Abstract
Introduction:
Cell-based therapies hold great promise for treating late-stage retinal degenerative diseases. However, photoreceptor cell transplantation has been limited by poor cellular integration, suggesting that an ideal population of donor cells has not yet been defined.
Methods:
Here, we utilized single-cell RNA-sequencing and lineage tracing to assess the heterogeneity of photoreceptor precursor cells during murine retinal development.
Results:
Subgroup analysis revealed three transcriptionally distinct populations of Crx + cells from postnatal day 2–6 retinas which are consistent with early (Dll1+), intermediate (Neurod4+), and late (Prom1+) photoreceptor precursor cells. Lineage tracing with subpopulation-enriched Cre mice showed that Dll1+, Neurod4+, and Prom1+ cells all generate photoreceptor cells, and captured dynamic changes in gene expression suggestive of sequential states of photoreceptor differentiation. Transcriptomic analysis revealed that similar CRX + subpopulations are present in maturing human retinal organoids.
Conclusion:
These findings highlight the heterogeneity of neonatal photoreceptor precursor cells, and could help inform future strategies to isolate an optimal population of cells for retinal regeneration.
Introduction
Retinal degeneration is a leading cause of blindness, affecting over two million people worldwide (; ; ). Vision loss is caused by the death of photoreceptor cells, the light sensitive neurons in the retina that perform phototransduction. Because mammalian photoreceptors do not regenerate, loss of these cells results in irreversible blindness. Cell-based therapies, in which healthy photoreceptor cells are transplanted into regions of retinal atrophy, are therefore an attractive approach to replace photoreceptors and restore visual function regardless of the underlying cause of disease. Numerous studies have demonstrated that donor- or induced pluripotent stem cell (iPSC)-derived photoreceptor precursor cells transplanted as dissociated cells (; ; ; ), sheets of cells (), or cells seeded on bioscaffolds () can survive in the subretinal space, form synaptic connections with host bipolar and horizontal cells, and result in detectable improvements in visual function (; ; ; ; ; ; ). However, progress in this field has been hindered by low cellular integration rates (<1%) and poor functional recovery (; ; ; ; ). In addition, many transplanted cells remain at the injection site as a subretinal mass; these unintegrated cells could impede the diffusion of choroidal oxygen and nutrients, or even lead to worsening atrophy and vision loss.
While there are many factors that undoubtedly affect the integration of transplanted cells - including the host environment, stage of disease, and immune response - one of the most critical factors is the identity of the donor cell itself. Venugopalan et al. have elegantly shown that neuronal cells that survive well in vitro also have the highest rates of in vivo transplantation survival (), suggesting that intrinsic characteristics of donor cells can directly impact the success of transplantation. Photoreceptor precursor donor cells are commonly selected using photoreceptor specific surface markers (CD73, CD133) (; ; ; ), or transgenic mice or cell lines (Nrl- or Crx-eGFP) (; ; ). However, these markers are expressed in both developing and mature photoreceptors and thus lack developmental specificity (; ; ). Prior transplantation studies have demonstrated that embryonic progenitor cells and mature photoreceptor cells have poor survival and integration potential compared to donor cells isolated from early (P2-P6) postnatal retinas (; ; ; ; ), but beyond the gross age of donor retinas, few studies have attempted to further define subpopulations of photoreceptor precursor cells for retinal regeneration (; ; ; ). Additionally, much of the literature defining this time window is confounded by cytoplasmic material transfer (CMT) (; ; ; ), in which donor cells form connections with recipient photoreceptor cells and exchange mRNAs, organelles, and proteins, including fluorescent labels (; ). Taken together, these findings suggest that an optimal population of donor cells with robust integration potential has not yet been defined, and that a better understanding of photoreceptor development may improve the efficacy of cell replacement strategies.
Photoreceptors, along with the other major retinal cell types (retinal ganglion, amacrine, horizontal, bipolar, and Müller glia cells) are derived from a common pool of retinal progenitor cells (RPCs) (; ; ) and develop in a highly stereotyped and conserved birth order (Figure 1A) (; ). Once retinal progenitor cells complete their terminal division, the nascent post-mitotic precursor cells become specified to their terminal cell fates. Photoreceptor precursor cells are marked first by the expression of the transcription factors Otx2 (; ) and Crx (; ), and foundational work has revealed the genes necessary for photoreceptor differentiation, maturation, and function including Nrl (; ; ), Rho (), and Nr2e3 (; ). In rodents, nascent photoreceptor precursor cells can take up to 2 weeks to initiate rhodopsin expression (), during which distinct developmental stages have been described (), though it is unclear if these represent progressive stages of photoreceptor precursors with functional differences. Additionally, since development occurs in temporal and spatial waves across the retina (; ; ; ; ; ), a single timepoint likely contains a heterogenous population of photoreceptor precursor cells at distinct stages of maturation that cannot be resolved by traditional markers. Such cellular heterogeneity has been shown to directly influence the differentiation potential in other stem cell systems (; ), and underscores the need to more precisely define the developmental dynamics of photoreceptor precursor cells.
FIGURE 1
Single cell RNA sequencing (scRNA-seq) has been successfully used to advance our understanding of the cellular heterogeneity of retinal bipolar (), ganglion (), and foveal versus peripheral retinal cells (). Recently, scRNA-seq analysis of fetal human retinal samples identified sub-states of immature photoreceptor precursors predisposed to form retinoblastoma tumors (), but thus far has not been extensively applied to understanding the diversity of maturing photoreceptor precursors. In this study, we utilized scRNA-seq and genetic lineage tracing to identify subpopulations of Crx + photoreceptor precursor cells in the neonatal mouse retina. These transcriptionally distinct cell populations contribute to photoreceptor cell development, and have comparable populations in human retinal organoids. Together, these findings reveal the heterogeneity of photoreceptor precursor cells in the developing retina which may help guide future retinal regenerative strategies.
Materials and methods
Mice
Dll1-GFP-IRES-Cre-ERT2 () mice were generated by Dr. Hans Clevers (Hubrecht Institute, Netherlands) and obtained from Dr. Yibin Kang (Princeton University). Neurod4-CreERT2 () mice were cryorecovered from the Riken Institute (CDB0511T). Prom1-CreERT2 () mice were cryorecovered from The Jackson Laboratory (B6N; 129S-Prom1tm1 (cre/ERT2)Gilb/J; strain 017743). Ai9 (B6.Cg-Gt (ROSA)26Sortm9 (CAG-tdTomato)Hze/J, strain 007909) () and WT C57BL/6J (strain 000664) mice were obtained from The Jackson Laboratory. For lineage tracing experiments, Cre recombination was induced with a single oral delivery of tamoxifen (; ) (50 mg/kg body weight) using flexible gavage needles (FTP-20-38-50, Instech Labs Inc) given to pups at P1 or P2. Tamoxifen free base (T5648, Sigma-Aldrich) was prepared at a stock concentration of 20 mg/mL in corn oil (C8267, Sigma). Eyes were harvested at P4, P6, P8 and P30 as indicated. Genotyping was confirmed prior to all experiments, and both male and female mice were used in these studies. Genotyping primers used in this study are listed in Supplementary Table S1. All animals were bred and maintained in the animal facilities of The University of Pennsylvania under a 12-h light/12-h dark cycle with free access to food and water.
Retinal dissociation
Eyes were enucleated and retinas were dissected in cold 1xDPBS (all buffers used in this study lacked Ca2+ and Mg2+ unless specifically noted otherwise). For single-cell RNA-sequencing, four to eight retinas from C57BL/6J mice were pooled for each timepoint. For flow cytometry analysis, one to two retinas were used for each sample and timepoint. Retinas were enzymatically digested in papain (0.6 mg/mL papain, LS003119, Worthington; 0.2 mg/mL L-Cysteine, C-7352, Sigma; 0.02 mg/mL DNase I, D4527, Sigma; 1xDPBS) at 37 °C for up to 10 min. Papain digestion was stopped by Lo-Ovo trypsin inactivator solution (1.5 mg/mL, LS003587, Worthington; 1.5 mg/mL BSA; 0.02 mg/mL DNase I, D4527, Sigma; 1xDPBS, pH 7.4). Retinas were then mechanically digested by gentle trituration and additional DNase I was added as needed to minimize tissue clumping. Dissociated cells were then filtered through a 40 μm cell strainer to remove aggregates and pelleted by centrifugation at 100 g for 10 min at room temperature.
