A secretory phospholipase D hydrolyzes phosphatidylcholine to suppress rice heading time

Abstract

Phospholipase D (PLD) hydrolyzes membrane phospholipids and is crucial in various physiological processes and transduction of different signals. Secretory phospholipases play important roles in mammals, however, whose functions in plants remain largely unknown. We previously identified a rice secretory PLD (spPLD) that harbors a signal peptide and here we reported the secretion and function of spPLD in rice heading time regulation. Subcellular localization analysis confirmed the signal peptide is indispensable for spPLD secretion into the extracellular spaces, where spPLD hydrolyzes substrates. spPLD overexpression results in delayed heading time which is dependent on its secretory character, while suppression or deficiency of spPLD led to the early heading of rice under both short-day and long-day conditions, which is consistent with that spPLD overexpression/suppression indeed led to the reduced/increased Hd3a/RFT1 (Arabidopsis Flowing Locus T homolog) activities. Interestingly, rice Hd3a and RFT1 bind to phosphatidylcholines (PCs) and a further analysis by lipidomic approach using mass spectrometry revealed the altered phospholipids profiles in shoot apical meristem, particularly the PC species, under altered spPLD expressions. These results indicate the significance of secretory spPLD and help to elucidate the regulatory network of rice heading time.

Fig 1. spPLD is a secretory PLD.

A. Sequence alignment was performed with BLAST and the top 30 proteins with protein functional annotation and identity at more than 70% were analyzed. The bootstrap consensus tree inferred from 1000 replicates is generated to represent the evolutionary history of the taxa analyzed. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates) is shown next to the branches. NCBI accession number, protein ID and species of analyzed protein are shown. B. Protein structural analysis showed the presence of two HKD motifs and one signal peptide (sp) at N-terminus of rice spPLD. C. qPCR analysis revealed the spPLD expression in various tissues including roots, tillers, stems, SAM, leaves, pulvini, flowers and panicles. Expression level was normalized to ACTIN1 transcript and relative expressions were calculated by setting spPLD expression in leaves as 1.0. Experiments were repeated three times and data were shown as mean ± SD (n = 3). D. Secretion of spPLD was confirmed by observing N. benthamiana plants expressing spPLD-GFP fusion protein. FM4-64 was used to highlight the plasma membrane. Plasmolysis was conducted by 1 M mannitol treatment for 10 min. The secreted sections were highlighted by arrows. DIC, bright field. Scale bar = 50 μm. E. Western Blotting analysis confirms the secretion of spPLD. Various fusion proteins were transiently expressed in rice protoplasts. After incubation for 48 h, proteins of supernatant (incubation medium) and precipitation (protoplast homogenate) were extracted and analyzed by Western Blotting using anti-GFP antibody. Arrows highlighted the GFP (sp-GFP) and spPLD-GFP (ΔPLD-GFP) proteins.

Fig 2. spPLD delays rice heading time.

A. Enzymatic assay showed that both spPLD and ΔPLD (spPLD deleting the signal peptide) present PLD activity. Purified spPLD or ΔPLD (1–5 μg) proteins were used for examination and choline was used as substrate. There is no choline in background (as negative control) and positive control is supplied in assay kit. Experiments were repeated three times and data were shown as mean ± SD (n = 3). B. Sequencing confirmed three mutation lines of spPLD (insertion or deletion of bases), by CRISPR/Cas9. The gRNA targeting site and PAM sequence, and the position of gRNA at spPLD gene, are indicated. C. Phenotypic observation of rice plants with altered spPLD expression under natural long-day condition at heading stage. Representative images were shown. Scale bar = 10 cm. D. Rice plants with altered spPLD expression were grown under natural long-day (left) or short-day (right) conditions and heading date were calculated and statistically analyzed using Tukey’s test (**, p < 0.01; ***, p < 0.001). NS, no significance. Days to flowering were scored when first panicle was bolted, and data were shown as mean ± SD (n = 50).

Fig 3. Altered PC species under spPLD overexpression or deficiency.

