Journal of Experimental Botany, Vol. 57, No. 11, pp. 2515–2523, 2006
doi:10.1093/jxb/erj173 Advance Access publication 1 August, 2006
FOCUS PAPER
Review of sorption and diffusion of lipophilic molecules in
cuticular waxes and the effects of accelerators on solute
mobilities
Lukas Schreiber*
Institute of Cellular and Molecular Botany (IZMB), University of Bonn, Kirschallee 1, D-53115 Bonn, Germany
Received 18 November 2005; Accepted 27 February 2006
Abstract
Many agrochemicals are applied to the leaf surfaces of
crop plants. Systemic chemicals have to penetrate
through the cuticle, which forms an effective transport
barrier. The barrier properties of cuticles are mainly
due to the cuticular waxes deposited as partially crystalline aggregates on the outer surfaces of leaves.
Substances increasing the mobilities of agrochemicals
in cuticular waxes are called accelerators and it is
shown that they act as plasticizers when absorbed by
cuticular waxes. They decrease the barrier properties
of the waxes and thus increase the mobilities of the
agrochemicals through them. In order to analyse the
efficiency of different accelerators, the sorption and
mobility of both agrochemicals and accelerators within
cuticular waxes was measured. Such information was
used to establish correlations between the internal
concentrations of accelerators and their mobilityenhancing effects on agrochemicals in the cuticle.
This, in turn, allowed the determination and comparison of the intrinsic effects of different accelerators and
to rationalize the effect of accelerators on the cuticular
permeability of agrochemicals. Results describing the
sorption (partition coefficients) and mobility (diffusion
coefficients) of lipophilic organic molecules in reconstituted cuticular waxes from different plant species,
and the effect of two different classes of accelerators
(alcohol ethoxylates and n-alkyl esters), on the mobility
of organic molecules are presented and discussed.
Key words: Accelerator, alcohol ethoxylate, diffusion coefficient, leaf surface, partition coefficient, plant cuticle, plasticizer,
reconstituted cuticular wax.
Introduction
Cuticular waxes of plants represent a complex mixture of
pentacyclic triterpenoids and linear long-chain aliphatic
compounds with different chain lengths and different
functions (Jenks and Ashworth, 1999). Pentacyclic triterpenoids can form the major fraction of cuticular waxes
within certain plant families amounting to 60% or more,
whereas cuticular waxes in other plant families are predominantly composed of linear aliphatic compounds with
none or only traces of triterpenoids present.
Cuticular waxes are synthesized in epidermal cells
(Kunst and Samuels, 2003) and deposited within (intracuticular waxes) and on the surface (epicuticular waxes) of
the cutin polymer membrane covering the outer epidermal
cell walls of stems, leaves, and fruits. At room temperature,
cuticular waxes form solid, partially crystalline aggregates
(Reynhardt, 1997) with their melting points normally being
higher than 90 8C (Sitte and Rennier, 1963). Thus, cuticular
waxes help to seal the amorphous cutin polymer and
they significantly contribute to increase cuticular barrier
properties (Riederer and Schreiber, 1995). Extraction of
cuticular waxes with organic solvents increases cuticular
permeability by factors varying from 10 to 1000, which
demonstrates the significance of cuticular waxes forming
the transport-limiting barrier of the plant cuticle (Schönherr
and Riederer, 1989).
* E-mail: [email protected]
Abbreviations: AE, alcohol ethoxylate; C12E8, octaethylene glycol monododecyl ether; C18AC, octadecanoic acid; C24AC, tetracosanoic acid; D, diffusion
coefficient; ESR, electron spin resonance; Kww, wax/water partition coefficient; Kcw, cuticle/water partition coefficient; NMR, nuclear magnetic resonance;
PCP, pentachlorophenol; PLS, phospholipid suspension.
ª The Author [2006]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved.
For Permissions, please e-mail: [email protected]
2516 Schreiber
Whereas ionic polar compounds apparently penetrate the
plant cuticle via aqueous polar pores (Schönherr, 2000,
2006; Schreiber, 2005), non-polar lipophilic compounds
diffuse across leaf surfaces via the lipophilic domains in the
cutin and wax (Buchholz and Schönherr, 2000; Buchholz,
2006). Since permeability of plant cuticles for lipophilic
compounds is largely determined by cuticular waxes, the
permeability of cuticular waxes has to be analysed in order
to understand the barrier properties of leaf structures. This
includes measurements of sorption into and diffusion by
penetrating compounds in cuticular waxes (Schreiber and
Schönherr, 1992, 1993). Furthermore, agrochemicals are
usually applied in the presence of adjuvants (Kirkwood,
1999). They improve the efficacy of agrochemicals by
increasing their solubility in the spray solution and by
improving adhesion and wetting of leaf surfaces by the
droplets. Furthermore, adjuvants have been shown to act
as plasticizers of cuticular waxes and, as a consequence,
the rates of cuticular penetration by non-polar lipophilic
compounds were increased (Schreiber, 1995). Compounds
with these plasticizing properties have been called accelerators (Schönherr et al., 2001). In order to understand the
mode of action of accelerators on cuticular penetration, and
to compare the activity of different accelerators, their
solubilities as well as mobilities in cuticular wax need to
be determined. The sorption and diffusion in cuticular wax
of lipophilic non-electrolytes, as well as accelerators, will
be presented and the effect of such accelerators relating to
increased rates of cuticular penetration will be discussed.
