Involvement of the phospholipid sterol acyltransferase1 in plant sterol homeostasis and leaf senescence

Abstract

Genes encoding sterol ester-forming enzymes were recently identified in the Arabidopsis (Arabidopsis thaliana) genome. One belongs to a family of six members presenting homologies with the mammalian Lecithin Cholesterol Acyltransferases. The other one belongs to the superfamily of Membrane-Bound O-Acyltransferases. The physiological functions of these genes, Phospholipid Sterol Acyltransferase1 (PSAT1) and Acyl-CoA Sterol Acyltransferase1 (ASAT1), respectively, were investigated using Arabidopsis mutants. Sterol ester content decreased in leaves of all mutants and was strongly reduced in seeds from plants carrying a PSAT1-deficient mutation. The amount of sterol esters in flowers was very close to that of the wild type for all lines studied. This indicated further functional redundancy of sterol acylation in Arabidopsis. We performed feeding experiments in which we supplied sterol precursors to psat1-1, psat1-2, and asat1-1 mutants. This triggered the accumulation of sterol esters (stored in cytosolic lipid droplets) in the wild type and the asat1-1 lines but not in the psat1-1 and psat1-2 lines, indicating a major contribution of the PSAT1 in maintaining free sterol homeostasis in plant cell membranes. A clear biological effect associated with the lack of sterol ester formation in the psat1-1 and psat1-2 mutants was an early leaf senescence phenotype. Double mutants lacking PSAT1 and ASAT1 had identical phenotypes to psat1 mutants. The results presented here suggest that PSAT1 plays a role in lipid catabolism as part of the intracellular processes at play in the maintenance of leaf viability during developmental aging.

Figure 1.

Isolation of loss-of-function mutants for PSAT1 and ASAT1. A, Schematic representation (based on The Arabidopsis Information Resource locus data) of T-DNA insertions in PSAT1 and ASAT1 genes. Gray boxes and white boxes indicate nontranslated sequences and exons, respectively. Black lines and dashed black lines indicate introns and other genomic sequences, respectively. T-DNAs appear as inserted frames. Seed stock accessions are as follows: psat1-1, SALK_037289; psat1-2, SALK_117091; asat1-1, GABI_123458. Light gray arrows indicate PCR primers. Primer sequences are given in Supplemental Table S1. B, RT-PCR analysis to confirm loss of function of genes considered here (with Platinum Taq DNA polymerase). Top, PCR with primer pair Pr5Pr6 for the detection of ASAT. Lane 1, the wild type; lane 2, psat1-1; lane 3, psat1-2; lane 4, asat1-1; lane 5, another wild-type individual plant; lanes 6 to 8, three different asatpsat1-1 double mutant individual plants; lanes 9 to 11, PCR on the same RT products as in lanes 6 to 8 using PSAT promoter-specific primers showing no amplification, as a control of genomic DNA-free samples. The 1.5-kb central band of the DNA ladder is marked with a white dot. Bottom, PCR with primer pair Pr1Pr4 for the detection of PSAT1. Lane 12, the wild type; lanes 13 and 16, two distinct psat1-1 individual plants; lanes 14 and 20, two distinct psat1-2 individual plants; lane 15, asat1-1; lanes 17 and 19, empty lanes; lane 18, asatpsat1-1 double mutant (showing the same band as in lanes 13 and 16). C, Another RT-PCR analysis (with standard Taq DNA polymerase) showing the amplification of PSAT1 and ACTIN2 mRNA using primer pairs Pr1Pr4 and Pr9Pr10, respectively. Lane 21, psat1-1; lane 22, psat1-2; lane 23, the wild type. Equal quantities of RNAs were used in the overall experiments presented here. The high abundance of ACTIN2 RT-PCR products compared with the low abundance of PSAT1 RT-PCR products found in young leaf material was confirmed in quantitative real-time RT-PCR experiments (see “Materials and Methods”; Fig. 4D; Supplemental Table S2).

Figure 2.

A to C, SE content of seeds (A), rosette leaves (B), and flowers (C) of wild-type (wt), asat1-1, psat1-1, asatpsat1-1, and psat1-2 lines and of a Pro-35S∷PSAT1 Arabidopsis line in the wild-type background. Data for asatpsat1-2 mutants, which are nearly identical to data for psat1-2, are not shown for simplification of the figure. D, Sterol composition of SE fractions from rosette leaves of wild-type, psat1-1, and psat1-2 lines. For the sampling, lipid extraction, sterol analysis, and sterol content determination, see “Materials and Methods.” n.d., Not determined. Biosynthetic precursors include cycloartenol, 24-methylene cycloartanol, Δ⁷-avenasterol, Δ⁷-sitosterol, 24-methylene cholesterol, and isofucosterol.

Figure 3.

