Molecular characterization of transesterification activity of novel lipase family I.1

Lipase are a group of hydrolase enzymes showing broad catalytic capabilities, including hydrolysis, esters synthesis, transesterification (acidolysis, interesterification and alcoholysis) and also aminolysis [1,2]. Lipase families were grouped into eight families and six sub-families of true lipase [3]. Today, lipase families comprise 35 families and 11 subfamilies of true lipase [4,5]. Lipases are used in a variety of industries, including agriculture, detergents, food, nutraceuticals and biodiesels [6].

Organic solvents are used in the majority of industrial-scale synthetic processes because of their ability to alter reaction balance toward synthetic direction and ease of product recovery, higher activity and stability, stereoselectivity and regiospecificity. [7]. Therefore, searching for organic solvent tolerance lipase is very important. Tolerance on organic solvents is positively correlated with thermostability [8]. These two characteristics make lipase can be fulfilled as industrial biocatalyst.

Thermostable and solvent tolerance enzymes could be obtained through protein engineering strategies in collaboration with discovery of new lipase [9]. However, protein engineering strategy requires a complete understanding of the structure of lipase using pre-characterized or commercial lipases. Hence, discovery of new lipase with both specific characteristic are important and needed to be explore. The source of a new lipases could be obtained by screening from cultivated microorganisms or metagenome approach [9–13].

Pseudomonas sp. lipase are emerging biocatalysts with thermostable and organic solvent tolerance character. Amano has marketed lipases from P. cepacia (Lipase PS) and P. fluorescens (Lipase AsK) [14]. The other widely used Pseudomonas lipase is Pseudomonas stutzeri lipase which have aminolysis and transesterification activity and also have enantiomer selectivity [15–18]. Studies of metagenomics related to P. stutzeri lipase have so far been rare.

This report investigated Lk2 and Lk3 thermostability and characterized transesterification activity in organic solvents. The enzymes are recombinant thermostable lipases isolated by metagenomic approach from domestic compost [19]. Based on homological analysis, the lipases are most closely related to P. stutzeri lipase belonging to family I.1. Both lipases have been characterized based on hydrolysis activity [13]. In Escherichia coli BL21(DE3), Lk2 and Lk3 are overexpressed and extracted as active soluble enzymes using the thermolysis technique. As the physiological and biochemical properties of Lk2 and Lk3 may differ despite being obtained from the same source, experimental and in silico analysis were carried out to probe a better understanding on molecular interactions of lipase–substrate, ion metal, and thermostability.

Lipase genes LK2 and LK3 were obtained from our collection. The recombinant plasmid pET-30a(+)-LK2 and pET-30a(+)-LK3 were transformed into E. coli strain BL21 (DE3) (15). pET-30a (+) plasmids vector were used as an expression vector. All of the chemicals were bought from Merck (Merck, Germany) and Sigma (Sigma, Chemicals, U.S.A.).

LB medium contained kanamycin sulfate (50 g/ml) was used to cultivate E. coli BL21 (DE3) with recombinant plasmid and shaken at 150 rpm at 37°C. IPTG was added at a final concentration of 1 mM, and the mixture was incubated at 37°C for 4 h when the 600 nm absorbance was 0.6–0.7. Centrifugation was employed to extract cells, which were then stored at −20°C until needed.

By adding 0.1% sodium dodecyl sulfate (SDS), The cells were resuspended in 50 mM sodium phosphate buffer (pH 8.0) and incubated for 30 min at 50°C. After centrifuging the mixture for 30 min at 11900 g, the supernatant was recovered. To avoid denaturation, the supernatant was settled with a 30 mM K₂HPO₄ buffer.

SDS-PAGE was done at 110 V on a 12% running gel with SDS running buffer. Protein bands were visualized by using 0.1% Commasie Brilliant Blue to stain the gel. The purified enzyme’s molecular mass was calculated using Protein marker III (pre-stained), peqGOLD Protein Ladder. The polyacrylamide gel-embedded lipase was renaturated with 50 ml of sodium phosphate buffer (50 mM, pH 8) containing 0.1% tritone overnight in preparation for the zymographic examination. Subsequently, 1-naphthalene laurate was incubated with lipase in 50 ml sodium phosphate buffer (50 mM, pH 8) at 50°C for 4 h. By adding 25 mg of fast blue dye, the staining activity was performed.

