ARCHIVES

Vol.

OF BIOCHEMISTRY

298, No. 2, November

AND

BIOPHYSICS

1, pp. 612-619,1992

Partial Purification and Characterization of Mannitol: Mannose I-Oxidoreductase from Celeriac (Apium graveolens var. rapaceum) Roots Johan

M. H. Stoop and D. Mason

Pharrl

Department of Horticultural Science and Plant Physiology Box 7609, Raleigh, North Carolina 27695-7609

Received

May

18, 1992, and in revised

form

July

Program,

North

Carolina

State University,

16, 1992

and of closely related celery (A. graueozens L. var. d&e). Both varieties belong to the family Apiaceae and can be crossed. Celeriac is often referred to as knob celery because the enlarged taproot and compressed swollen stem look like a knob. The knob represents the economic portion of celeriac and can function as a storage organ for carbohydrates. Both sucrose and mannitol are translocated through the phloem and stored in the petiole and/or knob. Mannitol is by far the most abundant polyol in nature, occurring in bacteria, fungi, algae, lichens, and vascular plants (1, 2). Although several physiological roles have been proposed for mannitol, such as carbohydrate storage, regulation of carbon partitioning, osmoregulation, and cofactor regulation (1, 3, 4), little information exists on the enzymatic pathways of mannitol synthesis and utilization. Most information has come from studies of fungi and bacteria which contain a mannitol dehydrogenase that catalyzes the NAD-dependent (5-8) or NADP-dependent (9-12) oxidation of D-mannitol to D-fructose. NAD-dependent mannitol-1-P dehydrogenase has been detected in bacteria (13-15), brown algae (16), and fungi (17) and catalyzes the conversion of mannitol-1-P to fructose-6-P. All three enzymes are 2-oxidoreductases and can catalyze the reaction in either direction, toward synthesis or utilization of the polyol, depending on the availability of oxidized or reduced cofactor and the pH (1). In celery, mannitol synthesis is catalyzed by a NADPHdependent mannose-6-P reductase which reduces mannose-6-P to mannitol-1-P (4) and the enzyme is strictly localized in the cytoplasm of photosynthetic tissue (18). Mannitol catabolism in higher plants is still poorly understood. Several labeling studies using [“C]mannitol suggestthat mannitol is utilized at a slow rate in vascular plants. When celery leaf discs were incubated in [14C]mannitol, mannitol utilization was restricted to young leaf tissue (19). Suspension cultures of Daucus carapaceum)

A mannitol:mannose 1-oxidoreductase was isolated from celeriac (Apium graveolens var. rapaceum) root tips by fractionation with (NH&SOI, followed by chromatography on a Fractogel DEAE column and then concentration with (NH&S04. This newly discovered mannitol dehydrogenase catalyzes the NAD-dependent oxidation of mannitol to mannose, not mannitol to fructose. The sugar product of the enzyme reaction was identified by three independent HPLC systems and by an enzymatitally linked system as being mannose and not fructose or glucose. Normal Michaelis-Menten kinetics were exhibited for both mannitol and NAD with K, values of 72 and 0.26 mM, respectively, at pH 9.0. The V,, was 40.14 pmol/h/mg protein for mannitol synthesis and 0.8 pmol/ h/mg protein for mannose synthesis at pH 9.0. In the polyol oxidizing reaction, the enzyme was very specific for mannitol with a low rate of oxidation of sorbitol. In the reverse reaction, the enzyme was specific for mannose. The enzyme was strongly inhibited by NADH and sensitive to alterations of NADNADH ratio. The enzyme is of physiological importance in that it is mainly localized in root tips (sink tissue) where it functions to convert mannitol into hexoses which are utilized to support root growth. Product determination and kinetic characterization were carried out on an enzyme preparation with a specific activity (SA) of 30.44 pmol/h/mg protein. Subsequently, the enzyme was further purified to a SA of 201 pmol/h/mg protein using an NAD affinity column. This paper apparently represents the first evidence of the existence of a mannitol:mannose 1-oxidoreductase and also the first evidence of the presence of a mannitol o 1992 Academic press, I,,~. dehydrogenase in vascular plants.

Sucrose and the acyclic polyol mannitol are the major photosynthetic products of celeriac (Apium graveolens var. 1 To whom

correspondence

should

be addressed.

Fax:

(919) 515.7747.

612 All

Copyright 0 1992 rights of reproduction

0003-9861/92 $5.00 by Academic Press, Inc. in any form reserved.

