See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/20559331

Rapid purification of pig heart NAD+-isocitrate dehydrogenase. Studies on the

regulation of activity by Ca2+, adenine nucleotides, Mg2+ and other metal ions

Article in Biochemical Journal · October 1989

DOI: 10.1042/bj2630445 · Source: PubMed

CITATIONS READS

25 77

2 authors:

Guy A Rutter Richard Denton

Imperial College London University of Bristol
612 PUBLICATIONS 27,023 CITATIONS 243 PUBLICATIONS 19,569 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Guy A Rutter on 16 October 2014.

The user has requested enhancement of the downloaded file.

Biochem. J. (1989) 263, 445-452 (Printed in Great Britain) 445

Rapid purification of pig heart NAD+-isocitrate dehydrogenase

Studies on the regulation of activity by Ca2+, adenine nucleotides, Mg2` and other metal ions
Guy A. RUTTER and Richard M. DENTON
Department of Biochemistry, School of Medical Sciences, University of Bristol, Bristol BS1 ITD, U.K.

1. A new procedure for purifying pig heart NAD+-isocitrate dehydrogenase from mitochondrial extracts has
been developed. This relies on the use of f.p.l.c. techniques and exploits the hydrophobic properties of the
gel-filtration medium Superose 6 at high ionic strength. A 300-fold purification to apparent homogeneity is
achieved within 5 h and with a yield of > 20 %. 2. The enzyme had an apparent native molecular mass on
gel filtration of 320 kDa. In agreement with previous studies [Ramachandran & Colman (1980) J. Biol.
Chem. 255, 8859-8864], three subunits (all close to 38 kDa) were separable by isoelectric focusing. 3. This
preparation was used to investigate the effects of adenine nucleotides, KCl and the required bivalent metal
ions, Mg2+ and Mn2+, on the regulation of the enzyme by Ca21. 4. In the presence of 1.5 mM-ADP, increasing
the concentration of Mg2" from 20 ,SM to 6.0 mm raised the concentration of Ca2+ required for half-maximal
effect (K0 5 value) from 1.2 /LM to 232 /LM. Similarly, in the presence of 2.5,uM-Mn2", a Ko.5 value for Ca2"
of 3.3 ,sM was obtained, and this value was increased to 8.9,M in the presence of 100 /tM-Mn2+. In the
presence of 1 mM-Mg2" and 1.5 mM-ADP, the K0.5 value for Ca2+ was raised from 4.7 ,uM to 10 UtM by
75 mM-KCl.

INTRODUCTION permeabilized mitochondria and mitochondrial extracts

(Rutter & Denton, 1988). Furthermore, these studies
NAD+-isocitrate dehydrogenase (NAD-ICDH; EC suggested that the sensitivity of the enzyme to Ca2+ is
1.1.1.41) is located exclusively within mitochondria in strongly influenced by the ADP/ATP ratio. It therefore
mammalian cells and represents an important control seemed important to investigate the effects of adenine
point of the citrate cycle (see Hansford, 1985; Williamson nucleotides and Mg2" or Mna+ on the regulation by Ca2"
& Cooper, 1980; Plaut & Gabriel, 1983; Gabriel et al., of purified NAD-ICDH and, in particular, to compare
1986). The activity of the enzyme, which is absolutely the Ca2l-binding properties of the purified enzyme with
dependent on the presence of Mg2", Mn2+ or Co2+ ions those of 2-oxoglutarate dehydrogenase and the pyruvate
(Plaut & Sung, 1954; Plaut, 1970), is exquisitely sensitive dehydrogenase complex.
to regulation by a number of metabolites (Plaut & However, although satisfactory methods exist for
Gabriel, 1983), including the allosteric activator ADP purifying these last two enzymes (McCormack & Denton,
(Chen & Plaut, 1963; Goebell & Klingenberg, 1964) and 1979; Cooper et al., 1974), those for NAD-ICDH (Plaut,
the inhibitors ATP and NADH (Chen & Plaut, 1963; 1969; Giorgio et al., 1970; Shen et al., 1974; Ehrlich
Plaut & Aogaichi, 1968). et al., 1981) are lengthy and can make the isolation of
At sub-saturating isocitrate concentrations, Ca2+ ions this unstable enzyme difficult.
(in the presence of an adenine nucleotide) enhance the In the present paper a more rapid and convenient
activity of NAD-ICDH up to 10-fold (Denton et al., means of purifying pig heart NAD-ICDH from mito-
1978; Aogaichi et al., 1980; Gabriel et al., 1985; Rutter chondrial extracts is described. This is based on the use
& Denton, 1988). The stimulation by Ca21 of this enzyme, of f.p.l.c. and exploits the hydrophobic properties of the
and also of two other mitochondrial dehydrogenases, gel-filtration medium, Superose 6, which are apparent at
2-oxoglutarate dehydrogenase (McCormack & Denton, high ionic strength. By this approach a 50-100-fold
1979) and the pyruvate dehydrogenase complex (Denton purification of the enzyme to near homogeneity can be
et al., 1972), may allow hormones and other extracellular achieved from a 35-65 00-satd.-(NH4)2SO4 fraction in
stimuli to directly influence mitochondrial oxidative a single step. Further purification and concentration
metabolism (for reviews see Denton & McCormack, by anion-exchange (Mono Q) chromatography yields
1985; Hansford, 1985; Denton et al., 1987; McCormack electrophoretically pure NAD-ICDH.
et al., 1989). The regulation of this preparation is described, in-
Recent studies have indicated that the Ca2+-sensitivity cluding the recognition that the Ca2+-sensitivity of the
of NAD-ICDH may be substantially less than that of the enzyme is critically dependent on the concentration of
other mitochondrial Ca21-sensitive dehydrogenases when Mg2 In the accompanying paper (Rutter & Denton,
.

the enzymes are assayed under identical conditions in 1989) the preparation is used to study Ca2+ binding.

