A composition, method of induction and use of chemical reprogramming to induce functional ciRPE cells

Abstract

The application discloses a chemical reprogramming induced functional ciRPE cell composition, an induction method and application. The application proposes a two-step chemical reprogramming strategy, which can transdifferentiate fibroblasts into functional RPE (ciRPE) cells through an eye field (EF) / optic vesicle (OV) intermediate state. The ciRPE cells are very similar to natural RPE cells in structure and function, have no tumorigenic risk after being transplanted into RD rat retinas, can be integrated into host tissues, effectively protect photoreceptor cells and significantly restore visual function.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of biotechnology, and particularly relates to a chemical reprogramming composition for inducing functional ciRPE cells, an induction method and application. BACKGROUND

[0002] Retinal pigment epithelium (RPE) is essential for retinal health and plays a key role in maintaining photoreceptor function and the blood-retinal barrier. RPE dysfunction or loss is a key driver of the pathogenesis of devastating retinal degenerative diseases (RDs), such as age-related macular degeneration (AMD), retinitis pigmentosa (RP) and Stargardt disease, which lead to progressive vision loss and, in the late stages, irreversible blindness. Given the important role of RPE cells in retinal function, repairing or replacing damaged RPE cells is a transformative strategy for treating these debilitating diseases. However, in regenerative medicine, generating functional and therapeutically safe RPE cells remains a major challenge.

[0003] Currently, there are two main methods for preparing RPE cells: isolating RPE cells from natural tissues or differentiating RPE cells from pluripotent stem cells. However, both strategies still have certain challenges. RPE cells derived from natural tissues are inherently scarce, and their function rapidly diminishes in culture, limiting their application in large-scale therapy. In contrast, pluripotent stem cell-derived RPE cells, while theoretically an ideal source, have tumorigenicity and profound ethical concerns. Another approach is to use ectopic transcription factors (TFs) for direct reprogramming, which can bypass the pluripotent stage and directly convert fibroblasts into RPE-like cells. Although promising, this method still faces challenges such as genomic integration risk, low reprogramming efficiency, high cost, which may limit its wider clinical application. SUMMARY

[0004] Determining the appropriate intermediate cell state is crucial for improving the efficiency and stability of RPE reprogramming. Eye-field (EF) cells or optic vesicle (OV) cell-derived progenitors naturally differentiate into RPE cells during retinal development, making them an ideal intermediate state. These cells have significant plasticity and differentiation potential, making them a biologically relevant step to guide somatic cells towards RPE fate. Using EF (or OV) as an intermediate state not only improves reprogramming efficiency, but also minimizes the risk of generating unwanted cell types, ensuring a more stable and reproducible conversion process.

[0005] To solve the technical problems existing in the prior art, the present application provides the following technical solutions.

[0006] The application provides a method for two-step reprogramming of fibroblasts into functional ciRPE cells using chemical small molecule compositions, comprising: treating fibroblasts with reprogramming medium 1 to induce fibroblasts to reprogram into EF-like or OV-like cells, and then treating the EF-like or OV-like cells with reprogramming medium 2 to induce the EF-like or OV-like cells to reprogram into ciRPE cells.

[0007] The reprogramming medium 1 comprises a chemical small molecule composition 1, and the reprogramming medium 2 comprises a chemical small molecule composition 2.

[0008] When the fibroblasts are of murine origin, the chemical small molecule composition 1 is specifically a combination of LDN193189, A 83-01, CKI-7, Hh-Ag1.5, CHIR-99021, BMS-345541, RG108, or pharmaceutically equivalent preparations, analogs, isomers, salts, hydrates or precursors thereof.

[0009] When the fibroblasts are of murine origin, the chemical small molecule composition 2 is specifically a combination of nicotinamide, retinoic acid, Activin A, or pharmaceutically equivalent preparations, analogs, isomers, salts, hydrates or precursors thereof.

[0010] Further, the compositions of the ciRPE cell reprogramming medium are all prepared by adding small molecule compounds to a basal medium.

[0011] Further, the basal medium 1 used in the reprogramming medium 1 is specifically a combination of equal volumes of Neurobasal and DMEM / F12 / GlutaMAX, 1% N2, 1% B27 without vitamin A, 7.5% BSA, 1% NEAA, 1% P / S, and 10 ng / mL of bFGF.

[0012] Further, the basal medium 2 used in the reprogramming medium 2 is specifically DMEM / F12 / GlutaMax supplemented with 10% KSR, 1% NEAA, 1% P / S, and 0.055 mM 2-mercaptoethanol.

[0013] Further, the proliferation culture uses a proliferation culture medium for ciRPE cells, and the components are as follows: DMEM / F12 / GlutaMAX as the substrate, supplemented with 1% N2, 2% B27 without vitamin A, 1% NEAA, 1% P / S, and 0.1 mM 2-mercaptoethanol.

[0014] Further, the proliferation medium for the ciRPE cells further comprises one or more of 10 ng / mL bFGF, 20 ng / mL EGF, 10 mM of Y-27632, 0.5 mM of A 83-01.

[0015] Further, the functional culture uses a functional maintenance medium for the ciRPE cells, the components of which are as follows: DMEM / F12 / GlutaMAX plus 1% N2, 2% B27 with vitamin A, 1% NEAA, 1% P / S, 0.1 mM 2-mercaptoethanol.

[0016] Further, when the fibroblasts are of murine origin, the functional maintenance culture is further supplemented with 0.2 mM of Activin A, 0.5 mM of retinoic acid, 1 mM of BMP4, and 10 mM of nicotinamide.

[0017] Further, when the fibroblasts are of murine origin, the concentration of LDN193189 in the chemical small molecule composition 1 is 0.1 mM, the concentration of A 83-01 is 0.5 mM, the concentration of CKI-7 is 5 mM, the concentration of Hh-Agl.5 is 0.5 mM, the concentration of CHIR-99021 is 3 mM, the concentration of BMS-345541 is 0.2 mM, and the concentration of RG108 is 10 mM.

[0018] Further, when the fibroblasts are of murine origin, the concentration of nicotinamide in the chemical small molecule composition 2 is 10 mM, the concentration of retinoic acid is 1 mM, and the concentration of Activin A is 0.2 mM.

[0019] In some embodiments, the LDN193189 is a small molecule compound that is a selective inhibitor of bone morphogenetic protein (BMP) signaling pathways.

[0020] In some embodiments, the A 83-01 is an effective inhibitor of TGF-beta type I receptor (ALK5-TD).

[0021] In some embodiments, the CKI-7 (dihydrochloride) is an ATP-competitive CKI inhibitor.

[0022] In some embodiments, the Hh-Agl.5 is an effective Hedgehog (Hh) agonist with an EC50 of 1 nM. Mediated reprogramming breaks the quiescent state of non-injured liver stem cells, thus rescuing liver failure.

[0023] In some embodiments, the CHIR-99021 (Laduviglusib, CT99021) is a GSK-3a and GSK-3b inhibitor.

[0024] In some embodiments, the BMS-345541 is a highly selective inhibitor of the catalytic subunit, acting on IKK-2 and IKK-1.

[0025] In some embodiments, the RG108 (N-Phthalyl-L-tryptophan) is a DNA methyltransferase inhibitor.

[0026] In the present invention, the term "μM" stands for micromolar concentration, μ: Greek letter "mu", stands for the International System of Units prefix "micro-", i.e. 10⁻ 6 (millionth part), M: stands for molar concentration (mol / L), i.e. the amount of substance of solute per liter of solution, therefore, 1 μM = 10⁻ 6 mol / L, i.e. there is 1 micromole of solute per liter of solution.

[0027] In the present invention, mM is used as a special symbol in the International System of Units to express concentration, its full name is millimolar per liter, which belongs to the solution concentration measurement method commonly used in the field of chemistry and biology. The definition of this unit is based on the basic amount of substance unit of mole (mol), where the prefix "m" stands for one thousandth (10 -3 ), therefore 1 millimole is equal to 0.001 mole. When applied to the solution system, mM specifically refers to the number of millimoles of solute per liter of solution, i.e. the ratio of the amount of substance of solute (in millimoles) to the volume of solution (in liters).

[0028] In some embodiments, the method specifically comprises the following steps:

[0029] First stage: chemical transformation of EF cells in MEFs (murine fibroblasts), specifically with the following steps: 1) 4 days before chemical transformation, Matrigel is thawed at 4°C overnight. The next day, the 6-well plate is pre-cooled at 4°C for at least 1 hour. Dilute Matrigel with DMEM / F12 / GlutaMAX solution 1:40, add 1 ml of diluted Matrigel solution to each well of the 6-well plate, and incubate at 4°C overnight (Note: To ensure optimal performance, Matrigel must be thawed at 4°C and should not be subjected to repeated freeze-thaw cycles).

