Origin of Thylakoid Membranes in Chiamydomonas reinhardtii ... - NCBI

1 downloads 0 Views 3MB Size Report
J. Kenneth Hoober*, Charles 0. Boyd and Laurie G. Paavola ...... Butler PJG, Kfihlbrandt W (1988)Determination of the aggre- gate size in detergent solution of ...
Received for publication November 15, 1990 Accepted March 8, 1991

Plant Physiol. (1991) 96, 1321-1328 0032-0889/91/96/1321 /08/$01 .00/0

Origin of Thylakoid Membranes in Chiamydomonas reinhardtii y-1 at 380C' J. Kenneth Hoober*, Charles 0. Boyd and Laurie G. Paavola Departments of Biochemistry (J.K.H.) and Anatomy and Cell Biology (C.O.B., L.G.P.), Temple University School of Medicine, Philadelphia, Pennsylvania 19140 We took advantage of this system to consider two questions regarding thylakoid membrane biogenesis. First, are the membranes assembled during the first few minutes of light at 380C photochemically active? Because the Chl-protein complexes we have examined serve to harvest light primarily for PSII, we chose to examine this question by employing a general assay for PSII activity, the photoreduction of DCI (2, 30). We also examined the fluorescence properties of LHCs to assess their integration into PSII units. Second, do thylakoid membranes originate from a specific location within the chloroplast? To answer the latter question, the distribution of initial thylakoid membranes, i.e. those made in cells greening at 380C for only several minutes, was examined by EM. It is reasonable to assume that the location of initially formed membranes should reflect their site of assembly. This algal system is particularly well suited to examine this question, as opposed to higher plants in which dispersion of the prolamellar body in etioplasts generates an initial population of membranes from preexisting components (10). This paper describes evidence for rapid assembly of functional thylakoid membranes in association with the chloroplast envelope.

ABSTRACT The origin of thylakoid membranes was studied in Chlamydomonas reinhardtii y-1 cells during greening at 380C. Previous studies showed that, when dark-grown cells are exposed to light under these conditions, the initial rates of accumulation of chlorophyll and the chlorophyll a/b-binding proteins in membranes are maximal (MA Maloney JK Hoober, DB Marks [1989] Plant Physiol 91: 1100-1106; JK Hoober MA Maloney, LR Asbury, DB Marks [1990] Plant Physiol 92: 419-426). As shown in this paper, photosystem 11 activity, which was nearly absent in dark-grown cells, also increased at a linear rate in parallel with chlorophyll. As compared with those made at 250C, photosystem 11 units assembled during greening at 380C were photochemically more efficient, as judged by saturation at a lower fluence of light and a negligible loss of excitation energy as fluorescence. Electron microscopy of cells in light for 5 or 15 minutes at 380C showed that these initial, functional thylakoid membranes developed in association with the chloroplast envelope.

Chl is not synthesized by cells of Chlamydomonas reinhardtii y-J in the dark (25). Consequently, yellow cells result from dilution of preformed Chl as cells grow and divide in the dark. These yellow cells accumulate Chl a and b, i.e. they "green," when exposed to light. Essentially all Chl b and an equivalent amount of Chl a (about half of the total Chl a) are incorporated into LHCs2 in association with LHCPs (13, 20). Each LHCP molecule binds 11 to 15 Chl molecules and several xanthophylls (1, 23, 27). When cells are exposed to light at 25°C, the normal growth temperature for this alga, rapid greening ensues following a lag period of 1 to 2 h (25). The lag reflects, at least in part, the kinetics of induction by light of expression of genes encoding LHCPs (19). However, expression of these genes is also induced in the dark by higher temperatures (12, 16). After a suspension of yellow cells is incubated at 38°C for 1 to 2 h, Chl and LHCPs accumulate without a lag and at linear rates when the cells are exposed to light (13, 20). ' This

research

was

MATERIALS AND METHODS Cells

Chlamydomonas reinhardtii y-J used for these experiments was a strain maintained in this laboratory. Strain cw-15 was kindly provided from the Chlamydomonas Genetics Center, Duke University, by Dr. Elizabeth H. Harris. Greening experiments were performed as described previously (20). Assay of PSII Activity Photoreduction of DCI was measured with a reaction mixture similar to that described by Cahen et al. (2). y-J cells were broken with a French pressure cell at 6000 p.s.i. and centrifuged at 2000g for 1 min to remove unbroken cells. Photoreduction of DCI was done with a reaction mixture (8 mL) that contained 10 mm Na2HPO4 (pH 7.2), 8.5 mM KC1, 0.1 mm DCI, and an aliquot of cell extract. The assay mixture was pumped from the reaction vessel through a flow cell in a spectrophotometer to continuously measure the change in absorbance at 590 nm. The reaction was monitored for 1 min before illumination to obtain a dark rate and then for 3 min while exposed to a light beam from a tungsten-halogen microscope lamp that was passed through a glass IR filter (Coming glass No. 4602). The incident fluence was 70 x 104 ergs/cm2.

