Journal of Experimental Botany, Vol. 52, No. 358, pp. 933±942, May 2001
Changes in fatty acid composition during development
of tissues of coconut (Cocos nucifera L.) embryos
in the intact nut and in vitro
Arturo LoÂpez-Villalobos1, Peter F. Dodds2 and Roland Hornung1,3
1
2
The T.H. Huxley School, Imperial College at Wye, University of London, Wye, Ashford, Kent TN25 5AH, UK
Department of Biology, Imperial College at Wye, University of London, Wye, Ashford, Kent TN25 5AH, UK
Received 19 September 2000; Accepted 28 November 2000
Abstract
Introduction
Intact coconuts were germinated in situ and compared with excised zygotic embryos germinated
in vitro. The growth of the embryonic tissue and their
fatty acid compositions were measured. Haustoria,
plumules and radicles of coconuts germinated in situ
grew continuously and proportionately throughout
the 120 d experiment with haustauria increasing to
45 g nut 1 and weighing 4±5-fold more than the other
two tissues. The plumules and radicles of the seedlings cultured in vitro also grew continuously but
the haustoria grew sporadically between 15 d and
75 d in culture and, at 250 mg nut 1 after 75 d, were
smaller than the other two tissues. All the tissues of
the nuts grown in situ contained significant amounts
of lauric acid, the acid characteristic of coconut oil,
as well as longer chain saturated and unsaturated
fatty acids. The content of medium and long chain
fatty acids increased in all growing tissues as the
experiment proceeded, especially the haustorium
which contained 24±35% of its fatty acid as lauric
acid; the fat content of solid endosperm reduced
during this period. Seedlings grown in vitro, on the
other hand, failed to accumulate lauric acid in any
of their tissues (haustorium contained 6±11% of its
fatty acid as lauric acid). The results may have
implications for the design of growth media for
growing zygotic and somatic cultures of coconut
and may provide a marker for successful germination.
The coconut palm (Cocos nucifera L.) plays an important
role as a subsistence crop and commercial oil crop in,
largely coastal, tropical regions (Persley, 1992). The
biology of this palm dictates that propagation can only
be achieved sexually, via the nut, and not by vegetative
means (Blake, 1995). Coconut is mainly a heterozygous
outbreeder with the result that the performance of the
progeny is dif®cult to predict. The controlled production
of hybrid nuts is labour-intensive and a long-term commitment. Ageing of existing plantations and the threat of
diseases such as lethal yellowing (Eden-Green, 1995)
makes it a priority to develop resistant, high-yielding and
locally-adapted hybrid material. Traditional propagation
techniques are a bottleneck in meeting the farmers'
demands. One promising approach would be the clonal
propagation of eÂlite palms selected in the ®eld.
Somatic embryogenesis in coconut was ®rst reported
in 1983 from immature in¯orescence tissue (Branton and
Blake, 1983). Subsequent efforts by a variety of teams
have considerably improved the ability to produce clonal
coconut palms but cannot yet be considered routine or
even reliable (Hornung and Verdeil, 1999). Further
advances have resulted from the use of plumular tissue
excised from mature zygotic embryos (Chan et al., 1998;
Hornung, 1995a). It is not yet possible, however, reliably
to propagate clonal coconut palms.
During germination, the haustorium of the coconut
embryo in the nut enlarges and grows towards the cavity
of the nut. This organ digests and absorbs nutrients which
it supplies to the growing shoot and root (Fig. 1B).
Principal among the nutrients are the triacylglycerols of
the oil which possess a distinctive fatty acid composition
Key words: Coconut germination, haustorium, tissue culture,
fatty acid composition, lauric acid.
3
To whom correspondence should be addressed. Fax: q44 20 7594 2919. E-mail: [email protected]
ß Society for Experimental Biology 2001
934
LoÂpez-Villalobos et al.
Fig. 1. (A) Developing seedling cultured in vitro after 75 d showing shoot (s), root (r) and undeveloped haustorium (h); ( 3 2.5). (B) Developing
seedling growing in situ after 60 d showing shoot (s), root (r) and developing haustorium (h); (1u3 of natural size). (C) Developing seedling in situ
showing emerging shoot (s), root (r) and haustorium (h) 30 d after germination; ( 3 3). (D) Developing seedling in vitro showing shoot (s), root (r) and
haustorium (h) 30 d after germination; ( 3 3).
Fatty acids in germinating coconut 935
Fig. 1. Continued.
936
LoÂpez-Villalobos et al.
characterized by an abundance of dodecanoic (lauric)
acid (12 : 0) (Rossell et al., 1985). Coconut seedlings
grown in vitro frequently have underdeveloped haustoria
which lack the stored lipids observed in situ. The results of
investigations into the changing fatty acid composition
of developing seedlings are reported and it is shown that
the content of lauric acid differs signi®cantly between
seedlings grown in vitro and in situ.
