Description

The Problem

There is a pressing need for systemic change in the fashion industry. Fast fashion has made it easy for us to keep up with the latest trends, allowing for constant self-expression through clothing. However, behind the appeal of cheap and trendy fashion lies a serious environmental problem. Fast fashion enables consumers to express their individuality through constantly changing clothing, yet the cost to the environment is massive. Each year, 92 million tons of textiles end up in landfills, while the fashion industry accounts for up to 10% of global CO₂ emissions (Niinimäki et al., 2020). This „buy and toss” mentality, driven by quick production cycles and low prices, creates an enormous amount of waste.

The production process itself is a significant environmental offender. The fashion industry is the second-largest consumer of water, utilising approximately 1.5 trillion litres per year and contributing around 20% of industrial water pollution from textile treatment and dyeing. Synthetic fabrics like polyester are cheap to produce but extremely harmful to the environment. Made from petroleum, polyester releases large amounts of CO2 during production and does not decompose easily, contributing to the growing problem of microplastic pollution (Niinimäki et al., 2020). Moreover, the complex global supply chains used to manufacture these clothes result in high carbon emissions. As supply chain expert Emilija Suvalova points out „The substantial carbon footprint from transporting raw materials, semi-processed goods, and finished products across the globe emphasises how the complex supply chains in fast fashion significantly contribute to environmental degradation.”

Professor Dr. Morone, a professor of Economic Policy at Sapienza University of Rome suggests that the solution lies in adopting more sustainable materials and production methods. Shortening supply chains, producing locally and using innovative technologies could drastically reduce the industry‘s impact on the environment. Professor Morone noted, „Educating consumers through tangible evidence of our product’s benefits is crucial, as many are unaware of their environmental impact demonstrating the advantages of sustainable practices over traditional methods can drive behaviour change.”

To truly reduce the fashion industry’s environmental footprint, we must rethink the way clothes are made, sold and consumed. The future of fashion needs to prioritise sustainability over speed because the planet can no longer afford the hidden costs of our fashion industry.

Our Project Description

Having this in mind, we reached out to Prof. Dr. Markus Pauly, head of Plant Cell Biology and Biotechnology Institute at Heinrich Heine University Duesseldorf, who also later became our Principal Investigator. He introduced us to Komagataeibacter xylinus (K. xylinus), a bacterial species known for its cellulose producing abilities. Bacterial cellulose (BC) pellicles hold great potential as a bio-based material with different applications. Previously, researchers used strains of Komagataeibacter to produce cellulose for industrial uses. Works of Prof. Tom Ellis from Imperial College London (Walker et al., 2024) and iGEM Vilnius 2023 and KCIS-Xiugang-Taipei 2023 are clear examples of this. Others have tried to improve the mechanical properties by infusing hemicellulose into growth media (Mikkelsen & Gidley, 2011). The Property Testing subgroup aimed to tackle a few questions: to what extent does hemicellulose such as xyloglucan improve mechanical properties? At which concentration do we see the optimal performance and does this influence the chemical composition of BC pellicles? To answer this question, we consulted experts in property testing: Dr. Amanda Staudt of DWI - Leibniz Institute for Interactive Materials. We gained insights into different types of mechanical testing and it was later conducted at Niederrhein University of Applied Sciences. We examined the change in mechanical properties, from different growth media, washing and drying procedures. We discovered that incorporating xyloglucan into the BC mats increased tensile strength at lower concentrations, though higher concentrations reduced elasticity. We also found that treating the mats with NaOH solution and drying in a cabinet at 50oC improved tensile strength. To conduct these tests on-site, we built a self-designed property testing machine, allowing us to perform mechanical tests such as tear and tensile strength tests. While most mechanical tests on BC pellicles were done at Niederrhein University of Applied Sciences, the machine provided data on cotton samples (see Property Testing). We also wanted to know the chemical properties of our mat and wanted to know if adding xyloglucan to media really had an effect. High performance anion exchange chromatography (HPAEC), experiments showed increased xylose and glucose for samples prepared with xyloglucan. We also tried to extract xyloglucan directly from tamarind seeds. However, after a long and laborious process, only a very low yield was collected.

