Post-translational transport into intact chloroplasts

Proc. Natl. Acad. Sci. USA Vol. 75, No. 12, pp. 6110-6114, December 1978 Cell Biology

Post-translational transport into intact chloroplasts of a precursor to the small subunit of ribulose-1,5-bisphosphate carboxylase (higher plants/ Chlamydomonas reinhardtii/cell-free translation/precursor processing/assembly of holoenzyme)

NAM-HAI CHUA AND GREGORY W. SCHMIDT The Rockefeller University, New York, New York 10021

Communicated by Philip Siekevitz, October 2, 1978

ABSTRACT A precursor to the small subunit of ribulose1,5-bisphosphate carboxylase [3-phospho-i-glycerate carboxylyase (dimerizing), EC 4.1.1.39] has been identified among the products of cell-free translation of polyadenylated RNA from spinach and pea. In both cases, the precursor is larger than the mature protein by 4000-5000 daltons. Upon incubation of postribosomal supernatants of the in vitro protein synthesis mixtures with purified intact chloroplasts, the pea and spinach precursors are transported interchangeably into the chloroplasts and processed to the mature size and charge. Moreover, the newly transported small subunits are found to assemble with endogenous large subunits to form the holoenzyme. In contrast, a precursor to the Chlamydomonas reinhardtii small subunit is not taken up by higher plant chloroplasts, indicating the specificity of the transport events. Together, these results demonstrate that the in vitro reconstruction of the post-translational transport of the higher plant precursors is physiologically significant.

Although chloroplasts are capable of protein synthesis (1), many stromal proteins and thylakoid membrane polypeptides are initially synthesized in the cytosol (2). The mechanism of transport of these proteins into the chloroplasts and the events surrounding it are problems of considerable current interest. Prominent among the transported proteins is the small subunit (S) of ribulose-1,5-bisphosphate carboxylase [RuP2Case; 3phospho-D-glycerate carboxy-lyase (dimerizing), EC 4.1.1.39] (3, 4). It has been reported that, in Chlamydomonas reinhardtii, S is synthesized as a larger precursor (pS) in a wheat germ cell-free system primed with poly(A) RNA (5). The precursor is 4000-5000 daltons larger than S and can be processed to the latter after translation by a specific endoproteolytic activity in the algal postribosomal supernatant. Because S is synthesized by free polysomes, it was proposed that pS is taken up by chloroplasts after chain completion-i.e., the transport is a post-translational event (5). However, the inability to isolate intact chloroplasts from C. reinhardtii has precluded the in vitro reconstitution of transport of the algal pS. This paper describes the in vitro synthesis and identification of pS in two higher plants from which intact chloroplasts can be readily isolated (6). We show that pS is transported into intact chloroplasts post-translationally and, either during or immediately after transport, pS is converted to S and the latter is assembled into the holoenzyme. While this work was in progress, Cashmore et al. (7) and Highfield and Ellis (8) reported the in vitro synthesis of a precursor to the small subunit of pea RUP2Case. The latter authors (8) also found that the precursor was taken up and processed to the small subunit by a crude chloroplast preparation in the absence of protein synthesis.

MATERIALS AND METHODS Pea (Pisum sativum var Alaska) seedlings were grown in a greenhouse; spinach (Spinacea oleracea) was obtained from local farms. RNA Extraction and Purification for Poly(A) RNA. Total nucleic acids were extracted from leaf sections of pea or spinach (5) and chromatographed on poly(U)-Sepharose column. Bound RNA was eluted from the column with 90% (vol/vol) formamide and the eluted fraction was designated poly(A) RNA. Poly(A) RNA from C. reinhardtii was purified as described

(5).

