lable at ScienceDirect

Plant Physiology and Biochemistry 73 (2013) 266e273

Contents lists avai

Plant Physiology and Biochemistry

journal homepage: www.elsevier .com/locate/plaphy

Research article

Overexpression, purification and enzymatic characterization of arecombinant plastidial glucose-6-phosphate dehydrogenase frombarley (Hordeum vulgare cv. Nure) roots

Manuela Cardi a,b,c, Kamel Chibani b,c, Daniela Castiglia a, Donata Cafasso a, Elio Pizzo a,Nicolas Rouhier b,c, Jean-Pierre Jacquot b,c, Sergio Esposito a,*

aDipartimento di Biologia, Università di Napoli “Federico II”, Via Cinthia, 80126 Naples, ItalybUniversité de Lorraine, Unité Mixte de Recherches 1136 Interactions Arbres Microorganismes, F-54500 Vandoeuvre-lès-Nancy, Francec INRA, Unité Mixte de Recherches 1136 Interactions Arbres Microorganismes, F-54280 Champenoux, France

a r t i c l e i n f o

Article history:Received 10 July 2013Accepted 4 October 2013Available online 17 October 2013

Keywords:OPPPG6PDHBarley rootsPlastidial isoform

* Corresponding author. Dipartimento di Biologia, UII”, Complesso Universitario di Monte Sant’Angelo, ENapoli, Italy. Tel.: þ39 081 679124; fax: þ39 081 679

E-mail address: sergio.esposito@unina.it (S. Espos

0981-9428/$ e see front matter � 2013 Elsevier Mashttp://dx.doi.org/10.1016/j.plaphy.2013.10.008

a b s t r a c t

In plant cells, the plastidial glucose 6-phosphate dehydrogenase (P2-G6PDH, EC 1.1.1.49) represents oneof the most important sources of NADPH. However, previous studies revealed that both native and re-combinant purified P2-G6PDHs show a great instability and a rapid loss of catalytic activity. Therefore ithas been difficult to describe accurately the catalytic and physico-chemical properties of these isoforms.

The plastidial G6PDH encoding sequence from barley roots (Hordeum vulgare cv. Nure), devoid of along plastidial transit peptide, was expressed as recombinant protein in Escherichia coli, either untaggedor with an N-terminal his-tag. After purification from both the soluble fraction and inclusion bodies, wehave explored its kinetic parameters, as well as its sensitivity to reduction.

The obtained results are consistent with values determined for other P2-G6PDHs previously purifiedfrom barley roots and from other land plants. Overall, these data shed light on the catalytic mechanism ofplant P2-G6PDH, summarized in a proposed model in which the sequential mechanism is very similar tothe mammalian cytosolic G6PDH.

This study provides a rational basis to consider the recombinant barley root P2-G6PDH as a goodmodel for further kinetic and structural studies.

� 2013 Elsevier Masson SAS. All rights reserved.

1. Introduction

The Oxidative Pentose Phosphate Pathway (OPPP) is found in allprokaryotes and eukaryotes [1], except Archaea [2], providingreducing power as NADPH for amino acid and fatty acid anabolicpathways; and precursors of nucleotides and aromatic amino acids[3]. Moreover, a role for OPPP in response to abiotic [4,5] and biotic[6e8] stresses has also been recently demonstrated [9e11].

The existence of cytosolic and plastidial OPPPs in plants hasbeen reported in leaves, roots and cultured cells [12,13] based onthe presence of cytosolic (Cy-G6PDH), chloroplastic (P1-G6PDH),and plastidial (P2-G6PDH) isoforms of glucose-6P dehydrogenase

niversità di Napoli “Federicodificio 7, Via Cinthia, 80126233.ito).

son SAS. All rights reserved.

(EC 1.1.1.49eG6PDH), the first andmain rate-controlling enzyme ofthis pathway [12].

Recently, Meyer et al. [10] demonstrated the existence of aperoxisome-localized hybrid G6PDH tetramer, functioning both asredox switch and phosphorylated sugar sensor.

In higher plants, the major isoform is Cy-G6PDH, representing80e95% of thewhole cellular activity. The remaining 5e20% of totalactivity is confined essentially in chloroplasts or plastids [14,19],where two different isoforms have been identified [12].

Chloroplastic P1-G6PDH is present in green, photosynthetictissues and it is strongly inhibited upon illumination [15,16]. Itsubstitutes to light reactions for NADPH production during thenight, thus assuring an adequate supply of reducing equivalents inthe chloroplasts in the dark.

Root P2-G6PDH plays a pivotal role in providing NADPH (andconsequently, reductants) for the anaplerotic metabolism, inparticular during nitrogen assimilation in non-photosynthetic cells[16,24,25], and fatty acid synthesis [26]. More recently, a specific

M. Cardi et al. / Plant Physiology and Biochemistry 73 (2013) 266e273 267

increase in the abundance and activity of P2-G6PDH has beendescribed in response to ABA in barley [11].

Plastidial P2-G6PDH transcripts and activities are detectablethroughout the plant, more abundantly in stems and roots [17e19],whereas P1-G6PDH is detected in leaves, stem, flowers and siliquesof A. thaliana, within chloroplasts [18].

