A Morpholino user, accustomed to using 25-base Morpholinos in Xenopus laevis, was concerned about a design she received for a 20-base Morpholino. She wrote to ask if we had designed many short oligos and if they worked well. Here is my response.
The key factor for performance of a Morpholino is not length but Tm. Organisms grown at different temperatures will have different optimal Tm ranges. If the Tm is too low then the oligo doesn’t stick well to RNA, too high and it will stick to off-target RNAs though high-affinity subsequences.
We like to make oligos at 25 base length because this keeps the per-base affinity low. For two oligos with the same Tm, a longer oligo has a lower per-base affinity and so is less likely to have a high-affinity subsequence that could be active against an off-target RNA.
However, there are a number of good reasons to design a shorter oligo. For instance, there might be sequences present at either side of the target that put stable self-complementarities into the oligo, causing the oligo to dimerize and lose antisense activity; sometimes it is only a 20-mer that can fit between these sequences. The target region might be flanked by lots of C bases, so that a 20-mer would have good aqueous solubility but a 25-mer would have too high a G content to dissolve well. Finally, the target sequence might make high-affinity oligos, with Tm too high for the organism they are to be used in; in this case, we will shorten an oligo to bring its Tm closer to the optimal RNA affinity.
We shorten oligos to 23 bases often, especially in organisms with zebrafish temperatures (26°C) and below. We also design shorter oligos if needed, but will typically annotate our designs to mention they are short. For bacterial work, oligos conjugated with cell-penetrating peptides are often designed as short as 12 bases (to allow them to pass though the cell envelope).
For your application, you are working in an ~18°C organism so short oligos are often needed to keep the Tm low enough to favor specificity of the Morpholino for its intended RNA target.
We have been designing many short oligos for Xenopus laevis since a 2018 paper (cited below) appeared showing there was an unacceptably high level of splice-modulation in Xenopus laevis with our usual designs (at that time, almost entirely 25-mers designed as we would for a zebrafish [26°C] transcript). Since decreasing our target Tm for Xenopus laevis designs (and often shortening the oligos to produce sequences in that Tm range), we have no longer heard from the Xenopus community about dissatisfaction with the specificity of Morpholinos (though of course, off-target effects are always possible and should be controlled for).
Gentsch GE, Spruce T, Monteiro RS, Owens NDL, Martin SR, Smith JC. Innate Immune Response and Off-Target Mis-splicing Are Common Morpholino-Induced Side Effects in Xenopus. Dev Cell. 2018;[Epub ahead of print] doi:10.1016/j.devcel.2018.01.022.
The innate immune debate continued:
Paraiso KD, Blitz IL, Zhou JJ, Cho KWY. Morpholinos Do Not Elicit an Innate Immune Response during Early Xenopus Embryogenesis. Dev Cell. 2019;49(4):643-50. doi:10.1016/j.devcel.2019.04.019.
Gentsch GE, Spruce T, Owens NDL, Monteiro RS, Smith JC. The Innate Immune Response of Frog Embryos to Antisense Morpholino Oligomers Depends on Developmental Stage, GC Content and Dose. Dev Cell. 2019;49(4):506-7. doi:10.1016/j.devcel.2019.05.004.