http://www.elsevier.com/locate/bba

Biochimica et Biophysica A

Review

Direct oxidative DNA damage, apoptosis and radio sensitivity by spermine

oxidase activities in mouse neuroblastoma cells

R. Amendolaa,T, A. Bellinia,b, M. Cervellib, P. Deganc, L. Marcoccid, F. Martinie, P. Mariottinib

aIstituto per la Radioprotezione, ENEA, CR Casaccia, Via Anguillarese 301, 00060 Roma, ItalybDipartimento di Biologia, Universita bRoma Tre Q, 00146 Roma, Italy

cIstituto Nazionale per la Ricerca sul Cancro (IST - IRCCS), 16132 Genova, ItalydDipartimento di Scienze Biochimiche bA. Rossi Fanelli Q, Universita bLa Sapienza Q, 00185 Roma, Italy

eIstituto Nazionale per le Malattie Infettive I.R.C.C.S. bL. Spallanzani Q, 00149 Roma, Italy

Received 31 August 2004; received in revised form 27 January 2005; accepted 16 February 2005

Available online 5 March 2005

Abstract

In mammals, the polyamines affect cell growth, differentiation, and apoptosis; their levels are increased in malignant and proliferating

cells, thus justifying an interest in a chemotherapeutic approach to cancer. The flavoprotein SMO is the most recently characterized catabolic

enzyme, preferentially oxidizing SPM to SPD, 3-aminopropanal and H2O2. In this report, we describe a novel functional characterization of

the recently cloned splice variant isoforms from mouse brain, encoding, among others, the nuclear co-localized spermine oxidase mSMOA.The over-expression of the active isoforms mSMOa and mSMOA, and the inactive mSMOy and mSMOg in mouse neuroblastoma cells,

demonstrated the first evidence of the direct oxidative DNA damage by the SMO activities, either alone or, in a higher extent, when

associated with radiation exposure, thus working as radio sensitizer. These effects were reverted by treatment with 50 AM and 100 AM doses

of the inhibitor of SMO activity MDL 72,527. The over-expression of all SMO isoforms failed to influence the expression of the regulating

enzymes of polyamines metabolism ODC and SSAT. Dealing with the unbalanced tissue specific SMO activities, these results could indicate

a new direction to tailor chemotherapy-associated radiotherapy, improving dose-rate protocol and allowing the modulation of deleterious side

effects on healthy tissues.

D 2005 Elsevier B.V. All rights reserved.

Keywords: Apoptosis; DNA damage; Neuroblastoma; Oxidative stress; Polyamine; Radiation

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

2. Materials and methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

2.1. Cell culture and radiation exposure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

2.2. RT-PCR analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

2.3. Spermine oxidase activity and polyamines content . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

2.4. Fraction of surviving and apoptotic cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

2.5. Flow cytometric (FCM) determination of DNA damage and apoptosis . . . . . . . . . . . . . . . . . . . . . . . . . . 17

2.6. Determination of 8-oxo-7,8-dihydroguanine (8-oxo-G) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

2.7. Statistical analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

0304-419X/$ - see front matter D 2005 Elsevier B.V. All rights reserved.

doi:10.1016/j.bbcan.2005.02.002

Abbreviations: MDL, MDL 72,527 (N,N9-bis[2,3]-1,4-butanediamine); ODC, ornithine decarboxylase; 8-oxo-G, 8-oxo-7,8-dihydroguanine; PUT

putrescine; SMO, spermine oxidase; mSMO, mouse SMO; SPD, spermidine; SPM, spermine; SSAT, spermidine/spermineN1-acetyltransferase

T Corresponding author. Tel.: +39 6 30486115; fax: +39 6 30483644.

E-mail address: amendola@casaccia.enea.it (R. Amendola).

,

cta 1755 (2005) 15–24

R. Amendola et al. / Biochimica et Biophysica Acta 1755 (2005) 15–2416

3. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

3.1. RNA expression of mSMO isoforms, mODC, mSSAT, and mPAO. . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

3.2. SMO activity in nucleus and polyamines content . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

3.3. mSMO activity and cell death, apoptosis and radiosensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

3.4. mSMO activity and oxidative damage to DNA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

4. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

1. Introduction

The polyamines putrescine (PUT), spermidine (SPD) and

spermine (SPM) are ubiquitous aliphatic polyamines,

carrying two, three, and four positive charges at physio-

logical pH, respectively, able to form an electrostatic bond

with negatively charged cellular macromolecules [1]. The

level of polyamines is strictly controlled by biosynthesis and

catabolism regulative patterns. Polyamine biosynthesis is

regulated by two key enzymes, ornithinedecarboxylase

(ODC) and S-adenosylmethionine decarboxylase (Ado-

MetDC). The polyamine level is responsible for feedback

homeostasis, as polyamines themselves act as down

regulators of both enzymes, and as up regulators of the

antizyme protein, an inhibitor of ODC and cellular poly-

amines uptake. Accordingly, in transgenic mice over-

expressing ODC, AdoMetDC, and spermidine synthase

separately, SPD and SPM were slightly increased [2,3].

The polyamine catabolism is a recycling pathway that

converts SPD and SPM back to PUT, via SPD/SPM N1-

acetyltransferase (SSAT), polyamines oxidase (PAO), and a

terminal catabolic pathway cupper containing diamino

oxidase (DAO) [4,5]. The flavoprotein SMO is the most

recent characterized catabolic enzyme, preferentially oxi-

dizing SPM to SPD, 3-aminopropanal and H2O2 [6–8]. The

degradation of polyamines depends mainly by SSAT

activity. In fact, in transgenic mice over-expressing both

SSAT and ODC, the polyamine content resembled only

SSAT overproduction [9]. In mammals, the polyamines

directly affect cell growth, differentiation, and apoptosis.

