Engineering

D is for Design

---

Design. The first part of the iGEM Engineering Success Cycle. A phase where everything looks perfect on paper, until you try to make it real. This was where our journey began. Our goal was simple enough: build a construct that could be delivered into cells in vitro and turn on antiviral protection when needed. Easy, right?

Inspired by existing gene therapy delivery systems, and advice from our Westlake professors, we developed a CRISPRa/TetON fusion system. By combining the two, we could fine-tune gene expression that promised quick activation when needed and silence when not, while avoiding the risk of leaky expression.

The system comprised three key components: dCas9-VP64 (the transcriptional activator), sgRNAs (our gene-targeting guides), and rtTA (the switch for the TetON promoter system). Simple ingredients. Complicated recipe. Together, they made up nearly 1700 amino acids, not even counting the three 100 nt sgRNAs. Delivering all that into a single cell would be no easy feat.

This was the first of many hurdles to come, but eventually, we decided to bite the bullet and made the decision to split the system: three separate parts, one for each major component. While it wasn’t perfect, this kind of setup meant more flexibility and control, letting us test individual and combined sgRNAs and explore how different ratios affected activation. The downside? Fewer cells would get the full trio, gene expression might vary more, and the cells would be put under serious metabolic strain. Regardless, every design starts as a compromise. What mattered was that we now had a clear direction for the next step of our journey: the Build Phase.

B is for Build

---

Of all the steps in our project, the Build Phase seemed like it would be the easiest, just basic bioengineering. Each component had its purpose, and all we had to do was assemble them according to the underlying mechanism. The TetON system operates through Doxycycline (Dox), which activates the rtTA protein. Once active, rtTA binds to a promoter known as TRE3G, enabling transcription of any downstream gene. This setup meant two important things: first, rtTA had to be expressed before Dox was added, and second, at least one of our other constructs needed to be under the control of a TRE3G promoter for inducibility.

As such, we built our system as follows. First, a plasmid containing rtTA, driven by an EF-1α promoter to ensure consistent expression and easy delivery. Then, a second plasmid containing dCas9-VP64, where we swapped the CMV promoter for TRE3G, making it Dox-inducible. And finally, the sgRNAs, delivered not as plasmids but as 2'-O-methylated ssRNA, to reduce metabolic burden and simplify delivery.

For this, we turned to lipid nanoparticles (LNPs), and after combing through literature, we settled on fairly simple composition, just DOTAP, a cationic lipid that complexes with nucleic acids, and cholesterol, to stabilize the particle. Previous studies had shown this mixture to be efficient for both mRNA and pDNA delivery, making it an ideal choice for our hybrid system.

With the plan in place, the workflow was straightforward, and for a brief, shining moment, it seemed like smooth sailing. We ordered the vectors, made the edits, and did simple cloning and isolation of the samples. What happened next, dear reader, was the problem.

T is for Test

---

To see if our LNPs actually worked, we started simple: an EYFP plasmid as a reporter for gene delivery efficiency. This was our first challenge. The paper we’d modeled our LNPs on reported robust gene expression even after 48 hours, but in our hands? Barely a flicker after 24 hours. Three separate transfections later, and our EYFP signal remained stubbornly minimal. It was a crushing contradiction to the goal of our system. Still, we held out hope: perhaps the problem was specific to EYFP, and our real construct would behave differently.

Armed with our understanding of hydrogel, LNP, and pDNA ratios, we moved on to test the actual system. We transfected cells and applied Dox induction across three concentrations and four 12-hour intervals, aiming to build a comprehensive map of how timing and dosage affected expression. The results? Uniform disappointment. dCas9 expression remained almost completely absent across all conditions. Western blotting confirmed that, despite a clear GAPDH reference signal in some blots, dCas9 was barely detectable. The experiment, as far as protein induction went, had failed. As a consequence, we were unable to even begin work with sgRNA testing, limiting our ability to effectively characterize the system in its full potential.

L is for Learn

---

So, what did we learn from all of this? Well, first and foremost, the major takeaway from this protocol is that our system is extremely bulky and hard to deliver. Although this was expected, we severely underestimated the extent to which is would inhibit our protein expression. Secondly, we learned that something in our system needs to change, most likely with Dox addition. Although it will usually take around 24 hours for expression to fully kick in, we should still be seeing some minimal expression at the 12-hour mark, but despite a clearly defined GAPDH band, we lacked any dCas9 expression. As such, we need to run some more thorough experiments in order to characterize our protein expression system, testing rtTA levels and dCas9 levels in a more refined way to examine the root cause of our issue.

Thirdly, and finally, we considered that with the difficulties of adding even just two plasmids, perhaps it would be better to create a barebones plasmid, synthesized almost fully from scratch, allowing us to pack the rtTA and dCas9 together. While this does go against our initial decision to split the system, we theorize that it may be necessary for us in order to achieve effective gene expression. However, the biggest take away is that our work lays the foundation for other groups to look into a fusion system like this, and further evaluate if it has the potential to become a genuine therapeutic.

Link to Part Registry Collection: https://registry.igem.org/collections/df8d0b8e-5baa-4573-986d-c8e995ada034