Eur. J. Biochem. 146,459-466 (1985)
0FEBS 1985
Isolation and characterization of an iron-containing superoxide dismutase
from tomato leaves, Lycopersicon esculentum
Jan KWIATOWSKI, Aleksandra SAFIANOWSKA, and Zbigniew KANIUGA
Institute of Biochemistry, Warsaw University
(Received May 4/August 1, 1984) - EJB 84 0483
A cyanide-insensitive superoxide dismutase was purified from tomato leaves (Lycopersicon esculentum, Mill.,
var. Venture) to apparent homogeneity. The enzyme had twofold higher specific activity (about 4000 standard
units) than ferric superoxide dismutases purified from Brassica campestris [Salin, M. L. and Bridges. S. M. (1980)
Arch. Biochem. Biophys. 201, 369-3741 and Nuphar luteum [Salin, M. L. and Bridges, S. M. (1982) Plant Physioi
69, 161 -1651.
The protein had a relative molecular mass of about 42000 and was composed of two equal subunits noncovalently joined. It was negatively charged (PI = 4.6) and contained about 1.45 mol Fe/mol dimer and negligible
amounts of Mn, Cu and Zn. Absorption spectrum and sensitivity to NaN3, HzOz and temperature are also
reminiscent of other ferric superoxide dismutases.
Comparison of amino acid composition indicated, however, a closer relationship to the Mn-containing enzymes
rather than to other Fe-containing superoxide dismutases. Two possible ways of Fe-containing superoxide
dismutase acquisition by vascular plants were suggested.
Superoxide dismutase is an important enzyme for aerobic
organisms [l], although it has also been found in several
anaerobic ones 121. It disproportionates potentially harmful
superoxide anion radicals to dioxygen and hydrogen peroxide
[31.
Three types of this enzyme have been recognized with
respect to the prosthetic metal present in it. Copper-and-zinccontaining superoxide dismutase is localized mainly in the
cytosol of eucaryotes [4]. However, it has also been found in
mitochondria [5] and higher plant chloroplasts [6 - 81. The
enzyme represents an independent line of descent while two
others, manganese and iron-containing superoxide dismutases, show a close evolutionary relationship [4, 9, 101. The former protein is present commonly in bacteria and mitochondria [4], suggesting an endosymbiotic origin of this
organelle [111.
Ferric superoxide dismutase was previously isolated from
many procaryotic sources and was believed to be confined
to them [4]. However, data on its existence in plants have
accumulated. The enzyme was isolated from the eucaryotic
green alga Euglenagracilis [I21 and two higher plants Brassica
campestris 1131 and Nuphar luteum [14]. It was also reported
that Fe-containing enzyme was present in only three families
of vascular plants out of the 43 surveyed 1151. However, we
have recently provided the evidence that ferric superoxide
dismutase (Fe-SOD) is present in two chilling-sensitiveplants :
tomato and bean [16], members of the next two families.
This is paper 14 of the series Photosynthetic apparatus of chillingsensitive plants.
Abbreviations. Fe-SOD, Mn-SOD, Cu,Zn-SOD iron, manganese
and copper-and-zinc-containing superoxide dismutase; SDS, sodium
dodecyl sulfate.
Enzymes. Superoxide dismutase (EC 1.15.1.l); xanthine oxidase
(EC 1.2.3.2).
In this work we describe ferric superoxide dismutase from
tomato leaves, which, according to their properties, is quite
similar to other Fe-SOD enzymes, but, on the grounds of
amino acid composition, seems to be closer to those containing Mn. Two possible ways of Fe-SOD acquisition by vascular
plants are suggested.
MATERIALS AND METHODS
Plant material and chemicals
Fresh tomato leaves (Lycopersicon esculentum, Mill., var.
Venture) were harvested from a greenhouse. The sources of
chemicals were as follows : xanthine oxidase, Boehringer; bovine serum albumin, riboflavin and sodium dodecyl sulfate
(SDS), BDH Chemicals Ltd; Sephadex G-100, Sephacryl
S-200, DEAE-Sephadex A-50 and DEAE-Sephacel,
Pharmacia Fine Chemicals; polyvinylpyrrolidone, K ' 0 ;,
NaN3 and 2-mercaptoethanol, Fluka AG; ampholines, LKB;
Coomassie brilliant blue R-250, International Enzymes Ltd;
acrylamide, bisacrylamide, dimethylsulfoxide, glycerol and
Titrisols (Fe, Cu, Zn and Mn standard solutions), Merck;
phenylmethylsulfonyl fluoride, Sigma Chemical Company;
ammonium persulfate and dithiothreitol, Roth KG; NBT and
KCN, Chemapol (Czechoslovakia); tetramethylethylenediamine and molecular mass standards, Serva Feinbiochemica.
