Characterization of muscle glyceraldehyde-3-phosphate dehydrogenase isoforms from euthermic and induced hibernating Jaculus orientalis

ELSEVIER Biochimica et Biophysica Acta 1243 (1995) 161-168

BB Biochi ~mic~a

et Biophysica A~ta

Characterization of muscle glyceraldehyde-3-phosphate dehydrogenase isoforms from euthermic and induced hibernating Jaculus orientalis

Abdelaziz Soukri a, Federico Valverde b, Nezha Hafid a, Mohamed S. Elkebbaj a,

Aurelio Serrano b,,

a Laboratoire de Biochimie, Biologie Cellulaire et Mol~culaire, Facult~ de Sciences-Ain Chock, Casablanca, Morocco b Instituto de Bioqulmica Vegetaly FotoMntesis, Consejo Superior de lnvestigaciones Ciendficas y Universidad de Sevilla, Sevilla, Spain

Received 18 May 1994; accepted 15 August 1994

Abstract

The specific activity of o-glyceraldehyde-3-phosphate (G3P) dehydrogenase (phosphorylating) (GPDH, EC 1.2.1.12) found in skeletal muscle of induced hibernating jerboa (Jaculus orientalis) was 3-4-fold lower than in the euthermic animal. The comparative analysis of the soluble protein fraction of these tissues by SDS-PAGE and Western blotting showed a significant decrease in the intensity of a protein band of about 36 kDa, the GPDH subunit, in hibernating jerboa. After using the same purification procedure, the GPDH from muscle of hibernating jerboa exhibited lower values for both apparent optimal temperature and specific activity than the enzyme from the euthermic animal. Non-linear Arrhenius plots were obtained in both cases, but the E~ values calculated for the GPDH from hibernating tissue were higher. Although in both purified enzyme preparations three isoelectric GPDH isoforms, exhibiting p I values in the range 8.2-7.5, were resolved by chromatofoeusing, clear differences were observed in these preparations concerning the relative contribution to the total enzymatic activity of the two main isoforms, named GPDH I (pI values, 8.1-8.2) and GPDH II (pI values, 7.8-7.9). Thus, whereas GPDH I was the major isoform purified from euthermic muscle, accounting for more than 90% of the total activity, the amount of activity due to GPDH II reached up to 65% in preparations of hibernating jerboa. All isoforms exhibited similar native and subunit molecular masses and cross-reacted with an anti-GPDH antibody raised against the GPDH I. However, the two muscle GPDH isoforms prevailing under hibernating conditions exhibited a decreased catalytic efficiency when compared with the corresponding major isoforms purified from euthermic animals, as indicated by their different specific activities and kinetic parameters, i.e. relatively high K m and low Vma ~ values. Since the glycolytic flow has been found to be widely reduced in skeletal muscle of induced hibernating jerboa, the changes in the GPDH isoforms described in the present study could provide a molecular basis to explain some of the metabolic changes associated with mammalian hibernation.

Keywords: Glyceraldehyde-3-phosphate dehydrogenase; Induced hibernation; Enzyme isoform; Chromatofocusing

1. Introduction

Glyceraldehyde-3-phosphate dehydrogenase (phospho- rylating) (GPDH, EC 1.2.1.12) is a key enzyme of the glycolytic pathway present in the cytosol of all organisms so far studied [1]. The glycolytic G P D H has been remark- ably conserved during evolution, having an homote- trameric structure with subunits of 3 5 - 3 7 kDa [1]. The

Abbreviations: G3P, D-glyceraldehyde-3-phosphate; GPDH, D- glyceraldehyde-3-phosphate dehydrogenase; PMSF, phenylmethylsulfo- nyl fluoride; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis

* Corresponding author. Fax: +34 5 4620154.

0304-4165//95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSDI 0 3 0 4 - 4 1 6 5 ( 9 4 ) 0 0 1 3 7 - 5

presence of enzyme isoforms of the glycolytic G P D H has already been described in skeletal muscle [2,3] but the physiological significance of this heterogeneity is not yet clear. The presence o f several G P D H s in photosynthetic cells and microorganisms has been also reported [4-7], but in this case they are actually different enzymes that cat- alyze different reactions, are located in diverse cellular compartments, and perform different physiological roles [4,5].