In Vitro Protein Synthesis and Immunoprecipitation of Translation Products. Poly(A) RNA was translated in a wheat germ cell-free system (9) containing 20 mM K Hepes (pH 7.5), 130 mM KCI, 2 mM Mg acetate, 2 mM dithiothreitol, 248 ,uM spermine, 150 ,tCi (1 Ci = 3.7 X 1010 becquerels) of [s5S]methionine per ml, 1 mM ATP, 120,uM GTP, 8.4 mM creatine phosphate, 19 amino acids to 24 ,M, 40 ,ug of creatine kinase per ml, and 1 A260 unit of RNA per ml. After incubation at 25°C for 1.5 hr, the translation mixture was centrifuged at 140,000 X gmax for 45 min and the postribosomal supernatant was used for immunoprecipitation. Antibodies to S of C. reinhardtii and of spinach were prepared as described (5). The anti-spinach S IgG fraction was found to crossreact with S of pea RUP2Case. The postribosomal supernatants were adjusted to 0.1I% sodium dodecyl sulfate (NaDodSO4), 1% Triton X-100, 0.3 M NaCl, and 0.1 M Tris-HCI (pH 8.6) and incubated overnight at 26°C with 1 mg of IgG per ml. Formalin-fixed Staphylococcus aureus (Cowan's strain) was then added to adsorb the IgG-antigen complex (9). The S. aureus was washed three times with 0.1% NaDodSO4/10 mM Tris-HCI, pH 8.6/0.3 M NaCl, and the IgG-antigen complex was eluted with 2%

NaDodSO4. Samples were analyzed by gradient polyacrylamide gel electrophoresis in NaDodSO4 (10) and gels were prepared for fluorography (11). Purification of Intact Chloroplasts. Intact chloroplasts were purified from pea and spinach according to Morgenthaler et al. (6) except that Ludox AM was replaced by Percoll and Na2P207 was omitted from the grinding medium and the silica sol gradients. Intact chloroplasts were washed once in 0.33 M sorbitol/50 mM K tris(hydroxymethyl)aminomethane (Tricine), pH 8.4, pelleted in a Sorval SS34 rotor by accelerating to 6000 rpm and then braking, and finally resuspended in the same medium; 70-90% of the chloroplasts were intact, as determined by phase-contrast microscopy. The purified chloroplasts were capable of light-dependent protein synthesis at rates comparable to those reported previously (12-14). Chlorophyll concentrations were determined according to Arnon (15).

The publication costs of this article were defrayed in part by page This article must therefore be hereby marked "adcharge payment. vertisement " in accordance with 18 U. S. C. §1734 solely to indicate this fact.

Abbreviations: NaDodSO4, sodium dodecyl sulfate; RuP2Case, ribulose-1,5-bisphosphate carboxylase; L and S, large and small subunits of RuP2Case, respectively; pS, precursor to S. 6110

Cell Biology: Chua and Schmidt Spinach

Pea

Proc. Natl. Acad. Sci. USA 75 (1978)

C.r.

fractions were considered to have been taken up by the intact

r

1r 3 4 5 66 77 8 9 1100 Xx 110-3 r 0-3 2 4 3

t iX

68 $ 1 -__60 --_52 -- t^ 40 n

.Ur

N.x4

6111

chloroplasts during the initial incubation period. Fractionation of Chloroplast Stromal Protein. To separate

