Galactolipids not associated with the photosynthetic apparatus in phosphate-deprived plants

Journal of Photochemistry and Photobiology B: Biology 61 (2001) 46–51 www.elsevier.com / locate / jphotobiol

Galactolipids not associated with the photosynthetic apparatus in phosphate-deprived plants b ¨ a ,1 , Peter Dormann ¨ Heiko Hartel , Christoph Benning a , * a

Department of Molecular Biology and Biochemistry, Michigan State University, 224 Biochemistry Building, East Lansing, MI 48824 -1319, USA b ¨ Molekulare Pflanzenphysiologie, Am Muhlenberg ¨ Max-Planck-Institut f ur 1, D-14476 Golm, Germany Received 12 January 2001; accepted 15 May 2001

Abstract The galactolipid digalactosyldiacylglycerol (DGGD) is one of the major constituents of thylakoids, accounting for about 25% of polar lipids found in these membranes. Although the presence of DGDG has frequently been correlated with the structural and functional integrity of the photosynthetic apparatus, it is still a matter of debate of what the in-vivo function of DGDG actually might be. To further the understanding of the role of DGDG within the photosynthetic apparatus, experiments were conducted on different Arabidopsis thaliana lines with altered DGDG content. The dgd1 mutant is characterized by a 90% reduction in the DGDG content, resulting in a severe dwarfism during growth. Complementation of the dgd1 mutant with a DGD1 cDNA completely restored the wild-type characteristics, while photosynthesis-related parameters were intermediate in transgenic plants with a partial reduction in DGD1 activity caused by post-transcription gene silencing due to over-expression of a DGD1 cDNA in wild-type plants. These data provide clear evidence for a causal relationship between the DGDG content, and the structure and function of the photosynthetic apparatus. However, a significant DGDG accumulation in the dgd1 /pho1 double mutant was without any detectable effect on photosynthetic activity, indicating that the molecular DGDG species synthesized upon phosphate deprivation in leaves cannot substitute for the DGDG species present under normal nutrient supply of plants. It is suggested that depending on the environmental growth conditions different pools of DGDG species exist in plants of which one is not associated with the photosynthetic apparatus.  2001 Elsevier Science B.V. All rights reserved. Keywords: Phosphate deficiency; DGD1 overexpression; Photosynthesis; Thylakoids

1. Introduction Lipids are integral components of thylakoid membranes and they are fundamental for the structural and functional integrity of the photosynthetic apparatus. Galactolipids are the predominant lipids (80 mol%) of photosynthetic membranes in organisms which carry out oxygenic photosynthesis [1–3]. About 50–60% of polar lipids in photosynthetic tissue are represented by monogalactosyldiacylglycerol (MGDG) and 20–25% are digalactosyldiacylglycerol (DGDG). In leaves of Arabidopsis thaliana and many other plant species, galactolipids are synthesized either directly in the plastid or are assembled from diacylglycerol moieties imported from the endoplasmic reticulum, giving rise to different molecular species that are characterized by a different fatty acid *Corresponding author. Fax: 11-517-353-9334. E-mail address: [email protected] (C. Benning). 1 Present address: BASF Plant Science L.L.C., Davis Drive 26, Research Triangle Park, NC 27709-3528, USA.

composition [1]. Because of their high abundance, in particular in thylakoid membranes, it has been frequently inferred that galactolipids have specific roles associated with the light reactions of photosynthesis [4–10]. However, due to the complex nature and the high variability in the membrane lipid composition an unambiguous assignment of the exact role of individual lipids based on in-vitro studies or comparative physiological studies has been notoriously difficult. The isolation of lipid head group mutants of the small cruciferous plant A. thaliana with an altered membrane lipid composition has shed more light onto this important problem. Two particular mutations, mgd1 and dgd1, have shown wide-ranging effects on growth and photosynthesis-related parameters. Both mutations affect enzymes of galactolipid biosynthesis. The 42% deficiency of MGDG in the mgd1 mutant has been correlated with chlorophyll deficiency and severe defects in the chloroplast ultrastructure [10]. The dgd1 mutant shows a 90% reduction of DGDG in leaves [7], resulting in a structural rearrangement of the photosynthetic apparatus [7,8]. A stop codon within the ORF suggests the complete

1011-1344 / 01 / $ – see front matter  2001 Elsevier Science B.V. All rights reserved. PII: S1011-1344( 01 )00144-0

