Control of progesterone receptor transcriptional synergy by SUMOylation and deSUMOylation

Abstract

Background: Covalent modification of nuclear receptors by the Small Ubiquitin-like Modifier (SUMO) is dynamically regulated by competing conjugation/deconjugation steps that modulate their overall transcriptional activity. SUMO conjugation of progesterone receptors (PRs) at the N-terminal lysine (K) 388 residue of PR-B is hormone-dependent and suppresses PR-dependent transcription. Mutation of the SUMOylation motif promotes transcriptional synergy.

Results: The present studies address mechanisms underlying this transcriptional synergy by using SUMOylation deficient PR mutants and PR specifically deSUMOylated by Sentrin-specific proteases (SENPs). We show that deSUMOylation of a small pool of receptors by catalytically competent SENPs globally modulates the cooperativity-driven transcriptional synergy between PR observed on exogenous promoters containing at least two progesterone-response elements (PRE2). This occurs in part by raising PR sensitivity to ligands. The C-terminal ligand binding domain of PR is required for the transcriptional stimulatory effects of N-terminal deSUMOylation, but neither a functional PR dimerization interface, nor a DNA binding domain exhibiting PR specificity, are required.

Conclusion: We conclude that direct and reversible SUMOylation of a minor PR protein subpopulation tightly controls the overall transcriptional activity of the receptors at complex synthetic promoters. Transcriptional synergism controlled by SENP-dependent PR deSUMOylation is dissociable from MAPK-catalyzed receptor phosphorylation, from SRC-1 coactivation and from recruitment of histone deacetylases to promoters. This will provide more information for targeting PR as a part of hormonal therapy of breast cancer. Taken together, these data demonstrate that the SUMOylation/deSUMOylation pathway is an interesting target for therapeutic treatment of breast cancer.

Figure 1

Modulation of PR transcriptional activity by SUMO-1 depends on the promoter context. A) schematic of PR-A and PR-B showing the location of hormone dependent phosphorylation sites; the Lys-388 SUMO conjugation site within an IKEE motif; and an acetylation consensus KxKK site (amino acids 638-641). BUS, B-upstream segment; DBD, DNA binding domain; LBD, ligand binding domain. B) HeLa cells were transfected with 2 μg of PRE2-Luc reporter, or C) MMTV-Luc, together with 5-1000 ng of wild-type PR-B or mutant PR-B K388R expression vectors and Renilla-Luc as an internal control. The cells were treated 24 hrs with 10 nM R5020, then harvested and lysed. The extracts were assayed for luciferase activities. Luciferase activity is expressed in relative light units (RLU). Data represent triplicates (± SD). Statistical significance was computed by unpaired student's t test. *p < 0.05.

Figure 2

SENP1 and SENP2 deSUMOylate PR-B and enhance their transcriptional activity. A) DeSUMOylation of PR-B by wt SENP1 and SENP2. HeLa cells were cotransfected with pSG5-PR-B, GFP-SUMO-1 and SENP1 or SENP2 as indicated. Cells were grown in the presence (+) or absence (-) of R5020. PR in cell extracts separated on SDS-PAGE, were detected with anti-PR 1294 monoclonal antibody. β-actin served as a loading control. B) HeLa cells were transfected with the PRE2-Luc reporter plasmid in the presence of pSV40-Renilla as internal control along with PR-B and increasing amounts (50-1000 ng) of SENP1, SENP1 mutant, or SENP2 expression vectors, or an empty vector control (-). Cells were treated without (-) or with (+) 10 nM R5020 for 24 hrs before being assayed for luciferase activity. The relative luciferase activity of wt PR-B in the presence of 10 nM R5020 is set as 100%.

Figure 3

The enhancement of PR transcriptional activity by SENP1 depends on an intact SUMO conjugation site. HeLa cells were transfected with the PRE2-Luc (A and B) or MMTV-Luc (C and D) reporter plasmids in the presence of pSV40-Renilla as internal control along with PR-B (A and C) or PR-B K388R (B and D) expressing vectors, and a Flag-SENP1 expression vector at doses of 20, 50, 100, 200 and 1000 ng of DNA or an empty vector control (-). Cells were treated without (-) or with (+) 10 nM R5020 for 24 hrs before being assayed for luciferase activity as in Figure 1. RLU of wt PR-B in the absence of hormone is set as 1. Statistical significance was computed by unpaired student's t test. *p < 0.05.

