Plant Physiol. (1990) 92, 710-7170032-0889/90/92/071 0/08/$01 .00/0

Received for publication June 9, 1989and in revised form October 5, 1989

Purification by lmmunoadsorption and ImmunochemicalProperties of NADP-Dependent Malic Enzymes from Leaves

of C3, C4, and Crassulacean Acid Metabolism Plants

Marion Fathi' and Claus Schnarrenberger*Institute of Plant Physiology, Cell Biology and Microbiology, Free University of Berlin, Konigin-Luise-Strasse

12-16a, D-1000 Berlin 33 (West), Federal Republic of Germany

ABSTRACT

NADP:malic enzyme from com (Zea mays L.) leaves was puri-fied by conventional techniques to apparent homogeneity asjudged by sodium dodecyl sulfate-polyacrylamide gel electropho-resis. Antibodies raised against this protein in rabbits were puri-fied, coupled covalently to protein A-Sepharose CL-4B, and usedas an immunoaffinity resin to purify the NADP:malic enzymes ofthe C3 plants spinach (Spinacia oleracea L.) and wheat (Triticumaestivum L.), of the Crassulacean acid metabolism (CAM) plantBryophyllum daigremontianum R. Hamed et Perr. de la Bathie andthe C4 plants corn, sugarcane (Saccharum officinarum L.), andPortulaca grandHflora L. Such procedures yielded homogeneousprotein preparations with a single protein band, as judged bysodium dodecyl sulfate-polyacrylamide gel electrophoresis, ex-cept for P. grandiflora L. with two bands. The specific activitiesof the purified proteins ranged between 56 and 91 units (milli-grams per protein). NADP:malic enzyme represented up to 1% ofthe total soluble protein in C4 plants, 0.5% in the CAM plant, andless than 0.01% in C3 plants. In immunotitration tests involvingimmunoprecipitation and immunoinhibition of activity by an anti-serum against the corn leaf enzyme, the NADP:malic enzymes ofcom and sugarcane shQwed virtually full identity of epitopes,while the NADP:malic enzymes of the C3 and CAM plants exhibiteda cross-reaction of one-twentieth and one-fourth by these tests,respectively. The NADP:malic enzyme of P. grandlflora exhibitedcharacteristics more closely related to the enzymes of C3 andCAM plants than to those of C4 plants.

The decarboxylation of malate via malic enzyme may beaccomplished by NADP orNAD as cofactor. In plants at leasttwo different enzymes are known. The NADP:malic enzyme(EC 1.1.1.40) reacts primarily with NADP, has only littleactivity with NAD, and needs Mg2e or Mn2" as cofactor (13,23). The NAD:malic enzyme (EC 1.1.1.39), which is knownto be mitochondrial, reacts predominantly with NAD, toabout 30% ofthis activity with NADP, and requires also Mg2'or Mn2' as cofactor (28). Another type ofNAD:malic enzymewas found in a particular subgroup of C4 plants (NAD:malicenzyme subgroup). This enzyme reacts primarily with NAD,to about 5 to 30% of this activity with NADP, and operateswith Mn2+, which cannot be replaced by Mg2' (14). In this

' Present address: Bundesgesundheitsamt, Thielallee 88-92, D-1000 Berlin 33 (West).

paper we will refer to the NADP:malic enzyme (EC 1.1.1.40)in green leaves.NADP:malic enzyme has an essential function in the me-

tabolism of photosynthetic leaf tissue, although the role isconsidered to vary in plants with different modes of photo-synthesis. Most importantly, this activity is needed in leavesof some C4 plants to split malate into pyruvate and CO2- C4plants have been subgrouped, depending on which enzymedegrades C4 acids, into a NADP:malic enzyme, a NAD:malicenzyme, and a phosphoenolpyruvate carboxykinase type (12).NADP:malic enzyme is present in high activity in theNADP:malic enzyme subgroup of C4 plants and only in lowactivity in the other two subgroups.CAM plants fix CO2 during the night, mainly into C4 acids,

which in turn become degraded into CO2 and a C3 compoundduring the subsequent day period before CO2 is refixed in theCalvin cycle. Either NADP:malic enzyme or phosphoenol-pyruvate carboxykinase may accomplish this reaction (21).Again, the activity of one enzyme is high in one group ofplants and low in the other, while the activity of the otherenzyme is present in the opposite pattern. Finally, C3 plants,which do not require NADP:malic enzyme directly for pho-tosynthetic CO2 fixation, exhibit NADP:malic enzyme activ-ity, too, although in likewise low amounts.NADP:malic enzymes of several plant species and tissues

have been shown to exhibit large differences in kinetic prop-erties. These differences have been discussed by Nishikidoand Wada (20) and Pupillo and Bossi (23) in that one type ofNADP:malic enzyme exists in C3 and CAM plants and innongreen tissues ofC4 plants and another type in green leavesof C4 plants only. Differences may be correlated with the factthat the NADP:malic enzymes of the two types appear to becompartmented in different subcellular locations. In this re-spect NADP:malic enzyme was found in the chloroplasts ofC4 plants (15). For CAM plants, however, reports vary. Inphylloclades of Opuntiaficus-indica, one isoenzyme each wasreported in the cytosol, the chloroplasts, and the mitochondria(19). In Bryophyllum calycinum and Crassula lycopodioides(24) and in Mesembryanthemum crystallinum (30), it wasreported mostly or totally in the cytosol. In C3 plantsNADP:malic enzyme was restricted to the cytosol (8, 27).A comparative study on NADP:malic enzymes was there-

fore initiated to elucidate differences of these enzymes inplant species differing in their C3, C4, and CAM type ofphotosynthesis. It was our goal to purify the NADP:malic