PARTIAL

PURIFICATION

OF

MANNITOL:MANNOSE

rota L. and Pinus radiata D. (20) or carrot root tissue (21) exhibited small uptake and metabolism of [“C]mannitol. Mannitol respiration, as measured by 14C02 evolution, was monitored in 15 higher plant speciesand ranged from very low rates (Arena satiua) to rates comparable to those of fructose or glucose (Fraxinus americana) (22). Exposure of white ash leaflets to [14C]mannitol resulted in the formation of a small amount of [14C]fructose after 2 days, suggesting that the first step of mannitol utilization in this species may be oxidation to fructose (22). Although these labeling studies support the presence of a mannitol utilizing enzyme in higher plants, apparently no mannitol oxidizing enzyme has been isolated from a higher plant. Here we present evidence of the presence of a mannitol dehydrogenase in celeriac. This paper apparently represents the first report of the existence of a mannitol:mannose 1-oxidoreductase in any living organism. The partial purification and characterization of the enzyme are described. MATERIALS

AND

METHODS

Materials Celeriac (A. graueolens var. rupaceum) cultivar Monarch was seeded on February 18, 1991 and reached transplant stage in May 1991. Some seedlings were kept in seeding flats (2 X 2-cm containers with a depth of 4 cm) for a period of 4 weeks in order to stress the young plants before being transplanted to 20-cm-diameter pots with a depth of 15 cm. Stress was evident from the fact that these plants grown in small containers were root bound and severely stunted. Celeriac cv. Marble Ball transplants, acquired from DNA-Plant Technology, Inc., were transplanted into 25-cm-diameter pots containing soilless media (Promix, Premier CT) and grown in a greenhouse from June until November 1991. Plants were watered daily and fertilized twice a week with a soluble fertilizer solution (200 ppm N, 100 ppm P206, 200 ppm KzO). Substrates and nucleotides were obtained from Sigma Chemical Co. or from Pharmacia Fine Chemical, Inc. Because the highest purity of mannitol obtained from Sigma Chemical Co. still contained about 1% sorbitol, mannitol was purified by two successive recrystallizations from H20 thereby eliminating sorbitol contamination in the mannitol substrate.

Enzyme Assays Spectrophotometric assay. Mannitol dehydrogenase (MDH): also referred to as mannitol oxidoreductase, activity was assayed by monitoring the oxidation-reduction of NADH/NAD spectrophotometrically at 340 nm. Assays were done at 25°C in a total volume of 1 ml. Optimum conditions (buffer, pH, and substrate concns) were determined. The reaction mixture contained 100 mM Bis-Tris propane (pH 9.0), 2 mM NAD+, enzyme extract, and 150 mM D-mannitol. The reactions were

’ Abbreviations used: Mops, 4-morpholinepropanesulfonic acid); Bis-Tris propane, (1,3-bis[tris(hydroxymethyl)-methylamino]propane; Hepes, 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid; DEAE, diethylaminoethyl; ADH, alcohol dehydrogenase; PMI, phosphomannose isomerase; MDH, mannitol dehydrogenase, or mannitokmannose l-oxidoreductase; MTT, 3,(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2~-tetrazolium bromide; PMI, phosphomannose isomerase; PMS, phenazine methosulfate; Triton X-100, octylphenoxy polyethoxyethanol; EDTA, ethylenediaminetetraacetate; FPLC, fast protein liquid chromatography; PAGE, polyacrylamide gel electrophoresis; U, unit, amount of enzyme catalyzing the formation of 1 pmol product/h.

l-OXIDOREDUCTASE

613

initiated with mannitol. One unit of activity was defined as the amount of enzyme which catalyzes the reduction of 1 pmol NAD+ per hour (pmol/ h). Initial rates of oxidation were used to calculate MDH activity. For the reverse reaction (mannose reduction) the enzyme extract was incubated in a mixture of 0.1 mM NADH, enzyme extract, 20 mM substrata (D-mannose, D-fructose, or D-glucose), and 100 mM Mops at pH 7.5 in a total volume of 1 ml. The alcohol dehydrogenase (ADH) assay was similar to the MDH assay using 5 mM ethanol as the substrate. The phosphomannose isomerase (PMI) assay contained the following reaction mixture: 100 mM Hepes (pH 7.5), 5 mM MgCl,, 0.5 mM NAD, 7.5 U/ml glucose-6-phosphate dehydrogenase, 6 U/ml phosphoglucose isomerase, enzyme extract, and 2.5 mM mannose g-phosphate. Detection of enzyme activity by staining. Enzyme extract was loaded on a 7.5% T, 2.6% C polyacrylamide gel, electrophoresed at constant voltage (200 V, 150 mA) for 40 to 45 min on a Mini-Protean II gel apparatus (Bio-Rad), and stained for MDH activity. MDH activity stain contained 50 ml of 0.2 M Bis-Tris propane, pH 9.0, 150 mM mannitol, 2 mM NAD+, 1 ml of MTT (3,(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (Sigma); 10 mg/ml), and 0.5 ml of PMS (phenazine methosulfate (Sigma);2 mg/ml).

