Results

Following the receipt of the cellulose-producing bacterial strain K. rhaeticus, we conducted cellulose production tests in various containers using YPS, YPG, and HS media. The highest cellulose yield was observed in YPG medium, as indicated in the referenced article (Gilbert et al., 2021). According to the article, K. rhaeticus does not grow in YPS since it cannot metabolize sucrose. These trials confirmed the capacity of the K. rhaeticus strain to produce cellulose (Figure 1). This strain was selected for subsequent experiments.

Co-culture protocol:

We chose to replicate the protocols from the article Living materials with programmable functionalities grown from engineered microbial co-cultures between the selected strain of K. rhaeticus and the wild-type (WT) S. cerevisiae BY4741 strain. Various containers were tested, including 60 mm Petri dishes, 90 mm Petri dishes, 24-well plates, and rectangular 4-well plates. Different media were also tested in accordance with the article's experiments, specifically YPS and YPG media. According to the article, YPS is the preferred medium for producing the patches because the bacterium requires an enzyme produced by the yeast to metabolize sucrose. Therefore, in YPS, the bacterium is less likely to outcompete the yeast, which is essential for its growth.

The YPS and YPG media were inoculated with K. rhaeticus and S. cerevisiae BY4741 according to the article's protocol: a mixture of 1/50 of YPS or YPG from a K. rhaeticus culture at an OD of 2.5, combined with 1/100 of YPS or YPG from a S. cerevisiae culture at an OD of 0.01. These mixtures were incubated for three days at 30°C, as recommended in the article.

Cellulose patches were obtained under all co-culture conditions (across all containers and media). However, the cellulose patches were more compact and resilient when produced in YPG medium (Figure 2). This result aligns with the observations from the article, as the bacterium cannot metabolize sucrose independently and relies on the yeast for growth in YPS, resulting in slower development and cellulose production. Conversely, the bacterium does not require the yeast for growth in YPG, leading to faster growth and cellulose production compared to YPS.

Passaging

Subsequent passages of the co-cultures were conducted according to the article's recommendations to determine whether it was possible to produce functionalized cellulose from the same co-cultures over several weeks. Given that the project's objective is to produce functionalized cellulose patches, it would be advantageous to generate an initial co-culture and then renew only the medium every three days without having to restart the cultures for new patches. After three days of co-cultivation, fresh YPS and YPG media were inoculated with a volume of co-culture corresponding to 1/100 of the final medium volume. These passages were incubated for three days at 30°C.

One of the goals of these passage tests was to verify that the quality of the patches did not decline over successive passages, allowing for extended use of the patches. A total of three passages were performed following the aforementioned protocol.

During the first passage, the texture of the cellulose patches was similar to that of the initial co-culture. However, from the second passage onward, the texture of the patches produced in YPG changed noticeably. Specifically, the patches lost rigidity, opacity, and firmness, becoming soft and translucent, with very little material remaining (Figure 3).

In contrast, the quality of the patches in YPS did not change significantly across passages, resulting in cellulose that remained soft and thin, making it difficult to handle.

In conclusion, the quality of cellulose patches is superior when the co-culture is performed in YPG, but this is only observed during the initial co-culture. A possible explanation is that the culture was not sufficiently homogenized before being introduced into the fresh medium.

The goal of this project is to produce functionalized cellulose patches through the co-culture of K. rhaeticus and recombinant S. cerevisiae expressing YFP and bioglue proteins. After generating and validating the genetic constructs for YFP and bioglue in the genome of S. cerevisiae, two clones have been selected (ScpD1 and ScpD7) we proceeded to test co-culture with K. rhaeticus. The aim was to confirm that cellulose quality would not be compromised by co-culturing with recombinant yeast. Cellulose patches were successfully produced (Figure 7), displaying similar texture and appearance to those obtained from wild-type yeast co-culture (Figure 2). In conclusion, cellulose patches retain their quality regardless of whether K. rhaeticus is co-cultured with wild-type or recombinant yeast. However, the cellulose produced was not yellow, and further analyses are required to assess the adhesive properties of the cellulose.

To preserve the cellulose patches produced from the co-culture, various drying methods were tested. Initially, absorbent paper was used to wrap the cellulose. However, this method made it difficult to detach the patches from the paper. Subsequently, following the article's recommendations, we employed the "sandwich" method. In this approach, the cellulose was encased in parafilm and then surrounded by absorbent paper. When the absorbent paper was changed after 24 hours, the cellulose was dry after 48 hours. However, detaching the cellulose without deforming the parafilm proved challenging, often resulting in the tearing of the patch.

Ultimately, the most effective and straightforward technique for recovering the patches was the sandwich method (Figure 4), which involved encasing the cellulose with grease-proof paper, as described in the article, along with absorbent paper.

