Arabidopsis peroxidase-catalyzed copolymerization of coniferyl and sinapyl alcohols: Kinetics of an endwise process

Article abstract of DOI:10.1016/j.phytochem.2010.06.011

In order to determine the mechanism of the earlier copolymerization steps of two main lignin precursors, sinapyl (S) alcohol and coniferyl (G) alcohol, microscale in vitro oxidations were carried out with a PRX34 Arabidopsis thaliana peroxidase in the presence of H₂O₂. This plant peroxidase was found to have an in vitro polymerization activity similar to the commonly used horseradish peroxidase. The selected polymerization conditions lead to a bulk polymerization mechanism when G alcohol was the only phenolic substrate available. In the same conditions, the presence of S alcohol at a 50/50 S/G molar ratio turned this bulk mechanism into an endwise one. A kinetics monitoring (size-exclusion chromatography and liquid chromatography-mass spectrometry) of the different species formed during the first 24 h oxidation of the S/G mixture allowed sequencing the bondings responsible for oligomerization. Whereas G homodimers and GS heterodimers exhibit low reactivity, the SS pinoresinol structure act as a nucleating site of the polymerization through an endwise process. This study is particularly relevant to understand the impact of S units on lignin structure in plants and to identify the key step at which this structure is programmed.

Full text of DOI:10.1016/j.phytochem.2010.06.011

Phytochemistry 71 (2010) 1673–1683
Contents lists available at ScienceDirect
Phytochemistry
journal homepage: www.elsevier.com/locate/phytochem
Arabidopsis peroxidase-catalyzed copolymerization of coniferyl and sinapyl
alcohols: Kinetics of an endwise process
Nathalie Demont-Caulet a,b, Catherine Lapierre a, Lise Jouanin a, Stéphanie Baumberger a, Valérie Méchin a,
*
a Institut Jean-Pierre Bourgin, UMR 1318 INRA/AgroParisTech, Pôle Systèmes Cellulaires, Morphogénèse et Signalisation, 78850 Thiverval-Grignon, Cedex, France
b UFR des Sciences du Vivant, Université Diderot-Paris 7, 75205 Paris, Cedex 13, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 22 April 2010
Received in revised form 11 June 2010
Available online 6 July 2010
In order to determine the mechanism of the earlier copolymerization steps of two main lignin precursors,
sinapyl (S) alcohol and coniferyl (G) alcohol, microscale in vitro oxidations were carried out with a PRX34
Arabidopsis thaliana peroxidase in the presence of H2O2. This plant peroxidase was found to have an
in vitro polymerization activity similar to the commonly used horseradish peroxidase. The selected poly-
merization conditions lead to a bulk polymerization mechanism when G alcohol was the only phenolic sub-
strate available. In the same conditions, the presence of S alcohol at a 50/50 S/G molar ratio turned this bulk
mechanism into an endwise one. A kinetics monitoring (size-exclusion chromatography and liquid chro-
matography–mass spectrometry) of the different species formed during the first 24 h oxidation of the S/
G mixture allowed sequencing the bondings responsible for oligomerization. Whereas G homodimers
and GS heterodimers exhibit low reactivity, the SS pinoresinol structure act as a nucleating site of the poly-
merization through an endwise process. This study is particularly relevant to understand the impact of S
units on lignin structure in plants and to identify the key step at which this structure is programmed.
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Coniferyl and sinapyl alcohols
Endwise
Bulk
Oligolignols structure
Lignin synthesis
Plant peroxidase
1. Introduction
alcohol oxidation in planta should be mediated by other phenolic
radicals. This hypothesis has been strengthened by in vitro copoly-
Lignin is a natural macromolecule resulting from the polymeri-
zation of three main monomers: p-coumaryl alcohol, coniferyl
alcohol and sinapyl alcohol. In vitro preparation of synthetic lignins
in the presence of oxidative systems such as peroxidase/H2O2 sup-
ports the idea that this polymerization mainly occurs via radical
couplings. Such a polymerization process implies that phenolic
groups of both the monolignol and the growing polymer are oxi-
dized into radicals, either directly by the catalyst or by the radical
transfer from another monolignol radical. The structure of the final
polymer is therefore highly dependent on the oxidative system, the
nature and proportion of monolignols and also on the physico-
chemical constraints. These parameters confer a high degree of
plasticity to native and synthetic lignins. Understanding the chem-
ical mechanisms of monolignols polymerization is thus of primary
importance for both controlling the structure of synthetic lignins
and the elucidation of lignification in plants.
merization experiments of sinapyl alcohol and coniferyl alcohol
and of their glucosides (Tobimatsu et al., 2008a,b, 2010). However,
some studies (Quiroga et al., 2000; Aoyama et al., 2002; Gabaldon
et al., 2005; Marjamaa et al., 2006) have highlighted that some
class III peroxidases display a highest affinity for sinapyl alcohol.
Such peroxidases, called syringyl peroxidases, seem to be rather
widespread in the plant kingdom (Barcelo et al., 2007). Whether
both radical mediation and direct oxidation by peroxidases can ac-
count for the incorporation of syringyl units in lignins, the detailed
mechanism of this incorporation has not yet been reported so far.
Similarly, there are few reports in which the whole process of
dehydrogenative polymerization of monolignols is monitored.
The objective of this study was to determine the impact of sina-
pyl alcohol on the polymerization mode of coniferyl alcohol and to
elucidate the bonding sequences leading to the incorporation of
sinapyl alcohol in the first steps of the oligomerization.
Class III plant peroxidases are secreted plant enzymes that
catalyze the oxidation of a wide variety of phenolic substrates,
including monolignols. However, peroxidase affinity for the three
different monolignols is different. While coniferyl alcohol is readily
oxidized by most peroxidases, sinapyl alcohol often appears to be a
poor substrate. Takahama and Oniki (1994) suggested that sinapyl
Using class III peroxidase purified from Arabidopsis thaliana, we
have investigated the oxidation of either a mixture of sinapyl and
coniferyl alcohol or sinapyl and coniferyl alcohols alone to deter-
mine the polymerization mode recruited in each condition. Kinet-
ics of the monomer conversion and of the subsequent formation of
di-, tri- and tetramers was carried out. We show that, in our in vitro
oxidation conditions, the presence of S monomers converted the
polymerization of G monomers from a bulk to an endwise mecha-
nism. This result allowed us to confirm the role of coniferyl alcohol
* Corresponding author. Tel.: +33 1 30 81 54 62; fax: +33 1 30 81 53 73.
E-mail address: valerie.mechin@grignon.inra.fr (V. Méchin).
0031-9422/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.phytochem.2010.06.011

1674
N. Demont-Caulet et al. / Phytochemistry 71 (2010) 1673–1683
as radical mediator for sinapyl alcohol oxidation and identify key
molecular structures associated with the incorporation of sinapyl
alcohol. This study brings new information on the compared reac-
tivity of S and S/G lignin model compounds.
1 h, alone or in mixture with G alcohol. We never detected in these
conditions any higher oligomers than polymerization degree (PD)
2. Even in combination with S alcohol, G alcohol is never used to
form dimers or other products.