Ni-NTA agarose matrix (1.5 ml) on column was balanced with 12 ml miliQ water, After that, 15 ml PBS 50 mM at pH of 8 (1% (V/V) Triton-X 100, 100 mM NaCl) were added. The recombinant enzymes were collected on a column after the thermolysis stage. Then, the column washed with PBS 50 mM at pH of 8 contained 100 mM NaCl. The elution buffer was used to elute the bound recombinant enzymes (50 mM PBS pH 8, 300 mM NaCl, imidazole 100 mM). The elution fraction was dialyzed overnight used PBS buffer pH 8 at 4°C to remove any remaining SDS and imidazole. The recombinant enzyme was then concentrated using diafiltration (Merck Millipore). Using the Bradford technique and a bovine serum albumin standard, the amount of protein in the sample was determined [20]. The concentrated enzymes were used for transesterification assay.

Transesterification activity of recombinant lipase Lk2 and Lk3 were measured using Pohnlein method with slight modification [21]. Transesterification was performed between methyl esters and para-nitrophenol in organic solvents at various temperature. The conversion of para-nitrophenol into para-nitrophenyl esters was monitored through decreasing absorbance at 400 nm. The reaction procedures as followed: 10 µl concentrated recombinant lipase was mixed with 990 µl organic solvents (contained 4 mM pNP and 24 mM methyl esters). In a 2 ml reaction tube, the reaction was carried out for 10 min at 150 rpm. Enzymes precipitation were done by centrifuged the tube at 11,900 g for 3 min, then 50 µl of the upper layer was removed and mixed with 1 ml of Tris Cl buffer pH7 (0.1 percent triton). UV-Vis spectrophotometry was used to measure the remaining pNP (Genesis 10S UV-Vis, Thermo Scientific). The amount of enzyme that released one mole of p-nitrophenyl ester in 1 min was defined as one enzymatic unit. All of the tests were carried out in triplicate, including the blanks (without the addition of enzymes).

Various methyl esters were used to test substrate specificity (varying fatty acids carbon chain length): laurate (C12), myristate (C14), palmitate (C16), stearate (C18), oleate (C18:1) and linoleate (C18:1) (C18:2). Various organic solvents, such as n-hexane, acetone and acetonitrile, were used to test the solvent specificity.

Shaking at 150 rpm for 10 min in n-hexane with 4 mM p-nitrophenol and 24 mM methyl palmitate as the substrate at various temperatures (35, 40, 45, 50, 55 and 60°C) determined the optimum temperature for transesterification activity. The pure recombinant lipase was pre-incubated at temperatures (50 and 55°C) for 24 h for stability experiments, with residual activity measured every 2 h. At 0 h, the activity level was recorded as 100%.

Bisubstrate kinetics were determined at different concentrations of methyl esters and pNP using n-hexane as a solvent at the optimal temperature of each lipase. The concentrations of pNP ranged from 0.6 to 4 mM and methyl ester from 3 to 24 mM. The K_m and V_max values were calculated from the Lineweaver–Burk linear regression plot. The k_cat values were calculated by dividing the V_max by the lipase concentration of the reaction mixture.

After 30 min incubation of pure recombinant lipase in the presence of 5 mM ZnCl₂, NiCl₂, FeCl₃ and CaCl₂, the residual activities were measured to investigated the influence of metal ions on transesterification activity. The activity obtained in the absence of metal ions was taken to be 100%.