Both P1- and P2-G6PDHs are reversibly inhibited in vitro byartificial reductants, such as dithiothreitol (DTT) [19,20]. In vivo, thismechanism is controlled by light, involving enzyme modificationby thioredoxins [21], as recently demonstrated for A. thaliana P1-G6PDH [22].

One major difference between P1- and P2-G6PDH is the sensi-tivity to inhibition by NADPH [10,23]. It should be underlined thatthis effect is not a product-inhibition consequence, but it involves adifferent regulatory NADPH binding site. The inhibition is muchhigher for P1-G6PDH (KiNADPH < 10 mM; KiNADPH << KmNADP

þ ) thanfor P2-G6PDH (KiNADPH > 50 mM; Ki z KmNADP

þ ).All the plastidial G6PDHs purified from plant tissues, as leaves

[30], roots [19,20] and cultured cells [13] or from green algae [27e29] show an intrinsic high instability. Notably, at least one of thetwo Arabidopsis P2-isoforms rapidly loses activity upon storage indifferent conditions [18].

Overall, this caused uncertainty in the establishment of kineticproperties of this isoform, and no kinetic model has been proposeduntil now, and likewise no reaction mechanism is available forplastidial G6PDHs.

Because previous attempts led essentially to unstable and/orinsoluble enzymes [23], we have used different strategies to obtain

Fig. 1. Comparison of HvP2-G6PDH deduced amino acid sequences vs other known G6PDH sebackground; similar amino acids are in black on grey background. Legend: HvP2(Nure): Ho(AM398980); StP2 Solanum tuberosum P2-G6PDH (CAB52708.1); StP1 Solanum tuberosum PSolanum tuberosum Cy-G6PDH (CAA52442.1).

an as stable as possible recombinant P2- G6PDH. This enzymepreparation has been used to define its kinetic properties, leadingto a model describing the regulation mechanisms of this isoform.

2. Results and discussion

2.1. Barley P2-G6PDH sequence

The sequence of a barley P2-G6PDH (HvP2-G6PDH) expressed inroots was previously obtained from Hordeum vulgare cv. Alfeo(Accession AM398980). Based on this sequence, specific primerswere designed to clone the sequence coding themature form of theorthologue in Hordeum vulgare cv. Nure, keeping in mind that thecleavage of the targeting sequence is crucial for the production and,most likely, stability of recombinant proteins.

The Fig. 1 shows the alignment of the HvP2-G6PDH sequencewith a few other known plastidial G6PDH isoforms, as well as withthe cytosolic barley G6PDH recently isolated in our laboratory(accession ACV97161).

A transit peptide region of about 50 amino acids which likelydirects the protein to organelles can be predicted by a bioinformaticanalysis with PCLR (http://www.andrewschein.com/cgi-bin/pclr/pclr.cgi).

However, several observations led us to use a different cleavagesite. First, the inspection of the nature of amino acids suggested alonger targeting sequence. Consistently, a significant identity be-tween plastidial and cytosolic G6PDHs was observed only afteramino acid 99 (sequence ASVSITV). Additionally, it should be

quences from barley and potato. The strictly conserved sequences are in white on blackrdeum vulgare P2-G6PDH, cv. Nure; HvP2 (Alfeo) Hordeum vulgare P2-G6PDH, cv. Alfeo1-G6PDH (CAA58775.1); HvCy Hordeum vulgareCy-G6PDH, cv. Nure (ACV97161.1); StCy

M. Cardi et al. / Plant Physiology and Biochemistry 73 (2013) 266e273268

underlined that the human G6PDH protein starts in this region aswell (not shown), thus suggesting that this portion of the sequenceis not necessary for a functional enzyme. Finally, Meyer and col-leagues reported that a peroxisomal G6PDH from A. thaliana,AtG6PD4, displays an unusual N-terminus located transit peptide of152 amino acids [10]. Based on these considerations, a G6PDHprotein, devoid of the first 99 amino acids, was expressed inEscherichia coli BL21(DE3), either untagged (HvP2-G6PDH) or withan N-terminal 6His tail (His6-HvP2-G6PDH).

2.2. Recombinant HvP2-G6PDH purification

The overexpression of the untagged recombinant proteingenerated an intense band around 56 kDa, in accordance with itspredicted molecular weight of 55,800 Da, but the largest partaccumulated in the pellet of bacteria lysates (Fig. 2A). Anyway, theresidual enzyme present in the soluble fraction was purified byanion exchange chromatography. The G6PDH activity eluted in asingle peak at about 100mMNaCl, and active fractions were furtherpurified by Blue-Sepharose affinity chromatography (Suppl. Fig. 1).

Fig. 2. SDS PAGE and Western blotting of samples at different purification steps of HvP2-GWestern blotting using anti P2-G6PDH antibodies; Panels C and F, Western blotting using anwith an “empty” plasmid); P, pellet; S, surpernatant of bacterial lysate; Pur, purified proteinthe subunit molecular weight (56 kDa) of the over-expressed G6PDH protein, calculated w

Alternatively, a procedure, adapted from protocols optimised forhuman G6PDH (see material and method section), was utilised torecover part of the recombinant protein present in the inclusionbodies with the aim of evaluating whether it can represent away toobtain pure protein for the kinetic analyses [32].

