var. the Sutton show a decline in the endogenous level of a-ketoglutarate ..... chemical Centre, Amersham, Bucks., England. NADP ..... synthesis in Plant Life.
Plant Physiol. (1970) 45, 624-630
Photoreduction of a-Ketoglutarate Chloroplasts'
to
Glutamate by Vicia faba Received for publication December 5, 1969
CURTIS V. GIVAN,2 ALICE L. GIVAN,2 AND RACHEL M. LEECH Department of Biology, University of York, Heslington, York, England ABSTRACT
Intact chloroplasts isolated from leaves of Vicia faba L. var. the Sutton show a decline in the endogenous level of a-ketoglutarate upon illumination. a-Ketoglutarate supplied to the chloroplasts is similarly utilized in this lightdependent reaction, and its consumption is paralleled by a concomitant increase in the level of glutamate. There is no
photostimulation of glutamate synthesis in chloroplasts broken by osmotic shock, but it can be somewhat restored by addition of ferredoxin and NADP. These results suggest that in the isolated chloroplast the synthesis of glutamate from a-ketoglutarate is regulated by the availability of reduced pyridine nucleotide generated by photosynthetic electron transport. This conclusion is supported by the finding of an apparent competition between the photoreduction of phosphoglycerate to triose phosphate and the photoutilization of a-ketoglutarate.
Bassham and Jensen (3) have recently speculated that a portion of the reduced pyridine nucleotide produced by photosynthetic electron transport might serve to reduce another keto acid, ca-ketoglutarate, in the reductive amination to glutamate. This amination reaction is considered important in mediating the assimilation of inorganic nitrogen into chloroplast amino acids (4), and the occurrence of an NADP-dependent glutamate dehydrogenase enzyme bound to the lamellae of Vicia faba chloroplasts has been previously reported (21). In view of the above results we wished to determine whether photosynthetically generated reductant does in fact drive the reductive amination of a-ketoglutarate. If the pool of photosynthetic NADPH can be so used, and if the conversion of a-ketoglutarate is regulated by the availability of reduced coenzyme, a light-dependent utilization of a-ketoglutarate would be predicted. The present experiments demonstrate such a photoutilization of a-ketoglutarate and a photostimulation of glutamate formation in V. faba chloroplast preparations.
MATERIALS AND METHODS Chloroplast Isolation Procedures. V. faba L. var. the Sutton plants were grown as described previously (27) in cabinets with controlled temperature and illumination conditions. The light It is well known that reductant generated by the electron source consisted of a bank of fluorescent lights (warm white transport reactions of photosynthesis can donate electrons to a Phillips Reflectalites, color 29), intensity 25,000 lux. To reduce wide variety of oxidants present in the chloroplast. In addition the amount of starch in the leaves, 16- to 20-day-old plants were to the ATP-dependent reduction of phosphoglycerate, which is darkened for 16 hr and then illuminated for 2 hr just before the reaction that probably utilizes most of the photosynthetic harvesting. The harvested leaves were darkened for 4 min and reducing power (cf. 2), the list of natural oxidants includes homogenized in an M. S. E. Atomix blender. Leaf laminae hydrogen ions (cf. 34), glyoxalate (25), and molecular oxygen (60-70 g) were homogenized in 200 ml of semifrozen buffer soluitself (6, 23). It is also thought that NADPH generated by photo- tion. The buffer used for homogenization of tissue contained synthetic electron transport can support the reductive steps in D (-) sorbitol, 0.33 M; tricine, 50 mm; disodium dihydrogen fatty acid synthesis within the chloroplast (24, 30). pyrophosphate, 10 mM; MgCl2, 1 mM; MnCl2, 1 mM; disodium Several years ago Graham and Walker (10) demonstrated that dihydrogen ethylenediaminetetraacetate, 1 mM; adjusted to pH leaf tissue readily reduces the keto acid oxaloacetate to malate in 7.2 with NaOH. The blender blades were first run slowly for a the light, whereas in the dark the reduction to malate is com- few seconds until all the leaves could be pushed below the paratively minor and the chief fate of oxaloacetate is transamina- surface of the liquid and then run at top speed for 5 sec. The tion to aspartate. With the notable exception of a report by crude homogenate was filtered through eight layers of muslin Harvey and Brown (11), most recent data have shown that