Functional Dipeptide Production by Immobilized Enzyme on Yeast Cell Surface

TL;DR

This study presents a clean and efficient method for producing the dipeptide L-Alanyl-L-glutamine using yeast cells with immobilized enzymes, suitable for industrial applications.

Contribution

A reusable whole-cell biocatalytic system using yeast with immobilized SsAET enzyme for efficient dipeptide production.

Findings

Optimized conditions increased Ala-Gln production by 6.7-fold compared to pre-optimization.

The biocatalyst retained over 60% activity after three reaction cycles with minimal conversion rate decline.

A maximum Ala-Gln concentration of 14.12 mM was achieved under optimal conditions.

Abstract

L-Alanyl-L-glutamine (Ala-Gln) is a high-value dipeptide with superior stability, solubility, and bioavailability, underscoring its potential for nutritional supplementation. Compared with conventional chemical catalysis, whole-cell biocatalysts offer a more efficient, simpler, and environmentally friendly alternative for peptide synthesis. Among these, enzyme cell-surface immobilization systems enable the stable display of target enzymes on yeast cells, thereby enhancing enzyme stability while simplifying catalyst recovery and reuse, which is particularly advantageous for large-scale industrial applications. In this study, an engineered Saccharomyces cerevisiae strain displaying α-amino acid ester acyltransferase (SsAET) from Sphingobacterium siyangensis SY1 on the cell surface was developed for clean and efficient biocatalytic production of Ala-Gln. Optimal reaction conditions were established (pH 8.0, 3 h, 20 g DCW/l, AlaOMe/Gln = 1:2), resulting in a 6.7-fold increase in Ala-Gln production compared with the pre-optimization conditions. Under these optimal conditions, repeated-batch reactions with an increased reaction volume of 20 ml achieved a maximum Ala-Gln concentration of 14.12 mM, representing the highest yield obtained in this study. The SsAET whole-cell biocatalyst retained more than 60% of its relative Ala–Gln production after three consecutive reaction cycles, while the conversion rate based on Gln consumption decreased only slightly from 32.08% to 29.65%, remaining essentially stable. Overall, this study demonstrates a clean, efficient, and reusable whole-cell biocatalytic system based on enzyme cell-surface immobilization for Ala-Gln production, highlighting its potential for industrial-scale peptide synthesis.

Figures

A](#F3)). A highly confident cleavage site was predicted between the 19^th^ and 20^th^ amino acids, with a probability of 97.8%. This result is consistent with previous reports showing that the Mfα1 signal peptide contains a 19–amino acid pre-sequence upstream of the cleavage site [[24](#ref24)], suggesting efficient secretion and proper processing of the SsAET fusion protein.

C](#F4)). At reaction times longer than 3 h, the activity decreased and subsequently remained approximately constant. The influence of whole-cell concentration was also examined, with the highest activity observed at 20 g DCW/l ([Fig. 4D](#F4)). Increasing the cell concentration beyond this level resulted in a reduction in biocatalytic activity. Lastly, the ratio of AlaOMe/Gln at 1:2 achieved the highest biocatalytic activity ([Fig. 4E](#F4)). Finally, the effect of substrate molar ratio was evaluated, and the highest activity was obtained at an AlaOMe/Gln ratio of 1:2 ([Fig. 4E](#F4)).

Full text

Introduction

L-Glutamine (Gln) is the most abundant amino acid in human plasma and plays a vital role in numerous physiological functions [1]. Although Gln is classified as a non-essential amino acid because it can be synthesized endogenously, a tightly regulated balance between its production and consumption is required under normal physiological conditions [2]. During periods of injury or illness, Gln consumption may exceed endogenous production capacity, disrupting this balance and rendering Gln a conditionally essential amino acid [3]. Consequently, supplementation through parenteral nutrition (PN) is particularly important for restoring Gln homeostasis in patients [4]. However, the pharmaceutical application of Gln is limited by its low water solubility (36 g/l), poor thermal stability, and the formation of toxic pyroglutamate during heat sterilization [5, 6].