Materials and methods
Plant material
Mature dehusked Malayan Yellow Dwarf coconuts were
obtained from the Coconut Industry Board (CIB), Jamaica.
The nuts were potted in gravel and germinated in the glasshouse
at 25 8C; this is referred to, hereafter, as germination in situ.
To prevent cracking of the nut shell, they were covered with
black capillary matting (Monro South, Aylesford, Kent, UK).
Mature zygotic embryos embedded in cores of endosperm
were also obtained from CIB, Jamaica. Embryos were excised
from the cores, sterilized and incubated in ¯at-bottomed soda
glass tubes (25 3 75 mm), containing 10 ml of liquid medium and
each covered with a polypropylene sheet (pore size 0.25 mm;
Cannings Packaging Ltd., Bristol, UK); this is referred to,
hereafter, as germination in vitro. Cultures were maintained at
27 8C in the dark and transferred at 4-weekly intervals.
Sterilization procedures
Prior to dispatch from Jamaica, the endosperm cores containing
the mature embryos were surface-sterilized using 1.85% (wuv)
sodium hypochlorite for 30 min. After excision, the embryos
were subjected to further sterilizations in sodium hypochlorite
solution. The ®rst was carried out using a 5% (wuv) solution for
10 min and the second in 1% (wuv) for 5 min. These treatments
were followed by immersion for 1 min in 70% (vuv) ethanol.
Each sterilization step was followed by three rinses in sterile
distilled water.
Preparation of medium
Liquid medium was prepared using powdered (Murashige and
Skoog, 1962) medium (Imperial Laboratories Ltd., Andover,
UK), supplemented with 0.116 M sucrose, adjusted to pH 5.7
with NaOH; 2.5 g of activated charcoal (acid-washed, Sigma,
Poole, UK) was added per litre of medium. Medium was
autoclaved at 103.5 kPa for 21 min.
Growth measurements
Samples were taken from coconut tissues grown both in situ and
in vitro at intervals of 30 d and 15 d, respectively, throughout
the germination process. At each sampling, embryos (or seedlings) from three nuts and from three culture vessels were
selected at random and dissected into plumule, cotyledonary
tissue (proximal end of the embryo that will produce the prostem; Haccius and Philip, 1979), haustorium and radicle. The
fresh weights of intact tissue and each of the organs were
recorded. In addition, the liquid and solid endosperms from the
sampled nuts were collected for analysis.
Analysis of tissue lipids
The plant tissues were homogenized in propan-2-ol using
a glassuglass homogenizer then incubated at 80 8C to inactivate
lipases. Lipid extraction was continued according to the method
described previously (Bligh and Dyer, 1959). The total weight of
lipid extract was recorded and de®ned amounts of tridecanoic
acid were added to the different tissues as internal standards.
Fatty acid butyl esters were prepared according to the method
of Iverson and Sheppard (Iverson and Sheppard, 1977) and
analysed by capillary gas chromatography using a HewlettPackard 5840A gas chromatograph (Hewlett-Packard, Palo
Alto, USA) ®tted with a 30 m 3 0.25 mm silica column coated
with SP2380 (Supelco, Poole, UK) to a ®lm-thickness of 0.2 mm.
The butyl esters were detected with a ¯ame-ionization detector
and the peaks integrated with an electronic integrator (SpectraPhysics, San Jose, USA). The injector and detector temperatures
were maintained at 250 8C, and the nitrogen carrier gas was
maintained at a pressure of 40 kPa. After injection of 1 ml of
butyl ester solution, the column temperature was held at 80 8C
for 1 min, then increased at 4 8C min 1 to 25 0 8C, which was
held for 15 min.
Statistical analysis
A completely randomized design was used throughout the
experiments. For germination in vitro, variables were measured
after 0, 15, 30, 45, 60, and 75 d in culture whereas, for germination in situ, samples were taken after 0, 30, 60, 90, and
120 d. At each sampling, three nuts in situ or three cultures
in vitro were chosen at random and analysed independently
of each other. Analysis of variance was carried out using the
General Linear Model Procedure (SAS, Cary, USA) with the
ages of the seedlings being considered as treatments. Means
were compared using the Tukey test (Steel and Torrie, 1960).
Results
Sixty-nine nuts were planted in the greenhouse and 49
germinated successfully to form part of this experiment.
Similarly, excised zygotic embryos were incubated in vitro
and a germination rate of 77% was recorded. At intervals
during their growth, samples from the growing nuts and
cultured seedlings were taken and weights of the dissected
organs were measured (Fig. 1A, B, C, D). The results are
shown in Fig. 2A for nuts grown in situ and in Fig. 2B for
seedlings in vitro. It is clear from the results that growth
in situ exceeds that in vitro by a considerable margin.
