Glucose-1-Phosphate Transport into Protoplasts and ... · Glucose-1-Phosphate Transport into Protoplasts and Chloroplasts from Leaves of Arabidopsis1 Joerg Fettke*, Irina Malinova,

Glucose-1-Phosphate Transport into Protoplasts andChloroplasts from Leaves of Arabidopsis1

Joerg Fettke*, Irina Malinova, Tanja Albrecht, Mahdi Hejazi, and Martin Steup

Mass Spectrometry of Biopolymers (J.F.) and Department of Plant Physiology (J.F., I.M., T.A., M.H., M.S.),University of Potsdam, 14476 Potsdam-Golm, Germany

Almost all glucosyl transfer reactions rely on glucose-1-phosphate (Glc-1-P) that either immediately acts as glucosyl donor oras substrate for the synthesis of the more widely used Glc dinucleotides, ADPglucose or UDPglucose. In this communication,we have analyzed two Glc-1-P-related processes: the carbon flux from externally supplied Glc-1-P to starch by either mesophyllprotoplasts or intact chloroplasts from Arabidopsis (Arabidopsis thaliana). When intact protoplasts or chloroplasts are incubatedwith [U-14C]Glc-1-P, starch is rapidly labeled. Incorporation into starch is unaffected by the addition of unlabeled Glc-6-P orGlc, indicating a selective flux from Glc-1-P to starch. However, illuminated protoplasts incorporate less 14C into starch whenunlabeled bicarbonate is supplied in addition to the 14C-labeled Glc-1-P. Mesophyll protoplasts incubated with [U-14C]Glc-1-Pincorporate 14C into the plastidial pool of adenosine diphosphoglucose. Protoplasts prepared from leaves of mutants ofArabidopsis that lack either the plastidial phosphorylase or the phosphoglucomutase isozyme incorporate 14C derived fromexternal Glc-1-P into starch, but incorporation into starch is insignificant when protoplasts from a mutant possessing a highlyreduced ADPglucose pyrophosphorylase activity are studied. Thus, the path of assimilatory starch biosynthesis initiated byextraplastidial Glc-1-P leads to the plastidial pool of adenosine diphosphoglucose, and at this intermediate it is fused with theCalvin cycle-driven route. Mutants lacking the plastidial phosphoglucomutase contain a small yet significant amount oftransitory starch.

The entire metabolism of eukaryotic cells consists ofdistinct reaction sequences within a given compart-ment and the action of metabolite transporters thatfunctionally interconnect the various compartments.Similarly, transporters located at the interface betweenthe cells and the apoplastic space (i.e. in the cellmembrane) permit an intercellular transport of me-tabolites and thereby integrate metabolic processesthat take place in various cells. Both metabolite-relatedenzymes and transporters can exert a second function,as they also are capable of integrating signaling pathsand thereby transferring information on the metabolicstatus of the cellular compartments or the entire cell(Smeekens, 1998; Rolland et al., 2002).A large group of metabolite transporters are func-

tionally related to the metabolism of sugars or sugarderivatives, and some of them act within pathways thateither synthesize or degrade polysaccharides. Amongthe plant polysaccharides, quantitatively most impor-tant are the plastidial starch and the apoplastic cell wall.Cell wall-related sugar transporters are mainly locatedin the endoplasmic reticulum and transport mostlynucleotide sugars (Seifert, 2004). Starch-related trans-

porters reside in the inner envelope membrane ofplastids. A Glc transporter of the envelope membranehas functionally been characterized (Weber et al., 2000).However, the quantitatively dominant product of theplastidial transitory starch degradation is maltose,which is exported into the cytosol by the recentlyidentified maltose transporter designated as MEX (formaltose excess; Niittyla et al., 2004; Weise et al., 2004).Like the other transporters listed below, this trans-porter is located in the inner chloroplast envelopemembrane. As MEX-deficient mutants from Arabi-dopsis (Arabidopsis thaliana) contain exceptionally highmaltose levels and exhibit a massive starch excessphenotype as well as a strong retardation in growth,MEX appears to exert an indispensable functionwithin the starch-Suc conversion (Niittyla et al., 2004).

Other metabolite transporters of the chloroplastenvelope membranes are functionally more closelyrelated to the reductive pentose phosphate cycle.Phosphate translocators mediate a strict counterex-change of phosphorylated sugars (or of 3-phospho-glycerate) and orthophosphate. The triose phosphate/phosphate translocator exports triose phosphate (and,to some extent, 3-phosphoglycerate) from the chloro-plast into the cytosol, mainly during the light period(Schneider et al., 2002). The xylulose-5-phosphate/phosphate translocator has been proposed to providethe plastidial pentose phosphate pathwaywith reducedcarbon compounds or reducing equivalents (Eicks et al.,2002; Weber, 2004). Depending on concentration gradi-ents, the phosphoenolpyruvate/phosphate transloca-tors (strongly expressed in photosynthesis-competent

1 This work was supported by the Deutsche Forschungsgemein-schaft (Teilprojekt no. B2 to J.F. and M.S.).

* Corresponding author; e-mail [email protected] author responsible for distribution of materials integral to the

findings presented in this article in accordance with the policydescribed in the Instructions for Authors (www.plantphysiol.org) is:Joerg Fettke ([email protected]).

www.plantphysiol.org/cgi/doi/10.1104/pp.110.168716

Plant Physiology�, April 2011, Vol. 155, pp. 1723–1734, www.plantphysiol.org � 2010 American Society of Plant Biologists 1723 www.plantphysiol.orgon May 20, 2020 - Published by Downloaded from

Copyright © 2011 American Society of Plant Biologists. All rights reserved.

http://www.plantphysiol.org

cells of C4 plants) as well as the Glc-6-P/phosphatetranslocators (prominent in heterotrophic cells) mediatethe transport of the respective phosphorylated metab-olite between the plastid and the cytosol. For each ofthe two translocators, two genes have been identifiedthat are differentially expressed in various Arabidopsisorgans (Weber, 2004). Recently, evidence has beenpresented for the occurrence of an evolutionarily con-served plastidial ADPglucose exporter (Colleoni et al.,2010). However, current knowledge of the metabolitetransporters is far from complete, as analysis of thegenome of Arabidopsis revealed that approximately140 putative metabolite transporters exist, all of whichare predicted to be located in the inner plastid envelopemembrane (Ferro et al., 2002; Schwacke et al., 2003).

None of the translocators described above is ableto transport another important metabolite, Glc-1-P(Kammerer et al., 1998; Eicks et al., 2002). This Glc esteris a key intermediate in several major carbon fluxes,such as starch, Suc, and cellulose biosynthesis. In thecurrent model of the photosynthesis-driven starch bio-synthesis, the plastidial phosphoglucomutase (pPGM)mediates the formation of Glc-1-P from Glc-6-P, whichis directly derived from a Calvin cycle intermediate,Fru-6-P. By the action of ADPglucose pyrophosphory-lase, Glc-1-P is then converted to ADPglucose, which isthe general glucosyl donor for a variety of starchsynthases. The dominance of the ADPglucose-depen-dent path of starch biosynthesis concurs with the factthat mutants from Arabidopsis lacking the plastidiala-glucan phosphorylase isozyme (PHS1) possess es-sentially the same starch content as wild-type plantswhen grown under normal conditions (Zeeman et al.,2004). However, transgenic potato (Solanum tuberosum)plants underexpressing both the plastidial and thecytosolic phosphoglucomutase unexpectedly possesstransitory starch levels similar to those of the wild-typecontrols. These phenotypic features are considered tobe inconsistent with this model (Fernie et al., 2002);therefore, the carbon fluxes toward starch appear to bemore complex than are often assumed.

Recently, we have shown that in heterotrophic tis-sues, such as potato tuber discs, Glc-1-P is selectivelytaken up and, subsequently, enters two paths. It ismetabolized via the cytosolic phosphorylase by trans-ferring the glucosyl residue to starch-related hetero-glycans (Fettke et al., 2008). Another portion of theimported Glc-1-P directly enters the amyloplasts andis converted to starch. Under normal conditions, thisconversion, which is mainly mediated by the plastidialphosphorylase isozyme, is not the dominant pathwayof reserve starch biosynthesis (Fettke et al., 2010).However, it offers an explanation of some features ofstarch metabolism as observed in various mutants (seeabove), because this route apparently relies neither onthe cytosolic nor the plastidial Glc-6-P/Glc-1-P inter-conversion.

The question remains whether or not this processoperates also in autotrophic tissues, as any evidence forthe transport of Glc-1-P into bothmesophyll protoplasts

and chloroplasts is lacking. In this communication, wehave studied the transport of Glc-1-P into autotrophiccells as well as possible implications of this process forthe entire transitory starch metabolism.

RESULTS

Uptake and Utilization of External Glc-1-P by Mesophyll

Protoplasts from Arabidopsis

To study the uptake of Glc-1-P by intact mesophyllprotoplasts, short-term experiments were performedusing varying concentrations of [U-14C]Glc-1-P. Fol-lowing an incubation of 30 s, the protoplasts werewashed and the total 14C content of the protoplastswas determined (Fig. 1, A and B). Uptake of the 14C-labeled Glc-1-P was clearly detectable at submicromo-lar concentrations, and saturation was achieved atapproximately 1 mM Glc-1-P. The apparent Km wasestimated to be 413 mM.

