The Arabidopsis transthyretin-like protein is a potential substrate of BRASSINOSTEROID-INSENSITIVE 1

Abstract

BRASSINOSTEROID-INSENSITIVE 1 (BRI1) is a Leu-rich-repeat (LRR) receptor kinase that functions as a critical component of a transmembrane brassinosteroid (BR) receptor. It is believed that BRI1 becomes activated through heterodimerization with BAK1, a similar LRR receptor kinase, in response to BR signal. A yeast two-hybrid screen using the kinase domain of BRI1 identified an Arabidopsis thaliana Transthyretin-Like protein (TTL) as a potential BRI1 substrate. TTL interacts with BRI1 in a kinase-dependent manner in yeast and is phosphorylated by BRI1 in vitro. TTL displays a similar expression pattern with BRI1 and is associated with the plasma membrane. Overexpression of the TTL gene results in a phenotype that was observed in weak bri1 mutants and null bak1 mutants. By contrast, two T-DNA insertional mutations in the TTL gene promote plant growth and enhance BR sensitivity. We hypothesized that TTL might directly regulate certain biochemical activities near the plasma membrane to control plant growth.

Figure 1.

Identification of TTL as a BRI1-Interacting Protein by Yeast Two-Hybrid Screening. (A) TTL showed specific interaction with BRI1 kinase domain (BRI1CK), examined by the growth on His-lacking synthetic media and the production of blue colonies on the 5-bromo-4-chloro-3-indolyl-β-d-galactoside (X-Gal)–containing media, compared with its interactions with nonrelevant baits, including CLV1 kinase domain (CLV1CK), the N-terminal portion of an Arabidopsis NADPH oxidase (N-gp91), the C-terminal segment of the Arabidopsis phytochrome A (C-PHYA), and the Gal4 binding domain (GAL4BD) alone. (B) Amino acid sequence alignment of TTL and its homologs from various organisms, including LeTTL from Lycopersicon esculentum, StTTL from Solanum tuberosum, GmTTL from Glycine max, TaTTL from Triticum aestivum, HvTTL from Hordeum vulgare, OsTTL from Oriza sativa, MmTRP from Mus musculus, DmTRP from Drosophila melanogaster, CeTRP from Caenorhabditis elegans, SpTRP from Schizosaccharomyces pombe, and EcTRP from Escherichia coli. Except TTL, other full-length plant TTL sequences were deduced from overlapping EST sequences as follows: LeTTL (AW649495, BE432987, NM535270, and BF114284), StTTL (BE920936, BM405599, and BE919514), GmTTL (AW100149, BE022630, BI468680, and BE824466), TaTTL (BE400327, BQ245413, and BE444381), HvTTL (BE602694, BU989926, and AL503126), and OsTTL (AC092075, AU067993, and CB214315). Sequence alignment was conducted using ClustalW software.

Figure 2.

TTL Is a Putative Substrate for BRI1. (A) Different bri1 mutations have different effects on the kinase activity of BRI1. In vitro autophosphorylation activity was measured using the E. coli expressed GST fusion proteins of the wild-type BRI1 cytoplasmic kinase domain and its three mutated counterparts containing bri1-101, bri1-104, and bri1-114, respectively. The top panel shows the amount of purified recombinant proteins visualized by Coomassie blue staining, and the bottom panel reveals the autophosphorylation level by autoradiography. (B) Yeast two-hybrid assay of the interaction between TTL and the three mutated BRI1 cytoplasmic kinases. (C) In vitro transphosphorylation assay of TTL by BRI1. The wild type and mutated BRI1 cytoplasmic kinase domains were fused to GST, whereas TTL was fused to MBP for recombinant protein expression in E. coli cells. Transphosphorylation assay was performed by mixing MBP-TTL with GST-fused wild type or mutated BRI1CK. Lane 1, GST-BRI1CK alone; lane 2, GST-BRI1 (one-fifth of the amount loaded on lane 1) mixed with MBP-TTL; lane 3, GST-BRI1CK (bri1-101) alone; lane 4, GST-BRI1CK (bri1-101) (one-fifth of the amount loaded on lane 3) mixed with MBP-TTL; lane 5, MBP-TTL alone; lane 6, MBP alone; lane 7, GST-BRI1CK mixed with MBP. The top panel displays the purified recombinant protein by Coomassie blue staining, and the bottom panel shows the level of protein phosphorylation by autoradiography.

Figure 3.

The Expression Pattern of the TTL Gene. (A) RNA gel blot analysis of the TTL gene expression. Total RNAs were isolated from various tissues of 5-week-old Arabidopsis plants, including young rosette leaves, old leaves, cauline leaves, floral buds, flowers, young siliques, old siliques, stems, and roots. (B) The expression of TTL is not regulated by BR. Total RNAs were isolated from 10-d-old seedlings of wild-type, det2, cpd, and bri1 mutants treated with or without 1 μM brassinolide. (C) to (G) Histochemical analysis of the GUS reporter gene expression driven by the TTL promoter. GUS signal was detected in root tips of 3-d-old seedling (C), hypocotyls (D), leaf primordia (E), expanding young leaves (F) of 10-d-old seedlings, and flowers (G) of mature plants.

Figure 4.

