Hydrophobic metabolites of 2,4-dichlorophenoxyacetic acid (2,4-D) in cultured coconut tissue

Article abstract of DOI:10.1016/j.phytochem.2004.08.034

Metabolites of 2,4-D, synthesised by coconut plumules cultured in vitro, included lipophilic chain-elongated forms with aliphatic chains of up to 16 carbons esterified to form triacylglycerol analogues. Cultures of inflorescence and plumular tissues of co

Full text of DOI:10.1016/j.phytochem.2004.08.034

PHYTOCHEMISTRY
Phytochemistry 65 (2004) 2763–2774
www.elsevier.com/locate/phytochem
Hydrophobic metabolites of 2,4-dichlorophenoxyacetic acid
(2,4-D) in cultured coconut tissue
1
Arturo Lopez-Villalobos , Roland Hornung , Peter F. Dodds
2
*
´
Department of Agricultural Sciences, Imperial College London, Wye campus, Wye, Ashford, Kent TN25 5AH, UK
Received 5 April 2004; received in revised form 27 July 2004; accepted 12 August 2004
Abstract
Cultures of inflorescence and plumular tissues of coconut palm (Cocos nucifera L.) were maintained in the presence of the auxin,
[14C]2,4-dichlorophenoxyacetic acid (2,4-D), so that its metabolic fate could be studied. Thin layer chromatography of methanol
extracts of the plumular tissue showed that four classes of metabolites, as well as the unchanged acid, were recovered in the extract.
In inflorescence tissue, only the unchanged acid and the most polar class of metabolites (metabolite I) were recovered. Metabolite I
was shown to consist mostly of a mixture of sugar conjugates and metabolite II (the next most polar) was an unidentified basic
metabolite. Metabolites III and IV were both novel triacylglycerol analogues in which one of the natural fatty acids was replaced
with a chain-elongated form of 2,4-D. Reversed-phase thin layer chromatography was used to identify the 2,4-D-derived acids and it
was found that metabolite III contained the 2,4-dichlorophenoxy-moiety attached to a chain-length of between 2 and 12 carbons,
whereas metabolite IV contained 12, 14 and 16 carbon chain lengths. In inflorescence tissue, and in plumular tissue at low sucrose or
2,4-D concentrations and after short periods in culture, metabolite I predominated. The other metabolites increased as a percentage
when plumular culture was prolonged or when sucrose or 2,4-D concentrations were raised. These changes correlated with better
development of the explant.
ꢀ 2004 Elsevier Ltd. All rights reserved.
Keywords: Coconut; Cocos nucifera L; Tissue culture; 2,4-Dichlorophenoxyacetic acid; Xenobiotic metabolism; Chain-elongation; Triacylglycerol;
Xenobiotic lipid
1. Introduction
propagated in vitro but the rate of conversion, in spite
of recent advances, into viable plantlets is still very
low (Chan et al., 1998; Verdeil et al., 1994).
There is a strong desire to develop a reliable method
for the clonal propagation of the coconut palm (Cocos
nucifera L.) for reasons of improvement of the farmed
phenotypes, disease avoidance and ease of production
at high quality (Blake, 1995; Buffard-Morel et al.,
1995). Both plumular and inflorescence tissue can be
One of the factors that is important in the induction
of callus formation, an important stage in coconut so-
matic embryogenesis, is the presence of an auxin: low
concentrations of the herbicide, 2,4-dichlorophenoxy-
acetic acid (2,4-D), are routinely added to culture media
to serve this function (Blake and Hornung, 1995; Buff-
ard-Morel et al., 1995). The effectiveness of 2,4-D de-
pends upon the concentration of 2,4-D in the culture
medium, its availability to the plant (i.e. the concentra-
tion in free solution rather than bound to activated char-
coal), the rate of uptake and its rate of metabolism by
the plant to form metabolites that are either inactive
*
Corresponding author. Tel.: +44 20 7594 2869; fax: +44 20 7594
2640.
E-mail address: p.dodds@imperial.ac.uk (P.F. Dodds).
Current address: Department of Biological Sciences, Faculty of
1
Sciences, University of Calgary, 2500 University Drive, N.W., Calgary,
Alta., Canada.
2
Current address: Salmanskirchen 18, 84539 Ampfing, Germany.
0031-9422/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.phytochem.2004.08.034

´
A. Lopez-Villalobos et al. / Phytochemistry 65 (2004) 2763–2774
2764
or are stored away from the site of action. After callus
formation is complete, it is necessary to reduce substan-
tially the concentration of 2,4-D which, otherwise, can
act as a mutagen and interfere with further ontogenesis
(Blake, 1995; LoSchiavo et al., 1989). Its removal from
the culture medium is straightforward but reductions
in intracellular concentrations depend upon conversion
into inactive metabolites.
1987; Sabater and Rubery, 1987) to be drawn. The effect
of 2,4-D concentrations may have been direct upon an
active transport system or indirect by stimulating an
intracellular transport or metabolism system, both of
which would have the effect of displacing the equilib-
rium of trans-membrane movement towards uptake.
2.2. Identification and quantification of 2,4-D metabolites
Studies of 2,4-D metabolism in a number of plants
have revealed many routes of metabolism (Roberts
et al., 1998) including ester conjugation with sugars, ami-
no acid conjugation, formation of conjugates of hyd-
roxyl metabolites, chain oxidation and even chain
elongation (to form ‘‘xenobiotic lipid’’ metabolites).
Here we report the results of studying metabolism of
2,4-D by plumular and inflorescence tissues in vitro
and the effects of sucrose concentration, 2,4-D concen-
tration and stage of development upon that metabolism.
We present evidence for the biosynthesis of chain elon-
gated 2,4-D acids, esterified to form analogues of triacyl-
glycerols (TAG) which have not, hitherto, been reported.
Fig. 2 shows a typical radio-scan of a silica gel thin
layer chromatography (TLC) plate used to separate
the 2,4-D metabolites extracted from plumular explants.
Five clear ‘‘peaks’’ of radioactivity can be observed, one
of which co-chromatographed with the (unchanged) free
acid form of 2,4-D. Even this preliminary analysis sug-
gested that the classes may contain more than one com-
ponent each. Metabolite I, which did not migrate from
the origin, and metabolite II were more polar than the
2,4-D precursor whereas metabolites III and IV were
hydrophobic and migrated close to the solvent front
(SF) in positions where neutral lipids might be expected.
Control calli, killed by freezing and thawing, contained
radioactivity corresponding only to unmodified, non-
esterified, 2,4-D.
2. Results and discussion
2.1. Uptake of 2,4-D by cultured tissues
2.2.1. Metabolite I
After recovery from the plate, metabolite I was acetyl-
ated and the products were subjected once more to TLC.
A fraction, representing about 17% of metabolite I,
co-chromatographed with the chemically synthesised
tetra-acetyl-glucose ester of 2,4-D (Fig. 3). The other
components of metabolite I all migrated further on the
plate than the non-acetylated ones. Incubation of metab-
olite I with the enzyme b-glucosidase, liberated about
14% of the metabolite as 2,4-D. Controls indicated that
about half of this conversion could be accounted for by
chemical decomposition, a finding that agrees with earlier
observations of the instability of chemically synthesised
2,4-D glucose ester (see Section 4). Incubation of metab-
olite I with protease also released about 18% of the
radioactivity in the form of 2,4-D indicating that a pro-
portion of metabolite I is a peptide-bonded conjugate
of 2,4-D. In summary, the evidence that suggests that
metabolite I comprises a mixture of carbohydrate conju-
gates is as follows: (i) TLC mobility (normal phase) is
consistent with a very polar metabolite; (ii) treatment
with b-glucosidase released a proportion of the radioac-
tivity as 2,4-D (non-esterified form); (iii) acetylation of
metabolite I reduced the polarity of all components, indi-
cating the presence of free hydroxyl groups; (iv) one com-
ponent, after acetylation, co-chromatographed with
Early stages of somatic embryogenesis were achieved
as described earlier (Blake and Hornung, 1995) in all
three tissues used (inflorescences at stages ꢀ4 and ꢀ5
and plumular tissue). This statement is based on visual
assessment and histological studies.
All tissues, cultured in the presence of [14C]2,4-D,
took 2,4-D up from the medium throughout the course
of the experiments. Fig. 1 (a)–(c) shows that the total
uptake of 2,4-D increased as a function of time in cul-
ture, of the concentrations of 2,4-D and sucrose in the
culture medium. The 2,4-D concentrations (0–0.6 mM)
used are those used in previous reports (Chan et al.,
1998; Hornung and Verdeil, 1999). Uptake was greater
in plumular tissue than in inflorescence tissue. In con-
trast, control incubations (using tissue that had been
killed by freezing and thawing) took up only insignifi-
cant quantities of radioactivity. When these results were
expressed ‘‘per mass of tissue’’ (not shown) the 2,4-D
content of the tissue showed relatively little response
to sucrose concentration but increased with increasing
2,4-D concentration in the medium. The increase in total
uptake in the presence of increasing sucrose concentra-
tions can, therefore, be interpreted as being secondary
to the effects of sucrose on tissue growth.
On their own, these results do not allow conclusions
about whether the uptake is by passive diffusion (Koens
et al., 1995; Shvetson and Gamburg, 1984) or by an ac-
tive transport (or facilitated diffusion) mechanism (Del-
barre et al., 1996; Minocha and Nissen, 1985; Rubery,
synthetic tetra-acetyl-2,4-D-glucose ester.
The possibility that some components were ether-
conjugates of sugars with a hydroxylated metabolite of
2,4-D cannot be totally excluded. Work with young
wheat plants has shown that 2,4-D can also form conju-