The jerboa (Jaculus orientalis), a small rodent from

desertic areas o f Moroccan Highlands, is an appropriate

organism to study metabolic regulation not only by its

remarkable tolerance to heat and dry diet but also because

it is one of the only certain small m a m m a l s that can

162 A. Soukri et al. /Biochimica et Biophysica Acta 1243 (1995) 161-168

undergo hibernation [8]. During hibernation, which can be induced in the laboratory when artificially cooled, the body temperature of this animal decreases to 6-9°C [9]. The glycolytic flow has been found to be widely repressed during this resting state [9]. This finding is in agreement with the fact that in muscle the main function of glycolysis is to provide ATP and the metabolite pool for respiration.

GPDH is a glycolytic allosteric enzyme that could be subjected to metabolic regulation [1]. Therefore, to explore the molecular mechanisms of the regulatory processes associated with hibernation we decided to undergo a com- parative study of the GPDHs from skeletal muscle of euthermic and induced hibernating jerboas.

In this paper we report that crude extracts from skeletal muscle of hibernating jerboa exhibit, when compared with euthermic animal preparations, a decreased GPDH specific activity concomitant with a reduction of the 36 kDa-sub- unit protein band in both SDS-PAGE gels and Western blots. Three GPDH isoforms, readily resolved by column chromatofocusing, were demonstrated in both physio- logical situations after using the same purification proce- dure. The change of the relative contribution of the two main GPDH isoforms to the total activity in these purified preparations and their different kinetic parameters could explain the repression of glycolysis found in the skeletal muscle of induced hibernating jerboa.

2. Materials and methods

2.1. Preparation of biological material

Jerboas (Jaculus orientalis) were captured in the sub- desert Moroccan East Highlands and kept in captivity, in a pre-acclimated room (22 + 2°C) with food and water, for about 3 - 6 mth. When necessary, animals were forced to hibernate in the laboratory by artificially cooling [10].

Briefly, they were placed in darkness at 4°C during 2-3 wk; then, food was removed during 1 week. Young adult jerboas of both sexes, of about 6 months old and weighting 130-150 g, were decapitated and skeletal muscles were immediately removed and frozen.

2.2. Enzyme purification

Unless otherwise indicated all steps were performed at 4°C. Centrifugations were carried out at 20 000 × g for 45 min.

Step 1. 100 g of skeletal muscle were ground and homogenized using a Sorvall mixer in 25 mM Tris-HC1 buffer, pH 7.5, containing 2 mM EDTA, 10 mM 2-mer- captoethanol and protease inhibitors (2 mM PMSF, 2 mM benzamidine, and 5 mM e-amino-n-caproic acid) at a ratio of 3 m l / g of fresh tissue. The resulting homogenate was centrifuged and the supernatant (soluble protein fraction) considered as the crude extract.

Step 2. The supematant from step i was brought to 66%

( w / v ) saturation with solid ammonium sulfate. After 1 h at 4°C, the suspension was centrifuged and the supernatant was precipitated with ammonium sulfate to a final satura- tion of 88% (w/v). The final pellet after centrifugation was dissolved in 25 mM Tris-HCl, pH 7.5, containing 0.1 mM EDTA. The protein solution was dialysed twice against 1 liter of the same buffer and eventually centrifuged.

Step 3. The supernatant from step 2 was chromato- graphed on a Blue Sepharose CL-6B column (1 × 6 cm) equilibrated with 2 bed volumes of buffer A (25 mM Tris-HC1, pH 7.5, 2 mM EDTA and 10 mM 2-mercapto- ethanol). The column was washed with 3 bed volumes of buffer A and subsequently with 2 bed volumes of the same buffer adjusted to pH 8.6 (buffer B). The enzyme was eluted with buffer B containing 10 mM NAD + at a flow rate of 20 m l / h . 2-ml fractions containing GPDH activity were collected, concentrated and washed with buffer B by ultrafiltration on a Diaflo PM-10 Amicon membrane.

Step 4. The active pool was dialysed against 25 mM Tris-HC1 buffer, pH 9.8, containing 1 mM EDTA and 5 mM 2-mercaptoethanol (starting buffer). Column chro- matofocusing in the pH range 9.0 to 5.5 was performed in a Polybuffer Exchancher PBE-94 column (1 × 18 cm) equilibrated with starting buffer. After application of the concentrated enzyme solution (about 10 ml), the column was washed with 5 ml of starting buffer. The enzyme was eventually eluted at a flow rate of 12 m l / h by washing the column with 10 bed volumes of a 10-fold diluted mixture of Polybuffer 96/Polybuffer 74 (30/70, v / v ) adjusted to pH 5.5 with acetic acid. The pooled active fractions were concentrated and equilibrated in standard buffer supple- mented with 0.1 M NaCI by ultrafiltration as described above.