Fluorescence-activated cell sorting (FACS)
For flow and qPCR analysis of neonatal lineage traced cells, dissociated retinal cells from Cre + pups were resuspended in FACS buffer (1% BSA, 1xHBSS) and labeled with DAPI (0.5 ug/mL), CD133-AF647 (1:100, AB_2566013, BioLegend), and CD73-PE/Cy7 (1:100, AB_2716103, BioLegend), and analyzed and sorted on a BD FACS Symphony S6 Sorter (BD Bioscience). Cells from WT littermates were labeled with DAPI only and used for compensation, along with AbC Total Antibody Compensation beads (A10513, Invitrogen) labeled with each antibody. tdTomato compensation beads were prepared by conjugating beads with anti-His antibody (1:500, AB_914704, GenScript) and His-tagged recombinant tdTomato protein (1 μg, TP790045, OriGene). GFP BrightComp eBeads Compensation Bead (A10514, Invitrogen) were also used for the Dll1-Cre;Ai9 sample analysis to account for the EGFP reporter. Samples were gated on forward and side scatter area, height, and width to remove debris and doublets, and DAPI to remove dead cells. For downstream qPCR analysis, 50,000-200,000 tdTomato + cells were isolated for each sample and timepoint. For flow analysis, 10,000 tdTomato + cell events were recorded, and data were processed using FlowJo (version 10.10.0). Full minus one (FMO) controls were used to set up Prom1-AF647 and CD73-PE/Cy7 negative gates.
Single cell RNA sequencing
Dissociated retinal cells were purified by FACS prior to sample submission. Cell pellets were resuspended in FACS buffer, labeled with DAPI (0.5 μg/mL), and analyzed and sorted on a BD FACS Aria Cell Sorter (BD Bioscience). Samples were gated on forward and side scatter area, height, and width to remove debris and doublets, and DAPI to remove dead cells. Following FACS purification, sorted cells were manually counted using Trypan Blue (25-900-CI, Corning) and a hemocytometer. Approximately 20,000 live cells were resuspended at 700–1220 cells/µL (0.04% BSA, 1xDPBS) and submitted to the Center for Applied Genomics (CAG) Core at the Children’s Hospital of Philadelphia (CHOP) for library preparation and single-cell RNA-sequencing using the 10x Genomics platform, and preliminary CellRanger processing. ScRNA-seq analysis was performed in RStudio (build 2024.04.2 + 764, R version 4.1.2) using the Seurat package () (version 4.3.0). Dying cells were filtered out by excluding cells with >15% mt-RNA. Additional filtering was performed to remove empty cells and potential doublets (300 < nFeature_RNA <4000). The clusters with the highest expression of Crx were identified and used for subset analysis of photoreceptor cells. Clustering for subset analysis used resolutions of 0.2 (mouse Crx + P2-P6 subset) and 0.25 (human organoid CRX + subset). Pseudotime analysis was performed using the Monocle3 package (; ; ; ) (version 1.2.9) and Ccnd1 or Olig2 () as the reference gene. Previously published scRNA-seq datasets for mouse and human retinal organoids were accessed from GSE118614 (), GSE153674 (), GSE142526 (), and GSE235577 (). Primary scRNA-seq data generated in this study have been deposited at the National Center for Biotechnology Information (NCBI) Gene Expression Omnibus (GEO) at GEO accession number GSE297036.
RNA isolation & qPCR analysis
For qPCR analysis of developing mouse retinas, eyes were harvested from WT mice at E14, E16, P0, P3, P4, P5, P6, P15, and P30. Two retinas from C57BL/6J mice were pooled for each timepoint. Following enucleation, retinas were isolated from the RPE and choroid, collected into 1 mL of TRIZOL reagent (15596026, Thermo Fisher), and gently digested with a Dounce homogenizer. For FACS isolated samples, cell pellets were resuspended in 1 mL of TRIZOL reagent, and gently digested by trituration. RNA was isolated following manufacturer’s instructions. cDNA synthesis was performed using the High-Capacity cDNA Reverse Transcription Kit (4374966, Applied Biosystems). Primers were designed using the NCBI Primer BLAST tool (). qPCR was performed in triplicate using SybrGreen reporter following manufacturer’s instructions (A25741, Applied Biosystems). Relative average fold changes were calculated using ΔΔCt analysis, using B2m as the housekeeping gene. E14 and P4 samples were used as the references for the developing mouse retina analysis and FACS isolated samples, respectively. The qPCR primers used in this study are listed in Supplementary Table S2.
Histology and immunofluorescence
Following enucleation, the cornea and lens were removed, and eye cups were fixed in paraformaldehyde (4%, 1xDPBS) for 1-2 h at room temperature or overnight at 4 °C. Eye cups were then placed into sucrose solution (30%, 1xDPBS) at 4 °C overnight, embedded in Tissue-Tek OCT freezing media (VWR), and processed on a cryostat (Leica) to prepare 12 µm sections. Slides were air dried and stored at −20 °C prior to immunofluorescence analysis. Sections were first rehydrated with 1xDPBS (x3 5-min washes), and then blocked in IF permeabilization buffer (5% horse serum, 1% goat serum, 3% BSA, 0.01% TritonX-100, 1xDPBS) for 1 h at room temperature. Sections were then washed in PBT buffer (0.1%Tween-20, 1xDPBS; x3 5-min washes), and incubated in primary antibodies diluted in IF buffer (5% horse serum, 2% goat serum, 3% BSA, 1xDPBS) overnight at 4 °C. Following 1xPBT washes (x3 5-min washes), sections were incubated in secondary antibody diluted in IF buffer for 1 hour at room temperature. Sections were washed again with 1xPBT (x3 5-min washes) and then counterstained with Hoechst-33342 (2.5 μg/mL, 1xDPBS) for 10 min at room temperature. Retinal sections were washed with 1xDPBS (x2 5-min washes) and covered with ProLong Gold Antifade Mountant (P36930, Invitrogen), and cover slips were sealed with nail polish. The following primary antibodies were used: anti-RFP (1:250, AB_2209751, Rockland), anti-CD73 (1:350, AB_1089066, BioLegend), anti-Rhodopsin (1:500, AB_10696805, abcam), anti-Cone Arrestin (1:500, AB_1163387, Millipore), anti-Pax6 (1:2000, AB_2565003, BioLegend), anti-Vsx2 (1:50, AB_2216010, Invitrogen), anti-Glutamine Synthetase (1:600, AB_1950421, GeneTex). The following secondary antibodies were used: Alexa Fluor 488 Goat anti-Rabbit (1:500, AB_2576217, Invitrogen), Alexa Fluor 647 Goat anti-Rabbit (1:500, AB_2535813, Invitrogen), Alexa Fluor 750 Goat anti-Rabbit (1:500, AB_2535710, Invitrogen), Alexa Fluor 647 Goat anti-Rat (1:500, AB_141778, Invitrogen), CF 488A Donkey anti-Sheep (1:500, AB_10583179, Biotium). Slides were imaged with a Zeiss 980 confocal microscope. For consistency, images were captured from central retinal sections proximal to the optic nerve. When necessary, images were taken from more peripheral areas to capture examples of rare bipolar and Müller glia cell labeling in the Neurod4-Cre + or Prom1-Cre + linage tracing samples. Images were processed and quantified in FIJI (ImageJ, version 1.54f) ().
RNA fluorescent in situ hybridization (RNA-FISH)
RNA-FISH was performed on 12 µm frozen retinal sections using the Advanced Cell Diagnostics RNAscope Multiplex Fluorescent Reagent Kit v2 (323100, Bio-Techne). Manufacturers’ instructions were followed with the following exception: target retrieval was performed manually, using 700 mL of 1x Target Retrieval Reagent (5 min at 100 °C). Probe signals were developed using Opal 520 (1:2000, FP1487001KT, Akoya BioSciences), Opal 570 (1:1500, FP1488001KT, Akoya BioSciences), and TSAVivid 650 (1:1500, 323273, Bio-Techne) reporters diluted in TSA Buffer.
