Fig 1.
Analysis of redox-dependent PAN DNA-binding.
(A) Arabidopsis wild-type flowers are composed of four sepals, four petals, six stamens and two fused carpels. Flowers of pan mutants form three pentameric whorls with five sepals, five petals, five stamens and two normal carpels. Scale bar 1 mm. (B) Sequence of the palindromic core TGA DNA-binding motif is shown on the top and the three used binding sites comprising the core motif below. The as-1-like motif from the Arabidopsis PR1 promoter comprises two TGA binding sites. The AAGAAT motif from the second intron of AG encompasses one central TGA binding site with one nucleotide exchange, the name-giving 5’ AAGAAT motif is underlined. Mutagenesis of two nucleotides (indicated in red) abolishes binding to the ΔbZIP motif. (C) EMSA analysis of the PAN interaction with the AAGAAT, as-1-like and ΔbZIP motifs. Radioactive-labeled DNA probes were incubated with PAN protein or with a mock translation under reducing conditions (0.9 mM DTT). The arrow indicates DNA-protein complex formation and the asterisk marks free DNA probe. D) Analysis of redox-dependent binding of PAN to the AAGAAT motif. For comparison of reducing and oxidizing conditions, PAN protein was incubated prior to DNA-binding with 0.9 mM or 20 mM DTT and with 2 mM diamide, respectively. Reversibility of the redox-dependent DNA-binding was analyzed by the addition of 20 mM DTT after a 2 mM diamide treatment. The control reaction was conducted without protein addition.
Fig 2.
Redox-dependent binding of selected Arabidopsis TGA TFs to the AAGAAT motif.
(A) Cladogram of Arabidopsis TGA TFs indicating clade memberships. (B) Schematic representation of selected TGA TF family members and respective cysteine residue positions. Orange boxes mark the bZIP domain and blue boxes two glutamine-rich regions. Cys340, Cys260 and Cys301 (red) are localized in a conserved position in the first glutamine-rich region. (C) Gel retardation analysis of selected TGA proteins incubated with the AAGAAT probe under reducing (0.9 mM DTT; red) or oxidizing (2 mM diamide; ox) conditions. Asterisk marks unbound AAGAAT probe.
Fig 3.
Analysis of posttranslational PAN modifications.
(A) Non-reducing SDS-PAGE analysis of MBP-PAN and MBP-PAN6xCysmut proteins. Proteins were incubated with 20 mM DTT, 2 mM diamide or with 2 mM diamide followed by 20 mM DTT. Triangles indicate MBP-PAN (~90 kDa) and MBP-PAN6xCysmut (~90 kDa) proteins and the asterisk marks MBP (~43 kDa). Dimerized and oligomerized fusion proteins are labeled with full circles. (B) Non-reducing SDS-PAGE analysis for identification of posttranslational cysteine modifications. MBP-PAN protein was incubated under different conditions (20 mM DTT, 2 mM diamide, 10 mM GSSG and 5 mM GSH/5 mM H2O2) and interesting bands, marked with an asterisk, were further processed for MS analysis. (C) Posttranslational PAN cysteine modifications as determined by the difference between predicted, unmodified and detected MS mass values.
Fig 4.
Redox-dependent DNA-binding of PAN is mediated via the N-terminal cysteines.
Redox EMSA analyses to investigate the influence of single (A) and multiple Cys-to-Ser substitutions (B and C) and the effect of a N-terminal PAN truncation (D). Schematic illustrations of the PAN variants indicate the bZIP domain in orange and the glutamine-rich regions in blue boxes. Cys-to-Ser substitutions are labeled in red. PAN proteins were incubated in 0.9 mM DTT (red) or 2 mM diamide (ox) prior to addition of the AAGAAT motif. Asterisks mark unbound free probe.
Fig 5.
Evolution of the N-terminal PAN extension and the AAGAAT motif.
Cladogram of evolutionary informative species from seed plants indicating AAGAAT motifs and N-terminal extensions of the respective PAN and AG homologs. Orange boxes indicate presence of an AAGAAT motif comprising the AAGAAT sequence and core TGA binding site. Beige boxes label partial AAGAAT motifs, which possess only the AAGAAT sequence and lack the core TGA binding sequence TGACG. Schematic representation of PAN protein homologs shows N-terminal extensions with significant (grey bar) or low sequence homology (striped bar) to the PAN N-terminus or its absence. Presence (C) or absence (X) of cysteines equivalent to the five N-terminal PAN cysteines is depicted. Stars indicate that the definition of a magnoliid PAN homolog start codon was not possible and that no genome sequencing data were available.