2.3. Determination of enzyme activity

Unless otherwise stated enzymatic activity in the oxida-

tive phosphorylation was determined spectrophotometri-

cally at 30°C by monitoring the appearance of NADH. The

1-ml reaction mixtures contained 0.1 M triethanolamine-

HC1 buffer (pH 8.9), 2 mM EDTA, 1 mM NAD ÷ and 2

mM o-G3P. Only initial rates were considered. Kinetic

parameters for NAD + and G3P were determined in the

conditions described by Ferdinand [11]. To determine opti-

mal pH, enzymatic activity was measured over a wide

range of pH (from 5 to 10) with different buffers (acetate,

imidazole, Tris and carbonate/bicarbonate) adjusted to the

same ionic strength than the standard reaction mixture. To

determine apparent optimal temperature, reactions were

carried out in a 5 to 65°C temperature range using a

thermostated cuvette holder connected with a refrigerated

bath circulator. One unit of enzyme is defined as the

amount which catalyzes the formation of 1 /zmol of

NADH per min under the conditions used. Protein was

determined by the method of Bradford [12]. Activity levels

in cell-free extracts were expressed as specific activity (mU/mg of protein).

2.4. Gel electrophoresis

Polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate (SDS-PAGE) was carried out in 12% ( w / v ) acrylamide slab gels (Mini-Protean Bio-Rad, Richmond, CA, USA) according to Laemmli [13]. Proteins were stained with 0.2% (w/v) Coomassie brilliant blue R-250 in methanol/acetic acid/water (4:1:5) for 30 min at room temperature. Stained protein bands were analyzed and quantified with an analytical imaging instrument (Bio Image, Millipore, Ann Arbor, MI, USA).

(Sigma Chemical Co., St. Louis, MO, USA). Immunos- rained protein bands were quantified by the analytical imaging system above described.

2.6. Chemicals

NAD +, D-G3P diethyl acetal, PMSF, benzamidine, e- aminocaproic acid, imidazole, trietanolamine, Tris, Tricine, EDTA were purchased from Sigma Chemical Co. (St.

Louis, MO, USA). Blue Sepharose CL-6B, Polybuffer Exchanger PBE 94, Polybuffer 96 and Polybuffer 74 were obtained from Pharmacia Fine Chemicals (Uppsala, Swe- den). All other chemicals were of analytical grade.

2.5. Western blotting 3. Results and discussion

The GPDH protein was detected immunologically, in either cell-free extracts or purified preparations from eu- thermic and hibernating jerboas, after SDS-PAGE (12%

acrylamide) and subsequent transfer to nitrocellulose. After blocking in non-fat milk, membranes containing samples were exposed to a 1:250 dilution of a monospecific poly- clonal antibody raised in rabbit against the chromato- focusing resolved GPDH I isoform purified from skeletal muscle of euthermic jerboa. Detection of the GPDH pro- tein was performed with a 1:1000 dilution of a goat anti-rabbit IgG antibody-horseradish peroxidase conjugate

The specific activity level of GPDH found in soluble protein fractions from skeletal muscle of euthermic jerboas (about 1.5 U / m g of protein) was 3-4-fold higher than that measured in preparations from the same tissue of induced hibernating animals. A relevant differential feature of the SDS-PAGE protein patterns of these crude preparations was the marked reduction (about 3-fold) in hibernating tissue of a major protein band corresponding to a 36 kDa molecular mass (Fig. 1A, lanes a and b). This difference was even more evident in protein fractions corresponding to the 66-88% saturation range of ammonium sulfate (Fig.

Marker Marker

molecular molecular

masses a b c d e masses

(kDa) (kDa)

9 7 ~ . . . 1 1 6 ,--,-

6 6 8 4

4 5 ~ 5 8 . . . .

4 5

31 - -

. . . . . . . 3 6 . 5 . . . - -

2 6 . 6 - - -

21

a b

Fig. 1. A. Coomassie-stained SDS-PAGE electrophoretogram showing the protein patterns corresponding to preparations of skeletal muscle from euthermic and hibernating J. orientalis. Lanes a and d represent, respectively, crude extract and 66-88% ammonium sulfate protein fraction from euthermic animal.