Purification

of Mannitol:Mannose

I-Oxidoreductase

Enzyme extraction. Celeriac root tips were harvested, pooled, and washed in distilled water. Fresh roots were ground in a chilled mortar using a 1:4 tissue:buffer ratio. The extraction buffer contained 50 mM Mops (pH 7.5), 5 mM MgClx, 1 mM EDTA, 5 mM DTT, and 1% Triton X-100. Homogenates were centrifuged at 20,OOOg for 15 min at 4°C. The crude extract consisted of supernatant that was desalted by centrifugal filtration on a Sephadex G-25-50 column equilibrated with 50 mM Mops-NaOH (pH 7.5), 5 mM MgClr, and 5 mM DTT prior to assay for MDH activity. Purification. Supernatant from the root extraction was brought to 45% saturation with ammonium sulfate at 4°C and centrifuged at 20,OOOg for 15 min. The supernatant was retained and brought to 65% saturation by further addition of (NH4).$S04. After centrifugation the supernatant was discarded and the pellet was suspended in a minimum volume of 20 mM Mops (pH 7.5), 2 mM MgClz, and 1 mM DTT. The dissolved pellet was filtered through a 0.4-pm filter and applied to a Fractogel DEAE ion-exchange column (10 X 1.5-cm) equilibrated with 20 mM Mops (pH 7.5), 2 mM MgClz, and 1 mM DTT. Protein eluting from the DEAE column was monitored by absorbance at 280 nm and protein concentrations in pooled fractions were determined by the Bradford method (23) using bovine serum albumin as a standard. Proteins were eluted with a linear KC1 salt gradient, 0 to 0.4 M KCl, in running buffer. Flow rate equaled 1.25 ml/min and 2-ml fractions were collected. Fractions containing MDH, which were free from ADH or PMI, were pooled and further concentrated by ammonium sulfate precipitation (90% saturation), and the pellet was resuspended in a small volume of 20 mM Mops (pH 7.5), 2 mM MgC&, and 1 mM DTT. The concentrated enzyme was desalted (by centrifugal filtration on a Sephadex G-25-50 column) and stored at -80°C. This enzyme preparation was utilized for kinetic analyses. Subsequently, 30 ~1 of the concentrated enzyme was further purified by affinity chromatography using a l-ml P-nicotinamide adenine dinucleotide agarose column (Sigma N 1008, attachment C-8, spacer 9 atoms). The column was washed with 10 bed vol of distilled water followed by 10 bed vol of running buffer (20 mM Bis-Tris propane, pH 9.0, containing 1 mM DTT) prior to loading of protein. At time of protein loading, elution solvent was collected in l-ml fractions and the column was washed with running buffer until protein concentration in fractions (as measured by A& returned to baseline levels. The column was then washed with eluting buffer (20 mM Bis-Tris propane, pH 9.0, 2 mM NAD+ containing 1 mM DTT) which eluted the oxidoreductase.

Product Purification Four different MDH ture (100 mM Bis-Tris

and Identification

reactions propane

containing (pH 9.0),

the complete 2 mM NAD+,

reaction mix10 ~1 partially

614

STOOP

AND

purified enzyme, and 150 mM mannitol), the complete reaction (-mannitol), the complete reaction (-NAD), or the complete reaction (-enzyme extract), were run for 2 h, heat killed in boiling water for 60 a, and centrifuged at 10,OOOg for 1 min. Because the buffer in the reaction mixture cochromatographed with and masked several sugars, including mannose and fructose, it was necessary to purify the reaction product away from its reaction buffer prior to analysis of the product by HPLC. An ion-exchange column which was composed of an anion-exchanger (Bio-Rad AG l-X8, H+-form) separated from a cation-exchanger (Dowex-50W, formate form) by Whatman No. 1 paper was developed. Total column volume equaled 2 ml. Columns were spun to dryness for 4 min at 1000 rpm in a table-top centrifuge. When this column was loaded with 1 ml of MDH reaction mixture, which was then centrifugally passed through, the column effectively exchanged the highly concentrated buffer, Bis-Tris propane, with a lower concentration of formic acid. The latter had the great advantage that on the chromatogram, formic acid had a different retention time (RT = 4.26 min) than mannose (RT = Il.20 min) or fructose (RT = 12.00 min). When mannose was added to the MDH reaction mixture, loaded on the ion-exchange column, and analyzed by HPLC on a Bio-Rad Fast Carbohydrate column (Pb column) a distinct mannose peak (RT = 3.64 min) could be observed. Because the Waters Sugar-Pak I column (Ca’+ column) had much longer retention times for most of the sugars, the Fast Carbohydrate column with shorter RT values was used to hasten the process of product purification. The MDH product peak was collected from numerous injections, freeze dried, and resuspended in 1 ml of dH?O. The purified MDH product resulting from the above protocol was then analyzed on three different HPLC systems: (i) A HPLC system containing a Bio-Rad Fast Carbohydrate column (Pb column); (ii) A HPLC system with high resolution conditions consisting of a Waters Sugar-Pak I Car’ column without either a Waters Bondapak C18/Corasil guard or anion and cation cartridges (Bio-Rad; elimination of the guards resulted in very high resolution chromatography with particular improvement in the resolution between mannose and fructose); and (iii) A Dionex high-performance anion-exchange chromatography system using pulsed amperometric detection. The purified product was also identified with an enzymatically linked system using commercial reagent enzymes so that the presence of mannose could be differentiated from that of fructose. This enzymatically linked spectrophotometric assay consisted of 5 U/ml hexokinase, 0.047 U/ml phosphomannose isomerase, 1 U/ml phosphoglucose isomerase, and 1 U/ml glucose-6-phosphate dehydrogenase, 1 mM NAD+, 1 mM ATP, 1 mM MgC12, 100 mM Mops buffer at pH 7.5 and the reaction was initiated with 20 pg substrate (mannose, fructose, or purified product). Assays were done at 25°C in a total volume of 1 ml. The product determination was based ultimately on the glucose-6-phosphate dehydrogenase reaction which reduces NAD to NADH and can be monitored spectrophotometrically at 340 nm. The key enzyme for mannose detection is PMI. By comparing the amount of NAD reduction in the presence (+PMI) and absence (-PMI) of PMI, it is possible to determine if fructose or mannose is present in a particular sample. If mannose is in the sample, the (-PMI) reaction will exhibit no NAD reduction while the (+PMI) will. If both (+PMI) and (-PMI) result in equal NAD reduction fructose and/or glucose is in the sample.