the and four layers of 25-I, nylon bolting cloth. The filtrate was then outer envelopes of intact chloroplasts are quite impermeable centrifuged for 15 sec at 5400 g, the centrifuge head being quickly to pyridine nucleotides (15, 28). Furthermore, a number of brought to rest by hand with a thick pad of cotton. The superreports have indicated that malate dehydrogenase occurs in natant phase was discarded, and the chloroplast pellets were the chloroplast (12, 14, 20, 29). The weight of the evidence resuspended in an ice-cold solution designated "resuspension therefore suggests not only that the pool of reduced coenzyme buffer" which contained D (-) sorbitol, 0.33 M; tricine, 50 mM; generated by photosynthesis is able to reduce oxaloacetate to MgCl2, 1 mM; MnCl2, 1 mM; disodium dihydrogen ethylenemalate, but also that this photoreduction probably occurs diaminetetraacetate, 1 mM; pH 7.6. These buffers and techniques used for isolating the chloroplasts were modifications of those within the chloroplast. employed by Cockburn and co-workers (8). We decided to use ISupported by a postdoctoral fellowship from the Agricultural these chloroplast isolation procedures because they have been found by other workers to yield, in pea and spinach, a high Research Council, Great Britain. 2Present address: Department of Biochemistry and Biophysics, percentage of morphologically intact chloroplasts capable of carrying out high rates of CO2 fixation (18, 31). In our studies, University of California, Davis, California 95616. 624
Plant Physiol. Vol. 45, 1970 CHLOROPLAST PHOTOREDUCTION OF a-KETOGLUTARATE
preparations containing 50 to 60% intact chloroplasts were routinely obtained, as judged by their shining appearance under phase contrast microscopy; this appearance indicates the presence of an intact limiting membrane (cf. 27). All subsequent experimental incubations were carried out in the buffered resuspension medium described above, at pH 7.6. When suspensions containing only broken chloroplasts (free of intact outer envelopes) were desired, chloroplasts were subjected to osmotic shock by adding chloroplasts to a large volume of the usual resuspension medium from which only sorbitol had been omitted. After 60-sec exposure to this medium of low osmotic strength, an equal volume of resuspension buffer containing 0.66 M sorbitol was added to yield a final sorbitol concentration of 0.33 M. After the osmotic shock treatment, virtually all the chloroplasts appeared enlarged, granular, and no longer shiny under phase contrast. Such an appearance indicates breakage and loss of the limiting envelope (cf. 27). This breakage treatment was carried out immediately prior to experimental incubations. Incubation Procedure. In experiments with radioactive tracer, incubations were carried out with a total volume of 2 ml in a glass cell, jacketed for circulation of water from a constant temperature apparatus. In experiments where metabolite pools were to be determined enzymatically, incubation mixtures varied from 6 to 30 ml and were contained in a glass vial held in a water bath. All experiments were done at 20 C, and constant magnetic stirring was employed. Illumination was provided by a white incandescent projection lamp at an intensity of 40,000 lux except where otherwise noted. Details of individual reaction mixtures and experimental protocols are presented in figure and table legends. Metabolite Pool Determinations. For measurements of total pool sizes of a-ketoglutarate and triose-P, 5.0-ml aliquots of chloroplast suspension were deproteinized by addition of 2.5 ml of 15%c (w/v) perchloric acid, yielding a final concentration of 5%c (w/v) perchloric acid containing the deproteinized suspension. After standing for 5 min at room temperature, each deproteinized aliquot was transferred to an ice bucket and cooled for 15 to 20 min. After centrifugation for 10 min at 1600g and 3 C, a small amount of floating debris was removed from the surface of the sample by gentle aspiration. The supernatant fraction was then decanted into 25-ml Erlenmeyer flasks and neutralized to pH 4.5 to 5.5 with 69% (w/v) K2CO3 in the presence of 2 drops of 0.05% (w/v) methyl orange indicator. After neutralization the samples were maintained at 0 C for at least 10 min, and the precipitated potassium perchlorate was removed by 5-min centrifugation at 1600 g and 3 C. The perchlorate-free samples were then kept at 0 C until assayed for metabolite concentrations, always within a few hours of the end of the experiment.
