Single-cell RNA-seq is a powerful tool in decoding the heterogeneity in complex tissues by generating transcriptomic profiles of the individual cell. Here, we report a single-nuclei RNA-seq (snRNA-seq) transcriptomic study on human retinal tissue, which is composed of multiple cell types with distinct functions. Six samples from three healthy donors are profiled and high-quality RNA-seq data is obtained for 5873 single nuclei. All major retinal cell types are observed and marker genes for each cell type are identified. The gene expression of the macular and peripheral retina is compared to each other at cell-type level. Furthermore, our dataset shows an improved power for prioritizing genes associated with human retinal diseases compared to both mouse single-cell RNA-seq and human bulk RNA-seq results. In conclusion, we demonstrate that obtaining single cell transcriptomes from human frozen tissues can provide insight missed by either human bulk RNA-seq or animal models.

26Gene expression profiling is an effective way to provide insights into cell function. 27 However, for heterogeneous tissues, bulk RNA-Seq can only provide the average gene 28 expression profile for all cells from the tissue, making the interpretation of the sequencing 29 result challenging. Single-cell RNA-seq, on the other hand, generates transcriptomic 30 profiles of individual cell and cell types, making it a powerful method to decode the 31 heterogeneity in complex tissues. 32 The retina is a heterogeneous tissue composed of multiple cell types with distinct 33 functions. Here we report the first single-nuclei RNA-seq transcriptomic study on human 34 neural retinal tissue to identify transcriptome profile for individual cell types. Six retina 35 samples from three healthy donors were profiled and RNA-seq data with high quality was 36 obtained for 4730 single nuclei. All seven major cell types were observed from the dataset 37 and signature genes for each cell type were identified by differential gene express 38 analysis. The gene expression of the macular and peripheral retina was compared at the 39 cell type level, showing significant improvement from previous bulk RNA-seq studies. 40Furthermore, our dataset showed improved power in prioritizing genes associated with 41 human retinal diseases compared to both mouse single-cell RNA-seq and human bulk 42 RNA-seq results. In conclusion, we demonstrated that feasibility of obtaining single cell 43 transcriptome from human frozen tissues to provide additional insights that is missed by 44 either the human bulk RNA-seq or the animal models. 45 46 48 cell type and state, and investigating human diseases 1-3 . Transcriptome can be more 49 powerful when combined with other 'omics' data to build prediction models on human 50 diseases 4 . For example, by combining transcriptomic data and proteomic data, a list of 51 candidate disease genes can be predicted with high specificity 5 . However, until recently, 52 vast majority transcriptome profiles are generated from profiling tissue samples 53 containing thousands to millions of cells. Thus, gene expression information of individual 54 cells would be lost. For tissues with high cellular heterogeneity, knowing the transcriptome 55 profiles of each cell type would be important for both identification of novel cell types and 56 understanding the functional organization of the tissue. Cell sorting would be required to 57 obtain transcriptome of a single cell type; not only was it not always practical, but also the 58 heterogeneities of many tissues were not fully revealed. This gap was met by the 59 development of the high throughput single-cell RNA-seq technology 6-8 . 60 61 Transcriptomic studies on the single cell level was first performed decades ago 9,10 , while 62 the first single-cell transcriptome study based on Next-Generation Sequencing was 63 reported ten years ago 11 . Since then, technologies have been dramatically improved in 64 scale and sensitivity. Development in single-cell isolation, such as microf...

Much of the complexity of the eukaryotic cell transcriptome is due to the alternative splicing of mRNA. However, knowledge on how transcriptome complexity is translated into functional complexity remains limited. For example, although different isoforms of a gene may show distinct temporal and spatial expression patterns, it is largely unknown whether these isoforms encode proteins with distinct functions matching their expression pattern. In this report, we investigated the function and relationship of the two isoforms of Reep6, namely Reep6.1 and Reep6.2, in rod photoreceptor cells. These two isoforms result from the alternative splicing of exon 5 and show mutually exclusive expression patterns. Reep6.2 is the canonical isoform that is expressed in non-retinal tissues while Reep6.1 is the only expressed isoform in the adult retina. The Reep6.1 isoform-specific knockout mouse, Reep6E5/E5, is generated by deleting exon 5 and a homozygous deletion phenotypically displayed a rod degeneration phenotype comparable to a Reep6 full knockout mouse, indicating that the Reep6.1 isoform is essential for the rod photoreceptor cell survival. Consistent with the results obtained from a loss-of-function experiment, overexpression of Reep6.2 failed to rescue the rod degeneration phenotype of Reep6 knockout mice while overexpression of Reep6.1 does lead to rescue. These results demonstrate that, consistent with the expression pattern of the isoform, Reep6.1 has rod-specific functions that cannot be substituted by its canonical isoform. Our findings suggested that a strict regulation of splicing is required for the maintenance of photoreceptor cells.

Hereditary retinal dystrophy is clinically defined as a broad group of chronic and progressive disorders that affect visual function by causing photoreceptor degeneration. Previously, we identified mutations in the gene encoding receptor expression-enhancing protein 6 (REEP6), in individuals with autosomal recessive retinitis pigmentosa (RP), the most common form of inherited retinal dystrophy. One individual was molecularly diagnosed with biallelic REEP6 mutations, a missense mutation over a frameshift mutation. In this study, we generated Reep6 compound heterozygous mice, Reep6, which mimic the patient genotype and recapitulate the early-onset retinal degeneration phenotypes observed in the individual with RP. To determine the feasibility of rescuing the Reep6 mutant phenotype via gene replacement therapy, we delivered Reep6.1, the mouse retina-specific isoform of REEP6, to photoreceptors of Reep6 mutant mice on postnatal day 20. Evaluation of the therapeutic effects 2 months posttreatment showed improvements in the photoresponse as well as preservation of photoreceptor cells. Importantly, guanylyl cyclase 1 (GC1) expression was also restored to the outer segment after treatment. Furthermore, rAAV8-Reep6.1 single treatment in Reep6 mutant mice 1 year postinjection showed significant improvements in retinal function and morphology, suggesting that the treatment is effective even after a prolonged period. Findings from this study show that gene replacement therapy in the retina with rAAV overexpressing Reep6 is effective, preserving photoreceptor function in Reep6 mutant mice. These findings provide evidence that rAAV8-based gene therapy can prolong survival of photoreceptors in vivo and can be potentially used as a therapeutic modality for treatment of patients with RP.