DNA Damage Response Screening
High-content DNA damage profiling with multiplexed SemaCyte® microcarriers
A 3-plex, image-based γH2AX assay demonstrating barcoded adherent-cell multiplexing, same-day deployment from frozen SemaCytes, and deconvolution of cell line-specific etoposide responses in a single well.
Abstract
DNA damage response (DDR) pathways are important targets in oncology drug discovery, yet their characterisation typically requires separate experiments across each cell model of interest — a throughput bottleneck as compound libraries and cell panel sizes grow. This case study describes a first-pass validation of SemaCyte®-enabled multiplexed DDR profiling, carried out in collaboration between Semarion Ltd and o2h Discovery. Three cancer cell lines (U2OS, A549, U87MG) were independently seeded onto optically barcoded SemaCyte® microcarriers, cryopreserved, and then pooled and dispensed together into a single 96-well assay plate. Following a two-hour etoposide dose-response challenge, cells were fixed and stained for γH2AX and RAD51, and imaged on a ThermoFisher Scientific CellInsight CX7 LED Pro. Semarion's image analysis pipeline (Semalyse) deconvoluted the mixed population by barcode to yield per-cell-line dose-response curves. EC50 values from the SemaCyte multiplex closely preserved cell-line rank ordering and absolute potency relative to well-plate controls, despite a roughly six-fold reduction in cells per condition. A pre-fixed barcoded U2OS population served as an embedded in-well reference control across every plate well. These results demonstrate the power and versatility of the SemaCyte® platform for scalable, multiplexed high-content DDR screening.
Contents
- 1. Background & Scientific Context
- 2. Objectives
- 3. Experimental Design
- 4. Workflow Overview
- 5. Results I — Assay Uniformity
- 6. Results II — Imaging and Dose-Response Profiling
- 7. Results III — EC50 Comparison: Multiplex vs. Plate Control
- 8. Conclusions
Background & Scientific Context
1.1 The DNA Damage Response in Drug Discovery
The DNA damage response (DDR) is a network of molecular pathways that detect and repair DNA lesions. Dysregulation of these pathways is a hallmark of cancer, making DDR a productive target class in oncology drug discovery. Key components — including ATM/ATR kinases, PARP, and the homologous recombination machinery — have already yielded approved medicines (notably PARP inhibitors), and DDR remains an active area of early-stage compound profiling.
Phosphorylation of histone H2AX at serine 139 (γH2AX) is among the most widely used surrogate markers for DNA double-strand breaks. Its detection by immunofluorescence, combined with high-content imaging (HCI), enables quantitative, cell-by-cell characterisation of DDR engagement across compound dose ranges. However, HCI-based DDR assays are conventionally run one cell line per plate — a configuration that limits throughput when multiple models must be profiled in parallel.
1.2 The Case for Multiplexed Cell Profiling
Cancer cell lines differ markedly in their intrinsic DDR capacity, making multi-model profiling essential for meaningful compound ranking. In conventional workflows, running three cell lines across an 11-point dose-response curve requires three separate assay plates — consuming three times the reagent volume, instrument time, and compound quantity, while introducing inter-plate variability that can confound cross-model comparisons.
SemaCyte® microcarriers address this elegantly by physically separating each cell population on its own barcoded microcarrier. Multiple cell lines can be pooled into a single assay well, treated identically under the same compound conditions, and subsequently deconvoluted by optical barcode during image analysis. This within-well co-culture design eliminates inter-plate variability while dramatically reducing the footprint of multi-model screens.
SemaCyte® Platform
SemaCyte® microcarriers are optically barcoded, chemically functionalised substrates onto which adherent cells are seeded and cultured. Each distinct barcode corresponds to a specific cell line, enabling post-imaging deconvolution of mixed populations. Cells remain on their carriers throughout seeding, freezing, pooling, drug treatment, fixation, staining, and imaging — preserving a full conventional adherent-cell workflow within a miniaturised, multiplexed format.
Objectives
This study was designed as a first-pass evaluation of whether the SemaCyte® platform could support a real-world HCI DDR assay in a multiplexed format, using frozen assay-ready cell stocks. The specific objectives were:
1. Validate frozen SemaCyte stocks. Confirm that cell-loaded SemaCytes cryopreserved prior to assay could be thawed, pooled, and deployed same-day with acceptable cell viability (≥75%).
