Rat Mitochondrial Manganese Superoxide Dismutase: Amino Acid PositionsInvolved in Covalent Modifications, Activity, and Heat Stability

Immacolata Castellano,1 Francesca Cecere,1 Alberto De Vendittis,1 Roberta Cotugno,1

Angela Chambery,2 Antimo Di Maro,2 Andzelika Michniewicz,3 Giuseppe Parlato,3

Mariorosario Masullo,1,4 Enrico Vittorio Avvedimento,5 Emmanuele De Vendittis,1

Maria Rosaria Ruocco1

1 Dipartimento di Biochimica e Biotecnologie Mediche, Universita di Napoli Federico II, Via S. Pansini 5, 80131 Napoli, Italy

2 Dipartimento di Scienze della Vita, II Universita di Napoli, Via Vivaldi 43, 81100, Caserta, Italy

3 Dipartimento di Medicina Sperimentale e Clinica ‘‘G. Salvatore’’, Universita di Catanzaro ‘‘Magna Graecia’’,

Via T. Campanella 115, 88100 Catanzaro, Italy

4 Dipartimento di Scienze Farmacobiologiche, Universita di Catanzaro ‘‘Magna Graecia’’, 88021 Roccelletta di Borgia (CZ), Italy

5 Dipartimento di Biologia e Patologia Molecolare e Cellulare, Universita di Napoli Federico II, Via S. Pansini 5, 80131 Napoli, Italy

Received 15 December 2008; revised 10 March 2009; accepted 3 April 2009

Published online 21 April 2009 in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/bip.21208

This article was originally published online as an accepted

preprint. The ‘‘Published Online’’ date corresponds to the preprint

version. You can request a copy of the preprint by emailing the

Biopolymers editorial office at biopolymers@wiley.com

Rat Mitochondrial Manganese Superoxide Dismutase: Amino Acid PositionsInvolved in Covalent Modifications, Activity, and Heat Stability

This article is in honour of Professor Lelio Mazzarella, on theoccasion of his 70th birthday.

Correspondence to: E. De Vendittis; e-mail: devendittis@dbbm.unina.it or M. R.

Ruocco; e-mail: ruocco@dbbm.unina.it

ABSTRACT:

The role of three amino acid residues (Q143, Y34, S82) of

rat mitochondrial superoxide dismutase (ratSOD2) in the

enzymatic activity, thermostability, and post-translational

modification of the enzyme was investigated through site-

directed mutagenesis studies. Six recombinant forms of the

enzyme were produced, carrying the Q143 or H143 residue

with or without the Y34F or S82A replacement. All proteins

bound manganese as active cofactor and were organized as

homotetramers. The greatest effect on the activity (sixfold

reduction) was observed in ratSOD2 forms containing the

H143 variant, whereas Y34F and S82A substitutions

moderately reduced the enzymatic activity compared to the

Q143 form. Heat inactivation studies showed the high

thermo-tolerance of ratSOD2 and allowed an evaluation of

the related activation parameters of the heat inactivation

process. Compared to Q143, the H143 variant was

significantly less heat stable and displayed moderately lower

enthalpic and entropic factors; the Y34F substitution caused

a moderate reduction of heat stability, whereas the S82A

replacement slightly improved the thermo-tolerance of the

Q143 variant; both substitutions significantly increased

enthalpic and entropic factors of heat inactivation, the

greatest effect being observed with S82A substitution. All

recombinant forms of ratSOD2 were glutathionylated in

Escherichia coli, a feature pointing to the high reactivity of

ratSOD2 toward glutathione. Moreover, the S82 position of

the enzyme was phosphorylated in an in vitro system

containing human mitochondrial protein extracts as source

of protein kinases. These data highlight the role played by

some residues in ratSOD2 and suggest a fine regulation of

the enzyme occurring in vivo. # 2009 Wiley Periodicals, Inc.

Biopolymers 91: 1215–1226, 2009.

Keywords: superoxide dismutase; rat SOD2; mutagenic

analysis; post-translational modifications; thermostability

Contract grant sponsor: MIUR.

Contract grant number: PRIN 2007.�C 2009 Wiley Periodicals, Inc.

Biopolymers Volume 91 / Number 12 1215

INTRODUCTION

The cell has developed enzymatic and nonenzymatic

systems aimed at the defence against reactive oxygen

species (ROS), mainly produced during the oxygen

metabolism. Indeed, ROS are highly reactive and

toxic compounds, as they can modify many cellular

targets, such as lipids, proteins, and DNA.1 On the other

hand, ROS have also a physiological role, being messengers

in many cellular processes, such as proliferation, differentia-

tion, apoptosis, and senescence.2,3 An early cellular defence is

provided by superoxide dismutase (SOD), a key antioxidant

enzyme that eliminates the toxic superoxide anion, the first

ROS produced through the univalent reduction of oxygen.4,5

Indeed, SOD is a metal enzyme that dismutates two super-

oxide anions into hydrogen peroxide and molecular oxygen

through a catalytic mechanism conserved in taxonomically

distant organisms. However, this ubiquitous enzyme is classi-

fied in various structurally unrelated families on the basis of

the type of metal present in its active site.6,7 The two major

groups include the Cu/Zn-SOD (SOD1), present in the cyto-

sol of eukaryotic cells, and the Fe- and Mn-SOD (SOD2),

found in bacteria, archaea and also in mitochondria and

chloroplasts of eukaryotic cells. Therefore, SOD2 is widely dis-

tributed in all living kingdoms and probably displays the most

crucial defence mechanism against the oxidative stress.

