Identification Of Transcription Factors Controlling The Expression Of BAS1 And Functional Analysis Of Four Alternative Spliced Isoforms Of SERK4

Brassinosteroids(BRs) are essential steroidal hormones for normal plant growth and development. Since BRs cannot be transported between plant tissues, their cellular levels must be controlled for proper developmental fates. BR homeostasis is maintained partially by a transcriptional feedback regulation loop that controls the expression of genes encoding for key metabolic enzymes, including the BR inactivating enzymes BAS1( CYP734A1, formerly CYP72B1), SOB7(CYP72C1),UGT73C5 and UGT73C6. Using a bioinformatic method, we identified several transcription factors which may be involved in BR homeostasis maintenance. To reveal their importance in regulating BAS1 expression, we generated a pBAS1::GUS transgenic line and then analyzed the expression of GUS in plant tissues including seedings, rosette leaves, inflorescence and siliques. We overexpressed these transcription factors in Col-0 and found two transcription factors, HAT1 and HAT3, can regulate the expression of BAS1. Compared to Col-0, 35S::HAT1-GFP and 35S::HAT3-GFP transgenic plants displayed a phenotype with dark green leaves and dwarfed statues, which was similar to BR deficiency mutants. When overexpressing HAT1 and HAT3 in pBAS1::GUS, we found that the expression of BAS1 was down regulated, which was consistent with the phenotype of the transgenic plants in Col-0.BAK1(SERK3)plays a key role in BR signaling pathway, but its closest homologous protein BKK1(SERK4)plays a redundant role with BAK1. In Arabidopsis, many genes have several isoforms generated by alternative splicing. It is important to study their functions. Searching sequences from RNA-seq data, we found SERK4 have four isoforms, and the presence of four different isoforms was verified by RT-PCRs. To determine their biological functions, we analyzed the expression levels of the four isoforms in different tissues and under different treatments, such as hormones, tempreture and salt. The results showed no difference in the various tissues or under those treatments. When overexpressing those four isoforms in bri1-5 or serk3 serk4, only the second isoform could suppress the deficient phenotypes of bri1-5 and the cell death phenotype of serk3 serk4. These results indicate that only the second isoform of SERK4 is functional in BR signaling and cell-death control pathways. The real roles of these isoforms need to be investigated in the near future.