Journal of Experimental Botany, Vol. 55, No. 401, pp. 1401±1410, June 2004
DOI: 10.1093/jxb/erh149 Advance Access publication 7 May, 2004
RESEARCH PAPER
Tomato fruit cuticular waxes and their effects on
transpiration barrier properties: functional characterization
of a mutant de®cient in a very-long-chain fatty acid
b-ketoacyl-CoA synthase
Gerd Vogg1,*, Stephanie Fischer1, Jana Leide1, Eyal Emmanuel2, Reinhard Jetter3, Avraham A. Levy2 and
Markus Riederer1
1
UniversitaÈt WuÈrzburg, Lehrstuhl fuÈr Botanik II, D-97082 WuÈrzburg, Germany
2
Department of Plant Sciences, Weizmann Institute of Science, Rehovot 76100, Israel
Departments of Botany and Chemistry, University of British Columbia, Vancouver, BC, V6T 1Z4 Canada
3
Received 4 February 2004; Accepted 27 February 2004
Abstract
Cuticular waxes play a pivotal role in limiting transpirational water loss across the plant surface. The
correlation between the chemical composition of the
cuticular waxes and their function as a transpiration
barrier is still unclear. In the present study, intact
tomato fruits (Lycopersicon esculentum) are used,
due to their astomatous surface, as a novel integrative approach to investigate this composition±
function relationship: wax amounts and compositions of tomato were manipulated before measuring
unbiased cuticular transpiration. First, successive
mechanical and extractive wax-removal steps allowed
the selective modi®cation of epi- and intracuticular
wax layers. The epicuticular ®lm consisted exclusively of very-long-chain aliphatics, while the intracuticular compartment contained large quantities of
pentacyclic triterpenoids as well. Second, applying
reverse genetic techniques, a loss-of-function mutation with a transposon insertion in a very-long-chain
fatty acid elongase b-ketoacyl-CoA synthase was
isolated and characterized. Mutant leaf and fruit
waxes were de®cient in n-alkanes and aldehydes with
chain lengths beyond C30, while shorter chains and
branched hydrocarbons were not affected. The
mutant fruit wax also showed a signi®cant increase
in intracuticular triterpenoids. Removal of the epicuticular wax layer, accounting for one-third of the total
wax coverage on wild-type fruits, had only moderate
effects on transpiration. By contrast, reduction of the
intracuticular aliphatics in the mutant to approximately 50% caused a 4-fold increase in permeability.
Hence, the main portion of the transpiration barrier is
located in the intracuticular wax layer, largely determined by the aliphatic constituents, but modi®ed by
the presence of triterpenoids, whereas epicuticular
aliphatics play a minor role.
Key words: Cuticle, epicuticular wax, intracuticular wax,
tomato, transpiration barrier.
Introduction
All primary aerial surfaces of plants are covered by a
cuticle. One of the major functions of the cuticle is the
formation of an ef®cient barrier against unregulated water
loss (Kerstiens, 1996a). It is a thin, hydrophobic, and
¯exible membrane (0.1±10 mm) that is composed of a
polymer matrix (cutin) and associated solvent-soluble
lipids (cuticular waxes). The cuticular waxes consist of
species-, organ- and tissue-speci®c homologous series of
very-long-chain aliphatics (e.g. fatty acids, alcohols,
aldehydes, alkanes, and esters) with characteristic chain
length distributions and varying proportions of cyclic
compounds (e.g. pentacyclic triterpenoids and/or phenyl
propanoids).
Despite considerable efforts, the relationship between
wax composition and the cuticular barrier properties is
* To whom correspondence should be addressed. Fax: +49 931 8886235. E-mail: [email protected]
Journal of Experimental Botany, Vol. 55, No. 401, ã Society for Experimental Biology 2004; all rights reserved
1402 Vogg et al.
poorly understood (Riederer and Schreiber, 2001). This
might be due to the fact that most wax analyses and
functional studies in the past did not differentiate between
cuticular substructures. Recently, methods have been
devised that allow the selective sampling of exterior
(epicuticular) and internal (intracuticular) wax layers for
chemical analyses (Jetter et al., 2000; Jetter and SchaÈffer,
2001). The permeability properties of epi- and intracuticular waxes might differ in relation to the chemical
gradients between these layers. Hence, it is necessary to
assess how much the cuticular compartments contribute to
the overall transpiration barrier.
Previous investigations into the composition±function
relationship of cuticular waxes relied on the description of
various plant species, i.e. on comparisons involving
variation of multiple characters (Riederer and Schreiber,
2001). In order to overcome this limitation, it would be
desirable to investigate wax biosynthetic mutants that
differ partially from the wild-type cuticle composition of a
selected species. On the one hand, several Arabidopsis
mutants with modi®cations in cuticular wax composition
have been chemically characterized (Koornneef et al.,
1989; Hannoufa et al., 1993; McNevin et al., 1993; Jenks
et al., 1995; Lemieux, 1996; Chen et al., 2003).
Unfortunately, the relatively small organs of Arabidopsis
make the functional analysis of its plant cuticle a dif®cult
task. On the other hand, the cuticular transpiration of
several plant species has been assessed in vitro using
isolated cuticular membranes (SchoÈnherr and Lendzian,
1981). This approach, however, is restricted to plants with
well-developed cuticles, and, for the relatively few species
thus investigated to date, wax biosynthetic mutants are not
available. For other plant species, cuticular transpiration
has to be measured using intact material. To this end,
astomatous tissues must be used because even under
conditions of minimal permeances (darkness, low air
humidity, high CO2) a residual component of stomatal
transpiration is likely (Kerstiens, 1996b).
Considering all these restrictions, tomato (Lycopersicon
esculentum) fruits appear suitable for this approach as they
lack stomata. The wax mixture is fairly simple, consisting
predominantly of alkanes and triterpenoids (Baker et al.,
1982). Furthermore, tomato is an important crop with
strong genetic and genomic tools available (Emmanuel and
Levy, 2002). Meissner et al. (1997, 2000) constructed an
Ac/Ds transposon-tagged mutant population of the dwarf
cultivar Micro-Tom and sequenced a large number of Ds
insertion sites. Based on sequence similarity to known wax
biosynthetic genes, mutants with potentially altered cuticle
properties can be selected from this collection.