Quantification and statistical analysis
Cell counts and quantification for the P30 lineage tracing samples were performed using maximum intensity projection images and the multipoint tool in FIJI. Cell counts were taken from a 353.33 µm2 area. The total number of tdTomato + cells was counted from 3-4 sections from 5-7 eyes per Cre line. The number of tdTomato + cells coexpressing INL cell markers was counted from 3-5 sections from 3-5 eyes. For counting the number of tdTomato + cells in the ONL of Neurod4-Cre + samples, a cropped image 1/10th the width of the imaged ONL was used and multiplied by 10 to estimate the total number of tdTomato + ONL cells. The number of tdTomato + cells in the ONL or INL was then divided by the total number of tdTomato + cells per image to calculate the proportion of labeled cells in each retinal layer. The number of tdTomato + cells coexpressing INL cell markers was divided by the estimated average number of tdTomato + INL cells to calculate the proportions of labeled amacrine, bipolar or Müller glia cells. To account for any inter-eye effects from the left and right eyes of our samples, we utilized generalized estimating equations (GEE) analysis to calculate the standard error of the means (SEM) (). All datasets were approximately normal (as visualized by Q-Q plots) except for the counts of Müller glia cells, which were analyzed with Poisson distribution. The proportions of labeled ONL and INL cells were analyzed by Least Squares Means (LSM) with Bonferroni post-hoc test (α = 0.05) for multiple comparisons. GEE and LSM analysis was performed in SAS Studio (version 3.81). For the flow cytometry analysis of the FACS isolated lineage tracing time course samples, proportions of Prom1+/CD73+ cells were assessed with one-way ANOVA with Tukey’s post-hoc test (α = 0.05) for multiple comparisons. For the qPCR analysis of the FACS isolated lineage tracing time course samples, gene expression changes were assessed with one-way ANOVA of the ΔCt values with post-hoc Dunnett’s test (α = 0.05) using the P4 timepoint as the reference to perform multiple comparisons. Data were compiled and graphed using GraphPad Prism version 10.4.1 for Windows (GraphPad Software, Inc; La Jolla, CA).
RESULTS
Generating a single-cell RNA-Sequencing dataset of murine retinal development
To better understand the fundamental heterogeneity of photoreceptor cells during normal development (Figure 1A), we characterized the transcriptional profile of the mouse retina at E14, E16, P0, P3, and P6 using scRNA-seq (Figure 1B; Supplementary Figure S1A). Given the relatively rapid nature of transcriptional changes that can occur during development, we selected these timepoints to complement existing scRNA-seq datasets during the early postnatal period, a critical window for rod photoreceptor development. By integrating our scRNA-seq data with additional timepoints from prior studies (), we developed a time resolved atlas of transcriptional changes every 24-48 h during this important developmental window that has been increasingly used as a source of cells for transplantation (Figure 1B, yellow box). Additionally, we incorporated data from adult (P60) retinas to include signatures from fully matured cells () (Figure 1B; Supplementary Figure S1B). ScRNA-seq data were integrated for analysis (Figure 1C) (), faceted by sample age (Figure 1D), by data source (Supplementary Figure S1B) and by cell cycle regression analysis (Supplementary Figure S1C). Panels of retinal cell type marker genes were used to assign cluster identities (Figure 1E). Retinal progenitor cells were most abundant in the early timepoints, and appeared along with early-born retinal ganglion, horizontal, and cone photoreceptor cells. Diminishing numbers of progenitor cells and increasing numbers of late-born bipolar, Müller glia, and rod photoreceptor cells were detectable in postnatal retinas (Figure 1D). In addition to the retinal progenitor and mature retinal cells, we also identified several intervening precursor populations based on marker gene expression and adjacent clusters (Figure 1C). Precursor 2 (Otx2+/Crx+) and Precursor 3 (Otx2+/Vsx2+) are photoreceptor and bipolar precursors, respectively, with Precursor 1 (Otx2+) potentially shared photoreceptor/bipolar precursors. Precursor 4 (Pou4f2+) and Precursor 5 (Tfap2b+) appear to be ganglion and amacrine precursors, respectively. These trajectories were further corroborated by pseudotime analysis (Supplementary Figure S1D) (; ; ; ). All major retinal cell types were captured and demonstrated expected developmental trajectories.
Subgroup analysis of Crx + cells in P2-P6 retinas reveals distinct subpopulations of photoreceptor precursor cells
Prior studies of photoreceptor precursor transplants have reported that cells collected from early postnatal retinas (P2-P6) have the highest rates of donor cell survival and integration compared to cells isolated from earlier or later timepoints (; ; ; ; ), though these reports warrant reevaluation due to cytoplasmic material transfer (; ; ; ). This time period coincides with the peak of rod photoreceptor genesis, and an enrichment of precursor cells expressing cone-rod homeobox (Crx), a transcription factor required during retinal development for photoreceptor specification and maturation () (Figure 1E; Supplementary Figure S1E). To investigate the transcriptional profile of developing photoreceptor cells during this stage, we performed a subgroup analysis of the early postnatal period (P2-P6) (Figure 2A). We identified the trajectory of photoreceptor cell development by identifying clusters with strong Crx and Otx2 expression (Figures 2B,C). Subgroup analysis of these cells in P2-P6 retinas (Figure 2A, dashed outline) revealed five clusters (Figure 2D). Based on differentially expressed marker genes (Figures 2E,F) and pseudotime analysis (Figure 2G), these include mature cone (purple) and rod (yellow) photoreceptor cells, as well as three clusters that correspond to early (green), intermediate (cyan), and late (magenta) stage photoreceptor precursor cells (Figure 2D). Cell cycle regression analysis showed that all clusters had low S and G2M scores, suggesting that these are post-mitotic cells (Supplementary Figure S2A). The earliest group (green) is enriched in genes associated with neural specification (Neurog2, Ascl1, Pax6) (Figure 2E; Supplementary Table S3), suggesting that these are precursor cells that have recently exited the cell cycle and begun differentiating. The intermediate group (cyan) is marked by genes associated with photoreceptor development (Neurod4, Abca4, Sox4), and cytoskeletal dynamics and axonal positioning (Tubb3, Robo2, Stmn2) (Figure 2E; Supplementary Table S4). The late group (magenta) is associated with late-stage precursors and maturing photoreceptor markers (Prom1, Nrl, Nr2e3) (Figure 2E; Supplementary Table S5).
FIGURE 2
We then identified candidate marker genes enriched in the early (Dll1+), intermediate (Neurod4+), and late (Prom1+) groups (Figure 2F). Gene expression was validated by qPCR of retinal lysates from E14-P30 mice (Figure 2H; Supplementary Table S6), and by RNA-FISH of retinal sections from P4 mice (Figure 2I). Interestingly, these markers have distinct spatial expression patterns in the developing neuroblast layer (dashed lines in Figure 2I). Dll1 (a marker of the early group) is expressed throughout the neuroblast layer in the P4 retina, Neurod4 (a marker of the intermediate group) is expressed in the medial neuroblast layer, and Prom1 (a marker of the late group) is expressed most strongly in the apical neuroblast (Figure 2I). Importantly, we observed expression of all three subpopulation markers in the outer neuroblast, consistent with Crx expression (Supplementary Figure S2B) and the location of the presumptive outer nuclear layer where mature photoreceptors ultimately reside. These data suggest that photoreceptor precursor cells are heterogeneous, and consist of several transcriptionally distinct subpopulations that coexist in the retina at the same developmental age.
