HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 282, Issue 7, 4613-4625, February 16, 2007

This Article Abstract Full Text (PDF) Supplemental Data All Versions of this Article: 282/7/4613    most recent M606295200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Fritz, M. Articles by Heinz, E. Search for Related Content PubMed PubMed Citation Articles by Fritz, M. Articles by Heinz, E. Social Bookmarking                      What's this?

Channeling of Eukaryotic Diacylglycerol into the Biosynthesis of Plastidial Phosphatidylglycerol^*

Markus Fritz, Heiko Lokstein^¶, Dieter Hackenberg^||, Ruth Welti^**, Mary Roth^**, Ulrich Zähringer, Martin Fulda¹, Wiebke Hellmeyer, Claudia Ott, Frank P. Wolter^¶¶, and Ernst Heinz

From the Biozentrum Klein Flottbek, Universität Hamburg, Ohnhorststrasse 18, D-22609 Hamburg, Germany, the Max-Planck-Gesellschaft, Generalverwaltung, Hofgartenstrasse 8, D-80539 München, Germany, the ^¶Institut für Biochemie und Biologie, Universität Potsdam, Pflanzenphysiologie, Karl-Liebknecht-Strasse 24–25, D-14476 Golm, Germany, the ^||Institut für Biologie/Pflanzenphysiologie, Humboldt-Universität zu Berlin, Unter den Linden 6, D-10099 Berlin, the ^**Division of Biology, Kansas State University, Kansas Lipidomics Research Center, Manhattan, Kansas 66506-4901, the Leibniz-Zentrum für Medizin und Biowissenschaften, Forschungszentrum Borstel, Parkallee 4, D-23845 Borstel, Germany, the Albrecht-von-Haller-Institut für Pflanzenwissenschaften, Georg-August Universität Göttingen, Biochemie der Pflanze, Justus-von-Liebig-Weg 11, D-37077 Göttingen, Germany, and the ^¶¶Bundesverband Deutscher Pflanzenzüchter, GVSmbH, Kaufmannstrasse 71–73, D-53115 Bonn, Germany

Received for publication, June 30, 2006 , and in revised form, December 8, 2006.

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Plastidial glycolipids contain diacylglycerol (DAG) moieties, which are either synthesized in the plastids (prokaryotic lipids) or originate in the extraplastidial compartment (eukaryotic lipids) necessitating their transfer into plastids. In contrast, the only phospholipid in plastids, phosphatidylglycerol (PG), contains exclusively prokaryotic DAG backbones. PG contributes in several ways to the functions of chloroplasts, but it is not known to what extent its prokaryotic nature is required to fulfill these tasks. As a first step toward answering this question, we produced transgenic tobacco plants that contain eukaryotic PG in thylakoids. This was achieved by targeting a bacterial DAG kinase into chloroplasts in which the heterologous enzyme was also incorporated into the envelope fraction. From lipid analysis we conclude that the DAG kinase phosphorylated eukaryotic DAG forming phosphatidic acid, which was converted into PG. This resulted in PG with 2–3 times more eukaryotic than prokaryotic DAG backbones. In the newly formed PG the unique 3-trans-double bond, normally confined to 3-trans-hexadecenoic acid, was also found in sn-2-bound cis-unsaturated C18 fatty acids. In addition, a lipidomics technique allowed the characterization of phosphatidic acid, which is assumed to be derived from eukaryotic DAG precursors in the chloroplasts of the transgenic plants. The differences in lipid composition had only minor effects on measured functions of the photosynthetic apparatus, whereas the most obvious phenotype was a significant reduction in growth.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
All subcellular membranes of plant cells contain phosphatidylglycerol (PG),² but it predominates in chloroplasts and, therefore, PG is characteristically associated with photosynthetically active cells (1). Its synthesis is confined to membranes of the endoplasmic reticulum (ER) and the inner membranes of the plastid envelope and mitochondria (2). In algae and higher plants, several mutants have been isolated that are affected in the synthesis of thylakoid PG, resulting in reduced or no detectable PG. The availability of such mutants enabled the investigation of the specific functions of PG in thylakoid membranes (2). A progressive loss of thylakoid PG is paralleled by a reduction in chlorophyll content and impairment of photoautotrophy. In algae, the loss of PG can be complemented by exogenous PG, whereas survival of corresponding Arabidopsis mutants requires sucrose in the growth medium. Detailed studies of these mutants have shown that PG of thylakoid membranes may have functions more specific than its contribution to the lipid matrix and its negative surface charge. Thylakoid PG specifically interacts with different pigment-protein complexes and favors oligomerization of photosystem I (PSI), photosystem II (PSII) and the main light-harvesting complex LHCII (2). The functional interactions of PG with PSI, PSII, and LHCII are supported by the identification of specifically bound PG molecules in the crystallized forms of the three complexes (3–6). Different functions have been ascribed to PG in envelope membranes. At the beginning of preprotein import into plastids, PG may favor the interaction of transit peptides with the outer envelope membrane (7). PG may also contribute to galactolipid synthesis by activation of the monogalactosyldiacylglycerol (MGD) synthase in the envelope (8). Furthermore, a critical contribution of PG to the physical state of plastidial membranes is deduced from the fact that the degree of fatty acid unsaturation in thylakoid PG, particularly the proportion of molecular species with pairs of the so-called high melting fatty acids palmitic, stearic, and 3-trans-hexadecenoic acid (16:0, 18:0, and 16:1t), is correlated with the occurrence of cold sensitivity in many plant species (9).

This last observation is related to a particular aspect of the structure of plastidial PG. Due to the fatty acid residues found in the sn-1 and sn-2 positions of its diacylglycerol (DAG) moiety, the hydrophobic backbones of PG show a small, but characteristic difference from DAG present in plastidial glycolipids. Because the origin of this difference is vital for the understanding of our approach, a short outline of the structure and assembly of DAG backbones in plant membrane lipids will be given. Depending on the organism, its taxonomic position, and developmental stage, plastidial glycolipids vary widely in the proportion of DAG species containing either C16- or C18-fatty acids at the sn-2 position. Because the corresponding DAG both carry a similar mixture of C18/C16:0 fatty acids at the sn-1 position, they only differ by the diagnostically relevant sn-2-bound fatty acids. DAG backbones with sn-2-C16-fatty acids are called "prokaryotic," and those with sn-2-C18-fatty acids are "eukaryotic," drawing attention to the similarity between plastidial and cyanobacterial prokaryotic lipid backbones (10), and the contrast between those and typical eukaryotic DAG. The fatty acid signature of DAG at the sn-2 position is due to the fact that the substrate specificity of isoenzymes of 1-acylglycerol-3-phosphhate acyltransferase differs significantly between the ER and the plastid. Whereas the plastidial enzyme introduces 16:0 at the sn-2 position to generate a prokaryotic signature the isoenzyme of the ER strongly prefers 18:1 to be incorporated at sn-2 position. Therefore, the sn-2-bound fatty acid indicates the subcellular origin and trafficking of DAG moieties of lipids finally targeted to thylakoid membranes (11, 12). The prokaryotic DAG moieties are assembled in the inner envelope membrane from fatty acids, which have never left the plastid (see Fig. 1, plastidial pool of primary DAG). Plant species with a predominance of prokaryotic DAG in MGD are called 16:3 plants, because 16:0, in the course of DAG assembly introduced into the sn-2 position, is desaturated as part of MGD to all-cis-7,10,13-hexadecatrienoic acid (16:3). In contrast, the only and characteristic modification of the sn-2-bound 16:0 in plastidial PG of all eukaryotic plants is desaturation to 16:1t (see Fig. 1) (13).

In angiosperms, 16:3 plants are a minority compared with 18:3 plants, which contain MGD with more or less exclusively eukaryotic DAG (14). Eukaryotic DAGs are assembled from 16:0 and oleic acid (18:1), which have been exported from plastids to ER membranes for incorporation by ER enzymes into phospholipids. This extraplastidial reaction sequence results in eukaryotic DAG backbones. As a constituent of ER phospholipids, the lipid-bound 18:1 may be desaturated by ER enzymes to linoleic (18:2) and linolenic acid (18:3), whereas 16:0, nearly exclusively found in the sn-1 position of eukaryotic phospholipids, is hardly desaturated by plant ER enzymes. A fraction of these ER-assembled lipids is transferred to plastids, where eukaryotic DAGs of different degrees of C18-desaturation become available for glycolipid formation in envelope membranes (plastidial pool of secondary DAG, see Fig. 1). Four different components have been suggested to function as the lipophilic metabolites transported between ER and chloroplast envelopes (11, 12): phosphatidylcholine (PC), DAG, lyso-PC (LPC (15)), and phosphatidic acid (PA (16)). Therefore, depending on the nature of the imported precursor, different reactions would be required for release or formation of eukaryotic DAG in envelope membranes (15–20). The resulting DAGs are incorporated into eukaryotic glycolipids by glycosyltransferases located in both inner and outer envelope membranes and finally transferred into thylakoids (21). Because of their small size and rapid turnover, plastidial pools of primary and secondary DAG in envelopes can hardly be detected in unlabeled form in vivo or in isolated chloroplasts (11).