Homology modeling was used to construct 3D protein structures using AlphaFold2 (https://colab.research.google.com/github/sokrypton/ColabFold/blob/main/AlphaFold2.ipynb) [22]. Validation of the resulting structure based on the Ramachandran plot using MolProbity [23]. Protein minimization was not performed to prevent amino acid residues from being present in the dissallowed area on the basis of the Ramachandran plot [24]. PubChem data were used to create the ligand structure (https://pubchem.ncbi.nlm.nih.gov/) [25]. The ligand was then minimized by using Orca [26] and recorded as PDB with the Avogadro software [27]. Flexible docking was performed using Autodock Vina [28]. Three catalytic and two oxyanion holes residues were set as flexible, while enzymes without these five flexible residues were set as rigid receptors. Preparation for docking and grid determination were carried out by Autodock MGLTools-1.5.6 [29]. Polar hydrogen and Kollman charges were added to the enzymes structure, while Gastaiger charges were computed for the ligand. Grid dimension covered conserved region and oxyanion hole. Grid spacing was kept at 1 Å. Docking visualization using Pymol and Ligplot [30,31]. PLIP server is used to analyze protein–ligand interactions (https://plip-tool.biotec.tu-dresden.de/plip-web/plip/index) [32].

The possible binding pockets for the 3D structure of Lk2 and Lk3 were predicted using Castp’s online server (http://sts.bioe.uic.edu/castp/calculation.html). The CASTp algorithm calculates the size and volume of projected pockets [33].

Solvent preference was confirmed by dynamic molecular simulation using structure minimization in various organic solvents. This is accomplished by defining minimization using the implicit mode (implicit solvent calculations Born generalized parallel with NAMD) [34]. The GBIS model treats polar solvent as a dielectric continuum and screens electrostatic interactions between solute atoms as a result. A total of 40,000 minimization steps were performed with VMD plugin QwikMD [34]. 3D protein structure was dissolved and minimize on various organic solvents (n-hexane, acetone and acetonitrile). The adjustment of the solvent dielectric constant was carried out manually in the configuration file.

The SCooP algorithm, a Gibbs–Helmholtz equation-based program that predicts protein stability assuming monomeric proteins and a two-state folding transition, was used to derive temperature-dependent protein stability predictions (http://babylone.ulb.ac.be/SCooP) [35]. Protein stability prediction is also analyzed by calculated hydrophobic cluster, hydrogen bond and salt bridge using ProteinTools website (https://proteintools.uni-bayreuth.de.) [36].

Binding potential prediction analysis for protein–metal ion were performed by uploaded .pdb file to the Metal Ion Binding servers (http://bioinfo.cmu.edu.tw/MIB/) [37].

Heterologous expression of lipase LK2 and LK3 genes were carried out by induction with 1 mM IPTG. The proteins were fused with his tag to assist for purification. Cells were lysed by thermolysis method following addition of 0.1% SDS to obtain soluble protein. Following SDS-PAGE analysis, Lk2 and Lk3 were expressed at size around 32 and 31 kDa, respectively (Figure 1A). The proteins still showed lipolytic activity following zymography using naphtyl laurate as substrate (Figure 1B).

IMAC Ni NTA purifications was used to purify the proteins. The purified proteins still shows lipolytic activity. The specific activity of purified Lk2 and Lk3 increased 13 and 12 times compared with that of the crude extracts, respectively (Table 1). Lk2 exhibit higher activity compared with that of Lk3.

Lipases are enzymes that are unique to a certain sort of substrate, such as carbon length [38] or a substrate with double bonds in a specific location [39]. The oxyanion hole of P. stutzeri lipase (Lk4) was mutated at H110F, which resulted in a shift toward a substrate with a longer carbon chain [40]. Lk2 and Lk3 activity’s were investigated on various methyl ester (C12-C18) and C18 containing double bonds (C18:1, C18:2). The results showed that Lk2 activity was higher than Lk3 activity in each of the substrates tested (C12-C18, C18:1, C18:2). Furthermore, Lk2 preferred C18:1 methyl oleate and Lk3 has similar preference for a range of substrates (Figure 2).

Lk2 and Lk3 are high homology with percent identical at 89%. Characterization of the structural models by alphafold2 showed that the closest homology to both lipases was the open lid structure of P. aeruginosa lipase with PDB ID: 1ex9 (Figure 3).

Both enzymes contains same catalytic triad, however, the geometry and catalytic pocket might different. Although the architecture of the catalytic triad is substantially conserved, the great diversity in the catalytic pocket region may result in varied substrate specificity [41].

Further characterization based on computational analysis showed that catalytic pocket volume of Lk2 is larger compared with that of Lk3 (Figure 4). This might due to the activity of Lk2 is higher compared with that of Lk3.