Dll1+, Neurod4+, and Prom1+ cells generate mature photoreceptors
To determine if the early (Dll1+), intermediate (Neurod4+), and late (Prom1+) subpopulations of Crx + cells identified by our unbiased scRNA-seq analysis contribute to the pool of mature photoreceptors, we performed genetic lineage tracing experiments using inducible Cre reporter lines driven by candidate marker genes. Each promoter was chosen based on a search of candidate genetic markers that were enriched in each population (hence the naming of clusters) (Figure 3A). Dll1-CreER (), Neurod4-CreER (), and Prom1-CreER () mouse lines were used to trace the early, intermediate, and late photoreceptor precursor groups, respectively. Cre mice were crossed with the Ai9 reporter mouse (Rosa26-CAG-LSL-tdTomato) (), so that upon tamoxifen administration, all cells expressing the subpopulation-enriched promoters were permanently labeled with tdTomato (Figure 3B). Tamoxifen was administered orally () at a single dose of 50 mg/kg () to one- or two-day old pups to allow for recombination to occur in the neonate retinas. Eyes were then harvested at P4 and P30 for histological analysis (Figure 3C). Dll1-CreER;Ai9, Neurod4-CreER;Ai9, and Prom1-CreER;Ai9 litters that did not receive tamoxifen all had minimal background signal at P4 and P30 timepoints, validating the fidelity of the inducible system (Supplementary Figure S3A,B). Mice treated with tamoxifen at P1 and P2 had normal retinal morphology (Figures 3D,E), suggesting that this tamoxifen dose does not grossly affect retinal development. Cells labeled with the tdTomato reporter were observed in P4 pups from crosses with all three Cre lines. To increase the signal-to-noise ratio of the tdTomato reporter, retinal sections were stained with anti-RFP antibody prior to imaging. The antibody signal strongly coexpressed with the endogenous tdTomato signal (Supplementary Figure S3C,D). Labeled cells in the Dll1-Cre + mice were located throughout the neuroblast layer. Labeled Neurod4-Cre + cells were found in a striking band medially to basally in the neuroblast layer as well as diffusely in the apical neuroblast layer. Labeled cells in the Prom1-Cre + mice were located apically and occasionally spanning the neuroblast layer (Figure 3D). The spatial patterns of these lineage traced cells in the P4 retina are consistent with the expression patterns of marker genes by RNA-FISH at this timepoint (Figure 2I). We next analyzed retinal sections from mice lineage traced to P30 to assess the terminal fates of the labeled cells in adult retinas. tdTomato + cells with photoreceptor cell morphology were present in the outer nuclear layer (ONL) of retinas from all three Cre lines (Figure 3E), and coexpressed with the photoreceptor marker CD73 (Figures 3F–I). The majority of labeled cells were rods based on cellular morphology (long, cylindrical outer segments) and co-staining with rhodopsin (Supplementary Figure S4A,C–E). Cone arrestin appeared to be less frequently expressed in tdTomato + cells compared to rhodopsin (Supplementary Figure S4B,F–H). Together, our lineage tracing experiments suggest that the early, intermediate, and late subpopulations all contribute (albeit not exclusively) to the photoreceptor fate in the mouse retina.
FIGURE 3
While a majority of labeled cells at P30 were photoreceptors, we also observed labeled cells in the inner nuclear layer (INL) (Figure 3E; Supplementary Figure S5). This is likely due to Cre expression in a small percentage of late RPCs and precursor cells outside of the photoreceptor lineage (Figure 3A). On average, Prom1-Cre labeled the greatest number of photoreceptor cells (96.6%) and the fewest non-photoreceptor cells (3.4%) compared to Dll1-Cre (76.7% PR/23.3% INL) and Neurod4-Cre (88.8% PR/11.2% INL) (Supplementary Figure S5M). The morphology and location of these labeled INL cells were most consistent with amacrine, bipolar, and Müller glia cells (Supplementary Figure S5A–C). We performed additional immunofluorescence analysis of the P30 lineage tracing samples to co-stain with known markers of amacrine cells (Pax6) (Supplementary Figure S5D–F), bipolar cells (Vsx2) (Supplementary Figure S5G–I), and Müller glia cells (Glutamine synthetase [GS]) (Supplementary Figure S5J–L). Of the non-photoreceptor cells, Dll1-Cre + retinal sections labeled amacrine, bipolar, and Müller glia cells, while the Neurod4-Cre+ and Prom1-Cre + samples labeled primarily amacrine and bipolar cells with rare Müller cells (0.3% and 4.0%, respectively) (Supplementary Figure S5N). No labeled retinal ganglion cells or horizontal cells were observed. Together, these data suggest that Dll1, Neurod4, and Prom1 are enriched but not exclusive to each photoreceptor precursor population, and that all three populations contribute to the mature photoreceptor pool.
Dll1+, Neurod4+, and Prom1+ subpopulations are consistent with sequential photoreceptor precursor states
We next sought to determine if the early, intermediate, and late subpopulations are sequential stages of photoreceptor precursor maturation as suggested by the pseudotime analysis (Figure 2G), or parallel pathways that can each directly form photoreceptors. To test this, we treated Dll1-CreER;Ai9, Neurod4-CreER;Ai9, and Prom1-CreER;Ai9 litters with tamoxifen at P2, and analyzed tdTomato + cells at P4, P6, and P8 by flow cytometry and qPCR (Figure 4A; Supplementary Figure S6A–F). Cells were stained with antibodies for Prom1 (CD133) (a marker of the late group) and CD73 (a marker of maturing photoreceptors). The proportions of Prom1 and CD73 double positive cells were measured at P4, P6, and P8 from tdTomato + cells (Figures 4B–C; Supplementary Figures S6A–D). Dll1-Cre + began with the fewest Prom1+/CD73+ cells at P4 (44.0%), which strongly increased by P6 (77.3%) and P8 (89.4%) (Figures 4B,C; Supplementary Figure S6E), suggesting that Dll1+ cell populations acquire Prom1 expression over time. In contrast, Prom1-Cre + samples did not have a significant increase in Prom1+/CD73+ expression from P4 to P8 (Supplementary Figure S6C–E), which would be expected if these cells were already enriched for Prom1 at P4. Interestingly, Prom1-Cre + samples showed a steady increase in CD73+ expression over time (Supplementary Figure S6C), which is consistent with our qPCR analysis (Figure 4L) and suggests that Prom1 cells are in the later stages of photoreceptor development and maturation, and that Prom1 marks an earlier cell state than CD73.
FIGURE 4
We then analyzed gene expression changes of representative panels of early, intermediate, and late gene markers in lineage traced cells at P4, P6, and P8 by qPCR (Figure 4D–L; Supplementary Table S7). The Dll1-Cre + samples showed the most dynamic gene expression from P4 to P8. Early gene markers were strongly downregulated (Figure 4D, blue boxes), while some intermediate and nearly all late gene markers were strongly upregulated over time (Figures 4E–F, orange boxes), suggesting that permanently labeled “early” population cells may subsequently acquire a signature of the intermediate and late groups. Similarly, the Neurod4-Cre + samples had low expression of early gene markers, accompanied by a gradual but strong upregulation of late marker genes from P4 to P8 (Figures 4G–I). Finally, Prom1-Cre + cells showed low expression of early and intermediate gene markers at P6 and P8 (Figures 4J–K) and strong upregulation of late gene markers associated with photoreceptor maturation (Figure 4L). Although these experiments were conducted on bulk-sorted tdTomato + cells and therefore cannot resolve the transition states of individual cells, these findings are consistent with–though not definitive evidence for–a sequential model of photoreceptor development in which Dll1+, Neurod4+, and Prom1+ cells progress through distinct developmental states along their trajectory to mature photoreceptors. It would be interesting to employ single-cell-resolution approaches in future studies to more precisely define these transition states.
CRX + photoreceptor precursor subpopulations are present in human retinal organoids
For future therapeutic applications, it would be advantageous to obtain photoreceptor precursor donor cells from cultured sources rather than primary retinal tissues. Human induced pluripotent stem cell (hiPSCs)-derived retinal organoids are three-dimensional structures that recapitulate the development of the human retina, and have become an increasingly popular source of donor cells for photoreceptor transplantation. To investigate the heterogeneity of photoreceptor precursor cells in retinal organoids and to determine if the Crx + subpopulations identified in this study are conserved across species, we next compared our mouse data to published scRNA-seq datasets of human retinal organoids (Supplementary Figure S7A,B) (; ). All expected retinal cell types were represented in human retinal organoids (Figure 5A; Supplementary Figure S7C). Subgroup analysis of CRX + cells revealed several clusters of maturing human photoreceptor precursor cells (Figures 5B,C; Supplementary Figure S7D). Differentially expressed gene analysis of the clusters identified cone (group 0, purple), and rod (group 1, yellow) photoreceptors (Figures 5C–E). Groups 2 (green) and 4 (magenta) showed enrichment for genes bearing some resemblance to the early and late groups from our mouse scRNA-seq data, respectively. Although expression of DLL1, the marker of the early mouse group, was not detected in the human CRX + subset, we did observe enrichment of PAX6, an additional marker of the early group (Supplementary Table S3) in group 2. Similarly, PROM1 was enriched in group 4, the presumptive late group (Figure 5D). Interestingly, while neither group had as many differentially expressed genes directly related to photoreceptor development, many genes were related to cellular migration and localization (MDK, FBLN1, KIF22). Lastly, group 3 (cyan) shared many transcriptional signatures with the intermediate group in the developing mouse retina such as NEUROD4 (Figure 5D,E), and showed strong enrichment for genes related to neurogenesis and axon development (NEUROD1, NFIB, ATOH7, PRDM1). Overall, these data suggest that human retinal organoids contain heterogeneous populations of photoreceptor precursor cells that may be evolutionarily conserved across species. Further studies validating the presence of these subpopulations and their utility in cell-based therapies will improve our understanding of retinal development and may refine a suitable source of donor cells for retinal regeneration.