In situ, balanced growth of the plumule and the root was
observed but the major proportion of the growth was
accounted for by the developing haustorium (Fig. 1B, C).
In contrast, in seedlings grown in vitro (Fig. 1A, D),
growth of the cotyledonary tissue, shoot and root was
observed with growth in the root appearing later and
proceeding more slowly than in the shoot. The haustorium in vitro grew very slowly after the ®rst 15 d in culture.
The fatty acid composition of the total lipid extracts
of all the excised tissues was determined by gas liquid
chromatography after conversion to butyl esters. Butyl
esters were used in preference to methyl esters so that
accurate determination of the more volatile short-chain
Fatty acids in germinating coconut 937
Fig. 2. The fresh weight of tissues of growing coconuts grown (A) in situ
or (B) in vitro at the indicated sampling dates. The symbols indicate
plumule (m), radicle (m), cotyledonary tissue (j) or haustorium (k).
Each value represents the mean"the standard error of the mean from
three individual coconuts.
acids could be made. The detailed results are presented
in Table 1A±D for tissues grown in situ and in
Table 2A±C for tissues grown in vitro; for greater clarity,
the results have been summarized by assigning the fatty
acids to four groups called `shorter-chain fatty acids',
`lauric acid', `longer-chain saturated acids', and `unsaturated fatty acids'. These are presented as graphs in
Figs 3±5, and de®ned in the legend to Fig. 3.
The composition of tissues dissected from nuts grown
in situ is shown in Fig. 3A, B, C. In all cases, growth of
the tissue was accompanied by an increase in the total
content of long-chain acids (both saturated and unsaturated) but the deposition of lauric acid was also apparent.
This was particularly evident in the haustorial tissue
at 120 d where lauric acid represented about 32% of the
total fatty acid content. Shorter-chain acids were relatively minor components. Samples of solid endosperm
were also analysed and the results shown as Fig. 4. It can
be seen that there was a steady decline in the mass of all
the fatty acids during the ®rst 90 d of the growth period
with the drop being particularly pronounced between
60 d and 90 d. No further fall was observed up to 120 d.
Throughout this period there was no major change in the
fatty acid composition expressed as a percentage of all
fatty acids. The major components were lauric acid
(45.3"0.38%, mean"standard error of the mean from
15 determinations) and the longer chain saturated acids
(30.3"0.36% (15)). Analysis of liquid endosperm showed
that its fatty acid content per nut was only approximately
0.15% of that found in the solid endosperm. In addition,
the results showed much more variability due, to some
extent, to the variable recoveries of liquid endosperm
from growing nuts at different stages of development.
Nevertheless, a large rise in the content of all the fatty
acids was observed between 0 d and 30 d of growth
which was maintained until 90 d before falling again.
Once again, the percentage fatty acid remained almost
constant with lauric acid representing 39.6"1.03% (15)
and long-chain saturated acids 34.1"1.46% (15).
The total content of long chain acids of roots and
plumules from seedlings in vitro also increased as the
tissue grew (Figs 2B, 5A, B). It may be noted that the
fatty acid contents of tissues in vitro, expressed as mg
of fatty acid per whole tissue is much lower than in those
in situ; this is accounted for by the greater growth rates
and tissue weights in situ. Expressed as mg of fatty acid
per gram of tissue, however, the difference is that tissues
grown in situ contain approximately 4-fold more fatty
acid. Haustorium did not grow consistently and the total
fatty acid content appeared also to fall slightly (Fig. 5C).
Once again the major fatty acids were the long-chain
acids. In all these cases, and in contrast to the nuts grown
in situ, lauric acid did not accumulate and usually
remained below 7% of the total fatty acid content.
Discussion
The coconut, which is botanically a drupe, is a very large
seed. It stores its energy reserves in the endosperm mainly
as lipids, which need to be digested and transported to the
haustorium of the developing seedling and thence to the
growing shoot and root. If not unique, the coconut is one
of very few seeds that store a major portion of its energy
reserves as the medium chain-length fatty acid, lauric acid
(12 : 0), oil palm being the other commercially important
seed that does the same.
The major ®ndings of the investigations were that
embryos germinated in situ followed expected ontogenic
patterns by growing haustoria and developing shoots and
roots (Fig. 1B, C). They contained signi®cant proportions
of 12 : 0 in all growing parts over the period examined
in the experiment. In contrast, in embryos germinated
in vitro, the haustorium failed to develop and 12 : 0 was
a minor component of all the tissues.
The results raise the question of whether the appearance of 12 : 0 in developing tissue can be used as a marker
938
LoÂpez-Villalobos et al.
Table 1. Fatty acid composition of tissues of coconut embryos germinated in situ
Results are expressed as mg of fatty acid per nut and shown as mean"SEM from determinations carried out on three independent embryos.