Figure 4

The enhancement effect of SENP-1 on PR transcriptional activity requires full-length PR-B but not the PR DBD. HeLa cells were transfected with the PRE2-Luc (A and B) or ERE2-Luc (C and D) reporter plasmids in the presence of pSV40-Renilla as internal control along with NT-B (A), PR DBD-LBD (B), a PR-B specificity mutant containing the ER DBD (C) or wild-type ER (D) expression vectors, and SENP1 or SENP1m (A, right panel) expression vectors at doses of 20, 50, 100, 200 and 1000 ng of DNA or an empty vector control (-). Cells were treated without (-) or with (+) 10 nM R5020 (A and C) or 1 nM 17β-estradiol (E2) (D) for 24 hrs before being assayed for luciferase activity. The values are expressed as relative luciferase units normalized to Renilla controls. Statistical significance was computed by unpaired student's t test. *p < 0.05.

Figure 5

Overexpression of SENP1 raises PR-B sensitivity to hormone. HeLa cells were transfected with 50 ng of the PR-B (A) or PR-B K388R (C) expression vectors, a PRE2-luc, and Renilla control plasmid in the presence or absence of 100 ng of SENP1 expression vector and treated with ethanol or various concentrations of R5020 for 24 h. The average was plotted as a percentage of the maximal induction by 10 nM R5020 above no hormone levels. Curve fitting was performed by Prism Graph as described under "Experimental Procedures". The S.D. of triplicate values is indicated by the error bars. The corresponding relative luciferase activities were plotted for PR-B (B) and PR-B K388R (D). Statistical significance was computed by unpaired student's t test. *p < 0.05.

Figure 6

The stimulatory effect of MEKK1 on PR-B transcriptional activity is independent of the SUMO conjugation site. HeLa cells were transfected with 2 μg of PRE2-luciferase reporters together with 50 ng of wild type PR-B (A), PR-B S294/345 phosphorylation mutant (B), or PR-B K388R SUMOylation deficient mutant (C) expression vectors and Renilla-Luc as an internal control in the presence or absence of 100 ng SENP1 and/or constitutively active MEKK1 expression vectors. The cells were treated for 24 hrs with the agonist R5020 (10 nM) then harvested and lysed. The extracts were assayed for luciferase activities as in Figure 1.

Figure 7

SRC-1 reverses the inhibitory effect of the HDAC inhibitor TSA on the PR transcriptional activity. A) HeLa cells were transfected with 2 μg of PRE2-luciferase reporters together with 50 ng of a PR-B and Renilla-Luc as an internal control. The cells were treated for 24 hrs with the agonist R5020 (10 nM), without (-) or with increasing amounts of trichostatin A (TSA). B & C) HeLa cells were transfected with 2 μg of PRE2-luciferase reporters together with 50 ng of a PR-B expression vector and Renilla-Luc as an internal control in the absence or the presence (+) of increasing amount of SRC1. The cells were treated for 24 hrs with the agonist R5020 (10 nM), without (-) or with (+) 100 nM (B) or 500 nM (C) of TSA then harvested and lysed. The extracts were assayed for luciferase activities as in Figure 1. (*) Compared with control and (+) compared with TSA treatment.

Figure 8

HDACs are not a major target for SENP1 action on PR transcriptional activity. A) TSA enhances PR-B protein stability. HeLa cells were transiently transfected with expression vectors encoding wild type PR-B. Cells were treated 24 hrs without (-) or with (+) 10 nM R5020 in the presence of increasing concentration of TSA. Western blot analysis was performed on cell extracts probed with the anti-PR1294 monoclonal and anti β-actin antibodies. B) HeLa cells were transfected with 2 μg of PRE2-luciferase reporters together with 50 ng of a PR-B (left), or the PR-B K388R mutant (right) expression vectors and Renilla-Luc as an internal control in the presence or absence of 100 ng SENP1 expression vectors. The cells were treated for 24 hrs with the agonist R5020 (10 nM), without (-) or with (+) 100 nM TSA then harvested and lysed. The extracts were assayed for luciferase activities as in Figure 1. Statistical significance was computed by unpaired student's t test. *p < 0.05.