Functional footprinting of regulatory DNA

Journal name:
Nature Methods
Volume:
12,
Pages:
927–930
Year published:
DOI:
doi:10.1038/nmeth.3554
Received
Accepted
Published online

Regulatory regions harbor multiple transcription factor (TF) recognition sites; however, the contribution of individual sites to regulatory function remains challenging to define. We describe an approach that exploits the error-prone nature of genome editing–induced double-strand break repair to map functional elements within regulatory DNA at nucleotide resolution. We demonstrate the approach on a human erythroid enhancer, revealing single TF recognition sites that gate the majority of downstream regulatory function.

At a glance

Figures

  1. Assessing effects of footprint-targeted genome editing of the BCL11A enhancer on fetal globin mRNA levels in human erythrocytes.
    Figure 1: Assessing effects of footprint-targeted genome editing of the BCL11A enhancer on fetal globin mRNA levels in human erythrocytes.

    (a) Overview of the functional footprinting approach to determine function of regulatory DNA. (b) Schematic of γ-globin gene regulation by BCL11A. An erythroid-specific intronic enhancer activates BCL11A expression, which represses γ-globin expression. (c) Per-nucleotide conservation, DNase I cleavage and computationally predicted protein-DNA interactions (motifs and footprints) within the +58 DNase I–hypersensitive site (DHS) of the BCL11A erythroid enhancer. (d) Effect of engineered zinc-finger nuclease (ZFN) editing on γ-globin mRNA expression relative to that of β-globin. Both γ-globin and β-globin were measured independently in replicate (each n = 2) and then combinatorially normalized to each other (total n = 4). Bars and error represent mean ± s.d. of the normalized data.

  2. Functional footprinting via genome editing and cell phenotyping reveals the precise boundaries of a GATA1 binding site in the BCL11A erythroid enhancer.
    Figure 2: Functional footprinting via genome editing and cell phenotyping reveals the precise boundaries of a GATA1 binding site in the BCL11A erythroid enhancer.

    (a) Genotypes derived from the alleles present in cells with low and high γ-globin protein expression edited with ZFN Z5. (b) Proportion of total alleles genotyped that retain a match to the GATA1 consensus sequence in cells with low and high γ-globin protein expression. (c) Meta-analysis of deep-sequencing data on cells containing high vs. low levels of γ-globin reveals the precise boundaries of a GATA1 binding site. The two bar charts plot the frequency for which a given base pair is deleted in erythroblasts with high and low levels of γ-globin (top and bottom, respectively). The WebLogos (bottom) show the normalized (high γ-globin – low γ-globin) nucleotide deletion frequency compared to the idealized GATA1 consensus sequence.

  3. Location-specific effects on fetal globin following genome editing of the BCL11A locus in CD34+ cells.
    Supplementary Fig. 1: Location-specific effects on fetal globin following genome editing of the BCL11A locus in CD34+ cells.

    (a) Bar chart showing the editing efficacy (blue bars) of ZFNs in primary human CD34+ cells targeting open reading frames and their effect on relative γ-globin mRNA expression (grey bars). mRNAs encoding ZFNs targeting coding sequence or control (GFP) were transfected into CD34+ cells and the relative level of γ-globin was measured following in vitro erythropoiesis. Editing efficiency is computed as the percentage of genotyped alleles bearing an indel. Both γ-globin and 18S rRNA were measured independently in replicate (each n = 2) and then combinatorially normalized to each other (total n = 4). Bars indicate the mean and error bars indicate the standard deviation of the normalized data. (b) Same as a except γ-globin expression relative to α-globin (orange bars) or β-globin (pink bars). (c) Deletion of individual DNase I hypersensitive sites (DHSs) from the BCL11A erythroid enhancer. Human CD34+ cells were electroporated with mRNA encoding control (GFP) or TALENs designed to delete the indicated DHS were genotyped by PCR following in vitro erythropoiesis. (d) Effect of TALEN-driven deletion of individual DHSs comprising the BCL11A enhancer on relative γ-globin mRNA levels following in vitro erythropoiesis. Both γ-globin and β-globin were measured independently in replicate (each n = 2) and then combinatorially normalized to each other (total n = 4). Bars indicate the mean and error bars indicate the standard deviation of the normalized data. (e) Same as c deleting specific stretches of the +58 DHS as indicated. (f) Same as d for regions indicated in e.

