Engineering Success

Cycle 1 – The Big Operon (Completed)

Design
At the very beginning, we envisioned constructing a large operon that would encode three different chitin-degrading enzymes: an endochitinase, an exochitinase, and an N-acetylglucosaminidase. Together, these would provide a “one-shot” solution for complete chitin hydrolysis.

Crossroad
As we planned this, it became clear that the operon would be too large to clone and express efficiently. Large constructs are unstable and difficult to maintain in plasmid backbones. In addition, literature suggested that many hosts already have endogenous N-acetylglucosaminidase activity, making one part of our design redundant.

Redesign
We shifted to a modular strategy. Instead of an oversized operon, we designed a fusion enzyme that combined one endochitinase and one exochitinase joined by a flexible linker.

Build
Fusion constructs were designed in silico and made compatible with E. coli and later Bacillus subtilis.

Test
Bioinformatic evaluations confirmed that the fusion constructs were stable and synthesizable, unlike the oversized operon.

Learn
We learned that more is not always better. Modularity and efficiency are critical for feasible engineering. This pivot defined the genetic backbone of Chitinator et al.

Cycle 2 – Lack of Secretion in pET-28a (Completed)

Design
We cloned our fusion enzyme into pET-28a(+) for expression in E. coli BL21. This vector has a strong T7 promoter and is widely used for protein overexpression.

Crossroad
Induction with IPTG led to strong expression, but the protein stayed inside the cells because pET-28a lacks a secretion signal. This blocked us from testing activity in the culture medium.

Redesign
We adapted our strategy:

• Use cell lysis followed by disk diffusion and zymogram assays to test activity.
• At the same time, prepare to move into alternative pET vectors that include secretion signals, such as pET-26b, which carry the pelB leader sequence for periplasmic export.

Build
Cultures were induced, lysed, and crude extracts were applied to functional assays. Cloning plans into secretion-capable vectors were also drafted.

Test
SDS-PAGE confirmed intracellular expression. Functional assays (disk diffusion and zymograms) gave indirect activity evidence.

Learn
We understood that secretion is essential for scalable testing. pET-28a was useful for initial expression but limited for functional screening, so we planned to switch to secretion-compatible vectors.

Cycle 3 – Expression Optimization (Completed)

Design
During expression trials in E. coli BL21, we encountered inconsistent solubility and yield depending on IPTG levels and sonication efficiency.

Crossroad
We had to decide whether to redesign the construct (e.g., adding solubility tags) or optimize culture and induction conditions first.

Redesign
We opted to fine-tune conditions:

• Test a gradient of IPTG concentrations.
• Vary induction times.
• Adjust sonication cycles to improve protein release without denaturation.

Build
Multiple cultures were induced with different IPTG doses and harvested at different time points. Sonication cycles were applied systematically.

Test
SDS-PAGE was performed across conditions to compare protein expression and solubility.

Learn
We identified more favorable conditions for protein expression and recovery, showing that optimization can solve issues without immediate genetic redesign.

Cycle 4 – PCR Optimization (Completed)

Design
To clone the insert into secretion-compatible plasmids (pWB26, pHT43), we needed to amplify it. However, our primers had large melting temperature (Tm) differences.

Crossroad
We had to decide whether to redesign primers (losing time and resources) or test amplification through optimization.

Redesign
We chose to run a gradient PCR, spanning a range of annealing temperatures, to empirically find conditions where both primers could work.

Build
Gradient PCR reactions were set up using the existing primer set.

Test
Clear bands of the expected size were observed at all tested temperatures.

Learn
Gradient PCR rescued the cloning process without primer redesign. This confirmed that the insert was robustly amplifiable, allowing us to move forward confidently.

Cycle 5 – Multi-Stage Transition to Bacillus subtilis with pHT43 and Beyond (Planned)

Design
While E. coli is excellent for rapid prototyping, it does not fully reflect our final application needs. To strengthen our screening pipeline, we plan a stepwise transition:

• Test secretion in E. coli using a vector such as pET-26b(+) which carries a secretion signal or pET-28a(+) (adding a secretion signal), allowing us to better screen activity directly in the culture medium.
• Transfer the construct into Bacillus subtilis 168 using pHT43, a secretion-competent plasmid with the amyQ signal peptide, for more realistic activity assays in the final chassis.
• In a later phase, adapt the construct into a food-grade Bacillus vector without antibiotic resistance markers, paving the way for agricultural use.

Crossroad
The choice was whether to jump directly into Bacillus expression, or to design a layered pipeline: first validate secretion in E. coli, then move into Bacillus subtilis, and finally prepare an antibiotic-free vector for safe agricultural deployment.

Redesign
We opted for this multi-step approach, since it combines the speed of E. coli with the realism of Bacillus subtilis, while ensuring long-term biosafety.

