Asthma is a chronic inflammatory disease of the airways. It is identified by airway hyperresponsiveness, variable airflow obstruction, and recurrent respiratory symptoms such as wheezing, coughing, and shortness of breath [1]. Besides, it results in the release of alarmins such as TSLP that induce the release of inflammatory cytokines (IL-4, IL-5, IL-13) and other mediators [2].
Thymic Stromal Lymphopoietin (TSLP) has a core function in asthma . It acts as a main "alarmin" that induces multifarious asthma -related pathways, such as Type 2 and non-Type 2 inflammatory pathways. This process occurs through the activation of dendritic cells and group 2 innate lymphoid cells (ILC2s) [2,3] This wide activation leads to airway inflammation and remodeling that plays a crucial role in the long term symptoms seen in severe asthmatic patients [3].
Its role in initiating and propagating the inflammatory cascade makes it a significant target of new therapeutic approaches[3].
Medical treatment of asthma still largely aims to control symptoms instead of targeting the disease pathway.For example, inhaled corticosteroids are the most widely prescribed as preventive agents used to reduce airway inflammation. Their prolonged use has been associated with significant adverse effects and shows limited efficacy in patients with non-type 2 (neutrophilic) asthma [4]. Other therapeutic options, such as the monoclonal antibodies and small-molecule drugs, face some limitations, such as rapid clearance from the lungs, systemic side effects, and reduced efficacy for cases of severe or corticosteroid-resistant asthma [5,6] for more details see Description page.
Figure(2) Pulsatile vs. continuous release:While corticosteroids provide only short bursts of activity, Lactobacillus plantarum enables continuous therapeutic release, offering a longer half-life and sustainable treatment.
Our project introduces an advanced system of therapeutic delivery: engineered probiotics (Lactobacillus plantarum) formulated for inhalation and designed to function as living therapeutics [7].
Unlike conventional small molecules or biologics that demand multiple administrations and often undergo rapid clearance from the lungs, living probiotics have the potential to colonize transiently. As they sense the diseased microenvironment, they adaptively release targeted therapeutic agents [8].
Figure(3) Design Advantages of Inhalable Engineered Lactobacillus plantarum Therapeutics
This design establishes a modular platform for drug delivery. Although our proof-of-concept focuses on severe asthma through the targeting of TSLP with siRNA, the same framework can be reprogrammed to address a variety of diseases. By swapping the therapeutic payload or the sensing circuit, this platform could be adapted to both Pulmonary diseases and Gastrointestinal diseases.
Figure(4) Therapeutic Versatility of the Modular Probiotic Delivery System: This illustration showcases the adaptability of the engineered Lactobacillus plantarum chassis for treating both pulmonary and gastrointestinal diseases.
Our goal is to design a therapeutic platform that unveils the potential of engineered probiotics for siRNA-based asthma treatment. Specifically, we focus on targeting thymic stromal lymphopoietin (TSLP), a cytokine that drives chronic inflammation and corticosteroid resistance in severe asthma [2,4].
To achieve that, we chose a modular Synthetic biology approach, where each one of the genetic circuits has a crucial function in the system. The genetically engineered Lactobacillus plantarum is programmed to sense inflammation, produce therapeutic siRNA, package it into Bacterial membrane vesicles, ensure its release into host cells, and ensure safety by using a double kill switch mechanism [7,9].
Together , these components form the foundation of our integrated therapeutic platform, which we call PRESS — Programmable Respiratory Microbiome as an Endogenous Sustainable System for drug delivery. On this page, we explain the elements of our design .
In recent years, probiotics lay the foundation for a broader platform for therapeutic delivery besides being dietary supplements. Their long history of safe use, colonization on mucosal surfaces, and their immunomodulatory properties position them as candidates for next-generation biotherapeutics. Engineered probiotics could be developed as “living medicines” capable of producing therapeutic molecules and delivering them directly to the disease sites in the gut or lungs. In addition such delivery helps to avoid side effects and allows for a more targeted modulation of immunoinflammatory pathways.
Among probiotic species, Lactobacillus plantarum is not only the chassis but also a therapeutic component of our project. This species was deliberately chosen as it is a Generally Recognized As Safe (GRAS) probiotic and is widely used in food and supplements. In addition, it has confirmed anti-inflammatory properties such as attenuation of NF-κB and MAPK signaling pathways. This causes a decrease in important markers of asthma such as IgE and pro-inflammatory cytokines, while promoting immune balance and enhancing the body's antioxidant defense mechanisms [10,11]. Therefore, we sought that L. plantarum may offer a beneficial immunomodulatory strategy.
