Agave Leaves as a Substrate for the Production of Cellulases by Penicillium sp. and the Obtainment of Reducing Sugars

Lignocellulosic biomass can be used to obtain fermentable sugars by enzymatic hydrolysis, and also it serves as a carbon source to produce cellulases by solid-state fermentation. In this study, we propose the use of leaves of Agave salmiana as a carbon source to produce cellulases by the fungus Penicillium sp., isolated from the same plant. $e crude enzymatic extract was used to obtain sugars from the hydrolysis of the parenchymal cells of the leaves. $e enzymes produced were characterized (endoglucanase 14.4U/g; exoglucanase 3.5U/g; β-glucosidase 4.14U/g). $e enzymes showed activities at elevated temperatures: 50°C for endoglucanase and exoglucanase and 70°C for β-glucosidase. Furthermore, the crude enzymatic extract obtained was able to hydrolyze the parenchyma in 51.6% in 48 h.$e evidence presented in this paper shows the potential of the agave leaves as a source of carbon in the production of enzymes by fermentation with the consequent production of reducing sugars. In addition, the enzymes produced by Penicillium sp. could be used in the production of bioethanol, since they work at high temperatures.


Introduction
With an annual production from land plants of more than 100 billion tons, lignocellulosic biomass has considerable potential to produce biofuels and high value-added byproducts [1]. To obtain these by-products, it is necessary to degrade lignocellulosic biomass from cellulose to glucose. Cellulases, an enzymatic system generally composed of endoglucanase, exoglucanase, and β-glucosidase, are the enzymes responsible for hydrolyzing the cellulose chain to glucose monomers [2]. Cellulases are the third most produced industrial enzymes, representing about 20% of the total market of enzymes in the world [3]. Nevertheless, the high price of cellulases and a large amount of these enzymes needed compromise the industrial degradations of cellulose cost. Strategies have been implemented to improve the efficiency of known cellulases and to search for new cellulase-producing microorganisms. Alternatively, cellulases can be produced by solid-state fermentation (SSF) [3]. Several microorganisms capable of producing cellulases have been reported using lignocellulosic residues as a carbon source [4][5][6].
Plants from the genus Agave represent a potential feedstock for enzymatic production of biofuels and high value-added by-products, with advantageous characteristics over other crops, such as low water requirement, high productivity in arid or semi-arid lands, adaptability to high temperatures, and resistance to drought [7]. ere are more than 200 species of the genus Agave, of which the majority are native to México [8].
Recently, Agave species have been found to be a good source of polysaccharides (i.e. cellulose and fructans), used to obtain different products such as nanocellulose and fermentable sugars and as a substrate for enzymes production [9][10][11][12][13]. During the industrial production of fermented beverages from A. salmiana, leaves are discarded. ese leaves contain a low amount of lignin (approximately 9-13 wt%) that provides low recalcitrance [14,15]. ese leaves are covered with a waxy layer of cutin and a thick layer of chlorophyll, that protects the plant (Figure 1(a)) from adverse environmental conditions. Inside, the cellulose fibers serve as support for the leaves, while the parenchyma cells are actively involved in photosynthesis, secretion, and food storage (Figure 1(b)). e cell wall in the parenchyma of the Agave leaves is thinner as compared with other plants [12,15].
Previously, we isolated cellulose layers in the form of nanoplatelets with a thickness of less than 90 nm from the parenchymal cell wall of A. salmiana leaves [16]. We have also reported that these cellulose nanoplatelets are easier to hydrolyze than larger cellulose, due to their low lignin content, poor crystallinity, and greater surface area [13]. Also, several strains of enzyme-producer fungi have been isolated from Agave species [17,18]. For instance, in earlier studies, we evaluated the enzymatic activity of different fungal strains previously isolated from A. salmiana leaves, with the highest enzymatic activity resulting from a fungus preliminarily classified as belonging to the genus Penicillium (unpublished data). For the above reasons, the aim of the present study was therefore to produce cellulases by solidstate fermentation using A. salmiana leaves as the carbon source. In addition, the enzymes obtained by this method were characterized and used to obtain fermentable sugars from the parenchymal cells of the same plant species.