¨ et al. / Journal of Photochemistry and Photobiology B: Biology 61 (2001) 46 – 51 H. Hartel

inactivation of DGDG synthase [11]. The availability of the DGD1 gene represents a unique tool to address the question of the role of DGDG in photosynthesis in vivo, permitting important insights with regard to the function of DGDG in the photosynthetic membrane, by constructing transgenic lines that exhibit a gradual decrease in the DGDG content. Different to the dgd1 mutant, the pho1 mutant of A. thaliana shows an increase in the DGDG amount to 170% of the wild-type level mainly at the expense of phospholipids [12,13]. The fact that the pho1 mutant lipid phenotype is due to a defect in phosphate translocation from roots to shoots resulting in phosphate deprivation in leaves [14] and not a mutation in a lipid biosynthetic pathway suggests that the observed lipid changes are caused by phosphate deprivation. In an attempt to further the understanding of the regulation of DGDG biosynthesis in plants we previously reported the generation of a double mutant, designated as dgd1 /pho1, as the result of a cross between the DGDG-deficient dgd1 mutant and the DGDG-enriched pho1 mutant [15]. This double mutant showed intermediate amounts of DGDG between wild type and dgd1 mutant. In this study, we compared different plants with altered DGDG pools to further unravel the role of DGDG in the plant membrane system. We report results demonstrating a causal relationship between the presence of DGDG and the structural and functional integrity of the photosynthetic apparatus under normal growth conditions. Moreover, our results also indicate that altered environmental conditions like phosphate deprivation can cause the accumulation of a DGDG molecular species that does not seem to affect photosynthetic processes.

47

2.2. Construction of transgenic plants A cDNA coding for the DGDG synthase (DGD1 ) was cloned into the binary vector pBINAR-Hyg (with CaMV 35S promoter) and used for the transformation into the dgd1 mutant as described [11]. The same construct was transferred into wild-type plants, and transgenic plants selected on Hygromycin B were screened by TLC and GLC. Harvested leaves were immediately frozen in liquid nitrogen, and lipids were extracted as previously described [13].

2.3. Biochemical analyses Lipids were extracted from leaves of 5-to-6-week-old plants and separated by thin-layer chromatography according to the previously described procedure [7] with some modifications as in Ref. [15]. Bands corresponding to each lipid class were isolated and used to prepare fatty acid methyl esters by incubating in 1 M HCl in methanol at 808C for 30 min. Methyl esters were quantified by gas chromatography by using myristic acid as an internal standard. Pigments were extracted from leaves and analyzed as described [17].

2.4. Chlorophyll fluorescence measurements In vivo room temperature chlorophyll fluorescence was monitored as described [18]. Fluorescence parameters used are as defined in Refs. [19,20]. Plants were dark-adapted for 1 h prior to the measurements.

3. Results and discussion 2. Materials and methods

2.1. Plant material Surface sterilized seeds of A. thaliana (ecotype Columbia-2), the multiple-times-backcrossed dgd1 [7], pho1 mutants [14] and the dgd1 /pho1 double mutant [15] were germinated on 0.8% (w / v) agar-solified MS medium [16] supplemented with 1% (w / v) sucrose. The seedlings were kept on agar for 10 d prior to the transfer to pots containing a soil mixture as described [15]. The dgd1 / pho1 double mutant was generated as described [15]. Plants were grown in growth chambers (AR-75L, Percival Scientific, Boone, IA, USA) under light of a photosynthetic photon flux density (PPFD, 400–700 nm) of 75 mmol photons m 22 s 21 . A 14-h light / 10-h dark regime was applied. The day / night temperature was controlled at 23 / 188C. PPFD incident upon the top of plants was determined using a quantum sensor (LI-189A, Li-Cor, Lincoln, NE, USA). All biochemical and physiological analyses were performed with rosette leaves of 5-to-6-week-old plants grown on soil.