In another series of short-term uptake experiments,the protoplasts were incubated with either 200 or 400mM [U-14C]Glc-1-P and an equal concentration of un-labeled Glc and Glc-6-P, respectively. As a control, theunlabeled compound was omitted (Fig. 1C). Neitherthe presence of Glc nor that of Glc-6-P affected theuptake of Glc-1-P. Thus, the putative Glc-1-P trans-porter is selective for both the phosphorylated Glc (asit does not react with Glc) and the anomeric position ofthe phosphate ester (as it does not act on Glc-6-P).

In order to determine whether or not the importedGlc-1-P is converted to starch (as is the case in heter-otrophic potato tuber cells; Fettke et al., 2010), meso-phyll protoplasts from Arabidopsis leaves wereincubated for 20 min with equimolar concentrations(20 mM each) of [U-14C]Glc-1-P or [U-14C]Glc. Duringincubation, the protoplasts were either illuminated ordarkened. Following incubation, starch was isolatedand the 14C content was quantified. Incubation withGlc-1-P results in a more than 10-fold higher incorpo-ration as compared with Glc (Fig. 2A). Thus, thesuperior efficiency of the Glc-1-P-dependent labelingof starch is not unique to heterotrophic cells (comparewith Fettke et al., 2010). Light stimulates the flux fromboth external Glc-1-P and Glc to starch, but even indarkened protoplasts, the Glc-1-P-dependent labelingof starch is higher than that of Glc in illuminatedprotoplasts. When comparing the incorporation intostarch in illuminated and darkened protoplasts, itshould be noted that in the dark starch biosynthesisvia the ADPglucose pyrophosphorylase is inhibited(Hendriks et al., 2003) and, in addition, net starchdegradation is initiated. The onset of starch mobiliza-tion, as observed under the conditions used, wasanalyzed by 14C labeling of protoplasts in the light,transfer into the dark, and determination of 14C label ofstarch during the dark phase (data not shown). There-fore, in darkened protoplasts, the 14C labeling of starchequals the total incorporation minus the release oflabel due to starch degradation.

Fettke et al.

1724 Plant Physiol. Vol. 155, 2011 www.plantphysiol.orgon May 20, 2020 - Published by Downloaded from



For a more detailed analysis of the Glc-1-P-depen-dent incorporation into starch, Arabidopsis mesophyllprotoplasts were incubated for 5 or 10 min with one ofthree mixtures each of which contained [U-14C]Glc-1-P

(20 mM each) and, in addition, one of three unlabeledcompounds (orthophosphate, Glc, or Glc-6-P; 10 mM

each). As a control, an aliquot of the protoplast sus-pension was incubated with [U-14C]Glc-1-P (20 mM)without adding any unlabeled compound. Protoplastswere illuminated during incubation. The addition ofany of the three unlabeled compounds increases the Glc-1-P-dependent incorporation into starch, but quantita-tively the enhancement differs depending upon theunlabeled compound. Glc-6-P ismore effective thanGlc,but orthophosphate is by far most efficient: It increasesthe labeling of starch approximately 20-fold as com-pared with the control (Fig. 2B). The stimulatory effectof orthophosphate clearly excludes the possibility thatthe Glc-1-P-dependent labeling of starch is due to adirect glucosyl transfer to the starch granules that iscatalyzed by a phosphorylase and takes place outsidethe intact protoplasts. In principle, any unnoticedbreakage of protoplasts could release both nativestarch granules and phosphorylase activity. However,the phosphorylase-mediated glucosyl transfer to starchgranules that occurs outside the protoplasts would beinhibited by orthophosphate. As a control, protoplastswere mechanically broken, and the conversion of Glc-1-P to starch within 10 min was monitored to be lessthan 1% as compared with that of the intact proto-plasts (data not shown). Furthermore, this type ofstarch labeling is not expected to be stimulated by light(Fig. 2B; see Fig. 3A below).

During incubation, the Glc-1-P-dependent carbonflux to starch is far from being constant, as most of the14C is incorporated into starch during the first 5 min ofincubation and the extension to 10min results in only avery small increment of the labeling (Fig. 2B). At themolecular level, this result is difficult to explain, as theentire flux is based on a series of reactions and in-cludes a variety of components, several of which couldbe limiting factors or steps but are difficult to estimate.Of special relevance appears to be a possible limitationexerted by the first step (i.e. the Glc-1-P import) by thecytosolic orthophosphate level. If so, the simultaneousaddition of orthophosphate and Glc-1-P would leadto a higher cytosolic orthophosphate concentration,which then would stimulate the import of the Glcphosphate.

In order to analyze a possible interdependenceof the transport of orthophosphate and of Glc-1-Pmore directly, two additional labeling experiments wereperformed. It should be noted that in these experi-ments, either the total label inside the protoplasts (Fig.2C) or the label released from the protoplasts into themedium (Fig. 2D) was monitored. In the first experi-ment, protoplasts were incubated in mixtures containing10 mM [33P]orthophosphate and, in addition, eitherunlabeled Glc-1-P or unlabeled Glc (20 mM each). Asa control, the incubation medium contained only[33P]orthophosphate. Following the incubation for 5 minin the light, the protoplasts were carefully washedto remove external 33P label and the amount of 33Pinside the protoplasts was monitored (Fig. 2C). The

Figure 1. Uptake of external Glc-1-P by Arabidopsis mesophyll pro-toplasts. A and B, Short-term uptake using varying concentrations ofGlc-1-P. Protoplasts were incubated for 30 s with [U-14C]Glc-1-P(concentrations as indicated) and washed, and the 14C label inside theprotoplasts was determined. The Michaelis-Menten (A) and the Line-weaver-Burk (B) plots are given. The mean of two independentlyperformed experiments (two replicas each) and the SD (n = 4) are given.C, Selectivity of Glc-1-P uptake by protoplasts. After 30 s of incubationof the protoplasts with 200 or 400 mM [U-14C]Glc-1-P, [U-14C]Glc-1-P,and equal concentrations of Glc or Glc-6-P, the protoplasts werewashed and the total 14C content of the protoplasts was determined.The mean of three independent experiments and SD are given.

Glc-1-P Transport

Plant Physiol. Vol. 155, 2011 1725 www.plantphysiol.orgon May 20, 2020 - Published by Downloaded from



addition of unlabeled Glc-1-P decreased the 33P labelinside the protoplasts as compared with the control.By contrast, equimolar external Glc did not affect the33P content of the protoplasts. This result is consistentwith the assumption that external Glc-1-P is importedinto the cell via a Glc-1-P/orthophosphate exchangethat takes place at the plasmalemma and therebydiminishes the orthophosphate-dependent prelabel-ing of the protoplasts.

In the second experiment, protoplasts were prela-beled by incubation with [33P]orthophosphate for 10min in the light. Subsequently, the residual externalorthophosphate was carefully removed by repeatedwashing steps and then the protoplast suspension wasdivided into two equal parts that were incubated for5 min in the light either in the presence or in theabsence (control) of 20 mM Glc-1-P. Finally, the proto-plasts were pelleted by centrifugation, and the 33Pcontent in the supernatant was monitored (Fig. 2D).External Glc-1-P increased the release of orthophos-phate into the medium more than 2-fold as comparedwith the control.

In summary, both labeling experiments clearly indi-cate that the mesophyll protoplasts import externalGlc-1-P by an antiport mechanism that utilizes ortho-phosphate as counter-ion.

Utilization of External Glc-1-P by Isolated

Arabidopsis Chloroplasts

The results shown in Figure 1 imply that Glc-1-P istaken up by the cells and is converted to starch. Thisraises the question whether or not Glc-1-P directlyenters the chloroplasts where the starch synthesistakes place. Until now, Glc-1-P has not been reportedto be imported into chloroplasts, but indirect evidencefor an uptake by nongreen plastids has been published(for potato tubers, Kosegarten and Mengel, 1994;Naeem et al., 1997; for wheat [Triticum aestivum],Tetlow et al., 1996; for soybean [Glycine max], Coatesand ap Rees, 1994). To answer the question mentionedabove, chloroplasts isolated from Arabidopsis leaveswere incubated with [U-14C]Glc-1-P. As a control, analiquot of the same chloroplast preparation was me-chanically broken by using a potter and, subsequently,the homogenate was treated identically. Isolated chlo-

