Delivery System

Abstract

Considering the highly acidic environment of gastric fluid and the location of Helicobacter pylori within the gastric mucosal layer, we aim to develop a Carbonate driving Yeast-Gel Micromotor to safely and precisely deliver engineered yeast through the gastric fluid to the gastric mucosal layer.

Based on a review of relevant literatureand through continuous improvement, we ultimately designed a Janus gel bead composed of half sodium alginate and half sodium alginate-calcium carbonate. Hereby, the carbonate molecule in alginate-calcium-carbonate module could react with the gastric acid, providing forces driving the whole yeast-containing gel penetrating the gastric mucosal layer, directly into where H. pylori locolizes. Along with the power module, the sodium alginate gel could protect the yeast from the acidic gastric fluid and depolymerize in the near-neutral gastric mucosal layer, releasing the yeast to inhibit H. pylori biofilm formation.

In wet lab experiments, we validated the pH-dependent stability of the sodium alginate gel and confirmed that yeast viability remains unaffected after gel depolymerization.

Additionally, we successfully produced larger-sized Janus gel beads as a macroscopic simulation of the propulsion mechanism. We aim to prepare micro-Janus gel beads and evaluate their properties in the future.

Introduction

Background & Design

To ensure the safe delivery of yeast through the acidic gastric fluid and into the near-neutral gastric mucosa layer, we plan to utilize a gel coating for yeast delivery. After literature reviewing, we identified a sodium alginate-calcium carbonate gel that remains stable under low pH conditions but dissolves in a neutral environment.¹⁶ Add yeast suspension when preparing the gel could easily complete the coating process.

As for the propulsion mechanism, we initially considered a direct preparation method involving magnesium powder. However, this approach is not feasible for our application. Therefore, we explored an alternative method inspired by microfluidic technology: creating Janus gel particles composed of sodium alginate and sodium alginate-calcium carbonate. The reaction between calcium carbonate and gastric acid would generate gas, propelling the particles.

For the preparation of Janus gel particles, we referred to a method described in a previous study.²⁴ This method involves using a microfluidic chip with two inlets for the aqueous phases (sodium alginate solution and sodium alginate-calcium carbonate solution) and one inlet for the oil phase (vegetable oil with surfactant). The design includes two cross-junctions to generate Janus droplets.

Goal

Develop a gel-coated delivery system that can safely protect engineered yeast through the gastric fluid and efficiently propell the yeast to the gastric mucosal layer to achieve highly targeted killing of Helicobacter pylori.

Result

We conducted a series of experiments to validate the pH-dependent stability of the sodium alginate gel and the viability of yeast after gel depolymerization. Additionally, we successfully produced larger-sized Janus gel beads and simulated their preparation using microfluidic methods.

Preparation and characterization of calcium alginate hydrogel beads

Utilizing the emulsification–internal gelation method³⁷ , we successfully prepared calcium alginate hydrogel microspheres. Optical microscopy revealed that the microspheres were uniformly spherical with a smooth surface. The diameter distribution analysis indicated that the majority of the microspheres had diameters ranging from 40 to 200 μm, with an average diameter of approximately 100 μm (Figure 1).

Figure 1: Calcium alginate hydrogel microspheres were prepared using the emulsification–internal gelation method. A, B: Optical microscope images of hydrogel microspheres under 10× magnification; C: Diameter distribution of the hydrogels.

Additionally, we employed scanning electron microscopy (SEM) to further characterize the surface morphology of the hydrogel microspheres. The SEM images confirmed that the microspheres successfully encapsulated yeast and formed as a spherical structure.

Figure 2: SEM image of hydrogel microspheres. A-C and D-F are different gel microspheres at different magnifications, Red circles indicated a very encapsulated yeast cell.The images show that the hydrogel microspheres encapsulate yeast and form a spherical structure.

Consequently, we assessed the pH-dependent stability of the hydrogels by immersing them in buffer solutions with varying pH levels. After approximately 200 seconds of immersion, we observed that the hydrogels remained intact in acidic conditions (pH = 1) but underwent significant depolymerization in neutral conditions (pH = 7.2), leading to the release of encapsulated yeast (Figures 2 and 3). This confirmed the pH-responsive behavior of the hydrogels.

Figure 3: A: Hydrogel treated with HCl solution at pH = 2.01, photograph at 200 s shows no signs of depolymerization; B: Hydrogel treated with phosphate buffer at pH = 4.02, photograph at 200 s shows that the hydrogel started depolymerizing; C: Hydrogel treated with phosphate buffer at pH = 7.44, photograph at 200 s shows significant depolymerization of the hydrogel and release of encapsulated yeast; D: Hydrogel treated with phosphate buffer at pH = 9.97, photograph at 200 s shows complete depolymerization of the hydrogel and release of encapsulated yeast.

Viability assessment of coated-and-uncoated yeast

To evaluate the viability of yeast after encapsulation and subsequent release from the hydrogel, we employed TTC staining as a viability marker. The results indicated that there was no significant difference in viability between freshly prepared yeast and yeast released from the hydrogel microspheres. This suggests that the encapsulation process does not adversely affect the normal biological activity of the yeast.

Preparation of hydrogel micromotors using microfluidic technology

We designed a microfluidic chip in collaboration with Dxfluidics Co., Ltd. to facilitate the preparation of hydrogel micromotors. The chip features two inlets for the aqueous phases (sodium alginate solution and sodium alginate-calcium carbonate solution) and one inlet for the oil phase (vegetable oil with surfactant). The design includes two cross-junctions to generate Janus droplets.

Figure 4: Microfluidic chip for preparing hydrogel micromotors, co-designed with Dxfluidics Co., Ltd.

We attempted to prepare hydrogel micromotors using the microfluidic chip.

Figure 5: Preparation of hydrogel micromotors using a microfluidic chip; the image shows the dispersion of the aqueous phase after the first cross-junction.

