Bannière de la page

R e s u l t s

Introduction

This section presents the experimental results of our strategies to engineer E. coli for controlled melanin production and biocontainment. We detail the successful construction and functional validation of the pAIDA-tyr1 plasmid for extracellular tyrosinase activity, leading to visible melanin synthesis. Furthermore, we report on the challenges faced in optimizing the tyrosine biosynthetic pathway through aroF mutagenesis and in establishing a functional ΔmurI auxotrophic system for biocontainment, alongside the successful creation of a complementary pMurI plasmid. Finally, we demonstrate the efficacy of alginate hydrogel encapsulation as a physical barrier, confirming its ability to confine cells while maintaining their viability and metabolic activity. These results collectively outline the construction and validation of a foundational platform for a safe and functional microbial system.

1/ Construction of the Melanin Production System pAIDA-tyr

Plasmid Design

Melanin is a natural pigment known for its antioxidant and radioprotective properties. Our project is to use melanin for its ability to scavenge free radicals and protect cells from oxidative stress.

To construct our melanin production system, we designed a plasmid based on the AIDA autotransporter system, which mediates the export of heterologous proteins to the bacterial surface or extracellular space.

In this construct, the tyr1 gene encoding tyrosinase was genetically fused to the aida gene sequence using a flexible linker to preserve the correct folding and activity of both protein domains.

To facilitate detection and purification, a MYC tag was added at the C-terminus and a His tag at the N-terminus of the tyrosinase domain.

This design is expected to enable the efficient extracellular export of active tyrosinase, thereby allowing melanin biosynthesis to occur outside the bacterial cell.

Plasmid Design

Figure 1.1: Plasmid Design - AIDA autotransporter system fusion with tyrosinase gene (tyr1) via a linker with addition of MYC tag (C-terminal) and His Tag (N-terminal) to facilitate detection and purification. Goal: Efficient extracellular export of tyrosinase → melanin production outside the cell.

Melanin Production System

Figure 1.2: Melanin production system based on tyrosinase exposure at the bacterial cellular surface through the AIDA autotransporter.

Cloning Strategy: Introduction of the tyr1 sequence into the pAIDA plasmid by insertional PCR cloning

To construct the pAIDA-tyr1 plasmid, we first amplified the tyrosinase gene (tyr1) by PCR. The PCR amplification produced the expected tyr1 fragment of ~1000 bp, as shown in the agarose gel figures 1.3.a and 1.3.b.

Several colonies were screened by colony PCR (Figure 1.3.a and 1.3.b). Colonies 1, 5, and 6 were positive for the presence of the pAIDA-tyr1 construct.

Colony PCR Results

Figure 1.3: Colony PCR showing amplification of the tyr1 gene, attesting successful cloning in the pAIDA plasmid. PCR signal from colonies 2 and 4 similar to the control (T- = pAIDA), indicating that cloning failed in the corresponding plasmids. However, PCR on colonies 1, 5 and 6 show 1kb amplification product, as expected for successful insertion of tyr1 in pAIDA.

Validation of Construction

After sequencing verification, only colonies 1 and 5 were confirmed to contain the correct tyr1 sequence within the pAIDA plasmid. The plasmids from colony 1 and colony 5 were selected for further experiments on tyrosinase expression and melanin production.

Validation of Protein Production

Then, the recombinant pTyr1-AIDA plasmid was transformed into E. coli W3110 and W3110 ΔompT super competent cells. Western blot analysis with anti-His and anti-Myc antibodies were used to confirm production of the pAIDA-tyr1 construct.

In the E. coli W3110 ΔompT strain, a clear band at the expected molecular weight (~97 kDa) was observed, corresponding to the full-length Tyr1-AIDA fusion protein using anti-Myc antibodies (Figure 1.4.a). In contrast, the anti-His antibody revealed a protein with an apparent molecular weight of 64 kDa corresponding to the full-length AIDA was observed.

These results allowed us to conclude that there is partial cleavage of construct at the C-terminus of the Tyrosinase domain, before the C-Myc tag. Very interestingly, AIDA-production was stronger and more stable in the ΔompT genetic background compared to the wild-type W3110 strain.

Expression Conditions

Induction by IPTG (200 µM and 500 µM) at different temperatures (16°C, 30°C, and 37°C) showed robust expression of tyrosinase, with optimal detection at 16°C and 37°C for colony 1. Therefore, the Tyr1-AIDA construct was successfully produced.