´
A. Lopez-Villalobos et al. / Phytochemistry 65 (2004) 2763–2774
2765
30
25
20
15
10
5
20
15
10
0
5
0
0
20
40
60
80
100
(a)
Period of culture (Days)
30
20
10
0
0
0.05
0.1
0.15
0.2
0.25
(c)
Concentration of sucrose (M)
0
0.2
0.4
0.6
(b)
Concentration of 2,4-D (mM)
Fig. 1. Graphs to show the uptake of [14C]2,4-D into methanolic extracts of plumules (a and b) or plumules and inflorescence tissue (c). The tissue
was grown in culture and incubated with [14C]2,4-D as described under Section 4. The standard conditions used, except as indicated on the horizontal
axes of the individual graphs, were 60 days in culture, 0.4 mM 2,4-D and 0.116 M sucrose. For inflorescence tissue the 60-day results are the sum of
the uptakes measured separately in the first 30 days period of culture and the second 30 days period of culture; plumular tissue was exposed to
radioactive 2,4-D continuously over the whole 60 days. The effects of duration of incubation (a), 2,4-D concentration in the incubation medium (b),
and sucrose concentration in the incubation medium (c) are shown. Symbols indicate plumular tissue (d), inflorescence tissue at developmental stage
ꢀ5 (m) or stage ꢀ4 (j). Error bars represent means standard errors of the mean from five individual cultures but are not shown where they are
smaller than the point.
gates with a number of mono-, di- and poly-saccharides
(Chkanikov et al., 1982). Conjugates of xenobiotic carb-
oxylic acids with 6-O-malonylglucose, glucosylxylose,
gentiobiose and triglucose have been noted in other
plant species (Mikami et al., 1984, 1985).
The finding that about 18% of metabolite I was liber-
ated as 2,4-D after proteolytic digestion suggests the
presence of peptide bonded 2,4-D. The evidence is not
consistent with the formation of simple amino acid con-
jugates because metabolite I is too polar on silica gel
TLC so a conjugate with a bipeptide or a longer peptide
chain is suggested. Conjugates of hydroxylated 2,4-D
with peptides between 2 and 12 amino acids in length
have been reported in wheat, barley and maize (Chkani-
kov et al., 1982). The same group, also working with
young wheat and barley, had earlier reported the forma-
tion of peptide linked conjugates of 2,4-D, its hydroxyl-
ated metabolites and the glycosyl conjugates thereof

´
A. Lopez-Villalobos et al. / Phytochemistry 65 (2004) 2763–2774
2766
with chains of up to 220 amino acids (Nazarova et al.,
1980).
2.2.2. Metabolite II
The possibility that metabolite II was an amino acid
conjugate, or a mixture of amino acid conjugates, of 2,4-
D was investigated by comparing its chromatographic
behaviour with that of a number of synthetic 2,4-D-ami-
no acyl conjugates in a variety of solvent systems. In an
acidic solvent system of low polarity, diethyl ether:pe-
troleum spirit (60–80 ꢁC):formic acid (70:30:2, by vol-
ume), metabolite II did not migrate from the origin
nor did the standards for the 2,4-D conjugates with gly-
cine or histidine. The aspartic acid, glutamic acid, trypt-
ophan, alanine, phenyl alanine, valine and leucine
conjugates all migrated away from the origin. In a more
polar acidic system, of toluene:2-butanone:acetic acid
(45:55:3, by volume), only the histidine conjugate stayed
on the origin with metabolite II. This evidence suggested
that metabolite II is a basic compound. Increasing the
polarity further to butanol:acetic acid:water (90:20:10,
by volume) (Fig. 4(a)) resulted in both metabolite II
and the histidine conjugate migrating from the origin
but the two had clearly different Rf values. There was
also evidence for at least one other (minor) component
of metabolite II. Using a basic solvent system confirmed
the basic nature of the main components of metabolite
II (Fig. 4(b)). Only a small proportion of metabolite II
remained on the origin with the two acidic amino acid
conjugates; the major proportion of metabolite II mi-
grated further than all the available standards of 2,4-D
amino acid conjugates. Attempted hydrolysis of metab-
olite II by protease action or by refluxing with ethanolic
KOH produced no change in the chromatographic
behaviour of the metabolite in several chromatographic
systems and no evidence for the release of a radioactive
acidic compound.
Fig. 2. Radiochromatogram of 2,4-D metabolites. Plumules were
incubated with [14C]2,4-D as described in the text. The tissue was
extracted with methanol and the extract subjected to TLC on a silica
gel plate developed in toluene:butanone:acetic acid (45:55:3, by
volume). The plate was visualized for non-radioactive standards and
radio-scanned for 30 min as described in Section 4. This figure
represents a ‘‘typical’’ scan and was obtained from tissue cultured in
0.4 mM 2,4-D and 0.116 M sucrose for 90 days. The position of the
origin (Or) and solvent front (SF) are indicated.
The chromatographic behaviour of the major
component of metabolite II suggests that it is a basic
metabolite. The metabolite did not, however, co-chro-
matograph with any of the synthetic amino acid conju-
gates of 2,4-D including the conjugate with histidine.
Conjugates with lysine or arginine were not available
to use as standards but published relative Rf values for
amino acid conjugates (Feung et al., 1973) suggest that
these also were not probabilities. The resistance of
metabolite II to proteolytic digestion rules out the
possibility that the conjugate was with a bipeptide (or
longer peptide chain) and suggested that the 2,4-D was
not conjugated via a peptide (amide) bond. The failure
to liberate an acidic metabolite after attempted chemical
hydrolysis appears to confirm this view but it is possible
that the conditions chosen for hydrolysis, 1 M KOH
which is appropriate for lipid hydrolysis, were not suffi-
ciently rigorous for amide bonds, with 7 M alkali being
used on occasion (Chkanikov et al., 1976; Nazarova
Fig. 3. Radiochromatogram of metabolite I. Metabolite I was isolated
and subjected to further TLC as described in Section 4. The plate was
developed in toluene:methanol:acetic acid (96:8:4, by volume). The
plate was scanned at 0.5 cm intervals as described and the resultant
scans assembled into the 2-dimensional representation. Radioactivity
is shown on a logarithmic grey-scale with the more active areas being
represented by the darker shading. The scale runs between 9 and 1250
counts accumulated over 30 min. Lane 1 shows unmodified metabolite
I; lane 2 is metabolite I after acetylation and lane 3 is a radioactive
standard for 2,4-D. Tick marks on the borders of the plate are 1 cm
apart. The position of the Or and SF are indicated.