Additionally, the drying of the cellulose was conducted to weigh the patches, allowing us to determine the weight variations associated with the different co-culture media tested.

The production of a functionalized cellulose patch through the co-culture of K. rhaeticus and recombinant S. cerevisiae involves the integration of genetically modified yeast within the patch. Given the necessity to avoid the use of cellulose containing these genetically modified microorganisms, UV sterilization of the produced cellulose was tested. It is crucial to identify a sterilization method that does not degrade the recombinant proteins, ensuring their continued functionality.

To quantify the yeast present in the cellulose patches, the patches were washed with phosphate-buffered saline (PBS) and subsequently immersed in a 2% cellulase PBS solution to degrade the cellulose. This mixture was left at room temperature until complete degradation of the patches occurred (24 to 48 hours). The yeast present in the digested patch solution was then enumerated using a Malassez cell. However, this counting method does not distinguish between dead yeast, resulting from sterilization, and viable yeast.

To further assess yeast viability, the digested patch solutions were centrifuged, and the pellet was diluted before being plated on YPG medium and incubated at 30°C. The count of yeast colonies obtained represents the cells that survived UV sterilization.

Preliminary results indicate that UV sterilization effectively reduces the number of yeast cells within the cellulose patches but does not eliminate them entirely. Future investigations should also focus on evaluating the functionality of the recombinant yeast proteins post-sterilization.

Perspectives for sterilization

The ultimate objective of our product is not to release genetically modified yeast into the environment, but rather to provide sterile, functionalized cellulose patches. To achieve this, various sterilization methods have to be tested to ensure the elimination of any genetically modified organisms (GMOs) without compromising the functionality of the proteins integrated within the patch.

Current efforts focus on UV sterilization, which has shown promising results, but other approaches, such as gamma irradiation, could also be explored to sterilize larger volumes of cellulose patches. Gamma irradiation, for instance, is commonly used in medical and cosmetic industries for its ability to treat entire palettes of products in a single process.

Moreover, a recent study (Chong et al., 2024) found a novel method to kill GMO S. cerevisiae: the kill-switch method. This method is composed of an ER destabilizing domain (ERdd) fused to an essential gene (spc110) in the chromosome of S. cerevisiae. ERdd protein is stabilized with estradiol in the cytosol of the cell and allows the survival of the yeast. If no estradiol is added to the extracellular medium, then ERdd protein fused to Spc110 protein will be destabilized and so will be destroyed by the proteasome of the cell, causing the cell death. The kill-switch method could be a promising device to produce a sterile functionalized cellulose. Further investigations into this and other methods are necessary to optimize the sterilization of cellulose patches for broader industrial applications.

BioGlue and chromoprotein ideas

To impart adhesive and visual properties to cellulose, we employed a biological system capable of producing proteins that bind to and functionalize cellulose. The selected system is Saccharomyces cerevisiae, a yeast widely used for heterologous protein production. Although several experts suggested using alternative yeasts, such as Pichia pastoris, we adhered to the protocol described in a reference article (Gilbert et al., 2021), as it provided a well-characterized and functional co-culture system.

The primary objective of this study is to enable S. cerevisiae to co-produce an adhesive protein (bioglue) and a colored protein, both designed to bind cellulose, rendering it both visually appealing and adhesive. We first identified adhesive proteins in nature, selecting MaSp1 (spider silk protein) and Cp19k (barnacle protein), which, when fused, exhibit enhanced adhesive strength, making them ideal for our bioglue (Ye et al., 2023).

After selecting the bioglue, we searched for colored proteins, or chromoproteins. Based on findings from the iGEM Uppsala 2014 team, we identified fwYellow (BBa_K1033910), a yellow chromoprotein suitable for attracting pest insects. At this stage, we had identified the proteins needed to functionalize cellulose. The next step involved optimizing S. cerevisiae to produce these proteins and ensuring their binding to cellulose.

Protein secretion and binding

To achieve this, we utilized a cellulose-binding domain (CBD) system described in a reference article (Gilbert et al., 2021). This domain binds directly to cellulose and, when fused to a protein of interest, facilitates its attachment to the cellulose surface.

Finally, we investigated protein secretion strategies in S. cerevisiae, selecting the well-documented alpha-factor signal sequence, which enables the secretion of fused proteins (O’Riordan et al., 2024). This ensures that both adhesive and colored proteins are secreted into the surrounding medium, promoting their interaction with cellulose.

