2.1. Materials

Groundnut shell was obtained from Surakarta, Central Java, Indonesia. A commercial culture of bio fungicide consisting of a mixture of Trichoderma harzianum, Trichoderma viride, Trichoderma sp., and rice as the carrier was purchased from the Indotani Lestari store. Commercial tempeh yeast (RAPRIMA) that consists of Rhizopus oligosporus. and tape yeast (CAP TAMPIR) that consists of Aspergillus sp., were purchased from online shops in Bandung. Phosphate, hydrochloric acid, sulfuric acid, sodium hydroxide, dinitrosalicylic acid (DNS), Rochelle salt, citric acid, acetic acid, nitric acid, and sodium citrate dihydrate were obtained from School of Life Sciences and Technology, Institut Teknologi Bandung (ITB), West Java, Indonesia.

2.2. Determination of Lignocellulose Composition in Groundnut Shell

Approximately, 1 g (weight a) sample of groundnut shell was dried. After drying, the sample was added with 150 ml of distilled water/benzene alcohol and then refluxed at 100°C using a water bath for 1 h. The reflux products were filtered and rinsed with 300 ml of hot water. The residue obtained from the filtering process was dried using an oven until a constant weight was obtained (weight b). The dried residue was added with 150 ml of 1 N H₂SO₄ and refluxed again using a water bath at 100°C for 1 h. After that, the reflux products were filtered, and the residue was rinsed with distilled water until the pH was neutral. The residue was then dried using an oven until a constant weight was obtained (weight c). The dried residue was then added with 100 ml of 72% H₂SO₄ and then soaked at room temperature (27°C) for 4 h. Next, the residue was refiltered. The filtered residue was added with 150 ml of 1 N H₂SO₄ and then refluxed using a water bath at 100°C for 1 h. After that, the reflux products were filtered and then rinsed with distilled water until the pH reached a neutral point. The residue was then dried using an oven at 105°C until a constant weight was obtained (weight d). The dried residue was then ashed using a furnace and weighed (weight e) [10]. The composition of cellulose, hemicellulose, and lignin content of the groundnut shell can be calculated using the following equations:

$\mathit{Hemicellulose}\mathrm{(\%)}=\frac{(\mathit{b}-\mathit{c})}{\mathit{a}}\times 100\%(1)$

$\mathit{Cellulose}\mathrm{(\%)}=\frac{(\mathit{c}-\mathit{d})}{\mathit{a}}\times 100\%(2)$

$\mathit{Lignin}\mathrm{(\%)}=\frac{(\mathit{d}-\mathit{e})}{\mathit{a}}\times 100\%(3)$

2.3. Solid-State Fermentation

Approximately, 5 g of groundnut shell mixed with 5 ml of water and 0.5 g of culture [11] particularly Trichoderma sp., tape yeast; tempeh yeast, and mixture of Trichoderma sp. with both yeast with a ratio of 1:1 (g/g). The fermentation was carried out in beaker glasses at room temperature (27°C), and sampling was carried out every 24 h in duplicate.

2.4. Determination of Fungal Biomass

Approximately, 1 g (weight a) of the mixture was dried using an oven until a constant weight was obtained. After drying, the mixture was weighed (weight b). The dried solids were then soaked in 5 ml of acetic acid reagent and boiled for 30 min using a water bath. Acetic acid reagent consists of mixture of acetic acid and nitric acid with ratio of 10:1 (v/v) and then cooled and centrifuged at 5500 rpm for 15 min. The solid phase obtained was then rinsed with distilled water and dried at 70°C using an oven until a constant weight was obtained (weight c) [12]. The dry weight (DW) of the fungal biomass can be calculated by equation (4).

$\mathrm{Fungal}\mathrm{(\%)}\mathrm{biomass}=\frac{(\mathrm{b}-\mathrm{c})}{\mathrm{a}}(4)$

2.5. Determination of Cellulase Activity (FPase)

After the fermentation process, extraction was carried out to separate the extract of a crude enzyme from the substrate by mixing the fermentation medium with phosphate buffer (pH 5) with a substrate to buffer ratio of 1:10 (weight/volume). The extraction was carried out at room temperature (27°C) using a shaker set at a speed of 120 rpm for 1 h [13] followed by filtration through a cloth filter. The filtrate obtained was centrifuged at 550 rpm for 20 min, and the resulting supernatant was separated for enzyme activity assay. Approximately, 0.5 ml of the crude enzyme was mixed with 1 ml of Na-citrate (pH 4.8) in a test tube. The test tube was immersed in a water bath at 50°C and incubated for 60 min. After that, 3 ml of DNS reagent was added to the tube. Test tube was then heated in a boiling water bath for 5 min followed with cooling by immersion in it water bath filled with water at room temperature (27°C). After that, 20 ml of distilled water was poured into the test tube and inverted several times until the mixture was homogenous. After that, the tube was settled down for at least 20 min, and the absorbance of the solution was measured at 540 nm [14]. One unit of enzyme activity is the required amount of enzyme to release 1 mol of reducing sugar per min per ml which can be calculated by equation (5).

