Overview

Our project employs E. coli Nissle 1917 (ΔnlpI) as the chassis organism. EcN possesses intrinsic anti-inflammatory tropism, enabling automatic targeting of hemorrhoidal inflammatory sites and facilitating the action of anti-inflammatory therapeutics. Additionally, it helps maintain a healthy gut microbiota balance. Most importantly, as a well-established probiotic strain with an excellent human safety profile, EcN poses no biosafety risks and demonstrates high compatibility with the HemorrEaser system. Furthermore, the nlpI-deleted mutant exhibits enhanced production of outer membrane vesicles(OMVs), thereby improving drug delivery efficiency.[23,24]

The circuit of HemorrEaser comprises four main modules:

1. The Global Regulation module utilizes a negative feedback loop to regulate the expression of downstream genes, ensuring a relatively constant dosage of the therapeutic agents.[1,2,3]
2. The Anti-angiogenesis module includes a bioPROTAC (named VHH-VHL) delivered by outer membrane vesicles (OMV) to target and degrade HIF-1α, and an anti-VEGF nanobody equipped with a masking peptide, allowing it to be secreted into the bloodstream to reach hemorrhoidal tissues and exert its effect.
3. The Anti-inflammatory module is activated by reactive oxygen species (ROS) present in inflamed hemorrhoidal tissue. It expresses an engineered melittin peptide with anti-inflammatory functions.
4. The Safety module (comprising the suicide and "Medicine-Food Collaboration" components) consists of an inhibitory riboswitch and a toxin protein. In the absence of a specific ligand derived from particular healthy foods, the engineered bacteria initiate the expression of the downstream toxin protein CcdB, leading to self-elimination and cessation of therapy. Furthermore, to meet specific requirements during treatment, we incorporated an "expiry-date" circuit to delay bacterial lysis, allowing the engineered bacteria to remain active for an extended period without the need for frequent dosing.

HemorrEaser comprises two administration routes:

1. Oral Formulation: In the oral administration mode, we plan to process the engineered bacteria into lyophilized powder that retains viability, delivering it via enteric-coated capsules to the small intestine for targeted therapeutic action. This primarily contains the anti-angiogenesis module, the medicine-food collaboration module, and the delivery module. We propose using enteric-coated capsules to deliver the engineered E. coli Nissle 1917 (ΔnlpI) to the gut for colonization, establishing the basis for "Medicine-Food Collaboration". It is intended for the long-term treatment and suppression of hemorrhoids, while simultaneously assisting patients in developing healthier eating habits through the "medicine-food collaboration mechanism."
2. Topical Formulation: ​​In the topical administration mode, the engineered bacteria are formulated into an ointment along with TPP, which serves as a survival signal. As TPP becomes depleted, the suicide circuit is activated in the engineered bacteria. This mainly consists of the anti-inflammatory module and the suicide module, designed for emergency management during flare-ups of hemorrhoidal inflammation. We propose an ointment containing TPP and the engineered bacteria as the specific form for topical application, aiming to achieve a rapid response to inflammation and ensure swift elimination of the bacteria post-task.

Global Regulation Module

To address the potential risks of engineered bacterial overproliferation in vivo and associated issues such as therapeutic module expression fluctuations and toxicity, we designed a regulatory system based on a LuxR/TetR dual-factor cascade. This system employs a quorum-sensing mechanism to monitor bacterial density and utilizes a negative feedback loop to regulate the expression of downstream genes, thereby achieving global control.

In this system, the genes for the quorum-sensing components (luxI and luxR) are constitutively expressed, leading to a baseline production of the signaling molecule AHL. As the bacterial population grows, the increasing concentration of AHL activates the LuxR protein. This active LuxR complex then binds to its target promoter (P_lux), triggering the expression of the TetR repressor protein. Finally, TetR protein binds to its cognate promoter (P_tet), thereby repressing the transcription of the downstream genes it controls.[4,5,6,7]

Anti-angiogenesis Module

HemorrEaser implements a dual-targeting strategy to suppress pathological angiogenesis in hemorrhoids by simultaneously targeting intracellular and extracellular signaling pathways. The intracellular pathway is inhibited through a bioPROTAC (VHH-VHL) that specifically binds to and mediates the ubiquitination and degradation of HIF-1α under hypoxic conditions. For extracellular signaling blockade, an anti-VEGF nanobody engineered as a MMP3-activatable probody provides localized inhibition of VEGF signaling at the hemorrhoid site. This dual approach enables coordinated inhibition of both hypoxic and cytokine-driven angiogenesis, enhancing treatment efficacy while minimizing off-target effects through spatial control of therapeutic activity.[8,10,11]

In preliminary research and literature reviews, we found that the occurrence and progression of hemorrhoids are closely associated with local inflammatory responses and abnormal vascular proliferation. Among these, angiogenesis is primarily regulated by vascular endothelial growth factor (VEGF), and the upregulation of VEGF is closely linked to hypoxia-inducible factor-1α (HIF-1α). As a core transcription factor in cellular responses to hypoxia, HIF-1α activates the expression of a series of inflammation- and angiogenesis-related genes, thereby driving the formation and maintenance of the inflammatory microenvironment. Therefore, selectively degrading HIF-1α holds promise for blocking inflammatory pathways at their source, reducing the expression of downstream inflammatory factors, and consequently alleviating inflammatory responses at the site of hemorrhoidal lesions.