Sorption in reconstituted wax
Sorption of lipophilic non-electrolytes has been measured
using reconstituted cuticular wax (Schreiber and Schönherr,
1992; Kirsch et al., 1997). Concentrations of the compounds (masscompound per massmedium) in the wax and in the
external donor solution (masscompound per masssolution) were
measured (either by counting radioactivity in the case of
labelled substances or by chemical analysis) and wax/
water partition coefficients (Kww) were calculated from
equation 1.
Kww =
masscompound
masswax
masscompound
masswater
ð1Þ
Wax/water partition coefficients are a measure of the
lipophilicity of the compounds and indicate the degree of
solubility of a compound in the lipophilic cuticular wax.
Wax/water partition coefficients of a series of different
non-electrolytes measured in reconstituted cuticular wax,
isolated from a series of different species, varied between
3.5 and 3500 (Table 1). Variation among the different
compounds was much larger than variation among the
waxes isolated from different species. In similar approaches, cuticle/water partition coefficients (Kcw; Riederer
and Schönherr, 1984) and octanol/water partition coefficients (Sangster, 1997) have been determined in the past. It
was shown that Kcw was very similar numerically to Kow
(Schönherr and Riederer, 1989). This was not true for Kww,
since Kww was generally found to be significantly smaller
than Kcw or Kow by factors between 2 and 10 (Fig. 1;
Schreiber and Schönherr, 1992; Kirsch et al., 1997). This
can be attributed to the fact that cuticular wax, forming
solid and partially crystalline aggregates at room temperature, offers fewer sorption sites compared with the
amorphous cutin polymer or solvents like octanol.
Sorption of different accelerators (alcohol ethoxylates
and n-alkyl esters) into cuticular wax of Hordeum vulgare
and Stephanotis floribunda was also determined (Table 2;
Simanova et al., 2005; Schreiber et al., 1996b; Burghardt
et al., 1998). As will be shown later, this information is
important in transport experiments, since it allows calculation of the amount of accelerator absorbed by the transportlimiting wax barrier in equilibrium with an external donor
Table 1. Wax/water partition coefficients (Kww) of organic molecules in reconstituted cuticular wax from Hordeum vulgare, Prunus
laurocerasus, Ginkgo biloba, and Juglans regia
Kww
H. vulgare
Metribuzin (MET)
4-Nitrophenol (4-NP)
Benzoic acid (BA)
Atrazine (AT)
Salicylic acid (SA)
Triadimenol (TRI)
2,4-Dichlorophenoxyacetic acid (2,4-D)
Tebuconazole (TB)
Bitertanol (BIT)
Pentachlorophenol (PCP)
a
From Burghardt et al. (1998).
From Schreiber and Schönherr (1992).
c
From Kirsch et al. (1997).
b
a
3.5
30a
32.3b
45.7b
640a
1047b
3548b
P. laurocerasus
c
12.8
14.0c
21.1c
28.2c
46.2c
136.3c
G. biloba
J. regia
22.5c
47.2c
139.4c
21.8c
47.6c
144.3c
Molecules in cuticular waxes and the effects of accelerators
2517
Table 2. Wax/water partition coefficients (Kww) of accelerators
(alcohol ethoxylates and n-alkyl esters) in reconstituted cuticular wax from Hordeum vulgare and Stephanotis floribunda
With n-alkyl esters, external donor solutions contained 10% 1,2-propanediol
instead of pure water, in order to make the n-alkyl esters soluble in an
aqueous solution.
Kww
H. vulgare
Fig. 1. Correlation between wax/water partition coefficients (Kww) and
cuticle/water partition coefficients (Kcw). Kww values for Prunus
laurocerasus, Ginkgo biloba, and Juglans regia and for the compounds
metribuzin, 4-nitrophenol, benzoic acid, atrazine, salicylic acid, and 2,4dichlorophenoxyacetic acid were taken from Table 1. Kcw values were
taken from Kirsch et al. (1997). Coefficient of determination r2=0.99.
solution. The Kww values in H. vulgare were, on average,
10 times smaller than in S. floribunda (Table 2). This can be
explained by the fact that H. vulgare wax is a homogeneous and highly crystalline wax, with nearly 90% primary
alcohols, offering significantly fewer sorption sites, which
is not the case for S. floribunda, which was characterized by
a much more heterogeneous wax composition (Simanova
et al., 2005). Nevertheless, more detailed studies trying to
relate measured partition coefficients with the chemical
composition of cuticular waxes have not been carried out.
Diffusion in reconstituted wax
Mobility of lipophilic non-electrolytes in isolated and
reconstituted cuticular wax has been measured using a
similar approach. Either a radiolabelled probe was reconstituted together with the wax, or reconstituted wax was
loaded with a radiolabelled probe from an external solution
(Schreiber and Schönherr, 1993). Subsequently, the radiolabelled probes were desorbed from the wax samples and
diffusion coefficients D (m2 s1) were calculated from
desorption kinetics by plotting the relative amounts desorbed (Mt/M0) versus the square root of time t1/2 using
equation 2 (Felder and Huvard, 1980):
rffiffiffiffi
Mt
4
D pffi
=
3
3 t
ð2Þ
p
M0 Dx
Dx (m) represents the thickness of the wax layer. D can be
calculated from the slopes of the linearized desorption
kinetics, which are proportional to (4/Dx)(D/p)1/2 (Felder
and Huvard, 1980).