Lanes b and e represent crude extract and ammonium sulfate protein fraction from hibernating animal. A similar amount of protein, about 50 g g , was

applied to each lane. Lane c corresponds to protein markers. The 36-kDa protein band, considered as the putative GPDH subunit, is indicated by the

arrows. B. Western-blot analysis of crude extracts of skeletal muscle of euthermic 0ane a) and hibernating (lane b) J. orientalis. Aliquots (about 50/~g of

protein) of cell extracts were subjected to SDS-PAGE and after electrophoretic transfer to nitrocellulose membrane, the Western-blot was developed as

indicated in the 'Materials and Methods' section. The arrow indicates the band corresponding to the 36 kDa GPDH subunit. The position and molecular

masses of the used prestained protein markers are also indicated.

164 A. Soukri et al. /Biochimica et Biophysica Acta 1243 (1995) 161-168

1.

"I" 0.8"

t) <

~,,

"6

0.4

0

00000~

0

i I I i i I I

2.9 3 3.1 3.2 3.3 3.4 3.5 3.6 3.7 l f r x l 0 a (K-l)

Fig. 2. Arrhenius plots of GPDH enzyme preparations purified from skeletal muscle of euthermic ( D ) and hibernating ( 0 ) J. orientalis. The arrows indicate changes in the slope of the plotted lines. The numbers adjacent to the plotted lines are the calculated apparent E a values in kcal/mol.

1A, lanes d and e). Since all glycolytic GPDHs so far studied are homotetrameric enzymes with 34-38 kDa sub- units [1] which are relatively abundant in muscle tissue where an intense glycolytic acitivity occurs, we initially assumed that the 36 kDa protein was the GPDH subunit.

Western blot analysis of these crude preparations using an

anti-GPDH antibody showed a 3-fold reduction of the immunostained 36-kDa band in hibernating tissue (Fig.

1B) demonstrating that this 36-kDa protein actually repre- sents the jerboa muscle GPDH subunit. Since, as shown in SDS-PAGE gels, most other protein bands remained sub- stantially unchanged in these crude extracts (see Fig. 1),

Table 1

Purification of GPDH isoforms from skeletal muscle of euthermic J. orientalis

Fraction Total protein Total act. Specific act. Purification factor Yield

(mg) (U) ( U / m g of protein) (fold) (%)

Crude extract 2571 3600 1,4 1

Ammonium sulfate (66-88%) 395 1700 4.3 3

Blue Sepharose CL-4B 67 1440 21.5 15

Chromatofocusing a

GPDH I (pl 8.12) 0.8 40.0 50.0 36

GPDH II (pl 7.80) 1.1 3.5 3.2 2

GPDH III ( p l 7.58) 0.5 0.4 0.8 -

100 47 40

a Only 180 U, about 8,4 mg of protein, were chromatofocused on the PBE 94 column.

Table 2

Purification of GPDH isoforms from skeletal muscle of hibernating J. orientalis

Fraction Total protein Total act. Specific act. Purification factor Yield

(rag) (U) ( U / r a g of protein) (fold) (%)

Crude extract

Ammonium sulfate (66-88%) Blue Sepharose CL-4B Chromatofocusing a GPDH I ( p l 8.15) GPDH II ( p l 7.85) GPDH III ( p l 7.55)

4000 1800 0.45 1 100

354 460 1.3 3 25

60 400 6.7 15 22

0.54 17.7 32.7 73

0.65 27.1 41.7 93

2.5 0.6 0.2 -

a Only 160 U, about 24 mg of protein, were chromatofocused on the PBE 94 column.

our results suggest a specific decrease in the amount of GPDH protein in the soluble protein fraction from hiber- nating animal. On the other hand, SDS-PAGE gels also showed changes in the amount of two other minor low- molecular mass proteins, namely the disappearance of a 26 kDa-protein and the appearance of a 22 kDa-protein, in crude muscle preparations of hibernating jerboa (see Fig.

1). In the same way it is significant to note that changes in levels of several blood proteins of similar low-molecular masses (in the range 20-27 kDa) are specifically associ- ated with hibernation in another mammalian hibernator [14]. Determining wether this is actually the case with the low-molecular mass proteins found in the skeletal muscle of jerboa needs further research work.