FIGURE 5
Discussion
Retinal development is a complex process, with all seven major retinal cell types emerging from a common pool of retinal progenitor cells in overlapping temporal and spatial waves. This results in a heterogeneous pool of progenitor, precursor, and maturing retinal cells. This is especially pertinent to photoreceptors, as many of the genes necessary for photoreceptor precursor cell development, such as Crx and Nrl, are also required for the maintenance and function of mature photoreceptors cells. While the genes regulating photoreceptor development have been extensively characterized, the developmental heterogeneity of photoreceptor precursor cells is poorly understood, particularly as it relates to isolating an optimal population of cells for regenerative strategies (; ; ). Recently, there have there been reports of precursor substates in developing fetal retinas (), as well as brains (), highlighting the heterogeneity of neural precursor cells. Reducing donor cell heterogeneity results in increased RPE cell engraftment and polarization (), enhanced cortical cell survival and cell fate control (; ), and improved cardiac functional recovery (; ), suggesting that a similar approach may be beneficial in the neural retina.
Previous reports have suggested that donor cells derived from the early neonatal period (P2-P6) exhibit improved integration upon transplantation compared to earlier or later timepoints (; ; ; ; ), therefore we focused our transcriptomic analysis on this important developmental window. Interestingly, we observed enrichment of Crx + precursor cells during this time, with a sharp decline by P8, suggesting that the presence of this population may be important for the survival and functional integration of transplanted cells. Of note, in light of cytoplasmic material transfer that may have confounded these studies, it would be interesting to validate that P2-P6 donor cells truly have the highest rates of integration (rather than the highest rates of material transfer). Most lineage traced photoreceptors were rods, rather than cones, which is consistent with the timing of tamoxifen administration during the peak of rod genesis in the neonatal period. It would be interesting to expand our transcriptomic analysis to investigate the heterogeneity of embryonic timepoints, which contain early-born retinal precursor cells that may be better suited for cone regeneration. Importantly, we show that even at a single (P4) timepoint, photoreceptor precursor cells have a great degree of heterogeneity, suggesting that the identification of an optimal cell identity, rather than optimal chronologic age, may provide a more useful strategy for achieving transplantation success for the desired cell type.
Our unbiased scRNA-seq analysis revealed three transcriptionally distinct subpopulations of Crx + photoreceptor precursor cells present within the neonate mouse retina. The early (Dll1+) group is predominantly marked by genes that promote neuronal differentiation (Pax6 (), Ascl1 (), Neurog2 (), Hes6 ()), suggesting that these are immature precursor cells committing to neuronal cell fates. The intermediate (Neurod4+) group is marked by pro-neural and visual function related genes (Neurod4, Dll3, Abca4). Notably, this group is also marked by the enrichment of genes associated with cytoskeletal dynamics and axon formation and guidance (Tubb3, Tubb2b, Robo2, Gap43), which suggests that these cells may be intrinsically poised for cell motility and migration–important prerequisites for transplantation success. Finally, the late (Prom1+) group is marked by genes related to photoreceptor maturation (Nr2e3, Nrl, Prdm1, Rom1) as well as synapse dynamics (Dnm3, Snap25), suggesting that they are committed photoreceptor precursors in the late-stages of differentiation. Interestingly, the early population acquired gene expression signatures of the intermediate and late groups over time, suggesting that these populations represent a spectrum of development with an increasing degree of photoreceptor maturation. These trends were further supported by our flow cytometry analysis, in which bulk-sorted tdTomato + samples showed sequential increases in the proportions of Prom1+/CD73+ cells over time. While cells from all three subpopulations contribute to the pool of mature photoreceptors, other retinal cell types were also labeled. To avoid inherent limitations of lineage tracing studies which rely on the fidelity of a single reporter (which in this study were enriched but not exclusive to the photoreceptor lineage) and that can reflect differences in promoter strength rather than true cellular abundance, it would be interesting in future studies to investigate only Dll1+, Neurod4+, and Prom1+ cells that co-express canonical photoreceptor markers, and to utilize single-cell resolution studies to definitively prove sequential maturation of these transition states.
Retinal organoids are an attractive source of photoreceptor precursor cells for retinal transplantation in human patients. Organoids recapitulate retinal development, can generate all retinal cell types, have scalable production, and can be generated from patient-derived cells to minimize immunogenicity. However, transplantation studies of cells from human retinal organoids exhibit similar limitations observed in mouse studies including limited donor cell integration and functional recovery. Furthermore, while many developmental patterns are shared between human and mouse, cross-species comparisons are critical to identify conserved pathways with therapeutic potential (). By analyzing previously published human scRNA-seq datasets (; ), we show that maturing human retinal organoids also have extensive heterogeneity of CRX + developing photoreceptor precursor cells that bear similarities to the Crx + subpopulations in the mouse. Our transcriptomic analysis suggests that these populations may be present as early as day (D) 45/60. Although organoids do not fully recapitulate in vivo retinal development, validating these findings in primary retinal organoids could reveal an earlier source of donor cells than the D200 or older retinal organoids commonly used for transplantation. Further investigation into the heterogeneity and dynamics of photoreceptor precursor cells in maturing human retinal organoids may help identify cell subpopulations with a greater capacity for integration.
Due to the overlapping nature of retinal development and the co-existence of the early, intermediate, and late photoreceptor cell populations at the same chronologic age, current donor cell isolation methods using broad photoreceptor markers cannot distinguish between these three groups. Ultimately, transplantation experiments using donor cells from isolated photoreceptor precursor populations will determine which has the highest rates of survival, synaptic connectivity, and functional integration within the retina. Based on our transcriptomic analysis, we hypothesize that the early (Dll1+) group may be ideally suited for neurogenesis and axon guidance, but these cells may require niche signaling and endogenous developmental cues that may be absent in diseased or degenerating retinas. The intermediate (Neurod4+) group in many ways represents an ideal donor cell population as it is marked by genes related to both photoreceptor development and neuronal cell migration and maturation. The late (Prom1+) group is most biased towards photoreceptor fate, but may have similar limitations as mature photoreceptors including poor cell survival following the mechanical stress of retinal dissociation and transplantation. The differentially expressed genes identified in our scRNA-seq data may also be leveraged to identify candidate genes to “prime” photoreceptor precursor cells prior to transplantation, or to direct the differentiation of human iPSCs or organoids towards an optimal cell type for neuronal integration. Further investigation into the genetic elements that regulate and drive the developmental transitions between states will improve our understanding of photoreceptor precursor maturation and may better inform future therapeutic strategies.
Statements
Data availability statement
Single-cell RNA-seq data have been deposited at the National Center for Biotechnology Information (NCBI) Gene Expression Omnibus (GEO) at GEO: accession number GSE297036, available at https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE297036.
Ethics statement
Ethical approval was not required for the study involving humans in accordance with the local legislation and institutional requirements. Written informed consent to participate in this study was not required from the participants or the participant’s legal guardians/next of kin in accordance with the national legislation and the institutional requirements. The animal study was approved by University of Pennsylvania IACUC protocol #806931. The study was conducted in accordance with the local legislation and institutional requirements.
Author contributions
JY: Conceptualization, Data curation, Formal Analysis, Investigation, Methodology, Writing – original draft, Writing – review and editing. ZW: Investigation, Methodology, Writing – review and editing. KG: Formal Analysis, Investigation, Methodology, Writing – review and editing. KB: Investigation, Writing – review and editing. EY: Investigation, Writing – review and editing. KU: Conceptualization, Data curation, Formal Analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Writing – original draft, Writing – review and editing.
Funding
The author(s) declared that financial support was received for this work and/or its publication. JJY was supported in part by NIH T32 HD083185. KTB and EY were supported by the National Foundation for Cancer Research (NFCR) and the Ware Bluegrass Foundation. KEU was supported by the NIH/NEI (K08EY031754 and R01EY036824), the Foundation Fighting Blindness (CD-CL-0823-0868-UPA), a Commonwealth of PA Health Research Formula Fund (HRFF) Award, the National Foundation for Cancer Research (NFCR), and an unrestricted grant from Research to Prevent Blindness to the University of Pennsylvania. ScRNA-seq samples generated in this manuscript were processed in the Penn Cytomics and Cell Sorting Shared Resource Laboratory at the University of Pennsylvania which is partially supported by the Abramson Cancer Center NCI Grant (P30 016520). The research identifier number is RRid:SCR_022376. Confocal microscopy was performed at the University of Pennsylvania Cell & Developmental Biology (CDB) Microscopy Core (RRID SCR_022373).
