Visualizing the Air-To-Leaf Transfer and Within-Leaf Movement and

However, most of the visualized phenanthrene reaching the leaves arrived via gas-phase transfer. Phenanthrene was found within the wax plugs of stomat...
0 downloads 0 Views 1MB Size
Download PDF
Environ. Sci. Technol. 2006, 40, 907-916

Visualizing the Air-To-Leaf Transfer and Within-Leaf Movement and Distribution of Phenanthrene: Further Studies Utilizing Two-Photon Excitation Microscopy EDWARD WILD,* JOHN DENT, GARETH O. THOMAS, AND KEVIN C. JONES Departments of Environmental and Biological Sciences, Lancaster University, Lancaster, LA1 4YQ, UK

Two-photon excitation microscopy (TPEM) was used to monitor the air-to-leaf transfer and within-leaf movement and distribution of phenanthrene in two plant species (maize and spinach) grown within a contaminated atmosphere. Phenanthrene was visualized within the leaf cuticle, epidermis, mesophyll, and vascular system of living maize and spinach plants. No detectable levels of phenanthrene were observed in the roots or stems of either species, suggesting phenanthrene entered the leaves only from the air. Phenanthrene was observed in both the abaxial and adaxial cuticles of both species. Particulate material (aerosols/ dust) contaminated with phenanthrene was located at the surface of the cuticle and became encapsulated within the cuticular waxes. Over time, diffuse areas of phenanthrene formed within the adjacent cuticle. However, most of the visualized phenanthrene reaching the leaves arrived via gas-phase transfer. Phenanthrene was found within the wax plugs of stomata of both species and on the external surface of the stomatal pore, but not on the internal surface, or within the sub-stomatal cavity. Phenanthrene diffused through the cuticles of both species in 24-48 h, entering the epidermis to reside predominantly within the cell walls of maize (indicative of apoplastic transport) and the cellular cytoplasm of spinach (indicative of symplastic transport). Phenanthrene accumulated within the spinach cytoplasm where it concentrated into the vacuoles of the epidermal cells. Phenanthrene was not observed to accumulate in the cytoplasm of maize cells. Phenanthrene entered the internal mesophyll of both species, and was found within the mesophyll cell walls, at the surface of the chloroplasts, and within the cellular cytoplasm. Phenanthrene was observed within the xylem of maize following 12 days exposure. The cuticle and epidermis at the edges of spinach leaves had a systematically higher concentration of phenanthrene than the cuticle and epidermal cells at the center of the leaf. These results provide important new information about how such compounds enter, move, and distribute within leaves, and suggest that contemporary views of such processes based on data obtained from traditional analytical methods may need to be revised.

* Corresponding author e-mail: [email protected]; tel: 44 01524 593300; fax: 44 01524 593985. 10.1021/es0515046 CCC: $33.50 Published on Web 12/16/2005

 2006 American Chemical Society

Introduction Vegetation plays a key role in the environmental fate and cycling of many semivolatile organic chemicals (SVOCs) and pesticides (1, 2), scavenging them from the atmosphere, storing and processing them, and transferring them into other environmental compartments (3-5). While associated with vegetation, SVOCs (including polycyclic aromatic hydrocarbons (PAHs), polychlorinated dibenzo-p-dioxins and -furans (PCDD/Fs), polychlorinated biphenyls (PCBs), and organochlorine (OC) pesticides) may be subject to various environmental fates, including photolysis (6-8), cellular metabolism (9, 10), seasonal uptake and storage (3), movement into terrestrial food chains (11, 12), and transport into the soil following leaf senescence (3). Their fate will be influenced by their location within the plant and the pathway by which they travelled to get there. For example, compounds stored in the near-surface cuticle may be more likely to undergo revolatilization, photodegradation, and cuticular shedding, whereas compounds which have reached the inner cells may be more readily metabolized (13). It is therefore important to identify exactly how these chemicals enter vegetation and where they become stored within it, to understand their potential fate within the environment as a whole. This is currently an issue of considerable uncertainty. The dominant transfer route for SVOCs to above-ground vegetation is thought to be from the atmosphere to the leaf; the surface area of leaves in contact with the air can be 6-14 times greater than the land the vegetation is growing upon (14). The leaf surface of vascular plants is protected by the waxy cuticle which can further increase the available surface for air-surface interactions by a factor of 1-2 by virtue of its microscopic structure (15). The leaf cuticle is a hydrophobic lipid structure synthesized by the epidermal cells. It has five generalized compartments: the epicuticular wax, the cuticle proper, the cuticle layer, pectinous layer, and cell wall (Figure 1) and is essential for the protection and waterproofing of the leaf (16). Its thickness can range from less than 1 µm (e.g., in Prunus laurolerasse) to tens of µm (e.g., Olea lancea) (17). Uptake of SVOCs by aboveground vegetation occurs through gaseous, dry particulate, and wet deposition. The relative importance of these pathways is dependent on local atmospheric conditions, the volatility of the chemical, and the morphology and physiology of the plant (1, 4, 5). Relatively volatile SVOCs with log octanol/air partition coefficients (Koa) 9 (e.g., benzo[a]pyrene) are more likely to be deposited to the leaf cuticle by particulate deposition (18). It is believed that the rate of compound transfer through the atmospheric boundary layer and/or the rate of diffusion through the epicuticular wax may provide limiting factors in the uptake of organic chemicals to the leaf (19, 20). Once particles adhere to the cuticle surface, SVOCs may redistribute to the waxy cuticle, influenced by the nature of their association with the particle and their affinity for cuticular waxes (19, 21). Particulates may also be removed from the cuticular surface by abrasion through wind and rain, or cuticular shedding. However, if the particulates become partially or wholly encapsulated into the cuticular matrix, the particle may not be so easily removed, aiding chemical movement into the cuticle. Vapor-phase compounds can be taken up by plants via the leaf cuticle or stomata, with the former generally believed to dominate (19). VOL. 40, NO. 3, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

