Proc. Nati. Acad. Sci. USA Vol. 85, pp. 4966-4970, July 1988 Biochemistry requirement and kinetic studies of solubilized UDP-galactose:diacylglycerol galactosyltransferase activity from spinach envelope membranes (monogalactosyldiacylglycerol biosynthesis/phosphatidylglycerol requirement/-lipid interactions/ UDP inhibition/Spinacia okracea) JACQUES COVES, JACQUES JOYARD, AND ROLAND DOUCE Laboratoire de Physiologie Cellulaire Vdgdtale, Unite Associ&e au Centre National de la Recherche Scientifique No. 576, D6partement de Recherche Fondamentale, Centre d'Etudes Nucldaires de et Universit6 Joseph Fourier, 85 X, 38041 Grenoble Cedex, France Communicated by Andrew A. Benson, February 19, 1988

ABSTRACT We have demonstrated a lipid requirement described lipid requirements for membrane-bound enzymes for the UDPgalactose:1,2-diacylglycerol 3-j8-D-galactosyl- (8-17), and, to our knowledge, no studies have been done on transferase (or monogalactosyldiacylglycerol synthase; EC enzymes involved in the biosynthesis of polar . 2.4.1.46), an enzyme involved in the biosynthesis of monoga- The purpose of this study was to investigate the role of the lactosyldiacylglycerol, solubilized from chloroplast envelope different envelope lipids on the MGDG synthase activity and membranes and partially purified by hydroxyapatite chroma- to investigate the kinetic mechanisms of the delipidated tography. The enzyme fraction was highly delipidated (<0.1 enzyme with respect to diacylglycerol and UDP-Gal, respec- mg of lipid per mg of protein), and addition of lipids extracted tively. from chloroplast membranes was necessary to reveal the activity. Acidic glycerolipids, and especially phosphatidylglyc- MATERIALS AND METHODS erol, were the best activators of the enzyme. The preparation of a delipidated enzyme fraction and the development of Purification of Chloroplast Envelope Membranes. Total optimal assay conditions were prerequisites for the determi- envelope membranes were prepared from intact spinach nation of the kinetic parameters for the hydrophobic substrate according to Douce et al. (18, 19). The envelope of the enzyme, diacylglycerol. In addition, we have demon- fraction was stored concentrated (10 mg of protein per ml) in strated the existence of two substrate-binding sites: a hydro- a medium containing 50 mM Mops-NaOH (pH 7.8) and 1 mM phobic one for diacylglycerol and a hydrophilic one for dithiothreitol (4-6) and in liquid nitrogen. UDP-galactose. Solubilization and Partial Purification of MGDG Synthase from Chloroplast Envelope Membranes. 3-[(3-Cholamido- The envelope membranes are involved in galactolipid propyl)dimethylammonio]-1-propanesulfonate (CHAPS) In was used to solubilize envelope membranes. Envelope mem- metabolism (1). spinach, the inner membrane is charac- branes (about 10 mg of protein) were incubated for 30 min at terized by the presence ofa UDPgalactose: 1,2-diacylglycerol 0°C under gentle agitation in 15 ml of solubilization medium 3-,/-D-galactosyltransferase (EC 2.4.1.46) (1, 2), or MGDG containing 50 mM Mops-NaOH (pH 7.8), 1 mM dithiothrei- synthase, which transfers a galactose from a water-soluble tol, 6 mM CHAPS, and 50 mM KH2PO4. After incubation, donor, UDP-galactose (UDP-Gal), to a hydrophobic acceptor the mixture was centrifuged for 15 min at 45,000 rpm molecule, diacylglycerol, for synthesizing the most abundant (Beckman L2 65B, rotor SW 50) and the supernatant (0.5-0.6 polar lipid in nature, monogalactosyldiacylglycerol (3): mg of protein per ml), which contained all the MGDG -* synthase activity, was recovered. The solubilized envelope UDP-Gal + 1,2-diacyl-sn-glycerol membranes were then loaded on top of a hydroxyapatite- 