SlDELLA interacts with SlPIF4 to regulate arbuscular mycorrhizal symbiosis and phosphate uptake in tomato

Abstract

Arbuscular mycorrhizal symbiosis (AMS), a complex and delicate process, is precisely regulated by a multitude of transcription factors. PHYTOCHROME-INTERACTING FACTORS (PIFs) are critical in plant growth and stress responses. However, the involvement of PIFs in AMS and the molecular mechanisms underlying their regulator functions have not been well elucidated. Here, we show that SlPIF4 negatively regulates the arbuscular mycorrhizal fungi (AMF) colonization and AMS-induced phosphate uptake in tomato. Protein-protein interaction studies suggest that SlDELLA interacts with SlPIF4, reducing its protein stability and inhibiting its transcriptional activity towards downstream target genes. This interaction promotes the accumulation of strigolactones (SLs), facilitating AMS development and phosphate uptake. As a transcription factor, SlPIF4 directly transcriptionally regulates genes involved in SLs biosynthesis, including SlCCD7, SlCDD8, and SlMAX1, as well as the AMS-specific phosphate transporter genes PT4 and PT5. Collectively, our findings uncover a molecular mechanism by which the SlDELLA-SlPIF4 module regulates AMS and phosphate uptake in tomato. We clarify a molecular basis for how SlPIF4 interacts with SLs to regulate the AMS and propose a potential strategy to improve phosphate utilization efficiency by targeting the AMS-specific phosphate transporter genes PTs.

Figure 1

SlPIF4 negatively regulates the AMF colonization in tomato. (a) Temporal expression pattern of SlPIF4 in roots of WT plants. Dpi: days-post-inoculation. Data are presented as the means of three biological replicates (±SD). Asterisks indicate a statistically significant difference from the control in the means (**P < 0.01; Student’s t test). (b) Immunoblots showing the SlPIF4 protein level in roots of WT plants at 10 dpi. The same membrane was split into two sheets and incubated with anti-PIF4 and anti-HSP70 antibodies, respectively. Representative pictures are shown on the left. The relative protein levels are shown on the right of the blots, and the relative protein levels of WT in NM roots were set to 1.00. Data are presented as the means of three biological replicates (±SD). Asterisks indicate a statistically significant difference from the control in the means (**P < 0.01; Student’s t test). (c) Representative images of WGA-AF488-stained roots of WT and pif4 mutant plants at 20 dpi. BF, bright-field image. Merge, merged WGA-AF488 and BF image. Scale bar = 100 μm. (d) The root length colonization of hyphae, arbuscules, vesicles, and hyphopodia in the roots of WT and pif4 mutant plants at 20 dpi. Data are presented as the means of four biological replicates (±SD). Each replication contains 80 root segments of two plants. Asterisks indicate a statistically significant difference from the control in the means (^**P < 0.01; Student’s t test). (e) Transcript of AMS-marker gene BCP1 in roots of WT and pif4 mutant plants at 20 dpi. Data are presented as the means of four biological replicates (±SD). Asterisks indicate a statistically significant difference from the control in the means (^**P < 0.01; Student’s t test). For (a–e), the plants were grown in a sterilized soil: quartz sand: vermiculite mixture (1:1:1) inoculating with (AM) or without (NM) R. intraradices under Pi deficient condition (without KH₂PO₄ but with the addition of 1 mM KCl).

Figure 2

SlDELLA physically interacts with SlPIF4 in vivo and in vitro. (a) Yeast two-hybrid assay to confirm the interaction between SlDELLA and SlPIF4. Yeast cells were grown on SD/−Leu/−Trp (-LW) for 2 days or SD-Leu/−Trp/−Ade/-His (-LWAH) medium with 1 μM 3-AT for 5 days. (b) BiFC assay showing the interaction of SlDELLA with SlPIF4 in Nicotiana benthamiana. Full-length SlPIF4 and SlDELLA were fused to YFPⁿ or YFP^c, respectively. Scale bar = 25 μm. Images were taken using a confocal microscope. H2B-mCherry was used as a nuclear marker. (c)–(d) Split luciferase complementation assay (SLCA) showing the interaction of SlPIF4 with SlDELLA. The cLUC-SlPIF4 and nLUC-SlDELLA constructs were co-transformed into N. benthamiana leaves, and the luminescence intensity was detected after 48 h. The luminescence intensity detected in N. benthamiana leaves co-expressed cLUC and nLUC as the control. Representative photographs are shown in (c), and luciferase activity is shown in (d). Data are presented as the means of six biological replicates (±SD). Asterisks indicate a statistically significant difference from the control in the means (^**P < 0.01; Student’s t test). (e) Pull-down assay showing that MBP-tagged SlPIF4, but not MBP alone, could pull down GST-tagged SlDELLA in vitro. Recombinant MBP-SlPIF4 and GST-SlDELLA were detected with anti-MBP and anti-GST, respectively. The red arrow represents MBP-SlPIF4.