In contrast, in isolated envelopes a large pool of tertiary DAG (see Fig. 1), containing highly unsaturated DAG of both pro- and eukaryotic structure is present (22). Its formation is ascribed to the action of galactolipid:galactolipid galactosyltransferase. This enzyme of the outer envelope membrane (23) is only activated in stress situations such as ozone fumigation of leaves (24), release of envelopes from isolated chloroplasts (22), and in tgd1 and tgd2 mutants of Arabidopsis (25). These envelope proteins, considered to be part of a larger lipid transfer complex (16, 25, 26), represent the only components identified so far as being involved in the transport of lipids between ER and thylakoids.

Most cyanobacteria, considered to represent phylogenetic ancestors of chloroplasts, contain exclusively prokaryotic thylakoid lipids, which are mainly imported from the plasma membrane (10). The advantage of detour through the ER in the biosynthetic pathway for plastidial lipids in eukaryotic plants is not clear yet, but its evolution necessitated significant additions and relocations of proteins, compared with the more simple cyanobacterial and prokaryotic-plastidial systems. In view of the complexity of the detour in lipid biosynthesis to and from the ER and the required cooperation that has evolved between the ER and the plastid, it is surprising that the use of eukaryotic DAG does not extend to the synthesis of plastidial PG. Although PG in chloroplasts of both 16:3 and 18:3 plants contains exclusively prokaryotic DAG with 16:0 and 16:1t in the sn-2 position (27), plant mitochondrial PG contains both eu- and prokaryotic DAG backbones (28), and PG in plasma membranes is nearly exclusively of eukaryotic structure (29). This raises the question as to whether only PG of prokaryotic nature with its characteristic fatty acids can fulfill the functions in plastids mentioned above, or whether other factors have prevented the establishment of a flow of eukaryotic DAG into plastidial PG.

Several of the aforementioned specific functions of PG can be fulfilled by eukaryotic PG as shown by the complementation of PG mutants of cyanobacteria resulting from the addition of eukaryotic dioleoyl-PG (30). This is in line with the formation of eukaryotic lipids by cyanobacteria in the presence of 18:2, which resulted in large proportions of eukaryotic backbones in both glycolipids and PG without affecting growth (31). On the other hand, only PG with 16:1t contributes specifically to the oligomerization of LHCII and possibly grana stacking (32). But this antenna system is absent from cyanobacteria, which do not form 16:1t in their prokaryotic PG (10). In the crystalline LHCII complex, the 16:1t group of prokaryotic PG was suggested to be bound in a hydrophobic pocket in close proximity and in parallel to violaxanthin, which is involved in plant photoprotection via the xanthophyll cycle (5). On the other hand, a 16:1t-free mutant of Arabidopsis had a normal growth phenotype at ambient temperature (33).

View larger version (33K): [in this window] [in a new window]   FIGURE 1. Major steps in the formation and the use of primary, secondary, and tertiary DAGs in the plastid envelope. Lipids of prokaryotic origin are on the left (green), and those of eukaryotic origin on the right (blue). The import of ER-derived precursors is indicated at the right. Green arrows indicate the reactions responsible for the biosynthesis of prokaryotic PG. Red arrows indicate possible reactions induced by the expression of the heterologous DAGK activity. These reactions result in the additional formation of eukaryotic PG (red). Pools of primary (green) and secondary DAG (blue, both as heavily lined boxes) are involved in steady-state lipid biosynthesis. Tertiary DAG (gray) and polygalactosyl diacylglycerols with the 18:3/16:3 species (here exemplified by trigalactosyldiacylglycerol = TGD) are only formed under specific conditions (dotted black arrows, see text). The positional distribution of fatty acids in the various lipids is indicated (sn-1 position/sn-2 position). Fatty acids are characterized by number of carbon atoms:number of double bonds (x = 1-3); t = one of the double bonds is the 3-trans-double bond. For several lipids, the most characteristic molecular species are given representing the fully desaturated forms (for example 34:4 for control PG). They may be compared with the profiles shown in Figs. 5 and 6. CDP-DAG, cytidine diphosphate diacylglycerol; PGP, phosphatidylglycerol phosphate. Further details are described in a recent review (12).

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Plant Material and Growth Conditions—Tobacco plants (Nicotiana tabacum L. var. Wisconsin-38) were grown in a Phytotron (16-h light of 50–180 µEm^-2 s^-1 at 22 °C/8 h dark at 18 °C, relative humidity 70–80%) for regeneration of transformants and in a greenhouse for generation of homozygous lines by repeated rounds of self-pollination after germination on selective agar plates (100 mg of kanamycin/liter) and final growth of plants on soil to be used for lipid analysis. Seeds of Tecoma stans (L.) H.B.K. (synonymous with Bignonia stans, Stenolobium stans, family Bignoniaceae) for the isolation of reference fatty acids were purchased from Sunshine Seeds.

Vector Construction and Plant Transformation—The coding region of the DAGK gene (dgk) from Escherichia coli was amplified by PCR from pVL1 (38) using the primers 5'-CCATGGCCAATAATACCACTG-3' and 5'-GTCGACTTATCCAAAATGCGACC-3' with the concomitant creation of an NcoI site and an SalI restriction site, both cut and blunted by filling in of the amplified fragment. The potato sequence encoding the promoter and the transit peptide from a ribulose-bisphosphate carboxylase/oxygenase small subunit gene was derived from p1H80 (39). The SphI site, located at the 3'-end of the transit peptide coding region, was cut and blunted by filling in. The linearized vector and the dgk sequence were blunt-end-ligated leading to an in-frame fusion of the transit peptide and the dgk coding regions as verified by DNA sequencing. An SalI fragment of this vector comprising the promoter/transit peptide/dgk sequence was inserted into pPCV 720 (40) upstream of the octopin synthetase terminator. The whole construct was then transferred as HindIII fragment into a modified pBI 121 vector (41), which contains the intron-interrupted -glucuronidase (GUS-int) sequence (42) for use as a screenable marker. In this pBI 121-based vector (pMF2) the dgk construct is flanked by the neomycin phosphotransferase II (NPTII) and the -D-glucuronidase (GUS) sequence (see Fig. 2C). As a control the corresponding vector without the chimeric dgk sequence was used (pMFc). Both were transformed separately into Agrobacterium tumefaciens GV3101.

The transformation of N. tabacum was carried out by cocultivation of leaf discs with A. tumefaciens (43) carrying either pMFc or pMF2. The regenerated plants (89 independent pMF2 transformants and 39 independent pMFc transformants) were grown in a Phytotron on agar medium in the presence of kanamycin (100 mg/liter) and carbenicillin (125 mg/liter) and transferred to fresh medium every 3–4 weeks. Representative plants from these groups were used for a two-step selection (see below) following growth in pots on soil under non-sterile conditions.

After growth establishment, transformants were subjected to a tissue assay for GUS activity. In the case of the pMF2 transformants, 68 positive plants were recognized from which 50 were used to assay DAGK activity with leaf extracts. Based on these results, 11 plants with the highest activities were selected for generation of homozygous single locus lines.

Green Fluorescent Protein Constructs and Biolistic Transformation—The constructs for transient expression of green fluorescent protein fusion proteins were established by using pCAT vectors as described before (44). The coding region of the DAGK gene (dgk) from E. coli, preceded by the transit peptide from a ribulose-bisphosphate carboxylase/oxygenase small subunit gene, was amplified by PCR from pMF2 by utilizing the primers 5'-TCCATGGCTTCCTCTGTTATTTCCTC-3' and 5'-TCCATGGATCCAAAATGCGACCATAACAGA-3'. This pair of primers introduced NcoI restriction sites on both ends and removed the stop codon of the original sequence. This fragment was cut with NcoI and ligated to an NcoI-digested pCAT vector generating an in-frame fusion of the DAGK to EYFP (Clontech, Heidelberg, Germany). The correct orientation of the fragment was confirmed by restriction digestion, and the correct sequence was confirmed by sequencing. The construct is shown schematically in Fig. 2D. Transient transformation of onion and tobacco epidermal cells was accomplished by particle bombardment as described before (44).