1,2-diacyl-3-0-,f-D-galactopyranosyl-sn-glycerol + UDP. Ultrogel (HA-Ultrogel, IBF Biotechnics, Villeneuve-la- Garenne, France) column (10 x 0.8 cm), equilibrated prior to In previous papers, we have described the solubilization of the experiment with 50 mM Mops-NaOH (pH 7.8)/6 mM envelope membrane by detergents (4, 5), the devel- CHAPS/1 mM dithiothreitol/50 mM KH2PO4. The proteins opment of a specific assay for the solubilized MGDG syn- were eluted by using a nonlinear KH2PO4 gradient (4, 6). The thase activity, and the partial purification of this enzyme by MGDG synthase-enriched fraction (peak 3; see refs. 4 and 6) hydroxyapatite chromatography (4, 6). The measurement of was recovered and stored in liquid nitrogen for further use. the activity requires the addition of envelope lipids (4, 6). Purification of Thylakoid Glycerolipids and Analyses of However, the role ofenvelope lipids can be complex. First, Lipids. In our standard assay conditions (6) we used envelope they provide substrate for the MGDG synthase, since they lipids-extracted from purified envelope membranes accord- contain about 10-15% diacylglycerol (7). According to San- ing to Bligh and Dyer (20)-to reveal MGDG synthase dermann (8), various mechanisms could also be involved in activity (6). Glycerolipids are almost identical in envelope the response of membrane enzymes to lipids: some enzymes membranes and thylakoids (21). Therefore, since thylakoids require lipid for activity; lipids can also be required for the are far more abundant than envelope membranes, we used solubilization of water-insoluble substrates or membrane- thylakoids as a source of individual chloroplast glycerolipids bound enzymes; and, finally, lipids may have direct kinetic (except for diacylglycerol; see below). Thylakoid lipids were effects on the enzymatic activity. All these aspects have been extracted with chloroform/methanol (1:2, vol/vol) according analyzed in numerous enzymes from animal tissues (for a to Bligh and Dyer (20) and separated on a silicic acid column review, see ref. 8). In plants, only a few studies have Abbreviations: UDP-Gal, UDP-galactose; MGDG synthase, mono- The publication costs of this article were defrayed in part by page charge galactosyldiacylglycerol synthase; PtdGro, phosphatidylglycerol; Ptd- payment. This article must therefore be hereby marked "advertisement" Cho, phosphatidylcholine; Ptdlns, phosphatidylinositol; CHAPS, 3- in accordance with 18 U.S.C. §1734 solely to indicate this fact. [(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate. 4966 Downloaded by guest on October 1, 2021 Biochemistry: Cove's et al. Proc. Natl. Acad. Sci. USA 85 (1988) 4967

(Bio-Sil HA -325 mesh, Bio-Rad) equilibrated in pure Table 1. Effect of envelope lipids on the MGDG synthase chloroform (100 g of gel per g of lipid). Pigments, free fatty activity of chloroplast envelope membranes after acids, and diacylglycerols were eluted with 5-6 vol of pure solubilization and partial purification chloroform and were discarded. Galactolipids were eluted MGDG synthase activity, with 5 vol of pure acetone, and finally phospholipids were nmol of galactose eluted with 5 vol of pure methanol. Galactolipids and phos- incorporated per hr per pholipids were then purified on thin-layer chromatography (TLC) plates (silica gel 60, Merck) developed, respectively, Lipid mg of protein with chloroform/methanol (80:20, vol/vol) and with chlo- content, Without With mg/mg envelope envelope roform/methanol/acetic acid (65:25:8, vol/vol). Lipids were Sample protein lipids lipids quantified by gas chromatography according to Allen and Good (22). Amounts of lipids were expressed as ug of fatty Envelope membranes 1.2 494 ND acid equivalent. Solubilized