Western Blot Anti-His

Figure 1.4.a: Western blot revealed with anti-His tag antibody showing production of His-tagged tyrosinase AIDA. T- = 64 KDA = control of Laetitia's team with His Tag. pAIDA-tyr1 = 97 KDA. Gel 12%. C-Myc (BSA).

Western Blot Anti-Myc

Figure 1.4.b: Western blot result with anti-Myc antibody.

Western Blot Optimization

Figure 1.4.c: Western blot result at different concentrations and different temperatures with antibody anti-Myc.

To verify partial cleavage of construct at the C-terminus of the Tyrosinase domain, before the C-Myc tag, we collaborated with David Hörnström's team in Sweden to compare their tyrosinase expression with ours.

Collaboration Results

Figure 1.5: Western blot results comparison with David Hörnström's team. Conclusion: we have same results.

Visual Tests

At the same time, enzyme activity tests confirmed the functionality of tyrosinase: cultures induced by IPTG and supplemented with L-tyrosine and Cu²⁺ showed visible black pigmentation, characteristic of melanin production. This phenomenon was absent in the non-induced controls.

Melanin Production Visual

Figure 1.6: Melanin production of bacteria containing our Tyr1-AIDA construction on plates supplemented with IPTG, tyrosine and Cu2+. Results: clear melanin production was observed in the bacterial strains.

Enzyme Activity Tests

In vivo tyrosinase activity measurements showing the strong increase in absorption at 400 nm after incubation of bacteria with IPTG, tyrosine and Cu2+ in W3110 bacteria containing our pAIDA1-Tyr plasmid after about 10 hours, and in the Delta OmpT strain (which grew less well) after about 30 hours.

Enzyme Activity Results

Figure 1.7: Enzyme activity tests showing tyrosinase functionality through melanin production measurement.

Conclusion

In conclusion, this first part enabled the design, construction, and functional validation of our melanin production system. We successfully designed a plasmid, pAIDA-tyr1, exploiting the AIDA autotransporter system for the extracellular export of tyrosinase. The cloning strategy was validated by PCR and sequencing, identifying clones 1 and 5 as carriers of the correct genetic construct.

The expression of the Tyr1-AIDA fusion protein was confirmed by Western blot, revealing the production of the protein at the expected molecular weight (~97 kDa). A key observation was the discovery of partial cleavage of the construct, releasing the AIDA domain (~64 kDa) from tyrosinase. In addition, we demonstrated that the E. coli W3110 ΔompT strain provided a much more stable and efficient genetic background for the expression of our construct.

2/ Enhancement of Tyrosine Biosynthesis

We wanted to introduce the mutation S181A into AroF gene which codes for the DAH7PS, the first enzyme of the shikimate pathway. According to Molecular basis for feedback inhibition of tyrosine-regulated 3-deoxy-D arabino-heptulosonate-7-phosphate synthase from Escherichia coli by Di Cui this mutation reduces the interaction between DAH7PS and tyrosine as a result the feedback self-regulation decreases. Thus we should have an increase in melanin production. To accomplish that we used the SLIC cloning method

Use of pKng101, suicide plasmid

First, we attempted to introduce our insert into the suicide vector pKng101 which bears streptomycin resistance and is only replicative in the strain CC118. We encountered three issues:

1. One of our flanking sequences for our insert was incorrect, it only led to empty vectors during colony PCR

Figure 2.1: Example of an empty vector

2. With the aid of our instructors we eventually managed to introduce the the insert into pKng101

Figure 2.2: Result of the SLIC realized by our PI: The insert in pkng101

3. Unfortunately the opmT- strain we were using carried streptomycin resistance so could not be used with the pKng101 plasmid

Therefore, we designed new oligonucleotides.