Other chemicals were from P.O.Ch. (Poland).
Electrophoretic procedures
SDS/polyacrylamide gel electrophoresis was performed by
the method of Weber and Osborn [17], using the following
molecular mass standards (MI): bovine serum albumin
(67000), egg albumin (45000), chymotrypsinogen (25000),
460
myoglobin (17800) and cytochrome c (12400). Electrophoresis of native protein in 7% and 12% gel was performed as
described by Davis [18]. Superoxide dismutase activity was
localized on gels by the method of Beauchamp and Fridovich
[19]. The isoelectric point was determined on acrylamide gel
containing ampholines (pH 3.5 - 10) using LKB 21 17
Multiphor apparatus. The corresponding pH values were
determined directly on the gel surface with a microelectrode.
Enzyme and protein assays
Superoxide dismutase activity, after each purification step
and when the purified enzyme was characterized, was measured by the method of McCord and Fridovich [3]. Activity was
calculated according to Asada et al. [20]. Column fractions in
the course of enzyme purification were assayed as described
by Henry et al. [21]. Protein content of crude fractions and
purified enzyme was determined according to Lowry et al.
[22] and Murphy and Kies I231 respectively using bovine serum
albumin as a standard for both methods.
Determination of molecular mass
Fraction number I15 ml 1
Fig. 1 . Separation of three superoxide dismutases of L. esculentum by
linear gradient DEAE-Sephadex A-50 chromatography. Leaf extract
was applied to a column following ammonium sulfate precipitation
and DEAE-Sephadex A-SO batch steps. Superoxide dismutase activity
is expressed as a percentage of inhibition, by a given aliquot of the
eluate, of the reaction of nitroblue tetrazolium (NBT) reduction by
K f O -[21]
from subsequent centrifugation (12000 x g, 1 h) was dissolved
and dialyzed for 40 h against 10 mM potassium phosphate,
pH 7.8, containing 5 mM 2-mercaptoethanol and 0.1 mM
EDTA. The dialysate was brought to 200 mM phosphate with
1 M potassium phosphate, pH 7.8, and DEAE-Sephadex A50 (20 m1/100 ml solution), pre-equilibrated with 200 mM
potassium phosphate containing 0.1 mM EDTA, was added.
After 0.5 h of stirring, the gel was centrifuged and washed
with the same buffer. Collected supernatants were dialyzed for
24 h against 10 mM potassium phosphate, pH 7.8, containing
0.1 mM EDTA and 0.3 mM dithiothreitol and then adsorbed
onto a DEAE-Sephadex A-50 column ( 5 x 35 cm) equilibrated
with
the same buffer. The column was washed with 50 mM
Amino acid composition and metal content
KCl in the above buffer until absorption at 280 nm decreased
Analysis was performed with a Jeol model JLC-6AH below 0.3. Then a linear gradient of KCl (50-250 mM, 3 1)
amino acid analyser. Protein samples, after exhaustive dialysis was applied (Fig. 1). Fractions with cyanide-insensitive
against water, were hydrolyzed for 24 h, 48 h and 72 h at superoxide dismutase activity were pooled, concentrated over
110°C under reduced pressure in 6 M HC1 containing an Amicon PM-10 membrane and dialyzed against 10 mM
2-mercaptoethanol (0.5 ml/l) to protect methionine and potassium phosphate, pH 7.8, containing 40 mM KC1,
tyrosine. Half-cystine was determined as cysteic acid, after 0.1 mM EDTA and 0.3 mM dithiothreitol. The dialysate was
oxidation with performic acid [25]. Tryptophan was deter- loaded to a column of DEAE-Sephacel(2 x 40 cm) equilibratmined spectrophotometrically by the method of Edelhoch ed with dialysis buffer. A linear gradient of KCl(40 - 150 mM,
1.5 1) in the above buffer was then applied. Fractions containP61.
Metals were determined by atomic absorption spectropho- ing superoxide dismutase activity were pooled and concentometry using an Instrumentation Laboratory model 551 trated and applied to a Sephadex G-100 column (2 x 140 cm),
video I apparatus.
which was equilibrated and eluted with 20 mM potassium
phosphate, pH 7.8, containing 100 mM KCI, 0.1 mM EDTA
and 0.3 mM dithiothreitol (Fig. 2A). Concentrated eluate
RESULTS
with superoxide dismutase activity was again subjected, in
three steps, to gel exclusion chromatography using a column
Purification of the enzyme
of Sephacryl S-200 (1 x 80 cm) equilibrated and eluted with
All operations were carried out at 0 - 2 "C. Buffers, except 50 mM potassium phosphate containing 0.5 M KC1 and
0.1 mM EDTA (Fig. 2B).
those for final dialysis, were protected with pHix (Pierce).
