.,anta.990,
Planta 9 Springer-Verlag1990
Phase transitions and thermal expansion coefficients of plant cuticles The effects o f temperature on structure and function Lukas Schreiber and J~rg Sch~inherr Lehrstuhl f/ir Botanik, Technische Universitgt Mfinchen, Arcisstrasse 21, D-8000 Mfinchen 2, Federal Republic of Germany Received 1 December 1989; accepted 24 March 1990
Abstract. The temperature-induced volume expansion of enzymatically isolated cuticular membranes of twelve plant species was measured. All cuticular membranes exhibited distinct second-order phase transitions in the temperature range of about 40 to 50~ C. Increases in the volumes of fruit cuticles (Lycopersicon, Cucumis, Capsicum, Solanum and Malus) were fully reversible, while leaf cuticular membranes (Ficus, Hedera, Nerium, Olea, Pyrus, Picea and Citrus) underwent irreversible structural changes. Below the phase-transition temperature, volumetric expansion coefficients ranged from 0.39"10 -6 m 3 - k g - l - K -1 to 0.62.10 -6 m 3 . k g - l . K -1, and above from 0 . 6 0 . 1 0 - 6 m 3 . k g - l . K -1 to 1.41. 1 0 - 6 m 3 . k g - l . K - 1 . Densities of cuticles at 25~ ranged from 1020 kg.m -3 to 1 370 kg.m -3. Expansion coefficients and phase transitions were characteristic properties of the polymer matrix as a composite material, rather than of cutin alone. It is argued that the sudden increase of water permeability above the transition temperature, is caused by an increase of disorder at the interface between the polymer matrix and the soluble cuticular lipids. Possible ecological and physiological consequences of these results for living plants are discussed. Key words: Cutin - Cuticle (membrane, density) - Expansion coefficient - Phase transition (cuticular membrane) - Wax
Introduction Cuticles are very effective barriers against passive water loss from plants (Sch6nherr 1982). Under extreme environmental conditions, such as water stress and high temperature, stomata will be closed and the water permeability of cuticular membrane determines the rate of Abbreviations: CM=cuticular membrane; CU=cutin; MX=polymer matrix; SCL = soluble cuticular lipids (waxes)
transpiration (Kamp 1930). The water permeability (P) of cuticles was found to increase with increasing temperature (Sch6nherr et al. 1979). Arrhenius plots (In P vs. l/T) in the range of 5 to 65~ C consisted of two linear portions which had an intersection around 45 ~ C. Moreover, a second measurement using the cuticular membranes previously heated to 65 ~ C displayed an increase in water permeability in the lower temperature range compared with the first measurement. It was concluded that temperatures higher than 45 ~ C caused irreversible structural changes in the membranes, which lead to an increase in water permeability. Plant cuticles are heterogeneous, as is evident from optical anisotropy (Sitte and Rennier 1963), transition electron microscopy (see review by Holloway 1982 a) and transport properties (Sch6nherr and Riederer 1989). Isolated cuticular membranes (CM) consist of an insoluble polymer matrix and soluble cuticular lipids associated with the polymer matrix (Riederer and Sch6nherr 1984). The polymer matrix itself is composed of covalentely linked cutin acids and "non-cutin components", mainly cellulose (Wattendorf and Holloway 1980) and, in minor amounts, amino acids, pectinaceous materials and phenols (Sch6nherr and Huber 1977; Hunt and Baker 1980). These definitions are mainly operational and have been found useful in previous studies into sorption in and diffusion across CM (see review by Sch6nherr and Riederer 1989). They differ to some extent from those used by Sitte and Rennier (1963), who distinguish an outer layer (the cuticle sensu stricto) which does not contain cellulose, and an inner cuticular layer, which does. Eckl and Gruler (1980) attributed the temperature effect on water permeability to a phase transition and reorientation of soluble cuticular lipids, which leads to hydrophilic holes in the barrier. Irreversible membrane alterations were thought to be due to a recrystallisation of soluble cuticular lipids upon cooling that results in a structure different from the original one and leads to a partial preservation of the hydrophilic holes in the barrier. However, our investigations of thermally induced volume expansion of cuticles revealed second-
L. Schreiber and J. Sch6nherr: Thermal expansion of plant cuticles o r d e r p h a s e t r a n s i t i o n s o f t h e p o l y m e r m a t r i x itself. T h i s fact necessitates a new look at the causes of irreversible increase in water permeability of cuticles with temperature.
Material and methods Cutieular membranes. Cuticular membranes of twelve species from nine different plant families were investigated. Adaxial, astomatous leaf cuticles were obtained from Citrus aurantium L., Ficus elastica Roxb. var. decora, Pyrus communis L., Hedera helix L., Nerium oleander L., Olea europaea L. and Pieea omorika Pancic. Fruit cuticles came from Capsicum annuum L., Lycopersicon esculentum L., Solanum melongena L., Malus pumila L. and Cucumis sativus L.. These membranes were subsamples of those used in earlier investigations by Riederer and Sch6nherr (1984, 1985), where additional details (origin, varieties etc.) can be found, with the exception of Citrus and Ficus cuticles. Citrus and Ficus leaves were obtained from trees grown in growth chambers (Geyer and Sch6nherr 1990). Enzymatic isolation of CM followed the procedure of Riederer and Sch6nherr (1986). Isolated cuticles will be referred to as (CM). Treatment of C M with a mixture of chloroform and methanol (1/1, v/v) for 24 h in a soxhlet appartus removed soluble cuticular lipids (SCL) quantitatively. The remainder is called polymer matrix (MX). Incubation of M X with 6 N HCI at 120 ~ C hydrolyzes all "non-cutin c o m p o n e n t s " (HY), such as cellulose, pectinaceous substances and polypeptides, and the insoluble residue remaining is cutin (CU; Sch6nherr and Bukovac 1973; Riederer and Sch6nherr 1984). Measurement of volume expansion. Determination of the volume expansion was carried out in the temperature range from 5 to 65 ~ C. About 300 mg of membrane material and water were added to a specific-gravity bottle (1 cm3). The filled bottle was aspirated extensively to remove air from water and cuticles, which would interfere with the measurements. The pressure of 6.7.103 Pa was sufficient to reduce the gas content of the water to a level which did not interfere with the measurements. A calibrated capillary was attached to the bottle, using groundglass joints sealed with high-vacuum silicone grease (Wacker, Munich, FRG). The bottle with the capillary attached was placed in a thermostatted water bath. Only the tip of the capillary protruded above the surface of the water. Temperature was controlled with an accuracy of + 0.01 K. The rise of the water column in the capillary with increasing temperature was measured using a cathetometer (Spindler and Hoyer, G6ttingen, F R G ; accuracy ___0.05 mm). The volume of the membranes was calculated by substracting the volume of water from the total volume of membrane plus water. Prior to the measurements, several calibration runs were carried out. Determination of the temperature difference between the interior of the bottle filled with CM and water, and the water bath, using a heating rate of 1 K in 4 min, revealed a constant lag of 0.2 K, which did not change over the whole range of 5 to 65 ~ C. The capillary itself was calibrated with mercury in the range of 5 to 65 ~ C (inner cross-sectional area 0.18.10 -6 m2), and no thermally induced volume expansion of the capillary itself could be detected. Calibration measurements, using deionized, degassed water, were carried out to determine the exact volume expansion of water in our apparatus. For three independent measurements the results could be repeated with a coefficient of variation of 0.5%. The volume expansion of a 0.01 M Mes buffer (2(N-morpholino)ethanesulfonic acid; Sigma, St. Louis, La., USA) adjusted with Ca(OH)2 to pH 6.0 was also determined. Results could be reproduced with good accuracy in the lower temperature range up to 4 5 ~ (coefficient of variation 0.5%). In the upper temperature range, however, the coefficient of variation of three callibration runs was 3%.
187
Calculation of volume expansion coefficients. The volume V (m 3) per kg of cuticle in the temperature range of 5 to 65 ~ C was calculated according to the equation t
5
V=(I-1)-L.M Mc
t
w
Ak
Eq. (1)
where 15 and It (m) denote the height of the water meniscus in the capillary at the experimental temperature T and at 5~ C, respectively, when the bottle was filled with cuticular membranes and water. L' ( m . k g - 1 ) is the height of the meniscus with pure water at the given temperature, as determined in the callibration procedure. M w and M c (kg) represent the masses of water and cuticular membrane, respectively, and A ~ (m 2) is the inner cross-sectional area of the capillary. Linear portions in the plot V (volume) versus T (temperature) were identified and regression lines were fitted to the lower and upper parts of the volume-expansion curves. Coefficients of correlation were always better than 0.99 in the lower, and better than 0.98 in the upper part of the curve. The slopes b of the regression lines give the absolute volume expansion coefficients in m3"kg - iK-~. The intersection of the two regression lines represents the midpoint of the phase transition. The region of the phase transition itself generally ranged over 10-15 ~ C. Relative volume expansion coefficients a (~ C-1) were calculated using Eq.2: a - (V'/VS)- 1 T-- 5
Eq. (2)
where V t (m 3) and V s (m 3) are the volumes of the cuticle at the experimental temperature and at 5~ C, respectively, and T (~ C) denotes the experimental temperature.