Figure 2. Utilization of external Glc-1-P by mesophyll protoplasts fromArabidopsis. A, Glc- or Glc-1-P-dependent incorporation into starch.Protoplasts were incubated for 20 min with either [U-14C]Glc-1-P(G1P; 20 mM) or [U-14C]Glc (Glc; 20 mM) in the light or in the dark.Subsequently, starch was isolated and the [14C]glucosyl incorporationwas monitored. The mean of two independently performed experi-ments (three replicas each) and the SD are given. B, Glc-1-P uptake andutilization by mesophyll protoplasts. Protoplasts were incubated with[U-14C]Glc-1-P (G1P; 20 mM), [U-14C]Glc-1-P (20 mM) plus unlabeledGlc (10 mM; G1P/Glc), [U-14C]Glc-1-P (20 mM) plus unlabeled Glc-6-P(10 mM; G1P/G6P), or [U-14C]Glc-1-P (20 mM) plus unlabeled ortho-phosphate (10 mM; G1P/Pi) for 5 or 10 min in the light. Subsequently,starch was isolated and the [14C]glucosyl incorporation was deter-mined. Themean of three independently performed experiments and SD

are given. C and D, Counterexchange of Glc-1-P and orthophosphate.C, Protoplasts were incubated for 5 min in the light with [33P]ortho-phosphate (10 mM; Pi), [33P]orthophosphate (10 mM) and unlabeledGlc-1-P (20 mM; Pi/G1P), or [33P]orthophosphate (10 mM) and unla-

beled Glc (20 mM; Pi/Glc). The protoplasts were then washed until no33P label could be detected in the washing solution. The 33P contentof the protoplasts was monitored. The mean of two independentlyperformed experiments (two replicas each) and the SD (n = 4) aregiven. D, Protoplasts were incubated for 10 min in the light with[33P]orthophosphate (10mM). Following removal of the residual external33P label by washing, the protoplast suspension was separated into twoequal parts. To one part, unlabeled Glc-1-P (20 mM) was added (+G1P).In the other part (control), Glc-1-P was omitted (2G1P). After 5 min ofillumination, the protoplasts of both aliquots were pelleted and the 33Pcontent in the supernatant was determined. Results from a typicalexperiment out of two independently performed experiments (threereplicas each) and SD are given.

Fettke et al.




roplasts are able to take up Glc-1-P and to convert it intostarch at a relatively high rate. This process requires,however, the intactness of the organelles (Fig. 3A).Thus, both the uptake of Glc-1-P and the flux towardstarch are functional in isolated intact chloroplasts.In an additional experiment, chloroplasts were in-

cubated with a mixture of [U-14C]Glc-1-P and (unla-beled) orthophosphate or with [U-14C]Glc-6-P. As acontrol, chloroplasts were incubated only with [U-14C]Glc-1-P. For the three incubation mixtures, the incor-poration of 14C into starch was monitored (Fig. 3B). Inthe presence of both [U-14C]Glc-1-P and orthophos-phate, incorporation into starch is decreased (but stillexceeds that observed with Glc-6-P), suggesting a Glc-1-P/orthophosphate antiport that is functional at theenvelope membranes of the chloroplasts. These dataare further supported by the fact that simultaneousincubation of chloroplasts with 33P-labeled orthophos-phate and unlabeled Glc-1-P results in a decreaseduptake of orthophosphate as comparedwith incubationonly with 33P-labeled orthophosphate (data not shown).This result is expected if the external Glc-1-P is im-ported via an exchange with internal orthophosphate.As the incubation of isolated chloroplast with equimo-lar [U-14C]Glc-6-P results in a minor labeling of starch(Fig. 3B), import and plastidial metabolism of the twoexternally supplied Glc phosphate esters differ.In another series of experiments, isolated chloro-

plasts were incubated with either [U-14C]Glc-1-P onlyor together with unlabeled Glc-6-P or Glc (16.67 mM

each). 14C incorporation into starch was monitoredafter 10 or 20 min of incubation. Neither unlabeledGlc-6-P nor free Glc significantly affected the incorpo-ration of 14C into starch, indicating a selective importof the anomeric Glc ester into the chloroplast (Fig. 3C).In summary, all the data shown in Figure 3 indicate

that mesophyll cells from leaves are capable of im-porting Glc-1-P from the cytosol into the chloroplaststroma.

Contribution of the Plastidial AtPHS1 to the Conversion

of Glc-1-P to Starch

Following the uptake into chloroplasts, Glc-1-Pcould be further metabolized by two distinct starch-synthesizing pathways. First, in a single reaction (me-diated by the plastidial AtPHS1), glucosyl residuescould be transferred directly to acceptor sites at thesurface of native starch granules. Alternatively, itcould undergo a more complex sequence: first, theconversion of Glc-1-P to ADPglucose via ADPglu-cose pyrophosphorylase (Lin et al., 1988), and sub-sequently, the glucosyl transfer from ADPglucose toglucans of the starch granule that is catalyzed by atleast five ADPglucose-dependent starch synthase iso-zymes.In order to test the existence and/or relevance of the

PHS1-dependent path, Arabidopsis insertion mutantswere used that are deficient in the plastidial phos-phorylase isozyme (AtPHS1; Zeeman et al., 2004). The

phosphorylase pattern from Arabidopsis wild-typeleaves consists of four bands of activity, all whichstrictly depend on Glc-1-P (Fig. 4A). On a glycogen-containing separation gel, the two slowly movingbands represent the cytosolic phosphorylase isoform(AtPHS2) that exists in two states differing in theapparent affinity toward the immobilized polyglucan(Fettke et al., 2005). Similarly, the plastidial phosphor-ylase isoform (AtPHS1) occurs in two distinct butfaster moving bands (Fig. 4A). In knockout mutantslacking the plastidial phosphorylase, these two bandsare undetectable, suggesting that they are products ofthe same gene. The structural and functional implica-tions of the heterogeneity of AtPHS1 and AtPHS2 areunknown.

As compared with the wild-type control, mesophyllprotoplasts from AtPHS1-deficient lines did not differin the Glc-1-P-dependent incorporation into starch(Fig. 4B). Therefore, in mesophyll cells, the directglucosyl transfer to starch, as mediated by AtPHS1,appears to be of no or minor relevance. By contrast, inpotato tuber discs, the conversion of Glc-1-P intostarch did reflect the level of the plastidial phosphor-ylase activity (Fettke et al., 2010).

ADPglucose-Dependent 14C Incorporation into Starch

As the plastidial Glc-1-P pool is not noticeablyused for any AtPHS1-mediated 14C labeling of starch(Fig. 4B), we tested whether or not the conversion ofGlc-1-P to starch does include both the action ofADPglucose pyrophosphorylase and the incorpora-tion into the plastidial ADPglucose pool. If so, theactivity of the Calvin cycle that also feeds into thisstarch-forming path is expected to affect the conversionof the externally supplied Glc-1-P toward starch. Inorder to test this possibility, we incubated protoplastswith 10 mM [U-14C]Glc-1-P in either the presence orabsence of 5 mM unlabeled HCO3

2. 14C labeling ofstarch is significantly reduced by the addition ofunlabeled hydrogen carbonate (Table I). These resultsstrongly suggest that the imported Glc-1-P and inter-mediates of the Calvin cycle enter the plastidialADPglucose pool and, subsequently, utilize the samereactions that transfer glucosyl residues to starch.

If so, incubation of protoplasts with [U-14C]Glc-1-Pshould result in a 14C labeling of ADPglucose. To testthis prediction, we incubated protoplasts derived fromwild-type leaves with [U-14C]Glc-1-P for 10 or 20 minin the dark or in the light. At intervals, aliquots of theincubation mixture were withdrawn and metabolicprocesses were terminated by the addition of ethanol(final concentration of 50% [v/v]) followed by heating.By this treatment, proteins were denatured and me-tabolites were extracted. Subsequently, the extractswere lyophilized, dissolved in water, and then reactedwith a mixture of native potato tuber starch andrecombinant starch synthase III (AtSSIII) derivedfrom Arabidopsis. Following careful washing, the 14Cincorporation into native starch granules was moni-

Glc-1-P Transport




tored (Table II). Due to the selectivity of the recom-binant starch synthase, this method permits the quan-tification of the 14C content of ADPglucose even in thepresence of a large excess of [U-14C]Glc-1-P. In addi-tion, aliquots of the protoplast suspension were usedto monitor the incorporation into starch by the intactprotoplasts during incubation with [U-14C]Glc-1-P(Table II).

During illumination of protoplasts, import of[U-14C]Glc-1-P results in 14C incorporation into bothstarch and ADPglucose. Both processes require intact-ness of the protoplasts. In the dark, almost no starchis labeled and, in addition, less 14C-ADPglucose isformed (Table II). This result is not unexpected, as thesynthesis of ADPglucose via ADPglucose pyrophos-phorylase has been reported to be light dependent(Hendriks et al., 2003).

Starch Synthesis in the Arabidopsis pgm1 Mutant

The Arabidopsis mutant that lacks a functionalpPGM (i.e. the pgm1 mutant) is incapable of perform-ing Calvin cycle-driven biosynthesis of assimilatorystarch (Caspar et al., 1985). For a long time, this mutanthas been considered to lack any leaf starch. However,some recently published data indicate that it doesindeed contain starch, although in very small quantities(Niittyla et al., 2004; Munoz et al., 2005). This impliesthat, although to a far lower extent, assimilatory starchcan be formed by an additional, yet unknown, paththat does not include the plastidial conversion of theCalvin cycle-derived Glc-6-P to Glc-1-P. The import ofGlc-1-P into the chloroplast, as analyzed in this study,could be an essential step in this path.