However, in the hypoxic tissue environment characteristic of hemorrhoids, the classical degradation pathway of HIF-1α is typically suppressed (Figure 3). Under normoxic conditions, prolyl hydroxylase domain proteins (PHDs) catalyze hydroxylation modifications at specific sites on HIF-1α, enabling its recognition by von Hippel-Lindau protein (pVHL) and recruitment of the Cullin2-Rbx1 ubiquitin ligase complex for subsequent ubiquitination and degradation. However, under hypoxic conditions, PHD activity is restricted, making it difficult to hydroxylate HIF-1α. Consequently, pVHL cannot recognize it, thereby blocking this efficient ubiquitin-mediated degradation pathway. This is also the fundamental reason why HIF-1α stably accumulates in hypoxic environments and drives inflammation and angiogenesis.[9]

Figure 3. Ubiquitination and degradation of HIF-1α under normoxic and hypoxic conditions. Figure is created by biogdp.com(https://biogdp.com).

Based on this, we aim to reactivate this suppressed degradation mechanism under hypoxic conditions. To achieve this, we designed a novel fusion protein strategy intended to directly mediate pVHL recognition of unhydroxylated HIF-1α, thereby bypassing PHD-mediated inhibition and restarting the ubiquitin-mediated degradation pathway of HIF-1α. Recent advancements in proteolysis-targeting chimeras (PROTACs) have opened new avenues for this approach. Inspired by this, we designed and engineered a bioPROTAC named VHH-VHL to achieve highly selective degradation of HIF-1α under hypoxic conditions. This strategy aims to block inflammatory pathways, inhibit angiogenesis, and alleviate inflammation associated with hemorrhoids.[12,13]To deliver the bioPROTAC via outer membrane vesicles (OMVs), we added an SRP signal peptide[30] to the N-terminus of VHH-VHL, facilitating its incorporation into OMVs.

Abnormal vascular proliferation is a recognized etiological factor in hemorrhoidal disease. To suppress vascular hyperplasia driven by vascular endothelial growth factor (VEGF) within hemorrhoidal tissue, we have engineered an anti-VEGF nanobody that can be expressed in vivo by a designed bacterial chassis.[14,15]

To preclude systemic diffusion of the therapeutic antibody and thereby minimize off-target safety liabilities, we appended a masking peptide to the C-terminus of the nanobody. The two moieties are connected by a short linker that contains a matrix-metalloproteinase-3 (MMP-3) cleavage site. Because MMP-3 is highly up-regulated in hemorrhoidal lesions, the linker is selectively proteolyzed at the disease site, releasing the masking peptide and unmasking the antigen-binding paratope. The liberated nanobody then competitively inhibits the interaction between VEGF-A and its cell-surface receptors, locally attenuating VEGF signaling.[11,16]

In addition, a P17 peptide was inserted at the N-terminus of the nanobody sequence to enhance solubility and in vivo stability. A PelB leading sequence was placed at the extreme N-terminus to direct Sec-dependent secretion of the recombinant protein from the engineered bacteria.[17]

Anti-inflammatory Module

To relieve the symptoms of hemorrhoids by treating the local inflammatory response and reducing swelling, pain, and exudation of hemorrhoidal tissue, we designed an anti-inflammatory module. This module works in synergy with the anti-angiogenesis module to enhance overall therapeutic efficacy.

We coupled the expression of our anti-inflammatory factor to an oxidative stress-responsive promoter, enabling engineered bacteria to sense the high-ROS environment of hemorrhoids and express the therapeutic peptide only under these conditions.

For this purpose, we selected the SoxR/P_SoxS oxidative stress response system. SoxR is a transcription factor containing a [2Fe-2S] cluster that collaborates with the P_SoxS promoter sequence. Under high ROS levels, oxidized SoxR binds between the -10 and -35 regions of the P_SoxS promoter, altering its conformation and facilitating the recruitment of RNA polymerase and additional transcription factors, thereby activating downstream gene transcription.[18,19]

As the therapeutic peptide, we chose di-melittin, designed by NKU-China,2024 and added a pelB signal peptide to enable its secretion into the periplasmic space. While melittin has strong anti-inflammatory potential, its high cytotoxicity limits practical application. Di-melittin overcomes this challenge by linking two melittin monomers via a flexible linker to form a hairpin structure, significantly reducing cytotoxicity and making prokaryotic expression feasible.[20]

By combining the SoxR/P_SoxS sensing system with di-melittin, our anti-inflammatory module enables ROS-responsive expression of the therapeutic peptide at the hemorrhoid site, thereby alleviating inflammation and related symptoms.

