To address the global challenge of waste oil recycling and support the enzymatic production of biodiesel—an environmentally friendly alternative to conventional diesel—we engineered the yeast Pichia pastoris (P. pastoris)as a robust cellular factory for high-level production of engineered lipases with enhanced tolerance and catalytic activity. Recognizing that natural microbial lipases often lack the stability and efficiency required for industrial-scale conversion, we employed a systematic protein and genetic engineering approach to optimize three key elements: lipase, promoter, and signal peptide. After selecting the best part, we combined them into a complete lipase production and secretion system, thereby obtaining a strain with the highest expression level and enzyme activity, which helps to achieve efficient production of lipase and lays a solid foundation for its application in biodiesel conversion.

Firstly, three promising lipases—CALB (from Candida antarctica), GGL (from Galactomyces geotrichum Y05), and LipZ01 (from oil-contaminated soil)—were selected for their distinct advantages in activity and stability (Table 1).

Table 1 Three lipase used in this project.

After the company synthesized three lipases that had been optimized for codons, we inserted them into a P. pastoris expression vector pPICZαA, which includes the methanol-inducible AOX1 promoter, the α-factor secretion signal, and a C-terminal His-tag, followed by the AOX1 terminator. The Zeocin resistance gene enables selection in transformants (Figure 1A). We firstly amplified the pPICZαA and three lipase gene, followed by agarose gel electrophoresis. The results showed the successful amplification of pPICZαA vector (3476 bp), CALB (954 bp), GGL (1692 bp), and LipZ01 (1428 bp)—with bands corresponding to their expected sizes under UV illumination (Figure 1B-C).

Figure 1 The amplification results of the DNA fragments.

(A) Plasmid diagram. (B-C) Agarose gel electrophoresis result.

Following the amplification of the lipase genes and plasmid vector, the transformation plates confirmed efficient plasmid introduction into E. coli DH5α competent cell (Figure 2A). Colony PCR analysis further validated the presence of the correct insert sizes, with distinct bands corresponding to CALB (1443 bp), GGL (2179 bp), and LipZ01 (1912 bp) observed across multiple clones (Figure 2B). Sequencing results confirmed the precise integration of each expression cassette (Figure 2C). These results collectively demonstrate the accurate assembly of all three plasmids.

Figure 2 Construction result of three plasmids.
(A) Transformation plates. (B) Colony PCR result. (C)Sequencing result.

The recombinant plasmids pPICZαA-CALB, pPICZαA-GGL, and pPICZαA-LipZ01 were successfully extracted and linearized by SacI enzyme prior to transformation, as confirmed by agarose gel electrophoresis (Figure 3A). Linearizing the plasmid is a critical step because it exposes homologous ends that facilitate targeted integration into the P. pastoris genome via homologous recombination, significantly increasing transformation efficiency compared to circular plasmids.

Next, we used the lithium chloride transformation method to transfer the linearized plasmid into P. pastoris. This method can generate active yeast cells with permeable cell membranes, enabling the linearized plasmid to smoothly enter the cells. For the pPICZαA vector, the DNA that has been linearized will specifically integrate into the yeast genome through homologous recombination at the AOX1 locus in the yeast genome. This process ensures that the target gene can be stably expressed under the regulation of the inducible AOX1 promoter.

As shown in Figure 3B, the linearized plasmids were individually transformed into P. pastoris strains, with successful transformants growing on selective YPD plate. The yeast colony PCR results further verified the correct genomic integration of each lipase gene, showing specific amplification products of 954 bp (CALB), 1692 bp (GGL), and 1428 bp (LipZ01) (Figure 3C). These results collectively demonstrate the successful construction of three recombinant P. pastoris strains capable of expressing the target lipases, providing a solid foundation for subsequent protein expression and functional studies.

Figure 3 Construction result of P. pastoris containing lipase.

(A) Plasmid Linearization result. (B) Transformation plates. (C) Yeast colony PCR result.

