Chlamydomonas reinhardtii (chloramphenicol/puromycin reaction/electron microscopy/synchronous cultures)

Transcription

1 Proc. Nat. Acad. Sci. USA Vol. 70, No. 5, pp , May 1973 Attachment of Chloroplast Polysomes to Thylakoid Membranes in Chlamydomonas reinhardtii (chloramphenicol/puromycin reaction/electron microscopy/synchronous cultures) NAM-HAI CHUA, G. BLOBEL, P. SIEKEVITZ, AND G. E. PALADE The Rockefeller University, New York, N.Y Contributed by G. E. Palade, March 16,1973 ABSTRACT Treatment of synchronous cultures of Chlamydomonas reinhardtii with chloramphenicol at 4 hr after the beginning of the light phase led to a preferential loss of 70S ribosomes from the 17,000 X gmax supernate. The "lost" 70S ribosomes were found associated with a thylakoid membrane fraction prepared from the 17,000 X gm. pellet. Electron microscopic examinations of this fraction revealed that the 70S ribosomes were bound to the unstacked regions of the thylakoid membranes as polygonal penta- and hexamers. These bound ribosomes were only released by treatment with 500 mm KCI and puromycin, suggesting that both ionic interactions and nascent peptide chains were involved in the ribosome-membrane attachment. Since growth of the thylakoid membranes occurs in the light, it is suggested that bound chloroplast ribosomes function in the synthesis of thylakoid membrane proteins. Eukaryotic ribosomes (80 S) occur in two distinct states within the cytoplasm of animal cells: they are either free or bound to the membranes of the endoplasmic reticulum (1, 2). The relative distribution of the two ribosome classes greatly depends on the cell type. Thus, cells active in the synthesis of secretory proteins contain a large proportion of bound ribosomes (3, 4), whereas cells synthesizing proteins primarily for intracellular use have most of their ribosomes in the free state (5, 6). In a limited number of cases, it has been shown that the free and membrane-bound ribosomes are indeed involved in the synthesis of different classes of proteins (3, 7-11), in agreement with the generalization mentioned above. There is, however, indirect evidence that in mammalian hepatocytes the bound ribosomes synthesize not only secretory proteins but also proteins of the endoplasmic reticulum membrane (12-16). Since chloroplasts contain their own ribosomes (70 S), and since their is evidence that these ribosomes synthesize some of the proteins of the thylakoid membranes (17-20), the question arises as to whether these ribosomes also exist as two distinct classes within the chloroplast. In this paper, we demonstrate that in synchronous cultures of Chlamydomonas reinhardtii a detectable fraction of the chloroplast ribosome population becomes bound to the thylakoid membranes during the light phase of the cell cycle at the time when growth of the thylakoid membranes occurs. The correlation suggests that these bound ribosomes are involved in synthesis of thylakoid membrane proteins. Abbreviations: 817, 17,000 X g. superate; P17, 17,000 X - gr..x pellet MATERIALS AND METHODS Culture Conditions; Cell Disruption. Cells of the wild-type strain (137 c, mt +) of C. reinhardtii were grown in the minimal medium of Sager and Granick (21) as described by Ohad et al. (22). The culture was synchronized by exposing the cells to a cycle of alternating 12-hr periods of light (light intensity, about 4000 lux) and dark at 250 (23). Experiments were performed when the cultures reached a cell density of 1 to 2 X 106 cells per ml. Each culture was divided into two equal portions 4 hr after the beginning of the light phase. To one sample, chloramphenicol was added from a stock solution of 100 mg of chloramphenicol per ml of ethanol to give a final concentration of 100 yg/ml, the control sample receiving an equivalent amount of ethanol (final concentration, 0.1%). Both samples were further incubated in the light for 1 hr, which brought the cultures to the 5th hr of the light phase. At this time, the cells were harvested by low-speed