To overcome these physicochemical limitations of Gln, Gln-containing dipeptides have been proposed as effective alternatives [7]. L-Alanyl-L-glutamine (Ala-Gln), a C-terminal glutamine dipeptide with high thermal stability in vivo and excellent water solubility (568 g/l), is considered one of the most promising Gln supplements [5, 8]. In addition to serving as an alternative to Gln, Ala-Gln exhibits improved bioavailability compared with Gln and does not cause cumulative toxicity in the body [9].

Recently, a novel enzyme, α-amino acid ester acyltransferase (SsAET), was identified in Sphingobacterium siyangensis SY1 [10]. This enzyme can directly synthesize Ala-Gln from L-alanine methyl ester hydrochloride (AlaOMe) and Gln. Currently, industrial production of Ala-Gln relies primarily on chemical synthesis methods, which involve complex reaction processes, the use of toxic reagents, and the generation of environmentally harmful by-products [11]. In accordance with the principles of green chemistry, there is an increasing demand for clean and efficient approaches to Ala-Gln production. Whole-cell biocatalysts offer an environmentally friendly alternative by improving enzyme accessibility to substrates, enhancing catalytic efficiency, and enabling facile recycling of the biocatalysts [12].

Enzyme cell surface immobilization is a type of whole-cell biocatalytic system in which target proteins are expressed and anchored on the yeast cell surface. This approach is particularly attractive for large-scale industrial applications because it enhances enzyme stability and simplifies enzyme recovery and reuse [13]. In this system, the target protein is genetically fused to an anchor protein and secreted under the guidance of a signal peptide for display on the cell surface [14]. Among the various signal peptides, the S. cerevisiae α-mating factor is the most widely used signal sequence for recombinant protein expression [15]. In a previous study, enhanced green fluorescent protein (eGFP) was successfully displayed on the surface of S. cerevisiae using a cell surface immobilization strategy, and the Sed1 anchor protein exhibited the highest activity among six anchor proteins (Aga2, Cwp1, Cwp2, Sed1, Pir1 and Tir1) [16]. Based on these findings, the Sed1 anchor protein was selected in this study for cell surface immobilization of the target enzyme.

In this study, the SsAET enzyme was expressed and immobilized on the surface of S. cerevisiae cells to enable the biocatalytic production of Ala-Gln from AlaOMe and Gln, and Ala-Gln production was further enhanced through optimization of the biocatalytic reaction conditions. To the best of our knowledge, this is the first study to report Ala-Gln production using a cell surface–immobilized SsAET enzyme in S. cerevisiae. This yeast-based expression system provides a promising foundation for the clean and sustainable industrial-scale production of Ala-Gln.

Materials and Methods

Strain and Growth Conditions

The recombinant yeast strains and plasmids used in this study are listed in Table 1. Yeast cells were pre-cultivated in YP medium (10 g/l yeast extract and 20 g/l peptone) supplemented with 20 g/l glucose (YPD) at 30°C with shaking at 250 rpm for 24 h. Following pre-cultivation, cells were sub-cultivated under the same conditions for an additional 6 h.

Escherichia coli TOP10 (Invitrogen, USA) was used for amplification of gRNA plasmids. E. coli cells were cultivated in Luria-Bertani (LB) medium supplemented with 100 μg/mL ampicillin at 37°C with shaking at 250 rpm for 18 h.

Strain Construction Using CRISPR-Cas9 System

Yeast transformation was performed using the lithium acetate/single-stranded carrier DNA-polyethylene glycol method [17] in combination with the CRISPR-Cas9 system, as described previously [18]. The guide RNA (gRNA) plasmids used in this study are listed in Table 1. The pRS42H-INT#1 gRNA plasmid was generated by PCR amplification using specific primers, and the resulting PCR product was transformed into E. coli DH5α.

