Choline Oxidation by Intact Spinach Chloroplasts - NCBI

7 downloads 5 Views 2MB Size Report
Jul 30, 1987 - MSU-DOE Plant Research Laboratory, Michigan State University, East ..... suspension with a magnetic stir bar (70 revolutions/min) during.

Plant Physiol. (1988) 86, 0054-0060 0032-0889/88/86/0054/07/$01 .00/0

Choline Oxidation by Intact Spinach Chloroplasts' Received for publication July 30, 1987 and in revised form September 14, 1987


MSU-DOE Plant Research Laboratory, Michigan State University, East Lansing, Michigan 48824 known, the enzymes involved are not. Thus, in vivo radiotracer work (6) has established that betaine is synthesized in leaves by a two-step oxidation of choline:

ABSTRACT Pl s ize ie by a two-step oidation of choline (choline betane aldehyde - betaine). Protoplast-derived chloroplasts of spinach (Spixacia olracea L.) carry out both resons, more rapidly in light than in dark (AD Hanson ct aL 1985 Proc NatlAcad Sci USA 82: 3678-3682). We inveapted the light-stimulted oxidation of choline, using spinch chWoropblst isoated direcdy from leaves. The rates of cholne oxidatio obtined (dark and light rates: 10-50 and 100-300 per her er m gm chlorophyll, respectively) were approxinately 20fold higher than for protoplast-derived chloroplasts. Betime aldehyde was the main product. Choline oxidation in darkss and liht Was supssed by hypoxia. Neither unonplers nor the Calvin cycle inhibitor glyceraldehyde greatly affected choline oxidation in the light, and maxima choline oxdatio was attained far below light saturamon of CO2 fixation. The light stimulation of choline oxidation was abolished by the PSII inhibitors DCIM and dibromothymoquinone, and was partlly restored by adding reduced dure, an electron donor to PSI. Both methyl viologen and phenazine methosulfate prevented choline oxidation. Adding dihydroxyacetone phosphate, which can generate NADPH ix orsixeio, doubled the dark rate of choline oxidation. These results iicate that choline oxidation in chloroplasts requires oxygen, aNd reducing power generated from PSI. Enzymic reactions consistent with these requirements are discussed.



Choline -. betaine aldehyde -- betaine. Both steps of choline oxidation in spinach have been localized to the chloroplast and the first step has been found to be lightstimulated and DCMU-sensitive (6). Recent evidence indicates that the second step in spinach is catalyzed by a specific betaine aldehyde dehydrogenase (28), as it is in other organisms (5). Very little is known about the first step, although the chloroplastic location and light-stimulation apparently distinguish it from the choline -- betaine aldehyde reactions already known from mammalian liver and several microorganisms. In liver, choline is oxidized by a flavoprotein dehydrogenase ofthe inner mitochondrial membrane (27). Some microorganisms contain membranebound flavoprotein choline dehydrogenases (17, 24), whereas others have soluble flavoprotein choline oxidase enzymes (11, 34). The known choline-oxidizing enzymes are readily detectable in crude membrane preparations or soluble fractions, with an artificial electron acceptor necessary in some cases. However, exploratory assays with lysed spinach chloroplasts showed little activity, reinforcing the inference that the plant choline-oxidizing enzyme is unlike those from other sources. Therefore, in this work intact chloroplasts were used to probe the nature of the choline -* betaine aldehyde reaction. Our aim was to form working hypotheses about the type of reaction and its regulation by light, to guide further biochemical dissection. Previous research on choline oxidation in spinach (6, 28) used protoplastderived chloroplasts, which present some limitations for biochemical studies: they are necessarily exposed to hypertonic and low light conditions for several hours, and protoplast preparation is feasible only on a small scale. We therefore sought to use chloroplasts isolated directly from spinach leaves for the present work. Such chloroplasts proved to have much higher cholineoxidizing activity than those prepared via protoplasts.