907

FIGURE 1. Schematic of a leaf section, depicting the cuticle, epidermis, stomata, mesophyll, and vascular system. Diffusion through the boundary layer above the leaf surface is affected by wind speed, because it controls the boundary layer thickness. The boundary layer is likely to be thinnest at the leaf edge and become thicker over the leaf surface; it is also affected by the leaf surface topography (22). Vaporphase organics may bypass the cuticle and directly enter the internal mesophyll of the leaf through the stomataspores within the epidermis which allow the free exchange of gases between the atmosphere and photosynthetically active mesophyll (19). In summary then, there are different pathways by which chemicals can enter the leaf. Chemicals must diffuse through the cuticle or pass through the stomata to reach the epidermis or internal mesophyll of the leaf (Figure 1). Movement through the cuticle will be affected by its composition and structure and the properties of the compound (20, 23). The generalized plant cuticle is composed of the aliphatic bio-polymers cutin and cutan and cuticular waxes (commonly composed of alcohols, acids, esters, aldehydes, and ketones). These waxes may be found at the cuticle surface, as epicuticular waxes, and within the polymer matrix, as intra-cuticular waxes. The proportions of crystalline waxes (which impede chemical movement) and amorphous waxes (which are more permeable to chemical movement) and the polarity of the wax and chemical are likely to influence diffusion through the cuticle into the leaf. It is believed that the cuticular waxes form an effective transport barrier into the leaf (20) where the mobility of organic substances within the plant cuticle is size selective, with an increase in molecular weight resulting in a decrease in mobility (20). Each subcompartment within the cuticle may play a different role in the uptake, movement, and ultimate fate of a chemical within the leaf (Figure 1). For instance, it is suggested that the epicuticular wax may be one of the major limiting factors in the uptake of chemicals into leaves (20). Following uptake to the leaf cuticle SVOCs may re-volatilize to the atmosphere, undergo photodegradation, become stored within the cuticle itself, or diffuse through it to the outer membrane and cell walls of the epidermal cells. Here they may enter the leaf apoplast, moving through and becoming stored within the cell walls and intercellular spaces of the leaf, or enter the symplast, moving through and becoming stored within the cellular cytoplasm and vacuoles of the leaf epidermis and mesophyll (24). 908