Polyamine depletion provokes an impaired synthesis of

DNA, proteins and DNA sensitivity to nuclease [10], as well

as an alteration of the p53/p21/p27 cell cycle regulatory

pathway [11,12]. Apoptosis is, in turn, enhanced or

negatively modulated [13,14]. Polyamine levels are

increased in malignant and proliferating cells, thus justifying

an interest in a chemotherapeutic approach to cancer. The

use of the 2-difluoro-methylornithine (DFMO) to inhibit

ODC, and the alteration of cellular content by polyamine

analogs play, however, a controversial role as inducers of

cell death by blocking the cell cycle, or, on the contrary, as

apoptotic protectors [14]. In the present work, we describe

the first evidence of direct oxidative DNA damage mediated

by both novel nuclear co-localized mSMOA and the

cytoplasmic mSMOa isoforms [15]. The mSMO activity

enhances oxidative DNA damage, either alone or in

association with radiation exposure. The over-expression

of nonactive isoforms, lacking some of the crucial FAD

binding regions, and the dose response reversion of cell

death mediated by the flavoprotein inhibitor MDL 72,527

(N,N9-bis[2,3-butadienyl]-1,4-butanediamine) (MDL) [16]

are indicative of a selective induction of DNA oxidative

stress by mSMO activities.

2. Materials and methods

All reagents were from Sigma-Aldrich (Sigma-Aldrich,

St. Louis, MO), unless otherwise specified. Taq polymer-

ase and M-MLV Reverse Transcriptase enzymes were

from Promega (Promega Corp., Madison, WI). MDL was

a gift from Hoechst Marion Roussel Inc. (Cincinnati,

OH). Plasticwares were from Nunc (Nunc A/S, Roskilde,

Denmark).

2.1. Cell culture and radiation exposure

The growth conditions of untransfected and transfected

with the pcDNA3/V5-His (Invitrogen Ltd, Paisley, Scot-

land, UK), pcDNA3/mSMOa/V5-His, pcDNA3/mSMOA/V5-His, pcDNA3/mSMOy/V5-His, and pcDNA3/mSMOg/

V5-His murine neuroblastoma (NB) N18TG2 cell line were

described elsewhere [15]. All experiments were performed

using a pool isolated from three separate transfections,

stably maintained in culture in the presence of 300 AMgeneticin. MDL was administered at 50 AM and 100 AMrespectively, for 24 h before experiments. The X-irradiation

of 2 and 4 Gy was delivered by a Gilardoni CHF 320 G Unit

(Gilardoni S.p.A., Mandello L., Italy) tested complying EU

standards by the manufacturer. Dose/rate was 0.99 Gy

min�1 at 250 KeV, with 0.5 mm Cu filter. Cells were

irradiated on ice, and fresh medium was replaced after

exposure. Control cells were treated similarly, without

irradiation. All experimental points were taken 6 h after

irradiation.

2.2. RT-PCR analysis

Total RNA was isolated by GeneElute system (Sigma),

and retro transcripted in cDNA by SuperScript First-Strand

R. Amendola et al. / Biochimica et Biophysica Acta 1755 (2005) 15–24 17

Synthesis System (Invitrogen), according to the manu-

facturer’s instructions. The mSSAT, mPAO, mSMO iso-

forms, and h-actin specific primer-pairs are described

elsewhere [15]. The murine ODC specific primer-pairs

are: 5V-TCCAGGTTCCCTGTAAGCAC-3V forward pri-

mer, and 5V-CCAACTTTGCCTTTGGATGT-3V reverse

primer (region between nt 708 and 728 and nt 1191 and

1211, respectively, of the murine ODC cDNA — GenBank

accession numberNM_013614). PCR samples were taken

after an increasing number of cycles to provide evidences of

linearity of amplification [15]. h-actin expression, as the

control housekeeping gene, and samples without RT enzyme

were prepared in parallel. Three separate experiments were

performed from each RNA preparation. Gel images were

taken with automatic exposure by Diana III dedicated

device (Raytest Italia, Cinisello Balsamo, Italy).

2.3. Spermine oxidase activity and polyamines content

At least 107 cells for each experimental point were

detached from plastic, and aliquots of cellular pellets were

flash-frozen in liquid nitrogen until analyses, or nuclear sub-

fractionated, in hypotonic extraction buffer, at 4 8C to avoid

polyamines dispersion [17]. Briefly, cells were Dounce

homogenized in 250 mM Sucrose, 2% Triton X-100, 2 mM

EDTA, and 20 mM Tris (pH 7.5) at 4 8C. Nuclei were

separated after dual sedimentation (10 s at 12,000�g, 4 8C),and 0.3 M perchloric acid was added before analysis. Nuclei

enrichment and integrities were checked by fluorometrical

DNA determination (DynaQuant 200, Hoefer, S. Francisco,

CA) versus protein amount of extracts in respect of standard

concentration of bovine albumin [18], and by phase contrast

microscopical observation (40� objective, Zeiss Axioskop,

Carl Zeiss, Milano, Italy). The enzyme activity of the cellular

and nuclear extracts was determined fluorometrically by

measuring the pmol H2O2 produced/min/mg protein upon the

oxidation of substrates as described elsewhere [5,19]. Poly-

amine concentration was determined as described elsewhere

[20] and expressed as nmol/mg protein for each sample, in

respect of standard concentration of bovine albumin.

2.4. Fraction of surviving and apoptotic cells

Cells were plated at 104 cells/well on four wells chamber

slide. 24 h later, randomly chosen chamber slides were

treated with MDL for 24 h. Cells analyzed for surviving

fraction were detached, gathered with floating cells, and

treated with 0.5% Trypan Bleu in PBS. Viable cells were

discriminated for dye exclusion in a Bqrker hematocytom-

eter. Cells analyzed for apoptosis were fixed in 3.7%

paraformaldehyde in PBS, 15 min at 48 C, and stained by

DAPI. At least 400 cells for each slide were screened for the

presence of apoptotic nuclei by a Zeiss Axioskop micro-

scope. Results derived from the series of the two independ-

ent pools were analyzed, representing three and six

replicated experiments.