centrifugation in a Sorvall GSA rotor at 00. The pelleted cells were washed once in TKMD buffer containing 25 mm Trise HCl (ph 7.5)-25 mm KCl-25 mm MgCl2-5 mm dithiothreitol. They were then suspended in the same buffer to a density of 1 to 2 X 108 cells per ml, and finally disrupted by passage through a chilled French pressure cell maintained at 4800 lb/ inch2. The efficiency of cell breakage approached 100% as monitored by electron microscopy. All subsequent procedures were performed at 4. Isolation of a Thylakoid Membrane Fraction. The disrupted cells were centrifuged at 17,000 X gr0a1 for 10 min, and the supernate was saved. In order to remove any free ribosomes that might have been trapped among the membranes during the preceding centrifugation, the pellet was dispersed in TKMD buffer with a Potter homogenizer and the suspension was centrifuged again at 17,000 X ge for 10 min. The supernates from this and the preceding centrifugation were pooled and are referred to hereafter as S17. The 17,000 X gma pellet (P17) contained, in addition to thylakoid membranes, eye-spot granules and other intracellular membranes, as well as cell-wall fragments and broken nuclei. A fraction enriched in thylakoid membranes was isolated from P17 by a flotation procedure. To this intent, P17 was suspended in 1.87 M sucrose in TKMD buffer to a chlorophyll concentration of about 300,gg/ml. 4.5 ml of the suspension were overlaid with 5 ml of TMKD buffer and centrifuged at 320,000 X gmar for 40 min at 40, which caused the thylakoid membranes to float to the sucrose-buffer interface, the eye-

2 Proc. Nat. Acad. Sci. USA 70 (1978) spot granules to move further up into the buffer layer, and the cell-wall fragments and nuclei to sediment at the bottom of the tubes. The membrane band was collected with a syringe fitted with a large-bore needle; it was dispersed by homogenization and diluted with three volumes of TKMD buffer. The suspension was centrifuged at 17,000 X gm.. for 10 min, and the resulting pellet was finally suspended in TKMD buffer to a chlorophyll concentration of about mg/ml. Electron microscopic examination of this pellet revealed that it contained primarily thylakoid membranes with a few contaminant rough microsomes. About 60% of the total chlorophyll of the initial sample was recovered in this fraction, which is referred to hereafter as the thylakoid membrane fraction. High Salt-Puromycin Treatment. The high salt-puromycin reaction was done according to Blobel and Sabatini (24). The reaction mixture contained: 0.5 ml of S17 (9-12 A260 units), or P17 ( mg chlorophyll per ml), or thylakoid membrane fraction ( mg of chlorophyll per ml); 0.1 ml of 5 mm puromycin (adjusted to ph 7.0 with KOH), and 0.4 ml of compensating buffer. Control samples received 0.1 ml of distilled H20 instead of puromycin. The composition of the compensating buffer was adjusted to give, in the final reaction mixture 50 mm Tris1HC (ph 7.5)-500 mm KCl-25 mm MgCl2-5 mm dithiothreitol. The reaction mixtures were incubated at 370 for 10 min, and suitable aliquots were layered onto 5-20% linear sucrose gradients containing the same salts in the same composition as the reaction mixture. For the S17 fraction, the aliquot contained about A260 units, whereas for the P17 and the thylakoid membrane fractions, the aliquots contained 250,ug of chlorophyll. The gradients were centrifuged at 39,000 rpm for 3 hr at 180 in an SB 283 rotor of an International Equipment Co. (JEC) centrifuge, model B 60. The absorbance at 254 nm was measured with an Instrumentation Specialities Co. gradient fractionator equipped with a UV analyzer. Electron Microscopy. Cells and representative fractions were first pelleted and then fixed by resuspension in 5% or 0.5% glutaraldehyde in 0.1 M cacodylate buffer (ph 7.4) in 25 mm MgCl2 for 30 min followed by 1% OS04 (in the same buffer- Mg solution) for 60 min. The specimens were stained in block in uranyl acetate and then embedded in epon. Sections, doubly stained with uranyl