Skip to content
rnaprog
This methodology article is published in English; the interface and SEO metadata are localised. Read the original English version.

ASO gapmer design: why 5-10-5 exists and what Tm is telling you

The gapmer geometry is a compromise between RNase H recruitment and nuclease stability. How to choose wing and gap lengths, what Tm window to aim for, and which sequence features quietly kill a gapmer.

rnaprog editorial 6 min read

An antisense oligonucleotide has to do two contradictory things: survive in the cell long enough to find its target, and recruit RNase H to cleave that target once it arrives. A gapmer splits the molecule into two jobs.

The geometry

The standard design is 5-10-5: five modified nucleotides at each end (usually 2′-O-methoxyethyl or 2′-OMe, sometimes constrained ethyl), and ten unmodified DNA bases in the middle. The wings resist nucleases and raise binding affinity; the DNA gap is the RNase H substrate — the enzyme requires a DNA:RNA duplex and will not cleave a fully modified strand.

The trade-off is real. Longer gap means better RNase H activity but less nuclease protection. Shorter gap protects better but may not recruit enough cleavage. Ten is the empirical sweet spot for 2′-MOE wings, and 5-10-5 remains the default in most clinical programmes.

Variants worth knowing: 3-10-4 and 4-8-4 appear when you need to tune total length for a specific target region. Fully modified steric-blocking ASOs (no gap) work by occupancy rather than cleavage and follow different rules entirely — do not apply gapmer heuristics to them.

Tm: the number that matters most

Gapmer activity correlates strongly with duplex melting temperature against the RNA target. The practical window is roughly 48–60 °C for a DNA:RNA hybrid, with 54 °C as a reasonable centre.

Below 48 °C the duplex is unstable at physiological temperature and you lose potency. Above 60 °C you start to see specificity problems: the ASO will bind near-matches, and RNase H will cleave them too.

Tm depends on length, GC content and chemistry. A 20-mer at 50% GC in 2′-MOE wings sits in the window; the same sequence at 35% GC does not, and no amount of wing modification fixes it. This is why GC filtering happens before scoring: an AT-rich region simply may not contain a viable gapmer, and the answer is a different target site, not a lower threshold.

Note what our Tm is: a nearest-neighbour approximation applied to the sequence context. It is good enough to rank candidates and to spot sequences far outside the window. It is not a measured Tm, and if you are filing an IND you will measure it.

Sequence features that quietly kill gapmers

CpG in the DNA gap. Unmethylated CpG motifs trigger innate immune activation through TLR9. In a gapmer, the DNA gap is exactly where the immune system looks. A CpG there will produce an inflammatory response that may swamp your readout — and in a therapeutic context, is a toxicity signal.

G-quadruplex motifs. Four or more consecutive G’s, or G-rich stretches capable of forming tetrads, produce structures that bind proteins and reduce availability. Flag and avoid where possible.

Homopolymer runs. Same problem as any synthesis: runs of six or more impair manufacture, and poly(G) is worse because of quadruplex formation.

Self-complementarity. A gapmer that can fold onto itself will not bind its target. Check for inverted repeats, especially in the wings.

Where to target

For an mRNA target, the practical hierarchy:

  1. Coding sequence, avoiding the start codon region and the first ~15 nt (translation initiation complexes occupy it).
  2. 3′ UTR, which is often more accessible than the CDS because it is less densely covered by ribosomes.
  3. Splice sites if you want to redirect splicing rather than degrade the transcript — but that is a different design goal and needs the junction sequence, not the exon.

Avoid the extreme 5′ and 3′ ends: they are bound by cap and poly(A) machinery respectively.

What to do with a candidate table

Order two or three, not one. Screen at two or three concentrations, and include a scrambled control with the same chemistry — chemistry-matched controls matter more for ASOs than for siRNAs, because the wings themselves have biological effects. A gapmer that shows activity at 100 nM but nothing at 10 nM is behaving like a non-specific oligo, not a specific antisense.

And read the flags column before the score column. A gapmer with CpG in the gap and a G-quadruplex in the wing can still rank highly on GC and Tm. The composition problems are what will show up in your assay.

Первый запуск начинается с 300 бесплатных кредитов

Хватит на 6 mRNA-конструкций или пул из 12 siRNA, включая QC и экспорт. Без карты.

300 credits · No card required · You keep the rights to your sequences