Notice: Wiley Online Library will be unavailable on Saturday 27th February from 09:00-14:00 GMT / 04:00-09:00 EST / 17:00-22:00 SGT for essential maintenance. Apologies for the inconvenience.
III Deposition and crystalline morphology of cuticular wax 474
IV Cuticular wax as a photoprotective layer 475
V Effects of irradiation and temperature on cuticular wax composition 478
VI Contact angles and wettability 481
VII Humidity effects 482
VIII Water, salinity and cold stress 482
IX Mechanical stress 485
X Altitude 486
XI Pollution 486
XII Genetic and environmental control of cuticular wax production 488
XIII Conclusions 493
Plants are subject to a wide range of abiotic stresses, and their cuticular wax layer provides a protective barrier, which consists predominantly of long-chain hydrocarbon compounds, including alkanes, primary alcohols, aldehydes, secondary alcohols, ketones, esters and other derived compounds. This article discusses current knowledge relating to the effects of stress on cuticular waxes and the ways in which the wax provides protection against the deleterious effects of light, temperature, osmotic stress, physical damage, altitude and pollution. Topics covered here include biosynthesis, morphology, composition and function of cuticular waxes in relation to the effects of stress, and some recent findings concerning the effects of stress on regulation of wax biosynthesis are described.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Plants have evolved to exist in conditions which are rarely ideal for maintenance of normal physiology and may be at the limit for survival. Abiotic stress arises from exposure to climatic extremes such as drought, heat, cold and frost, and the effects of radiation levels, shade, altitude, soil nutrient and water status and pollution are also significant. In response, plants can adapt to, avoid and overcome the stress by means of various physiological and biochemical mechanisms, including evolution of a resistance-conferring genotype, or by development of genes which can produce ecologically adapted phenotypes. Stress resistance mechanisms fall broadly into two categories – avoidance and tolerance – which may occur together. Avoidance involves establishment of internal conditions where the plant's cells are unstressed although external conditions may be stressful, for example, control of high leaf temperatures by transpiration and prevention of drought by water conservation. Tolerance involves endurance of the stress such that plants can function under extremes of both internal and external stress. Examples include plants endurance of desiccation during drought and revival on hydration. The development of specialised physiological mechanisms is more a characteristic of tolerance, whereas avoidance more often utilises the capability of general physiological processes to provide mechanistic and morphological devices to shield plants from the effects of extreme conditions. Stress resistance may be induced following exposure to sublethal levels of stress, and hardiness towards one form of stress may confer some resistance to other stresses. This illustrates the interrelationship between different stress factors; for example, resistance to drought and resistance to high temperatures are often linked, as are resistance to freezing and resistance to cellular dehydration.
Development of a water-resistant cuticle was fundamental to the successful colonisation of land by plants (Edwards et al., 1996) and such structures existed by approx. 400 million years ago. As the primary interface between the plant and its environment the cuticle plays a key role in maintaining the plant's integrity within an inherently hostile environment. Its outermost surface is covered in a hydrophobic layer of predominantly long chain aliphatic molecules, collectively referred to as cuticular wax, embedded in an underlying layer of the polymer cutin (Fig. 1). Before considering the functional significance of cuticular wax to plant stress, it is useful to summarise its origin and physicochemical nature.
II. Biosynthesis of cuticular wax
Synthesis of the major wax components occurs via sequential elongation of a C2 primer derived from acetyl-CoA with C2 units derived from malonyl-CoA. In this condensation-elongation process, acyl chains of up to C16 and C18 formed in the synthesis de novo are further extended to C30 or higher by a second elongation system. Modification of the acyl chain gives products including alkanes, aldehydes, primary alcohols, alkyl esters, secondary alcohols, ketones and various polyoxygenated compounds. The chemistry, biochemistry and molecular biology of cuticular wax biosynthesis have been extensively reviewed (Kolattukudy et al., 1976; Bianchi, 1995; von Wettstein-Knowles, 1995; Kolattukudy, 1996; Post-Beittenmiller, 1996; Kunst & Samuels, 2003; Shepherd, 2003).
1. Formation of malonyl-CoA
The initial step in wax synthesis is formation of malonyl-CoA from acetyl-CoA, catalysed by the multifunctional enzyme system acetyl-CoA carboxylase (ACCase; refer to Appendix A, Table A1 for full list of abbreviations). A molecule of CO2 derived from bicarbonate is added to the biotin moiety of biotin carboxylate carrier protein (BCCP) to form N-1′ carboxybiotin–BCCP, a reaction catalysed by biotin carboxylase (BC). Subsequently the CO2 is transferred to acetyl-CoA, forming malonyl-CoA, catalysed by carboxyltransferase (CT) (Fig. 2). Two types of ACCase occur: the plastidial form supplies malonyl-CoA for the synthesis de novo, whereas the cytosolic form supplies malonyl-CoA for further acyl chain elongation and synthesis of polketides and flavonoids (Schultz & Ohlrogge, 2002).
Table 1. Association of cuticular wax crystalline forms with specific classes of chemical component in the wax
Composition associated with morphology
β-Diketones or secondary alcohol nonacosan-10-ol, also diols
De novo acyl chain synthesis catalysed by fatty acid synthase (FAS) occurs in plastids (reviewed by von Wettstein-Knowles, 1995; Schultz & Ohlrogge, 2002). The enzyme complex includes acyl carrier protein (ACP), to which acyl groups are linked via a pantetheine-4′-phosphate prosthetic group. Chain formation starts with transfer of an acetyl group from acetyl-CoA to a cysteine thiol of 3-ketoacyl-ACP synthase (KAS), another key component of FAS, to serve as the C2 primer for elongation (Fig. 2, step a). The C2 elongating substrate consists of a malonyl group which is transferred from malonyl-CoA to the ACP pantetheine thiol (step b). Malonyl-ACP is decarboxylated by KAS to form a carbanion, which undergoes a Claisen condensation reaction with the acyl carbonyl group of the KAS-bound acetyl primer (step c). Overall, the acyl (acetyl) group is transferred to C-2 of malonyl-ACP, forming acetoacetyl-ACP and CO2, and the KAS cysteine thiol is regenerated. A three-reaction sequence follows, involving stereospecific reduction of 3-ketoacyl-ACP (step d), dehydration of D-3(R)-hydroxyacyl-ACP (step e) and reduction of Δ2(E)-2,3-enoyl acyl-ACP to the corresponding C4 acyl-ACP (step f), which completes the first elongation cycle. Transfer of the C4 acyl group from ACP to the KAS cysteine thiol starts the next cycle, freeing ACP to accept another malonyl group from malonyl-CoA (step g). Repetition of the cycle a further six times gives palmitoyl-ACP (16:0-ACP), then one further time to produce stearoyl-ACP (18:0-ACP). When the acyl chain is 16 or 18 carbons long, a double bond may be inserted between C-9 and C-10 by the action of acyl-ACP desaturase.
Acyl chains to be incorporated into cellular complex lipids, cutin, suberin and wax components, are hydrolysed by thioesterases and the resultant fatty acids are transported across the plastidial membrane (Fig. 2, step h). An acyl-CoA synthase at the outer plastidial membrane may then convert free acids to acyl-CoA derivatives.
3. Elongation and acyl chain modification
Further elongation of acyl chains derived from the synthesis de novo occurs in epidermal cells and involves a series of extra-plastidial elongation systems, collectively referred to as fatty acyl elongases (FAEs). Each FAE includes the condensing enzyme 3-ketoacyl-CoA synthase (KCS), 3-ketoacyl-CoA reductase, 3(R)-hydroxyacyl-CoA dehydrase and (E)-2,3-enoyl-CoA reductase (Millar & Kunst, 1997). Like FAS they catalyse a sequence of condensation, reduction, dehydration and reduction reactions, in which acyl chains are extended by C2 units derived from malonyl-CoA. The enzymes catalysing the last three reactions are believed to be expressed constitutively throughout the plant and are associated with the various cellular condensing enzymes (Millar & Kunst, 1997). FAE differs from FAS in that the extending acyl chain is linked to CoA rather than ACP and the C2 donor is malonyl-CoA, rather than malonyl-ACP. The major wax components arise via further modification of the elongation products by enzyme complexes associated with the elongation systems. Two distinct pathways occur and, based on their key reactions, these have been named the reductive and decarbonylative pathways, respectively (Fig. 3).