envelope The amounts ofglycerolipids in total envelope membranes, membranes 1.0 11 145 solubilized envelope membranes, and partially purified MGDG synthase- MGDG synthase fraction were determined according to the enriched fraction <0.1 9 535 same procedure. MGDG synthase activity was assayed in purified envelope mem- We also used phosphatidylglycerol (PtdGro) prepared by branes, and after solubilization of the membranes with CHAPS and transphosphatidylation from egg phosphatidylcholine partial purification of the enzyme by hydroxyapatite chromatogra- (PtdCho) (in the presence of free glycerol) and supplied by phy. The enzyme was assayed according to Coves et al. (6) in the absence or in the presence ofenvelope lipids (150 ,ug/300 Al ofassay Sigma (sodium salt). The fatty acid composition of thylakoid mixture). The lipid content in each fraction was determined as and commercial PtdGro was determined as described above. described in the text; in the MGDG synthase-enriched fraction it was Purification of 1,2-Diacyl-sn-glycerol. 1,2-Dioleoyl-sn- at the limit of detection. ND, not determined. glycerol (diacylglycerol) was purified from a mixture of 1,2- and 1,3-diacyl-sn-glycerol (Fluka) on TLC plates (silica gel about 2 Aumol of galactose incorporated per hr per mg of 60, Merck) developed with petroleum ether/diethyl ether/ protein (Fig. 1). acetic acid (80:20:1, vol/vol). Diacylglycerol was quantified Such a stimulation by total envelope lipids was expected according to Allen and Good (22). since diacylglycerol, the hydrophobic substrate of the en- Assay of MGDG Synthase. Envelope lipids or purified zyme, represents 10-15% of the envelope glycerolipids (7), thylakoid glycerolipids dissolved at various concentrations in when the chloroplasts used for envelope purification are not chloroform were introduced into glass tubes. After evapora- thermolysin treated (24). However, the stimulation observed tion of the solvent under a stream of argon, 150 1,l of could be due to a specific effect of some envelope lipids. incubation medium containing 50 mM Mops-NaOH (pH 7.8), Therefore, we purified glycerolipids to investigate their effect 6 mM CHAPS, 1 mM dithiothreitol, and 250 mM KH2PO4 on MGDG synthase activity. Table 2 shows that addition of was added and the tubes were vigorously mixed to resuspend acidic chloroplast glycerolipids (sulfolipid, PtdIns, and Ptd- lipids. Then 50 ,ul (1.2-1.6 ,ug of protein) of the MGDG Gro) led to a significant stimulation ofthe activity. The effect synthase-enriched fraction (peak 3; see ref. 6) was added and of PtdGro was most important: final rates higher than 7.5 the tubes were vigorously mixed again. After 5 min at room pumol ofgalactose incorporated per hr per mg ofprotein were temperature, the reaction was started by addition of UDP- obtained. PtdCho, a zwitterionic phospholipid, and un- [14C]Gal (New England Nuclear, 45.8 MBq/mmol) at various charged galactolipids (mono- and digalactosyldiacylglycerol) concentrations, as indicated. The reaction was stopped after were almost without effect on MGDG synthase activity. 10 min, the lipids were extracted, and the radioactivity of Therefore, in the presence of6 mM CHAPS, addition oflarge labeled galactolipids was determined by liquid scintillation amounts of PtdGro to the incubation medium was necessary counting, as described by Coves et al. (6). We have verified for maximal activity of MGDG synthase. To determine that the reaction was linear for at least 20 min under these whether the nature of the fatty acid chains of PtdGro was conditions. The activity was expressed as ,umol of galactose responsible for such a stimulation, we compared the effect of incorporated per hr per mg