the RuP2Case holoenzyme from other chloroplast stromal proteins, the 30,000 X g supernatant was layered on a 10-30% sucrose density gradient in 0.2 M NaCI/25 mM Tris-HCI, pH 7.5/0.5 mM EDTA/5 mM dithiothreitol and centrifuged at 170,000 x gave for 16-17 hr at 4VC (16). The RuP2Case holoenzyme sedimented as an 18S peak, whereas other stromal enzymes were distributed between the top of the gradient and about 12 S; chloroplast ribosomes were pelleted. Sources of Materials. All reagents used were analytical grade whenever available. [3-SIMethionine (500-1000 Ci/mmol) was obtained from Amersham/Searle; Percoll and poly(U)-Sepharose 4B was from Pharmacia; and the trypsin and a-chymotrypsin were from Worthington.

~~~~~-29

21

FIG. 1. Identification of a precursor to the small subunit of

RuP2Case in spinach, pea, and C. reinhardtii (C.r.). Analysis was by

NaDodSO4/polyacrylamide gel electrophoresis followed by fluorography. Lanes: 1, polypeptides synthesized in the wheat germ system without added mRNA; 2, polypeptides synthesized in the wheat germ system in the presence of poly(A) RNA from spinach; 3, immunoprecipitation of 2 with anti-spinach S IgG fraction; 4, [35S]RuP2Case from spinach; 5, polypeptides synthesized in the wheat germ system in the presence of poly(A) RNA from pea; 6, immunoprecipitation of 5 with anti-spinach S IgG fraction; 7, [35SJRuP2Case from pea; 8, polypeptide synthesized in the wheat germ system in the presence of poly(A) RNA from C. reinhardtii; 9, immunoprecipitation of 8 with anti-C. reinhardtii S IgG fraction; 10, [35S]RuP2Case from C. reinhardtii. Reconstitution of Protein Transport into Intact Chloroplasts In Vitro. For reconstitution of protein transport, a postribosomal supernatant (20,000 cpm/hl) derived from a translation mixture was adjusted to 0.33 M sorbitol and 0.2 M methionine. One volume of this was then mixed with an equal volume of intact chloroplasts (1 mg of chlorophyll per ml). After incubation at 25°C for 1 hr in the dark, the intact chloroplasts were pelleted by centrifugation as described above. The chloroplast pellet was resuspended in 0.33 M sorbitol/50 mM K tris(hydroxymethyl)aminomethane, pH 8.4, and TPCK-trypsin (N-tosylphenylalanine chloromethyl ketone treated) and achymotrypsin were added to final concentrations of 200 ,g/ml to digest any radioactive translation products that were trapped in the chloroplast pellet or bound adventitiously to the chloroplast envelopes. After incubation at 25°C for 30 min, Trasylol and phenylmethylsulfonyl fluoride were added to final concentration of 1000 units/ml and 2 mM, respectively, and the chloroplasts were pelleted by centrifugation. The supernatant was carefully removed and the pellet was resuspended in 1 ml of lysis buffer containing 5 mM Tris-HCI (pH 7.5), 1000 units of Trasylol per ml, and 2 mM phenylmethylsulfonyl fluoride. The suspension was mixed vigorously until all chloroplasts had lysed, as monitored by phase-contrast microscopy. The suspension was then adjusted to 0.2 M NaCl, 25 mM Tris-HCl (pH 7.5), and 0.5 mM EDTA and centrifuged at 30,000 X g for 10 min. The 30,000 X g supernatant consists of chloroplast stromal proteins whereas the pellet consists mainly of thiylakoid membranes. Any radioactive polypeptides present in these two

RESULTS

Addition of spinach poly(A) RNA to a wheat germ cell-free system stimulated the synthesis in vitro of a large number of polypeptides (Fig. 1, lanes 1 and 2). Of these, only the polypeptide at Mr 18,000 was precipitated specifically by antibodies raised against spinach S (Fig. 1, lane 3) but not by a preimmune IgG fraction (data not shown). Because the Mr 18,000 polypeptide was immunochemically related to spinach S (Mr 14,000) (Fig. 1, lane 4) but larger than the latter by 4000 daltons, it was identified'as a precursor (pS). By using similar techniques, a precursor (Fig. 1, lane 6) to S of pea RuP2Case (Fig. 1, lane 7) was also identified among the polypeptides translated from pea poly(A) RNA in the cell-free system (Fig. 1, lane 5). Com1