Figure 3

SlDELLA acts upstream of SlPIF4 to regulate the AMS in tomato. (a) Representative images of WGA-AF488-stained roots of WT-pTRV, WT-pTRV-SlDELLA, pif4#12-pTRV, and pif4#12-pTRV-SlDELLA plants at 20 dpi. BF, bright-field image. Merge, merged WGA-AF488 and BF image. Scale bar =100 μm. (b) The root length colonization of hyphae, arbuscules, and vesicles in the roots of WT-pTRV, WT-pTRV-SlDELLA, pif4#12- pTRV, and pif4#12-pTRV-SlDELLA plants at 20 dpi. Data are presented as the means of six biological replicates (±SD). Each replication contains 80 root segments of two plants. Asterisks indicate a statistically significant difference from the control in the means (**P < 0.01; Student’s t test); ns: no significant difference. (c)–(e) Transcripts of AMS-marker genes BCP1, PT4, and PT5 in roots of WT-pTRV, WT-pTRV-SlDELLA, pif4#12-pTRV, and pif4#12-pTRV-SlDELLA plants at 20 dpi. Data are presented as the means of four biological replicates (±SD). Asterisks indicate a statistically significant difference from the control in the means (^**P < 0.01; Student’s t test); ns: no significant difference. For (a)–(e), the plants inoculated with R. intraradices were grown in a sterilized soil: quartz sand: vermiculite mixture (1:1:1) under Pi deficient condition (without KH₂PO₄ but with the addition of 1 mM KCl).

Figure 4

SlPIF4-mediated AMS is partly dependent on SLs biosynthesis, accumulation and exudation in tomato. (a) Relative contents of SLs from WT and pif4 mutant plants roots at 10 dpi. Data are presented as the means of three biological replicates (±SD). Different letters indicate significant differences by One-way ANOVA followed by post hoc Tukey test (P < 0.05). (b)–(d) The expression of SlCCD7, SlCCD8, and SlMAX1 in roots of WT and pif4 mutant plants at 10 dpi. Data are presented as the means of four biological replicates (±SD). Different letters indicate significant differences by One-way ANOVA followed by post hoc Tukey test (P < 0.05). (e) Representative images of WGA-AF488-stained roots of in WT and SlPIF4-overexpressing (PIF4#89) plants applied with or without 1 μM rac-GR24 at 20 dpi. BF, bright-field image. Merge, merged WGA-AF488 and BF image. Scale bar =100 μm. (f) The root length colonization of hyphae, arbuscules, and vesicles in the roots of WT and SlPIF4-overexpressing (PIF4#89) plants applied with or without 1 μM rac-GR24 at 20 dpi. Data are presented as the means of four biological replicates (±SD). Each replication contains 80 root segments of two plants. Different letters indicate significant differences by one-way ANOVA followed by post hoc Tukey test (P < 0.05). (g) Transcripts of AMS-marker gene BCP1 in roots of WT and SlPIF4-overexpressing (PIF4#89) plants applied with or without 1 μM rac-GR24 at 20 dpi. Data are presented as the means of four biological replicates (±SD). Different letters indicate significant differences by One-way ANOVA followed by post hoc Tukey test (P < 0.05). For (a)–(d), the plants inoculated with (AM) or without (NM) Rhizophagus intraradices were grown in a sterilized soil: quartz sand: vermiculite mixture (1:1:1) under Pi deficient condition (without KH₂PO₄ but with the addition of 1 mM KCl). For (e)–(g), the WT and SlPIF4-overexpressing (PIF4#89) plants inoculated with R. intraradices were treated with ddH₂O (Control) or + rac-GR24 (1 μM) three times a week.

Figure 5

SlPIF4 directly binds to promoters of SLs biosynthesis genes and inhibits its transcription. (a) EMSA assay showing the binding of SlPIF4 to the promoters of SlCCD7, SlCCD8, and SlMAX1. Excess amounts (100×, 300×, 500×) of non-labeled or mutant oligo (mut) were set as the competitors. The biotin non-labeled oligo was used as a competitor, −: absence; +: presence. (b) ChIP-qPCR showing the binding of SlPIF4 to the SlCCD7, SlCCD8, and SlMAX1 promoters containing the G/E-box motifs in vivo. HA, hemagglutinin; OE, overexpressing. Values are percentages of DNA fragments that coimmunoprecipitated with anti-HA antibodies or anti-IgG relative to the input DNA. Data are presented as the means of three biological replicates (±SD). Different letters indicate significant differences by one-way ANOVA followed by post hoc Tukey test (P < 0.05). (c) Dual-luciferase assay showing the inhibition of SlPIF4 to the SlCCD7, SlCCD8 and SlMAX1 promoters in Nicotiana benthamiana. EV: empty vector. The LUC/REN ratios of the empty vector (EV) plus promoter were set at ‘1’. Data are presented as the means of six biological replicates (±SD). Asterisks indicate a statistically significant difference from the control in the means (^**P < 0.01; Student’s t test).