Images of EYFP-expressing cells were acquired using a confocal laser-scanning microscope (Zeiss LSM 510). Both chlorophyll and EYFP were excited with the 488 nm line of an argon laser (25 milliwatts) using 7 and 58% powers, respectively. Chlorophyll emission was monitored using a 560 nm long pass filter, whereas for EYFP a 505–530 band pass filter was used.

General Procedures—Total protein was determined according to Bradford (45), and GUS tissue assays (46) were carried out with small pieces (0.5 cm²) of leaves. For Southern blots with genomic DNA extracted from leaves (47) and cut by EcoRI and HindIII, the dgk fragment was amplified by PCR with primers slightly longer than those described above (5'-TTTTTCCCGGGATGGCCAATAATACCACTGGATT-3' and 5'-TTTTTCCCGGGTTATCCAAAATGCGACCATAACAG-3') to be used as digoxigenin-labeled probe following the instructions of the synthesis kit (Roche Applied Science). A digoxigenin-labeled NPTII probe (48) was prepared in the same way. For the data presented in Fig. 3, lipophilic pigments were extracted from leaf discs (1.3 cm²) with acetone/25 mM aqueous Na₂HPO₄ (4/1, v/v, 2 ml) (49) and chlorophylls (Chls) a and b measured spectrophotometrically (50). Intact chloroplasts were isolated on Percoll gradients and used for the subsequent isolation of thylakoids and envelope membranes by sucrose gradient centrifugation as described before (51). Lipids from leaves and chloroplasts were extracted and subjected to various analytical procedures as detailed before (22, 52–54). In particular, the lipid profiling method was identical to that employed by (54), except for the additional analysis of sulfoquinovosyl diacylglycerol (SQD), which was scanned using the parameters previously defined (55), using hydrogenated SQD species as internal standards.

Chlorophyll Fluorescence Analyses—In vivo Chl fluorescence was measured with attached leaves as previously described (56). Prior to fluorescence measurements, plants were dark-adapted and equilibrated to 10, 20, or 30 °C for 1 h and kept at the same temperature during the experiments. Fluorescence parameters assessed were: F_v/F_m = (F_m - F₀)/F_m, the maximum photochemical efficiency of PS II in the dark-adapted state. Photochemical quenching was quantified as: qP = (F_m'-F_t)/(F_m'-F₀'). Photon flux densities were measured using a quantum sensor (LI-189A, Li-Cor, Lincoln, NE).

Non-denaturing Gel Electrophoresis ("Green Gels")—Thylakoid membranes were isolated from freshly harvested leaves by grinding in buffer (0.4 M sorbitol, 20 mM Tricine, pH 7.8, 20 mM NaCl, 5 mM MgCl₂, and 250 µM phenylmethylsulfonyl fluoride), washed once in storage buffer (10 mM Tricine, pH 7.8, 10 mM NaCl, 10 mM MgCl₂), diluted to 1.1 mg of Chl a+b/ml with 7 mM Tris, 53 mM glycine, and 10% (v/v) glycerol, and then stored in liquid nitrogen. Pigment-protein complexes were separated following short solubilization with varying concentrations of n-dodecyl -D-maltoside (Glycon, Luckenwalde, Germany) on 7% polyacrylamide gels, pH 8.6, containing 12.4 mM Tris and 48 mM glycine. The stacking gel was 4% polyacrylamide. The running buffer (pH 8.3) contained 12.4 mM Tris, 96 mM glycine, and 0.2% (w/v) Deriphat 160 (disodium n-lauryl iminopropionate, a kind gift of Henkel Co.). Pre-cooled gels were run for 30 min at 100 V and room temperature.

Isolation of Reference 3-trans-C18 Polyunsaturated Fatty Acids—Lipids from seeds of T. stans were extracted as described before and used for preparation of methyl esters by acidic methanolysis (57). After extraction into petroleum ether and purification by TLC on Kieselgel 60 plates in petroleum ether/diethyl ether 85/15 (v/v), they were separated by preparative high-performance liquid chromatography on a Multospher column 100 RP18-5 (250 x 4 mm, www.cs-chromatographie.de). Components were eluted by a linear gradient of acetonitrile/water 8/2 (v/v) to pure acetonitrile in 30 min followed by 30 min of pure acetonitrile at a flow rate of 0.7 ml/min and manual collection of fractions based on detection at 204 nm. The purity of the isolated components was checked by gas-liquid chromatography (GLC). The same procedure was applied to the methyl ester mixture obtained from the sn-2 position of PG from transgenic tobacco. Proton nuclear magnetic resonance spectra (¹H NMR spectra) of the isolated methyl esters were recorded at 600 MHz (Bruker Avance DRX 600) in CDCl₃. For the determination and positional localization of double bonds, fatty acid methyl esters (200 µg) were first converted into their 4,4-dimethyloxazoline derivatives followed by GLC/electron impact-mass spectrometry as described before (53).

Assay of DAGK Activity—The activity measurement is based on the formation of ¹⁴C-labeled PA from [¹⁴C]acyl-labeled sn-1,2-DAG following a previously described method (58). The labeled substrate was prepared from di-[1-¹⁴C]oleoylphosphatidylcholine (specific radioactivity 105 µCi/µmol; from Amersham Biosciences) by enzymatic hydrolysis with phospholipase C (P7147 from Sigma) and subsequent purification by TLC in petroleum ether/diethyl ether 3/1 (v/v). The isolated [¹⁴C]DAG (237 dpm/pmol) was dissolved in chloroform/methanol 1/1 (v/v) (60,000 dpm/5 µl) and stored at -20 °C. Plant extracts were prepared by homogenizing tissue in buffer (1:2, w/v, 50 mM PIPES, 50 mM NaCl, 12.5 mM MgCl₂, adjusted to pH 6.8) with the aid of an Ultraturrax homogenizer. The homogenate was filtered through two layers of Miracloth and stored on ice until use. For the actual assay, solvent was evaporated from a solution of [¹⁴C]DAG (60,000 dpm) and cardiolipin (75 µg, Sigma) in an Eppendorf tube, and the lipids were suspended in 50 µl of buffer (double concentration as given above and including 102 mM n-octyl -D-glucopyranoside) by sonication (bath-type instrument, 10 min). The final assay volume was 100 µl and contained in addition the following components: dithiothreitol (2 mM), protein (0.1–5.0 µg), and ATP (5 mM), which was used to start the reaction. After 15 min at room temperature the reaction was stopped by addition of chloroform/methanol (1/1, v/v, 230 µl, containing 100 µg of carrier PA) and KCl solution (1 M in 0.2 M H₃PO₄ (52)). After vortexing and a short centrifugation, the subphase was withdrawn, blown to dryness, redissolved in chloroform/methanol (1/1, v/v, 50 µl), and subjected to TLC in chloroform/methanol/25% aqueous ammonia (7/4/1, v/v). After development and UV detection following spraying with anilinonaphthalene sulfonate (0.1% in methanol), the PA zone was scraped off and used for scintillation counting. In blanks and assays with control plants, 20–60 dpm was recovered in the PA zones. Under the given conditions the assay, linear up to 5µg of total protein, was useful for measuring widely varying activities. After incubation for 15 min, an activity of 1 nmol/min/mg of protein would correspond to a recovery of 3555 dpm in PA, assuming no dilution by endogenous DAG.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Transgenic Plants Are Smaller, Show DAGK Activity in the Envelope Fraction, and Contain Less Plastidial Lipids—The DAGK of E. coli (dgk) is a small protein of 121 amino acids that, due to its hydrophobic character, can insert spontaneously in active form into membranes (59). For expression of this protein in chloroplasts we transformed tobacco with a chimeric construct encoding a fusion of the bacterial DAGK with an N-terminal leader peptide derived from the small subunit of ribulose-bisphosphate carboxylase/oxygenase. The expression of the resulting preprotein was driven by the corresponding small subunit promoter. Both the leader peptide and its genuine promoter were derived from potato genomic DNA (39). In the finally transferred DNA the bacterial DAGK sequence was enclosed by the NPTII and the GUS sequence (Fig. 2C) to allow consecutive selection and screening steps. Regeneration and rooting of candidate transformants was carried out in the presence of kanamycin followed by a screening for GUS activity of kanamycin-resistant plants. Leaf extracts of these GUS-positive plants were subjected to in vitro DAGK assays. Based on these results, several independent transformants with the highest DAGK activity and a few control transformants (co) were selected.