710

PURIFICATION OF NADP:MALIC ENZYMES FROM PLANTS

enzymes from several C3, C4, and CAM plants and to comparetheir structural and kinetic properties. Their subunit structureand immunochemical properties will be analyzed in thispaper. In a subsequent study the kinetic differences of theseenzymes were evaluated (M Fathi, C Schnarrenberger, un-published results).

MATERIALS AND METHODS

Plant Material

Plants of corn (Zea mays L., type Badischer Landmais),sugarcane (Saccharum officinarum L.), and Bryophyllum dai-gremontianum R. Hamed et Perr. de la Bathie were raised ina greenhouse. Spinach (Spinacia oleracea L., type Norveto)was grown in the open. Portulaca oleracea L., Portulacagrandiflora L., and wheat (Triticum aestivum L., type Colibri)were kept in growth chambers with daily cycles of 10 h in 5.5kLux white light (Powerstar Lamps, Osram, HQI, 400 W) at22°C and of 14 h in the dark at 16°C.

Enzyme Assays

All enzyme assays were performed at 25°C in a 1 mLreaction mixture and recorded with an Eppendorf 1101 M(Netherer and Hinz, Hamburg, FRG) at 340 nm or with aCarl Zeiss (Oberkochen, FRG) photometer at 334 nm.NADP:malic enzyme (L-malate:NADP oxidoreductase [ox-aloacetate-decarboxylating], EC 1.1.1.40) from C4 plants wasassayed according to Hatch and Mau (13) in 1 mL of 50 mMTricine/KOH (pH 8.3), 0.1 mm EDTA, 2 mM MgC12, 0.3 mMNADP, and 5 mM L-malate. For the assay of NADP:malicenzyme from C3 and CAM plants, we used an assay mixtureconsisting of 50 mM Hepes/KOH (pH 7.2), 0.1 mM EDTA,0.5 mM MnC12, 0.3 mm NADP, and 5 mM L-malate (10).NAD:malic enzyme (L-malate:NAD' oxidoreductase [oxa-loacetate-decarboxylating], EC 1.1.1.39) was assayed in 25mM Hepes/KOH (pH 7.5), 5 mm dithioerythritol, 0.1 mMEDTA, 0.05 mm coenzyme A, 2 mm MnCl2, 2.5 mm NAD,and 5 mM L-malate (6). One unit of activity corresponds tothe reduction of 1 ,umol of NADP per min. Protein wasdetermined with the Coomassie brilliant blue G-250 reagentaccording to Bradford (4). If samples contained (NH4)2SO4 orpolyvinylpyrrolidone, proteins were treated with 2% sodiumdesoxycholate and precipitated with 24% TCA (2) and deter-mined by the method ofLowry et al. (17). Salt concentrationswere determined by conductimetry.

Conventional Enzyme Purification Procedures

For purification of malic enzyme from corn the procedureby Asami et al. (1) was followed with some modifications. Allhandling of enzymes was performed at 0 to 4°C and, forhomogenization, a buffer was used consisting of 200 mMpotassium phosphate (pH 8.0), 20 mM 2-mercaptoethanol,and 0.5% (w/v) polyvinylpyrrolidone. For preparing an ex-tract of corn and sugarcane leaves, portions of 100 g of leaftissue without midribs were cut with scissors into 1 cm broadstrips and homogenized with a five-fold excess of buffer for 1min in a Waring Blendor (Waring Products Division, New

Hartford, CT) and subsequently for 2 min in an Ultra-Turrax(Janke-Kunkel, Staufen, FRG), each at maximum speed. Thecrude homogenate was squeezed through cheesecloth andclarified by centrifugation for 45 min at 20,000g in a GSArotor of Sorvall RC 2B centrifuge (Du Pont de Nemours, BadNauheim, FRG). For plants other than corn and sugarcane,a twofold excess of buffer was sufficient and the homogeni-zation with the Ultra-Turrax was omitted.For (NH4)2SO4 gradient solubilization, 400 g Celite 545

(Serva, Heidelberg, FRG) and solid (NH4)2SO4 up to 70%final relative saturation were added to 4 L of crude extract.The pH was maintained at 7.0 to 8.0 with 0.5 M KOH. Thesuspension was settled in a 9 cm wide glass column andproteins were solubilized and eluted by a 2-L gradient of 70to 0% (NH4)2SO4 in 0.1 M potassium phosphate (pH 8.0).Fractions with more than 30% of the maximum NADP:malicenzyme activity were pooled, precipitated by 70% (NH4)2SO4,resuspended, and dialyzed in 10 mm potassium phosphateand 10 mm 2-mercaptoethanol (pH 8.0).