Table 1 summarises the results of the purification proceWashed, fresh leaves (about 8000 g) were thoroughly
ground in a blender containing: 100mM potassium phos- dure. Clarification of crude extract prior to ammonium sulfate
phate, pH 7.8, 5 mM 2-mercaptoethanol, 1 mM EDTA and salting-out did not improved the results of purification.
1 mM phenylmethylsulfonyl fluoride. 100 g polyvinyl- Although applying the DEAE-Sephadex A-50 batch did not
pyrrolidone was gradually added during homogenization. The raise the purity of the enzyme, it was essential for the good
resulting homogenate was squeezed through cheesecloth and development of subsequent ion-exchange chromatography. In
brought to 40% saturation with ammonium sulfate (0.242 g/ the DEAE-Sephadex A-50 step (Fig. 1) all cyanide-sensitive
ml). After 1 h of stirring, the precipitate was centrifuged superoxide dismutase activity was removed.
(I0000 x g, 20 min) and ammonium sulfate was added to the
The purified enzyme was homogeneous upon electrosupernatant to 90% saturation (0.365 g/ml). The precipitate phoresis in 7% and 12% gels (Fig. 3). It had the same electro-
The relative molecular mass was determined by gel exclusion chromatography using Sephadex G-100 and Sephacryl
S-200 gels. The molecular mass standards were the same as in
SDS/polyacrylamide gel electrophoresis.
Sedimentation equilibrium centrifugation was performed
in a MSE Centriscan 75 ultracentrifuge equipped with an
ultraviolet scanning system. The molecular mass of the
enzyme was estimated by the low-speed equilibrium method
according to a manufacturer. The partial specific volume was
calculated from the amino acid composition [24].
461
phoretic mobility as the enzyme in crude extract, indicating
that it was not modified during purification. Specific activity
was about 4000 U/mg, a value reported for some procaryotic
Fe-SOD enzymes [27, 281 but higher than those of other
vascular plants [13, 141.
Molecular properties
The relative molecular mass of the native enzyme was
determined by gel exclusion chromatography on Sephadex
G-100 (Fig. 2A) and Sephacryl S-200 (Fig. 2B and 4A) gel
columns. The value obtained from four determinations was
43 000 & 2 000. SDS/polyacrylamide gel electrophoresis, both
in the presence and absence of 2-mercaptoethanol, yielded
rl.0
I
a single band with a mobility corresponding to a relative
molecular mass of 22500 (Fig. 4B). These results indicate that
the enzyme is composed of two non-covalently joined subunits
of equal size.
Sedimentation equilibrium of the enzyme (0.5 mg/ml) was
carried out at 12000 rpm. The plot of lnAzsoversus the square
of the distance from the center of rotation was linear and
yielded a slope of 0.335 (Fig. 5). On the basis of a partial
specific volume of 0.744 cm3 g-', calculated from the amino
acid composition, the relative molecular mass was estimated
to be 41 000. For all further calculations a value of M , 42000
was established.
Absorption spectrum
Fig. 6 shows visible and ultraviolet spectra of purified
enzyme. It has an absorption maximum at 278 nm with slight
shoulders at 260 nm and 290 nm. A broad, weak absorption
in the range of 450-300 nm with a maximum at 350 nm is
3 ml 1
Froction number
0.8
0.6
0.5
C
0.4
20
Q
10
0
I
20
30 40
Fraction number
10
( 1.1 ml
I
Fig. 3. Polyacrylamide disc gel electrophoresis of the purified enzyme
Fig. 2. Gel exclusion chromatography of L. esculentum Fe-SOD; ( A )
Sephadex G-100 column, ( B ) Sephacryl S-200 Superfine column.
Superoxide dismutase activity is expressed as in Fig. 1
in 7% and 12% gels. 20 pg and 3 pg of the protein were stained with
Coomassie blue (right-hand gels) and nitroblue tetrazolium (left-hand
gels) for protein and superoxide dismutase activity respectively
Table 1. Purification of an iron-containing superoxide dismutase from leaves of Lycopersicon esculentum
Activity was determined according to McCord and Fridovich [3] in the presence of 1 mM KCN
Step
40-90% (NH4)ZSOd
DEAE-Sephadex A-50 batch
DEAE-Sephadex A-50,
KCI gradient
DEAE-Sephacel,
KCI gradient
Sephadex G-100
Sephacryl S-200, dialysed
Volume
Protein
concentration
Total
protein
Specific
activity
ml
mg/ml
mg
U/mg
580
640
473
11.4
6.9
0.61
6600
4730
284
220
0.29
63.6
0.98
20.9
9.3
21.4
11.4
1.22
Recovery
Purification
U
Y"
-fold
150
129
95.5
100
86
64
1
1.2
15
942
59.9
40
42
2040
4090
42.7
38.0
28
25
90
180
22.7
27.2
336
x
Total
activity
462
0.7
0.6
A
-
-
'
.
