Engineering

Content Overview

 To further confirm whether the enzymes designed by the dry lab group using AI tools possess genuine catalytic activity, we characterized the catalytic performance of DTE enzymes through wet lab experiments. During the formal experimental process, we primarily designed:

• A fructose biosensor based on pepper fluorescent RNA;
• A plasmid for expressing the DTE enzyme;
• Integration of both systems to measure DTE enzyme catalytic capacity.

 The entire process followed the iGEM-recommended DBTL (design, build, test, learn) cycle, where we continuously refined our working system through feedback. Below are the completed experimental steps and their underlying logic.

2. Design

2.1 Design of a novel DTE enzyme

 We aimed to redesign the DTE enzyme sequence while preserving catalytic activity.

 We first analyzed the structure of the native DTE enzyme (using PDB: 2OU4 as a template) to identify key regions that must be preserved as a conservative design strategy (Table 1). D-Tagatose-3-Epimerase (DTEase) is a key enzyme catalyzing C3-position epimerization of various ketosides, holding significant value in rare sugar biosynthesis. This study uses DTEase from Pseudomonas cichorii (PDB: 2OU4) as the design template. This enzyme exhibits a typical metal-dependent (β/α)₈ barrel structure. Its active site comprises a Mn²⁺ ion coordinated by four highly conserved residues (Glu152, Asp185, His211, Glu246), responsible for substrate deprotonation and protonation. Furthermore, 2OU4 forms a stable homodimer in its native state. Its active pocket is located on one side of the dimer interface, exhibiting an open, loose substrate-binding characteristic that enables it to efficiently catalyze the C3 epimerization reactions of both D-tagatose and D-fructose simultaneously.

 D-tagatose-3-epimerase (DTEase): 2ou4 sequence:

 MNKVGMFYTYWSTEWMVDFPATAKRIAGLGFDLMEISLGEFHNLSDAKKRELKAVADDLGLTVMCCIGLKSEYDFASPDKSVRDAGTEYVKRLLDDCHLLGAPVFAGL TFCAWPQSPPLDMKDKRPYVDRAIESVRRVIKVAEDYGIIYALEVVNRFEQWLCNDAKEAIAFADAVDSPACKVQLDTFHMNIEETSFRDAILACKGKMGHFHLGEAN RLPPGEGRLPWDEIFGALKEIGYDGTIVMEPFMRKGGSVSRAVGVWRDMSNGATDEEMDERARRSLQFVRDKLA

 DOI：10.1016/j.jmb.2007.09.033;10.1002/cbic.201402620

Table 1. Key Functional Regions and Design Strategy for DTE

 Using 2OU4 as a template, we performed structure-based sequence prediction with LigandMPNN:

1. Designated Mn²⁺-coordinating residues A152/A185/A211/A246 and critical sites like the dimer interface as fixed regions (non-mutable), while classifying the remaining ~90% of residues as redesignable regions;
2. Enhanced Val/Ile/Leu in the hydrophobic core and Asp/Glu/Lys/Arg in surface charge regions via residue-specific preference optimization, while globally shielding free Cys to mitigate oxidation risks;
3. Ensured synchronous optimization of both chains using symmetry constraints, incorporating ligand side-chain environment sensing and packing algorithms to prevent collisions between new sequences and metal ions/substrates;
4. Through iterative refinement, we optimized surrounding residues within the overall framework of the new enzyme's renew series. From 32×2 samples, we rapidly screened relatively stable variants DTE-1 and DTE-2 using the Expasy ProtParam tool.

 We performed molecular docking of D-fructose into DTE-1 and DTE-2 using CB-Dock2. Both enzymes exhibited well-defined binding pockets with favorable Vina scores (–5.1 and –5.0 kcal·mol⁻¹, respectively). Template-based refinement using 2qum further confirmed substrate-binding capability, with contact scores of 12.4 (DTE-1) and 54.7 (DTE-2). Although DTE-1 binds in a shifted cavity and DTE-2 shows moderate pocket similarity to the reference enzyme 2OU4, both retain key interactions necessary for substrate recognition. Complementary docking via DiffDockfurther revealed that DTE-2 displays a clear preference for D-tagatose over D-fructose,combined with our findings from atomic-interaction analysis and molecular dynamics with binding free energy calculations, suggesting enhanced catalytic pairing propensity toward the target substrate. Collectively, these results imply that both DTE-1 and DTE-2 possess detectable catalytic activity.