Subsequently, the enzyme was bound to a column (3 x 50cm) of DEAE-cellulose (DE32, Whatman, Maidstone, Eng-land) equilibrated with dialysis buffer (see above) and elutedby a linear gradient of 1200 mL of0 to 0.5 M KCI in the samebuffer. Fractions with malic enzyme activity were combined,precipitated by 70% (NH4)2SO4, and dialyzed overnight. Thescales of this and the previous step were changed accordingto the amounts of crude extracts used.The next purification step was performed on a 3 x 14 cm

column of hydroxylapatite (Bio-Rad, Munchen, FRG) equil-ibrated with 10 mm potassium phosphate and 10 mM 2-mercaptoethanol (pH 8.0). After loading the proteins,NADP:malic enzyme was eluted with a 350 mL gradient of 0to 0.4 M potassium phosphate in 10 mM 2-mercaptoethanolat pH 7.5.The last step of purification involved chromatography on

Blue-Sepharose 4B. The coupling of 2 g Cibacron Blue F3G-A (Serva, Heidelberg, FRG) onto 10 g Sepharose 4B (Phar-macia, Freiburg, FRG) was done according to the procedureby Bohme et al. (3). The resin was sedimented into a glasscolumn of 2 x 10 cm, equilibrated with buffer (10 mMpotassium phosphate and 10 mM 2-mercaptoethanol [pH8.0]). The enzyme solution was diluted with 10 mm 2-mer-captoethanol (pH 8.0), up to a conductivity of less than 5 xl0-' S and loaded onto the column. The column was washedwith one volume of column buffer. Elution of NADP:malicenzyme was achieved with a linear gradient of 0 to 0.7 M KCIin column buffer. Fractions with NADP:malic enzyme activ-ity were combined, concentrated with an Amicon ultrafiltra-tion cell (model 203, Amicon, Lexington, MA) and Diafloultrafilter PM 10, and dialyzed. The purified NADP:malicenzyme was stored as a solution of 1 mg protein per mL in10% (v/v) glycerol at -20°C.

Immunization and Purification of Antibodies

Prior to immunization, the purified NADP:malic enzymewas subjected to preparative SDS-PAGE in order to removeminor impurities. Two hundred gg of the purifiedNADP:malic enzyme with either Freund's complete or in-complete adjuvant were mixed and used for immunization of

711

FATHI AND SCHNARRENBERGER

two rabbits by standard procedures. After collection of theantisera (400 mL) antibodies were prepurified by precipitatingtwice with 0 to 50% (NH4)2SO4 at pH 7.5 and, after dialysis,by adsorption of other proteins to a DEAE-cellulose columnequilibrated with 30 mm potassium phosphate (pH 7.5). Pro-teins in the flow-through fractions were concentrated in anAmicon ultrafiltration cell (PM 10 membrane) and stored inbatches at -20°C.

Highly specific anti-NADP:malic enzyme IgGs2 wereobtained by immunoadsorption chromatography on aNADP:malic enzyme-Affi-Gel 10 column. About 2.5 mL ofwet packed Affi-Gel 10 was coupled with 4.2 mg of purifiedNADP:malic enzyme in 5 mL of 0.1 M Hepes/KOH (pH 8.0),for 4 h at 4°C and further processed according to the productinformation manual ofBio-Rad. Then, NADP:malic enzyme-Affi-Gel 10 was equilibrated with 40 mm Hepes/KOH (pH8.0), and agitated gently with 200 mg of prepurified anti-malic enzyme-IgGs in 12.5 mL of 40 mM Hepes/KOH (pH8.0), for 4 h. The material was sedimented into a glass column(1 x 6.5 cm) and washed with 50 mL of 0.4 M NaCl in 40mM Hepes/KOH (pH 8.0), to remove unbound IgGs. Specificanti-NADP:malic-enzyme IgGs were eluted with 4 M MgCl2in 100 mm Hepes/KOH (pH 8.0), at a flow rate of 30 mL h-'and were dialyzed immediately against 2 L of 20 mM Hepes/KOH (pH 8.0), for 2 h and against 2 x 2 L PBS overnight.The IgGs were concentrated to 1 mg mL-' with an Amiconultrafiltration cell (PM 10 membrane) and stored at -20°C.

Purification of NADP:Malic Enzyme by lmmunoadsorption

For the preparation of a resin, 10.6 mg of specific anti-NADP:malic-enzyme IgG in 20 mL PBS was coupled cova-lently to 3 mL of wet packed protein A-Sepharose CL-4B(Pharmacia/Freiburg) according to Gersten and Marchalonis(1 1). For removal of free binding sites, the gel was incubatedovernight in 0.1% (w/v) BSA in PBS. Excess proteins wereremoved by washing the gel in a glass column with 30 mL 3M MgCl2 in 100 mM Hepes/KOH (pH 8.0), and with 150 mL20 mm Hepes/KOH (pH 8.0). Aliquots (0.4 mL) of the gelmaterial were stored at 4°C in 20 mm Hepes/KOH (pH 8.0),and 0.02% merthiolate. The latter was removed by washingwith 40 mm Hepes/KOH (pH 8.0) prior to use.NADP:malic enzyme (enriched by (NH4)2SO4 gradient elu-