Acknowledgments
The authors thank Ahmara Ross, Ryan Passino, Steven Chomistek, and Robin Wilder for critical review of the manuscript, and Dr. Gui-shuang Ying for helpful discussions regarding the statistical analysis of correlated data arising from measurements of both eyes. We thank Hans Clevers and Yibin Kang for generously providing the Dll1-GFP-IRES-Cre-ERT2 mouse, and the Center for Applied Genomics at CHOP for their assistance with scRNA-seq data generation. We thank Joseph Altemus, Angela Borrelli and the University of Pennsylvania University Laboratory Animal Resources (ULAR) staff for the daily care of animals.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fcell.2026.1814134/full#supplementary-material
References
1
AavaniT.TachibanaN.WallaceV.BiernaskieJ.SchuurmansC. (2017). Temporal profiling of photoreceptor lineage gene expression during murine retinal development. Gene Expr. Patterns23-24, 32–44. 10.1016/j.gep.2017.03.001
2
AghaizuN. D.KruczekK.Gonzalez-CorderoA.AliR. R.PearsonR. A. (2017). Pluripotent stem cells and their utility in treating photoreceptor degenerations. Prog. Brain Res.231, 191–223. 10.1016/bs.pbr.2017.01.001
3
BaeS.BesshoY.HojoM.KageyamaR. (2000). The bHLH gene Hes6, an inhibitor of Hes1, promotes neuronal differentiation. Development127, 2933–2943. 10.1242/dev.127.13.2933
4
BaiY.HeH.RenB.RenJ.ZouT.ChenX.et al (2024). Sstr2 defines the cone differentiation-competent late-stage retinal progenitor cells in the developing mouse retina. Stem Cells Transl. Med.13, 83–99. 10.1093/stcltm/szad073
5
BartschU.OriyakhelW.KennaP. F.LinkeS.RichardG.PetrowitzB.et al (2008). Retinal cells integrate into the outer nuclear layer and differentiate into mature photoreceptors after subretinal transplantation into adult mice. Exp. Eye Res.86, 691–700. 10.1016/j.exer.2008.01.018
6
BassettE. A.WallaceV. A. (2012). Cell fate determination in the vertebrate retina. Trends Neurosci.35, 565–573. 10.1016/j.tins.2012.05.004
7
Belecky-AdamsT.CookB.AdlerR. (1996). Correlations between terminal mitosis and differentiated fate of retinal precursor cells in vivo and in vitro: analysis with the “window-labeling” technique. Dev. Biol.178, 304–315. 10.1006/dbio.1996.0220
8
BergerW.Kloeckener-GruissemB.NeidhardtJ. (2010). The molecular basis of human retinal and vitreoretinal diseases. Prog. Retin Eye Res.29, 335–375. 10.1016/j.preteyeres.2010.03.004
9
BrzezinskiJ. A. T.KimE. J.JohnsonJ. E.RehT. A. (2011). Ascl1 expression defines a subpopulation of lineage-restricted progenitors in the mammalian retina. Development138, 3519–3531. 10.1242/dev.064006
10
Bumsted O'brienK. M.ChengH.JiangY.SchulteD.SwaroopA.HendricksonA. E. (2004). Expression of photoreceptor-specific nuclear receptor NR2E3 in rod photoreceptors of fetal human retina. Invest Ophthalmol. Vis. Sci.45, 2807–2812. 10.1167/iovs.03-1317
11
ButchbachM. E.EdwardsJ. D.SchusslerK. R.BurghesA. H. (2007). A novel method for oral delivery of drug compounds to the neonatal SMNDelta7 mouse model of spinal muscular atrophy. J. Neurosci. Methods161, 285–290. 10.1016/j.jneumeth.2006.11.002
12
CaoJ.SpielmannM.QiuX.HuangX.IbrahimD. M.HillA. J.et al (2019). The single-cell transcriptional landscape of mammalian organogenesis. Nature566, 496–502. 10.1038/s41586-019-0969-x
13
Carter-DawsonL. D.LavailM. M. (1979). Rods and cones in the mouse retina. II. Autoradiographic analysis of cell generation using tritiated thymidine. J. Comp. Neurol.188, 263–272. 10.1002/cne.901880205
14
ChengH.KhannaH.OhE. C.HicksD.MittonK. P.SwaroopA. (2004). Photoreceptor-specific nuclear receptor NR2E3 functions as a transcriptional activator in rod photoreceptors. Hum. Mol. Genet.13, 1563–1575. 10.1093/hmg/ddh173
15
ClarkB. S.Stein-O'brienG. L.ShiauF.CannonG. H.Davis-MarcisakE.ShermanT.et al (2019). Single-cell RNA-seq analysis of retinal development identifies NFI factors as regulating mitotic exit and late-born cell specification. Neuron102, 1111–1126 e5. 10.1016/j.neuron.2019.04.010
16
DorgauB.CollinJ.RozanskaA.BoczonadiV.Moya-MolinaM.UnsworthA.et al (2024). Deciphering the spatiotemporal transcriptional and chromatin accessibility of human retinal organoid development at the single-cell level. iScience27, 109397. 10.1016/j.isci.2024.109397
17
EberleD.SchubertS.PostelK.CorbeilD.AderM. (2011). Increased integration of transplanted CD73-positive photoreceptor precursors into adult mouse retina. Invest Ophthalmol. Vis. Sci.52, 6462–6471. 10.1167/iovs.11-7399
18
FadlB. R.BrodieS. A.MalaskyM.BolandJ. F.KellyM. C.KelleyM. W.et al (2020). An optimized protocol for retina single-cell RNA sequencing. Mol. Vis.26, 705–717.
19
FlaxmanS. R.BourneR. R. A.ResnikoffS.AcklandP.BraithwaiteT.CicinelliM. V.et al (2017). Global causes of blindness and distance vision impairment 1990–2020: a systematic review and meta-analysis. Lancet Glob. Health5, e1221–e1234. 10.1016/S2214-109X(17)30393-5
20
FortinJ. M.AzariH.ZhengT.DariooshR. P.SchmollM. E.Vedam-MaiV.et al (2016). Transplantation of defined populations of differentiated human neural stem cell progeny. Sci. Rep.6, 23579. 10.1038/srep23579
21
FurukawaT.MorrowE. M.CepkoC. L. (1997). Crx, a novel otx-like homeobox gene, shows photoreceptor-specific expression and regulate photoreceptor differentiation. Cell91, 531–541. 10.1016/s0092-8674(00)80439-0
22
FurukawaA.KoikeC.LippincottP.CepkoC. L.FurukawaT. (2002). The mouse Crx 5'-upstream transgene sequence directs cell-specific and developmentally regulated expression in retinal photoreceptor cells. J. Neurosci.22, 1640–1647. 10.1523/JNEUROSCI.22-05-01640.2002
23
GaspariniS. J.TessmerK.RehM.WienekeS.CaridoM.VölknerM.et al (2022). Transplanted human cones incorporate into the retina and function in a murine cone degeneration model. J. Clin. Invest132, e154619. 10.1172/JCI154619
24
GustJ.RehT. A. (2011). Adult donor rod photoreceptors integrate into the mature mouse retina. Invest Ophthalmol. Vis. Sci.52, 5266–5272. 10.1167/iovs.10-6329
25
HaflerB. P.SurzenkoN.BeierK. T.PunzoC.TrimarchiJ. M.KongJ. H.et al (2012). Transcription factor Olig2 defines subpopulations of retinal progenitor cells biased toward specific cell fates. Proc. Natl. Acad. Sci. U. S. A.109, 7882–7887. 10.1073/pnas.1203138109
26
HamedA. A.KunzD. J.El-HamamyI.TrinhQ. M.SubedarO. D.RichardsL. M.et al (2022). A brain precursor atlas reveals the acquisition of developmental-like states in adult cerebral tumours. Nat. Commun.13, 4178. 10.1038/s41467-022-31408-y
27
HanI. C.BohrerL. R.Gibson-CorleyK. N.WileyL. A.ShresthaA.HarmanB. E.et al (2022). Biocompatibility of human induced pluripotent stem cell-derived retinal progenitor cell grafts in immunocompromised rats. Cell Transpl.31, 9636897221104451. 10.1177/09636897221104451
