Báo cáo khoa học: Barley polyamine oxidase isoforms 1 and 2, a peculiar case of gene duplication docx

Barley polyamine oxidase isoforms 1 and 2, a peculiar case
of gene duplication
Manuela Cervelli
1
, Marzia Bianchi
1
, Alessandra Cona
1
, Cristina Crosatti
2
, Michele Stanca
2
,
Riccardo Angelini
1
, Rodolfo Federico
1
and Paolo Mariottini
1
1 Dipartimento di Biologia, Universita
`
‘Roma Tre’, Rome, Italy
2 Istituto Sperimentale per la Cerealicoltura, Sezione di Fiorenzuola d’Arda (PC), Italy
Plant polyamine oxidase (PAO), a flavin adenine dinu-
cleotide-containing enzyme, catalyzes the oxidation of
spermidine (Spd) and spermine (Spm) to 4-aminobut-
anal and N-(3-aminopropyl)-4-aminobutanal, respect-
ively, plus 1,3-diaminopropane and H
2
O
2

[1–3].
Because these compounds cannot be converted directly
to other polyamines, plant PAO is considered to be
involved in the terminal catabolism of polyamines.
Two barley (Hordeum vulgare) paralogous PAO genes
(HvPAO1 and HvPAO2, formerly BPAO1 and
BPAO2) code for two protein isoforms which share
73% amino acid identity [4]. In particular, HvPAO2
isoform has been purified, characterized and compared
with the maize (Zea mays) counterpart PAO, ZmPAO,
Keywords
biochemical characterization; enzyme
isoform; gene duplication; polyamine
oxidase; tissue specificity
Correspondence
P. Mariottini, Dipartimento di Biologia,
Universita
`
degli Studi ‘Roma Tre’, Viale
Guglielmo Marconi 446, 00146 Roma, Italy
Fax: +39 06 55176321
Tel: +39 06 55176359
E-mail:
(Received 09 May 2006, revised 23 June
2006, accepted 30 June 2006)
doi:10.1111/j.1742-4658.2006.05402.x
Polyamine oxidases (PAOs, EC 1.5.3.11) are key enzymes responsible for
the terminal catabolism of polyamines in plants, bacteria and protozoa. In
barley, two PAO isoforms (HvPAO1 and HvPAO2) have been previously
analyzed as regards their tissue expression and subcellular localization.

Only the major isoform HvPAO2 has been biochemically characterized up
to now. In order to study the ear-specific expression of the HvPAO1 iso-
form in detail, RT-PCR analysis was performed in barley on the whole ear
and on various ear tissues. Moreover, HvPAO1promoter::GUS transient
expression was examined in barley developing caryopses at 30-day postfer-
tilization. Results from these analyses have demonstrated that the HvPAO1
gene is specifically expressed in all the ear organs analyzed (i.e. basal
lemma, rachis, awn, embryo-deprived caryopsis, embryo and sterile spike-
lets), at variance with the HvPAO2 gene, which is expressed at high levels
in sterile spikelets and at very low levels in embryos. We purified HvPAO1
from barley immature caryopses and characterized its catalytic properties.
Furthermore, we carried out in vitro synthesis of HvPAO1 protein in a
cell-free translation system. The HvPAO1 enzymes purified from immature
caryopses and in vitro synthesized showed the same catalytic properties, in
particular, an optimum at pH 7.0 for Spd and Spm oxidation and compar-
able K
m
values for both substrates, i.e. 0.89 · 10
)5
m and 0.5 · 10
)5
m for
Spd and Spm, respectively. It has been found that HvPAO1 enzyme activ-
ity significantly differs in substrate specificity and pH optimum when com-
pared with the major isoform HvPAO2. As a whole, these data strongly
suggest that, in barley, the two PAO genes evolved separately, after a
duplication event, to code for two distinct tissue-specific enzymes, and they
are likely to play different physiological roles.
Abbreviations
HvPAO1, barley polyamine oxidase 1; HvPAO2, barley polyamine oxidase 2; PAO, polyamine oxidase; Spd, spermidine; Spm, spermine;

Ubi, ubiquitin; ZmPAO1, maize polyamine oxidase 1; ZmPAO2, maize polyamine oxidase 2.
3990 FEBS Journal 273 (2006) 3990–4002 ª 2006 The Authors Journal compilation ª 2006 FEBS
the best characterized plant member of this class of
enzymes [2,5–8]. The maize PAO gene family is repre-
sented by a small number of copies; three genes enco-
ding polyamine oxidase (ZmPAO1, ZmPAO2 and
ZmPAO3, formerly MPAO1, MPAO2 and MPAO3)
and their upstream regions have been previously char-
acterized [9]. They show a highly conserved gene
organization and almost identical amino acid
sequences, indicating that they originated from dupli-
cation events. Molecular modeling of HvPAO2 shows
the same global fold of ZmPAO, but the two proteins
have different catalytic properties [4]. Both precursor
enzymes include a cleavable N-terminal leader; more-
over, HvPAO2 has an eight-residue-long carboxyl
extension (DELKAEAK) that directs this protein to
the vacuole [10]. Thus this C-terminus is responsible
for the different subcellular localization observed in
leaf tissues between the two enzymes, as HvPAO2
is symplastic in mesophyll cells, at variance with
ZmPAO, which is apoplastic in the same tissue
[4,10,11]. While HvPAO2 transcript is the major form
detectable in all barley plant tissues analyzed so far,
HvPAO1 gene transcription is tissue specific, being
observed by RT-PCR analysis only in the ear [4]. The
presence of the N-terminal signal peptide in HvPAO1
indicates the transit of this protein in the secretory
pathway, possibly targeting this protein to extracellular
compartment, as in the case of ZmPAO protein. The

amino acid identity shared by these two enzymes is
84% higher than the one shared by HvPAO2 and
ZmPAO (73%), indicating that HvPAO1 and
ZmPAO1-3 genes are orthologous [4]. The physiologi-
cal roles ascribed to ZmPAO relate mainly to poly-
amine homeostasis, as well as hydrogen peroxide
biosynthesis in the apoplast [3]. The latter functional
implication arises from the analysis of several experi-
mental results concerning: (i) the high specific activity
in extracellular fluids [12]; (ii) the correlation of PAO
expression with photomorphogenic growth regulation
and the hypersensitive response [13–15]; (iii) the inhibi-
tion of hydrogen peroxide release by PAO activity
inhibitors [16]; and (iv) the histochemical and ultra-
structural studies that demonstrated the association of
this enzyme with the cell wall [16,17].
Many of these studies highlighted the physiological
implications of PAO-mediated hydrogen peroxide syn-
thesis in the apoplast related to peroxidase-catalyzed
reactions or as a compound triggering signal transduc-
tion leading to hypersensitive response. However, the
vacuolar localization of HvPAO2 and the predicted
apoplastic localization of HvPAO1 draw a new scen-
ario to the possible role multiplicity of this enzyme
family. In fact, even in maize, this multiplicity may
have been underestimated. A recent immunogold ultra-
structural study has in fact shown that ZmPAO could
be detected in the cytoplasm of differentiating xylem
and rhizodermal cells of young root tissues [17]. This
localization has been correlated with two possible