O
1
-1.51
0.5 -
0.4
-
0.3 Bovine serum albumin
1
0
3
Fig. 5. Sedimentation equilibrium of L. esculentum Fe-SOD. Purified
enzyme at a concentration of 0.5 mg/ml in 10 mM potassium
phosphate, pH 7.5, was equilibrated at a rotor speed of 12000 rpm
after initial 5-h centrifugation at 16000rpm at 20°C. The ultracentrifuge was equipped with an ultraviolet scanning system and
InAZsowas plotted as a function of the square of the distance from
the centre of rotation
1.4
-
- 0.07
1.2
-
- 0.06
1.0-
-a05
0.8-
-0.04
Z 0.6-
-0.03
Myoglobin
Chymotrypsinogen
Cytocrome c
i
0.l
0.2 0.3 O b 0.5 0.6 0.7 0.8 0.9 1.0
Re lot ive
mobiIity
Fig. 4. Determination of relative molecular mass of L. esculentum
Fe-SOD. ( A )Molecular exclusion chromatography on a column of
Sephacryl S-200 Superfine, (B) Sodium dodecyl sulfate/polyacrylamide gel electrophoresis
C
0
n
ul
n
a
also observed. The A;:,,, at 278 nm was 17.9, corresponding
to a molar absorption coefficient of 7.5 x lo4 M-' cm-l. This
spectrum is reminiscent of Fe-SOD enzymes isolated from
eucaryotic and procaryotic sources [12, 14,29, 301.
0.4
-
- a02
0.2
-
-0.01
I
200
1
300
400
500
Wavelength (nrn)
Metal analysis
The purified enzyme, after extensive dialysis versus 10 mM
potassium phosphate, pH 7.8, containing 0.1 mM EDTA, was
subjected to atomic absorption spectrophotometry. The value
obtained from three measurements of different preparations
for iron was 1.45 0.12 atoms per dimer. Copper and zinc
contents were 0.06 0.01 atom of each per dimer and
manganese was below the detection level (0.00 atom per
dimer) .
Isoelectric point
Isoelectric focusing demonstrated a single band at pH 4.6.
This value is similar to those reported for other eucaryotic
Fe-SOD enzymes [12 - 141.
Enzyme stability and sensitivity to inhibitors
Frozen or refrigerated, concentrated solutions of a purified enzyme are quite stable. We have even observed some
increase in specific activity during first days of storage. It
is probably the effect of conformational changes within a
protein.
Fig. 6. Absorption spectrum of L. esculentum Fe-SOD. The enzyme,
at a concentration of 680 pg/ml in 10 mM potassium phosphate,
pH 7.8 was examined
Fig. 7 shows the temperature stability of Lycopersicon FeSOD. Inactivation at 60°C and 70°C is apparently first order
with respect to enzyme concentration and half-lives are
calculated to be 22 min and 4 min respectively. For comparison, procaryotic Fe-SOD enzymes,, except those from
luminous bacteria [31], are more thermostable 132- 351 while
a Euglena enzyme is more labile [12].
1 mM cyanide, as expected, had no effect on activity. The
enzyme was 50% inhibited by 5 mM NaN3, the concentration
predicted for Fe-SOD [36]. A Lycopersicon enzyme was inactivated by H z 0 2 as shown in Fig. 8. Pseudo-first-order
kinetics were observed for H 2 0 2concentration up to 5 mM.
For 0.5 mM H 2 0 2the half-life was 15.5 min and the secondorder rate constant was calculated to be 1.5 M - ' s-'. Again
a Euglena enzyme is more sensitive to H 2 0 2while Nuphar and
procaryotic enzymes, except the very resistant Plactonema one
[30], are inactivated in the same range (Table 4 from [12]) [2,
141.