Abbreviations used: NAD-ICDH, NAD+-isocitrate dehydrogenase; D,-IC, threo-D.-isocitrate; HEDTA, N-(2-hydroxyethyl)ethylenediaminetri-

acetate; PMSF, phenylmethanesulphonyl fluoride. Throughout this paper, [Ca2+] and [Mg2"] represent the concentrations of the free unbound
species of these metal ions.

Vol. 263
446 G. A. Rutter and R. M. Denton

EXPERIMENTAL Ca2l-free (< 0.5 /tM-Ca2+) medium was prepared by

Materials passing 20 mM-Mops/triethanolamine, pH 7.2, 2 mM-
NAD+ and 1.5 mM-ADP over a 60 cm x 1 cm column of
Sources of chemicals and biochemicals were given by Chelex ion-exchange resin. Assays of activity in this
Rutter & Denton (1988). In addition, pepstatin, antipain medium were made as described above, with additions as
and leupeptin were from Cambridge Research Biochemi- indicated.
cals, Harston, Cambs., U.K., and ultra-pure urea was Free concentrations of metal ions and of metal-ligand
from Bethesda Research Laboratories, Life Technologies complexes at pH 7.2 were calculated as described pre-
Inc., Gaithersburg, MD, U.S.A. Chelex resin (50-100 viously (Denton et al., 1978; Midgley et al., 1987; Rutter
mesh) was from Sigma Chemical Co., Poole, Dorset, & Denton, 1988). Kinetic constants were calculated by
U.K., and ampholytes were from Pharmacia LKB Bio- non-linear regression as described by Rutter & Denton
technology, Milton Keynes, U.K. Chromatography (1988). Data are given as parameter value + S.E.M. for the
media were from either Pharmacia LKB Biotechnology number of degrees of freedom in parentheses. One unit
or Sigma. Acetyl-CoA carboxylase was prepared as of activity is defined as the amount catalysing the
described by Borthwick et al. (1987), and the pig heart conversion of 1,umol of substrate/min at 30 'C.
pyruvate dehydrogenase complex as described by Cooper Protein was measured as described by Bradford (1976),
et al. (1974). with bovine serum albumin as standard.
Methods
Fast protein liquid chromatography (f.p.l.c). Automated RESULTS AND DISCUSSION
f.p.l.c. systems and columns as supplied by Pharmacia Purification of pig heart NAD-ICDH
LKB Biotechnology were used and maintained according
to the manufacturers' instructions. Elution of proteins All procedures were carried out at 0-4 'C.
was continuously monitored at 280 nm. All buffers were (1) Preparation of mitochondrial extracts. These were
passed through a 0.2 ,um-pore-size filter before use. prepared from 6-12 pig hearts essentially by the method
of Cooper et al. (1974). Hearts were obtained fresh from
Analytical gel chromatography. Native-molecular-mass the carcass and immediately packed on ice. The muscle
determinations were made at 4 °C on an analytical was cut into 1-2 cm cubes and dispersed with a Waring
Superose 6 column (20 ml bed volume) equilibrated blender into 30 mM-KH2PO4 (pH 7.6)/250 mM-sucrose/
in 50 mM-Mops/K' (pH 7.2)/0.2 M-KCI/ 1 mM-EGTA/ 1 mM-EDTA (400 ml/heart). Care was taken to ensure
1 mM-HEDTA/0. I mM-dithiothreitol/0.02 °0 NaN3, plus that the pH did not fall below 6.5, and adjustment was
Ca2" as indicated. If necessary, samples were concen- made with 10 M-KOH if necessary. The homogenate was
trated by (NH4)2S04 precipitation, and were trans- centrifuged at 2075 g for 25 min in a Mistral 6L centri-
ferred into the chromatography buffer by centrifugation fuge. The supernatant was filtered through muslin and
(2 min, 1000 g) through a 2 ml column of Sephadex G-25 retained; the pellet was re-dispersed with the blender
(fine grade), previously equilibrated in chromatography (300 ml of dispersal buffer/heart) and re-centrifuged.
buffer (McCarthy & Hardie, 1982). The combined supernatants were adjusted to pH 5.4
with 400 (v/v) acetic acid (over 10 min with continuous
SDS/polyacrylamide-gel electrophoresis and isoelectric- stirring). Precipitated mitochondria were separated by
focusing gels. SDS/polyacrylamide-gel electrophoresis centrifuging at 15 000 g for 20 min, washed once by
(Laemmli, 1970), with gels containing 100 (w/v) acryl- resuspension in water and collected by centrifugation.
amide, was performed as described by Belsham et al. The mitochondrial pellet was resuspended in 20 mm-
(1980), except that a Bio-Rad mini-gel system (Bio-Rad KH2PO4, pH 7.2, containing 1 mM-EDTA, 0.1 mM-ADP,
Laboratories, Watford, Herts., U.K.) was used. Slab 2 mM-benzamidine, 0.1 mM-phenylmethanesulphonyl
isoelectric-focusing gels, containing 9.3 M-urea and the fluoride (PMSF) and 1 mM-dithiothreitol (40 ml/heart).
ampholytes indicated, were prepared and run as described The suspension was shell-frozen and thawed three times,
by Wong et al. (1982). alternating between liquid N2 and a water bath (30-
A Joyce-Loebl Chromoscan was used for densito- 40 °C). The slurry was centrifuged for 2 h at 18 000 g and
metric scanning of Coomassie-Blue-stained gels at the supernatant retained.
700 nm. Areas under each peak were determined after
transmission of data to a Hewlett-Packard 9000/300 (2) Preparation of a 35-65 %-satd.-(NH4)2SO4 fraction.
computer (Brownsey et al., 1984). After the gradual addition of solid (NH4)2SO4 to 3500
saturation, maintaining a constant pH of 7.2, the extract
Assay of NAD-ICDH activity, measurement of protein was stirred for 20 min and then centrifuged (20000 g,
and handling of kinetic data. During the isolation pro- 15 min). The supernatant was adjusted to 650% satur-
cedure NAD-ICDH activity was measured by moni- ation with (NH4)2SO4, stirred and centrifuged as des-
toring the production of NADH at 340 nm and 30 °C in cribed above. The resulting pellet was dissolved in 5 mM-
50 mM-Mops/35.5 mM-triethanolamine (pH 7.2)/2 mM- KH2PO4, pH 7.1, containing 1 mM-ADP, 0.1 mM-EDTA,
NADI/ 1 mM-ADP/2 mM-MgCl2/5 mM-DL-isocitrate. 500 (w/v) glycerol, 2 mM-benzamidine, 0.1 mM-PMSF,
Buffers used for kinetic studies were as indicated. In all 0.1 mM-dithiothreitol and 0.020% NaN3 (buffer A) to
cases, sodium DL-isocitrate, containing 5000 threo-Ds- give 50-100 mg of protein/ml (equivalent to 1-4 ml/
isocitrate (D -IC) was used. Assays, with a Pye-Unicam original pig heart).
PU-8800 spectrophotometer, were carried out in a total
volume of 1 ml, and NAD-ICDH (3-10 munits; see (3) Gel filtration on Superose 6. After centrifuging at
below) was added to initiate reactions. Rates of NADH 180000 g for 20 min, the redissolved (NH4)2SO4 pellet
production were linear for at least 2 min. (10-20 ml) was loaded on a column (2.6 cm x 70 cm) of
1989
NAD+-isocitrate dehydrogenase 447