Betaine (glycine betaine) is a nontoxic osmolyte accumulated by many prokaryotic and eukaryotic organisms that face saline or dry environments (reviews: 21, 32, 35). For higher plants, Wyn Jones et al. (33) have proposed that during osmotic adjustment, betaine accumulates mainly in the cytoplasm whereas salts accumulate in the vacuole. Such compartmentation would free osmotic adjustment from the known adverse effects of salts on metabolic functions. Much evidence now supports this hypothesis (4, 31); particularly persuasive is the recent demonstration that chloroplasts of salt-stressed spinach contain up to 300 mm betaine, 20 times the concentration in the leaf as a whole (26). Although the steps of betaine biosynthesis in plants are well

MATERIALS AND METHODS Plant Material. All plants were grown in 8-h d in the following growth chamber conditions: 23°C d, 300 ,uE m-2 s-' PAR, 60% RH/20°C night. Irrigation was with half-strength Hoagland so' Supported by the United States Department of Energy under contract DE-AC02-76ERO-1338, by the United States Department of Agriculture lution. Spinach (cv Savoy Hybrid 612) and pea (cv Little Marvel) under grant 87-CRCR-1-2460 from the Competitive Research Grants plants were grown in flats of vermiculite and harvested when 4 Office, and by grants from the State of Michigan REED-Biotechnology to 5 weeks old and 9 to 12 d old, respectively. Sugar beet (cv Progam, from CIBA-GEIGY Corporation, and from the Beet Sugar Great Western D-2) plants were grown in flats of soil containing Development Foundation. P. W. and C. L., respectively, received fellow- slow-release fertilizer (Osmocote, Sierra Chem. Co., Milpitas, ships from the French Ministry of Foreign Affairs and CONACYT, CA; 1 g/140 g dry soil), and harvested at 3 weeks. Quinoa plants Mexico. Michigan Agricultural Experiment Station Journal Article No. (cv C0407) were grown singly in 7-cm pots of vermiculite containing Osmocote (1 g/20 g dry vermiculite), and used at 6 weeks 12406. of age after a 24-h dark treatment to remove starch. Lettuce (cv 2 Permanent address: Departamento de Bioquimica, Centro de InvesParris Island) plants were grown in 10-cm pots of soil (2 or 3 Av. Politecnico Instituto Nacional, tigaciones y Estudios Avanzados, plants/pot) in the growth chamber for 2 weeks, and then transInst. Politec. Ncl. No. 2508, Mexico, D.F. 54


ferred to the greenhouse (14-h d, 20-28C day/16C night) until they were harvested at 4 weeks. Betaine in freeze-dried leaf samples was determined as in Weigel et al. (28). Chloroplast Isolation. Deveined leaves (routinely 10 g) were cut into 2 mm strips and ground in 4 volumes of semifrozen grinding medium for 3 s with a Polytron homogenizer at about 70% full power. Grinding media (adjusted to pH 6.5 at room temperature) contained Mes/NaOH 30 mm, sorbitol 0.33 M, MgCl2 5 mm, Na2EDTA 2 mm, isoascorbate 2 mm (10 mm for pea), and sodium pyrophosphate 10 mm (omitted for pea). For lettuce, 0.5% PVP-40 was also added, and for sugar beet and quinoa, 0.1% BSA. Chloroplasts were pelleted at 5000 g for 15 s at full speed and resuspended in 3 ml of the following buffer (resuspension buffer): Hepes/KOH 50 mM (pH 7.6), sorbitol 0.33 M, MgCl2 1 mM, MnCl2 1 mm, and Na2EDTA 2 mM. Depending on the species, the resuspension buffer was supplemented with sodium pyrophosphate 5 mM (spinach, quinoa, and lettuce), with pyrophosphate 30 mM plus inorganic phosphate 1 mm plus ATP 0.4 mm (pea), with 0.2% BSA (sugar beet), or with 0.5% PVP-40 (lettuce). Except for pea, the resuspended chloroplasts (3 ml) were then layered in a 15-ml centrifuge tube onto a 4-ml cushion of 40% Percoll containing the appropriate resuspension buffer, and centrifuged at 3000 g for 2.5 min. The chloroplast pellet was resuspended in resuspension buffer and the Chl concentration adjusted to 1 mg/ml. Pea chloroplasts were used without a Percoll purification step. For isolation of chloroplasts from salinized spinach plants, the sorbitol concentration in the grinding medium and resuspension buffer was raised to 0.59 M, and 20 mM KCI was added to the resuspension


