CoChat

Genome-Wide Organization of GATA1 and TAL1 Determined at High Resolution

Open access

The DOI matches its Crossref record; 51 of the 51 references it cites were found in a registry or an index.

Abstract

Erythroid development and differentiation from multiprogenitor cells into red blood cells requires precise transcriptional regulation. Key erythroid transcription factors, GATA1 and TAL1, cooperate, along with other proteins, to regulate many aspects of this process. How GATA1 and TAL1 are juxtaposed along the DNA and their cognate DNA binding site across the mouse genome remains unclear. We applied high-resolution ChIP-exo (chromatin immunoprecipitation followed by 5'-to-3' exonuclease treatment and then massively parallel DNA sequencing) to GATA1 and TAL1 to study their positional organization across the mouse genome during GATA1-dependent maturation. Two complementary methods, MultiGPS and peak pairing, were used to determine high-confidence binding locations by ChIP-exo. We identified ∼10,000 GATA1 and ∼15,000 TAL1 locations, which were essentially confirmed by ChIP-seq (chromatin immunoprecipitation followed by massively parallel DNA sequencing). Of these, ∼4,000 locations were bound by both GATA1 and TAL1. About three-quarters of them were tightly linked to a partial E-box located 7 or 8 bp upstream of a WGATAA motif. Both TAL1 and GATA1 generated distinct characteristic ChIP-exo peaks around WGATAA motifs that reflect their positional arrangement within a complex. We show that TAL1 and GATA1 form a precisely organized complex at a compound motif consisting of a TG 7 or 8 bp upstream of a WGATAA motif across thousands of genomic locations.