1 (2',3',4') N 2 34
A 1 I 11
5.0r

4.0p 1 AU.V
n _

A .
-
I I
0
1-1 3.0 F 1, It 12.0 .cC
1-

2.0 F 8.0 >

I
1.0 F 4.0 Z
1:

0. L ; A
Iv Z
50 60 70 80
Volume after injection (ml)
Fig. 1. Separation of protein by gel filtration on Superose 6
Redissolved pellets (3.5 ml) from step 2 were chromatographed in the presence of 0.8 M-(NH4)2S04 on Superose 6
(1.6 cm x 60 cm column). Marked are the positions of elution of: 1, acetyl-CoA carboxylase (polymeric form; 4 x 107 Da;
Borthwick et al., 1987); 2, ferritin (460 kDa); 3, fl-galactosidase (540 kDa); 4, acetyl-CoA carboxylase (dimeric form; 460 kDa).
The position marked 2',3',4' indicates the corresponding region of elution of the above markers when the column was
equilibrated in 20 mM-Mops/K+, pH 7.2, containing 10 mM-sodium citrate, 10 mM-MgCI2, 2 mM-dithiothreitol and 5 % glycerol.
'N' indicates the position of elution of NAD-ICDH when the column was equilibrated in buffer B (see the text).

Superose 6 (prep grade) equilibrated in buffer A plus threitol and 0.0020% NaN3 (buffer B), and running at
0.8 M-(NH4)2SO4, and running at 90ml/h. Alternatively, approx. 10 ml/min.
3-4 ml of redissolved pellet was loaded on a 1.6 cm x
60 cm column of Superose 6, running at 30 ml/h. Under (4) Anion-exchange chromatography on Mono Q. De-
these conditions NAD-ICDH activity migrated as a salted fractions were loaded on a 10 ml (HR 10/10) (or
single band between two other well-separated protein 1 ml; HR 5/5) column of Mono Q, equilibrated in buffer
bands, which together contained more than 95 % of the B and running at 60 ml/h (30 ml/h for the 1 ml column).
loaded protein (Fig. 1). The column was developed with a gradient of buffer B
Separation of proteins is achieved at this high ionic plus 0-250 mM-KCI, and NAD-ICDH activity was
strength not merely on the basis of molecular mass. eluted isocratically at about 150 mM-KCI (Fig. 2). Peak
Hence, when fractions from step 2 were run on the fractions were pooled, giving 1-3 mg of NAD-ICDH/ml,
column in the absence of added (NH4)2SO4, most of the supplemented with 1 ,ug each of pepstatin, antipain and
protein was eluted as a single rather broad and complex leupeptin/ml, and stored in small samples at -70 'C.
band (results not shown). The effect of (NH4)2SO4 on the The enzyme was stable under these conditions for at least
column profile would seem to be the result of enhanced 6 months.
hydrophobic interactions with the column matrix. Thus The above procedure gives approx. 300-fold purific-
,-galactosidase, ferritin and the dimeric form of acetyl- ation of NAD-ICDH from mitochondrial extracts within
CoA carboxylase (Borthwick et al., 1987), which each hours. A typical example is detailed in Table 1, and an
have molecular masses close to 5 x 105 and closely co- SDS/polyacrylamide gel of fractions obtained at each
migrated on the column in lower-ionic-strength buffer, stage of the purification is shown in Fig. 3.
migrated at different rates in the (NH4)2SO4-containing The specific activity of NAD-ICDH prepared by this
buffer (Fig. 1). Moreover, the rate of migration of each method was in the range 25-35 units/mg, based on an
of the proteins was lower in the higher-ionic-strength assay at 30 'C and in the presence of Mg2" (see the
buffer. Experimental section). This range is identical with that
In contrast with the proteins mentioned above, the reported for the purified pig heart enzyme (Ehrlich &
rate of migration of NAD-ICDH activity was higher in Colman, 1981) and for the purified bovine heart enzyme
buffers containing 0.8 M-(NH4)2SO4 than in low-ionic- (Plaut, 1969; Giorgio et al., 1970).
strength buffer (Fig. 1). A likely explanation of this is
that the protein undergoes dimerization under these Subunit composition and native molecular mass