[0030] 2) 2 days prior to chemical induction, thaw frozen primary MEFs in a 37°C water bath and centrifuge to remove cryoprotectant. Resuspend cells in MEF media (DMEM media with 10% FBS) at a density of 20 x 10 4

[0031] 3) 1 day prior to chemical induction, replace MEF media with fresh media to allow cells to proliferate for 1 day. Check MEF cell growth and condition regularly to ensure cell health and density are optimal (Note: MEF cell growth condition is critical for successful conversion. Long term storage of MEFs in liquid nitrogen also reduces conversion efficiency. Freshly prepared MEF cells or MEF cells stored in liquid nitrogen for no more than 6 months were used in this study).

[0032] 4) On the day of chemical induction, prepare fresh reprogramming media 1, Stage 1, in the dark to ensure complete dissolution of all components, especially chemical reagents. After overnight incubation of MEF cells in MEF media, wash cells twice with 1x PBS and replace MEF media with freshly prepared reprogramming media 1, Stage 1. Replace reprogramming media 1, Stage 1 again the next day.

[0033] 5) During the first stage of induction, cells proliferate rapidly within the first 1-4 days, cell morphology changes from elongated spindle shape to small oval shape, and cell clonal colonies appear. After a week, the proliferation rate of the colonies slows down. At this point, mechanically remove surrounding non-clonal cells using yellow microscalpel tip tip end to promote continued expansion of the colonies. After about a week, cells will start to exhibit a cobblestone-like morphology. Thereafter, replace media daily until EF-like cells are formed.

[0034] Second Stage: EF cells generate ciRPE cells, specifically including: During the second stage of induction, cells exhibit initial RPE cell morphology but still lack maturity and pigmentation. At this stage, use reprogramming media 2 to induce pigmentation and promote further maturation of ciRPE cells. Replace media daily. After two weeks of treatment, cells are close to full maturation and exhibit clear pigmentation. Subsequently, ciRPE cells are passaged and expanded in maintenance media.

[0035] ​In some embodiments, the specific steps of the dynamic alternating culture of the ciRPE cells are: To balance proliferation and RPE-specific functions, MEFs-derived ciRPE cells are cultured using a dynamic alternating culture medium. After the first passage, during the initial expansion phase (0-7 days), the cells are cultured in a proliferation medium to maximize cell growth. Subsequently, the cells are briefly exposed to a maintenance function medium every 3-5 days for 1-2 days. This alternating approach effectively supports cell expansion and preservation of early RPE-like characteristics.

[0036] The present application provides a chemical small molecule composition, which is divided into the aforementioned chemical small molecule composition 1 and chemical small molecule composition 2.

[0037] Further, the chemical small molecule composition comprises a pharmaceutically acceptable carrier or excipient; preferably, the carrier or excipient comprises one or more selected from the group consisting of: water, saline, phosphate buffer or other aqueous solvent; DMSO, glycerol and ethanol or other organic solvent; microspheres, liposomes, microemulsions or high molecular surfactants; colloidal drug delivery systems or high molecular drug delivery systems; preservatives, antioxidants, flavorings, fragrances, cosolvents, emulsifiers, pH buffering substances, adhesives, fillers, lubricants or other pharmaceutical excipients; or, the chemical small molecule composition can be prepared into a pharmaceutical dosage form comprising: solid dosage forms, including: powder, granules, tablets, pills, capsules, sustained-release preparations, controlled-release preparations, or other solid dosage forms; liquid dosage forms, including: injections, infusions, suspensions, or other liquid dosage forms; gaseous dosage forms; or semi-solid dosage forms; or the small molecule composition is a reprogramming preparation or reagent, preferably further comprising an organic solvent, physiological saline, or other carriers or excipients.

[0038] The general requirements for excipients are stable in nature, have no incompatibility with the main drug, do not produce side effects, do not affect the therapeutic effect, are not prone to deformation, dry cracking, mold, insect infestation at room temperature, are harmless to the human body, have no physiological effects, do not produce chemical or physical effects with the main drug, do not affect the content determination of the main drug, etc. A full discussion of pharmaceutically acceptable carriers or excipients can be found in Remington’s Pharmaceutical Sciences (Mack Pub. Co., N.J. 1991). The carriers or excipients thereof include but are not limited to: aqueous solutions such as water, saline, phosphate buffer, etc.; organic solvents such as DMSO (dimethyl sulfoxide), glycerol and ethanol; microspheres, liposomes, microemulsions, high molecular surfactants; colloidal drug delivery systems, new high molecular drug delivery systems, new drug carriers and other pharmaceutical carriers; preservatives, antioxidants, flavorings, fragrances, cosolvents, emulsifiers, pH buffering substances in liquid preparations, adhesives, fillers, lubricants and other pharmaceutical excipients in tablets, etc.

[0039] The present application provides a reprogramming medium used in the above-mentioned method, the reprogramming medium comprising reprogramming medium 1 and reprogramming medium 2.

[0040] The reprogramming medium 1 comprises the above-mentioned chemical small molecule composition 1 and a basal medium 1, which is equal volume of Neurobasal and DMEM / F12 / GlutaMAX added with 1% N2, 1% B27 without vitamin A, 7.5% BSA, 1% NEAA, 1% P / S, bFGF.

[0041] The reprogramming medium 2 comprises the above-mentioned chemical small molecule composition 2 and a basal medium 2, which is DMEM / F12 / GlutaMax added with 10% KSR, 1% NEAA, 1% P / S, 0.055 mM 2-mercaptoethanol.

[0042] The present application provides an application of a chemical small molecule composition in preparing a product for chemically reprogramming fibroblasts into functional ciRPE cells, the chemical small molecule composition comprising the above-mentioned chemical small molecule composition.

[0043] The present application provides a kit or a reagent kit for inducing fibroblasts into ciRPE cells by two-step reprogramming with a chemical small molecule composition, the kit or the reagent kit comprising the above-mentioned chemical small molecule composition or the above-mentioned reprogramming medium.

[0044] Further, the fibroblasts comprise human or non-human mammalian fibroblasts.

[0045] Further, the fibroblasts comprise skin fibroblasts, liver fibroblasts, lung fibroblasts, kidney fibroblasts, and pancreatic fibroblasts.

[0046] Advantages and beneficial effects of the present application:

[0047] The present invention addresses the complex challenge of generating functional RPE cells by developing an innovative two-stage chemical induction protocol, implemented by a single-cell reprogramming compound screening platform (scRCF), an advanced platform that combines single-cell transcriptomic-driven computational predictions with a sophisticated drug sequence screening system. This platform systematically identifies the best small molecules to induce fibroblasts into an EF-like state and ultimately generate functional chemically induced RPE (ciRPE) cells. These ciRPE cells are very similar to native RPE cells in morphology, gene expression, and basic functional properties. In a rat model of RD, ciRPE cell transplantation integrates into the host RPE layer, significantly protects photoreceptors, and substantially restores visual function. Multi-omic analysis provides mechanistic insights into this dynamic. Mechanistic studies indicate that the compounds synergistically activate endogenous TFs, such as Ascll and Olig2, to reprogram fibroblasts toward an RPE cell phenotype. This study introduces a scalable, non-integrating, cost-effective chemical approach to generate functional RPE cells, providing a promising innovative strategy for cell replacement therapy against RD disease. BRIEF DESCRIPTION OF DRAWINGS

[0048] Figure 1Figure 1 is a schematic diagram of the scRCF system screening EF reprogramming small molecules results, wherein A is the workflow of scRCF. The input of scRCF includes scRNA-seq data from initial and target cell types, and three pre-established databases: 1) a small molecule perturbation database integrating GEO and LINCS L1000 data, retaining only TFs identified in AnimalTFDB3.0; 2) a small molecule target database and classification information, including information from STITCH, Drug Repurposing Hub, and MedChemExpress; 3) a signal network database derived from Reactome and OmniPath. scRCF contains three main steps: 1) identify differentially expressed transcription factors based on scRNA-seq data, and use signal network analysis to preliminarily screen candidate signal proteins; 2) community partitioning of pre-screened signal proteins, cumulative scoring according to the distribution of small molecule targets in the community. According to the classification of signal pathways, select the highest scoring small molecule from each pathway subset as the final output; 3) further screen candidate small molecules using DRUG-seq2. B is the UMAP visualization of scRNA-seq data from MEF and primary EF (pEF) cells, used to predict small molecules that promote reprogramming between the two cell types. C is a volcano plot showing differentially expressed TFs (DETFs) identified from scRNA-seq data of MEF and EFs, analyzed using Seurat. DETF is defined as P < 0.05, log2 > 1. D is the candidate small molecules determined by preliminary screening using scRCF, used to reprogram MEF into EF cells. E is a bar chart showing the Z-score of cells treated with LAC+1 small molecule combination, based on the gene set of neuroectoderm and EF-related genes. Z-score represents the relative gene expression change of each drug, with higher values reflecting stronger effects on the gene set. The bar chart is sorted from highest to lowest average Z-score, with the dashed line indicating a baseline score of zero. F is a heatmap of gene expression profiles of neuroectoderm and EF-related genes in cells treated with LAC+1 small molecule combination. The color scale represents log2-transformed expression values, with red indicating high expression, blue indicating low expression, and white indicating moderate level expression.

