Project Design

Our vision

Our intended product is a biocompatible material designed for burn wound healing, combining a structural scaffold with a hydrogel layer. The scaffold, composed of casein and propolis fibres, supports smooth tissue regeneration[1, 2], while the hydrogel provides a protective, moist, and antibacterial environment that promotes optimal healing.

Caseins

α-S1-casein

Fig. 2 - α-S1-casein

Initially we chose to use α-S1-casein from Homo sapiens for scaffold creation, as human-origin protein ensures higher biocompatibility than bovine caseins and a lesser risk for potential allergic reactions. Proteins derived from human sources are hypothesised to offer superior biocompatibility due to their intrinsic biological recognition, reduced immunogenicity, and enhanced integration with human tissues[3]. Besides, it has been determined before that bovine α-S1-casein, found in cow’s milk, is an allergen responsible for significant IgE-mediated allergic reactions[5]. Better biocompatibility of human-origin proteins and lowered immunogenic risk was also confirmed during our consultation with allergologist Dr.med. Māris Bukovskis.

α-S1-casein is a suitable choice for scaffolds for many reasons. First of them: mild aqueous processing. α-S1-casein can be processed in water under gentle conditions (pH 6–8, ≤ 37°C) and their well-defined isoelectric point (~pH 4.6), which provides a solvent-free handle for acid precipitation and pH-triggered aggregation/gelation during fabrication[6]. The second useful characteristic is phosphorylation and Ca- bridging. Compared with β-casein, α-S1-casein contains more phosphoserine residues arranged in clusters (two or more), conferring stronger Ca2+/calcium-phosphate binding and promoting tighter, mineral-bridged networks – this Ca2+ association is central to micelle formation and stability[7].

However, during our research, we came to a conclusion that Bos taurus or bovine-origin casein is significantly more researched than human-origin α-S1-casein. That prompted us to include it in our project alongside human-origin α-S1-casein as a well-characterised reference choice, which has been previously researched for biomaterial applications and electrospinning possibilities [8-10] [1-3]. This would allow us to better evaluate and optimize our experimental protocols, compare the properties of the resulting material and determine what advantages human-origin α-S1-casein has over bovine-origin one for creating scaffolds intended for tissue regeneration facilitation.

β-casein

Beta-casein is a phosphoprotein and a primary component of bovine milk casein micelles, characterised by its amphipathic structure and high proline content, which prevents extensive secondary structure formation. Its hydrophobic C-terminal region is crucial for micelle stabilisation, while its phosphorylation sites facilitate calcium binding and casein micelle assembly.

Human β-casein possesses several properties advantageous for mild gel fabrication and drug loading:

Bacterial strains for protein expression

For protein expression we used two different E. coli BSL-1 strains – BL21(DE3) and Rosetta 2 (DE3) – which were kindly shared with us from a partner laboratory at the Latvian Biomedical Research and Study Centre. Both of these strains are designed specifically for protein expression.

BL21 (DE3)

Genotype: fhuA2 [lon] ompT gal (λ DE3) [dcm] ∆hsdS

BL21(DE3) is one of the most widely used E. coli strains for recombinant protein production. It is derived from the B lineage of E. coli and engineered with the DE3 lysogen – a λ phage carrying the T7 RNA polymerase gene under lacUV5 promoter control. This strain is favoured for its deficiency in lon and ompT proteases, reducing degradation of heterologous proteins, and its lack of the hsdS gene, which prevents restriction of methylated foreign DNA. The T7 expression system (induced by β-D-1-thiogalactopyranoside - IPTG) enables tight, high-level transcription of target genes cloned under a T7 promoter, making it ideal for producing toxic or high-yield proteins[14].

Rosetta2 (DE3)

Genotype: F-ompT hsdSB(rB- mB-) gal dcm (DE3) pRARE2 (CamR)

Rosetta2 (DE3) is a cell line based on the BL21 (DE3) designed to enhance the expression of proteins coded by eukaryotic genes that contain codons rarely used in E. coli. It carries a pRARE2 plasmid (chloramphenicol-resistant) supplying tRNAs for seven rare codons (AGA, AGG, AUA, CUA, CCC, GGA, UUA), which are under-represented in E. coli but common in eukaryotic genes (e.g., human, plant, or fungal sequences). This supplementation enhances translation efficiency and solubility of proteins prone to premature termination or misincorporation, such as those from mammals or GC-rich organisms. Like BL21(DE3), it features the T7 RNA polymerase under lacUV5 control for IPTG-inducible expression while retaining the lon and ompT protease deficiencies to minimise protein degradation. Rosetta 2 (DE3) is particularly valuable for complex or GC-rich targets [15].

Although our constructed gene sequences are codon optimised for expression in E. coli, we decided to test both strains to see whether there are any advantages to using one specifically designed for proteins encoded by eukaryotic genes.