463
Table 2. Amino acid composition of higher plant ferric superoxide
dismutases
The values are averages of 24-h, 48-h and 72-h hydrolyses. Threonine,
serine and methionine were determined by extrapolation to zero hydrolysis time while valine and isoleucine were measured in 72-h
hydrolyzates. Half-cystine was determined as cysteic acid after
performic acid oxidation. Tryptophan was determined spectrophotometrically. The amino acid composition of the enzyme was
calculated assuming the relative molecular mass (M,) of 42000. The
numbers of residues as mol/mol subunit are given to the nearest
integer
~
~
Amino acid
~~
~~
~
~~
L. esculentum
5
40
20
60
80
~~
B. campestris
~ 3 1
~
~
N . luteum
~ 4 1
100
Time ( m i n )
Fig. I . Temperature stability of L. esculentum Fe-SOD. The purified
enzyme (700 pg/ml) in 50 mM potassium phosphate, pH 7.8, was
incubated at temperatures indicated. Small aliquots were withdrawn
and assayed for residual activity
Lysine
Histidine
Arginine
Aspartic acid
Threonine
Serine
Glutamic acid
Proline
Glycine
Alanine
Half-cystine
Valine
Methionine
Isoleucine
Leucine
Tyrosine
Phenylalanine
Tryptophan
mol/
100 mol
5.8
2.6
2.6
11.3
5.1
4.6
8.3
7.3
6.0
9.6
0.1
5.7
2.2
4.5
10.5
3.7
5.5
4.4
mol/mol subunit
10
4
4
19
9
8
14
12
10
16
0
10
4
8
18
6
9
7
11
4
5
18
9
11
16
10
16
17
2
11
1
6
14
6
7
4
12
5
6
23
6
15
19
9
22
20
2
5
5
4
16
6
9
7
I
5
10
15
25
20
Time
30
(min)
Fig. 8. Inactivation of L. esculentum Fe-SOD by Hz02. The enzyme
(1.67 pM) was incubated at 25°C in 50 mM potassium phosphate,
pH 7.8, containing 0.1 mM EDTA and the indicated concentration
of H z 0 2 .Small aliquots were taken and assayed for remaining activity
Amino acid composition and relatedness
to other superoxide dismutase enzymes
The amino acid composition of Lycopersicon and two
other vascular plant Fe-SOD enzymes is given in Table 2. The
tomato enzyme has the same high content of aspartic acid,
leucine and alanine and low content of cysteine and methionine as other Fe-SOD and Mn-SOD enzymes. However, it
contains distinctly more proline and less glycine than the
average.
To estimate the relatedness of tomato ferric superoxide
dismutase to other Fe and Mn-containing superoxide dismutases a statistical analysis of the chosen enzymes was performed by the method of Marchalonis and Weltman [37]. The
lowest SdQ values for tomato Fe-SOD, up to 50 (Table 3,
column a), were obtained mostly with Mn-containing
superoxide dismutases from organisms of different levels of
evolution. Especially indicative seem to be the relatively low
values for yeast and rat Mn-SOD enzymes. On the other hand,
ferric superoxide dismutases from many different organisms
gave greater values (from 51 to 103). An interesting example
is provided by Escherichia coli, possessing both Fe-SOD and
Mn-SOD, whose Mn-containing enzyme seems to be more
closely related (SAQ = 40) to tomato superoxide dismutase
than the ferric one (SdQ = 50). Noteworthy also are the high
values (82 - 84) for two eucaryotic plant Fe-SOD enzymes,
from Euglena [12] and Nuphar [14]. Such results indicate that
tomato Fe-SOD is more closely related to some Mn-containing enzymes than to other ferric ones. Another Fe-SOD from a
higher plant, Brassica campestris, closely related to the tomato
enzyme, has already been located among Mn-containing
superoxide dismutases [9].
DISCUSSION
The detailed data concerning Fe-SOD purified from leaves
of Lycopersicon allow the comparison of this enzyme with
those isolated from procaryotic sources, since Fe-SOD is
much better characterized in procaryotic cells [2, 27, 30, 32,
331 than in higher plants [13, 141. The specific activity of
purified Lycopersicon Fe-SOD (about 4000 standard units)
appears to be highest among those isolated from higher plants
[13,141and the eucaryotic alga Euglena gracilis [121and reaches the top values reported for procaryotic organisms [27,
281 and the protozoan Crithidia fasciculata [47]. The relative
molecular mass of the enzyme lies well in the range of 40000 46 000, reported for other dimeric ferric superoxide dismutases
[2, 12- 14,27, 30, 33,471. The same result is seen with respect
to the iron content, which varies from about 1 to 2 mol Fe/
464
Table 3. Comparison of SAQ values of iron-containing and manganese-containing superoxide dismutases
Values of SAQ were calculated from the following equation [37]: SAQ = C (xi, j - x k , j)’ using the amino acid composition of tomato Fej
SOD reported here and that of other superoxide dismutases taken from the literature. Subscripts i and k indicate the particular protein which