Estimation of experimental error. The reading accuracy of the cathetometer was 0.05 m m which is equal to a volume of about 0.01 m m 3. Experimental results thus vary in the second decimal point by about +0.01.10 -6 m z. For a volume expansion coefficient of 0.6.10-6 m 3 - k g - l . K - i , a deviation of + 0 . 0 1 . 1 0 - 6 m 3 corresponds to an error of 1.7%. Four successive measurements of the same sample of tomato fruit cuticle (Table 1) gave a mean volume expansion of 36.43. 10 - 6 m 3.kg 1 with a standard deviation of 0.46.10 - 6 , which corresponds to a coefficient of variation (cv) of 1.3%. The absolute volume-expansion coefficients of the same membranes below and above the phase transition were 0.57' 10- 6 m 3 . k g - 1. K - 1 + 0.005. 10 - 6 (CV= 1.0%) and 0.69.10 - 6 m 3 . k g - 1 - K - 1 ___0.02.10 - 6 (CV = 3.3%). The midpoint of the phase transition is 46.0 ~ C___0.6 (cv = 1.3%). It follows from these considerations that below the phase transition an experimental error of about 1.5-2.0% is associated with the data mainly due to the limits of the reading accuracy of the cathetometer. Above the phase transition the error of the volume expansion coefficient was about twice as large. Determination of water permeability. Water permeabilities of isolated Citrus leaf C M were determined according to the method described by Sch6nherr and Lendzian (1981), with the exception that the transpiration chambers were made of brass rather than Plexiglas. After the permeability of the native C M had been determined, the wet cuticles were stored for 20 h at 65 ~ C. The effect of heat treatment on water permeability was measured immediately after treatment. A third determination of water permeability with the same cuticles was conducted 100 d later, after they had been stored in dry form at room temperature. Permeance P ( m - s - 1 ) was calculated from the steady-state flow rate of water F (kg. s-1) across the cuticle according to Eq.3: F P=A-d'Aaw
Eq. (3)
where A (m 2) is the exposed membrane area, d ( k g . m - 3 ) is the density of liquid water at 25 ~ C and d a represents the gradient of water activity across the cuticle, w
188
L. Schreiber and J. Sch6nherr: Thermal expansion of plant cuticles
Results
Table 1. Volume expansion of isolated cuticular membranes from selected plant species in the temperature range of 5 to 65 ~ C
Volume expansion of cuticular membranes (CM). Plots of volume o f C M versus temperature exhibited two or three linear portions (Figs. 1, 2). The transition region from the lower to the upper volume-expansion curve was in most cases in the range f r o m 40 to 50 ~ C. The difference between the volume-expansion coefficient below and above the phase transition was significant at the 95% level in all cases. Fruit CM from five species responded similarly to temperature (Table 1). The absolute volume expansion from 5 to 65 ~ C ranged from 32.4.10 -6 m 3 .kg -1 (Capsicum CM 2) to 4 2 . 8 . 1 0 - 6 m 3 . k g -1 (Malus). The midpoints of the phase transitions reached from 40.2~ C (Capsicum CM 2) to 54.5 ~ C (Malus). Variation between different samples from the same species can be fairly
Transition Volume expansion midpoint coefficient (~ C) ( . 1 0 6 m 3 - k g - l - K -1)
Total volume expansion (-106 m3.kg -1)
Below Above transition transition Fruit CM Lycopersicon a
Measurement Measurement Measurement Measurement
1 2 3 4
45.9 46.8 45.5 46.0
0.57 0.57 0.57 0.56
0.70 0.69 0.66 0.71
36.9 36.4 35.8 36.6
46.2 40.2 42.3
0.57 0.51 0.56
0.63 0.60 0.60
35.5 32.4 34.5
(H20) (Ca z§
43.1 51.4
0.52 0.51
0.81 0.68
37.2 32.9
Malus
54.5
0.48
1.88
42.8
Cucumis
43.9
0.51
0.63
33.0
Measurement 1 48.8 Measurement 2 44.2
0.55 0.56
0.64 0.61
34.5 34.8
Pyrus
46.2
0.62
1.41
49.8
Nerium
38.5
0.39
0.62
29.4
Olea
55.3
0.45
0.70
29.1
Picea
49.3
0.49
0.68
32.2
Capsicum b
CM 1 CM 2 CM 3 Solanum"
Lycopersicon CH
60
"7
Pyrus CH
Citrus CH
50
Leaf CM
40
Hedera a
.~
30
m
20 > 10
i
0
20
40
20
60
40
60
i
20
40
i
60
" Different measurements using the same sample b Different samples of the same species
Temperature [~
Fig. 1. Volume expansion of the cuticular membranes (CM) of the species Lycopersicon, Pyrus and Citrus
60
Ficus CH
Ficus HX
Ficus CU
~- 50 ~, 40..2-
.~ 30 ~
20
o
> 10
0
,
,
,
20
40
60
0
20
40
60
0
20
40
60
Temperature [~
Fig. 2. Volume expansion of cuticular membrane (CM), polymer matrix (MX) and cutin (CU) of Ficus
high as results with Capsicum CM demonstrate. Temperature effects were reversible with fruit CM as shown for Lyr where results of four repeated determinations were identical within experimental error (Table 1). The only exception was Malus where the upper volume expansion coefficient in the second experimental run was significantly lower than in the first determination (data not shown). Leaf CM from seven species showed greater variation in their volume expansion behavior than fruit CM (Tables 1, 2). The absolute volume expansion ranged from 29.1-10 - 6 m3-kg -1 (Olea) to 49.8-10 - 6 m3-kg -1 (Pyrus) and the midpoint of the transition from the lower to the upper temperature extended from a minimum o f 38.5 ~ C (Nerium) to a maximum of 55.3 ~ C (Olea). Another striking difference between leaf and fruit CM was the irreversible effect of high temperatures on volume expansion. Repreated measurements using the same sample of Hedera CM (Table 1) showed a significant increase of the lower volume expansion coefficient from the first to the second measurement and a significant decrease of the volume expansion coefficient in the upper temperature range. The transitions became less distinct.
L. Schreiber and J. Sch6nherr: Thermal expansion of plant cuticles
189
Table 2. Volume expansion of leaf cuticular membranes of Citrus and Ficus between 5 and 65 ~ C. Temperature ranges = below the
first, between the first and the second, and above the second transition midpoint
Transition midpoints (o c)
Volume expansion coefficient (. 106 m3.kg -1 .K-i)
Total volume expansion (.10 6 m3.kg -i)
Below first transition
Below second transition
Above second transition
Citrus C M a
Adaxial C M Abaxial C M
17.1
42.3 44.8
0.47
0.53 0.50
0.74 0.60
36.2 32.0
14.0 24.6 19.1 18.1
46.0 48.3 50.7 49.9
0.64 0.60 0.66 0.64
0.61 0.58 0.57 0.55
0.92 1.00 1.00 0.95
43.8 42.4 42.0 40.6
Ficus C M ~
(Leaf (Leaf (Leaf (Leaf
1) 2) 7) 11)
a Different samples of the same plant
Table 3. Volume expansion of cuticular membrane (CM), polymer matrix (MX) and cutin (CU) of the species Ficus, Capsicum and Citrus
Transition midpoints
(o c)
Fieus C M Ficus M X Ficus C U
18.1
Capsicum CM Capsicum M X Capsicum C U Citrus C M Citrus M X Citrus C U
17.2 21.2
Volume-expansion coefficient (.106 mS.kg - i . K - i )
Total volume expansion (.106 m3.kg -1)
Below first transition
Below second transition
Above second transition
49.9 52.8
0.64 0.66 0.96
0.55
0.95 1.13
40.6 45.5 58.2
42.3 39.4
0.56 0.63 0.82
0.60 0.65
34.5 38.6 48.6
42.3 43.7
0.47 0.50 0.95
0.74 0.79
36.2 40.2 57.4
The other leaf CM, examined here, showed the same irreversibility, with the exception of Citrus and Pieea. Here the increases of the lower volume expansion coefficients in the second determination were not significantly different from the first determination (data not shown). Results obtained with Ficus and Citrus CM are less uniform (Table 2). The Citrus CM from the abaxial side o f the leaf responded to temperature like the CM o f the other six species. However the adaxial Citrus CM (Fig. 1) and Ficus CM (Fig. 2) showed a remarkable deviation. There were three distinct regions with three different volume expansion coefficients in the temperature range from 5 to 65 ~ C. The first transition occurred at about 15-20 ~ C and the second at about 40-50 ~ C. With Solanum fruit CM it was observed that volume expansion was affected by calcium ions. Determination of the volume expansion of CM in calcium buffer resulted in a smaller total volume expansion and the transition midpoint rose compared with the volume expansion of the same sample in water (Table 1). Ficus leaf CM of increasing age displayed a slight decrease in the total volume expansion and a slight in-
0.53 0.65
crease of the transition midpoint (Table 2). However, differences within the lower volume expansion coefficients o f the four stages of development were not significant at the 95% level. The same was true for the upper volume expansion coefficients.
