Tree Physiology 23, 443–452
© 2003 Heron Publishing—Victoria, Canada
Differences in wound-induced changes in cell-wall peroxidase activities
and isoform patterns between seedlings of Prosopis tamarugo and
Prosopis chilensis
GABRIELE LEHNER1–3 and LILIANA CARDEMIL1
1
Departamento de Biología, Facultad de Ciencias, Universidad de Chile, Santiago, Chile
2
Present address: GSF-National Research Center for Environment and Health, Institute of Soil Ecology, Environmental Engineering, Neuherberg,
Germany
3
Author to whom correspondence should be addressed ([email protected])
Received August 6, 2002; accepted October 9, 2002; published online April 1, 2003
Summary We determined changes in cell-wall peroxidase
activities and isoform patterns in response to wounding in
seedlings of Prosopis tamarugo Phil. (an endemic species of
the Atacama Desert) and Prosopis chilensis (Mol.) Stuntz (a
native species of central Chile), to assess tolerance to predation. In seedlings of both species, the maximal increase in
peroxidase activity occurred 48 h after wounding, reaching
three times the control value in P. tamarugo and twice the control value in P. chilensis. The activity of ionically bound
cell-wall peroxidases increased only locally in wounded embryonic axes, whereas the activity of soluble peroxidases increased systemically in unwounded cotyledons. Analysis of
ionic peroxidases by isoelectrofocusing revealed two groups of
peroxidases in the cell walls of both species: four distinct acidic
isoforms and a group of basic isoforms. In response to wounding, there was a large increase in activity of the acidic isoforms
in P. tamarugo, whereas there was an increase in the activity of
the basic isoforms in P. chilensis. In P. chilensis, the wound-induced increase in activity of the basic isoforms corresponded
with one of the two isoforms detected in P. tamarugo prior to
wounding. Experiments with protein and RNA synthesis inhibitors indicated that a preexisting basic peroxidase is activated
in P. chilensis after wounding. Assays of ionically bound
peroxidase activity with four different substrates corroborated
the differences found in isoform patterns between species. In
P. tamarugo, the largest increases in activity were found with
ortho-phenylenediamine and ferulic acid as substrates,
whereas in P. chilensis the largest increase in activity was found
with guaiacol as substrate. Because the same basic cell-wall
peroxidase that accumulated after wounding in P. chilensis was
present in P. tamarugo prior to wounding, and the activity of
acidic cell-wall peroxidases increased after wounding in
P. tamarugo but not in P. chilensis, we conclude that
P. tamarugo is more tolerant to wound stress than P. chilensis.
Keywords: cell wall reinforcement, coniferyl alcohol, cordycepin, cycloheximide, ferulic acid, guaiacol, o-PDA, predation.
Introduction
Plants are exposed to many types of mechanical stress caused
by biotic and abiotic factors. Biotic causes of mechanical injury include insect attack, parasite infestation and the many organisms that live or feed on plants. Abiotic causes of mechanical injury include wind and fires (Macheix et al. 1986). All
plants have developed mechanisms to protect themselves from
mechanical injury. Desert plants, which are heavily preyed
upon by animals, are especially well protected from mechanical injury.
Cell walls are the primary site of mechanical injury, and
many biochemical and structural responses to wounding lead
to the strengthening of cell walls. Cell walls also play important roles in recognition of pathogens and signaling of external
stimuli to the cell (e.g., Bowles 1998).
The cell wall can be strengthened by deposition of structural
proteins such as the hydroxyproline-rich and proline-rich proteins (Rodríguez and Cardemil 1994, Cassab 1998), and by deposition of other polymers such as lignin, callose and suberin
(Espelie et al. 1986, Chaman et al. 2001). The proline-rich and
hydroxyproline-rich proteins are cross-linked with themselves
and with other cell wall components. One form of cross-linkage among hydroxyproline-rich proteins is believed to be the
isodityrosine bridge, which is synthesized in situ by a cell-wall
peroxidase (Fry 1982, Cassab 1998). The synthesis of lignin
and suberin as well as the cross-linking of these substances
with other wall polymers is also catalyzed by cell-wall
peroxidases (Espelie et al. 1986, McDougall 1991, 1993).
Therefore, peroxidases are key enzymes for cell wall reinforcement (Chaman et al. 2001) and may also play an important role as intermediaries in signal cascades triggered by
external stimuli to the cell wall (Brownleader et al. 1997, Kiba
et al. 1997). A signaling role of peroxidases already has been
demonstrated in the hypersensitive response (Bolwell et al.
1995, Bestwick et al. 1998).
Peroxidases are heme enzymes that use H2O2 as an electron
acceptor to catalyze oxidative reactions with a variety of sub-
444
LEHNER AND CARDEMIL
strates (Li and Poulos 1994). Peroxidases are present in all organisms and in nearly all tissues and cell compartments
(McDougall 1991). Plants possess many different peroxidases
that are involved in, for example, general pathogen defense
mechanisms, cell wall reinforcement, ethylene biosynthesis,
plant growth regulation, and phenol and H2O2 metabolism
(Douroupi and Margaritis 2000).
To date, more than 60 plant peroxidase genes have been
identified, and about 20–30 more are expected to be found
with the help of ongoing gene projects such as the Arabidopsis
Genome Initiative (Tognolli et al. 2000). Many of the plant
cell-wall peroxidases belong to the type III peroxidases, a gene
family with a large number of isoenzymes with similar catalytic properties (Welinder 1992, Penel et al. 2000).
Prosopis chilensis (Mol.) Stuntz and Prosopis tamarugo
(Phil.) are native leguminous trees of Chile. Prosopis chilensis
is found in the semiarid region of central and northern Chile
(23–34°8′ S), whereas P. tamarugo is found in the Atacama
Desert (20°17′–20°50′ S), where it is the only native tree species and is exposed to heavy predation (Lehner et al. 2001). In
the natural environment of P. chilensis, in contrast, vegetation
is more abundant and diversified, so predatory stress may be
less severe for this species than for P. tamarugo. We predict,
therefore, that P. tamarugo is better adapted to wound stress
than P. chilensis, and is better able to reinforce its cell walls
against injury. Our study objective was to compare the magnitude of cell-wall peroxidase accumulation in these Prosopis
spp. in response to wounding.