Additional information was obtained by purification and subsequent analytical chromatofocusing of GPDH purified from skeletal muscle from both euthermic and hibernating jerboas. Tables 1 and 2 present respectively typical purifi- cations from euthermic and hibernating animals. As previ- ously reported for other NAD÷-dependent GPDHs [15], dye-ligand chromatography on Blue Sepharose is a very effective purification step. Both GPDH preparations thus purified from euthermic and hibernating jerboas exhibited under SDS-PAGE a major 36 kDa band, the GPDH sub- unit, and a few minor protein contaminants (data not shown). Noteworthy, the GPDH preparation purified by dye-ligand chromatography from hibernating tissue exhib- ited a lower specific activity than the corresponding prepa- ration of euthermic tissue (see Tables 1 and 2). The effect of temperature on the enzymatic activity of these purified GPDH preparations has also been studied. Differences have been found in the apparent optimal temperature val- ues, being respectively about 35 and 45°C for hibernating and euthermic GPDH preparations. Fig. 2 shows the Ar- rhenius plots, in which the logarithm of enzyme activity is plotted versus reciprocal absolute temperature, of these two GPDH preparations. Slope changes were observed in both Arrhenius plots over the temperature range investi- gated. These slope changes occur in both plots at the same temperature values (see Fig. 2). However, the calculated apparent E a values for the GPDH from hibernating tissue, namely 2.38 and 6.84 kcal/mol above and below the discontinuity point at 15°C, were clearly higher than those calculated for the enzyme from euthermic tissue, namely 1.59 and 3.14 kcal/mol (see Fig. 2). Thus, these data suggest that the conformation of the GPDH from euther- mic tissue is more favourable to the enzymatic reaction than that of the same enzyme from hibernating tisssue. It has been reported that non-linear Arrhenius plots may be caused by temperature-induced conformational changes of soluble oligomeric enzymes [16].

The subsequent use of column chromatofocusing, a high-resolution technique of protein separation according to pI's [17,18], allowed us to separate in both preparations three GPDH isoforms which were named going from the most basic to the most acidic form, GPDH I (pI, 8.1-8.2),

<1

i--

I.- t.) 14J

N Z

0 0

83 g6 102 ~ I

~ 4 5

• -- I ~ . . . . ~ 3 6

20 40 60 80

FRACTION NUMBER

12

10

E

¢ q

i/.1~

On"

en 2 O t/l If3

0

Fig. 3. Resolution by column chromatofocnsing of GPDH isoforms purified from skeletal muscle of euthermic J. orientalis. A sample containing about 8.4 mg of protein was applied to a Polybuffer Exchanger PBE 94 column (1 × 18 cm) and the enzyme was eluted by using a pH gradient ( O ) generated by 10 bed volumes of a 10-fold diluted mixture of Polybuffer 96/Polybuffer 74 (30/70, v / v ) adjusted with acetic acid to pH 5.5. One-ml fractions were collected. Absorbance at 280 nm and enzyme activity were measured. The photograph shows the Coomassie- stained SDS-PAGE electrophoretogram of the three activity peak frac- tions which have been marked with asterisks in the elution profile. The arrow indicates the 36-kDa protein band corresponding to the GPDH subunit. The positions and the molecular masses of the protein markers are also shown.

GPDH II (pl, 7.8-7.9) and GPDH III (pI, 7.5-7.6) fol- lowing the elution order under chromatofocusing (see Ta- bles 1 and 2). GPDH III was nevertheless a very minori- tary isoform, accounting for less than 2% of the total enzyme activity. Most significantly, whereas after chro- matofocusing of GPDH preparations from euthermic mus- cle the most basic form GPDH I clearly appeared as the single major isoform, accounting for more than 90% of the total activity (Fig. 3), this was not the case for preparations from hibernating animals, in which the amount of activity due to GPDH I was about 30% of the total and that of GPDH II reached up to 65% of the total activity (Fig. 4).

As it is also shown in Figs. 3 and 4, SDS-PAGE gels of the activity peak fraction corresponding to the isoform GPDH I always exhibited, both in euthermic and hibernat- ing preparations, only one protein band of about 36 kDa -the expected value for the enzyme subunit (1)- indicating a purification of the protein to electrophoretic homogene- ity. Concerning GPDH II isoform, the 36-kDa protein band (GPDH subunit) was also observed in SDS-PAGE, but sometimes co-purified together with a feeble minoritary protein band of either 22 kDa (in preparations of euthermic animals) or 11 kDa (in preparations of hibernating ani- mals) (data not shown). On the other hand, all isoforms of enzyme preparations from euthermic and hibernating tis- sues cross-reacted with a monospecific anti-GPDH anti- body raised against the GPDH I isoform, only one band of about 36 kDa being immunostained in Western blots (data not shown).

