Sample Prep Automation for 3D Culture Workflows

Team Dynamic Matrices | 3 August 2025

Getting organoids into culture at scale is one problem, and we cover it in our screening post. Getting them out again, into RNA, protein or single cells, in a form a robot can process without a technician watching, is a different problem and in our experience the harder one. Here is where automated 3D sample prep breaks and what chemistry has to look like to make it walk-away.

The four things that break

Matrix removal is unpredictable. Most organoids sit in basement membrane extract. Cold depolymerisation takes 30 to 60 minutes with washes and is incomplete on a variable schedule; enzymatic digestion degrades matrix inconsistently. Leftover matrix protein inhibits RT and polymerases, fouls capture chemistry, and clogs tips and filters. An automated protocol cannot look at the pellet and decide it needs another wash.

Structures vary in size. A well contains organoids at different stages and diameters. A digestion time set for 100 µm structures under-digests 300 µm ones and over-digests 50 µm ones. A technician adjusts; a robot does not.

The analytes are sensitive. 3D work often targets low-abundance transcripts, labile phosphorylation or native complexes. Extended processing at 37 °C and harsh lysis degrade exactly those, and automation tends to lengthen processing because it runs plates serially.

The material is awkward. Dissociated organoid lysate is viscous, contains matrix fragments and DNA, and clumps. Liquid handlers calibrated for aqueous transfers deliver inconsistent volumes and carry over debris.

What automation-ready chemistry has to do

• Remove the matrix completely, in a fixed time, without technician judgement
• Digest consistently across the organoid size range in one well
• Tolerate residual matrix and debris rather than requiring perfect cleanup
• Preserve RNA integrity and protein activity under gentle conditions
• Use additions and incubations, not timed stops or bead separations
• Stay stable on the deck for the length of an unattended run

Two design choices address most of this list. A matrix that dissolves on a defined trigger rather than by slow depolymerisation or variable digestion makes removal a fixed-time step. DyNAtrix dissolves in 10 to 15 minutes with a nuclease at 37 °C, completely and identically every time, which turns the least predictable step in the workflow into an addition and an incubation. And a capture chemistry that tolerates debris and does not need beads makes the downstream step robust to what the upstream step leaves behind.

Where LASSO fits

LASSO captures targets by hybridisation-triggered phase separation of a soluble polymer. In an automated context that gives four properties that bead and enzyme methods do not:

• Tolerance for matrix debris and residual protein. Contaminants stay in the supernatant rather than adsorbing to a bead surface, so upstream cleanup demands fall.
• Fixed-time completion. Capture goes to completion and holds; there is no digestion to stop on time.
• Gentle conditions. Physiological buffer throughout, so RNA integrity and protein activity survive a long run.
• Deck stability. No enzyme decay over the hours of a plate run; reagents stable at 4 °C for about a year.

The steps are addition, incubation, centrifugation and supernatant transfer, all of which automate directly. Two pilot labs are running riboLASSO on liquid handlers now.

A two-part validation plan

Validate the two halves separately, because they fail for different reasons.

Part 1: dissociation consistency. Take organoids across your real size distribution (sort or sieve into three bins), run the automated dissociation on each bin, and measure recovery, viability and, if headed to sequencing, a stress marker. You are looking for a protocol that gives similar numbers across bins. If it does not, the fix is usually matrix removal (make it complete and fixed-time) before it is digestion time.

Part 2: downstream tolerance. Take a clean lysate and deliberately contaminate it with matrix fragments, gDNA and cell debris at realistic levels. Run the capture or extraction step and measure yield and purity against the clean control. A chemistry that holds up under deliberate contamination will hold up on the deck. Testing only on clean preps tells you nothing about production.