Because of their intense respiratory activity, mitochondria are

the major cellular site for the production of superoxide

anions; therefore, the high efficiency of the mitochondrial

Mn-SOD is indispensable for cell viability, as confirmed by

the involvement of this enzyme in a number of diseases caused

by mitochondrial defects.8–12 The importance of SOD2 in

mammals is also highlighted by the observation that, in

contrast to SOD1, knock-out of its gene is lethal to mice.13,14

Mammalian mitochondrial Mn-SOD is encoded by the

nuclear genome as a precursor headed by a signal peptide

spanning 24 amino acid residues; this leader peptide is

removed for the entry of the enzyme in mitochondria.15 The

crystal structure of the human mitochondrial enzyme points

to its organization as a compact homotetramer, with an

interface made of two 4-helix bundles.16 Because of its com-

pactness, the human enzyme is endowed with a remarkable

thermal resistance.17 Another property shown by the human

Mn-SOD concerns its high reactivity toward peroxynitrite.18–20

Indeed, this highly toxic ROS causes the irreversible inactiva-

tion of the enzyme through the covalent modification of

Y34, a conserved residue in the channel that drives the super-

oxide anion to the active site. Other members of the Fe- and

Mn-SOD family display a similar high reactivity toward non-

physiological or physiological compounds, either on this

conserved tyrosine,21–25 or on cysteine residues.26–28

The rat mitochondrial Mn-SOD (ratSOD2), sharing 93%

amino acid identity with the human counterpart, is an

appropriate model to study the molecular and functional

properties of mammalian mitochondrial SOD2.29–31 This

investigation regards the role of some key amino acid posi-

tions of ratSOD2 regulating catalysis, reactivity, and thermal

resistance of the enzyme. The study was carried out through

the heterologous production of the mature form of ratSOD2

and mutants obtained by site-directed mutagenesis on the

corresponding gene. Among the three amino acid positions

considered in the mutagenic analysis, Q143 belongs to the

second coordination sphere of the metal ion. This residue

forms a hydrogen bond with a Mn-bound water molecule

and the hydroxyl side chain of Y34. In SODs belonging to the

same family, the position corresponding to Q143 is occupied

by glutamine or histidine.23,32 Furthermore, this position has

been hypothesized as a possible polymorphic site in rat-

SOD2, because of the occurrence of rat clones predicting the

H143 variant.29–31 The second amino acid considered in the

study is Y34, a conserved residue of the second coordination

sphere of the metal.16 This residue plays an important role in

the catalytic mechanism of superoxide dismutation and acts

as a powerful regulator of the activity of SOD2.17,21,32 Its

high reactivity toward peroxynitrite and/or phenylmethane-

sulfonyl fluoride has been proved even for bacterial and

archaeal Fe-SODs.21–25 The third residue considered is S82, a

potential target for post-translational modification of rat-

SOD2, as suggested by a phosphorylation prediction study. A

recent report, describing the phosphoproteomic screen on

proteins of the mitochondrial matrix, has identified

Mn-SOD as a target of phosphorylation.33

RESULTS

Molecular Properties of RatSOD2 Mutants

Using a heterologous expression system constituted by the

prokaryotic vector pT7-7 and the Escherichia coli host strain

BL21(DE3), a recombinant mature form of ratSOD2, called

SOD2(Q143), was produced and purified. In this recombi-

nant version, the initial Lys residue of the endogenous

mature enzyme (K25 in the premature form of ratSOD2) was

replaced by Met, representing the translation start codon in

the amplified ratSOD2 gene cloned in pT7-7. The purified

SOD2(Q143) was homogeneous when analyzed by SDS/

PAGE and showed an electrophoretic mobility corresponding

to a molecular mass of 24 kDa, slightly higher than the

expected size of the monomer subunit (22.3 kDa). The Mr of

SOD2(Q143) was also determined under nondenaturing

conditions by gel-filtration. The protein sample eluted as a

1216 Castellano et al.

Biopolymers

single symmetric peak with an apparent Mr of 90,000, a value

approaching the theoretical mass of a homotetramer.

To exclude that the small difference at the N-terminus

between endogenous ratSOD2 and SOD2(Q143) caused a

partial trimming of some residues in the recombinant pro-

tein, the molecular mass of the purified SOD2(Q143) was

analyzed by ESI/Q-TOF mass spectrometry on two different

preparations of the enzyme. Indeed, the major peak had an

experimental Mr of 22294.7, in good agreement with the the-

oretical value of 22294.3 derived from the amino acid

sequence of the expected heterologous product, including

the initial Met residue (Table I). Therefore, the purified

SOD2(Q143) was produced as an intact molecule, the only

difference with the endogenous ratSOD2 being at its N-ter-

minal residue.

The described expression system was productively used

for the obtainment of mutant forms of ratSOD2. To study

the possible variance between Gln and His at position 143 of

ratSOD2, a recombinant form of the enzyme was obtained,

carrying the H143 residue and thus called SOD2(H143).

Afterward, to investigate on the role of the Y34 position, the

Y34F replacement was realized on either SOD2(Q143) or

SOD2(H143) and the corresponding mutant forms were

called SOD2(Y34F-Q) and SOD2(Y34F-H), respectively. In a

similar way, the role of the S82 position was analyzed

through the S82A substitution and the corresponding mu-

tant forms were called SOD2(S82A-Q) and SOD2(S82A-H).

Variant and mutant forms of ratSOD2 had the same behavior

of SOD2(Q143), when analyzing their electrophoretic mobil-

ity on SDS/PAGE or elution under nondenaturing conditions

by gel-filtration. Therefore, all of them conserved the homo-

tetrameric structure of the enzyme. Moreover, all the

recombinant proteins were analyzed by ESI/Q-TOF mass

spectrometry and the experimental Mr values obtained from

two different preparations are reported in Table I. For all of

them, there was a good agreement with the theoretical Mr

derived from the amino acid sequence, taking into account

the presence of the initial Met residue.

Sulfhydryl Reactivity

The ratSOD2 amino acid sequence contains two cysteine res-

idues, C140 and C196 (sequence numbering of the mature

protein). Previous investigations on the closely related

human mitochondrial Mn-SOD, containing the same cyste-

ine residues, indicated a sulfhydryl reactivity toward thiol

reagents.26,27 MS analysis of SOD2(Q143), SOD2(H143),

and their mutants indicated that the sulfhydryl groups were

reactive. Indeed, besides the presence of potassium adducts,

an additional peak corresponding to a 305 Da extra mass was

clearly evident in all mass spectra of two different prepara-

tions of the six protein samples. As an example, Figure 1

reports the mass spectra of SOD2(S82A-Q) and SOD2(S82A-

H), containing the highest (32.0%) and lowest (2.7%)

amount of this extra peak, respectively. The 305 Da extra

mass can be accounted for a glutathionylation reaction

occurring on ratSOD2, likely due to a high sulfhydryl reactiv-

ity of its cysteine residue(s). The level of the glutathionylated

adduct was variable either among different or within the

same ratSOD2 form (Table I); however, no multiple gluta-

thionylated adducts were deduced from the mass spectra,

thus indicating that only one cysteine, among C140 and

C196, was covalently modified by glutathione. These findings

are intriguing, because a glutathionylated adduct on SOD2

was recently reported even for the eubacterial Fe-SOD from

P. haloplanktis.28 Hence, our data extend the occurrence of

this glutathionylation reaction, consistently detected in the

ratSOD2 forms analyzed, also to a mitochondrial SOD. The

great variability of the glutathionylation level observed in

preparations of the same ratSOD2 form do not allow any

conclusion concerning a possible relationship between amino

acid replacements and sulfhydryl reactivity of recombinant

ratSOD2.

Table I Molecular Mass Values (Mr) Obtained by ESI-Q/TOF

Mass Spectrometry Determined on Two Different Preparations of

Recombinant Wild Type and Mutant Forms of RatSOD2 and Rel-

ative Percentages of Glutathionylated Adduct, Calculated on the

Basis of the Relative Intensities of the Deconvoluted Mass Peaks

Protein

Experimental

Mra

Delta from

Theoretical Mr

Glutathionylated

Adduct (%)b

SOD2(Q143) 22294.7 10.4 5.1

10.0

SOD2(H143) 22303.8 10.4 7.5

5.7

SOD2(Y34F–Q) 22278.9 10.6 12.9

11.8

SOD2(Y34F–H) 22287.8 10.5 6.6

22.6

SOD2(S82A–Q) 22278.7 10.4 32.0

9.7

SOD2(S82A–H) 22287.8 10.5 2.7

30.0

a Mean value of the mass peak corresponding to the unmodified protein;

the variability among the two determinations never exceeded 50 ppm.b In the calculation of the total amount of unmodified and modified

protein, the potassium adducts revealed in high amount in almost all

samples have been included, whereas other poor represented adducts

have not been considered. Each first and second figure is referred to a

protein sample purified after the incubation at 708C in the absence or in

the presence of MnCl2, respectively, as indicated in the Materials and

Methods section.

Rat Mitochondrial Key Residues in Superoxide Dismutase 1217

Biopolymers

FIGURE 1 Electrospray mass spectrum of SOD2(S82A-Q) (panel A) and SOD2(S82A-H) (panel B).

The mixed disulfide with glutathione is indicated by a solid arrow, whereas the main potassium adduct

is indicated by a dashed arrow.