The comparison of kinetic parameters, obtained by sis-

tematic variation of substrates, of the two main muscle

166 A. Soukri et aL / Biochimica et Biophysica Acta 1243 (1995) 161-168

15

A

E lO ::) v

>.

I -

ILl

=E N z I,u

0 o

~b°°°d:°O°° ° ~ ° co o o o o o Oo q ~ o o ~ .

= g

F R A C T I O N N U M B E R

12

E

6 ' ~ o z z z ~ O ' r "

4 . ~ "

m n., 2 0

m <

0

Fig. 4. Resolution by column chromatofocusing of GPDH isoforms purified from skeletal muscle of hibernating J. orientalis. A sample containing about 8.4 mg of protein was chromatographed under the conditions described in the legend of Fig. 3. The photograph shows the Coomassie-stained SDS-PAGE electrophoretogram of the two activity peak fractions marked with asterisks in the elution profile, which corre- spond to the two main isoforms. The arrow indicates the 36-kDa protein band corresponding to the GPDH subunit. The positions and the molecu- lar masses of the protein markers are also shown.

isoforms GPDH I and GPDH II of euthermic and hibernat- ing jerboas showed marked differences (Table 3). Concern- ing euthermic isoforms, GPDH I exhibited significantly higher affinities for both substrates G3P and NAD + (i.e., lower K m values) and higher breakdown efficiencies of the enzyme-substrate complexes (i.e., higher Vma x values) than GPDH II. This was clearly shown by the Vraax/K m ratios which must be as high as possible to maximize the catalytic efficiency of the enzymes. The inverse situation was observed for the two major hibernating isoforms, GPDH II exhibiting a higher catalytic efficiency than GPDH I. These results are also in agreement with the specific activity values for purified preparations (see Ta- bles 1 and 2). Differences have also been found when comparing euthermic and hibernating isoforms. As is indi- cated by the corresponding V m a x / K m ratios, the catalytic efficiency clearly decreased in the hibernating GPDH I isoform when compared with the euthermic one, while the opposite was found for GPDH II. However, the catalytic efficiency of the major isoform (GPDH II) found under hibernating conditions is still lower than that of the pre-

dominant isoform (GPDH I) in the euthermic animal. As a whole, therefore, the GPDH isoforms found in skeletal muscle of hibernating jerboas exhibit lower catalytic effi- ciency than those from the same tissue of euthermic animals. It should be noted that these catalytic differences, namely lower V m a x / g m values for the hibernating GPDH isoforms, were also observed at the apparent optimal tem- peratures for both euthermic and hibernating GPDHs (data not shown). The relatively slight differences found be- tween the K m values for NAD + of the GPDH isoforms prevailing in both physiological situations could be due to the fact that GPDH is a NAD -containing enzyme, having one nucleotide molecule bound per enzyme subunit [19].

Thus, this cofactor may preserve in some extent the active site against possible structural changes induced by the dramatic decrease of body temperature during hibernation.

On the other hand, no significant differences have been observed in the pH dependence of the enzymatic activity of all muscle GPDH isoforms from both euthermic and hibernating jerboas, being about 8.5 the optimum pH value in every case. All GPDH isoforms from both euthermic and hibernating tissues were strongly inactivated by the sulphydryl modifying agent iodoacetamide, thus indicating the presence of thiol groups essential for the enzyme activity [19].

The results above presented are in sharp contrast with the low temperature adaptation of enzymes found in ec- tothermic organisms. Thus enzymes from low cell temper- ature species exhibit both higher specific activities and lower activation energies and enthalpies than the homolo- gous enzymes of high cell temperature species, conserving moreover proper apparent K m values at low temperatures -i.e., at 5°C they exhibit g m values for substrates similar to those of warm-blood animals at 35°C [20-23]. This discrepancy is not surprising because whereas enzymes of ectothermic organisms undergo a molecular adaptation in order to maintain an active metabolism at low tempera- tures, this should not be certainly the case for enzymes of hibernating animals in which basal metabolism must be drastically reduced. Since GPDH is a key component of a central metabolic pathway (glycolysis) that is strongly reduced during hibernation [9], it should be a good candi- date for a hibernation-triggered enzyme modification.

