Boosting Recombinant Bovine Chymosin in Komagataella phaffii via Fusion Protein and Constitutive Promoter Expression

TL;DR

This study improves the production of recombinant bovine chymosin in yeast using a fusion protein and a constitutive promoter, achieving high yields and better pH stability.

Contribution

A methanol-independent, high-yield platform for recombinant bovine chymosin production in Komagataella phaffii using a constitutive promoter and fusion tagging.

Findings

The engineered strain GH1 achieved a volumetric productivity of 105.03 SU/(mL·h), twice that of the inducible strain.

Optimized fermentation conditions increased enzyme activity to 12,000 SU/mL.

The recombinant chymosin showed similar enzymatic properties to native chymosin and broader pH stability (pH 2.0–6.0).

Abstract

Bovine chymosin is key for cheese production, yet its traditional sourcing is unsustainable. While microbial and plant-based alternatives exist, they often cause non-specific proteolysis, leading to bitter flavors in cheese. This study aims to develop a high-yield, methanol-independent platform for recombinant bovine chymosin production by engineering the expression system of Komagataella phaffii through multi-factorial optimization. Initially, the native bovine prochymosin gene (pcw) was codon-optimized (pcm14) and cloned, along with an mCherry-tag construct (clpcm14), into inducible vector pPIC9 for expression in Komagataella phaffii GS115. Screening identified the fusion-tagged strain clp2-91 as the highest producer. Subsequently, the inducible AOX1 promoter in the previously selected clp2-91 strain was replaced with a constitutive GAP promoter, yielding engineered strain GH1. Cultivated in a 3L fermenter, GH1 exhibited a volumetric productivity of 105.03 SU/(mL·h), twice that of inducible strain clp2-91 (53.59 SU/(mL·h)). The further optimization of fermentation conditions (pH 4.0, glucose as carbon source, fed-batch mode) boosted the enzyme activity of GH1 to 12,000 SU/mL. The recombinant chymosin exhibited enzymatic properties similar to those of the native enzyme and, importantly, demonstrated a broader pH stability (pH 2.0–6.0). This study demonstrates an efficient strategy for chymosin expression in K. phaffii, offering insights that may support the future development and optimization of heterologous protein production in this yeast.

Figures

Full text

1. Introduction

Cheese production heavily relies on the enzyme chymosin for milk coagulation. Bovine chymosin (EC 3.4.23.4), an aspartic protease synthesized in the abomasum of newborn calves, is highly valued in the dairy industry for its exceptional specificity in cleaving k-casein at the Phe105–Met106 bond, leading to efficient curd formation without any abnormal taste [1,2,3,4,5,6]. Traditionally, bovine chymosin is obtained from the stomachs of slaughtered calves. This source is insufficient to meet global demand and raises ethical concerns [7]. While alternative sources from plants and microorganisms exist, these alternatives often exhibit undesirable proteolytic activities, compromising cheese quality and taste [8,9,10].

To address these limitations and achieve higher yields of chymosin, recombinant expression technology has been employed for the production of bovine chymosin [11]. The yeast Komagataella phaffii (K. phaffii, formerly Pichia pastoris) emerged as a prominent expression host due to advantages such as high-density growth, post-translational modification capabilities, and Generally Recognized As Safe (GRAS) status, which applies to its use in production processes, although the final recombinant product still requires its own safety assessment [12,13,14,15].