On longer carbon chain of substrated up to C18, the activity of the enzyme tends to be more higher compared with that of Lk3. Moreover docking analysis result appeared in agreement with experimental data (Table 2).

Based on affinity energy and hydrogen bond interaction between substrate and the enzymes showed that Lk2 appeared to have the best interaction with methyl oleate (C18:1) as ligand. Eventhough, the affinity energy of the C18:1 is slightly lower compared with that of C18:2, hydrogen interaction occurred with Met¹⁹ and S⁸⁵ on C18:1 while on C18:2, the hydrogen interaction only occurred with S⁸⁵. Met¹⁹ and S⁸⁵ are known as one of oxyanion hole and catalytic triad on Lk2, respectively [19]. Comparison between interaction of C18:1 and C18:2 to Lk2 appeared that methyl oleate interacted with Met¹⁹ and Ser⁸⁵, while methyl linoleate only interacted with Ser⁸⁵ (Table 2). The interaction distance between methyl oleate and Ser⁸⁵ is longer compared with that of methyl linoleate to Ser⁸⁵ (Figure 5A). Meanwhile for Lk3, C18:2 was appeared as best ligand (Table 2). Eventhough, the affinity energy of C18 is slightly higher compared with that of C18:2; however, hydrogen bond distance between C18:2 to Ser⁷⁸ (0.16 Å) is shorter compared with that of the C18 to Ser⁷⁸ (0.18 Å) (Figure 5B). Ser⁷⁸ is one of catalytic triad on Lk3. Close orientation of catalytic triad to the substrate might cause better interaction and hence increase the activity. All of the computational results are in agreement with the experiment.

Some lipases’ activity is known to be stable in organic solvents [42]. The transesterification activity of Lk2 and Lk3 was determined using an organic solvent variant. On nonpolar liquids, both enzymes were more active (Figure 6).

The highest activity was on n-hexane. The relative activity was decreased up to 80% on acetonitrile compared with that of n-hexane, moreover on acetone the enzymes exhibited 50% activity. Similar result was reported for Thermomyces lanuginosus lipase [43]. Some lipases were reported to lose its activity on acetone [21]. The stability of some enzymes on organic solvent was considered to have positive correlation with thermal stability [44]. At high temperatures, the structure of lipase becomes more compact and is retained on native folding in non-polar solvents [45]. Study on the lid movements of P. stutzeri lipase (LipC) in water and THF showed that the lid opened wider on THF resulting of catalytic residue of serine exposed on medium [46]. Moreover, molecular dynamics simulation of LipMNK showed unfolded lipase structure on acetonitrile concentration at 80–100% [47]. The highest activity of Lk2 and Lk3 on n-hexane might due to the structure of the enzymes were keep on native folding, while on acetonitrile the structure of enzymes might be unfolded. To probe the above possibilities molecular dynamic simulation was carried out on both Lk2 and Lk3. The result showed that protein–solvent interactions were much stronger on more polar solvents (Figure 7A) in both Lk2 and Lk3 resulting on longer of radius gyration of proteins (Figure 7B).

Better interaction of solvent to protein might cause denaturation or conformation change of protein and hence reduce the activity. Previous research showed that the compact 3D structure of proteins in non-polar solvents maintained the conformation of active site [48]. The hydrogen bonding between OG (Ser)-NE2 (His) and OD2 (Asp)-ND1 (His), according to the lipase catalytic mechanism, was critical for the transition state’s stabilization [49]. On Lk2, hydrogen bond distance between OG (Ser⁸⁵)-NE2 (His²⁴⁶) and OD2 (Asp²³¹)-ND1 (His²⁴⁶) is closer in n-hexane (4.8 and 6.1 Å) compared with both in acetone and acetonitrile (5 and 6.2 Å) (Figure 8). This suggested that in n-hexane the enzymes maintained the conformation of the catalytic center as native structure.

Optimum temperature of Lk2 and Lk3 were assayed using methyl palmitate as substrate. The reactions were measured at variation of temperature range from 35 to 60°C. Lk2 and Lk3 showed temperature optimum at 50 and 55°C, respectively (Figure 9A). Most of lipases from thermophilic or thermotolerant bacteria appeared optimum lipolytic activity range at 50–70°C [15,50,51].