Concentrations of a-ketoglutarate and triose-P were determined in the same extract by conventional enzymatic techniques. Similar methods have previously been successfully employed in measuring levels of metabolic intermediates in isolated chloroplasts and in intact plant cells (cf. 9, 19). A dual monochromator spectrophotometer was used to monitor changes in optical density resulting from oxidation of reduced pyridine nucleotide in the assay. An Aminco-Chance instrument was employed, with a measuring wave length of 340 nm and a reference wave length of 380 nm. The standard reaction mixture for the assay cuvette contained 2.0 ml of extract; 1.0 ml of 0.4 M triethanolamine buffer, pH 7.6; and 50 ,l of 8 mM ,B-NADH in 1 % (w/v) sodium bicarbonate. After this assay mixture had been allowed to rise to room temperature, the cuvette was placed in the spectrophotometer and the base line was established. Five microliters of a-glycerophosphate dehydrogenase + triose-P isomerase (mixed crystals from
625
rabbit muscle, 10 mg/ml) were added, and the oxidation of NADH resulting from reduction of dihydroxyacetone-P and glyceraldehyde-P was recorded. The spectrophotometer was rebalanced, and 10 ,l of bovine liver glutamate dehydrogenase (20 mg/ml in ammonium sulfate) were added. The further oxidation of NADH corresponding to reduction of a-ketoglutarate was recorded. These assay procedures are modifications of those described by Bucher and Hohorst for dihydroxyacetone-P plus glyceraldehyde-P and by Bergmeyer and Bernt for a-ketoglutarate, as outlined in the compendium edited by Bergmeyer (5). The amount of a-ketoglutarate or triose-phosphates (dihydroxyacetone-P + glyceraldehyde-P) present was calculated by comparison of the total change in A transmittance (X34-X38) with that obtained upon addition of standard amounts of substrate. Full scale deflection on the recorder corresponded to values ranging from 17 to 170 m,moles according to the transmittance range selected for assaying the particular sample. Tracer Experiments with a-Ketoglutarate-5-'4C. Small changes in total glutamate pool size were not easily measured accurately, because the endogenous pool was relatively large, i.e., of the order of 10-fold higher than the endogenous a-ketoglutarate pool; radioactive tracer was therefore employed. In the tracer experiments, a-ketoglutarate-5-'4C was supplied as the labeled substrate. After incubation of the tracer with chloroplast suspensions in the light and dark for varying periods, 0.2-ml aliquots of suspension were deproteinized by addition to 0.2 ml of a saturated solution of 2,4-dinitrophenylhydrazine in 2N HCl. Samples (50 ,l) of the supematant from each deproteinized aliquot were spotted onto Whatman No. 1 chromatography paper (medium flow rate, basis weight 87 g/mm2, thickness 0.16 mm). Chromatograms were developed by descending chromatography for 18 to 24 hr in n-butanol-acetic acid-water (12:3:1) (Ref. 22). In a number of cases chromatograms were also developed for shorter periods in 88C% phenol-water-acetic acid (84:16:1) or n-butanol-propionic acid-water. The latter solvent was prepared by mixing equal parts of propionic acidwater (180:220) and n-butanol-water (375:25), as described by Pedersen et al. (26). At least one duplicate sample on each chromatogram was cochromatographed with glutamate or glutamine standards. After development in the butanol-acetic acid-water solvent system (22), chromatograms were cut into strips and scanned for radioactivity on a Nuclear-Chicago Actigraph III strip scanner, with a 6-mm collimating slit. Strips containing glutamate and glutamine standards were then sprayed with ninhydrin to locate the position of these standard compounds. From unstained chromatogram strips, 1.5- X 2.0-inch rectangles corresponding to the position of glutamate were cut out and counted in toluene base scintillation fluid in a liquid scintillation spectrometer. Background correction was made by counting a blank square in the same manner. Except for a small amount of radioactivity remaining at the origin, only a single peak of radioactivity was observed by the scanning technique when the DNPhydrazone derivative of a-ketoglutarate had been allowed to drip off the descending chromatogram. When not run off, the radioactive keto acid hydrazone was seen as a second peak running close to the solvent front. Whether or not the solvent front had been allowed to drip off the chromatogram, the radioactive peak nearest the origin was in a position corresponding to standard glutamate located by ninhydrin staining in all three solvent systems tested. There was no radioactive peak of any significance in the position corresponding to glutamine. Total radioactivity added to each reaction mixture was measured by the following procedure: 0.2 ml of the reaction mixture was deproteinized by addition to 0.2 ml of 2 N HCl without dinitrophenylhydrazine. A 50-Al sample of supernatant from this unfractionated sample was then spotted onto a rectangle of
GIVAN, GIVAN,
626
AND LEECH
Plant Physiol. Vol. 45, 1970
Table I. Endogenious Conitent of a-Ketoglutarate in Chloroplast Suspensions Illumintated for 4 miii or Kept in Darkness Data are mean values of assays i range of duplicate or triplicate determinations. Chloroplast suspensions were preincubated in the dark for 5 min in the reaction mixture and then illuminated or kept in the dark for a further 4 min before deproteinization with perchloric acid. The reaction mixture contained: salts and sorbitol as in the "resuspension buffer"; NH4Cl (where used), 60 ,moles; chloroplast suspension containing 1.6 to 2.6 mg chlorophyll. Total volume = 6 ml. In some experiments 12- or 24-ml total volume was used, with all components at the same concentration as given above. Endogenous a-Ketoglutarate
Exper-
iment
Dark
Dark + NH4C1
Light
Light + NH4C1
Inpy,noles,/lng chloroplhyll
I III III
IV
14.6
0.2 ..