´
A. Lopez-Villalobos et al. / Phytochemistry 65 (2004) 2763–2774
2767
yield basic conjugates and remain possibilities despite
never having been reported before in plant tissue. The
polyamines have been shown to be present in cultured
callus of soybean (Du Plessis et al., 1996).
2.2.3. Metabolites III and IV
From the chromatographic behaviour (Fig. 2) it ap-
peared that metabolite IV might be a triacylglycerol
and that metabolite III might be elongated forms of
2,4-D as has been reported in alfalfa (Linscott et al.,
1968; Linscott and Hagin, 1970). However, neither
metabolite migrated exactly with the standards of the
elongated acids. This impression was confirmed after
TLC in a basic solvent system (results not shown). After
alkaline hydrolysis, extracted acids were identified by
chromatography on reversed-phase TLC plates (Fig.
5). Under these conditions the elongated versions of
2,4-D from the 2 carbon side-chain (2,4-D itself) to the
16 carbon form (2,4-dichlorophenoxyhexadecanoic
acid) can be distinguished. Hydrolysed metabolite IV
showed three main products which co-migrated with
Fig. 4. Radiochromatograms of metabolite II. Metabolite II was
isolated and subjected to further TLC as described in Section 4. The
plate was developed in (a) butanol:acetic acid:water (90:20:10, by
volume) or in (b) chloroform:methanol:concentrated ammonia solu-
tion (75:35:2, by volume). The plate was scanned at 0.5 cm intervals as
described and the resultant scans assembled into the 2-dimensional
representation. Radioactivity is shown on a logarithmic grey-scale with
the more active areas being represented by the darker shading. The
scale runs (a) between 12 and 250 counts or (b) between 13 and 850
counts accumulated over 30 min. In (a, Lanes 1–6) and 8–10 show the
positions of non-radioactive standards visualized by quenching of the
fluorescent indicator under UV illumination. The standards are
conjugates of 2,4-D with the following amino acids: 1, aspartic acid;
2, glutamic acid; 3, valine; 4, leucine; 5, phenyl alanine; 6, alanine; 9,
tryptophan; 10, histidine. Lane 8 shows unmodified 2,4-D and lane 7 is
the radioactive scan of metabolite II. In b lanes, 1–5 and 7–11 show the
positions of non-radioactive standards. The standards are conjugates
of 2,4-D with: 1, glycine; 2, aspartic acid; 3, glutamic acid; 4, valine; 5,
leucine; 7, phenyl alanine; 8, tryptophan; 10, histidine.; 11, alanine.
Lane 9 shows unmodified 2,4-D and lane 6 is the radioactive scan of
metabolite II. Tick marks on the borders of the plates are 1 cm apart.
The position of the Or and SF are indicated.
Fig. 5. Radiochromatogram of metabolites III and IV. Metabolites III
and IV were isolated and subjected to further TLC as described in
Section 4. The reversed phase plate was developed in acetonitrile:meth-
anol:water (80:10:30, by volume). The plate was scanned at 0.5 cm
intervals as described and the resultant scans assembled into the 2-
dimensional representation. Radioactivity is shown on a logarithmic
grey-scale with the more active areas being represented by the darker
shading. The scale runs between 9 and 1600 counts accumulated over
30 min. Lanes 6 and 7 show the positions of non-radioactive standards
visualized by quenching of the fluorescent indicator under UV
illumination or by staining as described in Section 4. Lanes 1 and 2
are metabolite IV before and after hydrolytic treatment respectively;
lanes 3 and 4 are metabolite III before and after hydrolytic treatment
respectively; lane 5 is a mixture of radioactive lipid standards (TAG,
triacylglycerol; 16:0, palmitic acid; 14:0 myristic acid; 12:0, lauric acid;
8:0, octanoic acid); lane 6 is a mixture of 2,4-D homologues with the
number in parentheses indicating the length of the carbon side-chain
such that 2,4-D (2) is 2,4-D itself; lane 7 is the methyl ester of 2,4-D.
Tick marks on the borders of the plate are 1 cm apart. The position of
the Or and SF are indicated.
et al., 1980). An amide-linked conjugate of 2,4-D with a
non-protein amino acid like ornithine or with a polyam-
ine such as putrescine, spermidine or spermine would