Over the years, various strategies have significantly improved protein production in K. phaffii. Among these, codon optimization modifies coding sequences to align with host-preferred codon usage and tRNA availability, thereby enhancing translation efficiency and expression [16]. This method is widely applied to increase the yield of recombinant enzymes and therapeutic proteins [17]. Espinoza-Molina et al. successfully expressed a codon-optimized bovine chymosin gene in K. phaffii, yielding an enzyme with high milk-clotting activity and demonstrating the method’s effectiveness in this yeast system [18]. Furthermore, fusion tags—such as mCherry-tag, His-tag, and Strep-tag—facilitate protein purification, detection, and high-throughput screening, while also improving solubility and stability [19,20]. In one study, researchers inserted the optimized chymosin gene into a vector containing a C-terminal c-myc epitope and a His-tag (via the pPICZαA vector) for expression in K. phaffii, enabling successful purification using metal-chelating resin [21]. Promoter engineering is equally critical. Although the methanol-inducible AOX1 promoter enables high-level expression, its strict reliance on methanol introduces significant industrial limitations, including safety concerns related to inflammability, increased process control complexity, and regulatory constraints during large-scale operation [22]. Consequently, methanol-free expression strategies, such as the use of constitutive promoters or alternative carbon source regulation, have gained increasing attention. Among these, the constitutive GAP promoter has emerged as a robust and industrially relevant option for achieving stable, high-titer, and induction-free protein production [23,24]. For instance, the full-length camel chymosin gene was synthesized and expressed under the control of the GAP promoter, resulting in a milk-clotting activity of approximately 1412 U/mL and a yield of ~80 mg/L [25]. Numerous studies have reported the application of individual optimization strategies to improve recombinant protein production in Komagataella phaffii, including promoter engineering, gene optimization, and fermentation process control. However, only a limited number of studies have combined multiple genetic and process-level strategies in a integrated manner. As highlighted in recent systematic reviews, while diverse optimization approaches have been developed and successfully applied individually, their integrated implementation to simultaneously enhance enzyme yield and functional stability remains relatively underexplored in the literature [26]. This lack of integrated optimization may constrain the development of enzyme preparations that meet the combined performance requirements for industrial applications. Such an integrated approach may be essential to realize enzyme preparations that simultaneously satisfy high-yield and industrial robustness requirements. Additionally, the stability and functional robustness of the optimized recombinant proteins were not thoroughly evaluated, limiting the assessment of their suitability for large-scale industrial applications.

In this study, we optimized the bovine prochymosin gene (pcw) for codon usage in K. phaffii and fused it to an mCherry-tag to facilitate the monitoring of expression and activity [20,27]. Building upon this strategy, we further replaced the methanol-inducible AOX1 promoter with a constitutive GAP promoter to develop a methanol-free expression system, thereby avoiding the risk of methanol carry-over that could compromise cheese flavor and simplifying process control. We further refined fermentation parameters (pH, carbon source, feeding strategy) and evaluated the resulting recombinant chymosin to assess its similarity in enzymatic properties to the native enzyme. Collectively, this work successfully enhances chymosin production in K. phaffii. More broadly, it provides a systematic and potentially generalizable optimization strategy for the high-yield expression of diverse heterologous proteins in this system. While industrial-scale manufacturing remains a long-term goal that involves additional regulatory and economic considerations, the methodological framework established here is an important step towards overcoming the primary biological production hurdles.

2. Materials and Methods

2.1. Strains, Plasmids, and Chemicals

K. phaffii GS115, pPIC9 vector, and pGAPZα were sourced from Invitrogen (Carlsbad, CA, USA). Escherichia coli Trans1-T1 and TransStartTM Fast Pfu DNA polymerase were purchased from TransGen Biotech (Beijing, China). T4 deoxyribonucleic acid (DNA) ligase and restriction endonucleases were obtained from Thermo Fisher Scientific (Waltham, MA, USA). Endo-β-N-acetylglucosaminidase H (Endo H) was obtained from New England Biolabs (Ipswich, MA, USA). All other chemical reagents used in this study were analytical grade and commercially available.

2.2. Culture Media

E. coli was grown in Luria-Bertani (LB) medium (yeast extract 5.0 g/L, tryptone 10.0 g/L, NaCl 10.0 g/L) at 37 °C. YPD medium (yeast extract 10 g/L, tryptone 20 g/L, glucose 20 g/L) was used to culture yeast at 30 °C. Minimal dextrose (MD) agar medium (2% glucose, 2% agar, 1.34% yeast nitrogen base (without aa), 0.4 μg/mL biotin) was used to screen and selectively culture yeast cells. Buffered glycerol-complex medium (BMGY) (1.0% yeast extract, 2.0% tryptone, 1.0% glycerol, 100 mM potassium phosphate pH 6.0, 1.34% yeast nitrogen base (without aa), 0.4 μg/mL Biotin) was used to culture yeast cells. Buffered methanol-complex medium (BMMY) (1.0% yeast extract, 2.0% peptone, 1.0% pure methanol, 100 mM sodium citrate pH 3.0, 1.34% yeast nitrogen base (without aa), 0.4 μg/mL biotin) was used during the methanol-induced growth period of yeast cells. Basal salts medium (BSM) (potassium dihydrogen phosphate 5.0 g/L, ammonium dihydrogen phosphate 50.0 g/L, calcium sulfate 0.93 g/L, potassium sulfate 18.2 g/L, magnesium sulfate heptahydrate 14.9 g/L, potassium hydroxide 1.5 g/L) supplemented with trace metal solution (PTM1) and biotin was utilized as the liquid culture media for K. phaffi bioreactor experiments.

