Our experimental design covers the set of procedures we executed in order to successfully produce our recombinant endolysins. Covering everything from our plasmid construction to the bactericidal assays, this page explains our journey through the laboratory.

Round Zero: Plasmid Construction and Analysis

The initial round of experimentation encompassed the in silico design of our plasmids, their synthesis, assembly, transformation, and analysis of the constructs. This comprehensive phase integrates computational design with practical implementation, culminating in the evaluation of the resulting constructs. To design our endolysins coding sequences, we conducted an in-depth protein analysis. This helped us to identify the crucial domains required to produce highly effective and catalytically specific endolysins. The coding sequence (CDS) of the three recombinant intein-fusion proteins (LysK endolysin, LysCSA13, and LysSS) was in silico edited to incorporate BsaI enzyme recognition sites at the ends of the sequence. In the same way, each CDS design was evaluated through the protein structural model analysis tool, Alpha Fold 2, to identify the original functional protein structures and compare them with our final designs, ensuring their enzymatic activity. Afterwards, we proceeded with the selection of the bioparts to be utilized in the assembly of the transcriptional units (TUs) that would express our proteins.
The selected parts for the construction of the transcriptional units were:

1. Promoter

   BBa_J435350

2. RBS

   BBa_Z0262

3. Terminator

   BBa_J435371

4. Backbone

   BBa_J435330

These parts were assembled in SnapGene® software using the Golden Gate method, employing the type IIS restriction enzyme BsaI.The transcriptional units (TUs) of each of our endolysins (TU LysK, TU LysCSA13, and TU LysSS) were synthesized by Twist Bioscience, making this possible by the sponsorship of 10 kb of FREE Twist Gene Fragments for iGEM teams.

First Round: Golden Gate Assembly

From the iGEM 2023 plate, the BioBrick™ BBa_J435330 (plate 2, well A15) was resuspended, which was transformed into E. coli 5-alpha by heat shock, and subsequently extracted and purified with the PureLink™ Quick Plasmid Miniprep Kit (Invitrogen™, Lot #01260760). This part was chosen as the backbone due to its high replication origin, dual antibiotic resistance (kanamycin and ampicillin), and compatibility with the Golden Gate assembly method. Once the BBa_J435330 backbone was extracted and purified, Golden Gate assembly reactions were performed using the BsaI enzyme (NEB E1601, Lot #10182427) to insert each of the transcriptional units into the BBa_J435330 backbone.


After the Golden Gate assembly reaction, a bacterial transformation into E. coli 5-alpha was performed, followed by plating on LB agar with kanamycin (50 µg/mL). However, colonies were obtained both on the plates seeded with the transformation and in the negative controls. This led us to suspect that the plates from the Golden Gate reaction might contain a mixture of colonies transformed with the target construct and others transformed with the backbone alone, making them indistinguishable by visual inspection.


To confirm this, re-plating of several colonies from the same plate was conducted onto other plates with kanamycin and ampicillin. This was done in order to take advantage of the dual antibiotic resistance feature present in the used backbone. A colony from the reaction-seeded plate was selected and then plated onto two separate plates: one with kanamycin (50 µg/mL) and another with ampicillin (100 µg/mL). The same identifying number was assigned to both antibiotics. The next day, observation revealed that some colonies grew on both antibiotics, while specific colonies only grew on kanamycin (50 µg/mL) and on ampicillin there wasn't any growth (100 µg/mL). This suggested that these colonies might have been transformed with a construct carrying resistance only to kanamycin (50 µg/mL) and not to ampicillin. This led us to suspect that our construct might indeed be present in these colonies.