Bacteriophages

Bacteriophages – bacteria-infecting viruses – are the most abundant entities on earth, their total estimated count coming up to 10³¹. These viruses have been isolated from virtually every environment where bacteria are found, and it is believed that there are multiple phages capable of infecting each and every bacterial strain [16]. Phages are uniquely diverse in their complexity, size, shape and genomic constitution as a result of billions of years of evolution and adaptation to various environments [17]. While phages are appealing for many scientific and practical applications, one of their most promising roles lies in medicine, where they offer a potential solution to the growing problem of antimicrobial resistance. Current research on phages as antimicrobial agents focuses on lytic phages and their proteins, as during the life cycle of a lytic phage, host bacterial cells are broken and destroyed after replication to release new phages. This ability makes them ideal for phage therapy as they directly reduce bacterial populations [18]. Unlike lytic phages, lysogenic phages integrate into the bacterial genome as prophages, which replicate within the host cells until environmental stress triggers their induction into the lytic cycle, resulting in host cell lysis [19]. One of the most common approaches in phage therapeutics development is the formulation of a phage cocktail in order to reduce the potential for bacteria to develop phage resistance [20].

As highlighted by the phage researcher Dr. biol. Ņikita Zrelovs, phage therapy research mainly focuses on dsDNA tailed phages. Among them are viruses that have been long-time model organisms in fundamental research phage, namely, T4, T7 and λ phages. These three phages represent the main morphological groups of tailed dsDNA phages that are explored for therapeutic applications, which prompted us to choose them for our proof-of-principle studies.

Propolis

Artepillin C and fibroblast growth factors

A study by Kłósek et al. (2021) showed that ethanolic extract of Brazilian green propolis (EEP B) and its main compound, artepillin C, significantly increased secretion of acidic fibroblast growth factor (aFGF 1) by human gingival fibroblasts, while also decreasing E selectin levels [21]. This suggests both regenerative and anti inflammatory potential relevant to early wound healing stages.

Antibacterial spectrum and MIC values

Gomes et al. (2024) reported that green propolis extracts from Brazil exhibit strong antibacterial activity mainly against Gram positive bacteria (e.g., Staphylococcus aureus, Streptococcus mutans) with MICs in the low μg/mL range (e.g., the geographical origin (region, plant source) strongly influences the potency)[23].

Variation in phenolic profiles across bee species and places

A recent study on propolis from Brazilian stingless bees found that both phenolic and coumaric constituent levels vary with bee species and location [24]. For instance, formononetin was identified in samples where it was not previously reported, showing novelty in chemical profiles. This implies that sourcing Brazilian propolis for biomedical applications allows for selecting or standardising types with desired properties.

Hyaluronic acid

making it not only a structural component but also a signalling molecule involved in regenerative pathways.

The biological activity of HA is strongly dependent on its molecular weight. High molecular weight HA exhibits anti-inflammatory and immunosuppressive effects, whereas low molecular weight fragments can promote angiogenesis and immune cell recruitment. This dual behaviour enables the design of HA-based hydrogels with specific biological profiles. In recent reviews researchers emphasise that such tailored HA systems have shown clear benefits in wound healing, both in vitro and in clinical applications, particularly when applied as part of composite scaffolds or hydrogels with bioactive co-components [25].

Another research [26] further highlights that chemical modification of HA — for example, oxidation or methacrylation—can enhance its functionality, allowing precise control over degradation rate and mechanical properties. These advancements allow HA to be used not only as a passive carrier but as an active material tailored for regenerative medicine [26]. Moreover, there is recent research exploring how varying crosslinking approaches directly influence HAs performance in wound dressings, especially in terms of its integration with other biopolymers and ability to deliver drugs or cells to the wound site [27].

Together, these studies affirm that HA remains one of the most versatile and clinically relevant biopolymers in burn wound care, and its application continues to expand with the integration of synthetic biology tools and responsive material design.

Crosslinking

Despite the biological advantages of hyaluronic acid and other natural polymers, one major limitation is their insufficient mechanical strength and rapid enzymatic degradation. Crosslinking strategies are thus essential to stabilise hydrogels, extend their residence time in the wound, and modulate their physicochemical properties. Crosslinking not only improves the structural integrity of hydrogels but also determines their swelling behaviour, porosity, stiffness, and degradation kinetics—parameters that are critical for ensuring cell compatibility and therapeutic efficacy [28].

There are multiple crosslinking strategies currently applied in biomedical material science. Chemical crosslinking remains the most common, involving covalent bonding through agents such as EDC/NHS, DMTMM, or methacrylation followed by photo-initiated polymerisation. The choice of crosslinking method significantly affects antibacterial performance and cytocompatibility, indicating that optimisation is not only a structural task but also a biological one [29].

