ISSN 00268933, Molecular Biology, 2010, Vol. 44, No. 3, pp. 415–419. © Pleiades Publishing, Inc., 2010.

CELL MOLECULAR BIOLOGY UDC 577.21

Oligomerization Studies of Leuconostoc mesenteroides G6PD Activity after SDSPAGE and Blotting1 S. Ravera, D. Calzia, A. Morelli, and I. Panfoli Department of Biology, University of Genoa, Viale Benedetto XV3, Genova 16132, Italy; email: [email protected] Received September 11, 2009; in final form, October 5, 2009

Abstract—Glucose6phosphate dehydrogenase (G6PD) is a ubiquitous enzyme catalyzing the oxidation of Dglucose 6phosphate to Dglucono⎯lactone 6phosphate, in the first step of the pentose phosphate path way. Based on the currently available structural information on Leuconostoc mesenteroides G6PD, it is believed that the enzyme only works as a homodimer. Here we show that both after nondenaturing and after denaturing electrophoretic separation (SDSPAGE) and blotting L. mesenteroides G6PD retains its complete catalytic activity. In the two latter cases the molecular weight of the band corresponded to that of a G6PD monomer. Conversely, when the same technique was applied to G6PD from Saccharomyces cerevisiae, another fermentative organism, the monomer activity was not detectable after SDSPAGE and blotting. The results are discussed in terms of molecular evolution of the oligomeric state in the various G6PD sources. DOI: 10.1134/S002689331003009X Key words: blotting, dimer, glucose6phosphate dehydrogenase, Leuconostoc mesenteroides, monomer, SDSPAGE 1

INTRODUCTION

Glucose6phosphate dehydrogenase (G6PD, [EC 1.1.1.49]) is a ubiquitous housekeeping enzyme, present in prokaryotes, plants and animals [1]. It is the first enzyme of the pentose phosphate pathway regu lating its oxidative branch. G6PD catalyzes the oxida tion of glucose 6phosphate (G6P) to 6phosphoglu cono–δlactone with the reduction of NADP+ to NADPH, providing cells with pentose sugars for nucleotide and nucleic acid synthesis and NADPH which is the principal modulator of the intracellular redox potential and is required for lipogenesis and detoxification [2]. Leuconostoc mesenteroides, an anaer obic Gram positive bacterium, does not realize the complete glycolytic pathway, in fact it metabolizes glu cose by heterolactic fermentation. In this species the oxidative enzymes of the pentose phosphate pathway serve either for synthetic or for catabolic reactions, depending on the physiological conditions. L. mesenteroides G6PD (ExPASy entry: P11411) is a homodimer consisting of identical subunits with Mr 54.316 Da, with two active sites per dimer, putatively located in a pocket between the coenzyme binding domain and the large β + α domain in each subunit [2]. It is the only enzyme that utilizes both NAD+ and NADP+, in contrast to the enzymes from eukaryotic and most of prokaryotic cells, which are NADP+spe cific or preferring [1]. L. mesenteroides G6PD selects either NAD+ or NADP+ depending on the demands 1 The article is published in the original.

for catabolic or anabolic metabolism; the NAD+ reac tion is preferred at high G6P concentrations [3]. The NADH generated from the NAD+linked reaction is used for generating lactate, ethanol and ATP, while the NADP+linked reaction yields NADPH required for fatty acid synthesis. Multiple sequence alignment of over 100 currently known G6PDs from different organisms shows that the sequence identity varies from 30 to 94% [4]. All G6PDs reveal three conserved regions: a nineresidue peptide (RIDHYLGKE, resi dues 198–206 of the human enzyme), a nucleotide binding fingerprint (GxxGDLA, residues 38–44 of the human enzyme) and the sequence EKPxG (residues 170–174 of the human enzyme). Within the nineres idue peptide in L. mesenteroides G6PD the Asp, His and Lys have been shown to be important in G6P binding and catalysis [5]. Lys205 has been implicated in binding and catalysis in the human enzyme [6]. The nucleotide fingerprint has been associated with coen zyme binding [7] and the function of these residues has been elucidated thanks to the knowledge of the L. mesenteroides G6PD structure [2, 8]. However, apart from the conserved amino acids in the active site, var ious G6PDs differ in their overall amino acid compo sition and kinetic properties [9]. L. mesenteroides G6PD (485 residues in the subunit) has been cloned and expressed in E. coli as a soluble protein [10]. It was found that the L. mesenteroides enzyme is the only known G6PD that does not contain cysteine [11]. Several studies report G6PD activity after both nondenaturing and denaturing polyacrylamide gel