Materials and methods
Plant material and wound stress treatment of seedlings
In autumn of 1996, 1997 and 1998, fruits of P. chilensis were
collected in central Chile, near Colina, 35 km east of Santiago,
and fruits of P. tamarugo were collected in Canchones, part of
the Pampa del Tamarugal in the Atacama Desert, about 90 km
east of Iquique. Seeds were scarified and germinated as described by Rodríguez and Cardemil (1994).
Seedlings of P. chilensis and P. tamarugo were wounded 48
and 72 h after imbibition, respectively, by making five transverse cuts with a razor blade in both the hypocotyl and root of
each seedling. After wounding, seedlings were incubated for
the indicated times in growth chambers at 25 °C with a 12-h
photoperiod. Experiments were repeated three times and the
data shown are experiment means (with standard deviation).
Statistical analyses (t-tests) were performed with Sigma Stat
software (SPSS, Chicago, IL).
Antibiotic experiments
Wounded and unwounded seedlings were sprayed with
5 µg ml–1 cycloheximide in Buffer A (0.05 M Tris-HCl,
pH 7.0, containing 0.1% Tween-20) immediately after wounding and then twice a day until harvest (Nair and Showalter
1996). For the experiment with cordycepin, seedlings assigned
to the wounding treatment were sprayed with 0.4 mM
cordycepin in Buffer A for 2 h prior to wounding and then
twice a day until harvest (Ho and Varner 1976). Control seedlings were sprayed with buffer at the same times.
Cell wall preparation
Roots and hypocotyls (embryonic axes) of wounded and unwounded seedlings were separated from the cotyledons and
from each other and processed separately for preparation of
cell walls by the method of Cassab et al. (1985). For each cell
wall preparation, 5 g of tissue, corresponding to about 100
seedlings, was used. The supernatant from the first centrifugation step was designated the soluble protein fraction. Purity of cell walls was determined by assaying three marker
enzymes for cellular membranes (cytochrome c oxidase,
NADH cytochrome c reductase and 5′-AMP-nucleotidase) according to Briskin et al. (1987). Activities of these enzymes
were below the detection limit in purified cell wall preparations.
Cell-wall protein extraction
Ionically bound cell-wall proteins were extracted by incubating purified cell walls with 0.2 M CaCl2, 5 mM benzamidine
and 2 mM Na2S2O5, pH 5.0, first for 3 h, and then overnight.
Extracts were desalted and concentrated using Centricon filters (Amicon, Beverly, MA).
Protein and peroxidase activity assays
Protein concentration was determined by the method of Bradford (1976) at 595 nm with bovine serum albumin as a standard. All determinations were done in duplicate.
Peroxidase activity was determined spectrophotometrically
(Shimadzu UV-240, Kyoto, Japan) with ortho-phenylenediamine (o-PDA) as substrate. The assay buffer contained
5 mM o-PDA in 0.1 M sodium citrate, pH 4.5; 0.44 mM H2O2
was added immediately before starting the assay by adding
5 µl of sample. Product formation was followed at 450 nm.
Peroxidase activity is expressed as amount of product formed
per minute. This was calculated with an extinction coefficient
of 1.01 mM –1 cm –1, determined from a calibration curve obtained with horseradish peroxidase at saturating H2O2. All determinations of peroxidase activity were done in triplicate.
Besides o-PDA, natural peroxidase substrates were used for
some assays. Ferulic acid (50 µM) and p-coumaric acid
(35 µM) were dissolved in 0.1 M sodium citrate, pH 4.5, and
guaiacol (10 mM) was prepared in 0.1 M phosphate buffer,
pH 6.0. Hydrogen peroxide was also added prior to determination. For ferulic acid and p-coumaric acid, substrate oxidation
was followed at the respective absorption maxima, and
guaiacol product formation was followed at 470 nm. Extinction coefficients for peroxidase activity calculations were
taken from Gonzalez et al. (1999): ferulic acid: 16 mM –1 cm –1
at 310 nm; p-coumaric acid: 19 mM –1 cm –1 at 286 nm; and
guaiacol: 26.6 mM –1 cm –1 at 470 nm.
Electrophoretic analysis
Protein samples were incubated with electrophoresis buffer
for 1 h at 37 °C to avoid denaturation of the peroxidase en-
TREE PHYSIOLOGY VOLUME 23, 2003
WOUND STRESS AND CELL-WALL PEROXIDASES IN TWO PROSOPIS SPECIES
zymes, because some gels were stained for peroxidase activity
after separation. The SDS-PAGE was conducted in a Mini Protean II Cell (Bio-Rad, Richmond, VA) as described by
Laemmli (1970) with a 4% stacking and a 10% separating gel.
Ten micrograms of protein per well was loaded on the gel.
Isoelectrofocusing (IEF) was performed in a Mini IEF Cell,
Model 111 (Bio-Rad) according to the manufacturer’s instructions. One microgram of protein was loaded for each sample.
Gels were stained with Coomassie Blue R250 or silver
stained (Heukeshoven and Dernick 1985). To stain IEF gels
for peroxidase activity, immediately after IEF, the gels were
placed in buffer containing 5 mM 4-chloro-1-naphthol, 20%
methanol and 0.04% H2O2 in 0.1 M sodium citrate, pH 4.5 or
5 mM o-PDA and 0.44 mM H2O2 in 0.1 M sodium citrate,
pH 4.5, until peroxidase bands developed. To stop the reaction,
the gels were washed thoroughly with distilled water. For
SDS-PAGE gels, SDS was removed by the method of Mitsui et
al. (1993) before staining for peroxidase activity. Gels were
then stained as described for IEF gels.
Stained gels were photographed and the scanned photographs were analyzed densitometrically with Sigma Scan
(SPSS, Chicago, IL). All electrophoretic analyses were conducted in duplicate.
Separation of acidic and basic peroxidases
Extracts of ionically bound proteins were desalted by dialysis
against 0.05 M Tris-HCl, pH 6.8, and acidic and basic fractions were separated by preparative IEF in a Rotofor Cell
(Bio-Rad) according to the manufacturer’s recommendations.