Table II Manganese Content and Specific Activity of Recombinant Wild Type and Mutant Forms of RatSOD2

Protein

Manganese Contenta

(mol/mol ratSOD2 subunit)

Specific Activityb

(U/mg protein)

Specific Activity

Referred to Metal Content

(U/mg protein3 mol Mn 3 mol subunit21)

SOD2(Q143) 0.30 5350 17,800

SOD2(H143) 0.26 750 2900

SOD2(Y34F–Q) 0.22 2650 12,000

SOD2(Y34F–H) 0.16 450 2800

SOD2(S82A–Q) 0.26 3750 14,400

SOD2(S82A–H) 0.16 500 3100

a Determined by graphite furnace atomic absorption spectrometry, as indicated in Materials and Methods.b Measured through the inhibition of the cytochrome c reduction in the xanthine/xanthine oxidase system, as indicated in Materials and Methods.

1218 Castellano et al.

Biopolymers

Another evaluation of the glutathionylation reaction on

the heterologous ratSOD2 forms was obtained through

immunoblotting experiments, using anti-glutathione anti-

bodies. This experiment confirmed the presence of the gluta-

thionylated adduct in all samples examined, because of the

presence of specific immunoreactive bands (data not shown).

Metal Content and Activity

The analysis of the metal content of the purified proteins

revealed that all of them bound manganese as metal cofactor,

even though in low amounts in either wild type or mutant

forms of the enzyme. An improvement of this content was

obtained in protein samples purified after a moderate ther-

mal treatment of the protein soluble fraction in the presence

of 10 mM MnCl2 (see Materials and Methods). However, the

manganese content of the MnCl2-treated samples remained

understoichiometric, as shown in Table II. Other SODs

belonging to the same family, even when purified from their

endogenous source, displayed a similar low metal content.

The differences in manganese content among the protein

samples analyzed probably indicate a lower metal binding

capacity exhibited by the ratSOD2 variant containing His at

position 143. A previous analysis on several SOD2 primary

structures pointed to the relevance of position 143 for the

type of metal bound in the active site of the enzyme.34 This

position is semiconservatively occupied by Gln or His, and

this difference is usually predictive of the active metal cofac-

tor of SOD2, namely Mn or Fe with Gln or His, respectively.

To exclude that the lower Mn content of SOD2(H143) with

respect to SOD2(Q143) was due to its replacement by Fe, the

iron content of both protein samples was determined.

Interestingly, besides the Mn ion, 0.16 and 0.18 mol Fe/

mol ratSOD2 subunit were found in SOD2(Q143) and

SOD2(H143), respectively. The presence of iron bound to

ratSOD2 could be related to the heterologous production of

the enzyme and/or to the occupancy of some empty metal

binding sites, as resulting from the low content of the active

manganese. Indeed, the uptake of putatively inactive metals

was already described for another SOD of the same family.35

However, the presence of a similar Fe content in wild type

and H143 variant of ratSOD2 indicates that the Q143H

replacement did not significantly alter the specificity of

ratSOD2 for metal uptake.

The SOD activity of the six heterologous proteins

reported in Table II was measured on enzyme preparations

obtained when the thermal purification step described in

the Materials and Methods section was carried out in the

presence of MnCl2, in order to improve the manganese ion

bound to each ratSOD2 form. For a better comparison

among the different proteins, the values of specific activity

were also referred to the Mn content. All enzymes were

produced in their active state; however, significant varia-

tions among the different protein samples were found. The

highest value of specific activity was found with

SOD2(Q143), whereas a very low value was observed with

SOD2(H143), because only 16% of the activity displayed

by SOD2(Q143) was measured. Vice versa, in SOD2

(Y34F-Q) and even more in SOD2(S82A-Q), the Y34F and

S82A replacements caused a modest effect, because 67 and

81% of the activity sustained by SOD2(Q143) was deter-

mined, respectively. When the effect of the same replace-

ments was analyzed in the H143 variant of ratSOD2, the

observed decrease of activity almost exclusively reflected

the presence of His143 residue, because the activity of

SOD2(Y34F-H) and SOD2(S82A-H) was similar to that

exhibited by SOD2(H143).

Heat Stability

To evaluate the heat resistance of ratSOD2, heat inactivation

profiles of SOD2(Q143), SOD2(H143), SOD2(Y34F-Q), and

SOD2(S82A-Q) were realized, using the enzyme preparations

obtained with the thermal purification step carried out in the

presence of MnCl2. To this aim, the protein samples were

incubated for 10 min at different temperatures and after-

ward, their residual activity was measured and compared to

the corresponding untreated samples. The resulting heat

inactivation profiles indicate that all the recombinant forms

of the enzyme are endowed with a significant heat resistance,

because of the high values of half-inactivation temperatures

(T1/2), ranging between 77 and 818C (see Figure 2). An eval-

uation of the different heat stability among the proteins was

attempted. In particular, the T1/2 of 808C calculated for

SOD2(Q143) was slightly decreased to 798C in SOD2(Y34F-

Q), whereas a more pronounced decrease of heat stability

was found in SOD2(H143), whose T1/2 was 778C; on the

contrary, in SOD2(S82A-Q) the T1/2 increased to 818C. Fur-thermore, some differences were found in the shape of the

profiles; for instance, that related to SOD2(H143) had a very

broad shape.

To examine the effect of temperature on the inactivation

reaction of recombinant wild type and mutant forms of rat-

SOD2, kinetics were carried out in the temperature interval

75–838C. As an example, the time course of inactivation at

808C of the four proteins is shown in Figure 3. All reactions

followed first-order kinetics with rates depending on the rat-

SOD2 form analyzed. Indeed, compared to SOD2(Q143)

(Figure 3A), a moderately slower rate was measured with

SOD2(S82A-Q) (Figure 3C); vice versa, with SOD2(H143)

Rat Mitochondrial Key Residues in Superoxide Dismutase 1219

Biopolymers

(Figure 3B) and SOD2(Y34F-Q) (Figure 3D), a higher rate

of inactivation was observed. The rate constants of inactiva-

tion (kin) from the linear kinetics obtained at the various

temperatures (average squared correlation coefficient, 0.92

6 0.14) were treated according to the Arrhenius equation to

calculate the energetic parameters of the inactivation process

(see Figure 4). All plots were linear in the interval of temper-

ature investigated, with different slopes among the various

proteins. For instance, the slope related to SOD2(S82A-Q)

(Figure 4C) or SOD2(Y34F-Q) (Figure 4D) was greater than

that of SOD2(Q143) (Figure 4A) or SOD2(H143) (Figure

4B). Table III reports the values of activation energy, as well

as enthalpy, entropy, and free energy of activation, calculated

from the Arrhenius plots reported in Figure 4. Interestingly,

the Ea of SOD2(Q143) (299 kJ/mol) was significantly lower

than that observed upon the Y34F substitution (410 kJ/mol);

a more pronounced effect was observed with the S82A

replacement (476 kJ/mol); vice versa, in SOD2(H143) only a

modest decrease of Ea (270 kJ/mol) was observed. Striking

differences were also found in the DH* and DS* values calcu-

lated at 808C, accounting for diverse enthalpic and entropic

contribution to the inactivation process of each enzyme

sample. Indeed, compared to the values calculated for

SOD2(Q143), a significant increase of the enthalpic factor

was measured upon the Y34F substitution, counteracted by a

more favorable entropic behavior. The enthalpic factor cal-

culated upon the S82A replacement was even greater, again

counteracted by a significantly higher entropy. On the con-

trary, both DH* and DS* values of SOD2(H143) were lower

compared to those of SOD2(H143).

