Overview

Our project aimed to develop a novel, enzyme-based eyedrop for the early prevention of cataracts by engineering Manganese Superoxide Dismutase (Mn-SOD) as the core therapeutic agent. A significant challenge in this endeavor is the tendency of recombinant Mn-SOD to misfold and form insoluble inclusion bodies in prokaryotic production systems, alongside the need to rationally design superior variants and precisely predict their efficacy within the complex ocular environment. In order to scientifically design and evaluate the efficacy of Mn-SOD eye drops in practical application, we established an integrated engineering workflow that combines experimental optimization of protein solubility, computational rational design of enzyme variants, and multi-scale mathematical modeling of spatiotemporal drug efficacy. This comprehensive strategy not only ensured the successful production of active Mn-SOD but also provides the iGEM community with a reusable and scalable toolkit for developing and evaluating protein-based therapeutics, significantly accelerating the transition from conceptual design to practical application.

1 Optimizing Soluble Expression of Mn-SOD

The relevant parts have been registered (Parts ID: BBa_K4907023), and the represented experimental data are uploaded to the component library simultaneously.

A critical bottleneck in the bioproduction of therapeutic enzymes like Manganese Superoxide Dismutase (Mn-SOD) is their frequent misfolding and aggregation into insoluble inclusion bodies within prokaryotic expression systems. Our iGEM team systematically addressed this universal challenge through an integrated engineering approach, establishing a robust and reproducible method for achieving high-yield soluble expression of active Mn-SOD. This contribution provides future teams with a validated, dual-strategy toolkit to overcome protein solubility barriers, accelerating the development of enzymatic therapeutics.

1.1 Implementing a Chaperone Co-Expression System to Simulate a Eukaryotic Folding Environment

Recognizing that the reducing cytoplasm of E. colilacks the complex chaperone network required for the correct folding of many eukaryotic-derived proteins, we employed a rational auxillary strategy. We selected and co-expressed three distinct chaperone plasmid systems (pGro7, pKJE7, and pTf16) with our pET28a-Mn-SOD construct in E. coli BL21(DE3). This approach aimed to provide crucial folding assistance, mitigate aggregation, and increase the flux of properly folded, soluble protein. The success of this molecular chaperone strategy offers a powerful and generalizable first-line solution for iGEM teams facing similar protein solubility issues, effectively simulating a more hospitable folding environment within a prokaryotic host.

1.2 Optimizing Fermentation Conditions to Modulate the Protein Folding Kinetics

Beyond molecular intervention, we implemented a complementary physical strategy: fermentation temperature optimization. We hypothesized that reducing the rate of protein synthesis would provide nascent polypeptide chains more time to fold correctly before encountering other nascent chains, thereby minimizing aggregation. We tested a gradient of induction temperatures (37°C, 30°C, 25°C, and 16°C) to identify the optimal condition that balances protein yield with folding efficiency. This simple yet highly effective process parameter optimization is a universally applicable and cost-efficient method for any team seeking to improve the solubility of recombinant proteins, requiring no genetic modification of the host or product.

The ultimate validation of our solubility strategies was the successful production of a purified, active enzyme. The soluble Mn-SOD produced under optimized conditions (16°C induction) was purified and confirmed to be functionally active, with a specific activity of 552.131 U/mg. This critical outcome confirms that our engineering strategies not only solved the aggregation problem but also preserved the biological functionality of the enzyme, making it suitable for downstream applications as the core active ingredient in our antioxidant eye drops. For detailed results, please go to engineering success page.

In summary, our contribution transcends the specific expression of Mn-SOD. We have established and rigorously tested a standardized, two-pronged workflow for overcoming a central challenge in synthetic biology: 1) Molecular Assistance: Co-expression of chaperone systems (pGro7, pKJE7, and pTf16); and 2) Process Optimization: Lowering the induction temperature (16°C). This integrated approach provides a clear experimental roadmap for future iGEM teams. Teams can first implement low-temperature induction for a quick win, and if further enhancement is needed, proceed to chaperone co-expression. By sharing this validated strategy and its quantitative outcomes, we empower the entire community to more reliably produce soluble, functional proteins, thereby de-risking projects and accelerating the translation of synthetic biology designs into real-world applications.