$\mathrm{Enzymeactivity}=\left(\frac{\mathit{FPU}}{\mathit{ml}}\right)=\frac{\mathit{sugar}\mathit{concentration}}{\mathrm{MW}\mathrm{glu}\cos \mathrm{axt}}\times \frac{\mathrm{H}}{\mathrm{E}}(5)$

where MW is a molecular weight of glucose (180 g/mol), t is incubation time (min), H is total volume of enzyme-substrate (ml), and E is volume of enzyme (ml).

All the analyses were performed in triplicate and expressed as mean ± standard deviation. Differences were tested with one-way analysis of variance (ANNOVA) followed by Duncan test using MINITAB 17.

2.7. Estimation of Kinetic Parameters

Modified Monod and Luedeking-Piret equations were used to determine relevant kinetic parameters, particularly ∞_m, K_s, k_d, Y_x/s, m_s, k₁, k₂, dan K_i. for modeling cell growth, enzymatic activity, and consumption of substrate using equation (6), (7) and (8), respectively [15].

$\frac{\mathit{dX}}{\mathit{dt}}={\propto }_{\mathit{m}}\frac{\mathit{SX}}{{\mathit{K}}_{\mathit{s}}+\mathit{S}}-{\mathit{k}}_{\mathit{d}}\mathit{X}(6)$

where X is the total cell concentration of fungi in the medium (g/l), μ_m is the maximum growth rate (1/h), S is the concentration of substrate (g/l), K_s is the saturation constant of substrate (g/l), and k_d is the constant for cell death (1/h).

$\frac{{\mathit{dE}}_{\mathit{t}}}{\mathit{dt}}=\frac{{\mathit{k}}_1\mathit{X}}{1+\frac{\mathit{S}}{{\mathit{K}}_{\mathit{i}}}}-{\mathit{k}}_2\mathit{E}(7)$

where k₁ is the synthesis rate constant for cellulase (IU/mL h), k₂ is the decomposition rate constant for cellulase (1/h), and K_i is the coefficient for substrate inhibition.

$\frac{\mathit{dS}}{\mathit{dt}}=-\frac{1}{{\mathit{Y}}_{\mathit{X}/\mathit{S}}}\frac{\mathit{dX}}{\mathit{dt}}-{\mathit{m}}_{\mathit{s}}\mathit{X}(8)$

where Y_x/s is the yield of cell biomass per substrate (g/g), and m_s is the specific maintenance coefficient (1/h). The equations were solved using ode113 function in MATLAB and compared with secondary data obtained from the literature for SSF and SmF (Castillo et al., 1994; Ma et al., 2013)

3.1. Characterization of Lignocellulose Content in Groundnut Shell

Lignocellulose content of groundnut shell used in this study was determined using a Chesson-Datta method, and the results are shown in Table 1. Based on Table 1, the groundnut shell used in this study has a very low moisture content and primarily composed of hemicellulose; 50.15 ± 0.04% dw, followed by lignin; 33.70 ± 1.44% dw and cellulose; 12.24 ± 0.78% dw. The cellulose content obtained in this study is lower than the values reported in previous studies that can reach as high as 63.5% [16].

The differences may be due to the origin of the sample where the sample used in this study was obtained from Central Java, Indonesia meanwhile the values reported by other studies were obtained from different countries, particularly Japan, China, and Singapore [22,23]. Different cultivation locations may cause differences in soil temperature [24], altitude [21], light intensity [25], and soil moisture content [26,27] resulting in differences of lignocellulose content in the same varieties. Previous studies also have reported that drought stress has a great influence on the cellulose and hemicellulose content in biomass [27-30].

Table 1: Lignocellulose composition of groundnut shell. [Click here to view]

3.2. Effect of Fermentation Time on Enzymatic Activity of Cellulase

Fermentation was carried out for 5 days, and the enzymatic activity of cellulase was determined daily and the results are shown in Figure 1. The enzymatic activity of cellulase observed in this study lies in the range of 0.06–0.12 filter paper unit (FPU)/ml. The highest activity of cellulase was obtained from the fermentation of Trichoderma sp and tape yeast that reaches up to 0.12 FPU/ml. Mixed cultures have a higher cellulase activity as compared to the single cultures, probably due to growth symbiosis in mixed cultures [3] and synergism of enzyme complexes [31,32] which may occur due to the presence of 3 different enzymes that work synergistically, endo-β-glucanase, exo-β-glucanase, and β-glucosidase [4]. It has been reported that Trichoderma sp typically produces cellulase with high levels of endoglucanase and exoglucanase activity, but relatively low β-glucosidase activity [7].