View larger version (83K): [in this window] [in a new window]   FIGURE 2. A, size differences between wild-type (wt), empty vector control (co), and DAGK-expressing transgenic tobacco plants (t1 and t2) after growth for 6 weeks in soil. The transferred DNA (C) encodes as a central unit the octopin synthetase (OCS) terminator, the bacterial diacylglycerol kinase (dgk) linked in N-terminal fusion to the leader peptide (TP) of the small subunit of the ribulose-1,5-bisphosphate carboxylase/oxygenase and its genuine promoter (Pro). This block is flanked by units for constitutive expression of markers for selection (NPTII) and screening (GUS-int), all contained between left (LB) and right (RB) borders. In the text, t1 and t2 are referred to as transgenic transformants, whereas co (despite being transgenic, but lacking the central dgk block) is simply referred to as the control. B, subcellular localization of the DAGK as fusion protein with EYFP transiently expressed in a leaf hair cell of tobacco. The upper panel shows fluorescence of EYFP in green and autofluorescence of the chlorophyll in red. The lower panel shows the merged pictures. The scale bar represents 10 µm. The fusion construct, including the N-terminal extension by a plastid leader peptide and the C-terminal EYFP is shown in D. Details are given under "Materials and Methods."

The two transgenic lines showed significantly reduced and stunted growth as compared with the control (Fig. 2A). The content of total Chl per unit leaf area was reduced in both transformant lines (Fig. 3), which, particularly in one transformant (t2, Fig. 2), resulted in a pale green appearance as compared with the control. On the other hand, the Chl a/Chl b ratio was increased in both transgenic lines (Fig. 3) pointing to a possible decrease in antenna size. In leaf extracts of control plants, the proportions of individual lipids as well as of their fatty acid patterns were in agreement with studies carried out before (62–64), and details will not be presented again. Compared with the control, the leaf lipids of the two transgenic lines had reduced proportions of plastidial glycolipids and an increase in all eukaryotic phospholipids. The greater proportion of PC in the transgenic lines was particularly obvious (Fig. 3), but there were only minor changes in its fatty acid composition (data not shown). On the other hand, the MGD of the two transgenic lines was compositionally altered, because the proportion of 16:3 was reduced to about half of the control value (Fig. 3). An increase in the proportion of extraplastidial phospholipids has been observed before in several mutants impaired in the synthesis of plastidial MGD and PG (65–67). This effect has been attributed to the reduced contribution of plastidial lipids to total cellular lipids as a consequence of the reduction in the area of thylakoid membranes. A similar effect may contribute to the changes observed in our transgenic tobacco plants. Because the goal of the present investigation was modification of thylakoid PG, in the work described herein we focused on lipid analyses of thylakoid membranes. A detailed interpretation and a satisfactory understanding of changes observed in the extraplastidial lipids of the transgenic tobacco plants (data not shown) will require additional studies and the use of different techniques.

Expression of Heterologous DAGK in Fluorescent Form Results in Labeling of Plastids—To confirm the targeting of the heterologous DAGK into plastids and to see whether extraplastidial sites may contain significant proportions of DAGK, transient expression experiments were performed utilizing an EYFP fusion protein. The chimeric construct contained the bacterial DAGK with the N-terminal leader peptide of the small subunit of ribulose-bisphosphate carboxylase/oxygenase and EYFP fused to the C terminus (Fig. 2D). The construct was transferred into tobacco cells by biolistic transformation. The transient expression of the fusion protein revealed colocalization of the fluorescence of EYFP with those of the chlorophyll (Fig. 2B) indicating an effective targeting of the DAGK to plastids. Minor amounts of the EYFP signal were detected within the cytoplasm (Fig. 2B). Comparable transient expression experiments with onion epidermal cells revealed plastids with filamentous structures known as stromules as they are described for chlorophyll-free plastids (68) (data not shown). Therefore, the labeling studies suggest that the heterologous DAGK is concentrated in plastids and that extraplastidial localization may represent only a small proportion of the total cellular activity.

View larger version (16K): [in this window] [in a new window]   FIGURE 3. Comparison of leaves from control plants (white columns) and the two transgenic tobacco lines (gray columns, left t1, and right t2). The various parameters used were chl = micrograms of total chlorophyll/cm2 (n = 17; according to the t test, only t2 differs significantly from the control); a/b = ratio of chlorophyll a to chlorophyll b (n = 16; according to the t test, the two transgenic lines do not differ significantly from each other, but they differ from the control); MGD, DGD, and PC = three characteristic leaf lipids (as mol% summing up to 100%, n = 8); 16:3MGD = the proportion of 16:3 in MGD fatty acids (in mol%, n = 8).

We then looked at the molecular species of thylakoid glycolipids. In MGD of the transgenic lines the proportion of the prokaryotic 18:3/16:3 species (34:6, individual acyl species verified by positional analysis) was reduced, whereas the eukaryotic 18:3/18:3 species (36:6) was increased (Fig. 5). This is consistent with the decreased proportion of 16:3 in MGD fatty acids mentioned above (Fig. 3). In DGD the proportion of the eukaryotic 18:3/18:3 species was also increased (Fig. 5), whereas the reduction in SQD (Fig. 4) affected all species to similar extents.

View larger version (13K): [in this window] [in a new window]   FIGURE 4. Thylakoid lipids from control (white columns) and the two transgenic tobacco lines (gray columns). The data were obtained from a series of precursor and neutral loss scans by electrospray tandem mass spectrometry as described under "Materials and Methods." Individual lipids are shown in mol%. The absence of PE is an accepted marker for the purity of chloroplasts.

Phosphatidic Acid Molecular Species Differ Significantly between Control and Transgenic Plants—As outlined above, lipophilic metabolites from the ER are converted in the envelope into secondary DAG of eukaryotic structure (Fig. 1). The expression of the heterologous DAGK was intended to direct part of the eukaryotic DAG via PA into PG. Due to their small size, several of the pools involved in these conversions have never been determined in unlabeled form (11, 12). Based on previous labeling experiments (71), the presence of potential DAG-precursor phospholipids (PC, LPC, or PA) in the thylakoid fraction from control plants can be ascribed to contamination by envelope membranes. But despite the low abundance of some of these precursors in the thylakoid preparations (Fig. 4), the lipidomics technique allowed the identification of the predominant molecular species in these pools. The reliability of this approach was supported by the differences observed between control and transgenic samples.

View larger version (14K): [in this window] [in a new window]   FIGURE 5. Glycolipid molecular species from thylakoids of control (white) and the two transgenic tobacco lines (gray). The data were obtained from a series of precursor and neutral loss scans by electrospray tandem mass spectrometry as described under "Materials and Methods." The proportions (in mol%) of individual species in the profiles of each glycolipid sum up to 100%. The species are characterized by the sum of carbon atoms:double bonds of the two fatty acids. The 18:3/18:3 species (36:6) are of eukaryotic origin, whereas the 18:3/16:3 species in MGD (34:6) is prokaryotic. Positional analysis has shown (data not presented) that the 34:3-combination of SQD is predominant the eukaryotic 16:0/18:3 species (66% in the control, 80% in the transgenic lines). It should be pointed out that the 16:0/18:3 species (34:3) represents a minor proportion in the two galactolipids.

PA is the immediate precursor for the final steps of PG synthesis. Before discussing the PA molecular species data in detail, the differences between control and transgenic plants with regard to this metabolite should be recalled (see Fig. 1): in control plants, PA in the envelope represents a primary pool of plastidial origin (prokaryotic PA) to which a pool of directly imported eukaryotic PA has recently been added (16). In the transgenic plants, another pool of eukaryotic PA may be present. According to the strategy outlined above, it should result from DAGK-dependent phosphorylation of secondary DAG, which may be formed from any of the postulated transport metabolites. The characterization of the putative more upstream precursors LPC and PC, of which plastidial LPC has not been described before, is presented in the supplemental material.

