CoChat

Cis-regulatory landscapes of four cell types of the retina

Open access

The DOI matches its Crossref record; 75 of the 75 references it cites were found in a registry or an index.

Abstract

The retina is composed of ∼50 cell-types with specific functions for the process of vision. Identification of the cis-regulatory elements active in retinal cell-types is key to elucidate the networks controlling this diversity. Here, we combined transcriptome and epigenome profiling to map the regulatory landscape of four cell-types isolated from mouse retinas including rod and cone photoreceptors as well as rare inter-neuron populations such as horizontal and starburst amacrine cells. Integration of this information reveals sequence determinants and candidate transcription factors for controlling cellular specialization. Additionally, we refined parallel reporter assays to enable studying the transcriptional activity of large collection of sequences in individual cell-types isolated from a tissue. We provide proof of concept for this approach and its scalability by characterizing the transcriptional capacity of several hundred putative regulatory sequences within individual retinal cell-types. This generates a catalogue of cis-regulatory regions active in retinal cell types and we further demonstrate their utility as potential resource for cellular tagging and manipulation.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1093/nar/gkx923

Cites

75 of 75 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1. Transcriptome profiling of four cell types isolated from mouse retinas. (A) Scheme used to generate transcriptome and methylation datasets at the resolution of single cell types. Cell populations of rods, cones, horizontal cells and starburst amacrine cells are isolated by cell sorting from mouse transgenic lines carrying fluorescent markers that label these particular cell types of the retina. From cell isolates, whole genome methylation maps and expression datasets are generated. (B) Known cell specific expression markers reproducibly discriminate between cell isolates illustrating the reproducibility of the FACS procedure. Expression levels for markers of the studied retinal cell types. RPKM values for genic RNA-seq signal for samples issued from independent cell sorts. Side bar depicts the cell type associated with the marker (Siegert et al, 2009) (rods: dark green; cones: light green; HCs: orange; SACs: purple). Levels Rhodopsin in non-rod samples shows a particularly high degree of systematic contamination of cone samples with rods, as previously observed (Siegert et al, 2012; Mo et al, 2016). (C) The tested cell types show divergence in their expression profiles. Correlation heatmap comparing transcriptomes of the four studied cell types. Pearson correlation for genic RNA-seq signal merged across three biological replicates for each cell type. (D) Transcription factors showing differential expression between the tested cell types. Heatmap depicting the expression level of the most differentially expressed transcription factors between the tested cell types (top 10% variance). Shown are the RPKM values for genic RNA-seq merged across three biological replicates for each cell type. The heatmap was organized by hierarchical clustering.
Figure 1. Transcriptome profiling of four cell types isolated from mouse retinas. (A) Scheme used to generate transcriptome and methylation datasets at the resolution of single cell types. Cell populations of rods, cones, horizontal cells and starburst amacrine cells are isolated by cell sorting from mouse transgenic lines carrying fluorescent markers that label these particular cell types of the retina. From cell isolates, whole genome methylation maps and expression datasets are generated. (B) Known cell specific expression markers reproducibly discriminate between cell isolates illustrating the reproducibility of the FACS procedure. Expression levels for markers of the studied retinal cell types. RPKM values for genic RNA-seq signal for samples issued from independent cell sorts. Side bar depicts the cell type associated with the marker (Siegert et al, 2009) (rods: dark green; cones: light green; HCs: orange; SACs: purple). Levels Rhodopsin in non-rod samples shows a particularly high degree of systematic contamination of cone samples with rods, as previously observed (Siegert et al, 2012; Mo et al, 2016). (C) The tested cell types show divergence in their expression profiles. Correlation heatmap comparing transcriptomes of the four studied cell types. Pearson correlation for genic RNA-seq signal merged across three biological replicates for each cell type. (D) Transcription factors showing differential expression between the tested cell types. Heatmap depicting the expression level of the most differentially expressed transcription factors between the tested cell types (top 10% variance). Shown are the RPKM values for genic RNA-seq merged across three biological replicates for each cell type. The heatmap was organized by hierarchical clustering.CC BY-NC · PubMed Central
Figure 2. DNA methylation-based identification of putative CRE of the retinal system. (A) Correlation heatmap comparing methylation within low methylated regions detected in the retinal cell types. Pearson correlation for methylation of single CpGs located within the merged list of LMRs across cell types. Heatmap was subjected to hierarchical clustering. (B) Heatmap displaying average methylation of all putative CREs identified across the four studied retinal cell types. LMRs were grouped based on their binarized average methylation pattern (methylation < 60%). Sidebar indicates cell type specific clusters. Clusters were colored according to the specificity of the putative CRE (rods: dark green; cones: light green; HCs: orange; SACs: purple). (C) TF motifs enriched within subsets of putative CREs defined by their low methylated in individual cell types. Enrichment for known motifs was calculated for each set of cell-type specific LMRs and compared to a set of control sequences. Enrichments are displayed only for motifs significantly enriched in at least one cell-type specific subset, and for predicted transcription factors expressed in at least one of the cell types. Side-bar depicts the TF family associated with the enriched motif. (D, E) DNA methylation pattern of the four analyzed cell types of the retina at exemplified genomic regions. (D) Shown is the average WGBS signal around the Snx5 gene containing several SAC-specific low methylated regions (purple box) and the Faim gene containing a photoreceptor specific low methylated region (purple box). Black boxes denote regions displaying low methylation in at least one of the analyzed cell types, indicative of putative retinal CREs.