It has natural immunomodulatory effects that synergize with the synthetic siRNA-based approach, creating a dual-action therapeutic platform [10].
Gene silencing is a native regulatory process utilized by cells to control gene expression for selected genes. In the fields of Synthetic biology, gene silencing can be utilized to turn off disease-related genes to offer selective therapeutic applications [12].
● Transcriptional silencing – repression of gene transcription.
● Post-transcriptional silencing – destroying mRNA prior to translation.
● Translational silencing – blocking ribosome activity [12].
Among these, post-transcriptional gene silencing (PTGS) has been most effective with RNA-based therapy [13].
Figure(6): Key advantages of Post-Transcriptional Gene Silencing (PTGS) for therapeutic applications
Several molecular approaches exist for silencing target genes as the following:
Figure(7): Classification of major gene silencing techniques used in molecular biology. This schematic outlines four distinct approaches to gene silencing, each represent a mechanism for inhibiting gene expression
● Antisense oligonucleotides (ASOs) are short RNA sequences that bind to mRNA to block translation but must be chemically modified for stabilization [14].
● Ribozymes & DNAzymes: nucleic acids with a catalytic activity to cleave target RNA sequences, but are much less therapeutically utilized [12].
● Small interfering RNA (siRNA): Briefly double-stranded RNA that utilizes the native RNA interference (RNAi) pathway to induce efficient mRNA degradation [13].
● CRISPR-based silencing (CRISPRi): using dead Cas proteins fused to repressors to disrupt transcription at a DNA level [12].
Small interfering RNAs (siRNAs) are mainly capable of gene silencing due to their specificity and efficiency in targeting mRNAs via the cell's natural RNA interference (RNAi) pathway [13,15]. In some cases, siRNAs do not require chemical modifications to be stable, and therefore allow RISC binding to mRNA for degradation, unlike ASOs [14,15]. They are an easier form to design and more elegant than ribozymes and DNAzymes [12]. Unlike CRISPRi, siRNAs typically target molecules in the cytoplasm and are thus easier to deliver [12,39]. This makes siRNAs powerful and versatile therapeutic tools against aberrant gene expression [19,40].
siRNAs consist of 20–25 nucleotide-long RNA duplexes that guide gene silencing processes by mimicking the native RNAi processes. After entering the cytoplasm, double-stranded RNAs act as follows inside the host machinery of RNA interference:
● Incorporation into RISC
One of the two strands of the siRNA duplex (guide strand), is integrated into the (RISC) complex [15].
The other strand (passenger strand) is degraded.
● Target Recognition
The guide strand leads the RISC complex towards complementary sequences on target mRNA molecules.
● mRNA Cleavage and Silencing
The AGO (Argonaute) protein, the catalytic center of RISC, cleaves the mRNA that's been bound.
This process blocks the production of certain proteins by breaking down or silencing their RNA, making it a precise and effective tool in RNA-based therapies [15]
Figure(8):siRNA silencing pathway This diagram shows the intracellular processing of siRNA into cells.
CO-BERA (combinatorial Bio-Engineered RNA agent) is an advanced synthetic RNA therapeutic platform. Instead of using a single individual siRNA, CO-BERA is engineered as a long-stable RNA scaffold with multiple siRNA units embedded into a single transcript [17,18]. Upon expression within a delivery device or host, the transcript is cleaved by Dicer enzyme into functional siRNAs that can either target multiple mRNAs or synergize against a single gene to enhance therapeutic efficacy [16,17].
CO-BERA’s modular design enables broad applicability across cancer, viral infections, and chronic inflammatory diseases [19].
Figure(9): CO-BERA silencing pathway This diagram illustrates the intracellular processing of CO-BERA molecules delivered into lung cells.
Compared with conventional siRNA, CO-BERA offers:
● Higher biological stability due to its structured RNA scaffold [17].
● Multi-functionality, as one transcript yields several therapeutic RNAs [18].
● Scalability, since it can be expressed directly in engineered microbes, bypassing costly chemical synthesis [17].
Figure(10): Comparative overview of conventional siRNA and CO-BERA platforms.
In Our project, CO-BERAwas implemented as the core therapeutic scaffold to silence Thymic Stromal Lymphopoietin (TSLP), a cytokine that plays a pivotal role in driving airway inflammation and remodeling in asthma [20].
Structure in Our System: CO-BERA was rationally engineered to embed two distinct siRNA sequences targeting TSLP mRNA within a long and stable RNA framework. This design improves RNA stability and enables the simultaneous generation of multiple siRNAs from a single transcript [17,18].