Volume expansion of polymer matrix (MX) and cutin (CU). The MX of all three species examined (Ficus, Capsicum, Citrus) exhibited a volume-expansion behavior similar to that observed with CM (Table 3). The typical transitions found for the CM were also observed with the MX, whereas with CU they disappeared and volume expansion was linear between 5 and 65 ~ C (Ficus CM Fig. 2). The total volume expansions of Ficus, Capsicum and Citrus increased in the order CM, MX and CU (Table 3). The densities of CM and M X from Ficus, Capsicum and Citrus were all larger than 1.103 k g . m -3 (Table 4). The densities of CU of the species Ficus, Capsicum and Citrus were below 1-10 ~ k g . m -3. Cutin ( C U ) a m o u n t e d to about two-thirds of the total weight of the cuticular membrane (CM), while the hydrolyzable fraction (HY),
190
L. Schreiber and J. Sch6nherr: Thermal expansion of plant cuticles
Table 4. Densities of cuticular membrane (CM), polymer matrix (MX) and cutin (CU) at 25~ C. The densities are calculated from the density of the cuticle at 5~ C plus the measured volume expansion from 5 to 25~ C. Values are means of 3-5 measurements. Coefficients of variation were less than 1.5% in all cases
and especially the soluble cuticular lipids (SCL) varied appreciably among species (Table 5).
Density (.10 -3 kg.m 3) CM
MX
CU
Leaf cuticles Ficus Hedera Pyrus Nerium Olea Picea Citrus
1.02 1.09 1.07 1.17 1.05 1.37 1.04
1.04
0.92
1.03
0.94
Fruit cuticles Lycopersicon Capsicum Solanum Malus Cucumis
1.12 1.10 1.06 1.10 1.14
1.10
0.96
Table 5. Composition of Ficus, Capsicum and Citrus cuticular membranes (CM) as a percentage of the weight Composition (weight %)
Ficus Capsicum Citrus
CU a
SCL b
HY c
62 64 72
20 10 6
18 26 22
a Cutin (CU) fraction b Fraction of soluble cuticular lipids (SCL) ~ "Non-cutin components" (HY) obtained through hydrolysis
Table 6. Influence of heat treatment and storage on water permeabilities of isolated Citrus CM. A denotes the permeances of the untreated cuticles, B the permeances after treatment with 65~ and C the permeances after storage for a hundred days. The ratios B/A and C/A are a measure of the effects of heat and storage, respectively. Mean is the average of the six membranes and cv represents the coefficient of variation Permeance (- 101~m . s - 1) CM
A
B
B/A
C
C/A
1
2.81 2.67 2.35 2.71 1.89 1.90
8.18 5.10 3.72 9.06 7.91 5.77
2.91 1.91 1.58 3.34 4.18 3.04
3.31 3.30 1.83 2.88 2.99 2.56
1.18 1.39 0.78 1.06 1.58 1.35
2.39 17
6.62 31
2.83 34
2.81 20
1.17 23
2 3 4 5 6 Mean cv (%)
Water permeability o f Citrus ( C M ) . A heat treatment of wet CM for 20 h at 65 ~ C caused an average increase of the water permeability by a factor of 2.83 (Table 6). Storage of these cuticles at room temperature for 100 d led to a significant reduction of water permeability. The original permeabilities (prior to heat treatment) were almost reestablished. The variation of the water permeabilities between individual membranes as well as the influence of the heat treatment on the permeabilities was high.
Discussion
Phase transitions of plant cuticles, as they were already described for Citrus and Hedera (Sch6nherr et al. 1979; Eckl and Gruler 1980), could also be demonstrated using CM and M X of ten additional species. The phase transitions, described here, are transitions of the second order, which are characterised by a relatively wide region of transition over 10 or even 15 K (Seymour 1971). Secondorder phase transitions are typical properties of macromolecules, synthetic ones as well as biopolymers (Mandelkern 1983). Most leaf CM (except those of Citrus and Pieea) underwent irreversible changes when heated to 65 ~ C, while this was observed for only one fruit CM (Malus). This suggests differences in polymer structures between leaf and fruit CM. A correlation between cutin type (C16-, C18- and C16-/C18-type) according to Holloway (1982 b) and reversibility of temperature effects does not exist. Cutin types C16-, C18- and epoxy-fatty acids occur in both reversible and irreversible CM. An increasing number o f cross-links, due to the formation of ether bonds between epoxy- and hydroxylfunctions, occur during cutin development of Clivia (Schmidt and Sch6nherr 1982; Riederer and Sch6nherr 1988). This should result in an increase of the midpoint o f the phase transition and a decrease of the absolute volume expansion, as is frequently described with synthetic polymers (Sperling 1986). The Ficus MX contains epoxy-fatty acids and a slight decrease of the absolute volume expansion from the 1st to the 1lth leaf (Table 2) and a slight increase of the midpoint of the phase transition of 4 K were in fact detectable. Isolated CM and MX contain carboxyl groups which have a high selectivity for Ca 2 + (Sch6nherr and Bukovac 1973). This compensation of negative charges within the cuticle results in a physical form of cross-linking and an influence on volume expansion of CM can be expected and was indeed observed with Solanum CM (Table 1). In the presence of calcium buffer, absolute volume expansion decreased and the midpoint of the phase transition rose. How do individual classes of components such as cutin, soluble lipids or cellulose contribute to secondorder phase transitions? The volume expansion of the M X was not affected by the presence of SCL. This was
L. Schreiber and J. Sch6nherr: Thermal expansion of plant cuticles
191
expected, since SCL are not crosslinked to the polymer matrix. The absolute volume expansion as well as the lower and upper volume expansion coefficients of Ficus, Capsicum and Citrus MX increased by about 12%, compared with the CM. This can be attributed to a decrease in weight on extracting SCL (Table 5). The values o f the volume expansion are based on the weight o f the membrane material used in the experiment. The increase of the volume expansion of the MX, expressed in percent, is similar to the decrease in weight of the CM on extraction o f SCL. Comparing the volume expansions of cutin and the polymer matrix (Table 3) shows that phase transitions are a characteristic property of the MX rather than CU. The phase transitions of all three species at about 45 ~ C as well as the second phase transition of Citrus and Ficus between 15 and 20 ~ C disappeared and a linear volume expansion in the whole temperature range was obtained when cutin samples were studied. The cutin polymer (CU) showed an increase o f the absolute volume expansion of about 30-40% compared with the MX. The loss in weight o f the MX, due to hydrolysis, is too low fully to account for the distinct increase of the volume expansion of CU. The absolute volume expansion of cutin can be attributed to a significantly larger volume expansion coefficient (Table 3). The differences in thermal expansion of polymer matrix and cutin are most likely due to cellulose contained in the polymer matrix (Wattendorf and Holloway 1980). The low relative thermal-expansion coefficient of cellulose of about 4.10 - 4 (Brandrup and Imergut 1966) reduces the high thermal volume expansion of cutin, which is more than twice as large (Table 7). Relative expansion coefficients of CM and MX of these species had values located between those for cellulose and cutin. Thermalexpansion coefficients of CM and MX are mixed quantities, which result from the low volume expansion coefficient of cellulose and the high coefficient of cutin.