additional functions, reactive oxygen species-induced
programmed cell death of xylem elements [17] and
hydrogen peroxide-dependent cross-linking of polysac-
charides within the secretory pathway [13,17]. In par-
ticular, the early cross-linking of hemicellulose and
pectin which occurs in young cells or tissues could
result in the formation of large coagula that would
have a loosening effect within the cell wall due to their
scarce interactions with the cellulose microfibrils, this
being diverse from apoplastic polymer cross-linking,
which is thought to strengthen the cell wall inasmuch
as it occurs after the hemicelluloses and pectins have
already bonded to cellulose microfibrils [18]. Under
this view, it is reasonable to hypothesize that, in bar-
ley, different PAO isoforms, specifically expressed dur-
ing development in different tissues and organs, could
play different physiological roles.
This article describes a detailed analysis of HvPAO1
and HvPAO2 gene expression in barley ear and the
characterization of the main biochemical features of
purified and in vitro synthesized HvPAO1 enzyme.
RT-PCR analysis was also carried out on different
ear tissues and internodes. Constructs containing
HvPAO1, HvPAO2, ZmPAO1 and ZmPAO2 promoter
sequences [4,9], fused to the b-glucuronidase gene
(GUS), were transiently expressed in roots, leaves and
ears of barley, and in roots and leaves of maize with
the aid of a biolistic delivery system. In this study, we
present evidence that barley HvPAO1 and HvPAO2
genes represent an interesting evolutionary case of gene

duplication, since the orthologous HvPAO1 coding
sequence corresponds to ZmPAO1-3 genes, while the
paralogous HvPAO2 coding sequence could be consid-
ered as a more recently evolved gene with a different
physiological role.
Results
Accumulation of HvPAO1 and HvPAO2 mRNAs
in different barley tissues
The transcription level of HvPAO1 and HvPAO2
mRNAs has been examined in different stem inter-
nodes and whole ear at 30-days postfertilization and
in various ear tissues by RT-PCR analysis (Fig. 1).
PCR-amplified mRNAs have been probed with
primer-pairs specific for HvPAO1 and HvPAO2
isoforms and within the linear range of PCR
M. Cervelli et al. Barley polyamine oxidase isoform 1 gene expression
FEBS Journal 273 (2006) 3990–4002 ª 2006 The Authors Journal compilation ª 2006 FEBS 3991
amplification conditions. In details, the HvPAO1
transcript is accumulated only in the ear (Fig. 1A,
left panel); even with saturating conditions there
were no detectable PCR amplified products in any
stem internodes (not shown), thus confirming previ-
ous results reported by Cervelli et al. [4]. On the
contrary, HvPAO2 mRNA is accumulated in all the
tissues examined, including ear. A further RT-PCR
analysis has been carried out on different ear tissues
(Fig. 1A, right panel) and interestingly the accumula-
tion pattern shown by HvPAO1 transcript indicates
that this gene is expressed in basal lemma, rachis,
awn, embryo-deprived caryopsis, embryo and sterile

spikelets, at variance with the HvPAO2 gene, which
is expressed at a comparable level only in sterile
spikelets (Fig. 1A, right panel). The transcription
accumulation pattern of the ribosomal protein S12
(rp-S12) mRNA has been also analyzed as a control
housekeeping gene for RNA stability and the quan-
tity processed for each sample. As HvPAO2 promo-
ter was capable of driving GUS expression in barley
developing caryopsis, exclusively in the embryo (see
section below on HvPAO1, HvPAO2, ZmPAO1 and
ZmPAO2::GUS expression in different plants and
organs), to detect the presence of any low level of
HvPAO2 mRNA accumulation in the barley embryo,
a further PCR analysis was carried out in the non-
linear range of amplification (up to 35 cycles) using
the same cDNA sample. The results are shown in
Fig. 1(B); a PCR product of faint intensity specific
for HvPAO2 transcript was visible only after 35
cycles, demonstrating the presence of the HvPAO2
mRNA in embryonic organs albeit in relatively small
amounts.
A
B
Fig. 1. HvPAO1 and HvPAO2 transcript
detection by RT-PCR in different barley tis-
sues. (A) Total RNA isolated from different
stem internodes, whole ear and various ear
tissues was analyzed by RT-PCR amplifica-
tion procedure using specific primers for
HvPAO1, HvPAO2 and, as a control, rp-S12

transcripts in the linear range of amplifica-
tion (25 cycles). (B) cDNA from embryo was
analyzed in saturating PCR condition (35
cycles). The PCR products were fractionated
by 1.2% agarose gel electrophoresis. Expec-
ted size of PCR fragments are indicated at
right.
Barley polyamine oxidase isoform 1 gene expression M. Cervelli et al.
3992 FEBS Journal 273 (2006) 3990–4002 ª 2006 The Authors Journal compilation ª 2006 FEBS
Cis-elements search in upstream region of
HvPAO2 gene
The 5¢-flanking region of HvPAO2 gene was cloned by
inverse PCR using specific oligonucleotides designed
from HvPAO2 partial gene sequences [4] and analyzed
by searching for putative cis-acting elements known
in plants by using the PLACE cis-element database
[19] ( We found a
well-conserved TATA-box, located 25 nucleotides
upstream of the putative transcription start point, an
I-box localized at )191 ⁄ )186, a G-box at )303 ⁄ )296
and a CCAAT-box at )405 ⁄ )409 (Fig. 2A). These
putative light-response elements [20,21] are also present
in the promoter regions of HvPAO1, ZmPAO1 and
ZmPAO2 genes [4,9]. A sequence motif that is shared
only by HvPAO2 and HvPAO1 upstream regions
is the MYB1AT-box (consensus 5¢-WAACCA-3¢),
located at )348 ⁄ )353 (Fig. 2A). This is a dehydration-
stress response element present in Arabidopsis thaliana
but not in rice [22]. To summarize, in spite of the fact
that some sequence motifs are shared by HvPAO1,

HvPAO2, ZmPAO1 and ZmPAO2 genes, a comparat-
ive sequence analysis did not show any evident com-
mon cis-acting elements organization in their upstream
regions.
HvPAO1, HvPAO2, ZmPAO1 and ZmPAO2::GUS
expression in different plants and organs
GUS expression was obtained in developing caryopses,
roots and leaves of barley and in roots and leaves of
maize by means of biolistic inoculations with the GUS
gene under the control of different promoters.
Figure 2(B) shows a schematic representation of the
chimerical constructions used in these experiments.
Twenty-four hours after bombardment, the different
organs were assayed for GUS activity (Figs 3 and 4).
A
B
C
Fig. 2. Gene promoter::GUS constructs util-
ized in transient GUS expression experi-
ments. A. 5¢-flanking region of HvPAO2
gene. Upstream and exon nucleotide
sequences are indicated with lower-case
and upper-case letters (start translation
codon in bold), respectively. Putative promo-
ter sequence motifs are indicated with bold
underlined letters and marked. The restric-
tion site HindIII used in inverse PCR experi-
ments is indicated with italics underlined
letters. Exon sequence is numbered from
the putative tsp (+ 1), upstream sequence is