Carbohydrate

PHARR

transplanting to 20-cm pots, both leaf and petiole hexose, sucrose and especially mannitol concentrations decreased drastically 1 week after transplanting (Fig. 1). These transplants were severely stunted as compared to the control plants grown in 60-liter pots in the same greenhouse environment. Total shoot weight increased only slightly over the l-week period after transplanting, while root volume increased drastically (data not shown), suggesting the possibility that mannitol was translocated to the root where it could be metabolized. When root tips of these stressed plants were analyzed for mannitol dehydrogenase (MDH), an activity of 7.22 pmol/h/g fresh wt was observed, indicating that celeriac root tips contained a MDH which catalyzed the NAD-dependent oxidation of mannitol. No activity was observed if NADP was used as the substrate. In order to determine the tissue specificity of MDH, a crude extract of the root tip, older fibrous roots, young petioles, and mature petioles of

0 LEAF

PETIOLE

LEAF

PETIOLE

LEAF

PETIOLE

r--l

Analysis

Ethanolic extractions from each of the samples were used for measurements of sucrose, hexose sugars, and polyols. The procedures for ethanolic extraction were followed as described by Hubbard et al. (24). Carbohydrate content was determined on a HPLC system using a Waters Sugar-Pak I (Ca” column). The column was preceded by a Waters Bondapak C18/Corasil guard and a set of anion and cation cartridges (Bio-Rad). RESULTS

Discovery of Mannitol Dehydrogenase When celeriac cv. Monarch transplants were stressed by growing for a period of 4 weeks in seeding flats before

FIG. 1. Mannitol, sucrose, and hexose (glucose and fructose) concentration of stressed celeriac cv. Monarch plants. Plants were transferred to 20-cm pots at Week 0. Carbohydrate analysis was done at time of transfer (Week 0) and 1 week after transfer (Week 1). Data represent mean values of three replications. Error bars represent one standard error of the mean.

PARTIAL

YL

YL

YP

YP

ML

ML

PURIFICATION

MP

81

plant

part

MP

SI

plant

part

OF

TRI

OR

RT

TRI

OR

FIT

MANNITOL:MANNOSE

hydrate stored. The root and knob tissue stored sucrose while the shoot, which comprised young and mature leaves and petioles, stored mainly hexose. Only the root tips contained very low sucrose and mannitol and had almost no hexose (Fig. 2). Figure 3 is a gel showing the MDH and alcohol dehydrogenase (ADH) activity stain of different plant parts. Five microliters of extract for each tissue was electrophoresed on a nondenaturing PAGE gel. Figure 3 supports the previous finding that the MDH is most active in root tissue. A low activity was observed in young petioles and the knob. Alcohol dehydrogenase activity was observed when the gel was incubated with 25 mM ethanol. The ADH activity in this 45-65% ammonium sulfate fraction was lower than the MDH activity but was present in all parts of the celeriac plant. The MDH and ADH bands seemed to have a similar migration pattern, particularly in root tissue, indicating that these bands might represent a single bifunctional enzyme. Alternatively, the enzyme preparation might contain both MDH and ADH. In order to test these possibilities, the MDH extract was further purified. Purification

ii6

6

8 2

4

P

2 0

YL

YP

FIG. 2. Mannitol, sucrose, tration of different plant parts young leaf; YP, young petiole; innermost part of stem; TRI, RT, root tip). Data represent bars represent one standard

ML

MP

SI

plant

part

TRI

OR

RT

615

I-OXIDOREDUCTASE

of Mannitol Dehydrogenase

Initial purification efforts using affinity chromatography media (Sigma affinity media kit No. RDL-9, containing Cibacron Blue 3GA Agarose, Reactive Blue Agarose, Reactive Blue 72 Agarose, Reactive Brown 10 Agarose, Reactive Green 5 Agarose, Reactive Green 19 Agarose, Reactive Red 120 Agarose, Reactive Yellow 3 Agarose, Reactive Yellow 86 Agarose) were unsuccessful. Ammonium sulfate fractionations were carried out at 4565% saturation resulting in over 100% recovery of the MDH activity and an elimination of 88% of the protein (Table I). It became apparent that MDH was inhibited

and hexose (glucose and fructose) concenof nonstressed celeriac cv. Monarch (YL, ML, mature leac MP, mature petiole; SI, innermost part of taproot; OR, older root; mean values of three replications. Error error of the mean.

