Methods and systems for high-throughput screening of yeast secretions using molecular sensors on the surface of mother yeast cells

Abstract

A method for high-throughput screening of yeast metabolites is provided. The method includes providing a population of yeast cells and anchoring a plurality of molecular sensors to the cell membrane surface of a mother yeast cell to form a molecular sensor on the mother yeast cell membrane. The molecular sensor is selectively retained on the mother yeast cell during budding of the mother yeast cell without being transferred to daughter cells. The mother yeast cell is cultured under conditions that favor secretion of metabolites by the yeast cell, and the secreted metabolites are captured by the molecular sensor on the mother yeast cell membrane. A detectable signal resulting from an interaction between the molecular sensor and the captured metabolites is used to screen the yeast cells and identify cells exhibiting a desired metabolic secretion phenotype.

Description

Technical Field

[0001] This invention relates to the field of microbial screening technology. More specifically, this invention relates to a single-cell ultra-high-throughput screening method for yeast metabolites using membrane surface molecular sensors. Background Technology

[0002] Yeast plays a vital role in modern biotechnology, widely used in the production of various important metabolites in pharmaceuticals, food industries, environmental applications, and life sciences and biomedicine. Extracellular secretions of yeast are particularly important for key metabolic processes such as fermentation, chemical degradation, antibacterial activity, and drug development. Directed evolution of yeast strains typically relies on screening for mutants with superior performance from large-scale mutant libraries to achieve more efficient chemical production and pharmaceutical applications. This process requires the ability to target ultra-large-scale mutant libraries (10^6 MW). 6 -10 7 This approach offers a practical method for analyzing single-cell extracellular secretions in yeast mutants, thereby identifying rare, highly secretory strains. While molecular sensing technologies have emerged in recent years offering promising avenues for detecting intracellular compounds and extracellular secretions at the single-cell level, existing methods still struggle to achieve an ideal balance between sensitivity, throughput, and processing speed, thus significantly limiting their application in large-scale yeast metabolic secretion analysis and screening.

[0003] Immunosorbent assays offer a viable strategy for measuring yeast extracellular secretions by capturing target molecules for labeling and subsequent analysis. Enzyme-linked immunosorbent assays (ELISpot), developed from immunosorbent assays, produce visible spots for analyzing secretory activity at the cellular level. In recent years, microplate systems with nanoscale micropores have been developed to enable parallel single-cell assays. In these systems, secretory molecules are captured on the surface of glass slides pre-coated with antibodies or aptamers, allowing for multiplex analysis of secretory products. However, the throughput of these methods remains low, typically analyzing only about 10 samples per experiment. 3 -10 4 Single-cell secretion studies can be conducted using mass spectrometry techniques, such as gas chromatography-mass spectrometry (GC-MS) and high-performance liquid chromatography-mass spectrometry (HPLC-MS), to detect a variety of metabolites secreted by yeast. However, their detection throughput is extremely low, typically only about one cell per experiment, making them unsuitable for large-scale single-cell secretion studies.

[0004] Fluorescence-activated cell sorting technology offers a high-throughput alternative with an analysis speed of approximately 10. 3 -10 4Cells per second. However, traditional fluorescence-activated cell sorting technology mainly detects cell size, surface markers, and intracellular molecules, making it difficult to directly analyze extracellular secretions. Although intracellular staining methods can indirectly infer extracellular secretion characteristics, they are cumbersome, prone to artifacts, and may affect cell viability, thus they are not suitable for large-scale screening and identification of highly secretory yeast strains.

[0005] Currently, fluorescence-activated droplet sorting (FPSS) has become an important tool for high-throughput single-cell analysis. In FPSS systems, monodisperse water-in-oil droplets (approximately 10⁻⁴ pL) are generated using microfluidic droplet devices. Each droplet encapsulates a single cell and a chemiluminescence sensor. When a target metabolite is detected, a fluorescence signal is generated, thereby achieving single-cell detection and sorting based on metabolic activity. FPSS has been widely used in the field of metabolic engineering for screening yeast mutants, including the detection of intracellular molecules and extracellular secretions.

[0006] Intracellular single-cell molecular analysis is typically achieved by lysing cells within a droplet, releasing intracellular components, and then using a substrate reaction to generate a fluorescent signal to screen for superior enzyme variants. However, this method cannot recover live cells for subsequent functional analysis. Another approach assesses intracellular enzyme activity using cell surface display strategies, where enzymes representing intracellular molecules are expressed on the cell surface. These enzymes interact with substrates within the droplet, generating fluorescent signals, thus enabling directed evolution. However, this method requires genetic engineering of the cells.

[0007] In recent years, fluorescence-activated droplet sorting technology has also been used to directly measure extracellular secretions, including enzymes and small molecule metabolites, via intradroplet enzyme-coupled reactions. However, this method still faces several challenges in large-scale screening of secretory strains. First, its versatility is limited; detection typically relies on specific enzymatic reactions, thus restricting the types of metabolites that can be detected. Many valuable natural products, such as terpenes and phenolic compounds, remain difficult to detect using droplet enzymatic methods. Currently, only a few secretions, such as α-amylase, lactic acid, and ethanol, can be successfully measured.

[0008] Secondly, detection sensitivity remains a significant challenge. Although droplet enzyme reactions exhibit high sensitivity for certain low-abundance molecules, existing methods primarily focus on the detection of intracellular molecules. For extracellular secretions, except in rare cases where the detection limit for α-amylase can be as low as approximately 0.1 nM, the detection limit for most metabolic molecules remains around 10 μM.

[0009] Furthermore, detection throughput and screening speed remain limited. Single-cell encapsulation efficiency is typically low (<10%), resulting in processing speeds of only about 10-200 cells / second, while droplet screening rates are generally 10. 2 -10 3The droplet-per-second (PPS) limit makes fluorescence-activated droplet sorting technology difficult to apply to screening of ultra-large-scale mutant libraries.

[0010] The recently developed RNA-Aptamer-in-Droplet (RAPID) screening technology addresses some of the problems of traditional fluorescence-activated droplet sorting techniques by introducing suspended RNA aptamer sensors that generate fluorescence signals upon binding to target molecules. This method offers flexibility and potential for multiplex detection. However, it still suffers from limited sensitivity and insufficient stability of RNA aptamers during co-culture with yeast, leading to false positives, and a relatively high detection limit (approximately 260 μM).

[0011] In addition, there are methods that co-encapsulate genetically engineered live cell sensors with producing bacteria in droplets for the detection of secreted metabolites. Although this method has functional detection capabilities, its application in high-throughput metabolic screening is still limited due to the complexity of strain co-culture and low detection sensitivity (e.g., naringenin at approximately 70 μM).

[0012] More recent droplet graphene oxide (GO) aptamer sensors have improved detection performance by immobilizing aptamers on the surface of graphene oxide. This strategy reduces nonspecific binding, decreases false positives, and increases detection sensitivity to approximately 10 μM. However, its overall screening throughput remains limited by the low probability of single-cell droplet encapsulation (approximately 1-10%) and the limited processing speed (approximately 10 cells / second).

[0013] In summary, existing technologies for yeast secretion analysis have long suffered from a trade-off between sensitivity, throughput, and speed, making it difficult to simultaneously achieve all three. This severely limits the scalability and versatility of high-throughput, large-scale yeast variant screening platforms. Therefore, there is an urgent need to develop a new technical solution to address these issues. Summary of the Invention

[0014] The purpose of this invention is to provide systems and methods to solve the aforementioned technical problems.

[0015] According to a first aspect of the present invention, a method for ultra-high-throughput screening of yeast metabolites is provided. The method includes the following steps: providing a population of mother yeast cells; anchoring a plurality of molecular sensors to the cell membrane surface of the mother yeast cells to form molecular sensors on the surface of the mother yeast cell membrane, wherein the molecular sensors are selectively retained on the mother yeast cells during budding and are not transferred to daughter cells; culturing the mother yeast cells under conditions conducive to the secretion of metabolites by the mother yeast cells; capturing the secreted metabolites using the molecular sensors located on the surface of the mother yeast cell membrane; generating a detectable signal upon interaction between the molecular sensors and the captured metabolites; and screening the mother yeast cells based on the detectable signal to identify mother yeast cells exhibiting a metabolic secretion phenotype. In some embodiments, the secreted metabolites include target metabolites and / or proteins.

[0016] According to one embodiment, the molecular sensor includes a nucleic acid aptamer sensor capable of binding to the target metabolite and / or protein.

[0017] According to another embodiment, the aptamer sensor is immobilized on the cell membrane surface of the mother yeast cell through biotin-streptavidin interaction.

[0018] According to another embodiment, the density of the molecular sensor on the surface of each mother yeast cell is at least 10. 6 A molecular sensor.

[0019] According to another embodiment, the molecular sensor generates a fluorescent signal upon binding to secreted metabolites.

[0020] According to another embodiment, the method further includes encapsulating a mother yeast cell with molecular sensors on its individual cell surface in a microfluidic droplet to capture secreted metabolites within the droplet.

[0021] According to another embodiment, the method further includes breaking the droplet after culturing to release the mother yeast cell for subsequent screening.

[0022] According to another embodiment, the molecular sensor includes a structure-transformation aptamer that generates fluorescence upon binding to a target molecule.

[0023] According to another embodiment, the target metabolite includes vanillin, adenosine triphosphate, glucose, and zinc ions.

[0024] According to another embodiment, the target protein includes interferon-γ and glycosylated human serum albumin.

[0025] According to another embodiment, the mother yeast cell includes Saccharomyces cerevisiae and Pichia pastoris.

[0026] According to another embodiment, the method further includes constructing a yeast mutant library and performing directed evolution by screening yeast variants with enhanced secretion phenotypes.

[0027] According to another embodiment, the method further includes grafting different molecular sensors for different molecules onto the surface of the mother yeast cell membrane to achieve multiplex detection of multiple metabolites.

[0028] According to a second aspect of the invention, a functionalized yeast cell for detecting extracellular secretions is provided. The functionalized yeast cell includes at least one mother yeast cell having a membrane surface; and one or more molecular sensors fixed to the membrane surface of the mother yeast cell. The molecular sensors are configured to bind to secreted molecules produced by the functionalized yeast cell and generate a detectable signal, and the molecular sensors remain positioned on the mother yeast cell during budding and substantially do not migrate to any daughter cells.

[0029] According to one embodiment, the molecular sensor includes an aptamer sensor.

[0030] According to another embodiment, the molecular sensor is immobilized in the mother yeast cell via biotin-streptavidin binding.

[0031] According to another embodiment, the molecular sensor includes a fluorescence-based structure-transition aptamer.

[0032] According to a third aspect of the present invention, a system for ultra-high-throughput screening of microbial secretions is provided. The system includes: the aforementioned functionalized yeast cells; a droplet microfluidic device configured to encapsulate a single functionalized yeast cell within a droplet; a culture module configured to culture the functionalized yeast cells to secrete metabolites and capture them by the molecular sensor; and a detection and sorting device configured to detect the signal generated by the molecular sensor and separate yeast cells having a target secretion phenotype. Specifically, the system is configured to screen at least 10 microbial secretions in a single run. 7 The system is configured to use at least 10 yeast cells. 3 The cells are sorted at a rate of 1 cell per second.