In the present investigation, the focus was on how
cuticular wax composition is correlated with unbiased
cuticular water-barrier properties. The astomatous tomato
fruit system cv. Micro-Tom, which offers the possibility to
apply new genetic and phytochemical tools to manipulate
plant cuticles in an intact system was chosen. The
following questions were addressed. (1) Are cuticular
waxes on wild-type fruits arranged in layers with distinct
chemical composition and how do these layers contribute
to the permeation barrier? (2) If wax coverage and
composition are altered due to mutation of a particular
wax biosynthetic gene, a fatty acid b-ketoacyl-CoA
synthase, how will the overall barrier properties be
affected? (3) Employing a combination of genetic tools
and selective removal of wax layers, can the transpiration
barrier be localized within speci®c compartments and
components of the cuticular waxes?
Materials and methods
Plant material
Lycopersicon esculentum cv. Micro-Tom plants were transformed
with Ac/Ds constructs (Meissner et al., 2000). For the present
investigation, F4 to F6 progenies of the original F1 plant crosses
between Ac-transposase and Ds lines were used. Stabilized Ds lines
were selected for the isolation of transposon insertion sites (TIS) by
long-range Inverse-PCR (IPCR) adapted from Mathur et al. (1998)
as described below. The plants were cultivated in a growth chamber
with a 14/10 h day/night regime of 22/18 °C with constant 75%
relative humidity.
Identi®cation of the lecer6 insertional mutant
Genomic DNA was extracted from each Ds-insertion line (Dneasy
kit, Qiagen, Hilden, Germany), digested with restriction endonucleases and self-circularized with T4 DNA ligase (New England
Biolabs Inc., Beverly MA, USA). After heat inactivation the samples
were subjected to IPCR ampli®cation using speci®c primers of the
Ds-5¢, TTGTATATCCCGTTTCCGTTCCGTT, and Ds-3¢ termini,
TTTCGTTTCCGTCCCGCAAGTTAAATA. PCR reactions were
performed using the Expand Long Template PCR System (Roche
Diagnostics, Mannheim, Germany) according to the manufacturer's
instructions. PCR fragments were cleaned (Qiaquick PCR puri®cation kit, Qiagen) and used directly as templates for sequencing with
Ds-5¢ and Ds-3¢ speci®c primers (Gorbunova and Levy, 1997).
Each sequence was subjected to similarity searches using the
BLAST 2.0 package on the NCBI server, the TIGR database, and
the Lab-on-web services (Compugen, Tel Aviv, Israel). The
genotype of each individual plant (wild-type, homozygote, or
heterozygote for the Ds insertion into the target gene) was tested
by PCR using a compination of two gene-speci®c and one insertspeci®c primer within one assay (lecer6_forward: 5¢-TCCAGCTAAGAGGAAGTTG-3¢; lecer6_reverse: 5¢-GCTTCATTTCTAGCAGTCTCC-3¢; insert_reverse: 5¢-AGTTAAATATGAAAATGAAAACGG-3¢). In the case of template DNA from wildtype plants, PCR resulted in a single product of 695 bp; homozygote
mutant DNA template resulted in a single product of 386 bp,
heterozygote lines showed both PCR products. The mutant lines
were also characterized by DNA hybridization using a Southern blot
analysis as described (Bernatzky and Tanksley, 1986), with a
radiolabelled LeCER6 probe and genomic DNA digestion with
restriction enzymes (HindIII, BglII, and SpeI).
Expression analysis
Gene expression analysis was performed by semi-quantitative RTPCR. RNA extractions from Micro-Tom leaves and fruits (epidermis
6subepidermal layers) were carried out with the Trizol reagent
(Invitrogen, Karlsruhe, Germany). RNA integrity was controlled by
Cuticular wax composition and transpiration of tomato fruits 1403
agarose gel electrophoresis. cDNA ®rst-strand synthesis was carried
out according to the manufacturer's instructions (Superscript RTase,
Invitrogen).
Template concentration and linear ampli®cation range of PCR
cycles were tested in earlier experiments. 27 cycles with an
annealing temperature of 54 °C were chosen for LeCER6 ampli®cation. Target gene expression was analysed in parallel to actin in
a single assay. The following PCR primers were used: lecer6_1:
5¢-ATTTCTGTAACTCTGTTAAGC-3¢; lecer6_2: 5¢-ACCATAGCAGATAAAGACGG-3¢; actin_1: 5¢-CAAGTCATCATCCGTTTG-3¢; actin_2: 5¢-ATACCAGTGGTACGACC-3¢. To test the
reproducibility of reverse transcriptase ef®ciency between different
samples 13106 copies of synthetic RNA were added to the plant
RNA and checked for identical PCR ampli®cation (GeneAmplimer
pAW109 RNA; Applied Biosystems, Weiterstadt, Germany).
Wax analyses of leaves and fruits
An adhesive ®lm of gum arabic was used for the selective
preparation and analysis of surface waxes from intact tomato fruits,
according to the method of Jetter and SchaÈffer (2001). The whole
fruit surface was covered with a dispersion of gum arabic in water
(1 g ml±1; Roth, Karlsruhe, Germany). After drying, a thin whitish
layer of adhesive could carefully be removed with forceps.
This mechanical treatment was repeated once on the same
fruit surface. The gum arabic fractions were collected separately
and extracted with water/chloroform. After vigorous agitation
and phase separation the organic solution was removed and the
solvent was evaporated under reduced pressure. After mechanical
treatment, the remaining waxes were extracted for 2 min with 20 ml
chloroform (>99%; Roth) at room temperature. Only 80% of the fruit
were dipped to avoid solvent penetration at the attachment area of
the pedicel. Tetracosane was added to all extracts as internal
standard and the solvent was removed under reduced pressure. In a
similar way, untreated leaf blades and fruits were immersed to
extract total wax mixtures. Complete extraction was in all cases
achieved after 1±2 min. Lipids from internal tissues (chain lengths of
C16 and C18) could not be detected in these extracts. This was tested
in a series of experiments with extraction times ranging from 5 s to
10 min.