Using synthetic biology, we wanted to produce xyloglucan in yeast in a higher amount. Contacting a researcher, PhD Ronja Ronja Catharina Immelmann we learned, which difficulties research in the area of hemicellulose production in yeast has already faced. Based on that information we worked on a pathway, to express genes involved in hemicellulose production, in Saccharomyces cerevisiae. We conducted multiple transformation evaluations and digests, to construct transcriptional units, which are part of this pathway.

Created with BioRender.com

We found out that the part BBa J435281 which we planned to use as a backbone had some errors. Sadly, we noticed that the overhangs of the OYC-eforRed-dropout plasmid backbone (BBa_J435281) do not fit with the insert overhangs for level 1 assembly.

To clarify whether the plasmid sequence was actually incorrect, or only the annotation was faulty, we transformed the parts into E. coli, miniprepped them and had the plasmid sequenced twice. (See Backbone fixing study) It appears that the uploaded sequence is correct and thus the plasmid sequence is actually incompatible with level 1 assembly, making the whole Open Yeast Collection unusable. This error appears also using the original plasmid backbone sequence deposited on the Freegenes website. We proceeded to fix the backbone by running two self-ligating Golden Gate reactions to achieve a suitable level 1 construct afterwards. As this was not foreseen before planning the lab work, at this advanced point in time in relation to the project, we decided the completion of the hemicellulose aspect of the project was not realistic and would cost too many resources, so the focus was set onto the other aspects of the project.

After discussing with Prof. Dr. Danny Ducat, we learned more about co-culture. We aimed to make stable co-culture between our engineered K. xylinus and S. cerevisiae, work by investigating their growth behaviour and optimising ratios between both of them to achieve a consistent product. (See co-culture)

In addition, we wanted to express the cellulose binding domain also used for binding the chromoproteins to our mat on the yeast cell wall directly, hypothetically increasing S. cerevisiae concentration in our mats because the engineered yeast binds to the mat directly. We hope this would potentially further optimise incorporation of hemicellulose and chromoproteins in our BC mats.(See sticky yeast)

S. cerevisiae is not only useful for hemicellulose production but also for expressing dyes to colour our fabric. In our meeting with Sebastian S. Cocioba we had the opportunity to learn about chromoproteins. When mixing them in the right ratios using inducible promoters (Park et al., 2023) we can recreate a colour spectrum similar to the one found in a printer. The colouring system used in printers is also known as the CMYK colour scheme, since it uses the colours Cyan, Magenta, Yellow and “Key” (Black). CMYK is a subtractive colour model, meaning that when none of the colours is present the displayed colour is white while combining every colour results in a black colour being displayed. Since we also learned in the meeting that chromoproteins denature shortly after being expressed, we decided that pigments would be a much more stable and viable colouring agent for the future. We believe that future iGEM teams can take on the challenge of finding the perfect pigments and engineering a suitable pathway for them.

Created with BioRender.com

But how sustainable is our process? From our conversation with Global Organic Textile Standard, we found that sustainability models, such as life cycle assessment (LCA) can help in making such a judgement. We talked to an LCA expert, Dr. Pieter Nachtergaele, and conducted LCA experiments to determine the yield. However, the predicted yield was too poor to make our process economically viable. One way to overcome this is by lowering the substrate cost. We got in touch with iGEM Dresden, whose project focused on hydrolysing clothes to produce sugar substrates. If this sugar is fed to K. xylinus to produce new clothes, we have achieved a closed carbon cycle - solving both economic and sustainability issues. Another way is to make a metabolic model, which helps us understand different flux modes in K. xylinus. From this, different strategies to maximise cellulose flux can be devised. We were able to identify candidate reactions that can be knocked out and knocked down to optimise cellulose flux close to its theoretical maximum.