2.3. Gene Optimization and Vector Construction

The sequence of the bovine prochymosin gene (pcw) (GenBank: FJ768675.1) and cherry gene were optimized in codon sequences using the OptimumGene™ algorithm (GenScript, Piscataway, NJ, USA). The PCR products of optimized genes (pcm14 and clpcm14) were digested with EcoR I and Not I (Thermo Fisher, Scientific Waltham, MA, USA) enzymes and ligated into corresponding sites on the pPIC9 vector. The clpcm14 gene and pGAPZα vector were also ligated with T4 deoxyribonucleic acid (DNA) ligase (Thermo Fisher, Scientific Waltham, MA, USA). The recombinant chymosin genes were cloned downstream of the Saccharomyces cerevisiae α-mating factor secretion signal encoded by the pPIC9 or pGAPZα vectors to direct secretion of chymosin into the culture medium [28,29]. During the secretion process, the α-mating factor signal peptide is cleaved off by membrane-associated proteases. Consequently, the protein secreted into the culture medium no longer contains the α-mating factor sequence. Single clones were selected for restriction enzyme digestion verification and sent to Sangon Biotech (Shanghai, China) for sequence validation. The correctly verified recombinant expression vectors were named as recombinant plasmids pPIC9-pcm14, pPIC9-clpcm14, and pGAPZα-clpcm14 (Table 1).

2.4. Transformation and Selection of Recombinant Yeast

The expression vectors pPIC9-pcm14, pPIC9-clpcm14, and pGAPZα-clpcm14 were linearized with restriction enzymes Bgl II and Avr II (Thermo Fisher Scientific, Waltham, MA, USA). The linearized DNA was then transformed into K. phaffii GS115 via electroporation using a Gene Pulser (Bio-Rad Laboratories, Hercules, CA, USA) with 2 mm gap cuvettes under the following conditions: 2000 V, 200 Ω, 50 μF. For the inducible promoter strains, a two-step screening procedure was employed. For preliminary screening, transformants were cultured on MD plates at 30 °C. After colony formation, single colonies were inoculated into 500 μL of BMGY and grown at 30 °C with shaking at 200 rpm for 48 h. The cells were then harvested through centrifugation at 4000× g for 5 min (Hitachi, Tokyo, Japan), resuspended in 500 μL of BMMY, and induced at 30 °C with shaking at 200 rpm for 48 h. Methanol was added to a final concentration of 1% every 24 h to maintain induction. For re-screening, promising strains were inoculated at 1% (v/v) into 100 mL of BMGY and cultivated under the same conditions for 48 h. The cells were collected by centrifugation, resuspended in 50 mL of BMMY, and induced for 120 h with the addition of 1% methanol every 24 h. For the constitutive promoter strains, transformants were inoculated on YPD plates containing 100 μg/mL Zeocin and incubated at 30 °C for 24 h. Positive colonies were then transferred to 500 μL of YPD and cultivated in shake flasks at 30 °C and 200 rpm for 48 h. For re-screening, superior strains were inoculated at 1% (v/v) into 10 mL of YPD medium and grown under the same conditions for 48 h.