2

3

4

5

6

pS-

S-

m _ 1_

FIG. 2. Reconstitution of protein transport into intact chloroplasts in vitro does not require concomitant protein synthesis. In vitro polypeptide products of pea or spinach poly(A) RNA were incubated with intact chloroplasts from the same plant. Analysis was by NaDodSO4polyacrylamide gel electrophoresis and fluorography. Lanes: 1 and 2, 30,000 X g supernatant and pellet, respectively, of pea chloroplasts (30 jg of chlorophyll); 3, [35S]RuP2Case of pea; 4, [35SlRuP2Case of spinach; 5 and 6, 30,000 X g supernatant and pellet, respectively, of spinach chloroplasts (30 ,ug of chlorophyll).

Cell Biology: Chua and Schmidt

6112

Proc. Nati. Acad. Sci. USA 75 (1978) i 1

I

i11

Li

1

i C"'.

i

i

i

i

-L

i.

i

i

i

i-L

-L

'1.5

0 N

a) 0 cr_

* 1.0

0 0

.0 0.5

5

10

15

-s

20

Fraction No. FIG. 3. Fractionation of pea chloroplast soluble proteins by sucrose density gradient centrifugation and analysis of the gradient fractions. (a) Translation products of pea poly(A) RNA were incubated with intact pea chloroplasts, and then the 30,000 X g supernatant from the lysed chloroplasts (500 ,gg of chlorophyll) were layered onto a 10-30% sucrose gradient and centrifuged. The gradient was fractionated into 25 fractions with an ISCO gradient fractionator and the absorbance at 280 nm was monitored. (b) Radioactivity profile of the gradient fractions. Proteins in each gradient fraction were precipitated by adjusting to 10% trichloroacetic acid. Analysis was by NaDodSOSpolyacrylamide gel electrophoresis followed by fluorography. Figure shows only fluorograph of fractions 1-20.

of the precursors of spinach, pea, and C. reinhardtis revealed that they are larger than their corresponding mature S by about 4000-5000 daltons (Fig. 1). In all cases, S was not immunoprecipitated from the in vitro products, suggesting that the wheat germ extract lacks an enzyme that converts pS to S. The finding of pS in spinach and pea supports and extends the previous observation with C. reinhardta that S is synthesized as a larger precursor (5). To determine whether pS could be transported into intact chloroplasts after its synthesis, a postribosomal supernatant was prepared from a translation mixture containing polypeptide products of pea poly(A) RNA (Fig. 1, lane 5). The supernatant was incubated with purified intact pea chloroplasts in the dark to prevent chloroplast protein synthesis. Fig. 2 shows that after 1 hr at 250C a major radioactive polypeptide (Mr 14,000) was transported into the chloroplasts and was recovered in the 30,000 X g supernatant (lane 1) but not in the 30,000 X g pellet (lane 2). Based on its molecular weight, it was tentatively identified as S of pea RuP2gCase. Because pS but not S was present among the in vitro polypeptide products (Fig. 1, lane 6), these results strongly suggest that pS was taken up by pea chloroplasts and immediately processed to S. It should be emphasized that in these experiments pS was not found in the chloroplast soluble (Fig. 2, lane 1) or membrane fraction (Fig. 2, lane 2). Reconstitution of pS transport into chloroplasts in vitro was also achieved with homologous components derived from spinach (Fig. 2, lane 5 and 6). Because S is normally associated with L to form the RuP2Case holoenzyme in vivo, we considered the transport of pS into chloroplasts to be physiologically significant only if the newly transported S were assembled into the holoenzyme. To this end, we exploited the sedimentation of the holoenzyme at 18 S (17). Therefore, after uptake of pS by intact chloroplasts, the 30,000 X g supernatant was fractionated on a 10-0% sucrose gradient which resolved the supernatant proteins into two majr peaks, designated as the top fraction and 18 S (Fig. 3a). Analysis of the gradient fractions by NaDodSO4/polyacrylamide gel electrophoresis showed that the 18S peak consisted almost exclusively of the RuP2Case holoenzyme (data not shown). Fluorography parison