415

416

RAVERA et al.

electrophoresis (PAGE) in various aerobic and anaer obic bacterial species, [12–14]. Lacks and Spring horns [14] observed that the activity of L. mesenteroides G6PD is detectable by both nondenaturing and denaturing electrophoresis, even though in the second case the residual activity is decreased by about 80%. In this work first of all we would like to report the improvement of the detection of L. mesenteorides G6PD after SDSPAGE, thanks to the use of a proto col for SDS removal from gels [15]. Second, we show that the L. mesenteorides G6PD retains its catalytic activity after denaturing separation and blotting. Moreover, this capacity is not universal for G6PDs, because the G6PD from Saccharomyces cerevisiae, another unicellular fermenting organism, in the same conditions loses its activity. EXPERIMENTAL Chemicals. The nucleotides and the pure L. me senteroides glucose6phosphate dehydrogenase (G6PD) (no. G5760) were purchased from Sigma Aldrich (Sigma, St. Louis, MO). Pure S. cerevisiae G6PD was purchased from Roche (no. 11847321, Roche Diagnostic GmbH, Mannheim, Germany). Page Ruler Prestained Protein Ladder Plus from Fer mentas (no. 1811, Fermentas Life Sciences, Burling ton, Ontario) was used as protein molecular weight marker. All other chemicals were of reagentgrade purity. Ultrapure water (MilliQ, Millipore, Billerica, MA) was used throughout. Safety precautions were taken for chemical hazards in carrying out the experi ments described below. Native polyacrylamidegel electrophoresis. Native polyacrylamidegel electrophoresis was carried out according to Laemmli [16], except for the absence of SDS, βmercaptoethanol and EDTA. Considering that the isoelectric point (pI) of G6PD is around 4.5– 5.9, the pH of gels and of the Running Buffer was set to 8.3. The optimal acrylamide concentration for on gel G6PD detection was 5% and therefore the stacking gels were not used. The runs were performed at 4°C, at 20 mA for each gel, for 120–150 min in a Mini Pro tean III (BioRad, Hercules, CA, USA) apparatus [15]. Nondenaturing 4x Sample Buffer (40% (w/v) sucrose in 125 mM TrisHCl (pH 6.8)) was added to the samples (3 μg of the protein for each lane). The Running Buffer was: 0.05 M Tris, pH 8.0; 0.4 M gly cine and 1.8 mM EDTA. SDSPAGE electrophoresis. Denaturing electro phoresis was performed using the Laemmli protocol with minor modifications [16]. During the electro phoretic run the current was kept low (no more than 20 mA per gel) [15]. Samples (3 μg of protein for each lane) were electrophoresed in 10% polyacrylamide gel under reducing conditions in a Mini Protean III (Bio Rad, Hercules, CA, USA) apparatus. Polyacrylamide concentration for the Resolving (7 × 8.3 × 0.1 cm, at pH 8.8) and the Stacking gel (1.5 × 8.3 × 0.1 cm, at