is compared, x j is the content of a given amino acid of j type. 18 amino acids were used for calculations. For P. shermanii and N . asteroides
SAQ values were computed for 17 amino acids, omitting tryptophan and cysteine respectively
Source
a Lycopersicon
esculentum
b Propionibacterium
shermanii
C Pisum sativum
d Brassica campestris
e Escherichia coli
f Bacteroides fragilis
g Plectonema boryanum
h Streptococcus
faecalis
Paracoccus
1
denitrijkans
j Rat liver
k Saccharomyces
cerevisiae
I Methanobacterium
bryantii
m Chromatium vinosum
n Nocardia asteroides
0 Escherichia coli
P Pseudomonas ovalis
9 Thermoplasma
acidophilum
r Spirulina platensis
S Porphyridium
cruentum
t Chlorobium
thiosulfatophilum
U Rhodopseudornonas
spheroides
Nuphar luteum
V
W Euglena gracilis
isoenzyme I
X Crithidia
fasciculata
Refer-Metal
ence
Fe
a
b c
d e
f
g h i
k 1
j
m n o p q
r
s
t
u v
x
-
Fe/Mn
29
-
Mn
Fe
Mn
Fe/Mn
Fe
Mn
34
37
40
40
42
42
49
48
31
24
39
44
Mn
45 34 41 21 24 12 32 35 -
Mn
Mn
48 65 28 22 5 5 33 83 54 29 49 47 47 33 30 18 55 34 19 25 -
Fe
51 5 5 40 50 39 35 72 43 27 33 18 -
Fe
Fe/Mn
Fe
Fe
Fe
52
52
56
57
61
Fe
Mn
62 41 92 60 40 45 20 48 56 109 69 99 41 47 26 68 85 62 73 64 32 61 54 59 33 58 66 73 94 55 62 27 48 90 71 -
Fe
65 46 79 35 40 22 32 18 30 71 37 59 32 32 34 55 65
38 52 -
Mn
69 58 5 5 24 43 20 50 48 13 25 27 34 55 48 59 39 51
70 68 42
Fe
Fe
82 77 102 38 64 41 59 75 45 63 14 87 71 56 73 45 81 74 80 56 42 84 76 102 61 107 78 74 51 72 95 95 102 97 48 60 81 142 108 72 74 86 88
Fe
w
42
20
44
79
46
-
34
51
51
78
52
42
69
63
53
61
35
19
36
37
32
47
30
25
51
21 26 29 39 32 47
54
37
34
65
21
34
24
40
45
23
49
29
23
47
71
-
39
37
33
55
64
39
22
44
38
38
68
67
78
50
50
57
44
57
62
31
75
57
86
64
36
-
55
38
42
73
38 57 49 66 82 92 -
103 62 84 62 43 50 52 60 50 105 78 88 41 49 42 65 76
mol dimer. It should be assumed that all these proteins contain
1 mol Fe/mol subunit and intermediate values reflect partial
release of the metal during the separation procedure. Spectral
properties of the enzyme are comparable with those of other
Fe-SOD enzymes. Besides a strong absorption band with a
maximum at 278 nm possessing typical shoulders, a broad
band in the visible region, attributable to a ligand-to-metal
charge transfer [48], can be observed. This visible light band
was not reported for two eucaryotic plant Fe-SOD enzymes
isolated from Euglena [ 121 and Brassica [ 131. The temperature
stability of tomato Fe-SOD is in the same range as those
reported for procaryotic enzymes [32 - 351 although very
labile enzymes were described for luminous bacteria [31] and
Euglena [12]. Lycopersicon Fe-SOD shows the same sensitivity
to inhibitors, such as azide, cyanide and hydrogen peroxide,
as other enzymes of this type, from procaryotic as well as
eucaryotic organisms. It also exhibits a similar isoelectric
point and amino acid composition. However, with respect to
the latter feature the enzyme seems rather to resemble the
manganese-containing superoxide dismutases.
-
-
43 79 39 64 60 120
-
Isolation and characterization of Fe-SOD from tomato
leaves as well as identification of it in bean leaf extract [16]
add the two more families of land plants to those already
known to contain this enzyme. Thus, the existence of Fe-SOD
in plants does not seem to be so exceptional as was previously
believed [151.
Among hypotheses formulated on the evolution of FeSOD [15], the most plausible seems to be the one which
assumes that the enzyme is descended from more primitive
plants. In higher plants Fe-SOD might have originated from
blue-green algae, which became chloroplasts through endosymbiosis [49, 501. After elaboration of the transport into
chloroplasts of proteins encoded in the nuclear genome [51],
Cu, Zn-SOD could enter these organelles and replace FeSOD. However, organisms containing two superoxide dismutases with different properties, metal requirement and
possibly different locations within chloroplasts could have an
evolutionary advantage. Thus, primitive Ginkgo biloba
possess both types of chloroplast dismutases [ 151, while
accidental loss of Cu, Zn-SOD, as might be the case in
465
Nymphaceae [15], should cause a stabilization of an already
existing Fe-enzyme. Similarly, a temporary loss of Cu, ZnSOD from chloroplasts, caused by cold treatment of leaves
[52], while cyanide-insensitive superoxide dismutase activity
is retained [53], seems to indicate that under these conditions
Fe-SOD may take over the function of the Cu, Zn-containing
enzyme.
