Nicotinamide promotes formation of retinal organoids from human pluripotent stem cells via enhanced neural cell fate commitment

Retinal organoids (ROs) derived from human pluripotent stem cells (hPSCs) recapitulate key features of human retinogenesis and thus provide a promising platform to study the mechanisms of retinal development and disease in a human context and to evaluate therapies. Although multiple differentiation protocols are currently in use, hPSCs exhibit tremendous variability in differentiation efficiency, with some cell lines consistently yielding few or even no ROs thus limiting their utility in research. To improve the efficiency and robustness of RO generation, we set out to identify factors that promote neural retina differentiation from hPSCs. We report here that nicotinamide (NAM) treatment at the early stage of differentiation significantly improved RO yield across 8 hPSC lines from different donors. Importantly, NAM treatment enabled efficient production of ROs from cell lines that would otherwise fail to generate meaningful number of ROs. Further analyses revealed that NAM treatment promotes neural commitment of hPSCs at the expense of non-neural ectodermal cell fate, which in turn increases eye field commitment and RO generation. This effect is mediated, at least in part, through inhibition of BMP signaling. Thus, our modified protocol with a simple NAM treatment improves the yield of ROs in all lines tested, and importantly greatly benefits to cell lines that previously proved to be intractable in retinal differentiation. Our data should facilitate the broader use of human ROs for disease modeling applications requiring the use of multiple cell lines or a larger scale of cell source such as therapy screening. Overall design: Retinal organoids were generated two human induced pluripotent stem cell lines with and without nicotinamide supplementation. Total RNA was extracted at two time points (D4 and D7) using RNAeasy Plus Mini kit (Qiagen, Germantown, MD). Quality of isolated RNAs was assessed using Bioanalyszer RNA 6000 nano assays and only high-quality RNAs (RIN > 8) were used for construction of mRNA sequencing library. 100ng of total RNA were used to construct the strand-specific libraries using TruSeq RNA Sample Prep Kit-v2 (Illumina, San Diego, CA). Libraries were sequenced on two lanes of an Illumina HiSeq 2500 in paired-end mode (125 bp read length). Reads passing the Illumina chastity filter were demultiplexed then trimmed to remove low quality bases and adapters using Trimmomatic v0.36. Next, reads were aligned to the human transcriptome (GRCH38.p13, Ensembl 98) and quantified to the transcript-level using Kallisto v0.45.0.