In the control sample (Fig. 6A) PA is dominated by the prokaryotic 18:1/16:0 combination (34:1, 75 mol%). The location of 16:0 in the sn-2 position of PA and the prokaryotic nature were deduced from the high intensity of the released 16:0 acyl ion fragment as compared with the 18:1 acyl ion (Ref. 55, data not shown). This profile, observed in nearly identical form in labeling studies with isolated chloroplasts (71), is considered as the envelope pool of primary PA in its steady state. Because of its low proportion and similar to LPC (supplemental data), this PA was always difficult to analyze in unlabeled form by conventional methods. The 0.08 mol% PA in thylakoid glycerolipids as found here for controls also approach the limits of the lipidomics technique. Nevertheless, the large differences in the profiles between the PA from control and transgenic thylakoid preparations (Fig. 6, A and B) show the reliability of both the thylakoid isolation and the lipidomics analysis. The eukaryotic species (36:x) detected in low proportions in the PA from control thylakoids may represent part of the PA directly imported from the ER (16) and to be used for synthesis of eukaryotic glycolipids. The low proportion of these eukaryotic PA species may indicate an efficient channeling to the PA phosphatase preventing the build-up of a large pool of eukaryotic PA in the envelope.

In thylakoids of the two transgenic lines (Fig. 6B), the PA levels are increased severalfold (from 0.08 to 0.26 and 0.44 mol%) in line with DAG phosphorylation by the heterologous DAGK. Therefore and in contrast to the control sample, most of the PA from transgenic thylakoids may be considered as portraying the pool of secondary DAG in the phosphate-tagged form (secondary PA). This is supported by the profiles of molecular species, which are similar in the two transgenic lines but completely different from the control (Fig. 6). In the transgenic lines, the prokaryotic 34:1(18:1/16:0) species is still a significant component (as evident from the 18:1- and 16:0-acyl ion intensities). But it is accompanied by additional 16:0/18:x species of similar abundance (most likely eukaryotic) and a prominent block of eukaryotic 18/18 species. The detection of the 16:0/18- and 18/18-DAG backbones in PA reflects their presence in the pool of secondary DAG, which represents an independent and non-radioactive analysis of this important pool. Previous analyses had to be based on radiolabeling of isolated chloroplasts in the absence of UDP-galactose (71), whereas patterns deduced from labeling of galactolipid molecular species in leaves (72) were more complicated due to the desaturation-dependent changes in these patterns. From our lipidomics analysis of plastidial PA we conclude that the heterologous DAGK had access to the pool of secondary DAG and thus increased the eukaryotic species of plastidial PA. The extent of their incorporation into PG will be evident from the molecular species detected in PG as described in the following.

View larger version (24K): [in this window] [in a new window]   FIGURE 6. Molecular species of PA and PG from thylakoid preparations of control (A) and transgenic plants (B). The data in this figure were obtained from a series of precursor and neutral loss scans by electrospray tandem mass spectrometry as described under "Materials and Methods." The proportions (in mol%) of individual species of the different profiles sum up to 100%. The molecular species are characterized by the sum of carbon atoms:double bonds of the fatty acid pairs. In the panel at the bottom of B, the eukaryotic part of PG is shown (PGeu). It results from the subtraction of the calculated proportion of prokaryotic PG (see A) from the PG of the transgenic lines, which represent mixtures of both pro- and eukaryotic structure.

In the PG samples of the two transgenic lines, the prokaryotic fraction was significantly reduced. As seen from the high proportions of sn-2-bound C18 fatty acids (Fig. 7B), 66–82 mol% of the total PG consisted of eukaryotic species as analyzed from different preparations. Therefore, many molecular species with both 18/16 and 18/18 backbones were revealed by mass spectrometry (Fig. 6B). All of these species, except two fully desaturated species, i.e. the prokaryotic 34:4 as explained above and the eukaryotic 36:7 to be explained below, are present in the precursor PA. From these profiles we conclude that the heterologous DAGK has access to the pool of secondary DAG in the envelope and that the derived secondary PA is used for PG synthesis. Because PG is a substrate for desaturation, the PA species profile is shifted toward higher unsaturation in PG as also observed in the control.

With the knowledge of the profile of prokaryotic PG species in the control (Fig. 6A) and the assumption that the relative proportions of this block are not altered in the transgenic lines, the proportion of prokaryotic species can be calculated and subtracted from the transgenic PG samples, because the purely prokaryotic 34:4 species can be used as a marker (Fig. 6B). After subtraction of the prokaryotic from the total PG species, the resulting profiles represent the eukaryotic PG species formed in the transgenic lines (panel at the bottom of Fig. 6B). About one-third of this eukaryotic PG is represented by 16:0/18:x species and two-thirds by 18:x/18:x species. Similar proportions of these two eukaryotic blocks were found in the precursor PA of the transgenic plants. From this we conclude that hardly any selection interferes with the channeling of eukaryotic 16:0/18:x species from the pool of secondary DAG via PA into PG. This contrasts with the restricted use of this block of DAG precursors for the synthesis of eukaryotic galactolipids as evident from Fig. 5. A similar comparison with imported PA as the source of eukaryotic DAG (16) in control plants cannot be given, because the proportion of eukaryotic PA species was too low for a reliable characterization as mentioned above.

View larger version (18K): [in this window] [in a new window]   FIGURE 7. Positional distribution of fatty acids in thylakoid PG from control and transgenic plants as revealed by enzymatic cleavage and subsequent GLC analysis. A, fatty acids in the sn-2 position of control PG; B, fatty acids in the sn-2 position of PG from a transformant; and C, fatty acids in the sn-1 position of this transformant. The fatty acids are characterized by numbers of carbon atoms:double bonds. The sn-2 position of control PG contains the characteristic 3-trans-hexadecenoic acid (16:1t), whereas in this position from the transgenic lines new cis-unsaturated C18 fatty acids with an additional 3-trans-double bond are found (18:2t and 18:4t).

With regard to the functions of PG in thylakoids, and in particular the contribution of the 16:1t as discussed above, it has to be pointed out that the present approach has not completely eliminated the 18:3/16:1t species (34:4) as evident from Fig. 6B. Another very important detail is the absence of the 18:3/16:3 species (34:6) in PA and PG of the transformants (Fig. 6B). This indicates that a pool of tertiary DAG (Fig. 1), observed in isolated spinach envelopes with about equal proportions of 18:3/16:3 and 18:3/18:3 species (22), cannot be derivatized by the heterologous DAGK in situ under steady-state conditions in tobacco leaves. According to present understanding, a large pool of tertiary DAG does not exist under normal conditions and, therefore, the formation of polygalactolipids by the galactolipid:galactolipid galactosyltransferase (23) most likely does not represent the function of this enzyme in intact and unstressed leaves.

View larger version (24K): [in this window] [in a new window]   FIGURE 8. Mass spectra of polyenoic C18 fatty acids with a 3-trans-double bond. The spectra were recorded in the electron impact-mode during GLC separation of 4,4-dimethyloxazoline derivatives. The seed oil from T. stans contained all three components (A–C), whereas in the sn-2 position of PG from transgenic tobacco only two components (A and C) were identified. The structural details shown in the formulae were deduced/confirmed by 1H NMR spectroscopy. The spectrum of the 4,4-dimethyloxazoline derivative of 3,9,12-all-cis-18:3 (www.lipidlibrary.co.uk) differs significantly from the spectrum of the isomeric compound shown in B.

For a complete and independent structural identification, including the stereo- and regiochemistry of all olefinic bonds, the methyl esters of the three fatty acids (18:2t, 18:3t, and 18:4t) were subjected to ¹H NMR analysis. It should be pointed out that ¹H NMR analysis was able not only to determine the Z- and E-configuration, but also to identify the positions of the olefinic bonds, especially of those with low carbon numbers. As with the mass spectrometry analyses, identical NMR spectra were obtained for the corresponding components isolated from T. stans and the transgenic PG. One-dimensional (1D) and two-dimensional (2D) COSY ¹H NMR spectra of all three components gave a clear separation of the olefinic protons in positions C3/C4 from those at higher carbon numbers (supplemental Table S1). In addition, both proton resonances (H-3 and H-4) were clearly resolved and allowed a determination of the coupling constant of J_3,4 = 15.3 Hz, which indicates trans(E)-configuration. In 18:2t the second double bond had cis(Z)-configuration as evident from the smaller coupling constant J_9,10 = 5.6 Hz. The two other fatty acids 18:3t and 18:4t showed additional olefinic bonds assigned by GLC-mass spectrometry analysis to positions C12/C13 for 18:3t and to C12/C13 and C15/C16 for 18:4t. In contrast to the trans-double bonds at C3/C4, all other olefinic bonds showed coupling constants of J_9,10 (18:2t), J_12,13 (18: 3t), and J_15,16 (18:4t) of 5.6 Hz indicating cis-configuration. Thus, our reinvestigation fully confirmed the structure of the three unusual fatty acids present in the seed oil of T. stans (74, 75) from which two, 18:2^{E3, Z9} and 18:4^{E3, Z9,12,15}, were identified for the first time in a chloroplast lipid of transgenic tobacco.