2. Demonstrate barcode-mediated deconvolution. Show that optical barcodes could be reliably decoded by Semalyse image analysis to separate per-cell-line populations within a mixed well.
3. Generate cell-line-resolved dose-response data. Produce γH2AX-based dose-response curves for each of three cancer cell lines simultaneously from a single 96-well assay plate.
4. Benchmark EC50 values against conventional plate controls. Compare potency estimates from the SemaCyte multiplex format to historical well-plate control data to assess quantitative concordance.
5. Validate an embedded in-well reference control. Demonstrate the use of a pre-fixed, barcoded U2OS positive control population as a within-well anchor across all assay wells.
Experimental Design
3.1 Cell Lines
| Cell Line | Tumour Origin | SemaCyte Barcode | Role |
|---|---|---|---|
| U2OS | Osteosarcoma | Barcode 1 | Primary assay cell line; reference model |
| A549 | Non-small cell lung carcinoma | Barcode 2 | Primary assay cell line |
| U87MG | Glioblastoma | Barcode 3 | Primary assay cell line |
| U2OS (control) | Osteosarcoma | Barcode 4 | Pre-fixed in-well positive control (50 µM EP) |
3.2 Compound & Assay Format
| Parameter | Detail |
|---|---|
| Plate format | 96-well |
| Compound | Etoposide (EP) — topoisomerase II inhibitor |
| Dose points | 11-point DRC |
| Treatment duration | 2 hours |
| Multiplexed cell lines per well | 3 (+ 1 fixed reference) |
| SemaCyte barcodes | 4 (Barcodes 1–4) |
| Cell preparation | Frozen SemaCytes, thawed same-day |
| Primary endpoint | γH2AX mean nuclear intensity |
| Secondary endpoint | RAD51 (exploratory) |
| Imaging platform | ThermoFisher Scientific CellInsight CX7 LED Pro |
| Imaging parameters | 10× objective, 25 fields per channel, image stitching |
3.3 Fluorescent Probes
| Channel | Probe | Readout |
|---|---|---|
| Brightfield | — | SemaCyte barcode imaging & decoding |
| Blue | Hoechst 33342 | Nuclear staining; cell counting |
| Green | α-RAD51 (AF488) | Homologous recombination marker (exploratory) |
| Red | α-γH2AX (AF647) | Primary DDR endpoint; DNA double-strand breaks |
| Far-red / Orange | Phalloidin (iFluor 555) | Actin cytoskeleton; cell morphology & segmentation |
Workflow Overview
The experimental workflow followed four sequential stages, from cell preparation through to fixed-endpoint image acquisition. Each stage was designed to be compatible with standard laboratory equipment, while taking advantage of the logistical flexibility offered by cryopreservable SemaCyte stocks.
4.1 Cell Handling & Cryopreservation
Each of the four cell populations was seeded independently onto SemaCyte Seeding Dishes (SD20), each loaded with microcarriers bearing a distinct optical barcode. Following incubation to achieve appropriate confluency, cell-loaded SemaCytes were released, magnetically purified, and cryopreserved in separate vials. Post-thaw recovery across all three primary cell lines exceeded 75% viable cells, confirming excellent compatibility of the SemaCyte format with a freeze-and-store workflow.
On the day of the assay, frozen stocks were thawed and all four populations — the three primary cell lines together with the pre-fixed U2OS control — were pooled and dispensed as a single mixed suspension into the assay wells. This same-day setup from frozen stocks substantially simplifies assay scheduling and removes the need to maintain live cell cultures immediately before an experiment.
4.2 Image Acquisition & Barcode Deconvolution
Plates were imaged on a ThermoFisher Scientific CellInsight CX7 LED Pro using a 10× objective with 25 fields per channel and image stitching. In the brightfield channel, SemaCyte barcodes — functioning analogously to QR codes — were identified and decoded by Semalyse, Semarion's image analysis software. Decoded barcodes were used to generate cell-line-specific image masks, partitioning each field into regions belonging to each cell population. These masks were then passed to CellProfiler for per-cell feature extraction, yielding separate quantitative readouts for each cell line from a single well's image data.