conditions, as observed by Giorgio et al. (1970). On SDS/polyacrylamide-gel-electrophoretic analysis
Peak fractions of NAD-ICDH activity obtained after of purified NAD-ICDH (see Fig. 3), two or three closely
Superose 6 chromatography were pooled and rapidly spaced bands with molecular masses of approx. 38 kDa
(< 15 min) de-salted on a 5 cm x 17 cm column of Seph- were apparent, as previously observed for the pig heart
adex G-25 (coarse grade) equilibrated in 20 mM-Bistris, enzyme by Ramachandran & Colman (1978).
pH 6.5, containing 0.1 mM-ADP, 20 mM-KCI, 5 % gly- Three bands, however, were clearly resolved by isoelec-
cerol, 2 mM-benzamidine, 0.1 mM-PMSF, 0.1 mM-dithio- tric focusing in the presence of 9.3 M-urea (Fig. 4) with pl
Vol. 263
448 G. A. Rutter and R. M. Denton

0.5 k 260
e

0.4 /'I 220

10.0
180
0.3 8.0 I.--,
: 140
6.0 .,4->. E
0.2 In
4.0 I 100 °
a
0.1 1 u
2.0 aD 60
z
\ 0.0 20
0 10 20 30 40 50
Volume after injection (ml)
Fig. 2. Anion exchange on Mono Q
Desalted fractions (30 ml, 15 mg of protein) after Superose 6 chromatography were loaded on the 1 ml (HR 5/5) Mono Q
column, and NAD-ICDH activity was eluted with the KCl gradient shown.

Table 1. Purification of pig heart NAD+-ICDH

Values are taken for a preparation from 10 hearts. Details of the assay of activity and protein are given in the Experimental
section.

NAD-ICDH Specific
Protein activity activity Recovery
(mg) (units) (units/mg) (%)
Mitochondrial extract 11050 885 0.08 100
35-65 %-satd.-(NH4)2SO4 fraction 3100 380 0.12 43
Superose 6 chromatography 32 261 8.3 30
Mono Q 8 210 26.2 24

values of 6.1, 6.3 and 7.05. These represented 32 0, 25 % apparent value of 329 + 13 kDa (mean + S.E.M for three
and 17 % respectively of the Coomassie-Blue-stained observations), which was essentially unchanged by the
protein present on the gel. A further band (pl 6.8) was
-
presence of 100 /tM-Ca2+.
also apparent, representing about 10 % of the stained This value is comparable with those obtained by
protein. Giorgio et al. (1970) for the bovine heart enzyme and by
The resolution of three sharp bands in the present Cohen & Colman (1971) for the pig heart enzyme under
work is in contrast with earlier attempts to separate the similar conditions, and would appear to be most con-
subunits of pig heart NAD-ICDH by isoelectric focusing sistent with an octameric subunit composition (a2/Jy)2.
(Ramachandran & Colman, 1980). In those earlier stud- However, Ehrlich et al. (1981) have suggested that this
ies, three groups of three to six bands were apparent, apparent molecular mass obtained by gel filtration repre-
with estimated pl values of 5.7, 6.6 and 7.2, representing sents an overestimate of the true value, as a result of the
51 %, 280% and 21 % of the Coomassie-Blue-stained anomalously high Stokes radius of NAD-ICDH. Thus,
protein. These bands were taken to represent three through equilibrium ultracentrifugation and light-
subunits of NAD-ICDH, termed a, , and y, with a scattering studies, those authors obtained a value for the
probable stoichiometry of a2,fiy. However, the current native molecular mass of the enzyme of around 224 kDa,
data suggest that a stoichiometry of a2,/2y is also possible. and suggested that a rapid equilibrium exists between
The native molecular mass of the purified enzyme was tetramaric, 2lxy (160 kDa), and octameric, (a2fi)2
estimated by gel filtration on Superose 6 at an ionic (320 kDa), forms. An alternative explanation for this
strength where the migration rate was related directly to apparent native molecular mass value ofclose to 200 kDa
this parameter for a number of standards. This gave an is that the stoichiometry of the individual subunits of the
1989
NAD+-isocitrate dehydrogenase 449