Mechanism of Action: Once CO-BERA inside the cytoplasm, Dicer cleaves CO-BERA into two active siRNAs. These are loaded into the RNA-induced silencing complex (RISC), where Argonaute (AGO) mediates precise degradation of TSLP mRNA, suppressing its expression and reducing inflammation [21]. We validated AGO-siRNA binding through molecular dynamics simulations using OpenMM, demonstrating stable complex formation with an RMSD graph. See Model page for details.
Figure(11): CO-BERA silencing pathway This diagram illustrates the intracellular processing of CO-BERA molecules delivered into lung cells.
To ensure effectiveness, we used design tools such as siDirect and RNAxs, selecting siRNA sequences based on Ui-Tei rules to minimize off-target effects while maximizing silencing efficiency[22].
Figure(12): Rational Design of our siRNA based on Ui-Tei rules.
Figure(13): shows siDirect v2.1 tool, which is used for designing highly specific and effective siRNA sequences as it is able to predict functional candidates and that is based on established rules like Ui-Tei, Reynolds and Amarzguioui plus thermodynamic properties, in addition it evaluates each candidate's target position, sequence, and potential off-target effects against.
It also provides detailed metrics like seed-duplex stability and mismatch tolerance. The table shows and summarizes all of that . .
We tuned our therapeutic agent (CO-BERA system) to remain silent under healthy airway conditions and activate specifically during asthmatic inflammation. This conditioned activation minimizes off-target effects and improves therapeutic specificity [23].
To achieve this, CO-BERA expression is placed under the control of the oxidative stress–inducible promoter KatA, which responds to elevated hydrogen peroxide (H₂O₂) levels. Since H₂O₂ is abundantly released in the inflamed airway during asthma attacks, activation of the KatA promoter ensures that CO-BERA production occurs only in inflamed lung environments, thereby providing both spatial and conditional control over therapy [23].
In detail, the KatA promoter is normally kept repressed by the PerR regulator, a peroxide-sensing transcriptional repressor. Under non-inflammatory conditions, PerR binds to the promoter region and blocks transcription. However, in inflamed airways, elevated H₂O₂ levels act as a signal
● H₂O₂ oxidizes specific histidine and cysteine residues in the PerR protein.
● This oxidative modification reduces PerR’s DNA-binding affinity.
● As a result, PerR dissociates from the KatA promoter.
● Once PerR is removed, RNA polymerase can access the promoter and initiate transcription of CO-BERA.
Normal conditions: The transcriptional regulator PerR represses the KatA promoter, keeping CO-BERA silent.
Inflamed conditions: During asthma attacks, high levels of hydrogen peroxide (H₂O₂) are generated in the airway epithelium. H₂O₂ inactivates PerR, derepressing the KatA promoter, which then drives CO-BERA transcription [23].
Figure(14): It shows the regulation of gene expression via the KatA promoter under oxidative stress. Under normal conditions, the PerR repressor binds the promoter and blocks transcription. In inflamed environments, elevated H₂O₂ levels deactivate PerR, allowing CO-BERA to initiate transcription through the KatA promoter. This mechanism enables oxidative stress–responsive control of therapeutic gene expression.
During initial testing, our model showed that the pKatA promoter displayed detectable basal activity even under non-inflammatory conditions. Since unwanted expression of CO-BERA in healthy tissues could reduce safety and specificity, we introduced a dual-layer repression system to eliminate leakage as detailed on the Model page.
In bacteria, gene expression is often regulated by a simple and powerful mechanism: the operator–repressor system.
● Operator: a short DNA element located near or downstream of the promoter. It acts as a regulatory switch.
● Repressor protein: a DNA-binding protein that specifically recognizes the operator sequence. When bound, it physically blocks RNA polymerase from initiating transcription.
Figure(15): shows how a repressor protein binds to an operator downstream of the promoter to block RNA polymerase and prevent gene transcription.
operator–repressor systems are widely used as logic gates that allow precise control of engineered circuits. By adding or combining operators and repressors, one can design promoters that respond only under specific conditions, reducing unwanted background activity.
To achieve precise regulation of our therapeutic circuit and eliminate unwanted expression under healthy conditions, we implemented a multi-layer operator–repressor system.
At the core of This design is the Rep operator–repressor module. The operator sequence was placed downstream of the oxidative stress–responsive KatA promoter. Under basal conditions, the Rep repressor protein binds tightly to this operator, preventing RNA polymerase from initiating transcription [24]. This ensures that the KatA promoter remains silent in the absence of inflammatory signals, thereby avoiding unnecessary production of CO-BERA in healthy lung tissue.