2.5. Milk-Clotting Assay

According to the classic activation kinetics of bovine prochymosin, low-pH conditions promote limited proteolytic cleavage of the N-terminal pro-segment, thereby converting the inactive zymogen into its active form [30]. Therefore, protochymosin was activated through acid treatment prior to enzymatic activity determination. Culture supernatant was acidified to pH 2.5 with 1 M HCl incubated for 2 h at room temperature, then neutralized to pH 5.5 with 2 M NaOH. After acid treatment, no visible precipitate was observed in the solution, indicating that the chymosin remained soluble under these conditions. Milk-clotting enzymatic activity was determined as described by Arima and Kei [31] with slight modification. The substrate was prepared by mixing 10% skim milk and 0.01 mol/L CaCl_2_ in distilled water and kept at room temperature for 40 min. One milliliter of assay milk was transferred to a test tube and kept at 35 °C for 10 min. After mixing in 100 μL of appropriately diluted enzyme extract, the formation of curd was observed and timed. The clotting endpoint was defined as the first appearance of visible curd particles accompanied by a loss of fluidity upon gentle tilting of the test tube. One unit of chymosin activity is defined as the amount of active chymosin required to clot 1 mL of substrate in 40 min under these conditions. The milk-clotting activity was determined for all samples, and all experiments were performed in triplicate. The milk-clotting activity was expressed in standardized units defined by the assay conditions described above, enabling relative comparisons among samples within this study.

Notably, the milk-clotting activity of recombinant chymosin is commonly reported in SU or internationally recognized units for comparison with commercial standards [32,33,34]. Historically, this activity is defined by the time required to clot standardized milk under controlled conditions, and SU have long been used in rennet assays [35]. Although commercial enzyme references provide stronger benchmarks, all samples in this study were measured under identical conditions, enabling reliable relative comparisons among strains.

The milk-clotting activity (MCA) was calculated according to the following Formula (1):

$$[eqn]$$

where MCA was expressed as milk-clotting activity in Soxhelt units (SU). One SU was defined as the amount of enzyme required to clot 1 mL of milk substrate in 40 min under the assay conditions described above. This definition is consistent with the unit (U) commonly used in previous milk-clotting activity studies, and SU is therefore equivalent to U as reported in the cited literature [36]. V is the volume of milk substrate (mL), d is the dilution factor of the enzyme solution, E is the volume of the enzyme solution used (mL), and t is the milk-clotting time (s). The constant 2400 corresponds to the standard clotting time of 40 min expressed in seconds (40 × 60 s).

2.6. Expression of Chymosin on Shake Flask

For the inducible promoter strains, positive recombinant K. phaffii strains with the highest chymosin activity were obtained from the screening process and further grown in 200 mL of BMGY for 48 h. The cells were harvested through centrifugation and resuspended in 100 mL of BMMY for 120 h in which the concentration of methanol remained constant at 1%. For the constitutive promoter strains, K. phaffii strains with the highest chymosin activity were further grown in 200 mL of YPD medium for 48 h, then the supernatant was collected for activity assay. Culture samples were collected every 24 h and analyzed for cellular wet weight and enzyme activity. Chymosin production intensity (SU/mL·h) was calculated as the enzyme activity divided by the cultivation time. The C-terminus of the mCherry-tag contains a linker sequence, which is the Kex2 protease recognition site (EKREAEA). During the secretory expression of the fusion protein, this site is recognized and cleaved by Kex2 protease. Therefore, the prochymosin secreted into the culture medium does not contain the mCherry-tag.

2.7. Fermentation Culture and Optimization of Fermentation Conditions

According to the operation manual of K. phaffii (Invitrogen), the fermentation of the induced strain was performed in four phases: bacterial culture, carbon source feeding, mixed carbon source feeding, and methanol induction. Bioreactor cultivations were initiated with a 5% (v/v) inoculum of seed culture with an OD600 nm of about 40. The first phase consisting of glycerol (40 g/L) as the carbon source was completed in about 15–17 h in 2 L. When the glycerol was completely consumed in the BSM media, the dissolved oxygen (DO) suddenly increased. The second phase started with glucose feeding (20 mL/L/h) and lasted 6–8 h. The third phase was started with mixed carbon source feeding (25% glucose: 100% methanol = 8:1, v/v) and lasted 4–6 h at a 10 mL/L/h flow rate. The final induction phase began by feeding 100% methanol (supplemented with 12 mL/L PTM1). The methanol feeding began with a 1.5–3 mL/L/h flow rate and lasted 90 h. For the constitutive recombinant strain, a feeding solution containing 50% (v/v) glycerol, biotin, and PTM1 salts was administered during the fed-batch phase. Identical Dissolved Oxygen (DO) control strategies were applied throughout both fermentations. In both processes, DO was maintained at approximately 20% by dynamically adjusting agitation and aeration, and carbon feeding was regulated according to DO-stat control rather than a fixed exponential feeding model. Other operating parameters were set as follows: aeration was provided at 2 VVM (airvolume/culturevolume/min), meaning 300 L h^−1^, and dynamically adjusted in response to DO at approximately 20%; pH was controlled using ammonia; the feed rate was automatically regulated by a peristaltic pump to maintain pH at 5.0. Foaming was controlled through the addition of polydimethylsiloxane (PDMS) during fermentation. For the inducible strain, the specific feeding rate was set at 2.88 mL/(L·h), with a total methanol addition of 864 mL. For the constitutive strain, the specific feeding rate of glucose was set at 1.536 g/(L·h), with a total addition of 368.64 g. Samples were taken every 12 h to measure the wet cell weight and chymosin activity.