acetate and lead, were examined in a Siemens 101 electron microscope. Others. Chlorophyll concentrations were determined according to Arnon (25). Chloramphenicol and puromycin were purchased from Sigma Chemical Co. and Nutritional Biochemicals Corp., respectively. RESULTS Cells of C. reinhardtii contain two types of ribosomes with sedimentation coefficients of 70 S and 80 S. The first are derived from the chloroplast and the second from the cytoplasm (26-28). We have recently reported (29) that ribosomal monomers of both these types readily dissociate into their respective subunits upon exposure to 500 mmi KCl in the presence of 25 mm MgCl2, whereas 70 S and 80 S ribosomes with nascent polypeptide chains, i.e., monosomes and polysomes, dissociate under these conditions only after the chains have been released by puromycin. Therefore, by inducing dissociation in high-salt in the presence or absence of puromycin, it should be possible to estimate in a given preparation the proportion of active ribosomes, i.e., ribosomes with nascent Chloroplast Ribosomes Bound to Thylakoid Membranes 1555 chains. This technique was applied to the analysis of the S17 fraction prepared from a synchronous culture of C. reinhardtii harvested at the 5th hr of the light phase, and the results (Fig. la and b) showed that the addition of puromycin did not affect the amounts of 70S subunits displayed on the gradient and increased the amounts of 80S subunits by only about 10%. It appears, therefore, that about 90% of the 80S ribosomes and almost all of the 70S ribosomes in the S17 fraction were monomers. The occurrence of such a large proportion of ribosomes free of nascent chains in homogenates of C. reinhardtii was noted before and ascribed to the slow cooling of the cells during harvesting (29). Indeed, in other cases (30, 31), it Lo cmj Control S 17 P 17 TMF (a) -PM (b) + PM (c) + PM (d) + PM LP0 44i S co0s Chloromphenicol S 17 P 17 TMF (e) -PM (f) + PM (g) + PM (h) +PM L i OL 0.0 FIG. 1. Sucrose gradient analysis of ribosomal subunits in S17, P17, and the thylakoid membrane fraction (TMF). (a) S17 of control sample - puromycin, 2.06 A260 units; (b) S17 of control sample + puromycin, 2.06 A260 units; (c) P17 of control sample + puromycin, 250 ug of chlorophyll; (d) TMF of control sample + puromycin, 250,ug of chlorophyll; (e) S17 of chloramphenicoltreated sample - puromycin, 1.95 A260 units; (f) S17 of chloramphenicol-treated sample + puromycin, 1.95 A260 units; (g) P17 of chloramphenicol-treated sample + puromycin, 250,ug of chlorophyll; (h) TMF of chloramphenicol-treated sample + puromycin, 250,ug of chlorophyll. The direction of sedimentation is indicated by the horizontal arrows. The small peak marked by an asterisk in (a) and (e) is ribulose-1,5-diphosphate carboxylase. The vertical arrow in (b), (c), (d), (f), (g), and (h) indicates the position of L70. In all cases, the thylakoid membranes were sedimented to the bottom. The subunits were identified by cocentrifugation with purified subunits prepared by described methods (29). Representative experiment out of a series of 5. L80 and S80, large and small subunits of 808 ribosomes, respectively; L70 and S70, large and small subunits of 70S ribosomes, respectively.

3 1556 Cell Biology: Chua et al. has been demonstrated that slow cooling inhibits initiation of protein synthesis but allows both polypeptide chain elongation to continue and termination to occur, leading to an accumulation of free ribosomes in the corresponding in vitro systems. Hence, it can be assumed that a similar polysomal runoff occurs in C. reinhardtii. Were the above interpretation correct, it should be possible to prevent chloroplast polysomal runoff by treating the cells with antibiotics that specifically block peptide chain elonga- It toj 0.6 t 0 4 (a) (c) -PM LS( 25 mm KCI (b) 500mM KCI FIG. 2. Effects of KCl concentrations and puromycin on the release of ribosomes from thylakoid membrane fraction prepared from chloramphenicol-treated cells. The release