A comparison of the densities of CM, M X and CU provides further evidence of the decisive role o f cellulose in cuticular membranes. Cuticular membranes had densities about 1.1.103 k g . m -3 (Table 4), and M X membranes did not differ significantly from that value (Table 4). Hydrolyzed cuticles, lacking cellulose, had densities around 0.94-103 k g . m -3 (Table 4) and they exhibited a linear volume expansion in the temperature range between 5 and 65 ~ C. Density o f crystalline cellulose is about 1.4.103 k g . m -3 (Brandrup and Imergut 1966), and therefore the decrease of the density of CM after hydrolysis is partially due to the relatively high density of cellulose. But the increase o f the total volume expansion o f CU compared with MX is not fully compensated by the loss of weight of the MX, caused by hydrolysis. Cellulose associated with cutin seems to reduce the intermolecular distance of the polymer chains of the cutin. Only above 40-50 ~ C do cutin monomers gain sufficient intermolecular space to perform additional molecular motions. This results in a phase transition, as it is generally described for the occurrence of phase transitions in synthetic polymers (Aharoni 1974; Greiner and Schwarzl 1989). A moderate but significant correlation (r = - 0.91) between densities of CM, M X and C U and the respective volume expansions of the species Ficus, Capsicum and Citrus exists (Fig. 3). This again indicates that a lower density hinders the molecular motion o f the polymer chains less than a higher density. Below 50 ~ C, volume expansion of crystalline natural and synthetic waxes is linear and seven different samples had a mean relative volume expansion coefficient o f about 5.7-10 - 4 (CV = 3 0 % ) (Rosenberg and Brotz 1957). Above 50-60 ~ C, volume expansion coefficients increased steadily with rising temperature up to the solidliquid phase transition at about 80-100~ (Marsen 1957; Rosenberg and Brotz 1957). These results are in good agreement with cuticle behavior. Waxes (SCL) associated with the polymer matrix at physiological tem-
Table 7. Relative volume expansion coefficients of cuticular membrane (CM), polymer matrix (MX) and cutin (CU) of the species Ficus, Capsicum and Citrus and of cellulose and waxes. The relative volume expansion coefficient a is calculated according to Eq.2. The subscripts below, middle and above refer to the respective linear portions of the volume-expansion curves
60 CU
o
Capsicum
Citrus
6.4 5.5 9.5 6.6
5.6
11.0 9.6
6.5 8.2
4.7 5.4 7.4 5.0 6.5 7.9 9.8
x
CU
55
a (. 104 K - 1)
Ficus
9
S0 t,5
CM (below) CM (middle) CM (above) MX (below) MX (middle) MX (above) CU Cellulosea Wax b
4.0 5.7
6.0 6.2
m x
t~0
CHx
35 30 0.85
o. ,o
o. ,5
1. ,o
~+
.o5
CH
1. s
Density [.10 3 kg m-3]
Fig. 3. Correlation between absolute volume expansion and density
Brandrup and Immergut 1966 b Rosenberg and Brotz 1957
of cuticular membrane (CM), polymer matrix (MX) and cutin (CU) of the species Ficus, Capsicum and Citrus
192
L. Schreiber and J. Sch6nherr: Thermal expansion of plant cuticles
peratures are crystalline (Meyer 1938; Roelofsen 1952). They do not have a distinct melting point but indeed a wide melting range from 60 to 90 ~ C (Sitte and Rennier 1963). Relative volume expansion coefficients of SCL, CM and M X are similar in the lower temperature range. Above the phase transition the volume expansion of the polymer matrix exceeds that of the waxes because the moderately increasing expansion coefficients of the waxes cannot compensate for the sudden increase of the expansion coefficient of the polymer matrix at 40-50 ~ C. The difference between the two volume expansion coefficients above the phase transition will lead to defects between SCL and MX. These additional defects persist when the temperature is decreased again to physiological ranges, and they are responsible for the increase in water permeability o f Citrus CM (Table 6). However, determination of the water permeability 100 d later showed that these temperature-induced defects can heal (Table 6). Such a healing process was observed earlier by Fox (1958) using paper coated with paraffin wax. Water permeabilities of these membranes decreased steadily with increasing storage times because a change in the structure of the paraffin crystals led to fewer defects. We suggest that the healing of plant cuticles is also a consequence of recrystallisation of wax crystals and rearrangement of disordered waxes in the polymer matrix. In their natural environment, plants are sometimes exposed to temperatures o f 45-50 ~ C (Sapper 1935; Kuraishi and Nito 1980). They are able to tolerate those extreme conditions for an appreciable amount of time (Sapper 1935). Species having scleromorphic leaves (Nerium, Olea and Picea) had the lowest volume expansions (Table 1). This might be an ecological adaptation to dry and warm environments, because a low volume expansion o f the cuticle should minimize the eventually hazardous effect of an increased water permeability caused by thermally induced structural changes in the cuticle. A continuous synthesis and extrusion of cuticular lipids by mature leaves and fruits can be frequently observed (Schieferstein and Loomis 1959). This process helps to repair defects in the cuticle caused by natural influences such as wind, rain and temperature. Soluble cuticular lipids are responsible for the excellent barrier qualities of plant cuticles (Sch6nherr 1982). A continuous synthesis and extrusion of soluble lipids should be essential for a plant in order to maintain the functional integrity of the cuticle. Plant cuticles are not simply inert physical systems once synthesized. They are subject to permanent environmental stress. Both physical (rearrangement of wax molecules) and physiological (wax synthesis) processes cooperate and maintain the structural integrity of cuticles necessary for survival.
References
The authors greatfully acknowledge stimulating discussions with Drs. H. Gruler (Exp. Physik 3, Universitfit Ulm, FRG) and M. Riederer (Institut ffir Botanik und Mikrobiologie, Technische Universitfit Mfinchen, Mfinchen, FRG) and financial support by the Deutsche Forschungsgemeinschaft.
Aharoni, S.M. (1974) Thermal dilatation of polymers. J. Macromol. Sci. Phys. 94, 699-731 Brandrup, J., Immergut, E.H., eds. (1966) Polymer handbook. Wiley, New York Eckl, K., Gruler, H. (1980) Phase transitions in plant cuticles. Planta 150, 102-113 Fox, R.C. (1958) The relationship of wax crystal structure to the water vapor transmission rate of wax films. J. Tech. Assoc. Pulp Paper Ind. 41,283-289 Geyer, U., Sch6nherr, J. (1990) The effect of the environment on the permeability and composition of Citrus leaf cuticles. 1. Water permeability of isolated cuticular membranes. Planta 180, 147-153 Greiner, R., Schwarzl, F.R. (1989) Volume relaxation and physical aging of amorphous polymers. 1. Theory of volume relaxation after single temperature jumps. Colloid Polymer Sci. 267, 39-47 Holloway, P.J. (1982a) Structure and histochemistry of plant cuticular membranes: an overview. In: The plant cuticle, pp. 1-32, Cutler, D.F., Alvin, K.L., Price, C.E., eds. Academic Press, London Holloway, P.J. (1982b) The chemical constitution of plant cutins. In: The plant cuticle, pp. 45-85, Cutler, D.F., Alvin, K.L., Price, C.E., eds. Academic Press, London Hunt, G.M., Baker, E.A. (1980) Phenolic constituents of tomato fruit cuticles. Phytochemistry 19, 1415-1419 Kamp, H. (1930) Untersuchungen fiber Kutikularbau und kutikulfire Transpiration von Blfittern. Jahrb. Wiss. Bot. 72, 403465 Kuraishi, S., Noti, N. (1980) The maximum leaf surface temperatures of the higher plants observed in the inland sea area. Bot. Mag. Tokyo 93, 209-220 Mandelkern, L. (1983) A introduction to macromolecules. Springer, New York Berlin Heidelberg Marsen, H. (1962) Infrarotspektroskopische Untersuchungen zur Temperatur-Abhfingigkeit der Kristallinit/it yon Esterwachsen. Fette Seifen Anstrichm. 64, 41-44 Meyer, M. (1938) Die Submikroskopische Struktur der kutinisierten Zellmembranen. Protoplasma 29, 552-586 Riederer, M., Sch6nherr, J. (1984) Accumulation and transport of (2,4-dichlorophenoxy)acetic acid in plant cuticles: 1. Sorption in the cuticular membrane and its components. Ecotoxicol. Environ. Safety 8, 236-247 Riederer, M., Sch6nherr, J. (1985) Accumulation and transport of (2,4-dichlorophenoxy)aceticacid in plant cuticles: 2. Permeability of the cuticular membrane. Ecotoxicol. Environ. Safety 9, 196-208 Riederer, M., Sch6nherr, J. (1986) Plant cuticles sorb lipophilic compounds during enzymatic isolation. Plant Cell Environ. 9, 459-466 Riederer, M., Sch6nherr, J. (1988) Development of plant cuticles: fine structure and cutin composition of Clivia miniata Reg. leaves. Planta 174, 127-138 Roelofsen, P.A. (1952) On the submicroscopic structure of cuticular cell walls. Acta Bot. Neerl. l, 99-114 Rosenberg, G., Brotz, W. (1957) Beitrag zur Messung der W/irmeausdehnung yon Wachsen. Fette Seifen Anstrichm. 59, 589-594 Sapper, I. (1935) Versuche zur Hitzeresistenz der Pflanzen. Planta 23, 518-566 Schieferstein, R.H., Loomis, W.E. (1959) Wax deposits on leaf surfaces. Plant Physiol. 31,380-384 Schmidt, H.W., Sch6nherr, J. (1982) Development of plant cuticles: occurrence and role of non-ester bonds in cutin of Clivia miniata Reg. leaves. Planta 156, 380-384 Sch6nherr, J. (1982) Resistance of plant surfaces to water loss: transport properties of cutin, suberin and associated lipids. In: Encyclopedia of plant physiology, N.S., vol. 12B : Physiological plant ecology, pp. 153-179, Lange, O.L., Nobel, P.S., Osmond, C.B., Ziegler, H., eds. Springer, Berlin Heidelberg New York
L. Schreiber and J. Sch6nherr: Thermal expansion of plant cuticles
193
Sch6nherr, J., Bukovac, M.J. (1973) Ion exchange properties of isolated tomato fruit cuticular membranes: exchange capacity, nature of fixed charges and cation selectivity. Planta 109, 7393 Sch6nherr, J., Huber, R. (1977) Plant cuticles are polyelectrolytes with isoelectric points around three. Plant Physiol. 59, 145-150 Sch6nherr, J., Lendzian, K. (1981) A simple and inexpensive method of measuring water permeability of isolated plant cuticular membranes. Z. Pflanzenphysiol. 102, 321-327 Sch6nherr, J., Riederer, M. (1989) Foliar penetration and accumulation of organic chemicals in plant cuticles. Rev. Environ. Contamin. Toxicol. 108, 1-70