Subsequently, we inoculated the yeast transformants containing the CALB gene and activated them in the BMGY medium. Then, using the BMMY medium, we began to induce the secretion of CALB. Figure 4 presented the SDS-PAGE analysis of recombinant lipase secretion in P. pastoris after 72-hour fermentation. A distinct protein band at approximately 44 kDa, corresponding to the expected molecular weight of CALB (43.8 kDa), was clearly observed in the culture supernatant of all three tested strains at 48 h and 72 h, confirming successful extracellular secretion (Figure 4A). In contrast, no visible bands were detected at the predicted molecular weights for GGL (63.1 kDa) or LipZ01 (51.1 kDa) in any strain or time point (Figure 4B-C).

The successful secretion of CALB can be attributed to its compatibility with the α-factor secretion signal and efficient folding in the secretory pathway. The absence of GGL and LipZ01 bands suggests potential issues such as inefficient signal peptide recognition, protein misfolding leading to endoplasmic reticulum retention, or intracellular degradation. These results indicate that while the expression system supports CALB secretion, further optimization of the secretion signal or co-expression of chaperones may be necessary for efficient production of GGL and LipZ01.

Figure 4 SDS-PAGE results of (A) CALB, (B)GGL, and (C) LipZ01 secretion.

We found that the purity of the lipase secreted into the supernatant was not high enough. To avoid affecting the subsequent quantitative analysis of the lipase and the enzyme activity test, we carried out nickel column purification on it. Since the crude supernatant contained impurities that could interfere with accurate quantification and activity assays, we concentrated it using an ultrafiltration device and exchanged the buffer to PBS (pH 7.4) to optimize binding efficiency. Following washing and elution steps, a distinct band at approximately 44 kDa, corresponding to CALB, was clearly observed in the elution fractions (Figure 5, E1 and E2), confirming successful purification of the target protein to high purity. In contrast, no specific bands were detected at the expected molecular weights for GGL (63.1 kDa) or LipZ01 (51.1 kDa) in their respective elution lanes, despite applying the same concentration and purification procedure. This result is consistent with the earlier secretion data and further confirms that these two lipases were not successfully secreted in a soluble, His-tagged form, or were present at levels below the detection limit.

Figure 5 Purification results of three lipase.

Next, we use the BCA method to determine the concentration of the lipase. The BCA method relies on the reduction of Cu²⁺ to Cu⁺ by peptide bonds in an alkaline medium, with the resulting Cu⁺ ions forming a purple complex with BCA that can be measured at 562 nm. A series of bovine serum albumin (BSA) standards (25-500 μg/mL) were used to establish a linear calibration curve, which exhibited excellent linearity with the equation Y = 0.0009982*X + 0.1326 and a coefficient of determination (R²) of 0.9934 (Figure 6). The high R² value indicateed a strong correlation between absorbance and protein concentration, validating the reliability of this standard curve for subsequent quantification of lipase samples.

Figure 6 The BSA standard curve results of the BCA method

This standardized protocol was applied to determine the concentration of purified lipases in the culture supernatant, enabling accurate comparison of expression levels among CALB, GGL, and LipZ01 under different experimental conditions. The BCA assay results quantitatively confirmed the successful secretion of recombinant CALB, which achieved a concentration of 68.21 μg/mL, significantly higher than the negative control (0.96 μg/mL). In contrast, GGL and LipZ01 showed only minimal concentrations (4.89 and 1.61 μg/mL, respectively), consistent with earlier SDS-PAGE and purification data indicating their secretion failure (Figure 7). These results validated the effectiveness of the expression and purification system for CALB, while underscoring the need for further optimization to enable detectable secretion of GGL and LipZ01.

Figure 7 The quantitative results of the concentration of purified CALB.

We determined the activity of the lipase using the p-nitrophenol method. This assay is based on the hydrolysis of p-Nitrophenyl esters by lipase, releasing yellow-colored 4-Nitrophenol (4-NP) that absorbs at 450 nm. A series of 4-NP standards (0–0.1 mmol/L) were prepared, and their absorbance values were measured to establish a linear calibration curve. The fitted equation Y = 5.924x + 0.01800 exhibited excellent linearity (R² = 0.9857), confirming a strong correlation between absorbance and 4-NP concentration (Figure 8). This reliable standard curve enables accurate quantification of lipase activity in subsequent enzymatic assays.