When adding peroxidase to the initial reaction medium, we ob-
tained similar results while using either HRP or PRX34 peroxidase
and thus decided to focus the paper on PRX34-catalyzed polymer-
ization. If the only substrate is S alcohol, we observed as in the ab-
sence of peroxidase the appearance of dimers and the total
consumption of S alcohol within 4 h (Fig. 2a). G alcohol alone is
dimerized after 5 min (Fig. 3a) and is totally converted at 20 min
(Fig. 2a). Trimers begin to appear at 10 min and tetramers at
30 min (Fig. 3a). Monomer conversion rate is thus far higher for
G than for S alcohol (9-fold lower conversion time), which is in
contradiction with the respective redox potential of these species
(Tobimatsu et al., 2008c). This lower reactivity of S alcohol towards
peroxidase oxidation is thus commonly explained by the low
accessibility of S monomers to the peroxidase active site due to
steric hindrance and hydrophobic effects of the two OCH3 groups
on the aromatic ring (Nielsen et al., 2001; Tobimatsu et al., 2008a).
2. Results and discussion
Oxidative coupling was carried out in three different substrate
conditions: (i) coniferyl (G) alcohol alone, (ii) sinapyl (S) alcohol
alone or (iii) a 50/50 mixture of both G and S alcohols, to trace
the appearance and accumulation of oligomers synthesized with-
out peroxidase or in the presence of either horseradish peroxidase
(HRP) or PRX34 A. thaliana peroxidase (PRX34).
In order to over produce the protein PRX34 (gene At3g49120),
we used two different Arabidopsis mutant lines derived from the
Wassilewsija and Columbia ecotypes: (i) Atprx34 (FSTS51769) se-
lected in the Salk collection. Atprx34 is a deregulated line with
higher expression of the corresponding targeted gene at the stem
level compared to the wild type. (ii) In order to visualize and over
express PRX34, its cDNA under the control of the CaMV 35S pro-
moter was cloned in the GFP Gateway vector pGWB5 and intro-
duced in the Columbia ecotype.
2.2. Monomers consumption during peroxidase-catalyzed
copolymerization
The Atprx34 line over produced PRX34 three times whereas the
fusion Perx34-GFP allowed an over production of fifteen times. The
purified proteins have the same enzymatic activity and thus we
decided to use protein purified from the transformed Columbia be-
cause of the higher purification yield.
In the copolymerization conditions, all the monomers are con-
verted within the first 30 min (Fig. 2b). S monomers are completely
converted after 5 min (Fig. 2b). This result dramatically contrasts
with the four hours required for the total conversion of S in
homopolymerization conditions (Fig. 2a). Whereas the presence
of G accelerates the conversion of S monomers, the presence of S
slows down the conversion of G (10 additional minutes required
for total conversion; Fig. 2a and b). This result was repeatedly ob-
served using HRP peroxidase and is in agreement with the role of
radical mediator assigned to G in various studies (Takahama
et al., 1996; Aoyama et al., 2002; Fournand et al., 2003; Sasaki
et al., 2004; Tobimatsu et al., 2008a) that confirm the earlier obser-
vation that G monomers take part in S monomers conversion
(Freudenberg and Hubner, 1952). Once oxidized by peroxidase, G
would transfer its radical to S consistently with the lower S redox
potential. As proposed by Tobimatsu et al. (2008b), G monomers
could also act as a promoter in the stoichiometric quenching of
S-type quinone methide (QM) in the course of copolymerization
by displacing the equilibrium and thus accelerate the conversion
of stable S-type QM intermediates.
Stems (5–6 week-old) of these two lines were harvested and
used for peroxidase purification. Peroxidases were purified among
a set of N-glycosylated proteins selected on a Concanavalin-A Se-
pharose column (Minic et al., 2007). The fraction containing the
glycosylated proteins was further purified by carboxymethyl-Se-
pharose column. The activity peak was collected in each sample
and loaded on an acrylamide gel further stained with silver nitrate
or revealed with an antisera raised against bean peroxidase (Fig. 1).
Maldi-TOF MS was used to identify the purified PRX34 through
peptide mass fingerprinting after trypsin digestion of the bands.
The obtained peptide masses led to the identification of PRX34
using the MSDB database Accession No. T46118, with at least thir-
teen matching peptides, covering 42% of the protein and with a
mass accuracy inferior to 50 ppm. The protein has a calculated
molecular mass of 38.8 kDa. This purified protein was used to per-
form subsequent oxidative coupling of G and S alcohols.
In order to elucidate the mechanisms by which S monomers are
incorporated into the final polymer, we focused on the structure of
the dimers and further higher oligomers formed. The structure of
compound with a PD comprised from 2 to 4 were thus assigned
according to their mass and photodiode array (PDA) spectra as re-
ported by Morreel et al. (2004). As for the global monomer conver-
sion kinetics, the formation kinetics of the different dimers and
oligomers were found similar for the HRP and PRX34 peroxidase.
After a global comparison of the results obtained in homo- and
copolymerization conditions, we examine in detail the PRX34-cat-
alyzed copolymerization of S and G alcohols in order to dissect the
whole mechanism recruited in the first steps of lignin synthesis.
2.1. Monomers consumption during peroxidase-catalyzed
homopolymerization
If no peroxidase is added in the reaction medium (data not
shown), S alcohol begins to be consumed to produce dimers after
2.3. Homopolymerization and copolymerization context: a bulk versus
an endwise polymerization mechanism
Peroxidase-catalyzed oxidation of S or G alcohols allows to re-
cover a soluble fraction of dimers and oligomers (PD 6 12; Fig. 3)
that was analyzed for equivalent molar mass distribution by high
performance size-exclusion chromatography (HPSEC). After an or-
ganic-solvent extraction, HPSEC profiles of the G alcohol products
(Fig. 3a) and of the S and G alcohols copolymerization products
Fig. 1. SDS–PAGE and Western blot analysis of purified peroxidase using polyclonal
antibodies raised against peroxidases. Pooled active fractions from purification
stage were subjected to SDS–PAGE on 10% gels visualized by silver staining (left
part) or by Western blot analysis (right part). WT: wild type line, PRX34-prom:
over-expressed line, PRX34-GFP: fusionned protein PRX34 with GFP.

N. Demont-Caulet et al. / Phytochemistry 71 (2010) 1673–1683
1675
120
100
80
60
40
20
0
(a)
O min 5 min 10 min 20 min 30 min 60 min 90 min 120 min 4 h
24 h
120
100
80
60
40
20
0
(b)
O min 5 min
10 min 20 min 30 min 60 min 90 min 120 min 4 h
24 h
Fig. 2. PRX34-catalyzed oxidation kinetics of ( ) S alcohol and ( ) G alcohol under (a) homopolymerization (S or G alone) and (b) copolymerization (50/50 mol/mol S/G
mixture) conditions. Relative weight contents determined by LC-PDA are normalized with respect to the initial alcohol content. Chromatographic conditions: C18 column
(Highpurity, Thermo Electron Corporation, 5
signal 190–600 nm.
l
m, 150 ꢀ 4.6 mm) eluted with a 12–95 vol.% aqueous acetonitrile, 1‰ HCOOH gradient (45 min) and 1 ml minꢁ1 flow rate; PDA
PD 2
t=5 min
(a)
PD2/PD3=15
t=10 min
t=20 min
t=30 min
PD4/PD3=1.7
t=90 min
PD 12
PD 4
PD > 4
PD 3
PD2/PD3=2.4
(b)
PD4/PD3=0.5
13
13.5
14
14.5
15
15.5
16
Elution time (min)
Fig. 3. Time-dependant variations of the HPSEC-PDA profiles of the PRX34-catalyzed oxidation products in (a) homopolymerization (G) and (b) copolymerization (S + G)
conditions. The relative proportions of dimers (PD2) and tetramers (PD4) with respect to trimers (PD3) are estimated from the area of the corresponding elution peaks.