´
A. Lopez-Villalobos et al. / Phytochemistry 65 (2004) 2763–2774
2768
the 16 carbon and the 12 carbon elongated 2,4-D stand-
ards, the third migrating between these standards where
the 14 carbon elongated form would be expected (stand-
ard not available). A proportion (17%) remained on the
origin with natural TAG standards and very small
amounts migrated with the shorter forms, perhaps as a
result of incomplete separation of metabolites III and
IV on the original TLC plate. Metabolite III, after
hydrolysis, yielded a number of radioactive bands which
migrated with the shorter forms of elongated 2,4-D from
12 carbon to 4 carbon and 2,4-D itself. Again there is
evidence of unhydrolysed TAG on the origin and
incomplete separation as judged by the appearance of
some minor bands characteristic of metabolite IV. In
this reversed phase TLC system, 94.3% of non-hydro-
lysed metabolite III and 100% of non-hydrolysed metab-
olite IV remained on the origin. While the standard for
methyl-2,4-D also appears on the plate, results on nor-
mal-phase TLC show that it was not associated with
any radioactivity.
elongated forms of 2,4-D will exert any physiological ef-
fects in their own right or be b-oxidised back to active
2,4-D.
The findings here are of significance compared to the
previous reports and differ from those reports in three
respects: (i) the 2,4-D chains were elongated by as many
as 16 carbons, which is more than has been reported be-
fore; (ii) the elongated acids have been identified as con-
jugates analogous to natural TAG and (iii) they are
identified in a commercially exploited food-oil crop.
The formation of the TAG may suggest a quasi-physio-
logical role where the triacylglycerol acts as a sink for
2,4-D, when it is supplied in excessive (toxic or muta-
genic) quantities, and as a source of 2,4-D, which could
be supplied by the action of lipolytic and b-oxidation en-
zymes at times of lipid mobilisation when 2,4-D concen-
trations may be sub-optimal. The finding that TAG can
act as an end-product of xenobiotic metabolism suggests
that other xenobiotic compounds may also be metabo-
lised in this way and persist in the oil after the water-sol-
uble metabolites have been removed.
Absolute certainty in the identification of the compo-
nents of these metabolites could only come via mass
spectrographic (MS) analysis. This proved not to be fea-
sible because the techniques used for isolation of the
component acids, while very successful at separating
the radioactive (i.e. 2,4-D-derived) acids, were not capa-
ble of separating them from the non-radioactive natural
fatty acids, which were present in massive excess com-
pared to the 2,4-D metabolites. This makes analysis of
the resulting spectra almost impossible as had been
found on a previous occasion (Dodds et al., 1995).
The evidence presented is consistent with both metab-
olites III and IV being triacylglycerol (TAG) analogues,
with one of the three fatty acids being replaced with a
chain-elongated 2,4-D. At this stage we have no evi-
dence to enable us to distinguish the three possible sites
of esterification. The acids were liberated from the TAG
by hydrolysis and identified by a novel application of re-
versed-phase TLC. Similar but shorter acids have been
reported before in alfalfa (Linscott and Hagin, 1970)
fed 2,4-D and, in alfalfa (Linscott et al., 1968), soybean
and cocklebur (Wathana and Corbin, 1972) fed 2,4-
dichlorophenoxybutanoic acid; elongated forms, with
side chains in the range 4–10 carbons, have been identi-
fied and further (longer) acids were also observed but
not identified or quantified.
2.3. Effects of culture conditions on the proportions of 2,4-
D metabolites
Major differences between tissues in the metabolism
of 2,4-D were observed. Cultured inflorescence tissue
of either developmental stage produced metabolites of
only one of the four classes, the polar metabolite I,
whereas the plumular tissue produced measurable quan-
tities of all four metabolite classes. All tissues contained
some unchanged non-esterified 2,4-D. The results with
inflorescence tissue at stage ꢀ4 and ꢀ5, incubated under
standard conditions of 0.116 M sucrose and 0.15 mM
2,4-D, showed that the proportion of metabolite I de-
creased from 85.3 5.4% (mean SEM from five sepa-
rate explants) to 75.6 6.8% (stage ꢀ4) or from
91.6 1.8% to 80.9 2.6% (stage ꢀ5), the remainder
being unchanged 2,4-D, as the total 2,4-D taken up in-
creased between days 32 and 64 of incubation. In plu-
mular tissue (Fig. 6), cultured over a longer period
under standard conditions of 0.116 M sucrose and 0.4
mM 2,4-D, a steady increase in the total 2,4-D radioac-
tivity present in the tissue was observed. A progressive
decrease in the percentage (but not the total amount)
of metabolite I present (from about 55% of all radioac-
tivity at 30 days to about 37% at 120 days) was observed
as the total tissue uptake of 2,4-D increased. Metabo-
lites II–IV were each present at between 8% and 13%
up to 90 days of incubation but fell as a proportion
thereafter. Only the proportion of unchanged 2,4-D in-
creased throughout the incubation period. No attempt
was made to measure 2,4-D metabolites that may have
been released back into the medium. Since release was
likely to have been a low proportion of the approxi-
mately 3% of the total 2,4-D taken up from the medium,
One novel aspect of the findings is that two distinct
populations of elongated 2,4-D acids were identified.
Presumably, they arose by the action of fatty acid syn-
thase enzymes and the presence of two thioesterase
activities with different chain length specificities is sug-
gested. If this were the case, it would be consistent with
the normal functioning of the enzyme in coconut, which
produces both medium chain and long chain fatty acids.
TAGs may be subjected to lipolysis at a later stage of
development and it is not known whether the released

´
A. Lopez-Villalobos et al. / Phytochemistry 65 (2004) 2763–2774
2769
60
50
40
30
20
10
0
Our results are consistent with those published and
reviewed elsewhere inasmuch as the major metabolite
group in all our experiments were polar, probably sugar,
conjugates (Ba¨renwald et al., 1994; Edwards et al., 1982;
Schneider et al., 1984).
2.4. Relationships between metabolism and differentiation
of the tissue
In histological terms, these observations corre-
sponded to no observable de-differentiation of the plu-
mular tissue at 0.6 lM 2,4-D and to a high degree of
de-differentiation, as evidenced by a formation of meris-
tematic centres and by proliferation of parenchymal
cells in the callus, at 0.4 and 0.6 mM 2,4-D. The re-
sponse at 0.2 mM 2,4-D was intermediate between the
two extremes.
The capacity to metabolise 2,4-D by coconut explants
was associated with the de-differentiation of the tissue.
Higher absolute quantities and proportions of the novel
non-polar metabolites (TAG containing elongated
forms of 2,4-D) were contained in highly de-differenti-
ated tissue derived from plumules. Furthermore, inflo-
rescence explants failed both to produce callus and to
synthesise the TAG analogues. It is difficult to ascertain
whether or not this association represents a mechanistic
link between 2,4-D metabolism and its action on callus
proliferation; the possibility that the above metabolites
may just be products of callus metabolism, with no fur-
ther significance, cannot be excluded.
0
30
60
90
120
Time in culture (days)
Fig. 6. Effect of time in culture on the composition of metabolites of
2,4-D in coconut plumular tissue. Plumular tissue was obtained and
cultured in the presence of radioactive 2,4-D as described in Section 4.
Tissue was extracted in methanol and analysed by TLC also as
described in the text. The percentage of methanol-extracted radioac-
tivity appearing as 2,4-D (s), as metabolite I (j), metabolite II (d),
metabolite III (m) or metabolite IV (.) is shown. The data represent
radioactivity accumulated continuously from day 0 until the day of
analysis. The error bars represent the means standard error of the
mean from five separate explants. The standard errors for metabolites
II–IV cannot be shown because of the crowding of the points on the
graph but were all in the range 0.31–2.85%.
3. Concluding remarks
The findings presented herein demonstrate that 2,4-D
can be metabolised to novel chain-elongated forms
which can potentially be stored as TAG. The formation
of such conjugates responds to manipulations of the tis-
sue culture conditions and are associated with improved
performance of the cultures. The finding may provide
clues to the understanding of the variable efficacy of
methods used for embryogenesis. The precise physiolog-
ical and toxicological significance of these findings will
require further investigation.
it would represent a very low proportion of radioactivity
outside the explant.
Further experiments with plumular tissue, cultured in
the presence of increasing concentrations of sucrose (0–
0.233 M) or of 2,4-D (0.6 lM–0.6 mM) also showed that
metabolite I was the major metabolite present in cases of
low incorporation but decreased at higher sucrose con-
centrations as metabolites II–IV and unchanged 2,4-D
all increased and represented between 9% and 13% of to-
tal metabolites at 0.232 M sucrose. Similar findings were
obtained when the concentration of 2,4-D was in-
creased: the proportion of metabolite I fell as the other
metabolites increased.
The total absence of metabolites II–IV, in cases of
low uptake, suggests that the enzymes that catalyse their
synthesis were also being induced as the rate of uptake
increased. These may be enzymes that play a role in
the development of the tissue in culture.
4. Experimental
4.1. Materials
Activated charcoal (acid washed), amino acids, b-glu-
cosidase from almond, protease type XXI from Strep-
tomyces griseus, sucrose (from sugar cane, 99% pure),
Phytagel and lipid standards were obtained from Sigma
(Sigma-Aldrich Company Ltd., Poole, UK); 6-bromo-
hexanoic, 8-bromo-octanoic, 12-bromododecanoic and

