Channeling of Eukaryotic Diacylglycerol into the Biosynthesis of Plastidial Phosphatidylglycerol
2006; Elsevier BV; Volume: 282; Issue: 7 Linguagem: Inglês
10.1074/jbc.m606295200
ISSN1083-351X
AutoresMarkus Hsi-Yang Fritz, Heiko Lokstein, Dieter Hackenberg, Ruth Welti, Mary R. Roth, Ulrich Zähringer, Martin Fulda, Wiebke Hellmeyer, Claudia Ott, Frank P. Wolter, Ernst Heinz,
Tópico(s)Algal biology and biofuel production
ResumoPlastidial 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 Δ3-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. 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 Δ3-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. All subcellular membranes of plant cells contain phosphatidylglycerol (PG), 2The 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, β-d-glucuronidase; LHC, light-harvesting complex; NPTII, neomycin phosphotransferase II; PE, phosphatidylethanolamine; SQD, sulfoquinovosyl diacylglycerol. but it predominates in chloroplasts and, therefore, PG is characteristically associated with photosynthetically active cells (1.Benson A.A. Maruo B. Biochim. Biophys. Acta. 1958; 27: 189-195Crossref PubMed Scopus (51) Google Scholar). Its synthesis is confined to membranes of the endoplasmic reticulum (ER) and the inner membranes of the plastid envelope and mitochondria (2.Frentzen M. Curr. Opin. Plant Biol. 2004; 7: 270-276Crossref PubMed Scopus (111) Google Scholar). 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.Frentzen M. Curr. Opin. Plant Biol. 2004; 7: 270-276Crossref PubMed Scopus (111) Google Scholar). 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.Frentzen M. Curr. Opin. Plant Biol. 2004; 7: 270-276Crossref PubMed Scopus (111) Google Scholar). 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.Jordan P. Fromme P. Witt H.T. Klukas O. Saenger W. Krauss N. Nature. 2001; 411: 909-917Crossref PubMed Scopus (2081) Google Scholar, 4.Liu Z. Yan H. Wang K. Kuang T. Zhang J. Gui L. An X. Chang W. Nature. 2004; 428: 287-292Crossref PubMed Scopus (1391) Google Scholar, 5.Standfuss J. van Scheltinga T.A.C. Lamborghini M. Kühlbrandt W. EMBO J. 2005; 24: 919-928Crossref PubMed Scopus (648) Google Scholar, 6.Loll B. Kern J. Saenger W. Zouni A. Biesiadka J. Nature. 2005; 438: 1040-1044Crossref PubMed Scopus (1616) Google Scholar). 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.van't Hof R. van Klompenburg W. Pilon M. Kozubek A. de Korte-Kool G. Demel R.A. Weisbeek P.J. de Kruijff B. J. Biol. Chem. 1993; 268: 4037-4042Abstract Full Text PDF PubMed Google Scholar). PG may also contribute to galactolipid synthesis by activation of the monogalactosyldiacylglycerol (MGD) synthase in the envelope (8.Maréchal E. Block M.A. Joyard J. Douce R. J. Biol. Chem. 1994; 269: 5788-5798Abstract Full Text PDF PubMed Google Scholar). 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.Murata N. Plant Cell Physiol. 1983; 24: 81-86Crossref Scopus (198) Google Scholar). 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.Wada H. Murata N. Siegenthaler P.A. Murata N. Lipids in Photosynthesis: Structure, Function and Genetics. Kluwer Academic Publishers, Dordrecht, The Netherlands1998: 65-81Google Scholar), 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.Roughan P.G. Slack C.R. Annu. Rev. Plant Physiol. 1982; 33: 97-132Crossref Google Scholar, 12.Benning C. Xu C. Awai K. Curr. Opin. Plant Biol. 2006; 9: 241-247Crossref PubMed Scopus (64) Google Scholar). 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.Haverkate F. van Deenen L.L.M. Biochim. Biophys. Acta. 1965; 106: 78-92Crossref PubMed Scopus (129) Google Scholar). In angiosperms, 16:3 plants are a minority compared with 18:3 plants, which contain MGD with more or less exclusively eukaryotic DAG (14.Mongrand S. Badoc A. Patouille B. Lacomblez C. Chavent M. Bessoule J.J. Phytochemistry. 2005; 66: 549-559Crossref PubMed Scopus (53) Google Scholar). 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.Roughan P.G. Slack C.R. Annu. Rev. Plant Physiol. 