[0033] According to a fourth aspect of the invention, a mutant enoyl-CoA hydratase / aldolase (ECH) enzyme is provided, which, relative to the wild-type enoyl-CoA hydratase / aldolase, includes at least one amino acid substitution selected from I90N, Y169C, and N212Q. Attached Figure Description

[0034] Referring to the accompanying drawings, embodiments of the present invention will be described in more detail below, wherein:

[0035] Figure 1A-1D This demonstrates the application of molecular sensors on the surface of the mother yeast cell membrane in the detection of secretions in a single yeast cell, wherein: Figure 1A The illustration shows that molecular sensors located on the surface of the mother yeast cell membrane can selectively bind to the mother yeast cell and are not transmitted to the daughter cells during budding. Figure 1B The demonstration showcases the advantages of selectively coating mother yeast cells with sensors. The molecular sensor platform located on the surface of the mother yeast cell membrane exhibits high-sensitivity detection, high-throughput screening capabilities, and ultra-high-speed sorting performance, achieving over 10 [units / items] in approximately 40 minutes. 7 Analysis of individual yeast cells; Figure 1C The performance of a molecular sensor platform located on the surface of the mother yeast cell membrane was compared with that of various droplet-based screening methods. The results showed that the molecular sensor platform on the surface of the mother yeast cell membrane improved the sensitivity of extracellular secretion detection by more than 10 times and the speed of identifying and sorting secretory strains by more than 30 times; and Figure 1D A comparative analysis of various microbial extracellular secretion analysis techniques was presented, showing that the molecular sensor platform located on the surface of the mother yeast cell membrane achieves the best balance between sensitivity, screening throughput and sorting speed, and can be used to detect a variety of metabolic molecules secreted by a single microbial cell.

[0036] Figure 2A-2C This demonstrates that molecular sensors located on the surface of the mother yeast cell membrane do indeed exist on the yeast cell surface, wherein: Figure 2A Visualization of the binding of yeast cell walls to Alexa Fluor 488-tagged concanavalin A; Figure 2B Demonstrates that molecular sensors can be observed on the cell membrane surface of a single yeast cell using Cy5 fluorescence; and Figure 2C The combined image of Alexa Fluor 488 and Cy5 fluorescence shows that molecular sensors on the surface of the mother yeast cell membrane are only covered on the surface of the yeast cell.

[0037] Figures 3A-3D The study demonstrated the cell viability of yeast cells after being coated with biotin-VANapt-Cy5 and then treated with fluorescein diacetate, including: Figure 3A The coating of molecular sensors on the surface of the mother yeast cell membrane was observed under a confocal laser scanning microscope after Cy5 fluorescent labeling, confirming that the sensors were successfully coated on the yeast cells. Figure 3B Demonstrates the use of FDA live / dead staining to assess cell viability under a confocal laser scanning microscope; Figure 3C Showing a combined image of Cy5 fluorescence and stained fluorescein fluorescence; and Figure 3DThe flow cytometry analysis results show that the survival rate of the mother yeast cells with molecular sensors on their surface is 93%, demonstrating biocompatibility.

[0038] Figures 4A-4B Demonstrating the biocompatibility of molecular sensors on the surface of mother yeast cell membranes, including: Figure 4A The cell growth curve test shows that the molecular sensors on the surface of the mother yeast cell membrane do not affect the growth of yeast cells from different strains. Figure 4B The vanillin yield of mother yeast cells of wild-type and VAN-3 yeast strains was tested using the existing absorbance method under conditions with and without membrane surface molecular sensor coating.

[0039] Figure 5A-5L This demonstrates highly sensitive multiplex detection achieved through high-density surface molecular sensor coating, where: Figure 5A The study demonstrates how a biotin-tagged aptamer can be attached to the surface of yeast via a biotin-streptavidin binding strategy, thereby forming a molecular sensor on the surface of the parent yeast cell membrane. Figure 5B The study demonstrates the observation of Cy5-labeled membrane surface molecular sensors during budding of yeast cells stained with Alexa Fluor 488-ConA, proving that the membrane surface molecular sensors are localized only in the mother yeast cells. Figure 5C The study showed that yeast cells cultured with 0-3 μM membrane surface molecular sensors formed different coating densities, reaching saturation at 1.5 μM. Figure 5D The study showed that after more than 30 generations (3 days), about 77% of the membrane surface molecular sensors were still retained in the mother yeast cells, while the daughter cells had no membrane surface molecular sensors. Figure 5E Demonstrates a high-density surface molecular sensor (approximately 1.0 × 10⁻⁶). 7 (Number of cells / cell) persists on the mother yeast cell for 72 hours, although the proportion of mother cells decreases during reproduction; Figure 5F This demonstrates that high-density cell membrane surface molecular sensors enable highly sensitive multiplex detection in mother yeast cells; Figure 5G The display density is approximately 1.4 × 10⁻⁶ cells per cell. 7 The membrane surface molecular sensor of the aptamer has a higher detection sensitivity than the existing droplet-type aptamer sensor, with a detection sensitivity increase of more than 100 times and a lower detection limit of 100 nM from 50 μM. Figure 5H The high sensitivity of molecular sensors displayed on the cell membrane surface allows for the direct detection of ATP secreted by a single yeast cell; Figure 5I A molecular sensor with a membrane surface coated with targets vanillin, ATP and glucose is shown to enable multiple detection; Figure 5J The molecular sensor on the membrane surface is shown to be coated with yeast cells and vanillin (1.0 mM), ATP (1.0 mM), glucose (1.5 M), or Zn.2+ After (2.0 mM), the fluorescence intensity increased with time; Figure 5K The confocal microscope image shows the simultaneous triple aptamer coating on yeast cells to enable multiplex detection; Figure 5L The heatmap verification results confirm that the multiplex metabolite detection has 100% specificity;

[0040] Figure 6 The fluorescence calibration curves used to quantify aptamer concentrations are shown.

[0041] Figures 7A-7B This demonstrates a gating strategy for identifying mother yeast cells in flow cytometry analysis, including: Figure 7A This demonstrates how initial gating based on FSC-A and SSC-A signals can be used to screen for initial yeast cell populations from debris. Figure 7B The demonstration uses a negative control sample (unlabeled yeast cells) to establish a second gating mechanism to further distinguish between parent yeast cells and daughter cells coated with fluorescently labeled cell membrane surface molecular sensors;

[0042] Figure 8 Demonstrates the coating of molecular sensors on the cell membrane surface during yeast budding;

[0043] Figures 9A-9B The analysis of bud scars is shown, including: Figure 9A The aptamer sensor was shown to remain permanently localized to the mother cell, and multiple bud scars were observed, confirming that the sensor was retained after continuous division. Figure 9B The results of quantitative image analysis show that after multiple rounds of cell division within 24 hours, approximately 87% of the surface binding sensors remained stably anchored to the mother cell.

[0044] Figures 10A-10C This demonstrates a specific assay for detecting vanillin, glucose, and ATP using a cell membrane surface molecular sensor, including: Figure 10A The specificity of vanillin was assessed by comparing the responses of vanillin, vanillyl alcohol, vanillic acid, and ferulic acid. Figure 10B Demonstrates glucose specificity testing through a comparison of fructose and galactose; Figure 10C The demonstration confirms ATP specificity by comparing it with GTP, UTP, and CTP;

[0045] Figure 11 This demonstrates a sensitivity comparison between cell membrane surface molecular sensors and free suspended aptamer sensors;

[0046] Figures 12A-12C This demonstrates highly sensitive secretion measurements over time, including: Figure 12A This illustration demonstrates how cell membrane surface molecular sensors coated on yeast cells directly capture their own secretions on the cell surface. Figure 12BThe image shows a time-series image of ATP secretion in a single yeast cell captured using a CLSM-monitored cell membrane surface molecular sensor. Figure 12C The results of flow cytometry analysis show that yeast ATP secretion gradually increases over time;

[0047] Figures 13A-13D Demonstrates molecular detection at different concentrations, including: Figure 13A The experiment demonstrates vanillin detection, which uses a cell membrane surface molecular sensor to coat yeast cells to detect vanillin concentration in the range of 0 to 1.00 mM. Figure 13B The vanillin detection kinetics are shown, namely, monitoring changes in fluorescence intensity over 0 to 5 hours to assess the vanillin detection kinetics; Figure 13C An ATP detection experiment was shown, in which yeast cells coated with cell membrane surface molecular sensors were used to detect ATP concentrations in the range of 0 to 4.0 mM. Figure 13D An experiment for glucose detection was demonstrated, in which yeast cells coated with a cell membrane surface molecular sensor were used to detect glucose concentrations in the range of 0 to 2.0 M.

[0048] Figures 14A-14D Demonstrates a composite cell membrane surface molecular sensor coating on yeast cells, wherein: Figure 14A This demonstrates a cell membrane surface molecular sensor containing biotin-ATPapt-Cy5 coated on a yeast cell; Figure 14B A cell membrane surface molecular sensor containing biotin-VANapt-FITC is shown coated in yeast cells; Figure 14C This demonstrates a cell membrane surface molecular sensor containing biotin-GLUapt-Cy3 coated on a yeast cell; Figure 14D The combined fluorescence channel images show that different cell membrane surface molecular sensors can be simultaneously coated in the same yeast cell;

[0049] Figure 15A-15G High-throughput filtering is demonstrated, including: Figure 15A This demonstrates high-throughput screening using cell membrane surface molecular sensors, capturing single yeast cells and screening secretory strains using flow cytometry. Figure 15B Demonstrates that a cell membrane surface molecular sensor removes droplets before screening, enabling conventional flow cytometry (10⁻⁶) even under low droplet encapsulation conditions. 3 -10 4 Analyze using Hz; Figure 15C The monitoring results of vanillin secretion (VAN-3 > VAN-2 > VAN-1 ≫ WT) over time are shown, indicating that the signal gradually amplifies during the culture process; Figure 15D This demonstrates the formation of two populations when vanillin and ATP secretion are measured simultaneously: VAN-3 and a negative control. Figure 15EThis demonstrates that all yeast strains (VAN-1, VAN-2, VAN-3, and WT) share similar ATP secretion characteristics. Figure 15F When detecting vanillin secretion in a mixed yeast sample (VAN-3:WT=1:1), coupled detection can identify three populations (VAN-3, WT, and daughter cells); Figure 15G The different fluorescence peaks of the mixed wild-type and VAN-3 samples are shown to be used to sort VAN-3 strains;

[0050] Figures 16A-16C The selective cell membrane surface molecular sensor coating of mother yeast cells in a droplet is demonstrated, wherein: Figure 16A This demonstrates the budding process of yeast cells coated with cell membrane surface molecular sensors in droplets, observed using bright-field microscopy and fluorescence microscopy. Figure 16B The flow cytometry results demonstrate that the molecular sensor coating on the cell membrane surface exhibits high stability. Figure 16C The flow cytometry analysis results show that although the proportion of mother cells in the overall cell population decreases over time, the density of molecular sensors on the cell membrane surface of the mother yeast cells only decreases slightly.