supported by National Science Foundation

grant 8613585 to J.K.H. and National Institute of Child Health and Human Development grant HD-1730 1 to L.G.P. 2 Abbreviations: LHC, light-harvesting complex of photosystem II; DCI, 2,6-dichlorophenolindophenol; LHCP, light-harvesting Chl a/ b-binding apoprotein; F, fluorescence intensity; Fm, maximal fluorescence intensity of intact membranes; Fa, fluorescence intensity of Chl a after dissociation of LHCs with detergent.

1321

1 322

HOOBER ET AL.

s, measured with a Yellow Springs radiometer (model 65A). The rate of absorbance change was linear with time and with the amount of cell homogenate added to the reaction mixture. Fluence of incident light was varied by inserting a calibrated series of neutral density filters in the light beam.

Plant Physiol. Vol. 96, 1991

U1)

U)

6

I-ii = 0

w r-

.

0

)

O

Cell Fractionation Cells were suspended in 50 mM Tricene-KOH, pH 8.0, containing 1 mM MgCl2 at 2 x 108 cells/mL and broken by passage through a French pressure cell at 6000 p.s.i for y-J cells and 1000 p.s.i for cw-15 cells. The samples were centrifuged at 2000g for 1 min to remove unbroken cells. A series of pellets were obtained by successive centrifugation at 5000g for 5 min (25,000g/min), I0,000g for 10 min (I00,000g/min), and at 30,000g for 35 min (1,000,000g/min). The latter centrifugation was sufficient to completely sediment Chlcontaining membranes. Spectrofluonmetry and Other Analyses Excitation and emission spectra of membrane samples (0.51 mL of 50 mM Tricene-KOH (pH 8.0) containing 1 mM MgCl2, were obtained with a PerkinElmer model 650-105 fluorescence spectrophotometer. Spectra were determined after addition of 0.1% (w/v) octyl glucoside to reduce turbidity and then again after addition of Triton X-100 to 0.5% (w/v), which dissociated LHCs. HPLC and analysis of Chl were performed as described (20). 1 jg Chl), suspended in

Electron Microscopy

For EM, 2 mL was removed from a cell suspension at 38°C and immediately injected into 8 mL of ice-cold aqueous 2% OSO4 (25). After standing on ice for 3 h, the sample was centrifuged and the cell pellet was washed three times with cold growth medium adjusted to pH 7.0. Fixed cells were washed with 0.1 M sodium cacodylate (pH 7.4) containing 5% sucrose, mordanted in 1 % tannic acid to improve membrane visibility, stained en bloc with saturated aqueous uranyl acetate, dehydrated in graded ethanols, and embedded in low viscosity resin as detailed earlier (26). Thin sections were stained with uranyl acetate (3-5 min) and lead citrate (1-2 min) and viewed in a JOEL 100B electron microscope. RESULTS PS11 Activity Develops in Parallel with Newly Synthesized Chi After 3 d of growth in the dark, degreened y-J cells contain about 1 gg of residual Chl/107 cells (13, 20). When incubated in the dark for 1.5 h at 38°C and then exposed to light, these cells accumulate Chl at a linear rate of about 4 jg/107 cellsh (Fig. 1). To determine whether the thylakoid membranes assembled under these conditions were photochemically functional, PSII was assayed by the photoreduction of DCI. As shown in Figure 1, PSII activity in degreened cells was negligible. However, the activity increased at a linear rate, with no apparent lag, in greening cells during the hour in the light. The rate ofDCI reduction was measured versus light fluence

Ul)

0.

nL 0

(.)