  4. Effect of coding and regulatory DNA editing on erythroid development and maturation.
    Supplementary Fig. 2: Effect of coding and regulatory DNA editing on erythroid development and maturation.

    (a) FACS analysis of differentiating CD34+ cells for the erythroid cell-surface protein glycophorin-A with type of edit (none, BCL11A coding KO or enhancer ablation) vs. days in differentiation media. (b) Exemplary morphology of differentiating CD34+ cells as in a. Cells were centrifuged onto slides and stained using the Wright-Giemsa technique.

  5. Effect of coding and regulatory DNA editing on BCL11A and globin mRNA expression levels.
    Supplementary Fig. 3: Effect of coding and regulatory DNA editing on BCL11A and globin mRNA expression levels.

    (a) Levels of BCL11A mRNA relative to 18S rRNA following electroporation of human CD34+ cells with mRNA encoding ZFNs or control (GFP) and in vitro erythropoiesis. Both BCL11A and 18S rRNA were measured independently in replicate (each n = 2) and then combinatorially normalized to each other (total n = 4). Bars indicate the mean and error bars indicate the standard deviation of the normalized data. (b) Scatterplot of γ-globin vs. β-globin mRNA levels in CD34+-derived erythroid cells after exposure to ZFNs targeting DNase I footprints within the BCL11A erythroid enhancer (black) or the coding region of BCL11A (brown). Both γ-globin and β-globin were measured independently in replicate (each n = 2) and then combinatorially normalized to each other (total n = 4). Bars indicate the mean and error bars indicate the standard deviation of the normalized data. (c) Levels of γ-globin mRNA normalized to that of total β-like globin (γ-globin + β-globin) following electroporation of human CD34+ cells with mRNA encoding ZFNs or control (GFP) and in vitro erythropoiesis. Both γ-globin and β-globin were measured independently in replicate (each n = 2) and then combinatorially normalized to each other (total n = 4). Bars indicate the mean and error bars indicate the standard deviation of the normalized data.

  6. Fine-resolution unbiased reverse genetic tiling analysis of the +58 DHS.
    Supplementary Fig. 4: Fine-resolution unbiased reverse genetic tiling analysis of the +58 DHS.

    (a) Map of engineered TALENs tiling the +58 DHS with respect to per-nucleotide conservation, DNase I cleavage and computationally predicted protein-DNA interactions (motifs and footprints). (b) Per-nucleotide editing efficiencies of individual TALEN pairs determined by amplicon sequencing (Online Methods). Each panel shows different amplicons tiling the +58 DHS.

  7. Effect of TALEN-mediated editing on [gamma]-globin mRNA expression.
    Supplementary Fig. 5: Effect of TALEN-mediated editing on γ-globin mRNA expression.

    (a) Bar chart showing the editing efficacy of TALENs shown in Supplementary Figure 2. Editing efficiency is computed as the percentage of genotyped alleles bearing an indel in the overall cell population determined via amplicon sequencing (Online Methods). (b) Levels of γ-globin mRNA relative to that of β-globin following electroporation of human CD34+ cells with mRNA encoding TALENs (shown in Supplementary Fig. 4) and in vitro erythropoiesis. Both γ-globin and β-globin were measured independently in replicate (each n = 2) and then combinatorially normalized to each other (total n = 4). Bars indicate the mean and error bars indicate the standard deviation of the normalized data. (c) Relative γ-globin mRNA levels (to β-globin) (mean ± s.d.; from b) vs. TALEN editing efficiency (from a). Dotted grey line is the best-fit robust linear model (Online Methods). Dotted red line indicates the 95% CI of the linear model mean. (d) Residual plot of linear model in c used to determine TALEN edits significantly associated with increased γ-globin expression. The red dotted lines show the 1 standard deviation of the residuals; samples that exceed this threshold are associated with significant changes in their γ-globin expression.

  8. Cell sorting on [gamma]-globin protein expression levels.
    Supplementary Fig. 6: Cell sorting on γ-globin protein expression levels.

    Scatterplot of fluorescence-activated cell sorting (FACS) of in vitro generated erythrocytes following targeted disruption of footprint 5 within the +58 DHS using ZFNs. Boxes indicate the FACS gates used to isolate populations of erythroblasts bearing high and low levels of γ-globin protein.

