The TOPLESS interactome: a framework for gene repression in Arabidopsis

Abstract

Transcription factors activate or repress target gene expression or switch between activation and repression. In animals and yeast, Groucho/Tup1 corepressor proteins are recruited by diverse transcription factors to induce context-specific transcriptional repression. Two groups of Groucho/Tup1-like corepressors have been described in plants. LEUNIG and LEUNIG_HOMOLOG constitute one group and TOPLESS (TPL) and the four TPL-related (TPR) corepressors form the other. To discover the processes in which TPL and the TPR corepressors operate, high-throughput yeast two-hybrid approaches were used to identify interacting proteins. We found that TPL/TPR corepressors predominantly interact directly with specific transcription factors, many of which were previously implicated in transcriptional repression. The interacting transcription factors reveal that the TPL/TPR family has been coopted multiple times to modulate gene expression in diverse processes, including hormone signaling, stress responses, and the control of flowering time, for which we also show biological validation. The interaction data suggest novel mechanisms for the involvement of TPL/TPR corepressors in auxin and jasmonic acid signaling. A number of short repression domain (RD) sequences have previously been identified in Arabidopsis (Arabidopsis thaliana) transcription factors. All known RD sequences were enriched among the TPL/TPR interactors, and novel TPL-RD interactions were identified. We show that the presence of RD sequences is essential for TPL/TPR recruitment. These data provide a framework for TPL/TPR-dependent transcriptional repression. They allow for predictions about new repressive transcription factors, corepressor interactions, and repression mechanisms and identify a wide range of plant processes that utilize TPL/TPR-mediated gene repression.

Figure 1.

A summary of the TPL/TPR protein-protein interactions. A, Interactions detected in the whole plant yeast two-hybrid experiments. B, Interactions identified from the arrayed transcription factor library. In each case, the first column provides the Arabidopsis Genome Initiative (AGI) number for each factor. The second column indicates the identity of the factor, if known. Factors highlighted in gray in this column represent those that are known to act as transcriptional repressors. The third column indicates the transcription factor family to which each factor belongs, where dark green represents the AP2/ERF family, purple the MYB family, red the AUX/IAAs, pink the ARFs, light blue the C2H2 Zn fingers, dark blue the homeodomain family, brown the NAC proteins, orange the MADS box family, and yellow the TCP factors. Interactions between these factors and the TPL/TPR proteins are represented by dark gray boxes in the appropriate columns. For the whole plant yeast two-hybrid experiment, numbers represent the frequency (%) at which each factor was isolated. The RD column shows the sequence of any known functional RD found in the TPL/TPR-interacting factors. Black boxes to the right side of each table indicate those factors isolated in both the whole plant library and arrayed transcription factor screens. Factors and RDs highlighted in pale green in the arrayed transcription factor library table represent those proteins in which mutagenesis of the RD was performed to evaluate its requirement for interaction with the TPL protein.

Figure 2.

Short RD sequences are necessary for interaction with TPL. The predicted RDs of ERF3 (FDLNFPP to FAHNFPP), RAV1 (RLFGV to RSSGV), and AP2-L (MLFGV to MQCGV) were mutagenized in the context of the full-length prey proteins and tested for interaction with the TPL bait in a yeast two-hybrid assay. Transformed yeast were spotted onto medium that selects for protein-protein interactions (−His) or control medium without selection (+His). Wild-type prey proteins (WT) were compared with mutant prey proteins (M1 and M2) for interaction with TPL.

Figure 3.

Comparison of flowering-time phenotypes and FT expression in wild-type and tpl-1 plants constitutively expressing TOE1. A, Left to right: wild-type (Ler) plants, Ler plants overexpressing TOE1 (TOE1oe), tpl-1 mutant, and tpl-1 overexpressing TOE1. For Ler lines, photographs were taken 30 d after germination; for tpl-1 lines, photographs were taken 27 d after germination. B, qRT-PCR analysis of TOE1 expression in Ler, Ler TOE1oe, tpl-1, and tpl-1 TOE1oe leaves relative to the EIF4A control. C, qRT-PCR analysis of FT expression in Ler, Ler TOE1oe, tpl-1, and tpl-1 TOE1oe leaves relative to the EIF4A control. Values represent means of three independent biological replicates.

Figure 4.

Complementation of wus mutants. A, WUS constructs used in the complementation experiments. B, GUS staining patterns for plants expressing GUS from the WUS promoter (3,351 bp) used to prepare the constructs for the complementation experiment. The left panel shows staining in seedlings, and the right panel shows staining in an early flower and the anthers of a later flower. C, Complementation of the strong wus-1 allele. The first column shows mature wild-type plants, the second column shows wus-1 plants with the pWUS::WUS construct, the third column shows wus-1 plants with the pWUS::WUSΔTPL construct, the fourth column shows wus-1 plants with the pWUS::WUSΔ construct, and the fifth column shows wus-1 mutant plants. In all cases, the top panel shows mature plants, the middle panel shows typical inflorescences, and the bottom panel shows typical flowers. For wus-1 pWUS::WUS, the bottom panel shows a typical weak rescue flower on the left with a strong rescue flower on the right. D, Complementation of the weak wus-3 allele. The left panel shows a typical wus-3 mutant flower, the center panel shows a wus-3 flower with the pWUS::WUSΔ construct, and the right panel shows a wus-3 flower with the pWUS::WUSΔTPL construct.

Figure 5.

The TPL family acts as general corepressors in diverse biological pathways. The interactome data place the TPL/TPR family of corepressors at the center of many biological processes. TPL is implicated in hormone responses, with the interaction data suggesting that the TPL/TPR family is involved in novel mechanisms of auxin and JA signaling. TPL/TPRs appear to be involved in a broad range of stress and plant immune responses in addition to numerous developmental pathways, such as floral transition and leaf and flower development. Colored arcs represent the different processes in which the TPL family may act and show how aspects of these pathways overlap. TF, Transcription factor.