Phosphorylation at Ser82

A phosphorylation prediction study made with the NetPhos

program (www.expasy.org) on ratSOD2 indicated Ser82 as a

potential target of post-translational modification by protein

kinase(s) with a high prediction score of 0.974. To investigate

on the possible phosphorylation of this residue, an in vitro

kinase reaction was carried out. In particular, mitochondrial

protein extracts from the human neuroblastoma cell line

SHSY-5Y were used as a source of protein kinases and incu-

bated with [c–32P]ATP in the absence or in the presence of

recombinant SOD2(Q143) or SOD2(S82A-Q) (see Materials

and Methods). After fractionation of the reaction mixtures

FIGURE 2 Heat inactivation profiles of recombinant wild type

and mutant forms of ratSOD2. Samples of 10 lg/mL SOD2(Q143)

(panel A), SOD2(H143) (panel B), SOD2(S82A-Q) (panel C), and

SOD2(Y34F-Q) (panel D) in buffer A were incubated for 10 min at

the indicated temperatures. The tubes were then cooled on ice for 2

h, and aliquots were withdrawn and analyzed for residual SOD ac-

tivity, as indicated in the Materials and Methods section. Care was

taken to measure similar activity levels. Results were expressed as

percentage of the activity measured on untreated samples kept

at 08C.

FIGURE 3 Inactivation kinetics at 808C of recombinant wild

type and mutant forms of ratSOD2. A solution of 10 lg/mL

SOD2(Q143) (panel A), SOD2(H143) (panel B), SOD2(S82A-Q)

(panel C), and SOD2(Y34F-Q) (panel D) in buffer A was aliquoted

in different tubes and incubated at 808C. At the times indicated,

each sample was cooled on ice for 2 h, before measuring its residual

SOD activity, as reported in the legend to Figure 2. The data were

plotted according to a first order kinetics, where A0 or At represents

the activity measured at the time zero or time t, respectively.

1220 Castellano et al.

Biopolymers

by SDS/PAGE, an autoradiography was obtained (see Figure 5).

Interestingly, in the sample containing SOD2(Q143), a radio-

active signal was found with a size corresponding to that of

ratSOD2; vice versa, this signal was almost absent either in

the sample incubated with SOD2(S82A-Q), as well as in

mitochondrial extracts alone. This finding suggests that

the wild type ratSOD2 could be target of a phosphorylation

reaction on the Ser82 position.

DISCUSSIONIn the present article, we have analyzed the role of three

amino acid positions of ratSOD2 involved in catalysis, ther-

mostability, and also in post-translational modification of

the enzyme. In particular, Y34 and Q143, interacting residues

of the second coordination sphere of the metal, display a

crucial role in structure and activity of SOD2 in different

sources. Although Y34 is absolutely conserved, the position

corresponding to Q143 is occupied by a His residue in some

SOD2; moreover, the sequence of some rat clones predict a

H143 variant of ratSOD2. We have also analyzed S82 as a

potential target of phosphorylation in ratSOD2. This posi-

tion, although conserved in the highly homologous human

Mn-SOD, exhibits a lower phosphorylation prediction

score in this latter enzyme, probably because of the slightly

different consensus sequence.

To study the role of these residues on enzymatic activity

and heat stability of ratSOD2, we have first produced as

recombinant proteins the two putative variants of ratSOD2,

namely SOD2(Q143) and SOD2(H143), and afterward, four

mutant forms containing the Y34F or S82A replacement in

both variants of the enzyme, namely SOD2(Y34F-Q),

SOD2(Y34F-H), SOD2(S82A-Q), and SOD2(S82A-H). The

study of their molecular properties did not reveal any signifi-

cant difference in the homotetrameric structure of the

enzyme, thus excluding the involvement of these residues in

the organization of the quaternary structure.

As expected for a mitochondrial enzyme, the recombinant

ratSOD2 forms bound manganese as metal cofactor. The het-

erologous expression system used was unable to produce

SODs with an acceptable amount of this metal; the incuba-

tion of the expressing cell extract in the presence of Mn11,

although improving the metal uptake, did not allow a reach-

ing of the theoretical metal content. The lower amount of

manganese found in the recombinant ratSOD2 forms, con-

taining His at position 143, probably reflects the relevance of

this position for a correct metal binding in the active site;

indeed, in the 3D structure of the corresponding human

enzyme, the wild type Gln forms a hydrogen bond network

including a manganese-bound water molecule.16 The uptake

of a low and similar amount of putatively inactive iron in

both wild type and His143 variant of ratSOD2 excludes an

FIGURE 4 Arrhenius plots of the heat inactivation process of

recombinant wild type and mutant forms of ratSOD2. The values of

the kinetic rate constant of inactivation kin (min21) were obtained

from heat inactivations carried out at 75, 78, 80 (shown in Figure

3), and 838C. SOD2(Q143) (panel A), SOD2(H143) (panel B),

SOD2(S82A-Q) (panel C), and SOD2(Y34F-Q) (panel D). Squared

correlation coefficients of the fits are: 0.935 (panel A); 0.987 (panel

B); 0.999 (panel C); 0.973 (panel D).

Table III Activation Energetic Parameters of the Heat

Inactivation Process Evaluated at 808C for Recombinant

Wild Type and Mutant Forms of RatSOD2

RatSOD2 Form

Ea

(kJ/mol)

DH*

(kJ/mol)

DS*

(J/mol K)

DG*

(kJ/mol)

SOD2(Q143) 299 296 533 108

SOD2(H143) 270 267 459 105

SOD2(Y34F–Q) 410 407 853 106

SOD2(S82A–Q) 476 473 1032 109

Data were derived from the Arrhenius plots reported in Figure 4.

FIGURE 5 Phosphorylation of ratSOD2. Samples containing 20

lg of mitochondrial protein extracts from SHSY-5Y cells were used

in an in vitro kinase assay, realized as indicated in the Materials and

Methods section, in the absence (lane 3) or in the presence of 1 lgof SOD2(Q143) (lane 1) or SOD2(S82A-Q) (lane 2).

Rat Mitochondrial Key Residues in Superoxide Dismutase 1221

Biopolymers

alteration of the specificity for metal uptake upon the

Q143H replacement.

All the recombinant ratSOD2 forms were purified as

active enzymes. The different values of specific activity meas-

ured on the six protein samples reflect the specific role of the

analyzed amino acid positions in catalysis. The greatest

effects on the activity are related to the position 143. Indeed,

the lowest levels of specific activity were found with all rat-

SOD2 forms, containing His at position 143, with a sixfold

reduction of activity compared to SOD2(Q143). Also for the

human enzyme, amino acid replacements at the wild type

Q143 position cause a dramatic decrease of activity.36 At

present, we do not know the putative physiological role of

the His143 variant of ratSOD2.29–31 We can speculate that

the low activity displayed in vivo by SOD2(H143) is likely

compensated by other antioxidant mechanisms.

The effects of Y34F and S82A substitutions on the activity

are much less crucial. Indeed, the 1.5-fold reduction

observed when the activity of SOD2(Y34F-Q) was compared

with that of SOD2(Q143) is similar to that reported for the

corresponding mutation in the human Mn-SOD.17,37 It has

been proposed that the effect of this mutation on the activity

becomes more crucial when a highly efficient catalysis is

required by increased superoxide levels.17 A lower effect was

found upon the S82A replacement, because the activity of

SOD2(S82A-Q) was quite similar to that determined on

SOD2(Q143). This is not surprising, because S82 is far away

from the active site of the protein molecule. The role of this

residue, probably exposed to solvent, could be related to a

possible regulation of the SOD activity by post-translational

modifications, as suggested by the phosphorylation analysis

presented in this work. The specific activity measured on

mutants containing the Y34F or S82A replacement in the

H143 variant of ratSOD2 was very similar to that determined

on SOD2(H143). Clearly, the significant reduction of activity

only depends on the presence of His at position 143.