indicated by negative numbers. B. Schemes
of pHTT515 and pHTT-PAOs construct vec-
tors. C. Schematic representation of the
constructs utilized in particle bombardment
and detection (+) or absence (–) of their
expression in different organs and plants.
Numbering refers at the promoter
sequences (open boxes) jointed to the UB
intron region (grey boxes).
M. Cervelli et al. Barley polyamine oxidase isoform 1 gene expression
FEBS Journal 273 (2006) 3990–4002 ª 2006 The Authors Journal compilation ª 2006 FEBS 3993
Transient expression of HvPAO1, HvPAO2, ZmPAO1
and ZmPAO2promoter::GUS constructs, using
pHTT515 plasmid (UBpromoter::GUS) as a control
(Fig. 2B), revealed that only barley immature caryopsis
were competent for the HvPAO1 promoter driving
GUS expression, especially in the embryo, aleurone
layer and endosperm. On the contrary, it was inactive
in roots and leaves of both barley and maize. Interest-
ingly, HvPAO2 promoter was capable of driving GUS
expression in barley roots and leaves, as expected, but
also in developing caryopses, albeit exclusively in the
embryo, as well as in roots and leaves of maize. Fur-
thermore, ZmPAO1, ZmPAO2 and UB promoters
transpired to be active in all analyzed organs of barley
and maize. Results obtained in the transient GUS
expression experiments are summarized in Fig. 2(C).
Fig. 3. Histochemical localization of glucoronidase (GUS) activity in different barley tissues after gene bombardment. HvPAO1, HvPAO2,
ZmPAO1 and ZmPAO2promoter::GUS (pHTT-PAOs) and UBpromoter::GUS (pHTT515) expression vectors were biolistically delivered to bar-
ley roots, leaves and ears. Organs were histochemically reacted with X-Glu and examined for blue staining assessment with a Zeiss stereo-

microscope and photographed. Longitudinal section of ears is enlarged at right. Photographs are representative of three different
experiments each performed in triplicate.
Barley polyamine oxidase isoform 1 gene expression M. Cervelli et al.
3994 FEBS Journal 273 (2006) 3990–4002 ª 2006 The Authors Journal compilation ª 2006 FEBS
Purification of the HvPAO1 protein from
developing caryopses
HvPAO1 was extracted from immature caryopses with
a high ionic strength salt solution. The enzyme was
then partially purified from supernatant obtained after
centrifugation of the crude homogenate through a
fractionation in 70% saturated ammonium sulfate
and two chromatographic steps (hydroxylapatite and
SP-sepharose columns). By this procedure, a 58-fold
purification of the enzyme was achieved (Table 1). No
detectable HvPAO2 activity was revealed overall dur-
ing the entire procedure of HvPAO1 purification, nei-
ther in the purification steps reported in Table 1 (1–4
fractions), nor in hydroxylapatite and SP-sepharose
flow-through and washing fractions. In western blot
analysis, the SP-sepharose eluate (fraction 4; Table 1)
showed a band of 53 kDa molecular mass, corres-
ponding to the HvPAO1 expected mass (Fig. 5) when
probed against polyclonal anti-ZmPAO antibodies; it
has already been demonstrated that these cross-react
with the less conserved HvPAO2 protein [4].
In vitro HvPAO1 protein synthesis
In order to confirm that the PAO activity present in
the SP-sepharose eluate (fraction 4; Table 1) could be
ascribed to the HvPAO1 isoform, we carried out the
in vitro synthesis of HvPAO1 protein utilizing the

pET17b-HvPAO1 plasmid as a template in three
different cell-free translation systems and precisely
RTS-100 Roche (Roche Diagnostics, Monza, Italy),
Escherichia coli T7 S30 Extract System for circular
DNA (Promega Italia, Milano, Italy) and Pure-System
Classic (Post Genome Institute, Tokyo, Japan), as des-
cribed in the Methods section. The highest in vitro syn-
thesized HvPAO1 protein yield (0.1 U) was obtained
with the RTS-100 Roche translation system, as detec-
ted by enzymatic assay. Moreover, western blot analy-
sis of the in vitro translated product probed against
polyclonal anti-ZmPAO antibodies, showed a band of
53 kDa molecular mass, thus confirming the presence
of HvPAO1 protein (Fig. 5).
ZmPAO, HvPAO2 and HvPAO1 protein catalytic
properties
ZmPAO and HvPAO2 showed pH optima and K
m
val-
ues for Spd and Spm oxidation (Table 2) identical to
those previously reported by Cervelli et al. [4]. Cata-
lytic properties of the HvPAO1 enzyme purified from
immature caryopsis resulted identical to those exhib-
ited by the in vitro synthesized HvPAO1 recombinant
enzyme for both Spd and Spm substrates. In partic-
ular, HvPAO1 enzymatic activity showed an optimum
at pH 7.0 for Spm and Spd oxidation and K
m
values
of 0.89 · 10

)5
m and of 0.50 · 10
)5
m for Spd and
Spm substrates, respectively (Fig. 6; Table 2). More-
over, the V
max
ratio for Spd and Spm at pH 7.0 was
Fig. 4. Histochemical localization of glucoronidase (GUS) activity
in different maize tissues after gene bombardment. HvPAO1,
HvPAO2, ZmPAO1 and ZmPAO2promoter::GUS (pHTT-PAOs) and
UBpromoter::GUS (pHTT515) expression vectors were biolistically
delivered to maize roots and leaves. Organs were histochemically
reacted with X-Glu and examined for blue staining assessment with
a Zeiss stereomicroscope and photographed. Photographs are
representative of three different experiments each performed in
triplicate.
M. Cervelli et al. Barley polyamine oxidase isoform 1 gene expression
FEBS Journal 273 (2006) 3990–4002 ª 2006 The Authors Journal compilation ª 2006 FEBS 3995
found to be 1.3. The HvPAO1 biochemical features
exhibited by the in vitro synthesized recombinant pro-
tein and the native partially purified enzyme from bar-
ley ears are congruent, demonstrating that the PAO
activity detected in the SP-sepharose eluate (fraction 4;
Table 1) could be reasonably ascribed to the exclusive
presence of HvPAO1 enzyme with respect to HvPAO2
in developing caryopses. In fact, the presence of any
detectable HvPAO2 activity in the SP-sepharose eluate
(Table 1) would result in different PAO catalytic
parameters, with K

m
and pH optimum values for Spd
and Spm intermediate between those of purified
HvPAO2 and in vitro synthesized HvPAO1. HvPAO1
catalytic properties were very similar to those of
ZmPAO (Table 2), moreover HvPAO1 showed com-
parable affinity and identical pH optima values for
both Spd and Spm substrates (pH 7.0); analogously
ZmPAO showed comparable affinity and identical
pH optima values for both substrates (pH 6.5). On the
contrary, HvPAO1 enzymatic features differ from the
ones of HvPAO2 that preferentially oxidizes spermine
at pH 5.5 and spermidine at pH 8.0 with a ten-fold
lower V
max
[4].
Discussion
Our results definitely support the identification of the
HvPAO1 enzyme as the major product of HvPAO
gene expression in barley ear (Fig. 1A, left panel). In
fact, HvPAO1 mRNA was detectable by standard
RT-PCR analysis only in the ear ([4] and this work),
indicating that HvPAO1 gene expression is ear-specific.
Further ear-tissue dissection demonstrated that in all
the samples examined (basal lemma, rachis, awn,
embryo-deprived caryopsis and embryo) by RT-PCR,
the HvPAO1 gene is expressed at comparable levels
(Fig. 1A, right panel). The only tissue where it was
possible to detect HvPAO2 gene expression with stand-
ard PCR conditions (25–30 cycles) resulted to be the