To remove ampholytes after the separation, samples were
brought to 1 M NaCl, stirred for 15 min and dialyzed against
0.05 M Tris-HCl, pH 6.8. Peroxidase fractions yielded by this
method were used for measurements of activity with different
substrates and for further analysis of the basic isoenzymes of
P. chilensis.
To correlate isoelectric point and molecular mass of peroxidase isoforms, step-wise ammonium sulfate fractionation at
20, 40, 60, 80 and 100% was performed as described by Harris
and Angal (1989). Ammonium sulfate was dissolved in the
sample on ice and the samples were stirred for 1 h at 4 °C before centrifugation. Precipitates were resuspended in 0.1 M
Tris-HCl, pH 8.8, and electrophoretic analyses were conducted on all fractions and the supernatant of the 100% precipitation.
Results
Total ionically bound cell-wall peroxidase activity
In both species, there was a clear increase in ionically bound
cell-wall peroxidase activity in response to wounding (Figure 1). This increase was maximal 48 h after wounding, with
activities twice (P. chilensis) and three times (P. tamarugo) the
activities of the corresponding unwounded seedlings. The increase in peroxidase activity of wounded seedlings was first
observed 6 h after wounding in P. chilensis and 12 h after
wounding in P. tamarugo.
445
Figure 1. Relative activity of ionically bound cell-wall peroxidases.
The activity of ionic peroxidases relative to the activity of the unwounded controls is shown at different times after wounding in
P. chilensis (A) and P. tamarugo (B). Activities before wounding in
P. chilensis and P. tamarugo were 0.670 and 0.745 µmol o-PDA min –1
µg –1 protein, respectively (n = 3, ± SD).
Soluble peroxidase activity increased not only locally in the
wounded embryonic axis, but also systemically in unwounded
cotyledons of both species (Figures 2A and 2C). Ionically
bound peroxidase activity increased only locally in the embryonic axis 48 h after wounding (Figure 2B). There was no increase in ionically bound peroxidase activity in unwounded
cotyledons (Figure 2D).
Isoforms of ionic peroxidases
Analysis of ionic peroxidase isoforms by IEF over a pH range
of 3–10 revealed two groups of peroxidases in both species: a
group of four acidic isoforms and a group of basic isoforms
that could not could be resolved in this pH range (Figures 3A
and 3B). Densitometric analysis of the acidic and basic bands
indicated a strong increase in the amount of acidic isoforms
and a slight increase in the amount of basic isoforms in
P. tamarugo in response to wounding (Figure 3C). In
P. chilensis, wounding caused an increase in the amount of basic isoforms only (Figure 3D).
Analysis of the basic peroxidase isoforms by IEF in the pH
range 8–10.5 corroborated the difference in wound-induced
isoform patterns between species (Figure 4). In P. tamarugo,
two basic isoforms were distinguished in extracts of unwounded and wounded seedlings. In P. chilensis, there was
only one basic isoform in extracts of unwounded seedlings, but
two basic isoforms could be distinguished in extracts of
wounded seedlings. In P. tamarugo, no new isoforms were detected after wounding.
TREE PHYSIOLOGY ONLINE at http://heronpublishing.com
446
LEHNER AND CARDEMIL
Figure 2. Total specific activity of soluble (A, C) and ionically bound
(B, D) peroxidases in wounded embryonic axes (A, B) and unwounded cotyledons (systemic tissue, C, D) of P. chilensis and
P. tamarugo (n = 3, ± SD).
Figure 4. Analyses by IEF performed in the pH range 8–10.5 (A) followed by the densitometry of the bands (B) of the basic peroxidases
from P. chilensis and P. tamarugo. The densitometric analysis results
are shown as relative intensities. Abbreviations: M = markers; U = unwounded; and W = wounded. The arrow indicates the peroxidase
band induced by wounding in P. chilensis.
Figure 3. Accumulation of acidic and basic isoforms of cell-wall
peroxidases. Analyses were performed by IEF in the pH range 3–10
(A, B) followed by densitometry of the bands (C, D). Four acidic
isoforms (upper arrows) and a group of unresolved basic isoforms
(lower arrows) were detected in P. tamarugo (A) and P. chilensis (B).
The activity is relative to the values of the unwounded controls of
P. tamarugo (C) and P. chilensis (D). Abbreviations: M = markers;
U = unwounded; and W = wounded.
Correlation between isoelectric points and molecular
masses of ionic peroxidase isoenzymes showed that, in both
species, the acidic isoforms had higher molecular masses than
the basic isoforms. Staining of acid fractions for peroxidase
activity in the SDS-PAGE gels showed no distinct bands but
rather a broad smear of stain that may be associated with
glycosylation (gels not shown). The molecular masses of the
acidic isoforms were between 72 and 87 kDa in P. chilensis
and between 60 and 72 kDa in P. tamarugo (Table 1). Basic
peroxidase isoforms had molecular masses of 35 and 36 kDa
in both species (Table 1).
Activities of acidic and basic isoforms with different
substrates
After separation of the acidic and basic isoforms by preparative IEF, ionically bound peroxidase activity was assayed with
four different substrates, including the natural peroxidase substrates ferulic acid and p-coumaric acid (Figure 5). In
P. chilensis, only basic peroxidase activity increased markedly
in response to wounding, regardless of the substrate used (Figures 5A and 5C). However, the biggest increase in basic
peroxidase activity (seven times higher than the unwounded
control activity) was found when guaiacol was used as substrate (Figure 5C). Guaiacol corresponds to the phenolic part
TREE PHYSIOLOGY VOLUME 23, 2003
WOUND STRESS AND CELL-WALL PEROXIDASES IN TWO PROSOPIS SPECIES
Table 1. Isoelectric points and molecular masses of the ionically
bound cell-wall peroxidases from P. chilensis and P. tamarugo.
Species
Molecular mass range (kDa)
Isoelectric point
P. chilensis
72 (± 2)–87 (± 5)
≥ 5.0 ± 0.1
≥ 5.2 ± 0.3
≥ 5.8 ± 0.1
≥ 6.3 ± 0.1
≥ 10.0
35 (± 1), 36 (± 1)
P. tamarugo
60 (± 2)–72 (± 2)
35 (± 1), 36 (± 1)
≥ 4.4 ± 0.2
≥ 4.6 ± 0.3
≥ 4.7 ± 0.3
≥ 5.0 ± 0.3
≥ 10.0
of coniferylalcohol, a natural peroxidase substrate and monomer of lignin (Penel et al. 2000).