[0049] Figure 2Figure 1 is a graph showing the results of establishing a two-stage chemical reprogramming strategy to generate ciRPE cells, wherein A is representative morphological changes of cells at different time points after MEFs were exposed to reprogramming medium (RM) containing 10 small molecules (Stage 1). Scale bar, 400 μm. B is qRT-PCR analysis showing the expression of EF-related genes and early RPE development-related genes at the indicated time points. C is representative morphological images of cells after cells were exposed to differentiation and maturation medium (DM) containing three compounds (Stage 2). DMSO, dimethyl sulfoxide as a negative control. M3: NIC, RA, Activin A, scale bar, 200 μm. D is qRT-PCR analysis showing the expression of RPE-related marker genes at the indicated time points. E is a schematic diagram of the genetic lineage tracing strategy and chemical reprogramming of MEF-derived ciRPE cells. F is the morphological and tdTomato fluorescence expression changes of cells at different days during induction. Scale bar, 400 μm. G is the percentage of tdTomato + cells at different days. H is the percentage of tdTomato + cells treated with candidate medium (all 13 compounds), and the percentage of tdTomato + cells treated with candidate medium (all 13 compounds), and the percentage of tdTomato Figure 2 All representative examples of at least three independent experimental data are given. Data are Mean ± SD; *P<0.05, **P<0.01, ***P<0.001.

[0050] Figure 3is a profile of ciRPE cells, where A is immunostaining showing MEF-derived ciRPE cells express ZO-1, Pax6, Rpe65, Mitf, Bestl, and Cralbp. Scale bar, 50 μm. B is a Z-stack confocal micrograph showing ciRPE cells have polarized expression of typical RPE markers. ZO-1 (green) shows apical localization (top), while Bestl (red) shows basolateral localization (bottom). Scale bar, 10 μm. C is a transmission electron micrograph of ciRPE cells showing apical microvilli (yellow arrows), melanin granules (red arrows), and tight junctions (black arrows). Scale bar, 1 μm. D is a confocal micrograph showing photoreceptor outer segments (green) are phagocytosed by ciRPE cells. The apical side of ciRPE cells is stained with ZO-1 (purple), and nuclei are counterstained with DAPI (blue). Scale bar, 50 μm. E is ciRPE cells grown on Transwells polarize to secrete PEDF and VEGF on the apical and basolateral sides. F is a morphological image showing MEF-derived ciRPE cells form dome structures during in vitro culture. Red arrows indicate dome morphology observed under different phase contrast microscopy conditions. Scale bar, 50 μm. G is a measurement of ciRPE cell transepithelial electrical resistance (TEER) values over 30 days. Figure 3 All representative examples of at least three independent experiments are given. Data are Mean ± SD.

[0051] Figure 4A is a schematic of the multi-omics sequencing analysis strategy for the MEF to ciRPE cell reprogramming process. B is principal component analysis (PCA) of RNA-seq and CUT&Tag data (H3K4me3, H3K27ac, and H3K27me3) for samples collected at day 0, 7, 18, and 32 of reprogramming (ciRPE) with pRPE cells as controls. C is a heatmap of differentially expressed genes in MEF to ciRPE cell reprogramming samples at the indicated time points. Numbers on the heatmap represent independent biological replicates. Representative genes (left side of heatmap) and associated gene ontology (GO; right side of heatmap for each block) are shown. Red and blue colors represent up- and down-regulated genes, respectively. D is dynamic changes of CUT&Tag peaks (H3K4me3, H3K27ac, H3K27me3) for module-specific genes in (C). Red lines represent median CUT&Tag peak values over time, blue lines represent median RNA expression levels, and gray background lines represent peaks at individual time points. E is UMAP analysis of scRNA-seq data for cells collected at the indicated time points during MEF reprogramming to ciRPE cells. F is a UMAP plot of cell types identified in samples collected at the indicated time points during reprogramming. G is a bubble plot of representative marker gene expression between different cell types during reprogramming. H is a prediction of cell population transitions during reprogramming using scVelo method to plot RNA velocity streamlines. Arrows represent flow determined by the proportion of unspliced to spliced transcripts, predicting dynamic changes in cell identity. Black arrows represent RNA velocity streams based on unspliced to spliced transcript ratios, while gray to blue-green arrows are used for visual enhancement to highlight specific trajectories. I is a similarity analysis of gene expression between different cell types during reprogramming. pEF cells represent data used for small molecule prediction above, while pRPE data is from GSE183572 dataset.

[0052] Figure 5A is a volcano plot showing the differential regulatory activity of transcription factors (TFs) between different cell types during reprogramming, data derived from SCENIC. B is a column chart showing the MCC scores of the top 15 key TFs. C is a heatmap showing the expression of the top 15 TFs at the indicated time points during cell reprogramming. D is the normalized analysis of RNA-seq and CUT&Tag sequencing (for histone modifications H3K4me3, H3K27ac and H3K27me3) at the indicated time points for Ascl1, Olig2, Zic1, Pou3f2 and Lhx2 genomic loci during cell reprogramming. E-F are the relative reprogramming efficiency assessed by detecting the proportion of tdTomato positive cells at day 32 after knockdown of Ascl1 (E) or Olig2 (F) at the indicated time points under M7+M3 induction. The reprogramming efficiency of M7+M3 induced cells at day 32 was set as "1", and DMSO as negative control. WT represents wild type; Control KD represents scramble shRNA mediated knockdown. G is the scRCF network visualization of potential signal cascade induced by small molecules targeting Ascl1 and Olig2. Orange rectangles represent perturbed compounds, blue diamonds represent signal protein targets, white ovals represent intermediate signal proteins, and green hexagons represent TFs. Figure 5 All representative examples of at least three independent experimental data are given. Data are Mean ± SD; *P<0.05, **P<0.01.

[0053] Figure 6 A is a schematic diagram of subretinal ciRPE cell transplantation in RCS rats. B is OCT images of RCS rats after subretinal transplantation of ciRPE cells for 0, 1, 2 and 3 weeks. Scale bar: 600 μm. C is a representative photograph of in vivo imaging showing tdTomato-labeled mESCs (via lentivirus) and tdTomato + Tumor formation in nude mice after subretinal transplantation of ciRPE cells. The right histogram shows quantitative analysis data, N=10. D is a representative bright field (top) and immunofluorescence image (bottom) of eye tissue sections after 8 weeks of ciRPE cell transplantation. Dotted line box is the cell transplantation area. Nuclei are counterstained with DAPI (blue). Scale bar, 200 μm. E is immunofluorescence analysis of whole retinal sections at 12 weeks after transplantation showing tdTomato + Tumor formation in nude mice after subretinal transplantation of ciRPE cells. The right histogram shows quantitative analysis data, N=10. D is a representative bright field (top) and immunofluorescence image (bottom) of eye tissue sections after 8 weeks of ciRPE cell transplantation. Dotted line box is the cell transplantation area. Nuclei are counterstained with DAPI (blue). Scale bar, 200 μm. E is immunofluorescence analysis of whole retinal sections at 12 weeks after transplantation showing tdTomato +ciRPE cell clusters. Enlarged views of the areas depicted by the dashed boxes show: (i) the untransplanted region and (ii) the transplanted region. The statistical plot on the right shows the quantitative analysis data. N=10. Scale bar, 500 μm. FH is tdTomato + Immunostaining images of ciRPE cells co-expressing Mitf (F), Best1 (G), and Pax6 (H). Scale, 50 μm. I represents the sham transplantation group as a control. + Representative TUNEL-stained micrographs of frozen sections of the retina of RCS rats 12 weeks after ciRPE cell transplantation (left), with statistical results shown on the right. N=10. Scale bar, 50 μm. J represents tdTomato grafts transplanted under the retina of RCS rats. + Immunostaining images of ciRPE cells co-expressing with Rhodopsin after 12 weeks. Scale bar, 50 μm. K represents representative b-wave response data (left) at 4, 8, 12, and 16 weeks post-transplantation in the ciRPE transplantation group and sham transplantation group, with an fERG response intensity of 0.48 log cd*s / m² (dark 3.0). The right side shows the statistical analysis of b-wave amplitude in the ciRPE transplantation group and sham transplantation group. N=6 for each group. L is a schematic diagram of the quantitative optomotor response (qOMR) testing device (left). Quantitative evaluation of visual acuity (right) in the ciRPE transplantation group and sham transplantation group at 4, 8, 12, and 16 weeks post-transplantation, N=20. Figure 6 All representative examples of data from at least three independent experiments are provided. All data are expressed as Mean ± SD. *P<0.05, **P<0.01, ***P<0.001.