pH 6.8) gels was 10 and 5% (w/v), respectively. The samples were boiled for 5 min with 1/4 of the total vol ume of the 4x Sample Buffer (SB) (40% (w/v) sucrose, 8% SDS (w/v) in 125 mM TrisHCl (pH 6.8), 1.25% (v/v) βmercaptoethanol). Bromophenol blue (0.008%, w/v) was the tracking dye, that was allowed to get out of the gels before stopping the run. The run was performed at 4°C, at 20 mA for each gel, for 120– 150 min with the Running Buffer (0.05 M Tris, pH 8.0; 0.4 M glycine; 1.8 mM EDTA and 0.1% SDS). The gels were stained using the colloidal Coomassie G250 technique from Candiano et al. [17]. In some cases in order to stabilize the dimer during the denaturing PAGE, the sample was incubated with 5 mM ethylene glycolbis (succinic acid Nhydroxysuccinimide ester (EGS)) for 30 min, and then quenched with 0.2 M glycine before adding the SB [18]. SDS removal from gel after SDSPAGE. After the SDSPAGE run in order to remove the SDS from the gel, it was incubated for 1 h at 37°C with gentle agita tion on a shaker rotating at 50–60 rpm with 100 ml of the SDS removal buffer containing 2% w/v casein, 0.04 M TrisHCl, and 2 mM EDTA (pH 8.0) with fre quent washings [15]. Blotting. The samples separated by SDSPAGE as described above were transferred onto a nitrocellulose (NC) membrane by the electroblotting technique at 400 mA for 1 h in a Trisglycine buffer (50 mM Tris, 380 mM glycine) with 20% methanol. The NC was saturated with 5% BSA in a Tris Buffered Saline (TBS) over night at 4°C. After two washes with TBS the solu tion for G6PD activity detection could be added to the NC. Western blot analysis. After the blotting the NC was incubated with antibodies against G6PD (SigmaAld rich) (diluted 1 : 200 in PBS) for 3 h at RT. The sec ondary antibodies were also from SigmaAldrich. Detection of G6PD activity. In order to detect the G6PD activity the gels or the NC were incubated at 37°C in the dark in a Reaction Mixture containing: 2.5 mg G6F; 2.5 mg NADP; 2.5 mg phenazine meth osulfate, 3.75 mg Nitro Blue Tetrazolium; 0.2 M Tris, pH 8.0; 0.5 ml MgCl2 (total volume of 12 ml). After a few minutes of incubation dark blue bands appeared, the reaction was stopped and the gel was fixed with 50% ethanol [14]. RESULTS Figure 1 shows the detection of G6PD activity of the commercial homogeneous L. mesenteroides and S. cerevisiae G6PD ingel after nondenaturing PAGE and subsequent incubation with the Reaction Mixture (see Experimental). The dark bands represent the for mazan formed from the Nitro Blue Tetrazolium by the G6PD activity of the L. mesenteroides (lane 1), and S. cerevisiae enzymes (lane 2). It can be supposed that in each sample the band represents a dimer, because the two monomers did not separate in the nondena MOLECULAR BIOLOGY

Vol. 44

No. 3

2010

OLIGOMERIZATION STUDIES OF Leuconostoc mesenteroides G6PD ACTIVITY (a)

1

1

2

Fig. 1. Native PAGE for G6PD activity detection. Lane 1 —pure Leuconostoc mesenteroides G6PD (Sigma); lane 2 —pure Saccharomyces cerevisiae G6PD (Roche Diag nostic).

turing gel. In Fig. 2 panel (a) shows an SDSPAGE gel stained with colloidal Coomassie G250 with the sam ples: the purified S. cerevisiae G6PD (lane 1) and L. mesenteroides G6PD (lane 2), respectively. Molecu lar weight standards (Mr) were loaded in lane 3. Panel (b) shows the level of G6PD activity measured post the removal of SDS by several washings with casein after the SDSPAGE as previously reported [15]. Casein washing slightly reduced the color of the Mr standard. The L. mesenteroides G6PD (lane 2) was detected in gel, while in the S. cerevisiae (lane 1) the activity band was not visible. The apparent G6PD molecular weight of the activity band in lane 2 was 55 kDa, which is very close to that of the L. mesenteroides monomer reported in ExPASy (P11411). Panel (c) shows the activity of the L. mesenteroides G6PD after SDSPAGE without removing SDS. In this case the enzyme activity was not detectable. In Fig. 3 G6PD activity detection on NC is shown. Like in the case of the SDSPAGE gel, only the L. mesenteroides G6PD band (lane 2) was vis ible and its molecular weight corresponded to that of the L. mesenteroides monomer reported in ExPASy (P11411), i.e. about 55 kDa. Figure 4 presents the detection (panel (a)) of the L. mesenteroides G6PD activity and Western Blot analysis (panel (b)) con ducted after the linkage treatment with EGS. Lane 1 represents the sample without the linkagetreatment, while lane 2 shows the sample after the EGS addition. Both in panel (a) and in panel (b), the molecular weight of the sample in lane 2 corresponds to that of the L. mesenteriodes G6PD dimer, while the molecular weight of the band in lane 1 corresponds to that of the L. mesenteriodes monomer. MOLECULAR BIOLOGY