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 · 10.1128/mcb.00806-15

Cites

51 of 51 references verified0 citation errors · 0 not checkable

Statistics

Figures

FIG 1 Determination of binding locations using intersection of MultiGPS and union of peak pairs. Shown is a browser shot of ChIP-exo tag 5′ ends for GATA1 measured by ChIP-seq (CS) and ChIP-exo (XO) on chromosome 1 from coordinates 36,976,370 to 36,977,750. Tag locations were smoothed (20-bp moving average). The tag density on the lower (negative) strand is shown as an inverted plot.The locations of WGATAA (P < 10−3) and RefSeq genes are shown below through the Integrative Genomics Viewer (IGV) browser.
FIG 1 Determination of binding locations using intersection of MultiGPS and union of peak pairs. Shown is a browser shot of ChIP-exo tag 5′ ends for GATA1 measured by ChIP-seq (CS) and ChIP-exo (XO) on chromosome 1 from coordinates 36,976,370 to 36,977,750. Tag locations were smoothed (20-bp moving average). The tag density on the lower (negative) strand is shown as an inverted plot.The locations of WGATAA (P < 10−3) and RefSeq genes are shown below through the Integrative Genomics Viewer (IGV) browser.CC BY · PubMed Central
FIG 2 Peak-pairing and MultiGPS comparison of ChIP-exo location calling. (A) Workflow of ChIP-exo data analysis. Binding locations with high confidence were determined by obtaining the intersection of the union of significant peak pairs from multiple time points (left) and binding events determined by MultiGPS (right). Motif discovery analysis using MEME showed enrichment of WGATAA, E-box, and composite half E-box/WGATAA motifs on these stringent binding locations and validated both methods. To classify binding locations by the presence of motifs, FIMO and downstream analyses were conducted. (B) Venn diagrams showing overlap of GATA1 binding locations determined by the union of peak pairs and MultiGPS. MultiGPS binding locations positioned within a 100-bp window from the union of peak pairs were determined as final binding locations and percentages (n = 10,290). The most enriched MEME motifs and their occurrence frequencies from intersection and outersects of the two methods are shown below. (C) Binding locations detected by only one method, either peak pairing or MultiGPS. Binding locations called by the union of peak pairing and MultiGPS are shown in the Venn diagram. A browser shot of the 5′ end of TAL1 ChIP-exo tags surrounding the binding locations detected only by peak pairing is shown in the left browser diagram and that for MultiGPS in the right browser diagram. Binding locations at the intersection (blue) or outersection (red) of two methods are highlighted. WGATAA and E-box motifs are shown in black.
FIG 2 Peak-pairing and MultiGPS comparison of ChIP-exo location calling. (A) Workflow of ChIP-exo data analysis. Binding locations with high confidence were determined by obtaining the intersection of the union of significant peak pairs from multiple time points (left) and binding events determined by MultiGPS (right). Motif discovery analysis using MEME showed enrichment of WGATAA, E-box, and composite half E-box/WGATAA motifs on these stringent binding locations and validated both methods. To classify binding locations by the presence of motifs, FIMO and downstream analyses were conducted. (B) Venn diagrams showing overlap of GATA1 binding locations determined by the union of peak pairs and MultiGPS. MultiGPS binding locations positioned within a 100-bp window from the union of peak pairs were determined as final binding locations and percentages (n = 10,290). The most enriched MEME motifs and their occurrence frequencies from intersection and outersects of the two methods are shown below. (C) Binding locations detected by only one method, either peak pairing or MultiGPS. Binding locations called by the union of peak pairing and MultiGPS are shown in the Venn diagram. A browser shot of the 5′ end of TAL1 ChIP-exo tags surrounding the binding locations detected only by peak pairing is shown in the left browser diagram and that for MultiGPS in the right browser diagram. Binding locations at the intersection (blue) or outersection (red) of two methods are highlighted. WGATAA and E-box motifs are shown in black.CC BY · PubMed Central
FIG 3 GATA1 and TAL1 locations reside in clusters. (A) Percentages of GATA1 and TAL1 binding locations in homotypic clusters among the total binding locations. A cluster is defined as adjacent binding locations that are <500 bp apart. In the case of GATA1, 13.7% (1,414) of the binding locations form 663 clusters, while 64.4% (9,796) of the TAL1 binding locations form 2,905 clusters. (B) Expression change of the closest gene to a homotypic cluster or noncluster for GATA1 and TAL1. Expression change was calculated as the log2-fold change of gene expression between G1E-ER4 cell induction for 24 h relative to that for 0 h. Significant differences (P < 0.05) using the Mann-Whitney test are indicated by asterisks. Boxes denote the first and third quartiles. Horizontal lines, circles, and vertical "whiskers" denote median, mean, and the limits of the data, respectively.
FIG 3 GATA1 and TAL1 locations reside in clusters. (A) Percentages of GATA1 and TAL1 binding locations in homotypic clusters among the total binding locations. A cluster is defined as adjacent binding locations that are <500 bp apart. In the case of GATA1, 13.7% (1,414) of the binding locations form 663 clusters, while 64.4% (9,796) of the TAL1 binding locations form 2,905 clusters. (B) Expression change of the closest gene to a homotypic cluster or noncluster for GATA1 and TAL1. Expression change was calculated as the log2-fold change of gene expression between G1E-ER4 cell induction for 24 h relative to that for 0 h. Significant differences (P < 0.05) using the Mann-Whitney test are indicated by asterisks. Boxes denote the first and third quartiles. Horizontal lines, circles, and vertical "whiskers" denote median, mean, and the limits of the data, respectively.CC BY · PubMed Central