of protein. Experiments have PtdGro from chloroplasts with that of PtdGro prepared by been reproduced at least three times. transphosphatidylation from egg PtdCho. (Compositions are Protein Determination. Protein was measured according to LLJ Bradford (23), with bovine serum albumin as standard. w 0 (.) I--, 2 0 0 RESULTS < C 0 < o Lipid Requirement ofthe MGDG Synthase Activity. The use of a detergent for the solubilization of membrane proteins LD " leads also to the o E solubilization of membrane lipids. After 1 column chromatography, lipids and proteins can be eluted separately and enzymatic activity of proteins requiring hy- cr Z /0 drophobic substrates becomes difficult to measure. Table 1 demonstrates that this was the case for MGDG synthase. In 0 40' contrast to native and solubilized envelope membranes, the 0 z MGDG synthase-enriched fraction was highly delipidated 1 2 3 4 5 6 (the values obtained were at the limit of the detection by gas chromatography). Table 1 demonstrates that addition of ENVELOPE LIPIDS (mg/ml) envelope lipids is necessary to reveal the enzymatic activity FIG. 1. Effect of total envelope lipids on the activity of MGDG after solubilization and in the MGDG synthase-enriched synthase partially purified by hydroxyapatite chromatography. The fraction. The MGDG synthase activity was proportional to enzyme was purified from solubilized chloroplast envelope mem- the amount ofadded envelope lipids up to about 2 mg oflipids branes and assayed in the presence of increasing amounts of per ml of incubation mixture and reached a maximum of envelope lipids, in a final volume of 300 ul. Downloaded by guest on October 1, 2021 4968 Biochemistry: Cov.s et al. Proc. Natl. Acad. Sci. USA 85 (1988) Table 2. Lipid requirement of the MGDG synthase partially purified from chloroplast envelope membranes MGDG synthase activity, 10 of galactose incorporated jumol llJ Added lipids per hr per mg of protein C,) Diacylglycerol 0.16 0 Diacylglycerol + sulfolipid 1.60 < . _j a) Diacylglycerol + Ptdlns 2.25 < o Diacylglycerol + PtdGro 7.52

Diacylglycerol + PtdCho 0.60 E Diacylglycerol + MGDG 0.53 0 cm Diacylglycerol + DGDG 0.70 The MGDG synthase was purified by hydroxyapatite chromatog- raphy and assayed as described in the text. The amounts of diacylglycerol and other glycerolipids added to the incubation z mixture were 100 and 200 gg, respectively, in a total reaction mixture of 200 pl. Ptdlns, phosphatidylinositol; DGDG, digalactosyl- diacylglycerol. given in Table 3.) As shown in Fig. 2, no striking differences in the stimulation of the MGDG synthase could be noted, 100 200 300 400 500 although spinach chloroplast PtdGro-in contrast with egg PtdGro-contains large amounts ofthe polyunsaturated fatty PHOSPHATIDYLGLYCEROL (,Ug) acid linolenic acid (18:3) and a unique fatty acid, trans-3- FIG. 2. Effect of PtdGro on the activity of MGDG synthase hexadecenoic acid (16:1,) (25). Therefore, stimulation of the partially purified by hydroxyapatite chromatography. The enzyme MGDG synthase by PtdGro is due to a specific recognition of was purified from solubilized chloroplast envelope membrane and the chemical structure of the polar head group rather than assayed in the presence of diacylglycerol (100 ,ug) and increasing being due to a strict specificity for the fatty acid chains. amounts of PtdGro (PG) in a final volume of 200 ,ul. PtdGro from Whatever the source of PtdGro was, MGDG synthase activ- thylakoids was purified as described in Materials and Methods. Egg ity was proportional to the amount of PtdGro until 1 mg/ml PtdGro was prepared from egg PtdCho by transphosphatidylation in (i.e., 200 ,ug of PtdGro per tube) and reached a maximum of the presence offree