However, due to limitations in our microfluidic technique and time constraints, we were unable to produce micromotors of the desired size, functionality and quality.

Macroscopic simulation of microfluidic preparation of hydrogel micromotors

To simulate the microfluidic preparation of hydrogel micromotors, we employed a modified emulsification-internal gelation method. This involved using a homogenizer to create Janus droplets by mixing sodium alginate solution (containing yeast) and sodium alginate-calcium carbonate solution in vegetable oil with surfactant.

Figure 6: Hydrogel micromotors prepared using the simulated microfluidic method; the image shows micromotors with one half consisting of calcium alginate encapsulating yeast, and the other half consisting of sodium alginate encapsulating calcium carbonate.

We successfully produced larger-sized Janus gel beads, which serve as a macroscopic simulation of the intended microfluidic micromotors.

Motion of hydrogel micromotors in simulated gastric fluid

To evaluate the motility of the hydrogel micromotors in an acidic environment, we immersed them in a simulated gastric fluid (pH = 1.2). The reaction between calcium carbonate and gastric acid generated gas, propelling the micromotors. We recorded videos of their motion.

Discussion

pH-dependent stability

According to the method described in the reference, we prepared sodium alginate–calcium carbonate microgel beads. Subsequently, buffer solutions with pH values ranging from acidic to mildly alkaline were prepared. Time‐gradient sampling revealed that when these gel beads were immersed in the buffer solutions for about 200 seconds, dissolution of the beads and release of the encapsulated yeast were observed under neutral or mildly alkaline conditions, whereas the beads remained intact in acidic buffers. This result confirms that the gel beads possess pH-dependent stability.

In practical applications, the gel beads are expected to remain stable in the highly acidic gastric fluid (pH 1-2) and subsequently dissolve in the near-neutral gastric mucosal layer (pH 6-7), thereby releasing the encapsulated yeast.

For future experiments, we plan to conduct a more detailed assessment of the dissolution time of the gel beads across a broader pH range to better understand their stability profile.

Coated Yeast Viability Assessment

A viability assessment of the engineered yeast post-release from the gel microspheres was conducted to confirm the retention of their normal biological activity and cytotoxic capability. Using TTC staining as a marker for viability, we compared freshly prepared yeast with yeast released from gel beads. The observation of no significant decline in viability proves that the encapsulation process maintains the normal vitality of the yeast.

In future studies, we intend to perform additional assays to evaluate the functional activity of the released yeast, ensuring that their biofilm-inhibiting properties remain intact after encapsulation and release.

Janus Gel Beads & microfluidics

Although we did not achieve direct results in the fabrication of micromotors, our experimental design was acknowledged by engineers, provided the necessary equipment is available.

By conducting preparations under simulated microfluidic conditions and performing validation in a simulated gastric acid environment, we confirmed the feasibility of fabricating hydrogel micromotors using a microfluidic approach, as well as their motility in acidic environments. If conditions and time permit, we plan to fabricate hydrogel micromotors of the desired size and further verify their functionality in a simulated mucus environment.

In future experiments, we aim to optimize the microfluidic chip design and refine the preparation parameters to achieve the desired micromotor size and functionality.

Simulation of propulsion mechanism

To validate the propulsion mechanism of the hydrogel micromotors, we immersed them in a simulated gastric fluid (pH = 1.2). The reaction between calcium carbonate and gastric acid generated gas, propelling the micromotors. We recorded videos of their motion, which demonstrated their ability to move in an acidic environment.

By observing the motion of the micromotors in the simulated gastric fluid, we confirmed that the propulsion mechanism is effective. The generation of gas from the reaction provides sufficient force to propel the micromotors, which is crucial for their intended function of delivering yeast to the gastric mucosal layer.

In future studies, we plan to quantify the propulsion speed and distance of the micromotors under various conditions to better understand their performance in vivo and obtain parameters for drylab simulations.

Outlook

Owing to current limitations in microfluidic technology and time constraints, we have not yet been able to fully achieve micron-sized Janus gel microspheres. In the future, we will pursue further experimentation to obtain micron-scale Janus gel microspheres that more closely resemble the final product, followed by a comprehensive evaluation of their properties.

However, we have successfully prepared larger-sized Janus gel beads as a macroscopic simulation of the propulsion mechanism. We will continue to optimize our microfluidic techniques and explore alternative methods to achieve the desired micromotor size and functionality.

Overall, our results demonstrate the feasibility of using a gel-coated delivery system to protect and deliver engineered yeast to the gastric mucosal layer. We will continue to refine our methods and conduct further evaluations to ensure the effectiveness and safety of this delivery system.

References

Adhesion System

Abstract

Aiming to develop an adhesion system selectively targeting Helicobacter pylori without inducing an inflammatory response we expressed the N-terminal domain of the gastric-derived human protein CEACAM1 (C1ND) on our yeast cells as the HopQ-targeted adhesion site of H. pylori. Additionally, the soluble portion of HopQ (Re combinant HopQAD) was produced using an E.coli expression system to simulate and evaluate the adhesion effect.

Introduction

Background

We aimed to identify protein pairs with strong adhesion capabilities against Helicobacter pylori, which led us to consider simulating gastric cell targets to achieve specific adhesion. Through literature review, we selected several candidate adhesion molecules. However, due to concerns that certain protein interactions might trigger inflammatory or pathogenic responses, we ultimately chose the human gastric protein CEACAM1, which is known as the hunman gastric bind site of H. pylori’s outer membrane protein HopQ. Expressing this very protein in yeast is like “beating the bacteria at their own game”. On the other hand, as a human protein, this molecule is less likely to provoke an immune response, making it a safer choice for our application.

