. Furthermore, a positive effect of delignication has not been seen always. These inequalities occurred either due to weak delignication rate or as a result of substantial total solids loss 43 . Although higher coecient of conversion was not achieved in biologically pretreated samples the approach still remains justied, as results of this study indicate that the crucial moment can be the period of fungal culturing in which the effect of an up-concentration of carbohydrates should be stopped after partial delignication, to avoid substantial sugar consumption by fungi. Comparing with biologically pretreated wheat straw after chemical pretreatment with NaOH the higher amount of fermentable sugars was released. Considerably higher amounts of glucose and cellobiose and negligible presence of pentoses in hydrolysate of alkali pretreated wheat straw was the result of selective removal of lignin from biomass 44 . Process was followed by removal of acetyl and uronic acid groups on hemicelluloses and hydrolysis of ester linkages between xylan and hemicelluloses residues. All together this largely improved cellulose digestibility exhibiting minor cellulose loss comparing with fungal pretreated samples 44 . Previous studies also showed that chemical pretreatment is more ecient than biological pretreatment due to the partial loss of carbohydrates which are the carbon source for Another impact of enzymatic hydrolysis of pretreated accumulation of cellobiose. beta-glucosidase activity in enzymatic for saccharication. cellobiose 19 of culturing for ligninolytic enzyme extraction, established . The were for determination of hemicelluloses, cellulose and lignin contents and for enzymatic saccharication. declared All statistical analyses were performed using the statistical package and software STATISTICA, 6.0 Principal component analysis (PCA) showing the relationship between the degradation of lignin and ligninolytic enzyme activities of all taxa was performed, for each enzyme separately. The values of lignin degradation by selected fungal species were used as response variables, while the enzyme activity of the taxa and time variable showing the number of days after which the measurements were performed were used as supplementary ones. Multivariate analyses were done using Canoco 5 software 55 .


Introduction
The global demands for energy today are considerably dependent on fossil fuels but the opportunities for exploitation of renewable energy sources are increasingly growing. In this perspective, conversion of biomass, as renewable energy source, into second generation biofuels is a promising approach in green energy technologies, bio-based economy and society devoted to sustainable development 1 .
Lignocellulose is a major component of plant biomass and is primarily composed of cellulose, hemicellulose and lignin that form a complex composite structure 2 . Chemical composition recommends this feedstock for utilization in technological processes based on a sugar platform, where cellulose or holocellulose are hydrolyzed to monosaccharides which are further converted into bioethanol or other chemicals 3 . However, lignin as a high complex aromatic polymer with a three-dimensional structure is mostly responsible for the recalcitrance of this potentially valuable raw material, which is an obstacle that hinders economically feasible conversion of holocellulose to fermentable sugars 4,5 . Thus, deligni cation is an inevitable step in lignocellulose processing rendering the cellulose and hemicellulose more exposed to enzymatic sacchari cation 6 . Recently, several bio-re nery concepts have been established for converting renewable biomass to fuels but pretreatment of plant residues still remains mostly physical, physico-chemical and chemical. Contrary, biological pretreatment deserves much more attention as it is safe, environmentally friendly and less energy intensive compared to other pretreatment methods. This approach comprises of using different microorganisms to remove lignin with a minimal loss of carbohydrate polymers 7 . However, the rate of hydrolysis reaction is very low and requires a great improvement to be commercially applicable 8 .
Although some groups of bacteria are capable to degrade lignin, white-rot fungi are considered as the most effective lignin degrading organisms 9 . Fungal degradation of lignin is extracellular oxidative process catalyzed by ligninolytic enzymes, mostly laccase, lignin peroxidases and Mn-oxidizing peroxidases 10 . Some white rot fungi degrade lignin non-selectively depolymerising both cellulose and hemicelluloses, so holocellulose loss could range between 17 an 50%, resulting in decreased yield of sacchari cation products 11 . On the other hand, fungal species that exhibit selective lignin degradation have higher a nity for lignin and degrade it faster than holocellulose, though the loss of carbohydrates decreases. Relation between those two polymers losses is a crucial parameter characterizing e ciency of biological pretreatment of lignocellulose and varies among species and even among strains of the same species 11,12 . Previously has been shown that wheat straw can be considered as a promising substrate for ligninolytic enzymes production by cultivating various white-rot fungal species 12,13,14,15 . However, the effects of fungal culturing as a pretreatment method in lignocelluloses processing are not yet well understood. The aim of this study was to assess the white-rot fungal deligni cation mechanisms crucial to enhance e ciency of enzymatic sacchari cation of wheat straw after solid-state culturing.

