Introduction
The conversion of lignocellulosic biomass, derived from agricultural residues, into value-added compounds is a sustainable strategy for the development of many industries (Bharathiraja et al. 2017, Ravindran and Jaiswal 2016 and Ravindran et al. 2018). During the degradation of these materials by microorganisms, various compounds are produced, such as proteins, enzymes, organic acids, secondary metabolites and oligosaccharides (Knob et al. 2014).
Hydrolytic enzymes, obtained from fungi, bacteria and yeasts, stand out for their application in several industrial processes, such as extraction and clarification of fruit juices, the extraction of plant oils and pigments, pulp and paper bleaching and animal feed, among others. Among the microorganisms used for these purposes, Bacillus genus bacteria are considered promising for the development of agricultural industry, due to the variety of extracellular enzymes they produce and their stability at high temperatures (Chakdar et al. 2016).
The objective of this study was to evaluate wheat bran (Triticum aestivum L.), sugarcane bagasse (Saccharum officinarum L.), rice husk (Oryza sativa L.) and corn stubble (Zea mays L.), as substrates for the production of endocellulase, mannanase and endoxylanase enzymes by Bacillus subtilis E44 during solid state fermentation (SSF).
Materials and Methods
Chemical composition of agroindustrial waste. Wheat bran, sugarcane bagasse, rice husk and corn stubble were used as substrates to induce the activity of Bacillus subtilis E44 enzymes, from different municipalities in Matanzas province (table 1).
Table 1.
Origin of agroindustrial residues from Matanzas province used in the study
Wheat was imported from Germany in June 2015 and it is semi-hard. Bran was obtained in the successive stages of the milling and sifting process of wheat to produce meal. Collection was random, out of 83 bags containing the by-product.
Rice husk was collected at the end of the day, after the end of production, in a handcrafted mill. Sampling was also random, out of the seven bags containing the by-product.
Bran, like rice husk, was transferred to the laboratory in polyethylene bags weighing 1.98 and 1.53 kg, respectively.
Sugarcane bagasse was collected in the CAI bagasse facilities. Corn stubble was collected in the field, two days after harvest. The materials were collected at five points and all strata were taken from the surface to the soil. Samples were formed by quartering method and transferred to the laboratory in polyethylene bags, with 1.2 and 1.67 kg each. A batch of each substrate was analyzed, consisting of samples taken in each case.
Substrates were dried in an oven at 60 °C for 72 h. They were crushed in a hammer mill, Fritsch type, GMbH model with a maximum particle size of 2 mm. Cellulose, hemicellulose and lignin contents was determined by the sequential technique described by Van Soest et al. (1991), with modifications proposed by ANKOM (1998). For the analysis, polyester bags (ANKOM Corp # 57) were used, with a pore size of 30 µm and dimensions of 4.5 x 5.5 cm. Nitrogen was quantified using the Kjeldahl method (AOAC 2012), using a Kjeltec System 1002 (Tecator) distillation equipment. Crude protein was obtained by multiplying N content of the sample, according to the 6.25 conversion factor.
Microbial culture. Bacillus subtilis E44 strain, from the Microbiology Laboratory of the Faculty of Agricultural Sciences of the University of Matanzas, was used, preserved at -30ºC in glycerol.
Minimum salt culture medium (SM). The mineral solution was composed of NaCl (0.1%), KH2PO4 (0.3%), K2HPO4 (0.6%), MgSO4 (0.12%), peptone (0.5%) and yeast extract (0.3%). pH was adjusted to 7.5 with KOH (1 mol L-1) and sterilized at 121 °C for 15 min.
Microbial suspension preparation. A cell suspension was prepared from a Bacillus subtilis E44 culture of 16 h, in Erlenmeyer with 50 mL of nutrient broth. It was incubated at 28 °C in a shaker at 110 rpm, up to obtaining an optical density (OD 600nm = 0.8) equivalent to a concentration of 1x108 cfu mL-1.