  9. Fine-resolution analysis of the GATA1 motif within the +58 DHS using ZFN-driven genome editing.
    Supplementary Fig. 7: Fine-resolution analysis of the GATA1 motif within the +58 DHS using ZFN-driven genome editing.

    (a) Diagram of the engineered target specificity of ZFN pairs Z5 (top) and 1-bp shift (bottom). Blue and yellow indicate the predicted binding sites for the individual ZFN monomers. (b) Proportion of edited (containing an indel; e.g., non-reference) alleles genotyped that retain a match to the GATA1 consensus sequence following electroporation of either Z5 or 1-bp-shifted mRNA encoding ZFNs and in vitro erythropoiesis. (c) Levels of BCL11A mRNA relative to 18S rRNA following electroporation of human CD34+ cells with mRNA encoding ZFNs or control (GFP) and in vitro erythropoiesis. Both BCL11A and 18S rRNA were measured independently in replicate (each n = 2) and then combinatorially normalized to each other (total n = 4). Bars indicate the mean and error bars indicate the standard deviation of the normalized data. (d) Same as c for levels of γ-globin mRNA relative to that of β-globin.

References

  1. Maurano, M.T. et al. Science 337, 11901195 (2012).
  2. Carroll, D. Annu. Rev. Biochem. 83, 409439 (2014).
  3. Bauer, D.E. et al. Science 342, 253257 (2013).
  4. Bauer, D.E. & Orkin, S.H. Curr. Opin. Pediatr. 23, 18 (2011).
  5. Urnov, F.D., Rebar, E.J., Holmes, M.C., Zhang, H.S. & Gregory, P.D. Nat. Rev. Genet. 11, 636646 (2010).
  6. Miller, J.C. et al. Nat. Biotechnol. 29, 143148 (2011).
  7. Giarratana, M.-C. et al. Blood 118, 50715079 (2011).
  8. DeKelver, R.C. et al. Genome Res. 20, 11331142 (2010).
  9. Neph, S. et al. Nature 489, 8390 (2012).
  10. Ko, L.J. & Engel, J.D. Mol. Cell. Biol. 13, 40114022 (1993).
  11. ENCODE Project Consortium. et al. Nature 489, 5774 (2012).
  12. Shivdasani, R.A., Mayer, E.L. & Orkin, S.H. Nature 373, 432434 (1995).
  13. Chen, R.-L., Chou, Y.-C., Lan, Y.-J., Huang, T.-S. & Shen, C.K.J. J. Biol. Chem. 285, 1018910197 (2010).
  14. Findlay, G.M., Boyle, E.A., Hause, R.J., Klein, J.C. & Shendure, J. Nature 513, 120123 (2014).
  15. Klemm, S. et al. Nat. Methods 11, 549551 (2014).
  16. Miller, J.C. et al. Nat. Biotechnol. 25, 778785 (2007).
  17. Guschin, D.Y. et al. Methods Mol. Biol. 649, 247256 (2010).
  18. Doyon, Y. et al. Nat. Methods 7, 459460 (2010).
  19. Thorpe, S.J. et al. Br. J. Haematol. 87, 125132 (1994).
  20. Chang, K.-H. et al. Stem Cell Rev. 9, 397407 (2013).
  21. John, S. et al. Curr. Protoc. Mol. Biol. 103 21.27 (2013).

Download references

Author information

  1. These authors contributed equally to this work.

    • Jeff Vierstra &
    • Andreas Reik

Affiliations

  1. Department of Genome Sciences, University of Washington, Seattle, Washington, USA.

    • Jeff Vierstra,
    • Sandra Stehling-Sun &
    • John A Stamatoyannopoulos
  2. Sangamo BioSciences, Pt. Richmond, California, USA.

    • Andreas Reik,
    • Yuanyue Zhou,
    • Sarah J Hinkley,
    • David E Paschon,
    • Lei Zhang,
    • Yuri R Bendana,
    • Colleen M O'Neil,
    • Alexander H Song,
    • Andrea K Mich,
    • Pei-Qi Liu,
    • Gary Lee,
    • Michael C Holmes,
    • Edward J Rebar,
    • Philip D Gregory &
    • Fyodor D Urnov
  3. Department of Medicine, Division of Hematology, University of Washington, Seattle, Washington, USA.

    • Kai-Hsin Chang,
    • Nikoletta Psatha &
    • Thalia Papayannopoulou
  4. Boston Children's Hospital, Division of Hematology/Oncology, Boston, Massachusetts, USA.