2.2 Design of the DTE Enzyme Expression System

 To enable sustained DTE enzyme expression, we employed the high-copy pSEVA341 plasmid with a porin promoter, ensuring stable expression and functionality of the DTE enzyme in bacteria.

2.3 Design of an RNA-Pepper-Based Allulose Biosensor

  Validating DTE enzyme catalytic activity hinges on detecting allulose concentrations in the environment by coupling fluorescent signals to allulose levels. A 2019 study reported designing a GFP-fluorescent synthetic dye—(4-((2-hydroxyethyl)(methyl)amino)-subbenzylidene)cyanophenylacetonitrile (HBC), featuring a structurally rigid electron acceptor and a strong electron donor. Paired with the HBC-binding aptamer Pepper, it enhances HBC fluorescence by over 3000-fold in vitro and exhibits bright fluorescence when expressed in E. coli cells. Compared to other currently available fluorescent reporter groups, it offers an order-of-magnitude improvement in fluorescence intensity and activation factor, a one- to two-fold increase in affinity, enhanced thermal stability by approximately 20°C, superior pH tolerance, broader spectral range, and higher signal-to-noise ratio, providing an effective solution for characterizing allulose synthase activity. Therefore, based on the fluorescent RNA pepper-HBC system, we constructed a biosensor for DTE enzyme activity. By utilizing allulose to release the PsiRA repression effect on RNA pepper expression, we achieved biological detection of allulose concentration.

2.4 RNA-pepper-based DTE enzyme activity characterization system

 By integrating the two aforementioned systems, we constructed a dual-plasmid system to enable comparative analysis of different DTE enzyme activities.

 Using electroporation, p341-porin-DTE-WT, p341-porin-DTE-1, and p341-porin-DTE-2 were separately transformed into a strain containing pET28-porin-psiRA-pepper to establish the dual-plasmid system for enzyme activity characterization. Through the allulose-mediated PsiR repression release system, the allulose concentration was coupled with pepper expression levels and final fluorescence intensity to indirectly characterize the catalytic efficiency of DTE enzymes.

3. Build

3.1 Construction of the DTE enzyme expression system

 Utilizing the laboratory's existing DTE-WT sequence and PSEVA341 plasmid, we added restriction sites to both ends of the sequence via PCR. The resulting p341-porin-DTE-WT plasmid for DTE enzyme expression was constructed through Golden Gate assembly. We ordered the designed synthetic sequences for DTE1 and DTE2 from BGI-Hexa and obtained the p341-porin-DTE-1 and p341-porin-DTE-2 plasmids using the same method.

Schematic diagrams of DTE enzyme expression system plasmids(1:DTE-WT plasmid map; 2: DTE-1 plasmid map; 3: DTE-2 plasmid map)

3.2 Construction of an Allulose Biosensor Based on RNA-pepper

 Constructing the pET28-porin-psiRA-papper plasmid as a biosensor for DTE enzyme activity based on the fluorescent RNA pepper-HBC system. The pET28-porin-psiRA-papper plasmid was transformed into Escherichia coli BL21.

3.3 Construction of the RNA-pepper-based DTE enzyme activity characterization system

 Using electroporation, p341-porin-DTE-WT, p341-porin-DTE-1, and p341-porin-DTE-2 were separately transformed into strains containing pET28-porin-psiRA-papper to establish a dual-plasmid system for enzyme activity characterization.

 The successful construction of the above-mentioned systems was verified by colony PCR and DNA sequencing.

4. Test

4.1 DTE Enzyme Expression System Verification and Catalytic Activity Assessment

 To further confirm the catalytic activity of the DTE enzyme, we employed an Agilent 1260 liquid chromatograph for precise detection of allulose in the culture system. We first prepared monosaccharide and mixed sugar solutions of D-fructose and D-allulose, determining their retention times under experimental conditions. Subsequently, we tested E. coli DH5α not transformed with the DTE enzyme expression plasmid, confirming that conventional E. coli cannot catalyze the conversion of D-fructose to D-allulose.