[0054] Figure 7 This is a small molecule systematic screening diagram for EF fate reprogramming using scRCF. A is a visualization of the UMAP scRNA-seq dataset used to predict MEF and EF conversion small molecules for cell type transformation. B shows the gene co-expression network of signal proteins (upper layer) and TFs (lower layer). C shows the protein-protein interaction network using a random walk algorithm to partition the signal protein community, displaying the top 10 largest communities. D is a schematic diagram of the compound screening scheme used for DRUG-seq2 sequencing. LDN193189, A83-01, and CKI-7 (LAC) were used as the base compound combination, and predictive small molecules were added to form the LAC+1 combination. MEF was treated for 14 days before sequencing to assess changes in gene expression. E shows the small molecules targeting MEF conversion to EF selected using scRCF comprehensive screening.

[0055] Figure 8This diagram illustrates a two-stage chemical reprogramming strategy for generating ciRPE cells. A shows a bright-field image of an eye from an embryonic mouse with a Best1-Cre / ROSA26 tdTomato genetic background. Red fluorescence indicates the localization of Best1 in the eye tissue, tracked by tdTomato. Cell nuclei are reverse-stained with DAPI (blue). Scale bar: 750 μm. B shows an immunostaining analysis of whole retinal sections from an embryonic mouse eye with a Best1-Cre / ROSA26 tdTomato genetic background. Blue indicates DAPI staining, and red fluorescence indicates the localization of Best1 tracked by tdTomato. Scale bar: 500 μm. C shows FACS sorting of MEF cells from E13.5 mouse embryos with a Best1-Cre / ROSA26 tdTomato background. D shows qRT-PCR analysis of tdTomato with β-actin as a control. - Expression of RPE-specific genes in MEF and pRPE cells. E represents immunostaining analysis showing tdTomato - RPE-specific genes such as Mitf, Cralbp, Best1, and Rpe65 were negative in MEFs. Scale bar, 50 μm. F shows representative morphological images of different fibroblast types (C57BL / 6 MEF, 129 MEF, C57BL / 6 TTF, 129 TTF) induced by M7+M3 compound medium. Scale bar, 400 μm. G shows the percentage of Best1-positive cells after flow cytometry analysis of different fibroblast types (C57BL / 6 MEF, 129 MEF, C57BL / 6 TTF, 129 TTF) induced by compound medium.

[0056] Figure 9 These are characteristic images of ciRPE cells. A shows G-band karyotype analysis, indicating normal ciRPE cell karyotype. B shows morphological images of ciRPE and pRPE cells at different passages (P1, P3, and P6). Scale bar: 300 μm. C shows flow cytometry analysis of ethyldeoxyuridine (EdU) incorporation in ciRPE and pRPE cells at different passages (P1, P3, and P6). D shows the distribution of ciRPE and pRPE cells in the cell cycle (G1, S, and G2 phases) (left) and quantitative analysis of the percentage of cells at each phase (right). E shows confocal microscopy images of ciRPE and pRPE cells phagocytizing latex beads (green). The apical side of the cells is stained with ZO-1 (red). Scale bar: 10 μm.

[0057] Figure 10Figure 1 is a schematic diagram of the genetic lineage tracing strategy to confirm MEF-induced ciRPE cell results, where A is a schematic diagram of the genetic lineage tracing strategy. B is Bestl − / tdTomato + MEF cells. C is qRT-PCR analysis of Bestl - / tdTomato + expression of RPE-specific genes in MEF and pRPE cells. β-actin as control. D is immunostaining analysis showing that Bestl - / tdTomato + Bestl, Rpe65 and Cralbp are all negative in MEF. Scale bar, 50 μm. E is Bestl - / tdTomato + Brightfield and fluorescent images of MEF and ciRPE cells. Scale bar, 300 μm. F is immunostaining analysis showing that Bestl - / tdTomato + ZO-1, Rpe65, Mitf and Bestl positive in ciRPE cells derived from MEF. Scale bar, 20 μm.

[0058] Figure 11 Figure 2 is a molecular roadmap of ciRPE chemical reprogramming, where A is a heatmap of the expression of 16 fibroblast and 18 RPE cell-specific genes at designated time points during the reprogramming process. Representative genes for each population are listed (right). Red and blue colors indicate up- and down-regulated genes, respectively. B is hierarchical clustering analysis of cells at designated time points during the reprogramming of MEFs into ciRPE cells, including day 0, 7, 18, 32 and pRPE cells. C is a radar plot showing the average transcriptional activity of genes in ectoderm, mesoderm, endoderm, extraembryonic development and stem cell maintenance processes at designated time points during the reprogramming process. D is the changes in protein modifications at different time points. "Gain" indicates peaks of signal values that significantly increased between time points (logFC > 1 ), and "Loss" indicates peaks of signal values that significantly decreased (logFC < -1 ). E is enrichment curves of histone modifications H3K4me3 (left), H3K27ac (middle) and H3K27me3 (right) of differentially expressed genes at different time points during the reprogramming process. The x-axis represents the distance from the transcription start site (TSS) to the transcription end site (TES) (± 3 kb), and the y-axis represents the signal intensity. Each line corresponds to a specific time point, illustrating the dynamic changes in histone modification patterns during the reprogramming process.

[0059] Figure 12 is a graph of reprogramming scRNA-seq analysis results, wherein A is a UMAP density plot of the expression levels of marker genes in four different cell populations during reprogramming. Expression levels are represented by a color gradient, with darker red indicating higher gene expression. B is the proportion of cell types at different time points during reprogramming. C is GO-term biological process enrichment analysis of different cell types identified from scRNA-seq data. D is a UMAP density plot of the expression levels of proliferation-related genes in neural progenitor-like intermediate cells during reprogramming. Expression levels are represented by a color gradient, with darker red indicating higher gene expression. E is a pseudo-time trajectory showing the proportion of cell types during reprogramming. F is the pseudo-time of each cell in the trajectory predicted by Monocle 2.

[0060] Figure 13 is a graph of analysis results of key TFs during reprogramming, wherein A is a STRING network plot of interactions between TFs identified by SCENIC during reprogramming. Nodes represent TFs, color-coded by cell type: MEF (green), Intermediate cells (light blue), EF-like cells (blue), and ciRPE cells (red). B-C are the evaluation of gene knockdown and overexpression efficiency in MEFs after infection with shAscl1 / oeAscl1 (B) or shGli2 / oeGli2 (C) for 72 hours. Gene expression levels were normalized to wild-type (WT) levels (set as 1), with scramble shRNA (scramble)-infected cells as a control. D is the proportion of tdTomato + cell morphology and proportion at day 32 after knockdown of Ascl1 or Olig2 at the indicated time points during M7+M3 induction. WT, wild type; Control KD, scramble shRNA-mediated knockdown. Scale bar, 200 μm. E is the relative reprogramming efficiency after overexpression of Ascl1 or Olig2 under the indicated conditions. The reprogramming efficiency of the first stage M7 induction was set as "1", with DMSO as a negative control. F is a scRCF network visualization of the possible signal cascade induced by M7 targeting key TFs. Orange rectangles represent perturbation small molecules, blue diamonds represent signal protein targets, white ovals represent intermediate signal proteins, and green hexagons represent TFs.

[0061] Figure 14 is a graph of results of ciRPE cell transplantation to restore retinal function in RCS rats, wherein A is a schematic diagram of the process of subretinal injection of tdTomato + ciRPE cell transplantation into the subretinal space of 3-week-old RCS rats. B is a schematic diagram of the process of subretinal injection of tdTomato +Posterior fundus bright field images of ciRPE cells. Red arrows indicate the bumps formed after transplantation. C is teratoma test of ciRPE cells. Right bar graph shows quantitative analysis data. N=15. D is representative histological analysis of RCS rat retinal sections of un-injected group and subretinal injection of tdTomato + Representative histological analysis of RCS rat retinal sections of ciRPE cell group (left), and quantitative evaluation of the mean outer nuclear layer (ONL) thickness 12 weeks after injection (right). N=10. Scale bar, 50 μm. Figure 14 All representative examples of at least three independent experimental data are given. All data are expressed as Mean ± SD. **P<0.01, **P<0.001. DETAILED DESCRIPTION

[0062] The application will be further described in conjunction with the specific embodiments, and the examples given are only to illustrate the application, but not to limit the scope of the application. The materials, reagents and the like used in the following examples are commercially available unless otherwise specified. The experimental methods in the following examples are conventional methods unless otherwise specified. The cells in the following examples are all cultured at 37°C in an environment of 5% (volume percentage) CO2 unless otherwise specified.