Figure 8 Standard curve of 4-Nitrophenol

Finally, the lipase activity of the three recombinant strains was measured. The CALB-expressing strain demonstrated the highest enzymatic activity at 28.59 ± 0.37 U/mL, which was approximately 3.5-fold higher than the activities of GGL (8.15 ± 0.14 U/mL) and LipZ01 (8.41 ± 0.09 U/mL). Statistical analysis (​​​​p < 0.0001) confirmed that CALB activity was significantly higher than both GGL and LipZ01, while no significant difference (ns) was observed between GGL and LipZ01. The CALB-expressing strain showed significantly higher activity than the negative control (pPICZαA, 5.99 ± 0.70 U/mL), validating the successful expression of it (Figure 9). These results indicate that CALB exhibits superior catalytic efficiency under the tested conditions.

Figure 9 Lipase activity test result.

Summary:

Based on the successful plasmid construction and yeast transformation of lipases from three different sources, we obtained the corresponding recombinant strains. Subsequent fermentation and analysis of secretion levels and enzymatic activity revealed that only CALB was successfully secreted extracellularly, demonstrating measurable enzymatic activity (reaching 28.59 ± 0.37 U/mL). Therefore, in the subsequent optimization phase, we will focus on CALB, specifically targeting the enhancement of its expression and secretion efficiency through promoter and signal peptide optimization.

The choice and optimization of promoters in P. pastoris are critical for achieving high-level lipase expression, as promoters directly control transcriptional initiation and strength [4]. Signal peptides are essential for directing the secretion of heterologous lipases into the extracellular space, simplifying downstream purification and enhancing yield [5]. Thus, we further screened strong inducible promoters (AOX713, FDH1, FLD1) and efficient secretion signal peptides (0030α-factor, SWP1-α-factor, KRE1-α-factor), using the classic AOX1 promoter and α-factor signal peptide as the control, respectively, to maximize transcriptional activation and secretory efficiency (Table 2).

Table 2 Promoters and signal peptides used in this project.

Table 3 Signal peptides used in this project.

For the optimization of the promoter, we retained the α-factor signal peptide on the pPIC9K vector and replaced the original promoter with three candidates: AOX713, FDH1, and FLD1 (Figure 10A). For the optimization of the signal peptide, we retained the AOX1 promoter on the pPIC9K vector and replaced the signal peptide with three candidates: 0030α-factor, KRE1-α-factor, and SWP1-α-factor (Figure 10B). As shown in the agarose gel electrophoresis results, the DNA bands for the linearized vectors and the inserted fragments matched the expected sizes, verifying the successful amplification of all fragments.

Figure 10 The amplification results of the DNA fragments.

(A)Three promoters. (B) Three signal peptides.

Following the amplification of the different promoters, signal peptides, and plasmid vector, we combined them through homologous recombination. the transformation plates confirmed efficient plasmid introduction into E. coli DH5α competent cell (Figure 11). Colony PCR analysis demonstrated distinct bands of expected sizes: 2188 bp for AOX713-CALB, 2154 bp for FDH1-CALB, 2169 bp for FLD1-CALB, 2237 bp for AOX1-0030-α-CALB, 2246 bp for AOX1-SWP1-α-CALB, and 2243 bp for AOX1-KRE1-α-CALB. Sequencing results further verified the precise integration of each expression cassette with correct promoter-signal peptide-CALB organization. These results collectively confirm the accurate assembly of all six plasmid constructs, providing a reliable foundation for subsequent yeast transformation and protein expression studies.

Figure 11 The colony PCR and sequencing results of plasmids.

The six recombinant plasmids were then successfully extracted and linearized by SalI enzyme before transformation, as confirmed by agarose gel electrophoresis. Each plasmid displayed a single, clear band after single enzyme digestion, corresponding to the expected size of the linearized vector (Figure 12).

Figure 12 The linearization results of the plasmids.

After enzyme digestion, the linearized plasmids were transformed into P. pastoris competent cell, respectively. Figure 13A showed that transformants of P. pastoris growing on selective plates for six different constructs. Further colony PCR analysis confirmed genomic integration, with distinct bands observed at the expected size of 954 bp for all constructs when compared to the DNA ladder. These results demonstrated the successful integration of the CALB gene expression cassettes into the yeast genome, validating the effectiveness of the transformation procedure and providing a reliable foundation for subsequent protein expression studies.