Donor DNA fragments were prepared by PCR amplification using gene-specific primers, which are summarized in Table 2. The SsAET gene from Sphingobacterium siyangensis SY1 was codon-optimized for expression in S. cerevisiae and fused to the Sed1 anchor protein. The resulting expression cassette was integrated into the cg#1 intergenic locus of the S. cerevisiae P_CCW12_-Mfα1-cg#1-T_CYC1_ strain, which had been previously constructed [16], using the pRS42H-cg#1 gRNA plasmid.

Successful transformants were verified by PCR using the Kim094 and Kim497 primers to confirm integration of the P_CCW12_-Mfα1-SsAET-Sed1-T_CYC1_ cassette into the yeast genome.

Cell Growth Assessment

Yeast cells were pre-cultivated in YPD medium at 30°C with shaking at 250 rpm for 24 h. Cells were then harvested, washed with sterile distilled water, and resuspended in fresh YPD medium to an initial optical density at 600 nm (OD_600_) of 0.1. Cultivations were carried out in 50 ml flasks containing 10 ml of medium and incubated at 30°C with shaking at 250 rpm.

Prediction of Signal Peptide Cleavage Site and Structure of Protein

The specific cleavage site of the Mfα1 signal peptide in the engineered yeast was predicted using the SignalP 6.0 server (https://services.healthtech.dtu.dk/services/SignalP-6.0/). The 3D structure of the Mfα1-SsAET-Sed1 fusion protein, based on the designed sequences, was predicted using ESMFold [19], and the fusion interface between SsAET and Sed1 was visualized with PyMOL (https://www.pymol.org/).

Ala-Gln Production Assay

AlaOMe, Gln and Ala-Gln were purchased from Sigma-Aldrich. Ala-Gln production using AET-expressing yeast as whole-cell biocatalysts was performed under the following conditions. Cells were cultivated in YP medium containing 20 g/l glucose until reaching 0.5 g DCW/l (exponential phase), and then collected by centrifugation (4,000 ×g, 5 min). Reactions were conducted at a cell concentration of 5 g DCW/l. The enzyme-catalyzed reaction mixture was prepared in 100 mM sodium phosphate buffer (pH 8.5) to a final volume of 1 ml, containing 100 mM AlaOMe and 100 mM Gln. Reactions were maintained at 25°C and 100 rpm for 30 min and terminated sequentially by centrifugation (13,000 ×g, 1 min) and heat inactivation (100°C, 5 min). Supernatants were filtered through a 0.22 μm PES filter and used to determine Ala-Gln concentrations via HPLC.

Dry cell weight (DCW) was estimated from the correlation between optical density and cell concentration, where an OD_600_ of 1.0 corresponded to 3 × 10^7^ cells/ml, equivalent to 0.26 g DCW/l, as determined using a UV/Vis-spectrophotometer (Optizen NanoQ, Mecasys Co., Republic of Korea).

Optimization of Ala-Gln Production Conditions

The cultivation of recombinant yeast cells and the composition of the enzyme reaction solution were performed as described in Section 2.5. To enhance Ala-Gln productivity, the optimal enzyme reaction conditions were investigated. The variables optimized included reaction pH (6.0, 7.0, 8.0, and 8.5), AlaOMe addition method (single addition of 100 mM AlaOMe, stepwise addition of 10 mM AlaOMe , or addition after adjusting AlaOMe to pH 8.0 with 10 M NaOH), reaction time (0.5, 1, 3, 6, 9, and 12 h), reaction cell concentration (5, 20, and 50 g DCW/l), and the substrate ratios of AlaOMe to Gln (1:1, 1.5:1, 2:1, 1:1.5, 1:2, and 1:3). Ala-Gln production was subsequently evaluated under the optimized conditions. The conversion rate was calculated as the molar percentage of Ala–Gln produced relative to the molar amount of Gln consumed. Relative Ala–Gln production was defined as the normalized Ala–Gln concentration and is presented as a percentage of the maximum value observed across all tested conditions, which was set to 100%.