[0063] Example 1, scRCF system screens EF cell reprogramming small molecules

[0064] We established a two-step chemical induction strategy to reprogram fibroblasts into RPE cells. First, fibroblasts were converted into an intermediate EF-like state, and then differentiated into ciRPE cells. To address the challenge of identifying cell reprogramming small molecules, we developed scRCF ( Figure 1 A), which combines single-cell transcriptomics with gene co-expression networks to identify key TFs and target proteins, facilitating the rational selection of compounds. To refine their identification, we combined cell-level sequencing (DRUG-seq2) to improve the accuracy and efficiency of molecularly driven reprogramming.

[0065] Using this platform, we screened and analyzed scRNA-seq data of mouse embryonic fibroblasts (MEF) and EF cells ( Figure 1 B and Figure 7 A). Differential gene expression analysis identified 258 TFs associated with cell type conversion ( Figure 1 C). The improved SiPer20 framework calculated the similarity between these TFs and perturbation profiles from a small molecule database, pre-screening 489 signaling proteins. Further refinement by gene co-expression networks reduced the selection to 279 functionally relevant signaling proteins ( Figure 7B). This approach revealed key signaling relationships and transcriptional regulators necessary for cell conversion. To optimize the screening, a random walk algorithm for protein network modularization was used to identify protein modules that are enriched in the top-ranked genes of each pathway ( Figure 7 C) Effect scores were implemented and small molecules were prioritized according to pathway relevance. Ultimately, 41 high-potential candidate drugs were identified, including Wnt pathway modulators CKI-7 and CHIR-99021, RTK pathway inhibitors Orantinib and Rebastinib, epigenetic modulators Trichostatin-A and RG108, TGF-β / Smad inhibitor A 83-01, and NF-κΒ pathway inhibitor BMS-345541, among others ( Figure 1 D).

[0066] To further identify small molecules targeting EF reprogramming, we systematically evaluated 41 candidate compounds using DRUG-seq2. Considering the neuroectodermal origin of EFs, we prioritized molecules such as LDN193189 (BMP-I type receptor inhibitor) and A 83-01 (TGF-β-I type receptor) for their ability to inhibit mesoderm and endoderm differentiation. The addition of CKI-7, an ATP-competitive tyrosine kinase I inhibitor, promoted neuroectodermal formation. We selected LDN193189, A 83-01, and CKI-7 (LAC) as the base induction small molecules for generating EF cells and used DRUG-seq2 to identify other synergistic small molecules ( Figure 7 D). After 14 days of treatment with various small molecule combinations, DRUG-seq2 analysis revealed different molecular expression patterns and changes in target gene expression in different treatment groups, reflecting specific responses to each case ( Figure 1 E and Figure 1 F). Key regulatory small molecules that promote cell reprogramming were identified by effect scoring of gene sets associated with neuroectodermal and early EF development. Among them, Hh-Ag1.5, a Hedgehog signal antagonist, obtained the highest score. Other high-scoring compounds included GSK-3 inhibitors (CHIR-99021, 1-Azakenpaullone, Kenpaullone) and NF-κΒ pathway inhibitors (BMS-345541, WHI-P154) ( Figure 1 E). From the top-ranked candidate drugs, we selected seven molecules, Hh-Ag1.5, CHIR-99021, Golvatinib, Pirfenidone, BMS-345541, Masitinib, and RG108, based on gene expression activation profiles and removal of functionally redundant compounds ( Figure 1F), forming an optimized induction medium. Using this method, we identified 10 small molecules from 4319 candidate molecules capable of reprogramming MEF cells into EF cells. In summary, we established a systematic, single-cell transcriptome-driven compound screening system integrating computation and high-throughput. This method identified 10 candidate inducible components of small molecules that have the potential to drive effective cell reprogramming (F). Figure 7 E).

[0067] Example 2: Establishing a two-stage chemical reprogramming strategy to generate ciRPE cells

[0068] First, we optimized the serum-free, chemically defined culture medium containing these 10 small molecules from Example 1 and used it to treat MEFs. In Phase I, we observed epithelial-like cell colonies after 6 days of treatment. Figure 2 A). Then, by day 12, cell proliferation began to slow, but the clonal morphology showed more defined boundaries and distinct epithelial features (A). Figure 2 A). To enhance the expansion of cell clonal colonies, we mechanically removed some surrounding cells. As expected, the cell colonies gradually expanded. By day 18, the cells exhibited tight junctions, with some cells displaying a high nucleocytic ratio and typical early RPE features, such as cobblestone or hexagonal morphology. Figure 2 A). Further qPCR analysis showed that EF-related genes (Pax6, Sox2, Six3, and Vsx2) and early RPE development-related genes (Mitf and Best1) were significantly upregulated. Figure 2 B). These findings suggest that the 10 small molecule induction protocols we identified have the potential to convert MEFs into EF-like cells with early RPE characteristics.

[0069] To promote the further differentiation and maturation of EF-like cells into RPE cells, we introduced nicotinamide (NIC), retinoic acid (RA), and Activin A (hereinafter referred to as M3) during the second-stage induction process. NIC inhibited neurogenic differentiation and promoted RPE-specific gene expression by regulating epigenetic and metabolic states. RA promoted RPE differentiation and regulated the expression of RPE-specific genes to promote pigment production. Activin A promoted EF cell differentiation into RPE cells by activating the TGF-β / SMAD signaling pathway, regulating pigment production to maintain epithelial properties. After two weeks of treatment, cells exhibited typical hexagonal RPE morphology and pigmentation. Figure 2 C). Subsequent qPCR analysis confirmed significant upregulation of Best1 and mature RPE genes such as Rpe65, Tyr, Lhx2, Pmel, and Otx2. Figure 2(D) indicates that ciRPE cells were successfully generated. To further verify the effectiveness of reprogramming, we used a lineage tracing strategy to monitor the reprogramming process ( Figure 2 E). The Best1 gene is specifically expressed in RPE cells, and its promoter has been shown to effectively drive reporter gene expression. We used Best1-Cre / ROSA26tdTomato fluorescent reporter mice, and the results showed that tdTomato expression was stable and specific in RPE cells. Figure 8 A and Figure 8 B). We used fluorescence-activated cell sorting (FACS) to collect Best1-tdTomato negative (tdTomato) cells. - MEFs Figure 8 C). These tdTom - MEF is negative for the expression of RPE marker genes such as Mitf, Cralbp, Best1, and Rpe65. Figure 8 D and Figure 8 E), confirming the absence of residual RPE or progenitor cells. Then, we examined tdTom... - MEFs underwent a two-step chemical induction process (stage I with 10 small molecules, followed by stage II with M3). Following stage I induction (day 18), low levels of tdTomato expression were observed, with a positive rate of 2.51%. After stage II induction for further maturation into RPE cells (day 32), the tdTomato positive rate increased to 16.58%. Figure 2 (F and 2G). These results confirm that the constructed chemical induction system effectively reprogrammed MEF cells into RPE cells.

[0070] To minimize the potential cytotoxic effects of small molecules, we conducted a "-1" experiment to further optimize the chemical reprogramming system. The results showed that removal of Golvatinib, Pirfenidone, and Masitinib during the first induction phase had the least impact on the tdTomato positivity rate. Figure 2 H), the positive rate was 19.66% ( Figure 2 I). Therefore, we finally identified the reprogramming induction system, which consists of seven small molecules from stage I (M7: LDN193189, A 83-01, CKI-7, Hh-Ag1.5, CHIR-99021, BMS-345541, and RG108) and three compounds from stage ⅠⅠ (M3: NIC, RA, and Activin A). Figure 2 Finally, we validated the effectiveness of our reprogramming system in MEFs from different batches (n=5) and different genetic backgrounds (including C57BL / 6 and 129). Figure 8 F andFigure 8 G). In addition, this M7+M3 system successfully promoted the conversion of tail tip fibroblasts (TTFs) from newborn mice into ciRPE cells ( Figure 8 F and Figure 8 G). In summary, these results indicate that our optimized M7+M3 system can efficiently reprogram MEFs into ciRPE cells.