Chloroplasts were checked for photosynthetic activity in a Rank Brothers 02 electrode at 25°C and saturating light. The electrode medium was the resuspension buffer used for each species plus 5 mM NaHCO3. Rates of C02-dependent 02 evolution were routinely 50 to 120 smol/h. mg Chl for spinach chloroplasts; rates for other species are given in the text. Chloroplast integrity measured by ferricyanide reduction (19) was always >80% for all species. ['4CjCholine Oxidation Assays. [Methyl-'4C]Choline (52 sCi/ ,umol, DuPont-NEN) was purified by treatment with alkaline H202 followed by TLE3 in 1.5 M formic acid as described (6). The ['4C]choline was then converted from the formate to the chloride salt as follows. Choline (50 MCi) was passed onto a 2-ml Biorex-70 (H+) column, which was washed with 4 ml of water and eluted with 2.5 ml of 1 N HC1. The eluate was reduced to about 0.5 ml in a stream of N2 at 25 to 30°C; an equal volume of methanol was added and the mixture was reduced to about 0.2 ml, at which point a further 0.2 ml of methanol was added. The last two steps were repeated twice before taking the sample to dryness, and redissolving in water at a concentration of 0.8

MCi/S l. Standard assays were carried out in 12 x 75 mm glass tubes containing 50 Al of resuspension buffer, chloroplasts equivalent to 20 ug Chl, and 5 pl of ['4C]choline solution (0.8 ,uCi, final concentration 0.3 mM). For incubations in the light, 5 mm NaHCO3 was also included. When additional compounds were included in the assay, they were added from concentrated stock solutions; aqueous stock solutions were adjusted to pH 7.6. Incubation was at 26 ± 1°C in darkness or red light (Kodak FHS lamp, Corning filter 2424); the standard irradiance was 380 W/ m2, determined with a Kettering model 68 radiometer. Reactions

3Abbreviations: TLE, thin

layer electrophoresis; ACP, acyl carrier

protein; AOV, analysis of variance; CCCP, carbonylcyanide 3-chlorophenylhydrazone; DAD, diaminodurene; DBMIB, dibromothymoquinone; DHAP, dihydroxyacetone phosphate; FABMS, fast atom bom-

bardment mass spectrometry.