In parallel, in order to obtain a better yield, and possibly increasethe percentage of soluble recombinant protein, HvP2-G6DPH wasfused to an N-terminal 6His-tag. It should be underlined that pre-vious attempts to purify recombinant His-tagged potato P2-G6PDHfailed, and therefore crude bacterial lysates were used to measurethe properties of the enzyme [23].

Using this strategy, the amount of recombinant tagged proteinpresent in the supernatant of lysates was sensibly higher, being atleast 50% of the recombinant protein produced in bacteria, andallowing the purification of the enzyme from the soluble fraction(Fig. 2D). Therefore, the His6-HvP2-G6PDH was purified using animmobilized metal affinity chromatography (IMAC), having nickelas binding metal (Suppl. Fig. 2). Purified His6-HvP2-G6PDHmigrated on SDS-gels as single major band having the expectedsubunit molecular weight (around 56 kDa) (Fig. 2D).

6PDH (AeC) and His6-HvP2-G6PDH (DeF). Panels A and D, SDS PAGE; Panels B and E,ti Cy-G6PDH antibodies. Legend: M, Markers; C-, negative control (bacteria transformedfrom soluble fraction; Ref, refolded protein from inclusion bodies. The arrow indicatesith the relative mobility factor method.

M. Cardi et al. / Plant Physiology and Biochemistry 73 (2013) 266e273 269

The fact that severing the first 99 amino acids did not preventthe production of recombinant proteins and did not abolish proteinactivity and sensitivity to reduction (see next sections) indicatesthat this region is not functionally important for the kinetics or theregulation mechanism, and may indeed represent a transit peptidewith an unusual length [31].

All recombinant proteins either purified from soluble fraction orrefolded reacted with anti P2-G6PDH antibodies (Fig. 2B and E) butnot anti Cy-G6PDH antibodies (Fig. 2C and F). Moreover, thedigested peptides obtained from the mass spectrometry analysis ofthe purified His6-HvP2-G6PDH unequivocally confirmed the iden-tity of the purified recombinant protein (Suppl. Fig. 3).

2.3. Kinetic properties of the recombinant HvP2-G6PDHs

The kinetic parameters determined from recombinant HvP2-G6PDHs both purified from the soluble fraction and refolded frominclusion bodies, are very similar, giving a KmG6P of 0.8e1 mM, aKmNADP

þ of 6e8 mM and a KiNADPH of 75e80 mM. Similarly, His6-HvP2-G6PDH exhibited a KmG6P of 0.86mM, a KmNADP

þ of 15 mManda KiNADPH of 66 mM (Table 1). These values indicated that the His-tagaddition does not affect the kinetic parameters of the recombinantenzyme. It is noteworthy that these results are comparableto values observed for the plastidial G6PDH enzyme purified frombarley roots [16,25] and potato P2-G6PDH [23]. The solubleHis6-HvP2-G6PDHwas relatively stable, retaining more than 80% ofits activity after 24 h but losing 70% of the initial activity after 3 daysat 4 �C. Therefore, the tagged construct was preferred for the rest ofthe study as it allows obtaining more homogeneous preparations.

2.4. Determination of a possible reaction mechanism for P2-G6PDH

The reactionmechanism of G6PDH found inmammals and otherliving organisms does not follow a common kinetic scheme [33, 34].To our knowledge, no reaction mechanism for the plastidial G6PDHfound in higher plants has been proposed yet.

The analyses of double reciprocal plots obtained from the pu-rified His6-HvP2-G6PDH by varying substrate concentrations wereutilised in order to define a reaction mechanism for barley rootplastidial G6PDH.

The reciprocal plots of His6-HvP2-G6PDH activity vs G6P con-centrations converged above the x-axis (Fig. 3A), whereas thereciprocal plot vs NADPþ intersected exactly on the x-axis (Fig. 3B).These results are coherent with a sequential mechanism, in whichboth substrates must bind to the enzyme before product formation.

The intercepts and the slopes of primary plots with l/[G6P] asthe variable can be utilised to draw secondary plots of the slopesand intercepts against l/V (Fig. 3C and D) in order to estimate theinitial-rate parameters [33, 35] shown in Table 2.

Table 1Kinetic parameters of recombinant untagged HvP2-G6PDH purified from the solublefraction (a) or from inclusion bodies (b) or of recombinant His6-HvP2-G6PDH pu-rified from the soluble fraction (c). They were compared to those obtained withHvP2-G6PDH purified from barley roots (d) [19]; or from isolated barley root plastids(e) [23].

P2-G6PDH KmG6P (mM) KmNADPþ

(mM)KiNADPH(mM)

MW(kDa)

DTT inhibition

a Untagged 0.78 � 0.1 6.6 � 2.0 95 � 3.6 57 � 3 NDb Untagged

and refolded0.93 � 0.01 7.1 � 1.3 76 � 7.5 55 � 4 25% 1 h 30 mM

c His-tagged 0.86 � 0.2 15 � 4 66 � 1.5 56 � 2 60% 1 h 25 mMd Barley root 1.03 � 0.10 9 � 2 ND 52 � 1.5 40% 1 h 25 mMe Barley root

plastids0.96 � 0.14 9.2 � 2.2 59�13.5 53�1.5 ND

ND: not determined.