 To confirm the system's proper expression function, RT-qPCR was employed to detect the expression levels of the three DTE enzymes, using E. coli DH5α without the plasmid as a control. Results indicated that all three systems functioned well, exhibiting significant DTE enzyme expression levels.

 Subsequently, we analyzed the post-reaction culture supernatants from single-plasmid strains harboring p341-porin-DTE-WT, p341-porin-DTE-1, and p341-porin-DTE-2. However, the results showed that the allulose content produced by all three strains was extremely low, despite previous fluorescence level measurements indicating that all three DTE enzymes possessed a certain degree of catalytic activity. This suggests potential issues with the reaction conditions.

4.2 Performance testing of the RNA-pepper-based allulose biosensor

 The successfully constructed pET28-porin-PsiRA-Pepper plasmid was efficiently transformed into E. coli BL21. The recombinant E. coli was cultured in LB medium at 37°C for 10-12 hours. Substrate D-allulose was added at various concentrations: 0.001, 0.01, 0.1, 0.1, 10, 100, 1000 mM/L. After 2 h of continued incubation, HBC dye was added. Changes in fluorescence intensity following the addition of different D-allulose concentrations were measured. Results demonstrated the system functions effectively, capable of detecting allulose concentration changes.

 Through induction with varying concentrations of allulose, we demonstrated that this biosensor exhibits a significant response to allulose within the concentration range of 10 to 1000 mM. The relationship between the output signal intensity and the input concentration approximately follows a sigmoidal curve, which is consistent with theoretical predictions and confirms the proper functioning of the system.

4.3 Detection Using an RNA-Pepper-Based DTE Enzyme Activity Characterization System

 Next, the activity of different DTE enzymes was characterized using a dual-plasmid system. First, the optimal fructose induction concentration was determined. The p341-porin-DTE-WT and pET28-porin-psiRA-pepper dual-plasmid system was selected. Catalytic reactions were conducted at 100, 200, 300, 400, and 500 mM/L. HBC buffer was added to the supernatant, and fluorescence levels were measured using a microplate reader. The optimal catalytic concentration was confirmed to be 300 mM/L. To determine reaction duration, we employed the dual-plasmid system of pET28-porin-psiRA-pepper and pSEVA341-porin-DTE-WT under 300 mM/L catalytic conditions, measuring fluorescence intensity at 2, 4, and 6 hours. Considering the degradation curve of pepper mRNA obtained in the dry experiment, we concluded that fluorescence intensity beyond 6 hours no longer accurately reflects actual allulose concentrations. Therefore, subsequent measurements will focus on monitoring fluorescence intensity changes during the 6-hour catalytic process.

 In a 300 mM/L fructose environment, monitoring the fluorescence intensity changes over time for different dual-plasmid systems revealed no significant differences in DTE enzyme activity during the 2-hour reaction. However, at 4 hours, the fluorescence intensity corresponding to the newly designed DTE enzyme showed a certain degree of reduction. At 6 hours, the fluorescence intensity of DTE1 exhibited a relatively significant increase compared to both WT and DTE2. It can be reasonably inferred that both the newly designed DTE1 and DTE2 enzymes possess catalytic activity, with DTE1 potentially exhibiting a slightly higher conversion rate than the WT (wild-type) DTE enzyme.

Through our analysis, we have successfully constructed the system for DTE enzyme expression and indirectly measured the activity of different DTE enzymes. According to the fluorescence data, the novel DTE enzyme DTE1, which was designed with the assistance of AI tools, showed a higher conversion rate after 6 hours of catalysis. However, due to the inherent instability of pepper RNA, definitive conclusions cannot be drawn based solely on fluorescence levels. It is necessary to further refine the liquid chromatography experimental protocol to obtain precise quantitative data, which is essential for accurately calculating key indicators such as the enzyme's conversion rate.

5. Learn

 Subsequent analysis revealed: although fluorescence data indicated catalytic activity across different DTE enzymes, alonose sensor measurements suggested all three enzymes produced alonose levels around 10 mM—far below the 30% conversion rate reported in prior literature. Therefore, we conclude that insufficient bacterial biomass and overall DTE enzyme quantity prevent the production of adequate allulose. Consequently, for industrial allulose production, considerations must extend beyond the catalytic activity of DTE enzymes to include fundamental factors such as enzyme expression levels and cell density.