were stopped in liquid N2 immediately after adding 1 Amol of carrier betaine aldehyde; reaction mixes were held in liquid N2 until analysis as follows. Reaction mixes were divided; the first half received an additional 1 MAmol of carrier betaine aldehyde and was transferred to a 17 x 150 mm tube and treated with 1 ml of 0.17 M NaOH/10% H202 for 1 h at room temperature to convert betaine aldehyde to betaine. The sample was then passed under slight pressure through a 1-ml column of mixed-bed ion exchange resin (9), followed by a 5-ml water wash, collecting the effluent. The second half received 1 ;tmol of carrier betaine and was fractionated on a mixed-bed column as above. ['4C]Betaine in the column effluents was determined by scintillation counting in Safety-Solve fluor (RPI). Sets of experimental samples were accompanied by a blank sample stopped at zero time, to correct for traces of betaine aldehyde and betaine in the ['4C]choline. In the text, tables, and figures, the term choline oxidation denotes the sum of ['4C]betaine aldehyde and ['4C]betaine produced. Tests for 114CCholine Oxidation by Contaminating Bacteria. Bacterial counts on nutrient broth agar plates showed that chloroplasts equivalent to 20 ,ug Chl were contaminated by 2 x I03 colony forming units. Because the leaf microflora can include choline-oxidizing bacteria (16), we evaluated the possibility that bacterial choline oxidation was a significant factor, as follows. Leaf microorganisms were cultured aerobically at 37°C with choline (2.5 g/L) as the sole C and N source, in a liquid medium containing K2HPO4 1 g/L, MgSO4. 7H20 0.5 g/L, and microelements and vitamins as described (22). When 6 x 104 bacteria in the midlog phase of growth were incubated in 50 ul of cholinefree culture medium with 0.8 MCi of [14C]choline, there was no detectable conversion of ['4C]choline to betaine or betaine aldehyde after a 30-min incubation at 26C. Thus, bacterial [14C]choline oxidation was negligible, even at contamination levels 30-fold above those expected in a standard assay. Separation and Identification of ['4C]Choline Metabolites. Three protocols were used to confirm the presence of labeled betaine aldehyde and betaine, and to search for other labeled products, in reaction mixes from illuminated spinach chloroplasts. Protocol 1 was designed to detect labeled phosphatidylcholine. The reaction mix was extracted with 2 ml methanol/ chloroform/water (12/5/1), and phases were split by adding 0.5 ml of CHC13 and 0.75 ml H20. An aliquot of the organic phase was taken for scintillation counting. Protocol 2 was designed to detect (labile) water-soluble esters of ['4C]choline, by analyzing the reaction mix in two ways. (a) One portion was supplemented with appropriate authentic compounds, and subjected to TLE on cellulose plates in 70 mM sodium tetraborate (2 kV, 10 min). Radioactive compounds were located by autoradiography, and standards with the Dragendorff spray reagent. Phosphoryl- and CDP-choline, betaine and betaine aldehyde were resolved from choline by the TLE system. (b) A second portion was treated with excess choline oxidase (from Arthrobacter globiformis, Calbiochem) prior to analysis by the same methods. The enzyme treatment converted remaining substrate ['4C]choline to betaine, permitting detection of neutral esters of choline (e.g. acetylcholine) which would otherwise be masked by the nearby [14C]choline zone. Protocol 3 was designed to detect a range ofpossible metabolites, including ['4C]trimethylamine. The reaction products were applied to a set of three l-ml columns connected in series in the order Dowex-l (OH-), Biorex-70 (HI), and Dowex50 (HI). After washing each column with water, 1 N HCI was used to elute acidic compounds from Dowex-1 and basic ones from Biorex-70, and 4 N NH4OH was used to elute betaine from Dowex-50. Aliquots of the eluates were separated by TLE on polysilicic acid-glass fiber plates in 1.5 N formic acid (1.8 kV, 10 min), and by TLC on silica gel G in methanol/acetone/12 N HC1 (90/10/4). Radioactive products were detected by autoradiography and identified by comparison to migration of markers.



Conversion of Deuterium-Labeled Choline to Eletaine Aldehyde. Spinach chloroplast samples (100 ,ug Chl) vvere supplied with 20 nmol of deuterium labeled choline chloridle (trimethyldg, 98%) (Cambridge Isotope Labs.) in place of ["4CIcholine, and incubated for 15 or 30 min in the light. Reactions were stopped in liquid N2 after adding 225 nmol of y-butyrobetaijne as internal standard. Samples were diluted with 2 ml of wateir, centrifuged to clear, and applied to a l-ml column of Doiwex-l (OH-) connected in series with a 1 -ml column of Dowex-'50 (H+). Both columns were washed with 5 ml water, and the caLtion fraction (containing betaine aldehyde) was then eluted fron n the Dowex50 column with 5 ml of 2.5 N HCI. After freeze-drying, these fractions were derivatized with n-butanol:acetyl chloride and subjected to FABMS analysis as described (25); the signal of interest was the di-n-butyl acetai derivative of betaune aldehyde (m/z 241). '4C02-Fixation Assays. Conditions were the same as for ['4C]choline oxidation assays, except that 0.3 miM unlabeled choline replaced ['4C]choline, andD0.25 to 0.5 smol (ofNaH'CO3 (5.3 MACi/4mol) was added in place of unlabeled bicarbonate. The reaction was stopped after 5 min by adding 2 ml of 1.5 N formic acid, and the mixture was clarified by centrifugatiion. Incorporation of '4C into a acid-stable products was detemnined on 0.2ml aliquots of the supernatant; these were dried under an IR lamp and redissolved in 0.5 ml of water, to which IC ml of Safety Solve was added for scintillation counting. RESULTS AND DISCUSSION Time Courses and Products of Choline Oxidatioin. Representative results are shown in Figure 1; the rate in the ligWht was about 8-fold that in the dark, and betaine aldehyde w{as the main reaction product. Note that ['4C]choline oxidation was assayed