In P. tamarugo, acidic peroxidase activity increased in response to wounding with all substrates tested (Figure 5B). The
biggest increase was found with o-PDA (seven times higher
than the unwounded control activity), an artificial peroxidase
substrate, and ferulic acid (six times higher than the unwounded control activity). Ferulic acid is found in cell walls as
part of several polymers (pectin, hemicellulose, lignin) (Fry
1986, Penel et al. 2000). In P. tamarugo, basic peroxidase activity increased in response to wounding only when o-PDA
was the substrate (Figure 5D). However, basic peroxidase activity in unwounded seedlings of P. tamarugo was as high or
higher than basic peroxidase activity in wounded seedlings of
P. chilensis. This difference was observed with all substrates
tested and corroborates the finding that two basic peroxidases
were present in unwounded seedlings of P. tamarugo.
Figure 5. Specific peroxidase activities of acidic (A, B) and basic (C,
D) fractions of ionically bound peroxidases from unwounded and
wounded seedlings of P. chilensis (A, C) and P. tamarugo (B, D). The
activities were assayed with four different peroxidase substrates:
(1) 50 µM ferulic acid; (2) 35 µM p-coumaric-acid; (3) 10 mM
guaiacol; and (4) 5 mM o-PDA. The values for o-PDA have to be multiplied by 10 2 (*).
447
Effects of wounding and antibiotics on basic peroxidase
activity in P. chilensis
Analysis by SDS-PAGE of samples taken at different times after wounding showed that the wound-induced basic peroxidase could be detected 3 h after wounding (Figure 6), but not as
early as 2 h after wounding.
To determine if the appearance of the basic peroxidase was
the result of activation of a preexisting inactive isoform, we
conducted inhibitor studies with cycloheximide, a protein synthesis inhibitor, and with cordycepin, an RNA synthesis inhibitor. Unwounded and wounded seedlings sprayed with 5 µg
ml –1 cycloheximide had a lower protein concentration 48 h after wounding than unwounded and wounded seedlings not
sprayed with cycloheximide (unwounded and wounded controls) (Figure 7A). Also, peroxidase activity was slightly lower
in cycloheximide-treated seedlings than in control seedlings.
However, wounded cycloheximide-treated seedlings showed a
clear increase in peroxidase activity compared with the unwounded cycloheximide-treated seedlings, as did the
wounded control seedlings compared with the unwounded
control seedlings (Figure 7B). Analysis by IEF showed that
the increase in peroxidase activity was the result of an increase
in the activity of basic isoforms in wounded cycloheximidetreated seedlings and in wounded controls (Figure 7C).
Wounded seedlings sprayed with cordycepin had a higher
protein concentration 48 h after wounding than wounded seedlings not sprayed with cordycepin (wounded controls) (Figure 8A). Ninety-six hours after wounding, however, protein
concentration in wounded cordycepin-treated seedlings decreased below that in unwounded seedlings not sprayed with
cordycepin (unwounded controls), whereas protein concentration increased further in wounded control seedlings.
Peroxidase activity increased in wounded, cordycepintreated seedlings 48 h after wounding compared with the unwounded and wounded control seedlings (Figure 8B). However, 96 h after wounding, peroxidase activity decreased in
wounded, cordycepin-treated seedlings to the value in the unwounded control seedlings, whereas it increased further in the
wounded control seedlings.
Figure 6. Time course of appearance of ionically bound basic
peroxidases. The analysis was performed by SDS-PAGE of extracts
obtained from unwounded and wounded seedlings of P. chilensis at
different times after wounding. Times in h are given above the gel
lanes. Numbers at the right are molecular masses (kDa). An asterisk
(*) denotes wounded plants and M denotes molecular weight markers.
The arrow indicates the peroxidase band induced by wounding in
P. chilensis.
TREE PHYSIOLOGY ONLINE at http://heronpublishing.com
448
LEHNER AND CARDEMIL
Figure 7. Protein concentration, peroxidase activity and pattern of
peroxidase isoforms of cycloheximide-treated, unwounded and
wounded seedlings. Protein concentration (A) and ionically bound
peroxidase activity (B) 48 h after treatment are shown as values relative to those of untreated, unwounded controls. The absolute values of
unwounded controls were 0.03 µg protein mg−1
FW (A) and 0.04 µmol
o-PDA min –1 mg−1
FW (B). The pattern of isoforms was obtained by IEF
performed in the pH range 3–10 (C). Abbreviations: U = unwounded
seedlings; W = wounded seedlings; *U = unwounded, cycloheximide-treated seedlings; *W = wounded, cycloheximide-treated seedlings; and M = molecular weight markers. The arrow points to the
basic peroxidases in P. chilensis.
Analysis by IEF of peroxidase isoforms 48 h after wounding
showed that the increase in peroxidase activity of wounded
cordycepin-treated seedlings was the result of an increase in
the amount of the acidic isoforms (Figure 8C). Ninety-six
hours after wounding, basic peroxidases disappeared completely in wounded, cordycepin-treated seedlings, whereas
acidic peroxidases remained unchanged. At this time, there
was still a greater amount of basic isoforms in wounded controls than in unwounded controls.
Discussion
Effects of wounding on ionically bound cell-wall peroxidases
of Prosopis spp.
Increases in plant peroxidase activity have been found in response to various stresses in different species and cell fractions
(Smith and Hammerschmidt 1988, Loukili et al. 1997, Bestwick et al. 1998, Cipollini 1998, Wang and Liu 1999). Increases in cell-wall-bound peroxidases also have been
reported (Castillo et al. 1984, Espelie et al. 1986, Lee and Lin
Figure 8. Protein concentration, peroxidase activity and pattern of
peroxidase isoforms of cordycepin-treated, wounded seedlings. Protein concentration (A) and ionically bound peroxidase activity (B) 48
and 96 h after treatment are shown as values relative to those of untreated, unwounded controls. The absolute values of unwounded con–1
trols were 0.06 µg protein mg−1
FW (A) and 0.08 µmol o-PDA min
(B).
The
pattern
of
isoforms
was
obtained
by
IEF
performed
in
mg−1
FW
the pH range 3–10 (C). Abbreviations: U = unwounded seedlings;
W = wounded seedlings; *W = wounded, cordycepin-treated seedlings; and M = molecular weight markers. The arrow points to the basic peroxidases in P. chilensis.