Solid state fermentation. The experiment was carried out in 250 mL flasks, previously sterilized at 121 °C for 15 min and dried in the oven at 60 °C for 48 h. The flasks contained 1g of substrates (wheat bran, sugarcane bagasse, rice husk and corn stubble). In each Erlenmeyer, substrates were inoculated with 10% (w/v) of the microbial suspension and dry matter medium was added until obtaining 80% of humidity in the SSF. Flasks were placed in the incubator at 37 °C for 24 h. Each substrate was evaluated by triplicate.
Extraction of enzymatic crude. For the extraction buffer, 0.02 mol L-1 sodium phosphate and pH 7, 1:10 (w/v) ratio, were added to each flask. Each one was placed in a shaker at 110 rpm for 30 min. Subsequently, the substrate was filtered by gauze and centrifuged at 10,000 rpm at 4 °C for 15 min. Extracts were stored at -20 until evaluation.
Enzymatic activities. The activities of endoxylanase, endocellulase and mannanase enzymes were determined by triplicate in the obtained extracts. Reaction mixture was composed of 0.4 mL of substrate (1% beech xylan, 1% carboxymethyl cellulose and 0.5% galactomannan, respectively) and 0.1 mL of the enzyme extract. It was incubated at 50 °C for 10 min. The reaction was stopped by adding 0.5 mL of dinitrosalicylic acid. Samples were placed in a water bath at 100 °C for 10 min. An amount of 1.2 mL of distilled water was added and absorbance was measured at 540 nm. The equivalents of glucose, mannose and xylose were calculated from the corresponding standard curves. One unit of enzymatic activity was defined as the amount of enzyme required to produce 1µmol.mL-1 of glucose, mannose and xylose per minute under the test conditions. Enzyme production was determined from the enzymatic activities detected after the SSF with each of the substrates.
Statistical analysis. For the analysis of results of the chemical composition of substrates, descriptive statistics was used to determine mean, standard deviation (SD) and variation coefficient (VC). An analysis of variance was applied to data obtained from the production of enzymes, according to a completely randomized design, with a 3x4 factorial arrangement. Duncan test (1955) was implemented to establish differences among means. Data processing was carried out using the statistical package Insfostat (Di Rienzo et al. 2012).
Results and Discussion
The chemical composition of agroindustrial residues showed differences in the percentage of hemicellulose, cellulose and lignin (table 2). This result is due to the fact that used substrates come from different plant species. In addition to the genotype factor, cell wall compounds distribution may vary, due to the agro-cultural processes associated with sowing, harvesting, post-harvest events, crop age and physiological conditions of plants (Mussatto et al. 2012). According to Knob et al. (2014), in agroindustrial residues, cellulose is the dominant fraction in plant cell wall (35-50%), followed by hemicellulose (20-35%) and lignin (10-25%). The percentage of cellulose that wheat bran shows, does not correspond to statements of the latter authors. However, it is close to the range referred to by Babu et al. (2018), where 55% of fiber in this substrate is composed of arabinoxylanes, while cellulose occupies between 9 and 12% of its dry weight.
Table 2.
Chemical composition (%) of agroindustrial residues, used as substrates in solid state fermentation of Bacillus subtilis E44
Chemical composition of bran varies among different harvests and is closely related variety, cultivation conditions and methods used for wheat grain separation (Babu et al. 2018). Hatfield and Fukushima (2005) consider that polymerization degree, high content of proteins, minerals and other organic compounds present in the materials can affect the accuracy of chemical analyzes.
In this study, enzyme production is expressed in terms of their enzymatic activity, as international literature refers. Results of the production of endocellulases, mannanases and endoxylanases by Bacillus subtilis E44 during the SSF showed that there is interaction between substrates and enzymatic activity (P <0.0001) (table 3). The nature of carbon source in the culture medium influences the production of enzymes. These results indicate that all the evaluated residues induce endoxylanase expression by this strain. The highest values were obtained with wheat bran (25.08 IU.mL-1), followed by sugarcane bagasse (9.32 IU.mL-1). Rice husk and corn stubble did not differ from each other.