tion and chromatography on DEAE-cellulose) was applied toa column (0.4 mL) of anti-NADP:malic enzyme-IgG-proteinA-Sepharose CL 4B equilibrated with 40 mm Hepes/KOH(pH 8.0). Unbound NADP:malic enzyme was reapplied tothe column once again. The protein concentration was limitedto 4 mg mL-' maximum to prevent plugging. The flow ratewas 50 mL h-'. The column was washed with 20 mL of 0.4M NaCl in 40 mm Hepes/KOH (pH 8.0), and NADP:malicenzyme was eluted with 10 mL of 4 M MgC12 in 100 mMHepes/KOH (pH 8.0). The eluate was dialyzed immediatelyagainst 20 mM Hepes/KOH and 10 mM 2-mercaptoethanol(pH 8.0). The dialysate was concentrated with an Amiconultrafiltration cell (PM 10 membrane) and stored in 10% (v/v) glycerol at -20°C.

2 Abbreviation: IgG, immunoglobulin G.

Immunochemical Assays

For immunoprecipitation, 50 ,uL of NADP:malic enzyme,50 AL enriched antibodies, and 20 ,L immunoprecipitin (12%wet packed, formalin fixed Staphylococcus aureus Cowan Icells [Bethesda Research Laboratories, Neu Isenburg, FRG])were mixed in a 1.5 mL Eppendorf reaction vessel andincubated for 10 min at room temperature. After centrifuga-tion for 1 min at 5400g, an aliquot of 25 ,AL supernatant wasused to determine the activity of NADP:malic enzyme. Forimmunoinhibition studies the activity of NADP:malic en-zyme was assayed in the presence of enriched antibodies (orserial dilutions of it).

Electrophoresis

Disc-polyacrylamide-gel electrophoresis under nondenatur-ing conditions was performed according to Davis (9) in slabgels with a 14 x 8 x 0.1 cm separation gel and a 3 x 8 x 0.1cm collection gel on top at 80 to 120 V for 2 to 3 h. SDS-PAGE was done using the system developed by Laemmli (16).Samples were treated with 10 mm 2-mercaptoethanol, 1%SDS for 2 min in a boiling water bath prior to electrophoresis.Proteins were visualized by treatment with 0.07% (w/v) Coo-massie brilliant blue R-250 (Serva, Heidelberg) in 50% meth-anol and 10% acetic acid for 45 min and destaining with 5%methanol and 10% acetic acid.

Molecular Masses

Molecular masses of native enzymes were estimated byvelocity sedimentation centrifugation in 5 mL gradients of 5to 20% (w/w) sucrose in 20 mm potassium phosphate and 1mM dithiothreitol (pH 7.5), according to Martin and Ames(18). Markers were catalase (240 kD), glucose-6-phosphatedehydrogenase (1 16 kD), and hexokinase (51 kD). Centrifu-gation was performed for 13 h at 270,000g in a SW 41 Tirotor ofa Beckman L2/65 ultracentrifuge. Subunit molecularmasses were determined by SDS-PAGE (16) with phospho-rylase a (95 kD), human serum albumin (69 kD), carboan-hydrase (31 kD), and Cyt c (12.5 kD) as markers.

RESULTS

To purify NADP:malic enzyme from a variety of plants,we used a specific immunoadsorbent method as an essentialstep in the procedure. We therefore purified the NADP:malicenzyme of corn leaves according to a modified conventionalprocedure described by Asami et al. (1). A typical purificationprotocol is given in Table I. This procedure resulted in analmost homogeneous protein preparation as visualized bySDS-PAGE (Fig. 1).Such a preparation was further purified by preparative SDS-

PAGE and used for immunization of rabbits. In an Ouchter-lony double diffusion test, the respective antiserum showed asingle and confluent precipitation line with either purifiedNADP:malic enzyme or with an enzyme preparation enriched15-fold by chromatography on DEAE-cellulose. During im-munogelelectrophoresis, only a single precipitation line wasobserved with either probe (data not shown). Specific anti-

712 Plant Physiol. Vol. 92, 1990

PURIFICATION OF NADP:MALIC ENZYMES FROM PLANTS

Table I. Purification Protocol of NADP:Malic Enzyme from Green Corn LeavesPurification Step Protein Activity Specific Activity Purification Recovery

mg units units mg-1 -fold %

Crude extract 4280 5.2 1.2 1 100(NH4)2SO4 gradient solubilization 441 3.2 7.3 6 62DEAE-cellulose 108 2.0 18.3 15 38Hydoxyl-apatite 30 1.4 47.5 39 27Blue-Sepharose 5 0.45 93.7 77 9

M M

69 000

31 000

1 2 3 4 5

Figure 1. SDS-PAGE of corn leaf NADP:malic enzyme as purified byconventional procedures. Lanes 1 and 2: 20 and 10 Ag of purifiedmalic enzyme (see Table I); lanes 3 and 4: 30 and 50 Mg of eluatefrom a DEAE-cellulose column; lane 5: marker enzymes.