Design

We plan to express CEACAM1 on the yeast cell surface via a lipid anchor system Sed1, while HopQ will be expressed in E. coli. Considering the native transmembrane characteristics of both proteins, we selected the N-terminal domain of CEACAM1 (C1ND) and the soluble portion of HopQ (HopQAD). After adding a 6×His tag to each, we constructed the corresponding plasmids (pET-15b-HopQAD-FLAG and pESC-URA-pOpM-C1ND-Sed1). Following transformation into the respective microorganisms, we aim to confirm the successful expression of both proteins and subsequently verify their interaction using immunocytochemistry.

Goal

The goal of this module is to establish and validate the core biological mechanism that underpins our entire project: enabling Saccharomyces cerevisiae with C1ND to specifically bind to Helicobacter pylori. So this part serves as the foundational proof-of-concept for our adhesion strategy.

Results

hCEACAM1 N-terminal domain (C1ND) was successfully expressed on the surface of Saccharomyces cerevisiae.

We successfully induced the plasmid carrying the C1ND gene into Saccharomyces cerevisiae.

Figure 1: Recombinant Yeast Colonies. The image shows colonies of successful transformants growing on a SC-Ura selection plate following heat-shock transformation.

Then we performed SDS-PAGE to detect the expression level gradient of C1ND protein in yeast over time and to evaluate the efficiency of our protein purification.

Figure 2: Time Dependent C1ND Expression (SDS-PAGE). SDS-PAGE analysis of C1ND expression over a 0-2 hour period and the comparison between purified protein, yeast lysate, and supernatant. The results indicate the presence of a stable fragment, but not the full-length C1ND protein (due to our enzymatic protein extraction method). The C1ND protein was purified using IDA-Ni magnetic beads targeting the His-tag.

Respectively, we used Western Blot to further confirm the successful expression of C1ND protein.

Figure 3: Time Dependent C1ND Expression (Western Blot). Western Blot analysis of C1ND expression over a 0-2 hour period and the comparison between purified protein, yeast lysate, and supernatant. The C1ND protein was purified using IDA-Ni magnetic beads targeting the His-tag.

Adhesion receptor Recombinant HopQAD was successfully expressed in E.coli.

Plasmid carrying HopQAD gene marked by Amp resilience gene are delicately introduced into E.coli.

Figure 4: Recombinant E.coli Colonies. Transformed E. coli BL21(DE5) colonies selected on an LB agar plate supplemented with ampicillin. Competent cells were transformed via 42°C heat-shock treatment with HopQ recombinant plasmid.

Subsequently, we employed Western blot and SDS-PAGE to systematically analyze the expression gradient of the HopQ protein, evaluating both time-dependent and IPTG concentration-dependent changes.

Figure 5: Time Dependent HopQ expression (SDS-PAGE). SDS-PAGE analysis of purified HopQ samples taken at indicated time points (0-6h). The HopQ protein was purified using IDA-Ni magnetic beads targeting the His- tag.

The figure above shows the successful expression of the target protein HopQ, and the expression level increased significantly with the extension of the induction time. The band appeared between 45 kDa and 55 kDa, which is consistent with the expected size of the HopQ protein, approximately 49.3 kDa. There were almost no protein bands at 0 hours (before induction). Bands started to appear at 1 hour, and then the intensity (concentration) of the bands continued to increase at 2 hours, 4 hours, and 6 hours, indicating that the production of the protein within the cells continued to accumulate, reaching the highest level at 6 hours.

Figure 6: IPTG concentration Dependent HopQ expression (SDS-PAGE). SDS-PAGE analysis of HopQ protein expressed in E. coli under different concentrations of IPTG induction. Lysis: Total cell lysate; Supernatant: Soluble fraction after centrifugation; Purified: HopQ protein purified using Ni magnetic beads.

The figure above shows the results of optimizing the induction expression conditions of the target protein HopQ and purifying it. A low concentration of the inducer (IPTG) can efficiently express the HopQ protein, and the protein is soluble and was finally successfully purified to a relatively high purity.

The experiment tested the effects of different concentrations of IPTG (from 0 mM to 1.0 mM) on protein expression. As the concentration of IPTG increased, there was a relatively small increase in protein expression. This indicates that a low concentration of IPTG is sufficient for efficient expression.

In the right lane shown , there is a dominant and pure protein band at the ~49.3 kDa position, indicating that most of the contaminating proteins have been removed.

Respectively, we used Western Blot to further confirm the successful expression of HopQ protein.

Figure 7: Time Dependent HopQ expression (Western Blot). Western Blot analysis of purified HopQ samples taken at indicated time points (0-6h). The HopQ protein was purified using IDA-Ni magnetic beads targeting the His- tag.

Figure 8: IPTG concentration Dependent HopQ expression (Western Blot). Western Blot analysis of HopQ protein expressed in E. coli under different concentrations of IPTG induction. Lysis: Total cell lysate; Supernatant: Soluble fraction after centrifugation; Purified: HopQ protein purified using Ni magnetic beads.

A Modified Immunocytofluorescence (ICF) assay confirmed the Adhesion between C1ND-expressing Yeast and receptor HopQAD protein

To validate the adhesion between yeast with C1ND and HopQ target, we mixed yeast expressing the C1ND protein with a solution of HopQ protein, followed by treatment with a Cy3 fluorescently labeled secondary antibody and a primary antibody targeting the His-tag-HopQ. Simultaneously, we performed fluorescent staining of the yeast cell wall with CFW stains to observe whether the two fluorescent signals co-localized.

Figure 9: Modified Immunocytofluorescence (ICF) Assay Results. The first column (Fig A,D,G,J) shows results in CFW channel, locating yeast cell walls; the second column (Fig B,E,H,K) shows results in Cy3 channel, locating HopQ receptor protein; the third column (Fig C,F,I,L) shows merged images of the first two columns. The first row (Fig A-C) is the experimental group (C1ND +, HopQ +); the second and third rows (Fig D-I) is the background control group (C1ND -, HopQ +; C1ND +, HopQ - ); the fourth row (Fig J-L) is the blank group (C1ND -, HopQ -).