Early leaf senescence phenotype of psat1 mutants from bolting stage on. A, Two and a half months after sowing, three representative rosette leaves from the external verticils of psat1-1 plants (left) and wild-type plants (right). B, Three months after sowing, comparative view of psat1-2 (left) and wild-type (right) plant trays. C to E, Three months after sowing, top view of plants from the psat1-2, wild-type, and Pro-35S∷PSAT1-transformed psat1-1 mutant (named D128) lines, respectively. Stems were cut off for clarity. Note that asat1-1 mutant plants, which are not shown here, have the same morphological phenotype as the wild type (D). Note also that double mutants asatpsat1-1 and asatpsat1-2, which are not shown here, have the same morphological phenotype as psat1-1 and psat1-2 mutant plants, respectively. F, Relative PSAT1 expression measured by quantitative PCR in a Pro-35S∷PSAT1 line in the wild-type (wt) background and in the D128 Pro-35S∷PSAT1-transformed psat1-1 mutant, compared with the wild type.

Figure 4.

Progression of senescence in rosette leaves from wild-type (wt), psat1-1, and psat1-2 lines. Fully expanded rosette leaves of about 4 to 6 cm in length were detached before bolting. A set of 20 to 30 leaves was stored at −80°C (time 0 [t=0] control), and 20 to 30 leaves were placed upon deionized water in several petri dishes in the growth chamber. A, Representative image of this “detached leaf assay” at the end of the time frame chosen for the experiment, showing the top view of wild-type (left) and psat1-2 (right) leaves after 20 d in water. B, Progression of senescence determined by leaf imaging combined with computer-based quantification of greenness and nongreenness. The extent of senescence is expressed as a percentage of nongreen total leaf area (see “Materials and Methods”). A curve from one representative experiment is shown. C, Mutant psat1-1 and psat1-2 leaves were collected for SE measurement at the time they were detached (time 0) and when the nongreen surface reached 30% to 40% of the total leaf surface (10–15 d according to the experiment). Wild-type leaves were collected concomitantly. At that time (10–15 d), they were 10% to 15% nongreen in total surface. Data represent means and sd from three experiments where each sample was analyzed in duplicate. D, Expression in wild-type leaves of lipid biosynthetic genes considered in this study at the same stage (15 d) relative to time 0. Gene expression is calculated relative to ACTIN2 (see “Materials and Methods”).

Figure 5.

Effect of mevalonolactone on liquid medium-grown Arabidopsis plantlets. The psat1-1 and psat1-2 mutants and the wild type (wt) were germinated and grown for 2 weeks on solid Murashige and Skoog medium, then transferred to liquid medium in the absence or presence of mevalonolactone (1, 2, or 3 mm). Results shown are from a representative experiment from three similar ones. A, Top view of the wild-type and psat1-2 plantlets 15 d after transfer. B and C (separated for clarity), SE and free sterol (FS) contents of the three lines versus mevalonolactone concentration.

Figure 6.

Effect of squalene on rosette leaves of Arabidopsis plants. Two-month-old plants from the wild-type (wt), psat1-1, psat1-2, and asat1-1 lines were used. Squalene or squalane (10 μL) was applied three times at 5-d intervals on the rosette center and the five to seven youngest leaves. A, Top view of wild-type, psat1-2, and asat1-1 plants 8 d after the third application (stems were cut off for clarity; note that the psat1-1 plant phenotype is similar to that of psat1-2). B, SE content of control and squalene-treated plants. C, Free sterol (FS) content of control and squalene-treated plants.

Figure 7.

Effect of squalene on cauline leaves of Arabidopsis plants. In these experiments, special care was taken to restrict squalene applications to rosette leaves (excluding the rosette center). Two-and-a-half-month-old plants from wild-type, psat1-1, and psat1-2 lines bearing short growing stems (around 4 cm high) received one application of squalene (5 μL). A to C, Photographs of cauline leaves at the third internode from the bottom of the stem, 25 d after the treatment, of treated wild-type (A), control psat1-1 (B), and treated psat1-1 (C) plants. D to G, At the same stage, lower epidermis peels from cauline leaves were stained with Sudan IV, then observed by light microscopy. Micrographs are from control wild-type plant (D), treated wild-type plant (E), control psat1-1 plant (F), and treated psat1-1 plant (G). Images in A to C, F, and G are representative of both psat1-1 and psat1-2 mutant plants. H, SE, free sterols (FS), and squalene contents of cauline leaves. Two-month-old and 1-week-old plants from wild-type (wt) and psat1-2 lines showing 1- to 3-cm-high stems were treated three times at 5-d intervals by squalene (10 μL). Cauline leaves were collected 25 d after the first treatment, then freeze dried for analysis. Note that the black bars represent the squalene content divided by 10.