Figure 2. DNA methylation-based identification of putative CRE of the retinal system. (A) Correlation heatmap comparing methylation within low methylated regions detected in the retinal cell types. Pearson correlation for methylation of single CpGs located within the merged list of LMRs across cell types. Heatmap was subjected to hierarchical clustering. (B) Heatmap displaying average methylation of all putative CREs identified across the four studied retinal cell types. LMRs were grouped based on their binarized average methylation pattern (methylation < 60%). Sidebar indicates cell type specific clusters. Clusters were colored according to the specificity of the putative CRE (rods: dark green; cones: light green; HCs: orange; SACs: purple). (C) TF motifs enriched within subsets of putative CREs defined by their low methylated in individual cell types. Enrichment for known motifs was calculated for each set of cell-type specific LMRs and compared to a set of control sequences. Enrichments are displayed only for motifs significantly enriched in at least one cell-type specific subset, and for predicted transcription factors expressed in at least one of the cell types. Side-bar depicts the TF family associated with the enriched motif. (D, E) DNA methylation pattern of the four analyzed cell types of the retina at exemplified genomic regions. (D) Shown is the average WGBS signal around the Snx5 gene containing several SAC-specific low methylated regions (purple box) and the Faim gene containing a photoreceptor specific low methylated region (purple box). Black boxes denote regions displaying low methylation in at least one of the analyzed cell types, indicative of putative retinal CREs.CC BY-NC · PubMed Central
Figure 3. Parallelized reporter assay in specific retinal cell types. (A) Schematic representation of the procedure used to perform parallel reporter assays at the resolution of single cell types. Putative CREs are selected based on the detection of cell type-specific low methylation. Subsets of CREs are batch-cloned in front of a minimal promoter driving transcription of a GFP cassette followed by a unique barcode. These libraries are packaged into adeno-associated viruses and injected as pools in retinas of transgenic mice labeled for the cell types of interest. Three weeks following injection, cell populations are FACS sorted and RNA is extracted. CRE activity is determined as function of the barcode counts in the RNA normalized to the barcode counts in the AAV gDNA. (B) Histogram representing the distribution of activities observed for putative rod CREs (lowly methylated - green) and control regions (highly methylated - grey). Fragment activity was determined as ratio of barcode abundance in RNA sample versus abundance in the AAV pool used for infection. Displayed are average activity values derived from at least 3 biological replicates. CREs that were tested with an individual GFP reporter system are marked in the plot at their respective activity group. (C–E) Comparison of activity levels measured by PRA with trans-membrane GFP reporter signal for individual CREs. Immunohistochemical staining of whole mount and vibratome sections from wild type mouse retinas injected with individual constructs showing no (C), intermediate (D) or high (E) activity in the PRA assay. Green: GFP, white: Hoechst staining of DNA.
Figure 3. Parallelized reporter assay in specific retinal cell types. (A) Schematic representation of the procedure used to perform parallel reporter assays at the resolution of single cell types. Putative CREs are selected based on the detection of cell type-specific low methylation. Subsets of CREs are batch-cloned in front of a minimal promoter driving transcription of a GFP cassette followed by a unique barcode. These libraries are packaged into adeno-associated viruses and injected as pools in retinas of transgenic mice labeled for the cell types of interest. Three weeks following injection, cell populations are FACS sorted and RNA is extracted. CRE activity is determined as function of the barcode counts in the RNA normalized to the barcode counts in the AAV gDNA. (B) Histogram representing the distribution of activities observed for putative rod CREs (lowly methylated - green) and control regions (highly methylated - grey). Fragment activity was determined as ratio of barcode abundance in RNA sample versus abundance in the AAV pool used for infection. Displayed are average activity values derived from at least 3 biological replicates. CREs that were tested with an individual GFP reporter system are marked in the plot at their respective activity group. (C–E) Comparison of activity levels measured by PRA with trans-membrane GFP reporter signal for individual CREs. Immunohistochemical staining of whole mount and vibratome sections from wild type mouse retinas injected with individual constructs showing no (C), intermediate (D) or high (E) activity in the PRA assay. Green: GFP, white: Hoechst staining of DNA.CC BY-NC · PubMed Central