of ribosomes mediated by high-salt (500 mm KCl) with or without puromycin was done as described under Methods except that the final puromycin concentration was reduced to 40 AM. The low-salt (25 mm KCl) reactions were done similarly except that the KCl concentration in the reaction mixture was 25 mm. Aliquots from each reaction mixture containing 250 jg of chlorophyll were layered onto 10-45% linear sucrose gradients containing the same salts in the same composition as the reaction mixture. Gradients were centrifuged at 39,000 rpm for 1 hr at 18 in an SB 283 rotor of an IEC centrifuge. The absorbance on top of (a) and (c) was due to oxidized dithiothreitol and that on top of (b) and (d) was due to oxidized dithiothreitol and puromycin. Representative experiment out of a series of 5. Mb, membrane band; PM, puromycin. Proc. Nat. Acad. Sci. USA 70 (1973) tion or termination on 70S ribosomes. One such antibiotic is chloramphenicol, a known inhibitor of protein synthesis on chloroplast ribosomes in both higher plants and algae (see ref. 32). Surprisingly, however, analysis of an S17 fraction prepared from chloramphenicol-treated cells revealed a large reduction of 70S subunits relative to 80S subunits as compared to the S17 of the control sample (compare Fig. lb and f). Furthermore, there was very little difference in the ribosomal subunit profiles in either the absence (Fig. le) or presence (Fig. if) of puromycin. These results could be explained by assuming that treatment with chloramphenicol led to a shift of 70S ribosomes from S17 to P17, i.e., to membranes sedimenting at 17,000 X gmax. If this were the case, there should be a significant enrichment in 70S ribosomes in the P17 fraction derived from the chloramphenicol-treated sample as compared to its control. This prediction was borne out by the results presented in Fig. lc and g. The ribosomal subunit profiles in these two figures can be compared directly since the initial P17 samples contained the same amount of chlorophyll. It can be seen that although the amounts of 80S ribosomal subunits were about the same in both, there was a significant increase in the amounts of 70S ribosomal subunits in the chloramphenicoltreated sample. To investigate the structural connections of the sedimented 70S ribosomes, we isolated from P17 a thylakoid membrane fraction by the flotation procedure (see Methods). Analysis of this fraction by the high salt-puromycin technique showed that it contained a large amount of 70S ribosomes relative to 80S ribosomes (Fig. lh) when it was derived from the chloramphenicol-treated cells. The thylakoid membrane fractions prepared from the control sample contained a much smaller amount of ribosomes, and these were equally distributed between chloroplastic and cytoplasmic types (Fig. ld). So far our results suggested that treatment with chloramphenicol led to a specific association between chloroplast ribosomes and thylakoid membranes. The following experiments were designed to explore whether this association resembled that of eukaryotic ribosomes with the endoplasmic reticulum. The latter association is known to involve direct salt-sensitive bonds between the large ribosomal subunit and the membrane and an indirect linkage via the nascent chain TABLE 1. Relative amounts of ribosomes released from thylakoid membrane fraction at high or low KCI concenrration in the absence or presence of puromycin Relative amounts of ribosomes released Conditions Exp mm KCl + PM* mm KCl-PM mm KCl+ PM mm KCl-PM Experiments were carried out as described in Fig. 2. The relative amounts of ribosomes released from the thylakoid membrane fraction were assayed by the relative amounts of absorbance at 254 nm on the sucrose gradients. The thylakoid membrane fractions were prepared from chloramphenicol-treated cells and contained 20-30% of the total 70S ribosomes. PM, puromycin. * Ribosomes still left on membranes after 500 mm KC1 + PM 10% of the content of the original thylakoid membrane fraction sample.