of the gel revealed that, although a small amount of the newly transported S was present in the top fraction, most of it was recovered in the 18S region (Fig. 3b). Furthermore, the radioactivity profile of S coincided with the absorbance profile of the 18S peak, indicating that most of the newly transported S was in the holoenzyme. Radioactive polypeptides other than S were visible in gradient fractions 2-7. These unidentified polypeptides are minor in vitro translation products that also were transported into the chloroplasts. In another experiment, gradient fractions corresponding to the top fraction and the 18S peak (Fig. Sa) were pooled and analyzed separately. Gel bands corresponding to radioactive S were excised and the amount of radioactivity was measured. It was found that approximately 80%o of S was in the RuP2Case peak. Further evidence for assembly was obtained from electrophoresis of the 18S peak in polyacrylamide gels without NaDodSO4. Under nondenaturing conditions, the RuP2Case holoenzyme in the 18S peak migrated as a single band near the top of the 4-7% gradient gel and all the radioactivity was confined to this band (Fig. 4). These results, together with those presented in Fig. 3, demonstrate conclusively that the newly transported S found in the 18S peak was indeed assembled into the holoenzyme. The major radioactive polypeptide taken up by the intact pea chloroplast in vitro was tentatively identified as S on the basis of its electrophoretic mobility and its association with the pea S F

FIG. 4. Nondenaturing gel analysis of the 18S peak from intact pea chloroplasts after reconstitution of protein transport in vitro. An

aliquot of the 18S peak from Fig. 3a was analyzed by electrophoresis in a 4-7% gradient gel according to Laemmli (18) but without NaDodSO4. S, Coomassie blue-staining pattern; F, fluorograph of S.


Proc. Natl. Acad. Sci. USA 75 (1978)

(a) 2

(b) tW

t

t

t

FIG. 5. Two-dimensional gel analysis of the 18S peak from pea chloroplasts after reconstitution of protein transport in vitro. First dimension, isoelectric focusing (19); second dimension, NaDodSO4/ polyacrylamide gel electrophoresis (11). The gels were processed for fluorography. (a) [35S]RuP2Caselfrom Ipea; (b) 18S peakl from pea 30,000 X g chloroplasts supernatant (Fig. 3a). The pH gradient from right to left was 4.5 to 8.3. Arrows indicate the two major isoelectric species of S with pIs of 6.0 and 6.6.

RUP2Case holoenzyme. Definitive identification was provided by two-dimensional gel analysis according to O'Farrell (19). This polypeptide in the 18S fraction could be resolved into two isoelectric forms of identical molecular weight (Fig. 5b), and these two forms had the same pIs and molecular weight as the two major isoelectric species of pea RuP2Case S (Fig. 5a). Mixing experiments did not resolve any difference between the isoelectric species of S obtained in vitro and in vivo (data not shown). 1

S-

2

mo

3

4

5

6

-

.

-

7

8

9

10

FIG. 6. Homologous and heterologous reconstitution of protein transport into intact pea and spinach chloroplasts in vitro. Lanes 1-4: intact pea chloroplasts were incubated with translation products of poly(A) RNA from pea (lanes 1 and 2) or spinach (lanes 3 and 4). Lanes 5-10: intact spinach chloroplasts were incubated with translation products of poly(A) RNA from spinach (lanes 5 and 6), pea (lanes 7 and 8), or C. reinhardtii (lanes 9 and 10). The top fractions (lanes 1, 3, 5, 7, and 9) and the 18S peak (lanes 2, 4, 6, 8, and 10) were obtained as described in Fig. 3. Analysis was by NaDodSO4polyacrylamide gel electrophoresis followed by fluorography. Only the gel region containing S is shown.