4. Decarbonylation pathways
Acyl-CoAs are reduced to intermediate aldehydes from which alkanes are formed by decarbonylation (Fig. 3). Secondary alcohols are formed via stereospecific hydroxylation of alkanes, and oxidation of the alcohol gives the corresponding ketone. Further reactions can occur, including esterification of secondary alcohols with fatty acids and formation of diols, hydroxy ketones and diketones via additional hydroxylation and oxidation. In some species, such as cereals and grasses, a variation of the decarbonylation pathway is thought to be the source of β-diketones (von Wettstein-Knowles, 1995; Kolattukudy, 1996). During acyl chain elongation, 3-ketoacyl-CoA intermediates are formed by 3-ketoacyl-CoA elongase in a series of cycles where the usual reduction, dehydration and reduction steps are omitted following the condensation reaction. After three such steps, a polyketide is formed, from which a β-diketone is produced via a sequence of reduction, dehydration and reduction, more full cycles of elongation and finally decarbonylation or decarboxylation. Further reactions may include insertion of additional oxygenated substituents.
5. Reductive pathways
Acyl-CoA is reduced to an intermediate aldehyde and then to a primary alcohol (Fig. 3). A single enzyme carries out both reductions in some plants, and the aldehyde intermediate remains bound. Fatty acyl-CoA reductase has been purified from leaves of pea, Pisum sativum (Vioque & Kolattukudy, 1997), and jojoba, Simmondsia chinensis, embryo (Metz et al., 2000). Free fatty acids are formed by hydrolysis of acyl-CoA. Several mechanisms are possible for synthesis of long-chain esters, including direct esterification of acids with alcohols and transfer of acyl groups from phospholipids, glycerolipids or acyl-CoA to alcohols. Wax synthase (acyl-CoA alcohol transacylase) has been identified from several plants, and purified from jojoba (Lardizabal et al., 2000; Metz et al., 2000). Ketoaldehydes, ketoalcohols, diols and hydroxy fatty acids may be formed by introduction of oxygenated substituents into the acyl chain. Some of these compounds are found in the free form and as esters, while others only occur as components of polymers such as estolides, cutin and suberin.
III. Deposition and crystalline morphology of cuticular wax
The mechanism of wax deposition on the leaf surface is poorly understood (see review by Kunst & Samuels, 2003). Following synthesis at sites such as the endoplasmic reticulum (ER), Golgi apparatus and plasma membrane (PM), individual components must move to the PM, pass through the cell wall and cuticle and on to the cuticle surface. Direct movement between ER and PM where they are in very close proximity is one possibility and movement to the PM within intracellular vesicles is another, perhaps derived from the ER, Golgi and PM by exocytosis. Jenks et al. (1994) reported an increase in the local density of vesicles adjacent to the site of wax excretion when wax production in sorghum was induced by light. However, the nature and origin of the vesicles could not be determined. Passage of wax components through the cell wall and cuticle probably occurs via diffusion, possibly in a solvated form, through molecular-scale spaces and channels consisting mostly of temporary openings between polymer and wax chains, since there is little evidence for macroscopic pits, pores or excretory structures. Small lipid transport proteins (LTPs) associated with the cell wall may facilitate diffusion (Yamada, 1992; see also reviews by Post-Beittenmiller, 1996; Kunst & Samuels, 2003). Being hydrophobic, LTPs can probably move from the outer layer of the PM, through the cell wall and on to the cuticle surface, and such a protein has been found within broccoli cuticular wax (Pyee et al., 1994). However, there is currently no additional experimental evidence supporting this idea.
The existence of layers within cuticular waxes was verified using differential removal-extraction procedures. In earlier studies, for example with blackberry, Rubus fruticosus (Haas & Rentschler, 1984), the outer layer designated as ‘epicuticular wax’ was removed using materials like collodion, whereas more recently Jetter et al. (2000) used a cryoadhesive with Prunus laurocerasus for this purpose. Inner layers designated as ‘intracuticular wax’, which could not be removed mechanically, were extracted with solvent. The inter-layer boundary was considered to delineate the outer limit of the cutin matrix. The chemical basis and extent of such differentiation vary between species. Epicuticular wax from R. fruticosus consisted primarily of free alcohols, alcohol acetates and esters, with lesser amounts of free fatty acids and alkanes. Intracuticular wax had similar amounts of the alcohols, fewer esters and more free fatty acids, and triterpenoid acids were exclusive to this fraction (Haas & Rentschler, 1984). In the case of P. laurocerasus, epicuticular wax consisted entirely of aliphatic compounds, whereas intracuticular wax consisted primarily of triterpenoids (Jetter et al., 2000).
Seen under the electron microscope, epicuticular waxes usually have a microcrystalline structure, sometimes arising from an underlying amorphous layer (Jeffree, 1996). Numerous morphological forms classified by Barthlott et al. (1998) may be present, including rods, ribbons, filaments, tubes and plates. Some of these can be related to the presence of specific wax components (Table 1). Compounds with mid-chain oxy- substituents, such as β-diketones, hydroxy-β-diketones, diols and secondary alcohols, are associated with tubes, whereas primary alcohols with a terminal oxy- substituent are associated with platelets. Changes in the homolog distribution within a class of compound, for example the primary alcohols, are also associated with crystalline modifications (Table 1).
When Jeffree (1974) and co-workers (Jeffree et al., 1975) recrystallised leaf waxes from various solvents using a porous surface and wick-fed delivery system, the waxes formed similar structures to those found on leaves. They concluded that wax morphology was influenced more by the physicochemical properties of the constituents rather than by the underlying cuticular membrane or means of delivery to the surface. More recently, Kirsch et al. (1997) found that the transport characteristics (permeability) of recrystallised waxes from several species mirror the barrier properties of isolated cuticular membranes and intact leaves with wax in situ. They argued similarly that properly reconstituted wax layers have similar crystalline morphologies to those of intact waxes.
Temperature, light intensity and humidity influence wax morphology, and these parameters tend to act together so it can be difficult to distinguish between their respective effects. In general, higher temperatures favour structures parallel to the cuticle surface such as plates and flakes, while lower temperatures favour more vertical structures such as rods and tubes (Table 2). Complex structures may form at higher temperatures, for example the dendritic lattices observed in Brassica waxes which can develop layered structures reminiscent of rain forest canopies (Fig. 1c,d,g). Greater illumination often leads to shorter, less elaborate structures, although this is not always the case. Some examples of the effect of temperature and light on wax morphology are listed in Table 2, principally for members of the Brassicaceae. Established structures may also transform into other forms if conditions change. Tubular crystal forms are thermodynamically unstable because of their high surface area/volume ratio, and with an input of energy, for example heat, they transform into more compact planar and thermodynamically stable forms. This accounts for the observation of Baker (1974) that tube waxes on normal brussels sprouts turned into dendrites within 48 h of raising the temperature from 15 to 35°C, with little compositional change. Baker (1974) suggested that local temperature and crystallisation rate may be more significant than solute concentration and chemical composition in determining wax crystal structure. Slow crystallisation favours linear structures, whereas rapid cooling favours dendrite formation.
Table 2. Effect of temperature and irradiation on leaf wax crystal morphology
The ease of interconversion between crystalline forms is also more readily explained if waxes are carried to the cuticle surface in a solvent. Transportation of wax precursors through the cuticle where they are subject to final modification (von Wettstein-Knowles, 1974), or extrusion of waxes through the cuticle or via specialised pores, notwithstanding the lack of evidence for such structures, could not account for the crystallisation of different structures such as dendrites on the upper reaches of existing crystals. Plants produce a wide range of volatile organic compounds, including terpenes, short-chain aldehydes, ketones, alcohols and esters, etc., all of which could serve as a wax delivery solvent. Although the quantities involved are low, a steady rate of production would be sufficient to deliver the wax to the cuticle surface and volatiles may persist within the boundary layer long enough to aid in crystallisation and interconversion of wax structures under appropriate conditions.
