Limits...
A bacterial antirepressor with SH3 domain topology mimics operator DNA in sequestering the repressor DNA recognition helix.

León E, Navarro-Avilés G, Santiveri CM, Flores-Flores C, Rico M, González C, Murillo FJ, Elías-Arnanz M, Jiménez MA, Padmanabhan S - Nucleic Acids Res. (2010)

Bottom Line: Direct targeting of critical DNA-binding elements of a repressor by its cognate antirepressor is an effective means to sequester the repressor and remove a transcription initiation block.CarA and CarH repress the carB operon in the dark.Our findings uncover an unprecedented use of the SH3 domain-like fold for protein-protein recognition whereby an antirepressor mimics operator DNA in sequestering the repressor DNA recognition helix to activate transcription.

View Article: PubMed Central - PubMed

Affiliation: Instituto de Química-Física Rocasolano, Consejo Superior de Investigaciones Científicas (CSIC), Madrid, Spain.

ABSTRACT
Direct targeting of critical DNA-binding elements of a repressor by its cognate antirepressor is an effective means to sequester the repressor and remove a transcription initiation block. Structural descriptions for this, though often proposed for bacterial and phage repressor-antirepressor systems, are unavailable. Here, we describe the structural and functional basis of how the Myxococcus xanthus CarS antirepressor recognizes and neutralizes its cognate repressors to turn on a photo-inducible promoter. CarA and CarH repress the carB operon in the dark. CarS, produced in the light, physically interacts with the MerR-type winged-helix DNA-binding domain of these repressors leading to activation of carB. The NMR structure of CarS1, a functional CarS variant, reveals a five-stranded, antiparallel beta-sheet fold resembling SH3 domains, protein-protein interaction modules prevalent in eukaryotes but rare in prokaryotes. NMR studies and analysis of site-directed mutants in vivo and in vitro unveil a solvent-exposed hydrophobic pocket lined by acidic residues in CarS, where the CarA DNA recognition helix docks with high affinity in an atypical ligand-recognition mode for SH3 domains. Our findings uncover an unprecedented use of the SH3 domain-like fold for protein-protein recognition whereby an antirepressor mimics operator DNA in sequestering the repressor DNA recognition helix to activate transcription.

Show MeSH

Related in: MedlinePlus

CarS residues interacting with CarA identified by site-directed mutagenesis. (A) Yeast two-hybrid analysis of the interactions of each CarS mutant (CarS*) with CarA on glucose plates. C+: LexA–CarA and B42–CarS; C−: LexA–CarA only; others: LexA–CarA and the indicated B42–CarS*. (B) Color phenotypes for carotenogenesis in M. xanthus strains bearing the indicated carS allele. C+: wild-type; C−: ΔcarS; others are derived from introducing the indicated allele into the ΔcarS strain. (C) Elution profiles off a Superdex-200 analytical gel filtration column for the pure proteins (bottom) and CarANt mixed with CarS or CarS* (top). Dashed line is for pure CarANt; black, blue and red lines are for CarS, CarS* (F77A) and CarS* (D52A/D53A), respectively, with or without CarANt. Mr (in kDa) for each peak maximum is shown. (D) Top: schematic of the 130-bp EMSA probe spanning the carB promoter region, relative to the transcription start site (+1). The operator pI (–64 to −47) and pII (−40 to −26) sites are shown with their palindromic halves boxed. Bottom: EMSA for CarANt-binding to the 130-bp probe alone (lanes 2–5) or with increasing levels of CarS or CarS* added, as indicated.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2926617&req=5

Figure 2: CarS residues interacting with CarA identified by site-directed mutagenesis. (A) Yeast two-hybrid analysis of the interactions of each CarS mutant (CarS*) with CarA on glucose plates. C+: LexA–CarA and B42–CarS; C−: LexA–CarA only; others: LexA–CarA and the indicated B42–CarS*. (B) Color phenotypes for carotenogenesis in M. xanthus strains bearing the indicated carS allele. C+: wild-type; C−: ΔcarS; others are derived from introducing the indicated allele into the ΔcarS strain. (C) Elution profiles off a Superdex-200 analytical gel filtration column for the pure proteins (bottom) and CarANt mixed with CarS or CarS* (top). Dashed line is for pure CarANt; black, blue and red lines are for CarS, CarS* (F77A) and CarS* (D52A/D53A), respectively, with or without CarANt. Mr (in kDa) for each peak maximum is shown. (D) Top: schematic of the 130-bp EMSA probe spanning the carB promoter region, relative to the transcription start site (+1). The operator pI (–64 to −47) and pII (−40 to −26) sites are shown with their palindromic halves boxed. Bottom: EMSA for CarANt-binding to the 130-bp probe alone (lanes 2–5) or with increasing levels of CarS or CarS* added, as indicated.

