Limits...
Identifying synergistic regulation involving c-Myc and sp1 in human tissues.

Parisi F, Wirapati P, Naef F - Nucleic Acids Res. (2007)

Bottom Line: Dual sites show several distinct features compared to the single regulator sites: specifically, they exhibit overall higher degree of conservation between human and rodents, stronger correlation with TFIID-bound promoters, and preference for permissive chromatin state.Namely, the correlation with c-Myc expression in promoters harboring dual-sites is increased for stronger sp1 sites by strong sp1 binding and the effect is largest in proliferating tissues.Our approach shows how integrated functional analyses can uncover tissue-specific and combinatorial regulatory dependencies in mammals.

View Article: PubMed Central - PubMed

Affiliation: Swiss Institute for Experimental Cancer Research (ISREC) and NCCR Molecular Oncology, Lausanne, Switzerland.

ABSTRACT
Combinatorial gene regulation largely contributes to phenotypic versatility in higher eukaryotes. Genome-wide chromatin immuno-precipitation (ChIP) combined with expression profiling can dissect regulatory circuits around transcriptional regulators. Here, we integrate tiling array measurements of DNA-binding sites for c-Myc, sp1, TFIID and modified histones with a tissue expression atlas to establish the functional correspondence between physical binding, promoter activity and transcriptional regulation. For this we develop SLM, a methodology to map c-Myc and sp1-binding sites and then classify sites as sp1-only, c-Myc-only or dual. Dual sites show several distinct features compared to the single regulator sites: specifically, they exhibit overall higher degree of conservation between human and rodents, stronger correlation with TFIID-bound promoters, and preference for permissive chromatin state. By applying regression models to an expression atlas we identified a functionally distinct signature for strong dual c-Myc/sp1 sites. Namely, the correlation with c-Myc expression in promoters harboring dual-sites is increased for stronger sp1 sites by strong sp1 binding and the effect is largest in proliferating tissues. Our approach shows how integrated functional analyses can uncover tissue-specific and combinatorial regulatory dependencies in mammals.

Show MeSH

Related in: MedlinePlus

Relation between target expression, regulator expression levels and ChIP- binding strength for all genes. (A–B) Susceptibilities versus the strength of the ChIP- binding sites (t parameter) for c-Myc (A) and sp1 (B). Each dot is one TSS represented by the highest t score occurring in a fixed −1.5 to +1.5 kb window. Gray line show correlation for c-Myc (r2 = 0.26, P = 10−8) while that for sp1 is not significant. (C) Strength of c-Myc (tMyc) versus sp1 (tsp1) sites. Colored grid indicates the mean of the c-Myc susceptibility ag in each square. Red indicates positive and green negative mean values. Saturating colors represent absolute means ≥0.33. (D) c-Myc sites for two cutoffs (tMyc > 6 in black; tMyc > 9 in red) are binned according to sp1 binding. The smoothed mean (loess regression) of ag in function of tsp1 shows increasing average ag. The increase is more pronounced for stronger c-Myc sites (red). (E–F) Boxplots for the gene susceptibilities ag and bg stratified in groups. To emphasize the dependence on site strength we define groups as follows: the  group (Ø) has tMyc < 9 and tsp1 < 9; the S group has tsp1 > 9; the M group has tMyc > 9; the B group has both tMyc > 9 and tsp1 > 9. Groups are mutually exclusive and group size is indicated above the panels. (E) The distribution for ag shifts upwards: the B group has the highest median followed by the M group. (F) The distribution for bg shows no similar behavior. Comparable results are obtained with different processing of the raw ChIP data (Figures S8 and S9).
© Copyright Policy - openaccess
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC1851645&req=5

Figure 4: Relation between target expression, regulator expression levels and ChIP- binding strength for all genes. (A–B) Susceptibilities versus the strength of the ChIP- binding sites (t parameter) for c-Myc (A) and sp1 (B). Each dot is one TSS represented by the highest t score occurring in a fixed −1.5 to +1.5 kb window. Gray line show correlation for c-Myc (r2 = 0.26, P = 10−8) while that for sp1 is not significant. (C) Strength of c-Myc (tMyc) versus sp1 (tsp1) sites. Colored grid indicates the mean of the c-Myc susceptibility ag in each square. Red indicates positive and green negative mean values. Saturating colors represent absolute means ≥0.33. (D) c-Myc sites for two cutoffs (tMyc > 6 in black; tMyc > 9 in red) are binned according to sp1 binding. The smoothed mean (loess regression) of ag in function of tsp1 shows increasing average ag. The increase is more pronounced for stronger c-Myc sites (red). (E–F) Boxplots for the gene susceptibilities ag and bg stratified in groups. To emphasize the dependence on site strength we define groups as follows: the group (Ø) has tMyc < 9 and tsp1 < 9; the S group has tsp1 > 9; the M group has tMyc > 9; the B group has both tMyc > 9 and tsp1 > 9. Groups are mutually exclusive and group size is indicated above the panels. (E) The distribution for ag shifts upwards: the B group has the highest median followed by the M group. (F) The distribution for bg shows no similar behavior. Comparable results are obtained with different processing of the raw ChIP data (Figures S8 and S9).