To test this assumption, we prepared mesophyllprotoplasts from the pgm1 mutant and incubated theprotoplasts with [U-14C]Glc-1-P during illumination.For comparison, protoplasts were isolated from leavesof two types of Arabidopsis plants that had beengrown under essentially the same conditions and hadbeen treated identically: wild-type plants and a mu-tant having a strongly reduced level of the ADPglu-cose pyrophosphorylase activity (adg1; Lin et al., 1988).At intervals, aliquots of the three protoplast prepara-tions were withdrawn, and following the extraction ofstarch, the 14C content was monitored. In the pgm1mutant, 14C label was clearly detectable in the starchfraction and the 14C incorporation proceeded withtime (Fig. 5A). Thus, 14C derived from the externallysupplied Glc-1-P was taken up by the protoplasts,

Figure 3. Glc-1-P-dependent incorporation into starch by chloroplastsfrom Arabidopsis. A, Chloroplasts isolated from wild-type plants wereincubated with [U-14C]Glc-1-P (16.67 mM) in the light. As a control,chloroplasts were broken using a potter and were otherwise treatedidentically. After incubation, the starch was extracted and the 14Ccontent was monitored. The mean of two independently performedexperiments (two replicas each) and the SD (n = 4) are given. Intactindicates intact chloroplasts, and broken indicates mechanically dis-integrated chloroplasts were incubated. B, Selectivity of the Glc-1-P-dependent incorporation into starch. Illuminated chloroplasts wereincubatedwith [U-14C]Glc-1-P (16.67 mM each) or [U-14C]Glc-1-P plusunlabeled orthophosphate (6.67 mM) or [U-14C]Glc-6-P (16.67 mM

each). After isolation of the starch, the [14C]glucosyl incorporation wasdetermined. The mean of two independently performed experiments(two replicas each) and the SD (n = 4) are given. G1P, Incubation with[U-14C]Glc-1-P; G1P/Pi, incubation with [U-14C]Glc-1-P and unla-

beled orthophosphate; G6P, incubation with [U-14C]Glc-6-P. C, TheGlc-1-P-dependent incorporation into starch is unaffected by theaddition of unlabeled Glc-6-P or Glc. Illuminated chloroplasts wereincubated with [U-14C]Glc-1-P (16.67 mM each) only (G1P), with[U-14C]Glc-1-P plus unlabeled Glc-6-P (16.67 mM each; G1P/G6P), orwith [U-14C]Glc-1-P plus unlabeled Glc (16.67 mM each; G1P/Glc).Following 10 or 20 min of incubation, starch was isolated and the con-tent of 14C was quantified. The mean of two independently performedexperiments (two replicas each) and the SD (n = 4) are given.

Fettke et al.




imported into the chloroplast, and then incorporatedinto starch even in the absence of a functional pPGM.However, labeling of starch was lower as comparedwith the wild-type control. By contrast, the mutantfrom Arabidopsis having a lower ADPglucose phos-phorylase activity incorporated very little 14C intostarch, and labeling was essentially unchanged duringthe 5- to 20-min incubation (Fig. 5A). Thus, the resid-ual ADPglucose pyrophosphorylase activity of thismutant is insufficient to sustain the carbon flux fromexternally supplied Glc-1-P toward starch.Regarding the pgm1 mutant, the in vivo flux from

Glc-1-P into the plastids is minor and, therefore, un-able to permit normal starch accumulation. At the endof the light period, we monitored the leaf starchcontent to be 0.09 6 0.006 mg Glc g21 fresh weight,whereas the ecotype Columbia wild-type plants,grown under the same conditions, contained 6.843 60.427 mg Glc g21 fresh weight (n = 4). Despite the lowstarch content of the pgm1 mutant, we were able toisolate native leaf starch granules (Fig. 5C). Based onscanning electron microscopy (SEM) examination, themorphology of the mutant-derived starch particles issimilar to that obtained fromwild-type leaves (Fig. 5B;Streb et al., 2009).For several reasons, the relatively small contribution

of the Glc-6-P-independent flux to the total starchbiosynthesis is not surprising: First, in the cytosol, Glc-1-P is used in various (and, possibly, competing)reactions, among which the formation of UDPglucoseis most prominent. Subsequently, UDPglucose is usedfor the biosynthesis of Suc and cell wall polysaccha-

rides. Second, the total cellular content of Glc-1-P isvery low and undergoes only moderate changesthroughout the light/dark cycle (Schneider et al.,2002). Finally, provided an appropriate gradient isgiven, an efficient Glc-1-P/phosphate antiporter lo-cated in the chloroplast envelope membrane(s) willeven lower the plastidial Glc-1-P pool if mediatingbidirectional transport. Thereby, the transporter willdiminish the contribution of the Calvin cycle-indepen-dent path of starch biosynthesis.

DISCUSSION

In this communication, we provide evidence that pho-tosynthesis-competent mesophyll cells from leaves ofArabidopsis are capable of utilizing extracellularGlc-1-P. Following uptake, Glc-1-P passes the plastidialenvelopemembranes and, finally, the hexosyl residue isconverted to starch.

Figure 4. Impact of the plastidial phosphorylase isozyme on the incorporation into starch during Glc-1-P incubation. A,Phosphorylase pattern as revealed by native PAGE. Leaves were harvested during the light period. Buffer-soluble proteins wereextracted from Arabidopsis wild-type or AtPHS1 knockout plants. To each lane, 20 mg of protein was applied. The separation gelcontained 0.4% (w/v) glycogen. Following electrophoresis, the separation gel was equilibrated, incubated, and stained. I, Slowlymigrating cytosolic phosphorylase (AtPHS2); II, fast migrating cytosolic phosphorylase (AtPHS2); III, slowly migrating plastidialphosphorylase (AtPHS1); IV, fast migrating plastidial phosphorylase (AtPHS1). B, Impact of the plastidial phosphorylase (AtPHS1)on the Glc-1-P-dependent labeling of starch in Arabidopsis mesophyll protoplasts. Protoplasts from wild-type (wt) or plastidialphosphorylase knockout plants (lines Atphs1-1 and Atphs1-2) were incubatedwith [U-14C]Glc-1-P (20mM). At intervals, aliquotsof the protoplast suspension were withdrawn, and the incorporation of [U-14C]glucosyl residues into starch was monitored. Themean of two independently performed experiments (two replicas each) and the SD (n = 4) are given.

Table I. 14C incorporation into starch using Glc-1-P in the presenceof unlabeled hydrogen carbonate

Protoplasts were incubated in the presence or absence of 5 mM

HCO32 for 30 and 60 min, respectively. Protoplasts were illuminated

throughout the incubation. The starch was isolated, and 14C incorpo-ration was monitored. Values are given as nmol Glc mL21 (n = 3;6SD).

Incubation Time 5 mM HCO32 No HCO3

2 Added

30 min 1.05 6 0.04 1.72 6 0.0260 min 4.42 6 0.07 6.17 6 0.10

Glc-1-P Transport




Most of the experiments described here have beenperformed by using isolated mesophyll protoplaststhat, to some extent, are a nonphysiological systempossessing altered carbon fluxes such as the resyn-thesis of cell wall materials that may be favored overother biosynthetic paths. However, for this study, bothisolation and treatment of protoplasts had been care-fully optimized. Under the conditions used, more than85% of the protoplasts remained functional over aperiod of 24 h (data not shown). Nevertheless, controlexperiments were included ensuring that the biochem-ical processes studied are restricted to intact protoplasts.When using these precautions, functional protoplastsoffer some unique advantages, as, in short-term exper-iments, they permit a quantitative analysis of the uptakeof metabolites. For putative transporters, apparent Kmand Vmax values can be determined, which is impossibleif more physiological systems, such as leaf tissues, areapplied. As an example, due to both the geometry ofleaf discs and the structural complexity of the apo-plastic space, the uptake of a given metabolite into acell is unavoidably superimposed (and largely af-fected) by the diffusion of that compound to thevicinity of the respective cell; therefore, neither effec-tive extracellular concentrations of the metabolite noruptake rates can be determined.

Based on the experiments described above, wepropose the scheme of transitory starch biosynthesisoutlined in Figure 6.

The transfer of glucosyl residues to nonreducingends of native starch granules is mediated by variousisoforms of starch synthase (EC 2.4.1.21) that all utilizeADPglucose as donor. The glucosyl donor is formedby the action of the ADPglucose pyrophosphorylaseaccording to the following equation: ATP + Glc-1-P4ADPglucose + PPi.

However, two distinct routes lead to the plastidialpool of Glc-1-P. One is well established and driven by

the Calvin cycle intermediate, Fru-6-P, which, in a two-step reaction, is converted to Glc-6-P (mediated by theplastidial hexose phosphate isomerase) and, subse-quently, to Glc-1-P (mediated by the pPGM). InArabidopsis, this route is by far dominant; therefore,mutants lacking either functional plastidial hexosephosphate isomerase or pPGM are largely (but notcompletely) impaired in the biosynthesis of assimila-tory starch. A second path consists of the direct importof Glc-1-P into the chloroplast, where it joins the poolof the hexosyl phosphate used by ADPglucose pyro-phosphorylase. However, the import of Glc-1-P can-not compensate the biosynthetic route derived fromFru-6-P; therefore, the amount of native starch granulesthat is found in pPGM-deficient plants accounts foronly slightly more than 1% as compared with the wildtype (see above).

