Abstract
Diabetes Mellitus, an ever-growing epidemic, kills millions every year with ~50% of deaths attributable to diabetic heart disease (DHD). Despite large amounts of research dedicated to attenuating this issue, the epidemic shows no signs of slowing down. The development of adverse cardiac conditions in diabetics largely result from a prolonged loss of molecular homeostasis that stems from hyperglycaemia. Ultimately, the dysregulated molecular pathways result in pathogenic cellular apoptosis, pathogenic vascular remodelling and fibrosis of cardiac muscles. All of these inhibit normal heart function and eventually lead to heart failure. Research has found that small regulatory molecules, known as microRNAs (miRNA), which play a pivotal role in maintaining cellular homeostasis, are highly dysregulated in diabetic individuals. As such miRNA have been explored as both potential biomarkers and therapeutic targets/agents for DHD. However, the nature of miRNA therapeutics presents significant challenges. A singular miRNA has numerous targets, requires different levels in each cell type, and many therapeutics are prone to easy degradation, complicating their delivery. In this project we investigate the potential of new techniques to combat these issues.
First, attempts were made to harness an emerging technology CRISPR/Cas13, which has been shown to efficiently target and cleave single stranded mRNAs and has been used to detect the presence of miRNA using a system known as SHERLOCK. This project aimed to determine the feasibility of CRISPR/Cas13 as a novel tool for targeted miRNA knockdown. To this end, a cardiomyocyte-specific CRISPR/Cas13 expression plasmid was designed to target pro-senescent (miR-34a) and pro-apoptotic (miR-320) miRNAs that are upregulated in the diabetic heart. This plasmid was transfected into cultured cardiomyocyte cell lines and RT-qPCR was used to measure if the targeted miRNA were down regulated. A total of ten different guide RNA (gRNA) were tested targeting pre-miR-34a and pre-miR-320. All guides failed to regulate miRNA to the same degree as a commercial anti-miRNA control, suggesting that Cas13 is not suitable for miRNA knockdown in its current form. Unexpectedly, several guides resulted in the upregulation of target miRNA, suggesting potential stabilisation of target miRNA or the ability of expressed gRNA to be recognized as mature miRNA.
Next, this project aimed to explore formulation of lipid-based delivery vehicles for miRNA therapeutics. A recent strategy to improve nucleic acid delivery is to utilize lipid nanoparticles (LNPs), which confer increased stability against degradation and can enable targeted delivery to specific cell types via attachment of ligands to the LNPs. However, encapsulation of miRNAs in LNPs face two significant challenges. The first is low entrapment efficiency leading to high treatment costs, and the second is the long, difficult conventional manufacturing process which are not suited for large scale production. Recently the development of LNP formulation by microfluidics has largely addressed these two issues for mRNA encapsulation, however only few studies have determined the efficiency of this technique for miRNA encapsulation. This study explored the utilization of cationic LNPs, formulated using microfluidics to encapsulate a miRNA therapeutic cocktail. LNPs were formulated using DSPC, Cholesterol, DOTAP, and DSPE-PEG2000 in a molar ratio of 11:52.5:35:1.5 or DSPC CHOL DDA and DMG-PEG2000 in a molar ratio of 10:48:40:2. The miRNA mimics chosen for the encapsulation study were miR-199a-3p which promotes proliferation and miR-133a-3p which is anti-fibrotic. miRNA mimics were solubilized in pH 3.5 citrate buffer. Dynamic light scattering and cryo-transmission electron microscopy confirmed nanoparticles generated by microfluidics were highly uniform and consistently smaller than 150nm. Qubit analysis has confirmed miRNA encapsulation efficiencies as high as 85% can be achieved in a rapid, repeatable, fashion using microfluidics. Further, RT-qPCR results indicated delivery of high amounts of miRNA mimics to both AC16 cardiomyocyte cell lines and human umbilical vein endothelial cells (HUVECs) can be achieved within 24h post transfection.
These LNPs were also tested in a hypoxia model, to probe their effectiveness in the treatment of increased ROS, one of the main contributors to adverse cardiovascular events in diabetics. RT-qPCR showed miRNA levels were still highly regulated under hypoxic conditions, however downstream functional assays indicated regulation of target miRNAs was not sufficient to reverse the effects of hypoxia. Additionally, wound healing assays showed delayed healing in response to LNP treatment, regardless of the presence of miRNA mimics suggesting that LNP formulations while not demonstrating significant cytotoxicity, may not be beneficial to HUVECs. To further investigate the lack of functional changes observed despite marked miRNA upregulation in RT-qPCRs, mass spectrometry was performed. Mass spectrometry identified ~1000 regulated proteins, of which ~50 proteins were associated with miR-199a-3p and miR-133a-3p. However, many proteins typically associated with these target miRNAs were not significantly regulated. Nonetheless, STRING analysis and Gene Ontology (GO) analysis revealed changes in proteins that are predicted to enhance free metal ion abundance and ROS production. This could possibly be a result of cholesterol influx, as cholesterol production pathways were highly downregulated in LNP treated cardiomyocyte cell lines. Overall, this data suggests miRNA mimic treatments had little effect in this study, and raise questions of how suitable LNPs might be for treatment of DHD which is associated with increased free metal ion availability and ROS production.
In conclusion, this thesis marks a pioneering investigation into the potential of CRISPR/Cas13 as a novel tool for the targeted regulation of specific miRNAs dysregulated in diabetic heart disease. Additionally, it is the first study to successfully encapsulate a miRNA therapeutic cocktail within cationic lipid nanoparticles (LNPs) using microfluidics. These LNPs demonstrated stability, optimal size, high protective capabilities, excellent encapsulation efficiency, and successful transfection into both AC16 cardiomyocytes and HUVECs. Despite this, the functional outcomes of treatment were limited, indicating a potential issue between cellular uptake and functional effect. Additionally, proteomic analysis following treatment indicates that further research is necessary to fully comprehend the potential impacts and suitability of LNP vectors for the treatment of diabetes and cardiovascular diseases.