1995). The broad range of peroxidase responses to different
stimuli is probably associated with the large number of
isoforms of peroxidases and their numerous functions in the
cell (Campa 1991). The wound-induced increases in cell-wall
peroxidases that we observed in P. chilensis (twofold) and
P. tamarugo (threefold) are within the range of increases in
peroxidase activity found in other plants (Castillo et al. 1984,
Espelie et al. 1986, Loukili et al. 1997, Cipollini 1998, Wang
and Liu 1999). The wound-induced increase in activity was
detectable 6 h after wounding in P. chilensis and 12 h after
wounding in P. tamarugo. Because P. tamarugo is exposed to
greater predatory stress in its native habitat than P. chilensis,
we predicted a faster response to wounding in P. tamarugo, but
this was not the case. However, the initial absolute peroxidase
activity in P. tamarugo was slightly higher than in P. chilensis,
and the wound-induced increase in peroxidase activity was
also higher in P. tamarugo.
The wound-induced increases in peroxidase activity in
P. tamarugo and P. chilensis occurred not only in the wounded
tissues of the embryonic axis, but also in unwounded cotyle-
TREE PHYSIOLOGY VOLUME 23, 2003
WOUND STRESS AND CELL-WALL PEROXIDASES IN TWO PROSOPIS SPECIES
dons, indicating that a signal is transported from the wounded
site to the unwounded cotyledons. Many authors have reported
that responses to wound stress are restricted to the site of
wounding (Macheix et al. 1986). However, to account for the
induction of proteinase-inhibitor II genes—a model system
for wound-induced gene expression—systemic signals have
been described in the context of wound stress (Pearce et al.
1991, Pena-Cortés et al. 1995, Wildon et al. 1992). Only soluble peroxidases showed a systemic response to wounding in
P. chilensis and P. tamarugo. In contrast, ionically bound
cell-wall peroxidases increased only locally in the wounded
tissue, indicating that the systemic wound signal did not induce accumulation of peroxidases that were destined for the
cell wall, even 48 h after wounding. We note that 48 h should
be enough time to observe deposition of newly synthesized
proteins in the cell wall, because Cooper and Varner (1984)
observed deposition of structural proteins in carrots within this
time frame.
Effect of wounding on accumulation of isoforms of ionically
bound peroxidases
Isoforms of enzymes are widely characterized by their isoelectric points (pI). In P. chilensis and P. tamarugo, we distinguished two groups of ionically bound cell-wall peroxidases
based on their pIs: an acidic group and a basic group. The pIs
of the acidic group differed between species, as did the
amounts and activities of the isoforms in response to wounding. In P. chilensis, only the activity of basic peroxidases increased after wounding, whereas in P. tamarugo, wounding
mainly increased the activity of acidic peroxidases, although it
also resulted in a slight increase in the activity of basic peroxidases. Analyses by IEF of the basic isoforms confirmed
these species-specific differences. Thus, unwounded controls
of P. chilensis contained only one basic peroxidase isoform,
whereas two isoforms were found in wounded seedlings. In
unwounded and wounded seedlings of P. tamarugo there were
two basic isoforms with the same molecular masses and pIs as
those found in wounded seedlings of P. chilensis. Therefore,
we conclude that wounding leads to induction or activation of
a new, basic peroxidase isoform in P. chilensis that is present
prior to wounding in P. tamarugo. If this peroxidase has a role
in strengthening cell walls after wounding, P. tamarugo would
be protected sooner and more effectively than P. chilensis, because this peroxidase is constitutively expressed. The differential effect of wounding on peroxidase isoforms could also
explain why the time course of the wound-induced increase in
peroxidase activity differed between species. The acidic isoforms accumulated later in P. tamarugo than did the basic
isoforms in P. chilensis.
It was impossible to determine the exact pIs of the basic
isoforms in P. chilensis and P. tamarugo, because the most basic ampholytes available for IEF only cover the pH range of
8–10.5, and the basic isoforms of the two species clustered at
the very end of this range. Therefore, we assigned these peroxidase isoforms a pI ≥ 10 (cf. Hejgaard et al. 1991, Nair and
Showalter 1996).
449
Acidic cell-wall peroxidases of P. chilensis and P. tamarugo
have molecular masses between 70 and 90 kDa in P. chilensis
and between 60 and 75 kDa in P. tamarugo. These molecular
masses are high compared with previously reported molecular
masses of between 30 and 45 kDa for acidic peroxidases
(Lagrimini et al. 1987, Teichmann et al. 1997, Bernards et al.
1999, Carpin et al. 1999). The high molecular masses could be
a result of glycosylation of the peroxidase isoenzymes. Glycosylation has been demonstrated in other peroxidases (Hendriks
et al. 1991, Wan et al. 1994) and was confirmed for the acidic
peroxidases of P. chilensis and P. tamarugo by staining the
SDS-PAGE gels by the method of Eckhardt et al. (1976) (data
not shown). Basic peroxidases also stained with this method,
although not as strongly as the acidic peroxidases.
Activities of ionically bound peroxidases with different
substrates and the accumulation of peroxidase isoforms
The substrate study revealed that basic peroxidase activity in
unwounded seedlings of P. tamarugo was as high as or higher
than in wounded seedlings of P. chilensis, indicating the existence of two constitutive basic peroxidases in unwounded
seedlings of P. tamarugo. In P. chilensis, there was one constitutive and one inducible isoform. In response to wounding,
acidic peroxidase activity increased in P. tamarugo and the activity of constitutive basic peroxidases remained high,
whereas in P. chilensis, only the activity of basic peroxidases
increased. Therefore, absolute peroxidase activity after
wounding was higher in P. tamarugo than in P. chilensis, supporting our hypothesis that P. tamarugo, which is exposed to
severe predatory stress in its natural environment, has better
defenses against mechanical injury than P. chilensis.