The Rep repressor itself is produced constitutively from the p32 promoter, which provides a constant supply of repressor protein to maintain suppression. To make This system environmentally responsive, we embedded a Lac operator within the P32 promoter. This allows regulation of Rep repressor production depending on airway conditions.
In healthy airways (pH > 7), the p32 promoter remains active, continuously expressing the Rep repressor. The Rep protein binds the operator downstream of KatA, keeping CO-BERA silenced even if low background H₂O₂ levels are present [25].
Figure(16): Diagram illustrating a pH-responsive gene regulation system controlling CO-BERA expression. At high pH(>7), the P32 promoter drives Rep repressor production, which inhibits KatA-driven expression.
In contrast, during inflamed conditions (PH ≤ 6.9), the pH-sensitive promoter P170-CP25 becomes active and drives the production of LacR, a repressor that binds to the Lac operator within p32. This binding effectively inhibits p32, reducing Rep repressor expression. With less Rep available, the KatA promoter is relieved of repression. Now, in the presence of elevated H₂O₂ (a hallmark of inflamed lungs), KatA is fully activated and drives CO-BERA expression.
Figure(17): Under acidic conditions (pH ≤ 6.9), LacR is activated via the P170-CP25 promoter, suppressing p32 and reducing Rep levels. This relief of repression enables KatA to activate CO-BERA expression, ensuring precise control in inflamed, low-pH environments.
This design makes CO-BERA expression dual-gated, requiring both oxidative stress (H₂O₂) and acidic pH to be expressed. which controls and minimizes off-target activity, maximizes therapeutic specificity, and ensures activation only in the inflamed lung environment.
● Healthy airway (low H₂O₂, neutral pH): PerR represses KatA + Rep repressor continuously suppresses → No CO-BERA expression.
● Inflamed airway (high H₂O₂, low PH < 6.9): PerR inactivated by H₂O₂ + LacR suppresses Rep → KatA drives CO-BERA expression → Therapeutic siRNA produced for more details see Model page.
Figure(18):Schematic diagram illustrating the dual-input logic gate controlling CO-BERA expression. The system integrates two environmental signals—acidic pH(≤6.9) and oxidative stress (H₂O₂) to regulate gene activation via an AND gate. Acidic pH activates LacR, which supresses expression of Rep repressor, while H₂O₂ derepresses the activity of PerR. Both repressors converge to control the pKatA promoter, ensuring CO-BERA expression occurs only when both conditions are met. This design enables disease-context-specific activation, mimicking inflamed asthmatic lung environments.
Overall we could develop a dual conditioned silencing system but it is useless if it can't be loaded and delivered effectively From L. plantarum to lung epithelial cells through Bacterial membrane vesicles. So, we built a Precise Loading System.
Bacterial membrane vesicles (BMVs) are naturally secreted, nanoscale lipid bilayer particles that bacteria use to package and deliver proteins, nucleic acids, and signaling molecules. In recent years, they have gained increasing attention as biological delivery vehicles because of their unique advantages as shown in figure(19).
Building on these advantages, our design harnesses BMVs as a natural nanocarrier system to deliver the CO-BERA siRNA into lung epithelial cells. By engineering the loading mechanism, we ensure that therapeutic RNA molecules are selectively packaged into vesicles and efficiently released into target cells, maximizing delivery while minimizing off-target distribution.
| Feature | Gram-Negative OMVs | Gram-Positive MVs |
|---|---|---|
| Target Specificity | OMVs selectively bind host receptors and deliver hydrophobic signals to modulate host interactions. | MVs engage immune and epithelial cells via LTA and lipoproteins, often in a cell-specific manner. |
| Cargo Loading | OMVs sort cargo based on functional relevance; LPS composition influences vesicle content. | Cargo loading is less defined; vesicles may carry cytoplasmic and membrane components. |
| Release Mechanism | OMVs form via non-lytic blebbing, driven by membrane stress and lipid imbalance. | MVs bud through the thick peptidoglycan layer, aided by autolysins. |
| endocytosis Pathways | OMVs evade immune detection and enhance uptake by bypassing clathrin-mediated endocytosis. | MVs promote uptake through TLR2 engagement and lipid raft interactions. |
| Surface Composition | OMVs are coated with LPS, outer membrane proteins, adhesins, and lipoproteins. | MVs display LTA, lipoproteins, and phospholipids from the cytoplasmic membrane. |
loading RNA effectively into delivery Vesicles is representing one of the main challenges in developing RNA-based therapies. Without an active loading mechanism, only a small amount of therapeutic RNA naturally enters extracellular vesicles. This greatly limits the efficiency of treatment. Synthetic biology provides useful tools to address this issue. We can engineer Bacterial membrane vesicles (BMVs) with specific molecular systems that can selectively package the desired RNA [39].