2.5. Flow cytometric (FCM) determination of DNA damage

and apoptosis

To study DNA content, cells were treated with Pro-

pidium Iodide (PI), as described [21]. At least 2�105

cells were analyzed by a FACSCalibur flow cytometer

(Becton Dickinson, San Jose, CA), previously calibrated

by CaliBRITE 3 beads (Becton Dickinson), with laser

excitation set at 488 nm, and a 630-nm emission filter to

detect red fluorescence. To study apoptosis, cells were

treated to evidence the fluorescence of 3Vterminal deoxy-

transferase (TdT) by TUNEL method [22], following the

manufacturer’s instruction (In Situ Cell Death Fluorescent

Kit, Roche Diagnostic S.p.A., Monza, Italy). Laser

excitation was set at 488 nm, and emission was at

550 nm for FITC. As an auto-fluorescence control, a

sample treated with label solution but without TdT was

carried out for each set of analyses. Data acquisition and

analysis were performed by CellQuest software (Becton

Dickinson).

2.6. Determination of 8-oxo-7,8-dihydroguanine (8-oxo-G)

DNA was extracted from 3�106 exponentially growing

cells by a high-salt protein precipitation method. The

enzymatic digestion of the DNA was performed at 37 8Cwith nuclease P1 (Boehringer Mannheim S.p.A., Monza,

Italy) for 2 h and alkaline phosphates (Boehringer

Mannheim) for 1 h. The deoxyribonucleosides were

purified on a 30,000 Da cut-off UltraFree Millipore

Filtration system (Millipore, Billerica, MA) and separated

on LC-18-S column (Supelco, 15034.6 mm, Sigma)

equipped with an LC-18 guard column cartridge [23].

UV detection was at 254 nm and electrochemical analysis

was carried out by a ESA Coulochem II Electrochemical

Detector (ESA Inc., Chelmsford, MA, USA). Results are

expressed as number of 8-oxo-G residues/106 guanine

residues.

2.7. Statistical analysis

Data, presented as meanF95% confidential intervals

(95% CI), were analyzed using SPSS-11 statistical package

(SPSS Inc., Chicago, ILL); 95% CI of the differences of the

means not containing zero value denoted statistical signifi-

cance difference at PN0.05. All experiments were performed

in triplicate, unless otherwise indicated.

3. Results

3.1. RNA expression of mSMO isoforms, mODC, mSSAT,

and mPAO

In a recent work, SMO RNA has been described as

transcriptional activated after a bacterial stress [24]. To rule

R. Amendola et al. / Biochimica et Biophysica Acta 1755 (2005) 15–2418

out the possibility of an unspecific stress activation after

X-irradiation, we analyzed mSMOa and mSMOA RNA

level 6 h later at 2 and 4 Gy exposures in NB cells. Since

both endogenous enzymes were almost undetectable in our

cell model [15], we performed 25, 30, and 35 PCR cycles,

to provide a linearity range of amplification. In Fig. 1,

panel A, mSMOa and mSMOA were recognizable at 35

cycles showing no detectable differences between treat-

ments. The ectopical expression of different SMO isoforms

was studied for the effect on the transcription levels of

regulating enzymes. As shown in Fig. 1, panel B, the

expression of the active isoforms mSMOa and mSMOA,as well as of the inactive isoforms mSMOy and mSMOg,

do not alter the RNA expression of the regulating

biosynthetic enzyme mODC, of the regulating catabolic

enzyme mSSAT, and of mPAO, both in normal conditions

and after exposure to 2 or 4 Gy X-irradiation. As above,

an increasing number of PCR cycles have been performed

to provide linearity range of amplification. Our data are in

agreement with previous work, where ODC mRNA has

been found induced early after irradiation, and restored to

physiological level after 6 h [25]. In regard to SSAT

mRNA expression, following radiation, our data are

consistent with previous observations on SSAT promoter

induction 24 h later exposure [26].

Fig. 1. RT-PCR analyses. A representative RT-PCR experiments from three

independent replicas are shown. Samples from PCR reactions were taken at

increasing number of cycles to avoid signals saturation (as indicated on the

right side of the gel). Panel A: endogenous mSMOa and mSMOA were

evaluated after increasing doses of irradiation (0, 2, 4 Gy), in parental and

mock transfected cell lines. Abbreviations: M, fX174-HaeIII digested

DNA size marker; N, not transfected cells; P, mock transfected cells; +,

positive control (Ta-TA, transfected cells with pcDNA3/mSMOa, A, /V5-His plasmids); C, negative control without RT enzyme. Panel B: RNA

accumulation of mODC, mSSAT, mPAO in NB cells with increasing doses

of irradiation (0, 2, 4 Gy). Abbreviations: Ty-Tg, transfected cells with

pcDNA3/mSMOy, g, /V5-His plasmids; others as in panel A.

Fig. 2. mSMO activity. Results of mSMO activity are given as pmol H2O2

produced/min/mg protein, and as F95% CI on three replicas. Panel A:

mSMO activity evaluated after increasing doses of irradiation (0 Gy, white

bar; 2 Gy, grey bar; 4 Gy, black bar), from cellular extracts. Panel B:

mSMO activity evaluated from cellular and nuclear extracts (as indicated)

from parental and transfected cells. Symbols: closed square (n), parental

N18TG2 cells; closed circle (.), mock transfected cells; open square (5),

mSMOa transfected cells; open circle (o), mSMOA transfected cells.