In the experimental group (first row), we observed significant co-localization of the two fluorescent signals, indicating that HopQ successfully adhered to the yeast cell wall. In contrast, the control groups (second and third rows) showed no Cy3 fluorescence signal, demonstrating that in the absence of either C1ND or HopQ, no adhesion occurred. The blank group (fourth row) also showed no fluorescence signal, confirming the specificity of the staining. Overall, these results confirm the successful adhesion between yeast expressing C1ND and the HopQ protein.

Discussion

C1ND expression

By chemically transforming and inducing the expression of a plasmid carrying the C1ND gene fragment in yeast cells, C1ND was successfully expressed on the cell surface. Subsequently, the yeast cell walls were digested by lyticase, while cell wall proteins were extracted. SDS-PAGE and Western Blot analyse of the extracted proteins successfully detected the C1ND protein band, confirming the successful surface expression of C1ND.

However, judging by the SDS-PAGE band size, the C1ND protein we purified using the magnetic bead method appears to be only a fragment, rather than the expected full-length C1ND protein. The reason for this was find out to be the enzymatic digestion method we used to extract the cell wall proteins. The lyticase enzyme specifically digests the β-1,3-glucan component of the yeast cell wall, which may have led to partial degradation of the C1ND protein during the extraction process. This suggests that while our method was effective in isolating cell wall proteins, it may not be suitable for preserving the integrity of all surface-expressed proteins. Future experiments could explore alternative extraction methods that minimize protein degradation to obtain full-length C1ND.

Overall, the successful detection (especially Western Blot confirmation) of C1ND confirms the feasibility of our yeast surface display strategy.

HopQAD expression

After transforming a plasmid containing the HopQAD protein gene fragment into E. coli, its expression was induced. The E. coli cells were then lysed to extract soluble proteins. Detection by SDS-PAGE and Western Blot revealed the HopQAD band, validating its normal expression.

The SDS-PAGE results showed that the HopQAD protein band appeared between 45 kDa and 55 kDa, consistent with the expected size of approximately 49.3 kDa. The expression level increased significantly with longer induction times, reaching the highest level at 6 hours. Additionally, varying IPTG concentrations demonstrated that even low levels of IPTG were sufficient for efficient HopQAD expression, indicating that the protein can be produced effectively without requiring high inducer concentrations. The purified HopQAD protein exhibited a dominant band at the expected size, indicating successful purification with minimal contamination.

For future experiments, optimizing induction conditions such as IPTG concentration and induction time could further enhance yield and purity. Overall, the successful expression and purification of HopQAD confirm the effectiveness of our E. coli expression system.

Immunocytochemistry

Yeast cells expressing the C1ND protein on its surface was incubated with a purified HopQAD protein solution. After treatment with a fluorescently labeled antibody to mark HopQAD and Calcofluor White fluorescent dye to localize the yeast cell wall, fluorescence microscopy revealed co-localization of the two fluorescent signals. The control group showed no HopQAD fluorescence, demonstrating that HopQAD successfully adhered to the yeast cell wall and confirming the adhesion between C1ND-expressing yeast and HopQAD protein.

Considering the water solubility of secretory proteins, we expressed not the full-length HopQ protein but rather its soluble fragment. This may have a certain impact on its binding affinity with C1ND. It has been concluded that C1ND can bind to the HopQAD fragment. Future studies could further test its binding capacity with the full-length transmembrane HopQ protein.

Overall, the successful co-localization of fluorescence signals in the experimental group confirms the specific adhesion between yeast expressing C1ND and the HopQAD protein, validating our adhesion system design.

Outlook

The current results have successfully validated the expression of both proteins and their interaction on the surface of engineered yeast. However, due to time constraints of the iGEM competition, we could not obtain more data to confirm the adhesion effects under gastric fluid conditions or to obtain kinetic parameters of the protein interaction. In the future, additional experiments such as pull-down assays, co-immunoprecipitation (Co-IP) and surface plasmon resonance (SPR) shall be necessary to comprehensively characterize the functional interaction. We believe that with further validation and refinement, this system can achieve more precise and effective functionality.

Additionally, for technical feasibility and material accessibility, our adhesion system was tested using auxotrophic Saccharomyces cerevisiae as the chassis organism, rather than Saccharomyces boulardii as originally designed. In future studies, this system should be validated and refined through implementation in Saccharomyces boulardii.

Overall, while our current results provide a solid proof-of-concept, further experiments are needed to fully realize the potential of this adhesion system in practical applications.

