dicarboxylate transport - Europe PMC

5 downloads 0 Views 1MB Size Report
Nov 8, 1982 - dicarboxylate transport. (photorespiration/malate shuttle/glutamine transport) ... transporter, conducting malate-aspartate exchanges for the in-.
Proc. NatL Acad. Sci. USA

Vol. 80, pp. 1290-1294, March 1983 Botany

An Arabidopsis thaliana mutant defective in chloroplast dicarboxylate transport (photorespiration/malate shuttle/glutamine transport)

S. C. SOMERVILLE*t AND W. L. OGREN*t *Department of Agronomy, University of Illinois, Urbana, Illinois 61801; and tU. S. Department of Agriculture, Agricultural Research Service, Urbana, Illinois 61801 Communicated by Harry Beevers, November 8, 1982

ABSTRACT Reactions of the photorespiratory pathway of C3 plants are found in three subcellular organelles. Transport processes are, therefore, particularly important for maintaining the uninterrupted flow of carbon through this pathway. We describe here the isolation and characterization of a photorespiratory mutant of Arabidopsis thaliana defective in chloroplast dicarboxylate transport. Genetic analysis indicates the defect is due to a simple, recessive, nuclear mutation. Glutamine and inorganic phosphate transport are unaffected by the mutation. Thus, in contrast to previous reports for pea and spinach, glutamine uptake by Arabidopsis chloroplasts is mediated by a transporter distinct from the dicarboxylate transporter. Both the inviability and the disruption of amino-group metabolism of the mutant under photorespiratory conditions suggest that the primary function of the dicarboxylate transporter in vivo is the transfer of 2-oxoglutarate and glutamate across the chloroplast envelope in conjunction with photorespiratory nitrogen metabolism. The role commonly ascribed to this transporter, conducting malate-aspartate exchanges for the indirect export of reducing equivalents from the chloroplast, appears to be a minor one.

The constituent reactions of the photorespiratory pathway of higher plants occur in three organelles, the chloroplast, the mitochondrion, and the peroxisome. Thus, transport processes must intervene at several steps of the pathway (Fig. 1) (1, 2). To the extent that the transport of substrates, products, or cofactors is limiting, these transport processes may be expected to exert a strong regulatory influence on photorespiratory metabolism. The photorespiratory pathway is initiated by the oxygenation of ribulose bisphosphate by the bifunctional enzyme ribulosebisphosphate carboxylase/oxygenase (EC 4.1.1.39) (3, 4). 02 and CO2 act as competitive substrates for this enzyme, and the degree to which carbon is diverted from the Calvin cycle to the photorespiratory pathway is a function of the ratio of these two gases in the atmosphere (5). Experimentally, the flux of carbon through the pathway can be suppressed without adverse effect by placing plants in an atmosphere enriched in CO2 or enhanced by transferring plants to an atmosphere enriched in 02(6). Threequarters of the carbon entering the pathway is returned to the pool of Calvin-cycle intermediates as phosphoglycerate. The remaining carbon is lost as CO2 at the glycine decarboxylase step in the mitochondrion. Tightly integrated with the photorespiratory carbon cycle is the photorespiratory nitrogen cycle in which ammonia released during glycine deamination is refixed by the sequential action of glutamine synthetase and glutamate synthase (Fig. 1) (7). The resultant glutamate supports both photorespiratory ammonia refixation (8) and glyoxylate amination (9). In the absence of adequate glutamate pools, glyoxylate is oxidized nonenzymat-