Chromatographic conditions: PL-Gel column (5
normalized with respect to the internal standard.
l
m; 100 Å; 4.5 ꢀ 600 mm; Polymer Laboratories) eluted with stabilized THF (1 ml minꢁ1 flow rate); signal at 280 nm
(Fig. 3b) revealed two distinct global polymerization patterns. For-
mation of dimers and oligomers is often described as resulting
from radical couplings between two dehydrogenated compounds,
either the oxidized monomer and the growing polymer phenoxy
radical (endwise polymerization) or two oligomer phenoxy radi-
cals (bulk polymerization) (Brunow et al., 1998). The two patterns
observed here are typified by the respective proportion of odd- and
even-PD oligomers. Although no absolute quantification is possible
by SEC due to heterogeneity in the fractions corresponding to each
PD, the height ratio between elution peaks can be used for compar-
ative purpose to evaluate the relative proportion of these different
fractions. In our case, the result was that homopolymerization led

1676
N. Demont-Caulet et al. / Phytochemistry 71 (2010) 1673–1683
to a higher proportion of dimers and tetramers, with respect to tri-
mers, than copolymerization. Homopolymerization leads to a 6-
fold higher dimer-to-trimer ratio (15 versus 2.4; Fig. 3) and a
3.4-fold higher tetramer-to-dimer ratio (1.7 versus 0.5; Fig. 3).
The predominance of even-PD compounds in the absence of S
monomers indicate a bulk polymerization process based on the
coupling between dehydrodimers (Syrjanen and Brunow, 2000).
On the other hand, the presence of odd-PD compounds in S–G
copolymerization conditions suggests an endwise process based
on couplings between the oxidized monomer and the elongating
oligomer. These results show, in our in vitro oxidation conditions,
that the presence of S monomers is enough to convert the polymer-
ization of G monomers from a bulk to an endwise process.
In a previous paper (Méchin et al., 2007), we highlighted the
importance of peroxidase availability on the orientation towards
a bulk or an endwise polymerization process. Here, we underline
the importance of the available monomers to also direct the poly-
merization process. Taking into account (i) the low peroxidase-cat-
alyzed oxidation rate of S monomer in spite of its high potential
reactivity (Tobimatsu et al., 2008c) and (ii) the low reactivity of
S-type QM (Tobimatsu et al., 2008b) as proposed by these authors,
G monomer may be involved in both transferring its radical to S
monomer and quenching the stable S-type QM. The stability of
S-type QM may result from the reduced positive charge density
explain the switch from a bulk to an endwise mechanism when
adding S alcohol in the system.
This point is of importance when dehydrogenative polymeriza-
tion product (DHP) synthesis has the objective to modelize lignifi-
cations in angiosperm where both
incorporated.
G and S alcohols are
2.4. Dissection of the mechanism recruited in the first steps of aPRX34-
catalyzed copolymerization of S and G alcohols
Very schematically and according to Fig. 2b, the kinetics of oligo-
merization of the S–G mixture exhibit three distinct phases: the first
one (0–5 min) corresponds to the presence of both S and G mono-
mers; the second one (5–30 min) to the presence of G monomer only
and the last one (30 min–24 h) to the absence of monomers.
In the first phase, dimers accumulation is very fast within the
first 5 min reaction, and coincides with the S alcohol consumption
rate (Fig. 4a). This coincidence is in agreement with the fact that
the two main dimers formed, one SS homodimer and one SG het-
erodimer, involve S units. The SS dimer corresponds to a syring-
aresinol structure (bb bond, Fig. 5) and the SG dimer corresponds
to a phenyl coumaran (b5 bond, Fig. 6). After 5 min, all the S mono-
mers are consumed.
During the second phase, the SS dimer content starts to de-
crease as soon as there are no S monomers left (Fig. 5). Meanwhile,
G alcohol continues to be consumed during the next 25 min until
total conversion (Fig. 4a). In agreement with the absence of S
monomers after 5 min, SG dimer formation is no more observed
beyond this reaction time (Fig. 6). G is then involved in the elonga-
tion of the existing S-based dimers and, in parallel, to the forma-
tion of G bb, b5 and, to a lower extent, of b-O-4 homodimers.
b-O-4 coupling between the G radical and the SS and SG dimers ex-
plain the formation of trimers 614 (614 is the molecular weight of
at the a-position because of the presence of two electron-donating
OCH3 groups (Tobimatsu et al., 2010). Thus, when considering a
copolymerization context, G monomer first transfers its radical to
S monomer allowing the formation of SS QM. This QM being stable,
it is necessary that G monomer intercedes again to displace the
equilibrium and to allow the polymerization to go further. These
successive steps can explain that the polymerization process is
slower with a mixture of S and G alcohols than with G alcohol
alone. Such a slowing down of the polymerization process can
120
100
80
60
40
20
0
(a)
O min 5 min 10 min 20 min 30 min 60 min 90 min 120 min 4 h
24 h
120
100
80
60
40
20
0
(b)
O min 5 min 10 min 20 min 30 min 60 min 90 min 120 min 4 h
24 h
Fig. 4. PRX34-catalyzed oxidation kinetics of a 50/50 mol/mol S/G mixture showing (a) the coincidence of the monomer consumption, ( ) S alcohol and ( ) G alcohol, with
the total dimer (*) formation and (b) the slight time shift but similar profiles between the total ( ) trimers and ( ) tetramers formation curves. Relative weight contents
determined by LC-PDA are normalized with respect to the maximal content of each species. Chromatographic conditions: C18 column (Highpurity, Thermo Electron
Corporation, 5
l
m, 150 ꢀ 4.6 mm) eluted with a 12–95 vol.% aqueous acetonitrile, 1‰ HCOOH gradient (45 min) and 1 ml minꢁ1 flow rate; PDA signal 190–600 nm.