In previous studies with the dgd1 mutant of A. thaliana an apparent causal relationship has been observed between the lack of DGDG and the structural and functional integrity of the photosynthetic apparatus [7–9,21,22]. In an attempt to further understand the structure / function relationships in the photosynthetic apparatus with regard to the role of DGDG we started to compare the light response of the well-characterized dgd1 mutant with those of the pho1 mutant and the dgd1 /pho1 double mutant of A. thaliana. It is now well established that phosphate deprivation in leaves considerably affects the glycerolipid composition of bacterial and plant membranes [12,13,15,23,24]. When grown under the same light regime, there is a striking difference in the content of DGDG in these mutants ranging from 8% (dgd1 ), over 55% (dgd1 /pho1 ) to 175% ( pho1 ) of the wild-type level (Table 1). At the same time the content of MGDG, the precursor of DGDG and the most abundant galactolipid in plants, remained remarkably constant. In plants of the pho1 genetic background all phospholipids were found to be reduced, whereas the relative amount of sulfoquinovosyldiacylglycerol (SQDG)

¨ et al. / Journal of Photochemistry and Photobiology B: Biology 61 (2001) 46 – 51 H. Hartel

48

Table 1 Polar lipids from leaves of 5-to-6-week-old plants of wild type, pho1, dgd1 and dgd1 /pho1 double mutant of A. thaliana. Lipids were separated by one-dimensional TLC and individual fatty acids were quantified by GLC. All values are given as moles percent and represent the means of three different determinations with individual plants Lipid class a

Wild type

pho1

dgd1

dgd1 /pho1

MGDG PG DGDG SQDG PI PE PC MGDG /(PG1 DGDG1SQDG)b

47.461.2 9.660.7 16.560.8 1.260.4 1.860.5 9.060.3 15.461.5

46.561.1 5.060.3 28.860.3 7.460.3 0.860.3 4.361.2 7.360.9

45.560.7 14.861.0 1.460.2 1.160.2 2.060.3 12.960.3 24.161.5

44.861.4 11.660.4 9.160.5 4.761.1 1.360.2 8.261.4 20.361.9

1.73

1.13

2.63

1.76

a

PG, phosphatidylglycerol; SQDG, sulfoquinovosyldiacylglycerol; PI, phosphatidylinositol; PE, phosphatidylethanolamine; PC, phosphatidylcholine. b Ratio of non-bilayer-to-bilayer-forming lipids in chloroplasts.

was increased. Although the ratio of the non-bilayer forming thylakoid lipid MGDG to the bilayer-forming thylakoid lipids phosphatidylglycerol, SQDG and DGDG was increased from 1.7 to 2.6 in the dgd1 mutant, the ratio was apparently the same in the dgd1 /pho1 double mutant as in the wild type (Table 1). This is important to note, because growth was reduced to the same extent in pho1 and dgd1 /pho1 as compared to the wild type and dgd1, respectively [15]. Galactolipids usually are found to be located in thyla-

koid membranes, the site where photosynthetic light reactions take place. Pulse amplitude modulated chlorophyll fluorescence is a sensitive non-invasive method which allows the separation of photochemical and nonphotochemical events under in vivo conditions, and gives valuable information about changes in electron-transport reactions within thylakoids and the overall photosynthetic capability of leaves (for a review, see Ref. [25]). By using in-vivo chlorophyll fluorescence measurements as a very sensitive non-invasive tool we previously demonstrated striking differences in the PSII quantum efficiency (DF / Fm 9), the maximum intrinsic photochemical efficiency of PSII (Fv 9 /Fm 9) and the overall non-photochemical quenching (qN) between A. thaliana wild type and dgd1 [7–9]. As apparent from the light response curves depicted in Fig. 1A, the values for the parameter 1 2 qP, which indicates the oxidation state of the primary electron acceptor of PSII, QA , during steady-state light exposure were slightly lower in dgd1 as compared with the wild type in the high PPFD range. More striking differences with lower values for dgd1 are apparent for DF /Fm 9 in the low PPFD range (Fig. 1B) and Fv 9 /Fm 9 throughout all PPFDs applied (Fig. 1C), while qN was significantly enhanced especially in the low PPFD range (Fig. 1D). The values obtained for pho1 closely match the light response curves of the wild type and those for dgd1 /pho1 that of dgd1, indicating that neither the changes in lipid composition nor the phosphate deprivation have an immediate detectable effect on the photochemical efficiency, at least in short-term experi-