One of the major difficulties in researching and working with K. xylinus currently is the constitutive bacterial cellulose production which while quite performant also makes it difficult to properly cultivate in higher cell concentration, as K. xylinus tends to undergo spontaneous non-cellulose mutation during shaking cultivation (El-Gendi et al. 2022) reducing the viability of the culture to produce consistent mats unless expensive cellulase is added. In addition, we were also worried that to reach better incorporation of hemicellulose and dye the bacterial cellulose needs to be regulated. This is why we were interested in an inducible system and successfully generated a K. xylinus a knockout strain and an inducible strain. With that we could eventually reduce the need for cellulase necessary to increase cell count and optimise carbon intake by targeting knockouts or overexpression for different genes and or clone an inducible promoter in front of the gene we want to up or down regulate, giving us even more control over the bacterial cellulose synthesis. Through this we hope to push the scale in our favour to achieve a truly sustainable product.

Detailed Property Testing Description

Abstract

The goal and purpose of the property testing subgroup was to get information on how our produced textile behaves compared to commonly used textiles like cotton and polyester. For this, we tested the mechanical and chemical properties of the bacterial cellulose (BC) mats we created under varying conditions. In the context of composition analysis using HPAEC, we could prove that xyloglucan was implemented in the bacterial cellulose mat when growing it in media with 0,5% (w/v) of xyloglucan (XG) or washing it with a 0,5% (w/v) XG solution. There, the effect of growing the mat with XG was greater than washing it with a XG solution.We further hypothesized that the intercalation of hemicellulose in the BC mat would increase flexibility and tensile strength. This goal was partly achieved, as we found that a xyloglucan concentration of 0,25% increases the tensile strength, but reduces the elasticity. For higher concentrations, we found that both tensile strength and elasticity are reduced. Furthermore, we tested the effects of different washing and drying methods on the mats’ properties and found that washing the mat with a 1% (w/v) NaOH solution and drying it in the drying cabinet increases the tensile strength. In addition, we built a simplified property testing machine that can perform tear tests, as well as tensile tests, and may help future teams that need such tests but lack funding and/or contacts with professional machines.