[0071] Example 3, Characterization of ciRPE cells

[0072] To further confirm the phenotypic and functional properties of ciRPE cells, we first evaluated the expression of RPE markers by immunofluorescence staining. The results showed that ciRPE cells exhibited tight junction structures (ZO-1) and highly expressed RPE markers, including Pax6, Rpe65, Mitf, Bestl, and Cralbp ( Figure 3 A). In addition, these cells exhibited polarized morphology, with ZO-1 localized at the apical membrane and Bestl at the basal membrane ( Figure 3 B). This polar structure is crucial for RPE function, especially the apical microvilli, which mediate the phagocytosis and clearance of shed photoreceptor outer segments, ensuring normal photoreceptor turnover. Subsequently, we analyzed the structural features of ciRPE cells using transmission electron microscopy. These cells exhibited distinct apical microvilli, pigment granules, and tight junctions ( Figure 3 C), indicating that their morphology is highly similar to that of native RPE cells in vivo.

[0073] To assess the phagocytic ability of ciRPE cells, we introduced fluorescently labeled porcine photoreceptor outer segments (POS) and latex beads into the ciRPE cell culture medium. After a period of incubation, laser scanning confocal microscopy showed the presence of phagocytosed POS and latex beads within ciRPE cells ( Figure 3 D and Figure 9 E), confirming their phagocytic function. In addition, RPE cells are known to exhibit polarized secretion of growth factors, a process that is crucial for maintaining homeostasis between the retina and choroid. We further analyzed the polarized secretion of growth factors in ciRPE cells and found that these cells predominantly secreted VEGF from the basal side and PEDF from the apical side ( Figure 3 E). Moreover, ciRPE cells formed dome-like structures during in vitro culture ( Figure 3 F), indicating that their epithelial layer can efficiently transport fluid while maintaining the integrity of tight junctions. Measurement of transepithelial electrical resistance (TEER) further demonstrated the integrity of tight junctions and barrier function in ciRPE cells, with results showing a steady increase in TEER to approximately 80 Ω x cm2 within the first three weeks, followed by stabilization ( Figure 3G). In summary, these data indicate that ciRPE cells exhibit key functional characteristics of native RPE cells.

[0074] To further validate the proliferative potential of ciRPE cells, we purified them and passaged them continuously in expansion medium containing basic fibroblast growth factor (bFGF) and epidermal growth factor (EGF). The results showed that ciRPE cells could be passaged for at least 20 generations while maintaining stable RPE morphology and normal karyotype. Figure 9 A and Figure 9 B). In contrast, primary RPE (pRPE) cells gradually lost pigmentation and hexagonal morphology by the 6th generation, and their growth rate decreased significantly. Figure 9 B and Figure 9 C), which is consistent with previous research. In different passage batches, the proliferation capacity of ciRPE cells was significantly higher than that of pRPE cells, and the EdU incorporation rate consistently exceeded that of pRPE cells (C). Figure 9 C). Flow cytometry analysis showed that the cell cycle distribution of ciRPE cells (P3) was similar to that of pRPE cells, with 70%, 17.4%, and 6.87% of cells in the G0 / G1, S, and G2 / M phases, respectively. Figure 9 D). In summary, our results indicate that ciRPE cells are very similar to native RPE cells in morphology and function, exhibiting typical polarization characteristics, phagocytic capacity, and polarized secretion of growth factors. Furthermore, they possess strong proliferative capacity and stability.

[0075] Finally, we validated the process of MEFs reprogramming into ciRPE cells using a pedigree tracing strategy with Fsp1-Cre and ROSA26-tdTomato mice. MEFs were isolated from E13.5 transgenic mice with the Fsp1-Cre / ROSA26-tdTomato genetic background, and tdTomato cells were obtained through FACS sorting. + / Best1 - cell( Figure 10 A and Figure 10 B). These cells initially do not express the RPE marker (B). Figure 10 C and Figure 10 D). After chemical reprogramming, ciRPE cells began to express RPE-specific genes and co-express tdTomato ( Figure 10 E and Figure 10 F), further confirming that these cells originated from the original MEFs.

[0076] Example 4: Molecular route diagram of ciRPE chemical reprogramming

[0077] To enhance the resolution of the complex process of reprogramming MEFs into ciRPE cells, we performed a comprehensive multi-omics analysis ( Figure 4 A). RNA sequencing (RNA-seq) was performed on cells at day 0, day 7, day 18, and day 32 of the reprogramming process. Principal component analysis (PCA) of the RNA-seq results revealed significant transcriptional changes throughout the dynamic process, indicating a gradual transition from fibroblasts to ciRPE cells. Notably, the final ciRPE cells exhibited greater similarity to pRPE cells at the transcriptional level compared to earlier stages ( Figure 4 B). Time-series fuzzy clustering analysis of differentially expressed genes showed four distinct gene clusters corresponding to each stage of reprogramming: fibroblasts (day 0), intermediate state (day 7), EF-like cells (day 18), and mature ciRPE cells (day 32) ( Figure 4 C).

[0078] At the initial stage of transition from fibroblasts to the intermediate state, downregulated genes such as Fn1, Itga11, and Runx2 are associated with reduced fibroblast function and proliferative activity. In contrast, upregulated genes include Sox2, Ascl1, Gli2, Gli1, Gbx2, which are associated with neuroectodermal processes ( Figure 4 C). These results indicate that early chemical induction promotes the transition to a neuronal fate by activating neuroectodermal genes and signaling pathways while suppressing fibroblast characteristics. Additionally, at the intermediate stage, transient activation of genes such as Tfap2a, Rorb, Etv5, and others are associated with epithelial cell proliferation and inflammation, reflecting enhanced cell-cell interactions and adaptation to environmental signals ( Figure 4 C). By day 18, upregulated genes are primarily associated with eye development, such as Pax6, ID4, and Best1. As induction progresses toward mature RPE, early EF-related genes exhibit a trend of downregulation. By day 32, genes associated with RPE function (e.g., Rpe65, Lhx2, Tyr, Cralbp, Otx2) and pigmented cell differentiation (e.g., Pmel, Mlana, Slc24a5) are significantly upregulated ( Figure 4 C), confirming the transition of cells to mature pigmented ciRPE cells.

[0079] To confirm the successful transformation of fibroblasts into ciRPE cells, we analyzed 16 fibroblast-specific genes and 18 RPE-specific genes. During the reprogramming process, fibroblast markers were significantly downregulated, while RPE markers were upregulated, indicating that ciRPE cells acquired RPE characteristics ( Figure 11A). Ultimately, the gene expression profile of ciRPE cells highly matched that of pRPE cells, further confirming that these cells possessed RPE cell properties ( Figure 4 B, Figure 4 C, Figure 11 A and Figure 11 B). Remarkably, the reprogramming process did not involve an iPSC stage; instead, transcriptional activation was mainly focused on ectoderm-related genes ( Figure 11 C). This indicates that the combination of compounds we identified specifically directed fibroblasts towards neural ectoderm and eye developmental pathways, highlighting the effectiveness of our screening strategy.

[0080] Epigenetic remodeling plays a key role in cell fate reprogramming. We used CUT&Tag sequencing to track the dynamics of H3K4me3, H3K27ac and H3K27me3 histone modifications around the transcription start sites of genes. PCA analysis revealed significant changes in chromatin state during the ciRPE cell reprogramming process, indicating that ciRPE cells were very similar to pRPE cells in terms of histone modification patterns ( Figure 4 B). H3K4me3 gradually increased from MEFs to ciRPE cells, peaking at day 18 and day 32, associated with the activation of RPE genes. Loss of H3K4me3 decreased over time, indicating its role in establishing new gene expression patterns. H3K27ac varied at different stages, decreasing from day 7 to day 18, indicating that genes suppressed EF differentiation. The dynamic changes of H3K27me3 reflected gene silencing: a significant loss from MEFs to day 7 of induction promoted differentiation, while an increase from day 7 to day 18 established new silencing. By the ciRPE stage, H3K27me3 levels tended to stabilize, maintaining a mature RPE phenotype ( Figure 11 D). Further analysis found that CUT&Tag signals of H3K4me3 and H3K27ac were correlated with RNA expression levels, emphasizing their role in gene activation ( Figure 4 D). Although H3K27me3 was generally negatively correlated with RNA-seq data, there was a certain bias in the fibroblast stage, possibly due to the dominance of activation markers. From day 7 to day 18, the increase in H3K27me3 was associated with a decrease in RNA expression, while the stable trend in the ciRPE stage reflected the balance between gene silencing and activation ( Figure 4 D and Figure 11 E). These findings highlight the complex role of dynamic histone modifications in cell type transformation at different stages, emphasizing the key role of epigenetic regulation in MEFs reprogramming to RPE. By regulating gene activation and silencing, cells successfully reprogrammed to the RPE lineage.