The comparison of the amino acid composition of Fe-SOD
and Mn-SOD enzymes (Table 3) raises another possibility. FeSOD from tomato is strikingly closely related to Mn-SOD
enzymes, even more than to many other superoxide dismutases. This implies that the borderline between these two groups
of proteins is not sharp. They may even be regarded as
isozymes differing in the metal present in the active center.
Indeed, hybridization of Escherichia coli Mn-SOD and FeSOD subunits has been reported 1541and bacterial superoxide
dismutases, which may accept either iron or manganese, or
both, have recently been described [2,38,46]. Moreover, some
Fe-containing enzymes proved to be more closely related to
Mn-containing than to other ferric superoxide dismutases [9].
Thus, we are not able to exclude the possibility of reversed
evolution of some Fe-SOD enzymes from Mn-SOD. In spite
of earlier reports [4,6], Mn-SOD has never unequivocally been
localized within chloroplasts. The Mn-SOD isolated from the
higher plant. Pisum sativum, was proved to be located in
microbodies [55]. On the other hand, Fe-SOD was found in
chloroplasts of mustard [56] and tomato leaves (our
unpublished results). Thus, a reason might exist for precluding
Mn-SODcfrom chloroplasts, and adopting one enzyme rather
than the other may give the cell a substantial advantage. This
may also lie behind the origin of Fe-SOD in higher plants.
We wish to express our gratitude to Professors Helmut Sies
(Universitat Dusseldorf, FRG), Franz Miiller (Agricultural University, Wageningen, The Netherlands) and Achim Trebst (RuhrUniversitat, Bochum, FRG) for generous gifts of xanthine oxidase,
K,f 0 and Amicon Diaflo ultrafiltration membranes respectively.
We are also indebted to Dr A. Skladanowski (Department of Biochemistry, Medical Academy, Gdansk, Poland) for performing
ultracentrifuge sedimentation experiments, to Dr H. Debiec (Monument Hospital Child Health Centre, Warszawa, Poland) for help with
electrofocusing analysis and to Mr L. Kiedrowski (Nencki Institute
of Experimental Biology, Warszawa, Poland) for’ making available
the Amicon cell. Tomato leaves were kindly provided by Panstwowe
Gospodarstwo Ogrodnicze ‘Mysiadlo’. This work has been in part
supported by the Department of Plant Physiology, Polish Academy
of Science (grant MR 11/7).
~
REFERENCES
1. Fridovich, I. (1975) Annu. Rev. Biochem. 44, 147-159.
2. Gregory, E. M. & Dapper, Ch. H. (1983) Arch. Biochem. Biophys.
220,293 - 300.
3. McCord, J. M. & Fridovich, I. (1969) J . Biol. Chem. 244,60496055.
4. Asada, K., Kanematsu, S., Okaka, S. & Hayakawa, T. (1980) in
Chemical and Biochemical Aspects of Superoxide and Superoxide Dismutase (Bannister, J. V. & Hill, H. A. O., eds)
pp. 136 - 153, Elsevier North Holland, New York.
5 . Arron, G. P., Henry, L., Palmer, J. M. & Hall, D. 0. (1976)
Biochem. Soc. Trans. 4,618-620.
6. Lumsden, J. & Hall, D. 0. (1974) Biochem. Biophys. Res.
Commun. 58, 35-41.
7. Jackson, Ch., Dench, J., Moore, A. L., Halliwell, B., Foyer, Ch.
H. & Hall, D. 0. (1978) Eur. J. Biochem. 91, 339-344.
8. Foster, J. G. & Edwards, G. E. (1980) Plant Cell Physiol. 21,
895 - 906.
9. Martin, J. P., Jr & Fridovich, I. (1981) J . Biol. Chem. 256, 60806089.
10. Harris, J. I., Auffrat, A. D., Northrop, F. D. & Walker, J. E.
(1980) Eur. J . Biochem. 106, 297 - 303.
11. Fridovich, I. (1974) Life Sci. 14, 819-826.
12. Kanematsu, S. & Asada, K. (1979) Arch. Biochem. Biophys. 1Y5,
535 - 545.
13. Salin, M. L. & Bridges, S. M. (1980) Arch. Biochem. Biophys. 201,
369 - 374.
14. Salin, M. L. &Bridges, S. M. (1982) Plant Physiol. 69, 161 -165.
15. Bridges, S. M. & Salin, M. L. (1981) Plant Physiol. 68,275-278.
16. Kwiatowski, J. & Kaniuga, Z. (1984) Acta Physiol. Plant. 6,
197 - 202.
17. Weber, K. & Osborn, M. (1969) J . Biol. Chem. 244,4406-4412.
18. Davis, B. J. (1964) Ann. N . Y. Acad. Sci. 121, 404-427.
19. Beauchamp, C. 0. & Fridovich, I. (1971) Anal. Biochem. 44,
267 - 287.