stressed celeriac transplants was analyzed for MDH activity. The highest MDH activity was observed in root tips (7.22 pmol/h/g fresh wt). Fibrous roots, young petioles, and mature petioles had much lower activities of 1.15,0.82, and 0.33 pmol/h/g fresh wt, respectively. When celeriac cv. Monarch was grown under normal, nonstressed conditions, mannitol concentration was high in the foliage, knob, and older roots but very low in root tips. Foliar sucrose concentration was almost lo-fold lower as compared to mannitol concentrations and about 50% lower than the foliar hexose concentration (Fig. 2). Although mannitol content was high in both shoot and roots of celeriac plants, both tissues differed in type of carbo-

}-

Ethanol

+

+

Mannitol

-1

FIG. 3. Nondenaturing PAGE gel for: left (Ethanol), ADH activity stain of BSA (lane B), knob (lane K), mature petiole (lane MP), young petiole (lane YP), and root tip (lane RT); right (mannitol); MDH activity stain of BSA (lane B), knob (lane K), mature petiole (lane MP), young petiole (lane YP), and root tip (lane RT). Five microliters of extract was electrophoresed for each tissue.

STOOP AND PHARR

616

Purification

of NAD-Dependent

TABLE I Mannitol:Mannose

Total

Total

volume

Step

(ml)

Crude extract (NH,)*S04 45-65% saturation Fractogel DEAE ion-exchange

88.0 8.0 10.2

(NH,)*SOI 90%saturation Note.

See Materials

and Methods

activity

for a detailed

(md

72.16

139.48

79.02 description

of procedures.

due to someunknown circumstance(s) in the crude extract as indicated by greater than 100% recovery of MDH activity. When the Fractogel DEAE fractions were assayed for MDH, ADH, and phosphomannose isomerase (PMI) activity, three distinct peaks could be observed (Fig. 4). At high protein loading, ADH was not entirely retained on the column and eluted mainly at the start of the salt gradient. The MDH peak eluted at 0.16 M KC1 and PM1 eluted at 0.26 M KCl. It was clear from this profile that a MDH pool could be obtained free from ADH and PMI. Purification t,hrough the DEAE step resulted in a 53-fold purification of the enzyme from the crude extract. This was, however, an apparent value, since greater than 200% of the MDH activity in the crude extract was recovered upon ammonium sulfate precipitation of the enzyme. The MDH pool was then concentrated with a 90% (NH&SO, saturation step resulting in a final 59-fold purification of the enzyme from the crude extract. Subsequent purification by affinity chromatography using a ,&nicotinamide adenine dinucleotide agarose column (Sigma N 1008, attachment, C-8; spacer, 9 atoms) resulted in a specific activity of 201 U/mg. The enzyme was not retained on either of two other NAD affinity columns (Sigma N 8391, at-

A280 --

Total protein

NJ) 152.64 108.94

2.2

l-Oxidoreductase

KCI (M) - 0.5 - 0.4

16.58 4.01

Specific activity

Recovery

W/md

(%)

0.52

100 211 151 110

9.20 27.18

2.60 Crude

from Celeriac Roots

30.44 extract

was derived

from

Product Purification

18

53 59

22 g fresh wt of celeriac

root tips.

and Identification

The purified MDH product resulting from the protocol described under Materials and Methods was analyzed on three different HPLC systems: (i) A HPLC system containing a Bio-Rad Fast Carbohydrate column (Pb column), (ii) A HPLC system with high resolution conditions consisting of a Waters Sugar-Pak I Ca2+ column, and (iii) A Dionex high-performance anion-exchange chromatography system using pulsed amperometric detection. In all three chromatographic systems the product had a retention time indistinguishable from that of reagent D-mannose and different from that of D-fructose and D-glucose (data not shown). The purified product was also identified with an enzymatically linked system using commercial reagent enzymes so that mannose could be differentiated from fructose (see Materials and Methods). Results of the enzymatic analysis are shown in Table II. All four independent analyses indicated that the MDH product corresponded with mannose and not fructose or glucose. Furthermore, only mannose and not fructose or glucose was reduced by MDH when assayed in the reverse direction (Table III). Enzyme Stability The 45-65% (NH&SO4 fraction could be stored overnight at 4°C or at -80°C for several weeks without loss of activity. The partially purified enzyme was concenTABLE

II

Enzymatic Determination of Mannitol DehydrogenaseProduct

- 0.2 - 0.1 Substrate

k/ml)

- 0.0

FIG. 4. Fractogel DEAE ion-exchange column profile. Alcohol dehydrogenase (ADH), mannitol dehydrogenase (MDH), and phosphomannose isomerase (PMI) activities were analyzed. Protein content was monitored at Azm. Proteins were eluted with a linear KC1 gradient.

(-fold)

tachment, ribose hydroxyls; spacer, 11 atoms or Sigma N 1008, /3-nicotinamide adenine dinucleotide).