ically to CO2 at rates that significantly reduce net CO2 assimilation (9), and ammonia accumulates to toxic levels (8). Because glutamate is synthesized in the chloroplast (7) and consumed in the peroxisome, the transfer of glutamate and 2-oxoglutarate between these two organelles is a necessary component ofphotorespiratory nitrogen metabolism. Transport systems operating at the peroxisome-bounding membrane have not been characterized. However, studies with isolated chloroplasts have revealed the presence of a transporter, designated the dicarboxylate transporter, which catalyzes the counter-exchange of several dicarboxylic acids across the chloroplast inner membrane (10). These compounds include malate, 2-oxoglutarate, aspartate, and glutamate (10). Glutamine is also a reported substrate for this carrier (11, 12). Therefore, the chloroplast dicarboxylate transporter is implicated as an important component of the photorespiratory pathway. Because this transporter is also capable of effecting malate-aspartate exchanges, it has been ascribed the role of mediating the indirect export of reducing equivalents to the cytoplasm (13, 14). The isolation and characterization of a photorespiratory mutant defective in chloroplast dicarboxylate transport is reported here. Biochemical and physiological analyses of the mutant have proven useful in determining the specificity of this transporter and in ascertaining its primary in vivo role. MATERIALS AND METHODS Plant Material and Culture. Both the mutant line CS156, the subject of this study, and the previously described line CS113, a glutamate synthase-deficient mutant (8), were recovered in a screen for mutants of Arabidopsis thaliana (L.) Heynh. (race Columbia) with defects in photorespiratory metabolism (6). The basis of the mutant selection procedure is that strains with defects in the photorespiratory pathway cannot survive at atmospheric levels of CO2 and 02 but grow normally at 1% CO2, when photorespiration is suppressed. Plants were grown according to described methods and conditions (15). For most of this study, a line descended from a backcross of CS156 to the wild type was used. Experiments were conducted with plants at the rosette stage of development (34 wk from seeding). Procedures for making genetic crosses and measuring gas exchange have been described (15). Labeling Studies. Plants were labeled with 14CO2 for 10 min, and the distribution of label among products was determined by ion-exchange and thin-layer chromatography (15-17). Glutamate Synthase Assays. Glutamate synthase was assayed in crude extracts of leaf material after centrifugation at Abbreviation: Chl, chlorophyll. tPresent address: MSU-DOE Plant Research Laboratory, Michigan State Univ., East Lansing, MI 48824.

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

1290

Proc. Natl. Acad. Sci. USA 80 (1983)

Botany: Somerville and Ogren Krebs

Cycy HLOROPLAST

CH TP

APG ADP

L3-dPGA

I,-diPGA

TP

C02

PGA

Sucros P-Glycolate

I___________________ I__________

Mdote

tMabte

OAA

OAA -2064

Asp

'4 -Asp

I Glycerote GlyWcote OHPyruvate Qlyoxyote

>342: Sor

PEROXISOE T I

Ser

A

hb amg,

Gly

I x^A

* Gly NH3

T

MITOCHONDRION NAM

NOD

FIG. 1. Schematic presentation of thephotorespiratory carbon and nitrogen cycles. C1-THF, N5,N'0-methylenetetrahydrofolate; DT, dicarboxylate transporter; OAA, oxaloacetate; 20G, 2-oxoglutarate; PT, phosphate translocator; PGA, phosphoglycerate; RuBP, ribulose bisphosphate; THF, tetrahydrofolate; TP, triose phosphate.

30,000 x g and desalting by Sephadex G-25 column chromatography (8). Protein was determined by a dye-binding assay (18) with bovine serum albumin as the standard. Ammonia and Amino Acid Determinations. Plants were equilibrated. in darkness for 25 min and then illuminated (300 microeinsteins m-2se&- of photosynthetically active radiation) for various lengths of time in an open gas exchange system maintained at 25TC and 70% relative humidity. The photorespiratory gas regime was nitrogen containing 357 1.l of CO2 liter' and 49% 02. Methods for quantitating ammonia and amino acid levels in leaf tissue have been described (8). Amino acid values were normalized by using the amount of phenylalanine per mg of chlorophyll (Chl) as a basis to correct for losses that occurred during the preparation of some samples. Chloroplast Isolation and Transport Assays. Chloroplasts were prepared from Percoll gradient-purified protoplasts (19) and used immediately. The intactness of the chloroplast preparations was determined by the ferricyanide test (20). Oxoglutarate + glutamine-dependent 02 evolution by isolated chioroplasts was measured in a Hansatech 02 electrode at 25°C at a light intensity of 1,000 microeinsteins m 2sec- of photosynthetically active radiation at the surface of the cuvette