U m

CL 44o. ()

2 0 o

:1,

0

M in u te s Figure 1. Development of PSII activity during greening. Degreened cells were incubated 1.5 h at 380C in the dark and then exposed to light. At the time points indicated, photoreduction of DCI by brokencell samples was assayed (0). Fluorescence of samples at 678 nm was measured by excitation of Chi b (472 nm) in the presence of 0.1% octyl glucoside and expressed as fluorescence yield (), defined as total fluorescence of intact LHCs as a function of total Chi (0). Cells were transferred to the dark after 1 h in the light, as indicated by the bar across the top of the figure. to examine the photochemical efficiency of the PSII units. The rate at saturating light fluence was assumed to reflect the number of reaction centers, whereas the fluence of light at

half-maximal velocity should indicate the efficiency ofphoton capture and energy transfer to reaction centers. Figure 2 shows reciprocal plots with samples from cells greened at 38C for 1 h as compared to those from companion cells greened at 25C

for 5 h. At these time points, the cellular levels of Chl were similar. From plots for cells greened at the lower temperature, a saturation velocity of 30.3 ± 3.3 nmol DCI reduced/,ug Chl. min and a half-saturation fluence of 16.0 ± 0.2 x 104 ergs/ cm *s were obtained. For cells greened at 38C, 16.6 ± 1.3 nmol DCI reduced/jg Chl.min and 6.2 ± 2.2 x 104 ergs/ cm .s were obtained for the saturation velocity and halfsaturation fluence, respectively. These values, which were corrected for the residual Chl present in cells before exposure to light, were constant through the period of greening at the respective temperatures and demonstrate that PSII units were assembled in parallel with the accumulation ofChl. The lower maximal velocity of PSII units assembled at 38C suggests that they contain more Chl as compared with those made at 25°C and, perhaps as a result, were more efficient in capture of light energy. 2

2

Fluorescence Quenching during Greening We previously showed that the fluorescence intensity of Chl a in membranes from greening cells was the same whether it was excited through Chl b at 472 nm, by energy transfer, or directly at 438 nm (20). Such energy transfer indicated that the Chls were in LHCs. However, further analysis revealed that the fluorescence intensity did not increase along with Chl

1 323

ORIGIN OF THYLAKOID MEMBRANES

0)

4-

N

0.6

0.4

-

0.2 -0.0 -0.05

I 0

1

0.05 0.1

0.2

0.3

I/erg x 104 cm-2 s'1 Figure 2. Reciprocal plots of PS11 activity as a function of light fluence. Membranes from cells greened 1 h at 380C (0) or 5 h at 250C (U) were assayed for photoreduction of DCI. Rates were expressed as nmol DCI reduced/Ag Chl * min.

during greening at 38°C. Instead, fluorescence of Chl a in LHCs, assayed by excitation energy transfer from Chl b, remained approximately constant during an hour of greening. The ratio of fluorescence intensity to Chl (F/Chl) versus time of greening, consequently, decreased along a curve predicted by simple dilution (Fig. 1). Dissociation of protein-Chl complexes by addition of Triton X-100 to membrane suspensions increased fluorescence to a level proportional to the amount of Chl in the sample. As shown in Figure 3A, fluorescence intensity of Chl a in membranes from degreened cells increased only about 10% when LHCs were dissociated with Triton X- 100. This result implied that about 90% of the residual, intact LHCs in degreened cells were fluorescent and probably were not coupled to reaction centers (2, 9). However, after 15 min of light, a sufficient time for the cellular level of Chl to double, addition of Triton X- 100 resulted in more than a twofold increase in fluorescence (Fig. 3B). Because the spectra presented in Figure 3A and B were obtained with equal amounts of Chl, the fluorescence intensity of membranes before treatment with Triton X- 100 was less in Figure 3B than that shown in Figure 3A and corresponded to the 1 5-min point for the F/Chl ratio in Figure 1. These results suggest that newly formed LHCs, which accounted for about half the total Chl in the sample shown in Figure 3B, were coupled to reaction centers and quenched by a nonradiative energy decay process (2, 9, 17). In contrast, the residual LHCs apparently remained uncoupled to reaction centers. (Fig. 3 also illustrates the increase in the Chl a/b ratio [20] that occurred during the brief period of greening). The ratio F/Chl for residual Chl gradually increased during the preincubation period in the dark, i.e. after the temperature was raised from 25 to 38°C (data not shown). When cells were

transferred to the dark after an hour of greening, F/Chl also increased while PSII activity declined (Fig. 1). These observations suggest that at the higher temperature, the LHCs tend to become dissociated from reaction centers in the dark. A series of experiments were carried out to explore further the basis of the fluorescence quenching of LHCs during greening. An experiment similar to that described in Figure 1 was done in which cells were treated with gabaculine, an inhibitor of Chl synthesis (20). In this experiment, F/Chl remained nearly constant and was still 92 to 95% of the initial value after an hour in the light at 380C (data not shown). Thus, Chl synthesis, and not light alone, was required to cause a significant decrease in this ratio. The low F/Chl ratio achieved after an hour of greening was not affected when 10 ,gM DCMU was added to the assay mixture, a concentration of DCMU 10-fold greater than that required to completely inhibit photoreduction of DCI. DCMU also did not inhibit greening or affect the decrease in the F/Ch ratio during greening (data not shown), which indicated that electron flow through PSII units was not required for their assembly. In the presence of 200 ,ug/mL chloramphenicol, which strongly inhibits chloroplast protein synthesis, fluorescence of Chl a in intact membranes increased in parallel with the increase in