-c i




. -; ;--I


FIG. 1. Rates of ['4C]choline oxidation to betaine aldLehyde (0) and betaine (0) in light (A) and darkness (C) by spinach chli )roplasts (20 Chl). The rate of light '4C02 fixation in the chloroplast pireparation used in this experiment was 37 ;tmol/h * mg Chl. Panel B shoNws an autoradiograph of reaction mixes separated by TLE on cellulose plates in 70 mM Na borate, at zero time and after 15 min incubation in Hi



Plant Physiol. Vol. 86, 1988 in the presence ofphosphate and HCO3, and that parallel assays with H'4C03 demonstrated that the illuminated chloroplasts were photosynthesizing actively (Fig. 1, legend). Pooled data from 17 experiments similar to that of Figure 1 showed an average 8-fold stimulation of ['4C]choline oxidation by saturating light. Dark and light rates (mean ± SE) were: 32 ± 2 and 250 ± 13 nmol/h. mg Chl, respectively; these rates are approximately 20-fold higher than those obtained previously with chloroplasts isolated via a protoplast step (6). Figure lB shows only two metabolites of ['4C]choline in the light: a major product co-migrating with betaine aldehyde, and a minor product migrating with betaine. The identity ofthe main reaction product as betaine aldehyde was confirmed by supplying illuminated chloroplasts with deuterium (d9) labeled choline, and quantifying d9-betaine aldehyde by FABMS. The FABMS estimates of betaine aldehyde were in good accord (r = 0.95, n = 5) with estimates for sister samples given ['4C]choline. A thorough search for ['4C]choline metabolites of illuminated chloroplasts other than betaine aldehyde and betaine revealed none. The methods used would have detected (a) the phosphatidyl-, phosphoryl-, CDP-, and acetyl-esters of choline, and (b) trimethylamine, had these products represented 0.1 to 1% of the [14C]choline supplied. Since these are, respectively, activated forms and a likely breakdown product of choline, it is improbable that choline oxidation involved any activation step, or that it was accompanied by any degradative side-reactions. The accumulation of '4C in betaine aldehyde rather than in betaine was unexpected inasmuch as spinach chloroplasts have an active stromal betaine aldehyde dehydrogenase (150-250 nmol/h. mg Chl; Ref. 28), and labeled betaine aldehyde does not accumulate in spinach protoplasts (6) or leaf discs (25) supplied labeled choline. This anomaly was apparently due, at least in part, to efflux of ['4C]betaine aldehyde from the chloroplasts during incubation: when reaction mixtures were centrifuged gently (3 min, 500 g) to pellet chloroplasts after a 15-min incubation in light, >70% of the ["C]betaine aldehyde was found in the supernatant. Chloroplast integrity measured by the ferricyanide assay was similar before and after incubation. Since this excludes gross chloroplast breakage as the explanation for betaine aldehyde efflux, we speculate that a carrier system may be involved. A carrier system(s) for quaternary ammonium compounds would also account for the rapid effiux of endogenous betaine reported for spinach chloroplasts (26). Because salt-stress increases the in vivo rate ofbetaine synthesis several-fold in spinach (3), choline oxidation was measured in chloroplasts isolated from spinach plants salinized gradually to 200 mM NaCl. These chloroplasts showed satisfactory rates of C02-dependent O2 evolution (18 and 67 MAmol/h. mg Chl, in two independent experiments), but their rates of choline oxidation (75 and 120 nmol/h -mg Chl) were lower than those typical for chloroplasts from unstressed plants. Accordingly, we carried out all subsequent experiments with unsalinized plants. Influence of Assay Medium on Choline Oxidation. With the objective of optimizing the assay conditions, several parameters were varied. Increasing the amount of Chl in the range 20 to 80 Mg per 55 Ml assay progressively lowered the rate of choline oxidation expressed on a Chl basis, presumably by diminishing light transmission. Choline oxidation rate was a hyperbolic function of substrate concentration (Table I); the choline concentration giving half-maximum velocity was approximately 100 MM. This result argues against choline oxidation being a nonphysiological photooxidation process, because such a process would be unlikely to exhibit substrate saturation. When the pH of the incubation medium was modified to induce a change in stromal pH (29), choline oxidation in the light peaked at pH 7.6, as did '4CO2 fixation (results not shown). In the dark, no pH optimum was apparent; choline oxidation simply rose as the pH was

