Jun 22, 2000 - and David, H. L. (2000) FEMS Microbiol. Lett., 187, 95-1 0 I ... Velayos, A., Lopez-Matas, M. A,, Ruiz-Hidalgo, M. J. and. Eslava, A. P. ( 1997) ...
Biochemical Society Transactions (2000) Volume 28, part 6
3 Armstrong, G. A., Hundle. B. S. and Hearst, J. E. (1993) Methods Enzymol. 214,297-3 I I 4 Cunningham,Jr, F. X., Pogson, B., Sun, Z., McDonald, K. A,, DellaPenna, D. and Gantt, E. ( 1996) Plant Cell 8, I 6 13-1 626 5 Hugueney, P., Badillo, A., Cehn, H.-C., Klein, A., Hirschberg, J., Camara. B. and Kuntz. M. (1995) Plant J. 8,417424 6 Schnurr, G., Misawa, N. and Sandmann, G. ( 1996) Biochern. J. 3 15, 869-874 7 Bouvier, F.. Hugueney, P., d'Hadingue, A,, Kuntz. M. and Camara. B. ( 1994) Plant J. 6, 45-54 8 Bouvier, F., d'tladingue, A. and Camara, B. ( I 997) Arch.
pene cyclase gene must exist in this organism because p-carotene, which is the product of lycopene cyclization, was found in this cyanobacterium [19]. T h e genomic sequence data of one further completely sequenced cyanobacterium, Anabaena sp. PCC7120 (http ://www.kazusa.or.jp/cyano/ anabaena), did also lack similarities to the sequences of known lycopene cyclase genes.
Conclusion on the phylogeny of lycopene cyclase genes
Biochern. Biophys. 346, 53-64 9 Krubasik P. and Sandmann, G. (2000) Mol. Gen. Genet. 263,423432 10 Viveims. M.. Krubasik P., Houssaini-lraqui, M., Sandrnann, G. and David, H. L. (2000) FEMS Microbiol. Lett., 187, 95-1 0 I I I Ramakrishnan, L., Tran, H. T., Fedempiel, N. and Falkow, S. ( I 997) J. Bacteriol. 179,5862-5868 I 2 Krijgel, H., Krubasik P., Weber, K., Saluz, H. P. and Sandrnann, G. ( 1999) Biochim. Biophys. Ada 1439, 57-64 13 Botella, J. A.. Murillo, F. J. and Ruiz-Vazquez,R. ( 1995) Eur. J. Biochern. 233,238-248 14 Verdoes, J.C., Krubasik P., Sandmann, G. and van Ooyen, A. J. J. ( I 999) Mol. Gen. Genet. 262, 4 5 3 4 6 I 15 Schrnidhauser, T., Lauter, F., Schurnacher, M., Zhou, W., Russo, V. E. A. and Yanofsky, C. ( 1994) J. Biol. Chern. 269, 1206G12066 16 Velayos, A., Lopez-Matas, M. A,, Ruiz-Hidalgo, M. J. and Eslava, A. P. ( 1997) Fungal Genet. Biol. 22, 19-27 17 Ouchane, S.. Picaud, M.. Vemotte, C. and Astier, C. (1997) EMBO J. 16,47774787 Tanaka, A., Kaneko, T.. Sato, S., Sugiura, M. and 18 Kotani. H., Tabata, S, ( 1995) D N A Res. 2, I 33- I42 19 Brarnley, P. M. and Sandrnann. G. ( I 985) Phytochemistry 24, 29 19-2922
Two completely unrelated lycopene P-cyclases evolved independently in bacteria. T h e heterodimeric lycopene P-cyclase encoded by the genes crt Ye and crt Yd dominates among Gram-positive bacteria. From these genes the fungal lycopene P-cyclaselphytoene synthase fusion gene crt Y B derived. T h e completely unrelated crt Yof a monomeric P-cyclase, identified first in a Gram-negative enterobacterium, may be an ancestor of a cyanobacterial and prochlorophyte lycopene P-cyclase gene crtL. From this gene, a line of evolution can be drawn to the corresponding Icy-P genes in chlorophytes and higher plants. T h e genes encoding lycopene &-cyclase,ley-&,and Ccs, ccs, from plants are highly homologous with the P-cyclase genes and may originate from gene duplications.
References I Misawa, N., Nakagawa, M.. Kobayashi, K., Yarnano, S., Izawa, Y., Nakamura, K. and Harashirna. K. ( I 990) 1. Bacteriol. 172, 6704-67 I 2 2 Cunningharn, F. X., Sun, Z., Chamovitz, D., Hirschberg, J. and Gantt, E. ( I 994) Plant Cell 6, I 107- I I2 I
Received 22 June 2000
Effect of monogalactosyldiacylglycerol and other thylakoid lipids on violaxanthin de-epoxidation in liposomes D. Latowski, A. Kostecka and K. Strzaka' Department of Plant Physiology and Biochemistry, The Jan Zurzycki Institute of Molecular Biology, Jagiellonian University, Al. Mickiewicza 3, 3 I - I20 Krakow, Poland
Abstract In this study we present evidence that one of two reactions of the xanthophyll cycle, violaxanthin de-epoxidation, may occur in unilamellar egg
Key words: violaxanthin de-epoxidase, xanthophyll cycle. Abbreviations used: VDE, violaxanthin de-epoxidase; PC, phosphatidylcholine; MGDG, rnonogalactosyldiacylglycerol; DGDG, digalactosyldiacylglycerol; SQDG, sulphoquinovosyldiacyglycerol; PG, phosphatidylglycerol. 'To whom correspondence should be addressed (e-mail strzalka@ rnol.uj.edu.pl).
0 2000 Biochemical Society
810
phosphatidylcholine vesicles supplemented with monogalactosyldiacylglycerol (MGDG). Activity of violaxanthin de-epoxidase (VDE) in this system was found to be strongly dependent on the content of M G D G in the membrane; however, only to a level of 30 mol yo. Above this concentration the rate of violaxanthin de-epoxidation decreased. T h e effect of individual thylakoid lipids on VDEindependent violaxanthin transformation was also investigated and unspecific effects of phosphatidylglycerol and sulphoquinovosyldiacyglycerol, probably related to the acidic character of
Sterols and lsoprenoids
activity was determined by dual-wavelength measurements in a DW-2000 S L M Aminco spectrophotometer and separation of pigments was performed as described by Latowski et al. [9].
these lipids, were found. T h e presented results suggest that violaxanthin de-epoxidation most probably takes place inside MGDG-rich domains of the thylakoid membrane. T h e described activity of the violaxanthin de-epoxidation reaction in liposomes opens new possibilities in the investigation of the xanthophyll cycle and may contribute to a better understanding of this process.