Sch6nherr, J., Eckl, K., Gruler, H. (1979) Water permeability of plant cuticles: the effect of temperature on diffusion of water. Planta 147, 21-26 Seymour, R. (1971) Introduction to polymer chemistry. McGrawHill, New York Sitte, P., Rennier, R. (1963) Untersuchungen an cuticularen Zellwandschichten. Planta 60, 18-40 Sperling, L.H. (1986) Introduction to physical polymer science. Wiley, New York Wattendorf, J., Holloway, P.J. (1980) Studies on the ultrastructure and histochemistry of plant cuticles: the cuticular membrane of Agave americana L. in situ. Ann. Bot. 46, 13-28
Planta 9 Springer-Verlag1990
Phase transitions and thermal expansion coefficients of plant cuticles The effects o f temperature on structure and function Lukas Schreiber and J~rg Sch~inherr Lehrstuhl f/ir Botanik, Technische Universitgt Mfinchen, Arcisstrasse 21, D-8000 Mfinchen 2, Federal Republic of Germany Received 1 December 1989; accepted 24 March 1990
Abstract. The temperature-induced volume expansion of enzymatically isolated cuticular membranes of twelve plant species was measured. All cuticular membranes exhibited distinct second-order phase transitions in the temperature range of about 40 to 50~ C. Increases in the volumes of fruit cuticles (Lycopersicon, Cucumis, Capsicum, Solanum and Malus) were fully reversible, while leaf cuticular membranes (Ficus, Hedera, Nerium, Olea, Pyrus, Picea and Citrus) underwent irreversible structural changes. Below the phase-transition temperature, volumetric expansion coefficients ranged from 0.39"10 -6 m 3 - k g - l - K -1 to 0.62.10 -6 m 3 . k g - l . K -1, and above from 0 . 6 0 . 1 0 - 6 m 3 . k g - l . K -1 to 1.41. 1 0 - 6 m 3 . k g - l . K - 1 . Densities of cuticles at 25~ ranged from 1020 kg.m -3 to 1 370 kg.m -3. Expansion coefficients and phase transitions were characteristic properties of the polymer matrix as a composite material, rather than of cutin alone. It is argued that the sudden increase of water permeability above the transition temperature, is caused by an increase of disorder at the interface between the polymer matrix and the soluble cuticular lipids. Possible ecological and physiological consequences of these results for living plants are discussed. Key words: Cutin - Cuticle (membrane, density) - Expansion coefficient - Phase transition (cuticular membrane) - Wax
Introduction Cuticles are very effective barriers against passive water loss from plants (Sch6nherr 1982). Under extreme environmental conditions, such as water stress and high temperature, stomata will be closed and the water permeability of cuticular membrane determines the rate of Abbreviations: CM=cuticular membrane; CU=cutin; MX=polymer matrix; SCL = soluble cuticular lipids (waxes)
transpiration (Kamp 1930). The water permeability (P) of cuticles was found to increase with increasing temperature (Sch6nherr et al. 1979). Arrhenius plots (In P vs. l/T) in the range of 5 to 65~ C consisted of two linear portions which had an intersection around 45 ~ C. Moreover, a second measurement using the cuticular membranes previously heated to 65 ~ C displayed an increase in water permeability in the lower temperature range compared with the first measurement. It was concluded that temperatures higher than 45 ~ C caused irreversible structural changes in the membranes, which lead to an increase in water permeability. Plant cuticles are heterogeneous, as is evident from optical anisotropy (Sitte and Rennier 1963), transition electron microscopy (see review by Holloway 1982 a) and transport properties (Sch6nherr and Riederer 1989). Isolated cuticular membranes (CM) consist of an insoluble polymer matrix and soluble cuticular lipids associated with the polymer matrix (Riederer and Sch6nherr 1984). The polymer matrix itself is composed of covalentely linked cutin acids and "non-cutin components", mainly cellulose (Wattendorf and Holloway 1980) and, in minor amounts, amino acids, pectinaceous materials and phenols (Sch6nherr and Huber 1977; Hunt and Baker 1980). These definitions are mainly operational and have been found useful in previous studies into sorption in and diffusion across CM (see review by Sch6nherr and Riederer 1989). They differ to some extent from those used by Sitte and Rennier (1963), who distinguish an outer layer (the cuticle sensu stricto) which does not contain cellulose, and an inner cuticular layer, which does. Eckl and Gruler (1980) attributed the temperature effect on water permeability to a phase transition and reorientation of soluble cuticular lipids, which leads to hydrophilic holes in the barrier. Irreversible membrane alterations were thought to be due to a recrystallisation of soluble cuticular lipids upon cooling that results in a structure different from the original one and leads to a partial preservation of the hydrophilic holes in the barrier. However, our investigations of thermally induced volume expansion of cuticles revealed second-
L. Schreiber and J. Sch6nherr: Thermal expansion of plant cuticles o r d e r p h a s e t r a n s i t i o n s o f t h e p o l y m e r m a t r i x itself. T h i s fact necessitates a new look at the causes of irreversible increase in water permeability of cuticles with temperature.
Material and methods Cutieular membranes. Cuticular membranes of twelve species from nine different plant families were investigated. Adaxial, astomatous leaf cuticles were obtained from Citrus aurantium L., Ficus elastica Roxb. var. decora, Pyrus communis L., Hedera helix L., Nerium oleander L., Olea europaea L. and Pieea omorika Pancic. Fruit cuticles came from Capsicum annuum L., Lycopersicon esculentum L., Solanum melongena L., Malus pumila L. and Cucumis sativus L.. These membranes were subsamples of those used in earlier investigations by Riederer and Sch6nherr (1984, 1985), where additional details (origin, varieties etc.) can be found, with the exception of Citrus and Ficus cuticles. Citrus and Ficus leaves were obtained from trees grown in growth chambers (Geyer and Sch6nherr 1990). Enzymatic isolation of CM followed the procedure of Riederer and Sch6nherr (1986). Isolated cuticles will be referred to as (CM). Treatment of C M with a mixture of chloroform and methanol (1/1, v/v) for 24 h in a soxhlet appartus removed soluble cuticular lipids (SCL) quantitatively. The remainder is called polymer matrix (MX). Incubation of M X with 6 N HCI at 120 ~ C hydrolyzes all "non-cutin c o m p o n e n t s " (HY), such as cellulose, pectinaceous substances and polypeptides, and the insoluble residue remaining is cutin (CU; Sch6nherr and Bukovac 1973; Riederer and Sch6nherr 1984). Measurement of volume expansion. Determination of the volume expansion was carried out in the temperature range from 5 to 65 ~ C. About 300 mg of membrane material and water were added to a specific-gravity bottle (1 cm3). The filled bottle was aspirated extensively to remove air from water and cuticles, which would interfere with the measurements. The pressure of 6.7.103 Pa was sufficient to reduce the gas content of the water to a level which did not interfere with the measurements. A calibrated capillary was attached to the bottle, using groundglass joints sealed with high-vacuum silicone grease (Wacker, Munich, FRG). The bottle with the capillary attached was placed in a thermostatted water bath. Only the tip of the capillary protruded above the surface of the water. Temperature was controlled with an accuracy of + 0.01 K. The rise of the water column in the capillary with increasing temperature was measured using a cathetometer (Spindler and Hoyer, G6ttingen, F R G ; accuracy ___0.05 mm). The volume of the membranes was calculated by substracting the volume of water from the total volume of membrane plus water. Prior to the measurements, several calibration runs were carried out. Determination of the temperature difference between the interior of the bottle filled with CM and water, and the water bath, using a heating rate of 1 K in 4 min, revealed a constant lag of 0.2 K, which did not change over the whole range of 5 to 65 ~ C. The capillary itself was calibrated with mercury in the range of 5 to 65 ~ C (inner cross-sectional area 0.18.10 -6 m2), and no thermally induced volume expansion of the capillary itself could be detected. Calibration measurements, using deionized, degassed water, were carried out to determine the exact volume expansion of water in our apparatus. For three independent measurements the results could be repeated with a coefficient of variation of 0.5%. The volume expansion of a 0.01 M Mes buffer (2(N-morpholino)ethanesulfonic acid; Sigma, St. Louis, La., USA) adjusted with Ca(OH)2 to pH 6.0 was also determined. Results could be reproduced with good accuracy in the lower temperature range up to 4 5 ~ (coefficient of variation 0.5%). In the upper temperature range, however, the coefficient of variation of three callibration runs was 3%.
187
Calculation of volume expansion coefficients. The volume V (m 3) per kg of cuticle in the temperature range of 5 to 65 ~ C was calculated according to the equation t
5
V=(I-1)-L.M Mc
t
w
Ak
Eq. (1)
where 15 and It (m) denote the height of the water meniscus in the capillary at the experimental temperature T and at 5~ C, respectively, when the bottle was filled with cuticular membranes and water. L' ( m . k g - 1 ) is the height of the meniscus with pure water at the given temperature, as determined in the callibration procedure. M w and M c (kg) represent the masses of water and cuticular membrane, respectively, and A ~ (m 2) is the inner cross-sectional area of the capillary. Linear portions in the plot V (volume) versus T (temperature) were identified and regression lines were fitted to the lower and upper parts of the volume-expansion curves. Coefficients of correlation were always better than 0.99 in the lower, and better than 0.98 in the upper part of the curve. The slopes b of the regression lines give the absolute volume expansion coefficients in m3"kg - iK-~. The intersection of the two regression lines represents the midpoint of the phase transition. The region of the phase transition itself generally ranged over 10-15 ~ C. Relative volume expansion coefficients a (~ C-1) were calculated using Eq.2: a - (V'/VS)- 1 T-- 5
Eq. (2)
where V t (m 3) and V s (m 3) are the volumes of the cuticle at the experimental temperature and at 5~ C, respectively, and T (~ C) denotes the experimental temperature.