IV. Cuticular wax as a photoprotective layer
Electromagnetic radiation incident on plant cuticles covers a wide range of energies and wavelengths, from high-energy, short-wavelength ultraviolet (UV) through visible and near-infrared (IR) to lower-energy, longer-wavelength middle–far IR (Fig. 4a). In recent years quantities of UV-B radiation reaching the earth's surface have increased due to stratospheric ozone (O3) depletion caused by atmospheric pollutants such as chlorofluorocarbons (Farman et al., 1985; Kerr & McElroy, 1993; Webb, 1997; Holmes & Keiller, 2002). This is likely to continue due to the long-term persistence of chlorofluorocarbons within the upper atmosphere (Molina & Rowland, 1974). Both UV-B and UV-A are harmful to plant growth, UV-B being most damaging (Barnes & Cardoso-Vilhena, 1996; Holmes & Keiller, 2002).
Leaves may reflect the radiation via classical specular reflection and light scattering or they may absorb the energy. The key absorption event occurs in the photosynthetically active region (PAR) of the visible spectrum (400–700 nm), through the action of photosynthetic pigments, with further absorptions by water at longer wavelengths (Fig. 4a). The presence within the epidermis of UV-B-absorbing pigments, primarily flavonoids, with minimal PAR absorption, provides a major protective barrier to damage (Robberecht et al., 1980; Caldwell et al., 1983; Rozema et al., 1997; Cockell & Knowland, 1999; Liakoura et al., 2001). Where absorption is high, reflectance is generally low and vice versa (Barnes & Cardoso-Vilhena, 1996), and UV reflectance can range from < 10% in most plant species studied to date, to 70% in a few others (Caldwell et al., 1983; Barnes et al., 1996). In a study with four species, Robberecht et al. (1980) found that, whereas the amounts of UV-B reflected (5–40%) and absorbed (60–95%) varied between species, less than 1% was transmitted into the mesophyll.
Reflectivity is strongly influenced by the surface topography, which is primarily determined by leaf hairs and the cuticular wax layer, particularly at visible wavelengths. The importance of cuticular wax has been shown in various studies by removal of wax from waxy-leaved species. Reflectance was reduced and photosynthesis increased in glaucous Eucalyptus species, but had little effect in non-waxy species (Cameron, 1970). Similarly, reflectance from Kalanchoe pumila was greatly reduced and absorbance increased at visible and near-IR wavelengths (Eller, 1979). In a recent study using 45 different Eucalyptus and Kalanchoe species, Holmes and Keiller (2002) found that waxy leaves were more reflective at UV (330 nm) and photosynthetic (680 mn) wavelengths than various types of hairy leaves. Removal of wax generally reduced reflectance at both wavelengths, and the reduction was proportionally greater at 330 nm; however, waxless varieties showed little change in reflectivity. Reflectance was reduced over the range 270–500 nm in E. cinerea and E. gunnii, and mainly over 400–500 nm (PAR) in K. pumilla, showing that reflectance characteristics can vary widely. Light dispersion is particularly associated with glaucousness (Johnson et al., 1983; Juniper & Jeffree, 1983; Blum, 1986; Barnes & Cardoso-Vilhena, 1996; Febrero et al., 1998), and wax morphology (Table 3). Filamentous structures (tubes, rods, ribbons) common to glaucous lines lower incident radiation by increasing reflectance more than plates (Cameron, 1970; Sanchez-Diaz et al., 1972; Blum, 1975a,b; Juniper & Jeffree, 1983; Febrero et al., 1998). Reflectance can vary with temperature, for example, reflectance from Eucalyptus increased at higher day/night temperatures over the range 15/10–27/22°C, with a maximum at intermediate temperatures (21/16°C) (Cameron, 1970).
Table 3. Effect of water stress on leaf surface reflectivity and associated cuticular wax morphology
Reflection (wavelength, nm)
Associated crystal/surface morphology
W, water stress; S, salinity stress; NS, non-stressed; RES, resistant; SUS, susceptible or least resistant; GLC, glaucous; NG, non-glaucous (glabrous); adax, adaxial leaf surface; abax, abaxial leaf surface; ↓ reduction or ↑ increase in value; ≈, relatively unchanged.
Reflectance (PAR) by glaucous and non-glaucous lines of wheat was proportional to the amount of wax present, and elevated reflectance reduced light transmission to the underlying mesophyll cells in glaucous durum wheat by 12% (Johnson et al., 1983). Normally this is insufficient to affect photosynthesis, but might be a factor under non-saturating light conditions. Eller (1979) suggested that light reflected from glaucous surfaces may help to illuminate leaves in shaded locations lower in the canopy. Such an effect would benefit plants such as wheat with large leaf area indices, and increased photosynthesis at lower levels may balance reduced photosynthesis at higher locations. Reduced absorption at visible and near-IR wavelengths due to enhanced reflectance can lower tissue temperatures, thus reducing the vapour pressure difference between the tissue and the air, which helps to reduce transpirational water loss. Usually the reduction in transpiration is greater relative to any reduction in photosynthesis, causing an increase in transpiration efficiency (TE), net photosynthesis/transpiration. For example, in comparison with non-glaucous plants, photosynthetic tissue in glaucous durum wheat was up to 0.7°C cooler for droughted field-grown plants or 0.3°C cooler for well watered glasshouse-grown plants, photosynthesis and stomatal conductance were reduced, day and night transpiration by the ear were up to 50% less and overall TE was reduced. This can be beneficial at critical stages of development; for example, a 0.5°C fall in tissue temperature was calculated to equate with a 30 g per plant saving in water, sufficient to extend grain filling by at least 3 d (Richard et al., 1986).
Reflectance and photosynthesis are dependent on the angle of incidence of radiation. At angles between 90° and 60°, reflectance from non-glaucous E. regnans was constant and then increased rapidly at shallower angles (Howard, 1967). Photosynthetic rates fall at angles below 72°, to 70% of normal values at 45°, and approach zero at 5° (Kriedemann et al., 1964).
V. Effects of irradiation and temperature on cuticular wax composition
An increase in wax thickness is a response of many plants including various Brassica sp. (Baker, 1974; Reed & Tukey, 1982; Shepherd et al., 1995b), carnation (Reed & Tukey, 1982) and barley (Giese, 1975) to higher irradiation levels (Tables 2, 4). In her study using barley, Giese (1975) found that deposition rates and leaf-surface wax density in 4-d-old plants were 2.5 times greater in the light than in the dark. When 4-d-old dark-grown plants were exposed to light, rates of wax deposition increased 7.5-fold in the first 24 h after exposure, thereafter settling down to the same rate and wax density as for light-grown plants. While demonstrating the stimulatory effect of light, these findings also indicated that rates of wax deposition were density-limited. Temperature also affects wax production and in numerous plants, particularly Brassica sp., more waxes are produced at lower temperatures (Table 2), although this is not always the case.
Table 4. Effect of temperature and irradiation on leaf wax composition
In addition to affecting wax quantity, light and temperature influence composition, and, in some instances, such compositional changes have been studied in detail (Table 4). It is usually possible to relate compositional changes to the channelling of acyl precursors into free fatty acids and the products of the reductive and decarbonylation pathways, to changes in their chain-length (CL) distributions, or in some cases to both. In studies with brussels sprouts, B. oleracea, Baker (1974) found that normal (glaucous) plants and the gl4 mutant generally had more fatty acids and products of the reductive pathways at lower irradiation levels and higher temperatures, and more of the decarbonylation products at higher irradiation and lower temperatures. Patterns were more complex among several glossy (gl) mutants. In comparison, the effects were reversed for temperature (gl2 and gl3) or showed intermediate product distributions for temperature (gl1) and light (gl1, gl2 and gl3). In another study (Shepherd et al., 1995b, 1997) with the brassicas kale, B. oleracea, and swede, B. napus, glasshouse-grown plants (lower irradiation, higher temperature) were compared with plants grown outdoors (higher irradiation, lower temperature). Both species responded similarly under the different conditions, and there was also some similarity with the earlier findings of Baker (1974), decarbonylation products being relatively more abundant under higher irradiation/lower temperatures. The distribution of different compound classes within the major pathways can also change, suggesting that individual steps and transformations within the pathways (Fig. 3) may be differentially sensitive to the environment. However, this does not always happen. For example, Giese (1975) found with barley that the proportions of wax decarbonylation (1%) and reduction (89%) products or free fatty acids formed only by elongation (10%) were the same in the light and the dark. In their studies with Citrus aurantium, Riederer & Schneider (1990) showed that changes in daytime and night-time temperatures may differentially effect wax composition. Higher daytime temperature during leaf development reduced the quantities per unit area of alkanes, primary alcohols, fatty acids and alkyl esters, whereas, except for the esters, the amounts of these components increased with higher night-time temperatures.