[0051] Figures 17A-17B The metabolic pathway for vanillin production by engineered yeast is demonstrated, including: Figure 17A A schematic diagram illustrating the pathway of vanillin biosynthesis from ferulic acid; Figure 17B The gene flow diagram of plasmid pCYP-VAN is shown, which carries the 4cl and ech genes, which are essential for vanillin synthesis.

[0052] Figure 18 Saturation analysis of molecular sensor signals on the cell membrane surface shows that the fluorescence intensity on the surface of the mother cell increases proportionally with the increase of vanillin concentration in the droplet;

[0053] Figure 19A-19G Demonstrates high-speed sorting for rapid directed evolution, where: Figure 19A Demonstrates that cell membrane surface molecular sensors can efficiently screen mixed populations; Figure 19B This demonstrates the use of cell membrane surface molecular sensors to screen ECH enzyme mutant libraries for enzyme evolution. Figure 19C Time-series high-performance liquid chromatography analysis of fermentation supernatant of VAN-3 and the selected mutant strains is presented; Figure 19D A schematic diagram of the ECH mutants screened from strain S-10 is shown, which have amino acid substitutions of I90N, Y169C, N212Q and P213L. Figure 19E Demonstrates the use of cell membrane surface molecular sensors to screen 2×10 7 A library of yeast mutant strains was used to isolate high-performance strains with enhanced extracellular vanillin secretion capabilities. Figure 19FFour high-performance transporter protein variants were demonstrated and validated through batch fermentation. Figure 19G Showing two key mutations, K40R and V79D, identified in the optimal strain;

[0054] Figure 20 Demonstrates verification of vanillin production through absorbance measurement;

[0055] Figure 21 The time-series measurement results of vanillin concentration in the fermentation supernatant of VAN-3 and mutant strains (S-1, S-8 and S-10) are presented.

[0056] Figures 22A-22B The additional mutation sites of the ECH variants in the screened strains are shown, among which: Figure 22A A schematic diagram of the screening ECH mutant from strain S-1 with Y75H substitution is shown; Figure 22B A schematic diagram of the screening ECH mutant from strain S-8 with D80V substitution is shown;

[0057] Figure 23 Demonstrates the verification of ECH enzyme activity through vanillin production in E. coli;

[0058] Figures 24A-24C A quantitative analysis of ECH enzyme activity was presented, including: Figure 24A ECH enzyme was purified from E. coli strains expressing wild-type or mutant ech genes for quantification of enzyme activity; Figure 24B The purified enzyme (1 mg / mL) at the same concentration was incubated with 300 μM feruloyl-CoA and 7.5 μM β-NAD at 30 °C for 12 hours. Figure 24C Demonstrates enzyme activity calculated based on vanillin yield;

[0059] Figure 25 The results demonstrate the validation of the vanillin secretion ability of the selected transporter protein mutant strains;

[0060] Figure 26 Assessment of fluorescence signal variability related to yeast cell size and budding status;

[0061] Figures 27A-27B A robust gating strategy is demonstrated to reduce false fluorescence results caused by differences in yeast cell size and budding, wherein: Figure 27A High-performance strains were isolated from the first 0.05% of the vanillin transporter mutant library; Figure 27B The distribution of FSC-A signals in sorted and unsorted yeast populations is compared.

[0062] Figures 28A-28C The additional mutation sites of the PP_0179 vanillin transporter variant in the screened strains are shown, including: Figure 28A A schematic diagram showing the screening of the PP_0179 mutant from strain T-1, which has a G45C substitution; Figure 28B A schematic diagram of the screening PP_0179 mutant from strain T-2 with K48M substitution is shown. Figure 28C A schematic diagram of the PP_0179 mutant selected from strain T-3, which has T15I / G45R / Q98H substitutions is shown.

[0063] Figures 29A-29D This demonstrates the high flexibility of cell membrane surface molecular sensors for protein analysis, including: Figure 29A Demonstrates the dynamic detection range of IFN-γ from 1 to 1000 nM using aptamer-functionalized cell membrane surface molecular sensors; Figure 29B The specificity of the gHSA-surface molecular sensor was evaluated by comparing the responses of glycosylated human serum albumin, bovine serum albumin, and non-glycosylated human serum albumin. Figure 29C This demonstrates the preparation of antibody-functionalized cell membrane surface molecular sensors by coating biotinylated yeast cells with streptavidin-modified IFN-γ capture antibodies. Figure 29D This demonstrates the use of antibody-functionalized cell membrane surface molecular sensors, cultured with different concentrations of IFN-γ, and the detection of antibody-labeled captured proteins using APC labeling to generate quantitative fluorescence signals; and

[0064] Figure 30 This document demonstrates an example of a cell membrane surface molecular sensor platform adapted for use in Pichia pastoris cells. Detailed Implementation

[0065] In the following description, the systems and / or methods are listed as preferred embodiments. It will be apparent to those skilled in the art that modifications, including additions and / or substitutions, can be made without departing from the scope and spirit of the invention. Specific details may be omitted below to avoid obscuring the invention; however, this document is prepared to enable those skilled in the art to practice the techniques disclosed herein without excessive experimentation.

[0066] This invention provides an advanced molecular sensing platform, specifically a molecular sensor for the surface of a mother yeast cell membrane, for large-scale analysis of extracellular secretions from single yeast cells. This mother yeast cell membrane surface molecular sensor system employs an aptamer-based molecular sensor that selectively anchors to the cell membrane surface of the mother yeast cell. Figure 1A Furthermore, these molecules do not transfer to daughter cells during budding. This selective localization allows for the formation of a high-density molecular sensor coating on the surface of the mother cell, reaching approximately 1.4 × 10⁻⁶ per cell. 7This technology utilizes multiple sensors to enable precise detection and quantitative analysis of molecules secreted by individual yeast cells.

[0067] This yeast cell membrane surface molecular sensor system offers several key advantages. First, it exhibits high sensitivity, capable of detecting secreted metabolites at concentrations as low as approximately 100 nM (lower limit of detection). Second, the platform boasts high throughput, enabling efficient analysis of over 10... 7 Third, the system has a high processing speed, capable of processing approximately 3.0 × 10⁻⁶ single yeast cells. 3 A screening rate of [number] cells / second, capable of self-selecting cells containing approximately 2.2 × 10 [units] within approximately 12 minutes. 6 Rare, highly secretory strains (approximately 0.05%) were isolated from a large mutant library of several variant strains. Figure 1B ).

[0068] Compared to existing advanced droplet screening technologies, this yeast cell membrane surface molecular sensor platform exhibits significant performance improvements, including over 10-fold increased sensitivity, over 2-fold increased throughput, and over 30-fold increased processing speed when screening extracellular secretions from individual yeast cells. It is also suitable for the detection of a variety of metabolic compounds. Figure 1C The molecular sensor on the surface of the mother yeast cell membrane was compared and evaluated with other analytical techniques used for the analysis of microbial extracellular secretion. Figure 1D (See Table 1). The comparison results show that the yeast cell membrane surface molecular sensor platform achieves a better balance between sensitivity, screening throughput and sorting speed.

[0069] To demonstrate the practical application value of this mother yeast cell membrane surface molecular sensor, this invention applies the system to the directed evolution of yeast strains to enhance metabolite production. Using this method, an optimized vanillin-producing yeast strain was successfully screened, exhibiting a secretion rate more than 2.7 times higher than the parent strain. Therefore, the mother yeast cell membrane surface molecular sensor platform represents an advanced molecular sensing system for yeast metabolic analysis, combining excellent sensitivity, speed, and throughput. This platform can serve as a powerful tool for large-scale screening of extracellular secretions from single yeast cells and significantly expands its application capabilities in metabolic analysis and biomanufacturing.

[0070] Table 1: Comparative Analysis of Major Techniques for Single Yeast Cell Secretion Analysis

[0071] According to a first aspect of the present invention, a method for ultra-high-throughput screening of yeast metabolism is provided, comprising providing a population of mother yeast cells and anchoring a plurality of molecular sensors to the cell membrane surface of the mother yeast cells to form a mother cell membrane surface molecular sensor. The molecular sensors are selectively retained on the mother yeast cells during budding and do not migrate to daughter cells, thereby maintaining a stable population of sensors on the mother cell surface during continuous cell division. The mother yeast cells are cultured under conditions where yeast cells secrete metabolites. The secreted metabolites can be captured by the molecular sensors anchored to the mother yeast cell membrane surface. Upon interaction with the captured metabolites, the molecular sensors generate a detectable signal, which is then used to screen the yeast cells to identify cells with a target metabolic secretion phenotype. In some embodiments, the secreted metabolites include target metabolites and / or proteins.

[0072] In some embodiments, the secreted metabolite includes a target metabolite and a protein, and the molecular sensor includes a nucleic acid aptamer sensor that binds to the target metabolite and the protein. The aptamer sensor can be immobilized on the cell membrane surface of the mother yeast cell via biotin-streptavidin interaction, thereby achieving stable and high-density attachment of the molecular sensor to the cell surface. In some embodiments, the molecular sensor has at least 10 [units / units] on the surface of each yeast cell. 6 The density of individual sensors is fixed, thereby enhancing detection sensitivity by increasing the probability of capturing secreted molecules in the vicinity of the cell surface. In some embodiments, the molecular sensor generates a fluorescence signal upon binding to the secreted metabolite, thereby enabling optical detection and quantitative measurement of the secretory activity of a single yeast cell.

[0073] In some embodiments, the method further includes encapsulating a single yeast cell coated with cell membrane surface molecular sensors in a microfluidic droplet to capture secreted metabolites within the droplet. The droplet provides an isolated microenvironment to prevent cross-contamination between adjacent cells and promotes the accumulation of secreted products near the cell that produced the secretion. Upon culturing, the droplet can be disrupted to release the yeast cell for subsequent screening, such as fluorescence-based detection or sorting methods.

[0074] In some embodiments, the molecular sensor includes a structure-transforming aptamer that emits fluorescence upon binding to a target molecule. The aptamer undergoes a conformational change upon interaction with the target molecule, thereby generating a detectable fluorescent signal. In some embodiments, the target metabolite includes vanillin, ATP, glucose, and zinc ions. In other embodiments, the target protein includes interferon-γ and glycosylated human serum albumin. This method is applicable to yeast species including *Saccharomyces cerevisiae* or *Pichia pastoris*, as well as other budding yeasts.

[0075] In some embodiments, the method includes constructing a yeast mutant library and performing directed evolution by screening yeast variants with enhanced secretion phenotypes. These variants can be identified by analyzing detectable signals generated by the molecular sensor, thereby enabling the efficient enrichment of high-performance strains from large yeast populations. In some embodiments, multiplex detection of multiple metabolites is achieved by anchoring different molecular sensors targeting different molecules to the cell membrane surface of the mother yeast cell. This multiplex configuration allows for the simultaneous monitoring of multiple compounds secreted by a single yeast cell, thereby expanding the analytical capabilities of the screening platform.