Abbreviations: N.I., no inhibitor; 50 AM MDL, 100 AM MDL, concen-

tration of the MDL inhibitor treatment. Black arrowheads indicate statisti-

cally significant differences of the means at Pb0.05 between controls and

SMO transfected cell lines. White arrowhead indicates statistical significant

differences of the means at Pb0.05 between SMOa and SMOA transfected

cell lines.

3.2. SMO activity in nucleus and polyamines content

As mentioned above, SMO enzymatic activity has been

described to be activated after a bacterial stress [24]. To rule

out the possibility of a further unspecific post-transcriptional

stress activation, we analyzed mSMO activity 6 h later to 2

and 4 Gy exposures in NB cells. In Fig. 2, panel A, triplicate

experiments demonstrate any statistically significant aug-

mented level of activity. The spermine oxidase activity was

monitored in cellular and nuclear extracts of N18TG2 cell

line, and empty vector, mSMOa, and mSMOA transfected

cells, in the absence or presence of, respectively, 50 and 100

Fig. 3. Polyamines content. Polyamines content in cellular and nuclear

extracts (as indicated) from transfected cells. Results are given as nmol/mg

protein, and as F95% CI of three replicas. Symbols: closed square (n),

parental N18TG2 cells; closed circle (.), mock transfected cells; open

square (5), mSMOa transfected cells; open circle (o), mSMOA transfected

cells. Abbreviations: PUT, putrescine; SPD, spermidine; SPM, spermine.

Black arrowheads indicate statistically significant differences between

means at Pb0.05 of controls and SMO transfected cell lines. White

arrowheads indicates statistically significant differences of the means at

Pb0.05 between SMOa and SMOA transfected cell lines.

R. Amendola et al. / Biochimica et Biophysica Acta 1755 (2005) 15–24 19

AM MDL. As shown in Fig. 2, panel B, the empty vector

did not confer any difference from the parental cell line in

spermine oxidase activity. In the cellular extracts, a statisti-

cally significant increase of spermine oxidase was found

both in mSMOa and mSMOA transfected cell lines. In the

nuclear extracts, the nuclear co-localized isoform mSMOAproduced a higher activity, significantly different from the

cytoplasmic mSMOa isoform. A 50 AM dose of MDL

decreased the activity of all samples; only the mSMOA was

still significantly higher then the N18TG2 parental cells.

Moreover, a 100 AM MDL dose inhibited the activity in all

cell lines tested. Our results confirm the MDL inhibition of

Fig. 4. Over-expression of mSMO isoforms and radiation in NB cells. Trypan bleu

transfected N18TG2 cells, in the absence (0 Gy) and after 2 Gy of X-irradiation (as

square (n), parental N18TG2 cells; closed circle (.), mock transfected cells; open

cells; closed triangle (E), mSMOy transfected cells; open triangle (4), mSMOg

the means at Pb0.05 between controls and SMO transfected cell lines.

mSMOa and mSMOA purified enzymes in in vitro experi-

ments [27]. In Fig. 3, the polyamine content/mg of protein

levels is shown. Accordingly to the expected enzymes

activity, the cellular extracts of mSMOa and mSMOAtransfected cell lines showed a statistically significant

decrease in the SPM level, and an increase in PUT level.

In the nuclear extracts, only mSMOA caused a statistically

significant decrease in SPM level, thus confirming its

nuclear co-localization, while PUT level was statistically

higher for both active isoforms. The results related to

cellular extracts are in agreement with previously published

data by describing polyamines content in HEK293 cells

over-expressing human SMO [6]. Taken together, these

results show the first evidence of a nuclear mammalian

spermine oxidase activity.

3.3. mSMO activity and cell death, apoptosis and

radiosensitivity

In a preliminary experiment, the level of cell death and

apoptosis induced by the four mSMO splice variant

isoforms (mSMOa, mSMOy, mSMOg, and the active

mSMOA [15]) in transfected N18TG2 cells were inves-

tigated. The presence of empty vector and of the inactive

isoforms did not alter the surviving and apoptotic fractions

of parental cell line, both in the presence or absence of

irradiation. On the contrary, only the active mSMOa and the

mSMOA isoforms affected cell survival, enhancing the

adverse effects of an acute dose of 2 Gy X-irradiation. In the

absence of irradiation (panel B, first column), mSMOAshowed a slight statistical significant augmented level of

apoptotic cells (Fig. 4).

All four isoforms did not alter the cell proliferation and

differentiation when maintained in culture (data not shown).

The N18TG2 cell line, empty vector, mSMOa and mSMOAtransfected cell lines were then irradiated with 2 and 4 Gy

X-ray doses, in the absence or in the presence of 50 and 100

exclusion assay (panel A) and apoptotic bodies (panel B) determination of

indicated). Results are given asF95% CI of three replicas. Symbols: closed

square (5), mSMOa transfected cells; open circle (o), mSMOA transfected

transfected cells. Arrowheads indicate statistically significant differences of

R. Amendola et al. / Biochimica et Biophysica Acta 1755 (2005) 15–2420

AMMDL. In Fig. 5, cell death ratios determined for 6 h after

exposure are shown as dead cells/live cells ratio (panel A)

and percentage of apoptotic bodies vs. normal nuclei (panel

B). While the empty vector did not alter the survival and the

apoptotic fraction of parental cell line. Before irradiation,

both active isoforms show a level of dying cells that is not

statistically significant (panel A, first column). A dose

response adverse effect was registered for MDL treatments,

but independent of any transfection, probably caused by the

lysosomotropic effects of the inhibitor [28]. In the presence

of 2 Gy X-irradiation, the expressions of both mSMOa and

mSMOA were detrimental to cell survival in a statistically

significant way (panel A, fourth column), while a 4 Gy dose

provoked a not statistically significant augmented mortality

(panel A, seventh column). MDL administration showed a

dose response effect in repressing differences, accordingly,

in a less extent for 50 AM and 2 Gy X-irradiation (panel A,

fifth column). As for the counts of apoptotic bodies, both

active isoforms induced apoptosis in a statistically signifi-

cant way in the absence of irradiation (panel B, first

column). At 2 Gy X-irradiation, mSMOa and mSMOAstrongly enhanced the increase of apoptotic bodies (panel B,

fourth column). Accordingly to the trypan blue exclusion

determination, also the 4 Gy dose showed a not statistically

significant augmented mortality (panel B, seventh column).