Diffusion coefficients have been measured for different
lipophilic non-electrolytes (Table 3) and linear, long-
Diethylene glycol monobutyl
ether (C4E2)
Triethylene glycol monohexyl
ether (C6E3)
Tetraethylene glycol mono-octyl
ether (C8E4)
Pentaethylene glycol monodecyl
ether (C10E5)
Octaethylene glycol monodecyl
ether (C10E8)
Diethylene glycol monododecyl
ether (C12E2)
Triethylene glycol monododecyl
ether (C12E3)
Tetraethylene glycol monododecyl
ether (C12E4)
Pentaethylene glycol monododecyl
ether (C12E5)
Hexaethylene glycol monododecyl
ether (C12E6)
Heptaethylene glycol monododecyl
ether (C12E7)
Octaethylene glycol monododecyl
ether (C12E8)
Heptaethylene glycol monotetradecyl
ether (C14E7)
Octaethylene glycol monotetradecyl
ether (C14E8)
Octaethylene glycol monohexadecyl
ether (C16E8)
Dimethyl suberate (DMSU)
Diethyl suberate (DESU)
Diethyl sebacate (DES)
Dibutyl suberate (DBSU)
Dibutyl sebacate (DBS)
a
S. floribunda
0.083
a
0.83
4.9a
a
35.7
5.6b
b
2000
21125c
1300b
b
670
7909c
400b
4439c
237a
2626c
160b
a
962c
104
1561a
b
1200
12350a
c
3.1
5.9c
31.2c
503c
1821c
15.5c
55.7c
447c
5319c
28590c
a
From Schreiber et al. (1996b).
From Burghardt et al. (1998).
c
From Simanova et al. (2005).
b
chain, aliphatic molecules including one cyclic compound
(Table 4) in a series of cuticular waxes from different
species. Lipophilic molecules, representing typical pesticide molecules, had diffusion coefficients varying between
1018 and 1017 m2 s1 (Table 3). These are extraordinarily
low values. Comparable compounds of similar size and
similar physico-chemical properties have diffusion coefficients of about 1010 m2 s1 in solution (Cussler, 1984) and
about 1012 m2 s1 in cell membranes (Stein, 1967). This
is evidence that cuticular waxes forming the transportlimiting barrier of plant cuticles must be highly structured
and partially crystalline on a molecular level, leading to
these low diffusion coefficients. Diffusion coefficients of
typical pesticide molecules were not different when wax
was reconstituted together with the radiolabelled probe or
2518 Schreiber
Table 3. Diffusion coefficients (D) of organic molecules and one accelerator molecule (octaethylene glycol monododecyl ether) in
reconstituted cuticular waxes from Hordeum vulgare, Prunus laurocerasus, Ginkgo biloba and Juglans regia
D31017 (m2 s1)
H. vulgare
Pentachlorophenol (PCP)
Benzoic acid (BA)
4-Nitrophenol (4-NP)
Salicylic acid (SA)
2,4-Dichlorophenoxyacetic acid (2,4-D)
Atrazine (AT)
Metribuzin (MET)
Lindane (LI)
Tebuconazole (TB)
Triadimenol (TRI)
Bitertanol (BIT)
Octaethylene glycol monododecyl ether (C12E8)
a
1.96
3.34a
2.4b
1.63a
1.2b
0.555a
0.5b
0.409a
0.151a
0.89c
P. laurocerasus
G. biloba
J. regia
3.54d
3.23d
2.74d
1.63d
1.15d
0.91d
4.37d
3.21d
2.01d
5.59d
3.45d
2.22d
a
From Schreiber and Schönherr (1993).
From Burghardt et al. (1998).
c
From Schreiber (1995).
d
From Kirsch et al. (1997).
b
Table 4. Diffusion coefficients (D) of linear, long-chain, aliphatic molecules and cholesterol in reconstituted cuticular wax from
Hordeum vulgare, Fagus sylvatica, Picea abies and Vitis vinifera
D31017 (m2 s1)
H. vulgare
F. sylvatica
P. abies
V. vinifera
Hexadecane (C16AN)
Octadecane (C18AN)
Octacosane (C28AN)
Dotriacontane (C32AN)
1-Hexadecanol (C16OL)
Tetradecanoic acid (C14AC)
Hexadecanoic acid (C16AC)
Octadecanoic acid (C18AC)
Octadeca-9-enoic acid (C18:1AC)
Octadeca-9,12-dienoic acid (C18:2AC)
Octadeca-9,12,15-trienoic acid (C18:3AC)
Eicosa-5,8,11,14-tetraenoic acid (C20:4AC)
Tetracosanoic acid (C24AC)
Cholesterol (CHOL)
0.0000407a
1.51a
0.381a
0.109a
0.357a
0.791a
0.99a
0.00122a
–
0.099b
0.022b
0.00080b
0.000019b
3.35b
1.15b
0.78b
2.77b
2.83b
3.7b
4.5b
0.016b
0.043b
0.021b
0.00031b
0.000035b
1.51b
0.57b
0.47b
0.84b
0.96b
1.32b
0.92b
0.021b
0.0003c
a
b
c
From Schreiber and Schönherr (1993).
From Schreiber et al. (1996a).
From Casado and Heredia (1999).
after external loading of the wax (Schreiber and Schönherr,
1993).