Build
We will prepare PCR products and cloning strategies for each stage:

• Subcloning into pET-26b(+) or again in the initial vector with a secretion signal for secretion screening in E. coli.
• Cloning into pHT43 for secretion assays in Bacillus subtilis 168.
• Future design of an antibiotic-free Bacillus plasmid for real-world application.

Test
Each stage will involve activity assays:

• E. coli: secretion checked via supernatant assays (disk diffusion, zymograms).
• B. subtilis/pHT43: zone-of-clearance on colloidal chitin plates.
• B. subtilis antibiotic-free: performance testing under application-like conditions.

Learn
We expect this staged strategy to demonstrate that secretion is achievable in both chassis, while preparing a safe, antibiotic-free vector for agricultural deployment. This stepwise design balances scientific rigor, biosafety, and practical implementation.

Cycle 6 – Sprayable BioActivator for Garlic

Design
Our implementation goal is to develop a sprayable BioActivator tailored for garlic cultivation. Farmers already apply chemical fertilization to meet nutrient needs, but garlic requires additional spray treatments throughout its growth due to fungal vulnerability. Our BioActivator is designed not to replace fertilizers, but to complement them by acting as a biological spray with both growth-promoting and antifungal properties.

Crossroad
Farmers rely on repeated chemical sprays to protect garlic crops, but these increase costs, leave residues, and may harm soil health. We needed a product that integrates smoothly into their existing spraying schedules after chemical fertilization, offering a safer and more sustainable option.

Redesign
We plan to formulate a cell-free chitin hydrolysate (oligosaccharides and GlcNAc) that:

• Is sprayable with standard equipment (low viscosity, nozzle-safe).
• Stimulates garlic plant defense pathways, reducing fungal infections.
• Enhances soil vitality and crop resilience without replacing fertilizer inputs.
• Fits into the garlic crop cycle as a post-fertilization spray, within the same windows farmers already use for fungicide applications.

Build
Pilot bioreactor runs will generate BioActivator, followed by downstream steps (sterile filtration, buffering to pH 6–7). Formulations will be tested for sprayability and tank-mix compatibility with conventional sprays.

Test
Planned assays include:

• Analytical: GlcNAc quantification, oligosaccharide DP profiling, viscosity, sterility.
• Application: nozzle compatibility tests, leaf coverage, and mixing trials with chemical sprays.
• Biological: greenhouse garlic trials to monitor growth and fungal resistance (Fusarium proliferatum, Sclerotium cepivorum) under BioActivator treatment.

Learn
We expect to validate the BioActivator as a dual-purpose biological spray: boosting garlic growth and providing antifungal protection. By fitting into existing farming practices after chemical fertilization, it will increase farmer acceptance and demonstrate a clear real-world pathway for sustainable agriculture.

HP Cycle 1 – Biosafety & Farmer Concerns

Design
At the beginning, our plan was to engineer a living biofertilizer, releasing GMOs directly into the soil to promote crop growth.

Crossroad
Feedback from farmers and biosafety experts revealed strong resistance: “We do not want genetically modified organisms in our soil.” Concerns about ecological disruption and regulation made this approach socially and environmentally risky.

Redesign
We shifted toward a bioreactor-based approach, where engineered microbes are contained under controlled lab conditions and only their safe products (bioactivators) are applied to the soil.

Build
This decision shaped our Wet Lab workflow, focusing on maximizing enzyme expression and activity in vitro rather than engineering survival traits in soil microbes.

Test
Bioactivators could now be safely tested in agricultural scenarios without releasing GMOs, increasing farmer acceptance.

Learn
Human Practices taught us that biosafety and social trust are non-negotiable, pushing us toward a safer, more acceptable design.

HP Cycle 2 – Circular Bioeconomy Framing

Design
Initially, we presented our project mainly as a biotechnological tool for chitin degradation.

Crossroad
Discussions with sustainability experts highlighted that the project could be undervalued unless placed within a larger narrative of circular bioeconomy, where waste is recycled into value.

Redesign
We reframed Chitinator et al. as part of the circular bioeconomy, turning seafood and agricultural chitin waste into nutrient-rich bioactivators that enhance soil health and reduce reliance on synthetic fertilizers.

Build
All communication materials, presentations, and even experimental design began emphasizing the “waste-to-value” principle and environmental respect.

Test
When shared with stakeholders, this framing resonated strongly, making the project appear not only scientifically innovative but also economically and ecologically relevant.

Learn
HP guided us to understand that storytelling and framing are just as important as technical design in making biotechnology meaningful and impactful.

HP Cycle 3 – Choice of Garlic Cultivation

Design
Our system could, in theory, be applied to many crops.