Ala-Gln Production by Repeated-Cycle Batch Reactions

The cultivation and harvest of recombinant yeast cells were conducted as described in Section 2.5. The biocatalytic reaction mixture was prepared to a final volume of 20 ml, containing 100 mM AlaOMe and 200 mM Gln. The reaction was carried out at 25°C and 100 rpm for 3 h, with the pH maintained at 8.0 using 10 M NaOH. To terminate each reaction, a 1 ml aliquot was withdrawn, centrifuged (13,000 ×g, 1 min) to remove cells, and the supernatant was heat-inactivated at 100°C for 5 min prior to HPLC analysis. The remaining reaction mixture was centrifuged (4,000 ×g, 5 min) to harvest the cells, and fresh substrate solution was added to initiate the next reaction cycle. This procedure was repeated for a total of three cycles. Supernatants, filtered through a 0.22 μm PES membrane, were used to determine Ala-Gln concentrations by HPLC.

HPLC Analysis

The concentration of synthesized Ala-Gln was measured using a high-performance liquid chromatography (HPLC) system (Agilent, 1220 Infinity II LC system) equipped with an aminopropyl-silane bonded porous silica column (SUPELCOSIL LC-NH_2_, 4.6 × 250 mm^2^, 5 μm). The column was eluted with a mobile phase consisting of 65% (v/v) HPLC grade acetonitrile and 35% (v/v) 50 mM potassium dihydrogen phosphate solution (pH 5.0) at a flow rate of 1.0 ml/min and 30°C. The injection volume was 10 μl, and detection was performed at 210 nm using a UV detector.

Statistical Analysis

All results are presented as mean ± standard deviation (SD) from at least three independent experiments. Statistical analyses were performed using Student’s t-test (after confirming equal variances) or one-way analysis of variance (ANOVA) for multiple group comparisons. Tukey’s Honestly Significant Difference (HSD) test was applied for post-hoc analysis following ANOVA. Differences were considered statistically significant at p < 0.05. All analyses were conducted using IBM SPSS Statistics software (version 27; IBM Corp., USA).

Results

Construction of Recombinant S. cerevisiae with SsAET Enzyme Immobilized on Cell Surface

To construct a yeast strain displaying SsAET on the cell surface, an expression cassette consisting of the CCW12 promoter (P_CCW12_), the Mfα1 secretory signal peptide, a codon-optimized SsAET gene, the Sed1 anchor protein, and the CYC1 terminator (T_CYC1_) was designed (Fig. 1). The SsAET gene from Sphingobacterium siyangensis SY1 was codon-optimized for expression in S. cerevisiae. The Mfα1 signal peptide directed the SsAET fusion protein into the secretory pathway, while the Sed1 anchor enabled immobilization of the enzyme on the yeast cell surface.

The assembled P_CCW12_-Mfα1-SsAET-Sed1-T_CYC1_ expression cassette was integrated into intergenic region #1 (int#1) on chromosome VII of S. cerevisiae via CRISPR–Cas9–mediated homologous recombination (Fig. 1). This intergenic locus was selected as a neutral genomic site to allow stable expression without disruption of endogenous genes. Successful integration was confirmed by PCR analysis of the transformants, which produced bands of approximately 3.8 kb (Fig. S1).

To assess the functional expression of surface-displayed SsAET, whole-cell biocatalytic activity was evaluated using AlaOMe and Gln as substrates. The engineered AET strain produced approximately 0.65 mM Ala–Gln, whereas only trace levels were detected in the wild-type control strain (Fig. 2), indicating successful expression and catalytic activity of the immobilized SsAET enzyme. Furthermore, no significant differences in growth were observed between the wild-type and AET strains under standard culture conditions (Fig. S2), suggesting that surface display of SsAET had no detrimental effect on host cell viability.

