2.1. Optimization of Fermentation Parameters
In this study, eight variables i.e., inoculum density, moisture ratio, urea, incubation temperature, sugarcane bagasse (SB) solids, glucose, SB particle size and used cooking oil (UCO) ratio that affect lipase productivity were screened. The analysis of variance (ANOVA) was applied to test the interaction effects of the variables (
Table 1). The
p-value less than 0.05 indicated that the model terms are significant. Among the selected parameters, only moisture (
p-value: 0.0021) and UCO (
p-value: 0.0122) ratios were the significant variables. From these results, the correlations between the two factors and their effects on lipase production were evaluated using a 3-Level Factorial Design.
The model
F-value of 14.24 (
p < 0.0001) was obtained from the ANOVA analysis (
Table 2), which implies the model is significant. There is only a 0.01% chance that such a “Model
F-value” could occur due to noise. From the results, there is no correlation between the moisture and UCO ratios (
p-value > 0.05).
However, the moisture ratio was found to be the enhancing factor, while UCO ratio as the delimiting factor as shown in Equation (1) and
Figure 1. The moisture ratio acts as the enhancing factor as all organism need moisture for growth. Moisture was needed only up to a certain extend in SSF, otherwise it will be submerged fermentation. On the other hand, the UCO ratio acts as the delimiting factor as too much oil will cause fungus suffocation, limiting the lipase production. UCO is only needed in small amounts, as an inducer.
By selecting a moisture ratio of 3.0, the predicted lipase activity is calculated as 0.112 based on Equation (1). After fermentation, the lipase activity was observed to be 0.110 ± 0.00 U/gSB. The predicted and observed values were found to be similar and the model was thus validated.
2.4. Oil Hydrolysis and Glycerol Esterification
In this study, undefined or raw triacylglycerol substrates which are often used in industry were chosen for hydrolysis. As TAG is associated with three fatty acids, all three fatty acids must be released before glycerol is detectable. From the results obtained,
ScLipA showed the highest affinity towards long chain triacylglycerol (LCTG) hydrolysis. This was apparent in
Table 4 as the crude fish oil hydrolysis produced glycerol of 2.489 μmol/mg
protein/day. Fish oil is a source of LCTG, especially long chain essential fatty acids (C20 and C22) [
6]. The total crude fish oil from the extraction method was 13.2 g per 400 g of fish viscera and unwanted parts, which was used directly as crude fish oil. For this study, commercial coconut oil with 47.7% of lauric acid (C12:0) [
13] was used as a model to represent the medium chain triacylglycerol (MCTG). Lauric acid was found as an effective antimicrobial and antifungal material [
14]. As seen in
Table 4, coconut oil hydrolysis produced glycerol of 0.405 μmol/mg
protein/day. The result indicated that
ScLipA has an affinity towards MCTG as well. As for short chain triacylglycerol, unsalted commercial butter with significant amounts of butyric acid [
14] was chosen for this study and the glycerol production was 0.236 μmol/mg
protein/day. From the results, it was observed that
ScLipA showed affinity towards the hydrolysis of LCTG, MCTG and short chain triacylglycerol (SCTG).
ScLipB catalyzed glycerol production from hydrolysis of crude fish oil (0.460 μmol/mg
protein/day). However, glycerol production was not detected for the hydrolysis of commercial coconut oil and unsalted commercial butter. These indicated that
ScLipB has an affinity towards the hydrolysis of LCTG, but not MCTG and SCTG. These results also coincide with the findings of Singh et al. [
15] on fatty acid methyl ester (FAME) production from cyanobacterial endolith
Leptolyngbya ISTCY101 oil using immobilized purified 70% (NH
4)
2SO
4 precipitated lipase from
S. commune ISTL04. They reported that the cyanobacterial oil was mainly palmitic acid (C16:0), palmitoleic acid (C16:1), stearic acid (C18:0) and oleic acid (C18:1) fatty acids (more than 60%). The results from both studies verified that the 70% (NH
4)
2SO
4 fraction from
S. commune, which is similar to
ScLipB in this study, has affinity towards long chain fatty acids. This can also be observed as the
S. commune UTARA1 was cultured on milled SB impregnated with UCO of palm oil origin which is rich in palmitic acid.
Most lipases are either
sn-1,3 regiospecific or non-regiospecific [
16]. The
sn-1,3 regiospecific lipase act on the ester bonds from the
sn-1 or
sn-3 positions of the TAGs [
8], while non-regiospecific lipases hydrolyzes the ester bonds of the TAGs randomly [
16]. On the other hand, Lotrakul and Dharmsthiti [
17] stated that if the lipase is
sn-2 regiospecific, 1,3-DAGs and 1(3)-MAG might be produced. According to Tong et al. [
18], the TLC result obtained with
Candida antarctica lipase A (CalA) displayed a clear 1,2- (2,3-) diolein spot and a slightly more intense spot of 1,3-diolein, indicating that CalA was a
sn-2 regioselective lipase. For this study, similar scenario can also be observed on the TLC plate in Lanes 4–6 of
Figure 4, showing a significant amount of 1,3-DAGs, and less of 1,2- (2,3-) DAGs. Therefore, it was stipulated that
ScLipA is most probably a
sn-2 regioselective lipase.