Differences between columns (i.e. for a single fatty acid at different stages of development) were not signi®cant (P)0.05), according to ANOVAuTukey
test, when the entries bear the same letter.
Fatty acid
Days in culture
0
(A) Plumules
6:0
8:0
10 : 0
12 : 0
14 : 0
16 : 0
18 : 0
18 : 1
18 : 2
18 : 3
0.00"0.00
0.00"0.00
0.00"0.00
0.01"0.00
0.01"0.01
0.03"0.00
0.02"0.00
0.01"0.00
0.01"0.00
0.00"0.00
30
a
a
b
b
a
b
a
b
c
c
(B) Radicles
6:0
8:0
10 : 0
12 : 0
14 : 0
16 : 0
18 : 0
18 : 1
18 : 2
18 : 3
(C) Haustoria
6:0
8:0
10 : 0
12 : 0
14 : 0
16 : 0
18 : 0
18 : 1
18 : 2
18 : 3
(D) Solid endosperm
6:0
8:0
10 : 0
12 : 0
14 : 0
16 : 0
18 : 0
18 : 1
18 : 2
18 : 3
60
90
120
0.01"0.01
0.00"0.00
0.00"0.00
0.08"0.09
0.05"0.05
0.48"0.42
0.22"0.20
0.14"0.13
0.42"0.34
0.11"0.11
a
a
b
b
a
b
a
b
bc
bc
0.00"0.00
0.17"0.07
0.13"0.06
1.73"0.70
1.00"0.33
4.10"1.18
1.93"0.99
1.65"0.90
3.74"0.85
1.02"0.29
a
a
ab
b
a
ab
a
ab
abc
abc
0.00"0.00
0.16"0.08
0.25"0.22
2.20"1.90
1.26"1.13
4.86"3.40
2.06"1.96
1.97"0.94
4.47"1.99
1.34"0.76
a
a
ab
ab
a
ab
a
ab
ab
ab
0.00"0.00
0.20"0.16
0.48"0.21
5.71"2.05
3.43"2.44
6.69"1.83
2.39"0.87
3.61"1.40
5.40"1.23
1.83"0.43
a
a
a
a
a
a
a
a
a
a
0.01"0.00
0.01"0.00
0.01"0.00
0.06"0.03
0.04"0.02
0.53"0.57
0.17"0.13
0.05"0.04
0.76"0.93
0.12"0.14
a
b
b
b
b
b
b
b
a
a
0.00"0.00
0.08"0.03
0.07"0.02
0.72"0.12
0.47"0.08
2.58"1.11
1.32"0.63
0.67"0.31
1.95"0.98
0.32"0.14
a
b
ab
b
b
ab
ab
b
a
a
0.02"0.04
0.25"0.05
0.37"0.11
2.77"0.43
1.30"0.36
4.14"1.48
1.73"0.69
1.01"0.42
4.41"2.85
0.55"0.40
a
a
a
a
a
ab
ab
ab
a
a
0.00"0.00
0.12"0.04
0.35"0.16
2.45"0.79
1.58"0.18
7.11"3.64
4.17"2.25
1.93"0.66
4.61"0.91
1.41"0.91
a
b
a
a
a
a
a
a
a
a
0.00"0.00
0.07"0.00
0.06"0.01
0.55"0.22
0.29"0.12
0.55"0.04
0.10"0.02
0.55"0.02
0.16"0.02
0.00"0.00
a
a
ab
b
b
c
c
b
b
a
0.18"0.26 a
1.60"2.28 a
1.11"1.54 b
11.34"15.68 b
7.09"10.02 b
9.80"12.49 bc
4.07"5.55 bc
4.68"6.08 b
6.79"7.46 ab
0.60"0.74 a
1.00"0.05
11.25"1.94
8.17"1.24
68.15"9.56
36.85"3.81
41.44"1.41
15.80"3.15
24.00"6.44
24.54"9.11
1.24"0.16
a
a
a
a
a
ab
ab
ab
ab
a
0.48"0.25 a
6.87"5.73 a
5.86"3.00 ab
51.44"0.14 ab
29.34"14.21 ab
36.25"15.44 ab
14.18"5.66 abc
22.36"8.71 ab
23.47"13.43 ab
1.12"0.45 a
0.52"0.45 a
10.92"2.80 a
8.63"1.29 ab
78.94"10.75 a
44.57"7.44 a
54.45"6.56 a
21.81"4.30 a
35.69"13.55 a
27.78"1.56 a
1.71"0.46 a
0.66"0.08
6.45"1.40
5.01"1.08
43.97"9.49
19.40"4.63
10.85"3.50
3.50"0.94
7.96"3.21
2.02"0.68
0.01"0.01
a
a
a
a
a
a
a
a
a
a
0.57"0.11
5.82"1.25
4.38"0.85
39.45"6.04
18.97"2.05
10.48"1.06
3.25"0.47
7.18"0.90
1.84"0.12
0.01"0.01
1.15"1.00
6.10"1.14
4.43"1.01
37.72"7.98
17.40"2.09
9.43"0.70
2.98"0.30
6.10"0.43
1.76"0.26
0.01"0.00
a
a
a
a
a
a
a
ab
a
a
0.23"0.07
2.24"0.75
1.66"0.44
14.29"3.43
6.71"2.02
3.86"0.97
1.14"0.32
2.85"0.31
0.73"0.02
0.00"0.00
0.24"0.04
2.40"0.26
1.84"0.19
16.20"1.31
7.37"0.29
4.13"0.12
1.24"0.01
2.93"0.32
0.83"0.12
0.02"0.01
a
a
a
a
a
a
a
a
a
a
for successful germination. There appears to be a good
correlation between the accumulation of 12 : 0 and the
development of the haustorium which, in turn, is critical
to the germination process. The accumulation of 12 : 0