To some extent, the data reported here concur withthe metabolism of Glc-1-P as described for heterotro-phic tissues, such as potato tubers (Fettke et al., 2008,2010). Tuber parenchyma cells are capable of import-ing Glc-1-P and incorporating the glucosyl residuesinto starch. Evidence has been presented that thisprocess includes the action of two transporters locatedat the plasmalemma and the amyloplast envelope

Table II. Incorporation of [14C]ADPglucose formed by ADPglucosepyrophosphorylase into starch via AtSSIII

Protoplasts were incubated with 1 mM [U-14C]Glc-1-P in the light ordark. After 30 and 60 min of incubation, the metabolites wereextracted, and the incorporation of formed [14C]ADPglucose intostarch via AtSSIII was monitored. As controls, the AtSSIII was omitted(Minus SSIII), or 14C-labeled ADPglucose was added to the reactionmixture (Plus ADPglucose). In addition, from the protoplasts, the starchwas isolated, and the 14C incorporation during [U-14C]Glc-1-P incu-bation was monitored. Results from one experiment of two are shown(n = 3; 6SD).

Treatment 10 min 20 min

pmol ADPglucose mL21

Light 25.79 6 1.36 28.69 6 0.67Dark 1.50 6 0.12 1.67 6 0.02Plus ADPglucose 98,039.81 6 27.56Minus SSIII 0.48 6 0.21

nmol Glc mL21 incorporated into starch

Light 0.097 6 0.008 0.313 6 0.011Dark 0.013 6 0.009 0.017 6 0.0012

Figure 5. Starch synthesis in pgm1 mutants. A, Protoplasts were iso-lated from leaves of wild-type plants (wt), a mutant lacking the pPGMactivity (pgm1), or a mutant having reduced activity of ADPglucosepyrophosphorylase (adg1) and were incubated with [U-14C]Glc-1-P(20 mM) in the light. Subsequently, starch was extracted and the [14C]glucosyl incorporation into starch was determined. The data representmeans of three incubations 6 SD. B and C, Native starch granules iso-lated from leaves of wild-type plants (B) and the pgm1 mutant (C) asanalyzed by SEM. Bars = 20 mm.

Fettke et al.




membranes, respectively (Fettke et al., 2010). In addi-tion, the imported Glc-1-P has been shown to be incor-porated, via the cytosolic phosphorylase isozyme, intocytosolic heteroglycans (Fettke et al., 2008). However,this reaction seems to be of minor relevance for thecarbon flux directed to starch, as a reduced activityof the cytosolic phosphorylase did not affect the [U-14C]Glc-1-P-dependent labeling of starch. Nevertheless, thispathway could be involved in buffering the cytosolicGlc-1-P pool.Following the import of Glc-1-P into the amylo-

plasts, the subsequent incorporation into starch ismainly mediated by the plastidial phosphorylase iso-zyme (Pho1; Fettke et al., 2010). By contrast, in meso-phyll cells, Glc-1-P imported into the chloroplast leadsto the formation of ADPglucose, which then acts as aglucosyl donor for starch synthases (Table II). Theseresults are in agreement with the conclusion that inArabidopsis, the plastidial phosphorylase is not es-sential for starch metabolism under normal growthconditions (Zeeman et al., 2004).In potato tubers, the Pho1-dependent path seems to

be of minor relevance under normal in vivo condi-tions. Potato plants having a reduced pPGM activityaccumulate by far less reserve starch as comparedwithwild-type controls (Tauberger et al., 2000).Similarly, Arabidopsis plants (and potato leaves as

well) lacking a functional pPGM (Caspar et al., 1985)contain less starch than wild-type leaves. These mu-tants are capable of forming small amounts of leafstarch by a pathway that is not directly linked to thefunctional Calvin cycle (Fig. 6; Streb et al., 2009).However, potato lines possessing an antisense repres-sion of both the cytosolic PGM and the pPGM unex-pectedly exhibit a phenotype similar to the wild type(Fernie et al., 2002). Similarly, the sta5-1 mutant fromChlamydomonas reinhardtii, which is reported to lack thepPGM, accumulates 4% to 12% of the normal starchamounts (Van den Koornhuyse et al., 1996). In all these

cases, an import of Glc-1-P into the plastid can, at leastpartially, compensate the blocked Glc-6-P/Glc-1-Pconversion inside the plastid and, therefore, explainsthe described phenotypes.

For several reasons, it is not unexpected that the fluxoutlined above often permits only a partial restorationof starch biosynthesis. However, the quantification ofthe two Glc esters in several tissues indicates that theGlc-6-P levels exceed those of Glc-1-P (Tarnowsky et al.,1964; Alpers, 1968; Tetlow et al., 1998). Thus, the importof Glc-1-P into the plastids appears to be limited by thecytosolic concentration of the substrate. Interestingly, inmutants lacking a functional phosphoglucomutase, theGlc-6-P content is increased 10-fold but that of Glc-1-Pis only slightly higher as compared with the wild-typecontrols (Kofler et al., 2000). Furthermore, recentlypublished data strongly indicate that the expression ofmetabolite transporters located at the chloroplast enve-lope is affected by alterations in central carbon metab-olism. In pPGM-lacking mutants of Arabidopsis, theplastidial transport activity for both Glc-6-P and phos-phoglycerate is significantly increased (Kunz et al.,2010). Possibly, the flux from the plastidial Glc-6-Ppool into the cytosol is enhanced, which results in afaster formation of Glc-1-P by the cytosolic phospho-glucomutase activity. Furthermore, alterations in thecentral carbonmetabolism affect the expression of othertransporters of the chloroplast envelope as well. Inwild-type leaves, expression of the Glc-6-P transporteris weak (Niewiadomski et al., 2005). However, forthe plastidial Glc-6-P/phosphate translocator mutants,both a strongly increased expression of a second iso-form of this transporter and an enhanced transport ratehave been reported (Kunz et al., 2010).

In order to analyze the flexibility of the primarymetabolism, two double mutants from Arabidopsishave been generated in our laboratory that lack boththe plastidial phosphorylase (PHS1) and the ADPglu-cose pyrophosphorylase or AtPHS1 plus pPGM. Phe-notypic analyses of these mutants are in progress.

The uptake of Glc-1-P by isolated chloroplasts fromleaves of Arabidopsis, as shown in this study, appearsto contradict a previous study performed with spinach(Spinacia oleracea) chloroplasts. In that study (Quicket al., 1995), the conclusion was reached that externallysupplied Glc-1-P does not permit any significant starchbiosynthesis. However, in that study, chloroplasts wereisolated from spinach leaves that had been fed forseveral days with Glc. It remains to be clarifiedwhetheror not the long-term feeding of the spinach leaves withGlc affects the uptake and/or intrachloroplastidialmetabolism of Glc-1-P. In our experiments, the Glc-1-P-dependent incorporation into starch occurs at arelatively high rate that exceeds the rate of photosyn-thesis-dependent starch synthesis (data not shown).

The Glc-1-P-dependent starch labeling in mesophyllprotoplasts is much higher than that observed duringincubation with Glc (Fig. 2). Surprisingly, simulta-neous incubation of the protoplasts with [U-14C]Glc-1-P and orthophosphate results in a strongly increased

Figure 6. Proposed carbon fluxes in Arabidopsis leaves. ADPG,ADPglucose; AGPase, ADPglucose pyrophosphorylase; F6P, Fru-6-P;G1P, Glc-1-P; G6P, Glc-6-P; pPGM, plastidial phosphoglucomutase;SS, starch synthases; TP, triose phosphate.

Glc-1-P Transport




labeling of starch. This enhancement suggests an effi-cient limitation of Glc-1-P import by the cytosolicorthophosphate level. However, the simultaneous in-cubation with [U-14C]Glc-1-P and Glc or Glc-6-P alsoresults in an enhanced labeling of starch. Currently,this effect is difficult to explain. However, it should betaken into consideration that metabolic paths are oftensuperimposed by sugar-mediated signaling effectsthat alter carbon fluxes (see above).

Our results clearly demonstrate that Glc-1-P is takenup and, subsequently, is very efficiently metabolizedby mesophyll protoplasts from Arabidopsis. The directinterconnection between the cytosolic and plastidialGlc-1-P pools suggests so far unnoticed intracellularcarbon fluxes toward the plastidial starch that in-crease the flexibility of the plant primary metabolism.In planta under normal conditions, these pathwaysseem to be of minor relevance, but under someconditions (such as lower temperatures; Satoh et al.,2008) or distinct mutations, they enable the plant tobalance those particular situations. It remains to beclarified whether or not extracellular Glc-1-P and itsfast uptake by autotrophic cells will also constitute anefficient intercellular carbon flux.

MATERIALS AND METHODS

Plant Material

Arabidopsis (Arabidopsis thaliana) plants (ecotype Columbia, Wassilewskija,

pgm1, Atphs1-1, Atphs1-2) were grown in growth chambers using either a 12-h-

light (20�C)/12-h-dark (16�C) cycle or a 14-h-light (22�C)/10-h-dark (17�C)cycle. Throughout the light/dark cycles, relative humidity was 60%.