Activity of ligninolytic enzymes
Studied ve species of macromycetes have shown different levels of MnP, MnIP and laccase activities after 5, 10, 15 and 19 days of solid-state cultivation on wheat straw as a substrate (Fig. 1). Based on the results of ligninolytic enzyme activities it can be noticed that activities levels varied during the fungal cultivation.
The highest level of MnP activity was detected in Pleurotus ostreatus with the maximum activity recorded after 10 days of cultivation (996.2 ± 69.1 U L − 1 ), while the lowest activity was noticed in Cyclocybe cylindracea after 15 days of cultivation (8.5 ± 2.3 U L − 1 ).
The maximum value of MnIP activities was also noted in P. ostreatus (984.2 ± 84.7 U L − 1 ) after 10 days of cultivation, and the minimum activity was measured in C. cylindracea after 10 days of cultivation (9.5 ± 3.8 U L − 1 ). 68.8 ± 0.5 6.0 ± 0.7 The rates of polymers degradation in wheat straw differed among studied species (Table 1). In some occasions (T. versicolor, P. ostreatus and G. resinaceum) capacities of studied species to degrade hemicelluloses were signi cantly higher in comparison with their abilities to degrade cellulose. The highest rate of lignin, hemicelluloses and cellulose degradation was noted after pretreatment with I. lacteus (50.9 ± 4.1%, 53.9 ± 1.1% and 62.0 ± 3.3%, respectively). The lowest rate of lignin and cellulose degradation was noted after pretreatment with P. ostreatus (26.8 ± 1.7% and 29.6 ± 1.6%), while the minimum hemicelluloses degradation was achieved after pretreatment with C. cylindracea (26.8 ± 2.6%). Compared with other species, G. resinaceum achieved the maximum difference between the level of lignin and cellulose degradation rates (44.5 ± 5.4% vs. 34.1 ± 2.5%) which characterized this species as the most selective in deligni cation among all studied fungal species. One-way ANOVA was used to assess the difference between all fungal species considering rates of degradation, separately for lignin, hemicellulose and cellulose. Signi cant difference was con rmed between fungal species toward all three polymers (P < 0.01). Tukey post-hoc test was also performed and for lignin, signi cant differences (P < 0.01) were observed between P. ostreatus and all studied fungal species, except C. cylindracea, as well as between C. cylindracea and I. lacteus. Considering hemicelluloses, signi cant differences are observed between C. cylindracea and I. lacteus, C. cylindracea and T. versicolor, I. lacteus and P. ostreatus, P. ostreatus and T. versicolor. For cellulose, I. lacteus stands out considering the degradation of this polymer, because signi cant difference is observed between this and all other tested species.
The Vectors referring to laccase activities of fungal species are oriented toward the right side of the ordination diagram con rming high laccase activities in the initial stages of fungal cultivation (Fig. 2c). Contrary to the MnP and MnlP, laccase activity of all taxa was negatively correlated with the degradation of lignin.