[0076] According to a second aspect of the invention, a functionalized yeast cell for detecting extracellular secretions is provided. The functionalized yeast cell includes a mother yeast cell having a cell membrane surface and a plurality of molecular sensors immobilized on the cell membrane surface of the mother yeast cell. The molecular sensors can bind to secretory molecules produced by the mother yeast cell and generate a detectable signal upon interaction with the secretory molecules. By immobilizing the molecular sensors on the cell membrane surface of the mother yeast cell, secretory molecules released by the yeast cell can be captured near their source, thereby achieving highly sensitive detection of extracellular secretions. Notably, the molecular sensors remain localized on the mother yeast cell during budding and are substantially not transferred to daughter cells, maintaining a stable sensor density in the mother yeast cell during multiple rounds of cell division. This selective localization ensures that the signal generation is specific to correspond to the secretory phenomenon of the mother yeast cell and avoids dilution of the sensor density during cell proliferation.

[0077] In some embodiments, the molecular sensor includes an aptamer sensor capable of specifically recognizing target molecules. These aptamer sensors can be designed to bind to a variety of metabolites or proteins secreted by yeast cells, thereby enabling selective detection of a variety of extracellular molecules. In some embodiments, the molecular sensor is immobilized on the cell membrane surface of the mother yeast cell via biotin-streptavidin binding, which provides a stable and high-affinity interaction to anchor the sensor to the cell surface. Utilizing the biotin-streptavidin interaction facilitates efficient sensor anchoring while maintaining its functional conformation and molecular recognition capabilities.

[0078] In a further embodiment, the molecular sensor includes a fluorescence-based structure-transition aptamer. This aptamer undergoes a conformational change upon binding to a target molecule, thereby generating a detectable fluorescent signal. This fluorescence-based detection mechanism allows for real-time monitoring and quantitative analysis of extracellular secretions at the single-cell level. By employing a structure-transition aptamer as a molecular sensor, this functionalized yeast cell can serve as a highly sensitive platform for detecting extracellular metabolites and proteins, while maintaining the physiological activity and viability of the yeast cell.

[0079] According to a third aspect of the invention, a system for ultra-high-throughput screening of microbial secretions is provided. The system includes at least one of the aforementioned functionalized yeast cells, wherein the functionalized yeast cell comprises a mother yeast cell having a cell membrane surface, and a plurality of molecular sensors immobilized on the surface of the mother yeast cell membrane. The molecular sensors can bind secreted molecules produced by the yeast cell and generate detectable signals upon interaction with the captured molecules. In this system, the molecular sensors remain localized on the mother yeast cell during budding and substantially do not migrate to daughter cells, thereby maintaining a stable sensing interface on the mother yeast cell to monitor extracellular secretion events.

[0080] The system further includes a droplet microfluidic device to encapsulate a single yeast cell within a droplet. This droplet microfluidic device generates discrete droplets, serving as an isolated microenvironment for each individual yeast cell, thereby enabling single-cell secretion analysis. By encapsulating a single yeast cell within a droplet, the secreted molecules can accumulate within the confined droplet volume and be efficiently captured by molecular sensors immobilized on the surface of the mother yeast cell membrane.

[0081] The system also includes a culture module to enable molecular sensors to capture secreted metabolites. During culture, the encapsulated yeast cells are maintained under conditions suitable for cell growth and metabolic activity, thereby allowing the yeast cells to secrete metabolites or proteins. Molecular sensors immobilized on the surface of the mother yeast cell membrane capture the secreted molecules and generate a detectable signal corresponding to the secretory activity of a single cell.

[0082] The system further includes a detection and sorting device to detect the signal generated by the molecular sensor and separate yeast cells exhibiting the target secretion phenotype. This detection and sorting device analyzes the detectable signal generated by the molecular sensor and identifies yeast cells exhibiting enhanced or other targeted secretion characteristics. Based on the detected signal, the system can separate or sort yeast cells with the target secretion phenotype for further analysis, culture, or directed evolution.

[0083] This molecular sensor includes an aptamer sensor that binds to secreted metabolites. These aptamer sensors provide selective molecular recognition of target metabolites, enabling the system to detect specific metabolites produced by yeast cells. By employing aptamer sensors as molecular sensing elements, this system can achieve highly sensitive and selective detection of extracellular metabolites while supporting ultra-high-throughput screening of microbial secretion phenotypes.

[0084] Therefore, the system is able to achieve at least 10 7 Screening of yeast cells, with at least 10 3 Yeast cells are sorted at a rate of cells per second.

[0085] According to a fourth aspect of the invention, a mutant enoyl-CoA hydratase / aldolase (ECH) is provided, which, relative to the wild-type enoyl-CoA hydratase / aldolase, includes at least one amino acid substitution selected from I90N, Y169C, and N212Q.

[0086] Example

[0087] Table 2: Primers used in the examples

[0088] Example 1: Preparation of molecular sensors on the cell membrane surface of mother yeast cells

[0089] The preparation of molecular sensors on the surface of yeast cell membranes involves three main steps: (1) biotinylation of yeast cell surface; (2) preparation of aptamer-streptavidin complex; and (3) coating the aptamer complex onto the biotin-treated yeast cell surface.

[0090] In the first step, biotin molecules are covalently coupled to proteins on the surface of yeast cells. Yeast cells (2.0 x 10⁻⁶) 6One cell was suspended in 90 μL of Durbeco phosphate-buffered saline (DPBS) containing 5 mM magnesium chloride, followed by the addition of 10 μL of 10 mM sulfo-NHS-LC-biotin dissolved in DMSO. The reaction mixture was incubated at room temperature (approximately 25°C) with shaking at 150 rpm for 30 minutes to achieve surface biotinylation. The resulting biotinylated yeast cells were then washed three times with 150 μL of Durbeco phosphate-buffered saline containing 5 mM magnesium chloride to remove excess reagent.

[0091] In the second step, an aptamer-streptavidin complex was prepared by mixing biotin-labeled aptamers and streptavidin at a predetermined molar ratio (aptamer:streptavidin = 3:1). An aptamer solution (0.5, 1.0, 1.5, or 3.0 μL of 100 μM solution) was added to 99 μL of 20 μM streptavidin solution containing 5 mM magnesium chloride in Dürbeco phosphate buffer. The mixture was incubated at room temperature for approximately 5–10 minutes to form the aptamer-streptavidin complex. To stabilize the complex and reduce nonspecific interactions, 1 μL of herring sperm DNA at a concentration of 10 mg / mL was subsequently added.

[0092] In the third step, the prepared aptamer-streptavidin complex was coated onto the surface of biotinylated yeast cells. The biotinylated yeast cells were resuspended in 100 μL of the aptamer-streptavidin solution and cultured at room temperature with shaking at 150 rpm for 30 minutes. After culture, the cells were washed three times with 150 μL of Durbeco phosphate buffer containing 5 mM magnesium chloride. The resulting functionalized yeast cells were then ready for secretion assays.

[0093] To enable the detection of metabolic compounds, a double-stranded DNA aptamer sensor was introduced into a molecular sensor system on the surface of the yeast cell membrane. First, biotin- and fluorescently labeled aptamers (e.g., biotin-Xapt-Cy5, biotin-Xapt-Cy3, or biotin-Xapt-FITC, where X represents the target molecule) were hybridized with complementary DNA strands labeled with quenching groups (e.g., BHQ3 for Cy5 quenching, BHQ2 for Cy3 quenching, and BHQ1 for FITC quenching) to form a quenched double-stranded DNA structure. In one preparation, the biotin-labeled aptamer and approximately 2.5 to 4.0 equivalents of the quenched complementary DNA were incubated in 50 μL of Durbeco phosphate buffer containing 5 mM magnesium chloride at room temperature for more than 2 hours. The resulting double-stranded structure suppressed the aptamer fluorescence signal before activation of the target molecule.

[0094] Subsequently, the prepared DNA double strands were incubated with approximately three equivalents of streptavidin (relative to aptamer concentration) in 100 μL of Durbeco phosphate buffer containing 5 mM magnesium chloride and 0.1 mg / mL herring sperm DNA to generate a DNA-streptavidin complex. This complex was then coated onto the surface of biotinylated yeast cells using the method described above. The resulting functionalized yeast cells can serve as a versatile sensing platform for biochemical detection and fluorescence analysis.

[0095] In this invention, aptamer sensors for different molecules are prepared as follows: for a vanillin sensor, 3 μL of 100 μM biotin-VANapt-Cy5 or biotin-VANapt-FITC is hybridized with 7.5 μL of 100 μM BHQ3-VANQ or BHQ1-VANQ; for an ATP sensor, 3 μL of 100 μM biotin-ATPapt-Cy5 is hybridized with 7.5 μL of 100 μM BHQ3-ATPQ; for a glucose sensor, 3 μL of 100 μM biotin-GLUapt-Cy5 or biotin-GLUapt-Cy3 is hybridized with 12 μL of 100 μM BHQ3-GLUQ or BHQ2-GLUQ; for a Zn sensor... 2+ For the sensor, 3 μL of 100 μM biotin-ZINapt-Cy5 was hybridized with 12 μL of 100 μM BHQ3-ZINQ; for the IFN-γ sensor, 3 μL of 100 μM biotin-IFNapt-Cy5 was hybridized with 7.5 μL of 100 μM BHQ3-IFNQ; for the glycosylated human serum albumin (gHSA) sensor, 3 μL of 100 μM biotin-HSAapt-Cy5 was hybridized with 7.5 μL of 100 μM BHQ3-HSAQ. All aptamer sequences used in this invention are listed in Table 3.

[0096] Table 3: Aptamer Sequences and Details

[0097] In one embodiment, yeast cells are first treated with sulfonyl-NHS-LC-biotin to biotinylate proteins present on the yeast cell wall. Subsequently, streptavidin and a biotin-tagged DNA aptamer are sequentially conjugated. The charged sulfonyl group of the biotinylating agent prevents it from penetrating the cell membrane, ensuring that biotinylation occurs only on the cell surface and providing a stable fixation site for subsequent aptamer grafting. The aptamer sequence is specifically designed to recognize a variety of target molecules, including ATP, glucose, vanillin, and zinc ions, enabling molecular sensors on the surface of the mother yeast cell to capture a variety of secreted metabolites and proteins.

[0098] The presence of a molecular sensor coating on the yeast cell surface was visualized by anchoring a Cy5-labeled aptamer (excitation wavelength: 646 nm; emission wavelength: 664 nm) to the surface. Simultaneously, the yeast cell wall was stained with Alexa Fluor 488-concanavalin A (excitation wavelength: 495 nm; emission wavelength: 520 nm). Confocal laser scanning microscopy imaging showed that the aptamer sensor was present only on the cell surface, and no endocytosis was observed. Importantly, this coating process did not affect the physiological state or secretory capacity of the yeast cells. Figures 2A to 2C Therefore, the cell's inherent functions are preserved.

[0099] After completing the molecular sensor anchoring procedure on the surface of mother yeast cells, yeast cell viability was further assessed using fluorescein diacetate. fluorescein diacetate generates a fluorescent signal after hydrolysis by intracellular esterases. Confocal laser scanning microscopy and flow cytometry analysis showed a high proportion of viable cells (>93%). Figures 3A to 3D Furthermore, yeast cells functionalized with molecular sensors on their cell surface maintained normal proliferation and secretion behavior comparable to unmodified natural cells. These results confirm that modification of cell surface molecular sensors does not adversely affect cell health or function. Figures 4A to 4B ).