´
A. Lopez-Villalobos et al. / Phytochemistry 65 (2004) 2763–2774
2770
16-bromohexadecanoic acids, 2,4-dichlorophenol and
thionyl chloride were from Aldrich (Sigma–Aldrich
Company Ltd., Poole, UK); 10-bromodecanoic acid
was from Pfaltz and Bauer (Waterbury, USA); 2,4-
dichlorophenoxyacetic acid (2,4-D) and 2,3,4,6-tetra-
The extract was further washed and dried and recrystal-
lised from petroleum spirit (40–60 ꢁC boiling fraction).
A solution of the product in petroleum spirit was ap-
plied to a 70 cm by 2.5 cm (diameter) column of silica
gel, which had been prepared in petroleum spirit, and
eluted with dichloromethane:petroleum spirit (40–60ꢁ
fraction):diethyl ether (90:10:5, by volume). Fractions
containing product were identified by TLC on ‘‘micro-
scope-slide’’ silica gel plates developed in dichlorometh-
ane:petroleum spirit (40–60ꢁ fraction):diethyl ether
(90:10:5, by volume) and pooled. Structures were con-
firmed by NMR spectroscopy.
O-acetyl-a-D-glucopyranoyl bromide were obtained
from Acros (Geel, Belgium); 2-(2,4-dichlorophenoxy)-
2-14C-acetic acid ([14C]2,4-D) was obtained from ARC
inc, St Louis, MO, USA; Ecolite scintillation cocktail
was from ICN (Costa Mesa, USA); solvents and other
chemicals were obtained from the normal chemical
suppliers.
Plates for TLC were obtained from the following sup-
pliers: silica gel 60 F254 250 lm thick 20 cm · 20 cm or
7.5 cm · 2.5 cm (‘‘microscope-slide’’ sized) plates, from
Merck (Darmstadt, Germany); silica gel 60 A, 250 lm
thick, 20 cm · 20 cm ruled into 19 channels, with a
pre-adsorbant layer, from Whatman International Ltd.
(Maidstone, UK); reversed phase C-18 bonded silica
gel 60–250 lm thick, 20 cm · 20 cm were also from
Whatman.
Attempts to synthesise a glucose ester of 2,4-D
proved problematic. Both tetra-acetylated and tetra-
benzoylated derivatives were synthesised but various at-
tempts to remove the blocking groups, for instance by
hydrogenolysis as described for indol-3-acetic acid (Ja-
kas et al., 1993), also resulted in the hydrolysis of the
glucose-2,4-D ester bond. In the end, a different ap-
proach was adopted in which the possible glucose ester
metabolite was acetylated, and compared with a syn-
thetic tetra-acetyl-glucose ester of 2,4-D.
4.2. Chemical synthesis of 2,4-D metabolites and stand-
ards for chromatography
The tetra-acetyl-glucose ester of 2,4-D was prepared
according to the method of Stock (1979). 2,4-D, dis-
solved in acetone containing 4.7% (v/v) triethylamine,
Amino acid conjugates of 2,4-D were prepared by
reaction of the amino acids with 2,4-dichlorophenoxy-
acetyl chloride (2,4-D chloride) in a ‘‘Schotten-Bau-
man’’-type reaction (Wood and Fontaine, 1952). 2,4-D
chloride, made from 2,4-D and thionyl chloride (Freed,
was reacted with 2,3,4,6-tetra-O-acetyl-a-D-glucopyra-
noyl bromide at room temperature for 20 h, when a pre-
cipitate of triethylammonium bromide had formed. The
addition of ethyl acetate promoted more precipitation of
the bromide which was removed by filtration. Unreacted
2,4-D acid was extracted from the filtrate with sodium
hydrogen carbonate solution and the ethyl acetate phase
was washed, dried and filtered. The tetra-acetyl-glucose
ester of 2,4-D was further purified by recrystallisation
from 50% (v/v) ethanol. The identity of the product
was confirmed by NMR spectroscopy as above.
The methyl ester of 2,4-D was synthesised by the
method of Appleby et al. (1974).
1946), was reacted with
L
-(ꢀ)-alanine, aspartic acid,
glutamic acid, glycine, histidine, leucine, phenylalanine,
tryptophan or valine at room temperature for 3 h. After
extracting unreacted acyl chloride and acidifying with
2 M HCl to pH 2.0, the amino acid conjugate precipi-
tated as a white solid which was collected, washed three
times with diethyl ether, dried, recrystallised twice from
50% (v/v) ethanol and dried again giving 55–67% of the
theoretical yield. The structure was confirmed by (pro-
ton) nuclear magnetic resonance spectroscopy using a
Varian T-60 NMR spectrometer (60 MHz). The NMR
analyses were performed by Merlin Synthesis, Kent,
UK.
4.3. Plant material
Immature inflorescences of coconut palm (C. nucifera
L.) were obtained from Coconut Industry Board (CIB),
Jamaica. Cores of endosperm containing mature zygotic
embryos were obtained from The Coconut Plantation
for Hybrid Production, Colegio de Postgraduados en
Elongated forms of 2,4-D were prepared by reacting
the potassium salt of 2,4-dichlorophenol with an x-bro-
mo-1-alkanoic acid by a method modified from that of
Linscott et al. (1968). The reactants, 10 mol of
2,4-dichlorophenol, 10 mmol of 6-bromohexanoic acid,
8-bromooctanoic acid, 10-bromodecanoic acid, 12-
bromododecanoic acid or 16-bromohexadecanoic acid,
and 25 mmol of K2CO3, were dissolved in 50 mL of
anhydrous dimethylformamide. The mixture was ref-
luxed for 4 h then cooled and extracted with 100 mL
of diethyl ether to remove unreacted precursors. The
remaining phase was adjusted to pH 2.0 with 2 M HCl
and the product extracted into 100 mL of diethyl ether.
´
´
Ciencias Agrıcolas de Mexico, Tabasco, Mexico. The
ecotype used exclusively throughout the studies was Ma-
layan Yellow Dwarf (MYD).
4.4. Culture media
Full strength Murashige and Skoog medium (Mur-
ashige and Skoog, 1962) was supplemented with 3 g of
Phytagel and 2.5 g of activated charcoal per litre. When
required, the medium was adjusted to contain up to