B

A

10)

a

I'

I

I

0 S.

I~ II I I II

600

II I I

1lw4w

,a-

700 600 700 Wave length

Figure 3. Fluorescence properties of ChI from cells greened at 380C. Fluorescence emission spectra of Chi (0.8 ,g/mL) in membranes were determined in the presence of 0.1% octyl glucoside by excitation of Chi b at 472 nm (solid curves, maximum 678 nm). After dissociation of LHCs by adding Triton X-1 00 to 0.5%, emission spectra for Chi b (maximum 650 nm) and Chi a (maximum 672 nm) were determined by excitation at 472 and 438 nm, respectively (dashed lines). Samples were obtained (A) before exposure of cells to light and (B) after 15 min of light.

1 324

HOOBER ET AL.

1.0

8 A

C

61r

0 0

0

4 =

0.6

0

A

2

I

I

a

a

O

0.2

U

B 0. _0

0.8

. 0.

rn.P

0

EL

I E

ILL

a A a

4

~0

30.4 0

ol ~ ~~~~ I I

0

0

5

5 0 10 15 Minutes in light

0 10

15

Figure 4. Distribution of Chi and fluorescence yield in cell fractions. Degreened cells were exposed to light for 0, 5, 10, or 15 min and then cell fractionation was carried out as described in "Materials and Methods." A, B, Total Chi was measured in each fraction. C, D, Fluorescence spectra were measured as described under Figure 3 and expressed as Fm/Fg, where Fm and Fs are the fluorescence intensity of Chi a in membranes and in Triton X-1 00-dissociated complexes, respectively. Data for y-1 cells are shown in panels A and C, and for cw-15 cells in panels B and D. (U), 25,000g/min; (A), 1 00,000g/min; and (0), 1 ,000,000g/min fractions.

total Chl (data not shown), although at this concentration of inhibitor the rate of accumulation of Chl was reduced to 20% of that in control cells. Thus, proteins synthesized within the chloroplast may also be necessary to quench fluorescence of newly formed LHCs. We confirmed the observations of Cahen et al. (2) that during greening at 25C the fluorescence intensity of LHCs increased approximately in parallel with Chl. With samples treated with octyl glucoside, i.e. conditions as described under Figure 3, approximately 30% of the Chl in LHCs assembled during greening at 25C was fluorescent (data not shown). The incomplete quenching of LHCs by reaction centers at the lower temperature may explain the requirement of a higher intensity of light for half-saturation (Fig. 2). Distribution of Fluorescent and Nonfluorescent LHCs A reasonable interpretation of the lack of significant change in the level of fluorescence of LHCs during greening at 38C is that preexisting and new LHCs were localized in different, physically separate membranes and that preexisting LHCs did not become coupled to PSII units during greening. If this interpretation were correct, the existence of a separate membrane containing a pool of preexisting LHCs could compro-

