Gene therapy is an incredibly versatile therapeutic strategy that can be used to treat monogenic disorders. Gene therapy can inactivate or silence mutant gene transcripts, edit the genome to repair mutated genes, or express healthy copies of disease genes. Gene therapy is often delivered using a viral vector, but this can lead to various issues such as a lack of control over the levels of expression, alterations to the DNA or RNA other than the desired effects, and the viral vectors becoming part of the human DNA. These adverse effects could be minimized if the gene therapy expression could be tightly controlled. This is what the team led by Dr. Davidson set out to do: they developed a switch to control whether gene therapy is active or if it is switched off.

The group started their pursuit of a gene therapy switch by identifying small molecules capable of altering splicing. Splicing is the process in which a newly made messenger RNA transcript is converted into a mature messenger RNA. During splicing, sections of genetic code in the newly made messenger RNA called introns are removed from between sections called exons. By splicing the exons back together a mature mRNA is made. There are small molecules approved for treating the childhood-onset motor neuron disease, spinal muscular atrophy, that act by binding across a position where a newly synthesized mRNA transcript would normally be spliced to generate the mature transcript. The group took advantage of this system to control the inclusion or exclusion of genetic material that signals whether or not a protein should be made from the RNA transcript. If no protein is produced, then the gene therapy would not be expressed and would be in an off state, if a protein is made, then the gene therapy would be active or switched on.

The group first needed to find a genetic ‘context’ that was responsive to the splice-switching small molecule when used at a low dose. After treating cells with the compound called LMI070, the group looked for changes in splicing patterns compared to untreated cells. The group found 45 events changed in the splicing of control versus LMI070 treated cells. These events included a splicing change in the gene SF3B3. By re-analyzing published RNA sequencing data, they found that many of these events, including SF3BS, were consistently altered. Next, the group looked at how frequently these splicing changes occurred without LIM070 treatment. By analyzing these 45 events in 21,504 human RNA-seq datasets, the team found that the LMI070-induced change in SF3B3 rarely occurs naturally (without the presence of LMI070). Finally, the group found that treatment with LMI070, at the dose that leads to robust SF3B3 splicing changes, caused minimal changes in gene expression. Only five upregulated and nine downregulated genes were identified. This work showed that the SF3B3 splicing event is a highly selective genetic ‘context’ responsive to low doses of LMI070 and that LMI070 treatment caused minimal off-target effects – a highly sought-after characteristic for gene therapy switches.

To achieve control of gene therapy, the group made a synthetic exon that contains the signal for where to start making a protein. They placed this synthetic exon within the genetic context of SF3B3 which responds to LMI070. This combined genetic sequence was called X^on. In this system, LMI070 treatment led to the inclusion of the synthetic exon in the mature mRNA and the production of a reporter, green fluorescent protein (GFP). The GFP reporter enables the researchers to track the level of control they have over the system. Without LMI070, there is no production of GFP.