Our team at Project TaRNAdigrada is grateful to every iGEM team that came before us, with special love for those that worked with tardigrade proteins and extensively documented their parts and processes. Our work is built on the backs of those foundations. For instance, our choice of mammalian expression systems over an E. coli system was influenced by a previous iGEM team finding that tardigrade protein expression in E. coli under a strong promoter resulted in inclusion bodies, which complicates protein recovery and increases the frequency of misfolded proteins.

As such, we focused on situating our project in a broader, collaborative context, where future iGEM teams can pick up where we’ve left off with minimal hassle. While our project is centered narrowly on vaccine design applications, tardigrade proteins themselves are incredibly versatile, and their potential in research, industry, and healthcare settings has only begun to reveal itself. We hope future teams are just as excited as us to unlock that potential by iterating in different directions, with their launchpad being the foundations we’ve laid: our parts, expression system, and suggestions for future investigation.

We designed 10 genes spanning 5 major tardigrade protein classes implicated in stress tolerance, both natively in tardigrades and when synthetically expressed in eukaryotic systems. These genes constitute a representative selection of the diverse protective proteins present throughout tardigrades’ lifecycles and within different cellular components. Corresponding proteins from Ramazottius varieornatus and Hypsibius exemplaris were selected to investigate similarities and differences in protein structure and function between species. Each gene was codon-optimized for mammalian system expression, and fused with a 6xHis tag for purification through nickel affinity chromatography and a secretory signaling sequence to release protein into cell-conditioned media for easy collection.

The 5 protein classes in this collection include mitochondrial abundant heat-shock proteins (MAHS), cytosolic abundant heat-shock proteins (CAHS), and secretory abundant heat-shock proteins (SAHS), each localized to different cellular compartments and known collectively as tardigrade disordered proteins (TDPs) for their structural similarities and significant regions of disorder. The additional two classes are damage suppressor (Dsup) and Dsup-like proteins and late embryogenesis abundant proteins, both of which also sport intrinsically disordered regions.

This robust parts collection allows future teams the flexibility they need to customize their research. Teams may compare different proteins’ structure, aggregation, stress-protective properties, and more via expressing multiple genes across different cell lines; analyze the effects of expressing multiple proteins at once within a single cell line; or pursue evolutionary investigations by comparing orthologous genes between species. Each gene can be cloned traditionally for temporary protein expression or integrated into a viral or transposon system for host genome integration, and any required restriction sites can be added to each gene via PCR primer. The inclusion of BioBricks prefixes and suffixes on each gene allow for compatibility with other BioBricks-optimized parts and introduce multiple versatile restriction sites for future teams' cloning projects.

We invite all interested teams to learn more by visiting the respective parts pages.

Expressing and purifying proteins with significant disordered regions posed a challenge to our team, as intrinsically-disordered proteins tend to aggregate and activate cellular misfolding responses that result in degradation. We iteratively optimized our protein expression and purification protocols over time to troubleshoot and mitigate these challenges, which future teams can read about in detail on our Engineering and Experiments pages. Further optimization remains to be done; however, our optimization thus far has only succeeded in purifying the CAHS8 and SAHS2 proteins from R. varieornatus and H. exemplaris. We hope the efforts of future teams can develop more refined protocols that allow for efficient and economical recovery of all 10 proteins in our parts collection.

To make predictions about stress protection and nucleic acid binding for the proteins we were unable to collect, we modeled molecular interactions between each of the 10 proteins and sfCherry3C mRNA, both with and without the presence of vaccine-like buffer salts. These models can be found on our results page and were generated using AlphaFold 3. Future teams may find them useful in studying protein aggregation, comparing protein interactions with different biomolecules (how might the same proteins arrange differently around DNA or dsRNA?), and determining the extent of mRNA secondary structures’ impacts on protein interaction.

This iteration of AlphaFold does not allow for carbohydrates to be modeled, so these predictions do not account for the large presence of sucrose, a stabilizing sugar found in the BioNTech/Pfizer COVID-19 vaccine, in our experimental trials. Future teams are encouraged to investigate how carbohydrates may impact mRNA-tardigrade protein interactions and whether they affect the proteins’ capacity to protect against degradation.

These models, found on our Engineering page, can be useful in studying protein aggregation and comparing interactions with different biomolecules. Future teams are encouraged to investigate how other molecules, like stabilizing sugars, may impact these interactions.