Meis cofactors control HDAC and CBP accessibility at Hox-regulated promoters during zebrafish embryogenesis

Abstract

Hox proteins form complexes with Pbx and Meis cofactors to control gene expression, but the role of Meis is unclear. We demonstrate that Hoxb1-regulated promoters are highly acetylated on histone H4 (AcH4) and occupied by Hoxb1, Pbx, and Meis in zebrafish tissues where these promoters are active. Inhibition of Meis blocks gene expression and reduces AcH4 levels at these promoters, suggesting a role for Meis in maintaining AcH4. Within Hox transcription complexes, Meis binds directly to Pbx and we find that this binding displaces histone deacetylases (HDACs) from Hoxb1-regulated promoters in zebrafish embryos. Accordingly, Pbx mutants that cannot bind Meis act as repressors by recruiting HDACs and reducing AcH4 levels, while Pbx mutants that bind neither HDAC nor Meis are constitutively active and recruit CBP to increase AcH4 levels. We conclude that Meis acts, at least in part, by controlling access of HDAC and CBP to Hox-regulated promoters.

Figure 1

Meis proteins are required for histone H4 acetylation at Hox-regulated promoters. A. Diagram of hoxb1a and hoxb2a regulation by Hox, Pbx and Meis factors in zebrafish rhombomere 4 (PG1 = paralog group 1). B. Dissection of 14hpf zebrafish embryos (black line indicates position of cut) produces a posterior piece (Post) expressing hoxb1a (black shading) and an anterior piece (Ant) not expressing hoxb1a. C. ChIP analysis reveals that Pbx, Meis and Hoxb1a/b occupy the hoxb1a promoter in posterior (Post), but not anterior (Ant), tissues at 14hpf. D. ChIP analysis of zebrafish embryos reveals that AcH4 is higher in posterior than anterior tissues for the hoxb1a and hoxb2a promoters, but not for the pax2 promoter. E. Expression of a Meis dominant negative construct (PBCAB; right panel) blocks hoxb1a expression (dark blue stain) in zebrafish r4 (compare to control embryo; left panel). Panels show close-ups of the hindbrain with anterior to the top. F. ChIP analysis reveals that using a dominant negative construct (PBCAB) to interfere with Meis function reduces AcH4 at the hoxb1a and hoxb2a promoters, but not at the otx1 promoter, in zebrafish embryos. Data is expressed as the ratio of PBCAB-injected/control-injected embryos.

Figure 2

Meis proteins counteract HDAC activity at the hoxb1a promoter. A. A hoxb1a:luciferase reporter was co-transfected into HEK-293 cells with expression vectors as indicated below the graph. Data are reported as fold increase in luciferase activity over transfection with reporter construct alone and are normalized for transfection efficiency by inclusion of a renilla luciferase control plasmid. BMHoxb1b = Hoxb1b mutant that does not bind Pbx proteins; PBCAB = Meis dominant negative; BMMeis3 = Meis3 mutant that does not bind Pbx proteins; TSA = HDAC inhibitor Trichostatin A. B. Deletions of the Pbx4 protein generate constructs that bind HDAC but not Meis (ΔPBCA2), that partially bind HDAC but not Meis (ΔPBCA3) and that bind neither HDAC nor Meis (ΔPBCB5). GST pull-down data is shown in the bottom panel and a summary of the deletion constructs is shown in the top panel. C. A hoxb1a:luciferase reporter was co-transfected with expression vectors as indicated below the graph. Data is reported as fold increase in luciferase activity over transfection with reporter construct alone and is normalized for transfection efficiency by inclusion of a renilla luciferase control plasmid (ΔA2 = ΔPBCA2; ΔA3 = ΔPBCA3, ΔB5 = ΔPBCB5).

Figure 3

The ΔPBCA2 and ΔPBCB5 constructs modulate H4 acetylation and expression of Hox-regulated promoters in zebrafish embryos. A. Expression of the ΔPBCA2 construct has little effect on hoxb1a expression in wild type embryos, but reduces hoxb1a expression (dark blue stain) in r4 of lazarus (pbx4 mutant) embryos. Expression of a ΔPBCA2 mutant that cannot bind DNA (ΔA2*) has no effect. Panels show close-ups of the hindbrain with anterior to the top. B. ChIP analysis at 15hpf reveals reduced AcH4 at the hoxb1a and hoxb2a promoters, but not at a control promoter, in ΔPBCA2-injected embryos. Data is expressed as the ratio of ΔPBCA2-injected/control-injected embryos. C. ChIP analysis at 22hpf reveals increased AcH4 at the hoxb1a and hoxb2a promoters, but not at a control promoter, in ΔPBCB5-injected embryos. Data is expressed as the ratio of ΔPBCB5-injected/control-injected embryos. D. Expression of the ΔPBCB5 construct induces ectopic expression of hoxb1a and krox20 (white brackets) in zebrafish embryos. A ΔPBCB5 mutant (ΔB5*) that cannot bind DNA has no effect. Panels show flat-mounted embryos with anterior to the top. Dashed lines indicate where two images have been merged. E. Quantitative RT-PCR analysis reveals increased expression of hoxb1a and krox20, but not ODC1, in the tail region (white brackets in D) of ΔPBCB5-injected embryos. Data is presented as the ratio of ΔPBCB5-injected/control-injected embryos. F. ChIP analysis reveals increased AcH4 at the hoxb1a and krox20 promoters, but not a control promoter, in the tail region (white brackets in D) of ΔPBCB5-injected embryos. Data is expressed as the ratio of ΔPBCB5-injected/control-injected embryos.

Figure 4

Meis proteins compete with HDACs for binding to Pbx in vitro and modulate HDAC and CBP accessibility to Hox-regulated promoters in zebrafish embryos. A. Competitive co-immunoprecipitation experiment. HEK293 cells were transfected as indicated below the blots. After 48hrs, cells were lysed and either immunoprecipitated with anti-HA (to precipitate HA-Pbx4; lanes 1–4) or run as lysate (lanes 5–8), followed by Western blotting as indicated to the right. ΔNMeis3 = Meis3 mutant that cannot bind Pbx proteins. B. ChIP analysis reveals increased AcH4 at the hoxb1a and hoxb2a promoters, but not at the pax2 promoter, in Meis3-injected embryos (left panel). ΔNMeis3 has no effect on AcH4 levels (right panel). Data is expressed as the ratio of injected/uninjected embryos. C. ChIP analysis of zebrafish embryos reveals that Myc-HDAC1 (Myc-HDAC1 mRNA was injected to partially label the cellular HDAC1 pool) occupancy at the hoxb1a promoter is increased by ΔPBCA2, but decreased by Meis3 and ΔPBCB5. D. ChIP analysis reveals that CBP occupancy at the hoxb1a promoter in zebrafish embryos is increased by ΔPBCB5 and Meis3, but decreased by ΔPBCA2. E. GST pull-down experiments demonstrate that Pbx4, Hoxb1a and Hoxb1b, but not Myc-Meis3, bind CBP.