Functional regulation of Q by microRNA172 and transcriptional co-repressor TOPLESS in controlling bread wheat spikelet density

Abstract

Bread wheat (Triticum aestivum) spike architecture is an important agronomic trait. The Q gene plays a key role in the domestication of bread wheat spike architecture. However, the regulatory mechanisms of Q expression and transcriptional activity remain largely unknown. In this study, we show that overexpression of bread wheat tae-miR172 caused a speltoid-like spike phenotype, reminiscent of that in wheat plants with the q gene. The reduction in Q transcript levels in the tae-miR172 overexpression transgenic bread wheat lines suggests that the Q expression can be suppressed by tae-miR172 in bread wheat. Indeed, our RACE analyses confirmed that the Q mRNA is targeted by tae-miR172 for cleavage. According to our analyses, the Q protein is localized in nucleus and confers transcriptional repression activity. Meanwhile, the Q protein could physically interact with the bread wheat transcriptional co-repressor TOPLESS (TaTPL). Specifically, the N-terminal ethylene-responsive element binding factor-associated amphiphilic repression (EAR) (LDLNVE) motif but not the C-terminal EAR (LDLDLR) motif of Q protein mediates its interaction with the CTLH motif of TaTPL. Moreover, we show that the N-terminal EAR motif of Q protein is also essentially required for the transcriptional repression activity of Q protein. Taken together, we reveal the functional regulation of Q protein by tae-miR172 and transcriptional co-repressor TaTPL in controlling the bread wheat spike architecture.

Keywords: Q gene; miR172; spike architecture; transcriptional repression; wheat.

Figure 1

Identification of the tae‐MIR172 precursors in bread wheat. (a) The secondary structures of tae‐MIR172 precursors from the wheat A‐genome progenitor Triticum urartu, the bread wheat (T. aestivum) cultivar Chinese Spring (CS, chromosomes 1A, 1B and 1D) and cultivar Kenong199 (KN199, chromosome 1B). The mature tae‐miR172 sequences are highlighted in red. ΔG describes the minimum free energy (mfe) of the RNA structure. (b) Sequences of mature miR172 from different plant species. The single nucleotide variants were highlighted in red. tae, T. aestivum; osa, Oryza sativa; zma, Zea mays; bdi, Brachypodium distachyon; ath, Arabidopsis thaliana; gma, Glycine max.

Figure 2

Over accumulation of tae‐miR172 leads to the speltoid spike phenotype in bread wheat. (a) Whole plant view of wild‐type (WT) KN199 (left) and pUbi:tae‐MIR172 transgenic plants (right) at grain filling stage. Bar = 10 cm. (b) Quantification of tae‐miR172 in WT KN199 and pUbi:tae‐MIR172 transgenic plants by stem‐loop quantitative reverse transcription PCR (qRT‐PCR) at the heading stage. The accumulation levels of tae‐miR172 were normalized against TaU6. Error bars represent standard deviations (SDs) among three independent replicates. (c) Spike morphology of KN199 (left) and pUbi:tae‐MIR172 (right) at grain filling stage. Bar = 2 cm. (d) Spikelet densities (the number of spikelets per centimetre of spike length) of WT KN199 and pUbi:tae‐MIR172 transgenic plants at the mature stage. Error bars denote SDs (n = 5). #1, #3, #4 and #5 represent independent transgenic lines; **P < 0.01 (Student's t‐test).

Figure 3

Q is a target gene of tae‐miR172 in bread wheat. (a) Identification of cleavage site in Q by 5′ RACE. The red arrow indicates the cleavage site, and numbers below the arrow show the frequency of clones with matching 5′RACE products from this site out of total clones identified by sequencing. mQ , tae‐miR172‐cleavage resistant form of Q. (b) Determination of Q transcript levels by qRT‐PCR in WT KN199 and pUbi:tae‐MIR172 lines at heading stage. The flag leaves were collected for the determination, and the transcript levels of Q were normalized against the internal control gene TaGAPDH . Error bars represent SDs among three independent replicates. **P < 0.01 (Student's t‐test). (c) Transient expression assay in N. benthamiana confirming that Q is a target gene of tae‐miR172. 35S:Q‐Myc and 35S:mQ‐Myc were separately co‐expressed with 35S:tae‐MIR172 in N. benthamiana leaves, and the protein levels of Q‐Myc and actin were determined by Western blotting using α‐Myc or α‐Actin antibodies. Numbers below the blots represent relative protein levels as calculated by Image J software. In each experiment, four independent leaves were analysed, and three replicates were performed with similar results.