(21). Chloroplasts (20-40 ,ug of Chl) were added to the standard [300 mM sorbitol/28 mM HepesrKOH, pH 7.6/ 10 mM NaHCO3/2.5 mM EDTA/0.15 mM potassium phosphate/0. 1%. fatty acid-free bovine serum albumin/300 units of catalase per ml (19)]. D,L-Glyceraldehyde was added to 6-10 assay medium

m.M to inhibit C02-dependent 02 evolution (22). The silicone oil layer filter centrifugation technique was used to measure the transport of radiolabeled compounds into freshly isolated, intact chloroplasts (23). The standard assay medium contained, in addition to the standard components, 10-30 tig of Chl and 1-3 ,uCi (3.7-11.1 x 104 Bq) of 3H20. Glutamine transport was determined at pH 7.9. The silicone oil layer consisted of AR200/AR20 60:40 (wt/wt) (Wacker Chemie, SWS Silicones, Adrian, MI). Chloroplast volumes were determined with [14C]sorhitol in parallel experiments under the same conditions as the. transport assays (23). The average sorbitol-impermeable space was 38 ,ul/mg of Chl and 44 p.l/mg of Chl for wild-type and mutant chloroplasts, respectively: Transport assays, commonly 3-, 5-, or 10-sec duration, were performed in the dark at 50C. For some experiments, chloroplasts were preloaded with the compound to be assayed for uptake by adding a small volume of stock solution to freshly prepared chloroplasts to give a final concentration of 20 mM. After 15 min on ice in the dark, the chloroplasts were collected by centrifugation (270 X g at 40C for 40 sec) and resuspended in medium lacking the compound. Apparent Km and Vm.x values were estimated from Scatchard plots of the data presented in the text. Chl was determined spectrophotometrically in ethanol (24).

J L

,NWCOH THF

1291

RESULTS Mutant Isolation and Genetic Analysis. In an atmosphere that suppressed photorespiration (1% C02/99% air), the mutant line CS156 was capable of normal growth and development. However, in standard atmospheric conditions (N2 containing 0.03% CO2 and 21% 02), the mutant became yellow and lost vigor within 3-4 days. The F1 plants from a CS156 x wild-type cross were healthy in standard atmospheres, suggesting the mutant line carried a recessive, nuclear mutation. In a derivative F2 population, 196 plants exhibited the wild-type phenotype and 55 were yellow after 4 days in a normal atmosphere. Thus, the mutation in line CS 156 responsible for the growth requirement for high CO2 was inherited as a simple, recessive, nuclear mutation (X2 = 1.276; P > 0.25). The locus defined by this mutant was designated dct (dicarboxylate transport). The F1 plants from a cross of CS 156 with CS 113, a glutamate synthase-deficient line (8), were normal in appearance and capable of sustained growth in a normal gas regime. Because the mutation in CS156 complemented that in CS 113, the dct locus was genetically distinguished from the gluS locus. Gas Exchange Analyses. CO2 fixation was measured on individual plants in nonphotorespiratory (Fig. 2A) and photorespiratory (Fig. 2B) gas regimes. The photosynthesis rate of the mutant exceeded that of the wild type in an atmosphere of low 02 concentration (Fig. 2A). In contrast, photosynthesis was impaired in the mutant in atmospheres that promoted photorespiration. In the photorespiratory gas regime of nitrogen containing.49% 02 and 357 A.l of C02 liter', the final rate of CO2 -fixation in CS156.was reduced to 27% of the rate in wild type (Fig. 2B). CO2 evolution into C02-free 50% 02/50% N2, a measure of photorespiration, also was reduced in mutant plants (data not presented). Thus, the lesion in mutant CS156 disrupted photorespiratory metabolism specifically and had no detectable effect on photosynthesis and growth in nonphotorespiratory environments. Metabolite Distribution. A defect in photorespiratory metabolism was evident in the mutant from the altered distribution pattern of metabolites labeled with '4Co2 for 10 min in a photorespiratory gas regime (Table 1). Notably, the accumulation of 14C label in the basic fraction was reduced and that found in the acid-i fraction was increased in mutant plants compared to

