Network-based bioinformatics analysis of spatio-temporal RNA-Seq data reveals transcriptional programs underpinning normal and aberrant retinal development.

BACKGROUND: The retina as a model system with extensive information on genes involved in development/maintenance is of great value for investigations employing deep sequencing to capture transcriptome change over time. This in turn could enable us to find patterns in gene expression across time to reveal transition in biological processes. METHODS: We developed a bioinformatics pipeline to categorize genes based on their differential expression and their alternative splicing status across time by binning genes based on their transcriptional kinetics. Genes within same bins were then leveraged to query gene annotation databases to discover molecular programs employed by the developing retina. RESULTS: Using our pipeline on RNA-Seq data obtained from fractionated (nucleus/cytoplasm) developing retina at embryonic day (E) 16 and postnatal day (P) 0, we captured high-resolution as in the difference between the cytoplasm and the nucleus at the same developmental time. We found de novo transcription of genes whose transcripts were exclusively found in the nuclear transcriptome at P0. Further analysis showed that these genes enriched for functions that are known to be executed during postnatal development, thus showing that the P0 nuclear transcriptome is temporally ahead of that of its cytoplasm. We extended our strategy to perform temporal analysis comparing P0 data to either P21-Nrl-wildtype (WT) or P21-Nrl-knockout (KO) retinae, which predicted that the KO retina would have compromised vasculature. Indeed, histological manifestation of vasodilation has been reported at a later time point (P60). CONCLUSIONS: Thus, our approach was predictive of a phenotype before it presented histologically. Our strategy can be extended to investigating the development and/or disease progression of other tissue types.

In Vivo and Explant Electroporation of Morpholinos in the Developing Mouse Retina.

Neonatal in vivo electroporations and retinal explant electroporations have been widely employed in understanding the effects of loss or gain of function of protein-coding genes in retinal development. Here, we describe a rapid and efficient delivery of morpholinos to add another tool to perturb gene expression during mouse retinal development.

Loss of citron kinase affects a subset of progenitor cells that alters late but not early neurogenesis in the developing rat retina.

PURPOSE: To understand how loss of citron kinase (CitK) affects retinal progenitor cells (RPCs) in the developing rat retina. METHODS: We compared knockout (KO) and wild-type (WT) retinae by immunohistochemistry. The TdT-mediated dUTP terminal nick-end labeling (TUNEL) assay was performed to determine cell death. Pulse-chase experiments using 5-ethynyl-2'-deoxyuridine (EdU) were carried out to interrogate RPC behavior and in turn neurogenesis. RESULTS: Reverse transcription-polymerase chain reaction analysis showed that CitK was expressed at embryonic day (E)12 and was turned off at approximately postnatal day (P)4. Immunohistochemistry showed CitK being localized as puncta at the apical end of the outer neuroblastic layer (ONBL). Analyses during embryonic development showed that the KO retina was of comparable size to that of WT until E13. However, by E14, there was a reduction in the number of S-phase RPCs with a concomitant increase in TUNEL+ cells in the KO retina. Moreover, early neurogenesis, as reflected by retinal ganglion cell production, was not affected. Postnatal analysis of the retina showed that ONBL in the KO retina was reduced to half the size of that in WT and showed further degeneration. Immunohistochemistry revealed absence of Islet1+ bipolar cells at P2, which was further confirmed by EdU pulse-chase experiments. The CitK KO retinae underwent complete degeneration by P14. CONCLUSIONS: Our study showed that CitK is not required for a subset of RPCs before E14, but is necessary for RPC survival post E14. This in turn results in normal early embryonic neurogenesis, but severely compromised later embryonic and postnatal neurogenesis.

Minor splicing snRNAs are enriched in the developing mouse CNS and are crucial for survival of differentiating retinal neurons.

In eukaryotes, gene expression requires splicing, which starts with the identification of exon-intron boundaries by the small, nuclear RNA (snRNAs) of the spliceosome, aided by associated proteins. In the mammalian genome, <1% of introns lack canonical exon-intron boundary sequences and cannot be spliced by the canonical splicing machinery. These introns are spliced by the minor spliceosome, consisting of unique snRNAs (U11, U12, U4atac, and U6atac). The importance of the minor spliceosome is underscored by the disease microcephalic osteodysplastic primordial dwarfism type 1 (MOPD1), which is caused by mutation in U4atac. Thus, it is important to understand the expression and function of the minor spliceosome and its targets in mammalian development, for which we used the mouse as our model. Here, we report enrichment of the minor snRNAs in the developing head/central nervous system (CNS) between E9.5 and E12.5, along with enrichment of these snRNAs in differentiating retinal neurons. Moreover, dynamic expression kinetics of minor intron-containing genes (MIGs) was observed across retinal development. DAVID analysis of MIGs that were cotranscriptionally upregulated embryonically revealed enrichment for RNA metabolism and cell cycle regulation. In contrast, MIGs that were cotranscriptionally upregulated postnatally revealed enrichment for protein localization/transport, vesicle-mediated transport, and calcium transport. Finally, we used U12 morpholino to inactivate the minor spliceosome in the postnatal retina, which resulted in apoptosis of differentiating retinal neurons. Taken together, our data suggest that the minor spliceosome may have distinct functions in embryonic versus postnatal development. Importantly, we show that the minor spliceosome is crucial for the survival of terminally differentiating retinal neurons.