Use of a new vector thermosensitive plasmid Pko33

We decided to use Pko3, a thermosensitive plasmid only replicative at 30°C. This slowed our progress as our strain grew more slowly. Once again, there were two problems:

1. One of the oligonucleotides for the verification (3’VerifpKO3) was designed on the wrong version of Pko3 which led to insufficient amplification. At that time it was impossible to order new oligos

Figure 2.3: Electrophoresis showing that the amplified fragment was too small

C: Control, it’s the miniprep of Pko3. 1 and 2: Number of colonies

2. After testing three different types of primers pairs a, b, c (Cf:Notebook). Only the couple with olS 4.5/2246 seemed to give the correct amplification. We pursued the second recombination event only to find that all the vectors were empty according to the sequencing. This primer pair does not seem 100% reliable

Figure 2.4: PCR with the 3 different primer pairs shows that the first pairs of oligos seems rights. Even though of the sequencing all the vectors were empty

Tube a: oligo olS 4.5 and olS 2246 (it hybridizes on PKO3, gave by our PI). Tube b: oligo olS 4.8 and olS 4.5 Tube c: oligo olS 4.7 and olS 4.6

After further PCR on colonies, we found a vector that contained our insert but there were other unwanted mutations.

Final result

To reduce the number of mutations in our insert we decided to connect the two fragments of our insert by PCR using the Master Mix. The sequencing results were of higher quality, but there was a base deletion that was also found in our other sequencing results. This indicates that our template for making these fragments was mutated. Our time was limited, and we couldn’t obtain the mutation S181A.

Comparison of the insert sequencing made with oLS 4.7;

Figure 2.5: Direct assembling

Figure 2.6: PCR to connect the two fragments of the insert

Reference : Cui D, Deng A, Bai H, et al. (2019) Molecular basis for feedback inhibition of tyrosine-regulated 3-deoxy-d-arabino-heptulosonate-7-phosphate synthase from Escherichia coli. Journal of Structural Biology 206: 322–334

3/ The Biocontainment Strategy

Construction of W3110 Δ ompT ΔmurI for auxotrophy

Conception and validation of the murI gene deletion

In order to create a robust metabolic containment system, we targeted the chromosomal murI gene, which codes for glutamate racemase, an enzyme essential for peptidoglycan biosynthesis. This modification was intended to establish the first layer of our biocontainment system. Inactivation of this gene was designed to render the bacterium auxotrophic for D-glutamate, an amino acid not available in human tissues, thereby preventing proliferation of our bacterium outside of our control.

When designing the murI gene deletion, we deliberately designed a deletion that does not affect the contiguous vitamin B12 transporter gene. This strategic choice was made to create auxotrophy while maintaining the integrity of essential metabolic pathways unrelated to our containment goal.

Agarose gel electrophoresis results after murI deletion consistently revealed the presence of a single amplification product of approximately 1.2 kbp (Figure 3.1) for the majority of clones tested. We used a primer that hybridizes to the kanamycin resistance gene KanR and a primer that hybridizes to the conserved region of murI. This size corresponds precisely to that expected for the ΔmurI deletion, confirming that homologous recombination occurred and that the coding sequence of the murI gene was excised.

Figure 3.1: Genotypic validation of the murI deletion by colony PCR

Lane M: Molecular weight marker. Lane C: W3110 Δ ompT strain showing no PCR product because it lacks the KanR resistance cassette. Lane with the dot : clone that grew when DL glutamate was added during culture one hour after electroporation. Lanes 1 to 6: Different independent candidate clones showing a single PCR band at ~1.2 kbp, the molecular signature of the deletion.

Failure of phenotypic validation: absence of D-glutamate auxotrophy

Based on the literature, an E. coli strain lacking functional glutamate racemase should have a strict nutritional requirement for DL-glutamate and be unable to grow on standard media (Dougherty et al., 1993). To test this prediction, clones genotypically confirmed as ΔmurI were inoculated onto LB agar medium without any D-glutamate supplement (figure 3.2).

This prototrophic phenotype was consistently reproduced in several biological replicates, ruling out the possibility of accidental contamination

Figure 3.2 :D-glutamate dependency test on medium LB agar.

A/Clones growing on LB Kanamycin agar. B/ Clones growing on LB Kanamycin 20 mM DL-Glutamate agar

The dot: clone that grew when DL glutamate was added during culture one hour after electroporation. Clone 1 to 6:Different independent candidate clones, no DL glutamate added during the one hour culture after electroporation.

The absence of any visible difference in growth conclusively demonstrates that ΔmurI clones do not exhibit the expected auxotrophic phenotype.

The growth profiles of the mutant strains W3110 ΔompT and ΔmurI were evaluated by measuring optical density (DO600) in LB liquid medium without DL glutamate supplementation. No significant differences were observed, with both strains exhibiting superimposable growth profiles with no detectable delay. These results confirm that the murI deletion does not induce DL glutamate auxotrophy or even slow cell growth under these conditions.