Figure 13 Construction result of P. pastoris containing lipase.

(A) Transformation plates. (B) Yeast colony PCR results.

Next, we conducted fermentation on each of these six strains. For the supernatant from the culture medium that was fermented for 72 hours in the BMMY medium, we performed SDS-PAGE to determine the secretion of CALB. The SDS-PAGE results showed that all recombinant strains exhibited distinct protein bands at approximately 43.8 kDa after 72 hours of induction, which were absent in the pre-induction (0 h) samples and the negative control (NC, pPIC9k empty vector). Notably, the band intensities varied among strains, with pPIC9k-FDH1-α-CALB and the pPIC9k-AOX1-0030-α-CALB strains demonstrating particularly strong secretion. These results confirmed the successful secretion of recombinant CALB by all six engineered strains, with varying expression levels observed under different promoter-signal peptide combinations.

Considering that the CALB purity in the fermentation supernatant of these six strains was relatively high, we did not perform any additional purification and directly used the BCA kit to determine the concentration. Among promoters, FDH1 yielded the highest CALB concentration (93.82 ± 4.16 μg/mL), significantly outperforming AOX1 (68.21 μg/mL) and FLD1 (61.25 μg/mL). For signal peptides, the 0030-α variant achieved the highest secretion (102.37 μg/mL), markedly surpassing the standard α-factor (Figure 14B). These quantitative data align with band intensities observed in SDS-PAGE, collectively demonstrating that both promoter strength and signal peptide efficiency critically influence CALB secretion (Figure 14A). The optimal combination of the FDH1 promoter and 0030-α factor signal peptide provides an effective strategy for enhancing recombinant protein production in P. pastoris.

Figure 14 SDS-PAGE verification and quantification of CALB under different promoter regulation and signal peptide regulation.

Finally, we measured the lipase activity of the six recombinant strains using the p-nitrophenol method.  For the promoters, the FDH1 promoter yielded the highest enzymatic activity (31.68 U/mL), significantly outperforming AOX1 and FLD1 promoters (​​P ≤ 0.01). The secretion levels of the other two promoters (AOX713 and FLD1) were both lower than that of the AOX1 promoter (Figure 15A). For the signal peptides, the 0030-α-factor signal peptide led to the highest activity (31.25 U/mL), while the differences between the other two signal peptides (SWP1-α-factor and KRE1-α-factor) and the α-factor signal peptide were not statistically significant (ns).

Figure 15 Lipase activity test results of (A) different promoters and (B) signal peptide.

Summary:

In this part, we systematically evaluated the performance of three promoters (AOX713, FDH1, FLD1) and three signal peptides (0030α-factor, KRE1-α-factor, SWP1-α-factor) for the expression and secretion of CALB in P. pastoris. The recombinant plasmids were successfully constructed and linearized, followed by efficient transformation into P. pastoris, as confirmed by colony growth and PCR verification. Fermentation in BMMY medium and subsequent SDS-PAGE analysis revealed distinct protein bands at approximately 43.8 kDa for all six engineered strains, with varying secretion intensities. Quantitative BCA assays demonstrated that the FDH1 promoter yielded the highest CALB concentration (93.82 ± 4.16 μg/mL), while the 0030α-factor signal peptide achieved the highest secretion level (102.37 μg/mL). Enzymatic activity analysis further confirmed that the FDH1 promoter (31.68 U/mL) and the 0030α-factor signal peptide (31.25 U/mL) delivered superior lipase activity compared to other variants.

Based on the comprehensive evaluation of protein secretion levels, concentration, and enzymatic activity, the ​​FDH1 promoter​​ and ​​0030α-factor signal peptide​​ were identified as the most effective components. Consequently, this optimal combination will be utilized in all subsequent experimental phases to maximize recombinant CALB production.

