Limits...
A model of spatially restricted transcription in opposing gradients of activators and repressors.

White MA, Parker DS, Barolo S, Cohen BA - Mol. Syst. Biol. (2012)

Bottom Line: This model quantitatively predicts the boundaries of gene expression within OARGs.When trained on experimental data, our model accounts for the counterintuitive observation that increasing the affinity of binding sites in enhancers of Hedgehog target genes produces more restricted transcription within Hedgehog gradients in Drosophila.Because our model is general, it may explain the role of low-affinity binding sites in many contexts, including mammalian Hedgehog gradients.

View Article: PubMed Central - PubMed

Affiliation: Department of Genetics, Center for Genome Sciences and Systems Biology, Washington University School of Medicine, St Louis, MO 63108, USA.

ABSTRACT
Morphogens control patterns of transcription in development, often by establishing concentration gradients of a single transcriptional activator. However, many morphogens, including Hedgehog, create opposing activator and repressor gradients (OARGs). In contrast to single activator gradients, it is not well understood how OARGs control transcriptional patterns. We present a general thermodynamic model that explains how spatial patterns of gene expression are established within OARGs. The model predicts that differences in enhancer binding site affinities for morphogen-responsive transcription factors (TFs) produce discrete transcriptional boundaries, but only when either activators or repressors bind cooperatively. This model quantitatively predicts the boundaries of gene expression within OARGs. When trained on experimental data, our model accounts for the counterintuitive observation that increasing the affinity of binding sites in enhancers of Hedgehog target genes produces more restricted transcription within Hedgehog gradients in Drosophila. Because our model is general, it may explain the role of low-affinity binding sites in many contexts, including mammalian Hedgehog gradients.

Show MeSH

Related in: MedlinePlus

Repressor and activator occupancy curves demonstrate why repressors out-compete activators at high-affinity sites but not at low-affinity sites. Occupancy (y axis) of activator (red) and repressor (blue) shown as a function of the association constant, K (x axis), for a single position within the gradient (fixed [A] and [R]). [A] and [R] in the gradient range between 1 and 100, in arbitrary units, and cooperativity ω is fixed at 30. (A) At high levels of activator, activator occupancy always exceeds repressor occupancy, and all genes are activated regardless of enhancer affinity. (B, C) Within the intermediate zone of the gradient, repressor and activator occupancy curves intersect, due to the inherent steepness of the cooperative repressor binding curve. The curves intersect at a threshold affinity Kthreshold. Higher affinity enhancers (larger K) have higher repressor occupancy and are repressed, while lower affinity enhancers have higher activator occupancy and are activated. (D) When [R]>[A], repressor occupancy always exceeds activator occupancy, and all genes are repressed regardless of enhancer affinity. These results hold for enhancers with more than two sites (Supplementary Figure 1).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3472688&req=5

f3: Repressor and activator occupancy curves demonstrate why repressors out-compete activators at high-affinity sites but not at low-affinity sites. Occupancy (y axis) of activator (red) and repressor (blue) shown as a function of the association constant, K (x axis), for a single position within the gradient (fixed [A] and [R]). [A] and [R] in the gradient range between 1 and 100, in arbitrary units, and cooperativity ω is fixed at 30. (A) At high levels of activator, activator occupancy always exceeds repressor occupancy, and all genes are activated regardless of enhancer affinity. (B, C) Within the intermediate zone of the gradient, repressor and activator occupancy curves intersect, due to the inherent steepness of the cooperative repressor binding curve. The curves intersect at a threshold affinity Kthreshold. Higher affinity enhancers (larger K) have higher repressor occupancy and are repressed, while lower affinity enhancers have higher activator occupancy and are activated. (D) When [R]>[A], repressor occupancy always exceeds activator occupancy, and all genes are repressed regardless of enhancer affinity. These results hold for enhancers with more than two sites (Supplementary Figure 1).

Mentions: To understand how cooperative repressors can lose their advantage over activators at low-affinity sites, we examined what occurs at a single position in an OARG as the affinity of binding sites in an enhancer is decreased. We used our model to plot activator and repressor occupancy as a function of affinity K, at specific positions within the Gli OARG (Figure 3). The result is a ligand binding curve showing the response to changes in binding site affinity at fixed ligand concentration. In the region of the gradient with high levels of activator, activator occupancy is always greater than repressor occupancy, even as total enhancer occupancy decreases with affinity (Figure 3A). In the gradient region where repressor concentration is greater than activator concentration, repressor occupancy always exceeds activator occupancy (Figure 3D). In the middle zone of the gradient, with boundaries defined by Equations (6) and (7), concentrations of activator and repressor are at intermediate levels, but activator concentration is always greater than repressor concentration (Figure 3B and C). Within this OARG region, cooperatively binding repressors out-compete activators for occupancy at high-affinity sites, despite the greater concentration of activators. As binding site affinity decreases, overall occupancy of the enhancer decreases, and, as expected for a cooperatively binding ligand, the occupancy of cooperative repressors decreases more rapidly than the occupancy of non-cooperative activators (Figure 3B and C). Cooperative repressors lose their advantage over activators at low-affinity sites due to the increased steepness inherent in a cooperative ligand binding curve, compared with a non-cooperative binding curve.