Further investigation and consequences

The phenotype we observed shows that the deletion of murI is not sufficient to induce DL glutamate auxotrophy. We were therefore unable to establish the Δ murI base strain as functionally auxotrophic. As a result, the basis of our containment system could not be achieved.

However, faced with this contradiction between genotype and phenotype, we would have liked to have pursued the investigation further to identify the problem, but due to lack of time we were unable to do so. We would have liked to sequence the 1.2 kbp PCR product. Analysis of the sequence could confirm the absence of the murI coding sequence, which would validate the integrity of the deletion.

The planned steps of complementation with the pMurI plasmid, which carries a functional copy of murI, could not be carried out. Neither could the steps of mutation of the aroF pathway in our final strain, nor the melanin production test and encapsidation in the hydrogel. Thus, the first and main layer of our biological containment system could not be validated experimentally. This failure to reproduce the expected phenotype raises important questions about metabolic redundancy or the adaptation of E. coli strains.

Construction of complementation plasmid of Δ murI : pMurI

Blue-white screening of E. coli DH5ɑ with pMurI

In order to provide a functional copy of the murI gene to a genetically modified bacterial strain that is auxotrophic for D-glutamate, we attempted to construct a complementary plasmid pMurI using the Golden Gates assembly technique, then transformed the newly created plasmid into DH5α cells and verified its integration using white-blue screening.

Figure 4.1:Example of results for our DH5α E. Coli cells and NEB DH5α (C29871) E. Coli cells transformed with pMurI on an XGal+IPTG+Kanamycin plate as part of a blue-white screening.

After cell transformation and screening visualization, we first observe a higher number of white cells with NEB cells compared to our transformed cells, which produce more blue cells, i.e., cells that have not incorporated the plasmid. This means that NEB cells are more competent than the ones we created and that they have successfully integrated the plasmid.

Colony PCR, Enzymatic Digestion Electrophoresis

Figure 4.2:Example of PCR results and enzymatic digestion with EcoRI/XmaI and EcoRI/SpeI of pMurI after its transformation in DH5α E. Coli

We tested digesting our plasmid with two different pairs of restriction enzymes. In the case of EcoRI/XmaI, these enzymes cut within LacZ; if the vector is empty, we obtain two bands (2931 bp and 220 bp), otherwise we obtain a linearized band. In the case of EcoRI/SpeI, the opposite is true. That is, if the vector is empty, we obtain the size of lacZ, which is 460 bp; otherwise, we obtain two bands (2931 bp and 1134 bp).

Unfortunately, after electrophoresis, this is not very visible in terms of markers. We can only conclude that for EcoRI/XmaI there are no empty plasmids except for the control. As for the results of the second digestion, the results cannot be used without the marker. Since subsequent experiments showed the presence of our plasmid, we did not attempt another enzymatic digestion.

A PCR colony assay was performed to observe whether the DNA in our cells contained an empty plasmid (460 bp), lacZ, or our plasmid constructed by Golden Gates (1100 bp). We can see that among the colonies tested, colonies 3, 4, and 5 are the expected size in the case of the presence of the plasmid we integrated which means that the transformation worked well. (fig4.2A)

Western Blot of pMurI with an anti-HA antibody and sequencing of pMurI

Figure 4.3:Example of Western blot results for our pMurI plasmid with HA tag

For our initial results, we observed expression in two of the three plasmids tested (fig4.3.A), but visibility was poor. Assuming that the cause was the presence of disulfide bridges, we performed another Western blot, this time adding β-mercaptoethanol to the loading buffer. This time, the results are much clearer, as is the expression in the complementary plasmid tested.

Since the murI protein normally has a molecular weight of 32 kDa, we can clearly see a band between 28 and 38, which corresponds to the expected size. This experiment shows that our complementary plasmid produces the murI gene, enabling the complementation of a strain in which this same gene has been inactivated. It is assumed that the following bands correspond to remains.

Finally, our plasmid is sequenced and then aligned with the expected sequence.

Figure 4.4:Example of alignment of the expected cloning results with the sequenced plasmid after its transformation DH5α E. Coli

By aligning the sequenced fragment with the expected sequence, we observe a very good overall alignment. After manually checking the multiple sequences that were produced, we note that there are no gaps or mutations and that those shown in Figure 4.4 are in fact sequencing errors. Ultimately, the complementary plasmid we created is as expected and successfully produces the murI protein.