The study of the inactivation process of ratSOD2 has pro-

vided some important insights on the high thermo-tolerance

of this enzyme, probably reflecting a required ability for its

adaptation to different types of stress. Similarly, also the

human Mn-SOD displays a great heat resistance.17 The inac-

tivation studies on ratSOD2 led to interesting observations

on the role of the analyzed amino acid positions. Indeed, the

thermal inactivation profiles showed that SOD2(S82A-Q)

is slightly more resistant to heat inactivation than

SOD2(Q143); vice versa, a decrease in the half-inactivation

temperature is observed in SOD2(Y34F-Q) and even more in

SOD2(H143). Differential scanning calorimetry experiments

carried out on the human Mn-SOD highlighted the great im-

portance of the Q143 position for the thermal stability of the

enzyme, because of a significant reduction in the unfolding

transition temperature observed in Q143 mutants of the

enzyme.36 On the other hand, similar studies on the Y34F

substitution pointed to a significant improvement of the

thermal stability of the wild type human enzyme.17 Our inac-

tivation studies on ratSOD2 are in agreement with the pub-

lished observations on the structural role played by the

human Q143 position, thus suggesting that rat Q143 is also

important for keeping the correct structure of ratSOD2. Vice

versa, an apparent disagreement exists with the results

reported in this work for the rat Y34F mutant, possessing a

slightly reduced heat resistance; however, this feature could

be in part related to the different methodology used to mea-

sure the effect of a heat treatment on the enzyme. We have

also evaluated the possibility that the different thermal

behavior of the ratSOD2 forms could be related to a different

glutathionylation extent in the protein samples analyzed.

However, this seems unlikely, because the enzyme prepara-

tions used in this thermal inactivation studies possess a low

percentage of glutathionylation, not exceeding 11.8%. Fur-

thermore, the modification level is very similar in three out

of the four protein samples considered in the study.

We have also measured the activation parameters of the

heat inactivation process and found striking differences in

the enthalpic and entropic contribution to this reaction. The

calculated values of Ea fall in a large interval (270–476 kJ/

mol), the highest values being curiously unusual for a protein

from a mesophilic source. Among the different proteins,

SOD2(Y34F-Q) and even more SOD2(S82A-Q) exhibit a sig-

nificantly higher Ea compared to SOD2(Q143), whereas a

slight reduction of this energetic parameter is observed in

SOD2(H143). The enthalpic and entropic contribution to

the activation of the inactivation process greatly varies

among the different ratSOD2 forms. Indeed, the more favor-

able entropic contribution to the process was found with

SOD2(S82A-Q), followed by SOD2(Y34F-Q). These findings

suggest that these replacements render the active site more

disordered during the inactivation process. However, this

favorable entropy is counteracted by an unfavorable enthalpy

in these mutant forms, likely due to a greater difficulty to

reach the transition state. A different behavior was found

with SOD2(H143), because in this case both the enthalpy

and the entropy of the inactivation process were found

reduced.

To identify possible post-translational modifications in

the recombinant purified proteins, they were analyzed by

ESI/Q-TOF mass spectrometry. Surprisingly, the analysis

revealed the presence of a 305 Da extra mass, due to a

glutathionylated adduct, as confirmed by the immunoblot-

ting analysis. This modification was present in all proteins

1222 Castellano et al.

Biopolymers

produced, even though to a different extent. The gluta-

thionylation reaction observed in heterologous proteins

produced in E. coli points to the high reactivity of rat-

SOD2 toward glutathione, a physiological thiol normally

present in this host strain.38,39 Recent studies on the regu-

lation of protein function by ROS suggest that S-gluta-

thionylation is an important mechanism involved in cellu-

lar response to oxidative damage and redox signaling.40–46

An old study reported the presence of glutathionylated

forms for chicken Cu/Zn-SOD,47 but the role of this

modification was not clarified yet. Recently, we have

reported that the Fe-SOD from P. haloplanktis is gluta-

thionylated in vitro on Cys57 and that the enzyme under-

goes an in vivo S-glutathionylation when overproduced in

E. coli cells.28 Also in mammalian cells, protein glutathio-

nylation is involved both in defence against oxidative

damage and in redox signaling. In mitochondria, the high

concentration of protein thiols and glutathione has been

related to the presence of several glutathionylated pro-

teins.48,49 The human mitochondrial Mn-SOD is very sen-

sitive to reducing agents,26 and this susceptibility has been

linked to the high reactivity of Cys196 exposed on the

surface of the tetrameric enzyme. Moreover, the I58T iso-

form of the human enzyme, organized as a dimer rather

than a normal tetramer, exposes also Cys140, a residue

normally buried in the tetrameric structure.50 Interestingly,

these two cysteine residues are also conserved in the

mature ratSOD2. Our data indicate that only one cysteine

is target of glutathionylation; on the basis of the high

similarity between human and rat SOD2, the candidate

for this reaction is Cys196, because the amino acid

replacements realized on ratSOD2 did not alter its homo-

tetrameric structure and probably do not expose Cys140

to the solvent. The MS analysis has revealed great differ-

ences among the glutathionylation levels of the various

protein samples; however, these differences could not be

related to a change in the sulfhydryl reactivity of ratSOD2

caused by the amino acid replacements realized.

A phosphorylation prediction study on ratSOD2 indicated

Ser82 as a potential target of post-translational modification

by protein kinase(s). Indeed, a recent article has reported

that the cytoplasmic Mn-SOD from Listeria monocytogenes, a

facultative intracellular pathogen for humans and animals, is

phosphorylated on serine and/or threonine residues and that

this post-translational modification downregulates the enzy-

matic activity during the stationary phase.51 Moreover, a

comprehensive analysis of cytoplasmic proteins phosphoryla-

tion in Campylobacter jejuni indicated Fe-SOD as one of the

major phosphoprotein.52 To date, there is no information

available on the amino acid position(s) susceptible to this

modification. Ser82 is conserved in several SOD2 from differ-

ent organisms, including human, rat, and L. monocytogenes

Mn-SOD. To investigate on the possible phosphorylation of

this residue, we have used an in vitro kinase reaction, in

which mitochondrial protein extracts from a human cell line

were considered as a source of serine/threonine protein

kinases. The experiment showed that the recombinant

SOD2(Q143) is phosphorylated by mitochondrial kinases;

vice versa, the mutant SOD2(S82A-Q) is not, suggesting that

Ser82 is the likely target of this modification. In the crystal

structure of the corresponding human enzyme, this residue

is located in a region apparently exposed to solvent.16 It

remains to be determined the function of ratSOD2 phospho-

rylation and its relation with the enzymatic activity of the rat

enzyme.

CONCLUSIONSThe produced recombinant forms of ratSOD2 allowed an

investigation on the role of three amino acid positions of the

enzyme. Among them, the most crucial role is played by

Q143, either in catalysis or structure of ratSOD2. Moreover,

our results on the possible post-translational modification of

the enzyme suggest that ratSOD2 could be finely regulated

by physiological compounds, such as glutathione, or other

modifications, such as phosphorylation by protein kinases.

We wish to stress that the resistance to peroxynitrite of

eubacterial Fe-SOD from P. haloplanktis is significantly

improved upon glutathionylation of the enzyme.28 Also rat-

SOD2, as already demonstrated for the corresponding

human enzyme, is nitrated on Y34 and inactivated by perox-

ynitrite (unpublished observation). Future research should

speculate on the in vivo glutathionylation and/or phospho-

rylation of the endogenous ratSOD2 in mammalian cells and

on the possible functional role of such post-translational

events.