However, diffusion coefficients were significantly different between the two methods of loading the wax with the
radiolabelled probe when using linear, long-chain, aliphatic
molecules. Dotriacontane (Schreiber et al., 1997) and
tetracosanoic acid (Schreiber, 1995) had significantly lower
diffusion coefficients in H. vulgare wax when they were
reconstituted directly with wax. This observation fits the
model where cuticular waxes are viewed as a composite
transport barrier system with amorphous and crystalline
domains (Reynhardt and Riederer, 1991). Sorption and
diffusion of non-wax molecules is supposed to take place in
the amorphous wax phase. Linear, long-chain aliphatic
molecules of C24–C34 chain length (typical crystalline wax
molecules) and the cyclic compound cholesterol (as a model
for triterpenoids) had the lowest diffusion coefficients
(Table 4). This indicates that, although these molecules
are ‘trapped’ within the crystalline wax phase, they are not
completely immobile.
When the logarithms of measured diffusion coefficients
D (m2 s1) were plotted as a function of the molar volumes
MV (mol cm3) of the respective molecules, linear relationships were obtained (Fig. 2). This negative exponential
relationship between D and MV can be analysed applying
the following model (Potts and Guy, 1992).
D = D0 ebMV
ð3Þ
D0 (m2 s1) represent the diffusion coefficient of a molecule
having a hypothetical MV of zero and b (mol cm3)
represents the size selectivity characterizing the dependence of D on MV. Both parameters, D0 and b, in equation
Molecules in cuticular waxes and the effects of accelerators
2519
Table 5. Size selectivities (b) and D0 in reconstituted cuticular
wax from Prunus laurocerasus, Ginkgo biloba, Juglans regia,
Fagus sylvatica, Picea abies, and Hordeum vulgare
P. laurocerasus
G. biloba
J. regia
F. sylvatica (acids)
F. sylvatica (alkanes)
P. abies (acids)
P. abies (alkanes)
H. vulgare (PLS)
H. vulgare (C12E8)
H. vulgare (C8E4)
a
b
c
Fig. 2. Correlation between diffusion coefficients (D) and molar
volumes (MV). D values for Prunus laurocerasus, Ginkgo biloba, and
Juglans regia and for the compounds metribuzin, 4-nitrophenol, benzoic
acid, atrazine, salicylic acid, and 2,4-dichlorophenoxyacetic acid were
taken from Table 3. MV values were calculated according to Abraham
and McGowan (1987). Coefficient of determination r2=0.97.
3 have been related to the two different wax zones. It was
argued that b should reflect the amorphous wax phase
where diffusion takes place, whereas D0 was related to the
crystalline wax phase forming an exclusion volume for
diffusion (Schreiber et al., 1996a). A decrease in D0 reflects
an increase in crystallinity of the wax, and a decrease of b
indicates an increase in size selectivity. Both parameters
have been measured in a series of different plant waxes
(Schreiber et al., 1996a; Kirsch et al., 1997) and in isolated
plant cuticles (Buchholz et al., 1998). Size selectivities
measured in different wax samples varied between 0.016
and 0.037 (Table 5). In isolated cuticular membranes they
ranged from 0.017 to 0.044 (Schönherr and Baur,
1994). This is good evidence that the amorphous phase in
reconstituted wax and in the cutin polymer must have
a similar molecular composition.
Good linearities between D and MV can also be obtained
when the logarithm of D is plotted versus the logarithm
of MV (Schreiber and Schönherr, 1993). These doublelogarithmic plots are characterized by negative slopes
varying between 2.9 (cyclic organic compounds) and
15.7 (linear aliphatic compounds). Based on these
double-logarithmic plots, it was suggested (Fowler, 1999)
that a modified form (equation 4) of the Stokes–Einstein
equation be used to model the diffusion of organic
molecules in cuticular wax:
D=
/
kT
3
s 6pgr
ð4Þ
In addition to the terms D [diffusion coefficient (m2 s1)], k
[Boltzmann constant (J mol1 K1)], T [absolute temperature (K)], g [viscosity (Pa s1)], and r [radius of the
molecules (m)], known from the Stokes–Einstein equation
b (mol1 cm3)
D031016 (m2 s1)
0.017a
0.016a
0.019a
0.036b
0.037b
0.035b
0.036b
0.015c
0.0052c
0.0025c
1.62a
2.14a
1.49a
622b
53.9b
89.5b
6.68b
1.15c
1.41c
3.13c
From Kirsch et al. (1997).
From Schreiber et al. (1996a).
From Burghardt et al. (1998).
describing the diffusion of molecules in liquids, the ratio //
s is added as a new term (Fowler, 1999). This ratio should
account for the fact that diffusion is taking place in solid
wax, instead of a liquid, and characterized as a tortuous path
in a solid phase (Fowler, 1999). The term / is called
voidage and it represents the free volume available for
diffusion, whereas the term s represents the tortuosity of the
diffusional path around wax crystallites. The great advantage of this model is the fact that the process of diffusion
within wax could be reduced to simple physical parameters
such as viscosity, tortuosity, and voidage. However, this
model still awaits its validation for describing the diffusion
of lipophilic molecules in reconstituted cuticular wax.
Compared with approaches investigating the mobility of
non-electrolytes in reconstituted wax, much less effort has
been made to investigate the mobility of accelerators,
probably due to the fact that most of them are not available
as radiolabelled compounds. The only data available is the
diffusion coefficient of 14C-labelled C12E8 (octaethylene
glycol monododecyl ether) in reconstituted H. vulgare wax
(Schreiber, 1995). The diffusion coefficient for C12E8 was
8.531018 m2 s1, which is similar to that of other nonelectrolytes (Table 3). However, when the molar volume of
C12E8 is considered, the diffusion coefficient is far too high.