CHOLINE OXIDATION BY SPINACH CHLOROPLASTS Table I. Effect of Substrate Concentration on Choline Oxidation by Spinach Chloroplasts in the Light

Table II. Effect ofBicarbonate and Glyceraldehyde on Choline Oxidation by Spinach Chloroplasts in the Light

["C]Choline Oxidation

['4C]Choline Concentration

Betaine aldehyde

mM 0.03 0.10 0.30




8.2 15 16

121 153


"CO2 Fixation



0 1.5

+ +

HCO3- 5 mM HCO3- 5 mM + glyceraldehyde



0 I




..Rx ° 1.0

0 co 3.0








36 0.4


Table III. Effect of Uncouplers on Choline Oxidation and C02 Fixation in the Light Spinach chloroplasts (20 ,g Chl) were preincubated for 30 min on ice in the presence of 0.1% ethanol with or without uncouplers, in a total volume of 50 Ml, before assaying choline oxidation and CO2 fixation. The experiment was repeated three times, with similar results.




None None Ethanol only Nigericin 1 LM CCCP 10 ,M



191 133

10 mM


*0.5 0

["C]Choline Oxidation

nmol/h * mg Chl Amol/h. mg Chl

nmol/h. mg Chi 97




Light Light Light Light


14CO2 Fixation Oxidation nmol/h. mg Chl umol/h * mg Chl 10 0.6 195 185 122 68

24 27 0.7 0.2

ities is correct. Thus, addition of the Calvin cycle inhibitor glyceraldehyde (19), or omission of HCO3 , did not greatly Irradiance (W/m2) depress choline oxidation (Table II). Table III shows that nigerFIG. 2. Responses to irradiance of [1"C]choline oxidation and "CO2 icin and CCCP, which act on the electrochemical gradient in fixation by spinach chloroplasts. Data are pooled results from three different ways (13), totally suppressed photosynthetic CO2 fixaexperiments, each with duplicate samples, normalized by setting reaction tion. This confirms that they were effective in intact chloroplasts, rates at 6 W/m2 = 1.0 (arrows). The reaction rates at 6 W/m2 in each which other uncouplers tested (gramicidin, methylamine, NH4I) experiment were: `C02 fixation-8.9, 8.0, 8.9 Mmol/h -mg Chl; [1"C] were not. Neither nigericin nor CCCP abolished light-stimulated choline oxidation- 187, 254, 127 nmol/h-mg Chl. Vertical bars are choline oxidation, showing that choline oxidation can occur U

a~30 ~~ 60,-I. 90 120 150


180 "