The difference in activity of the enzyme complex causes the enzyme complex to interact synergistically and produces a higher activity compared to single culture enzymes. In addition, the growth of microorganism in mixed cultures mat performed better than in single cultures due to the synergism of metabolites produced by both cultures can support the growth of both cultures [33]. Higher biomass growth of microorganisms can support the production of cellulase enzymes with more optimum activity. Cellulase activity may vary depending on the microorganism, substrate, fermentation time and fermentation method as shown in Table 2. From the table, it can be noted that the cellulase activity determined in this study and the literature is quite different. Differences in enzyme activity can be caused by several factors, such as provision of pre-treatment, and cellulose content in the substrate. Pre-treatment of the substrate before the fermentation is highly recommended. Physical pre-treatment in the form of steam explosion [37,38] and ultrasonic [39] or chemical treatment in the form of alkali [40] is known to increase cellulase enzyme activity in the fermentation process. In this study, the groundnut shells used for the fermentation process were not pretreated. Meanwhile, in the study of Mojsov [35] and Afriyanti et al. [36], the fermentation substrate was pretreated. This could be the reason why the enzyme activity in both studies was higher than this study.

Pre-treatment of the fermented substrate is known to help loosen the lignocellulose structure, increase the solubility of hemicellulose, reduce the crystal structure, and increase the surface area and pore on the substrate so that the enzyme hydrolysis process becomes easier [41]. The high content of lignin can inhibit the hydrolysis of cellulose. In addition to protecting the cellulose structure from being hydrolyzed by the cellulase enzyme, lignin is also known to be bound to the cellulase enzyme so that the activity of the cellulase enzyme decreases [42]. Pre-treatment with H₂O₂ [43] or lignin-decomposing fungi [44] is known to dissolve lignin compounds in lignocellulose so that the activity of cellulase enzymes can be increased. Pre-treatment on the substrate is known to increase the activity of the cellulase enzyme in the fermentation process from 1.2 to 6.9 times compared to the fermentation process without pre-treatment [3,39].

Figure 1: Maximum enzyme activity of 5 days fermentation at room temperature condition. [Click here to view]

Table 2: Comparison of cellulase activity using different substrate. [Click here to view]

The cellulose content in the substrate can also affect the activity of the cellulase enzyme. Cellulase enzyme production by Trichoderma sp. is induced by the presence of inducer compounds such as pure cellulose, lactose, sophorose, and cellobiose [45,46]. Sophorose and cellobiose are compounds derived from cellulose [47]. Cellulase production by Trichoderma is also influenced by the presence of an inducer compound. The presence of the induction compound in the medium will be directly proportional to the production of cellulase. Because cellulase-inducing compounds are cellulose-derived compounds, the production of cellulase enzymes will be higher if the cellulose content in the substrate is also higher [45]. In this study, the groundnut shell substrate used has a cellulose content that is relatively small; 12.24 ± 0.78% dw. This could be the cause of low enzyme activity produced compared to Toscano et al. [34] that used a substrate with a higher cellulose content.

Nutrient source is one of the most important factors to produce metabolites. In this study, no additional nutrient sources were used, such as basal medium or Mendel and Reese’s medium which provided a source of nitrogen and vitamins for microorganisms. This simple medium is used by microbes as an early energy source in the initial condition to provide biomass growth. The nutrients obtained by microbes in this study were only provided by groundnut shells and carrier media, such as talc. This could be one of the causes of the lower cellulase activity. Adequate nutritional composition can maintain microbial growth so that microbes can grow optimally and produce cellulase activity more effectively. This study did not use an additional medium aimed at examining the production ability of Trichoderma sp and tempeh yeast using the lignocellulosic substrate as the only nutrient. On the other hand, the use of additional chemical mediums can increase production costs on a large scale. The cellulase production using lignocellulose as the sole nutrient can reduce production costs and reduce waste in the surrounding environment [48].

3.3. Effect of Fermentation Time on the Amount of Biomass

In this study, the growth of fungal biomass was calculated to determine the effect of different cultures on fungal growth. Fungal biomass was measured every 24 h for 5 days of fermentation. The result obtained from the fungal biomass measurement was tested using one-way ANOVA followed by Duncan post hoc test to analyze the results statistically. From the test, it can be concluded that fungal growth between each culture variations is significantly different (P < 0.05). The weight profile of fungal biomass throughout the fermentation period is presented in Figure 2.