Using the regression equation of Fig. 2 and a molar volume
of 458 cm3 mol1 for C12E8, the diffusion coefficients
should be 120-times lower than measured. This discrepancy
shows that C12E8 acts as an accelerator for its own
diffusion, as well as on the diffusion of other molecules,
by ‘plasticizing’ the wax phase where diffusion takes place.
Up to now only limited information has been available
on the extent to which diffusion in reconstituted wax
depends on the chemical composition of plant waxes. Some
insight was obtained using an artificial wax mixture of pure
alkanes with a defined mean chain and standard deviation
(C28.3363.04), to which other wax molecules with functional groups, normally found as characteristic components
of cuticular wax, were added in increasing amounts
2520 Schreiber
(Kirsch, 1996). Radiolabelled 14C-labelled octadecanoic
acid (C18AC) was used as probe. Addition of increasing
amounts of a C28 alcohol or C28 acid to this mixture of
alkanes led to a steep increase in the mobility of C18AC,
whereas at higher concentrations D decreased again
(Fig. 3). However, diffusion of C18AC was significantly
decreased, when a C44-ester or the triterpenoid oleanoic
acid, were added to the wax mixture (Fig. 3). A possible
interpretation of these results could be the following. At
low amounts added, the alcohol, and to a significantly
larger degree the acid, disturbed the physical structure of
the alkane wax, leading to an increase in D, whereas at
higher amounts (40% and above) these compounds increased crystallinity, eventually forming separated crystalline phases as has been described for the formation of
epicuticular wax crystallites in vitro (Jetter, 1993). The long
chain C44 ester and the pentacyclic triterpenoid probably do
not complement the alkane wax with a mean chain length of
C28, and it is argued that they from a disparate wax phase
from the very beginning, leading to an increase in overall
crystallinity of the wax. However, further experimental data
are needed to establish a convincing relationship between D
and wax chemistry.
Effects of accelerators on diffusion in
reconstituted wax
In the preceding section it has been shown that both
sorption and diffusion of lipophilic non-electrolytes, as well
as accelerators, can be measured in reconstituted cuticular
wax. Analysing the effects of accelerators on the mobility
of pesticides in cuticular wax has also been achieved, by
Fig. 3. Effect (D+add/Dadd) of adding increasing amounts of pure
compounds (octacosane (C28AN), octacosanol (C28OL), octacosanoic
acid (C28AC), docosanoic acid docosyl ester (C44EST), and oleanoic acid
(OleaAC) to a synthetic wax mixture [mixture composed of alkanes with
a mean chain length of C2863 (standard deviation)] on the diffusion
coefficient (Dadd) of 14C-labelled C18AC in that synthetic alkane wax.
Values were calculated by dividing the measured diffusion coefficients
(D+add) after the addition of the different compounds by the diffusion
coefficients (Dadd) prior to addition. Data from Kirsch (1996).
desorbing 14C-labelled compounds using solutions with the
accelerators at well-defined concentrations in the desorption media. Desorption media containing inert phospholipid
suspensions instead of the accelerators served as controls.
Diffusion coefficients were measured in the presence and
the absence of the accelerators and effects of the accelerators were calculated from equation 5:
effect =
Daccelerator
Dcontrol
ð5Þ
In this type of experiment care had to be taken that the
ratio between the amount of the external accelerators
solution (5–25 cm3) and the sorption capacity of the
reconstituted wax (104 cm3) is large, since accelerators
are absorbed by the wax at the beginning of the experiment
until equilibrium is reached. If the external reservoir is
large, fairly small amounts of accelerators absorbed by the
wax do not lead to a measurable decrease of the external
concentration. Thus, concentrations of accelerators in wax
can easily be calculated by multiplying the external
concentrations by their respective partition coefficients.
Using this approach experiments have been carried out
with reconstituted wax of H. vulgare and S. floribunda.
Accelerators are either homologous series of alcohol
ethoxylates (Schreiber et al., 1996b; Burghardt et al.,
1998) or n-alkyl esters (Simanova et al., 2005). The largest
data set exists with H. vulgare wax and alcohol ethoxylates.
In a very detailed approach, fundamental rules of accelerator/wax interaction have been worked out (Schreiber,
1995), using H. vulgare wax, octaethylene glycol monododecyl ether (C12E8) as accelerator molecule, pentachlorophenol (PCP) as a lipophilic organic molecule, and
tetracosanoic acid (C24AC) as a linear long-chain aliphatic
molecule. Classic dose–response curves were obtained, as
the effects of the accelerator on the diffusion of PCP and
C24AC increased linearly with the logarithm of C12E8
concentrations in the wax. At the same internal concentrations a much larger effect was obtained for C24AC (40fold increase of D) which has a much lower initial diffusion
coefficient (1.3231020 m2 s1) compared with PCP (4fold increase of D) with a much higher initial diffusion
coefficients (1.8331018 m2 s1). Most interestingly, the
effects of the accelerators on D were completely reversible
within 10 min of replacing the desorption medium containing C12E8 by the inert PLS solution. This led to the
conclusion that the enhancing effect of accelerators in
the cuticular wax barrier was due to a plasticizing effect of
the amorphous fraction of the wax leading to an increase
of D, whereas solubilization of the crystalline domains did
not occur.
