The long noncoding RNA RNCR2 directs mouse retinal cell specification

Abstract

Background: Recent work has identified that many long mRNA-like noncoding RNAs (lncRNAs) are expressed in the developing nervous system. Despite their abundance, the function of these ncRNAs has remained largely unexplored. We have investigated the highly abundant lncRNA RNCR2 in regulation of mouse retinal cell differentiation.

Results: We find that the RNCR2 is selectively expressed in a subset of both mitotic progenitors and postmitotic retinal precursor cells. ShRNA-mediated knockdown of RNCR2 results in an increase of both amacrine cells and Müller glia, indicating a role for this lncRNA in regulating retinal cell fate specification. We further report that RNCR2 RNA, which is normally nuclear-retained, can be exported from the nucleus when fused to an IRES-GFP sequence. Overexpression of RNCR2-IRES-GFP phenocopies the effects of shRNA-mediated knockdown of RNCR2, implying that forced mislocalization of RNCR2 induces a dominant-negative phenotype. Finally, we use the IRES-GFP fusion approach to identify specific domains of RNCR2 that are required for repressing both amacrine and Müller glial differentiation.

Conclusion: These data demonstrate that the lncRNA RNCR2 plays a critical role in regulating mammalian retinal cell fate specification. Furthermore, we present a novel approach for generating dominant-negative constructs of lncRNAs, which may be generally useful in the functional analysis of this class of molecules.

Figure 1

Genomic structure and evolutionary conservation of RNCR2. Schematic drawing showing the conservation and genomic position of RNCR2 based on the human March 2006, mouse July 2007, chicken May 2006 and Xenopus April 2005 assemblies of the UCSC Genome browser. Clusters of the conserved seven base repeat ACTAACC are shown below each RNCR2 transcript, with the sequence of one repeat cluster for each species is shown.

Figure 2

Expression of RNCR2 in the developing retina. Section fluorescent in situ hybridization (fISH) showing RNCR2 expression. RNCR2 is widely expressed in the retina at E14.5 (A), and in the INL and GCL at P8 (C). High-power magnification of E14.5 (B) and P8 (D) reveals that RNCR2 is localized to the nucleus but not associated with DNA. (E-G) RNCR2 expression examined by double dissociated-cell in situ hybridization at E16.5 (E), P0.5 (F), and P7 (G). Cellular expression pattern of each control probe is color coded by cell type. Some probes recognize more than one cell type. This is indicated by overlapping bars. For example Islet1 mRNA at E16.5 is present in developing amacrine and ganglion cells [34].

Figure 3

Overexpression of RNCR2 in the developing retina. CAG-GFP or CAG-RNCR2 construct was electroporated into P0.5 retina and harvested at P21, dissociated and stained for cell-specific markers. GFP-positive cells were counted to analyze the fraction of electroporated cells that expressed the markers in question. Error bars represent +/- SEM for at least three independent retinas. (A) CAG-RNCR2 overexpression has no effect on cell fate determination. The number of GFP-positive electroporated cells in ONL and INL were counted and compared (inset), with no difference was noted. (B-C) Section immunohistochemistry of retinas electroporated with either CAG-GFP or CAG-RNCR2 and harvested at P21 confirm no differences in cell fate or morphology of CAG-RNCR2 electroporated cells. Cell type specific markers used: rhodopsin (Rho4D2), rod photoreceptors, glutamine synthetase (GS), Müller glia; protein kinase C alpha (PKCα), rod bipolar cells; syntaxin (Syn), amacrine cells.

Figure 4

ShRNA-mediated knockdown of RNCR2 in the developing retina. Control shRNA or RNCR2-targeted shRNA construct was electroporated into P0.5 retina and harvested at P7 or P21, dissociated and stained for cell-specific markers. (A) RNCR2 knockdown led to a decrease in Müller glia at P7 and (B) an increase in Müller glia and amacrine cells at P21, *p < 0.05. A significant increase in GFP-positive electroporated cells in the INL was noted (B, inset). (C-D) Section immunohistochemistry of retinas electroporated with either control shRNA or RNCR2-targeted shRNA confirm an increase in Müller glia (white arrowheads) and amacrine cells (yellow arrowheads).

Figure 5

IRES-GFP fusion constructs mislocalize RNCR2 to the cytoplasm. (A, B) Model for a mechanism by which overexpression of IRES-GFP fusion-constructs can result in dominant-negative phenotypes in transfected cells. Fusion to the IRES-GFP sequence mislocalizes lncRNAs and associated effector proteins to the ribosome, and mislocalizes proteins that bind RNCR2 and mediate the biochemical effectors of lncRNA function away from endogenous nuclear-retained lncRNAs, effectively inhibiting their function. (C) RNCR2-IRES-GFP RNA is mislocalized from the nucleus to the cytoplasm. HeLa cells were transfected with RNCR2 or RNCR2-IRES-GFP fusion constructs and RNA location analyzed by fISH followed by immunohistochemistry against the cytoplasmic S6 ribosomal protein. White arrows indicate regions of prominent RNCR2 localization. The cell nucleus, defined by the region encompassed by DAPI-stained chromatin, is delineated by the white dashed line. The relative fraction of nuclear RNCR2 signal found within in cells transfected with CAG-RNCR2 and CAG-RNCR2-IRES-GFP is shown in the bar graph. *p < 1.5^-07.

Figure 6

Overexpression of RNCR2-IRES-GFP has dominant-negative effects in vivo. CAG-driven constructs encoding IRES-GFP or RNCR2-IRES-GFP were electroporated into P0 retina and harvested at either P7 or P21 and analyzed. (A-B) RNCR2-IRES-GFP expression results in a decrease in rhodopsin expression and an increase in the fraction of amacrine cells at P7 and P21, and an increase in the fraction of GFP-positive cells not stained by any marker tested at P21. *p < 0.02. (C-D) Analysis of electroporated retinas at P21 confirms the increase in amacrine cells (yellow arrowheads). RNCR2-IRES-GFP led to an increase in INL cells compared to IRES-GFP (B, inset).

Figure 7

Domain-specific RNCR2-IRES-GFP fusion constructs selectively inhibit either amacrine or Müller glia differentiation. Dominant-negative RNCR2 constructs were electroporated in vivo at P0.5 and analyzed at P21. (A-B) Section immunohistochemistry reveals that RNCR2-5'-IRES-GFP overexpression led to an increase in amacrine cells (yellow arrowheads), (C-E) RNCR2-middle-IRES-GFP overexpression led to an increase in Müller glia (white arrowheads) and amacrine cells and RNCR2-3'-IRES-GFP overexpression led to an increase in Müller glia. (F) Each of the RNCR2-IRES-GFP domain-specific constructs showed an increase in INL cells compared to control **p < 0.003.