28
HaoY.HaoS.Andersen-NissenE.MauckW. M.ZhengS.ButlerA.et al (2021). Integrated analysis of multimodal single-cell data. Cell184, 3573–3587.e29. 10.1016/j.cell.2021.04.048
29
HirataT.TohsatoY.ItogaH.ShioiG.KiyonariH.OkaS.et al (2021). NeuroGT: a brain atlas of neurogenic tagging CreER drivers for birthdate-based classification and manipulation of mouse neurons. Cell Rep. Methods1, 100012. 10.1016/j.crmeth.2021.100012
30
HoltC. E.BertschT. W.EllisH. M.HarrisW. A. (1988). Cellular determination in the xenopus retina is independent of lineage and birth date. Neuron1, 15–26. 10.1016/0896-6273(88)90205-x
31
HumphriesM. M.RancourtD.FarrarG. J.KennaP.HazelM.BushR. A.et al (1997). Retinopathy induced in mice by targeted disruption of the rhodopsin gene. Nat. Genet.15, 216–219. 10.1038/ng0297-216
32
KaiserT.FengG. (2019). Tmem119-EGFP and Tmem119-CreERT2 transgenic mice for labeling and manipulating microglia. eNeuro6 (4), ENEURO.0448-18.2019. 10.1523/ENEURO.0448-18.2019
33
KalargyrouA. A.BascheM.HareA.WestE. L.SmithA. J.AliR. R.et al (2021). Nanotube-like processes facilitate material transfer between photoreceptors. EMBO Rep.22, e53732. 10.15252/embr.202153732
34
KawamotoA.IwasakiH.KusanoK.MurayamaT.OyamadaA.SilverM.et al (2006). CD34-positive cells exhibit increased potency and safety for therapeutic neovascularization after myocardial infarction compared with total mononuclear cells. Circulation114, 2163–2169. 10.1161/CIRCULATIONAHA.106.644518
35
KohwiM.OsumiN.RubensteinJ. L.Alvarez-BuyllaA. (2005). Pax6 is required for making specific subpopulations of granule and periglomerular neurons in the olfactory bulb. J. Neurosci.25, 6997–7003. 10.1523/JNEUROSCI.1435-05.2005
36
KoikeC.NishidaA.UenoS.SaitoH.SanukiR.SatoS.et al (2007). Functional roles of Otx2 transcription factor in postnatal mouse retinal development. Mol. Cell Biol.27, 8318–8329. 10.1128/MCB.01209-07
37
KowalchukA. M.MaurerK. A.Shoja-TaheriF.BrownN. L. (2018). Requirements for Neurogenin2 during mouse postnatal retinal neurogenesis. Dev. Biol.442, 220–235. 10.1016/j.ydbio.2018.07.020
38
KuboF.TakeichiM.NakagawaS. (2005). Wnt2b inhibits differentiation of retinal progenitor cells in the absence of notch activity by downregulating the expression of proneural genes. Development132, 2759–2770. 10.1242/dev.01856
39
KumarR.ChenS.ScheurerD.WangQ. L.DuhE.SungC. H.et al (1996). The bZIP transcription factor Nrl stimulates rhodopsin promoter activity in primary retinal cell cultures. J. Biol. Chem.271, 29612–29618. 10.1074/jbc.271.47.29612
40
LaboissonniereL. A.SonodaT.LeeS. K.TrimarchiJ. M.SchmidtT. M. (2017). Single-cell RNA-Seq of defined subsets of retinal ganglion cells. J. Vis. Exp. (123), e55229. 10.3791/55229
41
LakowskiJ.HanY. T.PearsonR. A.Gonzalez-CorderoA.WestE. L.GualdoniS.et al (2011). Effective transplantation of photoreceptor precursor cells selected via cell surface antigen expression. Stem Cells29, 1391–1404. 10.1002/stem.694
42
LakowskiJ.Gonzalez-CorderoA.WestE. L.HanY. T.WelbyE.NaeemA.et al (2015). Transplantation of photoreceptor precursors isolated via a cell surface biomarker panel from embryonic stem cell-derived self-forming retina. Stem Cells33, 2469–2482. 10.1002/stem.2051
43
LakowskiJ.WelbyE.BudingerD.Di MarcoF.Di FoggiaV.BainbridgeJ. W. B.et al (2018). Isolation of human photoreceptor precursors via a cell surface marker panel from stem cell-derived retinal organoids and fetal retinae. Stem Cells36, 709–722. 10.1002/stem.2775
44
LuY.ShiauF.YiW.LuS.WuQ.PearsonJ. D.et al (2020). Single-cell analysis of human retina identifies evolutionarily conserved and species-specific mechanisms controlling development. Dev. Cell53, 473–491.e9. 10.1016/j.devcel.2020.04.009
45
LudwigA. L.GammD. M. (2021). Outer retinal cell replacement: putting the pieces together. Transl. Vis. Sci. Technol.10, 15. 10.1167/tvst.10.10.15
46
MaclarenR. E.PearsonR. A.MacneilA.DouglasR. H.SaltT. E.AkimotoM.et al (2006). Retinal repair by transplantation of photoreceptor precursors. Nature444, 203–207. 10.1038/nature05161
47
MadisenL.ZwingmanT. A.SunkinS. M.OhS. W.ZariwalaH. A.GuH.et al (2010). A robust and high-throughput Cre reporting and characterization system for the whole mouse brain. Nat. Neurosci.13, 133–140. 10.1038/nn.2467
48
MaslandR. H. (2001). The fundamental plan of the retina. Nat. Neurosci.4, 877–886. 10.1038/nn0901-877
49
MearsA. J.KondoM.SwainP. K.TakadaY.BushR. A.SaundersT. L.et al (2001). Nrl is required for rod photoreceptor development. Nat. Genet.29, 447–452. 10.1038/ng774
50
Molina-RuizF. J.SandersP.GomisC.AbanteJ.LondoñoF.BombauG.et al (2025). CD200-based cell sorting results in homogeneous transplantable striatal neuroblasts for human cell therapy for Huntington's disease. Neurobiol. Dis.209, 106905. 10.1016/j.nbd.2025.106905
51
MorrowE. M.BelliveauM. J.CepkoC. L. (1998). Two phases of rod photoreceptor differentiation during rat retinal development. J. Neurosci.18, 3738–3748. 10.1523/JNEUROSCI.18-10-03738.1998
52
MorrowE. M.FurukawaT.RaviolaE.CepkoC. L. (2005). Synaptogenesis and outer segment formation are perturbed in the neural retina of Crx mutant mice. BMC Neurosci.6, 5. 10.1186/1471-2202-6-5
53
NishidaA.FurukawaA.KoikeC.TanoY.AizawaS.MatsuoI.et al (2003). Otx2 homeobox gene controls retinal photoreceptor cell fate and pineal gland development. Nat. Neurosci.6, 1255–1263. 10.1038/nn1155
54
Ortin-MartinezA.TsaiE. L.NickersonP. E.BergeretM.LuY.SmileyS.et al (2017). A reinterpretation of cell transplantation: GFP transfer from donor to host photoreceptors. Stem Cells35, 932–939. 10.1002/stem.2552
55
Ortin-MartinezA.YanN. E.TsaiE. L. S.ComanitaL.GurditaA.TachibanaN.et al (2021). Photoreceptor nanotubes mediate the in vivo exchange of intracellular material. Embo J.40, e107264. 10.15252/embj.2020107264
56
PearsonR. A.BarberA. C.RizziM.HippertC.XueT.WestE. L.et al (2012). Restoration of vision after transplantation of photoreceptors. Nature485, 99–103. 10.1038/nature10997
57
PearsonR. A.Gonzalez-CorderoA.WestE. L.RibeiroJ. R.AghaizuN.GohD.et al (2016). Donor and host photoreceptors engage in material transfer following transplantation of post-mitotic photoreceptor precursors. Nat. Commun.7, 13029. 10.1038/ncomms13029
58
PengY. R.ShekharK.YanW.HerrmannD.SappingtonA.BrymanG. S.et al (2019). Molecular classification and comparative taxonomics of foveal and peripheral cells in primate retina. Cell176, 1222–1237.e22. 10.1016/j.cell.2019.01.004
59
QiuX.HillA.PackerJ.LinD.MaY. A.TrapnellC. (2017a). Single-cell mRNA quantification and differential analysis with census. Nat. Methods14, 309–315. 10.1038/nmeth.4150
60