The best-performing combinations were finally integrated into expression cassettes for recombinant strain construction. As shown in Figure 16A, the expression vector pPIC9K was engineered to contain the optimal FDH1 promoter and 0030α-factor signal peptide fused with the CALB gene. The agarose gel electrophoresis results (Figure 16B) confirmed successful amplification of both the vector backbone (Amp-FDH1, ~4402 bp) and the insert fragment (0030α-CALB-Kan, ~5954 bp), with bands matching expected sizes. Colony PCR verification (Figure 16C) demonstrated successful transformation of E. coli DH5α competent cell, showing specific bands of approximately 2495 bp corresponding to the target expression cassette in multiple transformants. These results validated the accurate assembly of the optimized expression system combining the FDH1 promoter and 0030α-factor signal peptide.

Figure 16 The amplification results of the DNA fragments.

(A) Plasmid diagram. (B) Agarose gel electrophoresis result. (C) Colony PCR and sequencing result.

We then extracted and linearized the correct plasmids for yeast transformation. The agarose gel electrophoresis (Figure 17 A) confirmed successful linearization, showing a single clear band for the linearized plasmid (lane 2) compared to the negative control (lane 1). The transformation plates (Figure 17 B) exhibited robust colony growth, indicating efficient plasmid introduction into P. pastoris. Furthermore, yeast colony PCR analysis (Figure 17 C) validated the genomic integration, revealing a distinct band at approximately 954 bp corresponding to the FDH1-0030α-CALB expression cassette. These results collectively confirm the successful construction of the recombinant P. pastoris strain.

Figure 17 Construction result of P. pastoris containing lipase.

(A) Plasmid Linearization result. (B) Transformation plates. (C) Yeast colony PCR result.

Next, we conducted fermentation on each of these strains (FDH1-0030α-CALB belongs to the experimental group, while FDH1-α-CALB and AOX1-0030α-CALB constitute the control group) and analyzed CALB secretion. The SDS-PAGE results (Figure 18A) clearly showed distinct protein bands at approximately 43.8 kDa in the experimental groups, corresponding to the expected molecular weight of CALB, while absent in the negative control. Quantitative analysis (Figure 18B) demonstrated that the combination of the FDH1 promoter and 0030-α-factor signal peptide yielded the highest secretion level (107.02 ± 3.95 μg/mL), significantly outperforming the other constructs (​​P ≤ 0.01). These results confirmed the synergistic effect of promoter and signal peptide optimization on CALB production.

Figure 18 SDS-PAGE verification of CALB secretion.

(A) SDS-PAGE result. (B) Quantitative results.

Similarly, we conducted an enzyme activity test. The results showed that compared with the negative control pPIC9K (NC), the lipase activity of all the recombinant strains was significantly increased. Among them, the combination of the FDH1 promoter and the 0030-α-factor signal peptide (FDH1-0030-α-factor-CALB) exhibited the highest enzyme activity (33.41 U/mL). Statistical analysis indicated that the enzyme activity of this combination was significantly higher than that of the strains that only optimized the promoter (FDH1-α-factor-CALB, 31.68 U/mL) or only optimized the signal peptide (AOX1-0030-α-factor-CALB, 31.25 U/mL) (P ≤ 0.05), confirming the positive effect of the synergistic optimization of the promoter and signal peptide on enhancing the catalytic efficiency of the lipase (Figure 19).

Figure 19 Lipase activity test results.

Finally, we investigated the temperature and pH values of the reaction to determine the optimal catalytic conditions of CALB. As shown in Figure 20A, the temperature optimization experiment indicated that CALB exhibited the highest enzymatic activity (approximately 40 U/mL) at 45°C, while the activity significantly decreased at low temperatures (25°C) or high temperatures (≥65°C). As shown in Figure 20B, the pH optimization results revealed that the enzyme maintained high activity within the range of 7.5–9.0, with the optimal pH being 8.0. It is noteworthy that CALB could still maintain over 70% relative activity within a wide pH range (6.0–10.0), indicating its good acid-base stability. These data clearly defined the optimal reaction conditions of CALB as 45°C and a neutral to weakly alkaline environment, providing key parameters for its industrial application.

Figure 20 Optimization of (A) reaction temperature and (B) pH for CALB.