Protoplast Preparation

Mesophyll protoplasts were prepared fromArabidopsis plants grown for 3

to 4 weeks in soil. Leaves (3–4 g each) that had been washed with water were

transferred into 400 mM mannitol dissolved in water. Following slicing

(approximately 4 mm thickness), the leaf material was incubated for 3 to

3.5 h at 25�C under continuous gentle shaking in an incubation medium

consisting of 5 mM MES-KOH, pH 5.6, 400 mM mannitol, 8 mM CaCl2, 1%

(w/v) cellulase Onozuka R-10 (no. 16419, from Trichoderma vivide; Serva), and

0.25% (w/v) macerozyme R-10 (no. 28302, from Rhizopus sp.; Serva). Follow-

ing incubation, protoplasts were separated from the residual leaf material by

filtration through a nylon mesh (350 mm). In the filtrate, the protoplasts were

pelleted by centrifugation (90g for 12 min at room temperature). After

resuspending in the resuspension medium (5 mM MES-KOH, pH 5.6, 400

mM mannitol, and 15 mM MgCl2), the protoplasts were washed three times

with the samemedium and centrifuged as above. Subsequently, the number of

protoplasts was determined by using a Thoma counting chamber, and the

protoplast suspension was diluted to 3 3 105 cells mL21 (Gandhi and

Khurana, 2001) by using the resuspension medium. The proportion of intact

protoplasts was determined by treatment with fluorescein diacetate (Larkin,

1976). For all experiments, protoplast preparations with an intactness of more

than 90% were used.

Labeling Experiments

14C Incorporation into Starch

The labeling experiments were performed by incubating 15 mL of the

protoplast suspension with [U-14C]Glc-1-P or [U-14C]Glc (final concentration

of 20 mM each, containing 111 or 46.25 kBq; GE Healthcare). Alternatively, the

protoplasts (7.5 mL each) were incubated with [U-14C]Glc-1-P (in each case,

final concentration was 20 mM, containing 37 kBq), [U-14C]Glc-1-P plus

unlabeled 10 mM Glc, [U-14C]Glc-1-P plus unlabeled 10 mM Glc-6-P , or

[U-14C]Glc-1-P plus 10 mM unlabeled orthophosphate. At intervals, aliquots of

the protoplast suspension were withdrawn, immediately frozen in liquid

nitrogen, and stored at280�C until use. Following thawing and centrifugation

(10,000g for 12 min at 4�C), the pellets were resolved in 20% (v/v) ethanol and

centrifuged as above. The resulting pellets were resuspended in 80% (v/v)

ethanol and decolorized at 70�C. The samples were centrifuged as above, and

the pellets were collected. Following the addition of 1 mL of water each, the

suspensions were mixed and centrifuged as above. Subsequently, 500 mL of

200 mM KOH was added to each pellet, and the mixtures were incubated for

1 h at 95�C. Following neutralization with 1 M acetic acid, the samples were

centrifuged as above, and the 14C content of the supernatants was monitored

by using a liquid scintillation counter (Beckman Coulter). As revealed by

treatment with amyloglucosidase, more than 98% of the 14C-labeled material

solubilized by KOH consists of a-polyglucans; therefore, this fraction is

designated as starch.

Exchange Experiments

The orthophosphate-related exchange was analyzed by two types of

experiments. Throughout both types of experiments, the protoplasts were

illuminated (80 mmol m22 s21, room temperature). For type 1, resuspended

protoplasts (each 15 mL) were incubated for 5 min in the presence of 10 mM

(185 kBq) [33P]orthophosphate or in a mixture containing 10 mM (185 kBq)

[33P]orthophosphate and 20 mM unlabeled Glc-1-P. The protoplasts were

then repeatedly washed with resuspension medium until the total radioac-

tivity in the washing solution was below 100 dpm. The pelleted protoplasts

were then resuspended in 1 mL of water, and the radioactivity was monitored

using a liquid scintillation counter. For type 2, protoplasts (30 mL) were

incubated for 10 min with 1.85 MBq [33P]orthophosphate (specific activity of

3,000 Ci mmol21) and, subsequently, were washed three times with resus-

pension medium. The protoplast suspension was then divided into two equal

parts. One part was incubated for 5 min in the presence of 20 mM unlabeled

Glc-1-P. The other part (control) was incubated for 5 min in the absence of Glc-

1-P. Following incubation, the protoplasts were pelleted by centrifugation (90g

for 12 min), and in the supernatant, the 33P content was determined using a

liquid scintillation counter.

Short-Term Uptake Experiments

For short-time uptake experiments, protoplasts (4 mL) were incubated

with Glc-1-P, Glc-1-P and Glc, or Glc-1-P and Glc-6-P (concentration as

indicated, and 37 kBq [U-14C]Glc-1-P was added) for 30 s at room temperature

and illumination. The protoplasts were then immediately centrifuged (90g for

3 min at 4�C), and the supernatant was discarded. The pelleted protoplasts

were resuspended in 8 mL of resuspension medium and centrifuged again

(90g for 3 min at 4�C). This washing step was repeated two times. Finally, 1 mL

of water was added to the pelleted protoplasts, and the total 14C content was

quantified by liquid scintillation counting.

Quantification of [14C]ADPglucose

Protoplasts were incubated in a resuspension medium containing 1 mM

unlabeled Glc-1-P and 74 kBq [U-14C]Glc-1-P in the light or in the dark. In the

latter case, protoplasts were predarkened for 1 h. After 10 or 20 min of

incubation, the entire suspension was immediately frozen in liquid nitrogen.

Following the addition of ethanol (final concentration of 50% [v/v]), the

suspension was heated (10 min at 90�C). Following cooling and centrifugation

(10,000g for 10 min), each pellet was used for isolation of starch and

monitoring of the 14C content (see above). Each supernatant (containing

soluble metabolites plus externally supplied compounds) was lyophilized and

was then dissolved in 3.5 mL of water. Then, 1.5 mL of each solution was

added to the reaction buffer (as final concentration, 40 mM Tricine, pH 8.0, 1.6

mM EDTA, 20 mM potassium acetate, 75 mM citrate, and 13 mg of native potato

[Solanum tuberosum] tuber starch). The glucosyl transfer from ADPglucose to

the native starch granules was started by the addition of 17 mg of recombinant

AtSSIII and was continued for 30 min at 30�C. Subsequently, 25 mmol of

unlabeled ADPglucose and 5 mg of AtSSIII were added, and the mixture was

incubated for an additional 1 h. As controls, either the AtSSIII was omitted

(negative control) or, alternatively, 8.3 kBq [U-14C]ADPglucose (positive

Fettke et al.




control) was added during incubation. Finally, the samples were centrifuged

(14,000g for 2 min), and the pelleted starch was washed six times by

resuspending in 1 mL of water each and centrifugation (as above). In the

pelleted starch, the total 14C content was quantified by liquid scintillation

counting.

Chloroplast Isolation and Labeling

Chloroplasts were prepared according to Shi et al. (2000) with minor

modifications. Approximately 2 g of Arabidopsis leaves was harvested in the

beginning of the light period, freed of midveins, cut into small slices, and

incubated in 50 mL of precooled isolation buffer containing 20 mM Tricine-

NaOH, pH 8.4, 300 mM sorbitol, 10 mM EDTA, 10 mM KCl, 0.25% (w/v) bovine

serum albumin, 5 mM sodium ascorbate, and 5 mM dithioerythritol (DTE). The

slices were homogenized two times for 3 s each using a Waring Blendor, and

the resulting homogenate was filtered through two layers of Miracloth

(Calbiochem-Novabiochem). The filtrate was centrifuged (750g for 1 min at

4�C), and the pelleted chloroplasts were resuspended in precooled isolation

buffer (as above). The isolated chloroplasts were incubated with [U-14C]Glc-

1-P (final concentration of 16.67 mM, containing 74 kBq; final volume of 4 mL).

As controls, the isolated chloroplasts were broken by using a potter, and the

homogenate was incubated under otherwise identical conditions. Alterna-

tively, the chloroplasts were incubated with [U-14C]Glc-1-P (16.67 mM,

containing 74 kBq; final volume of 4 mL) or [U-14C]Glc-1-P (16.67 mM,

containing 74 kBq; final volume of 4 mL) plus unlabeled 6.67 mM orthophos-

phate or Glc-6-P (16.67 mM, containing 74 kBq; final volume of 4 mL) or [U-14C]

Glc-1-P (16.67 mM, containing 74 kBq; final volume of 4 mL) plus unlabeled

Glc (16.67 mM) or [U-14C]Glc-1-P (16.67 mM, containing 74 kBq; final volume of

4 mL) plus unlabeled Glc-6-P (16.67 mM). Throughout the experiments, the

chloroplasts were illuminated (80 mmol m22 s21, room temperature). At

intervals, aliquots of the suspension were withdrawn and immediately frozen

in liquid nitrogen. Following thawing, the samples were centrifuged (10,000g

for 10 min at 4�C). The pellets were resuspended in 20% [v/v] ethanol and

were centrifuged again as above. The pellets were decolorized in 80% (v/v)

ethanol at 70�C (20 min). After centrifugation (as above), the pellets were

washed three times with 80% (v/v) ethanol. The resulting pellets were treated

with 200 mM KOH for 1 h at 95�C. Following neutralization with 1 M acetic

acid, the samples were centrifuged, and the 14C content in the supernatant (i.e.

starch; see above) was monitored.