FIG 4 Genome-wide structural organization of GATA1/TAL1 complexes. (A) Venn diagram showing the overlap of TAL1 and/or GATA1 binding locations further classified as TAL1-only, GATA1/TAL1-cooccupied, and GATA1-only binding locations. Cooccupancy was defined as having midpoints <40 bp apart. (B) Composite distribution of TAL1 (black trace) and GATA1 (red-shaded trace) ChIP-exo tag 5′ ends around the WGATAA motif (the underlined A is set to zero) at GATA1 binding locations. Tags that mapped to the reverse strand that contained the WGATAA motif are displayed as inverted traces. The data were plotted as a moving average of 5 bp. Above the plot are X-ray structure images of TAL1 and GATA1 bound to cognate DNA sites (Protein Data Bank [PDB] accession codes 3VEK for GATA1 [J. M. Matthews, unpublished data] and 2YPB for TAL1 [SCL:E47] [17]). The circled X indicates the deduced site of cross-linking of GATA1 (red) and TAL1 (gold).
FIG 4 Genome-wide structural organization of GATA1/TAL1 complexes. (A) Venn diagram showing the overlap of TAL1 and/or GATA1 binding locations further classified as TAL1-only, GATA1/TAL1-cooccupied, and GATA1-only binding locations. Cooccupancy was defined as having midpoints <40 bp apart. (B) Composite distribution of TAL1 (black trace) and GATA1 (red-shaded trace) ChIP-exo tag 5′ ends around the WGATAA motif (the underlined A is set to zero) at GATA1 binding locations. Tags that mapped to the reverse strand that contained the WGATAA motif are displayed as inverted traces. The data were plotted as a moving average of 5 bp. Above the plot are X-ray structure images of TAL1 and GATA1 bound to cognate DNA sites (Protein Data Bank [PDB] accession codes 3VEK for GATA1 [J. M. Matthews, unpublished data] and 2YPB for TAL1 [SCL:E47] [17]). The circled X indicates the deduced site of cross-linking of GATA1 (red) and TAL1 (gold).CC BY · PubMed Central
FIG 5 Uniformity of GATA1 and TAL1 cross-linking around WGATAA motifs. (A) Distribution of 5′ ends of GATA1 and TAL1 ChIP-exo sequencing tags around 7,927 WGATAA sites in GATA1 binding locations, comprised of 2,997 GATA1-enriched WGATAA sites (rows) having TAL1 and 4,930 sites lacking TAL1. Both GATA1- and TAL1-cobound WGATAA sites (top) and GATA1-only-bound WGATAA sites (bottom) were sorted by GATA1 occupancy in merged data for all time points. Occupancy was calculated as the sum of total tags around the motif reference point (WGATAA) from 25 bp upstream to 25 bp downstream and 40 bp upstream to 30 bp downstream for GATA1 and TAL1, respectively. (Left) Four-color plot showing the nucleotide compositions of 50-bp surrounding regions of the WGATAA motif: A (red), C (blue), G (gold), and T (green). The bar graphs on the right show the distances of the TG motif from WGATAA sites in GATA1- and TAL1-cobound locations (top) and for GATA1 binding sites with no/low TAL1 occupancy (bottom). (B) TAL1 and GATA1 cross-linking around WGATAA sites sorted by TAL1 occupancy. Shown is the distribution of the 5′ ends of GATA1 and TAL1 ChIP-exo sequencing tags around 7,927 WGATAA sites in GATA1 binding locations. The rows are linked across all the columns and sorted by TAL1 occupancy of merged data for all the time points. TAL1 occupancy was calculated as the sum of tags from 40 bp upstream and 30 bp downstream of the motif reference point (WGATAA). The far-left column shows the nucleotide compositions of regions 25 bp upstream and downstream of the GATA motif reference point.
FIG 5 Uniformity of GATA1 and TAL1 cross-linking around WGATAA motifs. (A) Distribution of 5′ ends of GATA1 and TAL1 ChIP-exo sequencing tags around 7,927 WGATAA sites in GATA1 binding locations, comprised of 2,997 GATA1-enriched WGATAA sites (rows) having TAL1 and 4,930 sites lacking TAL1. Both GATA1- and TAL1-cobound WGATAA sites (top) and GATA1-only-bound WGATAA sites (bottom) were sorted by GATA1 occupancy in merged data for all time points. Occupancy was calculated as the sum of total tags around the motif reference point (WGATAA) from 25 bp upstream to 25 bp downstream and 40 bp upstream to 30 bp downstream for GATA1 and TAL1, respectively. (Left) Four-color plot showing the nucleotide compositions of 50-bp surrounding regions of the WGATAA motif: A (red), C (blue), G (gold), and T (green). The bar graphs on the right show the distances of the TG motif from WGATAA sites in GATA1- and TAL1-cobound locations (top) and for GATA1 binding sites with no/low TAL1 occupancy (bottom). (B) TAL1 and GATA1 cross-linking around WGATAA sites sorted by TAL1 occupancy. Shown is the distribution of the 5′ ends of GATA1 and TAL1 ChIP-exo sequencing tags around 7,927 WGATAA sites in GATA1 binding locations. The rows are linked across all the columns and sorted by TAL1 occupancy of merged data for all the time points. TAL1 occupancy was calculated as the sum of tags from 40 bp upstream and 30 bp downstream of the motif reference point (WGATAA). The far-left column shows the nucleotide compositions of regions 25 bp upstream and downstream of the GATA motif reference point.CC BY · PubMed Central
FIG 6 GATA1 and TAL1 tag distribution around WGATAA sites sorted by the occupancy of individual peaks of GATA1 four-peak locations. Tags on the same strand as the motif (blue) and the opposite strand (red) and sequences in the sequence composition plot were sorted by occupancy of sense peak 1 (defined as 0 to 25 bp upstream of the WGATAA reference point) (top left), sense peak 2 (0 to 25 bp downstream) (top right), antisense peak 1 (25 bp upstream and 5 bp downstream) (bottom left), and antisense peak 2 (5 to 25 bp downstream) (bottom right).
FIG 6 GATA1 and TAL1 tag distribution around WGATAA sites sorted by the occupancy of individual peaks of GATA1 four-peak locations. Tags on the same strand as the motif (blue) and the opposite strand (red) and sequences in the sequence composition plot were sorted by occupancy of sense peak 1 (defined as 0 to 25 bp upstream of the WGATAA reference point) (top left), sense peak 2 (0 to 25 bp downstream) (top right), antisense peak 1 (25 bp upstream and 5 bp downstream) (bottom left), and antisense peak 2 (5 to 25 bp downstream) (bottom right).CC BY · PubMed Central