To verify this, a colony that exhibited only kanamycin resistance (50 µg/mL) was selected. DNA plasmid extraction and purification were performed using the PureLink™ Quick Plasmid Miniprep Kit (Invitrogen™, Lot #01260760).
With the aim of assessing the effectiveness of the Golden Gate assembly protocol, we conducted an enzymatic digestion to linearize the plasmids, followed by an analysis through agarose gel electrophoresis. We cut pUC-LysK and pUC-LysCSA13 to digestion using the SmaI enzyme, and pUC-LysSS underwent digestion using the EcoRI enzyme.

Second Round: Protein Expression

The next step of our recombinant protein production process was grouped into the second round of experimentation. All the selection, optimization, and standardization of our protein induction protocols were performed as follows:

Protein Production

Our constructs incorporate the biobrick BBa_J435350, featuring a T7 promoter with LacO regulatory regions. This choice was driven by its capability for high transcription levels and effective regulation. To induce protein production in our E. coli BL21 (DE3) transformants carrying pUC-LysCSA13 & pUC-LysK, we employed isopropyl β-D-1 thiogalactopyranoside (IPTG) as an inducer for the LacO regulation regions. This reagent activates the transformants, enabling them to synthesize our endolysins.

Protein Induction Kinetics

To determine the optimal protein induction parameters, we conducted protein induction kinetics. Following prior research, we established that the ideal conditions were approximately 6 hours at 37°C. Experiments were conducted at three IPTG concentrations—0.2 mM, 0.5 mM, and 1 mM—across all transformants using E. coli BL21 (DE3) strains transformed with plasmids obtained via Golden Gate Assembly.


Protein induction kinetics were carried out at 37°C for a maximum of 6 hours with vigorous agitation in an orbital shaker. Each endolysin underwent five treatments: a negative control with E. coli BL21 (DE3), a negative control with E. coli BL21 (DE3) transformed with the endolysin but without IPTG induction, and three E. coli BL21 (DE3) transformed with the endolysin induced at different IPTG concentrations—0.2 mM, 0.5 mM, and 1 mM. Samples were collected hourly, yielding 200 μL for each treatment.
Next, we performed SDS-PAGE analysis. Given that the molecular weight of LysCSA13-ABD is 35.098 kDa, and LysK-ABD-SH3B30 weighs 28.437 kDa, we used 12% and 6% concentrations for the separator and concentrator phases, respectively.

Fraction Identification

Finally, once the optimal concentration and induction time parameters were established for LysK and LysCSA13, an SDS-PAGE was performed using the soluble and insoluble fractions from each of the crude bacterial lysis extracts after induction. This step was crucial in assessing the solubility of both endolysins and, consequently, ensuring accurate purification.

Third Round: Protein Purification

After phase two, valuable information was gained regarding LysCSA13-ABD and LysK-ABD-SH3B30. Specifically, it enabled us to determine the optimal induction conditions and solubility of each. This paved the way for the subsequent round of experimentation: purification through metal affinity chromatography for His-tagged proteins (Thermo Scientific™, 88228).

Protein Purification

For this phase, we conducted a massive induction to produce our endolysins in greater quantities. Subsequently, we proceeded with purification through metal affinity chromatography using centrifugation columns.

Validation of the Endolysin Purification Process

In order to confirm the successful isolation of our recombinant proteins through the purification process, an analysis was conducted using SDS-PAGE. This involved verifying the eluate from the purification and contrasting it with two negative controls: non-transformed E. coli and non-induced transformed E. coli. Additionally, the molecular weights of the bands obtained in the polyacrylamide gel were confirmed.

Fourth Round: Bactericidal Assays

For this stage the main purpose was to corroborate the catalytic activity of our endolysins, the optimal parameters, and the real effects for each bacteria. In the same way, the chitosan-alginate delivery drug system was analyzed simultaneously with our enzymatic treatment.

Considering our protein purification results, we decided to work on the development of our treatment by using our crude protein extracts. To identify our optimal composition, we designed an experimental model which makes possible the enzyme simultaneous analysis and its effects on the pathogenic bacteria. The full protocol is described in our Experiments Page. Our ELISA plate designing was prepared as follows: