THE JOURNAL OF BIOLOGICAL CHEMISTRY Q 1993 by The American Society for Biochemistry and Molecular Biology, Inc. Vol. 268, No. 28, Issue of October 5, pp. 20892-20896,1993

Printed in U.S.A.

Cbr, an Algal Homolog of Plant Early Light-induced Proteins, Is a Putative Zeaxanthin Binding Protein*

(Received for publication, April 15,1993, and in revised form, June 16, 1993)

Haim Levy, Tamar Tal, Aviv Shaish, and Ada ZamirS From the Biochemistry Department, Weizmann Institute of Science, Rehovot 76100, Israel

The cbr gene, previously cloned from the unicellular green alga Dunaliella bardawil, is transcriptionally and translationally activated in parallel to accelerated carotenogenesis in response to light stress conditions. The product of cbr, structurally similar to Elips (early light-induced proteins of higher plants), is associated with a minor light harvesting complexes of photosys- tem I1 component (Levy, H., Gokhman, I., and Zamir, A. (1992) J. Biol. Chem. 267, 18831-18836). This study examines the relationship between the induction of Cbr and another plant response to light stress, the deepoxidation of violaxanthin to zeaxanthin. A paral- lel between the two processes was observed in cells exposed to high light, starved for sulfate, or treated with norflurazon, a herbicide inducing photooxidative damage by inhibiting de novo carotenoid biosynthesis. When highly illuminated cells were returned to normal light, Cbr decayed in parallel to the reepoxidation of zeaxanthin to violaxanthin. Evidence for the physical association of Cbr and zeaxanthin was provided by nondenaturing gel electrophoresis. In cells transferred from low to high light, zeaxanthin was associated with the faster migrating of two electrophoretically re- solved fractions of light harvesting complexes of pho- tosystem I1 that also contained Cbr. In cells growing under normal light, violaxanthin was bound equally to the two fractions. Based on these results we propose that Cbr/early light-induced proteins bind zeaxanthin to form photoprotective complexes within the light- harvesting antennae.

In recent years, increasing attention has been drawn to a widely conserved light stress response in plants and several algae, formation of the xanthophyll zeaxanthin via the light- regulated deepoxidation of violaxanthin with the monoepox- ide antheraxanthin as an intermediate (for reviews, see Dem- mig-Adams (1990) and Demmig-Adams and Adams (1992)). This conversion, together with its reversal upon resumption of favorable light conditions, is referred to as the xanthophyll cycle (Yamamoto et al., 1962). Correlations were observed between zeaxanthin accumulation and the nonphotochemical quenching of chlorophyll fluorescence under photon fluxes exceeding the photosynthetic capacity (e.g. Demmig et al., 1987; Bilger et al., 1989; Demmig-Adams et al., 1990). Such

*This study was supported in part by the Minerva Foundation (Munich, Germany) and the Leo and Julia Forchheimer Center for Molecular Genetics, The Weizmann Institute of Science. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked ‘‘uduer- tisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. + To whom correspondence should be addressed. Tel.: 972-8- 342788; Fax: 972-8-344118.

excessive fluxes are generated by high intensity illumination or, under lower light intensities, by environmental stress, e.g. nutrient starvation and water stress, that reduce the efficiency of photosynthetic energy conversions (Demmig-Adams and Adams, 1992). These observations have led to the proposal that zeaxanthin was effective in a protective mechanism enabling the dissipation of excessive light energy and thereby preventing photooxidative damage to the photosynthetic ma- chinery (Demmig-Adams and Adams, 1992). A correlation between the level of zeaxanthin, and xanthophylls in general, and the level of @-carotene was noted in a variety of plants (Demmig-Adams and Adams, 1992).

Another plant response to light stress conditions involves early light-induced proteins (Elips)’, originally characterized as products of genes transiently activated in etiolated seed- lings of pea and barley soon after their exposure to light (Kolnaus et al., 1987; Green et al., 1991). Transcriptional activation of elip was also noted during leaf development subject to circadian cycle regulation (Green et al., 1991). Recently, an elip-like gene was found to be activated during desiccation of the resurrection plant Craterostigma plantagi- neum (Bartels et al., 1992).

The predicted structure of Elips shows distinct similarity to apoproteins of light-harvesting complexes (LHC) of pho- tosystems I and 11. Thylakoid localization has been indicated for Elips from pea, barley (Grimm et al., 1989), and C. plan- tagineum (Bartels et al., 1992).

Previously, we cloned an elip-like gene from Dunaliella bardawil, a unicellular green alga that accumulates massive amounts of @-carotene under conditions potentially engen- dering photooxidative damage, e.g. high light intensity and nutrient deprivation (Lers et al., 1990). The D. bardawil gene was cloned on the basis of its co-activation with the induction of accelerated carotenogenesis and consequently designated cbr (for carotene biosynthesis :elated) (Lers et al., 1991).

The discovery of the coordinated activation of cbr and accelerated carotenogenesis provided a vital clue for the long sought function of Elips. Thus, we proposed that Cbr (as well as Elips) bound a carotenoid related to the coordinately synthesized @-carotene, forming a complex whose function is to counteract photooxidative damage during chloroplast de- velopment and under light stress conditions (Lers et al., 1991).

Analyses of the Cbr protein using antibodies raised against a synthetic oligopeptide matching a predicted sequence in Cbr demonstrated that Cbr co-fractionated with a minor LHCII component (Levy et al., 1992). This observation was consist- ent with a role of the presumed Cbr-pigment complex within the light-harvesting antennae. A similar Cbr-containing LHCII complex was also observed in Dunaliella salina, a strain incapable of massive accumulation of @-carotene yet activated

The abbreviations used are: Elip, early light-induced protein; LHC, light-harvesting complexes; E, einstein.

20892

Relationship between Cbr and Zeaxanthin 20893

for both carotenogenesis and Cbr synthesis under similar conditions as D. bardawil (Levy et al., 1992).

The relevance of the findings in the algae to higher plants was supported recently by the demonstration that high light intensities also induced elip expression in mature pea leaves (Adamska et al., 1992).

That Cbrplip and zeaxanthin might be interrelated in their biological function has been suggested by the similarity in the conditions inducing their accumulation. Furthermore, both protein and pigment were independently assigned pho- toprotective functions. Considered together with the carote- noid-binding role initially proposed for Cbr/Elip, these simi- larities suggested that Cbrplip possibly act as zeaxanthin binding proteins. To examine the possible relationship be- tween Cbr and zeaxanthin, the present study compared the kinetics of induction and decay of Cbr and zeaxanthin under several sets of conditions and localized Cbr and xanthophyll pigments in electrophoretically resolved LHCII complexes. The results support the conclusion that the function of Cbr/ Elip might involve the binding of zeaxanthin to form photo- protective complexes within the light-harvesting antennae.

MATERIALS AND METHODS

Algal Strains and Growth Conditions-The origin and clone puri- fication of Dunaliella salina and DunalieUa bardawil, the composition of complete and sulfate-depleted media, growth conditions, and sul- fate starvation were as described (Lers et al., 1990).

Light Conditions-Cells were normally grown under white light of 27 pE/mz/s (low light), 110 pE/m2/s (normal light), or 1650 pE/m2/s (high light) essentially as previously described (Lers et al., 1990).

Treatment with Inhibitors-Norflurazon was added to 0.5 p ~ , and sodium azide was added to 1 mM to algal cultures grown in complete media under normal light conditions.

Thylakoid Membranes-Preparation and solubilization of mem- branes were as described (Levy et al., 1992).

Nondenaturing and Denaturing Gel Electrophoresis-The prepa- ration of samples, composition of gels and buffers, and running conditions were as previously described (Levy et al., 1992).

Immunoblot Analysis-Analyses employing anti-Cbr antibodies were essentially as previously described (Levy et al., 1992). Instead of 'z61-protein A, immunoreactive bands were detected with alkaline phosphatase-conjugated anti-rabbit IgG from goat assayed according to McGadey (1970).

Zsohtion of Electrophoretically Resolved LHCZZ Fractions-Thyla- koid membranes from D. salina were solubilized and fractionated in the presence of mild detergents as previously described (Levy et al., 1992). Extracts made from 5 X 10' cells were loaded on a single-well gel. Gels were run in duplicate, and corresponding bands were excised, pooled, and homogenized with 1 ml of running buffer and centrifuged in a minicentrifuge to remove particulate matter.

Pigment Extraction and Analysis-Pigments were extracted from whole cells or electrophoretically resolved fractions with ethanokhexane, 2:l (v/v) as described (Shaish et al., 1991). The solvent was evaporated under Nz, and the pigments, redissolved in hexane, were analyzed by high performance liquid chromatography using a reverse phase C18 column (Vydac 201TP54) with a 80-100% gradient of organic solvent (acetonitrile:methanoktetrahydrofuran, 7515:lO (v/v) in Hz0 (Braumann and Grimme, 1981; Herrin et al., 1992). Pigments were identified and quantified spectrally (Braumann and Grimme, 1981).

RESULTS

Analysis of Xanthophyll Cycle Pigments in D. salina-D. salina was used in most of the present analyses, because massive accumulation of @-carotene and other carotenoids in D. bardawil sometimes interfered with pigment resolution and analysis. In the analysis shown (Fig. l ) , pigments were ex- tracted from (i) cells grown under normal light (to); (ii) following a 48-h exposure to high light; (iii) after a subsequent 4-h period in normal light. To facilitate comparison, the full absorbance scale in the elution profiles was set to correspond

L Chla

t o NL

48h HL

+4h NL

0 I O 20 30 40 50

Elution Time b i n )

FIG. 1. The induction of xanthophyll conversions in D. sal- ina in response to changes in light intensity. Pigments extracted from cells grown under normal or high light for the periods indicated were analyzed as described under "Materials and Methods." N, nea- xanthin; V, violaxanthin; A, antheraxanthin; L, lutein; 2, zeaxanthin; Chl, chlorophyll; Car, carotene.

to the height of the peak representing lutein, a xanthophyll whose total amount did not change drastically under the different light regimes. Clearly, exposure to high light brings about a decline in the level of violaxanthin and a rise in the level of zeaxanthin. High light also elicits a marked increase in the level of @-carotene (the shoulders surrounding the major, all-trans-@-carotene peak probably represent a-caro- tene and 9-cis-@-carotene). Also noted are a decrease in both chlorophyll a and b, consistent with a decrease in antenna size. Subsequent transfer to normal light conditions induces the reverse xanthophyll transition, i.e. the level of zeaxanthin decreases while that of violaxanthin increases. The latter transition occurs before the levels of @-carotene and chloro- phylls are significantly altered. These analyses clearly dem- onstrate the operation of the xanthophyll cycle in Dunaliella and reconfirm the induction of @-carotene accumulation under high light.

The Xanthophyll Cycle in Relation to the Induction and Decay of Cbr-Changes in the levels of Cbr and xanthophyll cycle pigments were examined in parallel in cells exposed first to high light and subsequently to normal light (Fig. 2). Similar to the results shown above, high light induced a relative rise in zeaxanthin and a concomitant decrease in violaxanthin. In agreement with previous observations with D. bardawil (Levy et al., 1992) no Cbr was detected in the to cells, but the protein was induced by exposure to high light. Upon resumption of normal light conditions, the gradual conversion of zeaxanthin back into violaxanthin occurred concurrently with a decline in the level of Cbr. Thus, once the algae were relieved from light stress, Cbr decayed in correspondence with the conver- sion of zeaxanthin into violaxanthin.

The correlation between xanthophyll conversions and Cbr accumulation was established also under another set of in- duction conditions, sulfate starvation under normal light in- tensity. The results (Fig. 3) indicate a correspondence between the course of Cbr induction, an increase in the level of

20894 Relationship between Cbr and Zeaxanthin

FIG. 2. Induction and decay of Cbr and zeaxanthin in response to high and low light. Cells grown under normal or high light for the periods in- dicated were analyzed for pigments and Cbr antigens as described under “Mate- rials and Methods.” The analysis of cells grown under high light for 72 h is shown on the right. The levels of violaxanthin (open bars) and zeaxanthin (filled bars) are expressed relative to the total amount of violaxanthin, antheraxan- thin, and zeaxanthin (cycle xantho- phylls).

e 0.8 I

9 0.8 - c 5 0.6 K

= 0.4 u” : E 0 0.2

L 0.0

Time (h)

I I 1-

High Light Normal Light Hlgh Light

Time (h) 0 72 96 120 rm

Normal light Low light

FIG. 3. Induction of Cbr and zeaxanthin in sulfate-starved cells. Cells were starved for sulfate and grown under normal or low light for the indicated periods. Pigments and Cbr antigens were determined as described under “Materials and Methods.” Analysis of cells starved for sulfate under low light for 120 h is shown on the right. The levels of violaxanthin (open bars) and zeaxanthin (filled bars) are expressed relative to the total amount of violaxanthin, antheraxanthin, and zeaxanthin (cycle xanthophylls).

zeaxanthin, and a reciprocal decrease in the level of violax- anthin. That the effect of sulfate starvation is related to light stress is directly indicated by the examination of sulfate- starved cells under low light conditions (Fig. 3). In this instance, both xanthophyll transitions and Cbr induction do not take place.

Norflurazon, an inhibitor of carotenoid biosynthesis, has been extensively used to experimentally enhance photooxi- dative damage in plants (Taylor, 1989; Susek and Chory, 1992). Treatment of a D. salina culture with norflurazon brings about a drastic decrease in the intracellular level of carotenoids. Examination of the relative levels of xanthophyll cycle pigments in norflurazon-treated cells (Fig. 4A) indicated that after an initial rise, the level of violaxanthin gradually decreased with a reciprocal rise in zeaxanthin. Cbr accumu- lation progressed in correspondence with the violaxanthin to zeaxanthin transition.

Our earlier studies indicated that induction of Cbr synthesis was correlated with accelerated carotenogenesis. One excep- tion to this correlation was noted previously in D. bardawil cultures treated with sodium azide under normal light, a considerable increase in @-carotene (Shaish et al., 1993) is not accompanied by the appearance of Cbr (data not shown). At the concentration used, sodium azide had no meaningful effect on cell proliferation. Examination of the effect of sodium azide on xanthophyll transitions (Fig. 4B) indicated that, despite a 10-fold increase in the cellular content of @-carotene (data not shown), treatment with azide did not induce the

a 0.8 - c h

a n

Blotting Control

FIG. 4. Effect of reagents affecting carotenoid biosynthesis on the induction of Cbr and zeaxanthin. Cells grown under normal light were treated with norflurazon ( A ) or sodium azide ( B ) for the indicated periods. The concentration of reagents and methods of analysis were as described under “Materials and Methods.” The levels of violaxanthin (open bars) and zeaxanthin (filled bars) are expressed relative to the total amount of violaxanthin, antheraxan- thin, and zeaxanthin (cycle xanthophylls).

violaxanthin to zeaxanthin conversion nor the induction of Cbr. Thus, in this case @-carotene accumulation is not accom- panied by a rise in both zeaxanthin and Cbr.

Analysis of Cbr and Xanthophylls in Electrophoretically Resolved LHCII Fractions-The correspondence in the course of their induction and decay suggested the possibility that Cbr and zeaxanthin were physically associated. This possibil- ity was examined by analysis of gently solubilized thylakoid membranes fractionated by nondenaturing gel electrophore- sis. As shown previously (Levy et al., 1992), this procedure resolves LHCII into a fraction containing Cbr and a more slowly migrating fraction containing little or no Cbr. In the present analysis, solubilized thylakoids from D. salina cells

Relationship between Cbr and Zeaxanthin 20895

continuously grown in low light or exposed to high light for 48 h were electrophoresed. The gel electrophoretic patterns are shown side by side with immunoblot determination of Cbr antigens in each of the two resolved LHCII fractions as well as in the starting, nonfractionated thylakoids (Fig. 5). The results agree with the previous analyses (Levy et al., 1992) in showing that Cbr synthesis is induced by high light and that the protein is associated with the faster migrating fraction of LHCII.

Following their excision from the gel, individual LHCII fractions were analyzed for their pigments (Table I). In the low light-grown cells, LHCII fractions I and I1 contain similar proportions of violaxanthin but not detectable zeaxanthin. Antheraxanthin, presumed to be an intermediate in the vio- laxanthin to zeaxanthin conversion, is detectable in fraction I but not in fraction 11.

In the cells exposed to high light, violaxanthin in both LHCII fractions is reduced to a marginal level. Zeaxanthin becomes a significant component of fraction I but is practi- cally absent in fraction 11. The relative levels of antheraxan- thin in the two fractions remain practically unchanged com- pared with the corresponding fractions from the low light- grown cells. In addition to the xanthophyll cycle pigments, the two LHCII fractions also differ somewhat in their content of other carotenoids as well as chlorophylls (data not shown).

Thus, in high light-exposed cells, zeaxanthin is nearly ex- clusively localized in fraction I of LHCII, which also includes the bulk of the Cbr protein. In the low light-grown cells, violaxanthin is distributed evenly between the two LHCII fractions. However, antheraxanthin, the partially deepoxi- dized intermediate, is considerably more abundant in fraction I than in fraction 11.

Low Light

Cbr in Total Membranes U

High Light

Cbr in Complexes

“,e. -.

FIG. 5. Electrophoretic resolution of LHCII fractions and distribution of Cbr. Thylakoid preparation and solubilization, non- denaturing gel electrophoresis, and isolation of LHCII fractions (com- plexes I and 11) were as described under “Materials and Methods.” Total membranes and isolated fractions were analyzed for Cbr by denaturing gel electrophoresis and immunoblot analysis as described under “Materials and Methods.”

TABLE I Xanthophyll cycle pigments in electrophoretically resolved LHCII

complexes from high and low light-grown cells Bands corresponding to LHCII fractions I and I1 from high or low

light-grown cells were eluted as described under “Materials and Methods” and Fig. 5. Data from three independent analyses are Dresented.

Relative to chlorophyll a

Pigment Low light High light

Fraction I Fraction I1 Fraction I Fraction I1 Violaxanthin 0.28 f 0.09 0.27 ? 0.06 C0.007 c0.003 Zeaxanthin <0.003 ~0.003 0.13 ? 0.04 c0.015 Antheraxanthin 0.02 +. 0.007 <0.007 0.016 f 0.01 <0.003

DISCUSSION

A correlation between the induction of Cbr and the rise in zeaxanthin, at the expense of violaxanthin, was observed under three different sets of conditions, high light intensity, sulfate starvation, and norflurazon treatment. While the first two conditions also induce the accumulation of /?-carotene, this process is inhibited in the norflurazon-treated cells. How- ever, the potential enhancement of the carotenoid biosyn- thetic capacity in such cells was previously indicated by their accumulation of phytoene under carotenogenesis-inducing conditions (Ben Amotz et al., 1987). Hence, the cellular re- sponses to norflurazon are closely similar to those actuated by the other treatments potentially generating photooxidative damage.

The absence of both xanthophyll conversions and Cbr accumulation in cells stimulated for accelerated caroteno- genesis by sodium azide further emphasizes the coupling between the first two responses. Stimulation of carotenoge- nesis by sodium azide then differs in some respects from induction by light stress.

Examination of the detailed course of change in Cbr and zeaxanthin accumulation in sulfate-starved or norflurazon- treated cells, as well as in cells exposed to high light (data not shown), suggests that considerable accumulation of zeaxan- thin occurs before Cbr becomes evident. This apparent dis- cordance might be partly explained by the different nature of the processes studied or differences in the threshold of detec- tion of the protein and pigment components. Additional pos- sible interpretations assume the existence of several separate pools of xanthophylls (c f . Demmig-Adams (1990)) that might vary in the rate of pigment influx or the operation of related but not necessarily simultaneous induction pathways.

A much closer correspondence between the course of change in Cbr and zeaxanthin is noted after transfer of the cells from high to normal light. Cbr decay and zeaxanthin reepoxidation occur simultaneously for several hours after the resumption of normal light conditions. The decrease in Cbr cannot be accounted for by dilution due to cell division and hence reflects the proteolytic degradation of the protein. Pea Elip has also been found to be unstable after removal of light stress conditions (Adamska et al., 1992).

Based on the possibility, discussed below, that Cbr and Elip function as zeaxanthin binding proteins, their instability might arise from the reepoxidation of the bound zeaxanthin into violaxanthin. If the binding affinity of Cbr for violaxan- thin is considerably lower than for zeaxanthin, the reepoxi- dation may result in the dissociation of the xanthophyll from Cbr rendering the protein more susceptible to proteolysis. Stabilization by bound pigments has been indicated for nor- mal Cab proteins (e.g. LaRoche et al. (1991) and Mukai et dl.

(1992)). The association of Cbr with antenna complexes was previ-

ously concluded from the co-fractionation of the protein with a minor LHCII component (Levy et al., 1992). The present pigment analyses provides an important new insight into the nature of this complex; it is probably also the preferential site of zeaxanthin binding.

In our earlier analyses, gently solubilized thylakoids were fractionated by sucrose gradient centrifugation before further resolution by gel electrophoresis. Under these conditions, Cbr co-migrated with a high mobility minor LHCII fraction (Levy et al., 1992). In the present study, the prefractionation step was omitted to minimize pigment loss but, as observed pre- viously, Cbr was localized nearly exclusively in the higher mobility LHCII fraction. Zeaxanthin was distributed similarly to Cbr; it was concentrated in fraction I and was practically

20896 Relationship between Cbr and Zeaxanthin

absent from fraction 11. In contrast, violaxanthin was distrib- uted evenly in the two LHCII fractions from the low light- grown cells.

On the basis of these observations we propose that Cbr functions as a zeaxanthin binding protein, although the bind- ing of additional pigments to Cbr cannot be ruled out. The conversion of violaxanthin to zeaxanthin will then entail an exchange of apoproteins from presumably Cab to Cbr. Among several possibilities, deepoxidation and reepoxidation reac- tions could take place on protein-bound xanthophylls followed by their release and rebinding to other apoproteins.

Complexes of zeaxanthin and Cbr probably form a part of LHCII during light stress. Our results do not exclude the presence of similar complexes in photosystem I, where the level of the protein might be below the level of detection by immunoblotting. Alternatively, in photosystem I a different member of the Cbr family might replace the one recognized by the antibodies (Lers et al., 1991; Levy et al., 1992). Elip was recently detected in photosystem I of pea (Cronshagen and Henfeld, 1990).

The binding of zeaxanthin to Cbr might serve to localize the pigment within the light-harvesting antennae. Therefore, by quenching of chlorophyll-excited states and preventing the formation of singlet oxygen, it would protect sensitive com- ponents of the photosytems. Binding of zeaxanthin to Cbr might also affect the spectral properties of the pigment to maximize the quenching efficiency.