Mentions: First, we monitored the effects of these CarS mutations on interactions with CarA by yeast two-hybrid analysis. In the system employed, the LexA DNA-binding domain fused to the N-terminus of CarA (LexA–CarA) served as ‘bait’, and CarS or any of its mutant forms (CarS*) fused to the C-terminus of the B42 transcriptional activation domain as ‘prey’. The B42–CarS hybrids were thus expressed from the GAL1 promoter, which is strongly activated by galactose and repressed by glucose. Physical interaction activates the reporter lacZ gene supplied in the yeast host employed. Control cells producing only one hybrid protein remained white on galactose plates even 24 h after the X-gal overlay. By contrast, cells producing LexA–CarA and B42–CarS turned blue within two hours, as also did cells expressing LexA–CarA and the other B42–CarS* fusions (Supplementary Figure S2A). Hence, under these conditions, no differences in interactions with CarA could be discerned between wild-type and any of the CarS* forms. However, on glucose plates, where the B42-fusions are expressed at basal levels, relative to the blue color of cells expressing LexA–CarA and B42–CarS, those expressing LexA–CarA and B42 fusions to the D52A/D53A, E79A/E80A, L63A, L75A or F77A CarS* were either white (like the negative control) or a very pale blue even after a prolonged (7–24 h) incubation with X-gal (Figure 2A). These CarS* are thus likely to be impaired in interactions with CarA. If this were indeed the case, introducing these mutations in M. xanthus could result in loss of light-induced carotenogenesis. Hence, this was examined next.Figure 2.


A bacterial antirepressor with SH3 domain topology mimics operator DNA in sequestering the repressor DNA recognition helix.

León E, Navarro-Avilés G, Santiveri CM, Flores-Flores C, Rico M, González C, Murillo FJ, Elías-Arnanz M, Jiménez MA, Padmanabhan S - Nucleic Acids Res. (2010)

CarS residues interacting with CarA identified by site-directed mutagenesis. (A) Yeast two-hybrid analysis of the interactions of each CarS mutant (CarS*) with CarA on glucose plates. C+: LexA–CarA and B42–CarS; C−: LexA–CarA only; others: LexA–CarA and the indicated B42–CarS*. (B) Color phenotypes for carotenogenesis in M. xanthus strains bearing the indicated carS allele. C+: wild-type; C−: ΔcarS; others are derived from introducing the indicated allele into the ΔcarS strain. (C) Elution profiles off a Superdex-200 analytical gel filtration column for the pure proteins (bottom) and CarANt mixed with CarS or CarS* (top). Dashed line is for pure CarANt; black, blue and red lines are for CarS, CarS* (F77A) and CarS* (D52A/D53A), respectively, with or without CarANt. Mr (in kDa) for each peak maximum is shown. (D) Top: schematic of the 130-bp EMSA probe spanning the carB promoter region, relative to the transcription start site (+1). The operator pI (–64 to −47) and pII (−40 to −26) sites are shown with their palindromic halves boxed. Bottom: EMSA for CarANt-binding to the 130-bp probe alone (lanes 2–5) or with increasing levels of CarS or CarS* added, as indicated.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2926617&req=5