The studies of Giese (1975) and Shepherd et al. (1995b, 1997) also showed the influence of environment on CL, with a shift towards shorter components under higher illumination (Figs 5, 6a). A shift to lower CL is seen for every class of component in barley wax, which is dominated by the products of the reductive pathways. In the light, one CL of a single distribution dominates a given wax class, whereas in the dark, two prominent CLs or groups are found (Fig. 5). The Brassica wax is dominated by decarbonylation products, which occur as a single distribution under the different conditions (Fig. 6a).
Environmental factors can also influence the distribution of isomers within a class of compound. Some waxes contain branched-chain (br-) components, formed by utilisation of a br-primer unit in the synthesis de novo (Baker & Holloway, 1975; Shepherd et al., 1995a, 1997; Shepherd, 2003). Higher levels of UV-B were associated with increased abundance of br-relative to n- alkanes in Nicotiana tabacum and an increase in the proportion of shorter (C< 30) to longer (C> 30) homologues (Barnes et al., 1996). Similarly, both free and esterified iso- (i-) and anteiso- (a-) br-acids and alcohols in waxes from kale and swede were generally more abundant than the n-components under greater illumination and lower temperature (Fig. 6b) (Shepherd et al., 1997). Environmental effects on CL were seen; at higher illumination n-acids, n-alcohols and i-acids showed a shift to shorter CL for kale but to longer CL for swede. One pair of kale and swede genotypes showed a shift to shorter i-alcohols, while the other pair showed a shift to longer CL. There were no apparent relationships between CL and environment for a-components. These effects may also have been attributable to differences in UV-B since plants with the fewest br-components were grown in a glasshouse while those with higher levels were grown outdoors.
VI. Contact angles and wettability
Leaf surface wettability is influenced by the physicochemistry of the cuticular wax and, to a lesser extent, leaf turgor (Butler, 1996; Cape, 1996). Wettability is measured in terms of the contact angle (CA), the angle subtended between the leaf surface and the plane of a tangent to the surface of a water droplet originating at the contact point for air, leaf and water (Fig. 4c). Values of CA vary widely from 120° for very hydrophobic Eucalyptus globulus (Hietala et al., 1997) to 29° for very hydrophilic Vicia faba. At angles < 90° water droplets tend to spread, ultimately forming water films, particularly when leaves are shaken by wind. Conversely, formation of isolated drops is a characteristic of angles > 90° (Cape, 1996). Contact angles are therefore indicative of water retention capacity, which is significant for surface-colonising insects and microorganisms (Butler, 1996; Huttunen, 1996) and for foliar deposition and uptake of nutrients, pollutants and agrochemicals (Cape, 1996). Droplets on wettable leaves can focus solar radiation up to 20 times (Brewer et al., 1991), possibly by acting as planoconvex lenses. This may increase the transmission of damaging UV-B to the epidermis and upper mesophyll. Some stress-induced changes in CA and wettability are associated with altered wax composition and morphology (Table 5). Reductions in CA of up to 30° were correlated with temperature-induced recrystallisation of conifer needle wax, largely composed of (S)-nonacosan-10-ol, from tubular to planar forms (Anfodillo et al., 2002), and increased abundance of br-alkanes in the wax from N. tabacum following exposure to elevated UV-B (Barnes et al., 1996).
Table 5. Association of contact angles with wax composition and crystalline morphology
Suppression of wax production by high humidity and low light intensity is a particular problem associated with tissue culture (Sutter & Langhans, 1979, 1982). Plants grown in vitro are susceptible to dessication due to the lack of waxes and require protective environments (humidity tents, intermittent mist and shade) for 8–10 d for hardening and survival before transfer to a glasshouse. Wax deposition on B. oleracea, E. gunni and Tropaeolum majus was increased by reducing relative humidity from 100 to 35% or from 98% to 20–30% (Baker, 1974; Sutter & Langhans, 1982; Koch et al., 2006). Higher humidity may affect wax morphology (Table 5); for example, formation of larger crystalline tubules in B. oleracea is restricted, leaving dendrites as the major morphological type (Baker, 1974; Koch et al., 2006), and the diameter and degree of clustering of branched rodlets in E. gunni are reduced. High humidity does not always affect wax coverage and composition, for example, as found by Riederer & Schneider (1990) for Citrus leaf cuticles, or structure, as found for T. majus, where the formation of branched tubules is unaffected (Koch et al., 2006).
VIII. Water, salinity and cold stress
1. Water stress
Leaf transpiration has stomatal (TRst) and cuticular (TRcu) components. TRst is controlled by stomatal conductance, largely determined by tissue water status at a given vapour pressure difference between the leaf surface and the air. TRcu is affected by the physiocochemical characteristics of the leaf surface, such as wax thickness and, to a greater extent, wax microstructure, which largely determine the leaf surface's hydraulic permeability and transport characteristics (Svenningsson, 1988; Xu et al., 1995, reviewed by Riederer & Schreiber, 1996; Schreiber et al., 1996). Under extreme water deficit, stomata close and stomatal conductance falls, and water loss by the cuticular route becomes significant (Fig. 4b,d). TRcu is technically difficult to measure accurately (see reviews by Kerstiens, 1996a,b), incomplete stomatal closure being the most significant problem, even under conditions promoting maximal stomatal closure. Therefore the term minimum conductance (gmin) is used to allow for any contribution from (TRst), and unless stomata-free preparations are used, this is the value determined experimentally.
Wax deposition is often a response to water stress, and this can occur rapidly within a few days (Bengston et al., 1978; Premachandra et al., 1991). Stress-resistant plants, for example those adapted to arid conditions, often have thicker waxes than those from more temperate locations or that are otherwise susceptible to stress. However, this is not always the case, Tischler & Burson (1995) found no relationship between geographical location and wax coverage or heat-tolerance in their study of several heat-tolerant bahiagrass (Paspalum notatum) cytotypes. Several studies have attempted to show correlations between gmin, reflectivity, morphology and specific wax components (Tables 3, 6); however, their findings have to be interpreted with caution due to uncertainty about stomatal contributions to the measured values. Although there is no simple relationship between gmin and the amount or composition of wax (Kerstiens, 1996a,b), there does appear to be a relationship with CL. Hauke & Schreiber (1998) reported that an increase in the mean CL of the constituents in ivy, Hedera helix wax from C27 to C33 over the first 60 d of growth was in greater accordance with the observed reduction in gmin than the actual wax composition.
Table 6. Effect of water or salinity stress compared with non-stressed plants on abundance and composition of cuticular wax
Effect on wax quantity
Composition or characteristic associated with adaptation
W, water stress; S, salinity stress; RES, resistant; SUS, susceptible or least resistant; ↓ reduction or ↑ increase in value; ≈ relatively unchanged.
Dryland conditions compared with well watered.
40 d old at start of stress treatment.
Shift in homolog distribution from shorter to longer chain lengths.
The ability of hydrocarbon chains to coalign and form a hydrophobic barrier is well known, for example the supression of evaporation by monolayers of long-chain alcohols (Langmuir & Schaefer, 1943; Mansfield, 1955). The model for the molecular organisation of wax molecules outlined by Riederer & Schneider (1990) and Riederer & Schreiber (1996) involves coalignment of the hydrocarbon backbone of wax constituents normal to the cuticle surface, forming monolayers extending laterally parallel to the cuticle (Sitte & Rennier, 1963), which constitute a solid crystalline zone. Many such sheets may be stacked, and between them are regions populated by the ends of alkyl chains, which could include the polar functional groups of acids, alcohols and aldehydes in addition to terminal methyl groups. If the constituents were of uniform CL, then this intermediate region would be indistinguishable from the crystalline zone. However, waxes usually have a skewed or often bimodal CL distribution, and consequently the intermediate region has a distinct volume fraction into which chain ends may protrude and is considered to be in a solid amorphous state. Adjacent layers may be bridged by intercalation of longer chains such as alkyl esters, and, in principle, secondary alcohols and ketones could also link the layers. The layered sheets are probably discontinuous, surrounded by further solid amorphous regions including components not present in the crystalline regions. Since the crystalline regions are considered impermeable in the barrier membrane model, water and solutes diffuse through the cuticular wax via the amorphous zones, and molecules travel many times further than the simple thickness of the wax, due to the large aspect ratios (width : thickness) of the crystalline regions. Using published crystallographic (Small, 1984) and density data for alkanes (Le Roux, 1969), Jetter et al. (2000) estimated that the cuticular wax of Prunus laurocerasus consisted of an alkane layer 130–160 nm thick, similar to a value of < 200 nm based on scanning electron microscopy (SEM) measurements of detached wax. Assuming molecular packing in cis configuration perpendicular to the cuticle (Sitte & Rennier, 1963), the wax was estimated to be 35–45 molecules thick.