Through binding of RNA-binding proteins (RBPs) with engineered RNA motifs, we can achieve accurate and efficient recruitment of therapeutic RNAs into vesicles. This method not only improves delivery but also increases stability and targeting. It is especially suitable for treating diseases like asthma [38].
To ensure that our therapeutic RNA (CO-BERA) is efficiently packaged into Bacterial membrane vesicles (BMVs), we designed a two-part loading system.
We used L7Ae, an archaeal RNA-binding protein that specifically recognizes the K-turn motif (C/D Box) in RNA. To anchor L7Ae to Bacterial membrane vesicles, we fused it to a transmembrane protein (TMP) [26].
● We tested different TMP candidates: Foldase PrsA and DUF4811.
● Using CHARMM-GUI to build the protein–membrane systems and GROMACS for molecular dynamics analysis as shown in Model page., we found that DUF4811–L7Ae was the most stable construct for membrane anchoring.This made DUF4811 the final choice for our loading system.
Figure(22): DUF4811–L7Ae fusion anchored in the bacterial membrane
To ensure strong and specific binding of CO-BERA RNA to L7Ae, we initially planned to add two copies of the C/D Box motif to the 3′ end of CO-BERA RNA, as this motif serves as a recognition site for L7Ae.
This strategy was based on findings from the iGEM 2024 HBMU-Taihe team, which showed that diagnostic RNAs with two C/D Boxes were most efficiently loaded into vesicles [27].
However, our molecular dynamics simulations Model page.suggested a potential conflict, indicating that a single C/D Box may provide greater accessibility and stronger binding compared to two copies more details in model page.
To resolve this debate and optimize RNA capture and packaging into bacterial membrane vesicles (BMVs), we ordered multiple variants of CO-BERA constructs with different numbers of C/D Box motifs (e.g., one, two, or more) for empirical testing and comparison.
Together, This system provides a robust and programmable way to load therapeutic RNA into bacterial vesicles, forming the foundation of our delivery strategy.
When therapeutic RNAs are delivered into mammalian cells, they are usually taken up by endocytosis. This means they enter the cell inside small vesicles called endosomes rather than directly into the cytosol.
1. Early Endosome (PH ~6.5):The RNA cargo is first enclosed in an early endosome soon after uptake.
2. Late Endosome (PH ~5.5):As time passes, the vesicle matures, the PH drops, and degradative enzymes start to accumulate.
3. Lysosome (PH ~4.5): Eventually, the cargo is delivered to lysosomes, where most RNA molecules are degraded [28].
So without an escape mechanism, most therapeutic RNA is destroyed before it can reach the cytosol. Experimental studies have mentioned that only 0.3% of delivered siRNA naturally escapes from endosomes into the cytosol; this low efficiency means that a large therapeutic dose is often needed, which increases costs and risks of side effects (28).
Endosomal entrapment is therefore considered one of the “grand obstacles" of RNA therapeutics, especially for vesicle-delivered RNAs targeting diseases [29].
Figure(25): showing three endosomal escape methods: cationic polymers (strong but cytotoxic), cell-penetrating peptides (require high doses), and pore-forming proteins (pH-sensitive, tightly regulated).
As part of our design , we chose to focus on pore-forming proteins given their promise to be engineered to be highly efficient for endosomal escape while having low off-target effects. As such, we incorporated Listeriolysin O (LLO) into our design . LLO is a well-characterized pore-forming protein from Listeria monocytogenes whose function is to cause phagosomal escape by forming pores in the vesicular membrane. However, using wild-type LLO raises safety concerns due to its pathogenic role in causing infection.
To address the limitations and side effects of wild-type LLO, we implemented two complementary strategies:
Instead of using the wild-type form, as its pathogenic role we engineered the L461T mutant (LLO-L461T).
● wild-type LLO: Activated only at acidic pH(late endosomes/lysosomes), but associated with high cytotoxicity and pathogenicity. In addition it is highly immunogenic as a key virulence factor, raising safety concerns in inflamed lung tissue (29). These adverse effects make wild-type LLO unsuitable in its native form, motivating us to adopt the safer, engineered LLO-L461T variant.
● Mutant LLO-L461T :Retains pore-forming activity, but becomes active at a higher, near-neutral pH. This means it triggers pore formation earlier, before RNA cargo is degraded, allowing rapid release of siRNA into the cytosol.
Thus, we decided to choose mutated LLO for our approach to ensure safe and efficient cytosolic delivery of CO-BERA [10, 29].
