Cytidinediphosphate diacylglycerol synthase-Mediated phosphatidic acid metabolism is crucial for early embryonic development of Arabidopsis

Abstract

Embryonic development is a key developmental event in plant sexual reproduction; however, regulatory networks of plant early embryonic development, particularly the effects and functional mechanisms of phospholipid molecules are still unknown due to the limitation of sample collection and analysis. We innovatively applied the microspore-derived in vitro embryogenesis of Brassica napus and revealed the dynamics of phospholipid molecules, especially phosphatidic acid (PA, an important second messenger that plays an important role in plant growth, development, and stress responses), at different embryonic developmental stages by using a lipidomics approach. Further analysis of Arabidopsis mutants deficiency of CDS1 and CDS2 (cytidinediphosphate diacylglycerol synthase, key protein in PA metabolism) revealed the delayed embryonic development from the proembryo stage, indicating the crucial effect of CDS and PA metabolism in early embryonic development. Decreased auxin level and disturbed polar localization of auxin efflux carrier PIN1 implicate that CDS-mediated PA metabolism may regulate early embryogenesis through modulating auxin transport and distribution. These results demonstrate the dynamics and importance of phospholipid molecules during embryo development, and provide informative clues to elucidate the regulatory network of embryogenesis.

Fig 1. Dynamics of phospholipids at different stages during microspore embryogenesis of Brassica napus.

(A-F) Microspore embryogenesis of Brassica napus, including vacuolated microspore (A), two-cells proembryo cultured for 48 h (B), four-cells proembryo cultured for 3 d (C), globular embryo with a suspensor cultured for 6 d (D), heart-shaped embryo with a suspensor cultured for 10 d (E), and cotyledonary embryos cultured for 28 d (F) after heat-shock treatment. Bars = 25 μm (A-E), or 1.5 mm (F). (G) Embryos at developmental stages A, D and E were collected and amounts of various phospholipids were measured by a lipidomics approach, which were shown as the percentage of total phospholipids. PC, phosphatidylcholine; PE, phosphatidylethanolamine; PG, phosphatidylglycerol; PA, phosphatidic acid; PS, phosphatidylserine. Experiments were performed with three technical replicates and data were mean ± SE. Statistical significance was determined by student’s t test (*, P<0.05).

Fig 2. CDS1 and CDS2 genes are transcribed during Arabidopsis embryo development.

Promoter-reporter gene (GUS) fusion study revealed that CDS1 and CDS2 were transcribed during Arabidopsis embryo development, including globular embryo (A, F), heart-shaped embryo (B, G), early torpedo embryo (C, H), late torpedo embryo (D, I) and cotyledon embryo (E, J). Representative images of three independent homozygous lines were shown. Bars = 50 μm.

Fig 3. Suppressed expression of CDS1 and CDS2 resulted in the defective embryonic development.

(A) RT-qPCR analysis of transcriptions of CDS1 and CDS2 genes in Col, various mutants, and cds1 cds2 plants with complemented expression of CDS2 driven by native promoter. ACTIN2/8 was amplified and used as an internal reference to normalize the expressions of CDS1 and CDS2. Experiments were biologically repeated for two times and data were mean ± SE. Statistical significance was determined by student’s t test (**, P<0.01). (B) Embryo development of Col, cds1, cds2, cds1 cds2 mutants and cds1 cds2 plants with complemented expression of CDS2 driven by native promoter (pCDS2:CDS2 in cds1 cds2). Images of cleared seeds were taken by DIC microscopy at 3, 4, 5, 6, 7, 8 days post anthesis (DPA). Representative images were shown. Bars = 50 μm. (C) Quantitative analysis of embryos at different developmental stages (4/8-cell, 16-cell or globular) of siliques at 3 DPA of Col and various mutants. Experiments were biologically repeated and at least 89 seeds were observed. (D) Quantitative analysis of embryos at different developmental stages (globular, triangular or mid-heart) of siliques at 5 DPA of Col, various mutants, and cds1 cds2 plants with complemented expression of CDS2 driven by native promoter. Experiments were biologically repeated for two times. At least 172 or 136 seeds were observed for embryos of cds mutants (upper) or embryos of cds1 cds2 mutants with complemented expression of CDS2 driven by native promoter (bottom), respectively.

Fig 4. Suppressed expression of CDS1 and CDS2 resulted in altered auxin distribution during embryo development.

(A-F) Auxin distribution in embryos at different developmental stages of Col and cds1 cds2 mutant, including globular embryo (A, D), heart-shaped embryo (B, E) and torpedo embryo (C, F). Arabidopsis marker line DR5::GFP or offsprings of DR5::GFP cross with cds1 cds2 (DR5::GFP × cds1 cds2) were analyzed. Representative images were shown. Bars = 25 μm. (G-I) Relative GFP fluorescence intensity of whole embryo (G), cotyledon primordia (H) and hypophysis and suspensor (I) at different developmental stages in DR5::GFP and DR5::GFP × cds1 cds2. Cotyledon primordia and hypophysis and suspensor are indicated by red or white boxes in A-F, respectively. Analyzed embryo numbers (n) from 2 independent experiments were indicated. Values were mean ± SE. Statistical significance was determined by student’s t test (*, p < 0.05; **, p < 0.01).

Fig 5. Disturbed polar localization of auxin efflux carrier PIN1 under suppressed expression of CDS1 and CDS2.

(A-L) Subcellular localization of auxin efflux carrier PIN1 in embryos at different developmental stages of Col and cds1 cds2, including globular embryo (A, G), heart-shaped embryo (B, H) and torpedo embryo (C, I). (D-F) Enlargement of A-C, respectively. Red arrows highlighted the polar localization of PIN1 at plasma membrane. (J-L) Enlargement of G-I, respectively. White arrows highlighted the dot structure of PIN1 in cytoplasm. Arabidopsis line harboring PIN1-YFP fusion protein or offsprings of PIN1-YFP cross with cds1 cds2 (PIN1-YFP × cds1 cds2) were analyzed. Observations were performed with 2 independent lines and representative images were shown. Bars = 25 μm. (M-N) Relative YFP fluorescence intensity (M) and ratio of YFP signal intensity at PM to cytoplasm (N) at different embryo developmental stages in PIN1-YFP and PIN1-YFP × cds1 cds2. Numbers of analyzed embryos or cells (n) from 2 independent experiments were indicated. Values were mean ± SE. Statistical significance was determined by student’s t test (**, p < 0.01).