These layered sheets, which may not appear crystalline in SEM images, constitute the basic structure out of which the microcrystalline forms protrude. In some cases, microcrystals may be vertical extensions of the basic structure, or partially detached regions. However, given the association between morphology and composition, it is likely that some microcrystals are outgrowths with a different chemical composition to the underlying sheets. Formation of tubular structures, characteristic of secondary alcohols and ketones, etc., implies a different mode of molecular packing, perhaps imposed by the stereochemistry in the region of the oxo-substituents. Alternatively, different crystal forms may arise from different crystallisation rates as suggested by Baker (1974). Such structures are likely to have amorphous zones, and the inner surface of tubes may be a channel for movement and deposition of wax constituents, perhaps in association with a solvent. This may aid morphological transformations such as dendrite formation on existing structures. If alkyl esters tend to bridge amorphous regions in the underlying layers rather than form ester-specific structures, it may be significant that increased abundance of esters in maize wax appears to reduce the efficiency and amount of visible crystal formation, as reported by Beattie & Marcell (2002).
The significance of alkane CL to the barrier properties of cuticular wax has been elegantly demonstrated by Dodd and co-workers in their studies of the geographic distribution of several tree species (Dodd et al., 1998; Dodd & Rafii, 2000; Dodd & Poveda, 2003). In a study of 27 populations of Austrocedrus chilensis from Chile and Argentina, mediterranean populations adapted to warmer and more arid conditions had higher proportions of longer (C34–C37) alkanes in their waxes than those of mesic populations adapted to cooler and more humid conditions at the rainforest margin. However, within the overall range (C21–C37), both populations had similar CL distributions for the shorter homologs (Dodd et al., 1998). These differences are thought to reflect increased bridging or protrusion of alkanes into the amorphous regions within the wax microstructure of mediterranean populations, impeding cuticular water loss. The annual rainfall related most closely to lower carbon CLs, and annual mean temperature to the longest CLs. It was concluded that the effects were primarily due to ecogenic adaptation.
In a second study, Dodd & Rafii (2000) compared three similar species, A. chilensis, Fitzroya cupressoides and Pilgerodendron uviferu. They span > 2500 km in range and are adapted to different conditions, but with overlapping range. The northernmost, A. chilensis, ranges from northerly mediterranean dry rock sites to the humid temperate rainforest margin and cool dry Patagonian steppe in the south. F. cupressoides occupies a narrow belt of moist coastal and Andean sites and P. uviferum extends to the southern tip of South America, in conditions ranging from warm humid and cool humid to continental. Alkane distributions in cuticular waxes can be analysed statistically to give the weighted mean (N) of CL Cn and dispersion (d) of Cn about the mean, and a lower value of d corresponds to a narrower CL distribution. Crystallinity is inversely proportional to d/N and mixtures with a narrow CL range have a narrower lipid phase transition temperature (Tm) range, retaining their crystalline structure for longer. Consequently, mixtures with higher mean CLs and narrower CL ranges have higher melting points and maintain crystallinity at higher ambient temperatures (Riederer & Schneider, 1990). A. chilensis and F. cupressoides had the highest average values of N (33.2, 32.2) and lowest values of d (1.9, 1.8), respectively, whereas P. uviferum had the lowest value of N (27.2) and highest value of d (3.0). This corresponds to higher melting points for A. chilensis than P. uviferum, where the difference in Tm is c. 12°C. Therefore, an elevated amount of the longest CL alkanes, with a reduced CL range in mediterranean populations of A. chilensis, helps to stabilize the wax's internal structure in the hottest and most arid conditions, and is another facet of an adaptation where reduced transpiration is the driving force. The parameters N and d for alkane CL distribution can also exhibit altitudinal variation. Populations of Juniperus communis from the Pyrenees growing at low and high altitudes had the highest values of N, while d increased with altitude (Dodd & Poveda, 2003). This was considered as indicative of possible adaptations to minimise gmin under hot summer conditions at low altitude and physiological drought caused by freezing at high altitude.
3. Salinity and cold stress
Increased wax deposition on plants such as peanut, Arachis hypogaea and salt-sensitive jojoba on exposure to salinity stress (Rao et al., 1981; Mills et al., 1997, 2001) appears to be primarily a response to water deficit. Wax deposition on leaves of salt-sensitive jojoba is also induced by exogenous abscisic acid (ABA) (Mills et al., 2001). The increase in wax loading caused by exposure of cabbage seedlings to moderate NaCl-induced water stress before transplantation was used to improve drought resistance in transplanted plants and is a useful hardening technique (Fujiwara et al., 2002). Increased leaf wax production by ornamental trees irrigated with treated sewage effluent (reuse water) has been cited by Jordan et al. (2001) as an indicator of levels of foliar damage and stress equivalent to use of saline solution. These findings were used to recommend limitations on spray irrigation with reuse water, which is common practice in municipal areas of the southwestern United States.
Cold stress varies widely in severity and effect. Exposure of maize to a 7-d-long cold spell reduced quantities of cuticular wax on the third leaf of four-leaf plants by 29%, resulting in increased wettability and increased herbicide retention (Gauvrit & Gaillardon, 1991). Under more severe conditions, frozen soil causes frost drought which lowers water potential and inhibits transpiration and CO2 uptake due to stomatal closure (Esch & Mengel, 1998b). Poor frost tolerance in willow, Salix sp., was correlated with increases in wax load and n-alkane content. Leaf surfaces were covered by more wax spheres and the increased hydrophobicity of the wax on a frost-susceptible high-producer clone was reflected in a CA value of 93°, approx. 20° higher than the other clones (Hietala et al., 1997).
IX. Mechanical stress
Under windy conditions, wax can be removed by fracturing of crystals and abrasion due to aerodynamic loading, impact of raindrops, dust and snow, etc., and by leaf-to-leaf contact. In wind tunnel experiments, wax crystals on the leaves of Picea stichensis became flattened and smeared after exposure to airflow at 11 ms−1 for 1 wk, and structures within and around the stomatal antechambers of P. stichensis and Pinus sylvestris were particularly vulnerable (Van Gardingen et al., 1991). Waxes can reform rapidly after mechanical removal, for example the fine structure of broccoli, B. oleracea wax regenerated within 3 d (Schwab et al., 1993). However, rates of replacement may not be fast enough to maintain normal levels (Hall & Jones, 1961). Latimer & Severson (1997) investigated the relative effects of wind, brushing (thigmic stress) and dehydrating moisture stress conditioning (MSC) on B. oleracea as mechanisms for conditioning and hardening glasshouse-grown plants before field planting. After 9 d treatment, reductions in wax levels due to brushing and wind were similar (31–38%), but that due to MSC was 11%. After 15 d, wax was reduced by 15% due to brushing and 6% due to wind, but had increased by 17% due to MSC, indicative of wax redeposition although the rate of removal by brushing was greater than the rate of renewal. There were no changes in wax composition during the treatments and after transplanting to the field, the plants were indistinguishable. A study of the seasonal development of wax on three plantain sp., Hosta plantaginea, H. lancifolia and the glaucous genotype Hosta‘Krossa Regal’, carried out by Jenks et al. (2002) illustrates the typical effects of weathering during a normal growing season. Wax coverage increased initially in the period from the expanding leaf stage to spring, then fell from spring to summer, increasing again from summer to autumn. The mid-season reduction was attributed to weathering by rain, wind and possibly heat stress. Specific compounds, β-diketones (C29, C31 10,12-diones), were abundant on both leaf surfaces of ‘Krossa Regal’ during development up to spring, then remained abundant on the inner abaxial surface, but became minor components on the exposed outer adaxial surface, possibly due to differential weathering. The extent of wax erosion during a growing season can vary greatly between species. Neinhuis & Barthlott (1998) found that wax on ginkgo eroded at a much slower rate than that on oak, Quercus robur, which eroded rapidly a few weeks after leaf expansion ceased. The reason for such differences is unclear but might relate to differences in the susceptibility of different crystal structures to damage, or may reflect a greater ability of ginkgo to regenerate its wax.