A model of spatially restricted transcription in opposing gradients of activators and repressors.

White MA, Parker DS, Barolo S, Cohen BA - Mol. Syst. Biol. (2012)

Repressor and activator occupancy curves demonstrate why repressors out-compete activators at high-affinity sites but not at low-affinity sites. Occupancy (y axis) of activator (red) and repressor (blue) shown as a function of the association constant, K (x axis), for a single position within the gradient (fixed [A] and [R]). [A] and [R] in the gradient range between 1 and 100, in arbitrary units, and cooperativity ω is fixed at 30. (A) At high levels of activator, activator occupancy always exceeds repressor occupancy, and all genes are activated regardless of enhancer affinity. (B, C) Within the intermediate zone of the gradient, repressor and activator occupancy curves intersect, due to the inherent steepness of the cooperative repressor binding curve. The curves intersect at a threshold affinity Kthreshold. Higher affinity enhancers (larger K) have higher repressor occupancy and are repressed, while lower affinity enhancers have higher activator occupancy and are activated. (D) When [R]>[A], repressor occupancy always exceeds activator occupancy, and all genes are repressed regardless of enhancer affinity. These results hold for enhancers with more than two sites (Supplementary Figure 1).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3472688&req=5

f3: Repressor and activator occupancy curves demonstrate why repressors out-compete activators at high-affinity sites but not at low-affinity sites. Occupancy (y axis) of activator (red) and repressor (blue) shown as a function of the association constant, K (x axis), for a single position within the gradient (fixed [A] and [R]). [A] and [R] in the gradient range between 1 and 100, in arbitrary units, and cooperativity ω is fixed at 30. (A) At high levels of activator, activator occupancy always exceeds repressor occupancy, and all genes are activated regardless of enhancer affinity. (B, C) Within the intermediate zone of the gradient, repressor and activator occupancy curves intersect, due to the inherent steepness of the cooperative repressor binding curve. The curves intersect at a threshold affinity Kthreshold. Higher affinity enhancers (larger K) have higher repressor occupancy and are repressed, while lower affinity enhancers have higher activator occupancy and are activated. (D) When [R]>[A], repressor occupancy always exceeds activator occupancy, and all genes are repressed regardless of enhancer affinity. These results hold for enhancers with more than two sites (Supplementary Figure 1).
Mentions: To understand how cooperative repressors can lose their advantage over activators at low-affinity sites, we examined what occurs at a single position in an OARG as the affinity of binding sites in an enhancer is decreased. We used our model to plot activator and repressor occupancy as a function of affinity K, at specific positions within the Gli OARG (Figure 3). The result is a ligand binding curve showing the response to changes in binding site affinity at fixed ligand concentration. In the region of the gradient with high levels of activator, activator occupancy is always greater than repressor occupancy, even as total enhancer occupancy decreases with affinity (Figure 3A). In the gradient region where repressor concentration is greater than activator concentration, repressor occupancy always exceeds activator occupancy (Figure 3D). In the middle zone of the gradient, with boundaries defined by Equations (6) and (7), concentrations of activator and repressor are at intermediate levels, but activator concentration is always greater than repressor concentration (Figure 3B and C). Within this OARG region, cooperatively binding repressors out-compete activators for occupancy at high-affinity sites, despite the greater concentration of activators. As binding site affinity decreases, overall occupancy of the enhancer decreases, and, as expected for a cooperatively binding ligand, the occupancy of cooperative repressors decreases more rapidly than the occupancy of non-cooperative activators (Figure 3B and C). Cooperative repressors lose their advantage over activators at low-affinity sites due to the increased steepness inherent in a cooperative ligand binding curve, compared with a non-cooperative binding curve.

Bottom Line: This model quantitatively predicts the boundaries of gene expression within OARGs.When trained on experimental data, our model accounts for the counterintuitive observation that increasing the affinity of binding sites in enhancers of Hedgehog target genes produces more restricted transcription within Hedgehog gradients in Drosophila.Because our model is general, it may explain the role of low-affinity binding sites in many contexts, including mammalian Hedgehog gradients.

View Article: PubMed Central - PubMed

Affiliation: Department of Genetics, Center for Genome Sciences and Systems Biology, Washington University School of Medicine, St Louis, MO 63108, USA.

ABSTRACT
Morphogens control patterns of transcription in development, often by establishing concentration gradients of a single transcriptional activator. However, many morphogens, including Hedgehog, create opposing activator and repressor gradients (OARGs). In contrast to single activator gradients, it is not well understood how OARGs control transcriptional patterns. We present a general thermodynamic model that explains how spatial patterns of gene expression are established within OARGs. The model predicts that differences in enhancer binding site affinities for morphogen-responsive transcription factors (TFs) produce discrete transcriptional boundaries, but only when either activators or repressors bind cooperatively. This model quantitatively predicts the boundaries of gene expression within OARGs. When trained on experimental data, our model accounts for the counterintuitive observation that increasing the affinity of binding sites in enhancers of Hedgehog target genes produces more restricted transcription within Hedgehog gradients in Drosophila. Because our model is general, it may explain the role of low-affinity binding sites in many contexts, including mammalian Hedgehog gradients.

Show MeSH
Related in: MedlinePlus