This suggests that i) the enzyme obeys to a sequential mecha-nism, therefore both substrates must bind to the enzyme before thereaction can occur; ii) the binding of NADPþ slightly increases thebinding of G6P, whereas, iii) the binding of G6P as first substratedoes not affect the binding of NADPþ. This would possibly suggestan ordered mechanism for root plastidial HvP2-G6PDH, althoughthe data presented here do not clearly indicate if there is a givenorder of product exit from the enzymeesubstrate complex.

G6PDHs from fungi (Saccharomyces carlsbergennis, Penicillumduponti) or mammals (rabbit erythrocytes, or rat breast; bovineadrenals), follow ‘random sequential’ mechanisms [36]. Similarly,studies on recombinant human G6PDH, suggested that the enzymefollows a randommechanism as well [33]. In contrast, other studiesproposed an ‘ordered sequential’mechanism, as for the bovine lensenzyme [34].

The inhibition pattern of His6-HvP2-G6PDH was investigated byassaying the enzyme in the presence of NADPH. Previously, it hasbeen shown that NADPH inhibits G6PDH activity in a mixed-typepattern for P2-G6PDH purified from barley root plastids(Ki ¼ 58.6 mM [25];). A similar behaviour is observed with the re-combinant enzyme used here, with a Ki of 58.2 � 1 mM (Table 1).Possibly, NADPH binds to the enzyme independently of the sub-strates, apparently with a greater affinity when one (or both)substrate(s) is (are) already present at the active site (the linesconverge near the x-axis; Suppl. Fig. 4).

2.5. Sensitivity of recombinant His6-HvP2-G6PDH to reductants

All plastidial G6PDHs are redox-regulated with two cysteineresidues present in the N-terminal region of the protein [37]mediating a dithiol-disulfide exchange reaction. When the regu-latory cysteines are fully oxidised, the enzyme adopts a stabletetrameric structure, exhibiting its maximal catalytic rate.Conversely when plastidial thioredoxins, in particular of them and ftype [22] reduce this disulfide bridge, enzyme activity is severelydiminished. Based on sequence comparisons, these residues arelikely to be Cys149 and Cys157 in barley P2-G6PDH (Fig. 1).

To verify if the recombinant enzyme is indeed susceptible toreduction, the activity of His6-HvP2-G6PDH was tested in thepresence of reduced dithiothreitol. As expected, after 1 h, the ac-tivity decreased by 50% with 10 mM DTT, and by 60% with 25 mMDTT (Fig. 4). These results are comparable to those obtained withnative P2-G6PDH purified from barley roots [19,20] and potato [23].However it should be noted that only very high DTT levels are ableto induce enzyme inhibition, presumably because it is a non-physiological electron donor. A faster effect could be obtained uti-lising reduced thioredoxins to mimic physiological conditions [22].It cannot be ruled out that P2-isoforms retain some activity evenwhen completely reduced. This should be further investigated inthe future.

3. Conclusions

To date, most if not all plastidial G6PDHs isolated from planttissues, and recombinant proteins so far purified, are recoveredwith a low yield and exhibit a high instability, thus causing manyproblems in describing the basic properties of this enzyme.

In this work, different strategies have been tested in order toproduce a stable recombinant P2-G6PDH from barley after cloningthe cDNA sequence in two different vectors, testing a variety ofpurification procedures. Soluble His-tagged protein purified frombacterial lysates was enough stable to allow kinetic investigations.The kinetic properties exhibited by the purified recombinantenzyme are similar to those determined for P2-G6PDH previouslypurified from roots of hydroponic grown barley and isolated root

Fig. 3. Kinetic properties of His6-HvP2-G6PDH. A e Primary double reciprocal plot of the effect of varied [NADPþ]: C, 6.125 mM; -, 12.5 mM; :, 25 mM ;, 50 mM; A, 100 mM; B,150 mM, by varying G6P concentration from 0.15 to 3 mM. B e Primary double reciprocal plot of the effect of varied [G6P]: C, 0.15 mM; - 0.3 mM; :, 0.75 mM; ;, 1.5 mM; A,2.25 mM; B, 3 mM, by varying NADPþ concentration from 1.5 to 150 mM. C e Secondary plots of slopes of primary plots vs. 1/[NADPþ]. Each point represents the average of threereplicates � standard error. D e Secondary plots of intercepts of primary plots vs.1/[G6P]. Each point represents the average of three replicates � standard error. The graphs arerepresentative of at least three independent purifications and kinetic parameter determinations.

M. Cardi et al. / Plant Physiology and Biochemistry 73 (2013) 266e273270

plastids (Table 1). From these reliable kinetic results, we havedetermined that the barley plastidial P2-type G6PDH exhibited anordered sequential reaction mechanism and that its activity ismodified upon reduction.

This recombinant HvP2-G6PDH will be suitable for furtherstudies investigating with higher accuracy the biochemical char-acteristics of this protein such as the sensitivity of P2-G6PDH iso-forms to the reduction by different classes of plastidialthioredoxins. It may also help determining the 3D structure of the

Table 2Initial-rate parameters [35] and their ratios for G6PDH reaction for NADPþ and G6P.