Table 1. Purification of the HvPAO1 protein from barley developing caryopses. HvPAO1 purification was performed from developing caryop-
ses at 30 days postfertilization, as described in the Methods section. The enzyme was partially purified from supernatant obtained after cen-
trifugation of the crude homogenate (fraction 1), through a fractionation in 70% saturated ammonium sulfate (fraction 2) and two
chromatographic steps (fraction 3 and 4).
Purification step
Total
volume
(mL)
Protein
(mgÆmL
)1
)
Total
protein
(mg)
Activity
(UÆmL
)1
)
Total
activity
(U)
Specific
activity
(UÆmg
)1
Æprotein)
Purification
fold
Recovery

(%)
Crude extract (fraction 1) 236 0.94 221.84 0.002 0.472 0.002 1.0 100.0
(NH
4
)
2
SO
4
70% sat. precipitation
(fraction 2)
104 1.24 128.96 0.004 0.416 0.003 1.5 88.1
Hydroxylapatite eluate (fraction 3) 44 0.24 10.56 0.005 0.220 0.021 10.5 46.6
SP eluate (fraction 4) 15 0.043 0.644 0.005 0.075 0.116 58.0 15.9
Fig. 5. Western blot analysis of ZmPAO, HvPAO2 and HvPAO1.
ZmPAO and HvPAO2 were purified as previously described [4,5].
HvPAO1 was purified and in vitro synthesized as described in
Experimental procedures. Analysis was performed running: 0.1 U
of ZmPAO and HvPAO2; 0.002 U of ear purified and RTS-100
Roche TS produced HvPAO1. Proteins were reacted, after deglyco-
sylation, with anti-ZmPAO polyclonal antibodies [4]. M, protein
molecular weight marker (Fermentas).
Table 2. K
m
values for ZmPAO, HvPAO2 and HvPAO1. For the
determination of the K
m
values, ZmPAO and HvPAO2 were purified
as previously described [4,5]. HvPAO1 was synthesized in vitro as
described in Experimental procedures. Data were obtained at 25 °C
with Spd and Spm as substrates at the specific pH optimum. K

m
values concerning Spd and Spm oxidation by ZmPAO and HvPAO2
were within the standard error of the values previously reported by
Cervelli et al. [4]. Ear purified and RTS-100 Roche TS produced
HvPAO1 showed identical K
m
values. All K
m
values, calculated from
Lineweaver-Burk plots, are means of three different experiments,
each performed in duplicate.
SD was 8%.
Enzyme pH Substrate K
m
(M)
ZmPAO 6.5 Spd 1.0 · 10
)5 a,b
ZmPAO 6.5 Spm 2.7 · 10
)5a,b
HvPAO2 8.0 Spd 56.0 · 10
)5a,b
HvPAO2 5.5 Spm 0.48 · 10
)5a,b
HvPAO1 7.0 Spd 0.89 · 10
)5a
HvPAO1 7.0 Spm 0.50 · 10
)5a
a
Present work;
b

Cervelli et al. [4].
Barley polyamine oxidase isoform 1 gene expression M. Cervelli et al.
3996 FEBS Journal 273 (2006) 3990–4002 ª 2006 The Authors Journal compilation ª 2006 FEBS
sterile spikelets. Using a higher number of PCR cycles,
we were also able to detect a very small amount of the
HvPAO2 transcript in the embryo tissue (Fig. 1B).
Thus, the HvPAO2 gene is also transcriptionally active
in the embryo, albeit at a very low level, probably rep-
resenting a basal transcriptional activity. This is in line
with the transient GUS expression experiments that
confirmed the specific and strong expression of the
HvPAO1 gene in barley ear, at variance with the weak
expression of the HvPAO2 gene, exclusively localized
in the embryo (Figs 2C and 3). As expected, the
HvPAO1 gene is silent in roots and leaves of maize
(Figs 2C and 3). Interestingly HvPAO2, ZmPAO1 and
ZmPAO2 promoters exhibit the same transcription
pattern, being able to drive GUS expression in all the
organs and plants analyzed in this study (Figs 2C, 3
and 4). Moreover, the ZmPAO1-2 promoters are also
active in barley embryo, albeit at a very low level, like
the HvPAO2 gene; it seems that the barley embryo
is able to allow a basal transcriptional level of these
promoter sequences. Sequence analysis of 5¢ flanking
regions of HvPAO2 (Fig. 2A), ZmPAO1 and ZmPAO2
genes albeit sharing some potential cis-acting elements,
do not show any obvious common promoter architec-
ture as expected by their identical transcription pattern
[4,9]. Furthermore, there are no evident sequence fea-
tures in the HvPAO1 and HvPAO2 promoters that

could explain their different gene expression profiles.
So, we are facing a puzzling gene duplication event
that occurs in barley, since the paralogous HvPAO2
and ZmPAO1-3 genes share a common tissue expres-
sion, which is at variance with the orthologous
HvPAO1 and ZmPAO1-3 genes that show a different
tissue regulation. The very similar catalytic properties
shown by HvPAO1 and ZmPAO could be ascribed to
the closer phylogenetic relationship existing between
them (84% identity), as compared with that between
HvPAO2 and ZmPAO1 (73% identity) (Table 2). It is
interesting to recall that, even if the global fold and
the flavin adenine dinucleotide-binding pocket are well
conserved in HvPAO1, HvPAO2 and ZmPAO, the
substitution of Phe403 of ZmPAO by a tyrosine resi-
due in HvPAO2 could probably play a relevant role in
the different substrate specificity and kinetic parame-
ters observed for this isoform [4]. Furthermore,
HvPAO1 amino acid sequence shows a different C-ter-
minus when compared with the HvPAO2 coding
sequence, which has an extra eight-residue long tail
(DELKAEAK) responsible for the symplastic localiza-
tion of this isoform [10]. On the contrary, the higher
similarity determined between HvPAO1 and ZmPAO,
strongly suggests an apoplastic localization of the
HvPAO1 isoform. Indeed, recent results have shown
that ZmPAO is also present at high levels in the cyto-
plasm, most probably in the secretory pathway of
young tissues undergoing or destined to programmed
cell death, such as developing xylem vessels and xylem

parenchyma of both the root and mesocotyl as well as
root cap cells [17]. Later, during cell maturation,
ZmPAO is found mainly in the cell wall [17]. On the
basis of these results, the authors hypothesized that
ZmPAO could play a dual role in these tissues being
involved either in programmed cell death or cell wall
differentiation through the action of its reaction
Fig. 6. HvPAO1 catalytic parameters for Spd
and Spm oxidation. HvPAO1 was purified
from developing caryopses at 30-day post-
fertilization, as described in Experimental
procedures. Data reported are the average
of three different experiments, each with
two replicates.
SD was 8%. (A) HvPAO1 cat-
alytic activity pH optima were determined at
25 °C, in 0.2
M sodium phosphate buffer
(pH range 4.5–8.5) with Spd or Spm as sub-
strates. PAO activity is expressed as per-
cent of the maximum value. (B) HvPAO1
(1 · 10
)3
U) K
m
values were determined at
25 °C, with Spd and Spm as substrates at
the respective pH optimum and then calcu-
lated from Lineweaver-Burk plots.
M. Cervelli et al. Barley polyamine oxidase isoform 1 gene expression