Plants growing at high altitude experience interacting stresses, including weathering, dehydration and low temperatures. For some conifers the altitude limit of the alpine treeline may be determined by the extent of winter needle dessication, resulting in increased gmin (reviewed by Kerstiens, 1996a; Körner, 1998). Lack of warmth in the short growing season may impair epidermal development, leaving the needles susceptible to water loss by TRcu during the following winter. Damage may arise from strong winds and abrasion by driven snow, resulting in increased water loss (Tranquillini, 1979; DeLucia & Berlyn, 1983; Hadley et al., 1991; Herrick & Frieland, 1991; Anfodillo et al., 2002). Grace (1990) attributed the treeline limit for Scots pine, Pinus sylvestris, in the Cairngorms, Scotland, to stomatal dysfunction and cuticular damage rather than impaired cuticular development. Increased leaf or needle wax coverage is a characteristic of growth at higher elevation in some species such as Norway spruce, Pinus abies and stone pine, Pinus cembra in southern alpine environments (Günthardt, 1984; Riolo, 1999).
Surface characteristics determined for conifer needles, such as CA, reflectance and gmin, can show altitudinal variation. Anfodillo et al. (2002) found for P. abies that CA both increased with altitude and exhibited a seasonal reduction from winter to summer, except at the lowest altitudes where the pattern was reversed. At high altitudes (treeline) CA ranged from > 113° in February to 83° in August (Table 4) as wax morphology changed from less wettable tubular forms to more wettable planar forms. Similar changes in CA were seen for P. cembra, for which variations were seen between adaxial and abaxial surfaces, the latter having CAs up to 30° greater than the former. Contact angles were also generally greater to windward than leeward. At low (valley bottom) and mid-altitudes, gmin rose from winter to summer, although gmin was unaffected by altitude. This was believed to indicate defective stomatal closure rather than changes in cuticle permeability.
Reflectance (PAR) for the co-occurring Alaskan conifers black spruce, Picea mariana and white spruce, P. glauca increased with both latitude and altitude over the range 60–930 m (Richardson et al., 2003). Both species had similar reflectance spectra, but reflectance was higher in black spruce than white spruce, probably due to differences in wax loading, which may represent a photoprotective adaptation in black spruce for colonisation of more stressful sites. Interpretation of the data suggested that a 1000 m increase in elevation was effectively similar to a 6° increase in latitude. Adaptation to different geographic locations can alter wax composition (see section VIII.2). In a study of the effects of UV-B on geographically adapted Poa sp., Pilon et al. (1999) found that differences in wax loading and composition, or production of UV-absorbing compounds such as triterpenes, were associated with location rather than exposure to UV-B. Wax loading on alpine plants was at the low end of the range found for all plants, although there was no clear altitudinal effect. Alkanes and aldehydes were far more abundant or were only found in alpine plant waxes, whereas primary alcohols were more abundant from lowland plants. Triterpenes (squalene, lupeol, β-amyrin) were found in waxes from lowland plants but were absent from the alpine varieties. Levels of esters were similar.
In high concentrations, and under prolonged exposure, acidic and oxidising gases and aerosols can help to degrade cuticular wax and impair stomatal function (Roberts & Cannon, 1989). This may impose chronic water stress, which can lead to defoliation and death. Various aspects of the interaction between atmospheric pollutants and the leaf cuticle have been reviewed by Heath (1980) and contributors in Percy et al. (1994a).
1. Acid rain
The wettability of cuticular wax influences the retention and spread of acidified droplets and potentially the extent of damage. Acidity is also a factor: greater damage is caused by rain or mist at pH 3 than at pH 5 (Knittel & Pell, 1991; Esch & Mengel, 1998a; Hoad et al., 1998). Formation of necrotic lesions within the leaf is characteristic of acid damage, which may be accompanied by damaged or altered wax morphology, resulting in an increase in gmin and water stress. Consequently, plants previously exposed to acid rain can become more sensitive to drought and dessication under water deficit, causing more pronounced stomatal closure and loss of turgor than for drought alone, as reported by Mena-Petite et al. (1999) for Pinus radiata, and by Esch and Mengel (1998a,b) for spruce, Picea abies, under frost drought conditions. However, this need not always apply; for example, the susceptibility of Zea mays to drought stress was unaffected by simulated acid rain, and the wax was morphologically unchanged (Knittel & Pell, 1991). The damaging effects of acid precipitation are enhanced when combined with wind action. Hoad et al. (1998) found that the greatest damage to birch occurred with a combination of direct or indirect wind action and acid mist at pH 3. By contrast, sheltered leaves were very resistant to the effects of the acid mist.
2. Wax erosion and stomatal damage
Airborne pollutants accelerate erosion of cuticular waxes; for example, aerosols containing antimony (Sb) eroded wax on Japanese cedar, Cryptomeria japonica 1.5 times faster than found for unpolluted trees (Sase et al., 1998; Takamatsu et al., 2001a,b). Erosion rates also depend on plant age and geographic location. In conifers, juvenile waxes are particularly vulnerable to damage; for example, wax degradation by sulphur dioxide (SO2) and absorption of S by Scots pine, Pinus sylvestris were greatest on the two youngest needle age classes (Manninen & Huttunen, 1995). Pollution-induced changes to wax morphology in the stomatal antechambers of some conifer needles are similar to ageing effects, starting with fusion of wax rods at the tip, ultimately leading to loss of crystalline structure (Bermadinger-Stabentheiner, 1994; Huttunen, 1994). Resistance to damage can vary between species exposed to similar pollution loads. Waxes on young needles of Norway spruce were damaged within a year of exposure, whereas those of silver fir, Abies alba required 3 years’ exposure before damage occurred (Bednarova, 2001a). Reduced pollution levels can lead to reduction in damage and ultimately to recovery (Bednarova, 2001b). Stomatal damage is common and may include collapse, depression, fusion or degradation of guard cells, and occlusion with wax clumps (Paoletti et al., 1997). In addition, acid rain and drought can leach calcium, causing formation of CaSO4 microcrystals in pine needles, particularly at the interface between stomatal antechambers and substomatal cavities (Huttunen et al., 1989; Turunen et al., 1994; Pritchard et al., 2000). This process effectively neutralises the acid and may provide protection from further damage.
Ozone is a common air pollutant and some of its effects may be related to membrane degradation via oxidation of double bonds in fatty acids and sulfhydryl groups in proteins. At 65 nl l−1, ozone severely reduced formation of new cuticular wax via the synthesis de novo, although the existing wax was unaffected (Carlsson et al., 1994; Hellgren et al., 1995). Ozone reduced wax synthesis in red spruce needles (Percy et al., 1992), degraded wax in Norway spruce (Barnes et al., 1988) and decreased chloroplast size and photosynthesis in Norway spruce (Sutinen et al., 1990; Wallin et al., 1990) and wheat (Ojanperäet al., 1992). Impaired wax synthesis could arise from a reduced supply of photosynthetically fixed carbon to the synthesis de novo and by disruption of the membrane integrity of the plastids where the synthesis occurs.
4. Carbon dioxide
According to some reports, climate change may double atmospheric CO2 levels by the end of the century (Long et al., 2004), accompanied by increasing temperatures, changes in rainfall patterns and increased UV-B exposure due to ozone depletion (see also section IV). Consequently, there is much interest in interaction between these parameters, particularly whether the generally beneficial effects of elevated CO2 may compensate for the deleterious effects of the other parameters, in particular increased UV-B. Effects of elevated CO2 are variable. Vanhatalo et al. (2001) found that elevated CO2 alone, and in combination with O3, increased wax coverage on the abaxial leaf surface of an inland clone of pubescent birch, Betula pubescens, but not on that of a coastal clone. By contrast, wax coverage fell when longleaf pine, Pinus palustris, was exposed to elevated CO2 in studies of the interactions between elevated CO2, water status and nitrogen supply (Prior et al., 1997), and carbon allocation under conditions of elevated CO2 and water stress (Runion et al., 1999). Similarly, wax coverage on the Crassulacean acid metabolism (CAM) species Agave deserti was reduced by 40% when CO2 levels were doubled, reflectance was 14% lower and transmittance 6% higher (Graham & Nobel, 1996). Elevated CO2 can partially ameliorate some of the adverse effects of UV-B radiation in canola, Brassica napus (Qaderi & Reid, 2005). Cuticular wax coverage similarly doubled under elevated levels of CO2 and UV-B alone. Although in combination there was a further slight increase, elevated CO2 increased wax deposition sufficiently to provide protection against UV-B. Percy et al. (2002) investigated the effects of elevated CO2, and O3 on Aspen, Pupulus tremuloides, singly and in combination. They found that both CO2 and O3 alone increased cuticular wax production (16 and 23%, respectively); however, in combination, these effects were negated.