A B C D enzyme (each with molecular masses close to 40 kDa) is

(kDa) ...-..
a2,f2y, rather than a2,fy. It is evident that further studies
are required to establish firmly the subunit stoichiometry.
Regulation of activity by Ca2": effects of adenine
116 _ nucleotides, Mg2+, Mn2+ and KCI
The sensitivity of the enzyme to Ca2+ ions was first
determined under conditions similar to those used to
66-* study rat heart NAD-ICDH in permeabilized mito-
chondria and extracts (Rutter & Denton, 1988), namely
55 -p. at close to physiological ionic strength and in the presence
of 1 mM-Mg2+. Kinetic parameters are given in Table 2
and are similar to the values obtained with the rat heart
48 -p ::: ....... ....
enzyme. In particular, Ca2+ lowered apparent Km values
39 --
for added DS-IC in the presence of either ADP or ATP.
As found in earlier studies using mitochondrial extracts
(Denton et al., 1987), no effect of Ca2+ was seen in the
absence of an adenine nucleotide (results not shown).
25.6 * The apparent Km values for D8-IC were lower in the
presence of ADP than of ATP in both the presence and
the absence of Ca2+, but the effect of the change of
adenine nucleotide was more marked in the presence of
Ca2+. Finally, the concentrations of Ca2+ required for
half-maximal effects (KO.5 values) were 3-6-fold higher in
the presence of ATP than of ADP.
KCI inhibited the enzyme uncompetitively with respect
Fig. 3. SDS/polyacrylamide-gel-electrophoretic analysis of to Ds-IC under a wide variety of conditions (Table 2).
fractions during the purification of NAD-ICDH Furthermore, 75 mM-KCl also increased Ko.5 values for
Electrophoresis was performed as described in the Ca2' approx. 2-fold in the presence of either ADP or
Experimental section. Samples were obtained after: A, ATP.
extraction of mitochondria (step 1); B, (NH4)2SO4 The effect of [Mg2+] on the sensitivity of NAD-ICDH
fractionation (step 2); C, Superose 6 chromatography activity to Ca2" ions was investigated in the experiments
(step 3); D, anion exchange on Mono Q (step 4). shown in Figs. 5 and 6. Previous studies (Cohen &

8.0
.
0
.
0

* 0

0.4 _ 0 7.0
0
-
0

0
CL
0

6.0
0.2 _

5.0
w 0~S
I I I I I I~~~~~~~
0.0
0 20 40 60 80 100 120

Distance from origin (mm)

Fig. 4. Isoelectric focusing of NAD-ICDH

Subunits were resolved in a gel containing 9.3 M-urea and 60% (w/v) acrylamide, with 300 (v/v) ampholytes pH 5-8 and
pH 3.5-10 in the ratio 4:1.

Vol. 263
450 G. A. Rutter and R. M. Denton

Table 2. Effects of ADP and ATP on the regulation of NAD-ICDH by Ca2+ ions
Measurement of enzyme activity and calculation of kinetic constants were as described in the Experimental section. The assay
buffer was 50 mM-Mops/triethanolamine, pH 7.2, with 2 mM-NAD+, 1 mm-EGTA, I mM-HEDTA, plus other additions as
indicated. [Mg2+] was 1.0 mm in all assays. When added, [KCI] was 75 mm. KO.5 values for Ca2+ were determined in the presence
of 0.6 mM-D -IC (in the presence of ADP) or 1.0 mM-DA-IC (in the presence of ATP).

1.5 mM-ADP 1.5 mM-ATP

Parameter and condition No KC1 + KCI No KCI + KCI

<1 nM-Ca2+
Km for D.-IC (/#M) 751 +28 (9) 643 +25 (9) 1196+23 (10) 983 +93 (9)
h 2.5 3.1 2.7 3.3
VM... (munits/mg) 34.2+0.7 22.7+0.5 24.6 +0.3 19.2+0.5
100 gUM-Ca2+
Km for D,-IC (,UM) 261 +13 (10) 155+8.0 (11) 710+27 (9) 499 +27 (9)
h 2.3 2.9 2.2 2.3
VMax. (munits/mg) 32.5 +0.6 26.1 +0.5 32.0 + 0.6 23.6 +0.5
Ko 5 for Ca2l (/LM) 4.7+0.3 (12) 10.8 + 1.5 (13) 24.8 +4.9 (12) 69.9+11.0(12)

100 Colman, 1974; Plaut et al., 1974; Wilson & Tipton, 1981)
have suggested that the true substrate of mammalian
(a) NAD-ICDH is the Mg2" chelate of DS-IC (Mg-Ds-IC),
75 whereas free Mg2" ions act as competitive inhibitors
against this substrate. The effects of Ca2" on the Km value
for Mg-D5-IC were therefore first examined over a wide
50 _ range of Mg2" concentrations (Fig. 5, Table 3). Consistent
with the above model, increasing [Mg2"] led to an increase
in the Km of the enzyme for Mg-D.-IC, in both the
presence (100 /UM-1 mM) and the absence (< 1 nM) of
(-)
25 _ Ca2", with essentially no effect on Va. values up to
6.0 mM-Mg2". However, the relationship between [Mg2"]
E
0 and the Km values for Mg-D1-IC was complex, precluding
0O an estimate of a K. value for Mg2+. Furthermore, at a
l -~
I x