in situ is apparent at 30 d suggesting enzymatic and gene
expression changes occurring even earlier. Such a marker
could be used to evaluate somatic embryos and their
further development. At the current stage, care must
be taken because other effects dependent on genotype
(Kulper, 1985), plant hormones (Mhaske and Hazra,
a
b
b
b
b
b
b
b
b
a
a
b
b
b
b
b
b
b
b
a
1994; Zou et al., 1995), phosphorylated proteins
(Islas-Flores et al., 1998), and non-cellulosic polysaccharides (White et al., 1989), have not been considered.
The data show a fall in the lipid content of solid
endosperm during the development in situ of the coconut
seedling as ®rst the haustorium and then the growing root
and shoot accumulate lipids. This pattern is consistent
with the proposed route of nutrition of the growing parts
which emphasizes the role of the haustorium in digesting
and absorbing lipids from the solid endosperm (Davies,
Fatty acids in germinating coconut 939
Table 2. Fatty acid composition of tissues of zygotic coconut embryos germinated in vitro
Results are expressed as mg of fatty acid per tissue and shown as mean"SEM from determinations carried out on three independent embryos.
Differences between columns (i.e. for a single fatty acid at different stages of development) were not signi®cant (P)0.05), according to ANOVAuTukey
test, when the entries bear the same letter.
Fatty acid
Days in culture
0
(A) Plumules
6:0
8:0
10 : 0
12 : 0
14 : 0
16 : 0
18 : 0
18 : 1
18 : 2
18 : 3
0.6"1.0 a
1.3"0.5 a
2.0"1.0 b
5.3"5.7 b
4.6"0.5 b
29.6"7.5 b
25.0"15.1 b
9.6"8.3 b
5.3"3.0 b
0.0"0.0 c
15
0.3"0.5 a
1.6"0.5 a
2.0"2.6 b
5.0"1.4 b
8.6"6.1 b
61.0"8.5 b
51.3"11.1 b
17.3"10.6 b
31.0"17.0 b
9.0"9.0 abc
(B) Radicles
6:0
8:0
10 : 0
12 : 0
14 : 0
16 : 0
18 : 0
18 : 1
18 : 2
18 : 3
(C) Haustoria
6:0
8:0
10 : 0
12 : 0
14 : 0
16 : 0
18 : 0
18 : 1
18 : 2
18 : 3
0.3"0.5 a
3.0"1.7 bc
7.3"2.5 a
84.6"7.2 a
100.3"8.0 a
598.6"18.7 a
126.0"38.5 a
609.6"43.0 a
260.3"18.8 a
0.0"0.0 a
0.0"0.0 a
2.0"2.0 c
7.0"1.7 a
61.6"23.7 a
80.3"31.5 a
701.6"206.1 a
113.0"13.0 a
594.3"157.3 a
444.0"120.4 a
36.0"13.0 a
30
45
60
75
4.5"3.5 a
9.0"7.0 a
12.5"4.9 ab
23.0"12.7 b
31.5"14.8 ab
364.0"193.7 ab
221.0"131.5 ab
141.0"96.1 ab
279.0"145.6 ab
2.5"3.5 bc
3.0"3.0 a
7.3"1.1 a
9.6"1.5 ab
17.6"1.5 b
22.3"2.5 ab
439.0"37.5 ab
188.6"28.3 ab
167.3"57.2 ab
459.6"51.4 a
69.6"17.5 ab
3.0"5.1 a
9.6"10.5 a
26.6"18.1 a
172.5"11.5 a
90.6"48.5 a
761.3"333.1 a
397.0"166.3 a
294.3"167.2 a
476.3"201.9 a
67.6"48.6 a
3.0"2.0 ab
4.6"3.0 a
6.3"2.0 b
8.6"2.0 a
9.3"4.1 a
141.0"49.1 c
94.0"28.1 c
19.3"10.7 b
125.5"9.1 b
6.3"7.7 a
1.3"2.3 b
8.6"5.7 a
11.3"5.7 a
18.3"9.2 a
29.3"15.0 a
315.0"71.3 ab
195.6"30.0 b
59.0"30.3 ab
150.6"35.9 ab
6.3"3.2 a
0.0"0.0 b
4.3"3.7 a
9.6"2.3 ab
22.5"4.9 a
36.6"22.7 a
248.6"10.4 bc
143.0"12.2 bc
42.0"9.5 ab
192.3"6.8 ab
13.3"3.5 a
7.6"2.5 a
10.3"1.5 a
13.0"2.0 a
37.3"17.3 a
42.3"10.9 a
436.3"65.8 c
269.0"26.0 a
84.6"31.5 a
234.6"55.6 a
21.6"8.7 a
2.6"2.3 a
3.3"1.5 bc
5.6"1.1 a
37.0"23.4 a
40.3"30.8 a
346.3"125.6 a
102.6"12.5 a
222.3"130.8 b
236.6"13.8 a
23.0"10.5 a
0.6"1.0 a
8.3"0.5 ab
