A 96-Well Plate Workflow for Building Block–DNA Conjugation in DEL Synthesis

Why the well format matters

Most DEL building blocks (BBs) are not commercially available as DNA conjugates — they are bi- or trifunctional fragments that have to be linked to a codon-specific oligonucleotide through a linker, one BB at a time, often thousands of times before initiating library synthesis. Doing that conjugation, purification, and deprotection one tube at a time doesn’t scale. The method described here keeps every manipulation — including HPLC purification and DNA precipitation — inside 96- or 48-well plates, so a plate of building blocks moves through the whole sequence with the same number of hands-on steps as a single reaction.

The five-stage process

1. Codon ligation and linker installation

The oligonucleotide backbone is assembled by splint ligation: a 16-nucleotide oligo carrying the PEG12 linker amine (Fmoc-protected, to keep it inert during ligation) is ligated to a 35-nucleotide hairpin oligo containing the codon, anticodon, and a conserved PCR-priming region. The anticodon base-pairs with the codon and the conserved regions basse-pairs with a splint throughout the process, protecting it from side reactions. Once ligation is complete, the Fmoc group is removed in situ with aqueous NaOH, exposing the primary amine that the building block will be conjugated to. The resulting amino-PEG12-oligonucleotide is isolated by ethanol precipitation in the plate and used without further purification.

2. Building block hydrolysis

Building blocks are stored as Boc- or pNs-protected amino esters in DMSO. Before conjugation, the ester is hydrolyzed with aqueous LiOH in a DMSO/water mixture and neutralized — the hydrolyzed acid is used directly as a crude mixture, with no intermediate isolation. Both protecting groups survive these aqueous basic conditions, which is what allows hydrolysis and conjugation to run as one pot instead of two.

3. DMTMM-mediated conjugation

The hydrolyzed building block is coupled to the amino-oligonucleotide using DMTMM (4-(4,6-dimethoxy-1,3,5-triazin-2-yl)-4-methylmorpholinium chloride) in a buffered water/DMSO mixture. DMTMM is an efficient, general-purpose amide-coupling reagent for aqueous, amine-functionalized oligonucleotides, and it performed well across almost every scaffold class we tested. A handful of slow-reacting heterocycles (some 4-carboxypyrazoles) likely need a different reagent — see the closing section below.

4. Purification by 96-well HPLC

Crude conjugates — still carrying their Boc or pNs protecting group, which is lipophilic enough to give good reverse-phase separation from unreacted oligonucleotide and byproducts — are purified by HPLC (0.1 M TEAA pH 7 / MeCN). A fraction collector distributes all 96 reactions from a plate automatically into eight 96-well deep-well plates, based on the retention time of the expected product. Fraction selection is done visually: an Excel template color-codes the wells to pool based on the collected retention windows, so the correct fractions are obvious at a glance rather than requiring 96 individual chromatogram reviews.

5. Deprotection

Pooled, precipitated conjugates are deprotected — Boc under basic buffer at elevated temperature, pNs under mild aqueous conditions at room temperature (see our companion note on choosing between the two) — then isolated a final time by ethanol precipitation and used directly in DEL synthesis.

What the numbers say

Yields are reported as three bands, based on the amount of starting oligonucleotide: low (under 9%, not enough material to support DEL synthesis), moderate (9–18%, usable but worth optimizing), and good (over 18%, ready to use without further work). Across the 278 building blocks and 15 scaffolds we tested, the large majority of pharmacologically relevant heterocycles — piperidines, pyrrolidines, azetidines, bicyclic pyrazoles — landed in the moderate-to-good range with no scaffold-specific optimization. All conjugates that passed this process were subsequently used in real DEL synthesis, and sequencing after selection confirmed both the DNA and the building block identity on every one.

Where the method generalizes

The workflow is built around codon-ligated oligonucleotides for yoctoReactor-based DEL synthesis, but nothing about the chemistry is specific to that format: the same hydrolysis/DMTMM-conjugation/plate-precipitation sequence applies to building block–oligonucleotide conjugates for split-and-mix DEL synthesis as well, since split/mix libraries also start from a BB-oligonucleotide starting material made the same way. It’s also directly transferable to bioconjugation work outside DEL entirely — anywhere an amine-functionalized oligonucleotide needs to be coupled to a carboxylic acid in water at scale.