In a further set of experiments, a homologous series of
AEs ranging from C4E2 through C6E3, C8E4, C10E5, C12E6,
C14E7 to C16E8 (for full names see Table 2) was investigated (Schreiber et al., 1996b). The average effects of
all AEs tested were again about 10-fold larger with C24AC
Molecules in cuticular waxes and the effects of accelerators
compared with PCP. When the effects on PCP or C24AC
were plotted as a function of the different internal concentrations of the respective accelerators in the wax, a linear
relationship was obtained (Fig. 4). This showed that there
were no differences in intrinsic effects among the different
AEs tested. The efficiency of increasing D by each single
AE on either PCP or C24AC was solely a function of its
internal concentration and not of the specific chemical
composition of the respective accelerator. It can be
concluded from this that the increase in D is caused by
a non-specific plasticizing effect by the accelerator molecules on the amorphous wax phase through which diffusion
takes place.
These findings were confirmed in a further set of
experiments analysing either the effect of two selected
AEs (C8E4 and C12E8) on the diffusion of six pesticide
molecules (salicylic acid, 2,4-D, metribuzin, triadimenol,
tebuconazole, and bitertanol) or the effect of 15 AEs (C4E2
through C6E3, C8E4, C10E5, C12E6, C14E7 to C16E8) on the
diffusion of only bitertanol (Burghardt et al., 1998). A
linear relationship was again obtained when the effects of
the AEs on the diffusion of bitertanol were related to the
internal concentrations of the accelerators. Furthermore,
those effects were again larger with molecules having lower
initial Ds. Analysing the effect of surfactants on D of 14Clabelled organic molecules, by applying equation 3, it
became evident that the size selectivity b was significantly
decreased in the presence of the accelerators, whereas D0
Fig. 4. Correlation between effects (D+CxEy/DCxEy) on the diffusion
coefficient (D) of pentachlorophenol (PCP) in reconstituted cuticular wax
of Hordeum vulgare and the internal concentrations of the alcohol
ethoxylates [diethylene glycol monobutyl ether (C4E2), triethylene glycol
monohexyl ether (C6E3), tetraethylene glycol monooctyl ether (C8E4),
pentaethylene glycol monodecyl ether (C10E5), hexaethylene glycol
monododecyl ether (C12E6), octaethylene glycol monododecyl ether
(C12E8), heptaethylene glycol monotetradecyl ether (C14E7), and octaethylene glycol monohexadecyl ether (C16E8)]. Effects were calculated
by dividing PCP diffusion coefficients measured in the presence of the
acoholethoxylates (D+CxEy) by PCP diffusion coefficients measured
without alcoholethoxylates (DCxEy). Data were taken from Schreiber
et al. (1996b). Coefficient of determination r2=0.92.
2521
was not significantly changed (Table 5). This again
confirms the conclusion already drawn previously that
accelerators have a plasticizing effect on the amorphous
wax phase, but they do not interact significantly with the
crystalline wax phase.
Spectroscopic techniques such as ESR (electron spin
resonance) and NMR (nuclear magnetic resonance) spectroscopy have been used to analyse the effect of accelerators on reconstituted cuticular wax at the molecular
level. Mobility of octadecanoic acid (C18AC), either
labelled for ESR (5-doxyl stearic acid) or 2H- NMR
(perdeuterated octadecanoic acid) experiments, was monitored in the absence and the presence of different AEs
within certain temperature ranges between 10 8C and
60 8C (Schreiber et al., 1996b, 1997). These spectroscopic
experiments showed convincingly (both ESR and NMR)
that the molecular environment of the ESR-probe as well as
the NMR-probe in the wax significantly increased in
fluidity in the presence of AEs, compared with controls
without AEs. A similar increase in fluidity in control
measurements without AEs could only be obtained when
temperature was increased by about 20–30 8C. This again
leads to the conclusion that the accelerators had a nonspecific plasticizing effect on the molecular environment of
the amorphous wax phase. This effect is similar to the
plasticizing effect obtained by increasing temperature.
Further experiments have been conducted comparing the
effect of two chemically very different classes of accelerators, AEs and n-alkyl esters, on the mobility of 2,4dibutyric acid (2,-DB) in reconstituted wax of H. vulgare
and S. floribunda (Simanova et al., 2005). Effects on the
diffusion of 2,4-DB within each class of accelerators again
increased linearly with increases in the internal concentration of the accelerators (Fig. 5). In H. vulgare these effects
were obtained at about a 10-fold lower internal accelerator
concentration compared to S. floribunda wax. This is
a function of the different sorption capacities of the two
different waxes and it shows that waxes from different
plants can behave very differently. Furthermore, at the
same internal concentrations of accelerators, n-alkyl esters
were 1–2 orders of magnitude more effective in increasing
D compared with AEs. Thus, different chemical classes of
accelerators can have very different effects, which shows
that the overall effects are a function of the specific
chemical composition of that class of accelerators, and
not only of their internal concentration.
Conclusions: advantages and limitations
Working with isolated and subsequently reconstituted cuticular wax has significantly contributed to our understanding on a molecular level of the interaction of accelerators
within the transport-limiting barrier of plant cuticles.