glycerol (Sigma). The fatty acid compositions of about 10 umol of galactose incorporated per hr per mg of the two phosphatidylglycerols are given in Table 3. protein for 1.2 mg of PtdGro per ml of incubation mixture. does not disturb the accessibility of the water-soluble sub- Determination of the Kinetic Parameters of the Reaction. strate or inhibitor to MGDG synthase. Moreover, the appar- Almost nothing is known about the relationship between the ent VmS, value was high: 10 pmol of galactose incorporated MGDG synthase and its hydrophobic substrate, diacylglyc- per hr per mg of because protein. erol, most of the previous studies have concerned Fig. 3 Lower shows that the reaction, as a function of the within membranes In enzyme envelope (1, 7, 26-32). diacylglycerol concentration and at saturating UDP-Gal con- these experiments, diacylglycerol was formed within the centration, is since the curves obtained are membrane either by envelope enzymes such as galactoli- complex, biphasic pid:galactolipid galactosyltransferase (26-28) and phospha- with respect to diacylglycerol concentration. The reason for tidic acid phosphohydrolase (28-30) or by exogenous phos- such a complex pattern is not yet clear. Nevertheless, the pholipase c, all of which produce diacylglycerol from phos- apparent Vms, value (10 ,umol of galactose incorporated per pholipids (31-33). Even in the early experiments (34, 35), hr per mg of protein) for the consumption of diacylglycerol which demonstrated the diacylglycerol requirement of the was, as expected, identical to the apparent Vm. value MGDG synthase by using acetone powder extracts of chlo- obtained with UDP-Gal as the variable substrate. The appar- roplasts, the kinetic parameters for diacylglycerol were not ent Km for diacylglycerol was estimated at 500 ,ug/ml (about determined. 1 mM). This high apparent Km value for diacylglycerol might Fig. 3 is Lineweaver-Burk plots for the inhibition by UDP be explained by the physical state in which this substrate is of the MGDG synthase activity, with UDP-Gal and diacyl- offered to the enzyme, as a mixture of diacylglycerol, glycerol as the changing substrates. Control experiments (in PtdGro, and detergent. However, because of the high appar- the absence of UDP) provide apparent Km and Vmax values ent Vmax value, it is likely that the Km value obtained reflects for each substrate. some reality within the envelope membranes, where high and Under these conditions, the apparent Km value for UDP- localized concentrations ofdiacylglycerol could be expected, Gal (about 100 ,LM) determined with the MGDG synthase- due to the functioning of the phosphatidic acid phosphohy- enriched fraction (Fig. 3 Upper) is very close to that previ- drolase (28-30). In the presence of UDP, the biphasic aspect ously reported for enzyme within the envelope (36). UDP of the curves was increased (Fig. 3 Lower). However, at inhibition, as expected, was competitive with respect to saturating diacylglycerol concentrations, UDP inhibition was UDP-Gal and the Ki value (8-10 uM) is similar to that clearly uncompetitive. The existence of two distinct types of previously reported for the nonsolubilized enzyme (27). inhibition by UDP toward UDP-Gal and diacylglycerol sug- These results indicate that the solubilization of the enzyme gests that the MGDG synthase possess two different sub- strate-binding sites, a hydrophilic one for UDP-Gal and a Table 3. Fatty acid compositions of phosphatidylglycerols from hydrophobic one for diacylglycerol. thylakoids and eggs PtdGro Composition, wt % DISCUSSION source 16:0 16:1, 18:0 18:1 18:2 18:3 An important limitation for the study ofMGDG synthase was Thylakoids 11.0 37.1 0.9 1.0 1.9 48.0 the determination of optimal and reproducible conditions to Egg 37.8 16.6 33.5 14.9 1.2 assay the enzyme after membrane solubilization and prepa- ration of a delipidated enzyme fraction. The demonstration of Downloaded by guest on October 1, 2021 Biochemistry: Cov6s et al. Proc. Natl. Acad. Sci. USA 85 (1988) 4969