N. Demont-Caulet et al. / Phytochemistry 71 (2010) 1673–1683
1677
120
100
80
60
40
20
0
SS ββ dimer
OMe
HO
OMe
OMe
OH
OMe
O
S
S
O
O min 5 min 10 min 20 min 30 min 60 min 90 min 120 min 4 h
24 h
614 trimer
644 trimer
OMe
OH
OMe
OH
OMe
O
OMe
O
S
O
H
O
H
S
HO
OMe
S
H
OMe
OH
HO
OMe
H
OMe
OH
S
O
O
S
G
MeO
OMe
OMe
OH
OH
Fig. 5. PRX34-catalyzed oxidation kinetics of a 50/50 mol/mol S/G mixture showing the simultaneous formation of (*) the syringaresinol (bb) S homodimer and ( ) the 644
SSS homotrimer and the delayed formation of ( ) the 614 SSG heterotrimer. Both trimers are formed by a b-O-4 coupling between the syringaresinol structure and one
monomer (G or S alcohol). Thirty minutes is a key reaction time corresponding to the total consumption of the ( ) S and ( ) G alcohol monomers, the S homodimers and the S
homotrimers and to the maximum formation of the heterotrimer. Relative weight contents determined by LC-PDA (monomers and dimers) and LC-ESI-MS (trimers and
tetramers) are normalized with respect to the maximal content of each species. Chromatographic conditions: C18 column (Highpurity, Thermo Electron Corporation, 5 lm,
150 ꢀ 4.6 mm) eluted with a 12–95 vol.% aqueous acetonitrile, 1‰ HCOOH gradient (45 min) and 1 ml minꢁ1 flow rate; PDA signal 190–600 nm; negative ion ESI-MS spectra
(120–2000 m/z) acquired using an ion trap spectrometer (Finnigan LCQ-DECA – Thermo Electron Corporation) setting the needle voltage at 4 kV and the desolvating capillary
temperature at 350 °C.
120
100
80
60
40
20
0
O min 5 min
10 min 20 min 30 min 60 min 90 min 120 min 4 h
24 h
SG β5 dimer
584 trimer
HO
HO
O
OMe
OH
OMe
OMe
O
OH
HO
HO
G
S
G
S
H
O
H
OH
OMe
OMe
OMe
G
OMe
OH
Fig. 6. PRX34-catalyzed oxidation kinetics of a 50/50 mol/mol S/G mixture showing the fast formation of the (*) b5 SG heterodimer symmetrically to the consumption of the
) S alcohol and the delayed formation of the ( ) 584 SGS heterotrimer symmetrically to the consumption of the ( ) G alcohol. Relative weight contents determined by LC-
(
PDA (monomers and dimers) and LC-ESI-MS (trimers) are normalized with respect to the maximal content of each species. Chromatographic conditions: C18 column
(Highpurity, Thermo Electron Corporation, 5
signal 190–600 nm; negative ion ESI-MS spectra (120–2000 m/z) acquired using an ion trap spectrometer (Finnigan LCQ-DECA – Thermo Electron Corporation) setting the
l
m, 150 ꢀ 4.6 mm) eluted with a 12–95 vol.% aqueous acetonitrile, 1‰ HCOOH gradient (45 min) and 1 ml minꢁ1 flow rate; PDA
needle voltage at 4 kV and the desolvating capillary temperature at 350 °C.

1678
N. Demont-Caulet et al. / Phytochemistry 71 (2010) 1673–1683
the formed trimer and will be used elsewhere to mention this spe-
cific trimer) and 584 (584 is the molecular weight of the formed
trimer and will be used elsewhere to mention this specific trimer)
respectively (Figs. 5 and 6 and Supplementary material). The 644
trimer (644 is the molecular weight of the formed trimer and will
be used elsewhere to mention this specific trimer) is an analogous
of the 614 trimer formed by a b-O-4 coupling between a S unit and
the SS dimer (Supplementary material). Very interestingly, the G
homodimers did not lead to any trimers, which show the higher
reactivity of the S-based structure towards copolymerization.
According to the respective dimer conversion rate, the SS b–b
structure turns out to be the most reactive one. The 5 min delay
by which tetramers appear after trimers (Fig. 4b and Supplemen-
tary material) indicate that the tetramers are formed by elongation
of the trimer by a monomer, according to an endwise process, and
not by homocoupling between dimers. The structure of the 810
and 840 tetramers (810 and 840 are the molecular weights of the
formed tetramers and will be used elsewhere to mention these
specific tetramers) confirm that they might be formed from the
614 and 644 trimers by elongation with one G unit, respectively
(Fig. 7). Here again, elongation proceeds by b-O-4 linkages. We
found no tetramer derived from the 584 trimer, which suggests
that the endwise elongation is favoured by the presence of one
terminal S structure. In parallel to this endwise elongation process
a bulk-type process seems to occur, involving the GG b–b homodi-
mer. Indeed, whereas no corresponding trimer or tetramer was
observed, we detected some pentamers with molar mass of
970 g molꢁ1. This mass corresponds exactly to the coupling be-
tween a GG homodimer and the 614 trimer and excludes the
involvement of a b-O-4 bond for this coupling. In view of the struc-
ture of the SSG trimer and GG dimer, the coupling possibilities are
the biphenyl or biphenyl ether one.
The third phase begins when all G units are consumed (Fig. 4a).
Amazingly, trimers and tetramers are apparently still produced
until 60–90 min, although there is no monomer left for endwise
elongation. This apparent increase in trimer and tetramer content
can be explained by some structural modification within the reac-
tion mixture. Indeed, at 30 min, trimers and tetramers of a new
type start to be produced, whereas there is no monomer available
anymore for the synthesis of trimer or tetramers. The mass spec-
tra of these compounds (characteristic fragments at m/z 433 and
629 respectively) suggest that they could correspond to the spec-
ulative structures incorporating a sinapaldehyde (S0) unit and pro-
posed by Morreel et al. (2004). These structures are trimers
(compound 31 GSS0 or GS0S; Morreel et al. (2004) and Supplemen-
tary material) and tetramers (compound 35 GGSS0 or GGS0S;
120
100
80
60
40
20
0
(a)
644 trimer
OMe
OH
OMe
O
S
O
H
H
S
OH
HO
OMe
O
OMe
S
MeO
OMe
OH
840 tetramer
OMe
OH
0min
O min
5min
10min 20min 30min 60min 90min 120min
4h
24h
24 h
5 min 10 min 20 min 30 min 60 min 90 min 120 min 4 h
OMe
HO
O
S
O
H
S
H
OMe
OH
H
O
H
614 trimer
O
HO
OMe
OH
OMe
S
OMe
O
G
MeO
OMe
S
O
H
H
OMe
OH
HO
OMe
OH
S
O
OH
OMe
G
120
100
80
60
40
20
0
OMe
(b)
OH
810 tetramer
OMe
OH
OMe
HO
O
O
H
S
H
OH
H
O
H
S
O
OMe
HO
OMe
G
OMe
G
OMe
OH
OH
O min 5 min 10 min 20 min 30 min 60 min 90 min 120 min 4 h
24 h
Fig. 7. PRX34-catalyzed oxidation kinetics of a 50/50 mol/mol S/G mixture showing the formation of the two tetramers (a; ) 810 GSSG and (b; ) 840 GSSS. The formation
curve of the 840 tetramer parallels the consumption curve of the (b; ) 644 trimers whereas the 810 tetramer formation curve follows the same shape as the (a; ) 614
trimer. Both tetramers reach their maximum formation at 30 min. Relative weight contents determined by LC-ESI-MS (trimers) are normalized with respect to the maximal
content of each species. Chromatographic conditions: C18 column (Highpurity, Thermo Electron Corporation, 5
lm, 150 ꢀ 4.6 mm) eluted with a 12–95 vol.% aqueous
acetonitrile, 1‰ HCOOH gradient (45 min) and 1 ml minꢁ1 flow rate; PDA signal 190–600 nm; negative ion ESI-MS spectra (120–2000 m/z) acquired using an ion trap
spectrometer (Finnigan LCQ-DECA – Thermo Electron Corporation) setting the needle voltage at 4 kV and the desolvating capillary temperature at 350 °C.