FIG 7 Sequence composition around cross-linking sites when sorted by GATA1 cross-linking levels in individual peaks of 4-peak patterns. Shown is the relationship of nucleotide compositions in bp −18 to +3 from the WGATAA reference point and GATA1 cross-linking levels in individual peaks of GATA1. For each position, the nucleotide compositions in low to high (left to right) cross-linking of an individual peak are shown through 4 adjacent bars.
FIG 7 Sequence composition around cross-linking sites when sorted by GATA1 cross-linking levels in individual peaks of 4-peak patterns. Shown is the relationship of nucleotide compositions in bp −18 to +3 from the WGATAA reference point and GATA1 cross-linking levels in individual peaks of GATA1. For each position, the nucleotide compositions in low to high (left to right) cross-linking of an individual peak are shown through 4 adjacent bars.CC BY · PubMed Central
FIG 8 Pie chart of the nucleotide composition of the E-box portion of the composite motif (TG[N7–8]WGATAA). When denoting the E-box motif as ZZNNTG, binding locations were first classified by their proportions of ZZ (left). Then, the most frequent CANNTG locations were further classified by their NN contents (right).
FIG 8 Pie chart of the nucleotide composition of the E-box portion of the composite motif (TG[N7–8]WGATAA). When denoting the E-box motif as ZZNNTG, binding locations were first classified by their proportions of ZZ (left). Then, the most frequent CANNTG locations were further classified by their NN contents (right).CC BY · PubMed Central
FIG 9 ChIP-exo versus ChIP-seq in location detection. (A) Venn diagram showing overlap of ChIP-exo and ChIP-seq binding locations for GATA1 (G1E-ER4 cells at 0 h, 3 h, and 24 h) and TAL1 (G1E and G1E-ER4 cells at 24 h). The upper numbers correspond to the number of ChIP-seq binding intervals that overlapped with at least one ChIP-exo peak pair location (window size = 80 bp), while the number below is the number of ChIP-exo peak pair locations overlapping with ChIP-seq intervals. The percentages in parentheses are the percentages of locations containing a WGATAA motif <40 bp from the ChIP-exo or ChIP-seq location. (B) Scatter plots with Spearman's correlation coefficient of log10-transformed ChIP-exo and ChIP-seq tag counts (window size = 400 bp) around all binding locations of ChIP-exo and ChIP-seq at GATA1 and G1E-ER4 cells at 3 h. The red to blue colors indicate the high to low densities of data points in the scatter plot. Similar results were obtained for the other time point data sets (not shown). (C) Composite distribution of G1E-ER4 3-h cell ChIP-exo reads (smoothing = 20 bp) around occupied segments detected only by ChIP-seq (n = 14,071). Similar results were obtained for other time points.
FIG 9 ChIP-exo versus ChIP-seq in location detection. (A) Venn diagram showing overlap of ChIP-exo and ChIP-seq binding locations for GATA1 (G1E-ER4 cells at 0 h, 3 h, and 24 h) and TAL1 (G1E and G1E-ER4 cells at 24 h). The upper numbers correspond to the number of ChIP-seq binding intervals that overlapped with at least one ChIP-exo peak pair location (window size = 80 bp), while the number below is the number of ChIP-exo peak pair locations overlapping with ChIP-seq intervals. The percentages in parentheses are the percentages of locations containing a WGATAA motif <40 bp from the ChIP-exo or ChIP-seq location. (B) Scatter plots with Spearman's correlation coefficient of log10-transformed ChIP-exo and ChIP-seq tag counts (window size = 400 bp) around all binding locations of ChIP-exo and ChIP-seq at GATA1 and G1E-ER4 cells at 3 h. The red to blue colors indicate the high to low densities of data points in the scatter plot. Similar results were obtained for the other time point data sets (not shown). (C) Composite distribution of G1E-ER4 3-h cell ChIP-exo reads (smoothing = 20 bp) around occupied segments detected only by ChIP-seq (n = 14,071). Similar results were obtained for other time points.CC BY · PubMed Central
FIG 10 Dynamics of GATA1 and TAL1 binding during erythroid development. (A) (Left) Heat map of GATA1 and TAL1 total-tag-count-normalized tags (strand merged) plotted around GATA1 binding locations, which were clustered by differential occupancy of GATA1 between time points. Differential occupancies between time points (G1E-ER4 cells at 3 h/0 h, 24 h/0 h, and 24 h/3 h) were determined as the log2-fold change in the ratio of GATA1 tags over the input control between time points using EdgeR and MultiGPS. (Right) For each binding location, a heat map of log2-fold expression change (Expr. Δ) of the closest gene between G1E-ER4 cells at 3 h/0 h, 24 h/0 h, and 24 h/3 h is shown, where increased, unchanged, and decreased expression are depicted as yellow, black, and blue, respectively. (B) Box plot of expression change (G1E-ER4 cells at 24 h/0 h) (left) and distance to the closest transcription start site (right). Differences that were insignificant (P > 0.05) after examining the differences among pairwise combinations of all clusters, using a Mann-Whitney test, are indicated (#).
FIG 10 Dynamics of GATA1 and TAL1 binding during erythroid development. (A) (Left) Heat map of GATA1 and TAL1 total-tag-count-normalized tags (strand merged) plotted around GATA1 binding locations, which were clustered by differential occupancy of GATA1 between time points. Differential occupancies between time points (G1E-ER4 cells at 3 h/0 h, 24 h/0 h, and 24 h/3 h) were determined as the log2-fold change in the ratio of GATA1 tags over the input control between time points using EdgeR and MultiGPS. (Right) For each binding location, a heat map of log2-fold expression change (Expr. Δ) of the closest gene between G1E-ER4 cells at 3 h/0 h, 24 h/0 h, and 24 h/3 h is shown, where increased, unchanged, and decreased expression are depicted as yellow, black, and blue, respectively. (B) Box plot of expression change (G1E-ER4 cells at 24 h/0 h) (left) and distance to the closest transcription start site (right). Differences that were insignificant (P > 0.05) after examining the differences among pairwise combinations of all clusters, using a Mann-Whitney test, are indicated (#).CC BY · 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