Plant Physiol. Vol. 96, 1991

mise morphological identification of newly assembled membranes. To test this possibility, homogenates of y-J cells, greened to allow a doubling of the Chl level, were fractionated by differential centrifugation. Figure 4A shows that most of the Chl was recovered in the 1,000,000g/min fraction from samples of y-J cells broken after 0, 5, 10, or 15 min of light. Fluorescence intensity of Chl a in membrane was measured by excitation of Chl b before addition of Triton X-100 (Fm) and again in the same sample by direct excitation of Chl a (Fa) after addition of sufficient detergent to dissociate LHCs. The ratio FmIFa was used as an index of the extent of quenching of LHCs, as illustrated in Figure 3. As shown in Figure 4C, the values of Fm/Fa and the pattern of decrease in the ratio with time were essentially the same in all fractions. Assays of photoreduction of DCI also showed that PSII activity was distributed among the fractions proportionally to the Chl content (data not shown). Analysis of Chl in each fraction by HPLC also showed that species of ChIs a and b with incompletely reduced side chains (20) were distributed proportionately among the fractions (data not shown). The rigorous conditions required to break y-J cells resulted in extensive fragmentation of membranes. Therefore, this experiment was repeated with cells of the wall-deficient strain cw-15, which also became yellow during growth in the dark and greened at 38°C with the same kinetics as y-J cells (data not shown). cw-15 cells were efficiently broken at 1000 p.s.i., conditions that reduced fragmentation of membranous structures. From cell-free homogenates of greening cw-15 cells, prepared in the same medium as was used with yI cells, 54 + 5% of the Chl was recovered in the 25,000g/min pellet, along with the bulk of the carotenoids, at each time point (Fig. 4B). Electron microscopic examination of the cell fractions revealed a complement of membrane structures consistent with the biochemical findings. The 25,000g/min fraction contained large membrane fragments, whereas the 1,000,OOOg/min fraction consisted of small vesicles. The 100,OOOg/min fraction contained a variety of membrane structures including numerous well-preserved mitochondria (not shown). Values for the ratio Fm/Fa decreased in all fractions with time of greening at the same rate found with yI cells (Fig. 4D). The similar fluorescence properties of the fractions, although the distribution of Chl was quite different in the two experiments shown in Figure 4, suggested that the old and new LHCs were not physically separate. We examined whether or not the distribution of Chl and its fluorescence properties in cw-15 cells were related to greening at elevated temperatures. Results similar to those shown in Figure 4B and D were obtained when cells were preincubated for 1.5 h in the dark at 38°C and then the temperature was dropped to 30°C during exposure to light. Also, the distribution of residual Chl in fractions prepared as described in Figure 4 did not change in cw-15 cells during the preincubation period at 38°C. These experiments showed that the fractionation results were not altered by the temperature of the greening experiments. Thylakoid Membrane Assembly Occurs at the Chloroplast Envelope The data presented above demonstrated that assembly of functional thylakoid membranes began immediately upon

ORIGIN OF THYLAKOID MEMBRANES

exposure of degreened cells to light. Therefore, an increase in membrane during the first few minutes of greening should mark the site of assembly. As shown by experiments with cw15 cells (Fig. 4B), the bulk of new Chl was associated with large membrane structures that sedimented in low gravitational fields. To investigate the source of these structures, cells were examined by EM. Degreened y-J cells were incubated at 38°C for 1.5 h in the dark and then fixed by injection into ice-cold 2% OS04. In preliminary experiments, we found that this procedure provided rapid fixation and better preservation of cellular structures than treating cells with 2% glutaraldehyde. Several cellular structures seemed notably affected by the higher temperature, some ofwhich are illustrated in Figure 5. The chloroplast was more irregular in shape than the cupshaped organelle found at lower temperatures (25) and lacked a pyrenoid body. In the cytoplasm, Golgi structures and associated vesicles were highly developed, and often five stacks were observed in a cross-section of the cell. Mitochondria were well preserved and exhibited extensive cristae. Samples were fixed after 0, 5, or 15 min of greening. Figure 6A shows that, in degreened cells, before exposure to light,

1 325

large areas of the chloroplast stroma were devoid of membrane material. A small amount of amorphous material was occasionally associated with the inner membrane of the chloroplast envelope (Fig. 6, arrows). However, a remarkable change occurred within 5 min of exposure to light. Arrays of membrane extended from the envelope, as shown in Figure 6B. Areas of the envelope engaged with these structures were joined by other regions that were free of thylakoid membranes, which suggested that membrane formation was localized to specific segments of the envelope. Figure 7 shows an area of the chloroplast at higher magnification from a cell that was exposed to light for 15 min. Developing thylakoid membranes appeared to be continuous with the inner envelope membrane, particularly at sites indicated by the arrows in Figure 7 (see also sites marked by arrowheads in Fig. 5, inset). The rate at which membranes were generated seemed to vary among cells in the sample, but in every case newly formed membranes were associated with the chloroplast envelope. A population of circular membrane profiles appeared in the stroma concomitant with thylakoid production (marked

Figure 5. Electron micrograph of a degreened y-1 cell incubated in the dark for 1.5 h at 38°C and then exposed to light for 15 min. Among the cellular structures included in this section of the cell are Golgi stacks (G) with associated vesicles, mitochondria (m), the nucleus (n), and glyoxysomes (g). The chloroplast (c) is irregular in ;; 4 shape and occasionally surrounds projections of the cytoplasm (*). Thylakoid membranes extend ;t iw into the stroma from the chloroplast envelope (arrowheads). Magnification x 30,000. Inset, The " area indicated by arrowheads is shown at higher t - | C r i0 l vi

HOOBER ET AL.

aIrian g ilie ,d. xplosure.k -\-: 4 -AeqreerlenpeLT

Plant Physiol. Vol. 96, 1991

L

*~~~~~~~~~~~~~-a *.r-i(i; A- h& *r t

-

r-r ar dk8L ~:

--pi ,

atec;

rirsntasi C,

r!I'..cylack;o ).ccasionaiix T ri:

:: A k t .

al t e a o-I 1. A -rdP, a d as e -let ,

v'!eigoh

M agnifiati-a3i

3)w;

*

e

rxpUsec

*%i.r.i

(t:!d; rOdndl(jarI adra: :tak. ri ;.