References

1. Bamba T, Inokuma K, Hasunuma T, Kondo A. Enhanced cell-surface display of a heterologous protein using SED1 anchoring system in SED1-disrupted Saccharomyces cerevisiae strain. J Biosci Bioeng. 2018 Mar;125(3):306-310.
2. Begum H, Murugesan P, Tangutur AD. Western blotting: a powerful staple in scientific and biomedical research. Biotechniques. 2022 Jun;73(1):58-69. doi: 10.2144/btn-2022-0003. Epub 2022 Jul 1.
3. Bussineau CM, Shuster JR. Genetic stability of protein expression systems in yeast. Dev Biol Stand. 1994;83:13-9.
4. Doohan D, Rezkitha YAA, Waskito LA, Yamaoka Y, Miftahussurur M. Helicobacter pylori BabA-SabA Key Roles in the Adherence Phase: The Synergic Mechanism for Successful Colonization and Disease Development. Toxins (Basel). 2021 Jul 13;13(7):485.
5. Froger A, Hall JE. Transformation of plasmid DNA into E. coli using the heat shock method. J Vis Exp. 2007;(6):253.
6. Goffeau A, Barrell BG, Bussey H, Davis RW, Dujon B, Feldmann H, Galibert F, Hoheisel JD, Jacq C, Johnston M, Louis EJ, Mewes HW, Murakami Y, Philippsen P, Tettelin H, Oliver SG. Life with 6000 genes. Science. 1996 Oct 25;274(5287):546, 563-7.
7. Matsuo Y, Kido Y, Yamaoka Y. Helicobacter pylori Outer Membrane Protein-Related Pathogenesis. Toxins (Basel). 2017 Mar 11;9(3):101.
8. Meilhoc E, Teissie J. Electrotransformation of Saccharomyces cerevisiae. Methods Mol Biol. 2020;2050:187-193.
9. Nguyen QA, Schmitt L, Mejías-Luque R, Gerhard M. Effects of Helicobacter pylori adhesin HopQ binding to CEACAM receptors in the human stomach. Front Immunol. 2023 Feb 20;14:1113478.
10. Qu S, Huang X, Song X, Wu Y, Ma X, Shen J, Zhu K. A Rigid Nanoplatform for Precise and Responsive Treatment of Intracellular Multidrug-Resistant Bacteria. Engineering, 15:57-66.
11. Shans N, Esmaeili M, Abraheh K, Asadi Hanjani N, Farrokhi M, Sardarpour N, Talebkhan Y, Kazemi-Lomedasht F, Mirabzadeh E, Mohammadi M. Development of Experimental Platforms to Assess Helicobacter pylori HopQ Interaction with Host CEACAM Molecules. Iran Biomed J. 2025 May 1;29(3):138-148.

Sensory System

Abstract

Random secretion of therapeutic proteins by our engineered yeast could lead to potentially harmful consequences. To address this, we designed several sensory modules based on modified endogenous yeast mating pathways, thereby equipping the engineered yeast with the ability to specifically recognize gastric H. pylori biomarkers and release therapeutic proteins in response. Specifically, we reengineered the yeast mating factor pathway by replacing its native agonist with one of the biomarkers, N^α-methylhistamine (NAMH). In response to NAMH, the pathway activates its intrinsic G-protein signaling cascade, thereby inducing gene expression under the control of the downstream promoter pFUS1. Collectively, our findings present an effective strategy for biomarker-responsive sensing and the targeted elimination of gastric H. pylori.

Background

H. pylori infection elevates specific biomarkers within the gastric niche. One is N^α-methylhistamine (NAMH), present in the mucosa of H. pylori infected patients and very low in non-infected patients or after eradication of H. pylori ¹.

We hope to develop a NAMH-activated genetic switch to trigger the downstream therapy system. AiiA is believed to effectively inhibit H. pylori biofilm formation and likely to block the virulence proteins on H. pylori ^{2 3}. In our yeast, AiiA is designed to be secreted as the main destructor against H. pylori biofilm.

The Sensing Module consists of a heterologous GPCR signaling pathway that allows for NAMH detection into targeted therapeutic action, coupling the histamine receptor to the endogenous yeast mating pathway, and finally drives the expression of the mating-responsive pFUS1 promoter, secreting the therapeutic AiiA enzyme.

Integrated Validation Strategy

We aimed to achieve mutual validation and supplementation between dry lab and wet lab results: wet lab experiments provide empirical validation of the biological system’s functionality, while dry lab simulations predict the structure and efficacy of different GPCRs.

Workflow for Biosensor Construction and Validation

The implementation of the sensing system followed a structured engineering workflow. Our overall strategy involved engineering the native yeast mating pathway by knocking out the endogenous receptor gene STE2 to eliminate background signaling and the inhibitory GTPase gene SST2 to improve the signalling efficiency, and then introducing our customized genetic modules to achieve a NAMH-responsive system.

We first designed and synthesized plasmid components encoding different histamine GPCRs, the modified chimeric Gα proteins (Gpa1), and the reporter or therapeutic genes (mCherry/AiiA), under the control of the pathway-responsive promoter pFUS1, all of which based on the sequences that literatures provide ⁴,⁵,⁶,⁷,⁸. Each plasmid was assembled using In-Fusion assembly methods and subsequently transformed into the corresponding chassis organism.

Positive clones were selected on appropriate antibiotic media or dropout media and initially validated by colony PCR to confirm the presence and size of the inserts, and then sequenced to ensure full identity.

In the preliminary validation experiments, histamine was employed as the inducer to activate reporter gene expression, because histamine functions as an agonist compatible with multiple histamine receptor subtypes and is more readily accessible compared with NAMH. Therefore, functional validation was performed by inducing the transformed yeast with histamine and quantitatively measuring the output response, which in the case of the reporter construct was the induction of mCherry fluorescence.

Results

Recovery and Plating of Core Strains and Plasmids

Core yeast strains and constructed plasmids were successfully recovered on plates.

Knockout Plasmid

Figure 1: E. coli library with pML104 plasmids.

Figure 2: E. coli library with d-sgRNA genes.

Sensory System Plasmid

Figure 3: Plate of the strain carrying the initial plasmids for modified vectors.

Figure 4: E. coli library for human histamine receptors. From A to D correspond to hH1R to hH4R, respectively.

Figure 5: Plate of the strain carrying the terminal plasmids.

Plasmid construction

Knockout plasmid

We successfully identified the initial vector and fragments required for knockout plasmid construction.

Figure 6: Electrophoresis results for KanMx4 and sgRNA.

We successfully linearized the initial vector, pML104, into two fragments using two primer pairs. These fragments were then joined with the corresponding inserts via In-Fusion cloning to assemble the final knockout plasmid. The assembled plasmid was successfully transformed into E. coli DH5α, yielding transformants. Sequencing confirmed the successful construction of the knockout plasmid.

Figure 7: Electrophoresis results for linearized pML104.

Figure 8: Transformation of the knockout plasmid into E. coli DH5α.

Figure 9: Sequencing results of the knockout plasmid.

Figure 10: Transformation of the knockout plasmid into yeast.

Sensory System Plasmid

We successfully identified the initial vector and fragments required for sensory system plasmid construction.