Summary

In this project, we successfully established an efficient system for recombinant CALB production in P. pastoris. The optimized gene construct, which combines the FDH1 promoter and the 0030α factor signal peptide of the CALB gene, was successfully integrated into the yeast genome. The secretion level of CALB reached 107.02 ± 3.95 μg/mL, and the enzyme activity was 33.41 U/mL, which were 1.84 times and 1.17 times higher than those of the original strain (AOX1-α factor-CALB, secretion: 68.21 μg/mL, enzyme activity: 28.59 U/mL), respectively. In addition, we determined that the optimal catalytic conditions for CALB were 45°C and a pH value of 8.0. Under these conditions, the enzyme activity could reach up to 44.96 U/mL, and the enzyme showed high stability within a wide pH range (6.0-10.0). In conclusion, the collaborative optimization of the FDH1 promoter and the 0030α factor signal peptide is the key strategy for achieving high-yield active CALB in P. pastoris. This recombinant strain and the clearly defined optimal reaction parameters have laid a solid foundation for the industrial application of this biocatalyst.

• About the Failure of GGL and LipZ01 Secretion

In the P. pastoris system, the failure of GGL and LipZ01 to be successfully secreted into the culture supernatant may result from multiple interrelated factors. First, the most fundamental reason likely lies in the inherent characteristics of proteins. GGL (63.1 kDa) and LipZ01 (51.1 kDa) possess larger molecular weights compared to the successfully secreted CALB (43.8 kDa), potentially exhibiting more complex spatial structures. The endoplasmic reticulum secretory pathway in P. pastoris may be unable to effectively recognize or correctly process the proper folding of these heterologous proteins. This leads to protein retention and aggregation within the endoplasmic reticulum, ultimately resulting in degradation by intracellular proteasomes [12].

Second, compatibility between the signal peptide and the target protein is critical. The α-factor signal peptide used in this experiment, while effective for many proteins, exhibits efficiency highly dependent on the N-terminal amino acid sequence of the guided protein. The N-terminal sequences of GGL and LipZ01 may be incompatible with the recognition and cleavage efficiency of the α-factor signal peptide. This incompatibility prevents the precursor proteins from being correctly processed by Golgi proteases, such as Kex2, thereby obstructing subsequent secretory steps [13].

• Effects of Promoters and Signal Peptides on CALB Production

Different combinations of promoters and signal peptides significantly influence CALB production and secretion levels. This primarily stems from their regulatory roles in two distinct yet critical processes: gene transcription and protein transport.

At the transcriptional level, promoter strength directly determines the amount of CALB mRNA transcribed, serving as the primary determinant of expression levels [14]. For instance, the FDH1 promoter may exhibit stronger constitutive or uninhibited transcriptional activity than the traditional AOX1 promoter under the fermentation conditions used in this experiment. This drives greater mRNA synthesis, laying the foundation for high yields.

At the post-translational level, signal peptide efficiency determines whether synthesized CALB protein precursors can be efficiently processed and transported to the extracellular space. Different signal peptides exhibit varying degrees of compatibility with the secretory machinery of P. pastoris, including endoplasmic reticulum transport, protein folding and modification, and vesicular transport systems [15]. An efficient signal peptide ensures the CALB protein precursor is rapidly and accurately translocated into the endoplasmic reticulum lumen, undergoes correct glycosylation and folding, and is ultimately released into the supernatant via exocytosis. Inefficient signal peptides may cause the protein precursor to accumulate in the cytoplasm, misfold, or become trapped in the endoplasmic reticulum, leading to degradation.

Furthermore, the physiological state of the host cell represents a deeper underlying cause. High-intensity transcription and translation driven by strong promoters, coupled with massive protein secretion facilitated by efficient signal peptides, impose substantial metabolic stress on the host cell. If this stress exceeds the cell's tolerance threshold, it can inhibit cell growth and activate quality control systems like endoplasmic reticulum (ER) stress, leading to extensive protein degradation [16]. Therefore, the optimal combination (such as the FDH1 promoter paired with the 0030 α-factor signal peptide) likely strikes the best balance between high expression drive and metabolically tolerable stress, thereby achieving maximum yield.

Based on the current experimental results, we have summarized the future project development plan, covering three key directions: expression system optimization, enzyme immobilization application, and hardware development, in order to further improve this project.

