The cuticles of plants provide a multifunctional interface between the plants and their environments. The cuticle, with its associated waxes, is a protective layer that minimizes water loss by transpiration and provides several functions, such as hydrophobicity, light reflection and absorption of harmful radiation. The self-healing of voids in the epicuticular wax layer has been studied in 17 living plants by atomic force microscopy (AFM), and the process of wax film formation is described. Two modes of wax film formation, a concentric layer formation and striped layer formation, were found, and the process of multilayer wax film formation is discussed. A new method for the preparation of small pieces of fresh, water-containing plant specimens for AFM investigations is introduced. The technique allows AFM investigations of several hours duration without significant shrinkage or lateral drift of the specimen. This research shows how plants refill voids in their surface wax layers by wax self-assembly and should be useful for the design of self-healing materials.
The epidermis, as the outermost cell layer of the primary tissues of all leaves and several other organs of plants, plays an important role in environmental interactions and surface structuring. The outer part of epidermis cells, schematically shown in figure 1, is an extracellular membrane called the cuticle. The cuticle is basically a biopolymer, composed of cutin and integrated and superimposed waxes. Waxes form the main transport barrier to reduce the loss of water by transpiration and reduce leaching of molecules from inside the cells (Riederer & Schreiber 1995, 2001). Wax on the cuticle is called epicuticular wax; wax located in the cutin network is called intracuticular wax. Epicuticular waxes play an important role in surface wettability (Barthlott & Neinhuis 1997; Koch et al. 2008). In some species, waxes reduce the adhesion of insects (Gorb et al. 2005) and, in others, they are responsible for the reflection of visible light and absorption of harmful UV radiation (Barnes & Cardoso-Vilhena 1996; Koch et al. 2009).
Most plant waxes are a complex mixture of long-chain aliphatic components, e.g. primary and secondary alcohols, aldehydes and ketones, while others contain high amounts of cyclic components, such as triterpenes or flavonoids (Jeffree 2006). Epicuticular waxes occur in different morphologies, such as tubules, platelets, threads and thin films or thicker crusts, which originate by self-assembly (Barthlott et al. 1998). The different morphologies are based on differences in chemical composition. The combination of three-dimensional wax crystals with an underlying wax film has been reported for many species (Jeffree et al. 1975; Barthlott et al. 1998; Koch & Ensikat 2008). The crystalline nature of the wax of many species has been verified by X-ray and electron diffraction (Ensikat et al. 2006). Wax films are often incorrectly referred to as an ‘amorphous’ layer, more a morphological description than a crystallographic one. On several plant surfaces, the wax film consists of only a few molecular layers, which is hardly visible in the scanning electron microscope (SEM). However, by mechanical isolation of the epicuticular waxes, e.g. freezing in glycerol (Ensikat et al. 2000), the waxes can be removed from the cuticle and transferred onto a smooth artificial substrate for microscopic investigations. By this method, the edges of the wax film can be detected, and the film thicknesses can be determined (Koch & Ensikat 2008).
Plant waxes are synthesized inside the living part of the epidermis cells, but the transport mechanisms of the wax molecules through the outermost hydrophilic cellulose wall and the cuticle are still under discussion. According to one hypothesis, epicuticular wax protrusions come out of pores that are continuous through the cuticle (Baker 1982; Anton et al. 1994), but such pores were later described as artefacts. More recently, it has been shown that lipidic compounds, such as waxes, diffuse through the cuticle via a lipidic pathway (Riederer & Schreiber 1995; Buchholz & Schönherr 2000), whereas water and polar molecules pass through the cuticle via a polar pathway called ‘aqueous pores’. Modelling and calculation of the molecular structure of the cuticular matrix revealed an average aqueous pore radius between 0.3 and 0.5 nm (Popp et al. 2005; Schönherr 2006). These values are much smaller than the accessible resolution of atomic force microscopy (AFM) on plant cuticles, and thus such pores have not been visualized until now. After the wax has moved through the plant cuticle, the different wax morphologies grow by crystallization or ‘self-assembly’ (Koch et al. 2004; Jeffree 2006). Most epicuticular waxes are soft and fragile structures and can easily be damaged. Thus, it is of interest to know whether plants are able to repair voids within the epicuticular wax layer and whether this process is based on wax self-assembly or other processes.