Feedstock and Reagents.
All reagents were of analytical grade. Salts and commercial carbohydrates were purchased from Sigma-Aldrich and culture media from BD Bioxon. e leaves were pulled away from the base of plant grown in soil and washed and decorticated to eliminate the epidermis and cuticle. e remaining material was cut into small pieces and frozen at −20°C until use. is biomass obtained from inside the Agave leaf was used to produce enzymes, nanocellulose, and sugars in the following steps.

Inoculum Preparation.
Previously, Penicillium sp. was isolated from a leaf of decaying A. salmiana. Potato dextrose agar (PDA) was used for the isolation and storage of the strain. Unless otherwise mentioned, the inoculum was prepared as follows: fungus was inoculated in tubes filled with 5 mL PDA and incubated by 6 days at 28°C. After, spores were recovered with saline 0.85% w/v. e inoculum was adjusted to a concentration of 1 × 10 6 spores/mL using a Neubauer chamber for counting the spores.

Agave salmiana Leaves as a Carbon
Source. Agave biomass was lyophilized to facilitate the mechanical separation of the fibers from the parenchyma. Fibers were washed with hot water to remove residual sugars, and the parenchyma was used unmodified. A subsample of the parenchyma was used to obtain nanocellulose by a method previously described [12]. e isolated Penicillium sp. strain was cultivated in 125 mL Erlenmeyer flasks containing 500 mg of fiber or parenchyma, 1 mg (NH 4 ) 2 SO 4 , and 1 mL acetate buffer pH 5. Flasks were sterilized by autoclaving and subsequently inoculated with 1 mL of the inoculum prepared previously. Flasks were incubated at 28°C under static conditions for 7 days. After incubation, enzymes were recovered by adding 25 mL acetate buffer pH 5 and transferred at 50 mL tubes. Tubes were centrifugated at 4,500 rpm for 20 min. e supernatants were separated and used as crude cellulase extracts to hydrolyze 3 substrates: fiber, parenchyma, and nanocellulose. In 2 mL tubes, 100 μL crude extract, 900 μL acetate buffer pH 5, and 10 mg of each substrate were added. Tubes were incubated at 50°C, 150 rpm for 24 h. After hydrolysis, sugars were measured for the 3,5 DNS acid method [19].

Scanning Electron Microscopy (SEM).
e morphology was observed by SEM using a HITACHI UHR Cold-Emission FE-SEM SU8020 with an acceleration voltage of 1 kV and a working distance of 9 mm. A subsample of the lyophilized biomass (previously obtained) was suspended in ethanol. A drop of the suspension was deposited on a silicon wafer and then dried at 40°C for 2 h. en, the silicon wafer was glued with an adhesive carbon tape to an aluminum holder.

Cellulase Production by SSF.
Agave biomass was defrosted and washed with hot water to remove inorganic compounds, colorants, and sugars (the reducing sugars were measured in the wash water until being completely removed from the biomass). e resulting bagasse was oven-dried for 2 days at 70°C and ground (Krups mill, model GX410011). Fermentations were carried out in 125 mL Erlenmeyer flasks, each having 3 g of the ground bagasse and 12 mL of a solution (g/L: 2.0 KH 2 PO 4 , 1.4 (NH 4 ) 2 SO 4 , 0.3 MgSO 4 , 0.3 CaCl 2 , 1.0 peptone, 1.0 yeast extract, 1 mL Triton X−100, and pH 5.5). e flasks were sterilized by autoclaving and subsequently inoculated with 3 mL of inoculum prepared previously at a concentration of 1 × 10 6 spores/mL. Flasks were incubated at 28°C for 0, 2, 4, 6, 8, 10, 12, and 14 days without agitation. Enzymes were recovered by adding 30 mL acetate buffer at pH 5 by the previously described method.
2.6. Enzymatic Activity Assays. Exoglucanase, endoglucanase, and β-glucosidase activities were determined by using microcrystalline cellulose type 20 (MCC20), carboxymethyl cellulose (CMC), and p-nitrophenyl-β-D-glucopyranoside (pNPG) as substrates, respectively. Assays to determinate exoglucanase and endoglucanase activities were carried out in 2 mL tubes containing 900 μL acetate buffer pH 5, 100 μL enzyme extract, and 10 mg substrate (MCC20 or CMC). Tubes were incubated at 50°C, 150 rpm for 2 h and 20 min for exoglucanase and endoglucanase, respectively. Sugars were measured for the 3,5 DNS acid method. β-glucosidase activity was determined by incubating 50 μL enzyme extract and 200 μL 0.2 mM pNPG (in acetate buffer pH 5) at 50°C for 10 min. After incubation, 1 mL of 0.2 M Na 2 CO 3 was added to stop the reaction, and the absorbance was measured at 405 nm. Once the optimal day of production of each enzyme was determined, the influence of pH and temperature on enzymatic activity was evaluated. Assays were performed at pH ranging from 3 to 10 and at temperatures ranging from 30 to 90°C. One unit of activity is defined as the amount of crude enzyme extract that releases 1 μmol of glucose or p-nitrophenol (pNP) per min per gram of biomass under assay conditions.

