To make our yeast extract-based skincare product rich in active substances, we built the oscillator in the Saccharomyces cerevisiae. The engineered strain contains a synthetic oscillator linking two genes, Hap4 and Sir2, which determine the lifespan. In WT yeast, the two genes inhibit each other. Thus, the down expression of either gene drives the yeast cell to the aging fate, either through DNA instability or mitochondrial dysfunction. The oscillator altered the regulation network between Sir2 and Hap4 by forming a new negative feedback loop. The oscillator led to a relatively sustainable oscillation and stable expression of Sir2 and Hap4. Thus, the yeast reached a relative stable metabolism, which in turn delayed aging. Thus, engineered yeast not only has a prolonged lifespan, but also has a potential to have better metabolic capacity compared to the WT yeast cell, providing the community a superior strain with oscillatory capabilities.

Adapted from Zhou Z. et al. [1], The diagram on the left shows how Sir2 and Hap4 interact in wild type yeast. The one on the right shows the rewired gene network by the oscillator. The arrows indicate promotion of expression while the arcs with bar on the head indicate inhibition.

For skincare ingredients, α-Bisabolol and ceramides are two potential chemicals we want to produce in yeast. However, the cytotoxicity property of these two substances poses a challenge for sustainable industrial production. The oscillator's role in gene regulation through negative feedback loops intrigued us, prompting us to investigate its potential for mitigating cytotoxicity in the production process. Consequently, we harnessed this oscillator to facilitate the production of ceramide and α-Bisabolol, aiming to enhance the safety and efficiency of our manufacturing methods... We have successfully integrated three critical enzymes, responsible for the biosynthesis of ceramide and bisabolol respectively, with Sir2 and Hap4. This strategic coupling allows us to regulate their abundance, ensuring a sustainable production process. (view the detail in project description).

In summary, we have effectively employed the oscillator to connect metabolic enzyme genes, which are pivotal for the production of cytotoxic substances. This innovative approach has not only reduced cytotoxicity but also introduced novel concepts for industrial-scale production of these critical compounds.

We improved and designed a microfluidic chip system to fix single yeast cells [2]. This chip features one hundred tiny units, which are called cell traps (view hardware for detail). The cell trap also features the “neck” that allows fixes yeast cells (mother cells) to pass through while restricting the movement of mother cells. The chip also enables us to support the by consistently perfusing the Yeast culture-medium (eg. YPD) to prevent growth compression from lack of nutrition. For every three cell traps, there is an submatrix to indicate position for tracking. The submatrix includes one positioning dot and five encoding dots. The positioning dot confirms the existence of the submatrix and provides absolute positioning information relative to adjacent “cell traps”. Encoding dots in each submatrix occupy five positions, corresponding to five digits in binary. Etched holes indicate “one”, while unetched places indicate “zero”, hence the 32 differentiated submatrix for the positioning of “cell traps”. Working with our software to recognize cells and process data, the hardware enables long-term observation of yeast cell as well as integrated follow-up data analysis.

Accessory plumbing system and chip jig are also designed to support the chip. The chip jig consists of the bottom plate and the midframe-cover assembly. The bottom plate is a thin plate to increase structural integrity while the midframe is the main load-carrying structure complex to reduce the stress on the PDMS material from pipes. Labs can also use the chip jig that helps to support the microfluidic chip, optimizing conditions to for better observation under the microscope. The Accessory plumbing system is user-friendly, simplifying operations and efficient in biological research.

We developed an image recognition model, LuminoSeg, to address the challenges encountered in the yeast oscillating system pattern research, particularly the handling of large datasets, the specific requirements in dot recognition, and the limited throughput of existing image processing software.

Our image recognition model can be divided into three parts, dot matrix segmentation, dot matrix recognition and single cell segmentation. Firstly, the images captured under a fluorescence microscope undergo image enhancement, then proceed through the YOLOv8, CNN, and the trained cellpose cyto3 model (view software for detailed information). The YOLOv8 model frames and selects each submatrix presenting in the image, as well as the adjacent three cell traps. Secondly, the selected submatrix and the three cell traps are segmented as a whole and processed within the CNN model. The CNN identifies the submatrix and labels the subsequent three cell traps. Subsequently, images are sent to our trained cellpose cyto3 model to segmentand recording the yeast cells within each cell trap. Finally, the average fluorescence intensity is “batch measured” by Imagej.

The model is capable of processing high-through data generated by the long observati. Additionally, the trained cellpose model enhances image segmentation accuracy, ensuring precise segmentation of yeast cells based on the specified size range.

Overall, LuminoSeg not only overcomes the limitations of existing methods but also offers a robust tool for researchers, significantly enhancing the efficiency and accuracy of their experimental workflows.

The AREA (Anticipate, Reflect, Engage, Act) framework, a cornerstone for guiding Responsible Research and Innovation (RRI), has been a staple among iGEM participants. To align with the beauty industry, we incorporate relevant ethical, cultural, and aesthetic considerations at every stage. For example, in the "Anticipate" phase, we could focus on the long-term cultural effects of beauty standards on product design. In the "Reflect" phase, we might look more closely at the current state of the beauty industry. The "Engage" phase would involve deeper interaction with stakeholders, not only those familiar with synthetic biology but also fashion influencers and social media communities. In the "Act" phase, the focus would be on quality control, legal regulations, ethical issues, and customer feelings.

iZJU-China's vision is to craft an AREA framework iteration that not only resonates in the present but also serves as an example for future iGEM teams delving into the fashion and cosmetics industry. We envision an ongoing dialogue where subsequent teams contribute to this framework's evolution, keeping pace with the industry's dynamic nature.

Below is an illustration of our refined AREA framework, specifically catered to the fashion and cosmetics village:

We also distributed our questionnaire about Synbio and cosmetics to individuals across various countries, age groups, and genders. During the first trial, we found that the understanding of this topic varies widely among individuals with different backgrounds. To design a more targeted questionnaire, we created a novel form of questionnaire that has branching logic and has question skip pattern. This design not only increased the specificity of questions for different populations but also enhances the accuracy of the data collected.

Below is an illustration of the question skip pattern:

1. Zhou Z, Liu Y, Feng Y, Klepin S, Tsimring LS, Pillus L, et al. Engineering longevity—design of a synthetic gene oscillator to slow cellular aging. Science. 2023 Apr 28;380(6643):376–81.

2. Li Y, Jiang Y, Paxman J, O’Laughlin R, Klepin S, Zhu Y, et al. A programmable fate decision landscape underlies single-cell aging in yeast. Science. 2020 Jul 17;369(6501):325–9.