[0100] To assess the biocompatibility of the molecular sensors on the cell surface, yeast cell viability and growth were further evaluated. In the viability test, yeast cells coated with fluorescently labeled single-stranded DNA (biotin-VANapt-Cy5) were suspended in 199 μL of DPBS, and 1 μL of 5 mg / mL fluorescein diacetate (dissolved in acetone) was added. The suspension was incubated at 30°C with shaking at 150 rpm for 15 minutes. After incubation, the yeast cells were analyzed using confocal laser scanning microscopy and flow cytometry. In the growth test, yeast cells with and without a molecular sensor coating were suspended in 5 mL of SD medium, with an initial optical density (OD) of 600 nm.600 The concentration was approximately 0.2. The culture was incubated at 30°C with shaking at 250 rpm for 72 hours, and the cell density was measured every 12 hours using a microplate reader.

[0101] In short, the flow cytometry method is performed as follows. An optimized gating strategy is employed in all flow cytometry operations. To distinguish the target yeast population from debris, an initial gate is first set based on the forward scattering area (FSC-A) and side scattering area (SSC-A) signals. Figure 7A To further distinguish between mother yeast cells and daughter cells with fluorescently labeled molecular sensors on their cell surfaces, a second gating region was established using a fluorescent negative control sample. Figure 7B When sorting highly secreting strains, a strict gating strategy was employed to reduce false positive signals caused by differences in cell size or budding. Specifically, only the top 0.05% of mutant yeast cells with the highest fluorescence intensity were sorted.

[0102] Example 2: High-density molecular sensor coating on the surface of mother yeast cells

[0103] Molecular sensors on the cell surface are coated with functional biotinylate groups on the yeast cell surface. These biotinylate groups bind to biotin-tagged aptamers via a streptavidin-mediated sandwich structure. Figure 5A By anchoring Cy5-labeled aptamers to the cell surface and further staining the yeast cell wall with Alexa Fluor 488-labeled concanavalin A (ConA), it was verified that the molecular sensors anchored to the cell surface selectively coat only the mother yeast cell. This dual-labeling result confirmed that during cell proliferation, the aptamer sensors are anchored only to the mother cell. As daughter cells bud and form new cell membranes, the molecular sensor coating on the cell surface remains only present on the original mother cell. This selective localization maintains a high density of molecular sensors on the mother cell surface and generates a clear fluorescent signal, thus clearly distinguishing between mother and daughter cells. Figure 5B ).

[0104] To quantitatively analyze the density of molecular sensors on the surface of yeast cells, a fixed concentration of yeast cells (2.0 x 10⁻⁶) was used. 7 Cells / mL) and different concentrations of aptamers were cultured in 100 μL of medium. Subsequently, flow cytometry was used to analyze the cells. Figure 5C )Analyze sensor density and use a fluorescence microplate reader ( Figure 6 Calibration was performed. As the aptamer concentration increased from 0.5 μM to 1.5 μM, the fluorescence intensity increased proportionally, representing a corresponding increase in cell surface sensor density. When approximately 1.4 x 10⁻⁶ was reached... 7A saturation point appears when the number of molecules / cells reaches a certain threshold, after which the fluorescence intensity remains constant, indicating that the available binding sites on the yeast cell surface have been almost completely occupied.

[0105] Further utilize flow cytometry ( Figure 5D The stability of the surface molecular sensor coating layer on the mother cells was examined. Using an appropriate gating strategy (Figure 7), mother cells coated with fluorescently labeled surface molecular sensors were distinguished from daughter cells. The results showed that the surface molecular sensors selectively coated only the mother yeast cells and did not migrate to daughter cells during budding. This selective localization was stably maintained for more than three days and over 30 cell divisions. Many mother cells retained a high-density surface molecular sensor coating layer and exhibited a strong Cy5 fluorescence signal, while daughter cells were largely uncoated.

[0106] After a prolonged budding culture (approximately 12 hours), a small percentage (approximately 16.4%) of molecular sensors on the cell surface were observed to detach, resulting in a slight decrease in fluorescence intensity. One possible reason is that some mother cells had already begun the budding process before molecular sensor anchoring; as the encapsulated buds continued to grow, they eventually detached from the mother cell surface during cytokinesis. Figure 8 Despite this partial loss, most molecular sensors (approximately 83.6%) remained on the mother cells, demonstrating high stability suitable for long-term secretion assays. Even after 72 hours of culture, a high sensor density (approximately 1.0 x 10⁻⁶) was maintained on the mother cells. 7 (molecules / cell), although the proportion of maternal cells in the population has decreased due to massive proliferation ( Figure 5E ).

[0107] Confocal laser scanning microscopy observation of mother yeast cells coated with Cy5-labeled aptamers further confirmed that the multi-generational budding process does not significantly affect the overall molecular sensor density on the yeast cell surface (Figure 9). Although the proportion of mother cells decreased over time, the molecular sensor density on the remaining mother cells only decreased slightly, thus ensuring long-term stable capture of secreted molecules. Importantly, the surface molecular sensor coating does not adversely affect the metabolic activity or proliferation of yeast cells, thus maintaining the cell's native physiological state and enabling accurate secretion phenotype analysis.

[0108] Example 3: Highly sensitive molecular detection using molecular sensors on the surface of mother yeast cells.

[0109] A highly sensitive detection method was established by grafting a high density of molecular sensors onto the surface of yeast cells to detect target compounds using fluorescence signals generated by structure-transition aptamer sensors. In this surface molecular sensor system, aptamers labeled with fluorescent probes are initially quenched by complementary DNA labeled with quenching groups. Upon binding to the target molecule, the aptamer undergoes a conformational change and dissociates from the complementary DNA, thereby recovering the fluorescence signal. The high-density surface molecular sensors enable the detection of low-abundance metabolic secretions, overcoming the insufficient sensitivity of existing sensing technologies. Figure 5F ).

[0110] To calibrate the detection sensitivity of the cell surface molecular sensor, a 2.0 x 10 6 Biotinylated yeast cells were mixed with 3 μM of a molecular sensor targeting a specific molecule in 100 μL of DNA-streptavidin solution to prepare yeast cells with a molecular sensor on their cell surface. The mixture was incubated at 30°C for 30 min; subsequently, the yeast cells with the molecular sensor on their cell surface were placed in a solution containing known concentrations of the target molecules (vanillin, ATP, glucose, and zinc ions) to determine the detection limit. To reduce the non-specific binding of quenched complementary DNA detached from the aptamer to the yeast cell surface, 0.1 mg / mL herring sperm DNA was added to the buffer as a blocking agent. Optimized buffer systems were used for different target molecules: 10 mM Tris, 150 mM sodium chloride, and 5 mM magnesium chloride were used for vanillin and ATP detection; DPBS supplemented with 5 mM magnesium chloride was used for glucose and zinc ion detection. After incubation at 30°C with shaking at 150 rpm for the specified time, the yeast cells were washed and analyzed by flow cytometry. In the specificity test, yeast cells coated with vanillin surface molecular sensors were cultured with 1 mM vanillin, vanillyl alcohol, vanillic acid or ferulic acid to evaluate their selective recognition ability.

[0111] To verify the detection specificity, aptamers targeting vanillin and structurally similar compounds (including vanillyl alcohol, vanillic acid, and ferulic acid) were tested. The cell surface molecular sensor platform produced a 3.86-fold fluorescence enhancement for vanillin, while the other compounds showed only weak signal changes (1.06-1.18-fold), resulting in a 3.45-fold specificity ratio. Figure 10A Similarly, in glucose-specific assays, cell surface molecular sensors exhibited strong fluorescence enhancement (approximately 3-fold) upon exposure to 1 M glucose, while fructose and galactose produced only slight signals, confirming the high molecular recognition specificity of the aptamer sensor. Figure 10BFor ATP detection, ATP-specific cell surface molecular sensors exhibited significant fluorescence enhancement (approximately 4-fold) upon exposure to 1 mM ATP, while structurally similar nucleotides (including GTP, UTP, and CTP) showed almost no fluorescence response. Figure 10C This further proves that the platform has excellent selectivity.

[0112] High-density surface molecular sensor coatings enable the detection of vanillin at concentrations as low as 100 nM, demonstrating a low detection limit. Similarly, the detection limits for ATP and glucose are approximately 100 nM and 1 μM, respectively. Figure 5G Compared to traditional droplet detection methods that use freely suspended aptamer sensors, the surface molecular sensor system significantly increases the local sensor density, reaching approximately 1.4 x 10⁻⁶. 7 One sensor per yeast cell. This high sensor density makes it more than 10 times more sensitive than droplet-based live cell sensors and more than 100 times more sensitive than aptamer sensors that are freely suspended in droplets.

[0113] To further confirm the enhanced sensitivity brought about by molecular sensors anchored to the cell surface, structurally identical ATP aptamer sensors were tested in a free-suspension state. These sensors (100 nM or 2 μM) were compared with uncoated yeast cells (2.0 x 10⁻⁶ m²). 7 The cells / mL were co-cultured with a blocking agent (0.1 mg / mL herring sperm DNA) and different concentrations of ATP, and the fluorescence signal was measured using a microplate reader. Figure 11 Unlike the pronounced fluorescence enhancement observed in cell-surface molecular sensor systems, freely suspended aptamers exhibited only fluctuating signal differences across the 100 nM to 10 μM ATP concentration range. This decrease in sensitivity may stem from nonspecific interactions between the aptamer and the yeast cell surface, leading to increased background signal. In contrast, aptamer immobilization in cell-surface-anchored molecular sensor configurations reduces nonspecific interactions, stabilizes sensor performance, and significantly improves detection sensitivity.

[0114] While cell-free droplet systems containing enriched cell extracts (such as tRNA, nucleotides, and cofactors) can reduce background noise and improve detection sensitivity by removing cell membrane structures, their encapsulation efficiency for single mutant DNA templates remains low (approximately 5% to 20%). Furthermore, many cell-free platforms are highly miniaturized, making it difficult to scale up to industrial biomanufacturing. Even when promising biosynthetic pathways are discovered in vitro, they ultimately need to be reintroduced into living cells for functional validation.

[0115] The high sensitivity of molecular sensors anchored to the cell surface enables visualization of secretions from individual Saccharomyces cerevisiae cells, as the secretions can be directly captured by aptamer sensors anchored to the yeast cell surface. Figure 12A ATP secretion kinetics were monitored using confocal laser scanning microscopy, with images taken every 5 minutes for a total of 40 minutes. Figure 12B Upon binding with secreted ATP, the molecular sensor aptamers on the cell surface undergo a conformational change and activate Cy5 fluorescence (excitation wavelength: 646 nm; emission wavelength: 664 nm). Yeast cells were dispersed in a diluted culture medium (approximately 5.0 x 10⁻⁶). 5 (cells / mL) reduces cross-contamination caused by diffusion, allowing the molecular sensor coating on each cell surface to immediately capture ATP molecules upon secretion. Under normal culture conditions, even low-abundance ATP secretion can be detected within 15 minutes and is clearly distinguishable from baseline signals at 0, 1, 5, and 10 minutes. Figure 5H Flow cytometry further confirmed that ATP secretion from individual yeast cells increases over time. Figure 12C ).