Another theory (Coley et al. 1985, Herms and Mattson
1992, Koricheva et al. 1998) proposes that slow-growing
plants with high concentrations of constitutive defensive components will grow in locations with few resources, whereas
fast-growing plants with low concentrations of defensive components or inducible defensive components will grow in resource-rich locations. Prosopis tamarugo, which is native to
habitats with few resources, grows more slowly, even under favorable laboratory conditions, than P. chilensis, which grows
in moderately resource-rich locations; however, P. tamarugo
has higher photosynthetic rates than P. chilensis (DelatorreHerrera 1996). Therefore, the differential accumulation of
cell-wall peroxidases in these species supports the theory that
defensive resources are inversely related to growth rate. However, other factors could explain the differences between species in peroxidase activity. For example, the light environment
differs between P. chilensis and P. tamarugo habitats (Lehner
et al. 2001). Alternatively, because cell-wall peroxidases play
an important role in the control of growth and lignification of
secondary cell walls (Cassab 1998), differences in peroxidase
activity and isoforms could arise because of different growth
rates and different lignification patterns in seedlings of these
species.
The largest increase in wound-induced basic peroxidase activity in P. chilensis was found with guaiacol as the substrate.
TREE PHYSIOLOGY ONLINE at http://heronpublishing.com
450
LEHNER AND CARDEMIL
Guaiacol corresponds to the phenolic part of coniferyl alcohol,
a monomer of lignin. Hence this finding could indicate a role
for this peroxidase in lignin formation (Penel et al. 2000). In
P. tamarugo, the largest increases in wound-induced basic
peroxidase activity were found with o-PDA and ferulic acid.
Ferulic acid, which is a natural substrate for peroxidases,
forms diferulate bridges making cross-links between various
kinds of cell-wall polysaccharides and cell-wall proteins,
thereby reinforcing the cell wall structure (Fry 1986, Penel et
al. 2000).
Protein synthesis and the wound-induced increase in the
basic ionically bound peroxidase in P. chilensis
The additional basic peroxidase found in P. chilensis after
wounding could be detected by SDS-PAGE as soon as 3 h after
wounding. Because no differences in protein concentration
were found between wounded and unwounded seedlings of
P. chilensis when SDS-PAGE gels were stained for proteins,
we postulated that the appearance of the new basic isoform in
wounded seedlings of this species was the result of activation
of a preexisting inactive isoform. Experiments with cycloheximide, a protein synthesis inhibitor, showed that, in extracts of
unwounded and wounded seedlings, total protein concentration and total peroxidase activity decreased compared with untreated controls, as expected when protein synthesis is
suppressed by cycloheximide. The concentration of cycloheximide used (5 µg µl –1) was sufficient to suppress protein synthesis (cf. Chapell et al. 1984, Lagrimini and Rothstein 1987,
Nair and Showalter 1996). In wounded, cycloheximide-treated
seedlings, however, peroxidase activity was higher than in unwounded, cycloheximide-treated seedlings, because of an increase in basic peroxidase activity, as indicated by IEF
analysis. Therefore, we conclude that protein synthesis is not
necessary for the wound-induced increase in basic peroxidase
activity in P. chilensis.
To confirm this finding, we studied the effect of 0.4 mM
cordycepin, an RNA synthesis inhibitor. Again, the concentration used should be sufficient to effectively suppress RNA synthesis in the seedlings, because other authors have found
effective inhibition with even lower concentrations of this inhibitor (Felix et al. 1991, Romera et al. 1998). However, 48 h
after wounding, protein concentration and peroxidase activity
in cordycepin-treated seedlings were higher than in untreated
seedlings. This result could be associated with high rates of
protein synthesis from mRNA that was present before treatment of seedlings with cordycepin (2 h before wounding). Alternatively, the high protein concentration and peroxidase
activity could be a response to the cordycepin treatment itself.
Concurrent analysis of peroxidase isoforms by IEF corroborated this hypothesis. In wounded, cordycepin-treated seedlings, activity of acidic peroxidases increased markedly,
whereas in wounded controls, only the wound-induced increase in activity of basic peroxidases was observed. Because
peroxidases appear in response to all types of stress, catalyzing
the oxidation of a broad variety of natural and artificial sub-
strates, the increase in acidic peroxidase activity might be a
response to stress caused by cordycepin.
Ninety-six hours after wounding, protein concentration and
peroxidase activity in cordycepin-treated seedlings decreased
notably, confirming the inhibitory effect of cordycepin on
RNA synthesis. The IEF analysis showed that basic peroxidases disappeared completely from extracts of wounded,
cordycepin-treated seedlings, indicating a high turnover of basic peroxidases, whereas acidic peroxidases seemed to be
more stable. We conclude that mRNA and protein synthesis
are not necessary for the initial increase in peroxidase activity
in P. chilensis after wounding, and that a preexistent basic
peroxidase is activated in response to wounding. However, to
maintain the elevated activity of basic peroxidase for more
than 48 h in wounded seedlings, mRNA synthesis is necessary.
Because peroxidases are heme proteins, activation of the inactive peroxidase isoform could occur through coupling of the
apoenzyme to a heme group. In response to wounding,
ionically bound cell-wall peroxidases appear to protect
P. tamarugo sooner and better than P. chilensis.
Prosopis chilensis possesses a preexisting inactive basic isoform of an ionically bound cell-wall peroxidase that is activated in response to wound stress. This peroxidase might be
involved in cell wall strengthening either through lignin biosynthesis or through cross-linking of cell-wall polymers, or
both. In P. tamarugo, in contrast, the same basic peroxidase is
present prior to wounding. Moreover, the activity of ionically
bound peroxidases was higher in P. tamarugo than in
P. chilensis. In response to wounding, acidic peroxidases increased in P. tamarugo but not in P. chilensis. These acidic peroxidases might also be involved in cell wall reinforcement
because they showed high activity with ferulic acid. As a consequence, P. tamarugo may tolerate wound stress better than
P. chilensis and, therefore, seems to be well adapted to its habitat in the Atacama Desert where it is exposed to heavier predation than P. chilensis, which is native to a more moderate
environment.
Acknowledgments
Fellowships from the Gottlieb Daimler and Karl Benz foundation and
DAAD to Gabriele Lehner are kindly acknowledged. We thank Angelica Vega for technical assistance, and Mrs. Margrita Moser for
English corrections to the text. Funding for this research was provided
by CONICYT Grant No. 195040-1 and by a Universidad de Chile
grant to Liliana Cardemil.