Various hypotheses have been tested on accelerator/wax
2522 Schreiber
8
effect (D+acc/D-acc)
6
n-alkyl esters
CxEy
4
2
0
0
20
40
60
80
100
120
140
concentration of accelerators in wax [g
160
180
kg-1]
Fig. 5. Correlation between effects (D+acc/Dacc) on the diffusion
coefficient (D) of 2,4-dichlorobutyric acid (2,4-DB) in reconstituted
cuticular wax of Stephanotis floribunda and the internal concentrations of
n-alkyl esters [dimethyl suberate (DMSU), diethyl suberate (DESU),
diethyl sebacate (DES), dibutyl suberate (DBSU), and dibutyl sebacate
(DBS)] and alcohol ethoxylates [diethylene glycol monododecyl ether
(C12E2), tetraethylene glycol monododecyl ether (C12E4), pentaethylene
glycol monododecyl ether (C12E5), hexaethylene glycol monododecyl
ether (C12E6), and octaethylene glycol monododecyl ether (C12E8)].
Effects were calculated dividing 2,4-DB diffusion coefficients measured
in the presence of the accelerators (D+acc) by diffusion coefficients
measured without accelerators (Dacc). Data were taken from Simanova
et al. (2005). Coefficient of determination r2=0.97 for n-alkyl esters and
r2=0.96 for alcohol ethoxylates.
interactions (e.g. solubilization versus plasticization) and it
has helped to establish fundamental principles of cuticular
transport at a molecular level. Although this system using
reconstituted wax is artificial, it has shown successfully for
different species that mobilities measured in reconstituted
wax were highly correlated with the transport properties of
isolated cuticles and even intact leaves (Schreiber and
Riederer, 1996; Kirsch et al., 1997). This was demonstrated
for water (cuticular transpiration) as well as for lipophilic
organic molecules. Furthermore, experiments measuring
the effects of accelerators on substances penetrating
isolated plant cuticles (Schönherr, 1993a, b) in general
fitted very well with results obtained using reconstituted
cuticular wax (Schreiber et al., 1996b; Burghardt et al.,
1998). Thus, there is obviously a spontaneous arrangement
of cuticular waxes at a molecular level during reconstitution, leading to a molecular arrangement similar to that
occurring in intact cuticles or intact leaf surfaces. One
major limitation of this experimental set-up using reconstituted cuticular wax is the fact that it only represents the
lipophilic transport path across the cuticle and, consequently, it only allows the measurement of the mobility of
lipophilic non-electrolytes. The polar path of diffusion
(Schönherr, 2000, 2006), allowing polar and even charged
molecules to diffuse through polar aqueous pores in
cuticles, cannot be analysed by this technique.
There are still some experiments, which have not yet
been carried out using this experimental set-up. Mobilities
of accelerators themselves have rarely been analysed,
although this information would be of significance. It can
be postulated that the maximum effect of accelerators on
a specific pesticide compound can only be obtained when
the mobilities of both are of a similar magnitude. This information could lead to more specific formulations, by
selecting accelerators and pesticides with comparable mobilities. Furthermore, experiments are missing on the effects
of temperature on D and K in reconstituted wax. Finally, the
major, and probably most difficult task is the combination
of: (i) transport (measurement of D) and sorption (measurement of K) experiments; (ii) approaches measuring
the effects of accelerators on D; and (iii) the analysis of the
chemical composition of different plant waxes, to find the
relationship between wax composition and transport properties of reconstituted cuticular wax and the effects of
accelerators. Ideally, such a relationship, whenever elucidated in future, could help to predict accelerator effects on
pesticide mobility from wax composition.
Acknowledgement
The author gratefully acknowledges financial support by the DFG
for most of the work presented.
References
Abraham MH, McGowan JC. 1987. The use of characteristic
volumes to measure cavity terms in reversed phase liquid
chromatography. Chromatographia 23, 243–246.
Buchholz A. 2006. Characterization of the diffusion of nonelectrolytes across plant cuticles: properties of the lipophilic pathway. Journal of Experimental Botany 57, 2501–2513.
Buchholz A, Baur P, Schönherr J. 1998. Differences among plant
species in cuticular permeabilities and solute mobilities are not
caused by differential size selectivities. Planta 206, 322–328.
Buchholz A, Schönherr J. 2000. Thermodynamic analysis of
diffusion of non-electrolytes across plant cuticles in the presence
and absence of the plasticiser tributyl phosphate. Planta 212,
103–111.
Burghardt M, Schreiber L, Riederer M. 1998. Enhancement of the
diffusion of active ingredients in barley leaf cuticular wax by
monodisperse alcohol ethoxylates. Journal of Agricultural and
Food Chemistry 46, 1593–1602.
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.
Cussler EL. 1984. Diffusion. Mass transfer in fluid systems.
Cambridge: Cambridge University Press.
Felder RM, Huvard GS. 1980. Permeation, diffusion, and sorption
of gases and vapors. In: Marton L, Marton C, eds. Methods of
experimental physics, Vol. 16. New York: Academic Press,
315–377.
Fowler JD. 1999. A discussion of the dependence of cuticular
mobility on molar volume. In: Tann RS, Nalewaja JD, Viets A,
eds. 19th symposium on pesticide formulations and application
systems: global pest control formulations for the next millennium,
ASTM STP 1373. West Conshohocken, PA: American Society for
Testing and Materials.
Molecules in cuticular waxes and the effects of accelerators
Jenks MA, Ashworth EN. 1999. Plant epicuticular waxes: function,
production, and genetics. In: Janick A, ed. Horticultural reviews,
Vol. 23. New York: John Wiley and Sons Ltd., 1–63.
