Selection Of The Specific DNA Recognition Motif Of Transcription Factor AINTEGUMENTA

APETALA2/ethylene response element binding proteins (AP2/EREBP)form a plant-specific large transcription factor super family in Arabidopsis.Every member possesses genetically conserved DNA binding domain (AP2domian) of either single (for EREBP subfamily) or double (for AP2 subfamily).Through AP2 domain, these transcription factors recognize specific cis-actingelement(s) located at the 5'-up stream region of the corresponding target gene(s),conducting a variety of gene regulation events relating to plant growth anddevelopment and, in responses to biotic and abiotic stresses. Proteins with asingle AP2 domain include EREBPs (or later named ERFs), CBFs/DREBs(C-repeat/dehydration response element binding proteins), ABI4, TINY,RAV1and RAV2. The solution structure of AtERF1 AP2 domain bound to theGCC box has been determined. This protein domain folds into a novel DNArecognition structure consisting of a three-stranded β-sheet packed against an-helix with 6 or 7 invariant amino acid residues in the β-sheet primarilyresponsible for DNA recognition.AINTEGUMENTA (ANT) is one of the members in AP2 subfamily and hastwo AP2 domains, namely Repeat1 (R1) and Repeat2 (R2), of an approximately70 amino acids for each repeat. Other proteins in this subfamily also includeAP2 and others gene products. It was reported previously that the dual AP2domain of ANT, i.e. R1+R2 was capable of binding specifically a sequenceconsensus of 5'-gCAC(A/G)N(A/T)TcCC(a/g)ANG(c/t)-3'. We noticed that thisconsensus sequence is different from the sequence of EREBP subfamily. It lacksthe core sequence like GCC box and, indeed, it lacks the 6 or 7 invariantDNA-contacting amino acids found in the single AP2 domain of AtERF1through sequence as conferring a loss of function phenotype in yeast. Thougheach AP2 domain of ANT exhibits similarity to that of AtERF1, little is knownabout the interactions between members of the AP2 subfamily of proteins andDNA. It is presumed that the three-dimensional structure of ANT AP2 domainsmay arise from a different distinct mode of DNA recognition. Structural studiesare needed to address the three-dimensional structure of the AP2 domain.To this end, we utilized SELEX approach for in vitro DNA binding siteselection to explore the specific DNA binding sites of either R2 or R1+R2 repeatof ANT. In in vitro selection, we used a randomized oligonucleotides sequencelibrary5'-CTTAGCGTGATGCATATGAAT(N)18GCTAAGGATCCTAATGTCGTT-3'.?Sequences bound by R1+R2 or R2 were recovered, amplified by PCR andsubjected to additional rounds of selection. After 5~8 rounds selection, PCRproduct were labeled and subjected to gel mobility shift assay. The results of gelmobility shift assay revealed that the PCR product from R1+R2 group couldform a special protein-DNA complex in the EMSA, while the PCR product fromR2 group could not be specially recognized by R2 or R1+R2. It is indicated thatANT may recognize its target DNA by using the dual AP2 domains and the AP2domain R2 of ANT may have not the capability to bind DNA as a independentmonomer.