Table 3.
Production of endocellulases, mannanases and endoxylanases by Bacillus subtilis E44 during the SSF, using different agroindustrial residues as carbon source
Residues | EA (IU.mL-1) | SE ± Sig. | ||
---|---|---|---|---|
Endocellulases | Mannanases | Endoxylanases | ||
Sugarcane bagasse | 2.58cd | 2.52bcd | 9.32f | ±0.14 P<0.0001 |
Wheat bran | 2.75d | 2.15bc | 25.08g | |
Rice husk | 0.07a | 2.07b | 6.92e | |
Corn stubble | 2.47bcd | 2.19bc | 6.95e |
Different letters indicate significant differences according to Duncan (1955) test (P<0.05)
Gowdhaman et al. (2014) reported wheat bran as the best endoxylanases inducer in a strain of Bacillus, compared with sugarcane bagasse, corn and rice husk. Similar results obtained Kaushik et al. (2014), Ho and Heng (2005) and Zhang and Sang (2015), in strains of Aspergillus lentulus, Penicillium chrysogenum QML-2 and Bacillus subtilis, respectively.
Wheat bran usefulness as a carbon source is attributed to its chemical composition, since 70% of non-starchy polysaccharides are arabinoxylanes (Maes and Delcour 2002). In addition, this residue has minerals, vitamins and other bioactive compounds that favor microorganism growth (Babu et al. 2018).
Physical characteristics of this by-product benefit its use as a substrate, as it easily degrades, since its particles have a large surface area and good humidity retention (Stevenson et al. 2015), which favors microbial attack.
The sugarcane bagasse is a good substrate for microbial development, because, in its composition, carbohydrates represent approximately 70%. Out of these, xylans are, after glucans, the most important (Batalha et al. 2015). For this reason, several authors use this residue as an inducer for the production of endoxylanases (Yang et al. 2015).
In the rice husk, from handcrafted mills, xylans occupy approximately 14% (López 2013), which also favors the production of endoxylanases. In this residue, arabinoxylanes are the major substitutes of xylan and are easily solubilized (Vegas et al. 2008).
Stubble and other corn residues are reported to be good endoxylanase inducers, since xylan content is approximately 40% (Knob et al. 2014). These residues are used for the production of these enzymes by different microbial genera, including Bacillus (Ling 2014). The difference among endoxylanase activity values, obtained during fermentation processes, can be related to substrate accessibility, hydrolysis rate, the amount of xyl-oligosaccharides and xylose, released during microbial metabolism (El-Sharnouby et al. 2012), as well as with xylan complex structure, which varies depending on the plant species (Wang et al. 2014).
Regarding endocellulases production, results show that there are no differences between the use of bagasse, bran and corn stubble, as carbon sources. These residues are reported in the international literature as good inducers of these enzymes. However, Sadhu et al. (2013), Kazemi et al. (2014) and Gaur and Tiwari (2015) agreed that sugarcane bagasse was the best source of carbon for endocellulase production during SSF by Acinetobacter sp. KKU44, Bacillus vallismortis and Bacillus sp, respectively.
With the use of rice husk, low enzyme activity was detected. However, Annamalai et al. (2013) reported the usefulness of rice husk and stubble, with good results in the production of endocellulases. Dhillon et al. (2011) found high endocellulolytic activity in Aspergillus niger, with wheat bran and rice husk, separate and combined.
Rice husk is a raw material of great interest for the production of cellulosic ethanol in Cuba (Martín 2006). López (2013) compared samples of this residue, from handcrafted and industrial mills, and found that glucose concentration of rice husk, obtained from a handcrafted mill, is higher than that of the industrial one (33.5 and 2.1 g.L-1, respectively). This author attributed this fact to the presence of easily hydrolysable glucans, and it reported the presence of hydroxymethyl furfural (0.3g. L-1) and formic acid (0.2g. L-1) in this residue.