Plot against: 40 (s) 4NADPþ (mM s) 4G6P (mM s) 4NADP

þG6P (mM2

1/G6P 0.352 � 0.028 0.179 � 0.023 0.348 � 0.018 7.51 � 0.3661/NADPþ 0.348 � 0.018 0.168 � 0.006 0.352 � 0.028 7.61 � 0.669Mean 0.35 0.1735 0.35 7.56

Kinetic data were determined by secondary plots of primary plots (double reciprocal) againdicated in row 3. Standard errors and linear regressions were calculated by Graphpad

regulatory plastidial enzymes pending that suitable crystals can beobtained.

4. Materials and methods

4.1. Cloning and expression of P2-G6PDH in E. coli BL21(DE3)

Three specific primers were designed for cloning the sequencecoding for the mature form of the plastidial G6PDH from barley

s) 4NADPþ

G6P/4NADPþ (mM) 4NADP

þG6P/4G6P (mM) 4NADP

þG6P/4G6P 4NADP

þ (s�1)

41.95 21.58 120.5545.29 21.62 130.0843.62 21.6 125.32

inst both reciprocal of G6P (row 1) and NADPþ (row 2). Themean values obtained arePrism 6.0 software.

Fig. 4. Effect of DTTred on the activity of purified recombinant His6-HvP2-G6PDH: B,control (no DTT); ,, 2 mM DTT;O, 5 mM DTT;P, 10 mM DTT; >, 25 mM DTT.Each point represents the average of three replicates � standard error.

M. Cardi et al. / Plant Physiology and Biochemistry 73 (2013) 266e273 271

roots (HvP2-G6PDH), i.e., devoid of 99 amino acids at the N-ter-minus, corresponding to the putative targeting sequence (seeSection 2.1). The restriction sites necessary for cloning are under-lined in the sequences.

HvP2-99-1-forward (NcoI) 50CCCCCCCATGGCGTCTGTTAGCATCACTGTG 30

(N ter MASVSITV) used to clone in pET-3d.HvP2-99-2-forward (NdeI) 50CCCCCCCCCATATGGCGTCTGTTAGC

ATC 30

(N ter MASVSI) used to clone in pET15-b.HvP2-reverse (BamHI) 50 CCCCGGATCCCTAGTGTTCCGAGCCGCCC

AG 30

(used both for pET-3d and pET-15b cloning).The nucleotidic sequence was first amplified by PCR using Pfu

DNA polymerase (Promega), barley root cDNAs and the appropriateprimers and subsequently cloned into the pET3d and pET15bexpression plasmids for the production respectively of an untaggedprotein and an N-terminally 6-His-tagged protein. The recombi-nant plasmids were used to transform the E. coli BL21(DE3) strain.Culture of 3 L deriving from one ampicillin resistant (50 mg/ml)colony first grown at 37 �C have been induced in the exponentialphase by 100 mM isopropyl-thiogalactopyanoside (IPTG) at 20 �Cfor 16 h.

Bacteria were harvested by centrifugation, suspended in bufferA (50 mM TriseHCl, 1 mM EDTA, 5 mMMgCl2 10% glycerol, pH 8.0)for pET3d, or buffer B (50mMKH2PO4e K2HPO4, 0.3MNaCl,10mMimidazole, buffered to pH 8.0) for pET15b.

4.2. Purification of untagged recombinant HvP2-G6PDH

Bacteria were lysed by sonication and the soluble and insolublefractions were separated by centrifugation (16,000 g, 30 min). Therecombinant HvP2-G6PDH was found in both the soluble fractionand inclusion bodies. The soluble protein fraction was loaded on aResource-Q column connected to the AKTA Prime plus system (GEHealthcare).

The columnwas washed with buffer A and proteins were elutedby applying a 0e400 mM NaCl linear gradient at a flow rate of1 ml min�1; 2 ml eluted fractions were assayed for G6PDH activity.

Active fractions were pooled and the partially purified enzymewas applied to a HiTrap-Blue HP column, equilibratedwith buffer A.The recombinant P2-G6PDH was eluted with buffer A plus 5 mMNADPþ and 1.5 M NaCl; fractions showing G6PDH activity werepooled and desalted in buffer A (thus removing NaCl and NADPþ) toestimate the kinetic parameters.

For the purification of the recombinant protein from the insol-uble fraction, a modified procedure from the method describedpreviously [32] was utilised: inclusion bodies were resuspended inbuffer A containing 20 mM reduced DTT and 7 M guanidine-HCl pH7.9, sonicated, kept under N2 and shaken for 4 h at 4 �C. The de-natured protein solution was diluted 20-fold in buffer A containing200 mM NADPþ, 0.5 M arginine, 10 mM oxidized DTT, pH 7.9. Thesolution was kept under N2 at room temperature for 5e10 days(usually one week). After centrifugation, the refolded protein wasconcentrated about 8e10 times by ultrafiltration under N2 (AmiconPW10 filters). The final solution was desalted and assayed forG6PDH activity and for determination of kinetic and immunologicalproperties.