-2.0 -1.0 0
higher concentration (20 mM), and in the presence of
-
1.0 Ca2+ ions, Mg2' also caused an apparent fall in Vma.X, to
0o 70 % of the value apparent at lower concentrations of the
I ion.
100 At each [Mg2+], Ca2+ caused a fall in the Km value for
Mg-D -IC of 2-7-fold, with the effect of Ca2+ increasing
z (b) at higher [Mg2+]. However, as shown in Fig. 6 and in
75 -
Table 3, increasing [Mg2+] caused a marked rise in K0.5
values for Ca2+, from 1.4,UM at 20#,M-Mg2+ to about
1 mm at 20 mM-Mg2+. Again, the relationship between
these parameters was complex, and could not be de-
50 _ scribed through a simple Ki value for Mg2+.
The above studies were performed with the chelators
EGTA and HEDTA to allow the precise control of the
25 _ free concentrations of Ca2' and Mg2+ ions. However,
these and other nitrogen-containing polycarboxylate
Ca2+ chelators have been reported to have a direct
0 _
inhibitory effect on NAD-ICDH (Gabriel & Plaut, 1985).
Although in our hands the effects are small (Denton
-2.0 -1.0 0 1.0 et al., 1978; G. A. Rutter, unpublished work), it seemed
10log{[Mg-Ds-lc] (mM))}
Fig. 5. Effect of IMg2+1 on the sensitivity of NAD-ICDH
to Mg-DS-IC concentration at (a) < 1 nM-Ca2+ and ions (see the Experimental section). Concentrations (mM)
(b) 100 /sM- (, A, *, 0) or 1.0 mM- (A) Ca2+ of Mg2+ were 0.02 (-), 0.2 (A), 2.0 (E1), 6.0 (0) and 20.0
Activity was measured in 50 mM-Mops/Tris, pH 7.2, with (-). The continuous lines are those obtained by fitting the
2 mM-NAD+, 1.5 mM-ADP, 1 mM-EGTA, 1 mM-HEDTA,
data by non-linear least squares regression analysis to the
supplemented with DL-isocitrate, MgCl2 and CaCl2 to give
foHlowing equation:
the required concentrations of Mg-D.-IC and free metal v= VMaax /{1 + (KMM gD IC/[Mg-DS.IC])h}
1989
NAD+-isocitrate dehydrogenase 451

Table 3. Effect of IMg2+j on the regulation of NAD-ICDH- by Ca2+ ions

Details of the assay and calculation of kinetic constants are given in the Experimental section. The buffer was 50 mM-Mops/Tris,
pH 7.2, with 2 mM-NAD+, I mM-EGTA, 1 mM-HEDTA, 1.5 mM-ADP, plus DL-isocitrate, MgCl2 and CaCl2 to give the required
[Mg-DS-IC], [Mg2+] and [Ca2+]. KO.5 values for Ca2+ were determined in the presence of 20.0, 50.0, 22.0, 100.0 and 500.0 ,UM-
Mg-DS-IC at 0.02, 0.2, 2.0, 6.0 and 20.0 mM-Mg2+ respectively. Abbreviation: N.D., not determined.
< I nM-Ca2+ 100 /uM-Ca2+
[Mg2+] (mM) K. Mg-Ds-Ic (1uM) h Km Mg-Ds-Ic (,LM) h Ko.5 for Ca2+ (4UM) h

0.02 107+7.7 (6) 1.0+0.1 61.0+ 5.9 (5) 0.9+0.1 1.24+0.21 (5) 1.8 +0.5
0.20 204+ 14.3 (7) 1.7 +0.2 70.6+ 5.8 (6) 1.3 +0.1 5.77+ 1.20 (10) 1.0+0.2
2.00 360+27.1 (6) 1.8 +0.2 80.0+ 5.0 (6) 1.7 +0.1 14.7+2.10 (6) 1.1 +0.2
6.00 537+ 52.0 (6) 1.6+0.2 181 + 16.0 (6) 1.7 + 0.3 232.0 +48.0 (7) 1.4+0.3
20.0 3300 +243.0 (6) 1.0+0.1 434+44.0* (6) 2.6+0.6 > 1000 N.D.
* Calculated assuming mixed (competitive plus non-competitive) inhibition (see the text); [Ca2"] = 1.0 mM.