Results Violaxanthin de-epoxidation carried out in liposomes made of a lipid mixture, the composition of which was close to that of the natural thylakoid membrane (50 mol yo M G D G / 2 5 mol yo D G D G / 1 3 mol o/o P G / 9 mol yo S Q D G / 3 mol?/, P C [7]) and which contained 5 p M violaxanthin, showed that apart from the expected deepoxidation products, i.e. antheraxanthin and zeaxanthin, additional unidentified peaks were detected by H P L C analysis. Formation of these additional peaks was not related to V D E activity as they were also formed in the absence of the enzyme. Studies on the effect of individual thylakoid lipids on violaxanthin without V D E addition revealed that their formation is related to the presence of PG and SQDG in the reaction mixture. T h e highest level of VDE-independent violaxanthin conversion was observed when the xanthophyll was incorporated into liposomes made only of P G or SQDG, and it amounted to about 30 and 1796 of the initial concentration of violaxanthin, respectively. Only negligible violaxanthin conversion (about 1 o/o of the initial concentration) was detected when the liposomes were made of D G D G . No conversion appeared when M G D G was the only lipid present in the reaction mixture. In such a system, however, liposomes are not formed because of M G D G ' s critical packing parameter value [lo]. Instead, large M G D G aggregates containing the majority of violaxanthin molecules were formed when an ethanolic solution of M G D G and violaxanthin was injected into buffer under vigorous bubbling with nitrogen. These yellow aggregates precipitated and remained adhered to the glass walls of the test tubes. Approx. 3 04 of the initial concentration of violaxanthin remained un-precipitated in the solvent. In the system of liposomes made of PC, neither VDEindependent violaxanthin conversion nor vesicle aggregation was observed. For these reasons, PC, as the lipid readily forming a bilayer [lo], was selected for further study of violaxanthin de-epoxidation in the liposome system. Violaxanthin present in P C liposomes was de-epoxided by VDE only if the system was supplemented with M G D G . With the increase in M G D G content in P C liposomes to the
Introduction One of the two known reactions of the xanthophyll cycle is de-epoxidation of violaxanthin to zeaxanthin via antheraxanthin as an intermediate [l]. Violaxanthin de-epoxidation is catalysed by the lumenal enzyme violaxanthin de-epoxidase (VDE). In the dark, when the pH in the thylakoid lumen is high, VDE is in the inactive state. Under strong light, the pH in the thylakoid lumen decreases, and VDE binds to the membrane, becomes active and converts violaxanthin into zeaxanthin [ 2 4 ] . Experiments carried out with the isolated VDE revealed that for optimal activity the enzyme requires the presence of the major thylakoid lipid, monogalactosyldiacylglycerol ( M G D G ) , as a cofactor [5-71. M G D G was found to be four times more efficient in precipitating VDE than the second most common thylakoid lipid, digalactosyldiacylglycerol ( D G D G ) , and up to 38 times more efficient than other lipids [S]. In the present study, we measured the de-epoxidation reaction in the system of unilamellar phosphatidylcholine ( P C ) / M G D G liposomes and in the presence of VDE as the only protein.
Materials and methods PC was purchased from Sigma. M G D G , phosphatidylglycerol (PG), sulphoquinovosyldiacyglycerol ( S Q D G ) and D G D G were obtained from Lipid Products (South Nutfield, Redhill, Surrey, U.K.). Liposomes were prepared by slow injection of the ethanolic solution of lipids with a Hamilton syringe into 0.1 M sodium citrate buffer, p H 5.2, under continuous bubbling with nitrogen. Subsequently, the liposome suspension was extruded through a polycarbonate membrane with a pore diameter of 100 nm. Violaxanthin was isolated from lucerne (Medicago sativa) by extraction with acetone and saponification followed by column chromatography on silica gel F254 (Merck) in petroleum ether/acetone (4: 1, v/v). VDE was isolated and purified from 7-dayold wheat leaves grown at 28 "C essentially as described by Hager and Holocher [4]. T h e enzyme
81 I
0 2000 Biochemical Society
Biochemical Society Transactions (2000) Volume 28, part 6
level of 30 mol o/o the percentage of transformed violaxanthin also increased. Above this concentration a decrease in the de-epoxidation rate was observed. At the M G D G concentration of 35 mol yo the liposome suspension became much more turbid and about 16 yo less zeaxanthin was formed compared with the system containing only 30 mol yo M G D G . Also, the ratio of zeaxanthin to antheraxantin was lower. T h e increase of turbidity was accompanied by aggregation and sedimentation of the lipid structures formed. These changes were probably the result of the appearance of hexagonal phases (inverted micelles) formed by M G D G , caused by its high levels in the liposomes [10,11]. T h e PC liposome suspension with an M G D G content of up to 30 mol”/, was transparent in appearance and showed no signs of aggregation.