Table 3

Kinetic parameters of the two main muscle GPDH isoforms of euthermic and hibernating J. orientalis

K m G3P Vma x G3P Vmax/Km G3P K m NAD +

(/zM) ( / z m o l / m i n ) (. 103 ) (/zM)

Vma x NAD + Vmax/K m NAD + (/xmol/min) (" 10 3)

euthermic GPDH I ( p l 8.12) 101 11.11 euthermic GDPH II ( p l 7.80) 160 2.85 hibernating GPDH I ( p l 8.15) 502 7.03 hibernating GDPH II ( p l 7.85) 250 8.55

110 69 11.38 164

18 160 3.03 19

14 161 7.11 44

34 50 8.50 170

a GPDH activity was determined according to the procedure of Ferdinand [11] for the oxidative phosphorylation. Kinetic parameters, which were

determined by Lineweaver-Burk double reciprocal plots, are means of three independent values (S.E. were in all the cases less than 15% of the mean

values).

Concerning the cause for the presence of different GPDH isoforms, it is interesting to notice that the partially purified preparations from both euthermic and hibernating tissues used for chromatofocusing, presented under gel filtration chromatography on Sephadex G-200 only one activity peak, which corresponded to the typical homote- trameric enzyme (about 150 kDa). Therefore, the possibil- ity of different aggregational states as the cause of the different GPDH isoforms could be ruled out. Moreover, since no significant differences have been observed in their subunit molecular masses, having a cocktail of protease inhibitors been used in the GPDH purification (see Materi- als and Methods section) a proteolytic origin is not proba- ble. Since no evidence of proteolytic digestion was found for phosphofructokinase (another glycolytic enzyme) puff- fled without protease inhibitors from muscle of euthermic and hibernating jerboas, namely the enzyme from both sources exhibits the same native molecular mass, immuno- chemical reactivity and C-terminal residue [24], we can assume that the GPDH isoforms purified from these tissues in the presence of protease inhibitors should not be degra- dation products. On the other hand, since all isoforms are recognized by an anti-GPDH antibody raised against one of them, namely the euthermic GPDH I, they should be very similar proteins. Therefore, the GPDH isoforms found in skeletal muscle of jerboa may probably be different charge isomers having similar native molecular masses. In this respect it should be noted that the three GPDH iso- forms isolated from flounder muscle are distinct conforma- tional forms of a multimeric enzyme composed of one subunit type [3]. However, our results do not discard the possibility of these isoforms being encoded by different genes. In fact, the length of time needed to artificially induce hibernation (several weeks) is sufficient long to allow changes in the expression of genes and, therefore, the replacement of the euthermic by the hibernating isoen- zymes if they were coded by different genes. Noteworthy, it has been recently proposed that hibernation must be the result of a reprogramming of existing mammalian capabili- ties through the differential expression of existing genes [25]. There are several cases of glycolytic GPDH isoen- zymes having different genetic origin. In this way, unicel- lular eukaryotes have several GPDH genes, i.e. three in yeast and two in Trypanosoma brucei, that codify for different GPDH isoenzymes [1]. In the thoroughly studied nematode Caenorhabditis elegans [26] two GPDH isoen- zymes -their subunits being encoded by four non-allelic genes- have been detected. However, this is not the case for mammals since although many GPDH gene copies (more than 200 in rat and mouse) have been found in mammalian genomes, only one is transcribed, the others being actually pseudogenes [27]. Noteworthy, it has been reported for rat pyruvate kinase, another glycolytic en- zyme, that isoenzymes are produced from a single gene by using different promoters [28]. Therefore, although the GPDH isoenzymes found in skeletal muscle of jerboa

could be due to post-transcriptional modifications, as is the case in rat [2], studies are currently in progress to explore a possible genetic origin.

Summarizing, mammals in hibernation are in a special physiological resting state the molecular basis of which is still poorly understood. Some physiological changes asso- ciated with hibernation are considered, however, under regulatory and genetic control [14,25]. Since glycolysis is markedly reduced during hibernation in skeletal muscle of jerboa, the changes that we describe here for muscle GPDH isoforms of this mammalian hibernator could help to understand some of the metabolic changes associated with the hibernating state.