9.3"2.3 a
62.0"39.8 a
79.0"37.6 a
525.0"201.4 a
195.3"40.1 a
293.0"98.4 ab
175.3"133.0 a
6.6"0.5a
2.3"4.0 a
8.6"3.2 ab
8.0"1.7 a
57.3"16.7 a
75.3"27.1 a
601.0"226.7 a
189.6"62.0 a
355.6"116.9 ab
349.6"291.3 a
33.6"30.2 a
0.0"0.0 a
11.6"2.5 a
10.0"1.0 a
105.0"50.4 a
77.0"19.5 a
468.6"82.4 a
179.6"75.0 a
177.6"161.3 b
207.0"78.4 a
8.0"5.0 a
2.0"1.0 a
2.6"0.5 a
2.6"1.0 b
6.0"2.6 b
9.6"4.5 b
86.6"23.0 b
60.6"9.4 b
23.3"8.1 b
63.6"20.3 b
14.6"6.5 abc
1983; Oo and Stumpf, 1983) and supplying them to the
shoot and root. The mechanism suggests an explanation
for the observed correlation between fatty acid composition and development.
The translocation of longer chain, saturated and
unsaturated, fatty acids to growing tissue has an obvious
application in the synthesis of new membrane phospholipids and glycolipids (Turnham and Northcote, 1984).
Lauric acid is presumably used as a source of energy, via
b-oxidation, to fuel the biochemical activities associated
with growth and to provide a source of carbon, in the
absence of photosynthesis, via the glyoxylate shuttle and
fatty acid synthesis de novo, for the growth. The question
of why the seedlings in vitro did not appear to be able to
use the sucrose supplied to carry out the same functions is
more perplexing. It may be that the enzyme activities
required to use the alternative substrate are not expressed
in suf®cient quantity until leaves develop and photosynthesis begins.
There are a number of possible reasons for the failure
of the haustoria to develop in vitro and the consequent
unbalanced germination of the embryo. One of them is
not that the supply of sucrose becomes limiting in vitro.
The total reducing sugar content of the medium
(measured after invertase digestion of the sucrose) falls
to 46.1"7.3% (mean"SEM from nine determinations)
of the original content after 4 weeks in culture, after
which the medium was changed. As suggested above, the
failure to supply preformed fatty acids for membranes or
a source of lauric acid at the critical early stages of
development may be important. Nevertheless, endosperm
does contain an equivalent amount of soluble sugars
(3.74% of solid matter) during the ®rst 10 weeks of
germination (Balachandran and Arumughan, 1995;
Balasubramaniam et al., 1973), so the seedling will
normally have a source of carbohydrate which could
be used for fatty acid synthesis. One important
function of haustorium in coconut germination is to
940
LoÂpez-Villalobos et al.
Fig. 4. Fatty acid composition of solid endosperm from mature nuts at
the indicated sampling dates; each value represents the mean"the
standard error of the mean from three individual coconuts, the symbols
are given and the terms de®ned in the legend to Fig. 3.
Fig. 3. Fatty acid composition of plumules (A), radicles (B) and
haustoria (C) of germinating coconuts in situ at the indicated sampling
dates; each value represents the mean"the standard error of the mean
from three individual coconuts, (j) shorter chain fatty acids; (k) lauric
acid; (m) longer chain saturated acids; (m) unsaturated fatty acids.