Prior to gas chromatography (GC) analysis, hydroxyl-containing
compounds in all samples were transformed to the corresponding
trimethylsilyl derivatives using BSTFA in pyridine (MachereyNagel, Dueren, Germany). After baseline separation with a
temperature controlled capillary GC (8000Top, Fisons
Instruments, Rodano-Milan, Italy) with He carrier gas inlet pressure
constant at 30 kPa, wax components were identi®ed using a mass
spectrometric detector (70 eV, m/z 50±850, MD1000, Fisons). GC
was carried out with temperature-programmed injection at 50 °C,
over 2 min at 50 °C, raised by 40 °C min±1 to 200 °C, held for 2 min
at 200 °C, raised by 3 °C min±1 to 320 °C, and held for 30 min at 320
°C. d-amyrin was identi®ed according to Bauer et al. (2002). The
quantitative composition of the mixtures was studied by capillary
GC (5890 II, Hewlett Packard, Avondale, PA, USA; 30 m DB-1,
0.32 mm i.d., df=1 mm, J&W Scienti®c, Folsom CA, USA) and ¯ame
ionization detection under the same conditions as above, but with H2
as carrier gas. Single compounds were quanti®ed against the internal
standard by manually integrating peak areas.
Characterization of cuticular water transport
Cuticular water permeances were determined for intact, mature, red
Micro-Tom fruits. The lack of stomata on the Micro-Tom fruit
surface was con®rmed microscopically by producing replicas of the
epidermis with Collodion 4% (Merck). The attachment site of the
pedicel was sealed with paraf®n (Merck). The amount of water
transpired versus time (5±8 data points per individual fruit; intervals
between measurements: 6±10 h) was measured using a balance with
a precision of 0.1 mg (Sartorius AC210S, Goettingen, Germany).
Cuticular water ¯ow rates (F, in g s±1) of individual fruits were
determined from the slope of a linear regression line ®tted through
the gravimetric data. Coef®cients of determination (r2) for a
representative data set averaged 0.999. Fluxes (J, in g m±2 s±1)
were calculated by dividing F by total fruit surface area as
determined from the average of vertical and horizontal diameters
by assuming a spherical shape of the fruit. During the measurements
the fruits were stored at constant 25 °C over dry silica gel. Under
these conditions, the external water vapour concentration is essentially zero. The vapour phase-based driving force (Dc) for
transpiration is therefore 23.07 g m±3. For calculating water
permeance (P, in m s±1) based on water vapour concentration, J
was divided by Dc.
For analysing the contribution of total cuticular waxes on water
permeability, the fruits were dipped for 1 min in CHCl3. The
in¯uence of the epicuticular wax layer was measured after mechanical removal with gum arabic.
Results
Identi®cation of a tomato mutant de®cient in a gene
homologous to CER6 in Arabidopsis
The transposon insertion sites (TIS) from the collection
described by Meissner et al. (2000) were subjected to
similarity searches on web-localized databases. One
line containing one TIS with assigned homology to
TC125065, a tentative consensus sequence of tomato
ESTs, showed 97% identity on the nucleotide level in the
overlapping region (Fig. 1A, B). This TC sequence shows
a signi®cant similarity to At1g68530, Arabidopsis verylong-chain fatty acid b-ketoacyl-CoA synthase CER6
(83% identity, 94% similarity). Hence, it was concluded
that the tomato gene encoded a CER6-like b-ketoacyl-CoA
synthase (condensing enzyme), and it was designated as
LeCER6.
Based on the expression pro®le of ESTs identical to
LeCER6 clustered in TC125065 (18 ESTs are available in
Lycopersicon species as of August 2003), it could be
deduced that LeCER6 is expressed in trichomes, ¯owers,
leaves, germinating seeds, and fruits. Gene expression of
LeCER6 in Micro-Tom leaves and epidermis of fruits was
tested by RT-PCR (Fig. 1C). No expression was detectable
in the homozygous mutant, whereas the heterozygous
plants did not differ signi®cantly under the experimental
conditions used in these RT-PCR experiments. The
complete molecular analysis of the mutant, including the
complementation of the mutated gene, is currently being
completed and will be reported separately.
Fruit characteristics were correlated with the mutant
genotype: while 49% (n=343) of the homozygote lecer6
fruits had wrinkled surfaces, only 11% (n=335) and 4%
(n=277) of the heterozygotes and the wild-type developed
this symptom, respectively (Fig. 2A, C). The wrinkling
effect only developed in red, mature fruits. This phenotype
was also connected to a reduction in fruit size (wild-type
fruit surface area 7.9462.95 cm2 fruit±1; homozygote
1404 Vogg et al.
Fig. 2. Phenotype of fruits from Micro-Tom wild-type and its mutant
lecer6. (A) Wild-type fruits of Micro-Tom keep their spherical and
immaculate shape at the state of ripeness. (B) Mature fruits from
homozygous mutants tend to have a wrinkled surface. (C) Close-up of
mature fruits from lecer6. (D) Micrograph of a replica of the fruit
epidermis illustrating the absence of stomata. The bars in (A) to (C)
represent a length of 10 mm.
Fig. 1. Characterization of the transposon insertion site (TIS) in
Micro-Tom mutant fruits. (A) Comparison of TIS obtained from the
IPCR with the tentative consensus sequence TC125065 of tomato
ESTs. The dotted line shows the start of translation, the triangle the
site of Ds insertion. (B) Nucleotide sequence alignment from the
overlapping region of the TIS and TC125065 (Clustal W 1.8). (C)
Expression of LeCER6 in leaves and skin of mature fruits determined
by RT-PCR for wild-type (WT), heterozygous (+/±), and homozygous
mutants (±/±). Expression was compared to actin in the same assay.