This result suggests that the plastidial desaturase fad4 (77), which normally converts the sn-2-bound palmitoyl residue of PG into the characteristic 3-trans-hexadecenoyl group, can also accept C18 fatty acids. From the presence of 3-trans-double bonds in the eukaryotic PG, we conclude that this PG has access to subplastidial desaturation sites similar to the prokaryotic PG (78).

Only Minor Differences in Photosynthetic Parameters between Control and Transgenic Plants—F_v/F_m, the maximum photochemical efficiency of PS II in the dark-adapted state, can be used as a measure of overall photosynthetic performance and as an indication of stress conditions. F_v/F_m values for the two transformants t1 and t2 (0.846 and 0.840) were slightly, but significantly (t test, n = 30) lower than for the control (0.859) at 20 °C. Also at 30 °C, F_v/F_m for both transformants (t1: 0.840; t2: 0.822) were significantly lower than the control value (0.850). Interestingly, at 10 °C t1 had a slightly, but significantly (t test, n = 30) higher F_v/F_m value than the control and t2 (0.854 versus 0.850 and 0.839, respectively). The F_v/F_m value of t1 at 10 °C was even higher than at 20 or 30 °C. Whether this effect can be attributed to the reduction of high melting PG species (9) requires further studies. Photochemical quenching (qP) was not significantly different between transformants and control, neither at the various temperatures nor under actinic light intensities (data not shown).

The impact of the fatty acid and lipid changes on the composition and stability of the pigment-protein complexes of the thylakoid membranes was investigated by non-denaturing polyacrylamide gel-electrophoresis ("green gels"). Complete solubilization of thylakoid membranes of both transformants required at least twice the detergent concentration (1.0 versus 0.5%, w/v, of n-do-decyl-D-maltoside) as was sufficient for control membranes. The resultant green gels revealed significant differences in patterns and intensities of resolved green bands (Fig. 9). This indicates that the large pigment-protein supercomplexes in the two transformants are less stable than in the control, which applies to both the peripheral antennae and the core complexes of the photosystems. The reduced contents of trimeric LHCII in the two transformants is accompanied by significantly higher proportions of LHCII monomers as compared with the control samples. A similar increase was observed for the partially disassembled photosystem II (represented by the two bands above trimeric LHCII) in both transformants. These observations are consistent with the previously noted role of PG in LHCII trimerization (33, 79). A reduced LHCII complement can readily explain the increased Chl a/b-ratios in the transformants (cf. Fig. 3). In summary, the assessment of the performance of the photosynthetic apparatus did not reveal significantly large or obvious differences between control and transgenic plants.

View larger version (47K): [in this window] [in a new window]   FIGURE 9. Separation of pigment-protein complexes from thylakoid membranes of WT and the transformants t1 and t2 by non-denaturing polyacrylamide gel electrophoresis. LHCII3 indicates trimeric light-harvesting complex II, whereas LHCmono comprises the monomeric LHCs and FP the free pigments (essentially xanthophylls). Bands above LHCII3 comprise partially disassembled photosystems. Despite the 50% increase of detergent concentration in the wt sample, it was far more resistant toward dissociation than the two transgenic samples at the lower detergent concentration.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

On the other hand, from our data we cannot identify the precursor of eukaryotic DAG. In contrast to PC (supplemental data), the proportion of eukaryotic PA species in control thylakoids was too low to confirm the recently suggested function of PA as source of eukaryotic DAG in envelopes (16). But as mentioned above, mass spectrometry depicts the size and the steady-state composition of a pool, whereas the detection of a possibly high turnover of a pool or of only specific components in a pool requires additional labeling experiments.

We have shown that the heterologous DAGK activity is present in the envelope, but we have not localized it separately in inner or outer envelope membrane. Based on the positive outcome of the PG modification as most convincingly demonstrated by the molecular species of PG in control and transgenic plants (Fig. 6, A and B), a more detailed answer of this question is not of major relevance in the present context. On the other hand, there is indirect evidence from several observations that suggest that DAGK activity is also localized in the inner envelope membrane as required for our approach. For example, the efficient conversion of secondary eukaryotic PA into PG suggests that at least part of DAGK is present in the inner envelope membrane, because the three enzymes converting PA into PG have been localized in this membrane (34), and in contrast to the mobility of DAG the exchange of PA between different membranes is a very slow process (80).

In this context, we would like to mention another implication of our data. Recent investigations suggest that eukaryotic DAG is formed from eukaryotic PA, which is imported directly into the inner envelope (16). From our results we would expect that some part of the eukaryotic PA after arrival in the inner envelope should be used directly for PG synthesis. But in wild-type plants eukaryotic species of thylakoid PG represent a negligible proportion. This apparent discrepancy may be resolved by the recent discovery that TGD2 is a PA-binding protein associated with TGD1 in the inner envelope (26). TGD2 may prevent the release of imported eukaryotic PA into the lipid phase of the inner envelope membrane by channeling it specifically to the PA phosphatase colocalized in this membrane. The subsequently released DAG would be available for glycolipid biosynthesis, but incorporation into PG would not be possible any more. It would be interesting to see if TGD2 mutants affected in the channeling of PA would accumulate some eukaryotic PG.

Apart from the differences in the lipid biochemistry, we must explain the reduced growth caused by DAGK expression, which is the most obvious phenotype. For this purpose, we measured the functionality of the photosynthetic apparatus. Most of the differences detected between transformants and control were only minor ones, and not a single parameter showed a relative reduction paralleling the massive growth impairment documented in Fig. 2. Therefore, it may be the combination of the small reduction in the performance of several photosynthetic parameters that may contribute to this visible phenotype. In addition, the putative futile cycle mentioned above and the slightly changed lipid composition may affect the overall performance and formation of chloroplasts. In the crystal structure of the LHCII trimers, DGD and PG are firmly bound in the contact zones between individual monomers and thus contribute to the functionally important oligomerization (5). In particular, the sn-2-bound 16:1t acyl group of prokaryotic PG is in intimate contact with the protein and buried in extended conformation in a hydrophobic channel. However, as mentioned above, the transgenic plants still contain the 18:3/16:1t species in appreciable proportion, which may be sufficient to occupy all specific protein binding sites. Despite our fluorescence labeling studies, we also cannot exclude the possibility that a small proportion of the dgk fusion protein, if inserted as catalytically active preprotein into other subcellular membranes, will produce extraplastidial PA, which may interfere with the function of PA in nuclear-cytoplasmic stress signaling (81). All these effects may sum up and result in the phenotype of reduced growth.

A further increase of eukaryotic PG may be reached by a reduced formation of prokaryotic PA as possible by antisense technology combined with DAGK expression in the act mutant of Arabidopsis, which has an impaired capacity to provide plastidial PA (82, 83). Plants with an increased proportion of eukaryotic PG would be useful for studying not only the function of prokaryotic PG, but it would also be interesting to see how the concomitant reduction of high melting PG species affects chilling sensitivity (9). With regard to this aspect of PG functions, plants such as tobacco (64, 84) and rice (85) may be more useful than the cold-tolerant Arabidopsis studied before (86, 87).

In summary, we have generated transgenic plants containing a modified thylakoid phospholipid. This change is combined with a reduction in leaf size, in the composition and contribution of thylakoid lipids to leaf lipids and in the Chl content per leaf area. The reduced growth of these plants can hardly be explained on the basis of the small differences observed for various photosynthetic parameters, but at present a straightforward attribution to the eukaryotic proportion of chloroplast PG is also not possible. On the other hand, the successful accumulation of eukaryotic PG in thylakoids encourages further studies of the question, why PG is the only thylakoid lipid for which the purely prokaryotic character was saved during evolution.

	FOOTNOTES

 
This paper is dedicated to Prof. Dr. A. Benson on the occasion of his 90th birthday.

^* This work was supported by the Gesellschaft für Technische Zusammenarbeit (Grant 91.7860.9-01-114) and the Bundesministerium für Bildung und Forschung (Napus 2000, FK 0312252F). Mass spectrometry performed at the Kansas Lipidomics Research Center (KLRC) Analytical Laboratory was supported by National Science Foundation (NSF) Grants MCB 0455318 and DBI 0521587 and by NSF EPSCoR Grant EPS-0236913, and by National Institutes of Health Grant P20 RR16475 from the National Center for Research Resources. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The on-line version of this article (available at http://www.jbc.org) contains supplemental Table S1, Fig. 1S, and additional references.