Prediction of Signal Peptide Cleavage Site and 3D Structure of Fused Protein

Efficient cleavage of signal peptides is a critical factor for successful protein expression and enzyme immobilization on the yeast cell surface, as it ensures accurate targeting to the endoplasmic reticulum (ER) and proper folding of the mature protein [20]. The pre-peptide region mediates ER targeting [21], while the pro-peptide region facilitates ER exit and correct protein folding [22, 23]. Based on these principles, the secretion efficiency of the Mfα1 signal peptide fused to SsAET was evaluated.

The Mfα1 signal peptide fused to SsAET was analyzed using SignalP 6.0, which predicted a typical signal peptide architecture comprising an N-terminal positively charged region (N), a central hydrophobic region (H), and a C-terminal cleavage region (C) (Fig. 3A). A highly confident cleavage site was predicted between the 19^th^ and 20^th^ amino acids, with a probability of 97.8%. This result is consistent with previous reports showing that the Mfα1 signal peptide contains a 19–amino acid pre-sequence upstream of the cleavage site [24], suggesting efficient secretion and proper processing of the SsAET fusion protein.

The three-dimensional structure of the Mfα1-SsAET-Sed1 fusion protein was further predicted using ESMFold (Fig. 3B). The predicted model indicates that SsAET retains its overall structural integrity upon fusion with the Sed1 anchor protein. A magnified view of the structure highlights the C-terminal fusion junction, where SsAET is linked to Sed1 at Asp687 (Fig. 3C), supporting the structural feasibility of yeast surface display without disruption of the enzyme architecture.

Optimization of Reaction Conditions to Produce Ala-Gln

To enhance Ala-Gln production by the whole-cell biocatalyst, the biocatalytic reaction conditions for the AET strain were systematically optimized. The effect of reaction pH on biocatalytic activity was first evaluated. The optimal pH was determined to be 8.0 (Fig. 4A), and approximately 80% of the maximum relative Ala-Gln production was retained within the pH range of 7.0-8.5, indicating stable performance under weakly alkaline conditions.

Because AlaOMe is supplied as a hydrochloride salt and may reduce the reaction pH upon addition, different substrate addition strategies were subsequently examined. Relative Ala-Gln production was compared between continuous addition of AlaOMe at a pH-sustainable concentration (10 mM per addition) and addition of AlaOMe after adjusting its pH to 8.0 using 10 M NaOH. The highest biocatalytic activity was observed when AlaOMe was pre-adjusted to pH 8.0 prior to addition (Fig. 4B).

The effect of reaction time on Ala–Gln production was then investigated. Biocatalytic activity increased in a time-dependent manner and reached a maximum at 3 h (Fig. 4C). At reaction times longer than 3 h, the activity decreased and subsequently remained approximately constant. The influence of whole-cell concentration was also examined, with the highest activity observed at 20 g DCW/l (Fig. 4D). Increasing the cell concentration beyond this level resulted in a reduction in biocatalytic activity. Lastly, the ratio of AlaOMe/Gln at 1:2 achieved the highest biocatalytic activity (Fig. 4E). Finally, the effect of substrate molar ratio was evaluated, and the highest activity was obtained at an AlaOMe/Gln ratio of 1:2 (Fig. 4E).

Based on these results, biocatalytic performance was evaluated under the optimized conditions: reaction pH 8.0, AlaOMe addition after pH adjustment with NaOH, reaction time of 3 h, cell concentration of 20 g DCW/l, and an AlaOMe/Gln ratio of 1:2. Under these conditions, Ala-Gln production increased by approximately 6.7-fold compared with that under non-optimized conditions (Fig. 5). In addition, the Gln conversion rate increased by approximately 3.4-fold following optimization of the biocatalytic reaction conditions (Table 3).

