Engineering Success

Design

Our team started brainstorming ideas in November 2023 the topic of hemophilia caught our interest early on, vesicle delivery was also an early concept, but the exact design of our project changed a lot during our literature research as we were learning more and more about hemophilia and the different types of vesicle delivery. Our project design started to crystallize at the beginning of march when we decided to use ionizable lipid nanoparticles (LNPs) for mRNA delivery. This method has already been well-researched in the last decades. Our design is largely based on a paper called “Engineered ionizable lipid nanoparticles for targeted delivery of RNA therapeutics into different types of cells in the liver” by Kim et al. 2021 This article gives instructions for formulating LNPs that can deliver siRNA as well as mRNA to specifically the liver sinusoidal endothelial cells, which was our original target, where factor VIII is normally produced. These LNPs contain Ionizable lipid (synthesized from 1,4-Bis(3-aminopropyl) piperazine and 1,2-epoxydecane) , DOPE, cholesterol, and DSPE PEG C16 ceramide in a molar ratio of 26.5:20:52:1.5. To specifically target LSECs, the LNPs include conjugated DSPE PEG-manose lipids that interact with the scavenger receptors on LSECs. Our goal was to recreate these LNPs with Factor VIII mRNA. In addition, we wanted to implement these LNPs for intra nasal delivery. Because the Factor VIII protein coding sequence was not yet in the registry, we chose to buy a plasmid by Robert Peters containing the full sequence for the human FVIII protein from addgene.

Build

To achieve our goal, we needed two things: the Factor VIII mRNA and the LNPs. We worked on these in parallel. The goal was to recreate the LNPs described by Kim et al. However, due to the cost of some components, we used a model LNP made out of cheaper alternatives. DSPE PEG C16 ceramide and DSPE PEG-manose were both substituted with the more affordable DSPE PEG-2000.

Producing the mRNA in a chassis organism was not feasible, so the plan was to synthesize the mRNA for Factor VIII using in vitro transcription and synthetically modifying it with a 5’-cap and 3’ poly(A)-tail. Amplifying the Factor VIII coding sequence before transcription proved to be difficult because of its length. After trying PCR with multiple sets of primers, the whole plasmid was amplified in E.coli to then be linearised. In our multiple tries of in vitro transcription, we managed to finetune our protocol and produce enough RNA to make LNPs. Before the RNA could be used in LNPs, it had to be purified, and modified with a 5’-cap and a poly-A tail. Purification was done with phenol-chloroform extraction, which also gave us some difficulties and it took us some time to troubleshoot the protocol. Working with Factor VIII RNA proved more difficult than expected, because the length made it prone to degradation and none of our team members had previous experience working with RNA.

Test

The LNPs by Kim et al. specifically targeted the sinusoidal endothelial cells, but several factors, including cost and availability, made it hard for us to test our LNPs on these cells. As an alternative, we used the HEK293T/17 cell line. To test our LNPs we tried to deliver mRNA encoding for the eGFP protein. We We tested the properties of our synthesized mRNA without LNPs by transfecting them into HEK cells to compare with the delivery of RNA with our LNPs. Analysis of the LNPs was done using dynamic light scattering (DLS) and transmission electron microscopy (TEM) to determine the size and zeta potential of the LNPs. The encapsulation efficiency was determined with the use of The concentration of RNA was also frequently measured using a NanoDrop spectrophotometer.

Learn

During our building stage, we were already confronted with some problems that made us reconsider our approach and rethink our design. The length of the Factor VIII RNA forced us to rethink our method of gene amplification. We also changed the concentration of the PBS buffer after measuring the zeta potential of our first batch of LNPs. Due to the cost of LSECs and some LNP components, we were forced to construct an affordable model that would still be able to prove our concept. We decided to use a generic HEK cell line and because there would be no specific targeting possible with these cells, we could also replace the expensive PEG-manose lipid with a more affordable option.