Research works printed but not linked: 1

  1. Mahony S, Edwards MD, Mazzoni EO, Sherwood RI, Kakumanu A, Morrison CA, Wichterle H, Gifford DK 2014 An integrated model of multiple-condition ChIP-seq data reveals predeterminants of Cdx2 binding. PLoS Comput Biol 10: e1003501. doi: 10.1371/journal.pcbi.1003501. 24675637 PMC3967921 · DOI · Paper page

Cited by

See all 57

Metadata

Authors
Han, G. C., Vinayachandran, V., Bataille, A. R., Park, B., Chan-Salis, K. Y., Keller, C. A., Long, M., Mahony, S., Hardison, R. C., & Pugh, B. F.
Affiliations
Pennsylvania State University (US)
Corresponding author
Pugh, B. Franklin
Venue
Molecular and Cellular Biology
Published
2016-01-01
Volume
36
Issue
1
Pages
157-172
ISSN
  • 0270-7306 (ISSN-L)
  • 1098-5549
OpenAlex
W2179073481
Cited by
48
References
51
Citations by year
2026: 1 · 2025: 6 · 2024: 6 · 2023: 5 · 2022: 4 · 2021: 7 · 2020: 9 · 2019: 3 · 2018: 2 · 2017: 4 · 2016: 1
FWCI
1.01
Type
journal article
Language
en
Open access
Locations
  • Molecular and Cellular Biology (publishedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
Funders
HHS | National Institutes of Health (R01DK065806, RC2HG005573, R56DK065806, U54HG006998, R01GM059055)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Genomics and Chromatin Dynamics
  • Health Sciences › Medicine › Genetics (clinical) › Hemoglobinopathies and Related Disorders
  • Health Sciences › Medicine › Hematology › Hematopoietic Stem Cell Transplantation
Keywords
GATA-1, TAL1, erythroid development, transcription factor binding sites, E-box, transcriptional regulation, erythroid differentiation, mouse genome, ChIP-exo
MeSH terms
T-Cell Acute Lymphocytic Leukemia Protein 1, Animals, Base Sequence, Cell Differentiation, DNA, Gene Expression Regulation, Proto-Oncogene Proteins, Transcription Factors, Sequence Analysis, DNA, Genomics, Chromatin Immunoprecipitation, GATA1 Transcription Factor, Mice, Basic Helix-Loop-Helix Proteins, Basic Helix-Loop-Helix Transcription Factors

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

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.