Obtainment of Sugars.
Ground bagasse was obtained from the frozen Agave biomass as previously described. is bagasse was sieved to separate the parenchyma from the fibers. e fibers were discarded, and the parenchyma was subjected to enzymatic hydrolysis. e assays were carried out in 2 mL Eppendorf tubes containing 1000 μL of acetate buffer (pH 5), 500 μL of crude cellulase extract, and 10 mg of parenchyma. e tubes were incubated at 50°C and 150 rpm for 0, 12, 24, and 48 h. After each time, sugars were measured using the 3,5 DNS acid method. e assays were performed in triplicate.

Agave salmiana Leaves as a Carbon
Source. Due to the morphological dissimilarities between the cellulose from the fibers and the parenchyma (Figure 1(b)), the ability of Penicillium sp. to use each type of cellulose as a carbon source was evaluated. e fungus grew faster using the parenchyma as the carbon source ( Figure S1); nevertheless, when evaluating enzymatic activity, the enzymatic extract alone was able to hydrolyze the parenchyma (Table 1). It is possible that the tested Penicillium strain produced inulinases to hydrolyze the fructans from the parenchyma instead of the cellulose; for this reason, the hydrolysis of the fibers or nanocellulose was absent. Using fibers as the carbon source, slowed Penicillium growth ( Figure S1), however, the enzymatic extract was able to hydrolyze both substrates (Table 1). It has to be noted that fibers were thoroughly washed before Penicillium was inoculated, avoiding the presence of any carbon source other than cellulose; therefore, it seems unlikely that the fungus produced alternative hydrolytic enzymes. e nanocellulose and parenchyma are known to be easier to hydrolyze as compared to fibers due to their lower crystallinity and small size that increases surface area and enzymes contact with the substrate, thus causing greater hydrolysis [13]. e slight differences in enzymatic activities between using either nanocellulose or parenchyma may be related by the structural differences among both carbon sources. Pretreatment given to the nanocellulose are reported to eliminate its lignin cover and makes nanocellulose easier to hydrolyze [20].
In natural habitats, microorganisms regulate their enzyme production for the most efficient usage of the available carbon source [21]. Previously, it has been reported that the addition of glucose to the culture medium favors rapid growth of fungi but suppresses the cellulolytic activity [22]. Since a parenchymal cell contains simple sugars rapidly used by fungi before cellulose, fungi production of cellulases was limited (Table 1). On the other hand, fungi were able to produce cellulases to hydrolyze fibers and grow. It is noted the importance of minimizing sugar presence in the biomass before carrying out fermentation to induce cellulose production by fungi from the culture start. Agave leaves contain high amounts of soluble sugars [10,14] which are mostly removed by squeezing the leaves after a freeze-thaw process. ese sugars were recovered and preserved for future experiments as they can be used directly for fermentation.

Cellulase Production.
e production of endoglucanase, exoglucanase, and β-glucosidase was determined in the SSF of Penicillium sp. at 28°C. Figure 2 shows that the activity of all enzymes was detected after 2 days of fermentation. Exoglucanase activity increased from day 2 and remained constant until 10 days, decreasing slowly after that. e constant activity of the exoglucanase was probably motivated by the presence of Agave fibers in the medium. e exoglucanases are responsible for the hydrolysis of the crystalline portions of cellulose [2]. Agave fibers have a high crystallinity index similar to commercial cellulose; therefore, enzymes are required to be active throughout the fermentation to hydrolyze the fibers [16,20]. e activities of endoglucanase and β-glucosidase increased until 6 and 8 days, respectively. e β-glucosidase showed a rapid decline after 8 days, while the endoglucanase remained active until 12 days. e synergy between endo-and exoglucanase generates cellobiose, which can be hydrolyzed by β-glucosidases [23]. is disaccharide is a strong inhibitor of endoand exoglucanases; therefore, a constant activity of β-glucosidase is necessary to keep the concentration of cellobiose to a minimum so that there is no inhibition of the other enzymes. e rapid decrease in activity of glucosidase from day 10 caused a slow loss of activity of the other enzymes (Figure 2), so it was not needed to extend the fermentation time, since the maximum activity point for each enzyme was already reached.