Enzymatic hydrolysis of wheat straw
Two crucial factors affecting the hydrolysis process with prede ned conditions, enzyme loading and particle size were investigated to determine the optimal parameters for further enzymatic hydrolysis of pretreated wheat straw.
As shown in Fig. 3, the reducing sugars production pro les in all investigated samples were similar i.e. continual increase occurred during the hydrolysis. Using the different concentrations of cellulase (0.5; 1.0 and 2.0 µL/mL) to determine the optimal concentration of enzyme required for wheat straw sacchari cation, continual increase of reducing sugars concentration in hydrolysate was noticed in the period of 72 h (Fig. 3a). Hydrolysis rate of untreated wheat straw depended on the cellulase concentration to the certain level, so the lowest value of released sugars was detected at enzyme loading of 0.5 µL/mL (5.3 ± 0.2 gGE L − 1 ). The highest content of reducing sugars was achieved at enzyme loading of 1.0 µL/mL (6.0 ± 0.1 gGE L − 1 ) after 72 h, and was not further increased with the enzyme loading of 2.0 µL/mL (Fig. 3a).
Obtained results showed that the particle size of wheat straw used for enzymatic hydrolysis signi cantly affect the rate of reducing sugars release (P < 0.01). Correspondingly, the highest content of reducing sugars in hydrolysate was detected after 72h in samples containing S fraction (6.0 ± 0.1 gGE L − 1 ), while the minimum was noticed in samples with L fraction (2.5 ± 0.0 gGE L − 1 ) (Fig. 3b).
Fungal culturing was further conducted with S fraction to investigate the effect of ligninolytic enzymes secretion by selected white-rot fungi on the release of reducing sugars during enzymatic hydrolysis of pretreated wheat straw. Pretreatment of wheat straw with selected white rot fungi differently affected the rate of reducing sugars release during the enzymatic hydrolysis (Fig. 4). Maximum amount of reducing sugars was released from wheat straw pretreated with G. resinaceum (7.3 ± 0.5 gGE L − 1 ), while the minimum was detected after pretreatment with C.
cylindracea (1.3 ± 0.1 gGE L − 1 ) (Fig. 4). Results indicates that among all tested fungal species only pretreatment with G. resinaceum increased total sugar yield for 21.7% compared with the control, while wheat straw processing with other tested species brought to lower sugar yields during sacchari cation.
Enzymatic hydrolysis of chemically pretreated wheat straw obtained signi cantly higher amount of reducing sugars compared to biologically pretreated straw, and the maximum amount was achieved after 72 h (18.2 ± 0.2 gGE L − 1 ) (Fig. 5). Signi cant differences considering time, as well as fungal species, were con rmed by one-way ANOVA (P < 0.01) when reducing sugars are used as a quantitative variable. Signi cant differences between all species pairs were assessed using Tukey post-hoc test, according to which C. cylindracea and G. resinaceum were separated, con rming that they have the most different effect on concentration of reducing sugars.
The e ciency of the pretreatment was evaluated by enzymatic convertibility of substrate expressed through hydrolysis yield and coe cient of conversion. The highest hydrolysis yield characterized chemically pretreated samples (70.2 ± 0.8%) ( Table 2). Obtained results also showed that differences in hydrolysis yields occurred depending on fungal species used for the pretreatment. Among pretreated samples a maximum convertibility was determined for samples pretreated with G. resinaceum with the highest hydrolysis yield (51.1 ± 4.7%), which was considerably higher than in untreated wheat straw (34.2 ± 0.3%). In all other samples lower values of hydrolysis yields were achieved. The minimum was noticed after sacchari cation of samples pretreated with C. cylindracea (9.4 ± 0.5%). Even thought high hydrolysis yield was achieved after pretreatment with G. resinaceum, coe cient of conversion that refers to a starting sample indicates that difference in enzymatic convertibility among this sample and the control is found negligible (0.335 and 0.341, respectively). The highest K c of chemically pretreated wheat straw (0.660 from theoretical maximum value) distinguishes this method as the most effective in comparison with pretreated samples (Table 2). Samples of the enzymatic hydrolysates of pretreated wheat straw were further analyzed by liquid chromatography to quantify the amount of solubilised sugars. As shown in Table 3 both hexoses and pentoses as well as disaccharides were detected in hydrolysates of samples pretreated with selected white rot fungi. Data show that arabinose and xylose are the predominant monosaccharides in enzymatic hydrolisates of biologically pretreated samples. The signi cant amount of glucose was also detectable in samples but in the form of disaccharide cellobiose.
The higher amount of glucose was released after chemical pretreatment of wheat straw with NaOH (18.4 ± 0.2 g L − 1 ) while the minimum was noticed after pretreatment with C. cylindracea (1.5 ± 0.0 g L − 1 ). Maximum of cellobiose was measured in samples pretreated with T. versicolor (19.2 ± 0.5 g L − 1 ) and the minimum was detected after chemical pretreatment with NaOH (4.0 ± 0.1 g L − 1 ). The maximal amount of arabinose was detected after pretreatment with T. versicolor (18.0 ± 0.3 g L − 1 ) while the highest amount of xylose was measured after enzymatic hydrolysis of untreated wheat straw (42.7 ± 1.2 g L − 1 ). The highest amounts of arabinose and xylose were measured after the pretreatmen with C.
cylindracea (1.2 ± 0.0 g L − 1 and 3.7 ± 0.0 g L − 1 , respectively). Detection of pentoses in hydrolysates of samples indicating that signi cant residual amount of degraded hemicellulose is still present in samples after biological pretreatment. In addition, those sugars became available even after enzymatic hydrolysis conducted by cellulases.
Finally, after simultaneous degradation of both cellulose and hemicellulose during biological pretreatment of wheat straw signi cant differences in amounts of released sugars can be observed. Thus, according to the sugar analysis pretreatment of wheat straw with T. versicolor caused the release of the highest amount of sugars (including hexoses and pentoses) during enzymatic hydrolysis.
As can be seen from Table 3 pretreatment with NaOH was strictly selective to cellulose degradation remaining hemicellulose none degraded. On the other hand, enzymatic hydrolysis of untreated wheat straw resulted in the highest amount of xylose.