¹ To whom correspondence should be addressed: Tel.: 49-551-39-5750; Fax: 49-551-39-5749; E-mail: mfulda{at}gwdg.de.

² The abbreviations used are: PG, phosphatidylglycerol; PIPES, 1,4-piperazinediethanesulfonic acid; TGD, trigalactosyl diacylglycerol; ER, endoplasmic reticulum; PSI, -II, photosystem I and II; MGD, monogalactosyldiacylglycerol; DAG, diacylglycerol; PC, phosphatidylcholine; LPC, lyso-PC; PA, phosphatidic acid; DAGK, DAG kinase; Chl, chlorophyll; Tricine, N-[2-hydroxy-1,1-bis(hydroxymethyl)ethyl]glycine; GLC, gas-liquid chromatography; DGD, digalactosyldiacylglycerol; EYFP, enhanced yellow fluorescent protein; GUS, -D-glucuronidase; LHC, light-harvesting complex; NPTII, neomycin phosphotransferase II; PE, phosphatidylethanolamine; SQD, sulfoquinovosyl diacylglycerol.

	ACKNOWLEDGMENTS

 
H. L. and D. H. thank Prof. B. Grimm for support. M. F. thanks Beate Preitz for assistance with the confocal laser-scanning microscopy. We thank I. Riedl for technical assistance.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Benson, A. A., and Maruo, B. (1958) Biochim. Biophys. Acta 27, 189-195[Medline] [Order article via Infotrieve]
2. Frentzen, M. (2004) Curr. Opin. Plant Biol. 7, 270-276[CrossRef][Medline] [Order article via Infotrieve]
3. Jordan, P., Fromme, P., Witt, H. T., Klukas, O., Saenger, W., and Krauss, N. (2001) Nature 411, 909-917[CrossRef][Medline] [Order article via Infotrieve]
4. Liu, Z., Yan, H., Wang, K., Kuang, T., Zhang, J., Gui, L., An, X., and Chang, W. (2004) Nature 428, 287-292[CrossRef][Medline] [Order article via Infotrieve]
5. Standfuss, J., van Scheltinga, T. A. C., Lamborghini, M., and Kühlbrandt, W. (2005) EMBO J. 24, 919-928[CrossRef][Medline] [Order article via Infotrieve]
6. Loll, B., Kern, J., Saenger, W., Zouni, A., and Biesiadka, J. (2005) Nature 438, 1040-1044[CrossRef][Medline] [Order article via Infotrieve]
7. van't Hof, R., van Klompenburg, W., Pilon, M., Kozubek, A., de Korte-Kool, G., Demel, R. A., Weisbeek, P. J., and de Kruijff, B. (1993) J. Biol. Chem. 268, 4037-4042[Abstract/Free Full Text]
8. Maréchal, E., Block, M. A., Joyard, J., and Douce, R. (1994) J. Biol. Chem. 269, 5788-5798[Abstract/Free Full Text]
9. Murata, N. (1983) Plant Cell Physiol. 24, 81-86[Abstract/Free Full Text]
10. Wada, H., and Murata, N. (1998) in Lipids in Photosynthesis: Structure, Function and Genetics (Siegenthaler, P. A., and Murata, N., eds) pp. 65-81, Kluwer Academic Publishers, Dordrecht, The Netherlands
11. Roughan, P. G., and Slack, C. R. (1982) Annu. Rev. Plant Physiol. 33, 97-132
12. Benning, C., Xu, C., and Awai, K. (2006) Curr. Opin. Plant Biol. 9, 241-247[CrossRef][Medline] [Order article via Infotrieve]
13. Haverkate, F., and van Deenen, L. L. M. (1965) Biochim. Biophys. Acta 106, 78-92[Medline] [Order article via Infotrieve]
14. Mongrand, S., Badoc, A., Patouille, B., Lacomblez, C., Chavent, M., and Bessoule, J. J. (2005) Phytochemistry 66, 549-559[CrossRef][Medline] [Order article via Infotrieve]
15. Mongrand, S., Cassagne, C., and Bessoule, J. J. (2000) Plant Physiol. 122, 845-852[Abstract/Free Full Text]
16. Xu, C., Fan, J., Froehlich, J. E., Awai, K., and Benning, C. (2005) Plant Cell 17, 3094-4011[Abstract/Free Full Text]
17. Andersson, M. X., Kjellberg, J. M., and Sandelius, A. S. (2004) Biochim. Biophys. Acta 1684, 46-53[Medline] [Order article via Infotrieve]
18. Bertrams, M., Wrage, K., and Heinz, E. (1981) Z. Naturforsch. 36c, 62-70
19. Miquel, M., Block, M. A., Joyard, J., Dorne, A. J., Dubacq, J. P., Kader, J. C., and Douce, R. (1987) Biochim. Biophys. Acta 937, 219-228
20. Kjellberg, J. M., Trimborn, M., Andersson, M., and Sandelius, A. S. (2000) Biochim. Biophys. Acta 1485, 100-110[Medline] [Order article via Infotrieve]
21. Benning, C., and Ohta, H. (2005) J. Biol. Chem. 280, 2397-2400[Abstract/Free Full Text]
22. Siebertz, H. P., Heinz, E., Linscheid, M., Joyard, J., and Douce, R. (1979) Eur. J. Biochem. 101, 429-438[Medline] [Order article via Infotrieve]
23. Heemskerk, J. W. M., Wintermans, J. F. G. M., Joyard, J., Block, M. A., Dorne, A. J., and Douce, R. (1986) Biochim. Biophys. Acta 877, 281-289
24. Sakaki, T., Kondo, N., and Yamada, M. (1990) Plant Physiol. 94, 773-780[Abstract/Free Full Text]
25. Xu, C., Fan, J., Riekhof, W., Froehlich, J. E., and Benning, C. (2003) EMBO J. 22, 2370-2379[CrossRef][Medline] [Order article via Infotrieve]
26. Awai, K., Xu, C., Tamot, B., and Benning, C. (2006) Proc. Natl. Acad. Sci. U. S. A. 103, 10817-10822[Abstract/Free Full Text]
27. Roughan, P. G. (1985) Plant Physiol. 77, 740-746[Abstract/Free Full Text]
28. Dorne, A. J., and Heinz, E. (1989) Plant Sci. 60, 39-46[CrossRef]
29. Lynch, D. V., and Steponkus, P. (1987) Plant Physiol. 83, 761-767[Abstract/Free Full Text]
30. Hagio, M., Gombos, Z., Várkonyi, Z., Masamoto, K., Sato, N., Tsuzuki, M., and Wada, H. (2000) Plant Physiol. 124, 795-804[Abstract/Free Full Text]
31. Quoc, K. P., and Dubacq, J. P. (1997) Biochim. Biophys. Acta 1346, 237-246[Medline] [Order article via Infotrieve]
32. Trémolières, A., Roche, O., Dubertret, G., Guyon, D., and Garnier, J. (1991) Biochim. Biophys. Acta 1059, 286-292[CrossRef]
33. McCourt, P., Browse, J., Watson, J., Arntzen, C. J., and Somerville, C. R. (1985) Plant Physiol. 78, 853-858[Abstract/Free Full Text]
34. Maréchal, E., Block, M. A., Dorne, A. J., Douce, R., and Joyard, J. (1997) Physiol. Plant 100, 65-77[CrossRef]
35. Gómez-Merino, F. C., Brearley, C. A., Ornatowska, M., Abdel-Hamiem, M. E. F., Zanor, M. I., and Mueller-Roeber, B. (2004) J. Biol. Chem. 279, 8230-8241[Abstract/Free Full Text]
36. Bovet, L., Müller, M. O., and Siegenthaler, P. A. (2001) Biochem. Biophys. Res. Commun. 289, 269-275[CrossRef][Medline] [Order article via Infotrieve]
37. Kjellberg, J. M., and Sandelius, A. S. (2004) Plant Sci. 166, 601-607[CrossRef]
38. Lightner, V. A., Bell, R. M., and Modrich, P. (1983) J. Biol. Chem. 258, 10856-10861[Abstract/Free Full Text]
39. Fritz, C. C., Wolter, F. P., Schenkemeyer, V., Herget, T., and Schreier, P. H. (1993) Gene (Amst.) 137, 271-274[CrossRef][Medline] [Order article via Infotrieve]
40. Koncz, C., Olssen, O., Langridge, W. H. R., Schell, J., and Szalay, A. A. (1987) Proc. Natl. Acad. Sci. U. S. A. 84, 131-135[Abstract/Free Full Text]
41. Bevan, M. (1984) Nucleic Acids Res. 12, 8711-8721[Abstract/Free Full Text]