Figure 6

SlPIF4 negatively regulates the AMS-induced phosphate uptake by inhibiting the transcription of AMS-specific PTs. (a)–(b) Total P content of WT, SlPIF4 overexpressing and pif4 mutant plants at 30 dpi. Data are presented as the means of four biological replicates (±SD). Different letters indicate significant differences by one-way ANOVA followed by post hoc Tukey test (P < 0.05). (c)–(d) Transcripts of PT4 and PT5 in roots of WT, SlPIF4 overexpressing (PIF4#89) and pif4 mutant plants at 20 dpi. Data are presented as the means of four biological replicates (±SD). Different letters indicate significant differences by one-way ANOVA followed by post hoc Tukey test (P < 0.05). For (a)–(d), the plants inoculated with (AM) or without (NM) R. intraradices were grown in a sterilized soil: quartz sand: vermiculite mixture (1:1:1) under Pi deficient condition (without KH₂PO₄ but with the addition of 1 mM KCl). (e) EMSA assay showing the binding of SlPIF4 to the promoters of AMS-specific PTs. Excess amounts (100×, 300×, 500×) of non-labeled or mutant oligo (mut) were set as the competitors. The biotin non-labeled oligo was used as a competitor, −: absence; +: presence. (f) ChIP-qPCR showing the binding of SlPIF4 to the AMS-specific PTs promoters containing the E-box motifs in vivo. HA, hemagglutinin; OE, overexpressing. Values are percentages of DNA fragments that coimmunoprecipitated with anti-HA antibodies or anti-IgG relative to the input DNA. Data are presented as the means of three biological replicates (±SD). Different letters indicate significant differences by one-way ANOVA followed by post hoc Tukey test (P < 0.05). (g) Dual-luciferase assay showing the inhibition of SlPIF4 to the AMS-specific PTs promoters in Nicotiana benthamiana. EV: empty vector. The LUC/REN ratios of the empty vector (EV) plus promoter were set at ‘1’. Data are presented as the means of six biological replicates (±SD). Asterisks indicate a statistically significant difference from the control in the means (^**P < 0.01; Student’s t test).

Figure 7

SlDELLA regulates the expression of SlPIF4 target genes by reducing the protein stability and attenuating the transcriptional activity of SlPIF4. (a) Transcript of SlCCD7 in roots of WT and pro plants at 10 dpi. Data are presented as the means of four biological replicates (±SD). Different letters indicate significant differences by one-way ANOVA followed by post hoc Tukey test (P < 0.05). (b) Transcripts of AMS-specific PTs in roots of WT and pro plants at 20 dpi. Data are presented as the means of four biological replicates (±SD). Asterisks indicate a statistically significant difference from the control in the means (^**P < 0.01; Student’s t test). (c) Immunoblots showing the SlPIF4 protein level in WT and pro plants at 10 dpi. A same membrane was split into two sheets and incubated with anti-PIF4 and anti-HSP70 antibodies, respectively. Anti-HSP70 was used as a loading control for the western blot analysis. Representative pictures are shown. Relative protein levels are shown on the right side of the blots. Data are presented as the means of three biological replicates (±SD). Asterisks indicate a statistically significant difference from the control in the means (^**P < 0.01; Student’s t test). (d) SlPIF4 degradation in cell-free degradation assay. Total proteins extracted from the roots of WT and pro plants inoculated with R. intraradices were incubated with or without MG132 (a 26S proteasome inhibitor) over the indicated time course, and the protein levels of SlPIF4 were detected using an anti-MBP antibody. Anti-HSP70 was used as a loading control for the western blot analysis. For (a)–(d), the plants inoculated with (AM) or without (NM) R. intraradices were grown in a sterilized soil: quartz sand: vermiculite mixture (1:1:1) under Pi deficient condition (without KH₂PO₄ but with the addition of 1 mM KCl). (e)–(f) Dual-luciferase assay showing the regulatory effect of SlPIF4 influenced by SlDELLA on the promoters of SlCCD7 and AMS-specific PT4. The ratio of firefly luciferase (LUC) and renilla luciferase (REN) of the empty vector (EV) plus promoter was set at ‘1’. Data are presented as the means of six biological replicates (±SD). Different letters indicate significant differences by one-way ANOVA followed by post hoc Tukey test (P < 0.05). (g)–(h) Electrophoresis mobility shift (EMSA) assay. The biotin-labeled SlCCD7 and AMS-specific PT4 oligos was used as SlPIF4-targeted DNA sequence. 1×, 3×, and 5× indicated the intensity of the GST-SlDELLA protein, and 1×, 2×, and 3× indicated the intensity of the GST protein. The protein purified from the empty vector was used as a negative control.

Figure 8

A proposed model showing SlDELLA-SlPIF4-SLs/PTs signaling pathway regulates AMS and phosphate uptake in tomato. SlDELLA interacts with SlPIF4 to reduce SlPIF4 protein stability and its transcriptional activity toward downstream target genes, including SLs biosynthesis genes and AMS-specific phosphate transporter genes, thus promoting SLs accumulation, AMS development, and Pi uptake in tomato.