TTL Is a Membrane-Associated Protein. (A) Confocal microscopic analysis of GFP fusion proteins in the root tips of 4-d-old transgenic plants containing the BRI1-BRI1:GFP or TTL-TTL:GFP transgene. The green fluorescent signal of the GFP itself driven by two copies of the strong 35S promoter of Cauliflower mosaic virus was used as a control for cytosolic and nuclear localization. (B) Subcellular fractionation analysis of the TTL localization. Cytosolic (lanes 1 to 4) and membrane fractions (lanes 5 to 8) were prepared from the 7-d-old seedlings of transgenic plants containing the pPZP212 vector alone, D35S-GFP, TTL-TTL:GFP, or BRI1-BRI1:GFP transgene. The similar amount of total proteins of each fractionation was analyzed for the presence of GFP or GFP-fusion proteins by immunostaining with anti-GFP antibodies (Molecular Probes, Eugene, OR). The asterisk indicates a nonspecific cross-reacting band, and the arrow indicates the TTL:GFP fusion protein.

Figure 5.

ttl Overexpression Leads to Growth Inhibition. (A) RNA gel blot analysis of ttl overexpression in eight independent transgenic lines with the ³²P-labeled full-length ttl cDNA probe. (B) Phenotypic comparison of two ttl-overexpressing transgenic lines with the corresponding wild-type (Col-0) plants. Plants shown here were 5 weeks old grown under long-day growth condition (16 h light/8 h dark). (C) Quantitative analysis of the hypocotyl growth of the two ttl-overexpressing transgenic lines grown for 5 d in the dark or under a short-day (8-h-light/16-h-dark) growth condition. Hypocotyl growth is expressed as a percentage of the hypocotyl length of the 5-d-old wild-type seedling grown in the dark, and each data point represents the average of ∼50 seedlings. (D) Quantitative analysis of the height and petiole length of the ttl transgenic plants relative to those of the wild-type plants. The wild-type control and the two ttl-overexpressing transgenic lines were grown in soil for 7 weeks under a long-day (16-h-light/8-h-dark) growth condition. The length of the main inflorescence stem was used as the height of an adult plant, and the average petiole length of the three longest rosette leaves of each analyzed plant was taken as the petiole length. Each data point represents the average of 14 plants, and the error bar denotes standard error.

Figure 6.

Two T-DNA Insertional Mutations Promote Plant Growth. (A) Schematic representation of two T-DNA insertion sites in the TTL gene. ttl-1 contains the T-DNA insertion in the 117th codon of the second exon, whereas ttl-2 has the T-DNA inserted 15 bp upstream of the presumed ATG start codon. (B) RNA gel blot analysis of the TTL gene expression in the two ttl mutants. (C) Phenotypic comparison between the two ttl mutants and the corresponding wild-type plant (Ws-0) grown for 5 weeks under the 16-h-light/8-h-dark growth condition. (D) Quantitative analysis of the hypocotyl growth of the ttl-1 mutants grown for 5 d in the dark or under a short-day (8-h-light/16-h-dark) growth condition. Hypocotyl growth is expressed as a percentage of the hypocotyl length of the 5-d-old wild-type seedlings grown in the dark, and each data point represents the average of ∼50 seedlings. (E) Quantitative analysis of the height and petiole length of the ttl-1 mutants relative to those of the wild-type control plants. Both the wild-type control and the ttl-1 mutants were grown in soil for 7 weeks under the 16-h-light/8-h-dark growth condition. The length of the main inflorescence stem was used as the height of an adult plant, and the average petiole length of the three longest rosette leaves of each analyzed plant was taken as the petiole length. For (D) and (E), each data point represents the average of 16 plants, and the error bar denotes standard error. (F) Shown are a 5-week-old ttl-1 mutant and a 5-week-old wild-type (Ws-0) plant.

Figure 7.

TTL Modulates BR-Mediated Growth Response. (A) Overexpression of the TTL gene suppresses the BRI1 overexpression phenotype. Shown (from left to right) are a 35S-TTL transgenic plant, a BRI1-BRI1:GFP transgenic line, and a F2 plant homozygous for both the 35S-TTL and BRI1-BRI1:GFP transgenes. (B) Quantitative analysis of the hypocotyl growth of the 5-d-old transgenic Arabidopsis seedlings grown in the dark or under the 8-h-light/16-h-dark growth condition. Hypocotyl growth is expressed as a percentage of the hypocotyl length of the dark-grown seedlings of the BRI1-overexpressing transgenic line, and each data point represents the average of ∼30 seedlings. (C) The ttl mutants display a brassinazole-resistant phenotype. Shown (from left to right) are a wild-type seedling, a ttl-1 mutant, and a BRI1-BRI1:GFP transgenic plant grown on 1 μM brassinazole-containing medium. (D) and (E) Quantitative analysis of BR sensitivity of the TTL-overexpressing plants (D) and the ttl-1 mutants (E). Seedlings were germinated and grown on medium containing increasing concentrations of brassinolide. Root elongation was measured 7 d after germination. Each data point represents the average root elongation of ∼50 seedlings. Inhibition of root growth by brassinolide is expressed as a percentage of the root elongation of the wild-type controls grown on medium containing the same volume of 80% (v/v) ethanol used to dilute brassinolide from a 2 mM stock solution. Error bar denotes standard error.