Mentions: Switching from a condition-centered to a gene-centered view, we systematically investigate associations between expression levels of genes and ChIP signals in their promoters. We model the expression levels of all genes in the atlas in function of c-Myc and sp1 mRNA levels using multilinear regression. We aim to test whether a correlation between gene expression and regulator activity reflects the strength of binding sites measured with ChIP. For this, the mRNA levels of the regulators are taken as best proxies for the activity levels of the proteins. The model (M1, methods section) assumes no indirect regulation and measures the gene-specific contributions for each transcription factor. To determine whether the susceptibilities reflect binding strength we use the nominal t scores for binding instead of fixed cutoffs as in Figure 3. We find that ag shows a significant correlation (Figure 4A) with the ChIP signal strength for c-Myc (tMyc) while not significant in the case of sp1 (Figure 4B), even though the sp1-bound promoters exhibit weak systematic positive bg. Analysis shows that for genes with dual sites, c-Myc susceptibility generally increases with the strength of sp1 binding (Figure 4C and D). This synergistic trend is confirmed in a stratified representation showing that ag for promoters with both strong c-Myc and sp1 ChIP sites (the B group) is higher than for promoters with weaker sites (Figure 4E). The regression coefficients for genes in group B are listed in Table S1. The susceptibility to sp1 mRNA level does not show similar differences, although the sp1 sites are subject to a slight increase in bg compared to c-Myc sites (Figure 4F). Turning to the significance of the regression parameters ag and bg, we find that the total fraction of genes that correlate significantly with c-Myc is about 65%, while only about 20% correlate with sp1 (Figure S10). For c-Myc, this fraction increases in the c-Myc only (76%, P = 0.09, hypergeometric test) and dual groups (87%, P = 0.025), while bg does not show large differences across groups. Interestingly, while there is overall bias for positive correlations (∼65% for both ag and bg), the fraction of positive ag is significantly enriched in the c-Myc only group (84%, P = 0.01) even more so for the dual sites (91%, P = 0.005). Finally the fraction with positive bg is highest (86%, P = 0.026) for the sp1-only sites.Figure 4.


Identifying synergistic regulation involving c-Myc and sp1 in human tissues.

Parisi F, Wirapati P, Naef F - Nucleic Acids Res. (2007)

Relation between target expression, regulator expression levels and ChIP- binding strength for all genes. (A–B) Susceptibilities versus the strength of the ChIP- binding sites (t parameter) for c-Myc (A) and sp1 (B). Each dot is one TSS represented by the highest t score occurring in a fixed −1.5 to +1.5 kb window. Gray line show correlation for c-Myc (r2 = 0.26, P = 10−8) while that for sp1 is not significant. (C) Strength of c-Myc (tMyc) versus sp1 (tsp1) sites. Colored grid indicates the mean of the c-Myc susceptibility ag in each square. Red indicates positive and green negative mean values. Saturating colors represent absolute means ≥0.33. (D) c-Myc sites for two cutoffs (tMyc > 6 in black; tMyc > 9 in red) are binned according to sp1 binding. The smoothed mean (loess regression) of ag in function of tsp1 shows increasing average ag. The increase is more pronounced for stronger c-Myc sites (red). (E–F) Boxplots for the gene susceptibilities ag and bg stratified in groups. To emphasize the dependence on site strength we define groups as follows: the  group (Ø) has tMyc < 9 and tsp1 < 9; the S group has tsp1 > 9; the M group has tMyc > 9; the B group has both tMyc > 9 and tsp1 > 9. Groups are mutually exclusive and group size is indicated above the panels. (E) The distribution for ag shifts upwards: the B group has the highest median followed by the M group. (F) The distribution for bg shows no similar behavior. Comparable results are obtained with different processing of the raw ChIP data (Figures S8 and S9).
© Copyright Policy - openaccess
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC1851645&req=5