2 Rational design of Mn-SOD based on Transformer deep learning

Machine learning (ML) and deep learning (DL) are transformative approaches in protein engineering, enabling the prediction of mutation effects from sequence, structural, and functional datasets. Unlike traditional methods that rely on manual feature engineering, ML/DL automatically extracts relevant features from raw sequences—avoiding bias and missed information—and models complex, nonlinear, high-dimensional sequence–function relationships. This scalability allows for rapid prediction across entire sequence spaces, far outperforming physics-based simulations in speed and efficiency.

In our project, we focused on Transformer models, a type of DL architecture that uses self-attention mechanisms to capture long-range dependencies between amino acids. This was critical for identifying interactions between distant residues (e.g., catalytic sites and distal loops) that influence protein function—relationships that traditional sequential models (like RNNs) might miss. By training Transformer models on large-scale Mn-SOD datasets, we could detect patterns associated with thermostability and catalytic activity. Our Transformer model has successfully identified variants, which were promising candidate for further experimental validation. The integration of ML/DL into our pipeline not only accelerated the discovery of beneficial mutations but also enhanced the precision of predicting their impact on SOD performance (Figure1).

Figure1. Transformer model architecture.

3 Mathematical Model

The multi-scale mathematical model framework constructed by our project provides a replicable and scalable rational design paradigm for the iGEM team in the future, and its core value is to transform complex biological systems into computable and predictable engineering problems. The framework systematically solves the complete prediction process from molecular mechanism to in vivo drug efficacy through the hierarchical strategy of microcatalytic kinetics, macroscopic 0D homogeneous system and 1D spatial diffusion. In the future, any team dedicated to developing enzyme therapies (such as antibacterial, metabolic disease treatment, antioxidant) can directly reuse this method, and only need to replace the core reaction equation to realize virtual screening, dose optimization, and spatiotemporal efficacy simulation of their specific systems, greatly reducing R&D costs and trial and error risks, and transforming the traditional trial and error process that relies on empirical exploration into an efficient computation-driven rational design. In addition, the framework significantly enhances the scientific depth and interdisciplinary integration capabilities of iGEM projects. It not only enables the team to give a quantitative conclusion of "Y% change in effector concentration at the target site at X dose", but more importantly, by introducing a 1D spherical coordinate diffusion model, it realizes the accurate dynamic simulation of the 3D absorption and distribution process of drugs in the eyeball, capturing the geometric dilution effect and spatial heterogeneity that cannot be described by traditional homogeneous models. Throughout the model development process, we paid particular attention to balancing biological rationality, computational feasibility, and practical applicability to ensure that the model outputs are both scientifically reliable and readily implementable. We hope that this model framework will provide a powerful template for all projects involving drug delivery and spatial modeling. For detailed information, please go to model page.

3.1Microscopic Catalytic Kinetic Model

Microscopic Catalytic Kinetic Model is based on the Ping-Pong Mechanism in which manganese superoxide dismutase (Mn-SOD) catalyzes the disproportionation reaction of superoxide anion (O2⁻). This mechanism accurately describes the cyclic catalytic process of manganese ions in the enzemic active center between the oxidation state (Mn³⁺-SOD, Eox) and the reducing state (Mn²⁺-SOD, Ered), with each step and a half of the reaction consuming a superoxide molecule and generating the corresponding enzyme-substrate intermediate complex (C1 and C2). The two-and-a-half-step reaction is shown in Figure 2 below.

Figure2. Chemical Reaction Mechanism of Mn-SOD Scavenging ROS.

The core purpose of this microscopic model is to build an accurate "biochemical reaction engine". It quantitatively describes the catalytic efficiency of Mn-SOD from the lowest molecular mechanism, providing the most fundamental and reliable chemical reaction terms. Without this microscopic engine, any macroscopic simulation would lose its foundation in biochemical authenticity.

3.2 Macroscopic Zero-Dimensional (0D) Homogeneous System Dynamics Model

The model simplifies the aqueous humor environment in the eyeball into an ideal, well-mixed continuous stirred kettle reactor (CSTR), and its core assumption is spatial uniformity, that is, the concentration of drug and ROS in the entire aqueous humor is uniform at any time. The model describes the changes of drug concentration and ROS concentration over time by establishing a coupled ordinary differential equation (ODE) system (Figure 3). Among them, the change of drug concentration is only determined by the purge process (such as degradation and loss), following the exponential decay law; while the change of ROS concentration is controlled by the dynamic balance of its endogenous generation rate and enzyme-catalytic clearance rate.