--

( i

pp

by arrowheads in Fig. 7). These structures may have been extensions of membrane from other regions of the chloroplast into the plane of the section or vesicles that developed along with the expanding thylakoid membranes. The fluorescence data from cell fractionation experiments, shown in Figure 4, indicated that the slowly sedimenting structures collected in the l,000,000g/min fraction, which by electron microscopic examination consisted of vesicles similar in size to those in the stroma, had similar properties to the larger, more rapidly sedimenting membranes. Thus, because of the small amount of Chl present in the 1,000,000g/min fraction (Fig. 4B), these structures were possibly vesicles derived from the developing thylakoids, as their morphology suggests. DISCUSSION Functional PSII units are assembled in parallel with accumulation of Chl and LHCPs in C. reinhardtii y-1 at 38°C. In this respect, greening at 38C is similar to the process at 25C (2, 7, 25). However, a significant difference between greening at the two temperatures is the linearity of the process at 38°C, which enables an experimental approach to initial events in membrane assembly. Kinetic studies of Chl accumulation at 380C showed that Chl b (and probably an equivalent amount of Chl a) is quantitatively complexed with LHCPs (20). Moreover, the rapid degradation of LHCPs under conditions in which Chl was not made indicated that the proteins must immediately associate with Chl for protection from proteolysis (13). Stabilization of Chl-binding proteins by Chl also occurs in other systems and seems to be a general phenomenon (15, 21, 24). We concluded from our kinetic studies that

interaction of the cytoplasmically synthesized LHCPs with Chl occurs concomitant with transport ofthe proteins through the chloroplast envelope. It seemed possible, therefore, that the envelope participates in assembly of LHCs and of the membrane. Rapid assembly of thylakoid membranes during greening of y-l cells provided the opportunity to directly determine the site of formation of the membrane. Electron micrographs of cells exposed to light for only 5 min revealed extensive arrays of membrane material extending from the inner envelope membrane, a feature not seen in cells before exposure to light. These results suggest that thylakoid membranes develop by local expansion and infolding of the inner membrane, as was proposed previously (6, 11, 22, 29, 32). Because the envelope is the site of synthesis of galactosyl diglycerides, the major thylakoid lipids (6, 8), light apparently also caused a large increase in lipid formation under these conditions. If these membranes contain newly synthesized Chl in LHCs, as the cell fractionation experiments suggest, it is reasonable to assume that these membranes also include newly synthesized LHCPs. In support of this conclusion, LHCPs were initially incorporated into membranes that had the same density as thylakoid membranes, with no evidence in this system for a soluble, stromal pool (13). Data obtained previously suggested that LHCPs were synthesized in the dark at 38°C, imported into the chloroplast, and subsequently degraded (13). A small fraction, however, escaped degradation and was recovered from dark-incubated cells in membranes as the correctly processed and integrated forms. Although these results imply that the amount of LHCP precursors imported into the chloroplast in the dark is roughly

ORIGIN OF THYLAKOID MEMBRANES

Figure 7. Section of the chloroplast from a cell exposed to light for 15 min. Extensive regions of the chloroplast envelope are associated with newly formed membranes. At several sites (single and double arrows) channels seem to extend through the envelope membranes. Numerous vesicles (large arrowheads) in the chloroplast stroma (c) are similar in morphology to membranes emanating from the envelope (small arrowheads). Mitochondria (m) are present in the surrounding cytoplasm. Magnification x 47,000.