increased from 6.5 to 8.8. The effect of stirring the chloroplast suspension with a magnetic stir bar (70 revolutions/min) during the incubation was also checked; this improved neither the rate of choline oxidation, nor the rate of CO2 fixation. Based upon these results, we adopted the standard assay conditions described in "Materials and Methods." Effect of Irradiance on Choline Oxidation and CO2 Fixation. Light-stimulated choline oxidation has been shown to be DCMU-sensitive, which relates it in some way to photosynthetic electron transport (6). As a first step to clarify this relationship, the responses to irradiance of choline oxidation and CO2 fixation were compared (Fig. 2). Note that each process was assayed in the presence of the other. Interestingly, choline oxidation reached half-maximal rate at a far lower irradiance (5 W/m2) than did CO2 fixation (30 W/m2). This high sensitivity to light led us to test the effect of preillumination on subsequent dark oxidation. Preillumination (5 min, 380 W/m2) did not increase choline oxidation under the standard dark assay conditions, but decreased it by about 40% (data not shown). Because light activation of some enzymic systems in chloroplasts can be mimicked by incubation with DTT in the dark (1, 2), DTT pretreatment (5 mM, 30 min) was also tested. DTT did not stimulate choline oxidation in the dark, but strongly (about 90%) inhibited it. Relationship of Choline Oxidation to the Calvin Cycle and the Electrochemical Proton Gradient. Two ways in which choline oxidation might depend indirectly upon photosynthetic electron transport are (a) through requirement for a Calvin cycle product, or (b) via some consequence of the electrochemical proton gradient. Tables II and III indicate that neither of these possibil-

when the electrochemical proton gradient is collapsed, or at least nearly so. The 35% reduction of choline oxidation seen with nigericin is consistent with the pH-respone results given above, as nigericin is expected to prevent stromal alkalization in the light. The stronger inhibition by CCCP may reflect inhibition of both stromal alkaliztion and photosynthetic electron flow (13). Relationship of Choline Oxidation to Photosynthetic Electron Transport. Table IV confirms our previous report (6) that DCMU prevents light-stimulated choline oxidation, and shows that DBMIB has the same effect. As these inhibitors block electron flow between PSII and PSI (12), the simplest interpretation is that choline oxidation requires electrons from PSI. This interpretation is strongly supported by the substantial (40%) restoration of light-dependent choline oxidation by reduced DAD, which donates electrons between PSII and PSI, after the DCMU and DBMIB blocks (Table IV). Note that DAD also partially restored CO2 fixation. Methyl viologen or phenazine methosulfate, both of which accept electrons directly from PSI, abolished choline oxidation (Table IV). Two main inferences follow from this result. First, in accord with data from uncoupler experiments (Table III), choline oxidation does not depend on the electrochemical proton gradient, which forms in the presence of either electron acceptor (12, 13). Second, choline oxidation requires reduction of native electron carriers on the reducing side of PSI. The redox states of NADP, Fd and other electron carriers on the reducing side of PSI can be modified in intact chloroplasts in darkness: adding DHAP leads to generation of NADPH in organello, whereas adding electron acceptors such as oxaloacetate and NO2- promotes NADPH oxidation (18, 29). The effect of these compounds on dark choline oxidation (Fig. 3) were con-

Plant Physiol. Vol. 86, 1988



Table IV. Effects of Inhibitors ofPhotosynthetic Electron Transport on Choline Oxidation and CO2 Fixation In experiment 1, spinach chloroplasts were preincubated for 30 min on ice in the presence of ethanol with or without the additions indicated. Final ethanol concentrations were 1.3% for samples containing DCMU, 4.8% for samples containing DBMIB, and for the 'ethanol only' control. Ascorbate (10 mM) was present in all samples containing DCMU or DBMIB. In experiment 2, chloroplasts were not preincubated with the inhibitors, and no ethanol was present. Additions Experiment Incubation (14C]Choline Oxidation 4C02 Fixation umol/h * mg Chl nmol/h * mg Chl 0.3 51 None Dark I 47 324 None Light 33 267 Ethanol only Light 0.1 37 DCMU 50 jiM Light 13 129 DCMU + DAD 0.5 mM Light 0.1 7.0 DBMIB 40gM Light 15 113 DBMIB + DAD 0.5 mM Light 0.3 29 Nohe Dark 2 38 212 None Light 0.1 0.3 PMS l mM Light 1.5 7.6 MVO.lmM Light Table V. Effect of a Nitrogen Atmosphere or an Oxygen Trap on Choline Oxidation by Spinach Chloroplasts In experiments of types 1 and 2, a stream of air or N2 was passed over the chloroplasts for 10 min before addition of ['4C]choline, and during the reaction. For experiments of type 2, DCMU, DAD and ascorbate concentrations were as in Table IV. All light treatments contained 5 mm NaHCO3. Experiments of type 3 used 0.6 unit of glucose oxidase, and either 20 ug Chl plus 50 mM glucose, or 60 jig Chl plus 100 mm glucose; 02 electrode measurements showed that 02 concentration remained below detection limits, and that H202 concentration remained below 5 mm, during the 15-min incubation.