4

3

2

- X 25miM UDP *

-. A CONTROL--\ Moo-o-o °. a I 10 20 30 40 0.01 0.02 0.03 0.04

1 1/UDP-GALACTOSE (mM) 1/DIACYLGLYCEROL (/Ag/ml) FIG. 3. Lineweaver-Burk plots of UDP inhibition of the purified MGDG synthase. The velocity of the reaction is expressed as Amol of galactose incorporated per hr per mg of protein. (Left) UDP-Gal was the variable substrate. The enzyme was assayed with increasing concentrations of UDP-Gal, in the presence ofdiacylglycerol (100 jzg) and PtdGro (200 ,ug) in a final volume of200 sul. This experiment (control) provides apparent values of Km and Vm. for UDP-Gal. The same experiment was done in the presence of UDP (25 or 100 ,uM) to determine the apparent Ki. (Right) Diacylglycerol was the variable substrate. The enzyme was assayed with increasing concentrations of diacylglycerol and in the presence of PtdGro (200 ,ug) and UDP-Gal (455 AuM) in a final volume of 200 p1. This experiment (control) provides apparent Km and V,,. values for diacylglycerol. The same experiment was done in the presence of UDP (25 or 100 ,LM) to determine the type of inhibition. a specific lipid requirement for the optimal activity of the and CHAPS. However, the solubilizing effect of phospho- enzyme was a major step toward solving this problem. lipids for water-insoluble substrates usually presents a broad The very high activity of the MGDG synthase-enriched selectivity toward phospholipids, and certain detergents are fraction obtained after hydroxyapatite chromatography often able to substitute for phospholipids (8). Obviously, this raised the problem of the significance of the purification was not the case here. However, we must keep in mind that factor for such enzymes. The optimal assay conditions for the although full activation was obtained only with PtdGro, enzyme in native and solubilized envelope membranes and in CHAPS was present at all stages of our experiments. the fraction prepared by hydroxyapatite chromatography are To discriminate between an obligatory PtdGro requirement very different. For instance, on the basis of optimal activity and a simple modulation ofthe enzymatic activity by PtdGro, in the solubilized envelope membranes, the purification we have to analyze whether all the steps in the enzymatic factor was (i) 6.3, (ii) 22, and (iii) 110 when the enzyme was process require PtdGro. For instance, is PtdGro necessary assayed, respectively, (i) under the same conditions as in the for UDP-Gal binding? A photoaffinity analogue of UDP solubilized membrane (6), (ii) with optimal envelope lipid (NAP4-UDP or N-4-azido-2-nitrophenyl-4-aminobutyryl-3'- concentration (Fig. 1), and (iii) with optimal PtdGro concen- UDP) requires, for efficient competitive inhibition of the tration (Fig. 2). These results demonstrate, in any case, that activity (37), the presence of octyl glucoside at a concentra- a significant purification ofthe MGDG synthase activity was tion (10 mM) that does not solubilize the envelope (5). achieved by hydroxyapatite chromatography. The high spe- Therefore, it is possible that the binding of UDP-Gal to the cific activity of MGDG synthase also raised the question of enzyme could be modulated by lipids. the specific activity in vivo, within the membrane. MGDG Tight lipid-protein interactions (the "boundary" lipid synthase is probably only a minor component ofthe envelope concept) are often necessary for the full activity