Once all target sequences were identified, they were fused using the SnapGene software as follows:

Final BioBrick construction

We then identified a viral peptide called P2A, which enables the generation of polycistronic RNAs in eukaryotic organisms (Mukherjee, M. & Wang, 2023, Liu et al., 2017). This approach was particularly suited to our project, as it allows the expression of two constructs under a single promoter. Consequently, we fused the two constructs using the P2A system and incorporated a strong constitutive yeast promoter, the GAP promoter:

Once our construct was finalized, the goal was to integrate it into S. cerevisiae to enable the expression of our two recombinant proteins. To achieve this, we added 500 bp homologous regions flanking the construct, targeting the URA3 gene locus. Additionally, we included the URA3 gene itself to facilitate the selection of transformants:

Transformation of plasmid D in S. cerevisiae

However, achieving the final plasmid, referred to as plasmid D, required several construction steps, which are summarized on the Engineering page. Once the plasmid was assembled, it was linearized and transformed into S. cerevisiae, resulting in successful transformants.

To confirm the correct integration of the fragment into the yeast genome, eight transformants were selected, and colony PCR was performed:

For the eight selected clones, a band of the expected size was observed, confirming the integration of the fragment into the S. cerevisiae genome. This result demonstrates the efficacy of plasmid D in integrating into the S. cerevisiae genome and its capability to insert heterologous genes.

While the genes have been successfully integrated into the genome, it remains uncertain whether they are being expressed and yet to produce a protein. To clarify this, further analyses are necessary. Consequently, positive PCR clones were isolated on selective plates, and additional assessments were conducted to validate the various mutant yeast strains.

SDS-PAGE

For each clone, it was necessary to determine whether the proteins of interest were produced and subsequently secreted by the yeast. To achieve this, first a SDS-PAGE was conducted for two clones confirmed by PCR (ScpD1 and ScpD7) under three different conditions: crude extract, supernatant, and precipitated supernatant.

No bands were detected in the supernatant or concentrated supernatant from cultures of clones 1 and 7 of S. cerevisiae transformed with plasmid D. Similarly, no bands were observed in the WT condition. Additionally, no apparent differences in protein bands were seen between WT and recombinant yeast for protein extraction from the cell pellets. In conclusion, no proteins were secreted in sufficient quantities to be detected by SDS-PAGE. A western blot should therefore be performed to specifically verify the production and secretion of YFP and bioglue.

Protein detection in supernatant via Western-Blot

Thus, a Western blot was conducted on the clones ScpD1 and ScpD7 under the same three different conditions as the SDS-PAGE: crude extract, supernatant, and precipitated supernatant. In each Western blot analysis, the presence of YFP, the product of the fwYellow gene, was detected using anti-GFP antibodies present in the plasmid D integrated into the S. cerevisiae genome.

The expected size for the alpha-factor-YFP-CBD fusion protein present in the plasmid D integrated into S. cerevisiae is 49.2 kDa. A band of this size was observed for clones 1 and 7 (YFP1 and YFP7) in the precipitated supernatant. This result suggests that the fusion protein is indeed expressed and secreted by the yeast. Furthermore, it indicates that the P2A system is functioning correctly, allowing for the translation of two recombinant proteins from the same messenger RNA.

However, numerous non-specific bands were observed in the precipitated supernatants, necessitating the performance of additional replicates of this experiment to confirm these results.

The objective of the fwYellow gene was to produce a yellow-colored protein that would be visible to the naked eye in order to color our trap. This protein appears to be expressed and secreted by the yeast, as previously observed in the Western blot analysis. However, no yellow coloration was detected, either visually on the yeast colonies or through fluorescence (data not shown). Thus, the product of the fwYellow gene is not functional when expressed in yeast. To address this issue, it will be necessary to identify other functional yellow chromoproteins in yeast and clone them as replacements for fwYellow.

To confirm the results observed in the Western blot using anti-GFP antibodies, this manipulation needs to be repeated.
In this study, we were unable to determine the expression, production, and secretion of the α-factor–BioGlue–CBD fusion protein because we did not have the necessary antibodies available.
We could utilize anti-CBD antibodies to detect our two proteins of interest: α-factor–YFP–CBD and α-factor–BioGlue–CBD. This would allow us to confirm the presence of each protein in the supernatant and further validate the P2A and α-factor systems.
To impart colored characteristics to cellulose, it will be necessary to identify new functional chromoproteins in yeast.
To address the issue of low protein production, one possible solution would be to use a stronger promoter to drive higher levels of gene expression. Replacing the current promoter with a well-characterized, high-strength promoter, such as the PGK1 or TEF1 promoters, could enhance the transcription of the fusion protein. This would likely increase protein production, making it easier to detect and evaluate its functionality in yeast. To confirm the effectiveness of our BioGlue, adhesion tests on cellulose should be conducted using a dynamometer.