Discussion
The substrate used for conversion to carbohydrates in this study was wheat straw pretreated with white-rot fungi applying fungal culturing as the method of pretreatment. Wheat straw is the material with the great potential as sustainable, abundant and the low cost biomass source in European countries 16 . Fungal culturing can be considered as environmental friendly and cost-effective method for pretreatment of lignocellulosic materials 17,18 .
So far, considerable efforts were done in applying white-rot fungal species to pretreat lignocellulosic feedstock even during submerged or solidstate cultivation. The most frequently used species were from genera Bjerkandera, Ceriporiopsis, Ganoderma, Irpex, Phanerochaete, Pleurotus and Trametes 17,18,19,20,21,22 . Even though all these species are recognized as potent producers of ligninolytic enzymes during the cultivation on various substrates, signi cant variations in enzyme activities may occur depending on substrate. That is why is important to elaborate a physiological response on different types of carbon source for every single species or even strain used for pretreatment 23 . Previous studies showed that wheat straw is a suitable substrate for production of ligninolytic enzymes, indicating that the rate of deligni cation depends on speci c enzymes production and the activity level 12,14,15 . Generally, peroxidises are predominantly responsible for lignin degradation and the level of their activity has to be signi cantly related to the rate of lignin degradation 24 . This is speci cally attributed to initial phase of wood degradation when laccase has limited diffusion into non-degraded plant cell wall primarily due to molecule size. Species used for pretreatment of wheat straw in this study exhibited various capacities to produce ligninolytic enzymes and contribution of individual enzymes in the enzymatic pool was evident. High peroxidase activity in G. resinaceum, P. ostreatus and T. versicolor was expected and is in line with results reported by other authors 12,18,25 . If not absent, laccase activity in all cases was higher at the beginning of cultivation expressing the trend of decrease in activity level after day 10 or day 15, which was also observed in other researches 26 . This can be highly related to intensive degradation of small molecular weight compounds by laccase in the initial phase of fungal growth 27 . Comparing all studied species it can be observed that I. lacteus was the only species without laccase activity. MnP has been considered the main ligninolytic enzyme in I. lacteus which is in agreement with previous studies 18 .
Degradation of lignin during fungal growth on lignocellulosic substrate is inevitable process that enables access to both cellulosic and hemicellulosic oligomers. This is a key assumption that candidates this mechanism as a promising tool to be used in lignocelluloses pretreatment for bioethanol production 28 . Contrary to biodegradation processes which apply extracted ligninolytic enzymes targeting the lignin and its derivates only, during white-rot fungal culturing on solid lignocellulosic substrate, deligni cation is followed by the loss of carbohydrates to some extent 18,29 . However, fungal species which selectively degrade lignocellulose have higher a nity for lignin and degrade it faster than carbohydrate components 30 . Previous studies showed that approach of using selective lignin degraders could be of a great importance in biotechnology since the most of holocellulose after pretreatment remains in substrate, and are available for sacchari cation 21 .
Besides lignocelluloses feedstock preparation methods and physical parameters the factor that strongly affects enzymatic hydrolysis of cellulose is the proper amount of applied cellulolytic enzymes 31 . Obtained results clearly showed that increase of cellulase concentration over an optimal level of loading cannot cause higher rate of reducing sugars release. The main reason for this is that cellulose accessibility limits the effectiveness of minimum cellulase loading on the hydrolysis e ciency leaving the part of enzyme molecules unassociated to the substrate and with no effect on hydrolysis yield 32 .
Apart from enzyme concentration, particle size of lignocellulosic feedstock is also a factor that in uences rate and e ciency of enzymatic hydrolysis 33 . This is mainly related to porosity and also substrate accessibility that has been discussed in literature for increased cellulosic conversions. Previous studies showed that particle size is correlated to hydrolysis e ciency indicating that smaller particles of lignocelluloses feedstock are more accessible to enzyme molecules due to lower degree of crystallinity and increased surface area 34 . Additionally, higher conversion can be attributed to improved mass and heat transfer in samples containing smaller particles 33 .