5. Direct effects on wax composition
There is a growing body of evidence (Table 7) that pollutants can modify wax composition, and that they may therefore have a direct effect on the biosynthesis of wax constituents, affecting both the distribution of products from the associated pathways and CL distribution. Percy et al. (1994b) have also shown, using recrystallised whole wax, that most wax constituents, once crystallised, are stable to treatment with ozone at 70 and 150 p.p.b. and acid at pH 3.0 and 1.0, levels of acidity to which plants can be exposed under some pollution scenarios. In exception to this, alkyl esters are subject to hydrolysis with increasing acidity, being converted to totally free fatty acids at pH 1.0.
Table 7. Effect of atmospheric pollutants on wax chemical composition
AR, simulated acid rain, mist or fog; CL, chain length; ↓ reduction or ↑ increase in value.
XII. Genetic and environmental control of cuticular wax production
Detection of, and response to, environmental stimuli, for example, osmotic stress induced by cold, drought and salinity, involve a sequence of events starting with detection of a stress-induced signal and leading via a cascade of signalling steps to the final response. Intermediate stages can involve release of secondary signalling molecules such as inositol phosphates, reactive oxygen species and ABA, modulation of intracellular Ca2+, initiation of protein phosphorylation cascades and induction of proteins involved in cellular protection and trancriptional control of stress-related genes (reviewed by Xiong et al., 2002; Zhu, 2002). Although numerous forms of stress influence cuticular waxes, they commonly induce changes in the amount and composition of wax, effects which are closely associated with the control of biosynthesis. Current understanding of this process comes mainly from the use of chemical and photoperiod inhibition (von Wettstein-Knowles, 1979) and mutants such as eceriferum (cer) in Arabidopsis and barley and gl in maize and various brassicas to block biosynthesis at particular points, leading to measurable accumulation of intermediates (Table 8a,b). Chain elongation and the channelling of acyl intermediates into the products of the reductive and decarbonylation pathways are shown schematically in Fig. 7. ACCase has a key role in the regulation of acyl chain production by supplying malonyl-CoA (Fig. 2) for elongation. Activation of partly purified ACCase from pea in vitro involves light-stimulated reduction of the enzyme by thioredoxin (Sasaki et al., 1997), and the activity of the CT β-subunit may be regulated by phosphorylation and dephosphorylation (Savage & Ohlrogge, 1999). Plastidial chain extension by FAS involves multiple tissue-specific isoforms of ACP, and in Arabidopsis, expression of the leaf-specific ACP4 is light-stimulated (Bonaventure & Ohlrogge, 2002), possibly based on the presence of I-box like motifs found in light-inducible promoters (Terzaghi & Cashmore, 1995). Post-transcriptional control is also exerted due to the greater association of ACP transcripts with polyribosomes in the light than in the dark.
Table 8. Effects of inhibitors, photoperiod and mutation (a) On elongation and (b) elongation and compound distribution within the associated pathways (Adapted from Shepherd (2003) with permission from Academic Press.)
The Claisen reaction catalysed by KAS is another key event during chain elongation by FAS (Fig. 2, step c), and multiple KAS isoforms with different CL specificities were revealed through the effect of inhibitors such as cerulenin and arsenite on fatty acid synthesis (Fig. 7). Formation of C4 chains involves KAS III, further elongation to C14 and C16 involves KAS I, and KAS II catalyses the final step to C18 (von Wettstein-Knowles, 1995; von Wettstein-Knowles et al., 2000; Carlsson et al., 2002). Genes encoding the major KAS isoforms have been cloned and expressed sequence tag (EST) clones for each KAS have been identified from various species. Extensive homology exists between KAS I and II, whereas KAS III is similar to another condensing enzyme, polyketide chalcone synthase (CHS). Transfer of acyl groups from acyl-ACP to KAS I (Fig. 2 step g) is inhibited by free ACP formed during the process and the inhibition is reversed by acylation of ACP, whereas ACP has little effect on KAS II or KAS III. However, C16-ACP is a feedback inhibitor of KAS III, and therefore overproduction of acyl chains by KAS I inhibits initiation of new chains by KAS III. Reduced KAS II activity in the Arabidopsis mutant fab1 results in accumulation of palmitic acid (C16), rendering the mutant susceptible to low temperature stress (Carlsson et al., 2002).
Defects at specific points in elongation by FAE are features of many of the cer, gl and other mutants, resulting in accumulation of shorter products than found for wildtype plants. Chemical inhibition and variation in photoperiod can also give similar effects, and, overall, impaired elongation can occur between C16 and C18, and at each step between C20 and C32 (Fig. 7, Table 8). These findings support the presence of sequential elongation systems with multiple isoforms of elongase, or subunits such as the condensing enzyme KCS, with specificity for a particular step or sequence of steps in elongation. The associated pathways may also be defective, changing the distribution of the different classes of compound (Fig. 7, Table 8b). Such alterations are often responsible for the phenotype of the mutation (von Wettstein-Knowles, 1995; Kolattukudy, 1996; Post-Beittenmiller, 1996; Kunst & Samuels, 2003), which bears a certain similarity with some of the stress-induced phenotypic modifications.
Of the many genes involved in wax biosynthesis, relatively few have known identities or function. Some involved with elongation and the associated pathways have been cloned, and their gene products studied. Genes encoding KCS have been identified from seeds of Arabidopsis, brassica and jojoba. Several Arabidopsis genes, KCS1 (Todd et al., 1999), CER 6 (CUT1) (Millar et al., 1999; Fiebig et al., 2000; Hooker et al., 2002), FDH (fiddlehead) (Yephremov et al., 1999; Pruitt et al., 2000) and HIC (high carbon dioxide) specific to stomatal guard cells (Holroyd et al., 2002; Hetherington & Woodward, 2003), encode KCS in leaves and stems. Up to 20 members of the Arabidopsis KCS gene family were recently found in epidermal tissues of stem tops during gene expression studies, with KCS1, CER6 and FDH having the greatest expression levels (Suh et al., 2005). The kcs1-1 and cer6 mutations affect each elongation step between C24 and C30, with reduced levels of the corresponding C26–C30 aldehydes and primary alcohols (kcs1-1, cer6) and C29 alkanes, secondary alcohols and ketones (cer6). The main effect of cer6 occurs between C24 and C26; however, for kcs1-1 no complete blockages in CL extension are evident, suggesting that there is considerable redundancy in elongase activity and CL specificity. Of the other enzymes associated with FAE, maize gl8 encodes a 3-ketoacyl reductase localised to the ER (Xu et al., 1997, 2002; Dietrich et al., 2005) and similar sequences have been found in Arabidopsis, barley and leek. Zheng et al. (2005) have identified from Arabidosis the first enoyl-CoA reductase associated with FAE. Although the biochemical function of Arabidopsis CER1 is unknown, it has been suggested that it encodes a fatty acid decarbonylase (Hannoufa et al., 1993; McNevin et al., 1993; Aarts et al., 1995; Jenks et al., 1995). TAA1a from bread wheat encodes a fatty acyl-CoA reductase which forms the alcohol moieties of wax esters present in the tapetum tissues of pollen (Wang et al., 2002). Arabidopsis CER24 and WAX2/YRE may be acyl-CoA reductases involved in alcohol (Rashotte et al., 2004), and alkane (Chen et al., 2003; Kurata et al., 2003) formation, respectively, and RST-1 (resurrection 1) also appears to be involved in the latter reaction (Chen et al., 2005), although its function is unknown. Maize GL1 has a high amino acid sequence homology with Arabidopsis WAX2, and is probably the maize WAX2 ortholog (Sturaro et al., 2005).