The fermentation conditions of the constitutive recombinant strain were optimized by single factor experiment. Three process parameters including pH (3.5, 4.0, 5.0), carbon source (glycerol, glucose), and feeding method (batch fermentation, batch feeding fermentation) were used to identify the optimal conditions for fermentation. In this study, batch and fed-batch fermentation were specifically defined as processes without a mixed carbon source feeding phase. Induction was initiated for all conditions using a 40% (w/v) glucose solution supplemented with PTM.

2.8. Purification of Recombinant Chymosin

To further measure enzymatic properties, the recombinant chymosin was purified. The induced supernatant was centrifuged at 12,000× g for 10 min at 4 °C. The supernatant after acidification was used as an enzyme source. Pall Minimate (PALL, USA) was then used for microfiltration and ultrafiltration. The resulting concentrate was applied to an anion-exchange column (pre-equilibrated with a low-sodium buffer, 20 mM Tris-HCl, pH 8.0) on an ÄKTA pure system. After extensive washing with the equilibration buffer to remove unbound proteins, the bound chymosin was eluted using a linear gradient of NaCl in the same buffer. The fractions containing chymosin activity were pooled for subsequent analysis. Purified recombinant chymosin was deglycosylated using 250 U of Endo-β-N-acetylglucosaminidase H (Endo H) for 2 h at 37 °C. The deglycosylated and untreated chymosin were analyzed using SDS-PAGE.

2.9. Determination of Enzymatic Properties of Bovine Chymosin

The optimal temperature for enzyme activity was determined by measuring activity at various temperatures (20, 25, 30, 35, 40, 50, and 60 °C) under otherwise standard assay conditions. The thermal stability of purified chymosin was determined by measuring residual enzymatic activity at 40 °C for different amounts of times (15 min, 30 min, 45 min, 60 min, 90 min, and 120 min).

The optimal pH for chymosin activity was determined by assaying the enzyme at 35 °C in 20 mM disodium hydrogen phosphate-citric acid buffers with pH values ranging from 4.5 to 8.0. To evaluate pH stability, the enzyme was first incubated at 30 °C for 1 h in buffers of varying pH (1.0 to 9.0). The residual activity was then measured under standard assay conditions (pH 6.0, 35 °C).

The effect of metal ions and chemical reagents on the activity of chymosin was determined by adding the final concentrations of 10 mM K^+^, Ca^2+^, Pb^2+^, Cu^2+^, Mg^2+^, Mn^2+^, Zn^2+^, SDS, and EDTA each to separate samples during enzyme assay. Enzyme activity was assumed to be 100% without metal ions, and the above samples containing different metal ions were tested.

2.10. Statistical Analysis

All data are presented as mean ± standard deviation (SD) based on three independent experiments (n = 3). Prior to statistical analysis, data were assessed for normality using the Shapiro–Wilk test and for homogeneity of variances using the Brown–Forsythe test. Rescreening of recombinant chymosin-producing strains was performed using one-way analysis of variance (ANOVA), followed by Tukey’s honestly significant difference (HSD) post hoc test. Comparisons of enzyme production intensity between strains clp2-91 and GH1 were conducted using an independent-samples t-test. A p-value < 0.05 was considered statistically significant.