Vol. 44

No. 3

2010

2

417

(b)

kDа 250 130

130

95

95

72

72

55

55

3

1

2

(c)

kDа 250

3

1

2

Fig. 2. SDSPAGE for G6PD activity detection. Panel (a)—Colloidal Coomassie G250 Staining of a SDSPAGE gel, representative of at least five experiments. Panel (b)— Ingel assay of Leuconostoc mesenteroides and Saccharo myces cerevisiae G6PD after SDSPAGE and SDS removal by Casein washing. Panel (c)—Detection of G6PD activity without casein washing after SDSPAGE. Lane 1—pure S. cerevisiae G6PD (Roche Diagnostic); lane 2—homogeneous L. mesenteroides G6PD (Sigma); lane 3—Molecular weight standard (Fermentas). The fig ure represents at least five experiments.

DISCUSSION L. mesenteroides G6PD, the first enzyme of the pen tose phosphate pathway, is pivotal for a fermenting organism. The activity of the L. mesenteroides G6PD from several sources was detected after nondenatur ing and denaturing electrophoresis [13, 14], but after SDSPAGE the activity decreased by about 80% [14]. We have improved the detection of enzyme activity after SDSPAGE thanks to the removal of the deter gent by washing with an alkaline Casein solution [15]. The G6PD activities from L. mesenteroides and S. cerevisiae were compared, since they are both fer menting unicellular organisms. The nondenaturing

kDа 250 130 95 72 55 1

2

3

Fig. 3. G6PD activity detection on NC after blotting. This figure shows inNC assay of Leuconostoc mesenteroides and Saccharomyces cerevisiae G6PD, after SDSPAGE and blotting. Lane 1—pure S. cerevisiae G6PD (Roche Diag nostic); lane 2—homogeneous L. mesenteroides G6PD (Sigma). The figure represents at least five experiments.

418

RAVERA et al. (a)

(b)

110 kDa

55 kDa

1

2

1

2

Fig. 4. G6PD activity detection after EGS treatment. This figure shows the detection of L. mesenteroides G6PD activity (panel (a)) and Western Blot analysis (panel (b)) using the samples treated with EGS. The loaded samples are: L. mesenteroides G6PD without the EGStreatment (lane 1) and L. mesenteroides G6PD after EGS addition (lane 2).

gel electrophoresis was taken as a positive control. Fi gure 1 shows that the two enzymes migrate in the gel in a similar manner maintaining their activity in the gel. On the other hand after the denaturing SDSPAGE in which the movement of the enzymes from both sources was visualized by the classical Coomassie staining (Fig. 2a), only L. mesenteroides G6PD activity was detectable ingel after casein washing for the removal of the excess of SDS as a single band around 55 kDa (Fig. 2b). The S. cerevisiae G6PD activity was not detectable in the same conditions (Fig. 2b). Nota bly even though it was reported that L. mesenteroides G6PD exerts its catalytic activity as a homodimer [2, 4], the activity signal in the gel at 55 kDa (Fig. 2b) cor responds to the molecular weight of the monomer. This unexpected result suggests that the L. mesenteroides G6PD monomer may retain the complete catalytical activity. On the other hand, the possibility that dimerization of the L. mesenteroides enzyme occurs after the removal of excessive SDS due to diffusion seems unlikely. In fact, the gel mesh would offer sterical hindrance to the reoligomerization of the protein. Also in the case of S. cerevisiae G6PD this reoligomerization does not seem to happen. Anyway it would not be possible to determine the oligomeric state of the enzyme ingel by any available technique, moreover once eluted from the gel the proteins would reoligomerize in any case. This data was confirmed by both WB analysis and by the EGS treatment of the sample (Fig. 4). The results obtained by blotting the proteins from the SDSPAGE gel to a nitrocellulose (NC) sheet (Fig. 3) without the removal of SDS are the same as in Fig. 2, in which the G6PD activity from L. mesenteroi