- 0.3

Fraction

Purification

Product Mannose Fructose Note. cedures. analysis

std std

20 20 20

-PM1 (ANO)

+PMI (A,,,)

0.007 0.003 0.708

0.624 0.702 0.722

See Materials and Methods for a detailed description of proThe product concentration determination was based on HPLC using mannose as a standard.

PARTIAL

PURIFICATION

TABLE Reverse

Reaction

bM)

MANNITOL:MANNOSE

l-OXIDOREDUCTASE

III 1-Oxidoreductase activity

(pmol/h/mg

Substrate

Specificity

pH 6.0

pH 7.5

pH 9.0

0.497 Na Na

0.810 0.000 0.000

0.124 Na” Na

20 20 20

Note. See Materials procedures. a Na, not analyzed.

and Methods

for a description

V

of Mannitol:Mannose

1-Oxidoreductase

protein)

MDH (amol/h/mg Substrate

Mannose Fructose Glucose

617

TABLE

of Mannitol:Mannose MDH

Substrate

OF

of experimental

pH 7.5

bM)

Mannitol Sorbitol Galactitol myo-Inositol

150 150 75

31.30 3.13 1.74

150

Effect of pH on Enzyme Activity

and Kinetic Constants

Initially, assays of MDH at 150 mM mannitol and 1 NAD+ revealed that the enzyme activity was not greatly affected by variations in pH between 7.5 and 9.0. Subsequently, the effect of pH on the kinetic constants Km and V,,, was investigated. Kinetic parameters were based on Lineweaver-Burk plots (highest R2 compared to other classic kinetic plots) and are summarized in Table IV. Normal Michaelis-Menten kinetics were exhibited for both mannitol and NAD with K,,, values of 72 and 0.26 mM, respectively, at pH 9.0. The V,,, was 40.14 pmol/h/ mg protein for mannitol synthesis at pH 9.0. At pH 7.5 the substrate saturation curve for mannitol did not reach a saturating level, resulting in a K,,, value of 228.69 mM of 85.64 pmol/h/mg protein. These values are and V,,, much higher than at pH 9.0, indicating that the saturating mannitol concentration at pH 9.0 was different and lower than that at pH 7.5. This is also reflected in the kinetic data of NAD at pH 7.5 where a higher K,,, value was observed as compared to pH 9.0. mM

pH 9.0

(100) (10)

30.96 6.77 2.20

(5)

0.00 (0)

galactitol (Table V). The apparent substrate specificity was higher at pH 7.5 as compared to pH 9.0. However, when a 99% pure sorbitol preparation was used at high concentration, a significant amount of contaminants could be detected by HPLC (data not shown). When sorbitol was used at 150 mM, the mannitol contamination was high enough to contribute to the apparent oxidation of sorbitol. Therefore, the actual specificity of the enzyme for mannitol might be greater than apparent from the data. NADP was not a substrate for activity in either the crude extract or the partially purified enzyme. MDH Effecters When the partially purified mannitol:mannose l-oxidoreductase was assayed under saturating conditions (150 mM mannitol and 2 mM NAD+) in the presence of potential effecters such as mannose (5 mM), sucrose (20 mM), mannose 6-phosphate (5 mM), fructose 6-phosphate (5 mM), glucose 6-phosphate (5 mM), and glucose l-phosphate (5 mM), no inhibition of the enzyme was observed (data not shown). The enzyme was strongly inhibited by NADH and was sensitive to alterations of NAD/NADH ratio (Fig. 5). Inorganic phosphate (1 mM), PPi (1 mM),

Mannitol:mannose l-oxidoreductase was very specific for mannitol but showed low activity with sorbitol and TABLE Parameters

2 ml4 NAD 5 mM NAD

IV

of Mannitol:Mannose

1-Oxidoreductase V msx

Substrate

PH

Mannitol NAD+ Mannitol NAD+

9.0 9.0 7.5 7.5

& 72.26 0.27 228.69 0.46

(pmol/h/mg

protein) 40.14 34.55 85.64 35.59

(21) (7)

of enzyme activity that commercial sor-

Substrate Specificity

Kinetic

(100)

0.00 (0)

a Numbers in parentheses equal the percentage observed relative to mannitol as a substrate. Note bitol is contaminated with mannitol.

trated with 90% (NH&SO, and remained stable at -80°C for several weeks. The enzyme was always stored in the presence of 1 to 2 mM DTT.

activity protein)”

Lo

I

I

0.1

0.2

NADH FIG.

5.

NADH

inhibition

as affected

0.3

(mM) by NAD

concentration.