Estimation of experimental error. The reading accuracy of the cathetometer was 0.05 m m which is equal to a volume of about 0.01 m m 3. Experimental results thus vary in the second decimal point by about +0.01.10 -6 m z. For a volume expansion coefficient of 0.6.10-6 m 3 - k g - l . K - i , a deviation of + 0 . 0 1 . 1 0 - 6 m 3 corresponds to an error of 1.7%. Four successive measurements of the same sample of tomato fruit cuticle (Table 1) gave a mean volume expansion of 36.43. 10 - 6 m 3.kg 1 with a standard deviation of 0.46.10 - 6 , which corresponds to a coefficient of variation (cv) of 1.3%. The absolute volume-expansion coefficients of the same membranes below and above the phase transition were 0.57' 10- 6 m 3 . k g - 1. K - 1 + 0.005. 10 - 6 (CV= 1.0%) and 0.69.10 - 6 m 3 . k g - 1 - K - 1 ___0.02.10 - 6 (CV = 3.3%). The midpoint of the phase transition is 46.0 ~ C___0.6 (cv = 1.3%). It follows from these considerations that below the phase transition an experimental error of about 1.5-2.0% is associated with the data mainly due to the limits of the reading accuracy of the cathetometer. Above the phase transition the error of the volume expansion coefficient was about twice as large. Determination of water permeability. Water permeabilities of isolated Citrus leaf C M were determined according to the method described by Sch6nherr and Lendzian (1981), with the exception that the transpiration chambers were made of brass rather than Plexiglas. After the permeability of the native C M had been determined, the wet cuticles were stored for 20 h at 65 ~ C. The effect of heat treatment on water permeability was measured immediately after treatment. A third determination of water permeability with the same cuticles was conducted 100 d later, after they had been stored in dry form at room temperature. Permeance P ( m - s - 1 ) was calculated from the steady-state flow rate of water F (kg. s-1) across the cuticle according to Eq.3: F P=A-d'Aaw
Eq. (3)
where A (m 2) is the exposed membrane area, d ( k g . m - 3 ) is the density of liquid water at 25 ~ C and d a represents the gradient of water activity across the cuticle, w
188
L. Schreiber and J. Sch6nherr: Thermal expansion of plant cuticles
Results
Table 1. Volume expansion of isolated cuticular membranes from selected plant species in the temperature range of 5 to 65 ~ C
Volume expansion of cuticular membranes (CM). Plots of volume o f C M versus temperature exhibited two or three linear portions (Figs. 1, 2). The transition region from the lower to the upper volume-expansion curve was in most cases in the range f r o m 40 to 50 ~ C. The difference between the volume-expansion coefficient below and above the phase transition was significant at the 95% level in all cases. Fruit CM from five species responded similarly to temperature (Table 1). The absolute volume expansion from 5 to 65 ~ C ranged from 32.4.10 -6 m 3 .kg -1 (Capsicum CM 2) to 4 2 . 8 . 1 0 - 6 m 3 . k g -1 (Malus). The midpoints of the phase transitions reached from 40.2~ C (Capsicum CM 2) to 54.5 ~ C (Malus). Variation between different samples from the same species can be fairly
Transition Volume expansion midpoint coefficient (~ C) ( . 1 0 6 m 3 - k g - l - K -1)
Total volume expansion (-106 m3.kg -1)
Below Above transition transition Fruit CM Lycopersicon a
Measurement Measurement Measurement Measurement
1 2 3 4
45.9 46.8 45.5 46.0
0.57 0.57 0.57 0.56
0.70 0.69 0.66 0.71
36.9 36.4 35.8 36.6
46.2 40.2 42.3
0.57 0.51 0.56
0.63 0.60 0.60
35.5 32.4 34.5
(H20) (Ca z§
43.1 51.4
0.52 0.51
0.81 0.68
37.2 32.9
Malus
54.5
0.48
1.88
42.8
Cucumis
43.9
0.51
0.63
33.0
Measurement 1 48.8 Measurement 2 44.2
0.55 0.56
0.64 0.61
34.5 34.8
Pyrus
46.2
0.62
1.41
49.8
Nerium
38.5
0.39
0.62
29.4
Olea
55.3
0.45
0.70
29.1
Picea
49.3
0.49
0.68
32.2
Capsicum b
CM 1 CM 2 CM 3 Solanum"
Lycopersicon CH
60
"7
Pyrus CH
Citrus CH
50
Leaf CM
40
Hedera a
.~
30
m
20 > 10
i
0
20
40
20
60
40
60
i
20
40
i
60
" Different measurements using the same sample b Different samples of the same species
Temperature [~
Fig. 1. Volume expansion of the cuticular membranes (CM) of the species Lycopersicon, Pyrus and Citrus
60
Ficus CH
Ficus HX
Ficus CU
~- 50 ~, 40..2-
.~ 30 ~
20
o
> 10
0
,
,
,
20
40
60
0
20
40
60
0
20
40
60
Temperature [~
Fig. 2. Volume expansion of cuticular membrane (CM), polymer matrix (MX) and cutin (CU) of Ficus
high as results with Capsicum CM demonstrate. Temperature effects were reversible with fruit CM as shown for Lyr where results of four repeated determinations were identical within experimental error (Table 1). The only exception was Malus where the upper volume expansion coefficient in the second experimental run was significantly lower than in the first determination (data not shown). Leaf CM from seven species showed greater variation in their volume expansion behavior than fruit CM (Tables 1, 2). The absolute volume expansion ranged from 29.1-10 - 6 m3-kg -1 (Olea) to 49.8-10 - 6 m3-kg -1 (Pyrus) and the midpoint of the transition from the lower to the upper temperature extended from a minimum o f 38.5 ~ C (Nerium) to a maximum of 55.3 ~ C (Olea). Another striking difference between leaf and fruit CM was the irreversible effect of high temperatures on volume expansion. Repreated measurements using the same sample of Hedera CM (Table 1) showed a significant increase of the lower volume expansion coefficient from the first to the second measurement and a significant decrease of the volume expansion coefficient in the upper temperature range. The transitions became less distinct.
L. Schreiber and J. Sch6nherr: Thermal expansion of plant cuticles
189
Table 2. Volume expansion of leaf cuticular membranes of Citrus and Ficus between 5 and 65 ~ C. Temperature ranges = below the
first, between the first and the second, and above the second transition midpoint
Transition midpoints (o c)
Volume expansion coefficient (. 106 m3.kg -1 .K-i)
Total volume expansion (.10 6 m3.kg -i)
Below first transition
Below second transition
Above second transition
Citrus C M a
Adaxial C M Abaxial C M
17.1
42.3 44.8
0.47
0.53 0.50
0.74 0.60
36.2 32.0
14.0 24.6 19.1 18.1
46.0 48.3 50.7 49.9
0.64 0.60 0.66 0.64
0.61 0.58 0.57 0.55
0.92 1.00 1.00 0.95
43.8 42.4 42.0 40.6
Ficus C M ~
(Leaf (Leaf (Leaf (Leaf
1) 2) 7) 11)
a Different samples of the same plant
Table 3. Volume expansion of cuticular membrane (CM), polymer matrix (MX) and cutin (CU) of the species Ficus, Capsicum and Citrus
Transition midpoints
(o c)
Fieus C M Ficus M X Ficus C U
18.1
Capsicum CM Capsicum M X Capsicum C U Citrus C M Citrus M X Citrus C U
17.2 21.2
Volume-expansion coefficient (.106 mS.kg - i . K - i )
Total volume expansion (.106 m3.kg -1)
Below first transition
Below second transition
Above second transition
49.9 52.8
0.64 0.66 0.96
0.55
0.95 1.13
40.6 45.5 58.2
42.3 39.4
0.56 0.63 0.82
0.60 0.65
34.5 38.6 48.6
42.3 43.7
0.47 0.50 0.95
0.74 0.79
36.2 40.2 57.4
The other leaf CM, examined here, showed the same irreversibility, with the exception of Citrus and Pieea. Here the increases of the lower volume expansion coefficients in the second determination were not significantly different from the first determination (data not shown). Results obtained with Ficus and Citrus CM are less uniform (Table 2). The Citrus CM from the abaxial side o f the leaf responded to temperature like the CM o f the other six species. However the adaxial Citrus CM (Fig. 1) and Ficus CM (Fig. 2) showed a remarkable deviation. There were three distinct regions with three different volume expansion coefficients in the temperature range from 5 to 65 ~ C. The first transition occurred at about 15-20 ~ C and the second at about 40-50 ~ C. With Solanum fruit CM it was observed that volume expansion was affected by calcium ions. Determination of the volume expansion of CM in calcium buffer resulted in a smaller total volume expansion and the transition midpoint rose compared with the volume expansion of the same sample in water (Table 1). Ficus leaf CM of increasing age displayed a slight decrease in the total volume expansion and a slight in-
0.53 0.65
crease of the transition midpoint (Table 2). However, differences within the lower volume expansion coefficients o f the four stages of development were not significant at the 95% level. The same was true for the upper volume expansion coefficients.