FEBS Journal 273 (2006) 3990–4002 ª 2006 The Authors Journal compilation ª 2006 FEBS 3997
products, hydrogen peroxide and aminoaldehydes
and ⁄ or modulation of polyamine levels [3,13,17]. One
can hypothesize that in barley PAO tissue-specific
functions are associated with the distinct isoforms
HvPAO1 and HvPAO2, an event that arose in the
course of evolution of C3 cereals. According to the tis-
sue distribution of HvPAO1 and HvPAO2, the prefer-
ential HvPAO2 substrate Spm has been detected at
higher level than Spd in barley leaves [23], whereas in
the developing grains a higher level of Spd than Spm
has been found [24]. Figure 7 summarizes the evolu-
tionary relationship among the cereal PAO genes stud-
ied in this work.
Is there any specific role for HvPAO1 in the ear
tissues and in particular in embryonic tissues and
aleurone, where the expression of HvPAO1 gene is
prominent, if not exclusive, compared with HvPAO2?
The available data suggest that HvPAO1 could have
a specific role in the aleurone cells. This is a tissue
that plays a key role during the germination of cereal
seeds. Aleurone cells secrete, under the stimulus of
embryo-synthesized gibberellin, amylase and other
hydrolytic enzymes involved in endosperm reserve
mobilization. Gibberellin also induces programmed
cell death in these cells, a process that is mediated by
hydrogen peroxide [25,26]. Thus the accumulation of
HvPAO1 during the development of barley caryopsis
could be functional to the production of hydrogen
peroxide needed in the programmed cell death process

taking place in the aleurone during germination.
However it should be recalled that hydrogen peroxide
production during germination could also have a
general protective role against microbial pathogens.
Alternatively, HvPAO1 could have a role in the regu-
lation of polyamine levels in the aleurone cells and in
the embryo as well. Indeed, it has been recently
reported that DNA synthesis early in development
and the advance in cell cycle ⁄ endocycle are tempor-
ally and spatially related to polyamine catabolism
and vascular development [27]. Moreover, polyamines
are active in triggering the synthesis of nitric oxide in
specific tissues of Arabidopsis thaliana seedlings [28].
This molecule is known to delay programmed cell
death in aleurone cells and also to have pleiotropic
effects on many facets of plant development and
defense [29].
Experimental procedures
Chemicals
Restriction and DNA-modifying enzymes and protein
molecular weight marker were purchased from MBI Fer-
mentas (MBI Fermentas, St. Leon-Rot, Germany). Sper-
midine (Spd) and spermine (Spm), horseradish peroxidase,
4-aminoantipyrine and 3,5-dichloro-2-hydroxybenzenesulf-
onic acid were purchased from Sigma-Aldrich-Fluka
(Sigma, Milano, Italy). TRIZOL reagent was from Invitro-
gen (Invitrogen, Milano, Italy). pGEM-Teasy vector and
the E. coli T7 S30 Extract System for circular DNA were
from Promega (Promega Italia, Milano, Italy). pET17b
vector and E. coli BL21 DE3 competent cells were from

Novagen (Novagen Inc., Madison, WI, USA). CHU(N
6
)
medium was from Duchefa (Duchefa Biochemie B. V.,
Haarlem, the Netherlands). Hydroxylapatite and gold par-
ticles (1.0 lm in diameter) were from Bio-Rad (Bio-Rad,
Milano, Italy). SP-sepharose was from Amersham
Biosciences (Amersham Biosciences, Milano, Italy). Carb-
oxymethylcellulose was from Whatman (Whatman, Maid-
stone, UK). Peroxidase-conjugated goat antirabbit IgG
was from Vector Laboratories (Burlingame, CA, USA).
The RTS-100 Roche translation system was from Roche
(Roche Diagnostics, Monza, Italy). The Pure-System Clas-
sic Mini Kit was from the Post Genome Institute (Post
Genome Institute, Tokyo, Japan). Other chemicals came
from Sigma-Aldrich-Fluka, Bio-Rad and J. T. Baker
(Baker Italia, Milano, Italy).
HvPAO2 HvPAO1
LEAVES ROOTS EARS
ZmPAO1,2
Extracellular
localization
?
Orthologous
gene
Paralogous
gene
HvPAO1 promoter
HvPAO2, ZmPAO1,2
promoters

?
Fig. 7. Schematic representation of the evolutionary relationship
among cereal PAO genes. Arrow width represents the promoter
expression level according to both RT-PCR analysis and biolistic
delivering experiments. The promoter boxes color reflects the
tissue specific expression.
Barley polyamine oxidase isoform 1 gene expression M. Cervelli et al.
3998 FEBS Journal 273 (2006) 3990–4002 ª 2006 The Authors Journal compilation ª 2006 FEBS
Plant material
Seedlings and adult plants of barley (H. vulgare) cultivar
Aura [30] were grown at the ‘Istituto Sperimentale per la
Cerealicoltura, Sezione di Fiorenzuola d’Arda, Italy’. Seeds
of barley cultivar Aura and of maize (Z. mays) cultivar
Corona (Monsanto Agricoltura, Italy) were soaked for 12 h
in aerated tap water, germinated at 20 °C in the dark and
grown aseptically in a growth chamber with a 16 : 8 h
light–dark cycle on Magenta vessels (Sigma-Aldrich-Fluka,
Milano, Italy) containing 0.8% agar. For ZmPAO and
HvPAO2 purification, maize and barley seeds were germi-
nated with 1 cm of fertile soil, at 20 °C in the dark. Two-
day-old barley seedlings were exposed to natural light for
4 days before harvesting; maize seedlings were kept in the
dark before protein extraction. Thirty-day postfertilization
ears were utilized for RT-PCR experiments, transient
expression assays and HvPAO1 purification.
RT-PCR analysis of HvPAO1 and HvPAO2 gene
expression in different tissues
Total RNA was isolated from different barley tissues by
TRIZOL reagent, according to the manufacturer’s instruc-
tions. Oligonucleotides utilized as primers for specific

amplification of HvPAO isoforms were: HvPAO-N, reverse
5¢-GTTATTACTTAGTACCTCTTAAT-3¢, HvPAO-O, for-
ward 5¢-GACGGAGATCTCCCACTC-3 ¢ and HvPAO-P,
reverse 5¢-GGTTGTCCGACTGCTGCTC-3¢ for HvPAO1;
HvPAO-Q, reverse 5¢-CTCGTCGGCGCGGTCCAT-3¢,
HvPAO-R, forward 5¢-GAGGGGAGAATTGAAGA
GAG-3¢ and HvPAO-S, reverse 5¢-GTCGTAGAGGCC
ACCGCT-3¢ for HvPAO2 as already described [4]. Oligo-
nucleotides for the control barley ribosomal protein S12
were: RPS12-A, reverse 5¢-ATTCTTCACCATAGTCCT-3¢,
RPS12-B, forward 5¢-GTGAGCCAATGGACTTGATG-3 ¢
and RPS12-C, reverse 5¢-ATGCAAGAGCAGCCTAC
AAC-3¢ [4].
HvPAO2 promoter isolation
Barley DNA was extracted and purified as described in
Cervelli et al. [4]. To clone 5¢- and 3¢-flanking regions of
HvPAO2 gene, amplified products were obtained by inverse
PCR using specific oligonucleotides designed from HvPAO2
partial gene sequences [4], in particular, HvPAO2-A,
reverse 5¢-TACTGTGTTAGCACTGCTAGC-3¢, and
HvPAO2-B, forward 5¢-GAGGGGAGAATTGAAGA
GAG-3¢, specific for the 5¢-end region. The gene-specific
primer couples were utilized on different samples of purified
barley total DNA, previously digested with HindIII and
self-ligated. A direct PCR to obtain the corresponding
HvPAO2 promoter sequence was performed on total
DNA with the oligonucleotides HvPAO2-C, forward
5¢-AAAAAGCTTACCAAAACTTGTGTAAACTT-3¢, and
HvPAO2-D, reverse 5¢-TTTAGATCTGCCCTGCTCTCC
GGCCCTGT-3¢, containing the HindIII and the BglII sites,