N. Demont-Caulet et al. / Phytochemistry 71 (2010) 1673–1683
1679
Morreel et al. (2004) and Supplementary material) involving one
sinapaldehyde unit linked with S alcohol. According to these
authors, these structures may result from the incorporation of a
sinapaldehyde unit during the oligomerization process. In our
study, the maximum contents of these two compounds are
reached at 60 for the trimer and at 90 min for the tetramers
(Fig. 8). The progressive accumulation of GSS0 and GGSS0 observed
in our kinetic approach at later stages (Fig. 9), together with the
absence of aldehydic structures of smaller PD suggest that these
unauthenticated structures may result from the oxidation of a
pre-existing oligomer rather than from the direct incorporation
of a sinapaldehyde monomer.
Also during this third phase, the consumption of the GG b5 di-
mer is observed together with the appearance of a series of oligo-
mers of PD > 5. Most of these oligomers include a pentameric
sequence corresponding to an ion of 969 amu. This sequence can
be explained by the cross-coupling between the 584 GSG trimer
and the 388 GS dimer or between a 358 GG dimer and the 810
GSSG tetramer with a fragmentation of the resulting oligomer by
cleavage of the end-sequence b-O-4 S–G bond. The highest DP oli-
gomer fragment detected exhibits a m/z of 1972 amu, which corre-
sponds theoretically to a decamer sequence involving six G
monomers, four S monomers and four b-O-4 bonds. This fragment
is accompanied on the same mass spectrum by a series of frag-
ments which we suspect to stem from an oligomer of PD 12
(Fig. 9). We propose for this oligomer a sequence which respect
both the mass of the fragments observed and the structure of the
oxidation products available at the time when these oligomers
start to be detected (Fig. 10; 90 min). Thus, the oligomer can tenta-
tively be assigned to a structure resulting from the coupling be-
tween two 584 GSG trimers, one 810 GSSG tetramer and one 358
GG dimer. The 584 GSG trimer, which does not take part to any
endwise elongation process and remains unreactive during the
first 90 min (Fig. 10), would then be recruited in the very last stage
of the kinetics once other more reactive species are already con-
sumed. The lower reactivity of the 584 trimer compared to 644
and 614 trimers could be explained by the fact that S species are
always more reactive. Thus, the more S units are involved in the
structure, the most reactive this structure appears. Given the far
highest consumption rate of the bb GG dimer compared to its b5
analogue (Fig. 10), it is most probable that the 358 dimer present
in the PD > 4 sequence is the bb one.
To summarize, the PRX34-catalyzed oxidation of a 50/50 mol/
mol S/G mixture leads mainly to four dimers (1 SS bb, 1 SG b5, 1
GG b5 and 1 GGbb), three trimers (1 SSS, 1 SSG and 1 SGG), two tet-
ramers (1 SSSG, 1 GSSG) and a series of oligomers of PD > 4. The
successive formation and consumption of all these compounds
along the oxidation reaction is schematized in Fig. 10. Such a
scheme is particularly useful to understand how a given oligomer
sequence can be formed at a given reaction time. For instance, it al-
lowed us to exclude an endwise elongation mechanism for the for-
mation of the PD > 4 oligomers since they are no monomer left
after 30 min, time at which the PD > 4 oligomers starts to appear.
The scheme also helps to illustrate the respective reactivity of
the different species as a function of the composition of the reac-
tion medium. The four dimers identified are precisely the one iden-
tified by Kishimoto et al. (2010) after thioacidolysis of a polymeric
fraction of Zutropf DHPs synthesized also from an equimolar S/G
alcohols mixture. These authors found a molar ratio between the
thioacidolysis dimers in favour of the SS syringaresinol dimer
(GG/GS/SS = 27/23/50). The SSS trimer and the corresponding SSSG
tetramer were the major oligomers identified by Evtuguin and
Amado (2003) in the low-molar-mass fraction of Eucalyptus globu-
lus dioxane lignin, together with a SGSSG pentamer that corre-
120
100
80
60
40
20
0
(a)
O min
5 min
10 min 20 min 30 min 60 min 90 min 120 min
4 h
24 h
120
100
80
60
40
20
0
(b)
O min
5 min
10 min 20 min 30 min 60 min 90 min 120 min
4 h
24 h
Fig. 8. PRX34-catalyzed oxidation kinetics of a 50/50 mol/mol S/G mixture showing the similar shift between the formation curves of the (a; ) 614 trimer and the (b; ) 810
tetramer and the speculative aldehydic structures (a; S0SG ) and (b; GS0SG ). Relative weight contents determined by LC-ESI-MS are normalized with respect to the
maximal content of each species. Chromatographic conditions: C18 column (Highpurity, Thermo Electron Corporation, 5
lm, 150 ꢀ 4.6 mm) eluted with a 12–95 vol.%
aqueous acetonitrile, 1‰ HCOOH gradient (45 min) and 1 ml minꢁ1 flow rate; PDA signal 190–600 nm; negative ion ESI-MS spectra (120–2000 m/z) acquired using an ion
trap spectrometer (Finnigan LCQ-DECA – Thermo Electron Corporation) setting the needle voltage at 4 kV and the desolvating capillary temperature at 350 °C.

1680
N. Demont-Caulet et al. / Phytochemistry 71 (2010) 1673–1683
Fig. 9. ESI-MS spectrum (elution time: 15.91 min, S–G copolymerization mixture recovered at 90 min) showing fragments that can be tentatively assigned to structural
sequences derived from an oligomer of PD 12 (fragments corresponding to PD > 10 cannot be detected given the 120–2000 amu m/z scan range of the spectrometer). The
sequence of these oligomers is in agreement with their formation by coupling of the lower PD oligomer present in the oxidation mixture at their time when they appear.
sponds to the GSSG tetramer coupled with a S monomer (data not
shown). The more complex poplar xylem extract analyzed by et al.
(2004) also comprises all the dimers, trimers and tetramers
identified here, including the speculative aldehyde analogues. No
oligolignol of PD > 4 could be detected by the authors.
precious information to better apprehend the complex lignification
process in plants. S-type lignins in plants, such as eucalyptus lignin
are known to be associated to higher b-O-4 contents. We show
here that the b-O-4 bonds are incorporated into the polymer as
soon as the trimerization step and that the formation of such bonds
is initiated by the presence of syringaresinol homodimer.
Most of the oligomers identified in the run of our in vitro S–G
copolymerization have been identified also in the oligomer frac-
tion recovered from plant extracts. This shows that in vitro poly-
merization remains a useful and reliable tool to investigate the
polymerization mechanisms of lignin. A specificity of our study
was to implement a mechanistic approach through a kinetic inves-
tigation. This kinetic investigation of the very first steps of oligo-
merization not only informs us on the respective reactivity of
the different structures, but also provides a way to monitor the
sequential formation of the different oligomers. Based on our re-
sults, we propose a general scheme for S–G copolymerization.