618

STOOP

AND

ADP (1 mM), and ATP (1 mM) did not substantially affect the mannitol:mannose 1-oxidoreductase when assayed under both saturating (200 mM mannitol and 2 mM NAD+) and limiting NAD (200 mM mannitol and 0.2 mM NAD+) or limiting mannitol (75 mM mannitol and 2 mM NAD’) conditions (data not shown). ADP (1 mM) may be a very weak inhibitor at limiting NAD concentrations (data not shown). DISCUSSION

This report constitutes the first evidence of the presence of a mannitol oxidizing enzyme in the higher vascular plant celeriac (A. graveolens var. rapaceum). Based on purification away from alcohol dehydrogenase and phosphomannose isomerase, product identification, and reverse reaction data, it is concluded that this enzyme is a NAD-dependent mannitol:mannose 1-oxidoreductase. The enzyme is not unique to celeriac but was also found to be present in celery, A. graveolens var. dulce (data not shown). The conversion of mannitol to mannose by this plant enzyme is uniquely different from the mannitol oxidation occurring in lower organisms. Fungi, bacteria, lichens, and algae oxidize mannitol to fructose either directly (11) or through a phosphorylated intermediate (17) by a NADor NADP-dependent 2-oxidoreductase. The kinetic characteristics of the 1-oxidoreductase are similar to the 2oxidoreductase in that the K,,, value for mannitol is relatively high. Mannitol synthesis is also different between lower organisms and vascular plants. Celery synthesizes mannitol by a NADPH-dependent mannose-6-phosphate reductase which reduces mannose 6-phosphate to mannitol-l-phosphate, whereas mannitol in lower organisms is synthesized from fructose or fructose 6-phosphate. It remains to be determined if mannitol to mannose interconversion may occur but has escaped detection in lower organisms. Partially purified mannitol:mannose l-oxidoreductase did not exhibit complex metabolite control, since activity was not altered by hexoses, hexose-phosphates, or energy-related compounds. Besides the very high concentrations of the polyol mannitol in all tissues except root tips, a rather drastic difference in carbon storage was observed between shoot and root tissue of celeriac. Sucrose was the main soluble sugar stored in the root, while hexoses were the main soluble sugars stored in the shoot. The physiological importance or implication of this remains unclear. When stressed plants, which were root bound and stunted due to growth in small containers, were transferred to a larger container, a dramatic increase in root growth occurred. This high root growth was accompanied with a higher MDH activity (7 pmol/h/g fresh wt) as compared to nonstressed plants (3 pmol/h/g fresh wt). This high activity of MDH in root tissues suggests that this enzyme might be of physiological importance in pro-

PHARR

viding the sink tissue with carbon by converting mannitol to hexose or sucrose which can be used to support rapid root growth. Root tips as well as young petioles constitute strong sinks. MDH activity was observed in young petioles, supporting the idea that sink growth might be partially dependent upon mannitol utilization catalyzed by MDH activity. The magnitude of mannitol utilization might be influenced by the sucrose or hexose pool in the sink tissues, as indicated by Fig. 2, where both root tip and young petiole had very low sucrose and hexose pools and relatively high MDH activities. It has been suggested in the past that the stored mannitol pool is utilized by the plant only after sucrose pools are depleted (25). Although the highest MDH activities were observed in tissues with low sucrose and hexose content, providing circumstantial evidence to support the above hypothesis, no direct evidence was obtained to determine the effect of internal carbon pools on MDH activity. Celery has exceptionally high photosynthetic rates and apparently produces more biomass per unit land area than other crops with the exception of sugar cane [(26) and references therein)]. The photosynthetic mannitol formation in source leaves may result in high COZ fixation which equals that of C-4 plants in this otherwise typical C-3 plant. One role postulated for mannitol synthesis is that utilization of extrachloroplastic reductant dissipates photosynthetically produced reductant, NADPH (26), into the cytoplasm by way of an indirect shuttle involving exchange of DHAP into the cytoplasm, oxidation to 3PGA generating cytoplasmic NADPH, and subsequent exchange for PGA back into the chloroplast. The cytoplasmic NADPH pool is then available for synthesis of mannitol l-phosphate by the cytoplasmic mannose 6phosphate reductase (4, 18). Furthermore, excess redox equivalents produced by photosynthetic electron transport can be exported from the chloroplast by malate-oxaloacetate shuttle or DHAP-PGA shuttle (27) to the cytosol and shuttled into mitochondria where NADPH can be oxidized by mitochondrial electron transport, thereby contributing to the ATP supply of the cell (28). In this regard, Kramer and Heldt (29) observed a 30-40% inhibition of photosynthesis when the mitochondrial oxidative phosphorylation in barley protoplasts and leaves was selectively inhibited by oligomycin. It was postulated that plant cells contained a mechanism by which surplus chloroplastic NADPH could ultimately be oxidized during mitochondrial oxidative phosphorylation. This mechanism could also prevent overreduction of the chloroplast redox carriers by oxidizing reductive equivalents generated by photosynthetic electron transport. This further supports the role of polyol synthesis postulated above. Polyol synthesis and mitochondrial oxidative phosphorylation may represent different means of accomplishing a similar result, e.g., enhancing COZ fixation indirectly by dissipating excess photosynthetically generated NADPH that cannot be used for reductive processes within the