Further investigation and consequences

As we were unable to finalize the DL-glutamate auxotrophic E. coli W3110 strain, it was not possible to test its complementation with our complementary pMurI plasmid. Nevertheless, a new composite part remains that can be tested by future teams and will enable future iGEM teams to complement a DL-glutamate auxotrophic E. coli strain.

4/ Hydrogel Encapsulation and Functionality

After setting up our protocols for hydrogel synthesis and characterization, we tested the system through a series of experiments. As a first step, we confirmed the difference between our motile and non-motile E. coli strains with a motility assay. Surprisingly, our W3110 ΔompT strain (the final strain used in our project) turned out to be motile (Figure 5.1). We had initially assumed it would be non-motile, since we believed that the ompT deletion might have been generated using the λ-Red recombination system, which is known to occasionally introduce secondary mutations in nearby loci. Such mutations can, in some cases, disrupt genes involved in flagellar assembly and result in a loss of motility.

Figure 5.1: Motility test results for E.coli W3110 ΔompT and E.coli W3110

The motility test successfully validated the difference between the motile ΔompT and the non-motile strains, allowing us to use them as a reliable model to evaluate containment.

Next, we transformed the pBAD33 plasmid (arabinose-inducible) into both strains, enabling us to track the cells during experiments using the Magic Box system and to monitor GFP fluorescence as a reporter of cellular activity (Figure 5.2).

Figure 5.2: Transformation results of pBAD33 transformation intoE.coli W3110, E.coli W3110 ΔompT induced by arabinose and E.coli W3110 non-motile.

We proceeded by developing our own alginate hydrogel synthesis protocol and evaluated both the ability of the cells to escape from the matrix and their survival rate over time. To demonstrate cell confinement, we designed a bacterial escape assay, which we repeated several times until the protocol was optimized (Figure 5.3)

Figure 5.3Our experiments confirmed the effectiveness of the alginate hydrogel in physically containing E. coli. We demonstrated that the matrix significantly limited bacterial escape, whether the hydrogel was carefully preserved or mechanically disrupted to simulate escape (in which case OD measurements increased slightly faster than under the intact condition). Escape curves revealed a clear difference between the two strains, with motile cells showing a higher propensity to move through the hydrogel compared to the non-motile strain. Nevertheless, the majority of encapsulated cells remained confined within the hydrogel throughout the tested time frames

These results are promising and suggest that introducing a motility mutation could be considered in future engineering cycles to further reduce bacterial escape for potential applications

A bacterial survival assay was also performed to assess the viability of cells within the hydrogel matrix over time (Figure 5.4).

Figure 5.4: Survival assays showed that encapsulation did not compromise cell viability. The survival histograms indicate that a substantial proportion of cells remained alive inside the matrix, demonstrating that the hydrogel provides a protective environment compatible with bacterial growth, even after 72 hours.

To better understand the structure of the hydrogel and the position of the cells inside it, we also performed SEM imaging of our hydrogel marbles (Figure 5.5.a, 5.5.b and 5.5.c)

Figure 5.5.a: SEM images of 1.5% alginate marbles. 10µL marbles containing E.coli W3110 non motile cells

Figure 5.5.b: SEM images of 1.5% alginate marbles. 10 µL marbles without cells

Figure 5.5.c: SEM images of 1.5% alginate marbles. 10 µL marble containing E.coli W3110 Δ ompT cells

SEM imaging further confirmed the structural integrity of the hydrogel and showed the embedded cells distributed throughout the matrix in agglomerates.

Finally, we used fluorescence microscopy (Figure 5.6) to confirm that our encapsulated engineered strains were still functional and able to express GFP inside the hydrogel.

Figure 5.6: Fluorescence microscopy provided additional confirmation that our engineered strains were successfully encapsulated and remained functional, as indicated by GFP expression inside the hydrogel.

Together, these results demonstrate that our alginate hydrogel is an efficient and biocompatible system for containing genetically modified E. coli, while preserving their viability and capacity for molecular production. However, its performance could be further improved by modifying certain aspects, as discussed in the Future Work section.