RT-PCR ef®ciency in the different samples was monitored by
ampli®cation of a synthetic RNA (pAW109) in a separate assay.
lecer6 6.2462.66 cm2 fruit±1; data are mean 6sd) and in
fruit fresh weight (wild-type 2.5561.34 g fruit±1; homozygote lecer6 1.9161.20 g fruit±1; data are mean 6sd).
These fruit phenotypes were observed in the growth
chamber and in the greenhouse with suf®cient irrigation of
all plants.
Fig. 3. Composition of leaf cuticular waxes from Micro-Tom wildtype (+/+), from heterozygous (+/±), and from homozygous (±/±)
lecer6 mutant plants. (A) Chain length distribution of n-alkanes. (B)
Chain length distribution of methyl-branched (iso- and anteiso-)
alkanes and (insert to B) total triterpenoids (a-, b-, and d-amyrin)
within the same samples. (mean 6sd; n=6; statistical evaluation:
Student±Newman±Keuls procedure, SPSS 10.0; different letters differ
signi®cantly (P <0.05) as determined by one-way ANOVA).
Leaf cuticular waxes
The substantial part of solvent-extractable, cuticular leaf
waxes consisted of very-long-chain alkanes and of cyclic
triterpenoids (Fig. 3). The complete series of n-alkanes
with chain lengths between C27 up to C34 was detected
(Fig. 3A). Hentriacontane (C31H64) was found to dominate
the wild-type leaf wax. Besides, small amounts of methylbranched alkanes with branching at the iso- or anteiso
Cuticular wax composition and transpiration of tomato fruits 1405
Fig. 4. Wax coverages on fruits of Micro-Tom wild-type,
heterozygous (+/±), and homozygous (±/±) LeCER6-de®cient mutants.
Two treatments with gum arabic and an extraction with chloroform
were performed successively on the same fruit surface. (mean 6sd,
n=3, statistical evaluation: Student±Newman±Keuls procedure, SPSS
10.0; different letters differ signi®cantly (P <0.05) as determined by
one-way ANOVA).
position and the triterpenoids a-, b-, and d-amyrin were
identi®ed (Fig. 3B). The wax composition of heterozygous
mutants did not differ signi®cantly from the wild-type. In
sharp contrast, homozygous LeCER6-de®cient leaves
showed a signi®cantly reduced coverage of C31 up to C34
alkanes, whereas shorter n-alkanes, branched alkanes, and
triterpenoids remained unaltered. An increased accumulation of pathway intermediates and of compounds with
shorter chain lengths could not be observed. The branched
alkane isomers were not affected in the homozygote
mutant (Fig. 3B).
Fruit cuticular waxes
Total wax load on the Micro-Tom wild-type fruits was
approximately 15 mg cm±2 and was only slightly reduced
for the LeCER6-de®cient mutant plants (Fig. 4). With a
®rst gum arabic treatment 4 mg cm±2 of wax could be
mechanically removed from the wild-type surface. A
second adhesive treatment released only small amounts of
wax, whereas solvent extraction yielded largely increased
quantities of 10 mg cm±2. From heterozygous fruits, the
same sequence of wax removal steps released similar wax
quantities. By contrast, the ®rst gum arabic treatment of
homozygous mutant fruits yielded less than 1 mg cm±2,
while the ®nal extraction step in this case again released 10
mg cm±2.
Gum arabic samples from tomato fruits were found
exclusively to contain aliphatic components of cuticular
waxes (Fig. 5A). n-Hentriacontane dominated the wax
mechanically removed from wild-type and heterozygous
lecer6 plants, whereas in homozygotes this alkane was
only detectable in trace quantities. Longer chain n-alkanes
(=31 carbons) and the C32 n-aldehyde were also missing,
Fig. 5. Wax composition on fruits of Micro-Tom wild-type (+/+), of
heterozygous (+/±), and of homozygous (±/±) LeCER6-de®cient
mutants. Waxes were (A) mechanically removed in two gum arabic
treatments and (B) with a subsequent chloroform extraction (mean
6sd; n=5; statistical evaluation: separate analysis for each compound,
Student±Newman±Keuls procedure, SPSS 10.0). (C) Relative portions
of total aliphatics and cyclic triterpenoids in the wax of intact fruits
(n=5) and of enzymatically isolated cuticles (n=3). (mean 6sd;
Student-Newman±Keuls procedure, SPSS 10.0; different letters differ
signi®cantly (P <0.05) as determined by one-way ANOVA).
while shorter-chain fatty acids and alkanes, as well as the
C31 methyl-branched alkane, were unaffected by the
mutation.
Wax extraction from wild-type and heterozygous fruits
(Fig. 5B) yielded the same aliphatics as the previous gum
arabic treatments. In the extracts of homozygous mutants,
n-alkanes with chain lengths beyond C30 were again
missing, while shorter chain-length aliphatics were
unaffected. The extracts in all cases were found to contain
the pentacyclic triterpenoids a-, b-, and d-amyrin in high
quantities. For homozygous fruits, the decrease in aliphatic
components was parallelled by an increase in all three
triterpenoids by factors of three to six. To exclude the fact
that the higher yields were only due to a difference in the
1406 Vogg et al.
for the LeCER6-de®cient fruits (Fig. 6B). A consecutive
mechanical treatment did not lead to a further increase.
After removal of the epicuticular layer, the permeances of
wild-type and heterozygous fruits did not reach the level of
homozygous fruits.
Discussion
Fig. 6. Cuticular water permeability of Micro-Tom wild-type (+/+),
heterozygous (+/±), and homozygous (±/±) LeCER6-de®cient fruits.