Replication-dependent histone genes are actively transcribed in differentiating and aging retinal neurons.

In the mammalian genome, each histone family contains multiple replication-dependent paralogs, which are found in clusters where their transcription is thought to be coupled to the cell cycle. Here, we wanted to interrogate the transcriptional regulation of these paralogs during retinal development and aging. We employed deep sequencing, quantitative PCR, in situ hybridization (ISH), and microarray analysis, which revealed that replication-dependent histone genes were not only transcribed in progenitor cells but also in differentiating neurons. Specifically, by ISH analysis we found that different histone genes were actively transcribed in a subset of neurons between postnatal day 7 and 14. Interestingly, within a histone family, not all paralogs were transcribed at the same level during retinal development. For example, expression of Hist1h1b was higher embryonically, while that of Hist1h1c was higher postnatally. Finally, expression of replication-dependent histone genes was also observed in the aging retina. Moreover, transcription of replication-dependent histones was independent of rapamycin-mediated mTOR pathway inactivation. Overall, our data suggest the existence of variant nucleosomes produced by the differential expression of the replication-dependent histone genes across retinal development. Also, the expression of a subset of replication-dependent histone isotypes in senescent neurons warrants re-examining these genes as "replication-dependent." Thus, our findings underscore the importance of understanding the transcriptional regulation of replication-dependent histone genes in the maintenance and functioning of neurons.

Expression analysis of an evolutionarily conserved alternative splicing factor, Sfrs10, in age-related macular degeneration.

Age-related macular degeneration (AMD) is the most common cause of blindness in the elderly population. Hypoxic stress created in the micro-environment of the photoreceptors is thought to be the underlying cause that results in the pathophysiology of AMD. However, association of AMD with alternative splicing mediated gene regulation is not well explored. Alternative Splicing is one of the primary mechanisms in humans by which fewer protein coding genes are able to generate a vast proteome. Here, we investigated the expression of a known stress response gene and an alternative splicing factor called Serine-Arginine rich splicing factor 10 (Sfrs10). Sfrs10 is a member of the serine-arginine (SR) rich protein family and is 100% identical at the amino acid level in most mammals. Immunoblot analysis on retinal extracts from mouse, rat, and chicken showed a single immunoreactive band. Further, immunohistochemistry on adult mouse, rat and chicken retinae showed pan-retinal expression. However, SFRS10 was not detected in normal human retina but was observed as distinct nuclear speckles in AMD retinae. This is in agreement with previous reports that show Sfrs10 to be a stress response gene, which is upregulated under hypoxia. The difference in the expression of Sfrs10 between humans and lower mammals and the upregulation of SFRS10 in AMD is further reflected in the divergence of the promoter sequence between these species. Finally, SFRS10+ speckles were independent of the SC35+ SR protein speckles or the HSF1+ stress granules. In all, our data suggests that SFRS10 is upregulated and forms distinct stress-induced speckles and might be involved in AS of stress response genes in AMD.

The expression analysis of Sfrs10 and Celf4 during mouse retinal development.

Processing of mRNAs including, alternative splicing (AS), mRNA transport and translation regulation are crucial to eukaryotic gene expression. For example, >90% of the genes in the human genome are known to undergo alternative splicing thereby expanding the proteome production capacity of a limited number of genes. Similarly, mRNA export and translation regulation plays a vital role in regulating protein production. Thus, it is important to understand how these RNA binding proteins including alternative splicing factors (ASFs) and mRNA transport and translation factors regulate these processes. Here we report the expression of an ASF, serine-arginine rich splicing factor 10 (Sfrs10) and a mRNA translation regulation factor, CUGBP, elav like family member 4 (Celf4) in the developing mouse retina. Sfrs10 was expressed throughout postnatal (P) retinal development and was observed progressively in newly differentiating neurons. Immunofluorescence (IF) showed Sfrs10 in retinal ganglion cells (RGCs) at P0, followed by amacrine and bipolar cells, and at P8 it was enriched in red/green cone photoreceptor cells. By P22, Sfrs10 was observed in rod photoreceptors in a peri-nuclear pattern. Like Sfrs10, Celf4 expression was also observed in the developing retina, but with two distinct retinal isoforms. In situ hybridization (ISH) showed progressive expression of Celf4 in differentiating neurons, which was confirmed by IF that showed a dynamic shift in Celf4 localization. Early in development Celf4 expression was restricted to the nuclei of newly differentiating RGCs and later (E16 onwards) it was observed in the initial segments of RGC axons. Later, during postnatal development, Celf4 was observed in amacrine and bipolar cells, but here it was predominantly cytoplasmic and enriched in the two synaptic layers. Specifically, at P14, Celf4 was observed in the synaptic boutons of rod bipolar cells marked by Pkc-α. Thus, Celf4 might be regulating AS early in development besides its known role of regulating mRNA localization/translation. In all, our data suggests an important role for AS and mRNA localization/translation in retinal neuron differentiation.