At the 4 Gy dose, MDL treatments confirmed not statisti-

cally significant repression differences among samples

(panel B, eighth and ninth columns).

In Fig. 6, panel A, the FCM determination of ipoG0/G1

cell fraction, a key event of necrosis and apoptosis [29], is

summarized as induction by the two mSMO splice variant

isoforms. The empty vector-transfected cell line produced

histograms overlapping those from the N18TG2 parental

cell line, and thus has been omitted in the merged

histograms. In the untreated samples, any difference has

been detected from the cell lines tested (Fig. 6, panel A, top

left). The 2 Gy irradiation caused an increased number of

ipoG0/G1 cells in the presence of both active isoforms (Fig.

Fig. 5. Over-expression of the active mSMO isoforms, radiation, and MDL treatme

(panel B) determination of transfected N18TG2 cells, in the absence (0 Gy) and aft

of six replicas. Symbols: closed square (n), parental N18TG2 cells; closed circle (.circle (o), mSMOA transfected cells. Abbreviations: N.I., no inhibitor; 50 and

arrowheads indicate statistical significant differences of the means at Pb0.05 betw

6, panel A, top right). The 4 Gy irradiation showed an

increased number of ipoG0/G1 cells in all the four mSMO

splice variant isoforms transfected cell lines, but with no

difference between groups (data not shown). The 50 AMMDL administration inhibited partially the mSMO-medi-

ated increase of ipoG0/G1 cells (Fig. 6, panel A, bottom

left) and was totally inhibitory at the 100 AM concentration

(Fig. 6, panel A, bottom right). The TUNEL technique is

considered more sensitive and apoptosis-related then ipoG0/

G1 determination, suitable to detect cells in any phase of the

cell cycle, and able to recognize DNA strand breaks beside

fragmentation [30]. The TUNEL test results are shown in

Fig. 6, panel B, where, as above, data coming from empty

vector-transfected cells were omitted, since they did not

show any difference from parental cells. A higher number of

apoptotic cells were seen in the mSMOa and mSMOAtransfected cell lines (Fig. 6, panel B, left). Consistent with

the FCM determination, these differences were statistically

significant also at 2 Gy of X-irradiation (Fig. 6, panel B,

middle). At the lower concentration of 50 AM, the MDL

abolishes difference, giving an evidence of the role played

by the spermine oxidase activity in mediating apoptosis

(Fig. 6, panel B, right). As expected, the 4 Gy X-irradiation

caused more apoptotic cells in all samples, thus covering the

spermine oxidase related radiosensitivity shown at a lower

irradiation dose.

3.4. mSMO activity and oxidative damage to DNA

Guanine is the preferential target of the DNA oxidative

damage mediated by direct one-electron oxidizing agent or

indirect Fenton type hydrogen peroxide reaction [31,32]. In

Fig. 7, 8-oxo-G residues levels are shown for cells trans-

fected with the two mSMO splice variant isoforms not

irradiated and irradiated. Among the samples that were not

irradiated, cell lines over-expressing the two active isoforms

presented a statistically significant higher number of 8-oxo-

G residues, almost 3 and 1.5 times in respect to the absence

nts in NB cells. Trypan bleu exclusion assay (panel A) and apoptotic bodies

er 2 and 4 Gy of X-irradiation (as indicated). Results are given as F95% CI

), mock transfected cells; open square (5), mSMOa transfected cells; open

100, respectively 50 AM and 100 AM MDL inhibitor treatment. Black

een controls and SMO transfected cell lines.

Fig. 6. FCM and TUNEL analyses. Mock transfected cells did not differ from N18TG2 and were omitted for sake of clarity. Panel A: FCM representative

merged histograms out of three experiments of the ipoG0/G1 cell fraction (as indicated): number of events (Counts) per relative amount of Propidium Iodide.

Symbols: grey closed histogram, N18TG2 cells; black line, mSMOa transfected cells; dotted line, mSMOA transfected cells. Panel B: TUNEL representative

merged histograms out of three experiments of the positive cell fraction (as indicated, M1 region); number of events (Counts) per relative amount of FITC.

Symbols: black closed histogram, auto-fluorescence; grey closed histogram, N18TG2 cells; black line, mSMOa transfected cells; dotted line, mSMOAtransfected cells. Abbreviations: 0 GY-N.I., untreated cells; 2 Gy-N.I., 2 Gy irradiation, no inhibitor; 2 Gy-50 AMMDL, 2 Gy irradiation with a 50 AMMDL; 2

Gy-100 AM MDL, 2 Gy irradiation with a 100 AM MDL.

R. Amendola et al. / Biochimica et Biophysica Acta 1755 (2005) 15–24 21

or presence of the increasing inhibitor doses (Fig. 7, 0 Gy

columns). Interestingly, the inhibitor caused an increment of

8-oxo-G residues in the parental and control empty vector-

transfected cell lines. After 2 Gy irradiation, N18TG2 and

mock transfected cell lines revealed a steady-state level of 8-

oxo-G residues, independent from inhibitor treatment.