1292

Botany: Somerville and Ogren

Proc. Natl. Acad. Sci.-, USA 80 (1983) In consideration of the difficulties associated with equating the distribution of radioactive label. and'mass flow of metabolites, we undertook a quantitative analysis of amino acid pools in the leaf under photorespiratory conditions. The results of this experiment confirmed that the mutant harbored a defect in glutamate metabolism (Fig. 3)'. Upon illumination, glutamate levels in the mutant declined, suggesting that utilization of this amino acid. greatly exceeded its synthesis (Fig. 3A)'. Glutamine levels in the mutant declined slightly and then stabilized at a relatively high level (Fig. 3B). Lack of depletion of glutamine pools and accumulation of '4C label in 2-oxoglutarate (Table 1) implied that the mutant was not able to convert these two compounds to glutamate. Ammonia accumulated to abnormally high levels in the mutant plants (Fig. 3C). A similar result was observed in the gluS. mutant, CS113, which. was unable to refix photorespiratory ammonia by the glutamine synthetase/glutamate synthase reactions (8). The-metabolism of glycine (Fig. 3D) and that of serine (Fig. 3E), both intermediates of the photorespiratory pathway, were also abnormal in the mutant line. These are considered secondary effects resulting from the disruption of photorespiratory metabolism.

30 20 10 0

cq

C, 0

bo

;

20

=

10

B

0~

e--l'

¢I

------------0

8.0 0

20

40

60

A

0

80

Time, min 4.0

2. NetCO2fixationbywild!typeandmutantArabidopsisplants under nonphotorespiratory (A) and photorespiratory (B) conditions. The gas regime in A was nitrogen containing 352 gl of C02 liter-' and 2% 02 and inB was nitrogen containing 357 AI of C02 liter-' and 49% 02. -* Response of wild type; ---, response of mutant CS156;-; dark conditions; m, light conditions. The response shown represents the average of two experiments. FIG.

wild-type plants. Labeling of all major constituents of the basic fraction was depressed. The increase of 14C label in the acid-i fraction was primarily due to an increased labeling of 2-oxoglutarate. In these respects the labeling pattern of the dct mutant closely resembled that of a glutamate synthase-deficient line, CS113 (Table 1), implicating a defect in glutamate synthesis as the cause for the high CO2 requirement of the dct mutant.

00 -

0

8.0

B.

-o

4.0 0

o

0

a0.2

0

0

0

Table 1. Percentage distribution of the products of 14c0 assimilation* Distribution (%) by strain CS113 Wild type CS156 Fraction 16.3 17.2 41.0 Basic 0.5 0.7 1.5 Glutamate 10.0 9.0 21.2 Glycine 2.2 3.6 5.8 Serine 22.6 29.8 6.3 Acid-i 2.4 2.6 3.6 Malate 0 0 1.0, Glycerate 0. 8.7 14.0 2-Oxoglutarate 33.5 27.3 24.0 Acid-2 13.1 11.3 8.9 Acid-3 10.8 11.0 14.0 Neutral 6.2 5.1 4.6 Insoluble

*

102.5 101.7 100.8 Recovery Values given are the percentage of total 14C incorporated into leaves. Plants were labeled with 14C02 at 5-15 min from the onset of, illumination. The photorespiratory gas regime was as described, andthe conditions during labeling were 25TC, 70% relative humidity, and 300 microeinsteins m 2sec' of photosynthetically active radiation. Each value is the mean of three determinations.

0/-

Q 4a)

A~~D

D

2 2.0. 1.0 0

j

E

0

1.0

0 10

20 30'

Time, min.

FIG. 3. Amino acid and ammonia levels in leaves of wild-type (o) and mutant CS156 (o) Arabidopsis plants under photorespiratory conditions. (A) Glutamate. (B) Glutamine. (C) Ammonia. (D) Glycine. (E) Serine.