5/ Futurs Works

Experiment to test the mutation on AroF

In order to test our mutation on AroF which is supposed to increase the production of tyrosine by E.Coli, we planned 2 experiences:

The first one is to analyze the production of tyrosine by HPLC ( High Performance Liquid Chromatography). This kind of chromatography allows us to measure the concentration of tyrosine. Here is the protocol we planned to use:

  • 1.Take 500 µl of bacterial culture
  • 2.Dilute the sample in 1 N HCl
  • 3.Incubate at 55°C for 30 minutes, shaking occasionally with a vortex mixer
  • 4.Centrifuge the sample and collect the supernatant
  • 5.Dillute the supernatant with the appropriate amount of distilled water
  • 6.Inject the prepared sample into an Agilent 1200 Series HPLC system equipped with a photodiode array detector (PDA) set at 210, 254, and 280 nm
  • 7.Perform seperation on a reverse phase C18 column(Inertsil; 2.1 x 250 mm, 3.5µm; GL Sciences) at a flow rate of 0.15mL/min
  • 8.Use a mobile phase of water (A)/ methanol (B) with the following gradient:
    • 0-8 min: 5 % B
    • 8-13min: 40 %B
    • 13-16min: maintain at 40% B
    • 16-21 min: 5% B
    • 21-31min: equilibrate at 5%
  • 9.Quantify L-tyrosine using a 5 points calibration curve (14 to 448 mg/L)
  • 10.Check the quality of calibration (R² = 0.99)

After the HPCL we’re going to do a measurement activity test to see if we obtain more melanin

Reference : Juminaga D, Baidoo EEK, Redding-Johanson AM, et al. (2012) Modular Engineering of l-Tyrosine Production in Escherichia coli. Appl Environ Microbiol 78: 89–98.

Plan for phenotypic characterization of the ΔmurI strain

If we had succeeded in obtaining the ΔmurI auxotrophic strain, we would have undertaken a phenotypic characterization.

We would first have performed a kinetic analysis of growth in LB medium. This study would have compared the growth profiles of four strains: the wild-type W3110 ΔompT strain as a control, the ΔmurI auxotrophic strain, the ΔmurI strain complemented with the pMurI plasmid, and the wild-type strain complemented with pmurI. We would have measured the growth curves in LB medium with and without 20 mM DL-glutamate supplement. And we would have performed three biological replicates for each condition. This analysis would have allowed us to determine essential parameters such as the growth delay or absence of growth without glutamate supplement and the final cell yield.

At the same time, we would have evaluated the survival profile of the strains under deficiency conditions. The protocol would have consisted of resuspending stationary cultures in nutrient-free PBS, then monitoring cell viability by counting CFU/mL for 48 hours. Analysis of the survival curves would have provided information on the respective mortality rates of the different strains in the absence of DL-glutamate.

A study of the response to different concentrations of D-glutamate was also conducted. We tested a full range of concentrations (0, 5, 10, 20, and 50 mM) to establish the minimum concentration required for normal growth. For each concentration, we produced complete growth curves and measured viability after 24 hours of deprivation.

To complete this characterization, we would have quantified the expression of the murI gene by real-time qPCR. This analysis would have compared the level of expression of the murI gene in the wild-type strain and in the strain complemented with pMurI. We would have measured this expression under different conditions in the presence and absence of D-glutamate and at different growth stages.

This characterization would have provided us with essential data for evaluating the potential of the metabolic containment system. Although the failure to obtain the auxotrophic strain prevented us from carrying out this work, this detailed protocol remains valid for future attempts to optimize the system.

Reference : Cabral, M. P. et al. Design of live attenuated bacterial vaccines based on D-glutamate auxotrophy. Nat Commun 8, 15480 (2017).

3/ Poly-DL-γ-glutamate (PGA) hydrogel for biocontainment

Building on our current results with alginate hydrogels, future work could explore the use of a poly-DL-γ-glutamate (PGA) hydrogel, as described in the scientific paper Li, J., Huang, Y., Wang, Y. & Han, Q. A Poly-γ-Glutamic Acid/ε-Polylysine Hydrogel: Synthesis, Characterization, and Its Role in Accelerated Wound Healing. Gels 11, 226 (2025). Since our engineered E. coli strains are auxotrophic for D-glutamate, a PGA-based matrix could provide an additional layer of biocontainment: cells would only survive and remain active within the hydrogel, while any escaped cells would be unable to grow in the absence of D-glutamate. This strategy could be combined with motility-reducing mutations to further minimize bacterial escape.