3. Results

3.1. Screening for High Expression of Recombinant Chymosin

The positive transformants were obtained through rapid preliminary screening. A total of 144 and 168 positive strains were screened out from 600 recombinant strains of pPIC9-pcm14 and pPIC9-clpcm14, respectively, and 188 positive strains were screened out from 200 recombinants of pGAPZα-clpcm14. From these three groups of positive strains, comprising 12, 18, and 17 strains of pPIC9-pcm14, pPIC9-clpcm14, pGAPZα-clpcm14, respectively, those with higher enzyme activity were selected for re-screening.

Among the pPIC9-pcm14 recombinant strains, strain 2-70 had the highest enzyme activity (2036 SU/mL), which was significantly higher than that of the other strains (p < 0.05). Among the pPIC9-clpcm14 recombinant strains, clp2-91 showed the highest enzyme activity, reaching 2918 SU/mL (p < 0.05) (Figure 1a,b). GH1 showed the highest enzyme activity of the pGAPZα-clpcm14 recombinant strains, with a value of 1100 SU/mL (p < 0.05) after rescreening (Figure 1c).

After a 120 h induction at the shake flask level, the growth rate of clp2-91 increased rapidly. The final cell wet weight reached 122 mg/mL, while the growth rate of 2-70 was slightly lower than that of clp2-91, concluding at 113 mg/mL (Figure 1d). The enzyme activity of 2-70 and clp2-91 were 2188 SU/mL and 3705 SU/mL, respectively, and the enzyme activity of clp2-91 was 1.7 times that of 2-70 (Figure 1d). In summary, the experimental data collectively demonstrate that the introduction of the mCherry-tag effectively enhanced the expression level of the enzyme.

Despite lower initial metrics during the first 24–48 h, the GH1 strain exhibited a surge in production by 72 h, reaching a final biomass of 125 mg/mL (Figure 1d) and a chymosin activity of 4544.67 SU/mL (Figure 1d) after 96 h of cultivation, ultimately exceeding the performance of the clp2-91. This demonstrates that the GAP promoter strategy effectively enhances overall production intensity while streamlining the process.

3.2. Analysis of the Secreted Protein Profile of Chymosin-Transformed Clone

The bovine prochymosin gene was inserted into the K. phaffii genome, and the protein expressed was an inactive recombinant bovine rennet. SDS-PAGE analysis of the purified recombinant protein revealed two distinct bands with molecular weights of approximately 40 kDa and 36.5 kDa (Figure 2a), a pattern consistent with glycosylated chymosin expressed in yeast systems. To investigate this, the protein was treated with Endo-β-N-acetylglucosaminidase H (Endo H) to remove N-linked glycans. This treatment resulted in the collapse of the 40 kDa band, leaving a single band at 36.5 kDa (Figure 2b). This finding confirms that the higher-molecular-weight species is a glycosylated form of the enzyme, with the glycan moiety accounting for a mass difference of approximately 4 kDa.

3.3. Different Promoters of Recombinant Bovine Chymosin in Bioreactor Production

The fermentation profiles of the two recombinant K. phaffii strains, carrying the methanol-inducible promoter (strain clp2-91) and the constitutive promoter (strain GH1), were evaluated in a 3L fermenter. For the inducible strain clp2-91, both the cell wet weight and enzyme activity increased over time. The cell wet weight reached a maximum of 300 g/L, while the chymosin activity peaked at 7084 SU/mL at 130 h. In contrast, the constitutive strain GH1 demonstrated a significantly different profile: cell growth achieved a maximum cell wet weight of nearly 500 g/L. More notably, the production of chymosin commenced earlier and reached a final activity exceeding 10,084 SU/mL at 94 h, which was considerably higher than that of the inducible strain (Figure 3a).

Strain GH1 achieved a production intensity of 105.03 SU/mL·h, which is substantially higher than the 53.59 SU/mL·h observed of strain clp2-91 (Figure 3b, p < 0.05). This result clearly demonstrates that the constitutive promoter strategy in GH1 enables a more efficient and productive fermentation process.