Crossroad
Conversations with Greek farmers revealed that garlic cultivation, particularly Greek white garlic, is economically important, culturally symbolic, and highly vulnerable to fungal infections such as Fusarium proliferatum and Sclerotium cepivorum.

Redesign
We selected garlic cultivation as our case study, focusing our bioactivators on this emblematic crop. This made our project locally relevant and globally recognizable.

Build
Wet Lab experiments and communication materials highlighted garlic as the first target application, linking bioactivators to plant defense and soil vitality.

Test
Feedback from farmers and experts confirmed that this choice made the project more convincing, showing a clear path from lab innovation to agricultural benefit.

Learn
HP directed us to root our innovation in a real-world agricultural challenge, ensuring that our design responds to actual farmer needs.

HP Cycle 4 – Farmers’ Needs for Nutrient Units

Design
We initially thought of presenting bioactivators mainly as sustainable alternatives to synthetic fertilizers.

Crossroad
Through discussions with agronomists and farmers, we learned that growers calculate their inputs in nitrogen units and expect every fertilizer or activator to clearly meet plant nutritional needs. Without quantifiable values, acceptance would be limited.

Redesign
We adapted our concept: instead of positioning bioactivators as a “replacement” for fertilizers, we framed them as complementary inputs that improve soil health and crop resilience while working alongside nitrogen-based fertilization.

Build
Communication materials and lab plans began to emphasize measurable impacts, e.g., not only degradation of chitin waste but also its contribution of carbon, nitrogen, and phosphorus back to the soil.

Test
Feedback confirmed that this framing made bioactivators more understandable and appealing to farmers, who now saw them as practical rather than abstract tools.

Learn
HP reminded us that real-world adoption depends on farmer language and expectations; to gain trust, we must present bioactivators in terms that align with agricultural practice, like nutrient units.

HP Cycle 5 – Complementary Use as Biological Sprays

Design
At first, bioactivators were considered as a possible substitute for chemical fertilizers.

Crossroad
Feedback from farmers and agronomists showed that this framing did not match reality: garlic cultivation already depends on chemical fertilization for nutrients, but the real problem is fungal vulnerability, which requires repeated chemical sprays. Farmers were interested in biological alternatives for protection, not in abandoning fertilizers altogether.

Redesign
We reframed bioactivators as biological sprays/boosters: applied to garlic crops not to replace fertilizers, but to complement them by strengthening plants against fungal infections and improving soil vitality.

Build
Our Wet Lab strategy and communication materials began to highlight garlic as a case study where bioactivators are applied after or alongside chemical fertilization, in order to reduce dependency on fungicides and increase resilience.

Test
Farmers responded positively to this repositioning, seeing bioactivators as a practical, biological solution that integrates with their current practices rather than replacing them.

Learn
HP reminded us that innovation succeeds when it fits into existing agricultural workflows. By presenting bioactivators as biological sprays, we aligned with farmer needs and increased the real-world applicability of our project.

Cycle 1 - Choosing the right enzyme

Design
We aimed to identify the optimal enzyme set for complete chitin degradation. In front of us stood a vast number of candidate chitinases, distributed across multiple glycoside hydrolase (GH) families. The key question was: which enzymes should we select?

Crossroad
The sheer abundance of options created uncertainty. Random selection risked inefficient or unsuitable enzymes for our project. We needed an objective, bioinformatics-based strategy to guide the decision.

Redesign
We turned to publicly available data from the European Nucleotide Archive (ENA). This allowed us to gather large collections of chitinase sequences and apply two complementary analyses:

• Ecological mapping of the samples (hosts, environments, clinical data).
• Phylogenetic reconstruction, to reveal evolutionary relationships among enzymes and identify patterns of convergence or specialization.

This dual approach provided a more reliable framework for judging which enzymes were truly suitable.

Build
By integrating ecological and phylogenetic insights, we concluded that the GH18 family represented the most reliable choice: it exhibits broad ecological distribution, strong diversification, and a history of use in biotechnological applications. We therefore selected specific GH18 enzymes (BBa_K4349000 – endochitinase, CfcI – exochitinase).

Test
This decision was validated by the convergence of both data streams: GH18 enzymes emerged as generalist, ecologically stable, and experimentally verified across diverse environments.

Learn
From this process, we learned that enzyme selection is not merely a matter of literature mining or chance, but must be grounded in the integration of ecological and phylogenetic data. This step marked a turning point, shaping our strategy for linking dry lab with wet lab.

Cycle 2 – From Proteins to DNA

Design
We initially planned to work with protein sequences of chitinases. The rationale was that protein information reflects function more directly and would provide high-quality phylogenetic trees, free from the noise of synonymous mutations.