14.6
14.4 0.4 18.5 ± 0.9 15.1 13.1 ± 0.7
7.4 i 1.6 ...
9.2 ± 2.0
6.2 ± 1.2 5.5 7.1 6.2 i 0.6
chromatography paper and counted with the scintillation counter. Quenching in this unfractionated sample was the same as for those in which the dinitrophenylhydrazine had been removed by chromatography. Reagents. a-Ketoglutarate-5-'4C was obtained from the Radiochemical Centre, Amersham, Bucks., England. NADP, 3-NADH, and the assay enzymes were obtained from Sigma Chemical Co., London. Ferredoxin was kindly donated by Dr. J. P. Thornber, formerly of Twyford Laboratories. Chlorophyll Determination. Throughout this work chlorophyll concentrations were measured by the method of Arnon (1). RESULTS AND DISCUSSION
Suspensions of V. faba chloroplasts were found to contain low but measurable amounts of ca-ketoglutarate. When suspensions were illuminated for 4 min, the endogenous level of a-ketoglutarate showed a definite decline (Table I). An exogenous source of ammonia was not required for the occurrence of this light-induced decrease in the level of ae-ketoglutarate, although in some cases the decrease seemed slightly greater when 10 mm NH4Cl was present. The endogenous level of a-ketoglutarate remained virtually constant over a 4-min period in darkness and was not appreciably affected by the presence or absence of ammonium chloride. Although endogenous a-ketoglutarate could be detected, and fluctuations in its level were studied to a limited extent, the endogenous levels were so low that it would have proved difficult to carry out detailed time course studies without having to prepare very large amounts of chloroplast suspension. Since Heber and collaborators (14) had found intact chloroplasts to be fairly permeable to a-ketoglutarate, we decided to carry out some experiments with exogenously supplied a-ketoglutarate. Except when high concentrations of a-ketoglutarate were being specifically tested, the amounts of a-ketoglutarate supplied were kept low, remaining within the same order of magnitude as the endogenous concentration of this keto acid. These levels of a-ketoglutarate were easily assayed with the techniques employed. a-Ketoglutarate supplied to the chloroplasts was found to be consumed in the light, whereas virtually none was utilized in darkness (Fig. 1). The rate of this consumption was higher than that of endogenous a-ketoglutarate and varied in nine experiments from 0.3 to 1.7 ,umoles x mg chlorophyll-' X hr-',
TIME (minutes)
FIG. 1. Consumption of a-ketoglutarate supplied exogenously to chloroplasts. Reaction mixtures containing the chloroplast suspension were preincubated in the dark for 5 min prior to time zero on the abscissa. a-Ketoglutarate and triose phosphates (dihydroxyacetone-P + glyceraldehyde-P) were then assayed in perchloric acid extracts from aliquots taken at the times indicated. Light was turned on after 5 min. Reaction mixtures contained salts and sorbitol as in the resuspension buffer; chloroplast suspension containing 14.4 mg of chlorophyll; phosphoglycerate, 2 ,umoles; supplied a-ketoglutarate, 2 ,umoles; NH4Cl (where used), 0.3 mmole. Total volume, 30 ml. Rate of ca-ketoglutarate consumption in the light is about 0.6 /mole X mg chlorophyll-' X hr-'; rate of triose-P synthesis (in absence of NH4Cl) is approximately 1.6 ,umoles X mg chlorophyll-' X hr-1.