(A) Effect of removal of total cuticular waxes. (B) Effect of removal
of epicuticular waxes using gum arabic. (mean 6sd, n=15, statistical
evaluation: Student-Newman±Keuls procedure, SPSS 10.0; different
letters differ signi®cantly (P <0.05) as determined by one-way
ANOVA).
extraction ef®ciency between mutant and wild-type fruits,
cuticular membranes were isolated enzymatically
(SchoÈnherr and Lendzian, 1981; SchoÈnherr and Riederer,
1986) and extracted exhaustively. The composition of the
resulting mixture closely matched the percentages in the
extracts from intact fruits (Fig. 5C).
Cuticular permeance for water
Cuticular water permeance (vapour-based) of wild-type
and heterozygote fruits were 2.0560.49310±5 m s±1
(transpiration rate: 4.7361.14310±4 g m±2 s±1), respectively, whereas in LeCER6-de®cient plants, cuticular water
loss was about four times larger (Fig. 6A). For analysing
the contribution of the total cuticular waxes, they were
removed by short-term solvent extraction. Permeance for
water increased by a factor of 10 for the wild-type plants
and a by factor of 2.5 for the homozygote mutant (Fig. 6A).
At these higher levels, no signi®cant difference could be
observed between the different lines.
In order to analyse the relevance of the epicuticular wax
to the overall transpiration barrier, the epicuticular layer
was selectively removed with gum arabic. The ®rst
mechanical treatment led to a signi®cant increase of
cuticular water permeance for the wild-type only, but not
The biosynthesis of aliphatic wax constituents, starting
with C18 fatty acid precursors, proceeds via sequential
elongation steps followed by modi®cation of functional
groups (von Wettstein-Knowles, 1993; Hamilton, 1995;
Kolattukudy, 1996; Post-Beittenmiller, 1996; Kunst and
Samuels, 2003).
Transposon mutagenesis offers a suitable tool to
manipulate tomato plants genetically (Emmanuel and
Levy, 2002), targeting genes and gene products relevant
for wax biosynthetic reactions. In the current study, this
reverse genetic approach was employed to identify a
mutant de®cient in a gene with high sequence homology to
Arabidopsis CER6, alias CUT1. The latter had been shown
to encode a b-ketoacyl-CoA synthase that is crucial for
wax biosynthesis (Millar et al., 1999; Fiebig et al., 2000;
Hooker et al., 2002). The enzyme is involved in early
elongation steps of very-long-chain fatty acids in
Arabidopsis, leading to the production of chain lengths
larger than 26 carbons. Based on its high sequence
homology, the corresponding tomato gene was designated
as LeCER6 and chosen for further investigations into the
chemical composition and physiological function of
mutant cuticular waxes. Plants carrying a heterozygous
mutation in LeCER6 did not differ signi®cantly from the
wild-type, neither in the expression level of the gene, nor
in the chemical and functional cuticle characteristics. In
sharp contrast, the gene was not expressed in the
homozygous mutant plants of the cultivar Micro-Tom
(Fig. 1C). Therefore, the mutation in LeCER6 is leading to
a loss of function and is recessive. This loss of LeCER6activity caused a reduction in the n-aliphatics with chain
lengths of 31 carbons and longer, both in leaf and fruit
waxes. The drastic change in chain length patterns clearly
indicates that LeCER6 is involved in chain length elongation rather than functional group modi®cation. It should be
noted that methyl-branched alkanes with chain lengths
beyond C30 were not affected by the mutation, thus
illustrating the substrate speci®city of the LeCER6
condensing enzyme.
Layered structure of tomato cuticular waxes
The chemical composition of tomato fruit cuticular wax
had been the topic of previous investigations (Hunt and
Baker, 1980; Baker et al., 1982). In accordance with these
reports, the current study con®rmed that cuticular waxes of
Micro-Tom fruits and leaves consist predominantly of
very-long chain alkanes and triterpenoids. In addition, in
Cuticular wax composition and transpiration of tomato fruits 1407
an attempt to distinguish between intra- and epicuticular
wax layers within the tomato fruit cuticle, mechanical and
extractive wax-sampling techniques were used sequentially. While consecutive adhesive treatments yielded
rapidly declining amounts of waxes, the following extraction step in all cases yielded relatively high amounts of
wax (Fig. 4). This proves that both techniques probed
different layers within the cuticular wax. The inner layer,
being concealed against adhesive sampling, must be
located within the mechanically resistant polymer matrix
of cutin and can, hence, be interpreted as intracuticular
wax. The outer layer, accessible to adhesive sampling, is
consequently equivalent to the epicuticular wax ®lm.
Because this ®lm was ®rst removed exhaustively, it can be
assumed that the intracuticular wax samples contained
little residual contamination of epicuticular material. In
conclusion, both wax layers were manipulated very
selectively for chemical analysis and later for physiological experiments (see below).
The epi- and intracuticular wax layers on wild-type
Micro-Tom fruits showed largely differing chemical
compositions (Fig. 5A, B). The epicuticular wax ®lm,
with a coverage of approximately 5 mg cm±2, consisted
exclusively of aliphatic compounds. A similar pattern of
aliphatics (c. 9 mg cm±2) was present in the intracuticular
wax compartment, but was mixed with 1.5 mg cm±2 of the
triterpenoid alcohols a-, b-, and d-amyrin. Interestingly,
similar gradients between both wax layers, also with
triterpenoids restricted to the internal part of the cuticle,
had previously been reported for leaves of Prunus
laurocerasus (Jetter et al., 2000). Homozygous mutant
Micro-Tom fruits, de®cient in LeCER6 expression, had
reduced amounts of aliphatic wax constituents with chain
lengths beyond C30. Consequently, both the overall
thickness of the epicuticular ®lm and the relative portion
of aliphatics in the intracuticular compartment were
reduced proportionally.
In the intracuticular layer, surprisingly, increased
extraction yields of pentacyclic triterpenoids (a-, b-, and
d-amyrin) compensated for the reduction in aliphatics. To
test whether this ®nding was due to differential extraction
effects, cuticular membranes were isolated from both wildtype and homozygous mutant fruits, and exhaustively
extracted with hot chloroform. The resulting triterpenoid
yields were similar to those from surface extraction,
con®rming the accumulation of relatively high amounts of
triterpenoids in the mutant cuticle (Fig. 5C). At the
moment, it is not clear whether this increase is due to
enhanced passive diffusion of pre-formed compounds into
the cuticle or to de novo synthesis.