´
A. Lopez-Villalobos et al. / Phytochemistry 65 (2004) 2763–2774
2771
0.232 M sucrose and up to 0.6 mM 2,4-D. For incuba-
tions containing radioactive 2,4-D, the final mixture
contained 1000 Bq of [14C]2,4-D per mL for inflores-
cences or 2650 BqmLꢀ1 for plumular tissue. The result-
ant specific radioactivities were 6.64 MBqmmolꢀ1 for
inflorescence tissue and a range from 2035 to 1.75
MBqmmolꢀ1 corresponding to non-radioactive 2,4-D
concentrations from 0 to 0.6 mM respectively for plu-
mular tissue. The media were autoclaved at 103.5 kPa
and 120 ꢁC for 20 min and used the following day. It
has previously been shown that autoclaving under these
conditions does not affect the integrity of 2,4-D (Ebert
and Taylor, 1990). The culture medium of cultures of
plumular tissue was removed after every 30 days and re-
placed with medium having a composition and a radio-
active content identical to the original medium. Half of
the cultures of inflorescence tissue were established in
the presence of radioactive 2,4-D and half in its absence.
After 32 days, the radioactive cultures were taken for
analysis and the remainder moved to a medium having
identical composition but now supplemented with radi-
oactive 2,4-D and maintained for a further 30 days.
Consequently, total uptake of [14C]2,4-D at time 62 days
reflects only the uptake between days 33 and 62 in inflo-
rescence tissue. In plumular tissue, in contrast, the re-
sults reflect continuous accumulation from day 0 to
the day of sampling.
face. The tissue was washed three times with distilled
water, to eliminate traces of 2,4-D, and dried with tissue
paper. They were transferred into 5 mL plastic cryogenic
vials in an ice-bath and cut into smaller pieces. Immedi-
ately 2 mL of ice-cold methanol was added to each vial,
which was then stored at ꢀ20 ꢁC for at least 48 h. After
this period, the frozen tissues were crushed with a glass
rod and a clear methanol supernatant was obtained for
further analysis after sedimentation of the tissue at 3000
rpm in a bench centrifuge. The residues, after a further
three washes in methanol, were transferred to counting
vials and the radioactivity associated with the tissue deb-
ris determined by liquid scintillation counting. The total
volume of methanol extract (including rinsings) was ad-
justed to 3.0 mL. Aliquots (200 lL) were taken for liquid
scintillation counting to determine the content of 2,4-D in
tissue.
Aliquots (500 lL) of the total methanol extract were
loaded onto individual lanes of 20 cm · 20 cm 19-chan-
nel TLC plates. The aliquots were applied to alternate
lanes with standards in the intervening lanes. The plates
were developed in toluene:2-butanone:acetic acid
(45:55:3, by volume). Appropriate authentic 2,4-D
metabolites and natural lipids were used as qualitative
standards. Plates were visualized by fluorescent quench-
ing under UV light (254 nm) and scanned for radioactiv-
ity. Alternatively, the plates were stained to visualise free
acids by spraying sequentially with 0.1% (w/v) 20,70-
dichlorofluoroscein in 95% (v/v) methanol, 1%(w/v)
AlCl3 in ethanol then 1% (w/v) aqueous FeCl3, warming
at 45 ꢁC between sprayings (Fried, 1989).
4.5. Tissue culture
Spathes sheathing immature inflorescences at both
stage ꢀ4 and stage ꢀ5 (Hornung, 1995b) were wiped
with 70% (v/v) ethanol before excision of the inflores-
cence materials. The inflorescences were cut into seg-
ments 0.5 cm in length prior to incubation for 72 h in
Individual radioactive areas, identified only as metab-
olites I–IV were scraped from the plates and pooled
from different samples for further analysis. Polar metab-
olites (I and II) were extracted from the silica gel in eth-
anol and non-polar metabolites (III and IV) were
extracted in diethyl ether.
´
liquid medium (Lopez-Villalobos et al., 2001). Subse-
quently, inflorescence segments of 0.5–1.0 mm length
were cut and transferred to gelled medium. Throughout
the incubation period, tissues were maintained in the
dark at 29 1 ꢁC.
4.6.1. Acetylation
It proved impossible to prepare a glucose ester of
2,4-D from its blocked chemical precursor, the
2,3,4,6-tetra-O-acetyl-D-glucopyranose ester of 2,4-D.
Mature zygotic embryos were excised from endo-
´
sperm cores and incubated as described by Lopez-Vill-
It was therefore decided to acetylate metabolite I, to
see whether any of the products shared properties
with the authentic tetra-acetyl-glucose ester. Pooled
metabolite I extracted from 10 samples was dissolved
in 1 mL of dry pyridine and reacted with 1 mL of
acetic anhydride at room temperature for 12 h. Then
2 mL of water at 4 ꢁC was added to stop the reaction
and acetylated products were extracted with 3.5 mL of
ethyl acetate. The extract was washed, dried and con-
centrated (Morillo et al., 2001). Concentrated samples
were applied to silica gel TLC plates and developed in
toluene:methanol:acetic acid (96:8:4, by volume).
Radioactive products and authentic standards were lo-
cated as described above.
alobos et al. (2001). After 10 days in culture, plumules
were excised following the procedure of Hornung
(1995a). The plumular tissue was transferred to gelled
medium and cultured as described for immature
inflorescences.
4.6. Extraction and analysis
The extraction of 2,4-D and its metabolites was carried
out using a method similar to that described by Schneider
et al. (1984). At the end of the culture period, plant tissues
were removed from the culture vessels and weighed after
carefully removing the gel medium adhering to their sur-