Figure3.Ordinary differential equation (ODE) system in Zero-Dimensional (0D) Homogeneous System Dynamics Model.

By numerically solving the above ODE system, we and other iGEMers can explore the following key questions:

1. Dose-Response Relationship Analysis: By fixing the ROS generation rate kgen and drug clearance rate kclear, and varying the initial dose E0,dose, we can simulate how low ROS levels can be suppressed (efficacy intensity) and for how long they can be maintained at a low level (duration of action). This helps determine the minimum effective concentration of the eyedrops.

2. Dosing Frequency Assessment: Simulations can determine the time it takes for ROS levels to rebound to a certain threshold. This time point can serve as a theoretical basis for recommending the dosing interval. We can further simulate periodic dosing to observe the long-term steady-state fluctuations of ROS levels.

3. Capability to Counter Different Stress Levels: By varying the value of kgen, we can simulate different levels of oxidative stress (e.g., normal aging vs. diabetes/UV exposure). The model can predict whether the current dosage remains effective under severe stress or if an increase in dose/frequency is required.

This model is designed to quickly assess the overall pharmacodynamic profile of dosing regimens and serves as a powerful tool for conducting efficient parameter scanning and preliminary screening. Its simple and fast calculations make it ideal for high-throughput virtual screening in the early stages of a project. However, this also makes it impossible to describe drug delivery processes and spatial concentration gradients, thus driving the development of high-dimensional models.

3.3 Macroscopic One-Dimensional (1D) Reaction-Diffusion Model

The target that drugs for preventing cataracts need to act on is the lens. Traditional models cannot answer the core questions of "how much drug will reach the lens" and "how long will it take?" Our spatial model directly and dynamically simulates the penetration process of drugs in the anterior chamber by setting physiologically reasonable boundary conditions (such as the delivery boundary at the cornea and the fluxless boundary at the lens).

This model introduces spatial dimensions and diffusion effects. The model describes this process through the reaction-diffusion partial differential equation (PDE), in which temporal evolution and spatial distribution are coupled (Figure 4). Tthe spherical coordinate system (Figure 5) can more realistically capture the geometric dilution effect caused by the spherical geometry of the eyeball due to the introduction of the (2/r) (∂C/∂r) term, that is, the natural decrease in concentration due to the increase in the area of the sphere through which the material passes when it diffuses outward.

Figure4. The reaction-diffusion partial differential equation.

Figure5. The spherical coordinate system equation.

The core purpose of this model is to make a leap from "temporal dynamic" prediction to "spatiotemporal dynamic" prediction to obtain simulation results with higher biological realism and prediction accuracy. It can accurately simulate the spatiotemporal process of how drugs penetrate aqueous humor and finally reach the target (lens) after dripping from the cornea, predicting its actual exposure concentration and history on the target surface, so as to provide the final and most reliable decision-making basis for designing precise dosing regimens (dosage and frequency) that can truly ensure the effectiveness of the target. At the same time, its visualizations, such as heat maps, provide an excellent representation of the spatial dynamics of drug penetration and ROS clearance.

4 Summary

In summary, our contributions extend beyond the successful engineering of a novel Mn-SOD variant for cataract prevention. We have established and rigorously validated a standardized, multi-pronged workflow that provides solutions to central challenges in synthetic biology:

1. In protein production, we deliver a validated dual-strategy toolkit—chaperone co-expression and low-temperature induction—to overcome the universal problem of insoluble inclusion body formation, enabling future teams to reliably obtain soluble, functional enzymes.

2. In protein design, we used Transformer-based deep learning to navigate the vast mutational landscape rationally, demonstrating a powerful paradigm for enhancing enzyme activity and stability without relying on high-throughput screening.

3. In efficacy prediction, we developed a multi-scale mathematical modeling framework that bridges from molecular catalysis to spatiotemporal drug distribution in vivo. This framework allows for the quantitative prediction of therapeutic outcomes and optimal dosing regimens, transforming a traditionally empirical process into a computation-driven, rational design.