To further reduce safety risks, we placed LLO-L461T under the control of an H₂O₂-inducible promoter. This ensures that the protein is expressed only under oxidative stress conditions, which are hallmarks of inflamed asthmatic lungs.
● In healthy tissue: No or low oxidative stress → LLO is not expressed → no unintended cytotoxicity.
● In inflamed tissue: Elevated H₂O₂ → promoter activated → LLO-L461T produced → precise, timely endosomal escape of siRNA.
This dual-layer design (mutated protein + conditional promoter) creates a tightly controlled Endosomal Escape System and improves therapeutic efficiency.
Toxin–antitoxin (TA) systems are natural genetic systems found in many bacteria, in which a stable toxin protein is neutralized by a less stable antitoxin. These systems have been widely repurposed in Synthetic biology for biocontainment, ensuring that engineered microbes cannot continue existing or spread beyond their intended environment [31,32].
Kill-switch: TA systems may work as suicide switches that induce cell death in the absence of an inducer or by way of an environmental cue [33].
Plasmid stability: : Another use in biotech involves plasmid maintenance via selection, killing host cells that lose the plasmid so we could avoid the use of antibiotic resistance markers, which are a concern in clinical applications due to horizontal gene transfer.[34].
Biocontainment in probiotics: Since toxin-antitoxin systems prevent the uncontrolled growth of engineered strains, they also help reduce the risk of horizontal gene transfer (HGT) [31,32].
Although our chassis bacterium (Lactobacillus plantarum) is generally regarded as safe and naturally associated with the gut–lung axis, there remains a potential risk of translocation into the bloodstream or unintended survival outside the host. To mitigate this, we integrated the PemK/I toxin–antitoxin system as the core biosafety mechanism of our design [31,33,36]. This safeguard prevents the persistence of engineered bacteria in unintended niches and lowers the risk of horizontal gene transfer (HGT), as mentioned in the Safety page.
The system is based on the natural imbalance between the stable toxin (PemK) and the less stable antitoxin (PemI), which is rapidly degraded by host proteases [33]. Under normal lung conditions, both toxin and antitoxin are expressed, with the antitoxin binding and neutralizing the toxin. However, when specific environmental signals are detected (phosphate and low temperature), antitoxin expression is repressed, leading to its degradation and a relative increase in toxin levels, causing bacterial cell death [36].
We designed two independent environmental triggers to ensure strict containment:
In the bloodstream, phosphate levels are significantly higher than in the lung. The PhoB-regulated promoter represses toxin-antitoxin production under these conditions [35]. Because PemI is rapidly degraded, the toxin PemK accumulates and kills the bacteria, preventing what is called bacteremia (bacteria in the blood).
Outside the host, lower environmental temperatures cause the thermosensor RNA to adopt a tightly folded conformation that occludes the ribosome binding site, thereby repressing translation of the toxin–antitoxin system [37]. Again, the rapid degradation of the antitoxin (PemI) shifts the balance toward accumulation of the stable toxin (PemK), ensuring that bacteria cannot survive in the external environment [33,36].
This dual-layered design guarantees that our therapeutic bacteria are only active in the lung microenvironment, while any escape into the blood or the outside environment triggers self-destruction.
Our Toxin-Antitoxin model predicts low PemK toxin levels without phosphate and specific temperatures, and high toxin levels in their presence which will kill the bacteria, reducing escape probability. This validates our robust PemI/PemK biocontainment design [30–33]. See Model page.
Finally, Our project demonstrates how Synthetic biology can integrate multiple smart design layers into one therapeutic platform. With inflammation-responsive siRNA expression, precise vesicle loading, controlled endosomal escape, and robust biosafety safeguards, we have built a system that is not only effective but also safe and tightly regulated. This proof of concept shows the potential of our PRESS.
1-Ji T, Li H. T-helper cells and their cytokines in pathogenesis and treatment of asthma . Front Immunol. 2023;14:1149203. Available from: Frontiers in Immunology.
2-Theofani E, Tsitsopoulou A, Morianos I, Semitekolou M. Severe asthmatic responses: the impact of TSLP. Int J Mol Sci. 2023;24(8):7581. Available from: MDPI Journal.
3-Gauvreau GM, Sehmi R, Ambrose CS, Griffiths JM. Thymic stromal lymphopoietin: its role and potential as a therapeutic target in asthma . Expert Opin Ther Targets. 2020;24(8):777–792. Available from: Taylor & Francis Online.
4-Cojocaru E, Arcana RI, Radu S, Trofor AC, Cojocaru C. Challenges and opportunities in achieving asthma remission. J Clin Med. 2025;14(8):2835. Available from: MDPI Journal.