des was detectable after blotting, while the activity of the enzyme from S. cerevisiae was not. Unlike the gel, proteins blotted onto NC cannot diffuse and remain stuck in the state in which they were during the run, so reoligomerization may be ruled out, suggesting that at least in this latter case the monomer retains its com plete catalytical activity. It may be supposed that each L. mesenteroides sub unit has a complete functional catalytic activity, unlike the eukaryotic G6PD for which dimeric or higher oli gomeric states are necessary for activity, i.e. to form the complete site for G6P binding [4]. Structural stud ies on the L. mesenteroides G6PD suggest that its active site lies entirely within the monomer [2, 5]. The reason for the monomer activity maintained in L. mesenteroides G6PD would be its amino acid com position which differs from the eukaryotic G6PD from both unicellular and multicellular organisms. In fact, it can be observed that the sequence of the 11residue peptide containing the reactive Lys in L. mesenteroides G6PD has very few similarities to the same structure around the same Lys in the S. cerevisiae [19] and in the human red blood cell enzyme [9]. In the compared sequences apart from the lysine residues placed at position 0, the only similarity are the leucine residues at the position 2. The differences in the amino acid structure of the active site may also allow the alternative use of NAD+ or NADP+, which is only typical for the L. mesenteroi des G6PD. The eukaryotic and most prokaryotic G6PD are in fact NADPspecific or NADPprefer ring [1]. The NADP+ reaction occurs in a specific order, with NADP+ binding before the G6PD, con versely the binding order in the NAD+ reaction is ran dom and there is evidence that a conformation change occurs when G6P or NAD+ binds [8]. These mecha nisms seem to provide a basis for the selection of the coenzyme and hence the direction of metabolism for the organism. Such flexibility, necessary for the partic ular metabolic requirements of L. mesenteroides, may represent an ancestral form of the enzyme, which would account for its ability to retain whole activity as a monomer. The subsequent evolution of the enzyme may have privileged a more specific coenzyme need and a mutual specialization of the two subunits. Such results would point to a possible structural evolution of the G6PD enzyme from prokaryotes to eukaryotes. It appears possible that further investigations of these enzymes will facilitate the assignment of the structurefunction relationships, as well as of the evo lutionary and metabolic considerations for glucose6 phosphate dehydrogenases. REFERENCES 1. Levy H.R. 1979. Glucose6phosphate dehydrogena ses. Adv. Enzymol. Relat. 48, 97⎯192. 2. Rowland P., Basak A.K., Gover S., Levy H.R., Adams M.J. 1994. The threedimensional strucuture of glucose6 MOLECULAR BIOLOGY

Vol. 44

No. 3

2010

OLIGOMERIZATION STUDIES OF Leuconostoc mesenteroides G6PD ACTIVITY

3.

4.

5.

6.

7.

8.

9.