20. Asada, K., Takahashi, M. & Nagate, M. (1974) Agric. Biol. Chem.
38,471 -473.
21. Henry, L. E. A,, Halliwell, B. & Hall, D. 0. (1976) FEBS Lett.
66, 303 - 306.
22. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J.
(1951) J . Biol. Chem. 193,265-275.
23. Murphy, J. B. & Kies, M. W. (1960) Biochim. Biophys. Acta 45,
382- 384.
24. Cohn, E. J. & Edsal, J. J. (1941) Proteins, Amino Acids and
Peptides, pp. 374- 376, Reinhold, New York.
25. Moore, S. (1963) J . Biol. Chem. 238, 235-237.
26. Edelhoch, H. (1967) Biochemistry 6, 1948- 1954.
27. Yamakura, F. (1976) Biochim. Biophys. Acta 422,280-294.
28. Misra, H. P. & Fridovich, I. (1977) J. Biol. Chem. 252, 64216423.
29. Yost, F. J., Jr & Fridovich, I. (1973) J . Biol. Chem. 248, 49054908.
30. Asada, K., Yoshikawa, K., Takahashi, M., Maeda, Y. & Enmanji,
K. (1975) J . Biol. Chem. 250, 2801 -2807.
31. Puget, K. & Michelson, A. M. (1974) Biochimie (Paris) 56,
1255- 1267.
32. Lumsden, J., Cammack, R. & Hall, D. 0. (1976) Biochim.
Biophys. Acta 438, 380-392.
33. Kanematsu, S. & Asada, K. (1978) Arch. Biochem. Biophys. 185,
473 -482.
34. Kanematsu, S. & Asada, K. (1978) FEBS Lett. 91, 94-98.
35. Searcy, K. B. & Searcy, D. G. (1981) Biochim. Biophys. Acta 670,
39-46.
36. Misra, H. P. & Fridovich, I. (1978) Arch. Biochem. Biophys. 189,
317-322.
37. Marchalonis, J. J. & Weltman, J. K. (1971) Comp. Biochem.
Physiol. 38B, 609 - 625.
38. Meier, B., Barra, D., Bossa, F., Calabrese, L. & Rotilio, G. (1982)
J. Biol. Chem. 257, 13977-13980.
39. Sevilla, F., Lopez-Gorge, J. & Del Rio, L. A. (1982) Plant Physiol.
70, 1321 - 1326.
40. Steinman, H. M. (1978) J. Biol. Chem. 253,8708-8720.
41. Bridgen, J., Harris, J. I. & Northrop, F. (1975) FEBS Lett. 49,
392- 395.
42. Terech, A. & Vignais, P. M. (1981) Biochim. Biophys. Acta 657,
411-424.
43. Salin, M. L., Day, E. D.. Jr & Crapo, J. D. (1978) Arch. Biochem.
Biophys. 187,223-228.
44. Ravindranath, S. D. & Fridovich, I. (1975) J . Bid. Chem. 250,
6107- 6112.
45. Kirby, T. W., Lancaster, J. R., Jr & Fridovich, I. (1981) Arch.
Biochem. Biophys. 210, 140- 148.
46. Beaman, B. L., Scates, S. M., Moring, S. E., Deem, R. & Misra,
H. P. (1983) J . Biol. Chem. 258,91-96.
47. Trant, N. L., Meshnick, S. R., Kitchener, K., Eaton, J. W. &
Cerami, A. (1983) J . Biol. Chem. 258, 125-130.
48. Barrette, W. C., Jr, Sawyer, D. T., Fee, J. A. & Asada, K. (1983)
Biochemistry 22,624-627.
466
49. Margulis, L. (1970) Origin qf’ Eucaryotic Cells, Yale University
Press, New Haven.
50. Doolittle, W. F. (1980) Trends Biochem. Sci.6, 146-151.
51. Ellis, R. J . (1981) Annu. Rev. Plant. Physiol. 32, 111 -137.
52. Michalski, W. P. & Kaniuga, Z. (1982) Biochim. Biophys. Acta
680,250-257.
53. Kaniuga, Z., Zgbek, J. & Michalski, W. P. (1979) Planta 145,
145- 150.
54. Dougherty, H. W., Sadowski, S. J. & Baker, E. E. (1978) J . Bid.
Chem. 253,5220 - 5223.
55. Del Rio, L. A,, Lyon, D . S., Olah, I., Click, B. & Salin, M. L.
(1983) Planta 158, 216-224.
56. Salin, M. L. & Bridges, S. M. (1980) 2. Pflunzenphysiol. 99,3745.
J. Kwiatowski, A. Safianowska, and Z. Kaniuga, Instytut Biochemii, Wydzial Biologii, Uniwersytet Warszawski,
Aleja Zwirki i Wigury 93, PL-02-089 Warszawa, Poland