Cuticular permeability of wild-type Micro-Tom fruits
In the present study, wild-type tomato fruits were found to
have a cuticular water permeability of 2.0560.49310±5 m
s±1. This value is slightly smaller than the data from four
previous investigations, ranging from 2.7310±5 m s±1 to
14310±5 m s±1 (SchoÈnherr and Lendzian, 1981; Becker
et al., 1986; Lendzian and Kerstiens, 1991; Schreiber and
Riederer, 1996). The fundamental differences between the
current and reported approaches add only little to the
overall variability of results. Therefore, this relatively
small difference between values clearly illustrates the
usefulness of both methods. The relatively large divergence between these results could, on the one hand, be due
to differences either in the growth conditions or in the fruit
maturation state. However, for tomato fruit cuticles to date
none of these factors has been investigated systematically,
and hence their potential in¯uence cannot be assessed. On
the other hand, the tomato varieties investigated in the
various studies might differ in cuticular permeability. To
test this possibility, several tomato cultivars were compared in the current study. They all showed fruit cuticular
transpiration characteristics very similar to wild-type
Micro-Tom (G Vogg, unpublished data), consequently
making genetic background variation an unlikely explanation for differences of previous results.
Finally, the methods used to characterize cuticular
barrier properties might account for differences in the
published results. It might also be expected that the barrier
properties will be affected by the harsh conditions during
the isolation of cuticular membranes. Accordingly, several
studies showed that storage of isolated cuticles prior to
physiological experiments decreased permeances signi®cantly, supposedly allowing time to heal defects in the wax
barrier (Geyer and SchoÈnherr, 1990; Kirsch et al., 1997).
Contribution of various cuticle substructures to the
transpiration barrier
A number of previous studies had focused on the
correlation between wax characteristics and the transpirational barrier properties of the cuticle. In one approach, the
effect of various growth conditions on the composition of
Citrus leaf cuticles and their water permeability had been
analysed (Geyer and SchoÈnherr, 1990; Riederer and
Schneider, 1990), but a direct composition/function relationship could not be deduced. In Sorghum a negative
correlation between wax load and cuticular transpiration
rate had been reported (Jordan et al., 1984). In the same
plant system, a single-locus mutation with strong negative
impact on cuticle thickness and wax deposition (bloomless) caused a 2.5-fold increase in the epidermal conductance for water vapour (Jenks et al., 1994). However, in
leaves, a signi®cant contribution of residual stomatal
transpiration cannot be excluded (Kerstiens, 1996b;
Muchow and Sinclair, 1989). Even though all these studies
showed a general effect of wax composition on permeability, they failed to delineate the contribution of distinct
components of cuticular waxes to the transpiration barrier.
In the present investigation, both the genetic manipulation of tomato fruit cuticles and the partial removal of
1408 Vogg et al.
wax layers offered the opportunity to study the in¯uence of
wax composition on water barrier properties. For LeCER6de®cient plants cuticular water loss was more than four
times larger than for wild-type fruits (Fig. 6). As this
difference could be abolished by exhaustive extraction of
total cuticular waxes, the altered permeation properties of
the mutant were exclusively due to qualitative and
quantitative changes of the cuticular waxes, caused by
the knock-out of a single gene involved in their biosynthesis.
The epicuticular wax layer of wild-type tomato fruits
was dominated by n-alkanes. When this layer was removed
by adhesive treatment, a signi®cant, but only moderate,
increase of water permeability resulted (Fig. 6B). This
proves that epicuticular waxes contribute to a small extent
to the transpiration barrier. This result is in accordance
with data on sweet cherry fruit surfaces (Knoche et al.,
2000), where the removal of epicuticular waxes led to only
a small increase in conductance compared with the effect
of removal of the epi- plus intracuticular waxes.
Unfortunately, the chemical composition of epi- and
intracuticular waxes was not investigated for that species,
and, hence, a composition±function correlation could not
be deduced.
For lecer6 mutants, an almost complete loss of C31±C34
compounds caused both a drastic reduction in epicuticular
wax quantity and substantial changes in surface composition. Homozygous mutant fruits showed signi®cantly
higher permeability for water than wild-type (and heterozygous) fruits after removal of their epicuticular layer.
The elevated transpiration rates of the mutant fruits
can, therefore, only be explained in part by changes in
the epicuticular waxes. The larger portion of the
transpiration barrier of tomato fruits must be located in
the intracuticular wax layer. Because homozygous and
wild-type fruits had similar quantities of intracuticular
waxes, de®cits in the transpiration barrier must be due to
changes in the wax composition. Based solely on the
chemical results, it can be inferred that the aliphatic
constituents of the intracuticular wax layer, especially their
chain length pro®le, play a pivotal role for fruit water
relations. Cuticular triterpenoids cannot compensate for
the loss of aliphatics.
It has been discussed that the structure of cuticular
waxes is a much more important determinant of permeability than either the wax amounts or the composition of
the wax mixture (Geyer and SchoÈnherr, 1990). Therefore,
from various biophysical and compositional data a model
for the molecular structure of plant cuticular waxes was
developed (Riederer and Schneider, 1990; Reynhardt and
Riederer, 1994). Based on this model the wax barrier
consists of impermeable clusters of crystalline zones
embedded in a matrix of amorphous material. Water
diffusion occurs only in the amorphous volume fractions
while crystallites are inaccessible. Based on X-ray dif-
fraction studies it has been postulated that triterpenoids are
localized exclusively in the amorphous zones (Casado and
Heredia, 1999). Therefore, for Micro-Tom mutant fruits a
reduction in n-aliphatics (=31 carbons), together with an
increase in cyclic triterpenoids, should create amorphous
zones at the expense of crystalline domains. This shift in
overall crystallinity might be the actual reason for the
increase in cuticular transpiration. SchoÈnherr and Riederer
(1988) found that sorption and desorption of lipophilic
compounds differs for the inner and outer surfaces of
isolated cuticular membranes, respectively. This asymmetry in cuticle properties led to the barrier membrane model
(Riederer and Schreiber, 1995), stating that the main
portion of the cuticular transport barrier is located at or
near the outer surface of the cuticle.