Table II. Enrichment of Anti-Malic Enzyme-lgGsThe presence of anti-malic enzyme-antibodies was assayed by

immunoprecipitation of 0.1 unit corn-leaf-NADP:malic enzyme (15-fold enriched by chromatography on DEAE-cellulose) with 15 /Igprotein of antibody preparation and by determining subsequently themalic enzyme activity which was not precipitated. Each value repre-sents the average of doublets. Controls with preimmune serum orwithout S. aureus Cowan cells showed no precipitation of malicenzyme (data not shown).

Purification Step Residual Activity Protein

units mL-1 mg

Control without antiserum 0.90Crude antiserum 0.80 16 400(NH4)2SO4 precipitation 0.78 2 600DEAE-cellulose 0.73 800Malic enzyme-Affi-gel 10 0.0 13

bodies were purified by conventional methods and by im-munoadsorption onto purified NADP:malic enzyme immo-bilized to Affi-Gel 10. During the latter step, the amount ofprotein was reduced by 97.9% while the antibodies becamehighly enriched (see Table II).

The highly purified antibodies against corn-leafNADP:malic enzyme were linked covalently to a protein A-Sepharose CL 4B and used as an immunoadsorbent to purifyNADP:malic enzyme from corn and several other plants. Toeliminate other interfering materials (polyphenols, membraneparticles, etc.), it was necessary to prepurify the NADP:malicenzymes by (NH4)2SO4 gradient solubilization and anion-exchange chromatography on DEAE-cellulose. NADP:malicenzymes of such preparations could be bound to the immu-noadsorbent with an efficiency of 28 to 62%. The remainingactivity was recovered in the eluate. The bound enzyme couldbe eluted by 4 M MgCl2 with a recovery of 27 to 45% relativeto the amount adsorbed to the column. Table III contains theprotocols for the purification of the enzymes from the C3plants spinach and wheat, from the CAM plant Bryophyllumdaigremontianum, and from the C4 plants corn, sugarcane,and Portulaca grandiflora. From Figure 2 it can be seen thatsuch enzyme preparations were homogeneous, showing asingle peptide, except for P. grandiflora which had two pep-tides in a ratio of about one to one.The recovery of only 27 to 45% of the bound activity from

the immunoadsorbent was largely due to the fact that elutionwas only achieved by high salt concentrations (4 M MgCl2)which, in part, inactivated the enzyme. Two M MgCl2 elutedonly half the activity of what 4 M MgCl2 did. It was thereforenecessary to use 4 M MgCl2 for elution and to remove MgCl2from the eluate immediately. Elution by other means wasineffective, i.e. by aqua bidestillata, by pH 4.0, pH 10.0, or 3M sodium rhodanide, pH 7.4. It is remarkable to note thatthe NAD:malic enzyme from P. oleracea was not bound tothe immunoadsorbent at all (see Table III bottom).The specific activities of the purified NADP:malic enzymes

of the six species ranged from 56 to 91 units (mg protein)-'with the trend that the specific activities of the enzymes fromC3 and CAM plants were somewhat lower than the specificactivities of the enzymes from C4 plants.To determine the immunochemical cross-reaction of the

NADP:malic enzymes from species of C3, C4, and CAMplants, we used enriched antibodies against the corn leafNADP:malic enzyme for immunotitration in a specific im-munoprecipitation procedure with a constant amount of pu-rified NADP:malic enzyme (Fig. 3). In another immunotitra-tion test, the inhibition of activity was determined in thepresence of antibodies (Fig. 4). In both sets of experiments,the titers for 50% residual activity or activity inhibition werealmost equally high for the enzymes of the C4 plants corn andsugarcane (Table IV). However, the titers were only 3 to 6%in the immunoprecipitation assay and 17 to 33% in theimmunoinhibition assay, respectively, with the NADP:malic

713

FATHI AND SCHNARRENBERGER

Table Ill. Purification Protocol of Leaf NADP:Malic Enzyme from Several Plant SpeciesP. oleracea Is a NAD:Malic Enzyme Type of C4 Plant and has a NAD:Malic Enzyme.

Enzyme Source Purification Protein Activity Specific Activity Purification RecoveryMethodi

mg units units mg-' -fold %S. oleracea Crude extract 12006 62.1 0.003 1 100

(NH4)2SO4 1078 50.3 0.047 9 81C3 species DEAE-cellulose 316 28.0 0.89 17 45

lmmunoadsorption 0.05 4.0 80.4 14889 6,4T. aestivum Crude extract 2054 66.5 0.032 1 100

(NH4)2SO4 1044 58.5 0.053 2 88C3 species DEAE-cellulose 210 29.4 0.138 4,6 44

Immunoadsorption 0,06 3.5 56.2 1874 5,3B. daigre- Crude extract 240 57.4 0.24 1 100montianum (NH4)2SO4 118 22.5 1.35 0,8 39

CAM species DEAE-cellulose 66 17.5 0.26 1 31Immunoadsorption 0,06 3.9 68.7 286 6,7

P. grandiflora Crude extract 486 44.4 0.09 1 100(NH4)2SO4 121 93.9 0.78 9 212

C4 species DEAE-cellulose 35 47.3 1.35 15 107lmmunoadsorption 0,04 3.6 81.8 909 8