Fig. 1. Light dependence of steady-state chlorophyll fluorescence parameters in fully expanded leaves of the wild type, pho1, dgd1 and the dgd1 /pho1 double mutant of A. thaliana. (A) Approximate QA photoreduction state (1 2 qP). The fraction of QA in the reduced state was estimated as 1 2 (Fs 9 2 Fo 9) /(Fm 9 2 Fo 9) (S.D.#12%; n $ 4). (B) Efficiency of PSII photochemistry (DF /Fm 9). PSII efficiency was calculated as (Fm 2 Fo 9) /Fm in the absence of actinic light and as (Fm 9 2 Fs 9) /Fm 9 in its presence (S.D.#13%; n $ 4). (C) Maximum PSII photochemical efficiency with the reaction center being in an open configuration. The maximum PSII photochemistry was calculated as (Fm 2 Fo 9) /Fm in the absence of actinic light and as (Fm 9 2 Fo 9) /Fm 9 in its presence (S.D.#11%; n $ 4). (D) Total non-photochemical quenching, qN, calculated as 1 2 (Fm 9 2 Fo 9) /(Fm 2 Fo 9) (S.D.#11%; n $ 4).

¨ et al. / Journal of Photochemistry and Photobiology B: Biology 61 (2001) 46 – 51 H. Hartel Table 2 Relationship between the DGDG content and structural parameters of mature leaves of 5-to-6-week-old A. thaliana wild-type plants, the mutant lines pho1, dgd1 and dgd1 /pho1, the transgenic lines R382 and R383 (DGD1 co-suppression), and R376, a DGD1 complemented dgd1 mutant line. All values represent the means (6S.D.) of at least three different determinations

WT pho1 dgd1 dgd1 /pho1 R383 R382 R376

DGDG (mol%)

oLipids oChl (mmol (g)21 fresh weight)

Chl a /b ratio

16.560.8 28.860.3 1.460.2 9.160.5 8.761.3 9.761.2 15.361.1

8.960.8 9.660.3 6.661.0 6.860.3 8.260.4 8.060.5 8.460.7

3.0860.06 2.8860.10 2.6060.09 2.5260.10 2.8860.11 2.8860.04 3.0860.01

2.1160.13 2.6060.23 1.6160.16 1.9160.07 1.8060.11 1.7660.09 2.0260.15

ments. This finding is of particular interest, because it does not support a relationship between DGDG abundance and photosynthesis-related processes in these particular mutant plants. To further explore the possible function of DGDG, transgenic plants expressing DGD1 in sense orientation in the wild-type background were generated. Some of the lines, designated here as R382 and R383, showed a reduced DGDG content, most probably due to post-transcriptional gene silencing, resulting in a similar amount of DGDG as in the dgd1 /pho1 double mutant (Table 2). Unfortunately, all efforts to recover transgenic lines with a more severe reduction in the DGDG content have so far been unsuccessful. We took also advantage of a DGD1

49

cDNA complemented dgd1 mutant line, designated as R376, which was phenotypically wild type with regard to growth and lipid composition [11]. These lines were compared to wild type and mutant lines for selected structural and functional parameters of the photosynthetic apparatus. The main results shown in Table 2 can be summarized as follows: (i) the total amount of lipid was reduced by about 30% in dgd1 and dgd1 /pho1 as compared with wild type and pho1, respectively, on a fresh weight basis; (ii) total lipid reduction correlates with total chlorophyll reduction; (iii) the chlorophyll a /b ratio, an indicator for structural assembly of the light-harvesting antenna complexes in PSI and PSII in dgd1 /pho1, resembles dgd1, and pho1 is similar to wild type, whereas the R383 and R382 values are intermediate between dgd1 and wild type; (iv) except for pho1 and dgd1 /pho1, the DGDG reduction correlated with a decline in the chlorophyll a /b ratio; (v) there is virtually no difference between the DGD1 complemented dgd1 mutant line and wild type. Fig. 2 illustrates the relationship between the DGDG levels and function of the photosynthetic apparatus monitored by chlorophyll fluorescence measurements. We have chosen a PPFD of 75 mmol m 22 s 21 for all fluorescence measurements for two reasons: (i) it is the PPFD that was used for plant growth; and (ii) differences between wild type and dgd1 are at their maximum (Fig. 1B–D). Strikingly, a gradual decline in the DGDG content was linearly related to processes of photosynthetic light reactions (Fig. 2A–D). All values closely fit to a first-order linear regression line with the strongest correlation found

Fig. 2. Relationship between the DGDG content and Fv /Fm (A), DF /Fm (B), Fv 9 /Fm 9 (C) and qN (D) in fully expanded leaves of 5-to-6-week-old A. thaliana wild-type plants, the dgd1 mutant, the transgenic lines R379, R382 and R383 (DGD1 co-suppression), and R376, a DGD1 complemented dgd1 mutant line. All values were obtained either after a 1 h dark incubation (A) or at the end of a 15-min period of exposure to a PPFD of 75 mmol photons m 22 s 21 (B–D). All values represent the means of at least five different determinations. The standard deviation of the mean is indicated. The continuous lines correspond to the first-order regression to the data. For comparison, the values obtained with the dgd1 /pho1 double mutant are indicated (n).