[0081] To accurately describe the reprogramming trajectory, we performed scRNA-seq at three key time points of MEFs reprogramming into ciRPE cells. We obtained single-cell transcriptome data of 9679, 13227, and 11282 cells from samples collected at day 7, 18, and 32, respectively. Figure 4 E). Specific marker gene clustering analysis showed that at day 7 of reprogramming, the cell population consisted of MEF cells, neural progenitor-like intermediate cells, and a small fraction of EF-like cells. By day 18, the proportion of intermediate cells decreased, and the proportion of EF-like cells increased. At day 32, EF-like cells decreased, and a large number of ciRPE cells appeared Figure 4 F, Figure 4 G, Figure 12 A and Figure 12 B). To further validate the reprogramming trajectory of ciRPE cells, we performed single-cell pseudotime trajectory analysis. The analysis revealed a trajectory from MEFs to neural progenitor-like intermediate cells, followed by differentiation into EF-like cells, and finally into ciRPE cells Figure 4 H). Unsupervised clustering highlighted dynamic molecular events during this process, including the transition from fibroblasts to neural progenitors, progression through stages of eye development, and establishment of RPE-specific functions (pigmentation and retinoid metabolism) Figure 12 C). Interestingly, among the two identified neural progenitor-like subpopulations, the differentiation trajectory of MEFs was mainly consistent with the subpopulation expressing neurogenic and proliferative markers (e.g., Mki67, Cdk1, Top2a), while the other subpopulation lacked these markers Figure 12 D). This suggests that proliferative neural progenitors are crucial for subsequent differentiation into EF-like cells. RNA rate analysis similarly reflected the progression toward ciRPE cell reprogramming, consistent with the trajectory observed in pseudotime analysis Figure 12 E and Figure 12 F). We further performed similarity analysis on the reprogrammed cell population and publicly available single-cell datasets. The induced cells scored 0.80 similarity to RPE cells and 0.64 similarity to EF-like cells Figure 4 I), indicating that the reprogrammed cells possessed EF-like and RPE cell characteristics. In summary, our two-step chemical reprogramming system effectively guided MEFs through a neural progenitor-like intermediate state and EF-like cell stage to generate functional ciRPE cells. This series of processes emphasizes the precise control of lineage-specific gene expression and epigenetic remodeling.

[0082] Example 5, Transcriptional activation of master regulators of neural and eye development controls ciRPE reprogramming

[0083] To determine the key TFs driving MEF reprogramming into ciRPE cells, we combined single-cell data with SCENIC analysis to systematically investigate the regulatory factors expressed in each cell type and their dynamics during reprogramming. Setting a logFC > 1.0 as a threshold, we identified 125 TFs in the four cell types, highlighting their distinct roles at different stages of reprogramming ( Figure 5 A). Using the STRING database, we constructed an interaction network of these TFs, revealing their synergistic effects during reprogramming ( Figure 13 A). Network analysis using the maximum clique centrality (MCC) identified 15 key TFs, including MEF-specific factors (Dlx2 and Dlx1), transiently activated factors at day 7 (Soxl l, Foxa2, and Pax2), factors persistently highly expressed from day 7 to day 18 (Sox2, Sox9, Olig2, Zicl, Ascll, Atohl, Pou3f2, and Gbx2), and ciRPE-associated factors (Lhx2 and Otx2) ( Figure 5 B and 5C). Notably, most of these key TFs are associated with neuroectoderm development, suggesting that the fate of developing into neuroectoderm is a critical transitional stage for MEFs to reprogram into EF-like cells and ciRPE cells.

[0084] To further explore the regulatory mechanisms of these TFs, we integrated RNA-seq and CUT&Tag-seq data to assess the activities of the top 5 TFs ( Figure 5 D). At day 0, neuroectoderm TFs, including Ascll, Olig2, Zicl, Pou3f2, showed weak activation marks (H3K4me3 and H3K27ac) and were mainly suppressed by H3K27me3, consistent with their repression in the stable fibroblast state. By day 7, when cells transitioned to an intermediate state like neural progenitors, these TFs showed elevated activation marks and RNA expression, indicating activation of the neuroectoderm pathway. This coincided with repression of fibroblast features and initiation of neurogenesis. By day 18, Ascll and Olig2 activity decreased, while Zicl and Pou3f2 remained highly activated. At day 32, in the ciRPE stage, Lhx2 exhibited robust activation and significant RNA upregulation, while early neuroectoderm TFs were silenced ( Figure 5D). Notably, while H3K27me3 played a critical inhibitory role at early stages (0-7 days), its role was attenuated at later stages, suggesting that other inhibitory mechanisms (such as DNA methylation or other histone marks) might be at play in regulating TF activity at this stage. These findings highlight the critical role of dynamic epigenetic modifications and TF expression in guiding cell fate transitions during MEFs reprogramming into ciRPEs. Overall, neural development-related TFs are the "switches" for early-stage differentiation into EFs, while RPE-related TFs dominate later stages, driving RPE maturation.

[0085] To determine the effect of Ascll and Olig2 on fibroblast-to-EF-like cell reprogramming, we investigated the effect of knocking down these genes individually using shRNAs on the efficiency of ciRPE induction. As expected, knockdown of Ascll or Olig2 significantly reduced reprogramming efficiency, confirming its positive regulatory role in reprogramming Figure 5 E, Figure 5 F, Figure 13 B- Figure 13 D). Importantly, knockdown of these genes had the most significant effect on reprogramming efficiency at early stages (9.1-fold, 16.7-fold, and 3.6-fold reduction when shAscll was transfected at day 0, 7, and 18, respectively, Figure 5 E; 12.5-fold, 8.3-fold, and 2.8-fold reduction when shOlig2 was transduced at day 0, 7, and 18, respectively, Figure 5 F), highlighting their critical role in initiating the neural ectoderm transition required for subsequent ciRPE differentiation. On the other hand, overexpression of Ascll or Olig2 enhanced M7+M3-mediated ciRPE reprogramming efficiency Figure 13 E), verifying that Ascll and Olig2 are directly involved in ciRPE reprogramming.

[0086] These findings highlight the importance of transitioning to neural ectoderm and EF-like cell states for successful reprogramming of MEFs into ciRPE cells, verifying the effectiveness of our first-stage small-molecule compound combination in orchestrating these transitions. To gain deeper insights into the mechanism by which these small molecules regulate key TFs, we constructed a complex regulatory network using scRCF Figure 5 G and Figure 13 E). Analysis revealed that activation of key TFs was a result of the collective action of these small molecules, which modulated TFs through their respective target proteins, coordinating activation of multiple signaling pathways, including Wnt, TGF-beta, and Hedgehog, collectively driving the reprogramming process and guiding the precise transition of cell fate towards ciRPE reprogramming.

[0087] Example 6. ciRPE cell transplantation restores retinal function in RCS rats

[0088] With the functional and safety advantages of ciRPE cells, we performed in vivo transplantation studies to evaluate their potential to treat RD disease. Royal College of Surgeons (RCS) rats, which develop retinal degeneration due to impaired phagocytosis of photoreceptor outer segments by RPE cells caused by a mutation in the Mertk gene, are a classic model for studying RD disease. We transplanted FACS-purified tdTomato + ciRPE cells into the subretinal space of 3-week-old RCS rats Figure 6 A, Figure 14 A and Figure 14 B). The contralateral non-transplanted eyes, PBS-transplanted eyes, and mESCs-transplanted eyes served as control groups. Postoperative optical coherence tomography (OCT) imaging showed a clear bulge at the transplantation site, which significantly decreased after one week and completely disappeared after three weeks, indicating successful transplantation and good tissue adaptation Figure 6 B). To evaluate the long-term safety of transplantation, we performed a 4-month follow-up evaluation. During this period, no signs of tumor formation were observed in the ciRPE-transplanted nude mice. In contrast, visible intraocular tumors were observed in 13 of 15 nude mice that received tdTomato-labeled mESCs Figure 6 C). This result is consistent with the results of the subcutaneous teratoma experimental study, further confirming that ciRPE cell transplantation has no risk of tumorigenicity, thus supporting its safety Figure 14 C).