Figure 11: Electrophoresis results for the initial plasmids used for modified vector construction.

Figure 12: Electrophoresis results for hH1R to hH4R.

Figure 13: Electrophoresis results for Chimeric GPCR.

Figure 14: Electrophoresis results for GAP1ΔC5.

Figure 15: Electrophoresis results for P2Y2-P2A-Gpa, hH3R-P2A-Gpai, hH3R-P2A-mCherry.

Following the synthesis schematic, the initial vector was linearized and assembled with the corresponding fragments via In-Fusion cloning to generate the intermediate vectors, pESC-URA-MatRep-pGAP and pESC-HIS-pGAP-αMF. Both assembled intermediate vectors were subsequently transformed into E. coli DH5α, yielding transformants. The successful construction of both intermediate plasmids was verified by PCR analysis.

Figure 16: Electrophoresis results for intermediate vector construction.

Figure 17: Transformation results of intermediate vector construction.

Following the synthesis schematic, the intermediate vectors were linearized and subsequently joined with their respective fragments via In-Fusion Assembly to yield the terminal plasmids. The assembled terminal plasmids were successfully transformed into E. coli DH5α, generating transformants. Gene sequencing results confirmed the successful construction of 10 out of the 14 target terminal plasmids.

Figure 18: Transformation of the terminal plasmids into E. coli DH5α.

Figure 19: Sequencing results of the terminal plasmids.

Plasmids were extracted from the transformants and transformed into yeast, followed by screening on the corresponding selective medium. The results demonstrated that we successfully obtained yeast single colonies with successful transformation, which can be used for subsequent screening and functional verification experiments.

Figure 20: The terminal plasmid was transformed into the knockout yeast strain.

Functional Validation

We induced the transformed yeast with histamine and quantitatively measuring the output response to confirm whether the sensory system functions as expected, and to select the most successfully modified GPCR for H. pylori detection.

Using mcherry as reporter gene, we tested strategies of modifying yeast signaling pathway by recording and comparing fluorescence data and OD₆₀₀ data of our yeast.

Growth

We also recorded OD 600 for each type of yeast, in case the growth of yeast would influence fluorescence result. However, the OD 600 curve of our yeast turned out to be normal.

Figure 21: The OD-time curves of yeast transformed with the sensory system plasmid were demonstrated under varying histamine concentrations, as well as the OD-time curves of yeast transformed with the P2Y2 plasmid under different dATP concentrations.

Fluorescence

We examined the fluorescence data every 20 minutes within 8 hours of cultivation. The result is shown as follows.

We also used the system of 2022 NEU iGEM team as a positive control, but the result did not come exactly as expected, either.

Figure 22: The RF-time curves of yeast transformed with the sensory system plasmid were presented under varying histamine concentrations, as well as the RF-time curves of yeast transformed with the P2Y2 plasmid under different dATP concentrations.

We can see the receptor did not show obvious response to histamine gradient, and this result will be later discussed.

Discussion

Molecular dynamics simulations performed in our dry-lab provided preliminary validation of the overall feasibility of our design. Nevertheless, the results obtained from subsequent experimental verification did not fully meet our expectations. Specifically, even in the absence of histamine, we observed a time-dependent increase in fluorescence intensity, and this trend appeared to be largely independent of histamine concentration. These findings suggest that the downstream promoter of our engineered GPCR exhibits undesired transcriptional leakage, resulting in basal gene expression under non-induced conditions. Such leaky expression likely contributed to substantial background noise, thereby obscuring the authentic histamine-induced signaling response.

In our current framework, mathematical approaches alone are insufficient to effectively eliminate this background signal. Consequently, future work will focus on enhancing the responsiveness and signal specificity of the system by mitigating promoter leakage. For chimeric GPCR, we learned that overexpression of the G protein α-subunit (Gpa1) has been validated as an effective strategy for suppressing background noise⁹. Gpa1 can attenuate the basal activity of the signaling pathway, thereby reducing background fluorescence and improving the signal-to-noise ratio. We plan to adopt this approach in future experiments to further optimize our system. Other potential strategy includes incorporating additional regulatory logic elements—such as CASwitch-based genetic circuits ¹⁰—which can strongly suppress basal transcriptional activity in the absence of an inducer. In parallel, we aim to explore alternative signaling pathways to identify optimized configurations with reduced noise and improved dynamic range. Screening additional combinations of G proteins with human GPCRs will also be included in our future research plans.

Moreover, the present dataset does not allow us to definitively exclude technical factors that may have contributed to the observed discrepancies, such as incomplete deletion of the Ste2 or Sst2 gene within the yeast mating pathway, which could result in unintended signaling. We are also considering knocking out additional genes involved in the yeast mating factor signaling pathway, such as FAR1, as the expression of these genes may interfere with the proper function of the engineered signaling pathway. Additional experiments will therefore be conducted to systematically investigate these possibilities and to clarify the underlying causes of the unexpected results. Through these efforts, we aim to iteratively refine the system toward a more robust and tightly regulated biosensing platform.

Outlook

Through the combined use of molecular dynamics simulations and preliminary histamine-based wet-lab screening, we have identified several potential engineering configurations. After further optimization of the sensory system to minimize background noise and achieve stable readouts, we will perform final validation using NAMH as the target biomarker.

Beyond NAMH, our design concept can be extended to a broader range of biomarkers to enhance detection reliability and sensitivity. Future work will also focus on refining the engineered signaling pathways for improved precision and dynamic control. Ultimately, we aim to establish a cost-effective and efficient engineering strategy that enables robust, biomarker-dependent gene expression in the yeast chassis under H. pylori–specific induction. This goal represents the central direction of our forthcoming experimental and system optimization efforts.