4.3. Purification of His-tagged recombinant HvP2-G6PDH

Cells were lysed and the samples treated as described before;the His-tagged recombinant HvP2-G6PDH was found in both sol-uble fraction and inclusion bodies.

The soluble protein fractionwasfiltered (0.22mm)and applied to aHiTrap FF crude column. After washing with buffer B, proteins wereeluted by applying buffer C (0.5 M imidazole, 50 mM KH2PO4 e

K2HPO4, 0.3MNaCl, pH8.0). Active fractionswerepooled, desalted (toremove imidazole) in 30mM TriseHCl,100mMNaCl, 5% glycerol, pH8.0, and used to estimate kinetic parameters.

Alternatively, the recombinant protein was extracted from in-clusion bodies, then refolded and purified. The inclusion bodieswere resuspended in cold isolation buffer: 20 mM TriseHCl con-taining 2 M urea, 0.5 M NaCl, 2% Triton-X100, pH 8.0; then soni-cated as described before, and centrifuged for 10 min at 4 �C. Thepellet was resuspended in 3 ml of cold isolation buffer. After afurther cycle of sonication and centrifugation, inclusion bodieswere washed in binding buffer: 20 mM TriseHCl containing 6 Mguanidine hydrochloride, 0.5 M NaCl, 5 mM imidazole, 1 mM b-mercaptoethanol, pH 8.0. After 1 h under gentle shaking at roomtemperature, the sample was centrifuged for 15 min at 25,000 g at4 �C, the supernatant was filtered (0.22 mm), pH adjusted at 8.0 andloaded on a 1 ml HisTrap FF column connected to an AKTA Primeplus (GE Healthcare), using stored on-column refolding and puri-fication programs, following manufacturer’s instructions. Therefolding buffer was 20 mM TriseHCl, 0.5 M of NaCl, 5 mM imid-azole, 1 mM b-mercapthoethanol, pH 8.0 and the buffer used toelute the proteinwas 20mM TriseHCl, 0.5 M NaCl, 0.5 M imidazole,1 mM b-mercapthoethanol, pH 8.0.

4.4. Electrophoresis and western blotting analysis

SDS-PAGE (180 V/40 mA) was performed using 10% acrylamideresolving gel with a 4% stacking gel.

For western blotting analysis, the separated polypeptides weretransferred (2 h e 25 V, 300 mA) on a Hybond membrane (GEHealthcare). After transfer, the membrane was incubated withprimary G6PDH antibody from potato for P1-, P2- and Cy-G6PDHisoforms [23]. These antibodies have proven to react with anddiscriminate the different G6PDH isoforms present in variousbarley organs, as described in previous studies [16,19,25]. Afterwashing, membranes were developed as previously described [29].

4.5. Enzyme activity assay and determination of the kineticparameters

G6PDH activity was assayed as previously described [16].The measurements were made within 36 h from purification,

using 5e10 ml of purified enzyme, desalted to remove the NADPþ togive a linear activity for at least the first 2 min of reaction. Catalytic

M. Cardi et al. / Plant Physiology and Biochemistry 73 (2013) 266e273272

rates were measured as nmoles of NADPþ formed per min, or persec, and expressed per ml of extract, or per mg of protein.

KmG6P and KmNADPþ were measured on purified enzyme by

varying (G6P) from 0 to 30 mM, and (NADPþ) from 0 to 150 mM at afixed concentration of the other substrate. The inhibition constantsfor NADPH (KiNADPH) were calculated varying NADPH concentrationfrom 0 to 150 mM at sub-saturating NADPþ concentration of 15, 30,50 mM, maintaining the saturating G6P concentration at 3 mM.

4.6. Determination of protein content

Protein concentrations were determined using the Bio-Radprotein assay based on the Bradford method [38] with bovineserum albumin as the standard. During the chromatography steps,proteins were monitored by their absorbance at 280 nm.

4.7. Software and statistical analyses

Data were analysed using functions available in GraphPadPrism� 6.0, Jandel SigmaPlot�, andMicrosoft Excel�. The results areexpressed as the average of three separate determinations (eachobtained in at least duplicate) � standard error, unless otherwisestated.

G6PDH activities were plotted, and kinetic parameters calcu-lated by both Jandel SigmaPlot� and GraphPad Prism� software,using kinetics formulae provided.

Gel images were acquired by a high resolution scanner andcomputed using the Corel Draw� suite.

Acknowledgements

The authors wish to dedicate this paper to Mauro Cardi.The Authors thank very much Antje von Schaewen (Muenster e

Germany) for the generous gift of G6PDH antibodies.Research supported by Legge Regionale della Campania 5/2002

(2007).Manuela Cardi acknowledges EGIDE funding from French Min-

istry of Foreign Affairs (grant 672700K; years 2009 and 2010/11);and Project FORGIARE (Formazione Giovani alla Ricerca) V-10/FORG/ST/2012/5 by Compagnia di San Paolo.

UMR Interactions Arbres Microorganismes is supported by theFrench National Research Agency through the Laboratory ofExcellence ARBRE (ANR-12-LABXARBRE-01)

Appendix A. Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.plaphy.2013.10.008.