Jetter R. 1993. Chemische Zusammensetzung, Struktur und Bildung
röhrenförmiger Wachskristalle auf Pflanzenoberflächen. PhD
thesis, University of Kaiserslautern, Germany.
Kirkwood RC. 1999. Recent developments in our understanding of
the plant cuticle as a barrier to the foliar uptake of pesticides.
Pesticide Science 55, 69–77.
Kirsch T. 1996. Charakterisierung der kutikulären Barriereeigenschaften ausgewählter Baumarten. Transportexperimente mit
intakten Blättern, isolierten Kutikularmembranen und rekonstituierten Wachsen. PhD thesis, University of Würzburg, Germany.
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.
Kunst L, Samuels AL. 2003. Biosynthesis and secretion of plant
cuticular wax. Progress in Lipid Research 42, 51–80.
Potts RO, Guy RH. 1992. Predicting skin permeability. Pharmaceutical Research 9, 663–669.
Reynhardt EC. 1997. The role of hydrogen bonding in the cuticular
wax of Hordeum vulgare L. European Biophysics Journal 26,
195–201.
Reynhardt EC, Riederer M. 1991. Structure and molecular
dynamics of the cuticular wax from leaves of Citrus aurantium
L. Journal of Physics D-Applied Physics 24, 478–486.
Riederer M, Schönherr J. 1984. Accumulation and transport of
2,4-dichlorophenoxyacetic acid in plant cuticles. 1. Sorption in
the cuticular membrane and its components. Ecotoxicology and
Environmental Safety 8, 236–247.
Riederer M, Schreiber L. 1995. Waxes: the transport barriers of
plant cuticles. In: Hamilton RJ, ed. Waxes: chemistry, molecular
biology and functions. Dundee: The Oily Press, 131–156.
Sangster J. 1997. Octanol–water partition coeffcients: fundamentals and physical chemistry. Chichester: John Wiley and Sons.
Schönherr J. 1993a. Effects of monodisperse alcohol ethoxylates on
mobility of 2,4-D in isolated plant cuticles. Pesticide Science 38,
155–164.
Schönherr J. 1993b. Effects of alcohols, glycols and monodisperse
ethoxylated alcohols on mobility of 2,4-D in isolated plant cuticles.
Pesticide Science 39, 213–223.
Schönherr J. 2000. Calcium chloride penetrates plant cuticles via
aqueous pores. Planta 212, 112–118.
Schönherr J. 2006. Characterization of aqueous pores in plant
cuticles and permeation of ionic solutes. Journal of Experimental
Botany 57, 2471–2491.
2523
Schönherr J, Baur P. 1994. Modelling penetration of plant cuticles
by crop protecting agents (CPA) and effects of adjuvants on rates
of penetration. Pesticide Science 42, 185–208.
Schönherr J, Riederer M. 1989. Foliar penetration and accumulation of organic chemicals in plant cuticles. Reviews of Environmental Contamination and Toxicology 108, 1–70.
Schönherr J, Schreiber L, Buchholz A. 2001. Effects of temperature and accelerator concentration on mobility of non-electrolytes
in plant cuticles. Pest Management Science 57, 17–24.
Schreiber L. 1995. A mechanistic approach towards surfactant/
wax interactions: effects of octaethylene glycol monododecyl
ether on sorption and diffusion of organic chemicals in reconstituted cuticular wax of barley leaves. Pesticide Science 45,
1–11.
Schreiber L. 2005. Polar paths of diffusion across plant
cuticles: new evidence for an old hypothesis. Annals of Botany
(in press).
Schreiber L, Kirsch T, Riederer M. 1996a. Transport properties of
cuticular waxes of Fagus sylvatica L. and Picea abies (L.) Karst.:
estimation of size selectivity and tortuosity from diffusion
coefficients of aliphatic molecules. Planta 198, 104–109.
Schreiber L, Riederer M. 1996. Determination of diffusion
coefficients of octadecanoic acid in isolated cuticular waxes and
their relationship to cuticular water permeabilities. Plant, Cell and
Environment 19, 1075–1082.
Schreiber L, Riederer M, Schorn K. 1996b. Mobilities of
organic compounds in reconstituted cuticular wax of barley
leaves: effects of monodisperse alcoholethoxylates on diffusion
of pentachlorophenol and tetracosanoic acid. Pesticide Science 48,
117–124.
Schreiber L, Schönherr J. 1992. Analysis of foliar uptake of
pesticides in barley leaves: role of epicuticular waxes and
compartmentation. Pesticide Science 36, 213–221.
Schreiber L, Schönherr J. 1993. Mobilities of organic compounds
in reconstituted cuticular wax of barley leaves: determination of
diffusion coefficients. Pesticide Science 38, 353–361.
Schreiber L, Schorn K, Heimburg T. 1997. 2H NMR study of
cuticular wax isolated from Hordeum vulgare L. leaves: identification of amorphous and crystalline wax phases. European Biophysics Journal 26, 371–380.
Simanova E, Shi T, Schönherr J, Schreiber L. 2005. Sorption in
reconstituted waxes of homologous series of alcohol ethoxylates
and n-alkyl esters of suberate and sebacate and their effects on
mobility of 2,4-dichlorophenoxybutyric acid. Pest Management
Science doi: 10.1002/ps.979.
Sitte P, Rennier R. 1963. Untersuchungen an cuticularen Zellwandschichten. Planta 60, 19–40.
Stein WD. 1967. The movement of molecules across cell membranes. New York, London: Academic Press.