Repeated-Cycle Batch Production of Ala-Gln in 20 ml Volume

Whole-cell biocatalysts by enzyme immobilization on the yeast cell surface have significant advantages such as enzyme stability and reusability, which make the process sustainable, cost-effective, environmentally friendly, and efficient [13]. These advantages reduce the need for expensive enzyme extraction and purification, minimize waste, and allow batch operation for enzyme reuse. AET was recycled every 3 h and the biocatalytic reaction was performed for three cycles by adding fresh substrate to the whole cells in repeated-batch operation, with the reaction volume scaled up from 1 ml to 20 ml. In the 1 ml reaction volume, the pH was adjusted using a sodium phosphate buffer, but in the 20 ml reaction volume, the pH was adjusted more delicately by continuously adding 10 M NaOH without using a buffer. Scaling up the reaction volume from 1 ml to 20 ml increased Ala-Gln production from 4.47 mM to 14.12 mM, an approximately 3.2-fold enhancement and the highest yield observed in this study (Fig. 6). Moreover, under repeated-cycle batch conditions, the biocatalytic activity of AET maintained more than 60% of the relative Ala-Gln production during the third cycle (Fig. 7), and the conversion rate of Gln decreased only slightly over three cycles, from 32.08% to 29.65% (Table 4). From these results, it can be concluded that AET biocatalytic activity during repeated-cycle batch operation maintained stable Ala-Gln production efficiency.

Discussion

In previously reported studies, Ala-Gln production has predominantly relied on heterologous expression of peptide-forming enzymes. L-Amino acid α-ligase (Lal) from Bacillus subtilis has been proposed to catalyze Ala–Gln synthesis from L-alanine and L-glutamine [25]. However, this reaction is ATP-dependent, and the reported Ala–Gln yields are relatively low, limiting its feasibility for large-scale applications. More recently, AET enzymes derived from S. siyangensis AJ2458 [26] and S. siyangensis SY1 [10] have been shown to directly synthesize Ala-Gln from AlaOMe and L-glutamine via an ATP-independent mechanism. These AET enzymes exhibit significantly higher catalytic activity compared to Lal, highlighting their potential for efficient dipeptide production. In particular, AET from S. siyangensis SY1 (SsAET) has demonstrated robust Ala–Gln production when heterologously expressed in both bacterial and yeast hosts [10, 27, 28].

High molar yields of Ala–Gln (67–94.7%) were obtained through overexpression of SsAET in E. coli [7, 10]. Despite this high efficiency, the use of E. coli poses significant limitations for pharmaceutical applications, including the risk of endotoxin contamination, dependence on toxin-inducing reagents such as IPTG, and the requirement for antibiotic selection. To address these concerns, a yeast-based system overexpressing SsAET in Pichia pastoris was subsequently developed, achieving molar yields of 63.5–80% [27, 28]. However, heterologous protein expression in P. pastoris is typically induced by methanol, and residual methanol can exert toxic effects on host cells and complicate downstream processing [29]. Moreover, the mandatory methanol removal step increases process complexity and overall industrial production costs.

To address these limitations, this study developed a new biocatalyst system based on heterologous expression of SsAET in S. cerevisiae, which offers distinct advantages as a production host by eliminating the need for external toxic inducers and enabling stable heterologous protein expression driven by endogenous promoters [30]. Moreover, enzyme immobilization on the yeast cell surface to generate whole-cell biocatalysts is widely recognized as an environmentally friendly and industrially suitable strategy, as it improves catalytic efficiency and operational stability while allowing facile recovery and reuse of the biocatalyst [12, 13]. Accordingly, engineered S. cerevisiae strains displaying heterologous SsAET on the cell surface were constructed to enable cleaner and more sustainable Ala–Gln production.

The efficiency of enzyme display on the yeast cell surface is strongly influenced by the choice of anchoring protein [14]. Previous comparative studies in S. cerevisiae demonstrated that the Sed1 anchor protein, which is predominantly localized on the external surface of the yeast cell wall, exhibited the highest surface-display efficiency among several candidates (Aga2, Cwp1, Cwp2, Sed1, Pir1, and Tir1) using eGFP as a reporter protein [16]. Based on these findings, the Sed1 anchor protein was selected in this study for cell surface immobilization of the target enzyme.

