Novel thiomorpholino oligonucleotides (TMOs) as a robust next generation platform for spice switching antisense therapies
Rakesh Naduvile Veedu, PhD, Associate Professor and Head, Precision Nucleic Acid Therapeutics Group
Dr. Rakesh Veedu started his talk with a diagram depicting the flow of information from DNA to proteins (DNA à pre mRNA à mature mRNA à protein) and highlighted that some diseases are triggered by the modulation and variations in the expression levels of proteins—overexpressed, under expressed, or not expressed at all. He highlighted the importance of nucleic acid-based therapeutic technologies such as antisense oligo, aptamers, siRNA, antimiR, and DNAzyme in targeting different levels of these processes. He noted that Milasen, a drug developed for the treatment of Batten Disease,went from idea to clinical use in just 10 months.
Dr. Veedu gave examples of several nucleic acid analogs developed over several years, few of which have been advanced to the clinic yet. The FDA-approved oligonucleotide therapeutics for Duchenne muscular dystrophy were made from phosphoramidite morpholino oligomers (PMOs), which are incompatible with the conventional oligonucleotide synthesis method, making batch production challenging. In 2016, Dr. Veedu and his team developed the phosphoramidite variants of morpholino nucleic acids to make 2’-OMe chimeric antisense oligos. They evaluated their potential to induce splice switching and demonstrated the efficiency in splice-switching as well as synthesis. With his mentor, Prof. Marvin Caruthers, they then designed and evaluated thiomorpholine oligonucleotides (TMOs), which are charged variants of PMOs.
Duchenne muscular dystrophy (DMD) was used as a model system to test the potential of TMOs. DMD is defined as a severe muscle-wasting disorder. Individuals with DMD lack functional dystrophin protein. Mutations and deletions within exons in the gene encoding dystrophin disrupt protein expression. Approved antisense drugs basically target the mutation containing exon and restore the reading frame to produce partially functional dystrophin protein. In their studies, they focused on exon 23 deletion in mdx mice myotubes in vitro.
Dr. Veedu and his team synthesized a 20-mer antisense oligonucleotidecomposed of TMOs and standard PMOs as well. They used a lipid-based transfection reagent, lipofectin, to transfect the oligos into a mouse model. A RT-PCR analysis was performed on isolated RNA, which showed TMOs efficiently inducing splice-switching at very low nanomolar concentrations while PMOs with a neutral charge that are not complex with lipid-based transfection reagents did not induce exon skipping. Next, they used electroporation to deliver TMOs and PMOs. The RT-PCR analysis indicated that TMOs were slightly better at exon skipping compared to PMOs. They also used naked transfection and found that TMO could actually induce exon skipping after five days whereas PMO could not.
They continued with truncation experiments by trying 20-mer, 18-mer and 16-mer TMOs and that showed 16-mer TMO could also efficiently induce splice switching. Following that, they prepared thiomorpholino-DNA chimeras in two classes: gapmer-like chimeric molecules consisting of DNA and TMO segments and mixedmers containing alternating TMO and DNA units in every two nucleotides. Dr. Veedu said that while all were able to induce splice switching, mixedmers worked more efficiently than gapmers. They also compared the performance of TMO along with the conventional negatively charged chemistries such as 2’OMePS and 2’MOEPS and found that TMOs were more efficient especially at low concentrations.
Dr. Veedu and his team wanted to find out the reason why TMOs were so efficient. Melting data could not explain the difference between the efficiency of TMOs and the other modalities. Cellular uptake studies with and without lipids indicated the efficient nuclear localization of TMOs, while the uptake of PMOs was limited consistent with the exon-skipping data. Their preliminary in vivo evaluation data demonstrated that both TMO and PMO induced exon 23 skipping and that TMO rescued dystrophin expression to some extent.
Dr. Veedu mentioned his other studies that focused on locked nucleic acid (LNA)-OMe mixmer antisense oligonucleotides that efficiently induced splice switching. They also performed experiments with alpha-L-LNA (α-L-LNA) oligonucleotides and α-L-LNA modified 2’-O-MePS mixmers. At all lengths (20-mer, 18-mer and 16-mer), α-L-LNA modified 2’-O-MePS mixmers induced splice switching more efficiently than fully modified 2’-O-MePS. They then evaluated the effect of DNA segments in splice switching ASOs. They constructed ASOs consisting of a 2’-OMe backbone with a segment of four nt DNA fragments on the sequence. The results showed that the construct with a DNA segment of four nucleotides at the 3’ end induced better exon skipping compared to the fully modified 2’-OMe oligo. The ASO with four-nucleotide DNA segments in the middle did not induce splice switching just like gapmer. The effect of the length of the DNA segment in the 3’ was evaluated and constructed by alternating ASOs containing two to seven DNA nucleotides in the 3’ end. Those with two and four DNA nucleotides in the 3’ end induced efficient splice switching.