1982; 33: 97-132Crossref Google Scholar, 12.Benning C. Xu C. Awai K. Curr. Opin. Plant Biol. 2006; 9: 241-247Crossref PubMed Scopus (64) Google Scholar): phosphatidylcholine (PC), DAG, lyso-PC (LPC (15.Mongrand S. Cassagne C. Bessoule J.J. Plant Physiol. 2000; 122: 845-852Crossref PubMed Scopus (53) Google Scholar)), and phosphatidic acid (PA (16.Xu C. Fan J. Froehlich J.E. Awai K. Benning C. Plant Cell. 2005; 17: 3094-4011Crossref PubMed Scopus (161) Google Scholar)). 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.Mongrand S. Cassagne C. Bessoule J.J. Plant Physiol. 2000; 122: 845-852Crossref PubMed Scopus (53) Google Scholar, 16.Xu C. Fan J. Froehlich J.E. Awai K. Benning C. Plant Cell. 2005; 17: 3094-4011Crossref PubMed Scopus (161) Google Scholar, 17.Andersson M.X. Kjellberg J.M. Sandelius A.S. Biochim. Biophys. Acta. 2004; 1684: 46-53Crossref PubMed Scopus (37) Google Scholar, 18.Bertrams M. Wrage K. Heinz E. Z. Naturforsch. 1981; 36c: 62-70Crossref Scopus (25) Google Scholar, 19.Miquel M. Block M.A. Joyard J. Dorne A.J. Dubacq J.P. Kader J.C. Douce R. Biochim. Biophys. Acta. 1987; 937: 219-228Crossref Scopus (32) Google Scholar, 20.Kjellberg J.M. Trimborn M. Andersson M. Sandelius A.S. Biochim. Biophys. Acta. 2000; 1485: 100-110Crossref PubMed Scopus (33) Google Scholar). The resulting DAGs are incorporated into eukaryotic glycolipids by glycosyltransferases located in both inner and outer envelope membranes and finally transferred into thylakoids (21.Benning C. Ohta H. J. Biol. Chem. 2005; 280: 2397-2400Abstract Full Text Full Text PDF PubMed Scopus (165) Google Scholar). 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.Roughan P.G. Slack C.R. Annu. Rev. Plant Physiol. 1982; 33: 97-132Crossref Google Scholar). 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.Siebertz H.P. Heinz E. Linscheid M. Joyard J. Douce R. Eur. J. Biochem. 1979; 101: 429-438Crossref PubMed Scopus (82) Google Scholar). Its formation is ascribed to the action of galactolipid:galactolipid galactosyltransferase. This enzyme of the outer envelope membrane (23.Heemskerk J.W.M. Wintermans J.F.G.M. Joyard J. Block M.A. Dorne A.J. Douce R. Biochim. Biophys. Acta. 1986; 877: 281-289Crossref Scopus (24) Google Scholar) is only activated in stress situations such as ozone fumigation of leaves (24.Sakaki T. Kondo N. Yamada M. Plant Physiol. 1990; 94: 773-780Crossref PubMed Scopus (69) Google Scholar), release of envelopes from isolated chloroplasts (22.Siebertz H.P. Heinz E. Linscheid M. Joyard J. Douce R. Eur. J. Biochem. 1979; 101: 429-438Crossref PubMed Scopus (82) Google Scholar), and in tgd1 and tgd2 mutants of Arabidopsis (25.Xu C. Fan J. Riekhof W. Froehlich J.E. Benning C. EMBO J. 2003; 22: 2370-2379Crossref PubMed Scopus (166) Google Scholar). These envelope proteins, considered to be part of a larger lipid transfer complex (16.Xu C. Fan J. Froehlich J.E. Awai K. Benning C. Plant Cell. 2005; 17: 3094-4011Crossref PubMed Scopus (161) Google Scholar, 25.Xu C. Fan J. Riekhof W. Froehlich J.E. Benning C. EMBO J. 2003; 22: 2370-2379Crossref PubMed Scopus (166) Google Scholar, 26.Awai K. Xu C. Tamot B. Benning C. Proc. Natl. Acad. Sci. U. S. A. 2006; 103: 10817-10822Crossref PubMed Scopus (159) Google Scholar), 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.Wada H. Murata N. Siegenthaler P.A. Murata N. Lipids in Photosynthesis: Structure, Function and Genetics. Kluwer Academic Publishers, Dordrecht, The Netherlands1998: 65-81Google Scholar). 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.Roughan P.G. Plant Physiol. 1985; 77: 740-746Crossref PubMed Google Scholar), plant mitochondrial PG contains both eu- and prokaryotic DAG backbones (28.Dorne A.J. Heinz E. Plant Sci. 1989; 60: 39-46Crossref Scopus (20) Google Scholar), and PG in plasma membranes is nearly exclusively of eukaryotic structure (29.Lynch D.V. Steponkus P. Plant Physiol. 