[0116] Therefore, the high sensitivity and temporal resolution of molecular sensors on the cell surface provide unprecedented insights into secretion dynamics research, enabling high-precision direct visualization and quantification of the metabolic activity of individual yeast cells.

[0117] In short, to verify the molecular sensor coating layer on the cell surface, yeast cells were labeled with fluorescently labeled single-stranded DNA. Specifically, 3 μM biotin-VANapt-Cy5 (excitation wavelength: 646 nm, emission wavelength: 664 nm) was mixed with 1.0 x 10⁻⁶ ppm of ... 6 Biotinylated yeast cells (wild-type Saccharomyces cerevisiae strain BY4742) were cultured in suspension for fluorescent labeling. After culturing, the labeled yeast cells were placed in 100 μL DPBS, and 2.5 μM Alexa Fluor 488-ConA (excitation wavelength: 495 nm, emission wavelength: 520 nm) was added. The mixture was then shaken at 150 rpm for 30 minutes at 37°C. The cells were then washed once and resuspended in DPBS or YPD medium for confocal laser scanning microscopy imaging.

[0118] To systematically quantify the molecular sensor coating density on cell surfaces, 2.0 x 10⁻⁶ 6 Yeast cells were cultured with different concentrations of biotin-VANapt-Cy5 (0.5 to 3.0 μM) and analyzed by flow cytometry. Yeast cells without fluorescently labeled DNA were used as negative controls. Each sample was analyzed at 3.0 x 10⁻⁶ cells / mL. 5Yeast cells were used. To estimate the number of sensor molecules on each yeast cell, the fluorescence signal of a biotin-VANapt-Cy5 solution of known concentration was measured using a microplate reader, and a standard curve was established. By measuring the residual biotin-VANapt-Cy5 concentration in the culture medium after coating, the molecular sensor density on the surface of each yeast cell could be calculated. Figure 6 ).

[0119] Example 4: Highly sensitive multiplex detection using a combined cell surface molecular sensor

[0120] By coating yeast cells with different surface molecular sensors, each containing a different aptamer sequence, multiple target metabolites can be detected simultaneously, thus achieving multiplex detection. Figure 5I ).

[0121] In multiplex detection, three surface molecular sensors with different aptamer sequences were simultaneously anchored to the surface of the same yeast cell. The molecular sensors included those specifically recognizing ATP, vanillin, and glucose, respectively, and were labeled with different fluorescent dyes: Cy5 (excitation wavelength: 646 nm, emission wavelength: 664 nm) for ATP, FITC (excitation wavelength: 490 nm, emission wavelength: 525 nm) for vanillin, and Cy3 (excitation wavelength: 555 nm, emission wavelength: 570 nm) for glucose. The molecular sensors (1.5 μM each) were then coupled with a 2.0 x 10⁻⁶ microarray. 6 Biotinylated yeast cells were co-cultured in a DNA-streptavidin solution containing 1.5 μM streptavidin and 0.1 mg / mL herring sperm DNA, with a total reaction volume of 200 μL. After multiplex molecular sensor anchoring, the yeast cells were cultured in solutions containing different target molecules, including 0.5 mM ATP, 1.0 mM vanillin, and / or 1.0 M glucose. The buffer composition was 10 mM Tris, 150 mM sodium chloride, 5 mM magnesium chloride, and 0.1 mg / mL herring sperm DNA. The culture conditions were 30°C with shaking at 150 rpm for 1 hour. After culture, the yeast cells with molecular sensors on their surface were washed, and their multiplex detection capability was analyzed using flow cytometry.

[0122] To evaluate the system's detection performance, yeast cells with molecular sensors on their surface were cultured with vanillin, ATP, glucose, or zinc ions under different conditions over time. Figure 5J ; Figures 13A to 13DIn vanillin detection, the fluorescence signal significantly increased within 1 hour of incubation and reached saturation after approximately 2 hours. A 3.82-fold signal amplification was observed at 1 mM vanillin concentration, with a dynamic detection range of 100 nM to 1 mM. For ATP detection, the fluorescence signal amplified 9.81 times at 4 mM ATP and reached saturation after approximately 2 hours, with a dynamic detection range of 100 nM to 4 mM. In glucose detection, the fluorescence signal gradually plateaued after approximately 2 hours, producing a 3.00-fold signal amplification at 1.5 M glucose, with a dynamic detection range of 0 to 2 M. For zinc ion detection, the fluorescence signal plateaued after approximately 3 hours, with a 2.78-fold signal amplification observed at 2 mM zinc ions. These results demonstrate that the surface molecular sensor platform is suitable for highly sensitive detection of a variety of small molecules and exhibits good versatility.

[0123] To further validate multiplex detection capabilities, three different types of surface molecular sensors (aptamers for vanillin, ATP, or glucose) were simultaneously grafted onto the surface of the same yeast cell. Figure 5K ; Figures 14A to 14D This combined surface molecular sensor can generate distinguishable fluorescence signals for each target molecule: ATP is detected using Cy5 fluorescence (excitation wavelength: 646 nm; emission wavelength: 664 nm), vanillin using FITC fluorescence (excitation wavelength: 490 nm; emission wavelength: 525 nm), and glucose using Cy3 fluorescence (excitation wavelength: 550 nm; emission wavelength: 570 nm). When exposed to a mixed solution containing all three target molecules, yeast cells with surface molecular sensors emit clear and distinguishable fluorescence signals corresponding to each metabolite. Flow cytometry analysis further confirmed that the system can distinguish different yeast populations based on fluorescence characteristics and effectively separate cells with multiple target signals, differentiating them from a low-fluorescence negative control population. Thermal analysis of mixtures of different target compounds further confirmed the accuracy of the surface molecular sensor detection, showing that multiplex detection has 100% specificity. Figure 5L ).

[0124] Example 5: High-throughput screening using cell surface molecular sensors

[0125] The high-throughput screening process mainly includes two steps: the capture step and the screening step. Figure 15A ).

[0126] Step 1: Capture. Single yeast cells grafted with cell surface molecular sensors were encapsulated using droplet microfluidics. Yeast cells with surface molecular sensors were suspended in YPD medium containing ferulic acid (as a precursor substrate for vanillin production) and encapsulated in approximately 33 pL water-in-oil droplets, which were stabilized by a biocompatible surfactant. The yeast cell concentration was approximately 1.5 x 10⁻⁶. 6 Cells / mL ensures a single-cell encapsulation probability of less than 5%, and according to Poisson distribution, a co-encapsulation probability of less than 0.13%. Encapsulated yeast cells are cultured in droplets for approximately 48 hours, during which time mother cells proliferate and produce daughter cells (…). Figures 16A to 16C Notably, the molecular sensor layer on the cell surface remains only present in the mother cell and maintains a high-density sensor distribution during budding. Simultaneously, secreted metabolites accumulate within the droplets and are successfully captured by the molecular sensors on the cell surface, generating strong fluorescent signals corresponding to the secretory characteristics of each individual cell.

[0127] Step 2: Screening. After culturing, the droplets are disrupted to release yeast cells, and the mother cells are analyzed using flow cytometry. Compared to traditional droplet-based secretion detection methods, which are typically limited by droplet sorting throughput (approximately 10 to 200 Hz at a cell density λ of 0.05), the molecular sensor method on the yeast cell surface of this invention bypasses the limitations of droplet encapsulation, achieving over 10... 3 Up to 10 4 With a screening throughput of Hz, the processing speed is increased by approximately 10 to 100 times. Figure 15B ).

[0128] To verify the capabilities of cell surface molecular sensors, an engineered Saccharomyces cerevisiae strain was constructed to produce vanillin. The biosynthesis of vanillin in yeast is catalyzed by 4CL and ECH enzymes, while the native metabolic pathway further converts vanillin into vanillyl alcohol or vanillic acid. Figures 17A to 17B To enhance vanillin accumulation and secretion, several engineered bacterial strains containing plastids encoding the 4cl and ech genes were constructed, including VAN-1 (Δgre2), VAN-2 (Δadh6), and VAN-3 (Δadh6 Δgre2). Deletion of these genes partially blocked the vanillin reduction pathway. After culturing in a fermentation system containing 1.0 g / L ferulic acid for 48 hours, the vanillin concentrations were: 9.6 mg / L for VAN-1, 12.3 mg / L for VAN-2, and 22.9 mg / L for VAN-3. During droplet culture, the fluorescence signal amplification increased from 1.11 to 1.56 for VAN-1, from 1.12 to 1.96 for VAN-2, and from 1.13 to 2.81 for VAN-3 within 20 hours. Figure 15CAfter 48 hours of droplet culture, the dynamic range of vanillin detection reached 1 to 1000 μM, indicating that the cell surface molecular sensor can accurately measure vanillin during long-term cell proliferation without signal saturation. Figure 18 ).

[0129] In one embodiment, the original ech gene is as described in SEQ ID NO:60; and the mutant ech gene with the amino acid substitutions I90N / Y169C / N212Q / P213L has the sequence described in SEQ ID NO:61.

[0130] To simultaneously analyze vanillin and ATP secretion, two cell surface molecular sensors were simultaneously anchored on the surface of yeast cells: a FITC-labeled molecular sensor for detecting vanillin and a Cy5-labeled molecular sensor for detecting ATP. Specifically, 1.5 μM of vanillin-targeting molecular sensor and 1.5 μM of ATP-targeting molecular sensor were anchored to 1.5 x 10⁻⁶ cells. 7 Biotinylated yeast cells (WT, VAN-1, VAN-2, or VAN-3) were mixed in 1.5 mL of a DNA-streptavidin solution containing 0.5 μM streptavidin and incubated at 30°C for 30 minutes. The negative control group used wild-type cells with DNA double-streptated sensors labeled with the same fluorescent probes but not recognizing vanillin or ATP (Table 3). These coated cells (1.5 x 10⁻⁶) 7 (Each cell) was suspended in 10 mL of YPD medium containing 1.0 g / L ferulic acid and encapsulated into water-in-oil droplets using a microfluidic droplet generator. The droplets were incubated at 30°C for 48 hours, allowing yeast cells to capture their own secreted vanillin and ATP through molecular sensors on the yeast cell surface. After incubation, the droplets were destroyed, and high-throughput multiplex single-cell secretion analysis was performed using flow cytometry.

[0131] Further, large-scale single-cell yeast populations (1.2 x 10⁻⁶) were constructed using yeast cells (containing different aptamers) anchored to various cell surface molecular sensors. 7 The high-throughput screening capability of the cells was demonstrated. Vanillin secretion was detected by FITC fluorescence (excitation wavelength: 490 nm; emission wavelength: 525 nm), while ATP secretion was simultaneously detected by Cy5 fluorescence (excitation wavelength: 646 nm; emission wavelength: 664 nm). Flow cytometry results clearly distinguished VAN-3 cells from the negative control group (NC), which consisted of wild-type (WT) yeast cells with DNA double-stranded sensors on their surface that did not recognize vanillin or ATP. Figure 15D ).