Figure 4

Q encodes an AP2 family transcription factor with transcription repression activity. (a) Schematic diagram of the domain structure of Q protein. The EAR (ethylene‐responsive element binding factor‐associated amphiphilic repression) motifs located at the N‐ or C‐terminus of Q were annotated as EAR1 (LDLNVE) and EAR2 (LDLDLR), respectively. (b) Subcellular localization of Q. The 35S:Q‐GFP was expressed in N. benthamiana leaves by Agrobacterium‐mediated infiltration. GFP signal was detected 48 h post infiltration (hpi). In each experiment, four leaves were analysed. Three replicates were performed independently with similar results. BF, bright field. Scale bars, 20 μm. (c) and (d) Transient expression assay in N. benthamiana protoplasts illustrating the transcriptional repression activity of Q. The reporters and effectors used in the assay were generated as shown in (c). The activities of firefly luciferase (LUC) and renilla luciferase (REN) were determined 16 h post‐transformation. The relative luciferase activities in control and Q‐expressed samples as shown in (d) were calculated by normalizing the LUC values against REN. Error bars indicate SDs among three independent replicates. **P < 0.01 (Student's t‐test).

Figure 5

Identification of TaTPL in bread wheat. (a) Sequence alignment of TaTPL, OsTPL (LOC_Os08g0162100) and AtTPL (At1g15750) proteins. The graphic view of alignment was generated by MegAlign using Clustal W method. The black or grey shade represents the similarity. The green line represents LiSH (lissencephaly type‐1‐like homology) domain, while the red line denotes CTLH (C‐terminal to LiSH) domain. The WD40 repeats are marked by orange boxes. (b) Schematic diagram of the conserved domains in TaTPL.

Figure 6

Determination of the transcript levels of Q and TaTPL in different tissues/stages in bread wheat KN199. R, root tip; St, stem; L, leaf; FL, flag leaf. The levels of Q and TaTPL transcripts were normalized against TaGAPDH . Error bars represent SDs among three independent replicates.

Figure 7

Q directly interacts with TaTPL. (a) Yeast two‐hybrid (Y2H) assay showing the interaction between Q and TaTPL. SD‐L/W, synthetic dextrose medium lacking Leu and Trp; SD‐L/W/H/A, synthetic dextrose medium lacking Leu, Trp, His and Ade. (b) Luciferase (LUC) complementation imaging (LCI) assay illustrating that Q could interact with TaTPL in N. benthamiana. Luciferase signals were detected 48 hpi. (c) Bimolecular fluorescence complementation (BiFC) assay confirming the physical interaction of Q with TaTPL in N. benthamiana. YFP fluorescence was detected 48 hpi. BF, bright field. Scale bars, 20 μm. In each experiment in (b) and (c), six leaves were employed for the analyses. All the above experiments were independently repeated for three times with similar results.

Figure 8

The N‐terminal EAR motif of Q mediates its interaction with TaTPL. (a) The truncated or mutated versions of Q employed in the interaction assays in N. benthamiana. Q‐N, 1–199 amino acids (aa); Q‐M, 120–275 aa; Q‐C, 276–447 aa; Q^mEAR1, the full‐length Q with mutation in EAR1 motif. (b) LCI assays mapping the interaction domain of Q with the full‐length TaTPL. The interaction signals were detected at 48 hpi. In each experiment, six N. benthamiana leaves were infiltrated for analysis, and similar results were observed. Three independent replicates were performed. EV, empty vector.

Figure 9

The CTHL motif of TaTPL is required for the physical association with Q protein. (a) The truncated versions of TaTPL used in the interaction assays. TaTPL‐N, 1–339 aa; TaTPL‐C, 340–1135 aa; TaTPL‐N∆CTHL, the TaTPL‐N domain without the CTHL motif (34–92 aa). (b) LCI assays illustrating the interaction of truncated versions of TaTPL with the full‐length Q protein in N. benthamiana. Signals were collected at 48 hpi. Six N. benthamiana leaves were analysed in each experiment, and three independent replicates were performed with similar results.

Figure 10

Transient expression assay in N. benthamiana protoplasts illustrating the transcriptional repression activity of wild‐type and mutated Q. (a) The effectors used in the assay. (b) Relative luciferase activities in effector‐expressed samples. The activities of LUC and REN were determined 16 h post‐transformation, and the relative luciferase activities in different samples were calculated by normalizing the LUC values against REN. Error bars indicate SDs among three independent replicates. **P < 0.01 (Student's t‐test).