Proc. Natl. Acad. Sci. USA 80 (1983)

Botany: Somerville-and Ogren Enzyme Analyses. Crude leaf extracts of mutant plants showed 63% of wild-type levels of ferredoxin-dependent glutamate synthase activity (data not presented). However, previous analyses of gluS mutants showed that F1 (wild-type x gluS) plants with 50% of wild-type levels of glutamate synthase activity were phenotypically normal (8). Thus, the observed reduction of enzyme

activity

in CS 156

was

not

sufficient to account for the

measured in isolated, intact chloroplasts as 2-oxoglutarate + glutamine-dependent 02 evolution (21). At low external concentrations of 2-oxoglutarate, no 2-oxoglutarate + glutaminedependent 02 evolution by mutant chloroplasts was detected (Table 2). However, when the 2-oxoglutarate concentration was raised to 100 mM, chloroplasts from both the mutant and wild type exhibited glutamate synthase activity. Thus, it appeared that the enzyme was present and active in chloroplasts from the dct mutant, confirming conclusions from in vitro enzyme assays. On the basis of these experiments, it was considered probable that chloroplasts from the mutant were substantially less permeable to 2-oxoglutarate than were wild-type chloroplasts. The uptake of 2-oxoglutarate and several other compounds by isolated chloroplasts was measured directly by using the silicone oil filter centrifugation technique (23). For wild-type chloroplasts, the uptake of four dicarboxylates, glutamine, and inorganic phosphate became saturated at high external substrate concentrations, a characteristic of facilitated transport (Fig. 4). To better characterize the transporters in wild-type Arabidopsis chloroplasts, kinetic parameters describing the concentration dependence of transport rates for these compounds were determined (Table 3). Apparent Km values were slightly higher than those reported for spinach (10, 25) and pea (12). Values of Vma, for the four dicarboxylates and glutamine were considerably higher (10, 12). For chloroplasts from the mutant, the transport of malate, 2-oxoglutarate, aspartate, and glutamate was severely reduced. However, the uptake of glutamine was indistinguishable from the wild type, suggesting that this amino acid was transported by a carrier distinct from the dicarboxylate transporter (Fig. 4E). The -uptake of inorganic phosphate was unaffected by the lesion at the dct locus in the mutant (Fig. 4F). Similar results were obtained at 250C. Thus, it is unlikely that the loss of dicarboxylate transport activity was an artifact of lowtemperature effects on the mutant chloroplast envelope. Also, the normal functioning of two carriers in the envelope of the mutant indicated that the lesion was specific for dicarboxylate transport and did notaffect thegeneral architecture of the chlo.roplast envelope. The chloroplast dicarboxylate transporter is reported to operate by a counterexchange mechanism (10). For some experTable 2. Glutamine + 2-oxoglutarate-dependent evolution by isolated chloroplasts 02 evolution, Amol/mg of Chl per hr Substrate, mM CS156 type Glutamine 2-Oxoglutarate strain mutant 5 3 4.77 0 5 100 4.00 10.60 The percentage chloroplast intactness was 88% for wild type and 76% for CS156. Wild

/-

B

C

120 _.4

0H

high

CO2 growth requirement in the mutant. Also, as noted above, the mutation in CS156 was not allelic to the gluS mutation in CS113. Transport Assays. Physiological studies of metabolite pools strongly indicated glutamate metabolism was disrupted in the dct mutant, but in vitro enzyme assays did not verify this conclusion. To examine this inconsistency, glutamate synthase was

02

A

1293

80 40 0-

/00 S-4

0

T.

-a)

1.0

2.0

3.0

2.0 0 4.0 8.0 0 0.4 External concentration, mM

0.8

1.2

1.0

!8 0

0

1.0

2.0 0

F

E

0

fI

1.0

2.0 D

o'e0~

FIG. 4. The uptake of several compounds into the sorbitol-impermeable space of chloroplasts from wild-type (e) and mutant CS156 (0) plants. All experiments were conducted at 5YC. Chloroplasts used in these experiments were >94% intact. The compounds assayed for uptake were 2-oxoglutarate (A), aspartate (B), glutamate (C), malate (D), glutamine (E), and inorganic phosphate (F). For experiments presented in A, B, C, and D, chloroplasts were preloaded with the compound to be assayed for transport.