Volume expansion of polymer matrix (MX) and cutin (CU). The MX of all three species examined (Ficus, Capsicum, Citrus) exhibited a volume-expansion behavior similar to that observed with CM (Table 3). The typical transitions found for the CM were also observed with the MX, whereas with CU they disappeared and volume expansion was linear between 5 and 65 ~ C (Ficus CM Fig. 2). The total volume expansions of Ficus, Capsicum and Citrus increased in the order CM, MX and CU (Table 3). The densities of CM and M X from Ficus, Capsicum and Citrus were all larger than 1.103 k g . m -3 (Table 4). The densities of CU of the species Ficus, Capsicum and Citrus were below 1-10 ~ k g . m -3. Cutin ( C U ) a m o u n t e d to about two-thirds of the total weight of the cuticular membrane (CM), while the hydrolyzable fraction (HY),
190
L. Schreiber and J. Sch6nherr: Thermal expansion of plant cuticles
Table 4. Densities of cuticular membrane (CM), polymer matrix (MX) and cutin (CU) at 25~ C. The densities are calculated from the density of the cuticle at 5~ C plus the measured volume expansion from 5 to 25~ C. Values are means of 3-5 measurements. Coefficients of variation were less than 1.5% in all cases
and especially the soluble cuticular lipids (SCL) varied appreciably among species (Table 5).
Density (.10 -3 kg.m 3) CM
MX
CU
Leaf cuticles Ficus Hedera Pyrus Nerium Olea Picea Citrus
1.02 1.09 1.07 1.17 1.05 1.37 1.04
1.04
0.92
1.03
0.94
Fruit cuticles Lycopersicon Capsicum Solanum Malus Cucumis
1.12 1.10 1.06 1.10 1.14
1.10
0.96
Table 5. Composition of Ficus, Capsicum and Citrus cuticular membranes (CM) as a percentage of the weight Composition (weight %)
Ficus Capsicum Citrus
CU a
SCL b
HY c
62 64 72
20 10 6
18 26 22
a Cutin (CU) fraction b Fraction of soluble cuticular lipids (SCL) ~ "Non-cutin components" (HY) obtained through hydrolysis
Table 6. Influence of heat treatment and storage on water permeabilities of isolated Citrus CM. A denotes the permeances of the untreated cuticles, B the permeances after treatment with 65~ and C the permeances after storage for a hundred days. The ratios B/A and C/A are a measure of the effects of heat and storage, respectively. Mean is the average of the six membranes and cv represents the coefficient of variation Permeance (- 101~m . s - 1) CM
A
B
B/A
C
C/A
1
2.81 2.67 2.35 2.71 1.89 1.90
8.18 5.10 3.72 9.06 7.91 5.77
2.91 1.91 1.58 3.34 4.18 3.04
3.31 3.30 1.83 2.88 2.99 2.56
1.18 1.39 0.78 1.06 1.58 1.35
2.39 17
6.62 31
2.83 34
2.81 20
1.17 23
2 3 4 5 6 Mean cv (%)
Water permeability o f Citrus ( C M ) . A heat treatment of wet CM for 20 h at 65 ~ C caused an average increase of the water permeability by a factor of 2.83 (Table 6). Storage of these cuticles at room temperature for 100 d led to a significant reduction of water permeability. The original permeabilities (prior to heat treatment) were almost reestablished. The variation of the water permeabilities between individual membranes as well as the influence of the heat treatment on the permeabilities was high.
Discussion
Phase transitions of plant cuticles, as they were already described for Citrus and Hedera (Sch6nherr et al. 1979; Eckl and Gruler 1980), could also be demonstrated using CM and M X of ten additional species. The phase transitions, described here, are transitions of the second order, which are characterised by a relatively wide region of transition over 10 or even 15 K (Seymour 1971). Secondorder phase transitions are typical properties of macromolecules, synthetic ones as well as biopolymers (Mandelkern 1983). Most leaf CM (except those of Citrus and Pieea) underwent irreversible changes when heated to 65 ~ C, while this was observed for only one fruit CM (Malus). This suggests differences in polymer structures between leaf and fruit CM. A correlation between cutin type (C16-, C18- and C16-/C18-type) according to Holloway (1982 b) and reversibility of temperature effects does not exist. Cutin types C16-, C18- and epoxy-fatty acids occur in both reversible and irreversible CM. An increasing number o f cross-links, due to the formation of ether bonds between epoxy- and hydroxylfunctions, occur during cutin development of Clivia (Schmidt and Sch6nherr 1982; Riederer and Sch6nherr 1988). This should result in an increase of the midpoint o f the phase transition and a decrease of the absolute volume expansion, as is frequently described with synthetic polymers (Sperling 1986). The Ficus MX contains epoxy-fatty acids and a slight decrease of the absolute volume expansion from the 1st to the 1lth leaf (Table 2) and a slight increase of the midpoint of the phase transition of 4 K were in fact detectable. Isolated CM and MX contain carboxyl groups which have a high selectivity for Ca 2 + (Sch6nherr and Bukovac 1973). This compensation of negative charges within the cuticle results in a physical form of cross-linking and an influence on volume expansion of CM can be expected and was indeed observed with Solanum CM (Table 1). In the presence of calcium buffer, absolute volume expansion decreased and the midpoint of the phase transition rose. How do individual classes of components such as cutin, soluble lipids or cellulose contribute to secondorder phase transitions? The volume expansion of the M X was not affected by the presence of SCL. This was
L. Schreiber and J. Sch6nherr: Thermal expansion of plant cuticles
191
expected, since SCL are not crosslinked to the polymer matrix. The absolute volume expansion as well as the lower and upper volume expansion coefficients of Ficus, Capsicum and Citrus MX increased by about 12%, compared with the CM. This can be attributed to a decrease in weight on extracting SCL (Table 5). The values o f the volume expansion are based on the weight o f the membrane material used in the experiment. The increase of the volume expansion of the MX, expressed in percent, is similar to the decrease in weight of the CM on extraction o f SCL. Comparing the volume expansions of cutin and the polymer matrix (Table 3) shows that phase transitions are a characteristic property of the MX rather than CU. The phase transitions of all three species at about 45 ~ C as well as the second phase transition of Citrus and Ficus between 15 and 20 ~ C disappeared and a linear volume expansion in the whole temperature range was obtained when cutin samples were studied. The cutin polymer (CU) showed an increase o f the absolute volume expansion of about 30-40% compared with the MX. The loss in weight o f the MX, due to hydrolysis, is too low fully to account for the distinct increase of the volume expansion of CU. The absolute volume expansion of cutin can be attributed to a significantly larger volume expansion coefficient (Table 3). The differences in thermal expansion of polymer matrix and cutin are most likely due to cellulose contained in the polymer matrix (Wattendorf and Holloway 1980). The low relative thermal-expansion coefficient of cellulose of about 4.10 - 4 (Brandrup and Imergut 1966) reduces the high thermal volume expansion of cutin, which is more than twice as large (Table 7). Relative expansion coefficients of CM and MX of these species had values located between those for cellulose and cutin. Thermalexpansion coefficients of CM and MX are mixed quantities, which result from the low volume expansion coefficient of cellulose and the high coefficient of cutin.