[0132] This screening system successfully distinguished between VAN-1, VAN-2, VAN-3, wild-type, and negative control groups based on secretion characteristics. Among the engineered strains, VAN-3 exhibited the highest vanillin secretion, followed by VAN-2 and VAN-1, all superior to the wild-type strain. The ATP secretion characteristics of the various yeast strains were similar, but significantly higher than the negative control group, further demonstrating that the molecular sensors on the yeast cell surface possess multiplex detection capabilities. Figure 15E ).

[0133] Furthermore, flow cytometry analysis of a 1:1 mixture of wild-type and VAN-3 strains revealed three distinct clusters ( Figure 15F VAN-3 mother cells with high fluorescence intensity (right cluster), wild-type mother cells with quenched fluorescence (middle cluster), and daughter cells with low fluorescence (left cluster). When analyzing mixtures of different WT:VAN-3 ratios (1:2, 1:1, 2:1, 5:1, and 1:0), independent fluorescence peaks for each population can be observed, thereby achieving accurate differentiation and efficient sorting of VAN-3 cells in heterogeneous mixed samples. Figure 15G ).

[0134] Example 6: High-speed analysis of rapid directed evolution using cell surface molecular sensors

[0135] This study investigates the high-speed sorting capabilities of highly secretory yeast strains by detecting the rapid, directional evolution of molecular sensors on cell surfaces. In this method, molecular sensors grafted onto the surface of a single mother yeast cell capture cell secretions. After droplet culture to accumulate the secreted metabolites, the droplets are disrupted, and the yeast cells are analyzed directly by flow cytometry. The fluorescence signal generated by the molecular sensors on the cell surface is approximately 3.0 x 10⁻⁶. 3 The screening rate per cell per second was measured. Although the single-cell encapsulation rate of the droplets was less than 5%, the screening speed of this method was still more than 30 times faster than traditional droplet screening techniques (typically about 10 to 200 Hz).

[0136] To validate this method, a mixed population containing 0.5% VAN-3 strain and 99.5% wild-type yeast strain was screened to isolate vanillin-secreting cells. Figure 19A Yeast cells grafted with surface molecular sensors were cultured in droplets for 48 hours to capture secreted vanillin. The droplets were then disrupted, and the cells were analyzed by flow cytometry at approximately 3.0 x 10⁻⁶ cells / mL. 3 Fluorescent yeast cells are sorted at a rate of cells / second. Using this high-speed screening capability, 1.2 x 10⁻⁶ cells can be analyzed and sorted in approximately 7 minutes. 6Approximately 100 vanillin-secreting yeast strains were successfully isolated from yeast cells. Vanillin production was verified using absorbance measurement (438 nm optical density, OD438) on randomly selected yeast strains from both pre- and post-sorting populations. After sorting, the percentage of VAN-3 strains with secretory capacity increased from 0.5% to approximately 90%, demonstrating the extremely high sorting efficiency of this invention.

[0137] To further demonstrate the application of this invention in directed evolution, a large mutant library (approximately 2.2 x 10⁻⁶) was constructed using error-prone PCR. 6 A variant of the library was developed to modify enoyl-CoA hydratase / aldolase (ECH; encoded by the ech gene) to enhance vanillin production. The library was transduced into yeast strain D-3, which had deletions of the gre2 and adh6 genes, and the resulting cells were yeast cells with vanillin-targeting molecular sensors on their cell surface. After 48 hours of droplet culture, fluorescence intensity was used as an indicator of vanillin secretion, from a total of 2.2 x 10⁻⁶. 6 Within approximately 12 minutes, the top 0.05% of cells were sorted out from each cell. Figure 19B This cell surface molecular sensor-assisted high-speed sorting process can effectively remove inactive mutants and enrich high-yield mutants.

[0138] The strains used in this invention are listed in Table 4 regarding the culture and transformation procedures for *E. coli* and yeast. *E. coli*-*Saccharomyces cerevisiae* shuttle plasmids were constructed and amplified in *E. coli* DH5α chemically competent cells. *E. coli* culture was incubated overnight in LB medium (5 g / L yeast extract, 10 g / L tryptone, and 10 g / L sodium chloride) at 37°C and 220 rpm with shaking. If necessary, 100 mg / L ampicillin was added to the medium for plasmid amplification. *E. coli* transformation was performed according to the manufacturer's instructions using a standard chemical transformation procedure. After transformation, the strains were plated on LB agar containing 100 mg / L ampicillin and incubated overnight at 37°C. Single colonies were inoculated into 5 mL of LB medium and incubated overnight. Subsequently, the *E. coli* culture medium (OD200) was... 600 = 2.0) extract plasmids.

[0139] Table 4: Strains used in this invention

[0140] For the construction of the yeast strain, *Saccharomyces cerevisiae* BY4742 was purchased from Beijing Zhuangmeng Biotechnology Co., Ltd., China, and cultured and amplified in yeast extract-peptone-glucose medium (10 g / L yeast extract, 20 g / L tryptone, 20 g / L glucose) at 30℃ and 250 rpm. To maintain plastids, the yeast strain was cultured in SD medium supplemented with DO supplement-Leu or DO supplement-Ura. Chemocompetent yeast cells were prepared using a standard kit. Subsequently, plastids were introduced using chemical transformation, and the engineered yeast cells were plated on SD agar medium and cultured at 30℃ for 2 to 3 days until single colonies were formed. These colonies were then inoculated into 5 mL of SD medium and cultured overnight to prepare a high-concentration cell suspension (OD). 600 = 2.0). The molecular characteristics of the transgenic yeast strain were confirmed using colony PCR to verify that the plasmids had been successfully introduced.

[0141] The plastids used in this invention are listed in Table 5. The *E. coli*-*Saccharomyces cerevisiae* shuttle vectors pYCP (VectorBuilder, China) and pML104 (Addgene, USA) were used for gene expression and gene deletion in the yeast strain, respectively. Recombinant plastids were constructed using the Gibson assembly method according to standard procedures. PCR amplification was performed using Q5® HighFidelity DNA Polymerase (NEB, USA), and the primers used are listed in Table 4. To construct plastids that enable yeast to produce vanillin, the target genes 4cl (derived from *Petroselinum crispum*) and ech (derived from *Pseudomonasputida* KT2440) were amplified by PCR using specific primers. The resulting DNA fragments were purified by colloidal recovery and then cloned into the pYCP vector to obtain the recombinant plastid pVAN. Gene deletion in yeast strain BY4742 was performed using the CRISPR / Cas9 gene editing method.

[0142] Table 5: Plasma used in this invention

[0143] After sorting, approximately 301 candidate strains were obtained and cultured on SD agar plates. 132 individual strains were randomly selected and cultured separately, and their vanillin production was verified using absorbance analysis. The results showed that the sorted strain population had an accuracy rate exceeding 68% in identifying rare high-secretion phenotypes. Figure 20 Among them, the top 10 variants ranked by vanillin production had an average production of approximately 1.93 times that of randomly selected variants from the unsorted mother population.

[0144] The vanillin production capacity of the sorted strains was further validated by batch fermentation using high-performance liquid chromatography (HPLC). Figure 19C Among the variants, strain S-10 exhibited the highest yield, producing 86.4 mg / L vanillin and 27.7 mg / L vanillin after 48 hours (equivalent to a total of 113.0 mg / L vanillin equivalent). In comparison, the control strain VAN-3 produced only 42.6 mg / L vanillin (equivalent to 42.0 mg / L vanillin), thus S-10 represented a 2.7-fold increase in yield. Notably, vanillin secretion was detectable in VAN-3 within the first 24 hours; however, the vanillin concentration decreased significantly after 36 hours, presumably due to conversion to vanillin, and became almost undetectable after 48 hours, at which point the vanillin concentration reached 42.6 mg / L. A comparison of the time histories of each strain is shown below. Figure 21 As shown.

[0145] Similar vanillin depletion was observed in ECH mutant strains S-1, S-8, and S-10, although the rate of depletion varied among strains. This suggests that vanillin may be converted to vanillyl alcohol via endogenous enzymes such as alcohol dehydrogenases (ADHs), aldehyde-ketone reductases (AKRs), and aldehyde reductases (ALDRs). Sequencing analysis of the high-secretion vanillin variants revealed three active ech sequences with complete open reading frames and no internal stop codons. Compared to the original ECH enzyme, the ECH variant identified in strain S-10 exhibited four mutations: I90N, Y169C, N212Q, and P213L. Figure 19D Furthermore, the ECH variants of strains S-8 and S-1 were identified with D80V and Y75H mutations, respectively. Figures 22A to 22B ).

[0146] To assess enzyme activity, wild-type and mutant ech genes were expressed in *E. coli* BL21(DE3). After incubation at 37°C for 24 hours with 200 μM feruloyl-CoA and 7.5 μM β-NAD, the vanillin concentration in the supernatant was determined by absorbance. The vanillin yield of all three mutant enzymes was significantly higher than that of the wild-type enzyme. The I90N / Y169C / N212Q / P213L mutant exhibited the highest vanillin yield (approximately 26 mg / L), about 11 times that of the wild-type. Figure 23 ).

[0147] To further quantify catalytic efficiency, wild-type and mutant ECH enzymes were purified by nickel affinity chromatography. Figure 24AAn equal volume of purified enzyme (1 mg / mL) was reacted with 300 μM feruloyl-CoA and 7.5 μM β-NAD at 30 °C for 12 hours. Vanillin yield was determined using the same absorbance method. Figure 24B ), and based on this, the relative enzyme activity was calculated ( Figure 24C All mutant enzymes consistently exhibited higher catalytic activity than the wild type, with the I90N / Y169C / N212Q / P213L mutant showing an approximately 2.3-fold increase in enzyme activity, confirming an improvement in its catalytic performance.

[0148] In addition to the directed evolution of ECH enzymes, the molecular sensor anchored to the surface of mother yeast cells in this invention is further applied to the engineering of transport proteins to enhance the efflux and secretion capabilities of yeast cells. A mutant library of the vanillin transport protein PP_0179 was constructed in the VAN-3 strain using error-prone PCR, with a total number of variants exceeding 2 x 10⁻⁶. 7 Species. Single-cell screening was performed using molecular sensors on the surface of mother yeast cells. By capturing and quantifying the vanillin secreted by each mutant, the top 0.05% of high-secreting strains were selected. Figure 19E ), and after plating and culturing, 108 colonies were randomly selected for verification. Figure 25 More than 78% of the strains showed better performance in transporting proteins compared to the wild-type PP_0179, demonstrating the high accuracy of this screening platform.

[0149] Potential fluorescence signal variations caused by differences in cell size or budding state ( Figure 26 It has been proven to have minimal impact on sorting accuracy. This is achieved through a strict gating strategy. Figure 27A This ensures that the cell size distribution is consistent before and after sorting. Figure 27B Subsequently, batch fermentation experiments were conducted on the top ten performing mutant strains, among which four mutant strains showed significantly enhanced vanillin secretion capacity. Figure 19F The best-performing variant was the T-4 strain-derived variant, which possesses both K40R and V79D mutations and exhibits a synergistic improvement in transport efficiency. Figure 19G Other key mutations include: G45C in strain T-1, K48M in strain T-2, and the T15I / G45R / Q98H complex mutation in strain T-3, such as... Figures 28A to 28C As shown above, the molecular sensor anchored to the surface of mother yeast cells in this invention has high versatility in the field of metabolic engineering, and is particularly suitable for the directed evolution of transport proteins.