Moreover, PGA hydrogels could be optimized to maintain cell viability and molecular production, while allowing controlled diffusion of molecules such as the melanin produced by our strain. Structural characterization of the PGA matrix, including porosity and degradation over time, would provide insights into cell confinement and metabolite exchange. Finally, future applications could include scaling up the hydrogel for safe bioproduction, designing hybrid matrices combining alginate and PGA, and integrating sensors to monitor metabolites in real time within the hydrogel.

4/ Genetic Modifications to Enhance Containment

To further improve the biocontainment of our engineered E. coli within the hydrogel, targeted genetic modifications could be considered. For example, deleting the fliC gene, which encodes the flagellin protein essential for flagellar assembly, would abolish motility and reduce the ability of cells to escape the hydrogel matrix (similar to our ΔompT strain).

5/ Biocompatibility and Biodegradability Testing

To further optimize hydrogel-based scaffolds, future work could focus on evaluating their biocompatibility and biodegradability. Biocompatibility can be assessed in vitro by measuring cell viability, proliferation, and adhesion using assays such as Live/Dead staining. For osteogenic applications, differentiation markers such as alkaline phosphatase (ALP) activity or mineralization can be monitored to confirm that the hydrogel provides a supportive environment for osteoblast function. Additionally, inflammatory responses could be evaluated by exposing immune cells to the hydrogel and measuring cytokine production.

Biodegradability could be tested by monitoring hydrogel mass loss or structural changes over time under physiological conditions (for example in PBS at 37°C) or in the presence of relevant enzymes such as lysozymes or proteases. Mechanical properties, porosity, and release of embedded molecules could provide complementary insights into scaffold degradation. Combined approaches, such as co-culturing cells on the hydrogel while tracking degradation, or using microscopy (SEM or confocal) to visualize structural changes and cell distribution, would provide a comprehensive understanding of both cell compatibility and controlled scaffold resorption.

6/ Injectability of the hydrogel

For potential applications, it is important that hydrogels exhibit injectability. This means that the gel should be in a liquid or semi-liquid state under specific conditions, such as temperature or pH, allowing for easy injection into the desired site. After injection, the hydrogel should rapidly solidify to form a stable gel, maintaining its structural integrity and providing a supportive environment for encapsulated cells.

Future work could focus on optimizing the physicochemical properties of the hydrogel to achieve this transition efficiently, while preserving cell viability and molecular production. Techniques such as rheological measurements could be used to assess the gel’s flow properties and gelation kinetics under different conditions.

7/ Melanin Diffusion Control

To ensure effective ROS reduction, future work could focus on controlling melanin release from the hydrogel. Simple diffusion assays could help us track melanin concentration in surrounding media over time, while ROS scavenging tests would help us verify antioxidant activity. Mathematical modeling could help predict release kinetics and optimize hydrogel properties. These tests together would guide the design of our hydrogel that maintains local ROS reduction while keeping cells contained and viable.

Protocol Overview:

  • Prepare hydrogel: Embed melanin-producing E. coli in alginate or PGA hydrogel. Place hydrogel in PBS or culture medium
  • Diffusion assay: Collect small aliquots of the surrounding medium at regular time points (e.g., 0, 2, 4, 8, 24, 48 h). Measure melanin concentration using UV-Vis or fluorimetry
  • ROS scavenging assay:Test supernatants on ROS-sensitive dye or cells to verify antioxidant activity over time.
  • Data analysis: Plot melanin release curves and ROS reduction activity. Compare hydrogel formulations to identify optimal conditions for controlled diffusion.

8/ Microstructure Optimization

To improve the functionality of our hydrogel in tissue engineering applications, advanced techniques such as microfluidics and 3D printing can be explored. These approaches allow the design of specific porosity patterns, optimizing both the controlled diffusion of therapeutic molecules, such as melanin, and the confinement of cells within the scaffold. For example, Wang et al. (2011) demonstrated the fabrication of highly organized alginate scaffolds using microfluidic technology, resulting in well-defined porous structures that promote cell proliferation and cartilage formation in tissue engineering (Wang, C.-C., Yang, K.-C., Lin, K.-H., Liu, H.-C. & Lin, F.-H. A highly organized three-dimensional alginate scaffold for cartilage tissue engineering prepared by microfluidic technology. Biomaterials 32, 7118–7126 (2011).). Additionally, 3D printing enables the creation of scaffolds with complex and controlled geometries, facilitating the customization of scaffold architecture according to the specific needs of the target tissue. Combining these technologies we could significantly enhance hydrogel performance, providing an optimal environment for cell adhesion, proliferation, and differentiation.