Epicuticular wax structures usually occur in a range from 0.2 to 100 μm in thickness; thus, the most suitable microscopy technique for studying the self-assembly process of waxes under ambient conditions is AFM (Koch et al. 2004, 2006a,b). During AFM investigation, loss of water from inside the plant material should be minimized to reduce specimen drift by cell shrinkage. Water loss can be reduced by investigation of intact leaves, as shown in figure 2. However, this precondition limits the selection of species, because most leaves and other organs of plants are much too large to mount in most AFM specimen chambers without cutting them.
In this paper, we introduce a new preparation method for AFM investigation of water-containing leaves and shoots, which allows the cutting of leaves and prevents desiccation of the specimen for several hours. Previous publications were limited to observations of a few plant species, which had been examined as entire intact plants (Koch et al. 2004). Here, we present the results of wax film formation on 17 species. Wax regeneration was studied in 10 species with intact leaves, and seven species have been investigated by using small excised pieces of leaves or shoots.
2. Material and methods
(a) Plant material
All 17 species investigated were cultivated in and provided by the Botanical Gardens of the Rheinische Friedrich-Wilhelms University of Bonn (BG Bonn). In table 1, species names, their accession numbers and wax morphologies of the upper leaf sides (adaxial) are provided.
(b) Atomic force microscopy
A NanoScope IIIa (Digital Instruments, Mannheim, Germany) with a z-piezo with 10 μm range was used. Tapping mode and silicon tapping-mode tips (Tap300Al) with a resonant frequency of approximately 300 kHz and force constant of 40 N m−1 were used (NanoWorld NCHR, JPK Instruments, Berlin, Germany). The tip radius was approximately 10 nm. To prevent thermal effects of the AFM laser beam on the plant surface, cantilevers with a reflective coating were used, and beam intensity was further reduced by integrating an attenuation filter above the cantilever. Plant waxes are usually relatively fragile and can be rapidly damaged when scanned at high magnifications, employing a scan size of less than 1 μm. Appropriate AFM conditions turned out to be a scan size of 3×3 to 20×20 μm2, a scan rate of 0.7–2 Hz (lines per second) and a setpoint near the upper limit to minimize the interaction between the tip and the sample.
(c) Image processing
The images presented here are ‘amplitude images’, which clearly show the finer details. For wax film thickness measurements, height images have to be used, which are recorded simultaneously. The accumulation of waxes has been visualized by calculating the difference in wax growth between two 16-bit greyscale images, recorded at different stages during wax regeneration. Therefore, one image was inverted (negative), adjusted to a matching position with the second (positive) image in a transparent view, and added using standard image processing software (Photoline 32, Computerinsel).
(d) Specimen preparation
The repair of induced voids in epicuticular wax layers was studied on adaxial (upper) leaf surfaces after removing the original wax layer in small areas (figure 1). A drop of two-component glue (UHU-plus-Schnellfest 2-Komponentenkleber, Henkel, Düsseldorf, Germany) was applied onto the leaf surfaces and was removed after hardening. The areas on which the glue was applied were approximately 3–5 mm in diameter. To obtain a completely clean surface, the wax removal procedure was repeated twice.
The Nanoscope IIIa has a quite small specimen chamber, which is designed to carry samples of approximately 15 mm in diameter. Therefore, two experimental set-ups of plant preparation were used. The first one is for small leaves that fit completely or with their apical leaf tips into the specimen chamber. In this preparation method, the leaf is still connected to the entire plant, which was placed close to the AFM, as shown in figure 2. The apical parts of the leaf were mounted on the sample holder plate with the same glue as used for the wax removal, placing the dewaxed area in the centre. After the second wax removal procedure, the AFM measurements started immediately, and first images could be obtained after a few minutes.
AFM investigations of species with larger leaves or stems of succulent plants required a different preparation technique. The specimen preparation is shown in figure 3. Pieces of plant leaves or stems were placed with their cut faces in wet cellulose fibres to avoid desiccation during long-term investigations for several hours. The cellulose fibres were produced by mechanical crushing of filter paper (grade 594; Schleicher and Schuell, Dassel, Germany). During the AFM measurements, a few droplets of water had to be added occasionally to moisten the cellulose fibres. Specimens were fixed at two edges with a drop of fast-drying glue (UHU-plus-Schnellfest 2-Komponenten Epoxidharz-Kleber, Henkel, Düsseldorf, Germany) on the sample holder. To avoid escape of water into the AFM body, the standard metal sample holders were hydrophobized with paraffin. To avoid drift of the specimen, a reliable mounting with the glue is important, but nevertheless some drift may occur. If the same position should be observed for several hours, the sample position needs readjustment.