Isolation of Native Starch Granules

During the light period, leaf material (35 g) from the pgm1mutant plants was

harvested and immediately frozen in liquid nitrogen. Following homogeniza-

tion using amortar, native starchwas isolated according to Ritte et al. (2000) with

minormodifications. Following a passage through a Percoll cushion (4,000g for 5

min at 4�C), the pelleted starch was washed twice with extraction buffer (Ritte

et al., 2000) and resuspended in 200 mL of the same buffer. Contaminating

compounds were removed by adding an equal volume of phenol:chloroform

mixture (1:1, v/v) to the starch suspension and centrifugation (9,000g for 1 min).

The aqueous phase containing the starch particles was mixed with 1 mL of

chloroform and centrifuged as above. Subsequently, the upper phase was

centrifuged for 5 min at 13,000g, and the pellet starch was collected.

Potato tuber starch was isolated according to Ritte et al. (2000). Potato tuber

tissue (15–20 g) was mixed with 50 mL of buffer (20 mM HEPES-KOH, pH 7.5,

and 0.05% Triton X-100) and was homogenized for 20 s using a Waring

blender. The homogenate was passed through a nylon net (100 mm mesh

width). In the filtrate, starch granules were allowed to settle for 20 min, and

the supernatant was discarded. The starch pellet was washed five times with

water and finally was lyophilized. SEM analyses of the native starch granules

were performed after coating with gold with a Quanta apparatus (Philips).

Extraction of Buffer-Soluble Proteins

Leaf material was frozen in liquid nitrogen and homogenized using a

mortar. Per 1 g fresh weight, 1 mL of precooled buffer A (100 mM HEPES-

NaOH, pH 7.5, 1 mM EDTA, 2 mM DTE, 10% [w/v] glycerol, and 0.5 mM

phenylmethylsulfonyl fluoride) was added. All subsequent steps were per-

formed at 4�C. The resulting homogenates were centrifuged (20,000g for 12

min), and the supernatants were passed through a nylon net (range, 60–100

mm). The filtrates were used for protein quantification and for native PAGE.

Quantification of Proteins and Starch

Buffer-soluble proteins were quantified by using the microassay of

Bradford (1976) with bovine serum albumin serving as the standard. Leaf

starch content was determined essentially as described by Abel et al. (1996).

Pooled leaf material from several plants was frozen in liquid nitrogen and

homogenized using a mortar. Samples (50–80 mg fresh weight of the homog-

enized frozen material) were extracted two times with 1 mL of 80% (v/v)

ethanol for 20 min at 80�C. Insoluble material was washed with 1 mL of water

and then lyophilized. After resuspension in 0.5 mL of 200 mM KOH and

incubation at 95�C for 1 h, the samples were neutralized by adding 1 M acetic

acid and centrifuged (10,000g for 10 min). Aliquots of the supernatant (50 mL

each) were mixed with 50 mL of amyloglucosidase solution (starch determi-

nation kit; R-Biopharm) and incubated at 50�C overnight. The enzymatic

quantification of Glc was performed following the instructions of the man-

ufacturer.

Native PAGE and Activity Staining

Native PAGE followed by phosphorylase activity staining was performed

as described elsewhere (Fettke et al., 2005).

Cloning and Expression of AtSSIII

RNA was isolated from 100 mg of Arabidopsis leaves, harvested after 3 h

of illumination, by using the total RNA purification kit for plant material

from Macherey-Nagel. First-strand cDNA from AtssIII was synthesized using

SuperScript II Reverse Transcriptase (Invitrogen) following the manufacturer’s

instructions and the 3# primer (5#-CTTGCGTGCAGAGTGATAGAGCTCA-

AGATA-3#). Subsequently, the cDNA served as a template for PCR. Both the

EcoRI- andXhoI-linked primers (5# forward primer, 5#-GAATCCGATGGAAGTG-

CTCAGAAAAGAAC-3#, and 3# reverse primer, 5#-CTCGAGCTTGCGTGCA-

GAGTGATAGAGCTC-3#) include the complete cDNA except the predicted

plastidial transit sequence (60 bp from the start). For amplification of the 3.0-

kb fragment, a Phusion Taq Polymerase (Finnzymes) was applied. In a 50-mL

reaction volume, 2 mL from the reverse transcription reaction was used as a

template (30 cycles, annealing temperature of 49�C, 60 s for extension). Except

where stated, the instructions of the manufacturer were followed. The 3.0-kb

AtssIII fragment was subcloned into pGEM T-Easy vector (Promega). Subse-

quently, the AtssIII fragment was restricted by EcoRI/XhoI and ligated to the

expression vector pET23b (Novagen). For heterologous expression, the AtssIII

clonewas transformed into Escherichia coli strain BL21. E. coli cells were grown at

37�C in 600 mL of culture in Luria-Bertani medium containing 100 mg mL21

ampicillin until an optical density at 600 nm= 0.8 was reached. Expression of the

AtSSIII protein (At1g11720) was then induced by isopropylthio-b-galactoside

(final concentration of 0.1 mM; 4 h at 30�C). Bacterial cells were collected by

centrifugation (5,000g for 10 min at 4�C), resuspended in 15 mL of extraction

buffer (20mM sodium phosphate buffer, pH 8.0, 500mM NaCl, 20 mM imidazole,

2.5 mM DTE, and protease inhibitor cocktail I; Calbiochem), and then broken by

ultrasonification (90-s short pulses). The homogenate was cleared by centrifu-

gation (13,000g for 15 min at 4�C), and the supernatant was loaded on a HisTrap

HP column (product no. 17-5319-01; GE Healthcare). Subsequently, the column

was washed with 15 mL of extraction buffer, and the His-tagged AtSSIII protein

was eluted stepwise by increasing concentrations of imidazole (100–500 mM; in

extraction buffer, pH 8.0). AtSSIII protein-containing fractions were identified by

western blotting using an anti-His antibody (Qiagen), pooled, and concentrated

by ultrafiltration (30 kD; Amicon Ultra; Millipore). Subsequently, the purified

AtSSIII preparation was equilibrated with a buffer containing 50 mM HEPES-

NaOH, pH 7.5, 1 mM EDTA, and 2 mM DTE. Aliquots of the protein preparation

were frozen in liquid nitrogen and stored at 280�C.

ACKNOWLEDGMENTS

We thank Ms. Ulrike Krause for excellent assistance, Ms. Henrike Brust

and Ms. Kerstin Pusch (all University of Potsdam) for assistance with the

recombinant AtSSIII experiments, and John Lunn (Max-Planck-Institut of

Molecular Plant Physiology) for kindly providing the ADPglucose pyrophos-

phorylase from E. coli. We are indebted to Prof. Dr. K. Hausmann (Free

University of Berlin) for his support of the SEM analyses of the starch granules.

Received November 1, 2010; accepted November 25, 2010; published Novem-

ber 29, 2010.

Glc-1-P Transport




LITERATURE CITED

Abel GJ, Springer F, Willmitzer L, Kossmann J (1996) Cloning and

functional analysis of a cDNA encoding a novel 139 kDa starch synthase

from potato (Solanum tuberosum L.). Plant J 10: 981–991

Alpers JB (1968) Studies on phosphomutases. II. Effects of 1,3-diphospho-

glycerate on phosphoglucomutase. J Biol Chem 243: 1698–1704

Bradford MM (1976) A rapid and sensitive method for the quantitation of

microgram quantities of protein utilizing the principle of protein-dye

binding. Anal Biochem 72: 248–254

Caspar T, Huber SC, Somerville C (1985) Alterations in growth, photosynthesis,

and respiration in a starchless mutant of Arabidopsis thaliana (L.) deficient in

chloroplast phosphoglucomutase activity. Plant Physiol 79: 11–17

Coates SA, ap Rees T (1994) Metabolism of glucose monophosphates by

leucoplasts and amyloplasts from soybean suspension cultures. Phyto-

chemistry 35: 881–883

Colleoni C, Linka M, Deschamps P, Handford MG, Dupree P, Weber

APM, Ball SG (2010) Phylogenetic and biochemical evidence supports

the recruitment of an ADP-glucose translocator for the export of pho-

tosynthate during plastid endosymbiosis. Mol Biol Evol 27: 2691–2701

Eicks M, Maurino V, Knappe S, Flugge UI, Fischer K (2002) The plastidic

pentose phosphate translocator represents a link between the cytosolic and the

plastidic pentose phosphate pathways in plants. Plant Physiol 128: 512–522

Fernie AR, Swiedrych A, Tauberger E, Lytovchenko A, Trethewey RN,

Willmitzer L (2002) Potato plants exhibiting combined antisense re-

pression of cytosolic and plastidial isoforms of phosphoglucomutase

surprisingly approximate wild type with respect to the rate of starch

synthesis. Plant Physiol Biochem 40: 921–927

Ferro M, Salvi D, Riviere-Rolland H, Vermat T, Seigneurin-Berny D,

Grunwald D, Garin J, Joyard J, Rolland N (2002) Integral membrane

proteins of the chloroplast envelope: identification and subcellular localiza-

tion of new transporters. Proc Natl Acad Sci USA 99: 11487–11492

Fettke J, Albrecht T, Hejazi M, Mahlow S, Nakamura Y, Steup M (2010)

Glucose 1-phosphate is efficiently taken up by potato (Solanum tuber-

osum) tuber parenchyma cells and converted to reserve starch granules.