One of the chemical cross-linkers that has a notable potential in biomedical applications is genipin. It is anatural chemical compound found in Genipa americana and Gardenia jasminoides fruits, which is itself recognised for various beneficial therapeutic effects as well as its crosslinking capabilities [30]. Unlike synthetic crosslinkers such as glutaraldehyde, which can release harmful byproducts, genipin is significantly less cytotoxic [31], which makes it an especially promising crosslinker in a medical context. Genipin’s crosslinking properties are based on its ability to form covalent bonds with amine groups. One of the hallmarks of a successful genipin-mediated crosslinking is the production of a characteristic blue pigmentation which can serve as a visual indicator of the reaction process [30].

Overall, the crosslinking method is not merely a technical detail—it is a key determinant of clinical success in hydrogel-based wound treatments.

Protein modification and gene fragment design

α-S1-casein

Firstly, at the N-terminal end we added a 6xHisTag which consists of six histidines. This tag is necessary to purify protein using Ni2+ metal affinity chromatography [33].

After 6xHisTag, we added a Tobacco Etch Virus (TEV) protease site to ensure that it is possible to cleave off the 6xHisTag after protein purification, as it is not needed in the scaffold formation.

The same modifications were applied to the human-origin α-S1-casein sequence as to the bovine-origin one, designing the recombinant proteins with a 6xHisTag and TEV cleavage site in the N-terminal end of the sequence.

β-casein

Similarly to α-S1-Casein, 6xHisTag and TEV-site were added to the protein structure in the N-terminal end. Since we intend to use β-casein for crosslinking, we introduced 2 additional cysteines into the protein sequence to facilitate crosslinking. Cysteine residues are commonly used as crosslinking sites in protein engineering, and in many proteins they play a crucial role in maintaining the protein’s structural stability [34, 35, 51]. However, it is crucial that non-native cysteines are introduced in a way that does not disrupt the native structure and function of the protein by forming undesirable covalent interactions, promoting protein misfolding or aggregation. To minimise such risks, cysteines can be introduced into flexible or terminal regions of the protein of interest [36, 37]. Out of these considerations and keeping in mind structural properties of the intrinsically disordered β-caseins, we inserted the cysteine residues in either N-terminal or C-terminal ends of the proteins, which would allow us to evaluate the differences in our approach.

Primer design

To simplify the workflow, it was decided that instead of designing separate primers for each gene fragment, we would create one primer pair that would be universally applicable. We made sure that the primer sequence is not found in our gene fragments and there is no complimentarity, so there would not be any unwanted target amplicons. This was done by consulting last year s iGEM Latvia-Riga team and our secondary PI to check if that is theoretically possible, and after some research, we found a primer pair that met our requirements and was then used in the gene fragment design. Further on, when designing our gene fragments, recognition sites for our chosen primers, which from now on shall be referred to as universal primers , were added to the start and the end of the gene sequence.

Other primers will also be designed to test whether cloning of our respective gene fragments into our plasmid backbone based on individual situations.

Cloning method

Since our plan involved cloning our gene of interest into a protein expression plasmid backbone, we chose to use restriction cloning. After consulting with local researchers, a double restriction reaction was recommended to avoid the possible issues of insert rotation when ligating into the plasmid backbone. It was also a suggestion that led us to choose NdeI and XhoI restriction sites and use their respective restriction enzymes. Conveniently, both were available in the laboratory as part of the Fast Digest enzyme collection we already had on hand. Following the advice we received and the results of our research, we added the NdeI restriction site between the Fw_universal_primer and our target gene fragment and the XhoI restriction site between the Rv_universal_primer and our target gene fragment.

Vector of choice – pET-24a(+)

To produce α-S1-Casein, the coding sequence of the protein was inserted into the pET-24a(+) plasmid. Our choice of vector was based on its proven reliability in protein expression and the presence of all essential genetic elements for efficient protein production. pET-24a(+) carries a lac operator – it inhibits lac repressor in the bacterial host, preventing protein production before induction. Regulated by the lactose analogue isopropyl β-D-1-thiogalactopyranoside (IPTG) inducible target gene control, it enables controllable protein expression [38]. Additionally, pET-24a(+) carries a specific MCS – multiple cloning site – which contains our selected restriction sites NdeI and XhoI for insertion of the gene of interest, allowing us to successfully clone our designed gene fragments into the backbone. The plasmid backbone also contains the T7 tag, which is an optional N-terminal tag for detection with anti-T7 antibodies. In our constructs it is purposefully omitted through restriction with NdeI and XhoI. Besides the T7 tag, the backbone also contains an optional C-terminal 6xHis tag, which is not expressed in our construct, as our design relies on the N-terminal 6xHisTag.