respectively. The PCR-amplified product was cloned in the
pGEM-Teasy vector. The promoter gene sequence has been
deposited in the EMBL database under EMBL accession
number AM231701.
DNA methodology and construction
of expression plasmids
The methods described by Sambrook et al. [31] were used
for the manipulation of plasmid DNAs and general DNA
in vitro procedures. In order to amplify HvPAO1, HvPAO2,
ZmPAO1 and ZmPAO2 promoter regions ([4,9] and
present work), gene-specific oligonucleotides containing
HindIII and the BglII sites were designed spanning from
the 5¢ end of the promoter sequence down to the transcrip-
tion start site [primer sequences used are available on
request from the first author (MC)]. HvPAO2, ZmPAO1
and ZmPAO2 promoter PCR products were cloned into
the expression vector pHTT515 utilizing the HindIII and
the BglII sites and replacing the original ubiquitin (Ubi)
promoter sequence. To clone HvPAO1 promoter sequence
[4], we used a different procedure because of the presence
of a BglII site within the gene sequence. A HvPAO1 pro-
moter sequence subclone inserted in pGEM-Teasy vector
was digested with EcoRI and filled in at its extremities, then
cloned in pHTT515 previously cut with HindIII and BglII
and blunt ended. The Ubipromoter::GUS expression
pHTT515 vector was used as a control plasmid expres-
sing the GUS reporter gene driven by the house-keeping
promoter Ubi. The HvPAO1, HvPAO2, ZmPAO1 and
ZmPAO2promoter::GUS expression plasmids were se-
quenced on both strands using the automated fluorescent

dye terminator technique (Perkin Elmer ABI model 373 A).
In order to clone the HvPAO1 cDNA (GenBank accession
number AJ298131) by PCR amplification, the full-length
cDNA was generated possessing modified 5¢- and 3¢-ends.
In particular, the two following synthetic oligonucleotides
were used to introduce NdeI and XhoI restriction sites at
the 5¢- and 3¢-ends of HvPAO1 cDNA: HvPAO1cdna-DIR,
5¢-CATATGGCCGGCCCCAGGGTCATCATC-3¢ and
HvPAO1cdna-REV, 5¢-CTGGAACTCGAGCTAGTCAA
ACTTGCCCGG-3¢
, respectively. The amplified PCR prod-
uct was restricted by NdeI and XhoI and ligated with the
restricted NdeI ⁄ XhoI pET17b vector, to obtain the genetic
construct encoding the mature form of HvPAO1 protein,
named pET17b-HvPAO1. The recombinant cDNA con-
struct was resequenced to check the accuracy of the nucleo-
tide sequence and then utilized to transform E. coli BL21
DE3 (Novagen, Madison, WI, USA) competent cells.
M. Cervelli et al. Barley polyamine oxidase isoform 1 gene expression
FEBS Journal 273 (2006) 3990–4002 ª 2006 The Authors Journal compilation ª 2006 FEBS 3999
Transient expression assay
Barley and maize transient transformation was performed
by particle bombardment with the expression plasmids
described above. In this study, the Biolistic PDS-1000 ⁄ He
Particle Delivery System from Bio-Rad (Bio-Rad, Milano,
Italy) was used. Cartridges containing plasmid-coated gold
microcarriers were prepared according to the manufac-
turer’s procedure with modification according to the plant
organ bombarded. Twenty-five micrograms of HvPAO1,
HvPAO2, ZmPAO1 and ZmPAO2promoter::GUS expres-

sion plasmids and the pHTT515 (Ubipromoter::GUS) con-
trol plasmid were precipitated onto 1.8 mg of gold particles
(1.0 lm in diameter, Bio-Rad) with ethanol. DNA-coated
gold particles were prepared according to the procedure
described by Sessa et al. [32] and finally suspended in 30 lL
of ethanol. Ten-microliter aliquots were used for one bom-
bardment. The barrel liner provided by the Biolistic PDS-
1000 ⁄ He system determined the distance from the gene gun
to the sample. This distance was 6.0 cm. Gold particles
coated with the plasmid of interest were used for bombard-
ment with a firing pressure of 900 psi for maize and barley
leaves and maize roots, 1.100 psi for barley roots and
developing caryopsis at 30-days postfertilization. After
bombardment, barley organs were placed on callus-induced
medium (4.3 g ÆL
)1
Murashige & Skoog medium (MS),
30.0 gÆL
)1
maltose, 1.0 mgÆL
)1
thiamine, 0.25 gÆL
)1
myo-
inositol, 1.0 gÆL
)1
hydrolyzed casein, 0.69 gÆL
)1
l-proline,
2.5 mgÆL

)1
dicamba, 5.0 mgÆL
)1
bialaphos, 3.5 gÆL
)1
phyta-
gel), while maize organs on 4.0 gÆL
)1
CHU(N
6
) medium
(Duchefa Biochemie B. V), then bombarded material
was incubated for 1-day at 25 °C in a growth chamber
with a 16 : 8 h light–dark programme. Three separate
bombardment experiments were carried out and a minimum
of three samples from each treated plant tissue was
examined.
Histochemical GUS activity assay
The histochemical localization of GUS in transiently trans-
formed plant organs was performed essentially as described
by Jefferson [33]. Plant samples from barley and maize were
immersed in a histochemical reaction mixture containing
3mgÆmL
)1
X-Glu (5-bromo-4-chloro-3-indolyl b-d-glucuro-
nide) in 100 mm sodium phosphate buffer, pH 7.0, 0.06%
Triton X-100, 5 mm potassium ferrocyanide. The histo-
chemical reaction was performed in the dark at 37 °C until
a blue indigo dye color as precipitate appeared at the site
of the enzymatic cleavage after 2–15 h. Roots and barley

immature caryopsis were rinsed several times in 50 mm
phosphate buffer to stop the reaction, finally kept in the
same buffer. After incubation, stained leaves were cleared
of chlorophyll by v ⁄ v 70% ethanol fixation. Samples were
examined for blue staining assessment with a Zeiss stereo-
microscope and photographed.
ZmPAO, HvPAO2 and HvPAO1 enzyme
purification
All purification procedures were performed at 0–5 °C. Low-
pressure chromatographic steps were carried out using the
GradiFrac System (Amersham Biosciences). As the three
ZmPAO genes encode for identical translation products [9],
it is not possible to distinguish the contribution of each iso-
form within the PAO enzymatic activity detectable in maize
and no numbering is used for ZmPAO enzymes. ZmPAO
protein was purified from shoots of 10-day-old etiolated
maize seedlings, as previously described [5] and HvPAO2
isoform was purified from shoots of 6-day-old de-etiolated
barley seedlings as previously described [4]. HvPAO1 pro-
tein was purified from 30-day postfertilization developing
caryopsis as follows. Plant material (100 g) was homogen-
ized in a Waring blender with three volumes of cold 0.5 m
NaH
2
PO
4
and the homogenate was filtered through nylon
mesh cloth. Filtrate was centrifuged at 20 000 g for 30 min
and the supernatant (236 mL, fraction 1) was fractionated
with 70% saturated ammonium sulfate and left 2 h at 4 °C.