Low cellulosic activity, found with the use of this by-product, could be associated with endocellulases inhibition, due to different factors: glucose is the final product of the action of the cellulases complex and it is, in turn, the repressor of its synthesis (Sukumaran et al. 2005). The presence of phenolic compounds at low concentrations, such as hydroxymethyl furfural, inhibits the enzyme complex and causes its precipitation and inhibition (Kim et al. 2011). The removal of these compounds, by means of treatments prior to rice husk from handcrafted mills, could increase the production of this enzyme.
Several research are carried out in order to improve the technologies for the production of microbial cellulases, due to their complex metabolic regulation mechanisms. Among the strategies developed for these purposes, the use of SSF and the use of molecular methods, such as mutagenesis, metabolic engineering and cellulase gene expression from different microbial domains, are highlighted, with the purpose of improving synthesis and catalytic properties (Kuhad et al. 2016).
The production of mannanases was favored during the SSF in the four residues, without differences among them. These results are consistent with those obtained by other researchers, who used wheat bran (Singh et al. 2010) and sugarcane bagasse (Chauhan and Gupta 2016) for the synthesis of these enzymes. Ravindran et al. (2018) referred to the ability of several species of Bacillus genus to produce enzymes that hydrolyze mannane. These are generally induced in the presence of galactomannane-rich substrates (Yamabhai et al. 2016).
Other agroindustrial and lignocellulosic wastes were evaluated in various researches. Yin et al. (2013) used a mixture of apple peel and cotton seeds to produce mannanase by Aspergillus niger strain SN-09 in SSF. Pangsri and Pangsri (2017) reported enzyme activity values of 0.80; 0.68 and 0.15 U.mL-1 with the use of tea and ground coffee residues. Generally, sugarcane bagasse, soybean residues, galactomannane, banana, mango and potato peel are used as excellent inducers for the production of these enzymes Onilude et al. 2012 and Almeida et al. 2015.
Mannanases are extracellular and inducible enzymes, and are considered second in importance during hemicellulose hydrolysis (Dhawan and Kaur 2007). They catalyze at random the hydrolysis of β-D-1.4 mannopyranoside bonds of β-1.4 mannane. Mannanases production is reduced to Gram positive bacteria, mainly some Bacillus species (Mabrouk and Ahwany 2008 and Meenakshi et al. 2010).
The results of this study show greater production of endoxylanases by B. subtilis E44, with respect to endocellulases and mannanases. Several authors highlight the predominance of xylanolytic microorganisms in different genera (Banka et al. 2014 and Gupta et al. 2015). However, the synthesis of these enzymes could be increased with the application of fermentation optimization methods. Many authors report enzyme productions, with a notable increase in activities, which vary between 10 and 80% with respect to the environment without optimization (Reis et al. 2015 and Zhang and Sang, 2015).
The possibility of having methodologies to produce these biomolecules, based on easily available lignocellulosic materials, favors the reduction of their production costs. The possibility of reusing agro-industrial residues reduces the polluting effects of the environment, associated with its accumulation.
Enzymes, which catalyze the hydrolysis of cellulose and hemicellulose, are used as zootechnical additives in animal production with favorable results. Its benefits include the decrease of anti-nutritional effects of non-starchy polysaccharides in diets and the increase of total digestibility. These enzymes also complement the activity of endogenous enzymes produced by the animal and lead to improvements in health, by reducing infections caused by pathogens such as Salmonella sp. and Clostridium sp. (Bedford 2018).
Conclusions
Results of this research indicate the potentials of wheat bran, sugarcane bagasse and corn stubble for the production of endocellulases, endoxylanases and mannanases. The production of endoxylanases and mannanases was induced with rice husk, but not that of endocellulases.
Bacillus subtilis E44 bacteria showed potentialities to produce greater amounts of endoxylanases. The enzyme extract obtained from this strain could be used as a zootechnical additive to improve animal feed quality.