1987; 83: 761-767Crossref PubMed Google Scholar). 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.Hagio M. Gombos Z. Va´rkonyi Z. Masamoto K. Sato N. Tsuzuki M. Wada H. Plant Physiol. 2000; 124: 795-804Crossref PubMed Scopus (162) Google Scholar). 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.Quoc K.P. Dubacq J.P. Biochim. Biophys. Acta. 1997; 1346: 237-246Crossref PubMed Scopus (21) Google Scholar). On the other hand, only PG with 16:1t contributes specifically to the oligomerization of LHCII and possibly grana stacking (32.Trémolie`res A. Roche O. Dubertret G. Guyon D. Garnier J. Biochim. Biophys. Acta. 1991; 1059: 286-292Crossref Scopus (29) Google Scholar). But this antenna system is absent from cyanobacteria, which do not form 16:1t in their prokaryotic PG (10.Wada H. Murata N. Siegenthaler P.A. Murata N. Lipids in Photosynthesis: Structure, Function and Genetics. Kluwer Academic Publishers, Dordrecht, The Netherlands1998: 65-81Google Scholar). 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.Standfuss J. van Scheltinga T.A.C. Lamborghini M. Kühlbrandt W. EMBO J. 2005; 24: 919-928Crossref PubMed Scopus (648) Google Scholar). On the other hand, a 16:1t-free mutant of Arabidopsis had a normal growth phenotype at ambient temperature (33.McCourt P. Browse J. Watson J. Arntzen C.J. Somerville C.R. Plant Physiol. 1985; 78: 853-858Crossref PubMed Google Scholar). As a first step toward answering some of the questions regarding the preservation of exclusively prokaryotic DAG backbones in plastidial PG, we thought it would be useful to generate higher plants with eukaryotic PG in chloroplasts. Due to the enzymatic equipment of chloroplast envelopes (34.Maréchal E. Block M.A. Dorne A.J. Douce R. Joyard J. Physiol. Plant. 1997; 100: 65-77Crossref Google Scholar), none of the plastidial DAG pools can be used for PG synthesis, because PA is the last common precursor shared by prokaryotic glycolipids and PG. Prokaryotic PG is formed from the precursor PA by three additional enzymes (Fig. 1) (2.Frentzen M. Curr. Opin. Plant Biol. 2004; 7: 270-276Crossref PubMed Scopus (111) Google Scholar). On the other hand, it may be possible to channel plastidial DAG of any pool into this final sequence of PG synthesis after conversion of DAG to PA. This phosphorylation step requires the expression of only one additional enzyme, a DAG kinase (DAGK), in chloroplast envelopes. Higher plants encode a gene family of DAGK (35.Go´mez-Merino F.C. Brearley C.A. Ornatowska M. Abdel-Hamiem M.E.F. Zanor M.I. Mueller-Roeber B. J. Biol. Chem. 2004; 279: 8230-8241Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar), but only very low activity of an unspecified member of this family has been measured in envelopes (36.Bovet L. Müller M.O. Siegenthaler P.A. Biochem. Biophys. Res. Commun. 2001; 289: 269-275Crossref PubMed Scopus (16) Google Scholar, 37.Kjellberg J.M. Sandelius A.S. Plant Sci. 2004; 166: 601-607Crossref Scopus (2) Google Scholar). In the genomes of cyanobacteria a nucleotide sequence with high similarity to bacterial DAGK is present (Cyanobase, www.kazusa.or.jp/cyanobase/). As reported in the following, the targeting of a bacterial DAGK into chloroplasts resulted in its incorporation into envelope membranes and in extensive channeling of eukaryotic DAG into thylakoid PG. This is demonstrated by detailed lipid analyses of the thylakoid fractions prepared from transgenic tobacco plants. In addition, the composition and functional parameters of the photosynthetic apparatus were assessed to find out whether plants having additionally large proportions of eukaryotic PG are affected in photosynthetic performance as compared with those with only prokaryotic PG. 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.Lightner V.A. Bell R.M. Modrich P. J. Biol. Chem. 1983; 258: 10856-10861Abstract Full Text PDF PubMed Google Scholar) 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.Fritz C.C. Wolter F.P. Schenkemeyer V. Herget T. Schreier P.H. Gene (Amst.). 