42. Vancanneyt, G., Schmidt, R., O'Connor-Sanchez, A., Willmitzer, L., and Rocha-Sosa, M. (1990) Mol. Gen. Genet. 220, 245-250[Medline] [Order article via Infotrieve]
43. Horsch, R. B., Fry, J. E., Hoffmann, N. L., Wallroth, M., Eichholtz, D., Rogers, S. G., and Fraley, R. T. (1985) Science 227, 1229-1231[Abstract/Free Full Text]
44. Fulda, M., Shockey, J., Werber, M., Wolter, F. P., and Heinz, E. (2002) Plant J. 32, 93-103[CrossRef][Medline] [Order article via Infotrieve]
45. Bradford, M. M. (1976) Anal. Biochem. 72, 248-254[CrossRef][Medline] [Order article via Infotrieve]
46. Jefferson, R. A., Kavanagh, T. A., and Bevan, W. M. (1987) EMBO J. 6, 3901-3907[Medline] [Order article via Infotrieve]
47. Rogers, S. O., and Bendich, A. J. (1985) Plant Mol. Biol. 5, 69-76
48. Beck, E., Ludwig, G., Auerswald, E. A., Reiss, B., and Schaller, H. (1982) Gene (Amst.) 19, 327-336[CrossRef][Medline] [Order article via Infotrieve]
49. Porra, R. J., Thompson, W. A., and Kriedemann, P. E. (1989) Biochim. Biophys. Acta 975, 384-394[CrossRef]
50. Lichtenthaler, H. K. (1987) Methods Enzymol. 148, 350-382
51. Tietje, C., and Heinz, E. (1998) Planta 206, 72-78[CrossRef]
52. Hajra, A. K. (1974) Lipids 9, 502-505[CrossRef][Medline] [Order article via Infotrieve]
53. Sperling, P., Lee, M., Girke, T., Zähringer, U., Stymne, S., and Heinz, E. (2000) Eur. J. Biochem. 267, 3801-3811[Medline] [Order article via Infotrieve]
54. Wanjie, S. W., Welti, R., Moreau, R. A., and Chapman, K. D. (2005) Lipids 40, 773-785[Medline] [Order article via Infotrieve]
55. Welti, R., Wang, X., and Williams, T. D. (2003) Anal. Biochem. 314, 149-152[CrossRef][Medline] [Order article via Infotrieve]
56. Lokstein, H., Tian, L., Polle, J., and DellaPenna, D. (2002) Biochim. Biophys. Acta 1553, 93-103
57. Domergue, F., Lerchl, J., Zähringer, U., and Heinz, E. (2002) Eur. J. Biochem. 269, 4105-4113[Medline] [Order article via Infotrieve]
58. Loomis, C. R., Walsh, J. P., and Bell, R. M. (1985) J. Biol. Chem. 260, 4091-4097[Abstract/Free Full Text]
59. Sanders, C. R., 2nd, Czerski, L., Vinogradova, O., Badola, P., Song, D., and Smith, S. O. (1996) Biochemistry 35, 8610-8618[CrossRef][Medline] [Order article via Infotrieve]
60. Dorne, A. J., Joyard, J., and Douce, R. (1990) Proc. Natl. Acad. Sci. U. S. A. 87, 71-74[Abstract/Free Full Text]
61. Malherbe, A., Block, M. A., Joyard, J., and Douce, R. (1992) J. Biol. Chem. 267, 23546-23553[Abstract/Free Full Text]
62. Siebertz, H. P., Heinz, E., and Bergmann, L. (1978) Plant Sci. Lett. 12, 119-126[CrossRef]
63. Koiwai, A., Matsuzaki, T., Suzuki, F., and Kawashima, N. (1981) Plant Cell Physiol. 22, 1059-1060[Abstract/Free Full Text]
64. Murata, N., Ishizaki-Nishizawa, O., Higashi, S., Hayashi, H., Tasaka, Y., and Nishida, I. (1992) Nature 356, 710-713[CrossRef]
65. Jarvis, P., Dörmann, P., Peto, C. A., Lutes, J., Benning, C., and Chory, J. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, 8175-8179[Abstract/Free Full Text]
66. Hagio, M., Sakurai, I., Sato, S., Kato, T., Tabata, S., and Wada, H. (2002) Plant Cell Physiol. 43, 1456-1464[Abstract/Free Full Text]
67. Babiychuk, E., Müller, F., Eubel, H., Braun, H. P., Frentzen, M., and Kushnir, S. (2003) Plant J. 33, 899-909[CrossRef][Medline] [Order article via Infotrieve]
68. Kohler, R. H., and Hanson, M. R. (2000) J. Cell Sci. 113, 81-89[Abstract]
69. Andersson, M. X., Goksör, M., and Sandelius, A. S. (2007) J. Biol. Chem. 282, 1170-1174[Abstract/Free Full Text]
70. Yu, B., and Benning, C. (2003) Plant J. 36, 762-770[CrossRef][Medline] [Order article via Infotrieve]
71. Heinz, E., and Roughan, P. G. (1983) Plant Physiol. 72, 273-279[Abstract/Free Full Text]
72. Williams, J. P., Imperial, V., Khan, M. U., and Hodson, J. N. (2000) Biochem. J. 349, 127-133[CrossRef][Medline] [Order article via Infotrieve]
73. Seifert, U., and Heinz, E. (1992) Bot. Acta 105, 197-205
74. Hopkins, C. Y., and Chisholm, M. J. (1965) J. Chem. Soc. 907-910
75. Plattner, R. D., Spencer, G. F., and Kleiman, R. (1976) Lipids 11, 222-227[CrossRef]
76. Luthria, D. L., and Sprecher, H. (1993) Lipids 28, 561-564[CrossRef][Medline] [Order article via Infotrieve]
77. Browse, J., McCourt, P. J., and Somerville, C. R. (1985) Science 227, 763-765[Abstract/Free Full Text]
78. Ohnishi, M., and Thompson, G. A., Jr. (1991) Arch. Biochem. Biophys. 288, 591-599[CrossRef][Medline] [Order article via Infotrieve]
79. Dubertret, G., Mirshahi, A., Mirshahi, M., Gérard-Hirne, C., and Trémolières, A. (1994) Eur. J. Biochem. 226, 473-482[Medline] [Order article via Infotrieve]
80. van Meer, G., Halter, D., Sprong, H., Somerharju, P., and Egmond, M. R. (2006) FEBS Lett. 580, 1171-1177[CrossRef][Medline] [Order article via Infotrieve]
81. Wang, X. (2004) Curr. Opin. Plant Biol. 7, 329-336[CrossRef][Medline] [Order article via Infotrieve]
82. Kunst, L., Browse, J., and Somerville, C. (1988) Proc. Natl. Acad. Sci. U. S. A. 85, 4143-4147[Abstract/Free Full Text]
83. Xu, C., Yu, B., Cornish, A. J., Froehlich, J. E., and Benning, C. (2006) Plant J. 47, 296-309[CrossRef][Medline] [Order article via Infotrieve]
84. Sakamoto, A., Sulpice, R., Hou, C. X., Kinoshita, M., Higashi, S. I., Kanaseki, T., Nonaka, H., Moon, B. Y., and Murata, N. (2003) Plant Cell Environ. 27, 99-105[CrossRef]
85. Ariizumi, T., Kishitani, S., Inatsugi, R., Nishida, I., Murata, N., and Toriyama, K. (2002) Plant Cell Physiol. 43, 751-758[Abstract/Free Full Text]
86. Wolter, F. P., Schmidt, R., and Heinz, E. (1992) EMBO J. 11, 4685-4692[Medline] [Order article via Infotrieve]
87. Wu, J., and Browse, J. (1995) Plant Cell 7, 17-27[Medline] [Order article via Infotrieve]

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				M. Chen, J. E. Markham, C. R. Dietrich, J. G. Jaworski, and E. B. Cahoon Sphingolipid Long-Chain Base Hydroxylation Is Important for Growth and Regulation of Sphingolipid Content and Composition in Arabidopsis PLANT CELL, July 1, 2008; 20(7): 1862 - 1878. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Supplemental Data All Versions of this Article: 282/7/4613    most recent M606295200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Fritz, M. Articles by Heinz, E. Search for Related Content PubMed PubMed Citation Articles by Fritz, M. Articles by Heinz, E. Social Bookmarking                      What's this?