¨ et al. / Journal of Photochemistry and Photobiology B: Biology 61 (2001) 46 – 51 H. Hartel

50

Table 3 Fatty acid composition of DGDG of mature leaves of 5-to-6-week-old A. thaliana wild type, the pho1 mutant, the dgd1 mutant, the dgd1 /pho1 double mutant and the transgenic lines R382 and R383 (DGD1 co-suppression), and R376, a DGD1 complemented dgd1 mutant line. Lipids were separated by one-dimensional TLC and individual fatty acids were quantified by GLC. All values are given as moles percent and represent the means of three different determinations with individual plants (S.D.#8%) Plant

16:0

16:1

16:2

16:3

18:0

18:1

18:2

18:3

16C / 18C ratio

WT pho1 dgd1 dgd1 /pho1 R382 R383 R376

16.6 20.6 31.4 37.7 15.5 15.9 14.4

1.5 2.0 10.5 5.9 2.9 2.9 1.7

0.2 0.3 0.8 1.1 1.2 0.5 0.6

2.6 2.0 2.0 1.3 2.3 2.3 3.0

1.4 1.8 5.7 2.7 1.6 1.5 1.4

1.9 2.0 3.8 3.2 2.2 1.9 1.7

4.6 6.8 5.4 8.6 3.9 3.5 3.4

71.2 64.5 40.3 39.5 70.4 71.5 73.8

0.26 0.33 0.81 0.85 0.28 0.28 0.25

for DF /Fm 9 versus DGDG content (r 2 5 0.955) and the lowest for qN versus DGDG content (r 2 5 0.820). As already apparent from the data depicted in Table 2, complementation of the dgd1 mutant with DGD1 (line R376) completely restored the wild-type characteristics, while photosynthesis-related parameters were intermediate in transgenic plants with a partial reduction in DGD1 activities (Fig. 2). For comparison, the values obtained with the dgd1 /pho1 double mutant are indicated (triangles). Irrespective of the similar DGDG content in the dgd1 /pho1 double mutant and in the transgenic lines R382 and R383 (about 9 mol%), the values obtained for the mutant do not fit the linear regression lines. This observation is fully consistent with the conclusion drawn from the data in Fig. 1 that the DGDG species that accumulates upon phosphate deprivation in the dgd1 mutant does not appear to have any restoring effect on photosynthesis. It suggests that the DGDG produced under phosphate deprivation does not substitute for the DGDG present in the wild type. A qualitative analysis of the fatty acid composition of DGDG reveals an increase in 16:0 carbon fatty acids in pho1 and more so in dgd1 /pho1 accompanied by a strong decline in 18:3 carbon fatty acids (Table 3). On the other hand, the fatty acid composition of DGDG in transgenic plants expressing cDNA for DGD1 almost completely matches the wild-type pattern. As a result, irrespective of the similar DGDG proportion, the ratio of 16 carbon / 18 carbon fatty acids in DGDG of R383 and R382 was significantly lower as in dgd1 /pho1, but it was the same as in the wild type. Thus, the structure of the molecular DGDG species in the transformants R382 and R383 versus the dgd1 /pho1 double mutant is different, which in turn may provide a conclusive explanation for the differences in the impact of DGDG on the photosynthetic apparatus.