We observe that an endwise and a bulk mechanisms are likely
to co-exist within the same system, the bulk mechanisms occur-
ring after total consumption of the monomers. Even if lignification
is a continuous process in plant cell wall, although not very
frequent, it can be imagined that such a situation of monomer
starvation in plants could occur at specific stages of plant develop-
ment and/or in specific tissues, which could also account for the
spatiotemporal variability of lignin structure. Results obtained in
this study could be completed in the future by an equivalent ap-
proach taking into account the continuous process of lignification
in plant cell wall and thus operating in Zutroph mode rather than
Zulauf mode.
3. Conclusions
The objectives of this study were simultaneously to test the
activity of a class III peroxidase purified for the first time from A.
thaliana and to use this plant-purified enzyme to dissect the oligo-
merization mechanisms of two main precursors of lignin, G and S
alcohols. We not only confirmed that the PRX34 peroxidase can
initiate the oxidation-catalyzed oligomerization of an equimolar
S/G mixture but also demonstrate that its in vitro activity is identi-
cal to that of the HRP. This similarity could be explained from the
high sequence homology (94%) observed between those two pro-
teins. As for HRP, the PRX34 shows a high substrate specificity
and seems unable to directly oxidize S monomers.
By monitoring the oxidation of a mixture of S and G alcohol in
comparison with the oxidation of S or G alcohol alone, we succeed
in evaluating the polymerization mode recruited in each condition.
We show that the presence of S monomers converted the polymer-
ization of G monomers from a bulk to an endwise mechanism. We
also confirm the role of G alcohol as radical mediator and identify
key structures associated with the incorporation of S alcohol. This
study brings new information on the compared reactivity of S and
S/G lignin model compounds and is supposed to provide some

N. Demont-Caulet et al. / Phytochemistry 71 (2010) 1673–1683
1681
Fig. 10. General schematic representation of the formation and consumption kinetics of the different species identified in the PRX34-catalyzed oxidation medium from a
50/50 mol/mol S/G mixture: (S) S alcohol; (G) G alcohol; (SS 418 bb) S syringaresinol homodimer; (SSS 644:bb;bO4) S homodimer derived from the syringaresinol dimer;
(GSSS 840:bO4;bb;bO4) heterotetramer derived from the S homotrimer; (SSG 614:bb;bO4) heterotrimer derived from the syringaresinol dimer; (S0SG) speculative
heterotrimer aldehyde; (GSSG 810:bO4;bb;bO4) heterotetramer derived from the SSG heterotrimer, (GS0SG) speculative heterotetramer aldehyde; (GG 358 bb) G pinoresinol
homodimer; (GG 358 b5) G phenyl coumaran homodimer and (GS 358 b5) its heterodimer analogue; (GSG 584:b5;bO4) heterotrimer derived from the GS b5 dimer; (PD > 4)
heterooligomers of polymerization degree higher than 4 derived only from dimers, trimers and tetramers present in the mixture at 30 min (at this reaction time, no monomer
are anymore available for an elongation through an endwise mechanism).
uble protein extract was loaded and the column washed
with 10 ml of buffer. The proteins were eluted with 0.2 M
4. Experimental
methyl-a-glucopyranoside in the same buffer. The eluates
4.1. Plant material
were collected (1 ml per fraction) and 3 or 5
ll samples from
each fraction were tested for peroxidase activity. Pooled
fractions showing peroxidase activity were equilibrated in
25 mM Tris–HCl buffer (pH 7.4) containing 5% glycerol (v/
v) and 0.015% Triton X-100 (v/v). Glycerol was added to
the buffer to prevent the partial enzymatic inactivation.
(ii) Cation exchange chromatography: The soluble protein extract
was loaded on a CM-Sepharose (Sigma) cation exchange col-
umn (1.5 ꢀ 2.5 cm). The proteins were eluted with the same
buffer, first alone and then with a 0.0–0.5 M NaCl discontin-
uous gradient using 2 ml of NaCl solution, increasing by
The ecotypes Wassilewsija (WS) and Columbia (ColO) were
used in this work. Two homozygous lines were used for PERX 34
overproducing. Wild type or mutant was identified in the A. thali-
ana collection of Versailles (Bouche and Bouchez, 2001) and in the
Salk Institute Genomic Analysis Laboratory (Alonso et al., 2003).
The Atprx34 line was selected using systematic border sequencing
program of the Salk Institute (San Diego, USA) on TAIR site.
A construct including cDNA of PRX34 (clone WS Versailles col-
lection) under the control of CaMV35S promotor and fusionned
with GFP (codon stop in-phase) was performed using the gateway
binary vector (pGWB5, Invitrogen). The transformed and the wild
type lines were grown together in greenhouse to ensure uniform
environmental conditions. Stems (5–6 weeks-old) were harvested
and used for peroxidase purification.
0.025 M. One-ml fractions were collected and 3 or 5
assayed for peroxidase activity.
ll
4.3. Identification of purified peroxidase by Maldi-TOF MS
4.2. Purification of PRX34 A. thaliana peroxidase
In order to identify the gene sequence of the Per34, the 44.7 kDa
protein band observed in the SDS–PAGE gel was in-gel digested with
Approximately 7 g of Arabidopsis stems were blended for 5 min,
suspended in 7 ml of extraction buffer (25 mM BisTris (pH 7),
0.5 lg of trypsin (Gold, Promega) over-night. The peptides were ex-
tracted with 2 ꢀ 0.1 trifluoro acetic acid (TFA) in acetonitrile/25 mM
200 mM CaCl2, 10% (v/v) glycerol, 4
l
M Na-cacodylate and 1/200
ammonium carbonate buffer (60/40, v:v) and concentrated. The
(v/v) protease inhibitor cocktail (P-9599; Sigma)) and blended
5 min. The material was centrifuged twice at 8 °C for 5 min at
3000g and one time at 4 °C for 5 min at 13,000g. The supernatant
was additionally centrifuged at 8 °C for 45 min at 15,000g. The pro-
tein purification was performed in two steps:
matrix solution contained 0.3 g/l of 4-hydroxy-
acid in ethanol/acetone (2/1, v:v).
Equal volumes of purified peptides sample or calibration stan-
dard (peptide calibration mixture Pepmix 1, LaserBio Labs, Sophia
Antipolis, France) were mixed with the matrix solution. One ll of
a-cyanocinnamic
this matrix/sample mixture was applied onto the target and let
(i) Affinity chromatography on a 0.5 ꢀ 3 cm column filled with
1 ml of Concanavalin-A Sepharose (Sigma) and washed with
3 ml of 20 mM Tris–HCl, 0.5 M NaCl buffer (pH 7.4). The sol-
to dry at room temperature.
Mass spectra were acquired on a Bruker Reflex III MALDI-TOF
instrument equipped with a nitrogen laser with an emission wave-

1682
N. Demont-Caulet et al. / Phytochemistry 71 (2010) 1673–1683
length of 337 nm. Spectra were obtained in the reflectron mode at
an accelerating voltage of 19 kV. FlexAnalysis software was used
for data analysis.