1 327

not increase as Chl accumulated during greening. Fluorescence measurements were made with highly dilute suspensions of membrane fragments, and thus quenching by subsequent photosynthetic reactions, such as occurs in intact cells or chloroplasts (2, 7), cannot explain our results. Aggregation of LHCs can lead to annihilation of excited states, which strongly attenuates fluorescence yields ( 14, 31). Fluorescence can also be quenched by synthesis in chloroplasts of zeaxanthin (5). However, these mechanisms should decrease efficiency of the photosynthetic unit, the opposite of what we found. Rather, our data suggest that new LHCs are connected to reaction centers and that absorbed light energy is transferred to other components of the PSII unit in a radiationless process (e.g. ref. 17). The quenching of fluorescence clearly was a consequence of the environment of the Chls, because dissociation of LHCs with Triton X-100 increased the fluorescence yield to the level expected with free Chl. The persistence of a population of fluorescent, and apparently uncoupled, LHCs during greening requires further study, but may reflect a tight coordination between synthesis of the components and assembly of the membrane. Greening at the higher temperature offers an advantage to this type of analysis because the kinetics are linear with time in light and membrane assembly can be examined over a time period of only a few minutes. This system should also allow analysis of the integration of specific components into the membrane. ACKNOWLEDGMENTS We gratefully acknowledge the technical assistance of Barbara Keller and Marie Hughes. We also thank Dr. Dawn Marks for critical reading of the manuscript. LITERATURE CITED

the same as in the light, electron micrographs showed that the morphology of the envelope was relatively unchanged by this flux (Fig. 6A). However, immediately adjacent to, and perhaps in contact with, scattered segments of the envelope were small amounts of amorphous material. Whether these structures were responsible for the low level of accumulation of LHCPs in the dark cannot be determined by this work. Extensive in vitro studies have shown that, following import, the LHCPs do not remain associated with the envelope. When integration of LHCPs, or their precursors, into membrane is impaired by deletion of segments of the protein near the carboxyl terminus or by addition of ionophores, the proteins in some cases accumulate in the chloroplast stroma or are degraded (3, 4, 18). Recently, by rapidly stopping integration into thylakoid membranes by use of HgCl2, Reed et al. (28) found a significant pool of soluble LHCP after transport into isolated chloroplasts. It remains an important question to be explored by future work if in the Chlamydomonas system insertion of LHCPs into membranes occurs within the envelope during import or whether insertion normally occurs deeper within the chloroplast at a rate too rapid to detect. Because PSII photochemistry was negligible at the time cells were exposed to light, we attributed the initial fluorescence of intact membranes to LHCs that were not connected to reaction centers. Interestingly, the level of fluorescence did

1. Butler PJG, Kfihlbrandt W (1988) Determination of the aggregate size in detergent solution of the light-harvesting chlorophyll a/b-protein complex from chloroplast membranes. Proc Natl Acad Sci USA 85: 3797-3801 2. Cahen D, Malkin S, Shochat S, Ohad 1 (1976) Development of photosystem II complex during greening of Chlamydomonas reinhardi y-J. Plant Physiol 58: 257-267 3. Clark SE, Oblong JE, Lamppa GK (1990) Loss of efficient import and thylakoid insertion due to N- and C-terminal deletions in the light-harvesting chlorophyll a/b binding protein. Plant Cell 2: 173-184 4. Cline K, Fulsom DR, Viitanen PV (1989) An imported thylakoid protein accumulates in the stroma when insertion into thylakoids is inhibited. J Biol Chem 264: 14225-14232 5. Demmig-Adams B, Adams WW III, Heber U, Neimanis S, Winter K, Kriiger A, Czygan F-C, Bilger W, Bjorkman 0 (1990) Inhibition of zeaxanthin formation and of rapid changes in radiationless energy dissipation by dithiothreitol in spinach leaves and chloroplasts. Plant Physiol 92: 293-301 6. Douce R, Block MA, Dorne A-J, Joyard J (1984) The plastid envelope membranes: their structure, composition, and role in chloroplast biogenesis. Subcell Biochem 10: 1-84 7. Gershoni J, Shochat S, Malkin S, Ohad I (1982) Functional organization of the chlorophyll-containing complexes of Chlamydomonas reinhardi. Plant Physiol 70: 637-644 8. Heemskerk JWM, Storz T, Schmidt RR, Heinz E (1990) Biosynthesis of digalactosyldiacylglycerol in plastids from 16:3 and 18:3 plants. Plant Physiol 93: 1286-1294 9. Henkin BM, Sauer K (1977) Magnesium ion effects on chloroplast photosystem II fluorescence and photochemistry. Photochem Photobiol 26: 277-286

1 328

HOOBER ET AL.