of many membranes (1, 6, 35). Thus the enzyme should be extremely membrane-bound enzymes (8, 38). Thus, the stimulation of active to support the biosynthetic requirements for chloro- MGDG synthase activity by PtdGro could reflect the in vivo plast biogenesis. situation, since (at least in spinach) the enzyme involved in The lipid requirement of the purified MGDG synthase for MGDG synthesis is localized in the inner envelope mem- PtdGro is remarkable: removal of envelope lipids during brane, which contains 7-9o PtdGro (2). Therefore, the enzyme purification by hydroxyapatite chromatography presence of PtdGro as a boundary lipid for the membrane- leads to a loss of activity; the activity is restored upon bound enzyme cannot be excluded. It is possible that solu- addition of envelope lipids and especially PtdGro. This bilization of MGDG synthase by CHAPS actually displaced corresponds to a typical lipid dependence, as discussed by glycerolipid molecules that bind to the hydrophobic surface Sandermann (8). Requirement for acidic lipids has been of proteins, at "binding" or "contact" sites, as discussed by described for several plant enzymes, such as NADH:cyto- Volwerk et al. (39). Then, the same process (39) would occur chrome c oxidoreductase (15) and UDP-glucose:sterol j-D- upon addition of PtdGro to the lipid-depleted protein, thus glucosyltransferase (13, 14, 17), but the stimulation of the resulting in the exchange of bound CHAPS molecules by activity by PtdGro is 2-3 times that obtained with PtdCho. PtdGro at the hydrophobic surface of the protein, and With MGDG synthase, stimulation by PtdGro was more than therefore in the restoration offull activity. However, nothing 10-fold that obtained with PtdCho. is known about the actual distribution of PtdGro within the It is difficult to determine why addition of PtdGro is membrane or about the possible specific association of necessary to assay the delipidated MGDG synthase. First, it PtdGro with any envelope proteins. In our experiments, the is possible that PtdGro behaves as a simple "cofactor" for amount of PtdGro necessary for full activation of MGDG solubilization of diacylglycerol. This hypothesis is supported synthase was high, compared to the amount of proteins by the observation that PtdGro was the only glycerolipid (about 200 ,ug of PtdGro per ,g of protein). On the contrary, forming an isotropic solution when added to diacylglycerol within a membrane, only about 20 boundary lipid molecules Downloaded by guest on October 1, 2021 4970 Biochemistry: Cov6s et A Proc. Natl. Acad. Sci. USA 85 (1988) are expected to bind on a 32,000-dalton protein (40), thus 14. Ullmann, P., Rimmele, D., Benveniste, P. & Bouvier-Nave, P. corresponding to a 0.5:1 weight ratio of lipid to protein. (1984) Plant Sci. Lett. 36, 29-36. forthe 15. Mazliak, P., Jolliot, A., Justin, A.-M. & Kader, J.-C. (1985) Finally, the most important lipid-protein interaction Phytochemistry 24, 1163-1168. envelope MGDG synthase is the binding of one of the 16. Kalinovska, M. & Wojciechowski, Z. A. (1986) Phytochem- substrates, diacylglycerol. We have demonstrated that the istry 25, 45-49. diacylglycerol binding site is distinct from that of UDP-Gal. 17. Ullmann, P., Bouvier-Nave, P. & Benveniste, P. (1987) Plant In addition, since the apparent Km for UDP-Gal is about 100 Physiol. 85, 51-55. whereas the K; for the competitive inhibition by UDP is 18. Douce, R. & Joyard, J. (1982) in Methods in Chloroplast ,uM, Molecular Biology, eds. Edelman, M., Hallick, R. & Chua, only 18 gM, our results suggest that UDP is the most N.