4.7. Size-exclusion chromatography (SEC) analysis of oligomeric
products
Proteins were identified on the using peptide mass fingerprints
and Mascot interface (http://www.matrixscience.com/) in May
2007. The A. thaliana database was queried, with no more than
one missed cleavage.
The extracted soluble fractions were analyzed by SEC, using a PL-
Gel column (Polymer Laboratories, 5
lm, 100 Å, 600 ꢀ 7.5 mm)
with tetrahydrofuran (THF) as an eluent (stabilized THF, JT Baker;
1 ml minꢁ1) and 280 nm UV detection. The polymerization degrees
(PD) were assigned according to the apparent molar masses of the
compounds based on a calibration with polyethylene oxide stan-
dards (Igepal, Aldrich) and purified lignin model compounds
(Baumberger et al., 2003).
4.4. In vitro peroxidase activity assays
A 20 mg mlꢁ1 solution of the substrate 3,30,5,50-tetramethylben-
zydine (TMB) was prepared in DMSO, and stored in aliquots at
ꢁ20 °C.
4.8. LC/MS analysis of oligomeric products
The extracted soluble fractions were analyzed by reverse-phase
LC with electro spray ionisation (ESI)-MS and photodiode array
(PDA) co-detection. The methanolic solutions were ultra filtrated
The reaction mixture contained 100 mM acetate citric acid buf-
fer (pH 6), 0.4
extract in a total volume of 800
room temperature for 1 min and stopped by the addition of 30
l
l of 6% (w/v) H2O2, 4
ll TMB and 1–3 ll of protein
ll. The reaction was carried out at
(0.45
lm – GHP Acrodisc – Gelman) and injected on a C18 column
l
l
(Highpurity, Thermo Electron Corporation, 5
lm, 150 ꢀ 4.6 mm)
concentrated H2SO4 to the assay mixture. The TMB oxidation was
using a 12–95 vol.% aqueous acetonitrile, 1‰ HCOOH gradient
(45 min) and 1 ml minꢁ1 flow rate. Negative ion ESI-MS spectra
(120–2000 m/z) were acquired using an ion trap spectrometer
(Finnigan LCQ-DECA – Thermo Electron Corporation) setting the
needle voltage at 4 kV and the desolvating capillary temperature
at 350 °C. The synthesized products were assigned according to
their PDA (190–600 nm) and mass spectra as reported by Morreel
et al. (2004). The polymerization degree and amount of b-O-4
bonds of the oligomers were determined according to the mass
of the deprotonated ion.
monitored by the increase of absorbance at 450 nm.
4.5. SDS–PAGE and protein gel blot analysis
Protein-denaturating SDS–PAGE was carried out using 10%
polyacryamide gels. Standard markers (molecular range 15–
100 kDa; Sigma) were used to determine the approximate molecu-
lar masses of purified proteins in silver stained gels.
Proteins were transferred onto 0.45 lm Hybond ECL membrane
(Amersham Biosciences, Piscataway, NJ) by electroblotting. Protein
detection was performed using first polyclonal antibodies raised
against peroxidases (Buffard et al., 1990) and second antibodies
phosphatase alkaline conjugate.
Acknowledgements
The authors sincerely acknowledge Dr. Bernard Cathala for his
help in synthesizing G and S monomers, François Perreau on the
Plateau Technique Spécifique de Chimie du Végétal of IJPB for Mal-
di-TOF analyses and Dr. Paul-Henri Ducrot for his critical reading of
the manuscript and fruitful discussions.
4.6. Oxidative coupling of coniferyl (G) and sinapyl (S) alcohols
The LC-MS equipment was obtained from a common financial
support of INRA and Région Ile de France (SESAME Grant).
This work was funded by the French Research National Agency
(ANR) in the context of the MAGIC program (ANR-08-BLAN-0307).
G and S alcohols were kindly synthesized by Dr. B. Cathala,
according to the method of Ludley and Ralph (1996). The internal
standard 3,4,5-trimethoxybenzoic acid (TMBA) was obtained from
Fluka (Buchs, Switzerland) and hydrogen peroxide (30 wt.% solu-
tion) from Acros (New Jersey). Horseradish peroxidase type II
(HRPII, 180 U/mg) was purchased from Sigma Chemical Co. (St.
Louis, MO). Phosphate buffer was prepared from sodium dihydro-
gen phosphate (NaH2PO4ꢂ2H2O) and di-sodium hydrogen phos-
phate (Na2HPO4ꢂ2H2O) (analytical reagents; Prolabo, France). The
3,30,5,50-tetramethylbenzydine substrate (TMB) was purchased
from Sigma (St. Louis, MO, USA).
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.phytochem.2010.06.011.
References
These oxidative coupling experiments were carried out at a
microscale level according to a modified version of Fournand
and Lapierre (2001). An internal standard of 2 mM TMBA
Alonso, J.M., Stepanova, A.N., Leisse, T.J., Kim, C.J., Chen, H., Shinn, P., Stevenson, D.K.,
Zimmerman, J., Barajas, P., Cheuk, R., Gadrinab, C., Heller, C., Jeske, A., Koesema,
E., Meyers, C.C., Parker, H., Prednis, L., Ansari, Y., Choy, N., Deen, H., Geralt, M.,
Hazari, N., Hom, E., Karnes, M., Mulholland, C., Ndubaku, R., Schmidt, I., Guzman,
P., Aguilar-Henonin, L., Schmid, M., Weigel, D., Carter, D.E., Marchand, T.,
Risseeuw, E., Brogden, D., Zeko, A., Crosby, W.L., Berry, C.C., Ecker, J.R., 2003.
Genome-wide insertional mutagenesis of Arabidopsis thaliana. Science 301,
653–657.
Aoyama, W., Sasaki, S., Matsumura, S., Mitsunaga, T., Hirai, H., Tsutsumi, Y., Nishida,
T., 2002. Sinapyl alcohol-specific peroxidase isoenzyme catalyzes the formation
of the dehydrogenative polymer from sinapyl alcohol. J. Wood Sci. 48, 497–504.
Barcelo, A.R., Ros, L.V.G., Carrasco, A.E., 2007. Looking for syringyl peroxidases.
Trends Plant Sci. 12, 486–491.
(800
to 1600
G and S alcohols in the same buffer. A solution of 4 mM hydrogen
peroxide (800 l) in ultrapure water was then added before the
addition of 1600 l of a 50 ng/ml solution of horseradish peroxi-
dase type II in the 10 mM NaH2PO4–Na2HPO4, pH 5.5, buffer or
solution of purified A. thaliana PRX34 peroxidase in the
l
l) in 10 mM NaH2PO4–Na2HPO4, pH 5.5, buffer was added
l
l of 2 mM G alcohol, S alcohol or a mixture 50/50 of both
l
l
a
Baumberger, S., Fournand, D., Pollet, B., Lapierre, C., Ducrot, P.-H., 2003. Peroxidase-
catalyzed polymerization of monolignols: HPSEC investigation of the
oligomerization step. In: Proceedings of the 12th ISWPC, Madison, pp. 165–168.
Bouche, N., Bouchez, D., 2001. Arabidopsis gene knockout: phenotypes wanted.
Curr. Opin. Plant Biol. 4, 111–117.