In some species, separate but parallel elongation systems appear to be linked with the different associated pathways, which may not share a common pool of precursors. This may also be characteristic of different developmental stages. In maize and barley, an elongation system, ED-I, active in seedlings, is responsible for formation of very long-chain alcohols, aldehydes, acids and alkanes (Fig. 7). A second system, ED-II, active over all stages of growth, produces shorter acyl chains primarily for ester formation, although some ED-I products may also be made (Mikkelsen, 1978; Avato et al., 1984, 1990). ED-1 is defective in some maize gl mutants or through the action of trichloroacetic acid (TCA), and a shift occurs in the acyl CL distribution away from C32 to shorter chains, particularly in primary alcohols (Avato et al., 1984; Bianchi et al., 1985; Avato et al., 1990; Beattie & Marcell, 2002). This has a strong parallel with the light-induced changes in CL found by Giese (1975) for the same components in barley wax (Fig. 5). In barley and wheat, ED-1 is paralleled by another system which elongates 3-ketoacyl chains for production of β-diketones (Fig. 3). A mutation in barley CER-CQU, which encodes a multifunction enzyme with 3-ketoacyl elongase activity, impairs formation of β-diketones, while alkane and alcohol formation is unaffected (von Wettstein-Knowles, 1995). A similar system to ED-I functions in wheat seedlings, whereas the 3-ketoacyl elongation system becomes active in mature plants. Such a switch between parallel pathways for formation of alkanes and β-diketones appears to occur in some species under water or salinity stress (von Wettstein-Knowles, 1972; Bengtson et al., 1978; Rao et al., 1981).
Branched constituents common to some species are thought to come from elongation systems with a specificity for br- precursors. In brassicas, br- (i-, a-) constituents are abundant in free and esterified fatty acids and alcohols; however, br- components are not usually found in the other products of the associated pathways. Defects in this specificity for n-compounds occur in some brassica mutants (gl4, gl6 and glNilla), resulting in the formation of br-alkanes and br-secondary alcohols (Netting et al., 1972; Baker & Holloway, 1975). Changes in the distribution of br- and n-components under different lighting regimes are features of tobacco and brassica waxes (Barnes et al., 1996; Shepherd et al., 1997), again suggesting that the stress, in this case exposure to UV-B, changes the activity of parallel elongation systems.
Developmental factors may have a key role in control of biosynthesis, since provision of a waterproof protective coat is essential at the earliest stages of growth. There is evidence from studies with leek, Allium porrum, that induction of microsomal elongation by FAE leads that of plastidial de novo synthesis by FAS, suggesting that the processes are under different developmental regulation (Rhee et al., 1998). Developmental patterns can vary between species, for example, onset of wax synthesis in leek follows cell elongation, whereas expression of Arabidopsis CER2 and CER3 is associated with elongating tissues (Xia et al., 1996) and cell division in the meristem (Hannoufa et al., 1996), respectively. Therefore, environmental factors are more likely to modulate normal developmental signals. How this occurs is not well understood. Stimulation by light is a key process in the synthesis de novo and probably constitutes the single greatest factor determining the light sensitivity of wax production. Some details have emerged about the influence of environmental factors on components of FAE and other enzymes involved in wax biosynthesis. Hooker et al. (2002) found that light is essential for expression of Arabidopsis CER6, accumulation of the CER6 transcript is also enhanced by osmotic stress and ABA, and overexpression of CER6 increases wax accumulation. The CER6 promoter region contains elements similar to I-box and GT1-binding sites found in light-inducible promoters (Terzaghi & Cashmore, 1995), and ABA-responsive cis-acting elements (ABRE) involved in ABA-regulated gene expression by drought and cold (Guiltinan et al., 1990). If this is more general of the other KCS condensing enzymes, it may contribute to the CL sensitivity of FAE and may form part of a signalling mechanism for regulating wax production under osmotic stress. By contrast, expression of the regulatory gene CER2 is not induced by light, temperature, drought or other stress conditions, but may be induced by cytokinins in young leaves (Xia et al., 1997) or by specific trancription factors.
Genes encoding three transcription factors, designated SHN (shine), including SHN1/WIN1 (wax inducer 1) (Aharoni et al., 2004; Broun et al., 2004), have recently been identified in Arabidopsis. From the observed phenotypes of the shn mutant and WIN1 overexpressors in leaves, they may influence early chain elongation, particularly between C18 and C22, which affects the overall quantities of all biosynthetic end-products. They may differentially regulate channelling of precursors towards decarbonylation rather than the reductive pathways (SHN/WIN1), or more specifically competing reactions utilising C30 acyl-CoA, that is, decarbonylation to form C29 products or further elongation to give C32 products (SHN). Within the decarbonylation pathway they appear to exert control on the hydroxylation and oxidation reactions that form C29 secondary alcohols and ketones (SHN). Overexpression of WIN1 in transgenic Arabidopsis induced some genes involved in wax biosynthesis (KCS1, CER1, CER2), whereas others (CER6/CUT1, FDH, CER3) were unaffected (Broun et al., 2004). This is consistent with the observed phenotype based on the known functions of the respective gene products (Fig. 7, Table 8a,b). However, a loss-of-function shn mutant has not yet been found, and therefore the regulation of wax biosynthesis by the three SHN transcription factors in epidermal cells remains to be confirmed. Another transcription factor possibly involved in regulating wax biosynthesis, WXP1 (wax production), was recently characterised from the model legume Medicago truncatula (Zhang et al., 2005). When overexpressed in transgenic alfalfa, Medicago sativa, WXP1 activated wax production, reduced water loss and enhanced drought tolerance. Expression of WXP1 in M. truncatula shoots was reversible, being induced by drought and ABA, but on removal of the stimuli, transcription levels returned to normal. Alcohols derived from the reductive pathway are the major components in alfalfa leaf waxes, whereas alkanes formed by decarbonylation predominate in stems. The main effect of WXP1 was to increase production of the C30 alcohol in leaves, whereas there was little change in stem wax composition, which is consistent with the specific involvement of WXP1 in the reductive pathway, in contrast with Arabidopsis SHN1/WIN1, which is mainly involved in the decarbonylation pathway. However, WXP1 also up-regulated three FAE genes, orthologs to Arabidopsis CER2, pointing to its possible involvement in regulation of acyl chain elongation as found for SHN1/WIN1. Greater resistance to cuticular transpirational water loss with an increase in the abundance of the longer chain C30 alcohol is consistent with mechanistic effects similar to those described previously for alkanes in cuticular waxes (section VIII.2).
The SHN/WIN proteins belong to the plant-specific family of APETALA (AP2) (DNA binding-domain)/ethlene-responsive element binding factor (EREBP or ERF) transcription factors associated with plant development and response to environmental stress (Singh et al., 2002). Different members of this family regulate genes involved in response to drought, cold, jasmonates and ethylene as part of the signalling cascade associated with plant sensing and response to environmental stimuli (Xiong et al., 2002; Zhu, 2002). WXP1 also belongs to the APT2 domain-containing group of transcription factors, but is distinct from most of the known examples on the basis of phylogenetic protein sequence analysis (Zhang et al., 2005).
Cuticular waxes have protected plants from the vigours of their environment for a very long time and it is only now becoming clear how this is achieved. There is an elegance about the ways in which plants have come to assemble a few simple molecules into the battery of protective structures and mechanisms that are deployed as their outermost line of defence. Our understanding of the interrelationship between wax composition and function and the basic mechanisms of their production is growing, but there is still much that needs clarification. Some of the major outstanding issues are the mechanisms of wax deposition on to the cuticle surface, control of composition at genetic, molecular and morphological levels, and how environmental pressures elicit an adaptive response. It is becoming apparent that in Arabidopsis, the CER6 gene may play a fundamental role in the plant's response to all types of environmental challenge, but it remains to be seen how general this may be. As of yet relatively little is known about what regulates the conversion of elongated acyl chains into the various products of the associated pathways, or indeed the mechanisms of such transformations. However, the identification of stress-inducible transcription factors in M. truncatula and A. thaliana, which appear to differentially regulate the utilization of acyl chains by the reductive and decarbonylation pathways, respectively, offers an insight as to how this might occur more generally. It has been suggested that overexpression of genes encoding transcription factors might be an effective general method for manipulation of wax biosynthesis in leaves (Zhang et al., 2005). This could be very useful both for mechanistic investigations and for transgenic production of stress-resistant clones of economically important crop species.
The authors would like to acknowledge the support of the Scottish Executive Environment and Rural Affairs Department.