• Expression System Optimization

To address the secretion failure of GGL and LipZ01, we will systematically optimize their expression systems in the future. Our approach will focus on the following aspects. First, rationally design or screen signal peptides to obtain novel ones with higher compatibility with GGL/LipZ01. Second, we will co-express molecular chaperones, such as endoplasmic reticulum oxidation 1(Ero1) and protein disulfide isomerase (PDI), to alleviate endoplasmic reticulum stress and promote proper folding of large-molecular-weight lipases. Concurrently, rational design of the GGL and LipZ01 genes (e.g., reducing hydrophobic clusters, optimizing N-terminal amino acids) will be pursued to enhance their secret ability. Finally, exploration of protease-deficient host strains (e.g., SMD1163) will be pursued to minimize intracellular degradation.

• Preparation of Immobilized Lipase

To enhance lipase stability, reuse efficiency, and reduce application costs, future efforts will focus on developing immobilization techniques for lipase. The plan involves employing covalent bonding methods using carriers such as amino-silica gel and magnetic nanoparticles, immobilizing purified lipase through crosslinking agents like glutaraldehyde [17]. This immobilization technique significantly enhances enzyme stability in organic solvents and at elevated temperatures. Concurrently, we will evaluate encapsulation methods, using sodium alginate-chitosan microspheres or metal-organic framework (MOF) materials to envelop lipase, creating a protective microenvironment particularly suited for complex waste oil substrates [18]. We will systematically evaluate the operational half-life and reuse batches of immobilized enzymes under simulated industrial conditions, aiming to maintain high activity even after continuous use.

Figure 21 Strategies for Lipase Immobilization [19].

• Test of Conversion of Waste Fats

We plan to apply the highly active lipase (CALB) developed in our project to test the conversion of waste oils. Specifically, we will establish a small-scale enzymatic reactor system using restaurant waste oil or gutter oil as feedstock. Under optimized conditions of temperature (~45°C) and pH (8.0), transesterification reactions will be conducted to efficiently produce biodiesel (fatty acid methyl esters). We will focus on evaluating the catalytic efficiency, methanol tolerance, and operational stability of lipases in complex waste oil systems. Additionally, we will explore enhancing enzyme reuse rates through immobilization techniques, providing data support for subsequent pilot-scale upscaling and process integration.

Figure 22 Conversion of Waste Oils Using Immobilized Lipase [20].

Furthermore, we plan to develop a prototype hardware system for enzymatic waste oil conversion (Figure 23). The core component will be a modular reactor featuring a transparent enzyme chamber packed with immobilized lipase carriers, allowing real-time observation of the catalytic process. A stainless-steel pre-filtration mesh at the inlet will ensure the removal of large impurities from waste cooking oil (WCO), while a dedicated glycerol separation layer at the bottom will enable efficient byproduct recovery. The integrated control unit will regulate key parameters such as temperature, flow rate, and reaction time to optimize biodiesel yield. This hardware will serve as a demonstrator for scalable, user-friendly biofuel production, aligning with iGEM's goal of converting synthetic biology designs into tangible engineering solutions.

Figure 23 Hardware system for enzymatic waste oil conversion design in our project.

[1] Zhang, N., Suen, W. C., Windsor, W., Xiao, L., Madison, V., & Zaks, A. (2003). Improving tolerance of Candida antarctica lipase B towards irreversible thermal inactivation through directed evolution. Protein Engineering, 16(8), 599-605.

[2] Yan, J., Yang, J., Xu, L., & Yan, Y. (2007). Gene cloning, overexpression and characterization of a novel organic solvent tolerant and thermostable lipase from Galactomyces geotrichum Y05. Journal of Molecular Catalysis B: Enzymatic, 49(1), 28-35.

[3] Zheng, J., Liu, C., Liu, L., & Jin, Q. (2013). Characterisation of a thermo-alkali-stable lipase from oil-contaminated soil using a metagenomic approach. Systematic and Applied Microbiology, 36(3), 197-204.

[4] Vogl T, Hartner FS, Glieder A. New opportunities by synthetic biology for biopharmaceutical production in Pichia pastoris. Curr Opin Biotechnol. 2013 Dec;24(6):1094-101.