With previously de ned optimal conditions referring to enzyme concentration and particle size, partial lignin degradation with white-rot fungi was investigated from the aspect of hydrolysis e ciency and overall convertibility. The effect of fungal pretreatment on enzymatic sacchari cation was evident according to obtained results. Studied fungal species exhibited various effectiveness in increasing sacchari cation which was mostly expected due to different physiological response of each fungus, in uenced by different physical and chemical factors during the cultivation period 35 . Thus, fungal culturing cannot always increase hydrolysis yield, and the positive effect depends on fungal species, the range of cultivation conditions and the substrate properties 36 . Besides our study, the positive effect of pretreatment with white-rot fungi has also been observed by other authors 17,18,36 . The increase of hydrolysis yields is directly correlated to a better accessibility to polysaccharides after fungal deligni cation. Pinto et al. 18 also reported that their strains of G. resinaceum and T. versicolor were the most effective among all tested species increasing sacchari cation about 2-fold. Results of biologically pretreated wheat straw clearly indicate that the higher rate of lignin in lignocellulosic feedstock decrees the e ciency of enzymatic hydrolysis which is in accordance with previous ndings 37 . However, results of sugar analysis lead us to assume that this can be compensated with higher cellulose amount in feedstock after biological pretreatment where hemicelluloses was predominantly degraded which was observed in the case of P. ostreatus HAI592.
Nowadays, various chemical and physical methods are available to fractionate, solubilise, hydrolyze and separate bre components of lignocelluloses. Those procedures are widely used in many re nery concepts for cellulosic ethanol production 28 . It is well known that different chemical methods used to process lignocelluloses promote the formation of various inhibitory compounds which are the main obstacle in subsequent fermentation with yeasts 38 . Thus, it is crucial to nd the optimal pretreatment process not only to obtain the highest possible amount of fermentable sugars but to avoid formation of inhibitors as much as possible. Therefore, the examination of effectiveness of biological pretreatment was conducted by comparison with chemical pretreatment as predominantly used for lignocelluloses feedstock processing. Earlier researches pointed out that chemical pretreatments of lignocelluloses are more effective comparing with biological pretreatment. This is mainly based on the partial loss of holocellulose which is consumed by fungi during their growth and is used as carbon source 39,40 . A selection of fungal species with lower cellulase activity and selective degradation of lignin can be approach to improve biological pretreatment with white-rot fungi 17 . As the potential way to increase the e ciency of biological pretreatment some authors suggest the use of genetically modi ed organisms or a combination of biological with chemical pretreatment 40 .
Even though carbohydrates loss is expected during pretreatment with white rot fungi the relative carbohydrate enrichment of the biomass, i.e. by comparing the ratio carbohydrates:lignin in pretreated wheat straw samples was relatively lower. Previous research also showed that the decrease in wheat straw lignin as the result of fungal culturing produced a low enhancement of carbohydrate component by increasing glucan ratio in the biomass 28 . Pretreatment of lignocellulosic biomass with white rot fungi such as I. lacteus and Ceriporiopsis subvermispora have been previously studied 41,42,43 . As in the occasion of G. resinaceum of our study, an e cient enzymatic conversion of pretreated lignocellulosic material was achieved, but the increased conversion came at the expense of considerable loss of holocellulose during pretreatment. Furthermore, a positive effect of deligni cation has not been seen always. These inequalities occurred either due to weak deligni cation rate or as a result of substantial total solids loss 43 . Although higher coe cient of conversion was not achieved in biologically pretreated samples the approach still remains justi ed, as results of this study indicate that the crucial moment can be the period of fungal culturing in which the effect of an upconcentration of carbohydrates should be stopped after partial deligni cation, to avoid substantial sugar consumption by fungi.
Comparing with biologically pretreated wheat straw after chemical pretreatment with NaOH the higher amount of fermentable sugars was released. Considerably higher amounts of glucose and cellobiose and negligible presence of pentoses in hydrolysate of alkali pretreated wheat straw was the result of selective removal of lignin from biomass 44 . Process was followed by removal of acetyl and uronic acid groups on hemicelluloses and hydrolysis of ester linkages between xylan and hemicelluloses residues. All together this largely improved cellulose digestibility exhibiting minor cellulose loss comparing with fungal pretreated samples 44 . Previous studies also showed that chemical pretreatment is more e cient than biological pretreatment due to the partial loss of carbohydrates which are the carbon source for fungal growth 40 . Another impact of enzymatic hydrolysis of pretreated wheat straw was the accumulation of cellobiose. This was due to limitation in beta-glucosidase activity in enzymatic solution applied for sacchari cation. However, the presence of cellobiose in hydrolysate cannot be considered disadvantageous because this disaccharide can easily be converted to glucose either enzymatic or by using engineered yeast strains 37 .