The investigations have been performed twice with different leaves or shoots.
3. Results and discussion
The new preparation technique provides a water reservoir to keep pieces of leaves or shoots hydrated. This prevents desiccation and allows observation of the plant material for 3–5 hours without shrinkage. Therefore, it was possible to observe the early stages of wax regeneration on the specimen surfaces. Wax regeneration has been investigated on pieces of the leaves of Citrus limon, Coffea arabica, Iris germanica and Thalictrum flavum glaucum. Three succulent plants, whose intact leaves (Kalanchoe daigremontianum, Lithops turbiniformis) or shoots (Aporocactus flagelliformis) were too thick for fixation in the AFM specimen chamber, were investigated by using the described new preparation technique. The observed wax regeneration patterns are listed in table 1. The modes of wax regeneration are discussed in §3a. In the other 10 species listed in table 1, wax regeneration was investigated on intact leaves. Large differences were found in the rate of wax regeneration, and two modes of wax layer formation were found. These differences are not related to the kind of specimen preparation. The study here focuses on the early stages of wax film formation on plant surfaces. However, in some species, the regeneration of three-dimensional wax crystals starts parallel to wax film formation (table 1).
(a) Concentric and stripe-shaped formation of wax films
Two different modes of wax layer formation, a concentric (planar) and a stripe-shaped (linear) formation of wax films, were found and are introduced here with representative species as examples. A concentric layer formation was observed for 11 species, indicated in table 1 as C-layer formation. Within these species, large differences were found in the amount of newly formed wax terraces (density of wax layer nuclei) and in the growth rate of the wax layer formation. In figure 4, the formation of a monomolecular layer of wax on a leaf of the succulent plant L. turbiniformis is shown. Figure 4a–c shows the increase in the wax layer at 12, 27 and 38 min after wax removal. The appearing wax forms almost concentric (C-layer) terraces, and the growth of these layers occurs by the addition of new wax at the outermost edges of the layers, as shown in figure 4d, which has been produced by overlaying and subtraction of figure 4b,c. Thus, figure 4d shows the wax added within 11 min.
In one investigation of wax regeneration of Taxus baccata, the removal of the original wax layer was incomplete, but in this initial situation it could be observed how regenerating wax films fill up the wax-free voids on the cuticle. The image series in figure 5 shows some thicker layers, which are the wax residues, and, in the centre between the residues, a growing new wax layer. Figure 5a represents the wax regeneration 23 min after partial removal of the original wax layer, figure 5b has been recorded after 33 min, figure 5c after 70 min and figure 5d after 138 min. The nearly concentric growth of the new wax layer changes its growth directions when it comes into contact with non-growing wax residues (figure 5c), and thereby effectively refills the spaces between the already existing wax residues. The continuous growth of the wax film occurs by a process of self-assembly.
In a second mode of layer formation, observed in seven species, the new wax material grows in a stripe-shaped pattern. Species with this type of layer formation are indicated in table 1 as S-layer formation. Ipheion uniflorum is the only species for which both modes of wax film formation have been observed. One representative species in which S-layer formation is shown is Lathyrus odoratus. The AFM images in figure 6a–c show that the stripe-shaped layers grew in different directions. Figure 6a has been recorded 109 min, figure 6b at 125 min and figure 6c at 149 min after removal of the wax. When growing layers come into contact with already existing layers, they start to overgrow them. The lateral sides of the stripes accumulate new wax material, as indicated by the closure of the hollows formed between the stripes, but, for a rapid re-covering of the cuticle with a complete wax layer, this kind of wax regeneration seems to be less efficient. Growth of these S-layers occurs by the addition of new wax preferentially at the edges of the terminal ends of the stripes, demonstrated by figure 6d, which shows the amount of new wax accumulated within 16 min. Figure 6d has been produced by overlaying of figure 6a,b and subtraction of figure 6a from b.