10. Henningsen KW, Boynton JE (1969) Macromolecular physiology of plastids. VII. The effect of a brief illumination on plastids of dark-grown barley leaves. J Cell Sci 5: 757-793 1 1. Hoober JK (1987) The molecular basis of chloroplast development. In MD Hatch, NK Boardman, eds, The Biochemistry of Plants, Vol 10. Academic Press, Orlando, pp 1-74 12. Hoober JK, Marks DB, Keller BJ, Margulies MM (1982) Regulation of accumulation of the major thylakoid polypeptides in Chlamydomonas reinhardtii y-J at 25°C and 38°C. J Cell Biol 95: 552-558 13. Hoober JK, Maloney MA, Asbury LR, Marks DB (1990) Accumulation ofchlorophyll a/b-binding polypeptides in Chlamydomonas reinhardtii y-J in the light or dark at 38°C. Plant Physiol 92: 419-426 14. Ide JP, Kiug DR, KuHlbrandt W, Giorgi LB, Porter G (1987) The state of detergent solubilised light-harvesting chlorophylla/b protein complex as monitored by picosecond time-resolved fluorescence and circular dichroism. Biochim Biophys Acta 893: 349-364 15. Jensen KH, Herrin DL, Plumley FG, Schmidt GW (1986) Biogenesis of photosystem II complexes: trancriptional, translational, and posttranslational regulation. J Cell Biol 103: 13151325 16. Kindle K (1987) Expression of a gene for a light-harvesting chlorophyll a/b-binding protein in Chlamydomonas reinhardtii: effect of light and acetate. Plant Mol Biol 9: 547-563 17. Knaff DB (1977) The primary reaction of plant photosystem II. Photochem Photobiol 26: 327-340 18. Kohorn BD, Tobin EM (1989) A hydrophobic, carboxy-proximal region of a light-harvesting chlorophyll a/b protein is necessary for stable integration into thylakoid membranes. Plant Cell 1: 159-166 19. Malnoe P. Mayfield SP, Rochaix J-D (1988) Comparative analysis of the biogenesis of photosystem II in the wild-type and y1 mutant of Chlamydomonas reinhardiii. J Cell Biol 106: 609616 20. Maloney MA, Hoober JK, Marks DB (1989) Kinetics of chlorophyll accumulation and formation of chlorophyll-protein complexes during greening of Chlamydomonas reinhardtii y-J at 38°C. Plant Physiol 91: 1100-1106

Plant Physiol. Vol. 96, 1991

21. Mishkind ML, Jensen KH, Bran AF, Plumley FG, Schmidt GW (1985) Roles of proteases in chloroplast biogenesis. Curr Top Plant Biochem Physiol 4: 34-50 22. Muihlethaler VK, Frey-Wyssling A (1959) Entwicklung und Struktur der Proplastiden. J Biophys Biochem Cytol 6: 507512 23. Mullet JE (1983) The amino acid sequence of the polypeptide segment which regulates membrane adhesion (grana stacking) in chloroplasts. J Biol Chem 258: 9941-9948 24. Mullet JE, Klein PG, Klein RR (1990) Chlorophyll regulates accumulation of the plastid-encoded chlorophyll apoproteins CP43 and Dl by increasing apoprotein stability. Proc Natl Acad Sci USA 87: 4038-4042 25. Ohad I, Siekevitz P., Palade GE (1967) Biogenesis of chloroplast membranes. II. Plastid differentiation during greening of a dark-grown algal mutant (Chlamydomonas reinhardi). J Cell Biol 35: 553-584 26. Paavola LG, Strauss JF III, Boyd CO, NestlerJE (1985) Uptake of gold- and [3H]cholesteryl linoleate-labeled human low density lipoprotein by cultured rat granulosa cells: cellular mechanisms involved in lipoprotein metabolism and their importance to steroidogenesis. J Cell Biol 100: 1235-1247 27. Plumley FG, Schmidt GW (1987) Reconstitution of chlorophyll a/b light-harvesting complexes: xanthophyll-dependent assembly and energy transfer. Proc Natl Acad Sci USA 84: 146-150 28. Reed JE, Cline K, Stephens LC, Bacot KO, Viitanen P (1990) Early events in the import/assembly pathway of an integral thylakoid protein. Eur J Biochem 194: 33-42 29. Schmidt GW, Mishkind ML (1986) The transport of proteins into chloroplasts. Annu Rev Biochem 55: 879-912 30. Sun ASK, Sauer K (1971) Pigment systems and electron transport in chloroplasts I. Quantum requirements for the two light reactions in spinach chloroplasts. Biochim Biophys Acta 234: 399-414 31. Van Grondele R (1985) Excitation energy transfer, trapping and annihilation in photosynthetic systems. Biochim Biophys Acta 811:147-195 32. Von Wettstein D (1961) Nuclear and cytoplasmic factors in development of chloroplast structure and function. Can J Bot 39: 1537-1545