QiuX.MaoQ.TangY.WangL.ChawlaR.PlinerH. A.et al (2017b). Reversed graph embedding resolves complex single-cell trajectories. Nat. Methods14, 979–982. 10.1038/nmeth.4402
61
RapaportD. H.WongL. L.WoodE. D.YasumuraD.LavailM. M. (2004). Timing and topography of cell genesis in the rat retina. J. Comp. Neurol.474, 304–324. 10.1002/cne.20134
62
RehemtullaA.WarwarR.KumarR.JiX.ZackD. J.SwaroopA. (1996). The basic motif-leucine zipper transcription factor Nrl can positively regulate rhodopsin gene expression. Proc. Natl. Acad. Sci. U. S. A.93, 191–195. 10.1073/pnas.93.1.191
63
Santos-FerreiraT.LlonchS.BorschO.PostelK.HaasJ.AderM. (2016a). Retinal transplantation of photoreceptors results in donor-host cytoplasmic exchange. Nat. Commun.7, 13028. 10.1038/ncomms13028
64
Santos-FerreiraT. F.BorschO.AderM. (2016b). Rebuilding the missing Part-A review on photoreceptor transplantation. Front. Syst. Neurosci.10, 105. 10.3389/fnsys.2016.00105
65
SchindelinJ.Arganda-CarrerasI.FriseE.KaynigV.LongairM.PietzschT.et al (2012). Fiji: an open-source platform for biological-image analysis. Nat. Methods9, 676–682. 10.1038/nmeth.2019
66
SeilerM. J.AramantR. B. (2012). Cell replacement and visual restoration by retinal sheet transplants. Prog. Retin Eye Res.31, 661–687. 10.1016/j.preteyeres.2012.06.003
67
ShaylerD. W. H.StachelekK.CambierL.LeeS.BaiJ.BhatB.et al (2025). Identification and characterization of early human photoreceptor states and cell-state-specific retinoblastoma-related features. eLife13, RP101918. 10.7554/eLife.101918
68
ShekharK.LapanS. W.WhitneyI. E.TranN. M.MacoskoE. Z.KowalczykM.et al (2016). Comprehensive classification of retinal bipolar neurons by single-cell transcriptomics. Cell166, 1308–1323.e30. 10.1016/j.cell.2016.07.054
69
SinghM. S.Charbel IssaP.ButlerR.MartinC.LipinskiD. M.SekaranS.et al (2013). Reversal of end-stage retinal degeneration and restoration of visual function by photoreceptor transplantation. Proc. Natl. Acad. Sci. U. S. A.110, 1101–1106. 10.1073/pnas.1119416110
70
SinghM. S.BalmerJ.BarnardA. R.AslamS. A.MoralliD.GreenC. M.et al (2016). Transplanted photoreceptor precursors transfer proteins to host photoreceptors by a mechanism of cytoplasmic fusion. Nat. Commun.7, 13537. 10.1038/ncomms13537
71
SridharA.HoshinoA.FinkbeinerC. R.ChitsazanA.DaiL.HauganA. K.et al (2020). Single-cell transcriptomic comparison of human fetal retina, hPSC-Derived retinal organoids, and long-term retinal cultures. Cell Rep.30, 1644–1659.e4. 10.1016/j.celrep.2020.01.007
72
TessmerK.GaspariniS. J.SchmidtkeK.AdhikariT.MichalkeL.HammerJ.et al (2024). Extensive monolayer formation depends on a subpopulation of transplanted human iPSC-derived RPE cells. bioRxiv10.15, 618147. 10.1101/2024.10.15.618147
73
TrapnellC.CacchiarelliD.GrimsbyJ.PokharelP.LiS.MorseM.et al (2014). The dynamics and regulators of cell fate decisions are revealed by pseudotemporal ordering of single cells. Nat. Biotechnol.32, 381–386. 10.1038/nbt.2859
74
TripathiH.PengH.DonahueR.ChelvarajanL.GottipatiA.LevitanB.et al (2020). Isolation methods for human CD34 subsets using fluorescent and magnetic activated cell sorting: an in vivo comparative study. Stem Cell Rev. Rep.16, 413–423. 10.1007/s12015-019-09939-7
75
TurnerD. L.CepkoC. L. (1987). A common progenitor for neurons and glia persists in rat retina late in development. Nature328, 131–136. 10.1038/328131a0
76
Van EsJ. H.SatoT.Van De WeteringM.LyubimovaA.Yee NeeA. N.GregorieffA.et al (2012). Dll1+ secretory progenitor cells revert to stem cells upon crypt damage. Nat. Cell Biol.14, 1099–1104. 10.1038/ncb2581
77
VentersS. J.MikawaT.HyerJ. (2013). Central and peripheral retina arise through distinct developmental paths. PLoS One8, e61422. 10.1371/journal.pone.0061422
78
VentersS. J.MikawaT.HyerJ. (2015). Early divergence of central and peripheral neural retina precursors during vertebrate eye development. Dev. Dyn.244, 266–276. 10.1002/dvdy.24218
79
VenugopalanP.WangY.NguyenT.HuangA.MullerK. J.GoldbergJ. L. (2016). Transplanted neurons integrate into adult retinas and respond to light. Nat. Commun.7, 10472. 10.1038/ncomms10472
80
Warre-CornishK.BarberA. C.SowdenJ. C.AliR. R.PearsonR. A. (2014). Migration, integration and maturation of photoreceptor precursors following transplantation in the mouse retina. Stem Cells Dev.23, 941–954. 10.1089/scd.2013.0471
81
WestE. L.PearsonR. A.MaclarenR. E.SowdenJ. C.AliR. R. (2009). Cell transplantation strategies for retinal repair. Prog Brain Res. 175, 3–21. 10.1016/S0079-6123(09)17501-5
82
WestE. L.PearsonR. A.BarkerS. E.LuhmannU. F.MaclarenR. E.BarberA. C.et al (2010). Long-term survival of photoreceptors transplanted into the adult murine neural retina requires immune modulation. Stem Cells28, 1997–2007. 10.1002/stem.520
83
WettsR.FraserS. E. (1988). Multipotent precursors can give rise to all major cell types of the frog retina. Science239, 1142–1145. 10.1126/science.2449732
84
YeJ.CoulourisG.ZaretskayaI.CutcutacheI.RozenS.MaddenT. L. (2012). Primer-BLAST: a tool to design target-specific primers for polymerase chain reaction. BMC Bioinforma.13, 134. 10.1186/1471-2105-13-134
85
YingG. S.MaguireM. G.GlynnR.RosnerB. (2017). Tutorial on biostatistics: Linear regression analysis of continuous correlated eye data. Ophthalmic Epidemiol.24, 130–140. 10.1080/09286586.2016.1259636
86
YoungR. W. (1985). Cell differentiation in the retina of the mouse. Anat. Rec.212, 199–205. 10.1002/ar.1092120215
87
ZhaosongM.NaF.ShulingG.JiachengL.RanW. (2021). Heterogeneity affects the differentiation potential of dental follicle stem cells through the TGF-β signaling pathway. Bioengineered12, 12294–12307. 10.1080/21655979.2021.2009974
88
ZhuL.GibsonP.CurrleD. S.TongY.RichardsonR. J.BayazitovI. T.et al (2009). Prominin 1 marks intestinal stem cells that are susceptible to neoplastic transformation. Nature457, 603–607. 10.1038/nature07589
Summary
Keywords
lineage tracing, neuroblast, photoreceptor, retinal development, retinal organoid, single cell RNA-sequencing
Citation
Yano JJ, Wei Z, Gajjar KJ, Barati-Stec KT, Yang E and Uyhazi KE (2026) Lineage tracing reveals photoreceptor precursor cell subpopulations that contribute to murine retinogenesis. Front. Cell Dev. Biol. 14:1814134. doi: 10.3389/fcell.2026.1814134
Received
20 February 2026
Revised
13 April 2026
Accepted
27 April 2026
Published
04 June 2026
Volume
14 - 2026
Edited by
Alan Marmorstein, Mayo Clinic, United States
Reviewed by
Eric Jong, The Francis Crick Institute, United Kingdom
Shida Chen, Sun Yat-sen University, China
Updates
Copyright
© 2026 Yano, Wei, Gajjar, Barati-Stec, Yang and Uyhazi.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Katherine E. Uyhazi, katherine.uyhazi@pennmedicine.upenn.edu
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.