Despite the inhibitor treatments, the two active isoforms

yielded a statistically significant higher level of 8-oxo-G

residues (Fig. 7, 2 Gy columns). The MDL treatment at 50

AM dose caused an almost 25% decrease of 8-oxo-G

residues for both active enzymes (Fig. 7, fifth column),

meanwhile, the 100 AM treatment further decreased 8-oxo-

G residues only in mSMOa transfected cells, almost 50% in

respect of the absence of inhibitor (Fig. 7, sixth column).

Noticeable, this level of oxidized residues was statistically

significantly lower if compared to mSMOA (Fig. 7, sixth

column). These results indicate that the spermine oxidase

activity is a direct oxidative stress inducer of DNA damage,

thus rendering cells more sensitive to radiation and

apoptosis. Interestingly, MDL provokes oxidative damage

probably as downstream effects of apoptosis induced by its

detrimental effects on lysosomal cellular fraction ([28]; this

paper). As a consequence, the steady-state level of 8-oxo-G

residues in the active isoforms transfected cell lines, in the

Fig. 7. Determination of 8-oxo-G residues. HPLC/EC analysis to determine

the number of 8-oxo-G residues/106 guanine residues, in the absence (0 Gy)

and after 2 Gy of X-irradiation (as indicated). Results are given asF95% CI

of measurements on three replicas. Symbols: closed square (n), parental

N18TG2 cells; closed circle (.), mock transfected cells; open square (5),

mSMOa transfected cells; open circle (o), mSMOA transfected cells.

Abbreviations: N.I., no inhibitor; 50 AM-100 AM MDL, concentration of

the MDL inhibitor treatment before analysis. Black arrowheads indicate

statistically significant differences of the means at Pb0.05 between controls

and mSMO transfected cell lines. White arrowhead indicates statistically

significant differences of the means at Pb0.05 between mSMOa and

mSMOA transfected cell lines.

R. Amendola et al. / Biochimica et Biophysica Acta 1755 (2005) 15–2422

absence of irradiation, could represent a balance between

mSMO inhibition and DNA damage, both exerted by MDL

treatment.

Fig. 8. SMO and PAO activities and H2O2 production in mammalian cell. SMO an

co-localized SMOA isoform. Abbreviations: CAT, catalase; FR, Fenton Reaction

glutathione peroxidase; GSSG, glutathione disulfide.

4. Discussion

The flavoprotein SMO specifically oxidizes SPM as one

of the regulatory enzymes in the animal cells polyamines

homeostasis [6–8]. Several splice variants of the canonical

cytosolic mSMOa were recently isolated and cloned from

mouse brain. Only the mSMOA isoform detains enzymatic

activity, but presents a nuclear co-localization [15]. Since

both nuclear and cytosolic SMO activity may modulate

DNA stability, and apoptosis, we decided to over-express

mSMO isoforms in murine NB cells, in the presence or

absence of the specific inhibitor MDL and X-irradiation.

N18TG2 cell line provides to be a suitable experimental

model since the level of endogenous mSMO and isoform

mRNAs expression are undetectable [15] and barely

detectable after 35 cycles of RT-PCR. We preferred to

compare a relatively low dose of irradiation of 2 Gy and a

higher dose of 4 Gy, trying to discriminate between the

irradiation induced saturating oxidative stress and the

oxidative stress only due to mSMO over-expression. To

this end, we analyzed experimental points at 6 h after

irradiation because the ODC mRNA accumulation and

enzyme activity have been described to be at physiological

level by 6 h after gamma ray exposure [25]. In this work,

we demonstrated that the over-expression of the mSMOa

and of the mSMOA causes an enhancement of the

oxidative damage to DNA, which leads to a marked cell

death and apoptosis phenomena after X-ray exposure.

Moreover, both endogenous mSMO isoform RNAs and

their enzymatic activities were not enhanced by radiation,

d PAO activities produce H2O2 in the cytoplasm and nuclear moiety, by the

(H2O2+Fe2+(Cu+)YOH�+OH8+Fe3+(Cu2+)); GSH, glutathione; GSH-PX,

R. Amendola et al. / Biochimica et Biophysica Acta 1755 (2005) 15–24 23

thus ruling out the possibility of an over-estimation of the

cellular damage due to a likely unspecific activation of

SMO metabolism. We described a first evidence that the

transfected mSMOA isoform retains enzymatic activity

inside the nucleus, thus potentially diminishing SPM

content that acts as a DNA shielding molecule [33,34]

and a scavenger against radicals species [35]. Furthermore,

since the catabolic oxidative metabolism of SPM produces

hydrogen peroxide, we estimated the DNA damage by

ROS measuring the number of 8-oxo-G residues [32].

Noticeable, both active isoforms were able to deliver a

direct oxidative stress to DNA in the absence of radiation,

in a greater extent if related to apoptosis. However,

although DNA lesions could be considered as one of the

most serious damage to cell, biological effects cannot be

strictly related to, in consideration of the complex DNA

repair system that is able to rescue cell life. The marked

higher level of DNA damage and apoptosis after 2 Gy

irradiation could reflect the overcame of DNA repair

threshold, in line with the low-LET irradiation delivered by

an X-ray generator, where a 2–3 times greater radio-

sensitivity is observed for cells exposed in the presence of

ROS, for the boxygen enhancement ratioQ (OER) effect

[36]. As a consequence of the present work, it should be

taken into account that the nuclear localization of the

mSMOA isoform could be relevant to the hydrogen peroxide

production in this compartment. In fact, several evidences

have been reported that a the highly reactive hydroxyl

radical, produced via H2O2 via Fenton reaction, in close

proximity of DNA, results in an oxidative damage of nucleic

acids [37,38]. In Fig. 8, a simplified scheme of H2O2

production by SMO and PAO enzyme activities and

detoxifying metabolism is presented. Although H2O2 is

diffusible, a nuclear over-production bymSMOAmetobolism

of the level of H2O2 is less hampered by the antioxidant

system, since only glutathione peroxidase has been

described in the nucleus [39]. In conclusion, we described

the first evidence of a nuclear co-localized active isoform of

the murine spermine oxidase enzyme, and we directly

correlate the activity with a DNA oxidative stress,

apoptosis, and radiosensitivity. The presence of the nuclear

co-localized isoform could play a regulative role for the

SPM level into the nucleus. Dealing with the chemo-

therapeutical approach to cancer, and taking into consid-

eration the tissues specific SMO activity [15], our work

could suggest a new tool to tailor radiotherapies associated

with chemotherapies, improving dose-rate protocol and

trying to modulate deleterious side effects on confining

healthy tissues.

