Abstract
CAR T cell therapy has revolutionized cancer treatment, but has also provided an opportunity for treating chronic viral infections such as HIV, HBV, and HCV. Despite the profound outcomes in the treatment of hematological malignancies, CAR T cell therapy for solid tumours has not been almost invariably unsuccessful in the clinic. Hostile conditions of TME, low tumour infiltration, lack of persistence, and absence of memory CAR T cell formation are the main obstacles ahead of CAR T cell therapy for solid tumours. This study aimed to improve Her2-CAR T cell persistence and TM (T memory) development.
TM cells have distinct mitochondria morphology and metabolism. TM cells have larger mitochondria (fusion) and rely on OXPHOS metabolism. In order to achieve the aim of this study, we selected Mcl-1 and miR429 as genes to overexpress in CAR T cells. Recently, several studies suggested that AICD (activation-induced cell death) induced by the CD95 pathway is the one of the main causes of low CAR T cell persistence in vivo. Mcl-1 is also well characterised for its role in OXPHOS metabolism, mitochondrial energetics and mitochondrial fusion. To complement this approach, the miRNA429 was selected as a means to enhance CAR T cell function through the suppression of genes that negatively affect T cell function, TM development, and mitochondrial fusion such as TCAIM, MFF, and TET-2.
The first aim of this study was to upregulate the endogenous level of Mcl-1. We tested eight small activating RNA (saRNA) targeting different regions of the Mcl-1 promoter, but none of them was able to induce Mcl-1. Further promoter analysis led to the identification and characterisation of the first antisense transcript (named mcl1-AS1) that modulate Mcl-1 expression. However, due to the late manifestation of gene regulation (at 48 - 72 hours) that was seen following mcl1-AS1 inhibition, it was not applicable for us to use this strategy to Mcl-1 expression (Chapter II).
The next strategy was the controlled expression of Mcl-1 using the Tet-On system. We used several approaches to improve the Tet-On system, including gene replacement, codon-optimisation of rt-TA, using G72V-rtTA, removing cryptic splice sites within rt-TA, creating an autoregulatory Tet-On system, and manipulating regulatory elements in TCE minimal promoter. Our final optimised construct showed high inducibility and a very low background expression compared to the original construct (Chapter III). However, due to the low transfection efficiency of SB system in primary T cells and lack of artificial antigen presenting cell (aAPC) at the time for expansion of T cells, we decided to create an inducible LV system. The lack of inducibility in low doxycycline concentration and low transduction efficiency made our inducible LV system not suitable for our study. Therefore, we decided to use a constitutive system to see the effects of Mcl-1 and miR429 overexpression in CAR T cells.
In order to express Mcl-1 and miR429 in a constitutive LV system, we examined the strength of four commonly used promoters, EF-1, CMV, RPBSA, and hPGK, in running short and long transcripts. EF-1 showed to be the best promoter in running short and long RNA in T cells (Chapter IV). As a result, we chose EF-1 to run the GFP-P2A-Her2CAR and hPGK to transcribe Mcl-1 or miR429.
For the first time, we showed that TCAIM, MFF, and TET-2 are direct targets of miR429. Overexpression of miR429 in CAR T cells slightly increased the number of TSCM and TCM in CD4+CAR T cells, while the number of Treg and TEMRA cells was marginally decreased. Upregulating Mcl-1 in CAR T cells promoted the TSCM and TCM development in both CD4+ and CD8+ CAR T cells, whereas the frequency of Treg and TEMRA cells declined. Mcl-1 overexpression also protected CAR T cells from CD95L-induced AICD. Although our study cannot provide a mechanism for the Mcl-1 role in memory CAR T cell development, an increase in mitochondrial mass and mtDNA suggest that Mcl-1 likely enhances mitochondrial energetics and fusion (Chapter V).
Lastly, we are in a SARS-CoV-2 pandemic era. When locked-down and unable to access the laboratory in early 2020, we investigated the effect of recurrent mutations on viral RNA secondary structure and host miRNA interaction. From an evolutionary point of view, mutations that arise several times independently (homoplasies) and lead to clade expansion are highly likely to increase viral fitness. The emergence of several mutations has resulted in the emergence of a G-clade responsible for 97% of cases around the globe. This clade consists of four mutations, two cause amino changes (C14408U and A23403G), while others (C241U, C3037U) are silent and are currently of unknown impact on viral fitness. Based on our bioinformatics survey, the C3037U mutation destroys the miR-197-5p binding site. Interestingly, miR-197-5p is highly expressed in SARS-CoV-2 target cells and has been reported to be upregulated in patients with cardiovascular disease. Interestingly, this miRNA also acts as defence against variety of viral infections such as HBV, HCV, HAV, Enterovirus 71, Ebola and H7N9. Further in vitro work is underway to test the significance of the C3037U mutation on miRNA inhibition in a SARS-2 pseudovirus and live virus assays.