Arabidopsis calcium-dependent protein kinase AtCPK32 interacts with ABF4, a transcriptional regulator of abscisic acid-responsive gene expression, and modulates its activity

The phytohormone abscisic acid (ABA) regulates stress-responsive gene expression during vegetative growth. The ABA regulation of many genes is mediated by a subfamily of basic leucine zipper class transcription factors referred to as ABFs (i.e. ABF1-ABF4), whose transcriptional activity is induced by ABA. Here we show that a calcium-dependent protein kinase is involved in the ABA-dependent activation process. We carried out yeast two-hybrid screens to identify regulatory components of ABF4 function and isolated AtCPK32 as an ABF4-interacting protein. AtCPK32 has autophosphorylation activity and can phosphorylate ABF4 in vitro. Mutational analysis indicated that serine-110 of ABF4, which is highly conserved among ABF family members, may be phosphorylated by AtCPK32. The serine-110 residue is essential for ABF4-AtCPK32 interaction, and transient expression assay revealed that it is also required for the normal transcriptional function of ABF4. The expression patterns and subcellular localization of AtCPK32 are similar to those of ABF4. Furthermore, its overexpression affects both ABA sensitivity and the expression of a number of ABF4-regulated genes. Together, our data demonstrate that AtCPK32 is an ABA signaling component that regulates the ABA-responsive gene expression via ABF4.

The p53 mutation which abrogates trans-activation while maintaining its growth-suppression activity

The p53 tumor suppressor protein functions as a transcription factor. It has, however, been previously reported that some p53 mutants are able to suppress cell growth independent of their transcriptional activity [Kaneuchi et al., (1999)]. In order to investigate the correlations between the trans-activation and growth-suppressive functions of p53, we have analyzed five p53 mutants by CAT reporter assay, colony formation assay, and growth-rate analysis. Five p53 mutants [Oh et al., (2000)]--199stop (Gly-->stop), 240ile (Ser-->Ile), 250ala (Pro-->Ala), 285lys (Glu-->Lys), and 291asn (Lys-->Asn)--were cotransfected with a reporter construct containing a p53-responsive element and then tested for their trans-activational activity in p53-null Saos-2 cells. As a result of a change in the protein structure, trans-activational activity was negated in 199stop, 240ile, 285lys, and 291asn, while 250ala retained its activity. Colony formation assay revealed that mutants 240ile and 250ala retained their growth suppression, while 199stop, 285lys, and 291asn did not. To study the features of these proteins, a group of isogenic cell lines that express mutant forms of p53 was generated from HeLa cells, and their growth rate was then examined: one group, containing 199stop, 285lys, and 291asn, showed a rapid growth rate, similar to that of the original HeLa cells; the other group, containing 240ile and 250ala, however, exhibited a slow growth rate. In conclusion, mutant p53 240ile, which completely lost its trans-activational activity, nevertheless continued to exhibit its growth-suppressive activity. Further work is required to understand how 240ile is involved in growth suppression.

Identification of p53 gene mutations in breast cancers and their effects on transcriptional activation function

Somatic mutations in the p53 tumor suppressor gene are the most common genetic alterations found in human malignancies. In the present study, we studied 36 primary human breast carcinomas, using a polymerase chain reaction-single-strand conformation polymorphism (PCR-SSCP) and sequencing analysis of exons 2 through 9 for the presence of p53 gene mutations. Six of 36 (17%) breast cancers contained mutations within the core domain of the p53 protein responsible for sequence-specific DNA binding (codons 102-292); all 5 missense mutations clustered between codons 240 and 291 (codons 240, 243, 250, 285, and 291), whereas one nonsense mutation occurred at codon 199. By using recombinant PCR in vitro mutagenesis, we introduced point mutations at codons 199 from Gly to stop (gly199stop), 240 from Ser to Ile (ser240Ile), 250 from Pro to Ala (pro250ala), 285 from Glu to Lys (glu285lys), and 291 from Lys to Asn (lys291asn), and all the p53 sequences were subcloned into the CMVneoBam vector under the control of the cytomegalovirus (CMV) promoter. To test whether the mutants p53 were functionally wild-type (wt) or mutant, we transfected them to p53-null Saos-2 cells with a reporter plasmid containing a p53-responsive element, and performed chloramphenicol acetyltransferase (CAT) assay. Transient CAT assay for transcriptional activation revealed that one group, including gly199stop, ser240ile, glu285lys, and lys291asn, abolished the transcriptional activity, whereas the other group, including pro250ala, retained stronger transcriptional transactivation activity than that of wt p53.