3.4. Optimization of Fermentation Conditions for GH1 Strain

The initial optimization of fermentation conditions for the GH1 strain focused on pH. As shown in Figure 4a, the highest enzyme activity of 7285 SU/mL was observed at pH 4.0, compared to 7058 SU/mL and 6400 SU/mL at pH 3.5 and 5.0, respectively. In contrast, the maximum cell wet weight (532 g/L) was achieved at pH 3.5. A pH of 4.0 was selected as the optimal condition for subsequent fermentations, as it afforded a high enzyme titer while maintaining substantial cell growth (477 g/L).

The choice of carbon source was evaluated for its impact on fermentation performance (Figure 4b). Throughout the process, cell growth was similar with both carbon sources, culminating in a comparable final cell wet weight of approximately 470 g/L. However, a divergence in enzyme activity was observed after 72 h, with glucose supporting a higher recombinant protein production. After 94 h, the enzyme activity in the glucose-based culture reached 12,000 SU/mL, compared to 10,584 SU/mL in glycerol. Furthermore, the total consumption of glycerol (1.5 kg) was higher than that of glucose (1 kg) over the course of the fermentation. Based on its superior enzyme yield and lower consumption, glucose was therefore selected as the preferred carbon source for the GH1 strain.

The fermentation performance of batch and batch-feeding strategies was compared (Figure 4c). In the batch-feeding process, 400 mL of BSM was added at 72 h following the harvest of an equivalent volume of fermentation broth. This supplementation resulted in a temporary decrease in both enzyme activity and cell wet weight by 84 h. However, after 12 h of continued cultivation, enzyme activity recovered to a level comparable to that of the conventional batch process (approximately 10,240 SU/mL). Although the final enzyme activity per unit volume was similar in both methods, the batch-feeding approach enabled the production of a larger total volume of enzyme solution, thereby increasing the overall product yield per fermentation cycle.

3.5. Characterization of Recombinant Chymosin and Native Chymosin

The optimal temperature and pH for the recombinant bovine chymosin were determined and compared with those of the native enzyme (Figure 5a,c). Both enzymes exhibited a shared optimal temperature of 35 °C (Figure 5a), with activity sharply declining at temperatures above 40 °C or below 30 °C. The recombinant bovine chymosin displayed maximal activity at pH 5.0 (Figure 5c), the same pH for optimal activity as the native enzyme.

The stability profiles of the two enzymes under different temperatures and pH conditions are shown in Figure 5b,d. The thermostability assay at 40 °C revealed that both enzymes maintained over 80% of their initial activity after 60 min of incubation, with recombinant chymosin showing a marginally slower rate of inactivation compared to the native enzyme over an extended period (Figure 5c). Recombinant chymosin retained a high activity across a broad pH range from 2.0 to 6.0, whereas the native enzyme was stable only within the narrower range of pH 3.0 to 6.0, indicating the recombinant enzyme’s superior tolerance to highly acidic conditions (Figure 5d).

The effects of various metal ions and chemical reagents on enzyme activity are summarized in Table 2. The recombinant chymosin and the native enzyme showed nearly identical sensitivity patterns. Strong inhibition was observed in the presence of Ni^2+^, Cu^2+^, CTAB, EDTA, and SDS, reducing the relative activity to below 30% in all cases. Conversely, ions such as Mg^2+^, Ca^2+^, Mn^2+^, Na^+^, K^+^, and Zn^2+^ had minimal to no inhibitory effects, with Ca^2+^ even exhibiting a slight activating effect on recombinant chymosin. This result confirms that the recombinant enzyme’s catalytic center maintains a ligand-binding profile highly similar to that of the native chymosin.

Taken together, these findings demonstrate that the recombinant chymosin closely mirrors the catalytic characteristics of the native enzyme in terms of optimal temperature and response to most reagents. Importantly, the recombinant chymosin possesses a distinct advantage in its significantly broader pH stability range, making it a robust and potentially superior candidate for applications under varying acidic conditions.

4. Discussion

This study established the high-yield production of bovine prochymosin in Komagataella phaffii using a multi-step optimization strategy. The resulting recombinant strain enabled the methanol-free, high-level expression of prochymosin. Within the scope of the parameters evaluated, the recombinant enzyme exhibited enzymatic properties similar to native chymosin and demonstrated an enhanced pH stability, supporting its potential as an industrially viable rennet alternative.