Figure 4. Characterization of the autonomous activity of CREs in distinct cell types of the retina. (A, B) Most tested sequences show similar activity in rods and cones. Scatter plot contrasting activity of CREs as measured by PRA in (A) rods or (B) cones against the remaining pool of unlabeled cells from the retina. For each experiment fluorescently labeled cells and remaining unlabeled cells were sorted from retinas infected by the AAV PRA library. Displayed is the normalized activity for each fragment averaged over biological replicates. (C) Scatter plot contrasting activity of CREs as measured by PRA in rods versus cones. (D, E) Most of the tested sequences show similar activity in HCs and SACs that largely differs from activity observed in whole retina. Similar scatterplot as in (A-B), contrasting PRA activity in (D) HCs or (E) SACs against the remaining pool of unlabeled cells from the retina. (F) Scatter plot contrasting activity of CREs as measured by PRA in HCs versus SACs. (G-H) Microscopical validation of the specificity of individual photoreceptor specific CREs in the tissue context. Shown is Chr2-GFP fluorescence for individual reporter constructs with indicated CRE. Immunohistochemical staining of vibratome sections from transgenic mouse (cone-labeled, mCAR) retinas injected with two CREs detected by PRA to be active in photoreceptor but not in interneurons. White, Hoechst; green, GFP; purple, cone marker mCAR. Scale bar, 10 μm. Side bar depicts the expected localization of the different studied cell types based on the considered retina layers. (I and J) Similar validation as in (G-H) for interneuron specific CREs. Shown is Chr2-GFP fluorescence for individual reporter constructs with indicated CRE. Immunohistochemical staining of retina mounts from transgenic mouse (SAC-labeled, ChAT) retinas injected with two CREs detected by PRA to be active in interneurons but not in photoreceptors. White, Hoechst; green, GFP; purple, cone marker ChAT. Scale bar, 10 μm. Side bar depicts the expected localization of the different studied cell types based on the considered retina layers.
Figure 4. Characterization of the autonomous activity of CREs in distinct cell types of the retina. (A, B) Most tested sequences show similar activity in rods and cones. Scatter plot contrasting activity of CREs as measured by PRA in (A) rods or (B) cones against the remaining pool of unlabeled cells from the retina. For each experiment fluorescently labeled cells and remaining unlabeled cells were sorted from retinas infected by the AAV PRA library. Displayed is the normalized activity for each fragment averaged over biological replicates. (C) Scatter plot contrasting activity of CREs as measured by PRA in rods versus cones. (D, E) Most of the tested sequences show similar activity in HCs and SACs that largely differs from activity observed in whole retina. Similar scatterplot as in (A-B), contrasting PRA activity in (D) HCs or (E) SACs against the remaining pool of unlabeled cells from the retina. (F) Scatter plot contrasting activity of CREs as measured by PRA in HCs versus SACs. (G-H) Microscopical validation of the specificity of individual photoreceptor specific CREs in the tissue context. Shown is Chr2-GFP fluorescence for individual reporter constructs with indicated CRE. Immunohistochemical staining of vibratome sections from transgenic mouse (cone-labeled, mCAR) retinas injected with two CREs detected by PRA to be active in photoreceptor but not in interneurons. White, Hoechst; green, GFP; purple, cone marker mCAR. Scale bar, 10 μm. Side bar depicts the expected localization of the different studied cell types based on the considered retina layers. (I and J) Similar validation as in (G-H) for interneuron specific CREs. Shown is Chr2-GFP fluorescence for individual reporter constructs with indicated CRE. Immunohistochemical staining of retina mounts from transgenic mouse (SAC-labeled, ChAT) retinas injected with two CREs detected by PRA to be active in interneurons but not in photoreceptors. White, Hoechst; green, GFP; purple, cone marker ChAT. Scale bar, 10 μm. Side bar depicts the expected localization of the different studied cell types based on the considered retina layers.CC BY-NC · PubMed Central
Figure 5. Dissection of the architecture of two photoreceptor CREs. (A, B) DNA methylation pattern of the four analyzed cell types of the retina around the two CREs selected for functional dissections (purple box); (A) Faim-Intra and (B) Intragenic 3 enhancers. Black boxes denote regions displaying low methylation in at least one of the analyzed cell types, representing putative retinal CREs. The TF architecture of the CREs is detailed, showing the TF binding sites identified (colored according to the TF family as in Figure 2C) and the PhyloP conservation track for the region. (C, D) Effects on activity of systematic deletion of TF motifs for two CREs. Each motif identified in the CRE sequence was iteratively replaced by a randomized sequence (shown in upper panel). Activity of the generated library of mutants was compared to the wild type sequence. Boxplots represents the distribution of individual measurements from different biological replicates. Statistical significance of changes in activity were tested using a bidirectional t-test (*P < 0.05; **P < 0.01).
Figure 5. Dissection of the architecture of two photoreceptor CREs. (A, B) DNA methylation pattern of the four analyzed cell types of the retina around the two CREs selected for functional dissections (purple box); (A) Faim-Intra and (B) Intragenic 3 enhancers. Black boxes denote regions displaying low methylation in at least one of the analyzed cell types, representing putative retinal CREs. The TF architecture of the CREs is detailed, showing the TF binding sites identified (colored according to the TF family as in Figure 2C) and the PhyloP conservation track for the region. (C, D) Effects on activity of systematic deletion of TF motifs for two CREs. Each motif identified in the CRE sequence was iteratively replaced by a randomized sequence (shown in upper panel). Activity of the generated library of mutants was compared to the wild type sequence. Boxplots represents the distribution of individual measurements from different biological replicates. Statistical significance of changes in activity were tested using a bidirectional t-test (*P < 0.05; **P < 0.01).CC BY-NC · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 75 references