´
A. Lopez-Villalobos et al. / Phytochemistry 65 (2004) 2763–2774
2772
4.6.2. Glucosidase
was concentrated in a stream of dry nitrogen to approx-
imately 1 mL which was applied to a reversed-phase
TLC plate, developed in acetonitrile:methanol:water
(80:10:30, by volume) and subjected to radioscanning.
The synthetic elongated 2,4-D standards and a variety
of natural lipid standards were also applied to the plate.
Metabolite I, obtained from 30 samples as above, was
divided into three tubes. They were subjected to diges-
tion by b-glucosidase as described by Scheel and Sander-
mann (1981). A solution of 0.1 mg b-glucosidase mLꢀ1
was made up in 100 mM sodium acetate buffer, pH
5.0, solution. To the three tubes, 1 mL of enzyme solu-
tion, 1 mL of buffer solution (control for non-enzymic
hydrolysis) or 1 mL of distilled water (control for
hydrolysis at acid pH) was added. The tubes were incu-
bated at 37 ꢁC. After 60 min, the reaction was stopped
by the addition of 20 lL of acetic acid followed by 4
mL of diethyl ether to each tube. The radioactive prod-
ucts from the ether or aqueous phase were applied to a
19-channel silica gel TLC plate, developed in tolu-
ene:methanol:acetic acid (96:8:4, by volume) then visual-
ised by radioscanning.
4.7. Determination of radioactivity
The radioactive content of samples was determined by
liquid scintillation counting in 22 mL plastic scintillation
vials (Meridian, Surrey, UK) containing 7 mL of EcoLite
scintillation fluid. Each vial could contain up to 1 mL of
methanolic or aqueous sample and was counted for 10
min. Homogeneous samples were counted using an exter-
nal standard ratio method to correct for quench and
non-homogeneous samples used a samples channel ratio
method. Colourless samples were corrected for chemical
quench by reference to quench-curves prepared using
quenched standards (Amersham Life Science, Bucks,
UK); coloured samples were corrected by reference to
quench-curves of coloured standards prepared in-house.
The scintillation counter was a LKB-Wallac Rackbeta
1211 (Perkin–Elmer Life Sciences, Cambridge, UK) and
counts-per-minute measurements were corrected for
quench using the software provided.
TLC plates were scanned for radioactivity using a Bio-
scan System 200 Imaging Scanner (Washington, D.C.,
USA). The percentage of total radioactivity detected
associated with individual bands was determined using
the software Win-Scan, version 1.6, (Lab Logic, Sheffield,
UK) provided. The data from a whole plate were viewed
as a single (‘‘2-dimensional’’) representation using Win-
Scan 2D, version.1.6, (Lab Logic, Sheffield, UK).
4.6.3. Protease
Metabolites I or II was obtained from TLC plates as
described for the glucosidase incubations and incubated
with a protease of low specificity in an attempt to hydro-
lyse peptide-like bonds between 2,4-D and the amino
group of amino acids. Pooled metabolites from 30 sam-
ples were again used for three incubations per metabo-
lite. The enzyme was dissolved to give 20 lg of
enzyme mLꢀ1 in 10 mM sodium acetate, 5 mM calcium
acetate solution adjusted to pH 7.5 with acetic acid. The
recovered metabolite was mixed with 2 mL of this en-
zyme solution or of buffer alone or of distilled water
and incubated at 37 ꢁC for 60 min. The tubes were ex-
tracted as described for glucosidase incubations and
subjected to TLC followed by radioscanning.
4.6.4. Chemical hydrolysis
Alkaline hydrolysis of metabolite II (obtained as
above) and of metabolites III and IV was carried out
so that the form of the radioactive acid could be identi-
fied. Metabolites III and IV were scraped from 14 lanes
of TLC plates and extracted with diethyl ether; metabo-
lite II was extracted with ethanol from silica scrapings
from 10 lanes. The solvent was removed in a stream of
dry nitrogen gas. Hydrolysis was carried out, as de-
scribed by Christie (1989) by adding 2 mL of 1 M
KOH in 95% (w/v) ethanol and refluxing for 60 min.
After cooling, 5 mL of water and 5 mL of hexane:diethyl
ether (1:1, v/v) were added, the tube contents mixed and
the phases separated in a bench centrifuge for 7 min at
3500 rpm. Liberated acidic compounds were extracted
as their potassium salts in the aqueous phase, which
was combined with two further 2.5 mL aqueous washes
of the organic phase. The resulting combined aqueous
phase was acidified with 0.4 mL of 6 M HCl and the,
now protonated, acids extracted three times with 4 mL
of hexane:diethyl ether (1:1, v/v). The combined extract
Acknowledgements
We acknowledge Mr. T. Gilkerson of Krown Synthe-
sis, Imperial College, London, Wye campus, for help
with the chemical syntheses and Dr. R. Nash of Merlin
Synthesis, Imperial College, London, Wye campus, for
help with the NMR spectroscopy. Dr. J.L. Garraway
helped with the IR spectroscopy and Mr. C. Ladley pro-
vided technical assistance with radiochemical methods.
Financial support for A.L.-V. was received from
CONACYT and Colegio de Postgraduados en Ciencias
´ ´
Agrıcolas de Mexico.
References
Appleby, R.S., Gurr, M.I., Nichols, B., 1974. Factors controlling the
biosynthesis of fatty acids and acyl lipids in subcellular organelles
of maturing Crambe abyssinica seeds. Eur. J. Biochem. 48, 209–216.