4 Proc. Nat. Acad. Sci. USA 70 (1973) Chloroplast Ribosomes Bound to Thylakoid Membranes 1557 I \ A -3.,; :1-1z..e. III, k. 14!, ' ?.9 p3 0@..~i FIGS. 3 and 4 show thylakoid membrane fractions prepared from synchronized cultures treated with chloramphenicol (100 Asg/ml) between the 4th and 5th hr of the light phase. In both figures, the arrows point to areas of persisting fusion between thylakoids. FIG. 3. Thylakoid membrane fraction washed with low salt-no puromycin. Grana cut at different angles appear at g. Polysomes can be seen on practically all unstacked thylakoid membranes. Penta- and hexamers in full-faced view on obliquely sectioned membranes are marked by arrows. X 40,000. FIG. 4. Thylakoid membrane fraction treated with high salt only. Bound polysomes appear on exposed thylakoid surfaces: a row of three particles, attached to a normally sectioned membrane, can be seen at pi, and another row-only partly included in the section-at p2; at p3 is a hexamer in full-faced view on an obliquely sectioned thylakoid membrane. X 180,000. (33). Thylakoid membrane fractions were therefore treated with KCl at low (25 mm) or high (500 mm) concentration in the presence or absence of puromycin. As is the case of the bound 80S ribosomes of mammalian hepatic rough microsomes, maximal release of 70S ribosomes from the thylakoid membranes required both puromycin and a high KCl concentration (Fig. 2d). High-salt treatment alone removed about 50% of the 70S ribosomes (Fig. 2c; Table 1), and low-salt treatment with or without puromycin removed only 20-30% (Fig. 2a and b; Table 1). These results are entirely comparable to those obtained under similar conditions with the 80S ribosomes of rough microsomes prepared from rat liver (33). By these criteria, therefore, the 70S ribosome-thylakoid membrane interaction is similar to that established between the 80S ribosomes and the endoplasmic reticulum membranes of mammalian hepatocytes. Electron microscopic examinations revealed that all thylakoid membrane fractions consisted essentially of swollen thylakoids, either free or still stacked into disorganized grana; they contained no recognizable mitochondria and only a few rough microsomes. All fractions derived from chloramphenicol-treated cellsexcept for the fraction treated with high-salt and puromycinhad 70S ribosomes (a 200 A) * attached to the membranes * The diameter of 80S ribosomes in Chlamydomonas is about A. of their thylakoids. These ribosomes were restricted to those parts of the membranes not involved in fusion into grana (Fig. 3). They appeared either singly or in short rows of 2-3 particles in normal sections, and as characteristic, closed, ringlike or polygonal penta- and hexamers in grazing sections (Fig. 4). Larger polymers were only occasionally encountered, but images suggesting partial detachments were rather frequent. Thylakoid membrane fraction treated with low salt with or without puromycin showed, in addition, a relatively large population of ribosomes dispersed either singly or in clusters among the membranes. These apparently free ribosomes were no longer present in thylakoid membrane fraction treated with high salt only, which fraction had essentially only attached polysomes. After high-salt puromycin, only occasional ribosomes were left on the membranes. Thylakoid membrane fractions prepared from control cells showed a considerably lower population of attached penta- and hexamers. Preliminary observations indicate that attached polysomes of the type described are present in relatively large numbers in intact chloramphenicol-treated cells and in the P17 fractions derived from them. DISCUSSION Although the association between 80S ribosomes and membranes has been established for some time (1, 2), there is as yet,:cb E-,.