[5] Huang J, Wang Q, Bu W, Chen L, Yang Z, Zheng W, Li Y, Li J. Different construction strategies affected on the physiology of Pichia pastoris strains highly expressed lipase by transcriptional analysis of key genes. Bioengineered. 2019 Dec;10(1):150-161.

[6] Vogl, T., & Glieder, A. (2013). Regulation of Pichia pastoris promoters and its consequences for protein production. New Biotechnology, 30(4), 385-404.

[7] Liu, Q., Li, Y.-h., Tao, L.-f., Yang, J.-y., Zhang, Y.-l., & Cai, M.-h. (2024). Rational design and characterization of enhanced alcohol-inducible synthetic promoters in Pichia pastoris. Applied and Environmental Microbiology, 91(1), e02191-02124. doi:10.1128/aem.02191-24

[8] HU Liyuan, LIU Luyao, BAI Zhonghu,et al.Cofactor specificity engineering and application of formate dehydrogenase from Komagataella phaffii[J].Food and Fermentation Industries,2023,49(13)：9-16.

[9] CREGG JAMES M. Promoter for the Pichia pastoris formaldehyde dehydrogenase gene FLD1. United States Reexamination Certificate First Reexamination US19990345828. 02 Jul 1999.

[10] Lv, X., Zhang, Y., Wang, L., Cui, S., Liu, Y., Li, J., Liu, L. (2024). Expression and antimicrobial activity of the recombinant bovine lactoferricin in Pichia pastoris. Synthetic and Systems Biotechnology, 9(1), 26-32.

[11] Richard Zahrl, Wien (AT); Oezge Ata Aykol, Wien (AT); Diethard Mattanovich,Wien (AT); Brigitte Gasser, Wien (AT)signal peptides for increased protein secretion_patent.US20240141363A1. 02 May 2024.

[12] Gu L, Zhang J, Liu B, et al. High-Level extracellular production of glucose oxidase by recombinant Pichia pastoris using a combined strategy[J]. Applied Biochemistry and Biotechnology, 2015, 175(3): 1429-1447.

[13] Wang X, Wang P, Li W, Zhu C, Fan D. Effect and mechanism of signal peptide and maltose on recombinant type III collagen production in Pichia pastoris. Appl Microbiol Biotechnol. 2023 Jul;107(13):4369-4380.

[14] Cereghino JL, Cregg JM. Heterologous protein expression in the methylotrophic yeast Pichia pastoris. FEMS Microbiol Rev. 2000 Jan;24(1):45-66.

[15] Daly R, Hearn MT. Expression of heterologous proteins in Pichia pastoris: a useful experimental tool in protein engineering and production. J Mol Recognit. 2005 Mar-Apr;18(2):119-38.

[16] Zepeda AB, Figueroa CA, Pessoa A, Farías JG. Free fatty acids reduce metabolic stress and favor a stable production of heterologous proteins in Pichia pastoris. Braz J Microbiol. 2018 Oct-Dec;49(4):856-864.

[17] Vasić K, Knez Ž, Leitgeb M. Immobilization of alcohol dehydrogenase from Saccharomyces cerevisiae onto carboxymethyl dextran-coated magnetic nanoparticles: a novel route for biocatalyst improvement via epoxy activation. Sci Rep. 2020 Nov 10;10(1):19478.

[18] Wróblewska-Krepsztul J, Rydzkowski T, Michalska-Pożoga I, Thakur VK. Biopolymers for Biomedical and Pharmaceutical Applications: Recent Advances and Overview of Alginate Electrospinning. Nanomaterials (Basel). 2019 Mar 10;9(3):404.

[19] Maghraby YR, El-Shabasy RM, Ibrahim AH, Azzazy HME. Enzyme Immobilization Technologies and Industrial Applications. ACS Omega. 2023 Jan 31;8(6):5184-5196.

[20] Binhayeeding, N., Klomklao, S., Prasertsan, P., & Sangkharak, K. (2020). Improvement of biodiesel production using waste cooking oil and applying single and mixed immobilised lipases on polyhydroxyalkanoate. Renewable Energy, 162, 1819-1827.