100 a K0 5 value for Ca2l of 3.3 +0.7 /tM (9); this value was
increased to 8.9+3.6 /M (4) at 0.1 mM-Mn2+.
These observations confirm that the enzyme is fully
75 F
sensitive to Ca2` in the absence of chelators, and in the
.P E
cc0 presence of Mn2+ as well as Mg2` ions. Furthermore,
0 x
increasing the concentration of Mg2` ions raised Ko05
c X
50 values for Ca2` under these conditions, consistent with
o E
the results in the presence of chelators, and this effect was
c
e O_
_
25
also apparent with Mn2" ions. Finally, since activation of
E

cn
the enzyme by Ca2+ occurred in the presence of up to
100 /tM-Mn2+, this suggests that Mn2+ ions are unable to
0
replace Ca2` as activators of the enzyme.
- General discussion
-7.0 -6.0 -5.0 -4.0 -3.0

log{[Ca2+] (M)} The method presented here for purifying pig heart
NAD-ICDH represents a marked improvement upon
Fig. 6. Effect of [Mg2+1 on the sensitivity of NAD-ICDH to Ca2+ previously published methods (Plaut, 1969; Giorgio
ions et al., 1970; Ehrlich et al., 1981). By this method it is
Details of the assays and concentrations of Mg2+ are as possible to obtain NAD-ICDH of equivalent purity, and
given in the legend to Fig. 5. In each case the stimulation with a similar or better yield, but in a matter of hours
of activity by Ca2+ was 5-10-fold. The continuous lines are rather than days. This improvement is due largely to the
those obtained by fitting the data by non-linear least- use of Superose 6 chromatography at high ionic strength
squares regression analysis to the following equation: as a means of achieving 50-100-fold purification of the
enzyme in a single step.
V-Vmin = Vmax /{1 + (KO.5/[Ca2])h} In the presence of a fixed concentration of Mg2+ ions
where V'min and V'max represent the activity of the enzyme (1 mM), the kinetic properties of the enzyme with respect
at zero and saturating [Ca2+] respectively. to Ca2+, ADP and ATP are very similar to those described
previously for rat heart NAD-ICDH (Denton et al.,
1978; Rutter & Denton, 1988). These studies confirm
that with the isolated enzyme K0 5 values for Ca2+, as well
important to demonstrate that the above effects of Ca2", as apparent Km values for (total) D.-IC, are higher in the
and the interaction between regulation by Ca2" and Mg2" presence of ATP than of ADP (Gabriel et al., 1985;
ions, were apparent in the absence of EGTA and Rutter & Denton, 1988).
HEDTA. Furthermore, the absence of these chelators In contrast with the other Ca2"-sensitive citrate-cycle
was also necessary to investigate the effects of Mn21 on enzyme, 2-oxoglutarate dehydrogenase, where Mg2` (up
the sensitivity of NAD-ICDH to Ca2+. This is because to 1 mM) appears to have essentially no effect on Ko 5
Mn2+ ions bind at least as strongly to the chelators as do values for Ca2+ (McCormack & Denton, 1979; G. A.
Ca21 ions (Denton et al., 1978). Rutter & R. M. Denton, unpublished work), the sen-
These studies were carried out at pH 7.2 in the presence sitivity of NAD-ICDH to Ca2+ is shown to be critically
of 1.5 mM-ADP and a total Ds-IC concentration of dependent on the concentration of Mg2+ (Fig. 6). Al-
0.3 mM. In the presence of added MgCl2 to give 10 gM- though studied over a narrower concentration range,
Mg2+, CaCl2 stimulated activity 3.5-fold, with a Ko5 value Mn2+ ions also appeared to decrease the sensitivity of the
for Ca2+ of 0.66 + 0.13 /tM (6); this value was raised to enzyme to Ca2+.
27.0 + 3.6 /tM (3) in the presence of 0.44 mM-Mg2+. Simi- The effects of Mg2+ (and Mn21) would seem to be best
larly, with added MnCl2 to give 2.6 /iM-Mn2+, addition of explained by the competition of these ions for an
CaCl2 stimulated the activity of the enzyme 5-fold, with activatory Ca2+-binding site on the enzyme. However,
Vol. 263
 452 G. A. Rutter and R. M. Denton