with the average value being 0.7. No added ammonia was required for this photoutilization. The main effect of 10 mm NH4Cl was that it severely inhibited the light-dependent synthesis of triose-phosphate; this inhibition is presumably a result of the uncoupling effect of NH4Cl (17). The uncoupling effect of ammonium chloride is deduced here on the basis of a several-fold acceleration of oxygen evolution in the presence of ferricyanide, and also by a similar acceleration in the rate of light-dependent 02 uptake by chloroplasts not provided with any exogenous oxidant other than 02. Because ammonium chloride does inhibit triose-P synthesis (Fig. 1) and since phosphoglycerate photoreduction occurs only in intact chloroplasts (cf. 13, 28), the failure of exogenous ammonium to promote a-ketoglutarate utilization cannot be ascribed to any failure of ammonium to enter intact chloroplasts. A more likely explanation is that the ammonium requirement for utilization of a-ketoglutarate and its conversion to glutamate (see below) is fully met from endogenous sources in isolated V. faba chloroplasts. This situation
Plant Physiol. Vol. 45, 1970 CHLOROPLAST PHOTOREDUCTION OF a-KETOGLUTARATE
627
10 1
Iigh ~~~~~~~ l t (intact)
t NADP
Fer redox n
8
0Z
°7
6
/dark (intact)
o/. ~~ ~ x-
E~
'
0
sions in the 50 ,onachrlsorlboec oolast -
z
0
0
i plast ( 6 0 0
2
light
(broken) dark
(broken) 0
x~
1
-
TIME (minutes)
FIG. 2B
2A i Accumulationinogutamate e nofnradioactiv dekt it trived soritoGa. hoolssi prxmtl Aoeo bytecrefrintc for fromi choolathsen andketuogluate-5fNAD preusreaddd.Pplintrored r1 gltmt X,gclrpyl-8 intritrdcdb the siuions lgtadihark.Chlropasdtions.pTenvalues plontaied
isprasntpriual upisn,sneteoe-l
mut
P0 o 6'intsr alcultchlrompdasts foral broken aditc chloroplastsasidct. Reaction mixtres A(ymutrcinguthradioactiveiktyogutarte werprtaaeincbthedar ino thearkforn 5 in. otheKegtogltarachepointmonmtheetimecourse pat Ahe on immedinitialy afterii was adedatctime (bastredmonu-INit mixtres, and TIMes ligte iRuminatedChlorel (inuoprtes) raiatvt mixture)represnted trce fthe ws d ed.Cmonenarts ofthe reaction byotal vourve) pror itoaddti chooflasktoltssaproiately14 were choleoro sionsainthe lIgt in the dark.tedChloroplastsupniscotne anemdposbeta orbgcroken)l~ cotinn 0o.5m fclrpyl at n plastsamnate
inothdprticuspensio prapsminhth srito aover-allutarants buffer;ising,asiactive clubaated (0mmls . c arksor mm. ra-Kyfredfoetoglutarate minsue are byarows in mtheses 0o1 xAmeogleAfter indicted ofxperiments wasd adedatthismeizeo, anidwudtherlihtwaotrned oen immediantel afdute doxuinated1 hmloell ofNDcells icroadded.Pit are orrectedfo ithe dilutioamatorIntrodued thatte nonradivacbie adintheiosolathed valuesoplastte reuspensaion. alkerntogvutaate, byssibebufer sorbiktogluasainate mighet activtymadded asynthesrailyoforutmaed from az-ketoglutarate. tarechsaminolatedfromldat thrfore Ponts broen and impotactchlroplasts utlzai n. rdoacivt ofth -ketoglutarate
the reactivablmixtue)represenfted
of a-ketoglutarate, namely reductive carboxylation to isocitrate, seemed much less probable; although isocitrate dehydrogenase does occur in isolated Vicia chloroplasts (20) as well as in other leaf tissues (e.g., 33), there seems to be little evidence for its functioning in the reductive direction in green leaf tissues. In the present instance no stimulation of a-ketoglutarate utilization by bicarbonate could be found. Had a-ketoglutarate consumption been proceeding via conversion to isocitrate, a stimulation by added bicarbonate would have been expected because this reaction is highly dependent on CO2 or bicarbonate availability (7). In addition, no rise in isocitrate was seen during illumination when this substrate pool was assayed. As noted under "Materials and Methods," changes in the over-all size of the glutamate pool were not easily assayed owing to the existence of a relatively large endogenous pool in the dark. Synthesis of glutamate was therefore best demonstrated with the aid of radioactive tracer. As illustrated in Figure 2A, an increasing amount of radioactivity was detected in glutamic acid when chloroplast preparations were incubated in the dark in the presence of a-ketoglutarate-5-'4C. Since in the dark there is virtually no net utilization of a-ketoglutarate (Fig. 1), the radioactivity accumulating in glutamate in the dark probably results chiefly from an exchange transamination with glutamate present in the endogenous pool; it is not, therefore, an indication of net glutamate synthesis. Exchange transamination activity is known to be very rapid in plant tissues (16), and exchange transamination activity has been localized in Vicia chloroplasts (P. R. Kirk, personal communication). While a progressive accumulation of radioactive label in glutamic acid does occur in the dark, the rate of glutamate labeling in the light is considerably faster, with close to a doubling of the rate occurring after the initial period in which dark exchange is very rapid. This degree of photostimulation is seen in those preparations containing a reasonable percentage (about
628
GIVAN, GIVAN, AND LEECH
Plant Physiol. Vol. 45, 1970
B 100
-
x""
--
low light
r
+PGA
~~~~0
,%*
hhigh light