Conclusions
Results of this study show the considerable potential of white-rot fungi in preparing lignocellulosic biomass for e cient production of fermentable sugars and further industrial scale processing to ethanol. Different fungal species have a various ability to degrade lignin in wheat straw and the rate of sugar consumption should be considered. A key mechanism of carbohydrate component convertibility enhancement is selective lignin removal in the biomass. Among tested species only Ganoderma resinaceum deserves to be considered as a promising lignin degrader after additional optimization of process parameters. Long time consumption, the low sugar yields and unpredictable fungal response to culturing conditions are the main challenge of the fungal pretreatment process in the conversion of lignocellulosic biomass.

Chemicals
All reagents and chemicals were of analytical grade and were purchased either from Sigma-Aldrich (St. Louis, MO, USA) or from Merck Millipore (Darmstadt, Germany), unless otherwise stated.

Organisms
Fungal isolates used in the study were obtained from the culture collection of the Institute of Botany, Faculty of Biology, University of Belgrade (BEOFB) or from the culture collection of the Institute of Evolution, University of Haifa, Israel (HAI): Cyclocybe cylindracea BEOFB1210 (MW176087, 100%), Ganoderma resinaceum BEOFB440 (MW176072, 100%), Irpex lacteus BEOFB1910 (MH671316, 99.83%), Pleurotus ostreatus HAI592 (MW176092, 91.99%), Trametes versicolor BEOFB321 (MW176038, 99.64%). Identi cation of fungal species was con rmed by ITS gene sequencing and the PCR ampli cation of this region was conducted as described by Savković et al. 45 . Gene bank accession numbers and homology percentages are given in brackets for each fungal isolate.

Preparation of lignocellulosic material
Wheat straw (Triticum aestivum L.) was washed with warm distilled water (T = 50°C), dried in a heating oven (BINDER ED53) at 65°C, to constant weight, grounded in the laboratory mixer (Waring 8010 S), and sieved to obtain pieces of 0.5-2.0 cm. So prepared wheat straw was subject to biological and chemical pretreatment.
Biological pretreatment

Cultivation conditions and extraction of ligninolytic enzymes
Solid-state culturing was carried out at 25°C in 250 mL asks containing 6.0 g of wheat straw and 30.0 mL of the modi ed synthetic medium (without glucose). Inoculation was done with 9.0 mL of inoculum per ask). Samples were harvested after 5, 10, 15 and 19 days of culturing and further used for ligninolytic enzyme extraction, according to previously established conditions 46 . The residues were used for determination of hemicelluloses, cellulose and lignin contents and for enzymatic sacchari cation.
Ligninolytic enzymes were extracted by stirring samples with 150.0 mL of cold dH 2 O on a magnetic stirrer at 4°C for 10 min. The extracts were ltrated using a laboratory sieve, the liquid fractions were centrifuged (4°C, 3000 rpm, 10 min) and the supernatants obtained were used for determining laccase and Mn-oxidizing peroxidases activities spectrophotometrically (CECIL CE2501 (BioQuest)). Solid phase remained after the extraction was dried in a drying oven at 65°C to constant weight.
Enzyme activity assay and determination of total protein content Enzymatic activity of 1 U is de ned as the amount of enzyme that transforms 1.0 µmol of substrate per min.