the complex relationship between [Mg2"] and KO.5 values Denton, R. M., Richards, D. A. & Chin, J. G. (1978) Biochem.
for Ca2" may suggest that the binding of Mg2" to another J. 176, 899-906
site, possibly the active site of the enzyme, is also involved. Denton, R. M., McCormack, J. G. & Edgell, N. E. (1980)
In the present studies, the observed KO.5 values for Biochem. J. 190, 107-117
Ca2" of NAD-ICDH in the presence of 1.5 mM-ADP fell Denton, R. M., McCormack, J. G., Midgely, P. J. W. & Rutter,
to values approaching those for 2-oxoglutarate dehydro- G. A. (1987) Biochem. Soc. Symp. 54, 127-143
genase (0.2-2.0 /M; McCormack & Denton, 1979; Den- Ehrlich, R. S. & Colman, R. F. (1981) J. Biol. Chem. 256,
ton et al., 1980; Lawlis & Roche, 1980; Rutter & Denton, 1276-1282
1988) only at concentrations of Mg2" (20 fM) well below Ehrlich, R. S., Hayman, S., Ramachandran, N. & Colman,
R. F. (1981) J. Biol. Chem 256, 10560-10564
those considered to occur in intact mitochondria (about Gabriel, J. L. & Plaut, G. W. E. (1985) Biochem. J. 229,817-822
0.3 mM; Jung & Brierly, 1986; Corkey et al., 1986). Gabriel, J. L., Milner, R. & Plaut, G. W. E. (1985) Arch.
However, the responses to Ca2" of two further mito- Biochem. Biophys. 240, 128-134
chondrial Ca2"-sensitive enzymes, pyruvate dehydrogen- Gabriel, J. L., Zervos, P. R. & Plaut, G. W. E. (1986) Metab.
ase phosphate phosphatase (which catalyses the de- Clin. Exp. 35, 661-667
phosphorylation and consequent activation of the pyru- Giorgio, N. A., Jr., Yip, A. T., Fleming, J. & Plaut, G. W. E.
vate dehydrogenase complex; Midgley et al., 1987; (1970) J. Biol. Chem. 254, 5469-5477
Rutter et al., 1989) and mitochondrial pyrophosphatase Goebell, H. & Klingenberg, M. (1964) Biochem. Z. 340,441-464
(Davidson & Halestrap, 1989) are also diminished with Hansford, R. G. (1985) Rev. Physiol. Biochem. Pharmacol.
increasing concentrations of Mg2". It is therefore evident 102, 1-72
that the intramitochondrial Mg2" concentration may Jung, D. W. & Brierly, G. P. (1986) J. Biol. Chem. 261,
have an important influence on the relative Ca2+-sensi- 6408-6414
tivities of the Ca2"-regulated enzymes within mito- Laemmli, U. K. (1970) Nature (London) 227, 680-685
chondria. Lawlis, V. B. & Roche, T. E. (1980) Mol. Cell. Biochem. 32,
147-152
McCarthy, A. D. & Hardie, D. G. (1982) FEBS Lett. 150,
These studies were supported by grants from the Medical 181-184
Research Council, U.K., and the British Diabetic Association. McCormack, J. G. & Denton, R. M. (1979) Biochem. J. 180,
We thank Dr. A. P. Halestrap for useful discussion, and 533-544
Dr. A. C. Borthwick and Mr. N. T. Redpath for assistance McCormack, J. G., Halestrap, A. P. &h Denton, R. M. (1989)
with f.p.l.c. and isoelectric-focusing experiments respectively. Physiol. Rev., in the press
Midgley, P. J. W., Rutter, G. A., Thomas, A. P. & Denton,
R. M. (1987) Biochem. J. 241, 371-377
REFERENCES Plaut, G. W. E. (1969) Methods Enzymol. 13, 34-42
Plaut, G. W. E. (1970) Curr. Top. Cell. Regul. 2, 1-27
Aogaichi, T., Evans, J., Gabriel, J. L. & Plaut, G. W. E. (1980) Plaut, G. W. E. & Sung, S.-C. (1954) J. Biol. Chem. 207,
Arch. Biochem. Biophys. 204, 350-360 305-314
Belsham, G. J., Denton, R. M. &Tanner, M. J. (1980) Biochem. Plaut, G. W. E. & Aogaichi, T. (1968) J. Biol. Chem 243,
J. 192, 457-467 5572-5583
Borthwick, A. C., Edgell, N. J. & Denton, R. M. (1987) Bio- Plaut, G. W. E. & Gabriel, J. L. (1983) in Biochemistry of
chem. J. 241, 773-782 Metabolic Processes (Lennon, D. L. F., Stratman, F. W. &
Bradford, M. M. (1976) Anal. Biochem. 72, 248-254 Zahlten, R. N., eds.), pp. 285-301, Elsevier, Amsterdam
Brownsey, R. W., Edgell, N. J., Hopkirk, T. J. & Denton, Plaut, G. W. E., Schramm, V. L. & Aogaichi, T. (1974) J. Biol.
R. M. (1984) Biochem. J. 218, 733-743 Chem. 249, 1848-1856
Chen, R. F. & Plaut, G. W. E. (1963) Biochemistry 2, Ramachandran, N. & Colman, R. F. (1978) Proc. Natl. Acad.
1023-1032 Sci. U.S.A. 75, 252-255
Cohen, P. F. & Colman, R. F. (1971) Biochim. Biophys. Acta Ramachandran, N. & Colman, R. F. (1980) J. Biol. Chem. 255,
242, 325-330 8859-8864
Cohen, P. F. & Colman, R. F. (1974) Eur. J. Biochem. 47, Rutter, G. A. & Denton, R. M. (1988) Biochem. J. 252, 181-189
35-45 Rutter, G. A. & Denton, R. M. (1989) Biochem. J. 263,453-462
Cooper, R. H., Denton, R. M. & Randle, P. J. (1974) Biochem. Rutter, G. A., Midgley, P. J. W. & Denton, R. M. (1989)
J. 143, 625-641 Biochim. Biophys. Acta, in the press
Corkey, B. A., Duszynski, J., Rich, T. L., Matchinsky, B. & Shen, W.-C., Manck, L. & Colman, R. F. (1974) J. Biol. Chem.
Williamson, J. R. (1986) J. Biol. Chem. 261, 2567-2574 249, 7942-7949
Davidson, A. M. & Halestrap, A. P. (1989) Biochem. J. 258, Williamson, J. R. & Cooper, R. H. (1980) FEBS Lett. 117,
817-821 K73-K85
Denton, R. M. & McCormack, J. G. (1985) Am. J. Physiol. Wilson, V. J. C. & Tipton, K. F. (1981) Eur. J. Biochem. 113,
249, E543-E554 477-483
Denton, R. M., Randle, P. J. & Martin, B. R. (1972) Biochem. Wong, S. T., Mastopaolo, W. & Henshaw, E. C. (1982) J. Biol.
J. 128, 161-163 Chem. 257, 5231-5237

Received 2 May 1989/12 June 1989; accepted 26 June 1989

1989
View publication stats