4. Conclusions Assignment of individual lipids to a specific function within the photosynthetic apparatus based on biochemical and physiological comparison analyses has been notorious-

ly difficult. In this study we took advantage of the availability of the DGD1 gene [11] and our knowledge about the plant response to phosphate deprivation [12,13,15] to further unravel the role of DGDG in the structural and functional integrity of the photosynthetic apparatus. Based on the comparison of structural and functional parameters in leaves of the dgd1 mutant, transgenic dgd1 plants expressing a wild-type DGD1 cDNA and wild-type plants with a partially reduced amount of DGDG due to post-transcriptional gene silencing following the expression of the DGD1 cDNA, the presence of DGDG is inferred to be an essential requirement for proper function of the photosynthetic apparatus. Thus, the observed structural changes in the photosynthetic apparatus of the dgd1 mutant [8], functional alterations in the water-oxidizing complex [21], changes in the chloroplast protein import characteristics [22], and alterations in the light utilization [9] can be traced back directly to the 90% DGDG deficiency in the mutant. The lack of a correlation in the genetically-induced phosphate deprived pho1 and more so in the dgd1 /pho1 double mutant suggests that the causal relationship between DGDG accumulation and the structural and functional integrity of the photosynthetic apparatus does not always hold true. Recent evidence demonstrated the enrichment of DGDG in non-thylakoidal membranes upon phosphate deprivation [15], providing a reasonable explanation why the DGDG that accumulates in the dgd1 /pho1 double mutant cannot substitute for DGDG formed through the DGD1 pathway in the wild type. It implies the existence of localized DGDG pools in plant membranes, similarly as proposed earlier [26]. Additionally, the distinct molecular structure of the phosphate-induced DGDG may prevent this DGDG from substituting for the photosynthesis-relevant molecular DGDG form. This highly complex nature of DGDG biosynthesis has to be taken into account in further considerations regarding an unambiguous assignment of the exact roles of individual lipids by comparative physiological studies. Our results may also provide a clue to explain the high variability in the membrane lipid composition found in the same plant species under different environmental conditions [27,28].

¨ et al. / Journal of Photochemistry and Photobiology B: Biology 61 (2001) 46 – 51 H. Hartel

Acknowledgements This work was supported by a grant from the U.S. Department of Energy (ER20305 to C.B.).

References [1] E. Heinz, G. Roughan, Similarities and differences in lipid metabolism of chloroplasts isolated from 18:3 and 16:3 plants, Plant Physiol. 72 (1983) 273–279. [2] J. Browse, C. Somerville, Glycerolipid synthesis: biochemistry and regulation, Annu. Rev. Plant Physiol. Plant Mol. Biol. 42 (1991) 467–506. ´ ` [3] J. Joyard, E. Marechal, C. Miege, M.A. Block, A.-J. Dorne, R. Douce, Structure, distribution and biosynthesis of glycerolipids from higher plant chloroplasts, in: P.-A. Siegenthaler, N. Murata (Eds.), Lipids in Photosynthesis: Structure, Function and Genetics, Kluwer Academic, Dordrecht, 1998, pp. 21–52. [4] K. Gounaris, D. Whitford, J. Barber, The effect of thylakoid lipids on an oxygen-evolving photosystem II preparation, FEBS Lett. 163 (1983) 230–234. [5] U. Pick, M. Weiss, K. Gounaris, J. Barber, The role of different thylakoid glycolipids in the function of reconstituted chloroplast ATP synthase, Biochim. Biophys. Acta 891 (1987) 28–39. ¨ ¨ [6] S. Nußberger, K. Dorr, N. Wang, W. Kuhlbrandt, Lipid–protein interactions in crystals of plant light-harvesting complex, J. Mol. Biol. 234 (1993) 347–356. ¨ [7] P. Dormann, S. Hoffmann-Benning, I. Balbo, C. Benning, Isolation and characterization of an Arabidopsis mutant deficient in the thylakoid lipid digalactosyl diacylglycerol, Plant Cell 7 (1995) 1801–1810. ¨ ¨ [8] H. Hartel, H. Lokstein, P. Dormann, B. Grimm, C. Benning, Changes in the composition of the photosynthetic apparatus in the galactolipid deficient dgd1 mutant of Arabidopsis thaliana, Plant Physiol. 115 (1997) 1175–1184. ¨ ¨ [9] H. Hartel, H. Lokstein, P. Dormann, R.N. Trethewey, C. Benning, Photosynthetic light utilization and xanthophyll cycle activity in the galactolipid deficient dgd1 mutant of Arabidopsis thaliana, Plant Physiol. Biochem. 36 (1998) 407–417. ¨ [10] P. Jarvis, P. Dormann, C.A. Peto, J. Lutes, C. Benning, J. Chory, Galactolipid deficiency and abnormal chloroplast development in the Arabidopsis MGD synthase 1 mutant, Proc. Natl. Acad. Sci. USA 97 (2000) 8175–8179. ¨ [11] P. Dormann, I. Balbo, C. Benning, Arabidopsis galactolipid biosynthesis and lipid trafficking mediated by DGD1, Science 284 (1999) 2181–2184. ¨ [12] B. Essigmann, S. Guler, R.-A. Narang, D. Linke, C. Benning, Phosphate availability affects the thylakoid composition and the expression of SQD1, a gene required for sulfolipid biosynthesis in Arabidopsis thaliana, Proc. Natl. Acad. Sci. USA 95 (1998) 1950– 1955.