Figure 4: Relation between target expression, regulator expression levels and ChIP- binding strength for all genes. (A–B) Susceptibilities versus the strength of the ChIP- binding sites (t parameter) for c-Myc (A) and sp1 (B). Each dot is one TSS represented by the highest t score occurring in a fixed −1.5 to +1.5 kb window. Gray line show correlation for c-Myc (r2 = 0.26, P = 10−8) while that for sp1 is not significant. (C) Strength of c-Myc (tMyc) versus sp1 (tsp1) sites. Colored grid indicates the mean of the c-Myc susceptibility ag in each square. Red indicates positive and green negative mean values. Saturating colors represent absolute means ≥0.33. (D) c-Myc sites for two cutoffs (tMyc > 6 in black; tMyc > 9 in red) are binned according to sp1 binding. The smoothed mean (loess regression) of ag in function of tsp1 shows increasing average ag. The increase is more pronounced for stronger c-Myc sites (red). (E–F) Boxplots for the gene susceptibilities ag and bg stratified in groups. To emphasize the dependence on site strength we define groups as follows: the group (Ø) has tMyc < 9 and tsp1 < 9; the S group has tsp1 > 9; the M group has tMyc > 9; the B group has both tMyc > 9 and tsp1 > 9. Groups are mutually exclusive and group size is indicated above the panels. (E) The distribution for ag shifts upwards: the B group has the highest median followed by the M group. (F) The distribution for bg shows no similar behavior. Comparable results are obtained with different processing of the raw ChIP data (Figures S8 and S9).
Mentions: Switching from a condition-centered to a gene-centered view, we systematically investigate associations between expression levels of genes and ChIP signals in their promoters. We model the expression levels of all genes in the atlas in function of c-Myc and sp1 mRNA levels using multilinear regression. We aim to test whether a correlation between gene expression and regulator activity reflects the strength of binding sites measured with ChIP. For this, the mRNA levels of the regulators are taken as best proxies for the activity levels of the proteins. The model (M1, methods section) assumes no indirect regulation and measures the gene-specific contributions for each transcription factor. To determine whether the susceptibilities reflect binding strength we use the nominal t scores for binding instead of fixed cutoffs as in Figure 3. We find that ag shows a significant correlation (Figure 4A) with the ChIP signal strength for c-Myc (tMyc) while not significant in the case of sp1 (Figure 4B), even though the sp1-bound promoters exhibit weak systematic positive bg. Analysis shows that for genes with dual sites, c-Myc susceptibility generally increases with the strength of sp1 binding (Figure 4C and D). This synergistic trend is confirmed in a stratified representation showing that ag for promoters with both strong c-Myc and sp1 ChIP sites (the B group) is higher than for promoters with weaker sites (Figure 4E). The regression coefficients for genes in group B are listed in Table S1. The susceptibility to sp1 mRNA level does not show similar differences, although the sp1 sites are subject to a slight increase in bg compared to c-Myc sites (Figure 4F). Turning to the significance of the regression parameters ag and bg, we find that the total fraction of genes that correlate significantly with c-Myc is about 65%, while only about 20% correlate with sp1 (Figure S10). For c-Myc, this fraction increases in the c-Myc only (76%, P = 0.09, hypergeometric test) and dual groups (87%, P = 0.025), while bg does not show large differences across groups. Interestingly, while there is overall bias for positive correlations (∼65% for both ag and bg), the fraction of positive ag is significantly enriched in the c-Myc only group (84%, P = 0.01) even more so for the dual sites (91%, P = 0.005). Finally the fraction with positive bg is highest (86%, P = 0.026) for the sp1-only sites.Figure 4.

Bottom Line: Dual sites show several distinct features compared to the single regulator sites: specifically, they exhibit overall higher degree of conservation between human and rodents, stronger correlation with TFIID-bound promoters, and preference for permissive chromatin state.Namely, the correlation with c-Myc expression in promoters harboring dual-sites is increased for stronger sp1 sites by strong sp1 binding and the effect is largest in proliferating tissues.Our approach shows how integrated functional analyses can uncover tissue-specific and combinatorial regulatory dependencies in mammals.

View Article: PubMed Central - PubMed

Affiliation: Swiss Institute for Experimental Cancer Research (ISREC) and NCCR Molecular Oncology, Lausanne, Switzerland.

ABSTRACT
Combinatorial gene regulation largely contributes to phenotypic versatility in higher eukaryotes. Genome-wide chromatin immuno-precipitation (ChIP) combined with expression profiling can dissect regulatory circuits around transcriptional regulators. Here, we integrate tiling array measurements of DNA-binding sites for c-Myc, sp1, TFIID and modified histones with a tissue expression atlas to establish the functional correspondence between physical binding, promoter activity and transcriptional regulation. For this we develop SLM, a methodology to map c-Myc and sp1-binding sites and then classify sites as sp1-only, c-Myc-only or dual. Dual sites show several distinct features compared to the single regulator sites: specifically, they exhibit overall higher degree of conservation between human and rodents, stronger correlation with TFIID-bound promoters, and preference for permissive chromatin state. By applying regression models to an expression atlas we identified a functionally distinct signature for strong dual c-Myc/sp1 sites. Namely, the correlation with c-Myc expression in promoters harboring dual-sites is increased for stronger sp1 sites by strong sp1 binding and the effect is largest in proliferating tissues. Our approach shows how integrated functional analyses can uncover tissue-specific and combinatorial regulatory dependencies in mammals.

Show MeSH
Related in: MedlinePlus