9/ In Vivo Testing

For future development, in vivo studies could be conducted to evaluate the safety, biocompatibility, and efficacy of the hydrogel system in animal models. These tests would provide critical information on how the scaffold behaves in a complex physiological environment, including degradation rate, immune response, cell viability, and controlled release of molecules such as melanin. In addition, in vivo studies would allow assessment of tissue regeneration, such as bone repair, and confirm that the hydrogel can effectively contain engineered cells while supporting host tissue integration. These experiments are essential before considering potential clinical applications.

10/ Implementation of a Toxin-Antitoxin Biocontainment System

To reinforce our multi-layered biocontainment strategy, a genetic "dead-man's switch" based on the CcdB/CcdA toxin-antitoxin system could be implemented. The CcdB toxin poisons DNA gyrase, leading to rapid cell death, while the CcdA antitoxin neutralizes CcdB.

Proposed Design:

Constitutive Toxin Expression: The ccdB gene would be placed under the control of a constitutive promoter, ensuring constant production of the lethal toxin to prevent any plasmid conjugation.

Validation Experiments:

  • Plasmid Construction: Clone the ccdB and ccdA genes into a bicistronic operon or on compatible plasmids with the specified promoters.
  • Killing Efficiency Test: Transform the system into our production strain and measure the loss of cell viability (by CFU count) over time after removal of the antitoxin inducer.
  • Containment Efficacy: Co-encapsulate the engineered strain with the toxin-antitoxin system in our hydrogel and measure the number of viable cells that escape into an inducer-free medium, comparing it to a control strain without the system.

This system would provide a potent, genetically encoded fail-safe mechanism, working synergistically with our physical (hydrogel) and metabolic (ΔmurI auxotrophy) containment layers to ensure maximum environmental safety.

11/ Integration of All Modules and System-Level Validation

The ultimate goal is to integrate all developed modules into a single, robust, and safe platform. This involves assembling the melanin production system, the optimized tyrosine pathway, and all biocontainment layers (metabolic, physical, and genetic) into a final chassis strain.

Integration Pipeline:

Strain Consolidation: Combine the successfully constructed genetic parts—pAIDA-tyr1 (melanin production), the feedback-resistant aroFS181A mutation (tyrosine overproduction), the ΔmurI deletion with pMurI complementation (metabolic containment), and the CcdB/A toxin-antitoxin system (genetic containment)—into the E. coli W3110 ΔompT background.

Functional Characterization:

Test the fully integrated strain to ensure that all systems function harmoniously without negative interference. Key assays would include:

  • Melanin Quantification: Measure the final yield of melanin production under optimized conditions.
  • Containment Verification: Perform a comprehensive escape assay, combining the hydrogel with all genetic safeguards, to quantify any viable cells in the external environment.
  • Growth and Stability: Monitor the strain's growth kinetics and genetic stability over multiple generations to ensure robust performance for potential applications.

Conclusion

In this project, we successfully designed and validated a melanin production system in E.Coli using the AIDA autotransporter to achieve extracellular export of tyrosinase, enabling efficient melanin synthesis.

Cloning, sequencing and expression analyses confirmed the functionality of our constructs, and enzyme activity tests demonstrated robust melanin production. Despite challenges in introducing the S181A mutation in the aroF gene and generating the Δ murI auxotrophic strain, the project established reliable methodologies and genetic tools that can be further optimized in future work.

Encapsulation of engineered bacteria in the alginate hydrogel provided effective physical containment while preserving cell viability and molecular activity, as confirmed by the survival tests, escape tests, SEM imaging and fluorescence microscopy. These results demonstrate the potential of therapies such as biodegradable hydrogel scaffolds for safe and functional deployment on engineered cells.

Future improvements described before will be crucial to validate Living Scaffolds performance and safety for potential therapy application. Overall, our work establishes a solid foundation for a modular, functional system that combines bacterial melanin production and hydrogel based confinement, providing a versatile platform for future synthetic biology and regenerative medicine applications.

Scroll to Top Scroll to Top