References

[1] U.K. Wendt, R. Hauschild, C. Lange, M. Pietersma, I. Wenderoth, A. vonSchaewen, Evidence for functional convergence of redox regulation inG6PDH isoforms of cyanobacteria and higher plants, Plant Mol. Biol. 40(1999) 487e494.

[2] I. Orita, T. Sato, H. Yurimoto, N. Kato, H. Atomi, T. Imanaka, Y. Sakai, Theribulose monophosphate pathway substitutes for the missing pentose phos-phate pathway in the archaeon Thermococcus kodakaraensis, J. Bacteriol. 188(2006) 4698e4704.

[3] D.T. Dennis, Y. Huang, F.B. Negm, Glycolysis, the pentose phosphate pathwayand anaerobic respiration, in: D.T. Dennis, D.H. Turpin, D.D. Lefebvre,D.B. Layzell (Eds.), Plant Metabolism, Longman, London, 1997, pp. 105e127.

[4] Y. Nemoto, T. Sasakuma, Specific expression of glucose-6-phosphate dehy-drogenase (G6PDH) gene by salt stress in wheat (Triticum aestivum L.), PlantSci. 158 (2000) 53e60.

[5] X. Wang, Y. Ma, C. Huang, Q. Wan, N. Li, Y. Bi, Glucose-6-phosphate dehy-drogenase plays a central role in modulating reduced glutathione levels inreed callus under salt stress, Planta 227 (2008) 611e623.

[6] L. Sindelar, M. Sindelarova, L. Burketova, Changes in activity of glucose-6-phosphate and 6-phosphogluconate dehydrogenase isozymes upon potatovirus Y infection in tobacco leaf tissue and protoplasts, Plant Physiol. Biochem.37 (1999) 195e201.

[7] J. Scharte, H. Schön, Z. Tjaden, E. Weis, Photosynthesis and carbohydratemetabolism in tobacco leaves during an incompatible interaction with Phy-tophthora nicotianae, Plant Cell Environ. 28 (2005) 1421e1435.

[8] J. Scharte, H. Schön, Z. Tjaden, E. Weis, A. von Schaewen, Isoenzymereplacement of glucose-6-phosphate dehydrogenase in the cytosol im-proves stress tolerance in plants, Proc. Nat. Acad. Sci. U. S. A. 106 (2009)8061e8066.

[9] R. Wang, M. Okamoto, X. Xing, N.M. Crawford, Microarray analysis of thenitrate response in Arabidopsis roots and shoots reveals over onethousand rapidly responding genes and new linkages to glucose,trehalose-6-P, iron and sulfate metabolism, Plant Physiol. 132 (2003)556e567.

[10] T. Meyer, C. Holscher, C. Schwoppe, A. von Schaewen, Alternative targeting ofArabidopsis plastidial glucose-6-phosphate dehydrogenase G6PD1 involvescysteine dependent interaction with G6PD4 in the cytosol, Plant J. 66 (2011)745e758.

[11] M. Cardi, K. Chibani, D. Cafasso, N. Rouhier, J.-P. Jacquot, S. Esposito, Abscisicacid effects on activity and expression of different glucose 6 phosphate de-hydrogenase isoforms in barley (Hordeum vulgare), J. Exp. Bot. 62 (2011)4013e4023, http://dx.doi.org/10.1093/JXB/ERR100.

[12] N.J. Kruger, A. von Schaewen, The pentose phosphate pathway: structure andorganisation, Curr. Op. Plant Biol. 6 (2003) 236e246.

[13] J. Krook, D. Vreugdenhil, C. Dijkema, L.H.W. van der Plas, Sucrose and starchmetabolism in carrot (Daucus carota L.) cell suspensions analysed by 13C-labelling: indications for a cytosol and a plastid-localized oxidative pentosephosphate pathway, J. Exp. Bot. 49 (1998) 1917e1924.

[14] P.M. Debnam, M.J. Emes, Subcellular distribution of enzymes of the oxidativepentose phosphate pathway in root and leaf tissues, J. Exp. Bot. 50 (1999)1653e1661.

[15] A. von Schaewen, G. Langenkamper, K. Graeve, I. Wenderoth, R. Scheibe,Molecular characterisation of the plastidial glucose-6-phosphate dehydroge-nase from potato in comparison to its cytosolic counterpart, Plant Physiol. 109(1995) 1327e1335.

[16] S. Esposito, G. Guerriero, V. Vona, V. Di Martino Rigano, S. Carfagna, C. Rigano,Glutamate synthase activities and protein changes in relation to nitrogennutrition in barley: the dependence on different plastidic glucose-6P dehy-drogenase isoforms, J. Exp. Bot. 56 (2005) 55e64, http://dx.doi.org/10.1093/jxb/eri006.

[17] J.S. Knight, M.J. Emes, P.M. Debnam, Isolation and characterisation of a full-length genomic clone encoding a plastidial glucose 6-phosphate dehydroge-nase from Nicotiana tabacum, Planta 212 (2001) 499e507.

[18] S. Wakao, C. Benning, Genome-wide analysis of glucose-6-phosphate dehy-drogenase in Arabidopsis, Plant J. 41 (2005) 243e256.

