Single Cell RNAseq of adult and stem cell derrived iris muscle

We performed single-cell RNA sequencing (scRNAseq) on iris tissue from three adult human donors and human embryonic stem cell (hESC)-derived whole eye organoids, termed “SEAM,” using the 10X Genomics Chromium platform. From the adult human datasets, we identified an iris muscle population based on the co-expression of PAX6 and smooth muscle proteins (ACTA2 and TAGLN). This dataset was further used to characterize the expression profile of the non-neuronal, crest-derived iris muscle. Additionally, we differentiated hESCs into whole eye organoids (SEAM) and performed scRNAseq after 7 weeks of differentiation. In the organoids, we identified cell populations corresponding to most ocular cell types, spanning neurons, distal optic cup, periocular mesenchyme, and ocular surface epithelium. Notably, we identified iris muscle-like cells in the developing eye organoids. We conducted an integrated analysis using the Seurat R package to compare the hESC-derived eye organoids with one of the adult donor irises (IRIS91 and SRAM48). This analysis revealed a shared iris muscle cluster containing cells from both datasets. The hESC-derived iris muscle represents a developing tissue with lower expression of mature muscle genes and higher expression of transcription factors, such as PAX6 and LHX2. Human globes from donors aged 50–74 years were obtained within 24 hours postmortem from the Eye-Bank for Sight Restoration, NY, and Miracle in Sight, NC, with informed consent provided by the donors or their next of kin. This study was reviewed by the Icahn School of Medicine at Mount Sinai IRB oversight and deemed not to be human subject research. We received approval for use of H9 cell lines by WiCell. This study involves single-cell RNA sequencing (scRNAseq) of two sample types: adult human iris tissue from three donors and human embryonic stem cell (hESC)-derived SEAM whole eye organoids collected at week 7 of differentiation. Sample Preparation: Frozen vials of freshly isolated adult human iris tissue and SEAM cultures were used. Cell viability was assessed using Trypan Blue staining, and only debris-free cell suspensions with >80% viability were processed. Approximately 10,000 cells were used as input for each sample. Library Preparation: Cells were processed using the 10X Genomics Chromium Controller (v3.16) and Chromium Single Cell 3’ Library and Gel Bead Kit (v3.0). Gel beads in oil microdroplets (GEMs) were created to encapsulate single cells. Reverse transcription was performed inside GEMs to generate cDNA tagged with unique molecular indices (UMIs) and cell barcodes. GEMs were broken, and cDNA was amplified and quantified using an Agilent Bioanalyzer High Sensitivity chip and Qubit analysis (Thermo Fisher). Sequencing: Libraries were sequenced on an Illumina NovaSeq 6000 platform using S4 flow cells (100 bp paired-end reads). Target sequencing depth was 50,000–100,000 reads per cell. Data Processing: FASTQ files were generated using the Cell Ranger Single-Cell Software Suite (v3.1). Reads were aligned to the GRCh38 human reference genome. Filtered matrices of unique gene counts were produced for downstream analysis. Data Filtering and Normalization: Low-quality cells (fewer than 500 unique genes) and potential doublets (more than 5,500 unique genes) were excluded. Final analysis included an average of 1,523 cells from adult human iris tissue and 6,039 cells from SEAM organoids. Data were normalized using a global-scaling method (LogNormalize) with a scale factor of 10,000. Analysis: Principal component analysis (PCA) was performed, and clustering was conducted. Dimensionality reduction was visualized using Uniform Manifold Approximation and Projection (UMAP). Differential expression analysis was performed to identify cluster-specific genes, and heatmaps were generated to visualize the top 20 differentially expressed genes per cluster. Annotation and Enrichment: Cluster annotation was based on manual curation of gene expression patterns and complemented with Enrichr gene set enrichment analysis (using the GO Biological Process 2018 database). *************************************************************** Submitter states that missing raw files are due to file loss. ***************************************************************