How Property Testing Started

"There was an idea, [...]. The idea was to [...] [evaluate and compare the textiles we grow to commonly used textiles like cotton and polyester] to see if they could become something more. See if [...] [we could improve them to be more durable and eco-friendly]"[1]. For this, it was crucial to know how durable our textile was in the first place, meaning we first had to agree on what properties we would deem essential to evaluate these in the form of tests. Searching through literature describing the art of property testing on textiles [2], we first collected properties that we regarded as crucial. These included mechanical properties [3] like abrasion, tensile strength, tear strength, axial torsion, and puncture resistance, chemical properties like composition, washability, and flammability [4], and also properties regarding the textile's general performance [5] like water repellency, breathability, and UV absorption. To evaluate our data we wanted to apply mathematical modeling. The goal was to predict the property outcomes from cultivation variables and find the "sweet spot" in terms of cultivation parameters and post-treatment. In terms of mathematical models, we wanted to make use of linear regression. To reduce the tested parameters for easier data comparison, we later reduced the tested properties to abrasion, tensile strength, tear strength, and chemical composition. We then researched the necessary tests. We chose the Martindale method [6] (ISO 12945-2:2020) to test the abrasion of our textiles and the Trouser Tear (Single Rip) Test [7] similar to the one described in ISO 34-1 [8] for testing the tear strength. For tensile strength, we tried to replicate tensile tests as described by Instron [9], for example, as it was hard to find a specific method to go by in this area. Some adjustments had to be made in mechanical testing as big pieces of our cloth were harder to make due to limited space in the incubator. That's why we, for example, chose the single rip method [7] (ISO 34-1 [8]), which is usually used for plastics and rubber to address the tear strength instead of the Tongue Tear (Double Rip) Test [7] (ISO 13937-2 [10]), which is the one that is usually used for textiles but needs larger samples. For composition analysis, we chose to do high-performance anion-exchange chromatography (HPAEC). The problem that arose then was that we didn't have the machines necessary for the mechanical testing at our university. That's why we searched for institutes at other universities with specialized machines. Searching for an institute with the necessary machinery, we contacted Amanda Staudt from the DWI - Leibniz Institute for Interactive Materials in Aachen [11], who helped us work out our ideas for mechanical and chemical property testing. She also advised us on what to look for in cultivation and mathematical modelling. From her, we also got the necessary measurements and the number of replicates for our mechanical tests (Abrasion, tensile strength, and tear strength). Then the Institute of Textile Technology at RWTH Aachen [12], which we contacted shortly after, told us that we could test to chemically treat our BC mats to increase their hydrophobicity (something that would be useful if we for example wanted to make outdoor clothing with our mats). They also gave us advice on testing the purity of our products, as their research on their LignoTex [13] project indicated, that removing impurities like lignin and hemicellulose results in stronger fibres (something that was in some sense contrary to our project of improving the flexibility of our BC mats by adding Hemicellulose). Furthermore, they advised us to try solubilising our Bacterial Cellulose to convert it into fibres and yarns as they could then be interwoven with other fabrics to get the best of both worlds. The problem there though, as we later found out, was that the hemicellulose in our textile would degrade in the process, which was why we scrapped that idea. While preparing our first round of BC mats to test in Aachen, we also built our own hardware. This act was inspired by the advice of one of our PIs, Prof. Guido Grossmann, who suggested that we could perform simplified versions of the mechanical property tests ourselves for easier and faster data access. What his suggestion led to, was the creation of our self-built property testing machine [14].

What we did

Once the first small bacterial cellulose mats had been grown in the lab, our first goal was to make bigger versions that could be used for mechanical testing. With this, we first wanted to explore their potential for the future of our project and get used to handling them right. This way, we grew our first round of mats in a baking dish and a 15x15 cm glass container at room temperature due to limited incubator space [15]. In the process of growing the mats, we could already see that they grew better on YPD Media [16] than on DSMZ [17] Media. We then tested how they would react to autoclaving and washing as we feared that they would not survive the harsh conditions of the autoclave [18]. The extraction of the mat grown on YPD before autoclaving can be seen in picture 1. The autoclaving process did not destroy the mats. The only visible change was that the edges appeared to be a little bit dried out. After later also extracting, washing, and drying the mat grown on DSMZ [19], we could see that it was a lot thinner and less robust than the one grown on YPD, as it already ripped apart in the cleaning process. On the flip side though it wasn’t stained by the media like the one grown on YPD and was more transparent (Picture 2). In the meantime, a big bacterial cellulose mat was grown for an extended time to see if it would change its properties [20]. The result was that the mat got thicker and stiffer with the extended cultivation time. The next step was to find out how the cultivation parameters could be perfected for the best properties of our mats [21] [22]. Following a paper [Mikkelsen et al., 2011; 23] on growing bacterial cellulose mats with hemicellulose, we grew these mats on HS media [24]. The media was then modified by switching from glucose to glycerol, as we wanted our carbon source to be more sustainable. If glycerol worked, we could grow our mats on waste products like crude glycerol in the future. To find the right amounts, we grew the mats at glycerol concentrations reaching from 1% to 4%. The other half of the experiment was to test the effects of different concentrations of the hemicellulose xyloglucan on the mechanical properties (in this case abrasion & tensile strength), with concentrations ranging from 0,1% to 0,5%. What we also wanted to find out with composition analysis via HPLC, was, if the hemicellulose would be incorporated in the BC mat and if the amount would increase with increasing concentrations in the media. Both plans for this experiment were combined by either making the glycerol or the xyloglucan concentration the variable with the other variable being at the maximum concentration for the experiment. While cleaning and drying the mats, we could already make out differences in the mats' properties like thickness and how easily they would break down, which was what happened to some of the mats to different degrees, making 3 of them unextractable. In general, samples grown in lower xyloglucan concentrations (0.1~0.3%) exhibited thinner and flimsy attributes than those grown in higher xyloglucan concentrations (0.4, 0.5%). The problem that then faced us, was that DWI cut our cooperation, meaning that we couldn’t do mechanical tests on our tests as planned. As the self-built property testing machine also wasn’t finished at that point, we again searched and reached out to other institutes to perform our mechanical tests. After some time we got in contact with Marc Neumann at Hochschule Niederrhein (HSN), who generously offered us to do these tests at their institute. We were also offered to perform our mechanical test at Hochschule Düsseldorf (HSD), which we didn't get to though, due to scheduling problems [25]. The first time we went to HSN, we performed Martindale Abrasion tests and Tensile Strength tests on the mats from the experiment with differing amounts of glycerol and xyloglucan as well as on mats from other experiments (Pic. 3) Additionally, we tested a cotton cloth we chose as our standard of comparison and pieces of Aramid/Kevlar.