Crossroad
In practice, significant obstacles arose:

• The number of curated protein sequences was limited compared to DNA.
• Annotations were heterogeneous, with incomplete metadata and unclear GH family assignments.
• Alignments proved unstable, especially across distant GH families, leading to questionable phylogenetic reconstructions.

This impasse forced us to reconsider our initial strategy.

Redesign
Instead of restricting ourselves to the few available protein sequences, we turned to DNA sequences from the ENA, which offered:

• Greater diversity and a larger number of samples.
• Better coverage across ecosystems and hosts.
• Integration with ecological metadata (country, isolation source, host).

This allowed us to build comprehensive datasets for each GH family, suitable for more reliable phylogenetic reconstruction.

Build
We developed pipelines to parse raw ENA records, extracting DNA sequences along with their associated ecological information. This approach enabled large-scale analyses without the data gaps we would have faced with proteins.

Test
Phylogenetic reconstructions based on DNA datasets produced much clearer trees, with well-supported clades and the ability to link evolutionary patterns to ecological contexts.

Learn
We learned that the “ideal” dataset is not always the most theoretically clean (proteins), but rather the one that is more complete and integrable with metadata. Shifting to DNA provided richer and reliable results, laying the foundation for connecting phylogenetics with ecology in our model.

Cycle 1 – The Linker Dilemma

Design
Initially, multiple linker strategies were envisioned to connect the endochitinase and exochitinase domains into a single, functional composite enzyme. These included both flexible (G₆S)₃ linkers and rigid motifs such as EAAAK, KEKE, Proline-rich, and ELP sequences. The goal was to determine which configuration would allow independent folding of both enzymes while preserving synergistic catalytic activity.

Crossroad
Early modeling revealed that rigid linkers, though structurally stable, restricted the natural movement required for cooperative activity between the enzymes. The Proline-rich and ELP linkers, in particular, caused excessive rigidity or instability. This forced a reconsideration of linker flexibility versus structural stability.

Redesign
The design focus shifted toward flexible linkers that could preserve both enzymatic integrity and motion. The (G₆S)₃ sequence was selected for further analysis, as its glycine and serine composition offered both flexibility and solubility.

Build
All linker variants were modeled in silico using Swiss-Model and ColabFold (AlphaFold2), followed by visualization in PyMOL and motif validation with ScanProsite. Each structure was evaluated for steric clashes, folding quality, and domain orientation.

Test
Validation metrics (Ramachandran plots, MolProbity scores, pLDDT) demonstrated that the (G₆S)₃ construct exhibited minimal steric hindrance and high stereochemical quality. The catalytic motifs remained accessible and unaltered.

Learn
This phase confirmed that flexibility is essential for cooperative chitin degradation. The (G₆S)₃ linker was finalized as the optimal design, offering the best balance between domain mobility, solubility, and folding precision.

Cycle 2 – Structural Validation and Comparative Analysis

Design
To ensure the predicted fusion protein was structurally feasible, detailed validation was performed combining homology modeling (Swiss-Model) and deep learning (AlphaFold2) predictions.

Crossroad
Initial models highlighted discrepancies in the linker regions of rigid constructs. Drops in IDDT confidence and poor local folding indicated strain. These limitations prompted a more rigorous validation of the flexible linker.

Redesign
Models were re-run and refined using hybrid pipelines, integrating multiple templates and quality metrics. Comparative visualization confirmed the flexible linker allowed optimal spatial proximity between domains.

Build
High-quality 3D models were constructed, colored by domain, and assessed for active-site accessibility. Clashscores and geometric deviations were recorded.

Test
Quantitative validation showed >90% residues in favored Ramachandran regions, MolProbity score ~2, and Clashscore at the 91st percentile, confirming realistic atomic geometry.

Learn
Independent folding with intact catalytic sites validated the core rationale of the fusion design. The flexible linker not only preserved activity but also improved predicted stability compared to rigid designs.

Cycle 3 – From Modeling to Implementation

Design
The final structural validation informed the future wet-lab phase by confirming that the construct could be expressed without steric or folding issues, paving the way for secretion optimization in Bacillus subtilis.

Crossroad
While the computational results were strong, the next bottleneck identified was the absence of a secretion signal for extracellular enzyme release, crucial for future implementation in the bio-activator system.

Redesign
A plan was proposed to integrate an appropriate secretion signal peptide into the construct, ensuring extracellular chitin degradation in B. subtilis.

Build
Future iterations will use the validated (G₆S)₃ fusion backbone, incorporating secretion and promoter elements within expression vectors optimized for Bacillus.

Test
Functional assays (chitin plate degradation, enzymatic kinetics) will evaluate real-world catalytic efficiency and confirm computational predictions.

Learn
The structural validation served as the foundation for the next experimental phase, transforming computational insights into tangible, testable synthetic biology applications.