`Shorter chain fatty acids' are saturated fatty acids from 6±10 carbons in
length and were mostly octanoic (8 : 0) and decanoic (10 : 0) acids with a
small proportion of hexanoic acid (6 : 0). Lauric acid is dodecanoic acid
(12 : 0). `Longer-chain saturated acids' were predominately palmitic acid
(16 : 0) with smaller amounts of myristic (14 : 0) and stearic (18 : 0) acids.
`Unsaturated fatty acids' consisted of oleic (18 : 1), linoleic (18 : 2) and
linolenic (18 : 3) acids with 18 : 1 being the most abundant and 18 : 3
seldom exceeding 6% of the total. Acids longer or more unsaturated
than 18 : 3 or shorter than 6 : 0 were not detected.
convert lipid into sugars via the glyoxylate cycle
(Balachandran and Arumughan, 1995). The continued
provision of sucrose, beyond the ®rst 10 weeks in the
culture medium, will be more readily available to the
growing seedling than that naturally occurring in endosperm. Paradoxically, this will result in the maintenance
of intracellular sucrose concentrations which may
have the effect of down-regulating the activity of the
glyoxylate cycle and, with it, the growth of the
haustorium. Failure to develop the haustorium would
then debilitate the subsequent development of roots and
shoots. Finally, plant hormones such as abscisic acid and
cytokinins (Kobayashi et al., 1997), contained in coconut
water and varying in composition throughout development (Hoad and Gaskin, 1980) but absent from the
culture medium, will be important. Plant growth regulators were omitted because recent results in these laboratories have shown no improvement in germination
when they were included (Hornung, 1995b).
The work presented here has shown marked differences in the way intact coconuts germinate in situ compared to the germination of zygotic embryos in vitro. In
particular, major differences in the fatty acid compositions have been identi®ed. The accumulation of lauric
acid provides a potential marker for use in micropropagation studies; more sensitive markers may come from
the use of probes for the mRNAs of key enzymes in the
synthesis of lauric acid-containing lipids (Davies, 1983;
Davies et al., 1995; Voelker et al., 1997). The work also
suggests modi®cations to the nutrient medium used
in vitro so that an adequate supply of important fatty
acids can be provided. Such modi®ed media may also
assist in the development of methods to recover excised,
transported or conserved embryos. In other species,
improvements in zygotic embryogenesis in vitro, have
led to similar improvements in somatic embryogenesis in vitro (Dutta et al., 1991; Taylor and Vasil,
1996; Zou et al., 1995). Experiments in progress are
testing the effects of supplementing culture media with
lauric acid.
Fatty acids in germinating coconut 941
References
Fig. 5. Fatty acid composition of plumules (A), radicles (B), and
haustoria (C) of germinating coconuts in vitro at the indicated sampling
dates; each value represents the mean"the standard error of the mean
from three individual cultures; the symbols are given and the terms
de®ned in the legend to Fig. 3.
Acknowledgements
We are grateful to the staff of the Plant Biotechnology
Laboratories (PBL), T.H. Huxley School, Imperial College at
Wye for assistance with tissue culture, Mr C Kempe for
technical assistance and Mr A Russell for assistance with the
glasshouse experiments. We thank Mr B Been of the Coconut
Industry Board, Jamaica for supplying the plant materials.
AL-V received ®nancial support from Consejo Nacional de
TecnologõÂa, Mexico and the Colegio de Postgraduados en
Ciencias AgrõÂcolas de MeÂxico.
Balachandran C, Arumughan C. 1995. Biochemical and cytochemical transformations in germinating coconut (Cocos
nucifera L.). Journal of Agriculture and Food Chemistry 72,
1385±1391.
Balasubramaniam K, Atokorala TMS, Wijesundera S,
Hoover AA. 1973. Biochemical changes during germination of coconut (Cocos nucifera). Annals of Botany 37,
439±445.
Blake J. 1995. A brief history of coconut tissue culture. In:
Oropeza V, Howard FW, Ashburner GR, eds. Lethal
yellowing: research and practical aspects. Dordrecht, The
Netherlands: Kluwer Academic Publishers, 217±223.
Bligh EG, Dyer WJ. 1959. A rapid method of total lipid
extraction and puri®cation. Canadian Journal of Biochemistry
and Physiology 37, 911±917.
Branton RL, Blake J. 1983. Development of organized
structures in callus derived from explants of Cocos nucifera
L. Annals of Botany 52, 673±678.
Chan JL, 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
Reports 17, 515±521.
Davies HM. 1983. Medium chain acyl-ACP hydrolysis activities
of developing oilseeds. Phytochemistry 33, 1353±1356.
Davies HM, Hawkins DJ, Nelsen JS. 1995. Lysophosphatidic
acid acyltransferase from immature coconut endosperm have
medium chain length substrate speci®city. Phytochemistry 39,
989±996.
Dutta PC, Appelqvist LA, Gunnarsson S, Von Hofsten A. 1991.