From the figure, the maximum weight of fungal biomass was obtained from mixed culture of Trichoderma sp. and tempeh yeast with a biomass weight up to 0.34 ± 0.00 g. That value was significantly higher than the others. This could be due to the synergism between the metabolites released by Trichoderma sp and tempeh yeast which mutually support the metabolic processes of the microorganisms [33]. However, Trichoderma sp. and tape yeast’s maximum biomass weight has a lower value than the others. This could be due to the competition that occurs between the microorganisms. Trichoderma is classified as a microorganism with high competitiveness characteristic and biofungicide activity [49]. However, this phenomenon needs to be investigated further by analyzing secondary metabolites produced during the fermentation process to obtain more conclusive explanation.

3.4. Mathematical Modelling for the Synthesis of Cellulase

In this study, mathematic modeling was carried out using secondary data from Castillo et al. [50] and Ma et al. [15] to estimate the value of kinetic parameters for production of enzyme using SMF and SSF. The value of kinetic parameters will differ from one method to another depending on microorganism species, fermentation conditions, and substrate characteristic. The secondary data from Castillo et al. [50] used a mixed culture of Trichoderma reesei and Aspergillus niger, with sorghum as the substrate, Mandel and Reese’s medium as a basal medium, and 10:1 carbon to nitrogen ratio. Fermentation process was performed at room temperature (25°C), and 70% humidity. Meanwhile in the second literature, T. reesei was cultivated in fed-batch fermentation system with 5 days of cultivation using pure cellulose as a substrate as reported b. Fermentation process was performed at 26°C, pH 5 and 30% dissolved oxygen [15].

The simulated parameters from both fermentation methods are shown in Table 3. From the results, there were differences among the parameters for SMF and SSF method. The differences could be due to the characteristic of the substrate, microorganism species, and fermentation conditions [51]. The μ_m value of SSF and SMF method differed from each other, although the same species of Trichoderma sp. was used in both methods. This might be related to the type of strain, fermentation method and substrate used that differ from both literatures [15]. Substrate saturation constant (K_s) of both methods also different from each other. This constant determines the saturation point of substrate at fermentation process. Both literatures use different substrates for their fermentation process. The first literature used sorghum as substrate for SSF method, meanwhile the second literature used pure cellulose as substrate for SMF method. The difference of substrate leads to different values of K_s for both simulations. Pure cellulose is well known as the best substrate for cellulase production through the fermentation process. Pure cellulose is more accessible by the microorganism to produce cellulase than another substrate such as sorghum or any agricultural waste that has another substance aside from cellulose such as lignin that could inhibit the cellulase production process by the microorganism [52].

Table 3: Estimated value of kinetic parameters for SSF and SMF method. [Click here to view]

Figure 2: Fungal biomass weight for 5 days of fermentation at room temperature [Click here to view]

The ccoefficient of substrate inhibition represents the concentration limit of substrate for inhibition to occur. Coefficient value is influenced by several factors such as type of substrate used, production of byproduct, and main product accumulation. Production of byproduct and accumulation of main product could inhibit enzyme production process and cell growth [53]. K_i value for SSF method was higher than SMF method. It could be influenced by the fermentation method. In solid-state fermentation, the enzyme produced by the microorganism tends to remain to stay at the substrate, whereas in SMF, the enzyme produced will be dissolved in the medium. This would lead to the higher substrate and product inhibition for the SSF method than the SMF method, resulting in a higher value of K_i at the SSF method [54].

Cell death constant, rate constant of cellulase synthesis, stoichiometric yield coefficient of biomass, and specific maintenance coefficient generally were influenced by species of microorganism used. Every microorganism has its own characteristic and metabolism pathway that could affect the value of kinetic parameters. Cell death constant describes the amount of degraded biomass. From the results presented in Table 3, a higher value of cell death constant would lead to higher value of cellulase degradation. It might be related to decreased amount of microorganism that could produce cellulase [55].

The kinetic parameters obtained from the simulation process could be used to build up mathematical model of cellulase production by SMF and SSF methods. The mathematical model consists of cell growth, substrate consumption and enzyme activity model. The simulated model compared to the experimental data to analyze the accuracy rate of the models. Comparison between simulated model and experimental data from literature for biomass, substrate, and cellulase activity was shown at Figures 3-5.

Biomass, substrate, and enzyme activity model fit the experimental data well with R² values of 0.97; 0.97, and 0.78, respectively, for SSF method. For SMF method, R² values for biomass, substrate, and enzyme activity model were 0.83, 0.99, and 0.93, respectively. With the high value of R², it can be concluded that the models constructed could represent data very well and have a high accuracy rate [56]. However, further studies are needed to assess the fermentation process better with include more aspects such as mass and energy transfer that occur within the fermentation process.

Figure 3: Simulated model of fungal growth based on secondary data with SSF [50] (a) and SMF [15] (b) method [Click here to view]

Figure 4: Simulated model of substrate consumption based on secondary data with SSF [50] (a) and SMF [15] (b) method [Click here to view]

Figure 5: Simulated model of enzyme activity based on secondary data with SSF [50] (a) and SMF [15] (b) method [Click here to view]