Determination of polymer content in a straw
Hemicellulose content determination Content of hemicellulose was determined by Van Soest bre analysis method by removing soluble sugars, proteins, pectin, lipids, and vitamins from the sample using neutral detergent and Na 2 SO 3 47,48 . Thus neutral detergent bres (NDF) were treated acid detergent solution to obtain acid detergent bres (ADF). Hemicellulose content is than expressed as NDF-ADF.
Cellulose and lignin content determination ADF were used for determining cellulose and lignin content using the Klason or 72% H 2 SO 4 method 49 . ADF were treated with 72% H 2 SO 4 and the Klason lignin residues were collected. Lignin content (LC) is expressed as a percentage of the original sample. Cellulose content is presented as the difference in weights of the samples treated with ADF and LC.

Enzymatic hydrolysis
Pretreated wheat straw was ground in the laboratory mixer and sieved to separate large fraction (L, 5-8 mm), medium fraction (M, 2-5 mm) and small fraction (S, 0.2-0.5 mm). Enzymatic sacchari cations were carried out in 250-mL asks containing 2 g of pretreated straw in 60 mL of acetate buffer (pH 4.8) with addition of streptomycin sulphate (500 mg L − 1 ) to prevent bacterial growth, and commercial cellulase (Cellulase from Trichoderma reesei, C2730-50ML, 700 U g − 1 ; Sigma-Aldrich) in loading of 0.5, 1.0 or 2.0 µL per mL of mixture prepared for hydrolysis. Incubation was carried out in a shaking incubator with horizontal rotary plate (Stuart SI500) (180 rpm, 50°C for 72 h). Immediately after hydrolysis samples were immersed in boiling water for 5 min to inactivate the enzymes and further used for determination of reducing sugar content 50 .
Autoclaved untreated wheat straw was used as a negative control while chemically pretreated wheat straw was used as a positive control. Chemical pretreatment of wheat straw was performed with 1 % (w/v) NaOH at 121°C for 90 min. The samples were than washed with dH 2 O till pH 6.8-7.5 was achieved and the solid phase was further dried to constant weight and used for enzymatic hydrolysis 51 .

Reducing sugars content determination
The content of reducing sugars was measured colorimatrically using 1,4-dinitrosalycilic acid, according to the method of Miller 52  The yield of enzymatic hydrolysis was calculated as follows: Hydrolysis yield (%) = Reducing sugars in hydrolysate (g) × 0.9 × 100 / Cellulose in sample (g) The correlation factor of 0.9 corresponding to hexoses was used in the calculations to compensate for the addition of a water molecule during hydrolysis of each glycosidic bound 53 .
To determine the impact of fungal cultivation on an overall wheat straw conversion to glucose, the coe cient of conversion (K c ) from the maximal theoretical value (1) of an initial sample was calculated as follows: K c = m f × 0.9 / Cellulose in an initial sample (g) Factor m f corresponds to the amount of reducing sugars (g) obtained after sacchari cation of entire pretreated sample.
Before analysis system was preconditioned at 15% A, 85% C; for 15 minutes. Each sample (25 µL) was injected with an ICS AS-DV 50 auto sampler (Dionex, Sunnyvale, CA, USA). The pulsed amperometric detection was performed on gold as working and Ag/AgCl as reference electrode.

Statistical analyses
The assays were carried out in three replicates and results are expressed as mean ± standard error. Normal distribution (Shapiro-Wilks test) and homogeneity of variance (Levene's test) were evaluated prior to further analyses where certain groups based on different quantitative parameters were compared to see are there signi cant differences between them. If data followed normal distribution and homogeneity of variance was met, one-way analysis of variance (ANOVA) was performed to test any signi cant differences among means, since the number of groups was always higher than two. If data followed a normal distribution, but the assumption of equal variances was violated, Welch Test or Welch ANOVA was considered for the analyses. Together with one-way ANOVA, the pairwise comparison option that runs Tukey's HSD (Honestly Signi cant Difference) post-hoc test was applied in some cases. Statistical signi cance was declared at P < 0.01. All statistical analyses were performed using Microsoft Excel, the statistical package XLSTAT 54    Time course of reducing sugars release from wheat straw during enzymatic sacchari cation: a. Effect of enzyme loading; b. Effect of particle size.

Figure 5
Time course of reducing sugars release from NaOH-pretreated wheat straw during enzymatic sacchari cation.