10.

phosphate dehydrogenase from Leuconostoc mesenteroi des refined at 2.0 Å resolution. Structure. 2, 1073⎯1087. Levy H.R. 1989. Glucose6phosphate dehydrogenase from Leuconostoc mesenteroides. Biochem. Soc. Trans. 17, 313⎯315. Kotaka M., Gover S., VandeputteRutten L., Au S.W.N., Lam V.M.S., Adams M.J. 2005. Structural studies of glucose6phosphate and NADP+ binding to human glucose6phosphate dehydrogenase. Acta Crystallogr. 61, 495⎯504. Cosgrove M.S., Naylor C.E., Paludan S., Adams M.J., Levy H.R. 1998. On the mechanism of the reaction cat alyzed by glucose 6phosphate dehydrogenase. Bio chemistry. 37, 2759⎯2767. Bautista J.M., Mason P.J., Luzzatto L. 1995. Human glucose6phosphate dehydrogenase: Lysine 205 is dis pensable for substrate binding but essential for catalysis. FEBS Lett. 366, 61⎯64. Levy H.R., Vought V.E., Yin X., Adams M.J. 1996. Identification of an arginine residue in the dual coen zymespecific glucose6phosphate dehydrogenase from Leuconostoc mesenteroides that plays a key role in binding NADP+ but not NAD+. Arch. Biochem. Biophys. 326, 145⎯151. Naylor C.E., Gover S., Basak A.K., Cosgrove M.S., Levy H.R., Adams M.J. 2001. NADP+ and NAD+ binding to the dual coenzyme specific enzyme Leu conostoc mesenteroides glucose6phosphate dehydroge nase: Different interdomain hinge angles are seen in different binary and ternary complexes. Acta Crystal logr. 57, 635⎯648. Bhadbhade M.M., Adams M.J., Flynn T.G., Levy H.R. 1987. Sequence identity between a lysinecontaining peptide from Leuconostoc mesenteroides glucose6 phospahte deydrogenase and an active site peptide from human erythrocyte glucose6phosphate dehydroge nase. FEBS Lett. 211, 243⎯246. Lee W.T., Levy H.R. 1992. Lysine21 of Leuconostoc mesenteroides glucose6phosphate dehydrogenase par

MOLECULAR BIOLOGY

Vol. 44

No. 3

2010

11.

12.

13.

14.

15.

16.

17.

18.

19.

419

ticipates in substrate binding through chargecharge interaction. Protein Sci. 1, 329⎯334. Lee W.T., Flynn T.G., Lyons C., Levy H.R. 1991. Clon ing of the gene and amino acid sequence for glucose6 phosphate dehydrogenase from Leuconostoc mesenteroides. J. Biol. Chem. 266, 13028⎯13034. Combe M.L., Lemeland J.F., PestelCaron M., Pons J.L. 2000. Multilocus enzyme analysis in aerobic and anaerobic bacteria using gel electrophoresisnitro cellulose blotting. FEMS Microbiol. Lett. 185, 169⎯174. Bischoff K.M., Shi L., Kenelly P.J. 1998. The detection of enzyme activity following sodium dodecyl sulfate polyacrylamide gel electrophoresis. Anal. Biochem. 260, 1⎯17. Lacks S.A., Springhorns S.S. 1980. Renaturation of enzymes after polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulphate. J. Biol. Chem. 255, 7467⎯7473. Ravera S., Calzia D., Panfoli I., Pepe I.M., Morelli A. 2007. A simultaneous detection of Mr, weight, and activity of adenylate kinases after electrophoretic sepa ration. Electrophoresis. 28, 291⎯300. Laemmli U.K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature. 227, 680⎯685. Candiano G., Bruschi M., Musante L., Santucci L., Ghiggeri G.M., Carnemolla B., Orecchia P., Zardi L., Righetti P.G. 2004. Blue silver: A very sensitive colloidal Coomassie G250 staining for proteome analysis. Elec trophoresis. 25, 1327⎯1333. Massague J., Guillette B.J., Czech M.P. 1981. Affinity labeling of multiplication stimulating activity receptors in membranes from rat and human tissues. J. Biol. Chem. 256, 2122⎯2125. Jeffery J., Hobbs L., Jornvall H. 1985. Glucose6 phosphate dehydrogenase from Saccharomyces cerevi siae: Characterization of a reactive lysine residue labelled with acetylsalicylic acid. Biochemistry. 24, 666⎯671.