Figure 2: CarS residues interacting with CarA identified by site-directed mutagenesis. (A) Yeast two-hybrid analysis of the interactions of each CarS mutant (CarS*) with CarA on glucose plates. C+: LexA–CarA and B42–CarS; C−: LexA–CarA only; others: LexA–CarA and the indicated B42–CarS*. (B) Color phenotypes for carotenogenesis in M. xanthus strains bearing the indicated carS allele. C+: wild-type; C−: ΔcarS; others are derived from introducing the indicated allele into the ΔcarS strain. (C) Elution profiles off a Superdex-200 analytical gel filtration column for the pure proteins (bottom) and CarANt mixed with CarS or CarS* (top). Dashed line is for pure CarANt; black, blue and red lines are for CarS, CarS* (F77A) and CarS* (D52A/D53A), respectively, with or without CarANt. Mr (in kDa) for each peak maximum is shown. (D) Top: schematic of the 130-bp EMSA probe spanning the carB promoter region, relative to the transcription start site (+1). The operator pI (–64 to −47) and pII (−40 to −26) sites are shown with their palindromic halves boxed. Bottom: EMSA for CarANt-binding to the 130-bp probe alone (lanes 2–5) or with increasing levels of CarS or CarS* added, as indicated.
Mentions: First, we monitored the effects of these CarS mutations on interactions with CarA by yeast two-hybrid analysis. In the system employed, the LexA DNA-binding domain fused to the N-terminus of CarA (LexA–CarA) served as ‘bait’, and CarS or any of its mutant forms (CarS*) fused to the C-terminus of the B42 transcriptional activation domain as ‘prey’. The B42–CarS hybrids were thus expressed from the GAL1 promoter, which is strongly activated by galactose and repressed by glucose. Physical interaction activates the reporter lacZ gene supplied in the yeast host employed. Control cells producing only one hybrid protein remained white on galactose plates even 24 h after the X-gal overlay. By contrast, cells producing LexA–CarA and B42–CarS turned blue within two hours, as also did cells expressing LexA–CarA and the other B42–CarS* fusions (Supplementary Figure S2A). Hence, under these conditions, no differences in interactions with CarA could be discerned between wild-type and any of the CarS* forms. However, on glucose plates, where the B42-fusions are expressed at basal levels, relative to the blue color of cells expressing LexA–CarA and B42–CarS, those expressing LexA–CarA and B42 fusions to the D52A/D53A, E79A/E80A, L63A, L75A or F77A CarS* were either white (like the negative control) or a very pale blue even after a prolonged (7–24 h) incubation with X-gal (Figure 2A). These CarS* are thus likely to be impaired in interactions with CarA. If this were indeed the case, introducing these mutations in M. xanthus could result in loss of light-induced carotenogenesis. Hence, this was examined next.Figure 2.

Bottom Line: Direct targeting of critical DNA-binding elements of a repressor by its cognate antirepressor is an effective means to sequester the repressor and remove a transcription initiation block.CarA and CarH repress the carB operon in the dark.Our findings uncover an unprecedented use of the SH3 domain-like fold for protein-protein recognition whereby an antirepressor mimics operator DNA in sequestering the repressor DNA recognition helix to activate transcription.

View Article: PubMed Central - PubMed

Affiliation: Instituto de Química-Física Rocasolano, Consejo Superior de Investigaciones Científicas (CSIC), Madrid, Spain.

ABSTRACT
Direct targeting of critical DNA-binding elements of a repressor by its cognate antirepressor is an effective means to sequester the repressor and remove a transcription initiation block. Structural descriptions for this, though often proposed for bacterial and phage repressor-antirepressor systems, are unavailable. Here, we describe the structural and functional basis of how the Myxococcus xanthus CarS antirepressor recognizes and neutralizes its cognate repressors to turn on a photo-inducible promoter. CarA and CarH repress the carB operon in the dark. CarS, produced in the light, physically interacts with the MerR-type winged-helix DNA-binding domain of these repressors leading to activation of carB. The NMR structure of CarS1, a functional CarS variant, reveals a five-stranded, antiparallel beta-sheet fold resembling SH3 domains, protein-protein interaction modules prevalent in eukaryotes but rare in prokaryotes. NMR studies and analysis of site-directed mutants in vivo and in vitro unveil a solvent-exposed hydrophobic pocket lined by acidic residues in CarS, where the CarA DNA recognition helix docks with high affinity in an atypical ligand-recognition mode for SH3 domains. Our findings uncover an unprecedented use of the SH3 domain-like fold for protein-protein recognition whereby an antirepressor mimics operator DNA in sequestering the repressor DNA recognition helix to activate transcription.

Show MeSH
Related in: MedlinePlus