Escherichia coli outer membrane vesicles (OMV) delivery system

As a well-studied delivery platform[22], E. coli outer membrane vesicles (OMV) can be modified to construct the intended anti-angiogenesis module delivery system.[29] To achieve precise targeted delivery of OMV to the Hemorrhoids site, we selected the OmpA-CAR fusion protein as the targeting element.

The positively charged region at the N-terminus of the OmpA signal peptide achieves initial anchoring through electrostatic interaction with the negative charge of the inner membrane, followed by the insertion of the middle hydrophobic α-helix into the lipid bilayer to form a "membrane insertion tag". This conformational tag is recognized by SRP/SecB and targeted to the SecYEG translocation channel, thereby precisely anchoring the outer membrane protein precursor to the inner membrane and initiating its transmembrane transport process.[25]

In the hemorrhoid microenvironment, ischemia, hypertonicity, and inflammatory mediators stimulate vascular endothelial cells and basal Keratinocytes of the anal cushion to significantly upregulate the expression of heparan sulfate proteoglycan syndecan-4 (SDC4). The literature indicates that scar homing peptide CAR can recognize the highly expressed SDC4 at the wound site, achieving selective binding and endocytosis by binding to the glycosaminoglycan chain of SDC4, thereby enriching at the re-epithelialization front and realizing the wound homing effect. Simultaneously, the CAR-SDC4 complex triggers the recruitment of CYTH2 guanine exchange factor, activates small GTPase ARF6, drives the cycling of α5β1 integrin and actin remodeling, significantly enhances the migration speed of keratinocytes, and ultimately, systemic administration accelerates wound re-epithelialization and healing.[26,27]

Schematic of E. coli outer membrane vesicles (OMV) engineering signal peptide was fused with the CAR as a single Fusion peptide (OmpA-CAR), whose N-terminal positively charged-hydrophobic region can drive Outer Membrane Vesicle (OMV) membrane anchoring, while the C-terminal CAR region is exposed on the vesicle surface; achieving the functional purpose of "membrane anchoring-specific recognition", which can effectively mediate the targeted transport of OMV to the Hemorrhoids site, laying the foundation for subsequent anti-vascular module delivery.

Suicide and Medicine-Food Collaboration Module

In order to avoid potential biosafety hazards from engineered-bacteria leakage inside the human body and into the external environment, we designed a suicide module that couples dietary metabolism to bacterial survival and clearance, forming a "Medicine-Food Collaboration" framework.

A riboswitch is a regulatory RNA element that can directly bind small metabolites and regulate downstream gene expression. For example, the thiM riboswitch specifically recognizes thiamine pyrophosphate(TPP). Upon ligand binding, it undergoes a conformational change that represses downstream gene expression, either by terminating transcription or by blocking translation initiation, depending on the regulatory context. This unique ligand-RNA interaction makes riboswitches powerful tools for connecting dietary metabolites to bacterial gene circuits.[27]

For functional validation, we first employed a well-characterized thiM riboswitch to regulate the expression of the toxin gene ccdB. When TPP is abundant, the riboswitch suppresses CcdB production, allowing engineered bacteria to survive; when TPP is absent, repression is relieved, ccdB is expressed, and bacterial death is induced.

Building on this proof-of-concept, we envisioned hippuric acid as a more clinically relevant dietary signal. Hippuric acid is a downstream metabolite of anthocyanins, enriched in foods such as Oryza sativa L. indica (black rice, “forbidden rice”). When hippuric acid is present, a corresponding riboswitch could suppress ccdB expression and maintain bacterial survival; once dietary intake stops and hippuric acid is cleared, the riboswitch derepresses, triggering ccdB expression and bacterial suicide. This design establishes a direct diet-bacteria interface for controlled containment.

In topical hemorrhoid treatment scenarios, this system can be directly applied for safe regulation. Once TPP is absent, engineered bacteria immediately trigger suicide, ensuring rapid clearance and safety. For oral administration, considering clinical needs, we integrated a genetic delay module, the expiry-date circuit[28]. After a single intake, even as intestinal TPP is metabolized and depleted, engineered bacteria can persist for about 40 hours to complete anti-inflammatory or repair tasks before self-destruction, thus avoiding frequent dosing. This design provides a safe and controllable solution for the clinical application of live biotherapeutics.

Since hippuric acid is a broad-spectrum metabolite that varies with diet and lacks tissue specificity, we expanded the concept to more candidate small molecules. By screening diet-derived metabolites using defined criteria—such as food-specific origin, gut persistence (>12 h), low systemic absorption, safety, and detectability by E. coli Nissle 1917, we identified multiple options including curcumin (turmeric), phloridzin/phloretin (apples), sulforaphane (broccoli), and raffinose family oligosaccharides (legumes). This strategy enables personalized adaptation: patients can flexibly choose foods aligned with their dietary habits while maintaining reliable suicide timing. Such diversification improves robustness, reduces off-target risks, and lays the foundation for a standardized "Medicine-Food Collaboration" framework.