Acknowledgements

This work is part of a Collaborative Research Project between CNR (Morocco) and CSIC (Spain), and has been partially supported by DGICYT (Spain) (grants PB 90-99 and PB 91-85) and the Autonomous Government of An- dalusia (Junta de Andalucla, Spain). The authors thank Profs. M. Losada and M.A. De la Rosa for their interest and help, Dr. N. Bourhim for providing induced hibernat- ing animals and Dr. A. Malki for his collaboration in antibody preparation.

References

[1] Fothergill-Gilmore, L.A. and Michels, P.A.M. (1993) Prog. Biophys.

Molec. Biol. 59, 105-235.

[2] Edwards, Y.H., Clark, P. and Harris, H. (1976) Ann. Hum. Genet.

40, 67-77.

[3] Marangos, P.J. and Constantinides, S.M. (1974) J. Biol. Chem. 249, 951-958.

[4] Iglesias, A., Serrano, A., Guerrero, M.G. and Losada, M. (1987) Biochim. Biophys. Acta 925, 1-10.

[5] Serrano, A., Mateos, M.I. and Losada, M. (1991) Biochem. Biophys.

Res. Comm. 181, 1077-1083.

[6] Mateos, M.I. and Serrano, A. (1992) Plant Sci. 84, 163-170.

[7] Lambeir, A.M., Loiseau, A.M., Kuntz, D.A., Velleux, F.M., Michels, P.A. and Opperdoes, P.A.M. (1991) Eur. J. Biochem. 198, 429-435.

[8] Hooper, E.T. and El Hilali, M. (1972) J. Mammal. 53, 574-593.

[9] Andjus, R.K., El Hilali, M., Veillat, J.P. and Baddouri, K. (1974) J.

Physiol., Lond., 68, 531-542.

[10] Bourhim, N., Elkebbaj, M.S., Ouafik, L.H., Chantard, T.H., Giraud, P. and Oliver, C. (1993) Comp. Biochem. Physiol. 104B, 607-615.

[11] Ferdinand, W. (1964) Biochem. J. 92, 578-585.

[12] Bradford, M.M. (1976) Anal. Biochem. 72, 248-254.

[13] Laemmli, U.K. (1970) Nature (London) 227, 680-685.

[14] Takamatsu, N., Ohba, K.-I., Kondo, J., Kondo, N and Shiba, T.

(1993) Mol. Cell. Biol. 13, 1516-152l.

[15] Thompson, S.T., Cass, K.H. and Stellwagen, E. (1975) Proc. Natl.

Acad. Sci. USA 72, 669-672.

[16] Massey, V., Curti, B. and Ganther, H. (1966) J. Biol. Chem. 241, 2347-2357.

[17] Serrano, A. (1986) Anal. Biochem. 154, 441-448.

[18] Serrano, A. and Llobell, A. (1993) Planta 190, 199-205.

[19] Soukri, A., Mougin, A., Corbier, C., Wonacott, A., Branlant, C. and

Branlant, G. (1989) Biochemistry 28, 2586-2592.

168 A. Soukri et al. / Biochimica et Biophysica Acta 1243 (1995) 161-168 [20] Low, P.S., Bada, J.L. and Somero, G.N. (1973) Proc. Natl. Acad.

Sci. 70, 430-432.

[21] Somero, G.N. (1975) J. Exp. Zool. 194, 175-188.

[22] Low, P.S. and Somero, G.N. (1977) Nature 266, 276-278.

[23] Dahlhoff, E. and Somero, G.N. (1991) J. Exp. Biol. 159, 473-487.

[24] El Hachimi, Z., Tuane, M., Boissonnet, G., Benjouad, A., Des- madril, M. and Yon, J.M. (1992) Comp. Biochem. Physiol. 102B, 507-513.

[25] Srere, H.K., Wang, L.C.H. and Martin, S.L. (1992) Proc. Natl.

Acad. Sci. 89, 7119-7123.

[26] Huang, X.Y., Barrios, L.A.M., Vonkhomporn, P., Honda, S., Albert- son, D.G. and Hecht, R.M. (1989) J. Molec. Biol. 206, 411-424.

[27] Piechaczyk, M., Blachard, J.M., Riaad-E1 Sabouty, S., Dani, C., Marty, L. and Jeanteur, P. (1984) Nature 312, 469-471.

[28] Noguchi, T., Yamada, K., Inoue, H. and Tanaka, T. (1987) J. Biol.

Chem. 262, 14366-14371.