The role of the transcription elongation factor Ell2 in gene expression control during lens development
Date
2025
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
The vertebrate eye is a multicomponent organ in which transparent tissues namely, the cornea and the lens, together refract and focus light on the retina to enable high-resolution vision. Formation of the mammalian eye is a complex process involving coordinate development of tissues. Perturbation of this process can cause ocular birth defects including microphthalmia (small eye), anophthalmia (absence of eye), or abnormalities in specific eye tissues, such as congenital cataract (defined as loss of lens transparency). To understand the pathology of ocular developmental defects, it is critical to identify key regulatory genes that play an important role in the developing eye, and to define their relationships with other factors in the form of networks. ☐ The lens comprises two principal cell types, the anteriorly localized epithelial cells, which differentiate posteriorly into fiber cells that make up the bulk of the tissue. Thus far, several genes involved in control of lens development have been identified and their deficiency or mutation has been associated with lens defects or congenital cataract. These include regulatory molecules such as members of signaling pathways (e.g., FGF, BMP etc.), transcription factors (TFs) (e.g., Pax6, Prox1, c-Maf, Sox1 etc.) and post-transcriptional regulatory RNA-binding proteins (e.g., Celf1, Tdrd7, Caprin2, Rbm24, etc.) – all distinct regulatory mechanisms that together determine the proteome–have been characterized. ☐ In the lens, the fiber gene expression is programmed toward building massive upregulation of select fiber cell expressed mRNAs (e.g., crystallins, membrane proteins, fibroblast growth factor receptors, etc.) and their subsequent translation into high amounts of proteins as they migrate towards the lens center and undergo terminal differentiation and lose organelles, which are essential for lens transparency. Therefore, fiber cells have the added challenge of achieving the proper transcript dosage prior to organelle/nuclear degradation. In human lenses, the concentration of crystallins can exceed 450 mg/ml. Presently, optimally high transcription of crystallins and other key lens genes is attributed to clusters of binding sites for specific groups of TFs (e.g., Sox1, c-Maf, Prox1, Pax6) in their promoter and/or enhancer region(s) that facilitates the formation of transcriptional complexes and thereby effectively initiate and regulate high levels of transcription. ☐ However, an unanswered fundamental question in lens research is whether our current understanding of lens transcription factors and the combinatorial control they exert on their target genes is sufficient to explain the extreme abundance of specific transcripts (e.g., crystallins) in fiber cells, or are there are other regulatory “general” transcription regulatory mechanisms that are necessary for achieving the specialized lens transcriptome? For example, are the regulators that control the ubiquitously important “transcription elongation” process specifically recruited for achieving proper dosage/levels of select highly expressed transcripts in development, particularly in the lens? My dissertation has addressed this key knowledge-gap and my new research findings hold a potential to have an broader impact, beyond lens development, on the mechanisms of gene expression control in other cells/tissue. ☐ I applied the bioinformatics tool termed iSyTE (integrated Systems Tool for Eye gene discovery) to identify the transcription elongation factor Ell2 (Elongation factor for RNA Polymerase II 2) (OMIM: 601874), based on its high absolute expression and high enriched expression in the mouse lens. Ell2 encodes a transcription elongation factor containing ELL domain and an occludin homology domain that allow interactions with RNA polymerase II (RNA Pol II) and other proteins of a large macromolecular protein complex called the super elongation complex (SEC). After transcription initiation, RNA Pol II transcribes between 30-80 nucleotides and then “pauses” because of the activity of two proteins, the negative elongation factor (NELF) and the DRB-sensitivity inducing transcriptional factor (DSIF). This “pause” must be relieved by SEC which contains positive transcription elongation factor b (P-TEFb) and an Ell family protein (e.g., at least one of Ell, Ell2, Ell3). Additionally, Ell2 prevents the backtracking of RNA pol II by keeping 3’OH of the nascent RNA aligned with the RNA Pol II catalytic site. Thus, Ell2 is important in control of transcription. ☐ To investigate the role of Ell2 in the lens, I first validated iSyTE’s prediction that Ell2 is expressed on both RNA and protein levels in mouse lens development by in situ hybridization (ISH) and immunofluorescence (IF), respectively. ISH shows that Ell2 mRNA is highly abundant in the lens transition zone (tz) and in fiber cells in early development. In agreement, IF assays show Ell2 protein is robustly present in tz and fiber cells at embryonic and early postnatal stages, and in later postnatal stages is also expressed in lens epithelial cells suggesting that Ell2 has temporal- and cell type-specific roles. Next, to gain insight into the impact of Ell2 deficiency in an autonomous manner in the lens, I generated lens-specific conditional Ell2 knockout mouse (termed Ell2cKO) using the Pax6GFPCre transgenic mouse line where Cre recombinase is expressed at embryonic day (E) 9.5 in lens placode formation. Lens-specific Ell2 KO in the Ell2cKO mice was confirmed on the gene, RNA and protein levels. ☐ I next performed phenotypic characterization of Ell2cKO mouse lenses. Light microscopy and grid imaging shows that Ell2cKO mice exhibit reduced sized lenses early in life starting from stage postnatal day (P) 15, and being fully penetrant with age after stage postnatal day (P) 30. Further, Ell2-deficient lenses exhibit abnormalities in lens shape and refractive properties and histological analysis by H&E shows lens tissue abnormalities. However, staining with a marker for proliferation, Ki-67, showed no appreciable differences between control and Ell2cKO lenses, suggesting that the observed lens defects were not due to changes in proliferation. ☐ To examine global gene expression changes associated with Ell2-deficiency-based lens defects, I performed bulk RNA-seq on Ell2cKO lenses at P30, when the lens defects are fully penetrant. RNA-seq analysis of Ell2cKO lenses at P30 identified 982 differentially expressed genes (DEGs). Of these DEGs, 457 were reduced, and 525 were elevated in Ell2cKO lenses. Cluster analysis of these DEGs using the Database for Annotation, Visualization, and Integrated Discovery (DAVID v6 .8) for functional annotation by gene ontology (GO) categories identifies GO categories that are relevant to vision and lens development. This suggests that deletion of Ell2 causes the misexpression of key genes that are involved in normal lens development. ☐ Interestingly, among the DEGs, several candidates are linked to lens development or cataract in animal models or in humans. Namely, genes exhibiting high expression in normal fiber cells, such as crystallins (e.g., Cryga, Crygf, Cryba1, Crybb1), Aqp0 (Mip), etc., as well as those exhibiting high expression in epithelial cells, (e.g., Aqp1), etc., are reduced in Ell2cKO lenses. These data indicate that Ell2 is necessary for optimal transcript levels of key genes in fiber cells and/or epithelial cells – a role perhaps made necessary as, particularly in the fiber cells, specific transcripts not only need to be made abundantly, but they need to achieve these high levels prior to nuclear and organelle degradation in fiber cell maturation, failure of which causes lens defects. However, some genes highly expressed in fiber cells (e.g., Ezr, Gja3, Gja8, Hspb1, Tdrd7) remain unaffected, suggesting that Ell2 functions in the specific upregulation of a subset of genes in the lens. Also, importantly, the key transcription factors (TFs) Maf (c-Maf), Prox1, and Sox1, which have been previously shown to be involved in fiber gene expression (e.g., crystallins) are unaltered in Ell2cKO lenses suggesting that the misexpression of fiber cell genes is not simply due to changes in the expression of these lens TFs, but likely a direct result of loss of Ell2. RT-qPCR showed a significant reduction of Crybb1, Cryge, Mip, Lgsn, Birc7, and Hmox1 transcripts in Ell2cKO lenses, thus offering independent validation of DEGs identified by RNA-seq in Ell2cKO lenses. ☐ In addition to bulk RNA-seq, I performed