TLC was recommended for qualitatively screening enzymatic TAGs hydrolysis reactions [
8]. As a result of their affinity to a variety of substrates, lipases possess specific characteristics for producing different products [
5]. Direct esterification of glycerol with acetic acid was carried out with
ScLipA in a solvent free system with the molar ratios of 1:9, and the results were evaluated using TLC.
Figure 5a shows the time course profile of direct esterification of glycerol and acetic acid using
ScLipA and triacetin (Nacalai tesque, Kyoto, Japan) was used as a control. It can be seen that glycerol was esterified to form monoacetin and diacetin from Day 1–4; while triacetin was formed on Day 5 with a reduction in the intensity of monoacetin spot. As for lauric acid, monolaurin was synthesized on Day 1 as in
Figure 5b Lane 1. Trilaurin can be observed on Day 2 for the molar ratios of 1:3 (glycerol:lauric acid); however, the trilaurin produced was degraded subsequently, as shown in
Figure 5b Lanes 3–5. Hydrolysis and esterification are reversible in lipase reaction [
9], where the water produced during esterification of lauric acid was most probably involved in the hydrolysis of trilaurin back to its original state, which is caused by the affinity of
ScLipA to hydrolyze MCTG. To ensure the reaction move towards the esterification process, water must be kept at a minimum amount or totally removed. This was also supported by Rosu et al. [
9], who stated that water removal during reaction is very important for moving the reaction equilibrium towards esterification processes. Esterification of lauric acid and glycerol (molar ratio of lauric acid/glycerol of 3) with 9% Lipozyme IM 20 at 80 °C was conducted by Langone and Sant’ Anna, Jr. [
19]. They found that 75% of trilaurin was produced after 26 h of incubation. Conversely, this study showed that trilaurin can be produced by Day 2 using 4% partial purified lipase at 30 °C; as the enzyme is more stable at this temperature. As for direct esterification of glycerol and oleic acid,
ScLipA did not show any triolein production (results not shown).
Figure 6a shows the time course profile of direct esterification of glycerol with acetic acid (molar ratios of 1:9) using
ScLipB. Triacetin (Nacalai Tesque, Kyoto, Japan) was used as a control. Similar with
ScLipA,
ScLipB also esterified glycerol forming monoacetin and diacetin from Day 1–4; while triacetin was formed only on Day 5, with a reduction in the intensity of monoacetin spot. Liao et al. [
20] were able to obtain 100% triacetin conversion with the same molar ratio using Amberlyst A-35 at 105 °C for 4 h. They also stated that the glycerol conversion increased with the increase of temperature and molar ratio [
20]. Conversely, this study took on a more subtle approach, using lipase at 30 °C.
In
Figure 6b, oleic acid (Bendosen, Norway) and triolein (Sigma, St. Louis, MO, USA) were used as control for Lanes 6 and 7, respectively. With the molar ratio of 1:9 (glycerol:oleic acid), 2-mono-olein was only apparent on Day 5 by
ScLipB as shown in
Figure 6b Lane 5, a distinct spot above 1-mono-olein. A
sn-2 regioselective lipase will incorporate fatty acid at the
sn-2 position in the glycerol molecule. Based on the results shown in
Figure 6b, it was stipulated that
ScLipB is also a
sn-2 regioselective lipase. Using
Aspergillus carneus lipase at 30 s high power microwave irradiations (800 W, 90 °C) under solvent free system, higher percentage of triolein (50%) was synthesized [
21]. Conversely, this study took on a more subtle approach, using lipase at 30 °C. By manipulating the reaction conditions and suitable enzyme selections, the resultant products can be controlled [
8], driving the reactions either towards hydrolysis or esterification. As for direct esterification of glycerol and lauric acid,
ScLipB did not show any trilaurin production (results not shown).
Sharma and Rathore [
22] stated that for bacteria lipases, ammonium sulfate precipitation can only purify the enzyme to a certain degree that is suitable for detergent formulations, but further purification is required for synthetic reactions. This might due to the fact that most bacterial lipases are intracellular, compared to the extracellular fungal lipases which may function well outside the cell. Based on our findings, the fungal lipases that were precipitated with ammonium sulfate can be used for several application purposes, namely TAGs hydrolysis and synthesis. To the authors’ best knowledge, this is the first time that
S. commune was reported to have two different lipases, namely
ScLipA and
ScLipB. The two partial lipases were tested on various TAGs and fatty acids of different chain lengths and showed different affinities towards different substrates.