New Phytol 185: 663–675

Fettke J, Eckermann N, Tiessen A, Geigenberger P, Steup M (2005) Identifi-

cation, subcellular localization and biochemical characterization of water-

soluble heteroglycans (SHG) in leaves of Arabidopsis thaliana L.: distinct

SHG reside in the cytosol and in the apoplast. Plant J 43: 568–585

Fettke J, Nunes-Nesi A, Alpers J, Szkop M, Fernie AR, Steup M (2008)

Alterations in cytosolic glucose-phosphate metabolism affect structural

features and biochemical properties of starch-related heteroglycans.

Plant Physiol 148: 1614–1629

Gandhi R, Khurana P (2001) Regeneration from leaf protoplasts of

Arabidopsis thaliana ecotype estland. Indian J Exp Biol 39: 705–709

Hendriks JH, Kolbe A, Gibon Y, Stitt M, Geigenberger P (2003) ADP-

glucose pyrophosphorylase is activated by posttranslational redox-

modification in response to light and to sugars in leaves of Arabidopsis

and other plant species. Plant Physiol 133: 838–849

Kammerer B, Fischer K, Hilpert B, Schubert S, Gutensohn M, Weber A,

Flugge UI (1998) Molecular characterization of a carbon transporter in

plastids from heterotrophic tissues: the glucose 6-phosphate/phosphate

antiporter. Plant Cell 10: 105–117

Kofler H, Hausler RE, Schulz B, Groner F, Flugge UI, Weber A (2000)

Molecular characterisation of a new mutant allele of the plastid phos-

phoglucomutase in Arabidopsis, and complementation of the mutant

with the wild-type cDNA. Mol Gen Genet 263: 978–986

Kosegarten H, Mengel K (1994) Evidence for a glucose 1-phosphate

translocator in storage tissue amyloplasts of potato (Solanum tuberosum

L.) suspension-cultured cells. Physiol Plant 91: 111–120

Kunz HH, Hausler RE, Fettke J, Herbst K, Niewiadomski P, Gierth M,

Bell K, Steup M, Flugge UI, Schneider A (2010) The role of plastidial

glucose-6-phosphate/phosphate translocators in vegetative tissues of

Arabidopsis thaliana mutants impaired in starch biosynthesis. Plant Biol

(Stuttg) (Suppl 1) 12: 115–128

Larkin RJ (1976) Purification and viability determinations of plant proto-

plasts. Planta 128: 213–216

Lin TP, Caspar T, Somerville C, Preiss J (1988) Isolation and character-

isation of a starchless mutant of Arabidopsis thaliana (L.) lacking ADP-

glucose pyrophosphorylase activity. Plant Physiol 86: 1131–1135

Munoz FJ, Baroja-Fernandez E, Moran-Zorzano MT, Viale AM, Etxeberria E,

Alonso-Casajus N, Pozueta-Romero J (2005) Sucrose synthase controls both

intracellular ADP glucose levels and transitory starch biosynthesis in source

leaves. Plant Cell Physiol 46: 1366–1376

Naeem M, Tetlow IJ, Emes MJ (1997) Starch synthesis in amyloplasts

purified from developing potato tubers. Plant J 11: 1095–1103

Niewiadomski P, Knappe S, Geimer S, Fischer K, Schulz B, Unte US, Rosso

MG, Ache P, Flugge UI, Schneider A (2005) TheArabidopsis plastidic glucose

6-phosphate/phosphate translocator GPT1 is essential for pollen maturation

and embryo sac development. Plant Cell 17: 760–775

Niittyla T, Messerli G, Trevisan M, Chen J, Smith AM, Zeeman SC (2004)

A previously unknown maltose transporter essential for starch degra-

dation in leaves. Science 303: 87–89

Quick WP, Scheibe R, Neuhaus HE (1995) Induction of hexose-phosphate

translocator activity in spinach chloroplasts. Plant Physiol 109: 113–121

Ritte G, Eckermann N, Haebel S, Lorberth R, Steup M (2000) Compart-

mentation of the starch-related R1 protein in higher plants. Starch 52:

179–185

Rolland F, Moore B, Sheen J (2002) Sugar sensing and signaling in plants.

Plant Cell (Suppl) 14: S185–S205

Satoh H, Shibahara K, Tokunaga T, Nishi A, Tasaki M, Hwang SK, Okita

TW, Kaneko N, Fujita N, Yoshida M, et al (2008) Mutation of the

plastidial a-glucan phosphorylase gene in rice affects the synthesis and

structure of starch in the endosperm. Plant Cell 20: 1833–1849

Schneider A, Hausler RE, Kolukisaoglu U, Kunze R, van der Graaff E,

Schwacke R, Catoni E, Desimone M, Flugge UI (2002) An Arabidopsis

thaliana knock-out mutant of the chloroplast triose phosphate/phos-

phate translocator is severely compromised only when starch synthesis,

but not starch mobilisation is abolished. Plant J 32: 685–699

Schwacke R, Schneider A, van der Graaff E, Fischer K, Catoni E,

Desimone M, Frommer WB, Flugge UI, Kunze R (2003) ARAMEM-

NON, a novel database for Arabidopsis integral membrane proteins.

Plant Physiol 131: 16–26

Seifert GJ (2004) Nucleotide sugar interconversions and cell wall biosynthesis:

how to bring the inside to the outside. Curr Opin Plant Biol 7: 277–284

Shi LX, Lorkovic ZJ, Oelmuller R, Schroder WP (2000) The low molecular

mass PsbW protein is involved in the stabilization of the dimeric photo-

system II complex in Arabidopsis thaliana. J Biol Chem 275: 37945–37950

Smeekens S (1998) Sugar regulation of gene expression in plants. Curr

Opin Plant Biol 1: 230–234

Streb S, Egli B, Eicke S, Zeeman SC (2009) The debate on the pathway of

starch synthesis: a closer look at low-starch mutants lacking plastidial

phosphoglucomutase supports the chloroplast-localized pathway. Plant

Physiol 151: 1769–1772

Tarnowsky W, Kittler M, Hilz H (1964) Die Wirkung von Cortisol auf die

Umwandlung von Hexose Phosphaten in Glycogen. Biochem Z 341: 45–63

Tauberger E, Fernie AR, Emmermann M, Renz A, Kossmann J, Willmitzer L,

Trethewey RN (2000) Antisense inhibition of plastidial phosphoglucomutase

provides compelling evidence that potato tuber amyloplasts import carbon

from the cytosol in the form of glucose-6-phosphate. Plant J 23: 43–53

Tetlow IJ, Blissett KJ, Emes MJ (1998) Metabolite pools during starch

synthesis and carbohydrate oxidation in amyloplasts isolated from

wheat endosperm. Planta 204: 100–108

Tetlow IJ, Bowsher CG, Emes MJ (1996) Reconstitution of the hexose

phosphate translocator from the envelope membranes of wheat endo-

sperm amyloplasts. Biochem J 319: 717–723

Van den Koornhuyse N, Libessart N, Delrue B, Zabawinski C, Decq A,

Iglesias A, Carton A, Preiss J, Ball S (1996) Control of starch compo-

sition and structure through substrate supply in the monocellular alga

Chlamydomonas reinhardtii. J Biol Chem 271: 16281–16287

Weber A, Servaites JC, Geiger DR, Kofler H, Hille D, Groner F, Hebbeker

U, Flugge UI (2000) Identification, purification, and molecular cloning

of a putative plastidic glucose translocator. Plant Cell 12: 787–802

Weber APM (2004) Solute transporters as connecting elements between

cytosol and plastid stroma. Curr Opin Plant Biol 7: 247–253

Weise SE, Weber APM, Sharkey TD (2004) Maltose is the major form of

carbon exported from the chloroplast at night. Planta 218: 474–482

Zeeman SC, Thorneycroft D, Schupp N, Chapple A, Weck M, Dunstan H,

Haldimann P, Bechtold N, Smith AM, Smith SM (2004) Plastidial

a-glucan phosphorylase is not required for starch degradation in

Arabidopsis leaves but has a role in the tolerance of abiotic stress. Plant

Physiol 135: 849–858

Fettke et al.




Documents

Glucose-1-Phosphate Transport into Protoplasts and ... · Glucose-1-Phosphate Transport into Protoplasts and Chloroplasts from Leaves of Arabidopsis1 Joerg Fettke*, Irina Malinova,