Reference

Therapy System

Abstract

This study aimed to investigate the effects of different yeast strains on Pseudomonas aeruginosa biofilm formation and to evaluate the degradation efficacy of AiiA protein against established biofilms. A reliable biofilm quantification protocol was established using crystal violet staining combined with microplate reader measurement at OD550. Experiments were conducted in M63 medium, comparing P. aeruginosa monocultures with co-cultures containing either Saccharomyces cerevisiae (auxotrophic strain) or Saccharomyces boulardii. Biofilm formation was dynamically monitored over a 24-hour period. Furthermore, exogenous AiiA protein was applied at concentrations ranging from 1 to 10 μg/mL to assess its inhibitory effects. Results indicated that both co-culture conditions and AiiA treatment significantly suppressed biofilm growth, providing experimental support for future engineering of yeast strains capable of secreting AiiA to target Helicobacter pylori biofilms.

Introduction

Background Information

Pseudomonas aeruginosa is a common Gram-negative opportunistic pathogen renowned for its environmental adaptability and robust biofilm-forming capacity. The ability to form complex biofilm structures on biotic and abiotic surfaces significantly contributes to its pathogenicity and antibiotic resistance. Biofilms, serving as a protective bacterial growth mode, considerably enhance bacterial survival against antimicrobial agents and host immune responses, presenting a major challenge in clinical infection management.

The crystal violet staining method represents a classical technique for biofilm quantification. This approach relies on the binding of dye to polysaccharides and proteins within the biofilm matrix, followed by solvent elution and spectrophotometric measurement to indirectly determine biofilm biomass. When combined with 96-well plate culture and microplate reader detection, this method offers simplicity, high throughput, and good reproducibility, making it a standard tool in in vitro biofilm research.

Studies on microbial interactions often utilize co-culture systems to explore cross-species dynamics. The two yeast strains selected for this study—auxotrophic S. cerevisiae and S. boulardii—are well-established model organisms. Their co-culture with P. aeruginosa may influence biofilm development through mechanisms such as nutrient competition, spatial exclusion, or signal interference.

AiiA protein, an acyl-homoserine lactonase derived from Bacillus species, degrades acyl-homoserine lactones (AHLs), which are key signaling molecules in bacterial quorum sensing (QS). Given the crucial role of QS in regulating P. aeruginosa biofilm formation, AiiA represents a promising quorum-quenching agent with potential anti-biofilm applications.

Figure 1: Conformational arrangement of conserved domains of AiiA protein. (a) Zinc ion coordinates with amino acid residues (b) The catalytic pocket of AiiA protein (Gopalakrishnan V, 2024)

Goal

This subproject aims to establish a robust in vitro method for quantifying P. aeruginosa biofilms and to systematically evaluate the inhibitory effects of yeast co-culture and exogenous AiiA protein treatment on biofilm formation. Through time-course and dose-response experiments, we seek to analyze biofilm dynamics under various conditions. The findings will provide a technical foundation for developing engineered yeast strains capable of secreting AiiA, supporting future strategies for targeted therapy against H. pylori biofilms.

Results

The growth curve of the biofilm produced by the co-culture of Pseudomonas aeruginosa and yeast

Since our experiment used four-deficient yeast and Brachlia yeast as carriers, in order to determine the effects of four-deficient yeast and Brachlia yeast on the formation of biofilms by Pseudomonas aeruginosa, we used M63 medium as a blank control. We co-cultured four-deficient yeast and Brachlia yeast with Pseudomonas aeruginosa and incubated them at 37°C. We set up a time gradient, performed crystal violet staining at corresponding time points, and used a plate washer to uniformly wash with 30% acetic acid to measure the absorbance at 550nm. We obtained three growth curves for the blank, four-deficient yeast, Brachlia yeast, and the co-culture of Pseudomonas aeruginosa and these three. Within a certain range, we successfully verified that the biofilm gradually increased with time, and the co-culture of four-deficient yeast and Brachlia yeast with Pseudomonas aeruginosa had no significant effect on the formation of the biofilm.

Figure 2: A representative result for biofilm formation assays performed for Pseudomonas aeruginosa, Pseudomonas fluorescens and Staphylococcus aureus. (A) A side view of the well with a biofilm of P. aeruginosa (8 hrs, 37°C). (B) A side view of the well with a biofilm of P. fluorescens (6 hrs, 30°C). (C) A top-down view of the biofilm formed by S. aureus in a flat-bottom microtiter plate (two wells, 24 hrs, 37°C). P. aeruginosa and P. fluorescens are both motile organisms and form a biofilm at the air-liquid interface. S. aureus is non-motile and forms a biofilm on the bottom of the well. ^[1](#ref1)

Figure 3: Effect of yeast co-culture on P. aeruginosa biofilm formation (with logarithmic fitted curve). We tested whether co-culturing P. aeruginosa with two different yeast strains would affect its biofilm formation over 24 hours. The bacteria's biofilm accumulated steadily over time. The presense of both yeast strains decreased the biofilm formation.

Utilizing N-acyl homoserine lactonase (AiiA) for Biofilm Degrading and Resistance Reversal

We next evaluated whether the AiiA protein, which disrupts bacterial quorum sensing, could inhibit biofilm formation. To quantify the efficacy of AiiA and determine its effective concentration, we treated P. aeruginosa cultures with increasing concentrations of purified AiiA enzyme. Biofilm formation was assessed by measuring the absorbance of crystal violet at 550 nm. As shown in Figure 3, AiiA treatment resulted in a significant and dose-dependent reduction in biofilm biomass. The inhibitory effect was apparent even at the lowest tested concentration (1 μg/mL). When the AiiA concentration was increased to 2 μg/mL and 5 μg/mL, the level of inhibition progressively enhanced. These results clearly demonstrate that AiiA effectively interferes with the quorum sensing system of P. aeruginosa, limiting its ability to form mature biofilms. And the anti-biofilm activity of AiiA is concentration-dependent, indicating a positive correlation between AiiA concentration and the inhibitory effect.