[0089] Histological and immunostaining analysis at 4 weeks post-transplantation showed the presence of tdTomato + transplanted cell clusters in the subretinal space Figure 6 D). By 12 weeks, these cell tissues had become a monolayer of ordered arrangement, indicating their successful integration into the host RPE Figure 6 E). In addition, the transplanted cells expressed mature RPE markers such as Mitf, Cralbp, Pax6, and Rpe65 Figure 6 F, Figure 6 G, Figure 6 H). Typically, RCS rats exhibit severe retinal dysfunction at 2 to 3 months of age, characterized by significant photoreceptor apoptosis and thinning of the outer nuclear layer (ONL). At 12 weeks post-transplantation, the ONL in the ciRPE-transplanted group was significantly thicker than in the non-transplanted and Sham groups Figure 6 E and Figure 14 D), indicating that ciRPE transplantation protected photoreceptors and slowed their degeneration, thereby maintaining and improving the structure and function of the retina. TUNEL staining further showed a significant reduction in apoptotic cells in the ONL of the ciRPE group compared to the Sham groupFigure 6 I), highlighting the protective effect of ciRPE cells on photoreceptor apoptosis. RCS rats have a loss of RPE phagocytic function due to a Mertk mutation, affecting the clearance of the outer segment of photoreceptors. To assess whether ciRPE cells could restore this function in vivo, we monitored their uptake of Rhodopsin, a major component of the outer segment of photoreceptors. Co-localization of tdTomato and Rhodopsin was observed in the subretinal space of rats 12 weeks after transplantation of ciRPE cells Figure 6 J), indicating that transplanted cells successfully integrated and restored the phagocytic function of RPE cells.

[0090] To fully assess the impact of ciRPE cell transplantation on retinal function, we performed dark-adapted flash electroretinogram (fERG) and behavioral tests. fERG data collected 4-16 weeks after transplantation showed that b-wave amplitudes were significantly higher in the ciRPE group than in the control group, especially in the first weeks after transplantation Figure 6 K). This indicates that ciRPE cells support photoreceptor survival and functional recovery. The sustained increase in b-wave amplitude reflects both short-term and long-term effects of cell integration. To assess whether these electrophysiological improvements translated into enhanced visual performance, we performed optomotor response (OMR) tests. The ciRPE group presented superior visual performance to the control group, such as faster responses to moving stripe stimuli, higher sensitivity to spatial frequency and changes in movement direction, which could be quantified by increased OMR scores Figure 6 L). These findings confirm the long-term survival, safety, and therapeutic effect of ciRPE cells in vivo, their ability to successfully integrate into host tissue, reconstitute functional RPE structures, and restore visual function, emphasizing their potential as a potential cell source therapy for RD disease.

Claims

1. A method for inducing embryonic fibroblasts into functional ciRPE cells through a two-step reprogramming process using a small chemical molecule composition for non-therapeutic purposes, comprising: Embryonic fibroblasts were treated with reprogramming medium 1 to induce them to be reprogrammed into EF-like or OV-like cells. Then, EF-like or OV-like cells were treated with reprogramming medium 2 to induce them to be reprogrammed into ciRPE cells. The reprogramming culture medium 1 includes a small chemical molecule composition 1, and the reprogramming culture medium 2 includes a small chemical molecule composition 2. The embryonic fibroblasts are mouse-derived embryonic fibroblasts; The chemical small molecule composition 1 specifically comprises LDN193189, A 83-01, CKI-7, Hh-Ag1.5, CHIR-99021, BMS-345541, and RG108; The chemical small molecule composition 2 specifically comprises nicotinamide, retinoic acid, and Activin A.

2. The method as described in claim 1, wherein the reprogramming culture medium 1 and the reprogramming culture medium 2 are respectively composed of a basal culture medium to which the chemical small molecule composition 1 and the chemical small molecule composition 2 are added for culturing.

3. The method of claim 2, wherein the reprogramming medium 1 uses basal medium 1, specifically an equal volume of Neurobasal and DMEM / F12 / GlutaMAX supplemented with 1% N2, 1% vitamin A-free B27, 7.5% BSA, 1% NEAA, 1% P / S and 10 ng / mL bFGF.

4. The method of claim 2, wherein the reprogramming medium 2 uses basal medium 2, specifically DMEM / F12 / GlutaMax supplemented with 10% KSR, 1% NEAA, 1% P / S and 0.055 mM 2-mercaptoethanol.

5. The method of claim 1, further comprising proliferating and culturing ciRPE cells.

6. The method of claim 5, wherein the proliferation culture uses a ciRPE cell proliferation medium with the following components: DMEM / F12 / GlutaMAX as substrate, supplemented with 1% N2, 2% vitamin A-free B27, 1% NEAA, 1% P / S and 0.1 mM 2-mercaptoethanol.

7. The method of claim 6, wherein the proliferation medium for ciRPE cells further comprises one or more of 10 ng / mL bFGF, 20 ng / mL EGF, 10 μM Y-27632, and 0.5 μM A 83-01.

8. The method of claim 1, further comprising functional culture of ciRPE cells.

9. The method of claim 8, wherein the functional culture uses a ciRPE cell function maintenance medium with the following components: DMEM / F12 / GlutaMAX plus 1% N2, 2% vitamin A-containing B27, 1% NEAA, 1% P / S and 0.1 mM 2-mercaptoethanol.

10. The method of claim 9, wherein the function maintenance culture medium is further supplemented with 0.2 μM Activin A, 0.5 μM retinoic acid, 1 μM BMP4 and 10 mM nicotinamide.

11. The method of claim 1, wherein the concentration of LDN193189 in the small molecule chemical composition 1 is 0.1 mM, the concentration of A83-01 is 0.5 mM, the concentration of CKI-7 is 5 μM, the concentration of Hh-Ag1.5 is 0.5 mM, the concentration of CHIR-99021 is 3 μM, the concentration of BMS-345541 is 0.2 mM, and the concentration of RG108 is 10 μM.

12. The method of claim 1, wherein the concentration of nicotinamide in the small molecule chemical composition 2 is 10 mM, the concentration of retinoic acid is 1 μM, and the concentration of Activin A is 0.2 mM.

13. A small molecule chemical composition comprising small molecule chemical composition 1 and small molecule chemical composition 2; The chemical small molecule composition 1 specifically comprises LDN193189, A 83-01, CKI-7, Hh-Ag1.5, CHIR-99021, BMS-345541, and RG108, and the chemical small molecule composition 2 specifically comprises nicotinamide, retinoic acid, and Activin A.

14. The small molecule chemical composition of claim 13, wherein the concentration of LDN193189 in the small molecule chemical composition 1 is 0.1 mM, the concentration of A 83-01 is 0.5 mM, the concentration of CKI-7 is 5 μM, the concentration of Hh-Ag1.5 is 0.5 mM, the concentration of CHIR-99021 is 3 μM, the concentration of BMS-345541 is 0.2 mM, and the concentration of RG108 is 10 μM; and the concentration of nicotinamide in the small molecule chemical composition 2 is 10 mM, the concentration of retinoic acid is 1 μM, and the concentration of Activin A is 0.2 mM.

15. A reprogramming culture medium assembly, the reprogramming culture medium assembly comprising reprogramming culture medium 1 and reprogramming culture medium 2; The reprogramming culture medium 1 comprises a small molecule chemical composition 1 and a basal culture medium 1. The small molecule chemical composition 1 specifically consists of LDN193189, A83-01, CKI-7, Hh-Ag1.5, CHIR-99021, BMS-345541, and RG108. The basal culture medium 1 specifically consists of equal volumes of Neurobasal and DMEM / F12 / GlutaMAX supplemented with 1% N2, 1% vitamin A-free B27, 7.5% BSA, 1% NEAA, 1% P / S, and 10 ng / mL bFGF. The reprogramming culture medium 2 comprises a small molecule chemical composition 2 and a basal culture medium 2. The small molecule chemical composition 2 specifically consists of nicotinamide, retinoic acid, and Activin A. The basal culture medium 2 specifically consists of DMEM / F12 / GlutaMax supplemented with 10% KSR, 1% NEAA, 1% P / S, and 0.055 mM 2-mercaptoethanol.

16. The reprogrammed culture medium combination as described in claim 15, wherein the concentration of LDN193189 in the small molecule chemical composition 1 is 0.1 mM, the concentration of A 83-01 is 0.5 mM, the concentration of CKI-7 is 5 μM, the concentration of Hh-Ag1.5 is 0.5 mM, the concentration of CHIR-99021 is 3 μM, the concentration of BMS-345541 is 0.2 mM, and the concentration of RG108 is 10 μM; and the concentration of nicotinamide in the small molecule chemical composition 2 is 10 mM, the concentration of retinoic acid is 1 μM, and the concentration of Activin A is 0.2 mM.

17. The use of a small chemical molecule composition in the non-therapeutic chemical reprogramming induction of embryonic fibroblasts into functional ciRPE cells, wherein the small chemical molecule composition is the small chemical molecule compound of claim 13 or 14; The embryonic fibroblasts are mouse-derived embryonic fibroblasts.

18. A kit for inducing two-step reprogramming of embryonic fibroblasts into ciRPE cells using a small chemical molecule composition, said kit comprising the small chemical molecule composition of claim 13 or 14, or the reprogramming culture medium composition of claim 15 or 16; The embryonic fibroblasts are mouse-derived embryonic fibroblasts.

Images