Learn

We learned that it extremely complicates the cloning and screening process and success if one of the L0 parts has the same resistance marker as the L1 you are aiming for. With that knowledge in mind, for all future reactions we switched to BBa_J428092 as a terminator because it has a chloramphenicol resistance and it does not pose the same problems as the previously mentioned terminator.

Future Applications

In our iGEM project, we developed many tools that are necessary to push forward innovation in using bacterial cellulose as a material alternative to those that rely on conventional synthetic-dependent processes. Ranging from the hardware that can be used to conduct property tests on-site, sticky yeast for co-culture, colour software to predict the colour of the textile, and genetic toolkits to implement genetic modifications: future iGEM teams can try to implement different aspects of our project that were unfinished. Namely, knocking out those candidate genes and using our inducible system to validate metabolic model prediction and validate for flux to ensure a comparable cellulose yield in engineered K. xylinus strain.

References

Niinimäki, K., Peters, G., Dahlbo, H., Perry, P., Rissanen, T., & Gwilt, A. (2020). The environmental price of fast fashion. Nature Reviews Earth & Environment, 1(4), 189–200. https://doi.org/10.1038/s43017-020-0039-9

Walker, K. T., Li, I. S., Keane, J., Goosens, V. J., Song, W., Lee, K.-Y., & Ellis, T. (2024). Self-pigmenting textiles grown from cellulose-producing bacteria with engineered tyrosinase expression. Nature Biotechnology. https://doi.org/10.1038/s41587-024-02194-3

Mikkelsen, D., & Gidley, M. J. (2011). Formation of Cellulose-Based Composites with Hemicelluloses and Pectins Using Gluconacetobacter Fermentation. In Z. A. Popper (Ed.), The Plant Cell Wall: Methods and Protocols (pp. 197–208). Humana Press. https://doi.org/10.1007/978-1-61779-008-9_14

Park, J. H., Bassalo, M. C., Lin, G.-M., Chen, Y., Doosthosseini, H., Schmitz, J., Roubos, J. A., & Voigt, C. A. (2023). Design of Four Small-Molecule-Inducible Systems in the Yeast Chromosome, Applied to Optimize Terpene Biosynthesis. ACS Synthetic Biology, 12(4), 1119–1132. https://doi.org/10.1021/acssynbio.2c00607

El-Gendi, H., Taha, T.H., Ray, J.B. et al. Recent advances in bacterial cellulose: a low-cost effective production media, optimization strategies and applications. Cellulose 29, 7495–7533 (2022). https://doi.org/10.1007/s10570-022-04697-1