Design on What Was Actually Done

 Quick Analysis Plan Copied

 Long Analysis Plan

Experimental Design: Validating the NR2F1–CYP26A1 Axis in Human Retinal Organoids

Grounding evidence. Three findings define the testable axis: (1) NR2F1 loss or mutation expands the Cyp26a1-expressing retinal progenitor population ~2-fold in E14.5 mouse retinae across three Nr2f1 models (Null, E397*, R139L) by scRNA-seq; (2) human NR2F1 directly binds a conserved 56-bp CYP26A1 regulatory region in native holdup assays, with patient mutations E400* (reduced binding) and R142L (abolished binding) creating a dose-response allelic series; (3) BBSOAS patients (n=7) show ~43–58% reductions in foveal pit depth, volume, and area vs controls (n=8) NR2F1 binds CYP26A1 promoter; foveal hypoplasia. Human CYP26A1 coding variants produce loss-of-function in cell assays (wild-type raises RA metabolite output 9-fold; g.3116delT frameshift is inactive), proving enzyme activity is assayable Functional CYP26A1 polymorphism assay.

Figure 1. Reported BBSOAS OCT values (volume/area scaled ×100 for visibility; original values annotated) BBSOAS foveal pit morphometry. Small cohort; values are group means ± SD.

Phase 1 — Model establishment and perturbation arm

• Isogenic organoid panel: CRISPR-edit a human iPSC line into: isogenic WT; NR2F1 null; NR2F1 R142L (DBD mimic); NR2F1 E400* (LBD truncation mimic) — the same allelic series validated in mice and nHU assays Allelic series design.
• CYP26A1 functional arm: CYP26A1 CRISPRi/knockdown, pharmacologic inhibition (talarozole or liarozole — sensitizers in CRC WNT/RA crosstalk data), and exogenous atRA dose-response. RA feeding induces CYP26A1 expression strongly (up to ~30-fold in CRC cells) and represses RA-synthesis enzymes, defining feedback expectations ATRA-induced CYP26A1 feedback RA feedback on RA-metabolism enzymes.
• Causal rescue logic: If NR2F1 loss drives CYP26A1 misexpression, then (a) CYP26A1 knockdown or inhibition should partially rescue the Nr2f1-mutant opsin-gradient/central-specialization phenotype, and (b) exogenous RA titration should phenocopy or dose-dependently modify it. Mirror the zebrafish rescue paradigm where cyp26a1 morpholino or RAMBA restored differentiation markers in mutants CYP26A1 inhibition rescue paradigm.

Phase 2 — Readouts

Molecular axis: (i) native holdup or ChIP-qPCR at the 56-bp CYP26A1 regulatory sequence across the NR2F1 allelic series; (ii) RNAscope/qPCR for CYP26A1 spatial gradient (dorsal/equatorial vs central domain); (iii) RA-responsive transcription (RARA/RARB targets) as functional RA readout; (iv) WNT interaction controls, since WNT/β-catenin can drive CYP26A1 independently of NR2F1 WNT-driven CYP26A1 induction. Foveal/specialization endpoints: central cone enrichment, M/S-opsin gradient (dorsal S-opsin ectopia is the sensitive mouse readout; note E397* HOM showed only a non-significant trend, p=0.08 — so power the organoid endpoint accordingly), RPC pseudo-spatial gene expression by scRNA-seq analyzed in Seurat with gseapy enrichment, following the mouse pipeline Opsin gradient readout and analysis stack.

Phase 3 — Bioinformatics plan

• Pseudo-spatial scRNA-seq: annotate RPCs along dorsoventral/apical-basal axes (RETIMAT-style), quantify CYP26A1+ fraction per axis bin; expect ~2-fold expansion in NR2F1-deficient organoids.
• Direct target confirmation: map NR2F1 ChIP-seq/CUT&RUN peaks overlapping the CYP26A1 regulatory region (MACS-style calling; NR2F1 ChIP-seq on chemo-resistant organoid lines demonstrates feasibility of the approach in 3D systems NR2F1 ChIP-seq in human organoids).
• Statistics: per-organoid (biological replicate) mixed models for OCT-analog morphometry; permutation testing for spatial gradient shifts; FDR correction across the 4-arm × 3-allelic design. Minimum n≥4 organoid batches per genotype given reported effect sizes and the marginal E397* result.

Blindspots and falsification

Mouse retina lacks a fovea; the cited evidence establishes the NR2F1–CYP26A1 binding and retinal-regionalization link but does not directly demonstrate CYP26A1 misexpression in human foveal tissue — the organoid experiment tests exactly this inferred link Stated limitations of the axis evidence. Falsification: if NR2F1-null organoids show no CYP26A1 domain shift, or CYP26A1 inhibition does not rescue opsin/specialization phenotypes, the axis is organoid-context-disconfirmed. Confounders to control: WNT-driven CYP26A1 induction independent of NR2F1, retinoid content of culture media, and batch-to-batch RA signaling variability in organoid protocols.