The Cbr-zeaxanthin complex would tend to lower the effi- ciency of the photosynthetic process (Demmig-Adams and Adams, 1992). The strict regulation of the induction and decay of both protein and pigment components in response to en- vironmental conditions can be understood in this light.

The Cbr-zeaxanthin complex, as well as the homologous

Elip complex in higher plants, can then be considered as modified forms of light-harvesting complexes evolved to serve as photoprotective agents.

Acknowledgments-We are indebted to Arie Segal for algal culti- vation and Dr. Avigdor Scherz and Paula Braun for helpful discus- sions.

REFERENCES Adamska, I., Kloppstech, K., and Ohad, I. (1992) J. Biol. Chem. 267, 24732-

Bartels, D., Hanks, C., Schneider, K., Michel, D., and Salamini, F. (1992) Ben Amotz, A., Gressel, J., and A w n , M. (1987) J. Ph col 18,529-537 Bilger W., B'orkman, O., and Thayer, S. (1989),Planf bhysiol 91,542-551 B r a d a n n , 'd., and Gnmme, L. H. (1981) Bwchrm Bwphys. Acta 637,8-17 Cronshagen, U., and Henfeld, F. (1990) Eur. J. Biochem. 193,361-366 Demmig-Adam!, B., and Adams, W. W., 111 (1992) Annu. Reu. P h t Physiol.

Demmig, B., Winter, K., Kruger, A., and Czygan, F.-C. (1987) Plant PhysioL

Demmig-Adams B. (1990) Biochim. Bioph s. Acta 1020,l-24 Demmig-Adams: B., Adams, W. W., 111, d e e r , U., Neimanis, S., Winter, K.,

Kruger, A., Czygan, F.-C., Bilger, W., and B~orkman, 0. (1990) Plant Physwl.

Green, B. R., Pichersky, E., and Kloppstech, K. (1991) Trendg Biochem. Sci.

Grimm, B., Kruse, E., and Kloppstech, K. (1989) Plant Mol. Biol. 13,585-593 Hemn, D., Battey, J. F., Greer, K., and Schmidt, G. W. (1992) J. Biol. Chem.

Kolnaus, W., Schamhorst, C., Kuhne, U., and Henfeld, F. (1987) MOL & Gen.

L a h h e , J., Mortain-Bertrand, A., and Falkowsky, P. G. (1991) Plant Physiol.

24737

EMBO J. 11,2771-2778

Plant Mol. BwL 43,500-526

84, 218-224

92,293-301

16,181-186

267,8260-8269

Genet. 209,234-239

Lers A., Biener, Y., and Zamir, A. (1990) Plant PhysioL 93,389-395 Lers' A Le H and Zamir A. (1991) J. Bwl. Chem. 266,13698-13705 Le ', H, GXhk, I., and Z b i i , A. (1992) J. BWL Chem. 267,18831-18836 M s a d e J. (1970) Histochemie 23, 180-184 Me er f?:, and Klo stech, K. (1984) Eur. J. Biochem. 138,201-207 Muiai; Y., Tazaki,%, Fujii, T., and Yamamoto, Y. (1992) Plant CeU Physwl.

Shaish, A,, Ben Amotz, A., and Avron, M. (1991) J. Phycol. 27,652-656 Shaish, A., A w n , M., Pick, U., and Ben Amotz, A. (1993) Planta, in press

Taylor, W. C., (1989) Annu. Reu. Plant Physiol. Plant MOL BWL 40,211-233 Susek, R. E., and Chory, J. (1992) A u t . J. Plant Physwl. 19,387-399

Yamamoto, H. Y., Nakayama, T. 0. M., and Chichester, C. 0. (1962) Arch.

97,147-153

33,859-866

Biochem. Biophys. 97, 168-173