    • Daniel E Bauer &
    • Stuart H Orkin
  5. Department of Medicine, Division of Medical Genetics, University of Washington, Seattle, Washington, USA.

    • George Stamatoyannopoulos
  6. Department of Medicine, Division of Oncology, University of Washington, Seattle, Washington, USA.

    • John A Stamatoyannopoulos

Contributions

K.-H.C., S.S.-S., Y.Z., S.J.H., D.E.P., L.Z., C.M.O., A.H.S., A.K.M. and N.P. performed the experiments and data collection. J.V., A.R., K.-H.C., Y.R.B., P.-Q.L., G.L., M.C.H., E.J.R., T.P., G.S. and F.D.U. analyzed the data. D.E.B. and S.H.O. provided critical insights. J.V., J.A.S. and F.D.U. wrote the manuscript with input from A.R., K.-H.C., E.J.R., P.D.G., T.P. and G.S.

Competing financial interests

All Sangamo authors (A.H.S., A.K.M., A.R., C.M.O., D.E.P., E.J.R., F.D.U., G.L., L.Z., M.C.H., P.-Q.L., P.D.G. S.J.H., Y.Z. and Y.R.B.) are full-time employees of Sangamo BioSciences, Inc. Patent applications have been filed covering certain aspects of this work.

Corresponding authors

Correspondence to:

Author details

Supplementary information

Supplementary Figures

  1. Supplementary Figure 1: Location-specific effects on fetal globin following genome editing of the BCL11A locus in CD34+ cells. (134 KB)

    (a) Bar chart showing the editing efficacy (blue bars) of ZFNs in primary human CD34+ cells targeting open reading frames and their effect on relative γ-globin mRNA expression (grey bars). mRNAs encoding ZFNs targeting coding sequence or control (GFP) were transfected into CD34+ cells and the relative level of γ-globin was measured following in vitro erythropoiesis. Editing efficiency is computed as the percentage of genotyped alleles bearing an indel. Both γ-globin and 18S rRNA were measured independently in replicate (each n = 2) and then combinatorially normalized to each other (total n = 4). Bars indicate the mean and error bars indicate the standard deviation of the normalized data. (b) Same as a except γ-globin expression relative to α-globin (orange bars) or β-globin (pink bars). (c) Deletion of individual DNase I hypersensitive sites (DHSs) from the BCL11A erythroid enhancer. Human CD34+ cells were electroporated with mRNA encoding control (GFP) or TALENs designed to delete the indicated DHS were genotyped by PCR following in vitro erythropoiesis. (d) Effect of TALEN-driven deletion of individual DHSs comprising the BCL11A enhancer on relative γ-globin mRNA levels following in vitro erythropoiesis. Both γ-globin and β-globin were measured independently in replicate (each n = 2) and then combinatorially normalized to each other (total n = 4). Bars indicate the mean and error bars indicate the standard deviation of the normalized data. (e) Same as c deleting specific stretches of the +58 DHS as indicated. (f) Same as d for regions indicated in e.

  2. Supplementary Figure 2: Effect of coding and regulatory DNA editing on erythroid development and maturation. (197 KB)

    (a) FACS analysis of differentiating CD34+ cells for the erythroid cell-surface protein glycophorin-A with type of edit (none, BCL11A coding KO or enhancer ablation) vs. days in differentiation media. (b) Exemplary morphology of differentiating CD34+ cells as in a. Cells were centrifuged onto slides and stained using the Wright-Giemsa technique.

  3. Supplementary Figure 3: Effect of coding and regulatory DNA editing on BCL11A and globin mRNA expression levels. (67 KB)

    (a) Levels of BCL11A mRNA relative to 18S rRNA following electroporation of human CD34+ cells with mRNA encoding ZFNs or control (GFP) and in vitro erythropoiesis. Both BCL11A and 18S rRNA were measured independently in replicate (each n = 2) and then combinatorially normalized to each other (total n = 4). Bars indicate the mean and error bars indicate the standard deviation of the normalized data. (b) Scatterplot of γ-globin vs. β-globin mRNA levels in CD34+-derived erythroid cells after exposure to ZFNs targeting DNase I footprints within the BCL11A erythroid enhancer (black) or the coding region of BCL11A (brown). Both γ-globin and β-globin were measured independently in replicate (each n = 2) and then combinatorially normalized to each other (total n = 4). Bars indicate the mean and error bars indicate the standard deviation of the normalized data. (c) Levels of γ-globin mRNA normalized to that of total β-like globin (γ-globin + β-globin) following electroporation of human CD34+ cells with mRNA encoding ZFNs or control (GFP) and in vitro erythropoiesis. Both γ-globin and β-globin were measured independently in replicate (each n = 2) and then combinatorially normalized to each other (total n = 4). Bars indicate the mean and error bars indicate the standard deviation of the normalized data.