51

¨ [13] H. Hartel, B. Essigmann, H. Lokstein, S. Hoffmann-Benning, M. Peters-Kottig, C. Benning, The phospholipid-deficient pho1 mutant of Arabidopsis thaliana is affected in the organization, but not in the light acclimation, of the thylakoid membrane, Biochim. Biophys. Acta 1415 (1998) 205–218. [14] Y. Poirier, S. Thoma, C. Somerville, J. Schiefelbein, A mutant of Arabidopsis deficient in xylem loading of phosphate, Plant Physiol. 97 (1991) 1087–1093. ¨ ¨ [15] H. Hartel, P. Dormann, C. Benning, DGD1-independent biosynthesis of extraplastidic galactolipids after phosphate deprivation in Arabidopsis, Proc. Natl. Acad. Sci. USA 97 (2000) 10649–10654. [16] T. Murashige, F. Skoog, A revised medium for rapid growth and bioassays with tobacco tissue cultures, Physiol. Plant. 15 (1962) 473–497. [17] H.K. Lichtenthaler, Chlorophylls and carotenoids: pigments of photosynthetic membranes, Methods Enzymol. 148 (1987) 350– 382. ¨ [18] H. Hartel, I. Reinhardt, B. Grimm, Relationship between energydependent fluorescence quenching and xanthophyll-cycle-pigments in transgenic chlorophyll-deficient tobacco grown under different light intensities, J. Photochem. Photobiol. Biol. 43 (1998) 136–145. [19] B. Genty, J.M. Briantais, N.R. Baker, The relationship between the quantum yield of photosynthetic electron transport and quenching of chlorophyll fluorescence, Biochim. Biophys. Acta 990 (1989) 535– 546. [20] O. Van Kooten, J.F.H. Snel, The use of fluorescence nomenclature in plant stress physiology, Photosynth. Res. 25 (1990) 147–150. ¨ [21] F. Reifarth, G. Christen, A.G. Seeliger, P. Dormann, C. Benning, G. Renger, Modification of the water oxidizing complex in leaves of the dgd1 mutant of Arabidopsis thaliana deficient in the galactolipid digalactosyl diacylglycerol, Biochemistry 36 (1997) 11769–11776. [22] L.-J. Chen, H. Li, A mutant deficient in the plastid lipid DGD is defective in protein import into chloroplasts, Plant J. 16 (1998) 33–39. [23] D.E. Minikin, H. Abdolrahimzadeh, J. Baddiley, Replacement of acidic phosphates by acidic glycolipids in Pseudomonas diminuta, Nature (London) 249 (1974) 268–269. [24] C. Benning, Biosynthesis and function of the sulfolipid sulfoquinovosyl diacylglycerol, Annu. Rev. Plant Physiol. Plant Mol. Biol. 49 (1998) 53–75. [25] P. Horton, A.V. Ruban, R.G. Walters, Regulation of light harvesting in green plants, Annu. Rev. Plant Physiol. Plant Mol. Biol. 47 (1996) 655–684. [26] J.P. Williams, M.U. Kahn, Lipid metabolism in leaves of an 18:4 plant, Echium plantagineum: a model of galactolipid biosynthesis in 18:3 and 18:4 plants, Plant Physiol. Biochem. 34 (1996) 93–100. [27] R.W. Pearcy, Effect of growth temperature on the fatty acid composition of the leaf lipids in Atriplex lentiformis (Torr.) Wats, Plant Physiol. 61 (1978) 484–486. [28] K.O. Burkey, R.F. Wilson, R. Wells, Effects of canopy shade on the lipid composition of soybean leaves, Physiol. Plant. 101 (1997) 591–598.