6113

Because L is synthesized inside the chloroplast (12-14), the uptake and processing of pS and the assembly of newly transported'S may be enhanced when chloroplast protein synthesis is active. Furthermore, protein transport might be stimulated by chloroplast ATP production. Therefore, intact pea chloroplasts were incubated with polypeptide products of pea poly(A) RNA in the light with and without chloramphenicol (100 ,ug/ml). Under these conditions, the amounts of radioactive S in the top fraction and 18S peak were nearly the same as those of the dark control (data not shown). Therefore, chloroplast ATP production and active chloroplast protein synthesis have no apparent effect on the transport of pS, its processing to S, and the assembly of the latter into the holoenzyme. The finding of pS in spinach and pea raises the possibility of reconstituting transport of pS into chloroplasts by using heterologous components. It also was interesting to know whether pS of C. reinhardtii, a unicellular alga, can be taken up by higher plant chloroplasts. Consequently, pea chloroplasts were incubated with in vitro polypeptide products of poly(A) RNA of spinach or C. reinhardtii under the same conditions given in Fig. 2. Spinach pS was taken up by pea chloroplasts and processed to S, and the S was assembled into the holoenzyme, presumably with the L of pea RuP2Case (Fig. 6, lanes 3 and 4). Similar results were obtained with pea pS and spinach chloroplasts (Fig. 6, lanes 7 and 8). In contrast, incubation of spinach chloroplasts with polypeptides synthesized in vitro by C. reinhardtii poly(A) RNA did not result in the appearance of any radioactive polypeptides in the top fraction (Fig. 6, lane 9), the 18S peak (Fig. 6, lane 10), or the 30,000 X g pellet (data not shown). Similar results were obtained with pea chloroplasts (data not shown). Thus, pea and spinach chloroplasts are unable to transport pS or any other C. reinhardtii polypeptides synthesized in vitro. The latter results provide unequivocal evidence that the proteolytic treatment used effectively removed polypeptides that were not transported into the chloroplasts in vitro.

DISCUSSION In the present study we confirm previous reports (5, 7,8) that pS is a major product of in vitro protein synthesis driven by poly(A) RNA from algae and higher plants. It has been reported that S is synthesized on free ribosomes in C. reinhardtii (5) as well as in pea (7, 20). Furthermore, the algal pS can be converted to S by a soluble protease after completed synthesis of the precursor in vitro (5). These observations have led to the proposal (5) that the transport of pS into chloroplasts is a posttranslational event, analogous to the entry of certain plant and microbial toxins through the plasma membranes of sensitive mamalian cells (21). This mode of transport contrasts with the cotranslational transfer of precursors of secretory proteins across microsomal membranes (22, 23). Highfield and Ellis (8) have shown recently that pea pS could be processed to S by a crude chloroplast preparation in the absence of protein synthesis. Because the newly formed S was protected from proteolysis, it was presumably transported into the chloroplasts. We consider that the transport of pS into intact chloroplasts can be unambiguously demonstrated only if the following three conditions are met. (i) The chloroplast preparation must contain a high proportion of intact plastids and be free of cytoplasmic proteins or other membrane-bound organelles. Such contaminants may contain spurious proteolytic activities, which would complicate interpretation of results. (ii) pS must be shown to be correctly processed. (iii) The newly transported S must be shown to assemble with L to form the holoenzyme. Assembly is the ultimate criterion that ensures that transport into the chloroplast stroma indeed has occurred in a physiological