MATERIALS AND METHODS

MaterialsRestriction and modifying enzymes, and [c–32P]ATP were from GE

Healthcare. Plasmid pGEM T-easy was from Promega. Vector pT7-7

and the E. coli BL21(DE3) strain were from Novagen. The Qiagen

kits (M-Medical) were used for purification of plasmids and DNA

fragments. Oligonucleotide synthesis and nucleotide sequencing

was carried out by Primm. Isopropyl-b-thiogalactopiranoside(IPTG), xanthine, xanthine oxidase, and cytochrome c were from

Sigma-Aldrich. Anti-glutathione monoclonal antibodies were from

Chemicon; horseradish peroxidase-conjugated anti-mouse antibod-

ies were from Santa Cruz. The chemiluminescent SuperSignal West

Rat Mitochondrial Key Residues in Superoxide Dismutase 1223

Biopolymers

Pico kit was from Pierce. Fuji films were used for autoradiography.

RPMI 1640 medium, fetal bovine serum (FBS), L-glutamine,

penicillin G, streptomycin, trypsin were purchased from Cambrex.

Protease inhibitor cocktail was obtained from Roche Diagnostics.

HPLC-grade solvents for mass spectrometry were obtained from

Carlo Erba. All other chemicals were of analytical grade. The follow-

ing buffers were used: buffer A, 20 mM Tris-Cl, pH 7.8; buffer B,

50 mM potassium phosphate, pH 7.2; buffer C, 20 mM Tris-Cl, pH

7.8, 50% (v/v) glycerol; buffer D, 100 mM potassium phosphate, pH

7,8, 0.1 mM Na2EDTA; buffer M, 5 mM Hepes, pH 7.4, 250 mM

mannitol, 0.5 mM EGTA, 5 mM Hepes, 0,1% BSA; PBS buffer,

10 mM sodium phosphate, pH 7.2, 150 mM NaCl; kinase buffer,

10 mM MOPS, pH 7.0, 7.7 mM MgCl2, 0.5 mM Na2EDTA, 0.5 mM

NaF, 0.5 mM Na3VO4.

MethodsSOD activity was measured at 258C in buffer D by the inhibi-

tion of cytochrome c reduction caused by superoxide anions

generated with the xanthine/xanthine oxidase method.35,53 One

unit of SOD activity was defined as the amount of enzyme

that caused 50% inhibition of cytochrome c reduction. The in

vitro kinase assay was realized by mixing mitochondrial protein

extracts with recombinant proteins in 30 lL kinase buffer sup-

plemented with 20 lM ATP and 1 lCi [c-32P]ATP. The reac-

tion mixture was incubated for 20 min at 308C and blocked by

adding the sample dilution buffer for SDS/PAGE. After frac-

tionation by SDS/PAGE, the gel was dried and radioactive signals

were visualized by autoradiography.

Protein concentration was determined by the method of Brad-

ford,54 using bovine serum albumin as standard. Purity of protein

samples was assessed by 14% SDS/PAGE according to standard pro-

tocols.55 The quaternary structure of wild type and mutant forms of

ratSOD2 was evaluated by gel-filtration on a Superdex 75 10/300

GL column (GE Healthcare). The metal content of the recombinant

protein samples was determined by graphite furnace atomic absorp-

tion spectrometry, as previously indicated.25 The presence of gluta-

thionylated adducts in the purified protein samples was revealed

through immunoblotting experiments realized with monoclonal

anti-glutathione antibodies, as previously reported.28

The energy of activation Ea and the Arrhenius constant A of the

heat inactivation process were calculated through the equation ln

kin 5 ln A 2 Ea/R 3 1/T, where kin is the heat inactivation rate con-

stant and R is the gas constant (8.314 J/mol K). The other thermo-

dynamic parameters of activation, namely enthalpy (DH*), entropy

(DS*), and free energy (DG*), were calculated at 808C according to

the equations: DH* 5 Ea 2 R 3 T; DS* 5 R 3 ln (h 3 NA 3 A/R

3 T 3 e); DG* 5 DH* 2 T 3 DS*, where h is the Planck constant

(6.624 3 10234J s), NA is the Avogadro’s number (6.023 3 1023

molecules/mol) and e is the Napier’s number, the base of natural

logarithm.

Expression Vectors and Site-Directed MutagenesisThe cDNA clone containing the ratSOD2 gene including flanking

regions has been previously described.56 Transformation of bacterial

strains, preparation of plasmids, and other details of DNA recombi-

nant technology were as previously described.57 To engineer a vector

for the heterologous production of recombinant ratSOD2 without

its leader peptide, a DNA segment was amplified by PCR, using the

rat cDNA clone as a template and the following oligonucleotides as

primers: 50d-C63TCCCGGCAT-ATG-CACAGCCT83-30 (forward

primer) and 50d-C709ACCCACCCCGGGCCTGAC691-30 (reverse

primer). Numbering in primers begins from the starting codon of

the complete ratSOD2 gene, including the segment encoding the

leader peptide; italicized letters indicate the base substitutions to

create the NdeI and XmaI cloning sites in direct and reverse primer,

respectively; the new engineered initiation codon is between two

hyphens. The 647 bp amplification product was subcloned in

pGEM T-easy vector and sequenced to confirm the identity with the

ratSOD2 gene reported in databases.29 After digestion of the

recombinant plasmid with NdeI and XmaI, the resulting 627 bp

fragment containing the ratSOD2 gene was cloned into the prokary-

otic expression vector pT7-7, previously digested with the same en-

donucleases. The vector was controlled by restriction analysis and

nucleotide sequencing, and to our surprise it contained the C499AC

codon, predicting the H143 variant of ratSOD2. Therefore, in order

to obtain the Q143 variant of ratSOD2, a site-directed mutagenesis

on the pT7-7 derivative vector was realized.58 In particular, the

following pairs of oligonucleotides were used, in which the base

mismatch introduced to create the desired amino acid replacement

is italicized: 50d-G491CTCTAATCAGGACCCACTG510-30 and 50d-

C510AGTGGGTCCTGATTAGAGC491-30 (forward and reverse

primer for the production of the Q143 variant of ratSOD2). The

PCR reaction, containing the engineered vector as template, the

specific primers and the PfuTurboTM DNA polymerase, was carried

out according to the protocol of the QuickChangeTM Site-Directed

Mutagenesis Kit (Stratagene). The correct substitution, as well as

the absence of other undesired mutations, was confirmed by

sequence analysis. In conclusion, the new engineered vectors, called

vSOD2(Q143) and vSOD2(H143), respectively, predicted the pro-

duction of Q143 and H143 variant of ratSOD2, called SOD2(Q143)

and SOD2(H143), respectively. Using these vectors, recombinant

forms of the mature ratSOD2 were produced, in which the initial

Lys residue of the endogenous mature enzyme (K25 in its premature

form) was replaced by Met; vice versa, the original stop codon of

the ratSOD2 gene was retained.

The mutant forms of ratSOD2, containing the replacements

Y34F and S82A were obtained through a site-directed mutagenesis

of the ratSOD2 gene present in vSOD2(Q143) and vSOD2(H143).

The following pairs of oligonucleotides were used: 50d-A164CGCGACCTTCGTGAACAAT183-3

0 and 50d-A183TTGTTCAC-

GAAGGTCGCGT164-30 (forward and reverse primer for the Y34F

replacement); 5d0-A307CAAACCTGGCCCCTAAGGGT327-30 and

50d-A327CCCTTAGGGGCCAGGTTTGT307-30 (forward and reverse

primer for S82A replacement). Therefore, four new vectors were

obtained, called vSOD2(Y34F-Q), vSOD2(Y34F-H), vSOD2(S82A-

Q) and vSOD2(S82A-H).

Expression and Purification of Recombinant ProteinsThe E. coli BL21(DE3) strain was transformed with the prokary-

otic expression vector vSOD2(Q143) or its mutant forms and

grown at 378C in Luria-Bertani medium containing 0.1 mg/mL

ampicillin in the absence or in the presence of 1 mg/L MnSO4.