The present ®ndings for tomato fruits not only con®rm
the existing models, but also add further aspects to it:
(i) the cuticle is a mechanically and chemically asymmetric layered barrier, (ii) the main portion of the water
barrier is located in the intracuticular wax layer, i.e.
underneath the surface, (iii) the relevant barrier properties
are due to the speci®c ratio of aliphatics and triterpenoids
of the wax mixture. All the arguments taken together, it is
suggested that, at least for tomato fruits, the cuticular
transport barrier consists of crystalline aliphatic domains
that are located asymmetrically in the intracuticular wax
layer.
Conclusions
In conclusion, it could be shown that the tomato system
offers the possibility for an integrative approach to
correlate cuticle biosynthesis, composition, and its function as a transpiration barrier using an intact system. It has
been shown how the mutation of a wax biosynthetic gene
(LeCER6), with a known function in very-long-chain fatty
acid elongation, leads to an alteration of the chemical
cuticular wax composition and, consequently, of the
cuticular water permeability. The epi- and intracuticular
waxes were found to contain a similar composition of
aliphatics. In sharp contrast, triterpenoids were located
exclusively in the intracuticular layer of the tomato fruit
cuticle. Finally, the genetic manipulation of wax composition and the mechanical manipulation (removal) of the
epicuticular wax layer were combined to study the
contribution of cuticle substructures to the overall
transpiration barrier. The main portion of the transpiration
barrier of tomato fruits is located in the aliphatic constituents of the intracuticular wax layer, but modi®ed by the
presence of variable amounts of triterpenoids. This result
substantiates models that describe the cuticular transport
barrier as a thin layer, consisting of crystalline aliphatic
domains, and located a short distance underneath the
cuticle surface.
Cuticular wax composition and transpiration of tomato fruits 1409
Acknowledgements
This project was ®nancially supported by the German Research
Foundation (VO 934-1), the Sonderforschungsbereich 567, and the
Fonds der Chemischen Industrie. The authors are grateful to Stefan
Bauer (University of Muenster, Institute for Food Chemistry) for
identi®cation of d-amyrin by comparison with their reference
standard, Bianka Pink for support in analytical questions, and Julia
Blass-Warmuth and Alex Gall for excellent technical assistance.
References
Baker EA, Bukovac J, Hunt GM. 1982. Composition of tomato
fruit cuticle as related to fruit growth and development. In: Cutler
DF, Alvin KL, Price CE, eds. The plant cuticle. London:
Academic Press, 33±44.
Bauer S, Schulte E, Thier HP. 2002. Untersuchung des
Ober¯aÈchenwachses von Tomaten. Lebensmittelchemie 56, 13.
Becker M, Kerstiens G, SchoÈnherr J. 1986. Water permeability of
plant cuticles: permeance, diffusion and partition coef®cients.
TreesÐstructure and function 1, 54±60.
Bernatzky R, Tanksley SD. 1986. Methods for detection of single
and low copy sequences in tomato on Southern blots. Plant
Molecular Biology Reporter 4, 37.
Casado CG, Heredia A. 1999. Structure and dynamics of
reconstituted cuticular waxes of grape berry cuticle (Vitis
vinifera L.). Journal of Experimental Botany 50, 175±182.
Chen X, Goodwin SM, Boroff VL, Liu X, Jenks MA. 2003.
Cloning and characterization of the wax2 gene of Arabidopsis
involved in cuticle membrane and wax production. The Plant Cell
15, 1170±1185.
Emmanuel E, Levy AA. 2002. Tomato mutants as tools for
functional genomics. Current Opinion in Plant Biology 5, 112±
117.
Fiebig A, May®eld JA, Miley NL, Chau S, Fischer RL, Preuss D.
2000. Alterations in CER6, a gene identical to CUT1,
differentially affect long-chain lipid content on the surface of
pollen and stems. The Plant Cell 12, 2001±2008.
Geyer U, SchoÈnherr J. 1990. The effect of the environment on the
permeability and composition of Citrus leaf cuticles. I. Water
permeability of isolated cuticular membranes. Planta 180, 147±
153.
Gorbunova V, Levy AA. 1997. Circularized Ac/Ds transposons:
formation structure and fate. Genetics 134, 571±584.
Hamilton RJ. 1995. Analysis of waxes. In: Hamilton RJ, ed.
Waxes: chemistry, molecular biology and functions, Vol. 8.
Dundee: The Oily Press, 311±342.
Hannoufa A, McNevin J, Lemieux B. 1993. Epicuticular waxes of
eceriferum mutants of Arabidopsis thaliana. Phytochemistry 33,
851±855.
Hooker TS, Millar AA, Kunst L. 2002. Signi®cance of the
expression of the CER6 condensing enzyme for cuticular wax
production in Arabidopsis. Plant Physiology 129, 1568±1580.
Hunt GM, Baker EA. 1980. Phenolic constituents of tomato fruit
cuticles. Phytochemistry 19, 1415±1419.
Jenks MA, Joly RJ, Peters PJ, Rich PJ, Axtell JD, Ashworth
EN. 1994. Chemically-induced cuticle mutation affecting
epidermal conductance to water vapour and disease
susceptibility in Sorghum bicolor (L.) Moench. Plant
Physiology 105, 1239±1245.
Jenks MA, Tuttle HA, Feldmann KA. 1995. Changes in
epicuticular waxes on wild-type and eceriferum mutants in
Arabidopsis during development. Phytochemistry 42, 29±34.