6114


fashion. Otherwise, it could be argued that S is lodged in the chloroplast envelope or trapped in the intermembrane space. All the above criteria were satisfied in the reconstitution experiments reported here. Moreover, using heterologous components, we found that pS of pea can be transported into spinach chloroplasts and vice versa. In contrast, C. reinhardtll pS is not taken up into chloroplasts of either higher plant. Therefore, chloroplast envelopes of higher plants are impermeable to polypeptides of M. 8000-100,000 which are present in the in vitro translation products of the algal poly(A) RNA. The transport of pS into chloroplasts in pea and spinach suggests the presence of specific chloroplast envelope receptor. Because incubation of intact chloroplasts with proteases compIetely abolishes subsequent pS transport (data not shown), the envelope receptor is likely to be a protein. The processing enzyme that converts pS to S remains associated with intact chloroplasts after their purification on a silica sol gradient. Therefore, this enzyme must be localized in the chloroplasts. It has been suggested that the processing enzyme of higher plaints is bound to the chloroplast envelope (24) but direct evidence is still lacking. Our results do not provide direct evidence that the precursor form of S is required for transport. Nevertheless, we are confident that the in vitro experiments described here accurately reconstruct the molecular events that occur in vivo. Thus, it is now possible to examine in detail the factors that regulate the synthesis, transport, and assembly of RuP2Case subunits. . This work We thank Ms. Sally Liang for expert technical a was supported in part by NIH Grants GM-21060 and GM-25114. N.-H. C. is a recipient of National Institutes of Health Research Career Development Award 1KQ4 GM-00223. G.W.S. is supported by National Institutes of Health National Research Service Award Postdoctoral Fellowship 4F32 GM-05390. 1. Ellis, R. J. (1977) Biochim. Biophys. Acta 463,185-191.

Proc. Natl. Acad. Sd. USA 75 (1978) 2. Gillham, N. W., Boynton, J. E. & Chua, N.-H. (1978) Cufr. Top. Bioenerg. 8,209-259. 3. Gray, T. C. & Kekwick, R. G. 0. (1974) Eur. J. Biochem. 44, 491-500. 4. Roy, H., Patterson, R. & Jagendorf, A. T. (1976) Arch. Biochem.

Biophys. 171,64-74. 5. Dobberstein, B., Blobel, G. & Chua, N.-H. (1977) Proc. Nat!. Acad. Sd. USA 74,1082-1085. 6. Morgenthaler, J.-J., Marsden, M. P. F. & Price, C. A. (1975) Arch. Biochem. Biophys. 168,289-01. 7. Cashmore, A. R., Broadhurst, M. K. & Gray, R. E. (1978) Proc. Nat!. Acad. Sci. USA 75, 655-659. 8. Highfield, P. E. & Ellis, R. J. (1978) Nature (London) 271, 420-424. 9. Kessler, S. W. (1976) J. Immunol. 117,1482-1490. 10. Chua, N.-H. & Bennoun, P. (1975) Proc. Nat. Acad.Sd. SUSA 72,2175-2179. 11. Bonner, W. M. & Laskey, R. A. (1974) Eur. J. Biochem. 46, 83-88. 12. Morgenthaler, J.-J. & Mendiola-Morgenthaler, L. (1976) Arch. Biochem. Biophys. 172, 51-58. 13. Blair, G. E. & Ellis, R. J. (1973) Biochim. Bioph*;. Acta 319, 223-234. 14. Bottomley, W., Spencer, D. & Whitfield, P.t'. (1974) Arch. Biochemn. Biophys. 164,106-117. 15. Arnon, D. I. (1949) Plant Physo. 065 4754)0. 16. Iwani, V., Chua, N.-H. & Siekevitz, P. (19%X) Biochim. Biophys. Acta 358,329340. 17. Kawashima, N. & Wildman, S. G. (1970) -Annu. Rev. Plant Physiol. 21,325-38. 18. Laemmli, U. K. (1970) Nature (London) New Biol. 227,680685. 19. O'Farrell, P. H. (1975)1. Biol. Chem. 250,4007-4021. 20. Roy, H., Terenna, B. & Cheong, L. C. (1977) Plant Physaol. 60, 532-57. 21. Pappenheimer, A. M., Jr. (1977) Annu. Rev. Biochem. 46,6994. 22. Blobel, G. & Dobberstein, B. (1975) J. Cell Biol. 67,835-851. 23. Blobel, G. & Dobberstein, B. (1975) J. Cell B1o1. 67,852-862. 24. Roman, R., Brooker, J. D., Seal, S. N. & Marcus, A. (1976) Nature (London) 260,359-360.