1993; 137: 271-274Crossref PubMed Scopus (12) Google Scholar). 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.Koncz C. Olssen O. Langridge W.H.R. Schell J. Szalay A.A. Proc. Natl. Acad. Sci. U. S. A. 1987; 84: 131-135Crossref PubMed Scopus (90) Google Scholar) upstream of the octopin synthetase terminator. The whole construct was then transferred as HindIII fragment into a modified pBI 121 vector (41.Bevan M. Nucleic Acids Res. 1984; 12: 8711-8721Crossref PubMed Scopus (1813) Google Scholar), which contains the intron-interrupted β-glucuronidase (GUS-int) sequence (42.Vancanneyt G. Schmidt R. O'Connor-Sanchez A. Willmitzer L. Rocha-Sosa M. Mol. Gen. Genet. 1990; 220: 245-250Crossref PubMed Scopus (755) Google Scholar) 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 β-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.Horsch R.B. Fry J.E. Hoffmann N.L. Wallroth M. Eichholtz D. Rogers S.G. Fraley R.T. Science. 1985; 227: 1229-1231Crossref PubMed Scopus (3809) Google Scholar) 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.Fulda M. Shockey J. Werber M. Wolter F.P. Heinz E. Plant J. 2002; 32: 93-103Crossref PubMed Scopus (141) Google Scholar). 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.Fulda M. Shockey J. Werber M. Wolter F.P. Heinz E. Plant J. 2002; 32: 93-103Crossref PubMed Scopus (141) Google Scholar). 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.Bradford M.M. Anal. Biochem. 1976; 72: 248-254Crossref PubMed Scopus (217508) Google Scholar), and GUS tissue assays (46.Jefferson R.A. Kavanagh T.A. Bevan W.M. EMBO J. 1987; 6: 3901-3907Crossref PubMed Scopus (8351) Google Scholar) were carried out with small pieces (0.5 cm2) of leaves. For Southern blots with genomic DNA extracted from leaves (47.Rogers S.O. Bendich A.J. Plant Mol. Biol. 1985; 5: 69-76Crossref PubMed Scopus (1560) Google Scholar) 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.Beck E. Ludwig G. Auerswald E.A. Reiss B. Schaller H. Gene (Amst.). 1982; 19: 327-336Crossref PubMed Scopus (696) Google Scholar) 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.Porra R.J. Thompson W.A. Kriedemann P.E. Biochim. Biophys. Acta. 1989; 975: 384-394Crossref Scopus (4743) Google Scholar) and chlorophylls (Chls) a and b measured spectrophotometrically (50.Lichtenthaler H.K. Methods Enzymol. 1987; 148: 350-382Crossref Scopus (9417) Google Scholar). 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.Tietje C. Heinz E. Planta. 1998; 206: 72-78Crossref Scopus (35) Google Scholar). Lipids from leaves and chloroplasts were extracted and subjected to various analytical procedures as detailed before (22.Siebertz H.P. Heinz E. Linscheid M. Joyard J. Douce R. Eur. J. Biochem. 1979; 101: 429-438Crossref PubMed Scopus (82) Google Scholar, 52.Hajra A.K. Lipids. 1974; 9: 502-505Crossref PubMed Scopus (77) Google Scholar, 53.Sperling P. Lee M. Girke T. Zähringer U. Stymne S. Heinz E. Eur. J. Biochem. 2000; 267: 3801-3811Crossref PubMed Scopus (73) Google Scholar, 54.Wanjie S.W. Welti R. Moreau R.A. Chapman K.D. Lipids. 2005; 40: 773-785Crossref PubMed Scopus (53) Google Scholar). In particular, the lipid profiling method was identical to that employed by (54.Wanjie S.W. Welti R. Moreau R.A. Chapman K.D. Lipids. 2005; 40: 773-785Crossref PubMed Scopus (53) Google Scholar), except for the additional analysis of sulfoquinovosyl diacylglycerol (SQD), which was scanned using the parameters previously defined (55.Welti R. Wang X. Williams T.D. Anal. Biochem. 2003; 314: 149-152Crossref PubMed Scopus (90) Google Scholar), using hydrogenated SQD species as internal standards. Chlorophyll Fluorescence Analyses−In vivo Chl fluorescence was measured with attached leaves as previously described (56.Lokstein H. Tian L. Polle J. DellaPenna D. Biochim. Biophys. Acta. 2002; 1553: 93-103Google Scholar). 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: Fv/Fm = (Fm - F0
Referência(s)