[0150] Example 7: Highly Flexible Application of Molecular Sensors on Yeast Cell Surfaces in Protein Analysis

[0151] The molecular sensor on the surface of yeast cells is a highly versatile platform that can accommodate various types of sensing components via biotin-streptavidin interactions, enabling the detection of diverse molecular targets. This system is compatible with high-affinity aptamers for metabolite detection (Table 6) and high-affinity aptamers for protein detection (Table 7). For example, when using an aptamer targeting human interferon-γ (IFN-γ), the aptamer-type cell surface molecular sensor (aptamer-MOMS) can detect interferon-γ in a dynamic range of 1 to 1000 nM, with a detection limit of approximately 1 nM. Figure 29A Similarly, specific aptamers targeting glycosylated human serum albumin (gHSA) enable aptamer-type cell surface molecular sensors to selectively recognize glycosylated pharmaceutical proteins. Figure 29B For target molecules for which no aptamers have yet been validated, emerging configuration-change aptamer technology also offers a viable alternative to further expand the sensing capabilities of molecular sensors on the surface of yeast cells.

[0152] In addition to adapting to body size sensing modes, molecular sensors on the surface of yeast cells can also be modified to an antibody-based architecture. This involves coating the cell surface with streptavidin-conjugated antibodies to achieve the detection of a wider range of protein targets. For example, when yeast cells are coated with streptavidin-modified interferon-γ antibodies, uniform surface coverage can be achieved at antibody concentrations higher than 5 μg / mL. Figure 29C It also features a detection limit of approximately 1 nM and a wide dynamic range of 1 to 1000 nM. Figure 29D This modular architecture allows molecular sensors on the cell surface to be quickly customized for detecting different analytes, from small molecules to large proteins, simply by changing the sensing element.

[0153] Table 6: Aptamers used for metabolite analysis

[0154] Table 7: Aptamers used for protein analysis

[0155] It is worth noting that, in addition to Saccharomyces cerevisiae, molecular sensors on the cell membrane surface can also be applied to other budding fungi, such as Pichia pastoris, for secretion characteristic analysis. Figure 30One key advantage of cell membrane surface molecular sensor strategies lies in their utilization of the asymmetric budding process in yeast cells, allowing surface-bound sensors to remain persistently localized on the mother cell surface during cell division. This mechanism maintains a stable sensor density across multiple generations of proliferation, enabling robust fluorescence detection of extracellular secretions. In contrast, for organisms that undergo symmetrical division (e.g., *E. coli*), surface-bound sensors are evenly distributed among daughter cells, leading to signal dilution and decreased detection stability, making them unsuitable for cell membrane surface molecular sensor platforms. Table 8 summarizes fungal species for which cell membrane surface molecular sensors are suitable or unsuitable.

[0156] Table 8: Fungal Species Applicable to Cell Membrane Surface Molecular Sensors

[0157] In summary, this invention provides a molecular sensing platform that can selectively anchor molecular sensors to the surface of mother yeast cells without transferring them to daughter cells during budding. This selective positioning allows for the formation of a high-density sensor layer (approximately 1.4 x 10⁻⁶) on the surface of the mother yeast cells. 7 (Each sensor / cell) can maintain its functionality during multiple cell divisions over a period of up to three days.

[0158] Molecular sensor platforms immobilized on cell membrane surfaces offer three main advantages. First, high sensitivity: High-density sensor coverage and an extremely short diffusion distance between the sensor and the cell surface enable highly sensitive detection with a detection limit below 1 μM. Second, high throughput: This is achieved through a two-step process. First, yeast cells with molecular sensors covering their cell membranes are encapsulated in droplets to capture secreted molecules, generating a fluorescent signal on the mother cell surface. Then, the droplets are disrupted, and flow cytometry analysis is performed, allowing for the detection of over 10... 7 High-throughput phenotypic analysis of yeast cells. Third, high speed: yeast cells can be directly sorted during the analysis stage without the need for droplet encapsulation again, thus overcoming the limitation of low single-cell encapsulation efficiency (about 1-10%) of traditional droplet methods, and increasing the sorting speed by more than 30 times.

[0159] To verify the molecular sensor of the present invention, approximately 10 [units of something] derived from the parent strain VAN-3 were used. 6A mutant library was used for screening. High-speed sorting was employed to isolate the top 0.05% of mother yeast cells with the highest vanillin secretion. The secretory activity of the selected population reached a plateau, approximately 2.7 times that of the parental strain. Sequencing analysis identified four unique variants, among which the ech gene contained mutations leading to amino acid substitutions. In particular, strain S-10, with mutations in I90N, Y169C, N212Q, and P213L, exhibited the highest vanillin yield and demonstrated excellent potential for directed evolution.

[0160] Furthermore, industrial-scale vanillin production concentrations (e.g., greater than 2.5 g / L) can be achieved by inhibiting endogenous vanillin degradation pathways (including ADHs, AKRs, and ALDRs) and combining this with a repetitive, directed evolution program of molecular sensors on the cell membrane surface targeting key enzymes such as ECH and 4CL.

[0161] It is worth noting that the molecular sensor of this invention is highly versatile and can be applied to various metabolic engineering fields. For example, it can be used for the directed evolution of extracellular transport proteins (such as PP_0179), and also for the detection of extracellular proteins, including IFN-γ and glycosylated proteins. This platform can also be used to screen high-yielding strains for subsequent optimization, and can be further extended to analyze protein heterogeneity, such as immunoglobulins with different glycosylation patterns.

[0162] Therefore, molecular sensors on the surface of yeast cell membranes are an advanced analytical tool that enables highly sensitive, high-throughput screening and rapid sorting of single-cell yeast secretions. They can effectively identify rare, high-performing strains in large populations and provide new opportunities for metabolic analysis and strain engineering.

[0163] As used herein, the terms “approximately,” “substantially,” “essentially,” and “about / approximately” are used to describe and explain a small variation. When used in conjunction with an event or situation, the term can refer to the exact occurrence of the event or situation, or approximately the occurrence of the event or situation. The term “about,” as used herein with respect to a given value or range, typically refers to a range of ±10%, ±5%, ±1%, or ±0.5% of the given value or interval, which can be understood herein as ranging from one endpoint to another or between two endpoints. Unless otherwise stated, all ranges disclosed in this disclosure include endpoints. Furthermore, unless explicitly stated otherwise, the term “a” as used herein should be understood to include one or more; additionally, terms such as “first,” “second,” and “third” are used only as illustrative purposes and are not intended to give numerical meaning or to order them according to their importance.

[0164] The above description is provided for the purpose of illustrating and describing the invention, and is not intended to be exhaustive or to limit the invention to the precise forms disclosed. Many modifications and variations will be apparent to those skilled in the art.

[0165] The above embodiments were chosen and described in order to best explain the principles of the invention and its practical application, thereby enabling other skilled in the art to understand the various embodiments of the invention and the various modifications suitable for the intended particular use.

Claims

1. A method for ultra-high throughput screening of yeast metabolites, characterized in that, include: Provide mother yeast cell population; Multiple molecular sensors are anchored to the cell membrane surface of a mother yeast cell to form molecular sensors on the surface of the mother yeast cell membrane, wherein the molecular sensors are selectively retained on the mother yeast cell during the budding process of the mother yeast cell and are not transferred to the daughter cells. The mother yeast cells were cultured under conditions that facilitated the secretion of metabolites by the mother yeast cells; The metabolic products secreted by the mother yeast cells are captured using molecular sensors on the surface of the mother yeast cell membrane; The molecular sensors on the surface of the mother yeast cell membrane interact with the captured metabolites to generate a detectable signal. as well as The mother yeast cells are screened based on the detectable signal to identify mother yeast cells with metabolic secretion phenotypes; The metabolites include target metabolites and / or proteins.

2. The method of claim 1, wherein the molecular sensor comprises a nucleic acid aptamer sensor that binds to the target metabolite and / or protein.

3. The method according to claim 2, wherein the nucleic acid aptamer sensor is immobilized on the cell membrane surface of the mother yeast cell via biotin-streptavidin interaction.

4. The method according to claim 1, wherein the surface of the mother yeast cell has at least 10 6 The molecular sensor.

5. The method of claim 1, wherein the molecular sensor generates a fluorescence signal upon binding to the secreted metabolite.

6. The method of claim 1, wherein the method further comprises encapsulating a single mother yeast cell having the molecular sensor on its surface in a microfluidic droplet to capture secreted metabolites within the droplet.

7. The method of claim 1, wherein the method further comprises breaking the droplets after culturing to release the mother yeast cells for subsequent screening.

8. The method of claim 1, wherein the molecular sensor comprises a structure-transforming aptamer that generates fluorescence upon binding to the target metabolite.

9. The method according to claim 1, wherein the target metabolite comprises vanillin, adenosine triphosphate, glucose, and zinc ions.

10. The method according to claim 1, wherein the target protein comprises interferon-γ and glycosylated human serum albumin.

11. The method according to claim 1, wherein the yeast cells comprise Saccharomyces cerevisiae or Pichia pastoris.

12. The method according to claim 1, wherein the method further comprises constructing a yeast mutant library and performing directed evolution by screening the mother yeast cell mutant strain with enhanced secretion phenotype.

13. The method of claim 1, wherein the method further comprises enabling multiplex detection of multiple metabolites by immobilizing different molecular sensors for different molecules on the surface of the mother yeast cell membrane.

14. A functionalized yeast cell for detecting extracellular secretions, characterized in that, include: At least one mother yeast cell with a cell membrane surface; as well as One or more molecular sensors fixed on the surface of the mother yeast cell membrane; The molecular sensor is configured to bind to secreted molecules produced by the mother yeast cell and generate a detectable signal, and the molecular sensor remains positioned on the mother yeast cell during budding and does not substantially migrate to any daughter cells.

15. The functionalized yeast cell of claim 14, wherein the molecular sensor comprises an aptamer sensor.

16. The functionalized yeast cell according to claim 14, wherein the molecular sensor is fixed on the cell membrane surface of the mother yeast cell by binding to biotin-streptavidin protein.

17. The functionalized yeast cell of claim 14, wherein the molecular sensor comprises a fluorescence-based structure-transition aptamer.

18. A system for ultra-high throughput screening of microbial secretions, characterized in that, include: The functionalized yeast cells according to claim 14; A droplet microfluidic device configured to encapsulate a single functionalized yeast cell within a droplet; A culture module configured to culture the functionalized yeast cells to secrete metabolites and capture them by the molecular sensor; and A detection and sorting device is configured to detect the signal generated by the molecular sensor and separate yeast cells with a target secretion phenotype; The system filters at least 10 samples per run. 7 Yeast cells, and the system is configured to use at least 10 3 The cells are sorted at a rate of 1 cell per second.

19. The system of claim 18, wherein the molecular sensor comprises an aptamer sensor for binding the metabolite.

20. A mutant enoyl-CoA hydratase / aldolase, comprising at least one amino acid substitution selected from I90N, Y169C, and N212Q relative to the wild-type enoyl-CoA hydratase / aldolase.

Images