single nucleus (sn) multiome analysis (snRNA-seq and snATAC-seq (assay for transposase-accessible chromatin using sequencing)) on P10 and P15 Ell2cKO lenses to gain insight into spatiotemporal gene expression in lens development on the single cell-level (represented by single nucleus), and to examine how it is impacted by loss of Ell2. snRNA-seq identified alterations in gene expression in the epithelial, intermediate and fiber populations in Ell2cKO lenses. The snRNA-seq analysis reveals a significant reduction in the expression of several cohorts of crystallin genes, including Cryaa, Cryba1, Cryba2, Crybb2, Cryga, Crygf, Cryge, Crygd, Crygc, Crygb, Crygs and many other lens expressed genes in Ell2cKO lens cell populations. These data are consistent with bulk RNA-seq data, thus offering independent validation. ☐ Further, to examine if Ell2 loss results in alteration of the lens chromatin and to understand how this correlates with transcriptional changes, I performed snATAC-seq. Overall, the comparison of snRNA-seq and snATAC-seq data showed correlation between chromatin accessibility changes and RNA levels changes in Ell2cKO lens. In detailed analysis focusing on specific genes, snATAC-seq identified specific alterations in chromatin accessibility in distinct cell populations in Ell2cKO lenses. For example, in Ell2cKO lenses, snATAC-seq showed reduced promoter accessibility across loci for Cryge, specifically in the lens cell population termed as “intermediate cluster”, which can be considered as that representative of cells that are in the transition zone or are in early differentiating stages. Interestingly, in cell clusters that represent later stages of differentiating fiber cells, the promoter accessibility for Cryge was similar between control and Ell2cKO lenses. However, on the RNA level, in both cell populations Cryge mRNAs were found to be reduced in Ell2cKO lens. These data suggest that Ell2 is critically necessary – in transition zone cells or in early differentiating fiber cells – for achieving optimal chromatin accessibility and RNA abundance at specific genic loci. Furthermore, these data show that in the absence of Ell2, even if the chromatin accessibility is re-gained, as in later stages of fiber cells, the RNA abundance levels cannot be recovered to normal and remain overall reduced. Together, these data, based on single nucleus multiomics analysis, uncover the requirement of Ell2 in a critical early phase of fiber differentiation, perhaps correlating with a critical spatiotemporal window when there is optimal stochiometric levels of the transcription factors (e.g., Pax6, c-Maf, Prox1, Sox1) for achieving optimal transcriptome. ☐ To determine Ell2’s preferential localization to lens genomic loci, I performed chromatin immunoprecipitation (ChIP) in wild type (WT) lenses using Ell2 antibody followed by genomic DNA PCR which demonstrates that Ell2 protein is enriched in Cryge and Crygf gene loci within 100-150 bp from the transcriptional start (TSS). Together, these data (snRNA-seq, snATAC-seq, ChIP-PCR) help define Ell2’s selective role in regulation of genes important in lens development and cataract. ☐ To gain insights into the regulation of Ell2 itself, I analyzed RNA-seq data generated on lenses deficient for the RNA-binding protein (RBP) encoding genes Celf1 or Elavl1, which are linked to lens defects and cataract. This analysis shows that Ell2 mRNA is mis-expressed in Celf1cKO and Elavl1cKO lenses, suggesting that post-transcriptional regulatory mechanisms have evolved to ensure that Ell2 mRNA levels are optimally controlled in the lens. These data also suggest a potential cross-talk between transcriptional and post-transcriptional networks in the lens. ☐ In sum, I utilized a rigorous multi-disciplinary approach using the tools of genetics, cell biology and state-of-the-art omics to characterize a novel function of the conserved regulatory protein, Ell2, in the lens. My research has made far-reaching impact by showning that proteins that have been historically considered as “general” regulators of transcription can specifically contribute to the regulation of a subset of key genes in tissue development, perturbation of which can result in developmental defects.
Description
Keywords
Biological science, Mammalian eye, Microphthalmia, Transparent tissues