Zea mays Crude extract 109 148 0.72 1 100(NH4)2SO4 14,1 118 8.4 6 80

C4 species DEAE-cellulose 6,4 70 10.9 8 47lmmunoadsorption 0,22 20 91.1 66 13

S. officinarum Crude extract 413 196 1.38 1 100(NH4)2SO4 61 123 2.02 4 63

C4 species DEAE-cellulose 31,4 67 2.42 5 39Immunoadsorption 0,16 13.5 84.3 178 6,9

P. oleracea Crude extract 549 15.9 0.03 1 100(NH4)2SO4 82 5.0 0.06 2 32

C4 species DEAE-cellulose 25 3.5 0.14 5 22lmmunoadsorption 0 0 - - 0

enzymes of the C3 plants spinach and wheat and of the CAMplant B. daigremontianum as compared with the titer of theNADP:malic enzyme from corn leaves. The enzyme of theC4 plant P. grandiflora behaved rather like the enzymes of theC3 and CAM plants.The molecular masses of the native NADP:malic enzymes

were determined for the enzymes of corn, spinach, and B.daigremontianum as representatives for C4, C3, and CAMplants by velocity sedimentation centrifugation in a sucrosegradient (Fig. 5). These analyses indicated molecular massesof 220 kD for the enzymes from corn and B. daigremon-tianum and of 250 kD for the enzyme from spinach. Themolecular masses of the subunits of all purified NADP:malicenzymes were estimated to be 64 to 68 kD in size except theenzyme of P. grandiflora which showed two peptides of 67 or69 kD. From this one may conclude that all NADP:malicenzymes are homotetrameric proteins under native conditionsexcept the enzyme of P. grandiflora which could also be aheterotetrameric protein.

DISCUSSIONThe purification of NADP:malic enzyme by immunoaffin-

ity, as applied in the course of this paper, proved highly

efficient and reliable with the enzymes of several C3, CAM,and C4 plants. Five out of six enzyme preparations wereobtained with a single protein band as judged by SDS-PAGE.Only the enzyme ofthe C4 plant Portulaca grandiflora showedtwo bands. This was observed in two independent experi-ments. One reason for this finding could be that one of thetwo bands is an impurity. Alternatively, the NADP:malicenzyme in this species could consist of two different subunitsor there could be two NADP:malic enzymes. Nevertheless, allthese NADP:malic enzymes appear to be of tetrameric struc-ture when compared with the mass of the native proteins asdetermined for corn, spinach and B. daigremontianum.

Purification of NADP:malic enzymes from plants by non-conventional methods have already been described whenenzymes of grapes (26) and corn (7) were bound to ADP-Sepharose and eluted with NADP/NaCl or NADP. Similarly,the enzyme of rat liver could be bound to Procion Red-Agarose and be eluted with NADP/NaCl (31). Attempts inour laboratory to purify the corn leaf NADP:malic enzymeby affinity chromatography on ADP-Sepharose or Blue-Seph-arose, failed since it was not possible to elute bound enzymewith NADP/NaCl or NADP.The antibodies against the corn leaf NADP:malic enzyme

714 Plant Physiol. Vol. 92, 1990

PURIFICATION OF NADP:MALIC ENZYMES FROM PLANTS

az _ |

-0

~ _. <

c

120

oo

.80

0

0

40

a

_:--9500C

69

3 1'co

3 21:64 1:1024 1:16384 1:262144

1gG dilution

G-IAN(;5000Y -

*6900C

4 3 2 2 4 3e f2 3

Figure 2. SDS-PAGE of NADP:malic enzymes from corn (a), sugar-cane (b), P. grandiflora (c), spinach (d), wheat (e), B. daigremontianum(f). Lane 1, Eluate from anti-NADP:malic enzyme-lgG-protein A-Seph-arose CL 4B; lane 2, markers (phosphorylase, human serum albumin,and carboanhydrase); lanes 3 and 4, eluate from a DEAE-cellulosecolumn.

crossreacted sufficiently with the enzymes of the other fiveplants to act as immunoadsorbent. It is remarkable that theNAD:malic enzyme from the NAD:malic enzyme type ofC4 plant, P. oleracea, failed to do so. Therefore, this enzymemust be immunochemically very different from the NADP:malic enzyme.The specific activities of the purified malic enzymes from

leaves of the six different C3, C4, and CAM plants are on theorder of 56 to 91 units (mg protein)-'. These values are in thesame order as the highest reported values, e.g. for the purifiedenzymes of corn leaves (1, 22) and B. tubiflorum (5). It is notclear to what extent, if at all, the values in our preparationswere affected by MgCl2 inactivation during elution from theimmunoadsorbent resin.

It is interesting to note from our data that the degree ofpurification over crude extracts correlates closely with the factthat the activity ofNADP:malic enzyme is considerably lowerin C3 plants than in C4 and CAM plants known to have suchhigh activities (12, 21). Thus, NADP:malic enzyme represents

Figure 3. Immunotitration of purified malic enzyme from C3, C4, andCAM plants with serial dilutions of enriched antibodies. A constantamount of malic enzyme (2.8 Ag per incubation mixture) was used.Immune complexes were removed by adsorption to protein A-coatedS. aureus Cowan I cells and by subsequent centrifugation. Nonbound

95O9f enzyme was determined in the supernatant. (0), Corn; (E), sugarcane;(0), P. grandiflora; (A), B. daigremontianum, (0), spinach; (A), wheat.