´
A. Lopez-Villalobos et al. / Phytochemistry 65 (2004) 2763–2774
2773
Ba¨renwald, G., Schneider, B., Schutte, H.R., 1994. Metabolism of bis-
¨
Hornung, R., Verdeil, J.-L., 1999. Somatic embryogenesis in
coconut from immature inflorescence explants. In: Oropeza,
C., Verdeil, J.L., Ashburner, G.R., Carden˜a, R., Santamaria,
J.M. (Eds.), Current Advances in Coconut Biotechnology.
Kluwer Academic Publishers, Dordrecht, The Netherlands, pp.
297–309.
ethylene glycol esters of 2,4-D and MCPA in excised plants and cell
suspension cultures of tomato. J. Plant Physiol. 144, 396–399.
Blake, J., 1995. A brief history of coconut tissue culture. In: Oropeza,
V., Howard, F.W., Ashburner, G.R. (Eds.), Lethal Yellowing:
Research and Practical Aspects, vol. 5. Kluwer Academic Publish-
ers, Dordrecht, The Netherlands, pp. 217–223.
Jakas, A., Magnus, V., Horvat, S., Sanderg, G., 1993. Synthesis of the
Blake, J., Hornung, R., 1995. Somatic embryogenesis in coconut
(Cocos nucifera L). In: Mohan, S.J., Gupta, P.K., Newton, R.J.
(Eds.), Somatic Embryogenesis in Woody Plants, vol. 2. Kluwer
Academic Publishers, Dordrecht, The Netherlands, pp. 327–340.
Buffard-Morel, J., Verdeil, J.L., Dusset, S., Magnaval, C., Huet, C.,
Grosdemange, F., 1995. Initiation of somatic embryogenesis in
coconut (Cocos nucifera L.). In: Oropeza, C., Howard, F.W.,
Ashburner, G.R. (Eds.), Lethal Yellowing: Research and Practical
Aspects, vol. 5. Kluwer Academic Publishers, Netherlands, pp.
217–223.
b-D
-glucosylester of [carbonyl-13C]-indole-3-acetic acid. J. Labelled
Compd. Radiopharm. 33, 933–939.
Koens, K.B., Nicoloso, F.T., Van Vliet, T.B., Harteveld, M., Boot,
C.J.M., Van Iren, F., Mulder, P., Libbenga, K.R., Kijne, J.W.,
1995. Kinetics of 2,4-dichlorophenoxyacetic acid content in auxin-
dependent suspension culture of Nicotiana tabacum cells. J. Plant
Physiol. 147, 383–390.
Linscott, D.L., Hagin, R.D., 1970. Additions to the aliphatic moiety of
chlorophenoxy compounds. Weed Sci. 18, 197–198.
Linscott, D.L., Hagin, R.D., Dawson, J.E., 1968. Conversion of 4-
(2,4-dichlorophenoxy) butyric acid to homologs by alfalfa.
Mechanism of resistance to this herbicide. J. Agric. Food Chem.
16, 844–848.
Chan, J.L., Saenz, L., Talavera, C., Hornung, R., Robert, M.,
Oropeza, C., 1998. Regeneration of coconut (Cocos nucifera L.)
from plumular explants through somatic embryogenesis. Plant Cell
Rep. 17, 515–521.
´
Lopez-Villalobos, A., Dodds, P.F., Hornung, R., 2001. Changes in
Chkanikov, D.I., Makeev, A.M., Pavlova, N.N., Grygoryeva, L.V.,
Dubovoi, V.P., Klimov, O.V., 1976. Variety of 2,4-D metabolic
pathways in plants; its significance in developing analytical
methods for herbicides residues. Arch. Environ. Contam. Toxicol.
5, 97–103.
fatty acid composition during development of tissues of coconut
(Cocos nucifera L.) embryos in the intact nut and in vitro. J. Exp.
Bot. 52, 1–10.
LoSchiavo, F., Pitto, L., Guliano, G., 1989. DNA methylation of
embryogenic carrot cell cultures and its variation as caused by
mutation, differentiation, hormones and hypomethylating drugs.
Theor. Appl. Genetics 77, 325–331.
Chkanikov, D.I., Makeev, A.M., Pavlova, N.N., Nazarova, T.A.,
1982. Metabolism of 2,4-D in cultivated cereals. Sov. Plant Physiol.
29, 425–431.
Mikami, N., Wakabayashi, N., Yamada, H., Miyamoto, J., 1984. New
conjugated metabolites of 3-phenoxybenzoic acid in plants. Pestic.
Sci. 15, 531–542.
Christie, W.W., 1989. Gas Chromatography and Lipids. The Oily
Press, Ayr, Scoland.
Delbarre, A., Muller, P., Imhoff, V., Guern, J., 1996. Comparison of
mechanisms controlling uptake and accumulation of 2,4-dichloro-
phenoxyacetic acid, naphthalene-1-acetic acid, and indole-3-acetic
acid in suspension-cultured tobacco cells. Planta 198, 532–541.
Dodds, P.F., Chou, S.-C., Ranasinge, A., Coleman, R.A., 1995.
Metabolism of fenbufen by cultured 3T3-L1 adipocytes: synthesis
and metabolism of xenobiotic glycerolipids. J. Lipid Res. 36, 2493–
2503.
Mikami, N., Wakabayashi, N., Yamada, H., Miyamoto, J., 1985. The
metabolism of fenvalerate in plants – the conjugation of the acid
moiety. Pestic. Sci. 16, 46–58.
Minocha, S.C., Nissen, P., 1985. Uptake of 2,4-dichlorophenoxyacetic
acid and indolacetic acid in tuber slices of Jerusalem artichoke and
potato. J. Plant Physiol. 120, 351–362.
Morillo, M., Lequart, V., Grand, E., Goethals, G., Usubillaga, A.,
Villa, P., Martin, P., 2001. Synthesis of peracetylated chacotriose.
Carbohydr. Res. 334, 281–287.
Du Plessis, S., Stirk, W.A., Cress, W.A., Van Staden, J., 1996.
Biochemical comparisons of habituated and non-habituated callus
lines of Glycine max (L.) cv. Acme. Plant Growth Regul. 18, 223–
231.
Murashige, T., Skoog, F., 1962. A revised medium for rapid growth
and bioassays with tobacco tissue cultures. Physiol. Plant 15, 473–
497.
Ebert, A., Taylor, H.F., 1990. Assessment of changes of 2,4-dichlo-
rophenoxyacetic acid concentrations in plant tissue culture media
in presence of activated charcoal. Plant Cell Tiss. Organ Cult. 20,
165–172.
Nazarova, T.A., Pavlova, N.N., Makeev, A.M., Chkanikov, D.I.,
1980. Amino acid conjugates of 2,4-D in cereals. Sov. Plant
Physiol. 27, 557–561.
Roberts, T.R., Hutson, D.H., Lee, P.W., Nicholls, P.H., Plimmer,
J.R., 1998. Hebicides and Plant Growth Regulators, vol. 1. Royal
Society of Chemistry, Cambridge..
Edwards, V.T., McMinn, A.L., Wright, A.N., 1982. Sugar conjugates
of pesticides and their metabolites in plants – current status. In:
Hutson, D.H., Roberts, T.R. (Eds.), Progress in Pesticide Bio-
chemistry, vol. 2. John Wiley and Sons, Norwich, pp. 71–125.
Feung, C.S., Hamilton, R.H., Mumma, R., 1973. Metabolism of 2,4-
dichlorophenoxyacetic acid. IV. Mass spectra and chromato-
graphic properties of amino acid conjugates. J. Agric. Food Chem.
21, 632–637.
Rubery, P.H., 1987. Auxin transport. In: Davies, P.J. (Ed.), Plant
Hormones and Their Role in Plant Growth and Development.
Martinus Nijhoff Publishers, Boston, pp. 341–362.
Sabater, M., Rubery, P.H., 1987. Auxin carriers in Cucurbita vesicles.
Planta 171, 507–513.
Scheel, D., Sandermann, H., 1981. Metabolism of 2,4-dichlorophen-
oxyacetic acid in cell suspension cultures of soybean (Glycine max
L.) and wheat (Triticum aestivum L.). I. General results. Planta 152,
248–252.
Freed, V.G., 1946. Preparations and reactions of 2,4-dichlorophen-
oxyacetic acid. J. Am. Chem. Soc. 68, 2112.
Fried, B., 1989. Lipids. In: Sherma, J., Fried, B. (Eds.), Handbook of
Thin-Layer Chromatography. Marcel Dekker Inc., New York,
USA, pp. 583–623.
Schneider, B., Schutte, H.R., Tewes, A., 1984. Comparative investi-
¨
gations on the metabolism of 2-(2,4-dichlorophenoxy)isobutyric
acid in plants and cell suspension cultures of Lycopersicon
esculentum. Plant Physiol. 76, 989–992.
Hornung, R., 1995a. Micropropagation of Cocos nucifera (L.) from
plumular tissue excised from mature zygotic embryos. Plantations
´
Recherche Developpement 2, 38–41.
Shvetson, S.G., Gamburg, K.Z., 1984. Kinetic characteristics of 2,4-
D conjugation in corn cell culture. Sov. Plant Physiol. 31, 905–
911.
Hornung, R., 1995b. Initiation of callogenesis in coconut palm (Cocos
nucifera L.). In: Oropeza, C., Howard, F.W., Ashburner, G.R.
(Eds.), Lethal Yellowing: Research and Practical Aspects, vol. 5.
Kluwer Academic Publishers, Netherlands, pp. 203–215.
Stock, M., 1979. Vergleichende Untersuchungen zur Biotransforma-
tion von 2,2-diphenylssigsa¨ure, Nitrofen and Fluorodifen in

´
A. Lopez-Villalobos et al. / Phytochemistry 65 (2004) 2763–2774
2774
Ganzpflanzen und Zellsuspensionskulturen von Agrostema githago
Wathana, S., Corbin, F.T., 1972. Metabolism of 4-(2,4-dichlorophen-
oxy) butyric acid in soybean and cocklebur. J. Agric. Food Chem.
20, 23–26.
und Datura innoxia. Martin-Luther-Universita¨t, Halle.
Verdeil, J.L., Huet, C., Grosdemange, F., Buffard-Morel, J., 1994.
Plant regeneration from cultured immature inflorescences of
coconut (Cocos nucifera L.): evidence for somatic embryogenesis.
Plant Cell Rep. 13, 218–221.
Wood, J.W., Fontaine, T.D., 1952. Synthetic plant-growth regulators.
III. 2,4-Dichlorophenoxyacetyl derivatives of amino acids. J. Org.
Chem. 17, 891–896.