iments, chloroplasts were incubated in high concentrations of substrate prior to measuring uptake to ensure that the internal

supply of dicarboxylate did not limit the uptake of exogenously supplied compounds by the counterexchange transporter. Preloading wild-type chloroplasts with malate, 2-oxoglutarate, and glutamate stimulated malate, 2-oxoglutarate, and glutamate uptake, respectively, =2-fold. This enhancement of transport rate cannot be attributed to significant carryover of substrate from the preloading to the assay step. The amount of carryover was calculated to be 31-90 nmol (equivalent to a 0.010 to 0.028 mM increase in the substrate concentration in the assay mixture) by comparing the rate of glutamate or aspartate uptake at low external concentrations not stimulated by preloading with the rate expected when a linear relationship between transport rate and concentration is assumed. In addition to enhancing the rate of uptake and, hence, the observed Vmax, preloading led to a reduction by 0.5 mM of the apparent Km of 2-oxoglutarate and malate for wild-type chloroplasts. Uptake of 2-oxoglutarate and glutamate by mutant chloroplasts was unaffected by the preloading step. However, in the absence of preloading, malate uptake by the mutant could not be detected. A low rate of aspartate uptake, which did not saturate at 2.0 mM, was observed when the preloading step was omitted. With preloading, a reduced level of apparently facilitated aspartate transport was measured (Fig. 4B). Table 3. Kinetic constants of uptake of several compounds by wild-type chloroplasts VmLax, PMo0/mg Substrate Km x 103 of Chl per hr Malate 1.3 130 2-Oxoglutarate 0.5 176 111 Aspartate 0.5 Glutamate 2.7 150 Glutamine 2.4 42 Inorganic phosphate 0.5 90

1294

Botany: Somerville and Ogren

DISCUSSION Mutant line CS156 carries a recessive nuclear mutation responsible for the specific loss of chloroplast dicarboxylate transport. activity. Thus, although.the dct mutant showed substantial in vitro glutamate synthase activity, in vivo analyses indicated that the enzyme was essentially inoperative because of substrate limitation (Table 2 and Fig. 3). The lethality of the dct mutant in atmospheres promoting photorespiration can be explained by the in vivo loss of glutamate synthase function. This in turn restricts glyoxylate amination and photorespiratory ammonia refixation (Fig. 3C) by limiting the supply of glutamate (Fig. 3A). If glyoxylate is not transaminated, it does not accumulate but undergoes nonen' zymatic oxidation, resulting in, a greatly enhanced loss of recently fixed carbon. As a result, Calvin-cycle intermediates are depleted, leading to a decline in net CO2 fixation (Fig. 2B). Similar explanations have been offered for the high CO2 growth requirement of four other classes of mutants with defects in photorespiratory nitrogen metabolism (8, 9, 26, 27).- These results emphasize the interdependence of the photorespiratory carbon and nitrogen cycles. The chloroplast dicarboxylate transporter is generally considered an essential component of a malate shuttle for transporting reducing equivalents across the chloroplast inner membrane (13, 14). Because mutant CS 156 was recovered in a screen for photorespiratory mutants and was capable of normal photosynthesis (Fig. 2A) and growth in nonphotorespiratory regimes, the major role of the chloroplast dicarboxylate transporter in vivo must be as a component of the photorespiratory nitrogen cycle. Conversely, the proposed role for this transporter as a part of the chloroplast malate shuttle appears to be of minor physiological importance. The chloroplast phosphate translocator is the only other known route for indirectly exporting reducing equivalents from the chloroplast at significant rates (14, 25). However, both ATP and NAD(P)H are exported in an obligate manner by this transporter. It seems likely that some mechanism must exist for modulating the cytosolic levels of ATP and NADH. The low level of dicarboxylate transport activity observed in the dct mutant may be adequate to meet this requirement. The dct mutant has been useful for determining the specificity of the dicarboxylate transporter. Previous studies based on back-exchange (11) and competition (12) experiments suggested glutamine was transported across the chloroplast envelope by the dicarboxylate transporter. This appears to be inconsistent with the observation that glutamine transport is unaffected by the lesion at the dct locus in the mutant. Clearly a separate transporter for glutamine, distinct from the dicarboxylate carrier, must occur in the chloroplast envelope. The discrepancy between previous and present results regarding the specificity of the dicarboxylate carrier can be resolved if two carriers with overlapping specificity exist in the chloroplast en. velope. The carrier defined by the dct mutant recognizes dicarboxylates but not glutamine (Fig. 4), whereas the second carrier presumably transports dicarboxylates at a low rate and glutamine. The low level of malate, 2-oxoglutarate, glutamate,