When OD600 reached 0.6, 0.4 mM IPTG was added for induction,

and cultures were grown up for additional 3 h. Bacterial cells

were harvested by centrifugation at 5000 rpm for 15 min and pel-

1224 Castellano et al.

Biopolymers

lets were resuspended in buffer A. Cell lysis was obtained by a

cell disruption system (Constant System, UK) at 1.5 kbar; after-

ward, the cell homogenate was incubated at 378C for 30 min in the

presence of 40 U/mL DNAasi A to remove nucleic acids that some-

times form complexes with proteins. Cell homogenate was then

centrifuged at 100,000g for 2 h and supernatant was incubated at

708C for 30 min to denature most of the E. coli host proteins,

removed by centrifugation at 30,000g for 30 min. This purification

strategy, suggested by the high thermal stability of the closely

related human SOD2,17 was validated by the heat stability experi-

ments of ratSOD2 shown in this work. To improve the metal uptake

of all heterologous ratSOD2 forms, the incubation at 708C was car-

ried out in the presence of 10 mM MnCl2. The supernatant was

then extensively dialyzed against buffer B and loaded onto an

Econo-Pac cartridge (5 mL) hydroxyapatite column (Bio-Rad),

connected to a FPLC system (Pharmacia), equilibrated at room

temperature with buffer B. Protein elution was achieved by a linear

50–500 mM potassium phosphate gradient and the identification of

the recombinant product was made by the SOD assay and/or elec-

trophoretic mobility on SDS/PAGE. The recombinant protein was

eluted as a single symmetrical peak at around 200 mM potassium

phosphate; however, an additional purification step was necessary,

and the active fractions were concentrated by Aquacide IIA (Calbio-

chem) and loaded on a Superdex 75 HR10/30 column to remove

minor protein contaminants. Sometimes, during this gel-filtration

experiment we observed an additional peak with an absorbance

maximum at 260 nm; this material was identified as nucleic acids

copurifying with the heterologous product on the hydroxyapatite

column. Fractions containing single protein bands were pooled

together and concentrated. Care was taken to avoid precipitation of

the purified protein, when its concentration was higher than 1 mg/

mL. The concentrated material was then dialyzed against buffer C

and stored at 2208C. Under these conditions, the final yield was

nearly 1 mg from 1 L culture. No significant differences emerged in

the purification procedure among wild type or mutant forms of rat-

SOD2. The presence of MnSO4 during growth of the bacterial cul-

ture did not improve the metal uptake of ratSOD2. Other details of

the purification procedure were as previously described.24

Mass Spectrometry AnalysisThe relative molecular mass (Mr) of wild type and mutant forms of

ratSOD2 was determined by a Q-TOF Micro mass spectrometer

(Waters, Milford, MA), fitted with a Z-spray electrospray ion

source. The capillary source voltage and the cone voltage were set at

3000 and 35 V, respectively. The source temperature was kept at

808C and nitrogen was used as a drying gas (flow rate about 50 L/

h). Before ESI/MS analysis, proteins were desalted by RP-HPLC, as

previously reported.28 Using a mixture of acetonitrile/0.1% formic

acid in water (50/50, v/v), the sample was brought to a concentra-

tion of 1 pmol/lL and infused into the system at a flow rate of

5 lL/min. The acquisition and deconvolution of data were per-

formed by MassLynx software.

Mitochondrial ExtractsThe human neuroblastoma cell line SHSY-5Y was obtained from

American Type Culture Collection. SHSY-5Y cells were grown in

RPMI 1640 medium supplemented with 10% FBS, 2 mM L-gluta-

mine, 100 IU/mL penicillin G and 100 lg/mL streptomicyn in

humidified incubator at 378C under 5% CO2 atmosphere. Cells

were split and seeded in plates (75 cm2) every 3 days and used for

assays during exponential phase of growth.

For the preparation of mitochondrial protein extracts, SHSY-5Y

cells were plated at a density of 2 3 106 cells/plate (75 cm2) and

grown to subconfluency. Cells were then harvested, washed in PBS

buffer and then suspended in buffer M supplemented with a prote-

ase inhibitors cocktail, and homogenized. The homogenate was cen-

trifuged at 800g for 10 min at 48C and the supernatant was than

centrifuged at 12,000g for 30 min at 48C. The resulting pellet (mito-

chondrial fraction) was suspended in buffer M, whereas the final

supernatant represented the cytosolic fraction.

REFERENCES1. Finkel, T. Curr Opin Cell Biol 2003, 15, 247–254.

2. Adler, V.; Yin, Z.; Tew, K. D.; Ronai, Z. Oncogene 1999, 18,

6104–6111.

3. Sauer, H.; Wartenberg, M.; Hescheler, J. Cell Physiol Biochem

2001, 11, 173–186.

4. McCord, J. M.; Boyle, J. A.; Day, E. D., Jr.; Rizzo, L. J.; Salin, M.

L. In Superoxide and Superoxide dismutases; Michelson, A. M.;

McCord, J. M.; Fridovich, I., Eds.; Academic Press: London,

1977, pp 129–138.

5. Fridovich, I. J Biol Chem 1989, 264, 7761–7764.

6. Bannister, J. V.; Bannister, W. H.; Rotilio, G. CRC Crit Rev

Biochem 1987, 22, 111–180.

7. Zelko, I. N.; Mariani, T. J.; Folz, R. J. Free Radic Biol Med 2002,

33, 337–349.

8. Wispe, J. R.; Warner, B. B.; Clark, J. C.; Dey, C. R.; Neuman, J.;

Glasser, S. W.; Crapo, J. D.; Chang, L. Y.; Whitsett, J. A. J Biol

Chem 1992, 267, 23937–23941.

9. Hirose, K.; Longo, D. L.; Oppenheim, J. J.; Matsushima, K.

FASEB J 1993, 7, 361–368.

10. Li, Y.; Huang, T. T.; Carlson, E. J.; Melov, S.; Ursell, P. C.; Olson,

J. L.; Noble, L. J.; Yoshimura, M. P.; Berger, C.; Chan, P. H.;

Wallace, D. C.; Epstein, C. J. Nat Genet 1995, 11, 376–381.

11. MacMillan-Crow, L. A.; Cruthirds, D. L. Free Radic Res 2001,

34, 325–336.

12. Vincent, A. M.; Russell, J. W.; Sullivan, K. A.; Backus, C.; Hayes,

J. M.; McLean, L. L.; Feldman, E. L. Exp Neurol 2007, 208, 216–

227.

13. Lebovitz, R. M.; Zhang, H.; Vogel, H.; Cartwright, J., Jr.;

Dionne, L.; Lu, N.; Huang, S.; Matzuk, M. M. Proc Natl Acad

Sci USA 1996, 93, 9782–9787.

14. Melov, S.; Coskun, P.; Patel, M.; Tuinstra, R.; Cottrell, B.; Jun,

A. S.; Zastawny, T. H.; Dizdaroglu, M.; Goodman, S. I.; Huang,

T. T.; Miziorko, H.; Epstein, C. J.; Wallace, D. C. Proc Natl Acad

Sci USA 1999, 96, 846–851.

15. Wispe, J. R.; Clark, J. C.; Burhans, M. S.; Kropp, K. E.;

Korfhagen, T. R.; Whitsett, J. A. Biochim Biophys Acta 1989,

994, 30–36.

16. Borgstahl, G. E.; Parge, H. E.; Hickey, M. J.; Beyer, W. F., Jr.;

Hallewell, R. A.; Tainer, J. A. Cell 1992, 71, 107–118.