Effect of pH and Temperature on Enzymatic Activity.
e optimal pH for each enzymatic activity was 4 for endoand exoglucanases and 6 for β-glucosidase. Variations in pH decreased these activities (Figure 3). Endo-and exoglucanases exhibited their highest activity at 50°C, while β-glucosidase up to 70°C. No enzyme showed activity above 80°C (Figure 4). e optimal pH values of each enzyme reported in this work are within the range of those already reported in other studies [24,25]. Industrial enzymatic hydrolysis carried out at <50°C shows slow hydrolysis rates, generating low amounts of sugars from lignocellulosic biomass; in addition, they are prone to microbial contamination. In such cases, high doses of enzymes are required. For this reason, enzymes that work at high temperatures have been suggested to be used to overcome these limitations [26]. Microorganisms producing thermophilic enzymes have already been reported [25][26][27]. Enzymes produced by the tested Penicillium fungus have shown to have activity at high temperatures, so they could be used in enzymatic hydrolysis technologies. e highest enzymatic activities obtained for endoglucanase, exoglucanase, and β-glucosidase were 14.4, 3.5, and 4.14 U/g of biomass, respectively, under the ideal conditions of pH and temperature. Table 2 compares the enzymatic activities of cellulases produced by different microorganisms under SSF using different carbon sources. e fermentation conditions, as well as the microorganism and the biomass source used, are the main factors that influence the enzymatic activity and the amount of cellulases produced under solid-state fermentation [3,24,28]; therefore, a realistic comparison of cellulase production by different other microorganisms and carbon sources as reported in the literature may not be possible due to the different fermentation conditions described by each research group [24]. Nevertheless, the enzymes produced by the isolated Penicillium fungus had a higher enzymatic activity compared with other Penicillium species (Table 2). e physical processes of cutting and grinding are useful for processing the lignocellulosic biomass as it manages to decrease its particle size, increasing the surface area [32]. Nevertheless, these physical pretreatments have no impact to eliminate other components such as waxes, lignin, and sugars nor do they improve the digestibility of cellulose considerably; this is why chemical pretreatments (alkaline or acid) have become necessary to improve the demineralization of cellulose [33]. However, it has been shown that these chemical pretreatments generate toxic compounds for yeasts during bioethanol production [34].

Enzymatic Hydrolysis of the Parenchyma.
To favor the biomass hydrolysis to obtain a greater quantity of sugars, a high amount of crude enzyme extract was used, with which the parenchyma was hydrolyzed in 51.6% in a time of 48 h ( Figure 5).
Although the parenchyma used to obtain sugars had no pretreatment, a high conversion of the biomass was achieved. e obtained percentage of hydrolysis was similar to that reported by other authors using Agave species [11,13,32]. As previously demonstrated, parenchyma was easier to hydrolyze than fibers due to its greater surface area.      erefore, in addition to selecting the most appropriate pretreatment method and enzyme complex, it would also be convenient to consider the most digestible portion of the biomass as another factor to consider during enzymatic hydrolysis.

Conclusions
In this study, Agave salmiana leaves were evaluated both to produce enzymes and for obtaining sugars. e portion of the leaves corresponding to the parenchyma cells proved to be a material with promising potential to be used in obtaining reducing sugars. Nevertheless, for enzyme production, it is necessary to optimize the fermentation conditions to improve the yield. In addition, the Penicillium fungus isolated by the research group has not been evaluated or characterized in depth; however, it was observed that its enzymes have activity at high temperatures, so it will be convenient in the future to characterize the enzymes more thoroughly to be used in industrial enzymatic hydrolysis.

Data Availability
e data used to support the findings of this study are included in the article.