[19] S. Esposito, G. Massaro, V. Vona, V. Di Martino Rigano, S. Carfagna, C. Rigano,Ammonium induction of a novel isoform of glucose-6P dehydrogenase inbarley roots, Physiol. Plant. 113 (2001) 469e476.

[20] S. Esposito, S. Carfagna, G. Massaro, V. Vona, V. Di Martino Rigano, Glucose-6Pdehydrogenase in barley roots: kinetic properties and localisation of theisoforms, Planta 212 (2001) 627e634.

[21] P. Schurmann, J.-P. Jacquot, Plant thioredoxin systems revisited, Annu. Rev.Plant Physiol. Plant Mol. Biol. 51 (2000) 371e400.

[22] G. Née, M. Zaffagnini, P. Trost, E. Issakidis-Bourguet, Redox regulation ofchloroplastic glucose-6-phosphate dehydrogenase: a new role for f-type thi-oredoxin, FEBS Lett. 17 (2009) 2827e2832.

[23] U.K. Wendt, I. Wenderoth, A. Tegeler, A. von Schaewen, Molecular charac-terization of a novel glucose-6-phosphate dehydrogenase from potato (Sola-num tuberosum L.), Plant J. 23 (2000) 723e733.

[24] C.G. Bowsher, D.P. Hucklesby, M.J. Emes, Nitrite reduction and carbohydratemetabolism in plastids purified from roots of Pisum sativum, Planta 177 (1989)359e366.

[25] S. Esposito, G. Massaro, V. Vona, V. Di Martino Rigano, S. Carfagna, Glutamatesynthesis in barley roots: the role of the plastidic glucose-6-phosphate de-hydrogenase, Planta 216 (2003) 639e647.

[26] D. Hutchings, S. Rawsthorne, M.J. Emes, Fatty acid synthesis and the oxidativepentose phosphate pathway in developing embryos of oilseed rape (Brassicanapus), J. Exp. Bot. 56 (2005) 577e585.

[27] K. Honjoh, A. Mimura, E. Kuroiwa, T. Hagisako, K. Suga, H. Shimizu, R.S. Dubey,T. Miyamoto, S. Hatano, M. Iio, Purification and characterization of two iso-forms of glucose 6-phosphate dehydrogenase (G6PDH) from Chlorella vulgarisC-27, Biosci. Biotechnol. Biochem. 67 (2003) 1888e1896.

[28] S. Esposito, G. Guerriero, V. Vona, V. Di Martino Rigano, S. Carfagna, C. Rigano,Glucose-6P dehydrogenase in Chlorella sorokiniana (211/8k): an enzyme withunusual characteristics, Planta 223 (2006) 796e804, http://dx.doi.org/10.1007/s00425-005-0110-2.

[29] M. Ferrara, G. Guerriero, M. Cardi, S. Esposito, Biochemical characterisation ofglucose-6P dehydrogenase from the psychrophilic green alga Koliellaantarctica, Extremophiles 17 (2013) 53e62, http://dx.doi.org/10.1007/s00792-012-0492-6.

[30] R. Scheibe, A. Geissler, K. Fickenscher, Chloroplast glucose-6-phosphate de-hydrogenase: Km shift upon light modulation and reduction, Arch. Biochem.Biophys. 274 (1989) 290e297.

M. Cardi et al. / Plant Physiology and Biochemistry 73 (2013) 266e273 273

[31] D.W. Lee, J.K. Kim, S. Lee, S. Choi, S. Kim, I. Hwang, Arabidopsis nuclear-encoded plastid transit peptides contain multiple sequence subgroups withdistinctive chloroplast-targeting sequence motifs, Plant Cell 20 (2008) 1603e1622.

[32] X.T. Wang, P.C. Engel, An optimised system for refolding of human glucose-6-phosphate dehydrogenase, BMC Biotechnol. 19 (2009), http://dx.doi.org/10.1186/1472-6750-9-19.

[33] W.T. Wang, W.N.S. Au, V.M.S. Lam, P.C. Engel, Recombinant humanglucose-6-phosphate dehydrogenase. Evidence for a rapid equilibriumrandom-order mechanism, Eur. J. Biochem. 269 (2002) 3417e3424.

[34] N.N. Ulusu, B. Tandogan, F.E. Tezcan, Kinetic properties of glucose-6-phosphate dehydrogenase from lamb kidney cortex, Biochimie 87 (2005)187e190.

[35] K. Dalziel, Initial steady state velocities in the evaluation of enzyme-coenzyme-substrate reaction mechanisms, Acta Chem. Scand. 11 (1957) 1706e1723.

[36] H.R. Levy, Glucose-6-phosphate dehydrogenases, in: A. Meister (Ed.),Advances in Enzymology, vol. 48, John Wiley and Sons, New York, 1979,p. 97.

[37] I. Wenderoth, R. Scheibe, A. von Schaewen, Identification of thecysteine residues involved in redox modification of plant plastidialglucose-6-phosphate dehydrogenase, J. Biol. Chem. 272 (1997) 26985e26990.

[38] M.M. Bradford, A rapid and sensitive method for the quantitation of micro-gram quantities of protein utilizing the principle of protein-dye binding, AnalBiochem. 72 (1976) 248e254.