Synthesis and evaluation of biosilicified Pseudomonas fluorescens lipase
Based on previous reports [31–33,35], biosilicified Pseudomonas fluorescens lipases were synthesized with different theoretical amounts of enzyme (0, 1, 5, 10, 50, 100 and 150 mg) and the activity of the obtained biocatalysts to produce biodiesel was evaluated.
Initial essays were carried out employing the encapsulated lipases as biocatalysts, and sunflower oil and food-, pharmaceutical-, and cosmetic-grade ethanol (96 v/v % and available in most local grocery stores) as raw materials. The latter was chosen for three reasons: (1) the enzyme needs the water to be active (4% v/v H2O is adequate; the water acts as a lubricant allowing the enzyme active site to open and interact with the substrates) [22,25,36], (2) it is produced from renewable sources, and (3) ethanol has lower flammability and toxicity than methanol.
Transesterification activity was detected for all biocatalysts analyzed (Figure 1). Moreover, the activity of each biocatalyst increased together with the amount of immobilized enzyme. However, only protein loadings higher than 67 mgprotein/gsupport produced acceptable fatty acid ethyl esters (FAEE) contents (Table 1). Moreover, the immobilization efficiency of lipases achieved during the biosilicification process was higher than 90% with respect to the theoretical content. On the other hand, biodiesel production was measured under initial rate conditions (2 h in our system and the enzymatic reaction showing linear behavior) in order to compare the different FAEE contents obtained. As can be observed, with the sole exception of LOBE6, the FAEE content showed to be proportional to the lipase content (Table 1).
Table1. Protein loading, FAEE content and specific activity of the synthesized biocatalysts.
Biocatalyst
|
Theoretical protein amount (mg)
|
Protein loading (mgprotein/gsupport)
|
Immobilization efficiency (%)
|
FAEE content (wt%)
|
Specific activity (U/mglipase)
|
Free lipase
|
50
|
-
|
-
|
30.17
|
5.03
|
LOBE1
|
1
|
1.17
|
98.60
|
0.45
|
25.59
|
LOBE2
|
5
|
6.04
|
99.70
|
2.43
|
26.97
|
LOBE3
|
10
|
11.80
|
99.82
|
3.87
|
22.12
|
LOBE4
|
50
|
66.85
|
96.88
|
62.39
|
66.37
|
LOBE5
|
100
|
120.66
|
97.66
|
68.61
|
42.48
|
LOBE6
|
150
|
141.72
|
96.03
|
57.82
|
31.05
|
Based on FAEE content results, LOBE5 appeared to have the best biocatalytic activity. However, when the specific activity was determined, LOBE4 showed to have the best performance: the specific activity was 57% higher than that of LOBE5. Therefore, LOBE4 was chosen to proceed with its physicochemical characterization.
It should be highlighted that when the free lipase was used, significantly fewer contents of FAEE were produced (Table 1). That was probably since free enzyme forms aggregates in the organic reaction mixture, which reduces the number of exposed active catalytic sites, and therefore, making it less efficient [25].
As control, when material synthesis in the absence of enzyme was evaluated, no FAEE production was detected, indicating that the lipase is responsible for the transesterification activity.
Physicochemical characterization of the LOBE4 biocatalyst
The structural and textural properties of the biosilicified enzymes were studied by X-ray diffraction (XRD), transmission electron microscopy (TEM), N2 adsorption/desorption to obtain the specific areas by the BET method and Fourier Transform Infrared Spectroscopy. The typical ordering of mesoporous silicates was detected when the control and LOBE4 were analyzed by small angle XRD. As shown in Figure 2, experimental spectra exhibit two maxima peaks which can be assigned to the typical (2 1 1) and (2 2 0) diffraction planes of a three-way pore structure such as that of Mobil Composition of Matter No. 48 (MCM-48) siliceous solid. In addition, the ratio value d220/d211 was approximately 0.87, which is also in accordance with the cubic structure of MCM-48 and indicates a structural organization in the architecture of biosilicified enzymes [37,38]. n-dodecylamine (DDA) is a molecule with a polar head and a hydrophobic chain, which can form micelles and act as a surfactant. In solution, DDA gives rise to a liquid crystal micellar phase that can be employed as structure directing agent to form mesoporous structures (such as MCM-48) when TEOS is mineralized [35,37,39,40]. During this process, the enzyme can insert itself into the micelles without interfering with the formation of the cubic phase (Scheme 1).