Lipid bodies in tissue culture, somatic and zygotic embryo of
Daucus carota L.: a qualitative and quantitative study. Plant
Science 78, 259±267.
Eden-Green S. 1995. A brief history of lethal yellowing
research. In: Oropeza C, Howard FW, Ashburner GR, eds.
Lethal yellowing: research and practical prospects, Vol. 5.
Dordrecht, The Netherlands: Kluwer Academic Publishers,
229±244.
Haccius B, Philip VJ. 1979. Embryo development in Cocos
nucifera L.: a critical contribution to a general understanding
of palm embryogenesis. Plant Systematics and Evolution 132,
91±106.
Hoad GV, Gaskin P. 1980. Abscisic acid and related compounds
in phloem exudate of Yucca ¯accida Haw. and coconut (Cocos
nucifera L.). Planta 150, 347±348.
Hornung R. 1995a. Micropropagation of Cocos nucifera (L.)
from plumular tissue excised from mature zygotic embryos.
Plantations Recherche DeÂveloppement 2, 38±41.
Hornung RKW. 1995b. Developmernts in the micropropagation
of Cocos nucifera L. PhD thesis: University of London.
Hornung R, Verdeil J-L. 1999. Somatic embryogenesis in
coconut from immature in¯orescence explants. In: Oropeza
C, Verdeil JL, Ashburner GR, CardenÄa R, Santamaria JM,
eds. Current advances in coconut biotechnology. Dordrecht,
The Netherlands: Kluwer Academic Publishers, 297±309.
Islas-Flores I, Oropeza C, HernaÂndez-Sotomayor SMT. 1998.
Protein phosphorylation during coconut zygotic embryo
development. Plant Physiology 118, 257±263.
Iverson JL, Sheppard AJ. 1977. Butyl ester preparation for
gas-liquid chromatographic determination of fatty acids in
butter. Journal of the American Oil Chemists' Society 60,
294±298.
Kobayashi H, Morisaki N, Tago Y, Hashimoto Y,
Iwasaki S, Kawachi E, Nagata R, Shudo K. 1997. Structural
identi®cation of a major cytokinin in coconut milk as
942
LoÂpez-Villalobos et al.
14-O-{3-O-wb-D-galactopyranosyl-(1 £ 2)-a-D-galactopyranosyl-(1£3)-a-L-arabinofuranosylx-4-O-(a-L-arabinofuranosyl)b-D-galactopyranosyl}-trans-zeatin riboside. Chemical and
Pharmaceutical Bulletin 45, 260±264.
Kulper PJC. 1985. Environmental changes and lipid metabolism
of higher plants. Physiologia Plantarum 64, 118±122.
Mhaske VB, Hazra S. 1994. Appearance of storage lipid
(triglycerides) in somatic embryos of peanut (Arachis
hypogaea L). In vitro Cell and Developmental Biology
30P, 113±116.
Murashige T, Skoog F. 1962. A revised medium for rapid
growth and bioassays with tobacco tissue cultures. Physiologia
Plantarum 15, 473±497.
Oo KC, Stumpf PK. 1983. The metabolism of the germinating
oil palm (Elaeis guineensis) seedling. Plant Physiology 73,
1033±1037.
Persley GJ. 1992. Replanting the tree of life: towards
an international agenda for coconut palm research. Oxford,
UK: CAB: ACIAR, 156.
Rossell JB, King B, Downes MJ. 1985. Composition of oil.
Journal of the American Oil Chemists' Society 62, 221±230.
Steel RGD, Torrie JH. 1960. Principles and procedures of
statistics, New York, USA: McGraw-Hill, 109±110.
Taylor MG, Vasil IK. 1996. Quantitative analysis of ultrastructural changes during zygotic and somatic embryogenesis
in pearl millet (Pennisetum glaucum (L.) R. Br.). Sexual Plant
Reproduction 9, 286±298.
Turnham E, Northcote DH. 1984. The incorporation of
w1-14Cxacetate into lipids during embryogenesis in oil palm
tissue cultures. Biochemistry 23, 35±39.
Voelker TA, Jones A, Cranmer AM, Davies HM, Knutzon DS.
1997. Broad-range and binary-range acyl-acyl-carrier-protein
thioesterases suggest an alternative mechanism for mediumchain production in seeds. Plant Physiology 114, 669±677.
White AR, Elmore HW, Watson MB, Gill JP. 1989. Puri®cation
and partial characterization of polysaccharides from coconut
milk. Annals of Botany 64, 205±209.
Zou J, Abrams GD, Barton DL, Taylor DC, Pomeroy MK,
Abrams SR. 1995. Induction of lipid and oleosin biosynthesis
by (q)abscisic acid and its metabolites in microspore-derived
embryos of Brassica napus L. cv. Reston. Plant Physiology
108, 563±571.