Acknowledgements

The authors are indebted to Prof. R. Federico (Universita

Roma, Italy) and Dr. D. Horejsh (INMI, I.R.C.C.S. bL.SpallanzaniQ, Roma, Italy) for criticisms and scientific help,

and Dr. P. Altavista (ENEA, Italy) for the statistical

analyses. P.M. has been partially funded by PRIN 2003

(M.I.U.R., Italy). F.M. has been funded by the Ministry of

Health 2003 Current Research project.

References

[1] H.M. Wallace, A.V. Fraser, A. Hughes, A perspective of polyamine

metabolism, Biochem. J. 376 (2003) 1–14.

[2] R. Heljasvaara, I. Veress, M. Halmekytf, L. Alhonen, J. J7nne, P.Laajala, A. Pajunen, Transgenic mice overexpressing ornithine and

S-adenosylmethionine decarboxylases maintain a physiological

polyamine homoeostasis in their tissues, Biochem. J. 323 (1997)

457–462.

[3] L. Kauppinen, S. Myfh7nen, M. Halmekytf, L. Alhonen, J. J7nne,Transgenic mice overexpressing the human spermidine synthase gene,

Biochem. J. 293 (1993) 513–516.

[4] R.A. Casero Jr., A.E. Pegg, Spermidine/spermine N1-acetyltransfer-

ase the turning point in polyamine metabolism, FASEB J. 7 (1993)

653–661.

[5] N. Seiler, Oxidation of polyamines and brain injury, Neurochem. Res.

25 (2000) 471–490.

[6] S. Vujcic, P. Diegelman, C.J. Bacchi, D.L. Kramer, C.W. Porter,

Identification and characterization of a novel flavin-containing

spermine oxidase of mammalian cell origin, Biochem. J. 367 (2002)

665–675.

[7] Y. Wang, T. Murray-Stewart, W. Devereux, A. Hacker, B. Frydman,

P.M. Woster, R.A. Casero Jr., Properties of purified recombinant

human polyamine oxidase, PAOh1/SMO, Biochem. Biophys. Res.

Commun. 304 (2003) 605–611.

[8] M. Cervelli, F. Polticelli, F. Rodolfo, P. Mariottini, Heterologous

expression and characterization of mouse spermine oxidase, J. Biol.

Chem. 278 (2003) 5271–5276.

[9] Y. Wang, W. Devereux, P.M. Woster, R.A. Casero, Cloning and

characterization of the mouse polyamine-modulated factor-1 (mPMF-

1) gene: an alternatively spliced homologue of the human tran-

scription factor, Biochem. J. 359 (2001) 387–392.

[10] J.O. Fredlund, S.M. Oredsson, Normal G1/S transition and prolonged

S phase within one cell cycle after seeding cells in the presence of an

ornithine decarboxylase inhibitor, Cell Prolif. 29 (1996) 457–466.

[11] D.L. Kramer, S. Vujcic, P. Diegelman, J. Alderfer, J.T. Miller, J.D.

Black, R.J. Bergeron, C.W. Porter, Polyamine analogue induction of

the p53-p21WAF1/CIP1-Rb pathway and G1 arrest in human

melanoma cells, Cancer Res. 59 (1999) 1278–1286.

[12] R.M. Ray, B.J. Zimmerman, S.A. McCormack, T.B. Patel, L.R.

Johnson, Polyamine depletion arrests cell cycle and induces inhibitors

p21(Waf1/Cip1), p27(Kip1), and p53 in IEC-6 cells, Am. J. Physiol.

276 (1999) 684–691.

[13] R. Poulin, J.K. Coward, J.R. Lakanen, A.E. Pegg, Enhancement of the

spermidine uptake system and lethal effects of spermidine over-

accumulation in ornithine decarboxylase overproducing L1210 cells

under hyposmotic stress, J. Biol. Chem. 268 (1993) 4690–4698.

[14] R.G. Schipper, L.C. Penning, A.A. Verhofstad, Involvement of

polyamines in apoptosis. Facts and controversies: effectors or

protectors? Semin. Cancer Biol. 10 (2000) 55–68.

[15] M. Cervelli, A. Bellini, M. Bianchi, L. Marcocci, S. Nocera, F.

Ponticelli, R. Federico, R. Amendola, P. Mariottini, Mouse spermine

oxidase gene splice variants. Nuclear subcellular localization of a

novel active isoform, Eur. J. Biochem. 271 (2004) 1–11.

[16] P. Bey, F.N. Bolkenius, N. Seiler, P. Casara, N-2,3-butadienyl-1,4-

butanediamine derivatives: potent irreversible inactivators of mamma-

lian polyamine oxidase, J. Med. Chem. 28 (1985) 1–2.

[17] A. Sjoholm, P. Arkhammar, P.-O. Berggren, A. Andersson, Poly-

amines in pancreatic islets of obese-hyperglycemic (ob/ob) mice of

different ages, Am. J. Physiol., Cell Phisyol. 280 (2001) 317–323.