69000

120

> SO0 .

3 40 .U

a

(

0

1:2 1:8 1:32 1:128 1:512 1:2048IgG dilution

Figure 4. Inhibition of activity of malic enzyme from C3, C4, and CAMplants by adding enriched antibodies (or serial dilutions of them) inthe photometric assay. In all experiments 2.8 ug of malic enzymeprotein was used. (0), Corn; (a), sugarcane; (0), P. grandiflora; (AL),B. daigremontianum; (>), spinach; (A), wheat.

Table IV. Percent Cross-Reaction of NADP:Malic Enzyme fromSugarcane, P. grandiflora, Wheat, Spinach, and B. daigremontianumwith the Corn Leaf Enzyme as Determined by Immunotitration in anImmunoprecipitation Assay or by Inhibition of Enzyme Activity byEnriched IgGs against the Corn Leaf NADP:Malic Enzyme

Percent cross-reaction was defined as the titer for 50% precipita-tion or inhibition of a particular malic enzyme relative to the titer for50% precipitation or inhibition of corn leaf malic enzyme.

Percent Cross-ReactionSource of Enzyme

Immunoprecipitation Inhibition of activity

Corn 100 100Sugarcane 79 93P. grandiflora 6 20Wheat 6 26Spinach 3 17B. daigremontianum 4 33

715

I

dI I

FATHI AND SCHNARRENBERGER

4

2

X(,)

xCo.

4 \N

a)

00

2 EL.

O 0

4 2

2

0 . *08 I,-&̀ . ,

--

10 15 20 25

Fraction number

Figure 5. Velocity sedimentation of NADP:malic enzymes (dashedlines) from corn (top), spinach (middle), and B. daigremontianum(bottom). Markers used were (A), catalase; (x), glucose-6-phosphatedehydrogenase; and (0), hexokinase.

up to 1% of the total protein in C4 plants and 0.5% in theCAM plant but less than 0.01% in the C3 plants.The characteristics of the NADP:malic enzymes, as de-

scribed by the present analyses, are distinct from a NADP-dependent activity associated with a NAD:malic enzyme. Thetwo enzymes were already separated during (NH4)2SO4gradient solubilization whereby the NADP:malic enzymeeluted at 47% (NH4)2SO4 and the NAD:malic enzyme elutedat 41% (NH4)2SO4. In addition, the specific activity of theNADP:malic enzyme is severalfold higher than the NADP-dependent activity of the purified NAD:malic enzyme: Thehighest reported specific activity of the NAD:malic enzymewith NAD is 88 units (mg protein)-' for the enzyme fromSolanum tuberosum (29) and, thus, 26 units (mg protein)-'for the NADP-dependent activity if the latter accounts for30% of the NAD-dependent activity (28).As already mentioned in the introduction the results on the

intracellular compartmentation of NADP:malic enzymes indiffering plant tissues and species appear to be contradictory.On a simplified basis one may suppose a localization ofNADP:malic enzyme in the chloroplasts of C4 plants and inthe cytosol of C3 and CAM plants (for references see intro-duction). Comparing this with other cell compartment specificisoenzyme systems like the cytosol and plastid isoenzymes ofsugar phosphate metabolism (25), one would expect very lowcross-reactivity of NADP:malic enzymes even if compart-mented differently in different plant species. The immuno-chemical differences between the NADP:malic enzymes ofthe C4 plants corn and sugarcane and the NADP:malic en-zymes of the C3 and CAM plants seem to be in accord withsuch a view. The NADP:malic enzyme of the C4 plant P.grandiflora does not fit this view. This enzyme displayed animmunochemical cross-reaction more like the enzymes of theC3 and CAM plants than that of the C4 plants. This finding,however, could be erroneous since protein was used as a basisfor both immunotitration tests. This may not have beenappropriate when one of the two protein bands in the SDSgels were not immunochemically reactive. If this were thecase, the enzyme of P. grandiflora would be immunochemi-cally between C3 and CAM plants and the two other C4 plants.It is therefore hard to judge whether the NADP:malic enzymesof C4 plants form a group of immunochemically closelydefined enzymes or not.

ACKNOWLEDGMENT

The authors thank Mrs. Vivienne Amery for helpful comments inpreparing the manuscript.

LITERATURE CITED

1. Asami S, Inoue K, Matsumoto K, Murachi A, Akazawa T (1979)NADP-malic enzyme from maize leaf: purification and prop-erties. Arch Biochem Biophys 194: 503-510

2. Bensadoun A, Weinstein D (1976) Assay of proteins in thepresence of interfering materials. Anal Biochem 70: 241-250

3. Bohme HJ, Kopperschlager G, Schulz J, Hofmann E (1972)Affinity chromatography of phosphofructokinase using ciba-cron blue F3G-A. J Chromatogr 69: 209-214

4. Bradford MM (1976) A rapid and sensitive method for thequantitation of microgram quantities of protein utilizing theprinciple of protein-dye binding. Anal Biochem 72: 248-254

5. Brandon PC, van Boeckel-Mol TN (1973) Properties of purifiedmalic enzyme in relation to crassulacean acid metabolism. EurJ Biochem 35: 62-69