-H. (Elsevier/North-Holland, Amsterdam), pp. 239-256. important part of UDP-Gal for binding to the active site. 19. Douce, R., Holtz, R. B. & Benson, A. A. (1973) J. Biol. Chem. However, although binding of UDP-Gal and binding of 248, 7215-7222. diacylglycerol are separate events, it is not yet known which 20. Bligh, E. G. & Dyer, W. J. (1959) Can. J. Biochem. Physiol. 37, substrate binds first, and whether UDP is released from the 911-917. site after cleavage of galactose and transfer to the 21. Siebertz, H. P., Heinz, E., Linscheid, M., Joyard, J. & Douce, active R. (1979) Eur. J. Biochem. 101, 429-438. diacylglycerol moiety or whether cleavage of UDP-Gal oc- 22. Allen, C. F. & Good, P. (1971) Methods Enzymol. 23, 523-547. curs during galactosylation of diacylglycerol. 23. Bradford, M. M. (1976) Anal. Biochem. 72, 248-254. 24. Dome, A.-J., Block, M. A., Joyard, J. & Douce, R. (1982) We are indebted to Dr. A. J. Dome for his generous gift ofpurified FEBS Lett. 145, 30-34. sulfolipid, forfruitful discussions, and for constant advice. Dr. M. A. 25. Haverkate, F. & Van Deenen, L. L. M. (1965) Biochim. Block is also gratefully acknowledged for her collaboration in the Biophys. Acta 106, 78-92. early stages of this work. 26. Van Besouw, A. & Wintermans, J. F. G. M. (1978) Biochim. Biophys. Acta 529, 44-53. 1. Joyard, J. & Douce, R. (1987) in The Biochemistry ofPlants: 27. Van Besouw, A. & Wintermans, J. F. G. M. (1979) FEBS Lett. Lipids, ed. Stumpf, P. K. (Academic, New York), Vol. 9, pp. 102, 33-37. 215-274. 28. Dome, A.-J., Block, M. A., Joyard, J. & Douce, R. (1982) in R. J. Biochemistry and Metabolism of Plant Lipids, eds. Winter- 2. Block, M. A., Dome, A.-J., Joyard, J. & Douce, (1983) & (Elsevier, Amsterdam), Biol. Chem. 258, 13273-13280. mans, J. F. G. M. Kuiper, P. J. C. 378- pp. 153-164. 3. Gounaris, K. & Barber, J. (1983) Trends Biochem. Sci. 8, 29. Joyard, J. & Douce, R. (1977) Biochim. Biophys. Acta 486, 273- 381. 285. 4. Coves, J., Pineau, B., Block, M. A., Joyard, J. & Douce, R. 30. Joyard, J. & Douce, R. (1979) FEBS Lett. 102, 147-150. (1987) in Plant Membranes: Structure, Function, Biogenesis, 31. Dome, A.-J., Block, M. A., Joyard, J. & Douce, R. (1985) J. eds. Leaver, C. & Sze, H. (Liss, New York), pp. 103-122. Cell Biol. 100, 1690-1697. 5. Coves, J., Pineau, B., Joyard, J. & Douce, R. (1988) Plant 32. Heemskerk, J. W. M., Jacobs, F. H. H., Scheijen, M. A. M., Physiol. Biochem. 26, 151-163. Helsper, J. P. F. G. & Wintermans, J. F. G. M. (1987) Bio- 6. Coves, J., Block, M. A., Joyard, J. & Douce, R. (1986) FEBS chim. Biophys. Acta 918, 189-203. Lett. 208, 401-406. 33. Oursel, A., Escoffier, A., Kader, J.-C., Dubacq, J.-P. & 7. Joyard, J. & Douce, R. (1976) Biochim. Biophys. Acta 424, 125- Tremolieres, A. (1987) FEBS Lett. 219, 393-399. 131. 34. Mudd, J. B., Van Vliet, H. H. D. M. & Van Deenen, L. L. M. 8. Sandermann, H., Jr. (1978) Biochim. Biophys. Acta 515, 209- (1969) J. Lipid Res. 10, 623-630. 237. 35. Eccleshall, T. R. & Hawke, J. C. (1971) Phytochemistry 10, 9. Cocucci, M. & Ballarin-Denti, A. (1981) Plant Physiol. 68, 377- 3035-3045. 381. 36. Joyard, J. (1979) These de Doctorat d'Etat (Univ. of Grenoble, 10. Jolliot, A., Demandre, C. & Mazliak, P. (1981) Plant Physiol. France). 67, 9-11. 37. Coves, J. (1987) These de Doctorat (Univ. of Grenoble, 11. Jolliot, A., Justin, A.-M., Bimont, E. & Mazliak, P. (1982) France). Plant Physiol. 70, 206-210. 38. Sandermann, H., Jr. (1983) Trends Biochem. Sci. 8, 408-411. 12. Heiniger, U. (1983) Plant Sci. Lett. 32, 35-41. 39. Volwerk, J. J., Mrsny, R. J., Patapoff, T. W., Jost, P. C. & 13. Bouvier-Nave, P., Ullmann, P., Rimmele, D. & Benveniste, P. Griffith, 1. H. (1987) Biochemistry 26, 466-475. (1984) Plant Sci. Lett. 36, 19-27. 40. Marsh, D. (1983) Trends Biochem. Sci. 8, 330-333. Downloaded by guest on October 1, 2021