The suspension was centrifuged at 20 000 g for 30 min. The
pellet was dissolved in 0.2 m NaH
2
PO
4
, dialyzed overnight
against the same buffer and then centrifuged at 20 000 g
for 30 min (104 mL, fraction 2). The dialyzed sample was
applied to a hydroxylapatite column equilibrated with the
dialysis buffer. The column was washed with 0.2 m
NaH
2
PO
4
containing NaCl 0.1 m, until the A
280
decreased
to zero. The enzyme was eluted with 0.5 m sodium phos-
phate buffer, pH 6.5, containing 0.1 m NaCl (44 mL, frac-
tion 3) and dialyzed overnight against 50 mm sodium
phosphate buffer, pH 5.0. The dialyzed sample was applied
to an SP-sepharose column equilibrated with the dialysis
buffer. The column was washed with 50 mm sodium phos-
phate buffer, pH 5.0, containing 0.1 m NaCl, until the A
280
decreased to zero. The enzyme was eluted with 50 mm
sodium phosphate buffer, pH 5.0, containing 0.3 m NaCl
(15 mL, fraction 4).
Enzyme assay, protein determination and
western blot analysis

Enzyme activity was measured spectrophotometrically by
following the formation of a pink adduct (e
515 nm
¼
2.6 · 10
4
m
)1
.cm
)1
) as a result of the H
2
O
2
-dependent
oxidation of 4-aminoantipyrine catalyzed by horseradish
peroxidase and the subsequent condensation of oxidized
4-aminoantipyrine with 3,5-dichloro-2-hydroxybenzenesulf-
onic acid [34]. The measurements were performed in 0.2 m
sodium phosphate buffer, pH 6.5, with Spd as a substrate
for ZmPAO; in 0.2 m sodium phosphate buffer, pH 5.5,
with Spm as a substrate for HvPAO2; in 0.2 m sodium
phosphate buffer, pH 7.0, with Spd as a substrate for
HvPAO1. Enzyme activity was expressed in IU. Protein
content was estimated by the method of Bradford [35]
Barley polyamine oxidase isoform 1 gene expression M. Cervelli et al.
4000 FEBS Journal 273 (2006) 3990–4002 ª 2006 The Authors Journal compilation ª 2006 FEBS
with bovine albumin as a standard. SDS ⁄ PAGE was
performed according to the method of Laemmli [36]. West-
ern blot analysis was performed after protein deglycosyla-

tion [37], in order to eliminate cross-reactivity due to
glycan moieties, using a rabbit polyclonal anti-ZmPAO
antiserum [38].
Analysis of catalytic properties
ZmPAO, HvPAO2 and HvAO1 pH optima for catalytic
activity were determined at 25 ° C, in 0.2 m sodium phos-
phate buffer (pH range 4.5–8.5) with Spd or Spm as sub-
strates. K
m
values were determined for the three enzymes
(1 · 10
)3
U) at 25 °C, with Spd and Spm as substrates at
the respective pH optimum and calculated from Lineweav-
er-Burk plots. Spd or Spm concentrations range was from
4.0 to 20.0 · 10
)6
m for ZmPAO enzyme and from 2.0 to
8.0 · 10
)6
m for HvPAO1 enzyme. In the case of HvPAO2
enzyme, Spd concentration range was from 3 to
20 · 10
)4
m, while Spm concentration range was from 5.0
to 60.0 · 10
)7
m. Data here reported are the average of
three different experiments, each with two replicates. sd
was 8%.

Reaction conditions for in vitro HvPAO1 protein
synthesis
For in vitro protein synthesis, we followed the standard
reaction conditions that were basically described in the
manufacturer’s instructions. In the RTS-100 (Roche Diag-
nostics, Monza, Italy) case, the standard reaction solution
(50 lL) contained the following components: 12 lL E. coli
lysate, 10 lL reaction mix, 12 lL amino acids, 1 lL methi-
onine, 5 lL reconstitution buffer and 2.5 lg of circular
template (pET17b-HvPAO1 plasmid) in 10 lL of water.
The reaction mixtures were incubated at 30 °C for 3 h. The
E. coli T7 S30 Extract System for circular DNA (Promega
Italia, Milano, Italy) contained the following components:
4 lg of DNA template, 5 lL amino acid mixture, 20 lL
S30 premix, 15 lL T7 S30 extract, and nuclease free water,
which came to a final volume of 50 lL. The reaction incu-
bation was carried out at 37 ° C for 3 h. In the Pure-System
Classic Mini Kit (Post Genome Institute, Tokyo, Japan),
the reaction mixtures were prepacked in two tubes, Solu-
tions A and B. Before the reaction, we thawed both tubes
on ice and, using new tubes we gently mixed 0.5 lgof
DNA template, 25 lL of Solution A, 10 lL of Solution B
and added RNase free water to a final volume of 50 lL.
The tubes were then placed in an incubator for 3 h at
37 °C. In order to improve the enzyme activity we add fla-
vin adenine dinucleotide (15 mgÆL
)1
final concentration) in
each tube. In vitro synthesized HvPAO1 protein was ana-
lyzed for enzymatic activity and catalytic properties as des-

cribed above.
Acknowledgements
This research was partially supported by the grant
PRIN 2005 from ‘Ministero dell’Istruzione, dell’Uni-
versita
`
e della Ricerca’ (MIUR).
References
1 McIntire WS & Hartman C (1993) Principle and Appli-
cation of Quinoproteins (Davison, VL, eds), pp. 97–171.
Marcel Dekker Inc., New York.
2 Seiler N (1995) Polyamine oxidase, properties and func-
tions. Prog Brain Res 106, 333–344.
3 Cona A, Rea G, Angelini R, Federico R & Tavladoraki
P (2006) Functions of amine oxidases in plant develop-
ment and defence. Trends Plant Sci 11, 80–88.
4 Cervelli M, Cona A, Angelini R, Polticelli F, Federico
R & Mariottini P (2001) A barley polyamine oxidase
isoform with distinct structural features and sucellular
localization. Eur J Biochem 268, 3816–3830.
5 Federico R, Alisi C & Forlani F (1989) Properties of
polyamine oxidase from the cell wall of maize seedlings.
Phytochemistry 28, 45–46.
6 Tavladoraki P, Schinina
`
ME, Cecconi F, Di Agostino
S, Manera F, Rea G, Mariottini P, Federico R &
Angelini R (1998) Maize polyamine oxidase: primary
structure from protein and cDNA sequencing. FEBS
Lett 426, 62–66.