R. Amendola et al. / Biochimica et Biophysica Acta 1755 (2005) 15–2424

[18] M. Bradford, A rapid and sensitive method for the quantitation of

microgram quantities of protein utilizing the principle of protein–dye

binding, Anal. Biochem. 72 (1976) 248.

[19] N. Seiler, Polyamine oxidase, properties and functions, Prog. Brain

Res. 106 (1995) 333–344.

[20] J.M. Mates, J. Marquez, M. Garcia-Caballero, I. Nunez de Castro, F.

Sanchez-Jimenez, Simultaneous fluorimetric determination of intra-

cellular polyamines separated by reversed-phase high-performance

liquid chromatography, Agents Actions 36 (1992) 17–21.

[21] I. Nicoletti, G. Migliorati, M.C. Pagliacci, F. Grignani, C. Riccardi, A

rapid and simple method for measuring thymocyte apoptosis by

propidium iodide staining and flow cytometry, J. Immunol. Methods

139 (1991) 271–279.

[22] Y. Gavrieli, Y. Sherman, S.A. Ben-Sasson, Identification of pro-

grammed cell death in situ via specific labeling of nuclear DNA

fragmentation, J. Cell Biol. 119 (1992) 493–501.

[23] M.T. Russo, M.F. Blasi, F. Chiera, P. Fortini, P. Degan, P.

Macpherson, M. Furuichi, Y. Nakabeppu, P. Karran, G. Aquilina,

M. Bignami, The oxidized deoxynucleoside triphosphate pool is a

significant contributor to genetic instability in mismatch repair-

deficient cells, Mol. Cell. Biol. 24 (2004) 465–474.

[24] R. Chaturvedi, Y. Cheng, M. Asim, F.I. Bussiere, H. Xu, A.P. Gobert,

A. Hacker, RA. Casero Jr., K.T. Wilson, Induction of polyamine

oxidase 1 by Helicobacter pylori causes macrophage apoptosis by

hydrogen peroxide release and mitochondrial membrane depolariza-

tion, J. Biol. Chem. 279 (2004) 40161–40173.

[25] E. Grassilli, M.A. Desiderio, E. Bellesia, P. Salomoni, F. Benatti, C.

Franceschi, Is polyamine decrease a common feature of apoptosis?

Evidence from grays- and heat shock-induced cell death, Biochem.

Biophys. Res. Commun. 216 (1995) 708–714.

[26] H. Tomitori, M. Nenoi, K. Mita, K. Daino, K. Igarashi, S. Ichimura,

Functional characterization of the human spermidine/spermine N(1)-

acetyltransferase gene promoter, Biochim. Biophys. Acta 1579 (2002)

180–184.

[27] A. Bellelli, S. Cavallo, L. Nicolini, M. Cervelli, M. Bianchi, P.

Mariottini, M. Zelli, R. Federico, Mouse spermine oxidase: a model of

the catalytic cycle and its inhibition by N,N1-bis(2,3-butadienyl)-1,4-

butanediamine, Biochem. Biophys. Res. Commun. 322 (2004) 1–8.

[28] H. Dai, D.L. Kramer, C. Yang, K. Gopal Murti, C.W. Porter, J.L.

Cleveland, The polyamine oxidase inhibitor MDL-72,527 selectively

induces apoptosis of transformed hematopoietic cells through lysoso-

motropic, Cancer Res. 59 (1999) 4944–4954.

[29] M.J. Arends, R.G. Morris, A.H. Wyllie, Apoptosis: the role of the

endonuclease, Am. J. Pathol. 136 (1990) 593–608.

[30] V. Ehemann, J. Sykora, J. Vera-Delgado, A. Lange, H.F. Otto,

Flow cytometric detection of spontaneous apoptosis in human

breast cancer using the TUNEL-technique, Cancer Lett. 194 (2003)

125–131.

[31] T. Lindahl, Instability and decay of the primary structure of DNA,

Nature 362 (1993) 709–715.

[32] J. Cadet, T. Douki, D. Gasparutto, J.L. Ravanat, Oxidative damage to

DNA: formation, measurement and biochemical features, Mutat. Res.

531 (2003) 5–23.

[33] A.U. Khan, P. Di-Mascio, M.H. Medeiros, T. Wilson, Spermine and

spermidine protection of plasmid DNA against single-strand breaks

induced by singlet oxygen, Proc. Natl. Acad. Sci. U. S. A. 89 (1992)

11428–11430.

[34] C. Muscari, C. Guarnieri, C. Stefanelli, A. Giaccari, C.M. Caldarera,

Protective effect of spermine on DNA exposed to oxidative stress,

Mol. Cell. Biochem. 144 (1995) 125–129.

[35] E. Lovaas, G. Carlin, Spermine: an anti-oxidant and anti-inflammatory

agent, Free Radic. Biol. Med. 11 (1991) 455–461.

[36] J.-P. Pouget, S.J. Mather, General aspects of the cellular response

to low- and high-LET radiation, Eur. J. Nucl. Med. 28 (2001)

541–561.

[37] B.N. Ames, M.K. Shigenaka, T.M. Hagen, Oxidants, antioxidants,

and the degenerative diseases of aging, general aspects of the

cellular response to low- and high-LET radiation, Proc. Natl. Acad.

Sci. U. S. A. 90 (1993) 7915–7922.

[38] T. Kurz, A. Leake, T. von Zglinick, U.T. Brunk, Relocalized redox-

active lysosomal iron is an important mediator of oxidative-stress-

induced DNA damage, Biochem. J. 378 (2004) 1039–1045.

[39] L.K. Rogers, S. Gupta, S.E. Welty, T.N. Hansen, C.V. Smith, Nuclear

and nucleolar glutathione reductase, peroxidase, and transferase

activities in livers of male and female Fisher-344 rats, Toxicol. Sci.

69 (2002) 279–285.