6. Chapman KSR, Hatch MD (1977) Regulation of mitochondrialNAD-malic enzyme involved in C4-pathway photosynthesis.Arch Biochem Biophys 184: 298-306

7. Collins PD, Hague DR (1983) Light-stimulated synthesis ofNADP malic enzyme in leaves of maize. J Biol Chem 258:4012-4018

8. Davies DD, Patil KD (1974) Regulation of malic enzyme ofSolanum tuberosum by metabolites. Biochem J 137: 45-53

9. Davis BJ (1964) Disc electrophoresis-Il. Method and applicationto human serum proteins. Annals NY Acad Sci 121: 404-427

10. Garnier-Dardart J, Queiroz O(1974) Malic enzyme in the leavesof Bryophyllum daigremontianum. Phytochemistry 13: 1695-1702

11. Gersten DM, Marchalonis JJ (1978) A rapid, novel method forthe solid-phase derivatization of IgG antibodies for immuneaffinity chromatography. J Immunol Methods 24: 305-309

12. Hatch MD, Kagawa T, Craig S(1975) Subdivision ofC4-pathwayspecies based on differing C4 acid decarboxylating systems andultrastructural features. Aust J Plant Physiol 2: 111-128

10

.)4.

cJ

nL

L

%a

54.'-

00

Ea)Ncli

5

010

5

010

5

716 Plant Physiol. Vol. 92, 1 990

PURIFICATION OF NADP:MALIC ENZYMES FROM PLANTS

13. Hatch MD, Mau S-L (1977) Association of NADP- and NAD-linked malic enzyme activities in Zea mays: Relation to C4pathway photosynthesis. Arch Biochem Biophys 179: 361-369

14. Hatch MD, Mau S-L, Kagawa T (1974) Properties of leafNADmalic enzyme from plants with C4 pathway photosynthesis.Arch Biochem Biophys 165: 188-200

15. Hatch MD, Osmond CB (1976) Compartmentation and trans-port in C4 photosynthesis. In CR Stocking, U Heber, eds,Encyclopedia of Plant Physiol (New Series), Vol 3. Springer-Verlag, Berlin, pp 144-184

16. Laemmli MK (1970) Cleavage of structural proteins during theassembly of the head of bacteriophage T4. Nature 227: 680-685

17. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ (1951) Proteinmeasurement with the folin phenol reagent. J Biol Chem 193:265-275

18. Martin RG, Ames BN (1961) A method for determining thesedimentation behaviour of enzymes: Application to proteinmixtures. J Biol Chem 236: 1372-1379

19. Mukerji SK, Ting JP (1968) Malate dehydrogenase (decarbox-ylating) (NADP) isoenzymes of Opuntia stem tissue. Mito-chondrial, chloroplast, and soluble forms. Biochim BiophysActa 167: 239-249

20. Nishikido T, Wada T (1974) Comparative studies of NADP-malic enzyme from C4- and C3-plants. Biochem Biophys ResCommun 61: 243-249

21. Osmond CB (1978) Crassulacean acid metabolism: a curiosity incontext. Annu Rev Plant Physiol 29: 379-414

22. Persanov WM, Woronowa EA, Oparina LA, Karpilow JC (1976)Isolation and purification ofNADP "malic-enzyme" from cornleaves. Biochimija 41: 921-925

23. Pupillo P, Bossi P (1979) Two forms of NADP-dependent malicenzyme in expanding maize leaves. Planta 144: 283-289

24. Schnarrenberger C, Gross D, Burkhard Ch, Herbert M (1980)Cell organelles from crassulacean acid metabolism (CAM)plants. II. Compartmentation of enzymes of the crassulaceanacid metabolism. Planta 147: 477-484

25. Schnarrenberger C, Herbert M, Kruger 1 (1983) Intracellularcompartmentation of isoenzymes of sugar phosphate metabo-lism in green leaves. In MC Rattazzi, JG Scandalios, GS Whitt,eds, Isozymes: Current Topics in Biological and Medical Re-search, Vol 8. Liss, New York, pp 23-51

26. Spettoli P. Bottacin A, Zamorani A (1981) Affinity chromatog-raphy of malic enzyme from grape berries. Phytochemistry 20:29-30

27. Tomaszewska B, Schramm RW, Kokot A, Mazurowa H (1983)Isolation and characterization of cytoplasmic NADP-malicenzyme from Lupinus luteus leaves. Physiol Plant 59: 641-646

28. Wedding RT (1989) Malic enzymes of higher plants. Character-istics, regulation, and physiological function. Plant Physiol 90:367-371

29. Willeford KO, Wedding RT (1987) Evidence for a multiplesubunit composition ofNAD malic enzyme. J Biol Chem 84:1084-1087

30. Winter K, Foster JG, Edwards GE, Holtum JAM (1982) Intra-cellular localization of enzymes of carbon metabolism inMesembryanthemum crystallinum exhibiting C3 photosyn-thetic characteristics or performing crassulacean acid metabo-lism. Plant Physiol 69: 300-307

31. Wise LS, Rubin CS (1984) Rapid purification and radioimmuno-assay of cytosolic malic enzyme. Anal Biochem 140: 256-263

717