7 Binda C, Coda A, Angelini R, Federico R, Ascenzi P &
Mattevi A (1999) A 30 A
˚
long U-shaped catalytic tunnel
in the crystal structure of polyamine oxidase. Structure
7, 265–276.
8S
ˇ
ebela M, Radova
´
A, Angelini R, Tavladoraki P, Fre
´
-
bort I & Peˆ c P (2001) FAD-containing polyamine oxi-
dases: a timely challenge for researchers in biochemistry
and physiology of plants. Plant Sci 160, 197–207.
9 Cervelli M, Tavladoraki P, Di Agostino S, Angelini R,
Federico R & Mariottini P (2000) Isolation and charac-
terization of three polyamine oxidase genes from Zea
mays. Plant Physiol Biochem 38, 667–677.
10 Cervelli M, Di Caro O, Di Penta A, Angelini R, Fede-
rico R, Vitale A & Mariottini P (2004) A novel C-term-
inal sequence from barley polyamine oxidase is a
vacuolar sorting signal. Plant J 40, 410–418.
11 Li ZC & McClure JW (1989) Polyamine oxidase of pri-
mary leaves is apoplastic in oat but symplastic in barley.
Phytochemistry 28, 2255–2259.
12 Federico R & Angelini R (1986) Occurrence of diamine
oxidase in the apoplast of pea epicotyls. Planta 167,
300–302.

13 Cona A, Cenci F, Cervelli M, Federico R, Mariottini P,
Moreno S & Angelini R (2003) Polyamine oxidase, a
hydrogen peroxide-producing enzyme, is up-regulated
by light and down-regulated by auxin in the outer
M. Cervelli et al. Barley polyamine oxidase isoform 1 gene expression
FEBS Journal 273 (2006) 3990–4002 ª 2006 The Authors Journal compilation ª 2006 FEBS 4001
tissues of the maize mesocotyl. Plant Physiol 131, 803–
813.
14 Cooley T & Walters D (2002) Polyamine metabolism in
barley reacting hypersensitively to the powdery mildew
fungus Lumeria graminis f.sp. hordei. Plant Cell Environ
25, 461–468.
15 Yoda H, Yamaguchi Y & Sano H (2003) Induction of
hypersensitive cell death by hydrogen peroxide produced
through polyamine degradation in tobacco plants. Plant
Physiol 132, 1973–1981.
16 Laurenzi M, Rea G, Federico R, Tavladoraki P &
Angelini R (1999) De-etiolation causes a phytochrome-
mediated increase of polyamine oxidase expression in
outer tissues of the maize mesocotyl: a role in the
photomodulation of growth and cell wall differentiation.
Planta 208, 146–154.
17 Cona A, Moreno S, Cenci F, Federico R & Angelini R
(2005) Cellular re-distribution of flavin-containing poly-
amine oxidase in differentiating root and mesocotyl of
Zea mays L. seedlings. Planta 221, 265–276.
18 Fry SC, Willis SC & Paterson AE (2000) Intraproto-
plasmic and wall-localised formation of arabinoxylan-
bound diferulates and larger ferulate coupling-products
in maize cell-suspension cultures. Planta 211, 679–

692.
19 Higo K, Ugawa Y, Iwamoto M & Higo H (1998)
PLACE: a database of plant cis-acting regulatory DNA
elements. Nucleic Acids Res 26, 358–359.
20 Argu
¨
ello-Astrolla GR & Herrera-Estrella LR (1998)
Evolution of light-regulated plant promoter. Annu Rev
Plant Physiol Plant Mol Biol 49, 525–555.
21 Terzaghi W & Cashmore AR (1995) Light-regulated
transcription. Annu Rev Plant Physiol Plant Mol Biol
46, 445–474.
22 Yazaki K, Shimatani Z, Hashimoto A, Nagata Y, Fujii
F, Kojima K, Suzuki K, Taya T, Tonouchi M, Nelson
C et al. (2004) Transcriptional profiling of genes respon-
sive to abscisic acid and gibberellin in rice: phenotyping
and comparative analysis between rice and Arabidopsis.
Physiol Genomics 17, 87–100.
23 Walters D, Cowley T & Mitchell A (2002) Methyl jas-
monate alters polyamine metabolism and induces sys-
temic protection against powdery mildew infection in
barley seedlings. J Exp Bot 53, 747–756.
24 Asthir B, Duffus CM, Smith RC & Spoor W (2002)
Diamine oxidase is involved in H
2
O
2
production in the
chalazal cells during barley filling. J Exp Bot 369, 677–
682.

25 Bethke PC & Jones RL (2001) Cell death of barley
aleurone protoplasts is mediated by reactive oxygen spe-
cies. Plant J 25, 19–29.
26 Palma K & Kermode AR (2003) Metabolism of hydro-
gen peroxide during reserve mobilization and pro-
grammed cell death of barley (Hordeum vulgare L.)
aleurone layer cells. Free Radic Biol Med 35, 1261–
1270.
27 Paschalidis KA & Roubelakis-Angelakis KA (2005)
Sites and regulation of polyamine catabolism in the
tobacco plant. Correlations with cell division ⁄ expansion,
cell cycle progression, and vascular development. Plant
Physiol 1382174, –2184.
28 Tun NN, Santa-Catarina C, Begum T, Silveira V,
Handro W, Floh EI & Scherer GF (2006) Polyamines
induce rapid biosynthesis of nitric oxide (NO) in Arabi-
dopsis thaliana seedlings. Plant Cell Physiol 47, 346–354.
29 Beligni MV, Fath A, Bethke PC, Lamattina L & Jones
RL (2002) Nitric oxide acts as an antioxidant and
delays programmed cell death in barley aleurone layers.
Plant Physiol 129, 1642–1650.
30 Delogu G, Terzi V, Cattivelli L & Stanca AM (1993) Le
Varieta
`
Di Orzo Cultivate in Italia, Editoriale Bortolazzi
Stei, San Giovanni Lupatoto, Verona.
31 Sambrook J, Fritsch EF & Maniatis T (1989) Molecular
Cloning: a Laboratory Manual, 2nd edn. Cold Spring
Harbor Laboratory, Cold Spring Harbor, NY.
32 Sessa G, Borello U, Morelli G & Ruberti I (1998) A

transient assay for rapid functional analysis of transcrip-
tion factors in Arabidopsis. Plant Mol Biol Report 16,
191–197.
33 Jefferson RA, Kavanaghi TA & Bevan MW (1987)
GUS fusions: beta-glucuronidase as a sensitive and ver-
satile gene fusion marker in higher plants. EMBO J 6,
3901–3907.
34 Artiss JD & Entwistle WM (1981) The application of a
sensitive uricase-peroxidase couple reaction to a
centrifugal fast analyser for the determination of uric
acid. Clinica Chimica Acta 116, 301–309.
35 Bradford MM (1976) A rapid and sensitive method for
the quantitation of microgram quantities of protein uti-
lising the principle of protein-dye binding. Anal Biochem
72, 248–254.
36 Laemmli UK (1970) Cleavage of structural proteins dur-
ing assembly of the head of bacteriophage T4. Nature
277, 680–685.
37 Woodward MP, Young WW & Bloodgood RA (1985)
detection on monoclonal antibodies specific for carbo-
hydrate epitopes using periodate oxidation. J Immunol
Methods 78, 143–153.
38 Federico R, Pini C, Raiola A, Tinghino R & Angelini
R (1995) Immunological cross-reactivity of plant glyco-
proteins: maize polyamine oxidase, a cell wall protein.
Physiol Mol Biol Plants 1, 5–12.
Barley polyamine oxidase isoform 1 gene expression M. Cervelli et al.
4002 FEBS Journal 273 (2006) 3990–4002 ª 2006 The Authors Journal compilation ª 2006 FEBS