(b) Formation of multilayered wax films
Multilayer wax formation has been described for Galanthus nivalis (Koch et al. 2004) and Euphorbia lathyris (Dommisse 2007). In these studies, it has been shown that the wax films on plant surfaces are built by several molecular layers. AFM height measurements at the edges of the growing layers represent the heights of vertically orientated long-chain hydrocarbons, with a chain length between C20 and C40. In this study, the average thickness of the wax layers of L. odoratus (figure 6) is 8 nm and represents a bilayer formation of aliphatic hydrocarbon molecules of approximately 30–32 carbon atoms. In T. baccata (figure 5d), the average heights of the steps in the wax layers are 6.5, 9.0 and 12.1 nm, indicating that the wax film is composed of two, three and four layers of molecules with chain lengths of approximately 3 nm (C24–C26 molecules). In L. turbiniformis (figure 4), the average heights of the wax layers are 4.2 and 6 nm, representing the chain length of vertically orientated long-chain hydrocarbons, with a chain length of approximately 32 and 46 carbon atoms. In I. uniflorum (figure 6), an independent development of two different layers was observed. While a ‘normal’ stripe-shaped layer grew slowly, a layer with variable thickness (from 3 to 12 nm) spread over the observed area in less than 1 hour. Probably one of the layers grows under an already existing wax layer, the other one on top of it. During the formation of multilayered wax films and for the formation of three-dimensional wax crystals on the films, diffusion of new wax material through already existing wax layers is assumed (Koch et al. 2004). However, in I. uniflorum, already existing wax structures were pushed up by a new layer of regenerated wax below. This process of wax formation is shown in figure 7. Here, a formation of S-layers, denoted in figure 7a by the upper arrow, and a fast growing layer below, denoted by the lower arrow, is shown. Height measurements made at the edge of this layer show a step height between 3 and 12 nm, which indicates that this layer is composed of more than one molecular layer. Figure 7b, recorded only 4 min later, shows that the wax layer has grown underneath the S-layer structures. In figure 7c, 44 min after the original wax has been removed, the wax layer has covered nearly 80 per cent of the scanned area, whereas the S-layers grew very slowly. After a total time of 53 min (figure 7d), the scanned surface area is completely covered with the new wax layer. Even the higher structure in the cuticle, visible at the right side of the figures, was overgrown by the new layer.
The data presented here show that all 17 species started to regenerate a wax film immediately after removal of the wax, but differences in the intensity of wax regeneration were found, and in some investigations no wax regeneration could be observed (data not shown). The fact that some species showed no wax regeneration is a strong indication that wax regeneration is not based on pure diffusion of intracuticular wax to the surface. If this were the case, on each specimen at least a few layers of wax would have been found. The ability of self-healing seems to correlate with the new production of wax. Wax synthesis is generally high during the development of the leaves (Baker 1982; Post-Beitenmiller 1996). Thus it can be assumed that the lower intensity of wax regeneration found in some older leaves is mainly based on differences of wax synthesis. Differences in wax regeneration in leaves of different ages have been found by SEM studies (Neinhuis et al. 2001). In this study, mature leaves from 6 out of 16 species showed no wax regeneration, whereas 13 of 16 species showed a high intensity of wax regeneration when leaves were in development. However, external factors such as changes in temperature and air humidity can also influence the dynamics of wax transport through the plant cuticle (Schreiber & Schönherr 1993; Schreiber 2001) and the epicuticular movement of the wax molecules.
4. Mechanisms of wax formation
(a) C- and S-layer formations of wax films
Two modes of wax film formation have been observed: concentric (C) layer formation, in which the new wax spreads out in a concentric or planar way; and stripe-shaped (S) layer formation, in which the wax film grows in the form of stripes or lines. Both growth modes are a process of molecular self-assembly, in which new wax molecules are preferentially added to the edges of existing layers, resulting in a layer structure. The stripe-like wax layer results from a faster accumulation of new material at the terminal ends of the stripes, whereas in the concentric growing wax layers the new wax is added approximately equally onto all sides of the layer. When the layer growth continues, a complete monolayer, as schematically shown in figure 8a, is formed, and, later, a multilayered wax film is formed.