Initially, a codon-optimized pcm14 gene was expressed in K. phaffii, and screening identified a high-producing strain, 2-70 (2188 SU/mL). Subsequently, fusing pcm14 to an mCherry-tag reporter yielded strain clp2-91 with a titer of 3705 SU/mL at the shake flask scale. The observed increase likely reflects the following effects: codon optimization improved translation, while mCherry-tag correlated with enhanced folding and secretion, reducing misfolding and degradation [20,37,38,39]. These genetic modifications not only elevated titers in shake flasks but also translated into markedly improved volumetric productivity under controlled bioreactor conditions. Subsequently, based on clp2-91, the inducible AOX1 promoter was replaced with a constitutive GAP promoter, yielding engineered strain GH1 (4544.67 SU/mL). When the two selected high-activity strains were scaled up in a 3L fermenter, strain GH1 achieved a production intensity of 105.03 SU/mL·h, nearly double the 53.59 SU/mL·h productivity of strain clp2-91. These results indicate that replacing the methanol-inducible AOX1 promoter with a constitutive GAP promoter increased productivity, simplified the production process, reduced safety risks, shortened the production time, and decreased the risk of residual methanol and off-odors in the cheese.

The subsequent optimization of fermentation conditions revealed that a pH of 4.0 was optimal for enzyme production, yielding 7285 SU/mL. At pH 3.5 and pH 5.0, the enzyme activities were 7058 SU/mL and 6400 SU/mL, respectively. Although this pH was lower than the optimal range for host cell growth, it likely suppressed the activity of host proteases, thereby minimizing product degradation [40], consistent with the recommendation of Ramon et al. for the expression of sensitive proteins [41]. Furthermore, the glucose-based culture reached an enzyme activity of 12,000 SU/mL and required only 1 kg of substrate, whereas the glycerol-based culture yielded 10,584 SU/mL and consumed 1.5 kg of substrate. This combination of higher product yield and lower resource consumption directly translates to superior economic feasibility, making glucose the optimal choice for scaling up the fermentation process industrially [42].

Enzyme characterization revealed that the recombinant bovine chymosin exhibited a milk-clotting performance comparable to that of the native enzyme under the tested conditions. Both share an optimal temperature of 35 °C and exhibit maximal activity at pH 5.0, with steep activity loss outside the ranges of 30–40 °C and beyond pH 6.0. Thermostability assays at 40 °C showed that both enzymes retained >80% of initial activity after 60 min, although the recombinant form displayed a slightly slower inactivation rate over prolonged incubation. Notably, the recombinant chymosin exhibited a markedly broader pH stability (pH 2.0–6.0) than the native enzyme (pH 3.0–6.0), indicating a superior tolerance to highly acidic conditions—an advantageous trait for cheese-making processes where low pH prevails. Collectively, the recombinant chymosin demonstrates potential as a robust alternative to the native enzyme, combining comparable catalytic behavior with enhanced pH stability. This profile may be suitable for future applications under acidic conditions.

Future work will focus on evaluating additional industrially relevant parameters, such as κ-casein cleavage specificity and the milk-clotting activity to proteolytic activity (MCA/PA) ratio, in order to more rigorously assess the catalytic equivalence between recombinant and native chymosin. In parallel, future studies will evaluate the glycosylation profiles of recombinant chymosin using LC-MS and conduct standardized cheese-making trials, including measurements of curd firmness, proteolysis index, and sensory attributes, to benchmark its performance against commercial rennets. From an economic standpoint, the use of ÄKTA chromatography in this study may present cost-related challenges for large-scale implementation. Accordingly, future studies will explore more cost-effective and operationally simpler purification strategies, including ultrafiltration and precipitation, to balance production efficiency and product purity in industrial applications.

5. Conclusions

This study demonstrates that chymosin production in Komagataella phaffii can be effectively enhanced through a combination of rational genetic and process engineering. In particular, the fusion strategy contributed to improved expression and secretion efficiency, while the use of a constitutive promoter enabled a methanol-free production system with simplified operation. The recombinant chymosin exhibited functional properties, indicating a potential for future applications. Moreover, the strategies presented here provide useful insights for the efficient production of other heterologous proteins in K. phaffii.