Results I — Assay Uniformity
Prior to interpreting dose-response data, assay uniformity across the 96-well plate was assessed to confirm that signal variability arose from compound-driven biology rather than systematic positional artefacts. γH2AX mean nuclear intensity was measured across all plate columns in vehicle control wells.
Assay Quality
Uniform γH2AX signal across all 12 plate columns under vehicle control conditions confirms that the SemaCyte dispensing process does not introduce positional bias — a key quality indicator providing confidence that EC50 values faithfully reflect genuine pharmacological responses.
Results II — Imaging and Dose-Response Profiling
Following barcode deconvolution, imaging data and γH2AX quantitation were analysed across all wells and cell lines. The SemaCyte platform delivered high-quality, cell-line-resolved fluorescence data from a single pooled assay, spanning from raw field images through to fitted dose-response curves — all without physical separation of the three cell models at any point during the experiment.
(SemaCyte format)
per assay well
per experiment
control (554 vs 84)
Results III — EC50 Comparison: Multiplex vs. Plate Control
To assess quantitative concordance between the SemaCyte multiplex format and conventional plate-based assays, EC50 values derived from the multiplexed γH2AX dose-response curves were compared against historical well-plate control data for the same three cell lines. The results demonstrate excellent agreement across all models.
Rank ordering of cell-line sensitivity was perfectly preserved: U2OS was consistently the most sensitive model, followed by U87MG, then A549. Absolute EC50 values showed excellent concordance, with SemaCyte multiplex values within 1.2-fold of plate control estimates throughout — a striking result given the approximately six-fold reduction in cells per condition (84 ± 42 nuclei per SemaCyte replicate vs. 554 ± 245 for plate controls).
| Cell Line | Well Plate Control EC50 (µM) | SemaCyte Multiplex EC50 (µM) | Fold Difference | Rank Concordance |
|---|---|---|---|---|
| U2OS | 16 | 15 | 1.1× | 1 → 1 ✓ |
| U87MG | 49 | 43 | 1.1× | 2 → 2 ✓ |
| A549 | 57 | 48 | 1.2× | 3 → 3 ✓ |
Key Quantitative Finding
EC50 rank ordering was preserved with 100% concordance across all three cell lines, with absolute values within 1.2-fold of plate controls — using approximately 6× fewer cells per condition. This demonstrates that the SemaCyte format delivers fully pharmacologically interpretable compound potency data at substantially reduced cell number, without any compromise in data quality.
Conclusions
This collaborative study between Semarion Ltd and o2h Discovery successfully demonstrated proof-of-concept for multiplexed, SemaCyte-enabled DDR profiling using a γH2AX high-content imaging assay. The results are highly encouraging across all evaluated criteria:
1. Same-day frozen stock deployment is fully viable. All three cell lines were successfully thawed and deployed directly from cryopreserved SemaCyte stocks, with post-thaw viability ≥75% across all models. This removes the need for live cell culture prior to assay setup, offering substantial logistical flexibility and enabling scalable assay scheduling.
2. Barcode deconvolution is reliable and accessible. Semalyse successfully decoded SemaCyte barcodes from brightfield images in mixed-population wells, generating per-cell-line image sets fully compatible with standard downstream analysis in CellProfiler. The process opened new options for organising and controlling HCI experiments without additional workflow complexity.
3. Three cell lines are profiled simultaneously from a single well. Clear, sigmoidal γH2AX dose-response curves were generated for U2OS, A549, and U87MG in parallel from the same 96-well plate. Differential sensitivity across these models was captured within a single experiment — data that would otherwise require three separate plate-based assays.
4. EC50 concordance with plate controls is excellent. Cell-line sensitivity ranking was identical between the SemaCyte multiplex and plate controls, with absolute EC50 values within 1.2-fold throughout — despite using approximately 6× fewer cells per condition.
5. Embedded in-well controls work as designed. The pre-fixed U2OS population correctly intersected the high-signal plateau of the primary U2OS dose-response curve, providing a reliable cross-well anchor and validating the in-well positive control strategy without requiring additional plate resources.