Two factors can induce differences in C- and S-layer formation. One is differences in the chemical composition of the wax and the other is the unknown molecular structure of the underlying cuticle. In most species, the wax film occurs together with three-dimensional wax crystals, but until now it has not been possible to separate both structures for separate chemical analysis. The AFM images show that the regenerated wax layer thicknesses vary around 4 nm (the average molecule length of typical aliphatic wax compounds with a chain length of 30 carbon atoms). The mass of a single molecular wax layer, as observed with the AFM, is not sufficient for a chemical analysis by gas chromatography and/or mass spectroscopy. Thus the general lack of knowledge about the chemistry of the thin wax film makes an interpretation difficult. By comparison, a 1 μm thick wax layer on a leaf consists of approximately 250 molecular layers of wax. Current knowledge about the chemistry of the underlying wax films is based on conclusions made from recrystallization experiments with waxes and single wax compounds. For wax tubules and some platelets, it has been shown that only one or two compounds of the wax mixtures are responsible for the formation of the three-dimensional structures (Jeffree 2006; Koch & Ensikat 2008). As a consequence, it is assumed that the majority of wax compounds are located in the wax film.
For some species where only wax films and no three-dimensional waxes exist, chemical data are known. A well-investigated species is Prunus laurocerasus, which showed a C-layer formation. Jetter & Schäffer (2001) showed that these films are composed of approximately 39 different aliphatic wax compounds, but they also gave evidence that the chemical composition varies during the ontogeny of the leaves. Thus, the next step for understanding the differences in wax film formation could be in vitro recrystallization of the single wax components and their mixtures.
(b) Formation of multilayered wax films
The mechanisms of formation and growth of plant waxes have been discussed for many years. Since the first AFM analysis of wax regeneration (Koch et al. 2004), further in vitro studies of wax crystal formation have been carried out (Koch et al. 2006a,b). Based on these and the data presented here, it can be concluded that two mechanisms of layer formation lead to multilayered wax films, as schematically shown in figure 8b,c.
In the first mode (figure 8b), new wax material moves through the existing, complete wax layers. With only two exceptions (discussed below), this mode was found in all species after the first layer of waxes was completely regenerated. However, this mode predicts a certain degree of permeability of the already existing wax layers to provide the wax molecules necessary for the growth of three-dimensional waxes on the wax films. Schreiber & Schönherr (1993) performed transport studies with reconstituted wax layers and showed that aliphatic molecules are mobile within the wax mixture. Reynhardt & Riederer (1994) emphasized the importance of crystalline and amorphous zones in plant waxes for the diffusion of molecules through the cuticle. Reynhard (1997) demonstrated that a mixture of short and long chains form amorphous zones of higher fluidity. These amorphous zones could be the pathways for the new wax molecules on their way through a wax film. However, the molecule mobility, and therefore the intensity of wax regeneration, can be reduced when the wax film is in a stable crystallized form.
In the second case, schematically shown in figure 8c, existing waxes are lifted up by the growing new wax layer underneath. This kind of multilayer formation has been observed only during wax regeneration in two plants, I. uniflorum and T. baccata. The I. uniflorum AFM micrographs demonstrate that existing striped wax layers were not influenced by the much faster growing second wax layer. Compared with other species, these species showed a very fast regeneration of the wax layer, and it can only be speculated that the amount of new wax that appeared on the cuticle surface was too great to diffuse quickly through the already existing wax layers.
The preparation technique for AFM for small pieces of living plant material, described in this paper, provides the opportunity to study plant surface structures, or wax regeneration, over several hours with minimized material shrinkage by water loss. The comparative study of wax film formation in 17 different species showed two different kinds of wax film formation, described as concentric layer formation and striped layer formation. The AFM examinations presented here showed that regeneration of epicuticular wax films on living plant surfaces is a highly dynamic and comparatively fast process, reflecting the importance of a continuous wax coverage of leaves. This study showed that most plants are able to refill voids within the epicuticular wax layer. The process of wax film formation occurs by self-assembly of the wax molecules into layered structures, but it seems to be limited in some plants by non-existing or low wax synthesis inside the cells. The self-healing process found in plants might be an interesting system for the development of coatings of materials that are permeable for small molecules. In future, AFM investigations of recrystallization of isolated film-forming waxes might give a deeper understanding of the correlation between the chemistry and the observed different modes of layer formation.
The authors thank the German Science Foundation (Deutsche Forschungsgemeinschaft), Deutsche Bundesstiftung Umwelt (DBU) and the Bundesministerium für Bildung und Forschung (BMBF) for the financial support of their research. Special thanks go to Dr H. Bargel for his comments and discussions during the preparation of the manuscript.
One contribution of 9 to a Theme Issue ‘Biomimetics II: fabrication and applications’.
- © 2009 The Royal Society