As can be observed in Figure 3, TEM analysis revealed that LOBE 4 has a nanotubular structure like a nanofiber network with channels. Similar results were previously reported by Garcia et. al., indicating that the siliceous mineralization has been carried out over the enzyme (to facilitate the observation of nanofiber network architecture, LOBE4 images are showed in original and inverted colors) [31]. Noteworthy, and as shown in Figure 1, the biosilicification process does not affect the activity of the enzymes or the flow of substrates and products.
Infrared spectroscopy is a widely used technique to study the substructure of peptides and proteins and it can be used to monitor the presence of the proteins on the supports [41,42]. Figure 4a shows the different functional groups characteristic of the free Pseudomonas fluorescens lipase evidenced by FT-IR. The stretching C–H vibrations of –CH2 and –CH3 groups, the stretching vibrations of carbonyl groups, and stretching vibration of ≡C–O– groups were detected. The signal of the –OH deformation vibrations appears at 1400 cm-1 [43]. Finally, the signals for most characteristic functional groups of the pure enzymes, amide I and amide II, were also observed (Figure 4a) [44,45]. The decrease of the intensities and wavenumbers of amide bands in LOBE4 FT-IR spectrum indicates that the immobilization of the lipase inside the silica matrix was successful (Figure 4b) [46]. Furthermore, the presence of the amide I and amide II bands evidences that the secondary structure and bioactivity of the enzyme are conserved when the protein is incorporated into the nanostructure (Figure 1) [47]. The presence of silicon could be corroborated by the ≡Si–O bond vibration band [48–50]. Bands between 3500 and 3300 cm-1 correspond to primary amine (–N–H bond) of DDA (Figure 4b). Additionally, the bands assigned to C–H stretching of the saturated –CH2 and –CH3 groups increase their intensity in comparison to the free enzyme, due to the surfactant presence in the biocatalyst (Figure 4b). These bands disappeared when the biocatalyst was calcined at 500 °C for 8 h, indicating that the organic material was removed (Figure 4c).
The FT-IR spectrum of the calcined biocatalyst reveals only the characteristic bands of the siliceous matrix such as the ≡Si–O bond vibration, the ≡Si−OH bending, and ≡Si–O–Si≡ symmetric stretching and bending (Figure 4c) [48–50,53].
Since the FT-IR spectrum of LOBE4 showed the presence of typical functional groups of the enzyme and the siliceous material, it confirmed the effective immobilization of lipase through biosilicification. Then, control material and LOBE4 (calcined and non-calcined) specific surface was determined (Table 2).
Table 2. Specific areas of control and LOBE4 before and after calcination.
Sample
|
Specific areas m2/g (C)
|
Specific areas
m2/g (NC)
|
Control
|
221.01
|
4.97
|
LOBE4
|
289.54
|
6.12
|
C=Calcined, NC= Non calcined
After calcination at 500 °C, the specific area was expected to increase due to the elimination of the organic phase (surfactant and lipase). Nevertheless, the XRD pattern of the calcined sample did not show the permanence of an ordered structure (Figure 5), indicating that the architecture of the siliceous network generated by the biosilicification process is unstable at high temperatures.
These results indicate the formation of an ordered but incomplete siliceous-organic hybrid structure, further suggesting that silicification fails to cover the entire organic structure. Consequently, by removing the organic template, the nanostructure collapses. Nonetheless, the incomplete mineralization gives the biocatalyst a certain flexibility that allows the diffusion of substrates and products to the active sites of the enzymes. Due to these characteristics, these materials were denominated as Low Ordered Biosilicified Enzyme (LOBE) (Scheme 1).