PARTIAL

PURIFICATION

OF

MANNITOL:MANNOSE

chloroplast. Thus, the extra step for cytoplasmic NADPH utilization in polyol synthesizing plants constitutes an additional outlet for carbon reduction. In mannitol forming and translocating plants, oxidation of translocated mannitol in sink tissues yields reduced pyridine nucleotide, which theoretically might impart an additional energetic efficiency as compared to sucrose or stachyose translocating plants which actually require phosphorylation of the translocated sugar upon entry into metabolism in sink tissues. Mannitol oxidation at sink tissues yields one NADH or the theoretical equivalent of three ATPs. Only one of these is required for phosphorylation of mannose to mannose 6-phosphate, which can subsequently be isomerized to enter central metabolism, resulting in a theoretical net yield of two ATPs for each molecule of mannitol oxidized to mannose and phosphorylated for further metabolism. Mannitol metabolizing plants might therefore represent a unique advantage over nonpolyol metabolizing plants in that both mannitol synthesis and mannitol utilization might result in a more efficient overall metabolism. Although only speculative, genetic transformation of plants for expression of the mannose 6-phosphate reductase in photosynthetic tissues and the mannitol:mannose l-oxidoreductase in sink tissues might ultimately prove to be a novel way to increase photosynthetic rates and biomass production in agricultural crops. ACKNOWLEDGMENTS The authors thank Dr. Jim Burton and Rosemary B. Sanozky for technical assistance and helpful discussions. We also thank Dr. Phil Kerr and D. B. Dempsey of E. I. DuPont for HPLC pulsed amperometric detection of mannose.

l-OXIDOREDUCTASE

619

4. Rumpho, 5. 6.

M. E., Edwards, G. E., and Loescher, W. H. (1983) Plant Physiol. 73, 869-873. Quain, D. E., and Boulton, C. A. (1987) J. Gen. Microbial. 133, 1675-1684. Martinez, G., Barker, H. A., and Horecker, B. L. (1963) J. Biol.

Chem. 238,1598-1603. 7. Yamanaka, K. (1975) in Methods

in Enzymology (Wood, W. A., Ed.), Vol. XLI, pp. 138, Academic Press, New York. 8. Ueng, S. T. H., Hartanowicz, P., Lewandoski, C., Keller, J., Holick, M., and McGuinness, E. T. (1976) Biochemistry l&1743-1749. 9. Ruffner, H. P., Rast, D., Tobler, H., and Karesch, H. (1978) Phytochemistry 17,865-868. 10. Niehaus, W. G., and Jiang, W. (1989) Mycopathologia 107, 131-

137. 11. Morton, N., Dickerson, A. G., and Hammond, J. B. W. (1985) J. Gen. Microbial. 131, 2885-2890. 12. Strobel, G. A., and Kosuge, T. (1965) Arch. Biochem. Biophys. 109,

622. 13. Wolff, 869. 14. Liss,

J. B., and Kaplan, M., Horwitz,

N. 0. (1956)

S. B., and Kaplan,

J. Biol.

Chem.

N. 0. (1962)

218, 849-

J. Biol.

Chem.

237,1342-1350. 15. Horwitz,

S. B., and Kaplan,

N. 0. (1964)

J. Biol.

Chem.

239,830-

838. D. E. F., and Kirst, G. 0. (1987) 16. Richter, W. G. (1981) 17. Kiser, R. C., and Niehaus,

Planta 170, 528-534. Arch. Biochem. Biophys.

211,613-621. W. H., Tyson, R. 18. Loescher, Bieleski, R. L. (1992) Plant J. K., and Loescher, 19. Fellman, 341. M. R., Douglas, 20. Thompson, Thorpe, T. A. (1986) Physiol. 21. Cram, W. J. (1984) Physiol. G., and 22. Trip, P., Krotkov,

H., Everard, J. D., Redgwell, R. J., and Physiol. 98, 1396-1402. W. H. (1987) Physiol. Plant. 69,337T. J., Obata-Sasamoto, Plant. 67, 365-369. Plant. 61, 396-404. Nelson, C. D. (1964) Am.

T. J., and

J. Bot.

51,

828-835.

1. Bieleski, R. L. (1982) in Encyclopedia of Plant Physiology (Loewus, F., and Tanner, W., Eds.), New Series, Vol. 13A, pp. 158192, Springer-Verlag, Berlin.

23. Bradford, M. (1976) Anal. Biochem. 72, 248. 24. Hubbard, N. L., Huber, S. C., and Pharr, D. M. (1989) Plant Physid. 91, 1527-1534. 25. Obaton, F. (1929) Reu. Gen. Bot. 41,622-633. 26. Fox, T. C., Kennedy, R. A., and Loescher, W. H. (1986) Plant Physiol. 82,307-311. 27. Heineke, D., Burgi, R., Grosse, H., Hoferichter, P., Peter, U., Fliigge,

2. Lewis, D. H. (1984) in Storage Carbohydrates in Vascular Plants. Distribution, Physiology and Metabolism (Lewis, D. H., Ed.), pp. 158179, Cambridge Univ. Press, Cambridge.

S., Stitt, 28. Kramer, 352-356.

3. Loescher,

29. Kramer,

REFERENCES

W. H. (1987)

Physiol.

Plant.

70,553-557.

U., and Held&

H. W. (1991) Plant M., and Heldt,

S., and Heldt,

H. W. (1991)

Physiol. 95, 1131-1137. H. W. (1988) FEES L&t. Plant

Physiol.

95,1270-1276.

226,