5 1558 Cell Biology: Chua et al. no convincing evidence that 70S ribosomes are involved in a comparable association. Previous morphological (34, 35) and biochemical (36) studies have raised the possibility that ribosome attachment to thylakoid membranes might occur in chloroplasts. The experiments reported here establish ununequivocally that an association of 70S ribosomes with thylakoid membranes indeed exists and that, under conditions in which chloroplast polysomal runoff is prevented by treatment with chloramphenicol, it is possible to isolate a thylakoid membrane fraction containing bound chloroplast polysomes. The association appears to be limited in time and to correspond to periods of membrane synthesis. Thus, it was detected at the 4th-5th hr of the light phase, a time at which there is extensive growth of thylakoid membranes and synthesis of at least two membrane proteins, cytochromes 553 and 559 (20, 37). Moreover, our preliminary results indicate that there are no membrane-bound ribosomes during the dark phase of the cell cycle when there is no growth of thylakoid membranes (37). Taken together with data that show that chloramphenicol inhibits the synthesis of certain thylakoid membrane proteins (17-20), our results strongly suggest that membranebound chloroplast ribosomes are involved in the synthesis of such proteins. The association is also limited in space; it appears to involve at any given time only membrane areas exposed to the chloroplast stroma. The fact that the dissociation is obtained by high salt-puromycin definitely implicates a nascent polypeptide chain in the association of the 70S ribosomes with the thylakoid membrane, but it remains to be seen whether the newly synthesized protein remains embedded in the membrane or is segregated in the intradisc space (in analogy with secretory proteins). The implication of bound chloroplast ribosomes in the biogenesis of thylakoid membranes raises the question whether similarly bound ribosomes are involved in the synthesis of other membranes in both prokaryotic and eukaryotic systems. The conditions for detecting these bound ribosomes, if they exist, may be difficult to control since ribosome association with the corresponding membranes may be restricted in both time and space. In the case of the chloroplast-bound ribosomes, the detection was made possible only by combining synchronization of the cell population with the arrest of chloroplast polysomal runoff. Similar manipulations may have to be used for the demonstration of bound ribosomes in other systems. We thank Mrs. Louise Castle for technical assistance in electron microscopy. This work was made possible by NIH Grant 2 RO1HD GM to P.S. 1. Palade, G. E. & Siekevitz, P. (1956) J. Biophys. Biochem. Cytol. 2, Proc. Nat. Acad. Sct. USA 70 (1973) 2. Palade, G. E. (1958) in Microsomal Particles and Protein Synthesis, ed. Roberts, R. B. (Pergamon Press, New York), p Siekevitz, P. & Palade, G. E. (1960) J. Biophys. Biochem. Cytol. 7, Blobel, G. & Potter, V. R. (1967) J. Mol. Biol. 26, Attardi, B., Cravioto, B. & Attardi, G. (1969) J. Mol. Biol. 44, Andrews, T. M. & Tata, J. R. (1971) Biochem. J. 121, Redman, C. M. (1968) Biochem. Biophys. Res. Commun. 31, Redman, C. M. (1969) Fed. Proc. 28, Ganoza, M. C. & Williams, C. A. (1969) Proc. Nat. Acad. Sci. USA 63, Hicks, S. T., Drysdale, J. W. & Munro, H. N. (1969) Science 164, Takagi, M., Tanaka, T. & Ogata, K. (1970) Biochim. Biophys. Acta 217, Dallner, G., Siekevitz, P. & Palade, G. E. (1966) J. Cell Biol. 30, Dallner, G., Siekevitz, P. & Palade, G. E. (1966) J. Cell Biol. 30, Sargent, J. R. & Vadlamudi, B. P. (1968) Biochem. J. 107, Omura, T. & Kuriyama, Y. (1971) J. Biochem. 69, Ragnotti, G., Lawford, G. R. & Campbell, P. N. (1969) Biochem. J. 112, Hoober, J. K., Siekevitz, P. & Palade, G. E. (1969) J. Biol. Chem. 244, Hoober, J. K. (1970) J. Biol. Chem. 245, Eytan, G. & Ohad, I. (1970) J. Biol. Chem. 245, Armstrong, J. J., Surzycki, S. J., Moll, B. & Levine, R. P. (1971) Biochemistry 10, Sager, R. & Granick, S. (1953) Ann. N.Y. Acad. Sci. 56, Ohad, I., Siekevitz, P. & Palade, G. E. (1967) J. Cell Biol. 35, Bernstein, E. (1964) J. Protozool. 11, Blobel, G. & Sabatini, D. D. (1971) Proc. Nat. Acad. Sci. USA 68, Arnon, D. I. (1949) Plant Physiol. 24, Hoober, J. K. & Blobel, G. (1969) J. Mol. Biol. 41, Goodenough, U. W. & Levine, R. P. (1970) J. Cell Biol. 44, Bourque, D. P., Boynton, J. E. & Gillham, N. W. (1971) J. Cell Sci. 8, Chua, N.-H., Blobel, G. & Siekevitz, P. (1973) J. Cell Biol., in press. 30. Friedman, H., Lu, P. & Rich, A. (1971) J. Mol. Biol. 61, Morimoto, T., Blobel, G. & Sabatini, D. D. (1972) J. Cell Biol. 52, Boulter, D., Ellis, R. J. & Yarwood, A. (1972) Biol. Rev. 47, Adelman, M. R., Sabatini, D. D. & Blobel, G. (1973) J. Cell Biol. 56, Falk, H. (1969) J. Cell Biol. 42, Wellburn, F. A. M. & Wellburn, A. R. (1971) J. Cell Sci. 9, Chen, J. L. & Wildman, S. G. (1970) Biochim. Biophys. Acta. 209, Schor, S., Siekevitz, P. & Palade, G. E. (1970) Proc. Nat. Acad. Sci. USA 68,