  4. Supplementary Figure 4: Fine-resolution unbiased reverse genetic tiling analysis of the +58 DHS. (184 KB)

    (a) Map of engineered TALENs tiling the +58 DHS with respect to per-nucleotide conservation, DNase I cleavage and computationally predicted protein-DNA interactions (motifs and footprints). (b) Per-nucleotide editing efficiencies of individual TALEN pairs determined by amplicon sequencing (Online Methods). Each panel shows different amplicons tiling the +58 DHS.

  5. Supplementary Figure 5: Effect of TALEN-mediated editing on γ-globin mRNA expression. (141 KB)

    (a) Bar chart showing the editing efficacy of TALENs shown in Supplementary Figure 2. Editing efficiency is computed as the percentage of genotyped alleles bearing an indel in the overall cell population determined via amplicon sequencing (Online Methods). (b) Levels of γ-globin mRNA relative to that of β-globin following electroporation of human CD34+ cells with mRNA encoding TALENs (shown in Supplementary Fig. 4) and in vitro erythropoiesis. Both γ-globin and β-globin were measured independently in replicate (each n = 2) and then combinatorially normalized to each other (total n = 4). Bars indicate the mean and error bars indicate the standard deviation of the normalized data. (c) Relative γ-globin mRNA levels (to β-globin) (mean ± s.d.; from b) vs. TALEN editing efficiency (from a). Dotted grey line is the best-fit robust linear model (Online Methods). Dotted red line indicates the 95% CI of the linear model mean. (d) Residual plot of linear model in c used to determine TALEN edits significantly associated with increased γ-globin expression. The red dotted lines show the 1 standard deviation of the residuals; samples that exceed this threshold are associated with significant changes in their γ-globin expression.

  6. Supplementary Figure 6: Cell sorting on γ-globin protein expression levels. (45 KB)

    Scatterplot of fluorescence-activated cell sorting (FACS) of in vitro generated erythrocytes following targeted disruption of footprint 5 within the +58 DHS using ZFNs. Boxes indicate the FACS gates used to isolate populations of erythroblasts bearing high and low levels of γ-globin protein.

  7. Supplementary Figure 7: Fine-resolution analysis of the GATA1 motif within the +58 DHS using ZFN-driven genome editing. (70 KB)

    (a) Diagram of the engineered target specificity of ZFN pairs Z5 (top) and 1-bp shift (bottom). Blue and yellow indicate the predicted binding sites for the individual ZFN monomers. (b) Proportion of edited (containing an indel; e.g., non-reference) alleles genotyped that retain a match to the GATA1 consensus sequence following electroporation of either Z5 or 1-bp-shifted mRNA encoding ZFNs and in vitro erythropoiesis. (c) Levels of BCL11A mRNA relative to 18S rRNA following electroporation of human CD34+ cells with mRNA encoding ZFNs or control (GFP) and in vitro erythropoiesis. Both BCL11A and 18S rRNA were measured independently in replicate (each n = 2) and then combinatorially normalized to each other (total n = 4). Bars indicate the mean and error bars indicate the standard deviation of the normalized data. (d) Same as c for levels of γ-globin mRNA relative to that of β-globin.

PDF files

  1. Supplementary Text and Figures (1,059 KB)

    Supplementary Figures 1–7

Excel files

  1. Supplementary Table 1 (38,106 KB)

    ZFNs used in this study.

  2. Supplementary Table 2 (9,969 KB)

    TALENs used in this study.

  3. Supplementary Table 3 (1,669 KB)

    Genotype frequencies of TALEN induced edits within the +58 DHS.

  4. Supplementary Table 4 (36,954 KB)

    Genotype frequencies in high and low γ-globin expressing cells after exposure to ZFN Z5.

  5. Supplementary Table 5 (99,358 KB)

    Genotype frequencies from ZFN Z5 or ZFN 1-bp shift.

Additional data