Jetter R, SchaÈffer S. 2001. Chemical composition of the Prunus
laurocerasus leaf surface. Dynamic changes of the epicuticular
wax ®lm during leaf development. Plant Physiology 126, 1725±
1737.
Jetter R, SchaÈffer S, Riederer M. 2000. Leaf cuticular waxes are
arranged in chemically and mechanically distinct layers: evidence
from Prunus laurocerasus L. Plant, Cell and Environment 23,
619±628.
Jordan WR, Shouse PJ, Blum A, Miller FR, Monk RL. 1984.
Environmental physiology of Sorghum. II. Epicuticular wax load
and cuticular transpiration. Crop Science 24, 1168±1173.
Kerstiens G. 1996a. Plant cuticles: an integrated functional
approach. Oxford: Bios Scienti®c Publishers Ltd.
Kerstiens G. 1996b. Cuticular water permeability and its
physiological signi®cance. Journal of Experimental Botany 47,
1813±1832.
Kirsch T, Kaffarnik F, Riederer M, Schreiber L. 1997. Cuticular
permeability of the three tree species Prunus laurocerasus L.
Ginkgo biloba L., and Juglans regia L.: comparative
investigation of the transport properties of intact leaves, isolated
cuticles and reconstituted cuticular waxes. Journal of
Experimental Botany 48, 1035±1045.
Knoche M, Peschel S, Hinz M, Bukovac MJ. 2000. Studies on
water transport through the sweet cherry fruit surface:
characterization conductance of the cuticular membrane using
pericarp segments. Planta 212, 127±135.
Kolattukudy PE. 1996. Biosynthetic pathways of cutin and waxes,
and their sensitivity to environmental stresses. In: Kerstiens G,
ed. Plant cuticles: an integrated functional approach. Oxford:
Bios Scienti®c Publishers, 83±108.
Koornneef M, Hanhart CJ, Thiel F. 1989. A genetic and
phenotypic description of eceriferum (cer) mutants in
Arabidopsis thaliana. Journal of Heredity 80, 118±122.
Kunst L, Samuels AL. 2003. Biosynthesis and secretion of plant
cuticular wax. Progress in Lipid Research 42, 51±80.
Lendzian K, Kerstiens G. 1991. Sorption and transport of gases
and vapors in plant cuticles. Reviews of Environmental
Contamination and Toxicology 121, 65±128.
Lemieux B. 1996. Molecular genetics of epicuticular wax
biosynthesis. Trends in Plant Science 1, 312±318.
Mathur J, Szabados L, Schaefer S, Grunenberg B, Lossow A,
Jonas-Straube E, Schell J, Koncz C, Koncz-Kalman Z. 1998.
Gene identi®cation with sequenced T-DNA tags generated by
transformation of Arabidopsis cell suspension. The Plant Journal
13, 707±716.
McNevin JP, Woodward W, Hannoufa A, Feldmann KA,
Lemieux B. 1993. Isolation and characterization of eceriferum
(cer) mutants induced by T-DNA insertions in Arabidopsis
thaliana. Genome 36, 610±618.
Meissner R, Chaque V, Zhu Q, Emmanuel E, Elkind Y, Levy
AA. 2000. A high throughput system for transposon tagging and
promoter trapping in tomato. The Plant Journal 22, 265±274.
Meissner R, Jacobson Y, Melamed S, Levyatuv S, Shalev G,
Ashri A, Elkind Y, Levy AA. 1997. A new model system for
tomato genetics. The Plant Journal 12, 1465±1472.
Millar AA, Clemens S, Zachgo S, Giblin EM, Taylor DC, Kunst
L. 1999. CUT1, an Arabidopsis gene required for cuticular wax
biosynthesis and pollen fertility, encodes a very-long-chain fatty
acid condensing enzyme. The Plant Cell 11, 825±838.
Muchow RC, Sinclair TR. 1989. Epidermal conductance, stomatal
density and stomatal size among genotypes of Sorghum bicolor
(L.) Moench. Plant, Cell and Environment 12, 425±431.
Post-Beittenmiller D. 1996. Biochemistry and molecular biology
of wax production in plants. Annual Review of Plant Physiology
and Plant Molecular Biology 47, 405±430.
Reynhardt EC, Riederer M. 1994. Structures and molecular
dynamics of plant waxes. II. Cuticular waxes from leaves of
1410 Vogg et al.
Fagus sylvatica L. and Hordeum vulgare L. European
Biophysical Journal 23, 59±70.
Riederer M, Schneider G. 1990. The effect of the environment on
the permeability and composition of Citrus leaf cuticles. II.
Composition of soluble cuticular lipids and correlation with
transport properties. Planta 180, 154±165.
Riederer M, Schreiber L. 1995. Waxes: the transport barriers of
plant cuticles. In: Hamilton RJ ed. Waxes: chemistry, molecular
biology and functions. Dundee, Scotland: The Oily Press, 130±
156.
Riederer M, Schreiber L. 2001. Protecting against water loss:
analysis of the barrier properties of plant cuticles. Journal of
Experimental Botany 52, 2023±2032.
SchoÈnherr J, Lendzian K. 1981. A simple and inexpensive method
of measuring water permeability of isolated plant cuticular
membranes. Zeitschrift fuÈr P¯anzenphysiologie 102, 321±327.
SchoÈnherr J, Riederer M. 1986. Plant cuticles sorb lipophilic
compounds during enzymatic isolation. Plant, Cell and
Environment 9, 459±466.
SchoÈnherr J, Riederer M. 1988. Desorption of chemicals from
plant cuticles: evidence for assymetrya. Archives of
Environmental Contamination and Toxicology 17, 13±19.
Schreiber L, Riederer M. 1996. Ecophysiology of cuticular
transpiration: comparative investigation of cuticular water
permeability of plant species from different habitats. Oecologia
(Berlin) 107, 426±432.
von Wettstein-Knowles PM. 1993. Waxes, cutin and suberin. In:
Moore TS, ed. Lipid metabolism in plants, Chapter 4. Boca
Raton: CRC Press, 127±166.