Lk2 and Lk3 were tested for thermostability at 50 and 55°C, respectively. The reaction was carried out using n-hexane with three different substrates (methyl laurate, methyl oleate and methyl linoleate). After 2 h of incubation at 50°C, Lk2 activity increased by 1.2- and 1.6-fold each with methyl linoleate and methyl oleate. There is no increasing activity when Lk2 incubated at 55°C. After 24 h of incubation at 50 and 55°C, the highest Lk2 activity was observed on the methyl laurate substrate in comparison with methyl oleate and methyl linoleate (Figure 9B). Meanwhile, after 2 h incubation at 50 and 55°C, Lk3 activity increased 1.2- and 1.6-fold with methyl linoleate. After a 24 h incubation period, the highest activity of Lk3 remained approximately 80% with methyl linoleate (Figure 9C).

Further characterization on thermal stability of the enzymes based on thermal melting prediction analysis [35] showed that Lk3 exhibited a higher T_m value (72.6°C) compared with that of Lk2 (70.6°C) (Table 3).

There are many factors influenced on thermostability of protein such as hydrophobic cluster, salt bridge and hydrogen interactions [52–54]. Structure prediction analysis of Lk2 and Lk3 showed that Lk3 contains more hydrophobic cluster and hydrogen bond interaction compared with that of Lk2, respectively (Table 4). All of the in silico data is agreement with the experiment that Lk3 is more thermostable compared with that of Lk2.

The constant Michaelis (K_m) and the V_max were determined from the Lineweaver–Burk double reciprocal plot. These transesterification kinetics were bisubstrate between methyl oleate and p-nitrophenol and also methyl linoleate and p-nitrophenol (Table 5).

Kinetics mechanism for Lk2 and Lk3 that were categorized as ping-pong double displacement was showed from the plot in Figure 10. Ping-pong (or double displacement) reactions have been highlighted by the formation of parallel lines on a double reciprocal plot [55,56].

The Km values for the Bulkholderia cepacia lipase were 580 mM with triolein and ethanol as substrates. Commercial lipase Novozyme 435 displayed K_m values of 29 mM with waste cooking oil and ethanol as substrates. The K_m and V_max values for Candida antartica lipase A using soybean oil and methanol as substrates were 481 mM and 68.5 U/min. All these enzymes also have ping-pong Bi–Bi mechanisms as kinetics mechanism [57–59].

The activity of most enzyme is influenced by the present of metal ion, some of the ions increase the activity [60], some of them inhibit the activity [61]. Lipase family I.1 are known as Ca²⁺ dependent enzyme [3], the presence of Ca²⁺ enhanced the enzyme’s activity considerably. The enzymes contain a conserved region to bind Ca²⁺. In the presence of a variety of metal ions, the enzymes were studied to see how they affected Lk2 and Lk3 activity. Ca²⁺ and Zn²⁺ were shown to inhibit Lk2, while the present of Fe³⁺ and Ni²⁺ increased the activity of the enzyme (Figure 11). The same effect was also exhibited on the activity of Lk3 except for the present of Zn²⁺. The activity of both lipases is slightly affected by EDTA added at a concentration of 1 and 5 mM. Activation and deactivation of both lipases were dose dependent. All metal ion variations and EDTA inhibited lipase activity at a concentration of 10 mM. Increasing the activity of same lipases on the present of Fe³⁺ were reported on Geobacillus sp. TW1 lipase [62] and lipase from P. aeruginosa [63]. Previous report proposes that redox-inert metal ions tend to stabilize negative charge on the enzyme and activating substrate, while the redox active ions might play a role as Lewis acids and as redox centers [64].