5-del-Barrio-Buesa S, Narrillos-Moraza Á, García-de-Pedro J, et al. Monoclonal antibodies in severe asthma : outcomes from real-world data. Front Med. 2025;12:1635688. Available from: Frontiers in Medicine.
6-Menzies-Gow A, Corren J, Bourdin A, et al. Tezepelumab in adults and adolescents with severe, uncontrolled asthma. N Engl J Med. 2023;388(22):2069–2081. Available from: https://doi.org/10.1056/NEJMoa2207989
7-Kleniewska P, Pawliczak R. Can probiotics be used in the prevention and treatment of bronchial asthma ? Pharmacol Rep. 2024;76:740–753.
8-Le Noci V, Guglielmetti S, Arioli S, Camisaschi C, Bianchi F, Sommariva M, Storti C, Triulzi T, Castelli C, Balsari A, Tagliabue E, Sfondrini L. Modulation of Pulmonary Microbiota by Antibiotic or Probiotic Aerosol Therapy: A Strategy to Promote Immunosurveillance against Lung Metastases. Cell Rep. 2018 Sep 25;24(13):3528-3538. doi: 10.1016/j.celrep.2018.08.090. PMID: 30257213.
9-Kim HJ, Lee JH, Kwon SY, et al. Effect of a combination of Lactiplantibacillus plantarum KC3 and Leonurus japonicus extracts in respiratory discomfort: a randomized, double-blind, placebo-controlled trial. Nutrients. 2024;16(13):2128.
10-Hao R, Liu Q, Wang L, et al. Anti-inflammatory effect of Lactiplantibacillus plantarum T1 cell-free supernatants through suppression of oxidative stress and NF-κB- and MAPK-signaling pathways. Appl Environ Microbiol. 2023;89(10):e00608-23. Available from: Applied and Environmental Microbiology.
11-Stankovic M, Veljovic K, Popovic N, et al. Lactobacillus brevis BGZLS10-17 and L. plantarum BGPKM22 exhibit anti-inflammatory effects by attenuating NF-κB and MAPK signaling in human bronchial epithelial cells. Int J Mol Sci. 2022;23(10):5547. Available from: International Journal of Molecular Sciences.
12-Malakondaiah S, Julius A, Ponnambalam D, et al. Gene silencing by RNA interference: a review. Genome Instab Dis. 2024;5:225–241. Available from: Springer article.
13-Collotta D, Bertocchi I, Chiapello E, Collino M. Antisense oligonucleotides: a novel frontier in pharmacological strategy. Front Pharmacol. 2023;14:1304342. Available from: Frontiers in Pharmacology.
14-Kilanowska A, Studzińska S. In vivo and in vitro studies of antisense oligonucleotides – a review. RSC Adv. 2020;10:34501–34516. Available from: RSC Publishing.
15-Kim YK, Kim VN. Processing of intronic microRNAs. EMBO J. 2007;26(3):775–783. Available from: EMBO Journal.
16-Vergani-Junior CA, Tonon-da-Silva G, Inan MD, Mori MA. DICER: structure, function, and regulation. Biophys Rev. 2021 Nov 16;13(6):1081-1090. doi: 10.1007/s12551-021-00902-w. PMID: 35059029; PMCID: PMC8724510.
17-Yu AM, Batra N, Tu MJ, Sweeney C. Novel approaches for efficient in vivo fermentation production of noncoding RNAs. Appl Microbiol Biotechnol. 2020;104(5):1927–1937. https://doi.org/10.1007/s00253-020-10350-3.
18-Yu AM, Jian C, Tu MJ, Ho PY. Bioengineering of a single long noncoding RNA molecule that carries multiple small RNAs. Appl Microbiol Biotechnol. 2019;103(3):1053–1063. https://doi.org/10.1007/s00253-019-09934-5.
19-Batra N, Tu MJ, Yu AM. Molecular engineering of functional siRNA agents. ACS Synth Biol. 2024;13(5):1123–1135.
20-Yuan T, Klossowski S, Afrazi M, et al. Silencing TSLP expression with a lung-targeted RNAi molecule suppresses pulmonary allergic inflammation. Eur Respir J. 2023;62(Suppl 67):PA1305. https://doi.org/10.1183/13993003.congress-2023.PA1305
21-Meister G. Argonaute proteins: functional insights and emerging roles. Nat Rev Genet. 2013;14(7):447–459. https://doi.org/10.1038/nrg3462.