References
Bernards, M.A., W.D. Fleming, D.B. Llewellyn, R. Priefer, X. Yang,
A. Sabatino and G.L. Plourde. 1999. Biochemical characterization
of the suberization-associated anionic peroxidase of potato. Plant
Physiol. 121:135–145.
Bestwick, C.S., I.R. Brown and J.W. Mansfield. 1998. Localized
changes in peroxidase activity accompany hydrogen peroxide generation during the development of a nonhost hypersensitive reaction in lettuce. Plant Physiol. 118:1067–1078.
Bolwell, G.P., V.S. Butt, D.R. Davies and A. Zimmerlin. 1995. The
origin of the oxidative burst in plants. Free Radical Res. 23:
517–532.
TREE PHYSIOLOGY VOLUME 23, 2003
WOUND STRESS AND CELL-WALL PEROXIDASES IN TWO PROSOPIS SPECIES
Bowles, D. 1998. Signal transduction in the wound response of
tomato plants. Philos. Trans. R. Soc. Lond. B Biol. Sci. 353:
1495–1510.
Bradford, M.M. 1976. A rapid and sensitive method for the
quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72:248–254.
Briskin, D.P., R.T. Leonard and T.K. Hodges. 1987. Isolation of the
plasma membrane: membrane markers and general principles.
Methods Enzymol. 190:542–568.
Brownleader, M.D., P.E. McNally, G.E.A. Davies, M. Trevan and
P.M. Dey. 1997. Elicitor-induced extensin insolubilization in suspension-cultured tomato cells. Phytochemistry 46:1–9.
Campa, A. 1991. Biological roles of plant peroxidases: known and
potential function. In Peroxidases in Chemistry and Biology. Vol.
II. Eds. J. Everse, M.B. Grisham and K.E. Everse. CRC Press,
Boca Raton, pp 25–50.
Carpin, S., M. Crèvecoeur, H. Greppin and C. Penel. 1999. Molecular
cloning and tissue-specific expression of an anionic peroxidase in
zucchini. Plant Physiol. 120:799–810.
Cassab, G.I. 1998. Plant cell-wall proteins. Annu. Rev. Plant Phys.
49:281–309.
Cassab, G.I., J. Nieto-Sotelo, J.B. Cooper, G.-J. van Holst and
J.E. Varner. 1985. A developmentally regulated hydroxyprolinerich glycoprotein from the cell walls of soybean seed coats. Plant
Physiol. 77:532–535.
Castillo, F.J., C. Penel and H. Greppin. 1984. Peroxidase release
induced by ozone in Sedum album leaves. Plant Physiol. 74:
846–851.
Chaman, M.E., L.J. Corcuera, G.E. Zúñiga, L. Cardemil and
V.H. Argandoña. 2001. Induction of soluble and cell-wall peroxidases by aphid infestation in barley. J. Agric. Food Chem. 49:
2249–2253.
Chapell, J., K. Hahlbrock and T. Boller. 1984. Rapid induction of ethylene biosynthesis in cultured parsley cells by fungal elicitor and its
relationship to the induction of phenylalanine ammonia-lyase.
Planta 161:475–480.
Cipollini, D.F., Jr. 1998. The induction of soluble peroxidase activity
in bean leaves by wind-induced mechanical perturbation. Am. J.
Bot. 85:1586–1591.
Coley, P.D., J.P. Bryant and F.S. Chapin, III. 1985. Resource availability and plant antiherbivore defense. Science 230:895–899.
Cooper, J.B. and J.E. Varner. 1984. Cross-linking of soluble extensin
in isolated cell walls. Plant Physiol. 76:414–417.
Delatorre-Herrera, J.G. 1996. Efecto de la temperatura y del déficit
hídrico sobre la fotosíntesis de Prosopis chilensis (Mol.) Stuntz y
Prosopis tamarugo (Phil.). Tesis de Magister, Universidad de
Chile, Santiago de Chile, 78 p.
Douroupi, T. and L.H. Margaritis. 2000. Biochemical and molecular
data on Urtica dioica plant peroxidase. Plant Peroxidase Newsletter 14:13–20.
Eckhardt, A.E., E.H. Colleen and I.J. Goldstein. 1976. A sensitive fluorescent method for the detection of glycoproteins in polyacrylamide gels. Anal. Biochem. 73:192–197.
Espelie, K.E., V.R. Franceschi and P.E. Kolattukudy. 1986. Immunocytochemical localization and time course of appearance of an anionic peroxidase associated with suberization in wound healing
potato tuber tissue. Plant Physiol. 81:487–492.
Felix, G., D.G. Grosskopf, M. Regenass, C.W. Basse and T. Boller.
1991. Elicitor-induced ethylene biosynthesis in tomato cells: characterization and use as a bioassay for elicitor action. Plant Physiol.
97:19–25.
Fry, S.C. 1982. Isodityrosine, a new cross-linking amino acid from
plant cell-wall glycoproteins. Biochem. J. 204:449–455.
451
Fry, S.C. 1986. Cross-linking of matrix polymers. In The Growing
Cell Walls of Angiosperms. Annu. Rev. Plant Physiol. 37:
165–186.
González, L.F., F. Perez and M.C. Rojas. 1999. Indole-3-acetic acid
control on acidic oat cell-wall peroxidase. J. Plant Growth Regul.
18:25–31.
Harris, E.L.V. and S. Angal. 1989. Protein purification methods: a
practical approach. The Practical Approach Series. IRL Press at
Oxford University Press, Oxford, New York, 317 p.
Hejgaard, J., J.F. Petersen, N.C. Veitch, B.J. Pedersen and
K.G. Welinder. 1991. Three cationic peroxidases of barley grain. In
Biochemical, Molecular and Physiological Aspects of Plant
Peroxidases. Eds. J. Lobarzewski, H. Greppin, C. Penel and
T. Gaspar. University M. Curie-Sklodowska, Lublin, Poland and
Univ. Geneva, Switzerland, pp 49–53.
Hendriks, T., H.J.W. Wijsman and L.C. van Loon. 1991. Petunia
peroxidase a: isolation, purification and characteristics. Eur. J.
Biochem. 199:139–146.
Herms, D.A. and W.J. Mattson. 1992. The dilemma of plants: to grow
or to defend. Q. Rev. Biol. 67:283–335.