mControl frnrrai r3 o1.0




[14C]Choline Oxidation



nmol/h. mg Chl 1 ['4Ccholine DHAP, oxaloacetate and N02a oxidation in the dark by spinach chloroplasts. Final concentrations FIG. 3.









mms for


the Li

and oxaloacetate, 2 mm for N02. DHAP was salt. Data:are pooled results from 11I experiments,




nomlzdby setting the control reaction

rates at

1.0; the



for control reaction rates ± SE was 25.8 ± 3.8 nmol/h .mg Chl. Difrerences between the control and the DHAP and oxaloacetate treatments, and the control and- the

0.05, respectively (t



tests on


sinficant at P


0.01 and

paired data).

sistent with the view that light stimulates choline oxidation via electrons from PSI. Thus, the dark rate of choline oxidation more than doubled in the presence of'DHAP, but was lowered by oxaloacetate and N02-. Although DHAP generates ATP as well as NADPH in organello (29), the results of Tables III and IV make it improbable that ATP formation mediates the promotive effect of DHAP. Effects of Low Oxygen Tension and of an H202-Generating Oxygen Trap. When the 02 tension was lowered by continuously passing a stream of N2 over the reaction mix, dark choline oxidation was almost completely inhibited, and light oxidation was markedly suppressed (Table V). The remaining activity in the light is attributable to a low residual level of photosynthetically generated 02, because in chloroplasts treated with DCMU and reduced DAD (to allow choline oxidation without 02 evolution) the N2 treatment totally eliminated choline oxidation (Table V). The effect of a glucose/glucose oxidase 02 trap on dark choline

Air gassed N2 gassed

DCMU + DAD, air gassed DCMU + DAD, N2 gassed

36 3.8*

166 46** 48 0.3*

15 Control 19 + Glucose only 12 + Glucose oxidase only 0* +-Glucose + glucose oxidase 'Each type of experiment was repeated three times, and the data pooled for analysis by AOV. * ** Differences from the corresponding control treatment significant at P = 0.05 or 0.01, respectively.


oxidation was also examined, both to check the results obtained with N2 flushing and to test the effect of the H202 generated by the trapping reaction (Table V). The observed inhibition of choline oxidation confirms an 02-dependence, and also indicates that the stimulatory effect of light is not due to H202 producfion. Two other observations are inconsistent with an involvement of H202. First, that methyl viologen inhibited choline oxidation in the light (Table IV), because methyl viologen promotes formation of superoxide and H202 (12). Second, the addition of catalase (200 units/assay) did not affect ['4C]choline oxidation in the light (not shown). Choline Oxidation by Chloroplasts of Other Species. To evaluate the correlation between chloroplast choline-oxidizing activity and in vivo betaine accumulation, four other plants were tested: the chenopods sugar beet and quinoa, which accumulate betaine, and pea and lettuce which do not (Table VI). Chloro-



Table VI. Choline Oxidation by Chloroplastsfrom Betaine-accumulating and Nonaccumulating Species Incubations with ['4CJcholine were for 15 min (spinach, quinoa, and sugar beet) or 30 min (lettuce and pea), using 20 lAg Chl per assay in all cases except quinoa, where 100 tg Chl was used. Except for quinoa, all choline oxidation assays included NaHCO3. All species were tested in two or more experiments, with similar results.

['4C]Choline Oxidation Species

Betaine Content




0BEvolutione BetaieeBeaiine e Betaine

S s i n


Amol/g fresh wt





- mg

nmol/h mg Chl

Chi 61 20a

Spinach 3.8 22 1.0 232 Sugar beet 6.9 114 36 9.3 19a 3.9