Brunow, G., Kilpelainen, I., Sipila, J., Syrjanen, K., Karhunen, P., Setala, H.,
Rummakko, P., 1998. Oxidative coupling of phenols and the biosynthesis of
lignin. In: Lewis, N.G., Sarkanen, S. (Eds.), Lignin and Lignan Biosynthesis.
American Chemical Society, Washington, DC, pp. 131–147.
10 mM NaH2PO4–Na2HPO4, pH 5.5, buffer. A reference was made
by replacing peroxidase solution by 10 mM NaH2PO4–Na2HPO4,
pH 5.5, buffer. Mixtures were kept at 30 °C, and aliquots of the
oxidation reaction medium (600 ll) were regularly taken, mixed
with HCl 0.1 N and extracted with a 50/50 v/v dichloromethane/
ethyl acetate mixture before evaporation of the organic phase to
dryness.

N. Demont-Caulet et al. / Phytochemistry 71 (2010) 1673–1683
1683
Buffard, D., Breda, C., van Huystee, R.B., Asemota, O., Pierre, M., Ha, D.B., Esnault, R.,
1990. Molecular cloning of complementary DNAs encoding two cationic
peroxidases from cultivated peanut cells. Proc. Natl. Acad. Sci. USA 87, 8874–
8878.
Evtuguin, D.V., Amado, F.M.L., 2003. Towards a better understanding of lignin
assembly: the assessment of the sequential order of building blocks and inter-
units linkages using electrospray ionisation mass spectrometry. In: Proceedings
of the 12th ISWPC, Madison, pp. 277–280.
Nielsen, K.L., Indiani, C., Henriksen, A., Feis, A., Becucci, M., Gajhede, M., Smulevich,
G., Welinder, K.G., 2001. Differential activity and structure of highly similar
peroxidases. Spectroscopic, crystallographic, and enzymatic analyses of
lignifying Arabidopsis thaliana peroxidase A2 and horseradish peroxidase A2.
Biochemistry 40, 11013–11021.
Quiroga, M., Guerrero, C., Botella, M.A., Barcelo, A., Amaya, I., Medina, M.I., Alonso,
F.J., de Forchetti, S.M., Tigier, H., Valpuesta, V., 2000. A tomato peroxidase
involved in the synthesis of lignin and suberin. Plant Physiol. 122, 1119–1127.
Sasaki, S., Nishida, T., Tsutsumi, Y., Kondo, R., 2004. Lignin dehydrogenative
polymerization mechanism: a poplar cell wall peroxidase directly oxidizes
polymer lignin and produces in vitro dehydrogenative polymer rich in beta-O-4
linkage. FEBS Lett. 562, 197–201.
Syrjanen, K., Brunow, G., 2000. Regioselectivity in lignin biosynthesis. The influence
of dimerization and cross-coupling. J. Chem. Soc. Perkin Trans. 1, 183–187.
Takahama, U., Oniki, T., 1994. Effects of ascorbate on the oxidation of derivatives of
hydroxycinnamic acid and the mechanism of oxidation of sinapic acid by cell
wall-bound peroxidases. Plant Cell Physiol 35, 593–600.
Fournand, D., Lapierre, C., 2001. Capillary zone electrophoresis of coniferyl alcohol
oxidation products. J. Agric. Food Chem. 49, 5727–5731.
Fournand, D., Cathala, B., Lapierre, C., 2003. Initial steps of the peroxidase-catalyzed
polymerization of coniferyl alcohol and/or sinapyl aldehyde: capillary zone
electrophoresis study of pH effect. Phytochemistry 62, 139–146.
Freudenberg, K., Hubner, H.H., 1952. Hydroxycinnamyl alcohols and their
dehydrogenation polymers. Chem. Ber. 85, 1181–1191.
Gabaldon, C., Lopez-Serrano, M., Pedreno, M.A., Barcelo, A.R., 2005. Cloning and
molecular characterization of the basic peroxidase isoenzyme from Zinnia
elegans, an enzyme involved in lignin biosynthesis. Plant Physiol. 139, 1138–
1154.
Kishimoto, T., Chiba, W., Saito, K., Fukushima, K., Uraki, Y., Ubukata, M., 2010.
Influence of syringyl to guaiacyl ratio on the structure of natural and synthetic
lignins. J. Agric. Food Chem. 58, 895–901.
Ludley, F.H., Ralph, J., 1996. Improved preparation of coniferyl and sinapyl alcohols.
J. Agric. Food Chem. 44, 2942–2943.
Marjamaa, K., Kukkola, E., Lundell, T., Karhunen, P., Saranpaa, P., Fagerstedt, K.V.,
2006. Monolignol oxidation by xylem peroxidase isoforms of Norway spruce
(Picea abies) and silver birch (Betula pendula). Tree Physiol. 26, 605–611.
Méchin, V., Baumberger, S., Pollet, B., Lapierre, C., 2007. Peroxidase activity can
dictate the in vitro lignin dehydrogenative polymer structure. Phytochemistry
68, 571–579.
Takahama, U., Oniki, T., Shimokawa, H., 1996.
A possible mechanism for the
oxidation of sinapyl alcohol by peroxidase-dependent reactions in the apoplast:
enhancement of the oxidation by hydroxycinnamic acids and components of
the apoplast. Plant Cell Physiol. 37, 499–504.
Tobimatsu, Y., Takano, T., Kamitakahara, H., Nakatsubo, F., 2008a. Studies on the
dehydrogenative polymerizations (DHPs) of monolignol beta-glycosides: Part 4.
Horseradish peroxidase-catalyzed copolymerization of isoconiferin and
isosyringin. Holzforschung 62, 495–500.
Tobimatsu, Y., Takano, T., Kamitakahara, H., Nakatsubo, F., 2008b. Studies on the
dehydrogenative polymerization of monolignol beta-glycosides: Part 5. UV
spectroscopic monitoring of horseradish peroxidase-catalyzed polymerization
of monolignol glycosides. Holzforschung 62, 501–507.
Tobimatsu, Y., Takano, T., Kamitakahara, H., Nakatsubo, F., 2008c. Studies on the
dehydrogenative polymerizations of monolignol beta-glycosides. Part 3:
Horseradish peroxidase-catalyzed polymerizations of triandrin and
isosyringin. J. Wood Chem. Technol. 28, 69–83.
Tobimatsu, Y., Takano, T., Kamitakahara, H., Nakatsubo, F., 2010. Studies on the
dehydrogenative polymerization of monolignol beta-glycosides. Part 6:
Monitoring of horseradish peroxidase-catalyzed polymerization of monolignol
glycosides by GPC–PDA. Holzforschung 64, 173–181.
Minic, Z., Jamet, E., Negroni, L., der Garabedian, P.A., Zivy, M., Jouanin, L., 2007.
A
sub-proteome of Arabidopsis thaliana mature stems trapped on
Concanavalin is enriched in cell wall glycoside hydrolases. J. Exp. Bot.
58, 2503–2512.
A
Morreel, K., Ralph, J., Kim, H., Lu, F.C., Goeminne, G., Ralph, S., Messens, E., Boerjan,
W., 2004. Profiling of oligolignols reveals monolignol coupling conditions in
lignifying poplar xylem. Plant Physiol. 136, 3537–3549.