Heukeshoven, J. and R. Dernick. 1985. Simplified method for silver
staining of proteins in polyacrylamide gels and the mechanism of
silver staining. Electrophoresis 6:103–112.
Ho, D.T.-H. and J.E. Varner. 1976. Response of barley aleurone layers
to abscisic acid. Plant Physiol. 57:175–178.
Kiba, A., C. Miyake, K. Toyoda, Y. Ichinose, T. Yamada and
T. Shiraishi. 1997. Superoxide generation in extracts from isolated
plant cell walls is regulated by fungal signal molecules. Phytopathology 87:846–852.
Koricheva, J., S. Larsson, E. Haukioja and M. Keinänen. 1998.
Regulation of woody plant secondary metabolism by resource
availability: hypothesis testing by means of meta-analysis. Oikos
83: 212–226.
Laemmli, U.K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680–685.
Lagrimini, L.M. and S. Rothstein. 1987. Tissue specificity of tobacco
peroxidase isozymes and their induction by wounding and tobacco
mosaic virus infection. Plant Physiol. 84:438–442.
Lagrimini, L.M., W. Burkhart, M. Moyer and S. Rothstein. 1987.
Molecular cloning of complementary DNA encoding the ligninforming peroxidase from tobacco: molecular analysis and tissuespecific expression. Proc. Natl. Acad. Sci. 84:7542–7546.
Lee, T.-M. and Y.H. Lin. 1995. Changes in soluble and cell-wallbound peroxidase activities with growth in anoxia-treated rice
(Oryza sativa L.) coleoptiles and roots. Plant Sci. 106:1–7.
Lehner, G., J. Delatorre, C. Lütz and L. Cardemil. 2001. Field studies
on the photosynthesis of two desert Chilean plants: Prosopis
chilensis and Prosopis tamarugo. J. Photochem. Photobiol. B Biol.
64:36–44.
Li, H. and T.L. Poulos. 1994. Structural variation in heme enzymes: a
comparative analysis of peroxidase and P450 crystal structures.
Structure 2:461–464.
Loukili, A., F. Limam, C. Thonat, W. Chaibi, A. Ayadi, N. Boyer and
L. Quelhazi. 1997. Involvement of phenylalanine ammonia-lyase
and cationic peroxidase in tomato thigmomorphogenesis. Plant
Peroxidase Newsletter 11:3–12.
Macheix, J.J., A. Fleuriet and M.P. Quessada. 1986. Involvement of
phenols and peroxidases in wound healing and grafting. In Molecular and Physiological Aspects of Plant Peroxidases. Eds. H.
Greppin, C. Penel and T. Gaspar. Univ. Geneva, Switzerland, pp
267–286.
McDougall, C.J. 1991. Cell-wall-associated peroxidases and lignification during growth of flax fibers. J. Plant Physiol. 139:182–186.
TREE PHYSIOLOGY ONLINE at http://heronpublishing.com
452
LEHNER AND CARDEMIL
McDougall, C.J. 1993. Accumulation of wall-associated peroxidases
during wound-induced suberization in flax. J. Plant Physiol. 142:
651–656.
Mitsui, T., Y. Ueki and I. Igaue. 1993. Biosynthesis and secretion of
α-amylase by rice suspension-cultured cells: purification and characterization of α-amylase isozyme H. Plant Physiol. Biochem. 31:
863–874.
Nair, A.R. and A.M. Showalter. 1996. Purification and characterization of a wound-inducible cell-wall cationic peroxidase from carrot
roots. Biochem. Biophys. Res. Commun. 226:254–260.
Pearce, G., D. Strydom, S. Johnson and C.A. Ryan. 1991. A polypeptide from tomato leaves induces the synthesis of wound-inducible
proteinase inhibitor proteins. Science 253:895–898.
Peña-Cortés, H., J. Fisahn and L. Willmitzer. 1995. Signals involved
in wound-induced proteinase inhibitor II gene expression in tomato
and potato plants. Proc. Natl. Acad. Sci. 92:4106–4113.
Penel, C., S. Carpin, M. Crèvecoeur, P. Simon and H. Greppin. 2000.
Binding of peroxidases to Ca2+-pectate: possible significance for
peroxidase function in cell wall. Plant Peroxidase Newsletter 14:
33–40.
Rodríguez, J.G. and L. Cardemil. 1994. Cell-wall proteins in seedling
cotyledons of Prosopis chilensis. Phytochemistry 35:281–286.
Romera, F.J., E. Alcántara and M.D. de la Guardia. 1998. The induction of the “turbo reductase” is inhibited by cycloheximide,
cordycepin and ethylene inhibitors in Fe-deficient cucumber
(Cucumis sativus L.) plants. Protoplasma 205:156–162.
Smith, J.A. and R. Hammerschmidt. 1988. Comparative study of
acidic peroxidases associated with induced resistance in cucumber,
muskmelon and watermelon. Physiol. Mol. Plant Pathol. 33:
255–261.
Teichmann, T., C. Guan, P. Kristoffersen, G. Muster, O. Tietz and
K. Palme. 1997. Cloning and biochemical characterization of an
anionic peroxidase from Zea mays. Eur. J. Biochem. 247:826–832.
Tognolli, M., S. Overney, C. Penel, H. Greppin and P. Simon. 2000. A
genetic and enzymatic survey of Arabidopsis thaliana peroxidases.
Plant Peroxidase Newsletter 14:3–12.
Wan, L., M. Gijzen and R.B. van Huystee. 1994. Heterogeneous
glycosylation of cationic peanut peroxidase. Biochem. Cell Biol.
72:411–417.
Wang, W. and Z.-H. Liu. 1999. Harpin PSS-induced peroxidase and
lignin accumulation in tobacco during the hypersensitive response.
Aust. J. Plant Physiol. 26:265–272.
Welinder, K.G. 1992. Superfamily of plant, fungal and bacterial
peroxidases. Curr. Opin. Struct. Biol. 2:388–393.
Wildon, D.C., J.F. Thain, P.E.H. Minchin, I.R. Gubb, A.J. Reilly, Y.D.
Skipper, H.M. Doherty, P.J. O’Donnell and D.J. Bowles. 1992.
Electrical signalling and systemic proteinase inhibitor induction in
the wounded plant. Nature 360:62–65.
TREE PHYSIOLOGY VOLUME 23, 2003