A. Fat-western immunoblot analysis revealed the binding of rice Hd3a (left) and Hd3a with mutated PC binding sites (Hd3a^M, right) to phospholipids. The phospholipid type of each dot is indicated. PC, phosphatidylcholine; PE, phosphatidylethanolamine; PS, phosphatidylserine; PA, phosphatidic acid; LPA, lysophosphatidic acid; LPC, lysophosphatidylcholine; S1P, sphingosine 1-phosphate; PI, phosphatidylinositol; PI3P, PI 3-monophosphate; PI4P; PI5P; PI(3,4)P₂, PI 3,4-bisphosphate; PI(3,5)P₂; PI(4,5)P₂; PI(3,4,5)P₃, phosphatidylinositol 3,4,5-trisphosphate. B. Liposome binding assay (left) and quantitative analysis (right) confirmed the preferential binding of Hd3a to PC. His-tag fused Hd3a and Hd3a with mutated PC binding sites (Hd3a^M) were purified and incubated with liposomes containing different PC:PA or PE:PA ratios. After collecting the liposomes, the portion of proteins bound to liposomes was detected by western blotting using anti-His antibody. Nonbinding protein was detected in the supernatant (bottom). Band density is measured by Image J and relative density was calculated by setting the intensity under PC:PA ratio 1:1 as 1.0. Data were presented as means ± SD (n = 3, right). C-D. Relative content of predominant PCs with different saturation status, PC (32:1), PC (36:2), PC (36:3) and PC (38:3) (C); PC (36:5), PC (36:6) and PC (38:6) (D), in ZH11 and various lines with altered spPLD expressions. Ten rice shoot apical meristem before bolting were collected and used for lipids extraction. Phospholipids were profiled by a lipidomic approach using mass spectrometry. Experiments were biologically repeated three times and data were shown as mean ± SD (n = 3). Statistical analysis was performed by Tukey’s test (*, p <0.05; **, p < 0.01 ***, p < 0.001, compared to ZH11).

Fig 4. Expression of OsMADS14, 15, 18 and 34 were decreased in spPLDox lines while increased in spPLD-RNAi/Cas9 lines.

Total RNAs were extracted from SAM (~ 1 cm in length) of five-week-old rice plants (~ 7 days before bolting) and expressions of OsMADSs genes were examined by qPCR analysis. Rice plants were grown under LD condition (14-h light / 10-h dark cycle, 28°C) with 70% humidity. Expression levels of examined genes were normalized to ACTIN1 transcript. Experiments were repeated for three times and data were shown as mean ± SD (n = 3). Statistical analysis was performed by Tukey’s test (*, p <0.05; **, p < 0.01; ***, p < 0.001, compared to ZH11).

Fig 5. Conserved function of spPLD.

A. Hd3a with mutated PC binding sites (Hd3a^M) did not promote flowering time of Arabidopsis as Hd3a. Transgenic Arabidopsis overexpressing Hd3a showed early flowering under long-day condition while those overexpressing Hd3a^M did not (left). Number of rosette leaves and flowering time of plants were calculated (right) and statistically analyzed by Tukey’s test (lower, *, p < 0.05; ***, p < 0.001, compared to Col-0). NS, no significance. Representative images were shown (left, bar = 1 cm) and data were shown as mean ± SD (n = 15). Protein sequence of Arabidopsis FT and rice Hd3a were shown, and amino acids for PC binding of Arabidopsis FT, which is conserved in rice Hd3a and used for mutation analysis, were highlighted by red color (upper). B. Phylogenetic analysis of spPLDs in different plant species. Sequence alignment and the rectangular cladogram were generated with ClustalX 2.0 or TreeView respectively. Amino acid pairwise identity (percentage) between different protein sequences was shown. Scale bar represents 0.05 amino acid substitutions per site. C. spPLD delays flowering time of Arabidopsis. Transgenic Arabidopsis overexpressing spPLD showed delayed flowering under long-day condition. Number of rosette leaves and flowering time of plants were calculated and statistically analyzed using Tukey’s test (right, ***, p < 0.001, compared to Col-0). Representative images were shown (left, bar = 1 cm) and data were shown as mean ± SD (n = 15). D. A hypothetical model (tissue model and cellular model respectively) illustrating how secretory spPLD functions in suppressing rice heading time. spPLD showing a diurnally rhythmic and JA-upregulated expression, is secreted and hydrolyzes phosphatidylcholine (PC) at apoplast to reduce the levels of light period predominant PC species (less unsaturated PCs) in shoot apical meristem (SAM), which may lead to the reduced activity of primary pump (H⁺-ATPase) at plasma membrane, resulting in the less driving force for Hd3a/RFT1 entering cell at SAM, hence the decreased Hd3a/RFT1 activity and delayed heading. Other phospholipids may involve in heading time regulation through binding with and regulating the activity of Hd3a/RFT1.