Figure 4: Dose-dependent inhibition of P. aeruginosa biofilm formation by AiiA. P. aeruginosa was incubated with different concentrations of AiiA. Biofilm formation was quantified by crystal violet staining and measured at OD550. A clear negative correlation between AiiA concentration and biofilm formation was observed.

Discussion

We successfully characterized the biofilm formation dynamics of Pseudomonas aeruginosa and evaluated the inhibitory effects of yeast co-culture and AiiA protein treatment.

Our basic experiments confirmed that the two strains of yeasts did inhibit the biofilm formation ability of Pseudomonas aeruginosa when co-cultured with it. This confirmed the yeast itself has a certain inhibitory effect on the biofilm formation of Pseudomonas aeruginosa, as them reported as probiotics.

The key finding was that when AiiA was present, it exerted a significant and sustained inhibitory effect on the biofilm formation of Pseudomonas aeruginosa (Figure 3). AiiA can degrade the key autoinducers (AHLs) of Pseudomonas aeruginosa, thereby interrupting its quorum sensing (QS) signaling pathway. Along with that, AiiA functioned through a still unknown mechanism thus inhibiting the virulence factor of P. aeruginosa and H. pylori. Our results indicate that the AiiA protein can effectively penetrate and act on the microenvironment and ultimately weaken the ability of Pseudomonas aeruginosa to form structured biofilms. In conclusion, we have successfully verified the feasibility of the strategy of “utilizing engineered yeast as a AiiA protein delivery platform to inhibit bacterial biofilms”. This system presents a potential and novel anti-biofilm approach, with the advantage of achieving targeted signal interference through biological means rather than direct killing, which may help reduce the selective pressure brought by traditional antibiotics.

Outlook

Based on the concept feasibility verified in this study, future work will focus on deepening mechanism research and promoting practical applications. Firstly, the concentration dynamics of AHLs in the co-culture system can be quantitatively detected at the molecular level, and the yeast vector can be optimized to enhance the delivery efficiency of AiiA. Further, the anti-biofilm efficacy of this strategy should be evaluated in more complex in vivo models or real infection environments (such as on the surface of biomaterials). Additionally, the potential of combining this strategy with conventional antibiotics to test its synergistic potential in disintegrating mature biofilms and reversing drug resistance can be explored, which will provide key evidence for the development of novel combination therapies.

References

1. Pais P, Almeida V, Yılmaz M, Teixeira MC. Saccharomyces boulardii: What Makes It Tick as Successful Probiotic?. J Fungi (Basel). 2020;6(2):78. Published 2020 Jun 4. doi:10.3390/jof6020078 ↖

2. Gopalakrishnan, V., Saravanan, V., Mahendran, M. I. M. S., & Kumar, M. P. N. (2024). Helicobacter pylori biofilm interference by N-acyl homoserine lactonases: in vitro and in silico approaches. Molecular biology reports, 51(1), 1106. https://doi.org/10.1007/s11033-024-10013-w

3. Biradar, B., & Devi, P. (2011). Quorum sensing in plaque biofilms: challenges and future prospects. The journal of contemporary dental practice, 12(6), 479–485. https://doi.org/10.5005/jp-journals-10024-1080

4. Rajesh, P., & Rai, V.R. (2016). Inhibition of QS-regulated virulence factors in Pseudomonas aeruginosa PAO1 and Pectobacterium carotovorum by AHL-lactonase of endophytic bacterium Bacillus cereus VT96. Biocatalysis and agricultural biotechnology, 7, 154-163. https://doi.org/10.1016/j.bcab.2016.06.003.

5. Rajesh, P. S., & Rai, V. R. (2015). Purification and antibiofilm activity of AHL-lactonase from endophytic Enterobacter aerogenes VT66. International journal of biological macromolecules, 81, 1046–1052. https://doi.org/10.1016/j.ijbiomac.2015.09.048

6. Rajesh, P. S., & Ravishankar Rai, V. (2014). Quorum quenching activity in cell-free lysate of endophytic bacteria isolated from Pterocarpus santalinus Linn., and its effect on quorum sensing regulated biofilm in Pseudomonas aeruginosa PAO1. Microbiological research, 169(7-8), 561–569. https://doi.org/10.1016/j.micres.2013.10.005

7. Smith, R. S., Harris, S. G., Phipps, R., & Iglewski, B. (2002). The Pseudomonas aeruginosa quorum-sensing molecule N-(3-oxododecanoyl)homoserine lactone contributes to virulence and induces inflammation in vivo. Journal of bacteriology, 184(4), 1132–1139. https://doi.org/10.1128/jb.184.4.1132-1139.2002

8. Dong, Y. H., Xu, J. L., Li, X. Z., & Zhang, L. H. (2000). AiiA, an enzyme that inactivates the acylhomoserine lactone quorum-sensing signal and attenuates the virulence of Erwinia carotovora. Proceedings of the National Academy of Sciences of the United States of America, 97(7), 3526–3531. https://doi.org/10.1073/pnas.97.7.3526

9. Aybey, A., & Demirkan, E. (2016). Inhibition of quorum sensing-controlled virulence factors in Pseudomonas aeruginosa by human serum paraoxonase. Journal of medical microbiology, 65(2), 105–113. https://doi.org/10.1099/jmm.0.000206

10. Tuon, F. F., Dantas, L. R., Suss, P. H., & Tasca Ribeiro, V. S. (2022). Pathogenesis of the Pseudomonas aeruginosa Biofilm: A Review. Pathogens (Basel, Switzerland), 11(3), 300. https://doi.org/10.3390/pathogens11030300

11. Sompiyachoke, K., & Elias, M. H. (2023). Engineering Quorum Quenching Acylases with Improved Kinetic and Biochemical Properties. bioRxiv : the preprint server for biology, 2023.09.01.555929. https://doi.org/10.1101/2023.09.01.555929