17. Guan, Y.; Hickey, M. J.; Borgstahl, G. E.; Hallewell, R. A.;

Lepock, J. R.; O’Connor, D.; Hsieh, Y.; Nick, H. S.; Silverman,

D. N.; Tainer, J. A. Biochemistry 1998, 37, 4722–4730.

Rat Mitochondrial Key Residues in Superoxide Dismutase 1225

Biopolymers

18. MacMillan-Crow, L. A.; Crow, J. P.; Kerby, J. D.; Beckman, J. S.;

Thompson, J. A. Proc Natl Acad Sci USA 1996, 93, 11853–

11858.

19. MacMillan-Crow, L. A.; Crow, J. P.; Thompson, J. A. Biochemis-

try 1998, 37, 1613–1622.

20. Yamakura, F.; Taka, H.; Fujimura, T.; Murayama, K. J Biol

Chem 1998, 273, 14085–14089.

21. Sorkin, D. L.; Duong, D. K.; Miller, A. F. Biochemistry 1997, 36,

8202–8208.

22. Soulere, L.; Claparols, C.; Perie, J.; Hoffmann, P. Biochem J

2001, 360, 563–567.

23. Ursby, T.; Adinolfi, B. S.; Al-Karadaghi, S.; De Vendittis, E.; Boc-

chini, V. J Mol Biol 1999, 286, 189–205.

24. De Vendittis, E.; Ursby, T.; Rullo, R.; Gogliettino, M. A.;

Masullo, M.; Bocchini, V. Eur J Biochem 2001, 268, 1794–1801.

25. Castellano, I.; Di Maro, A.; Ruocco, M. R.; Chambery, A.;

Parente, A.; Di Martino, M. T.; Parlato, G.; Masullo, M.; De

Vendittis, E. Biochimie 2006, 88, 1377–1389.

26. Matsuda, Y.; Higashiyama, S.; Kijima, Y.; Suzuki, K.; Kawano,

K.; Akiyama, M.; Kawata, S.; Tarui, S.; Deutsch, H. F.;

Taniguchi, N. Eur J Biochem 1990, 194, 713–720.

27. Hernandez-Saavedra, D.; McCord, J. M. Cancer Res 2003, 63,

159–163.

28. Castellano, I.; Ruocco, M. R.; Cecere, F.; Di Maro, A.;

Chambery, A.; Michniewicz, A.; Parlato, G.; Masullo, M.; De

Vendittis, E. Biochim Biophys Acta 2008, 1784, 816–826.

29. Ho, Y. S.; Crapo, D. J. Nucleic Acids Res 1987, 15, 10070.

30. Ho, Y. S.; Howard, A. J.; Crapo, D. J. Am J Respir Cell Mol Biol

1991, 4, 278–286.

31. Hurt, J.; Hsu, J. L.; Dougal, W. C.; Visner, G. A.; Burr, I. M.;

Nick, H. S. Nucleic Acids Res 1992, 20, 2985–2990.

32. Gogliettino, M. A.; Tanfani, F.; Scire, A.; Ursby, T.; Adinolfi, B.

S.; Cacciamani, T.; De Vendittis, E. Biochemistry 2004, 43,

2199–2208.

33. Hopper, R. K.; Carroll, S.; Aponte, A. M.; Johnson, D. T.;

French, S.; Shen, R. F.; Witzmann, F. A.; Harris, R. A.; Balaban,

R. S. Biochemistry 2006, 45, 2524–2536.

34. Wintjens, R.; Gilis, D.; Rooman, M. Proteins 2008, 70, 1564–

1577.

35. Dello Russo, A.; Rullo, R.; Nitti, G.; Masullo, M.; Bocchini, V.

Biochim Biophys Acta 1997, 1343, 23–30.

36. Leveque, V. J. P.; Stroupe, M. E.; Lepock, J. R.; Cabelli, D. E.;

Tainer, J. A.; Nick, H. S.; Silverman, D. N. Biochemistry 2000,

39, 7131–7137.

37. MacMillan-Crow, L. A.; Thompson, J. A. Arch Biochem Bio-

phys 1999, 366, 82–88.

38. Masip, L.; Veeravalli, K.; Georgiou, G. Antioxid Redox Signal

2006, 8, 753–762.

39. Hondorp, E. R.; Matthews, R. G. PLoS Biol 2004, 2, 1738–1753.

40. Rahman, I.; Biswas, S. K.; Jimenez, L. A.; Torres, M.; Forman,

H. J. Antioxid Redox Signal 2005, 7, 42–59.

41. Biswas, S.; Chida, A. S.; Rahman, I. Biochem Pharmacol 2006,

71, 551–564.

42. Wang, J.; Tekle, E.; Oubrahim, H.; Mieyal, J. J.; Stadtman, E. R.;

Chock, P. B. Proc Natl Acad Sci USA 2003, 100, 5103–5106.

43. Pan, S.; Berk, B. C. Circ Res 2007, 100, 213–219.

44. Adachi, T.; Weisbrod, R. M.; Pimentel, D. R.; Ying, J.; Sharov, V.

S.; Schoneich, C.; Cohen, R. A. Nat Med 2004, 10, 1200–1207.

45. Chen, C. L.; Zhang, L.; Yeh, A.; Chen, C. A.; Green-Church, K.

B.; Zweier, J. L.; Chen, Y. R. Biochemistry 2007, 46, 5754–5765.

46. Chen, Y. R.; Chen, C. L.; Pfeiffer, D. R.; Zweier, J. L. J Biol Chem

2007, 282, 32640–32654.

47. Schinina, M. E.; Carlini, P.; Polticelli, F.; Zappacosta, F.; Bossa,

F.; Calabrese, L. Eur J Biochem 1996, 237, 433–439.

48. Beer, S. M.; Taylor, E. R.; Brown, S. E.; Dahm, C. C.; Costa, N.

J.; Runswick, M. J.; Murphy, M. P. J Biol Chem 2004, 279,

47939–47951.

49. Hurd, T. R.; Costa, N. J.; Dahm, C. C.; Beer, S. M.; Brown, S. E.;

Filipovska, A.; Murphy, M. P. Antioxid Redox Signal 2005, 7,

999–1010.

50. Borgstahl, G. E.; Parge, H. E.; Hickey, M. J.; Johnson, M. J.;

Boissinot, M.; Hallewell, R. A.; Lepock, J. R.; Cabelli, D. E.;

Tainer, J. A. Biochemistry 1996, 35, 4287–4297.

51. Archambaud, C.; Nahori, M. A.; Pizarro-Cerda, J.; Cossart, P.;

Dussurget, O. J Biol Chem 2006, 281, 31812–31822.

52. Voisin, S.; Watson, D. C.; Tessier, L.; Ding, W.; Foote, S.; Bhatia,

S.; Kelly, J. F.; Young, N. M. Proteomics 2007, 7, 4338–4348.

53. McCord, J. M.; Fridovich, I. J Biol Chem 1969, 244, 6049–6055.

54. Bradford, M. Anal Biochem 1976, 72, 248–254.

55. Laemmli, U. K. Nature 1970, 227, 680–685.

56. Ginsberg, M. D.; Feliciello, A.; Jones, J. K.; Avvedimento, V. A.;

Gottesman, M. E. J Mol Biol 2003, 327, 885–897.

57. Maniatis, T.; Fritsch, E. F.; Sambrook, J. Molecular Cloning;

Cold Spring Harbor: New York, 1982.

58. Braman, J.; Papworth, C.; Greener, A. Methods Mol Biol 1996,

57, 31–44.

Reviewing Editor: Laurence Nafie

1226 Castellano et al.

Biopolymers