Assessment of LOBE4 activity with alternative raw materials
To analyze the versatility of LOBE4, its capacity to catalyze the production of biodiesel was tested with the following feedstocks: soybean oil, waste frying oil, Jatropha Excisa oil, acid oil from soapstock and pork fat (Figure 6a). The selection of these raw materials was based on the reasons detailed below. In 2018, Argentina was the world leading exporter of soybean oil and the world third largest exporter of sunflower oil [54]. Besides, the use of soybean oil as raw material for the production of biodiesel does not produce a detrimental effect on the local population food availability since sunflower oil is mostly consumed in Argentina. Furthermore, biocatalytical conversion of soybean oil into fuels and chemicals of commercial interest would foster local industrialization and employment generation [55]. In addition, the process of soybean oil purification generates an acid oil side-product containing a large amount of free fatty acids (50-80wt% of FFA approximately) and, to a lesser extent, a mixture of phospholipids, tocopherols, sterols, degraded and oxidized residues, pigments, salts, color bodies, triglycerides, diglycerides and monoglycerides [56,57]. Converting this acid oil into biodiesel represents an attractive added value strategy for a more efficient use of agricultural production and recycling.
Moreover, waste frying oils are a domestic and gastronomic industry waste with high energy content. These waste oils are available in large quantities at a minimal cost and are often discarded in drains, causing obstructions in the sewer system and polluting water resources. Recycling them for the production of biodiesel could be a greener alternative to substantially reduce the price of biofuel [58–60].
In addition, the use of non-edible oilseeds for the production of biofuel is also an attractive alternative as a raw material in place of oils intended for food. Jatropha excisa is an endemic and non-conventional oilseed species from the semiarid and arid northwest region of Argentina with an average oil concentration of 34 wt%. Although this oil is presumably toxic, native people have used it for centuries in traditional medicine as purgative and emetic [61]. It does not represent a competition for agricultural food crops and diversifies farmland, emerging as an alternative feedstock for biofuel production with high economic potential [62,63].
Table 3. Characterization of raw materials.
Feedstock
|
Density (g/cm3)
|
Kinematic Viscosity (mm2/s) 1
|
Acid Value (mgKOH/goil)
|
FFA Content (wt %) 2
|
Water Content (ppm)
|
Triglyceride content (wt %)
|
|
Soybean oil
|
0.93
|
18.38
|
0.13
|
0.07
|
626
|
97.70
|
|
Waste Frying oil
|
0.94
|
20.48
|
0.21
|
0.11
|
671
|
95.73
|
|
Jatropha Excisa oil
|
0.92
|
15.75
|
1.55
|
0.78
|
980
|
94.60
|
|
Acid oil from soybean soapstock
|
0.96
|
10.94
|
153.72
|
76.91
|
5221
|
9.18
|
|
Pork Fat
|
0.90
|
21.85
|
0.67
|
0.33
|
590
|
97.64
|
|
1 At 60 °C (reaction temperature). 2 Calculated from the acid value (EN 14104: 2003) and expressed as oleic acid [64].
Animal fats are other alternative raw material source for biodiesel production since their cost is considerably lower than that of vegetable oils. These feedstocks are currently added to pet food and animal feed or used for industrial soap production. Nevertheless, many researches have shown that these raw materials can effectively be transformed into a high-quality biodiesel that meets the ASTM specifications [65,66].
Excluding soybean oil, the remaining raw materials detailed represent interesting sources for the production of second-generation biofuels. Nevertheless, their high content of free fatty acids and water does not allow their direct use in the conventional homogenous process of biodiesel production (Table 3). In turn, free fatty acids must be esterified with sulfuric acid and methanol (homogeneous acid process); then, the acid catalyst in solution must be neutralized, and the product should be washed and dried. Only after this pretreatment, the resulting raw material (a mixture of fatty acids esters and triglycerides) can be employed in the transesterification reaction with a homogeneous basic catalyst. Finally, neutralization, washing and drying steps must be performed again to obtain the product that will be used as fuel [67,68]. All these steps substantially increase the cost of the process.
As shown in Figure 6b, the five oily raw materials analyzed were successfully converted into biodiesel in the presence of ethanol and LOBE4, which indicates that the biocatalyst was able to mediate the transesterification of acylglycerols and the esterification of the FFA without any previous treatment. Therefore, when the reaction is complete, purification steps are reduced to just separating the biocatalyst from the reaction mixture by simple filtration and removing excess ethanol (which can be also recovered and reused in the next reaction). Then, the biocatalyst can be reactivated by washing with acetone (to eliminate the organic residues) and dried at room temperature to be reused in the next cycle. Overall, the sustainability of this process lies in the use of alternative, renewable, and low-cost raw materials to produce second-generation biofuels that reduce the emission of carbon dioxide.