PS Proc. Natl. Acad. Sci. USA 80'(1983)

and aspartate uptake by chloroplasts of the dctmutant may represent transport activity by this second carrier. Aspartate (10) and 2-oxoglutarate (M. 0. Proudlove and D. A. Thurman, personal communication) are recognized by more than one chloroplast carrier. In each case, the second transporter was described as having a relatively low activity. Mutants harboring defective transporters generally have been identified as being resistant to an exogenously supplied toxic substance. For this reason, transport processes operating across the plasma membrane of simple organisms have been the most amenable to analysis with. mutants. The dot mutant of Arabidopsis demonstrates that intracellular transport in eukaryotes also can be subject- to mutant analysis. 1. Chollet, R. & Ogren, W. L. (1975) Bot. Rev. 41, 137-179.

2. Tolbert, N. E. (1979) in Encyclopedia of Plant Physiology, New Series, eds. Gibbs, M. & Latzko, E. (Springer, New York), Vol. 6, pp. 338-352. 3. Bowes, G., Ogren, W. L. & Hageman, R. H. (1971) Biochem. Biophys. Res. Commun. 45,716-722: 4. Lorimer, G. H. (1981) Annu. Rev. Plant Physiol. 32, 349-383. 5. Laing, W. A., Ogren, W. L. & Hagemnan, R, H. (1974) PlantPhysiol,

54, 678-685. 6. Somerville, C. R. & Ogren, W. L. (1979) Nature (London) 280, 833-836. 7. Keys, A. J., Bird, I. F., Cornelius, M. J., Lea, P. J., Wallsgrove R. M. & Miflin, B. J. (1978) Nature (London) 275, 741-743. 8. Somerville, C. R, & Ogren, W. L. (1980) Nature (London) 286, 257-259. 9. Somerville, C. R. & Ogren, W. L. (1981) Plant Physiol 67, 666671. 10. Lehner, K. & Heldt, H. W. (1978) Biochim. Biophys. Acta 501, 531-544. 11. Gimmler, H., Schafer, G., Kraminer, H. & Heber, U. (1974) Planta 120, 47-61. 12. Barber, D. J. & Thurman, D. A. (1978) Plant Cell Environ. 1, 297303. 13. Heber, U. (1974) Annu. Rev. Plant Physiol. 25, 393-421. 14. Heldt, H. W. (1976) in The Intact Chloroplast, ed. Barber, J. (Elsevier/North-Holland, Amsterdam), pp. 215-234. 15. Somerville, C. R. & Ogren, W. L. (1982) in Methods in Chloroplast Biology, eds. Edelman, M., Hallick, R. B. & Chua, N. H. (Elsevier, Amsterdam), pp. 129-138. 16. Cossins, E. A. & Sinha, S. K. (1966) Biochem. J. 101, 542-549. 17. Block, R. J., Durrum, E. L. & Zweig, G. (1955) A Manual of Paper Chromatography and Paper Electrophoresis (Academic, New York), p. 168. 18. Spector, T. (1978) Anal. Biochem. 86, 142-146. 19. Somerville, C. R., Somerville, S. C. & Ogren, W. L. (1981) Plant Sci. Lett. 21, 89-96. 20. Lilley, R. M., Fitzgerald, M. P., Rientis, K. G. & Walker, D. A. (1975) New Phytol. 75, 1-10. 21. Anderson, J. W. & Done, J. (1977) Plant Physiol 60, 354-359. 22. Stokes, D. M. & Walker, D. A. (1972) Biochem. J. 128, 1147-1157. 23. Heldt, H. W. (1980) Methods Enzymol 69, 604-613. 24. Wintermans, J. F. G. M, & Demots, A. (1965) Biochim. Biophys. Acta 109, 448-453. 25. Fliege, R., Flugge, U. I., Werden, K. & Heldt, H. W. (1978) Biochim. Biophys. Acta 502, 232-247. 26. Somerville, C. R. & Ogren, W. L. (1980) Proc. Natl Acad. Sci. USA 77, 2684-2687. 27. Somerville, C. R. & Ogren, W. L. (1982) Biochem. J. 202, 373380.