may also occur in the lipid matrix of thylakoid membranes and not necessarily only within pigment-protein complexes, as suggested by Thayer and Bjorkman [13]. However, the occurrence of de-epoxidation in such complexes in addition to that occurring in the lipid phase cannot be totally excluded. Although membrane proteins and their complexes are not necessary for transformation of violaxanthin to antheraxanthin and zeaxanthin in model lipid membranes, the presence of M G D G in PC liposomes was found to be indispensible for this reaction to occur. When the effect of M G D G on violaxanthin de-epoxidation kinetics was investigated in PC liposomes a strong dependence on the percentage content of M G D G in these liposomes was found. It is plausible to assume that VDE binds only to certain membrane domains that are rich in M G D G and that the deepoxidation reaction takes place in these domains.
Discussion
This work was financially supported by grant no. 6P04A028 I9 from the Committee for Scientific Research (KBN) of Poland. W e express our thanks to Marta Skrzynecka-Jaskier for her help in preparing this paper.
T h e results presented show that the deepoxidation reaction of violaxanthin can be studied in artificial lipid bilayers, which are a better approximation of natural membranes than the commonly used, undefined system of M G D G aggregates. However, our attempt to use liposomes with lipid contents mimicking the natural lipid composition of thylakoid membranes [7] was not successful because of the unspecific partial conversion of violaxanthin to unidentified products, which complicated the quantitative measurements. T h e VDE-independent conversion of violaxanthin occurring in the presence of P G and S Q D G could be explained by the acidic character of these lipids. It is well known that an acidic environment promotes conversion of violaxanthin into auroxanthin and that some isomerization products can be also formed [12]. No such effect was observed in PC liposomes, probably due to the buffering properties of this lipid. Therefore, PC liposomes were selected for further studies. Another reason for the use of PC was that the M G D G / P C mixture forms liposomes more easily than the M G D G / D G D G mixture [ l l ] . VDE was the only protein in the system, and efficiently converted violaxanthin present in the M G D G / P C liposomes into antheraxanthin and zeaxanthin. This is evidence that the de-epoxidation reaction
0 2000 Biochemical Society
References I Yamamoto, H. Y.. Nakayama, T. 0. H. and Chichester, C. 0. ( I 962) Arch. Biochem. Biophys. 97, I 68- I73 2 Bratt, C. E., Arvidson, P.-O., Carlsson, M. and Akerlund. H.-E. ( I 995) Photosynth. Res. 45, 169- I75 3 Hager, A. ( 1969) Planta 89, 224-243 4 Hager. A. and Holocher, K. ( I 994) Planta 192, 58 1-589 5 Yamamoto, H.Y., Chenchin, E. E. and Yarnada, D. K. ( I 974) Proceedings of the Third International Congress of Photosynthesis, pp. 1999-2006, Elsevier, Amsterdam 6 Siefierman-Harms. D., Ninnernann, H. and Yamamoto H. Y. ( I 987) Biochim. Biophys. Acta 892, 303-3 I 3 7 Webb, M. and Green, B. ( I 99 I) Biochim. Biophys. Acta 1060, 133-158 8 Havir, E. A,. Tausta, L. 5. and Peterson, R. B. (1997) Plant Sci. 123, 57-66 9 Latowski, D., Burda, K. and Strzaka. K. (2000) J, Theor. Biol. 206, 507-5 I 4 10 Israelachvili,J. N. and Mitchell, D. J. ( 1975) Biochim. Biophys. Acta 389, 13- I9 I 1 Sprague, G. 5. and Staehelin, L. A. ( 1984). Biochirn. Biophys. Acta 777,306-322 I 2 lackman, L. M. and Liaaen-Jensen,5. ( 1964) Acta Chem. Scand. 18. 1403 13 Thayer, 5.5. and Bjorkman, 0.( 1992) Photosynth. Res. 33, 2 I 3-225 Received 20 June2000
812