22-Naito Y, Yoshimura J, Morishita S, Ui-Tei K. siDirect 2.0: updated software for designing functional siRNA with reduced seed-dependent off-target effect. BMC Bioinformatics. 2009;10:392.
23-Levi S, iGEM Technion-Israel. pKatA promoter for Bacillus subtilis: hydrogen peroxide–responsive regulation via PerR. iGEM Registry. 2019. Available from: https://parts.igem.org/Part:BBa_K293400
24-Blanch-Asensio M, Tadimarri VS, Wilk A, Sankaran S. Discovery of a high-performance phage-derived promoter/repressor system for probiotic Lactobacillus engineering. Microb Cell Fact. 2024;23:42. https://microbialcellfactories.biomedcentral.com/articles/10.1186/s12934-024-02302-7.
25-SMadsen SM, Arnau J, Vrang A, Givskov M, Israelsen H. Molecular characterization of the pH-inducible and growth phase-dependent promoter P170 of Lactococcus lactis. Mol Microbiol. 1999;32(1):75–87. https://doi.org/10.1046/j.1365-2958.1999.01326.x.
26-Rozhdestvensky TS, Tang TH, Tchirkova IV, Brosius J, Bachellerie JP, Hüttenhofer A. Binding of L7Ae protein to the K-turn of archaeal snoRNAs: a shared RNA binding motif for C/D and H/ACA box snoRNAs in Archaea. nucleic acids Res. 2003;31(3):869–877. https://doi.org/10.1093/nar/gkg175.
27-iGEM HBMU-Taihe 2024. GPC3 CDS–2×C/D Box RNA packaging construct. iGEM Registry Part:BBa_K5073025. Available from: BBa_K5073025 – iGEM Registry
28-Dowdy SF. Endosomal escape of RNA therapeutics: How do we solve this rate-limiting problem? RNA. 2023. https://rnajournal.cshlp.org/content/early/2023/01/20/rna.079507.122.full.pdf.
29-Plaza-GA I, Manzaneda-González V, Kisovec M, et al. pH-triggered endosomal escape of pore-forming Listeriolysin O toxin-coated gold nanoparticles. J Nanobiotechnol. 2019;17:108. https://jnanobiotechnology.biomedcentral.com/articles/10.1186/s12951-019-0543-6.
30-Pantoja Angles A, Valle-Pérez AU, Hauser C, Mahfouz MM. Microbial biocontainment systems for clinical, agricultural, and industrial applications. Front Bioeng Biotechnol. 2022;10:830200.
31-Wright O, Stan GB, Ellis T. Building-in biosafety for Synthetic biology. Microbiology. 2013;159(Pt 7):1221–1235.
32-Yamaguchi Y, Park JH, Inouye M. Toxin–antitoxin systems in bacteria and archaea. Annu Rev Genet. 2011;45:61–79.
33-Kroll J, Klinter S, Schneider C, Voß I, Steinbüchel A. Plasmid addiction systems: perspectives and applications in biotechnology. Microb Biotechnol. 2010;3(6):634–657.
34-iGEM UNILausanne 2020. PhoB-regulated promoter responsive to extracellular phosphate levels. iGEM Registry Part:BBa_K3482013. Available from: https://parts.igem.org/Part:BBa_K3482013.
35-iGEM MSP-Maastricht 2024. Optogenetic kill switch for marine biocontainment using toxin–antitoxin systems. Available from: MSP-Maastricht Kill Switch – iGEM 2024.
36-Marti T, iGEM UNILausanne 2020. Heat-repressible RNA-thermosensor F2 for temperature-sensitive control of toxin expression. iGEM Registry Part:BBa_K3482001. Available from: https://parts.igem.org/Part:BBa_K3482001.
37-Velez, M., & Arluison, V. (2025). Does the Hfq Protein Contribute to RNA Cargo Translocation into Bacterial Outer Membrane Vesicles? Pathogens, 14(4), 399. https://doi.org/10.3390/pathogens14040399.
38-DFabbiano, F., Notarangelo, M., Corsi, J., et al. (2020). RNA packaging into extracellular vesicles: An orchestra of RNA-binding proteins? Journal of Extracellular Vesicles, 9(1), 12043. Access the article.
39. Gilbert LA, Horlbeck MA, Adamson B, et al. CRISPR-mediated modular RNA-guided regulation of transcription in eukaryotes. Cell. 2013;154(2):442–451. Available from: https://doi.org/10.1016/j.cell.2013.06.044.
40. Setten RL, Rossi JJ, Han SP. The current state and future directions of RNAi-based therapeutics. Nat Rev Drug Discov. 2019;18(6):421–446. Available from: https://doi.org/10.1038/s41573-019-0017-4.
