ZmMADS47 Regulates Zein Gene Transcription through Interaction with Opaque2

Abstract

Zeins, the predominent storage proteins in maize endosperm, are encoded by multiple genes and gene families. However, only a few transcriptional factors for zein gene regulation have been functionally characterized. In this study, a MADS-box protein, namely ZmMADS47, was identified as an Opaque2 (O2) interacting protein via yeast two-hybrid screening. The N-terminal portion of ZmMADS47 contains a nuclear localization signal (NLS), and its C-terminal portion contains a transcriptional activation domain (AD). Interestingly, the transcriptional activation activity is blocked in its full length form, suggesting conformational regulation of the AD. Molecular and RNA-seq analyses of ZmMADS47 RNAi lines revealed down regulation of α-zein and 50-kD γ-zein genes. ZmMADS47 binds the CATGT motif in promoters of these zein genes, but ZmMADS47 alone is not able to transactivate the promoters. However, when both O2 and ZmMADS47 are present, the transactivation of these promoters was greatly enhanced. This enhancement was dependent on the AD function of ZmMADS47 and the interaction between ZmMADS47 and O2, but it was independent from the AD function of O2. Therefore, it appears interaction with O2 activates ZmMADS47 on zein gene promoters.

Fig 1. Protein interaction between ZmMADS47 and O2.

A. Yeast two-hybrid analysis of the interaction between O2-2 (O2-bZIP) and a MADS protein. AD: GAL4 activation domain; BD: GAL4 DNA binding domain; O2 (bZIP): the O2-2 bZIP fragment that does not contain the transactivation domain and C terminal. B. GST pull-down analysis of ZmMADS47 and O2. The pull-down sample was detected using GST antibody (Abcam) and His antibody (Abcam). GST and GST-ZmMADS47 were used as bait and incubated with 6XHis fused O2 protein. After incubation, GST and GST-ZmMADS47 were both eluted with glutathione. IB: immunoblot. C. Co-immunoprecipitation assay to confirm O2-ZmMADS47 interaction. Immatured kernel extracts were incubated with O2 antibody and precipitated by Protein A Sepharose Beads (GE Healthcare). Total protein and IPed protein immunoprecipitated by O2 antibody (purified from rabbit serum) were blotted by ZmMADS47 and O2 (purified from guinea pig serum) antibodies. D. O2 complex seperated by gel filtration and detected by O2 and ZmMADS47 antibodies. The 500 μl immatured kernel extracts were extracted by the protocol used for coIP and injected into a Superdex 200 10/300 GL (GE Healthcare) Column. The protein eluate was collected at 0.4 ml/min speed. T: total extract.

Fig 2. Sub-cellular localization and transactivation identification of ZmMADS47.

A. Schematic representation of multiple truncated ZmMADS47 constructions. B. Fluorescence signal resulting from expression of ZmMADS47 (FL)-CFP, ZmMADS47 (MADS)-CFP and CFP alone. Signal from CFP, DIC (differential-interference microscope) and merging the two signals are shown in these panels. Bar represents 50μm. C. The β-galactosidase activity resulting from the TF transactivation. O2 was used as a positive control, while the pGBK-T7 vector alone was used as a negative control. Error bars represent SD (n = 3) (***P < 0.001, Student’s t test).

Fig 3. Zein content in ZmMADS47 RNAi transgenic lines.

A. Real-time PCR assay to detect zein transcripts in RNAi lines 3 and 6 in. ZmUBQ was used as an internal control. All the assays were repeated three times. Error bars represent SD (n = 3) (*P < 0.05, Student’s t test). B. Comparison of zein, nonzein, and total proteins from wild type and RNAi lines 3 and 6 mature endosperm. The measurements were done on per mg of dried endosperm. Error bars represent SD (n = 3) (*P < 0.05, Student’s t test). C. SDS-PAGE detection of zein accumulation in RNAi lines 3 and 6. D. Immunoblot detection of zein proteins in RNAi lines 3and 6 with specific zein antibodies. E. Cross section of segregating wild type and RNAi mutant kernels from ears of lines 3 and 6 at maturity.

Fig 4. Transmission electron microscopy of wild type, ZmMADS47 RNAi, o2, and ZmMADS47 RNAi; o2 starchy endosperm.

A-D. Protein bodies in the fourth starchy endosperm cell layer at 18DAP stage were observed by transmission electron microscopy. Each genotype is labeled above the corresponding TEM image. PB: protein body; CW: cell wall; SG, starch granule. Bars = 1μm. E. Quantitative comparison of PB number and size in the fourth and fifth cell layer from aleurone between wild type, ZmMADS47 RNAi, o2, and ZmMADS47 RNAi;o2 endosperm. Error bars represent SD (***P < 0.001, Student’s t test).

Fig 5. Characterization of the ZmMADS47 binding motif.

A. Schematic representation of z1A promoter fragmentation. Black arrow points to the ZmMADS47-indused shift bands. B. Deletion analysis to narrow down the conserved domain recognized by ZmMADS47. The upper black arrow identifies the shifted bands due to the ZmMADS47/DNA complex. The bases highlighted by red letters represent the sharing sequences in both two fragments. C. Mutagenesis assay of CATGT motifs. The upper black arrow identified the shifted bands due to the ZmMADS47/DNA complex. The bases highlighted by red letters represent the mutational bases compared with wild-type bases. D. Effect of competitive probe on the band shift reaction. The upper black arrow identifies the shifted bands due to ZmMADS47/DNA complex. E. The binding ability of His-ZmMADS47 and His-O2 to CATGT motifs in the z1A promoter and 50-kD zein promoter. The experiment was performed by adding ZmMADS47 or O2 protein. The upper black arrow identifies shifted bands due to ZmMADS47/DNA complex or O2/DNA complex. F. Schematic representation of ZmMADS47 binding site (M) and O2 DNA binding site (O) in down-regulated zein promoters.

Fig 6. Relative transactivation of ZmMADS47 to different zein genes in onion cells.

LUC/REN indicates the ratio of the firefly luciferase activity and the renilla reniformis activity. Error bars represent SD (n = 3) (*P < 0.05, **P < 0.01, ***P < 0.001, Student’s t test). NC respersent to those with the reporter genes (LUC/REN) alone.

Fig 7. Transactivation ratio of O2 and ZmMADS47.

A. Schematic representation of AD-truncated O2 construction. LUC/REN indicates the ratio of the firefly luciferase activity and the renilla reniformis activity. Error bars represent SD (n = 6) (*P < 0.05, **P < 0.01, ***P < 0.001, Student’s t test). B. Schematic representation of the 50–2 motif deletion in 50-kD γ-zein promoter. LUC/REN indicates the ratio of the firefly luciferase activity and the renilla reniformis activity. Error bars represent SD (n = 6) (*P < 0.05, **P < 0.01, ***P < 0.001, Student’s t test). C. Schematic representation of seven amino acid deletion at the C-terminus of ZmMADS47. LUC/REN indicate the ratio of the firefly luciferase activity and the renilla reniformis activity. Error bars represent SD (n = 6) (*P < 0.05, **P < 0.01, ***P < 0.001, Student’s t test).

Fig 8. Illustrtion of how the O2/ZmMADS47 complex may regulate zein gene expression.

The upper illustration shows ZmMADS47 alone (square) binds the CATGT motif but does not induce transactivation. The middle illustration shows O2 binding is able to transactivate its targets. The bottom illustration shows ZmMADS47 (oval) and O2 bind their respective motifs and the O2/ZmMADS47 complex enhances transactivation of gene targets.