BHLH32 modulates several biochemical and morphological processes that respond to Pi starvation in Arabidopsis

Abstract

P(i) (inorganic phosphate) limitation severely impairs plant growth and reduces crop yield. Hence plants have evolved several biochemical and morphological responses to P(i) starvation that both enhance uptake and conserve use. The mechanisms involved in P(i) sensing and signal transduction are not completely understood. In the present study we report that a previously uncharacterized transcription factor, BHLH32, acts as a negative regulator of a range of P(i) starvation-induced processes in Arabidopsis. In bhlh32 mutant plants in P(i)-sufficient conditions, expression of several P(i) starvation-induced genes, formation of anthocyanins, total P(i) content and root hair formation were all significantly increased compared with the wild-type. Among the genes negatively regulated by BHLH32 are those encoding PPCK (phosphoenolpyruvate carboxylase kinase), which is involved in modifying metabolism so that P(i) is spared. The present study has shown that PPCK genes are rapidly induced by P(i) starvation leading to increased phosphorylation of phosphoenolpyruvate carboxylase. Furthermore, several Arabidopsis proteins that regulate epidermal cell differentiation [TTG1 (TRANSPARENT TESTA GLABRA1), GL3 (GLABRA3) and EGL3 (ENHANCER OF GL3)] positively regulate PPCK gene expression in response to P(i) starvation. BHLH32 can physically interact with TTG1 and GL3. We propose that BHLH32 interferes with the function of TTG1-containing complexes and thereby affects several biochemical and morphological processes that respond to P(i) availability.

Figure 1. P_i starvation induces PPCK gene expression and increases the phosphorylation state of PEPC

(A) and (B) 7-day-old cells were transferred to medium containing the indicated concentration of KH₂PO₄ for 3 h. (A) Cells were collected, RNA was isolated and gene expression was analysed by RT-PCR. (B) Cells were treated as in (A). The malate sensitivity of PEPC in cell extracts is expressed as the percentage inhibition given by 1 mM L-malate. Error bars are S.D. for n=5 measurements in one representative experiment. Relative to the value at 2.4 mM P_i, malate sensitivity is significantly reduced at 0.6 mM P_i and at zero P_i (P<0.01 and P<0.001 respectively, by Student's t test. (C) 7-day-old seedlings were transferred to MS micronutrient medium containing 50 mM sucrose and 5 mM KNO₃ without P_i for the times indicated. Seedlings were collected, RNA was isolated and gene expression was analysed by RT-PCR. (D) Left-hand panel, seedlings were harvested after growth for 7 days with 1.25 mM P_i or after a further 3 h without P_i and PPCK protein was detected by Western blot analysis following SDS/PAGE of extracts (15 μg protein); right-hand panel, purified tagged soybean PPCK resolved on SDS/PAGE and stained with Coomassie blue. The M_r of this protein is slightly bigger than those of the two Arabidopsis PPCKs.

Figure 2. Gene expression in mutants under P_i starvation

Seedlings (7 days) were transferred to 1/5×MS without KH₂PO₄, 1×MS micronutrient medium containing 50 mM sucrose and 2.5 mM Mes (pH 6.0) for 12 h prior to harvesting. (A) Quantitative real-time PCR analysis of PPCK transcript abundance relative to that of UBQ10, values are means±S.D. (n=4). (B) Total P_i contents, means±S.D. (n=4). WT, wild-type.

Figure 3. Increased expression of PPCK and DFR genes in bhlh32

(A) Seedlings were harvested 3 h after transfer to medium with (+) or without (−) P_i. Expression of BHLH32 and DFR was assessed by RT-PCR. The T-DNA insertion in bhlh32 (Salk_013517) is immediately after the end of the first exon of At3g25710 (results not shown). (B) Seedlings were transferred to medium without P_i for the indicated times. PPCK expression was assessed by quantitative real-time PCR as described for Figure 2. WT, wild-type.

Figure 4. Expression of GFP–BHLH32 corrects the root hair phenotype of bhlh32

Seedlings were grown as described in the Experimental section. (A) Representative images of Col0, bhlh32 and bhlh32 transformed with the GFP–BHLH32 fusion. The scale bar represents 100 μm. (B) Root hairs were counted from 0.75 mm to 1.75 mm from the root tip, values are means±S.D. (n=5). The bhlh32 line contains significantly more root hairs in the presence of P_i than the wild-type or bhlh32 expressing GFP–BHLH32 (P<0.001, Student's t test). (C) Western blot analysis of the location of GFP–BHLH32. Left-hand panel, seedlings of bhlh32 transformed with the GFP–BHLH32 fusion were grown as described in the Experimental section and harvested before (+P_i) or after (−P_i) 3 h of P_i starvation. Extracts were centrifuged at 12000 g and equal amounts (15 μg protein) were analysed by SDS/PAGE and Western blot analysis with an anti-GFP antibody. Lanes a, supernatants; lanes b, pellets. Right-hand panel, nuclei from seedlings in P_i-sufficient medium were prepared and subjected to high salt extraction (see the Experimental section); SN and P represent the soluble and particulate material from this extraction respectively.

Figure 5. bhlh32 contains increased anthocyanin and P_i

Seedlings were grown as described in the Experimental section. (A) Anthocyanin content and (B) P_i content, values are means±S.D. (n=4). Relative to the wild-type, bhlh32 contains significantly more anthocyanin at both 0.625 mM P_i and 0.04 mM P_i, P<0.001 and P<0.01 respectively, and significantly more P_i at both concentrations, P<0.001 and P<0.05 respectively.

Figure 6. Physical interactions between components involved in the P_i starvation response

(A) and (B) show interactions analysed in the yeast-two-hybrid system. (A) The indicated combinations of BD (DNA-binding domain) and AD (activation domain) vectors were co-transformed into yeast strain AH109 and plated as described in the Experimental section. Positive interactions are shown by the growth of blue colonies. (B) The strength of positive interactions was assessed by α-galactopyranosidase activity assays. Values are means±S.D. from four individual yeast colonies. (C) In vitro pulldown assay. BHLH32 was expressed as a GST-fusion protein in E. coli and immobilized onto glutathione-agarose beads and then incubated with ³⁵S-labelled proteins expressed using the in vitro transcription–translation system. As a negative control, beads were loaded with GST. Bound proteins were resolved by SDS/PAGE and visualized using a Fuji phosphorimager.