Research works printed but not linked: 3

  1. 2012, The accessible chromatin landscape of the human genome, Nature, 489, 75, 10.1038/nature11232 · DOI · Paper page
  2. 2013, Systematic dissection of motif instances using a massively parallel reporter assay, Genome Res.
  3. 2015, Massively parallel cis-regulatory analysis in the mammalian central nervous system, Genome Res.

Cited by

See all 37

Metadata

Authors
Hartl, D., Krebs, A. R., Jüttner, J., Roska, B., & Schübeler, D.
Affiliations
  • University of Basel (CH)
  • Friedrich Miescher Institute (CH)
Corresponding author
  • Krebs, Arnaud R.
  • Roska, Botond
  • Schübeler, Dirk
Venue
Nucleic Acids Research
Published
2017-10-20
Volume
45
Issue
20
Pages
11607-11621
ISSN
  • 0305-1048 (ISSN-L)
  • 1362-4962
OpenAlex
W2765448062
Cited by
37
References
75
Citations by year
2026: 2 · 2025: 1 · 2024: 3 · 2023: 5 · 2022: 5 · 2021: 3 · 2020: 7 · 2019: 10 · 2018: 1
FWCI
2.01
Type
journal article
Language
en
Open access
Locations
  • Nucleic Acids Research (publishedVersion, cc-by-nc)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by-nc)
  • PubMed Central (publishedVersion, cc-by-nc)
  • PubMed (publishedVersion)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Genomics and Chromatin Dynamics
  • Health Sciences › Medicine › Ophthalmology › Retinal Development and Disorders
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Single-cell and spatial transcriptomics
Keywords
cis-regulatory elements, retinal cell types, rod photoreceptors, cone photoreceptors, starburst amacrine cells, transcription factor binding sites, cell type-specific gene regulation, mouse retina
MeSH terms
Animals, Female, Mice, Inbred C57BL, Mice, Transgenic, Regulatory Sequences, Nucleic Acid, Retina, Retinal Rod Photoreceptor Cells, Retinal Cone Photoreceptor Cells, DNA Methylation, Regulatory Elements, Transcriptional, Mice, Transcriptome

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Cis-regulatory landscapes of four cell types of the retina. https://app.cochat.ai/papers/10.1093/nar/gkx923
The paper
Hartl, D., Krebs, A. R., Jüttner, J., Roska, B., & Schübeler, D. (2017). Cis-regulatory landscapes of four cell types of the retina. Nucleic Acids Research. https://doi.org/10.1093/nar/gkx923

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.