A comparison of the densities of CM, M X and CU provides further evidence of the decisive role o f cellulose in cuticular membranes. Cuticular membranes had densities about 1.1.103 k g . m -3 (Table 4), and M X membranes did not differ significantly from that value (Table 4). Hydrolyzed cuticles, lacking cellulose, had densities around 0.94-103 k g . m -3 (Table 4) and they exhibited a linear volume expansion in the temperature range between 5 and 65 ~ C. Density o f crystalline cellulose is about 1.4.103 k g . m -3 (Brandrup and Imergut 1966), and therefore the decrease of the density of CM after hydrolysis is partially due to the relatively high density of cellulose. But the increase o f the total volume expansion o f CU compared with MX is not fully compensated by the loss of weight of the MX, caused by hydrolysis. Cellulose associated with cutin seems to reduce the intermolecular distance of the polymer chains of the cutin. Only above 40-50 ~ C do cutin monomers gain sufficient intermolecular space to perform additional molecular motions. This results in a phase transition, as it is generally described for the occurrence of phase transitions in synthetic polymers (Aharoni 1974; Greiner and Schwarzl 1989). A moderate but significant correlation (r = - 0.91) between densities of CM, M X and C U and the respective volume expansions of the species Ficus, Capsicum and Citrus exists (Fig. 3). This again indicates that a lower density hinders the molecular motion o f the polymer chains less than a higher density. Below 50 ~ C, volume expansion of crystalline natural and synthetic waxes is linear and seven different samples had a mean relative volume expansion coefficient o f about 5.7-10 - 4 (CV = 3 0 % ) (Rosenberg and Brotz 1957). Above 50-60 ~ C, volume expansion coefficients increased steadily with rising temperature up to the solidliquid phase transition at about 80-100~ (Marsen 1957; Rosenberg and Brotz 1957). These results are in good agreement with cuticle behavior. Waxes (SCL) associated with the polymer matrix at physiological tem-
Table 7. Relative volume expansion coefficients of cuticular membrane (CM), polymer matrix (MX) and cutin (CU) of the species Ficus, Capsicum and Citrus and of cellulose and waxes. The relative volume expansion coefficient a is calculated according to Eq.2. The subscripts below, middle and above refer to the respective linear portions of the volume-expansion curves
60 CU
o
Capsicum
Citrus
6.4 5.5 9.5 6.6
5.6
11.0 9.6
6.5 8.2
4.7 5.4 7.4 5.0 6.5 7.9 9.8
x
CU
55
a (. 104 K - 1)
Ficus
9
S0 t,5
CM (below) CM (middle) CM (above) MX (below) MX (middle) MX (above) CU Cellulosea Wax b
4.0 5.7
6.0 6.2
m x
t~0
CHx
35 30 0.85
o. ,o
o. ,5
1. ,o
~+
.o5
CH
1. s
Density [.10 3 kg m-3]
Fig. 3. Correlation between absolute volume expansion and density
Brandrup and Immergut 1966 b Rosenberg and Brotz 1957
of cuticular membrane (CM), polymer matrix (MX) and cutin (CU) of the species Ficus, Capsicum and Citrus
192
L. Schreiber and J. Sch6nherr: Thermal expansion of plant cuticles
peratures are crystalline (Meyer 1938; Roelofsen 1952). They do not have a distinct melting point but indeed a wide melting range from 60 to 90 ~ C (Sitte and Rennier 1963). Relative volume expansion coefficients of SCL, CM and M X are similar in the lower temperature range. Above the phase transition the volume expansion of the polymer matrix exceeds that of the waxes because the moderately increasing expansion coefficients of the waxes cannot compensate for the sudden increase of the expansion coefficient of the polymer matrix at 40-50 ~ C. The difference between the two volume expansion coefficients above the phase transition will lead to defects between SCL and MX. These additional defects persist when the temperature is decreased again to physiological ranges, and they are responsible for the increase in water permeability o f Citrus CM (Table 6). However, determination of the water permeability 100 d later showed that these temperature-induced defects can heal (Table 6). Such a healing process was observed earlier by Fox (1958) using paper coated with paraffin wax. Water permeabilities of these membranes decreased steadily with increasing storage times because a change in the structure of the paraffin crystals led to fewer defects. We suggest that the healing of plant cuticles is also a consequence of recrystallisation of wax crystals and rearrangement of disordered waxes in the polymer matrix. In their natural environment, plants are sometimes exposed to temperatures o f 45-50 ~ C (Sapper 1935; Kuraishi and Nito 1980). They are able to tolerate those extreme conditions for an appreciable amount of time (Sapper 1935). Species having scleromorphic leaves (Nerium, Olea and Picea) had the lowest volume expansions (Table 1). This might be an ecological adaptation to dry and warm environments, because a low volume expansion o f the cuticle should minimize the eventually hazardous effect of an increased water permeability caused by thermally induced structural changes in the cuticle. A continuous synthesis and extrusion of cuticular lipids by mature leaves and fruits can be frequently observed (Schieferstein and Loomis 1959). This process helps to repair defects in the cuticle caused by natural influences such as wind, rain and temperature. Soluble cuticular lipids are responsible for the excellent barrier qualities of plant cuticles (Sch6nherr 1982). A continuous synthesis and extrusion of soluble lipids should be essential for a plant in order to maintain the functional integrity of the cuticle. Plant cuticles are not simply inert physical systems once synthesized. They are subject to permanent environmental stress. Both physical (rearrangement of wax molecules) and physiological (wax synthesis) processes cooperate and maintain the structural integrity of cuticles necessary for survival.
References
The authors greatfully acknowledge stimulating discussions with Drs. H. Gruler (Exp. Physik 3, Universitfit Ulm, FRG) and M. Riederer (Institut ffir Botanik und Mikrobiologie, Technische Universitfit Mfinchen, Mfinchen, FRG) and financial support by the Deutsche Forschungsgemeinschaft.
Aharoni, S.M. (1974) Thermal dilatation of polymers. J. Macromol. Sci. Phys. 94, 699-731 Brandrup, J., Immergut, E.H., eds. (1966) Polymer handbook. Wiley, New York Eckl, K., Gruler, H. (1980) Phase transitions in plant cuticles. Planta 150, 102-113 Fox, R.C. (1958) The relationship of wax crystal structure to the water vapor transmission rate of wax films. J. Tech. Assoc. Pulp Paper Ind. 41,283-289 Geyer, U., Sch6nherr, J. (1990) The effect of the environment on the permeability and composition of Citrus leaf cuticles. 1. Water permeability of isolated cuticular membranes. Planta 180, 147-153 Greiner, R., Schwarzl, F.R. (1989) Volume relaxation and physical aging of amorphous polymers. 1. Theory of volume relaxation after single temperature jumps. Colloid Polymer Sci. 267, 39-47 Holloway, P.J. (1982a) Structure and histochemistry of plant cuticular membranes: an overview. In: The plant cuticle, pp. 1-32, Cutler, D.F., Alvin, K.L., Price, C.E., eds. Academic Press, London Holloway, P.J. (1982b) The chemical constitution of plant cutins. In: The plant cuticle, pp. 45-85, Cutler, D.F., Alvin, K.L., Price, C.E., eds. Academic Press, London Hunt, G.M., Baker, E.A. (1980) Phenolic constituents of tomato fruit cuticles. Phytochemistry 19, 1415-1419 Kamp, H. (1930) Untersuchungen fiber Kutikularbau und kutikulfire Transpiration von Blfittern. Jahrb. Wiss. Bot. 72, 403465 Kuraishi, S., Noti, N. (1980) The maximum leaf surface temperatures of the higher plants observed in the inland sea area. Bot. Mag. Tokyo 93, 209-220 Mandelkern, L. (1983) A introduction to macromolecules. Springer, New York Berlin Heidelberg Marsen, H. (1962) Infrarotspektroskopische Untersuchungen zur Temperatur-Abhfingigkeit der Kristallinit/it yon Esterwachsen. Fette Seifen Anstrichm. 64, 41-44 Meyer, M. (1938) Die Submikroskopische Struktur der kutinisierten Zellmembranen. Protoplasma 29, 552-586 Riederer, M., Sch6nherr, J. (1984) Accumulation and transport of (2,4-dichlorophenoxy)acetic acid in plant cuticles: 1. Sorption in the cuticular membrane and its components. Ecotoxicol. Environ. Safety 8, 236-247 Riederer, M., Sch6nherr, J. (1985) Accumulation and transport of (2,4-dichlorophenoxy)aceticacid in plant cuticles: 2. Permeability of the cuticular membrane. Ecotoxicol. Environ. Safety 9, 196-208 Riederer, M., Sch6nherr, J. (1986) Plant cuticles sorb lipophilic compounds during enzymatic isolation. Plant Cell Environ. 9, 459-466 Riederer, M., Sch6nherr, J. (1988) Development of plant cuticles: fine structure and cutin composition of Clivia miniata Reg. leaves. Planta 174, 127-138 Roelofsen, P.A. (1952) On the submicroscopic structure of cuticular cell walls. Acta Bot. Neerl. l, 99-114 Rosenberg, G., Brotz, W. (1957) Beitrag zur Messung der W/irmeausdehnung yon Wachsen. Fette Seifen Anstrichm. 59, 589-594 Sapper, I. (1935) Versuche zur Hitzeresistenz der Pflanzen. Planta 23, 518-566 Schieferstein, R.H., Loomis, W.E. (1959) Wax deposits on leaf surfaces. Plant Physiol. 31,380-384 Schmidt, H.W., Sch6nherr, J. (1982) Development of plant cuticles: occurrence and role of non-ester bonds in cutin of Clivia miniata Reg. leaves. Planta 156, 380-384 Sch6nherr, J. (1982) Resistance of plant surfaces to water loss: transport properties of cutin, suberin and associated lipids. In: Encyclopedia of plant physiology, N.S., vol. 12B : Physiological plant ecology, pp. 153-179, Lange, O.L., Nobel, P.S., Osmond, C.B., Ziegler, H., eds. Springer, Berlin Heidelberg New York
L. Schreiber and J. Sch6nherr: Thermal expansion of plant cuticles
193
Sch6nherr, J., Bukovac, M.J. (1973) Ion exchange properties of isolated tomato fruit cuticular membranes: exchange capacity, nature of fixed charges and cation selectivity. Planta 109, 7393 Sch6nherr, J., Huber, R. (1977) Plant cuticles are polyelectrolytes with isoelectric points around three. Plant Physiol. 59, 145-150 Sch6nherr, J., Lendzian, K. (1981) A simple and inexpensive method of measuring water permeability of isolated plant cuticular membranes. Z. Pflanzenphysiol. 102, 321-327 Sch6nherr, J., Riederer, M. (1989) Foliar penetration and accumulation of organic chemicals in plant cuticles. Rev. Environ. Contamin. Toxicol. 108, 1-70
Sch6nherr, J., Eckl, K., Gruler, H. (1979) Water permeability of plant cuticles: the effect of temperature on diffusion of water. Planta 147, 21-26 Seymour, R. (1971) Introduction to polymer chemistry. McGrawHill, New York Sitte, P., Rennier, R. (1963) Untersuchungen an cuticularen Zellwandschichten. Planta 60, 18-40 Sperling, L.H. (1986) Introduction to physical polymer science. Wiley, New York Wattendorf, J., Holloway, P.J. (1980) Studies on the ultrastructure and histochemistry of plant cuticles: the cuticular membrane of Agave americana L. in situ. Ann. Bot. 46, 13-28