Although the AET strain constructed in this study exhibited a significant increase in Ala–Gln production compared with the wild-type strain, the overall production level remained limited. To further enhance Ala–Gln production, the biocatalyst reaction conditions were systematically optimized. The assessment of AET biocatalytic activity was complicated by the presence of hydrochloride (HCl) in the AlaOMe substrate, which caused a rapid decrease in the pH of the reaction mixture. Previous studies have reported that SsAET expressed in P. pastoris exhibits optimal activity under weakly alkaline conditions, with an optimal pH of approximately 8.5 [28]. Consistent with these findings, the AET strain developed in this study showed maximal activity within a weakly alkaline pH range of 7.0–8.5, with the highest activity observed at pH 8.0. In addition, the effect of AlaOMe addition strategy on biocatalytic activity was evaluated. Among the tested approaches, adjusting the reaction mixture to pH 8.0 prior to AlaOMe addition was most effective in preventing rapid acidification and maintaining a stable reaction environment.

Subsequently, the effects of reaction time, cell concentration, and substrate ratio on Ala–Gln production were investigated to further enhance product formation. The optimal conditions were determined to be a reaction time of 3 h, a cell concentration of 20 g DCW/l, and a substrate ratio (AlaOMe/Gln) of 1:2. Biocatalytic activity increased in a time-dependent manner up to 3 h, after which it declined and stabilized at longer reaction times. In addition, cell concentrations exceeding 20 g DCW/l resulted in reduced biocatalytic activity, suggesting that excessive biomass negatively affects enzyme performance. This decrease may be associated with the uptake of produced Ala–Gln by yeast cells via endogenous peptide transporters, leading to reduced extracellular product levels. In S. cerevisiae, Ptr2p has been identified as a major transporter responsible for the uptake of di- and tripeptides [31]. Similar effects of intracellular uptake and degradation of dipeptides have been reported in E. coli, where intracellular peptidases encoded by pepA, pepB, pepD, and pepN, as well as the dipeptide ABC transporter dpp, contribute to dipeptide degradation and uptake; deletion of these genes resulted in increased dipeptide accumulation [32].

Whole-cell biocatalysis via enzyme immobilization on the yeast cell surface offers several important advantages, including enhanced enzyme stability, reusability, and process robustness, making it a sustainable, cost-effective, and environmentally friendly strategy [13]. To evaluate these advantages, the AET strain was recycled every 3 h and applied for three consecutive biocatalytic cycles, with the reaction volume scaled up from 1 ml to 20 ml. At the larger reaction scale, pH control was achieved through continuous and delicate adjustment using 10 M NaOH, which effectively maintained the optimal reaction environment. Under these conditions, Ala–Gln production reached approximately 14.12 mM, representing the highest titer achieved in this study and highlighting the critical role of precise pH control in maximizing AET biocatalytic activity. In repeated-batch experiments, the immobilized AET retained more than 60% of its initial activity after the third cycle, while the conversion rate to Ala–Gln decreased only slightly from 32.08% to 29.65%, indicating good operational stability and reusability of the surface-displayed enzyme.

To our knowledge, this is the first report demonstrating Ala–Gln production via surface display and immobilization of SsAET on S. cerevisiae. Although the Ala–Gln titer achieved in this study (14.12 mM) is lower than that reported for some E. coli–based systems (>100 mM) [33] and P. pastoris–based systems (63.5–227.9 mM) [27, 28], the present yeast-based platform offers distinct practical advantages, including the use of a generally recognized as safe (GRAS) host, stable heterologous expression without toxic inducers, direct enzyme immobilization on the cell surface, and efficient biocatalyst reuse. Further improvements in conversion efficiency could be achieved through strategies such as minimizing product uptake by deleting endogenous peptide transporters [31, 32], increasing surface display density and improving expression and secretion efficiency of the target enzyme [14], and fine-tuning reaction engineering parameters [34].

Supplemental Materials

Supplementary data for this paper are available on-line only at http://jmb.or.kr.