Deactivating of Lk2 and Lk3 activities in the present of Ca²⁺ were surprising since Lk2 and Lk3 were belonged to Family I.1 known as Ca²⁺-dependent enzyme. In the present of Ca²⁺, interaction between Ca²⁺ and amino acid residues close to histidine (one of triad catalytic) contributed to keep His at proper orientation in the catalytic center (His loop) and hence activating the enzyme [65]. Conformation or orientation of triad catalytic residues on active center of P. aeruginosa lipase was analysed. Pseudomonas aeruginosa lipase was used as model enzyme from family I.1 lipase. The result showed that the distance of Ser⁸⁵-His²⁵¹ and His²⁵¹-Asp²²⁹ are 2.3 and 3.2, respectively. Moreover, the distance between Ca²⁺-amino acid residues is within 2.2 and 2.4 Å (Figure 12). Analysis on the interaction between Ca²⁺-amino acid residues on Lk2 and Lk3 showed that the interaction distance was longer compared with that of interaction on P. aeruginosa lipase (Figure 12). The phenomenon affected orientation of Histidine on active center becoming improper (Figure 13). The distance between Ser⁸⁵-His²⁵³ and His²⁵³-Asp²³¹ in Lk2 were 4.2 and 3.9 Å, respectively. Moreover the distance between Ser⁷⁸-His²⁴⁶ and His²⁴⁶-Asp²²⁴ on Lk3 were 3.0 and 4.5 Å, respectively. Orientation shift on catalytic residue of Lk2 and Lk3 caused dramatically decreased on the activity of Lk2 and Lk3. Several previous studies of lipase from metagenomics isolation appeared to have different properties from their related family [66–69]. The above data suggested that Lk2 and Lk3 are unique or novel lipase on family I.1.

Transesterification activity of Lk2 and Lk3 on various organic solvents was successfully characterized. Lk2 exhibited higher activity compared with that of Lk3 in various carbon length, C12-C18, C18:1 and C18:2. 3D structure prediction of Lk2 contained larger catalytic pocket compared with that of Lk3. Most preference substrate of Lk2 was methyl oleate (C18:1), while for Lk3, the highest activity was appeared on methyl linoleate (C18:2). The preference of substrate on Lk2 and Lk3 were confirmed by computational analysis. The activity of both Lk2 and Lk3 was preferentially increased when a nonpolar organic solvent was present. The enzyme–solvent interaction in n-hexane was weaker than in acetone or acetonitrile. Lk2 was less thermostable compared with that of Lk3. The folding parameter of Lk3 showed that the enzyme is more compact. The activity of both enzymes were activated by the present of Fe³⁺; however, in the present of Ca²⁺ the activity were inhibited. This is contradictive for lipase from family I.1, known as Ca²⁺ dependent enzyme, suggesting that Lk2 and Lk3 are novel lipase on family I.1.

Lipase gene LK2 and LK3 available in GenBank (https://www.ncbi.nlm.nih.gov/) with accession number of KP204884 and KP204885. All data generated or analyzed during the present study are included in this article and supplementary file.

The authors declare that there are no competing interests associated with the manuscript.

This work was supported by SIMLITABMAS PDUPT research project program, Ministry of Research, Technology and Higher Education [grant number 5E/IT1.C02/TA.00/2022] and SIMLITABMAS PDD doctoral research project program, Ministry of Research, Technology and Higher Education [grant number 2/E1/KPPTNBH/2021].

Titin Haryati: Conceptualization, Data curation, Software, Formal analysis, Funding acquisition, Investigation, Visualization, Methodology, Writing—original draft, Writing—review & editing. Made Puspasari Widhiastuty: Conceptualization, Data curation, Formal analysis, Supervision, Validation, Investigation, Project administration, Writing—review & editing. Fida Madayanti Warganegara: Conceptualization, Resources, Data curation, Formal analysis, Supervision, Funding acquisition, Validation, Investigation, Methodology, Project administration, Writing—review & editing. Akhmaloka Akhmaloka: Conceptualization, Resources, Data curation, Formal analysis, Supervision, Funding acquisition, Validation, Investigation, Methodology, Writing—original draft, Project administration, Writing—review & editing.

EDTA

ethylenediaminetetraacetate

IPTG

isopropyl β-D-1-thiogalactopyranoside

NAMD

Not (just) Another Molecular Dynamics

PAGE

polyacrylamide gel electrophoresis

PBS

phosphate-buffered saline

SDS

sodium dodecyl sulfate