Propolis extract

For research purposes propolis is usually used in the form of an extract. For analytical purposes this is usually done with organic solvents, like ethanol. Since we are interested in the antimicrobial properties of propolis, using an ethanol extract is not recommended since ethanol will inhibit microbial growth by itself, and the results will not be trustworthy. In order to test the antimicrobial properties of propolis, we found that PEG 400 solution could also be used to prepare a propolis extract [3]. To test the phenolic compounds present in the extract, the Folin-Ciocalteu method will be used with gallic acid being used for the calibration curve [39]. This extract can be used in antimicrobial tests since the concentration of PEG 400 in the final culture will be well below the threshold of having significant microbial growth inhibitory effects.

A 96-well plate assay was designed to make sure that our propolis extract has the antimicrobial properties that are reported in literature. To achieve this, growth patterns in liquid culture with E. coli and propolis extract were cultivated in 96-well plates. In order to obtain more reliable results, different propolis extract concentrations were used. Additionally, since a 20% PEG solution was used for propolis extraction, control repetitions were added, using 20% PEG solution at the same volumes as the propolis extract, to make sure we take into account any inhibitory effects from it.

Hydrogel crosslinking

To investigate the potential of casein hydrogels as a biocompatible material for controlled phage delivery, we will prepare casein hydrogels using commercially available casein and genipin as a chemical cross-linker, also attempting to incorporate selected bacteriophages into the internal structure of the casein hydrogel.

By systematically varying the concentrations of casein and genipin during hydrogel synthesis, we will create a range of hydrogel formulations with distinct physical properties. The hydrogels will be characterised for their rheological properties and biocompatibility. These tests will confirm their potential for use in treating burn wounds.

To produce casein hydrogels, we will follow a protocol that has already been described in literature. After producing the hydrogels with varying concentrations of casein and genipin, they will be tested for mechanical strength by evaluating their rheological properties in a partner institution – Baltic Biomaterials Centre of Excellence.

Bacteriophage integration

We chose three Enterobacteriaceae phages for proof-of-principle studies of phage survivability in hydrogels:

• Teseptimavirus T7 (T7 phage) - a virus with an icosahedral capsid, short tail with fibers and a lytic life cycle [40]. It belongs to the Autotranscriptaviridae family (previously Podoviridae) [41] and is commonly used as a model phage in microbiology [42];
• Tequatrovirus T4 (T4 phage) – a widely studied phage with a prominent role in molecular biology discoveries [43], from the Straboviridae family (previously Myoviridae) [44]. The virus has an icosahedral head and a contractile tail [40].
• Lambdavirus lambda (λ phage) – a popular model organism, notable for its ability to switch between lysogenic and lytic cycles [45]; from the Caudoviricetes class (previously Siphoviridae family) [46]. The phage has an icosahedral head and a long, flexible, noncontractile tail [40].

It is possible to get phages in the hydrogel by mixing the solution for the hydrogel with the solution containing the phage cocktail before the crosslinking reaction has occurred [47], [48].

Since the volume of bacteriophage solution added to the hydrogel will be relatively small, the phages should first be concentrated to ensure that their final concentration in the hydrogel is high enough to provide the desired antibacterial properties without rendering the protein concentration too low.

To do this, concentration by ultra-centrifugation will be performed to increase the amount of active phages in the volume unit of the solution. This process will also help us to change the media in which phages usually reside after filter purification.

A potential challenge when integrating bacteriophages with propolis extract is the reported antiviral activity of propolis [49]. This concern was also raised during one of our consultations with Dr. sc. ing. Zane Zelča. This points to the possibility that phages may not remain viable or effective in lysing host cells when in close proximity to propolis compounds.

To address this, we designed a 96-well plate assay. The assay evaluates the impact of different concentrations of propolis extract and phage culture on host bacteria that have already entered the exponential growth phase. By introducing propolis and phages at this stage, we ensure that any reduction in optical density reflects phage-mediated lysis, rather than propolis-induced growth inhibition during earlier bacterial growth phases. This approach will allow us to specifically assess phage survivability and lytic activity in the presence of propolis.

Cytotoxicity with
CellTox™ Green Cytotoxicity Assay:

After successful experiments and tests with each different component, to test if our product is not toxic to mammalian cells, we are planning to test the hydrogel material with a cytotoxicity assay provided to us by PROMEGA.

We plan to perform the cytotoxicity assays to ensure that our created material is not harming the cells. For that we would test the Hs68 cell line with CellTox Green Cytotoxicity Assay (catalog code G8741) from Promega using their protocol. When a cell dies it releases (because the membrane is impaired) biomarkers which this kit can detect in real time in exposures up to 72 hours [50]. Viable cells would emit lower fluorescence than non-viable cells.