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 3, 2006

Following transport via these routes, the chemical may also approach, enter, and travel through the vascular system of the leaf (24). However, it is generally believed that SVOCs remain/partition within the lipid-rich portions of the plant (25). As the preceding sections have shown, plant leaves are structurally complex, and the way chemicals reach them, move through them, and become stored/processed by them will affect their environmental fate. Our existing knowledge of these issues has often relied on inference, because the traditional methods used to study the uptake and compartmentalization of organic chemicals within leaves involve destructive analysis of the leaf or leaf parts, through the extraction of intact cuticles or cuticular waxes, 14C uptake studies, and adsorption/desorption experiments. These approaches often cannot definitively differentiate between or separate compartments of the living plants. Recently, however, we reported the development and use of a new technique utilizing two-photon excitation microscopy (TPEM) and plant and chemical autofluorescence to visualize the in situ uptake, movement, storage, and degradation of organic chemicals within unmodified living plants (26-28). The technique allows the direct visualization of organic chemicals within whole living plants, in real time and at the cellular level. Our previous work has reported the diffusion of PAHs through the cuticle into the epidermis of living maize leaves (26), real-time PAH photodegradation on and within living maize leaves (27), and the uptake, storage, and degradation of PAHs in roots of maize and wheat (28). The previous leaf studies on compound diffusion and photodegradation had been undertaken by directly applying compound in small quantities of solvent to the leaf surface, to introduce compound to the leaves in sufficient quantities to be detectable by TPEM. However, as explained above, relatively volatile PAHs, and other important SVOCs normally reach leaf surfaces via the gas phase. This study therefore extends the application of this pioneering technique to visualize the uptake, accumulation, and compartmentalization of phenanthrene by the leaves of maize (a C4 monocotyledon) and spinach (a C3 dicotyledon) when grown within a contaminated atmosphere over a 12-day period. Phenanthrene was chosen because it is the most abundant PAH in the air and

vegetation (29), and it has good fluorescence properties suited to TPEM.

Materials and Methods Maize (Zea mays) and spinach (Spinacia oleracea) plants were grown within a chamber containing elevated levels of vaporphase phenanthrene for 1-12 days. The plants were grown hydroponically. Chamber Preparation. A 54-dm3 glass chamber was used. Pure phenanthrene (99.9%) was obtained from Aldrich Chemical Co., and 30 mg was dissolved in 10 mL of acetone in an amber container. This was coated onto the inner glass walls of the chamber using a glass pipet and allowed to dry to produce an even layer. Air was circulated with a 12-V fan within the chamber to simulate ambient turbulent conditions and hence encourage gaseous exchange. The plants were grown within the center of the chamber with at least 5 cm separating the leaves from the chamber walls, to ensure no direct transfer of phenanthrene occurred. Plant Preparation. Maize and spinach seeds were obtained from Unwins Seeds Ltd (Chester, UK). They were germinated on moist cotton wool in a darkened environment at 25 °C for 2-6 days, then placed into individual glass growth containers, and grown hydroponically. Hydro Tops Bioponic Grow & Bloom, and Essentials Hydroponic Oxygen Increase were used in the proportions specified by the manufacturers. Plants were grown under a 16-h photoperiod illuminated by a 400-W Na solar light. The contamination chamber was maintained at ∼25 °C. Plants of either species were added to the chamber only after leaves 2-3 had developed. Plants were grown within the chamber for 1-12 days, and analyzed using TPEM after 1, 2, 3, 4, 5, 6, and 12 days. Three leaves, taken from between leaves 2-4 (maize) and leaves 3-8 (spinach), were analyzed from each living plant, and at least 3 plants were analyzed for each time period with repeat experiments. The plant roots were also analyzed after 12 days to ensure that phenanthrene was not entering the plants via any means other than the air. Immediately prior to analysis by TPEM a plant would be removed from the contaminated chamber, the roots were wrapped in moist cotton wool and foil, and the leaves were sealed in a foil envelope. The whole living plant was then immediately transferred to the TPEM for analysis. Control plants were prepared as stated above, but not added to the contaminated chamber. After analysis, the living plants could be returned to the chamber to continue growing if required. Determination of Air and Plant Concentrations. Air and plant concentrations were determined by reverse-phase highperformance liquid chromatography (HPLC) using water and acetonitrile mobile phase with fluorescence detection. Six air samples, each of ∼10 dm3, were taken directly through Florisil (6-g) traps. Gas-phase phenanthrene was retained on the Florisil and eluted with DCM. Plant samples were ground in liquid nitrogen and Soxhlet extracted in 250 mL of DCM and cleaned on columns containing silica gel and alumina. Five individual leaves of each plant species were analyzed for their concentrations after 5 days exposure in the chamber. Three unexposed control plants were also analyzed. Dibenzo[ah]anthracene was used as an internal standard. Recoveries of dibenzo[ah]anthracene ranged from 50 to 90%; all phenanthrene concentrations were corrected to 100% recoveries. Instrumentation. A Bio-Rad Radiance 2000 MP scanning system was used with a Spectra Physics Tsunami/millennia tuneable laser (690-1050 nm) and a Nikon Eclipse TE300 inverted microscope fitted with a Nikon 60x/1.20 Plan Apo D.I.C water immersion lens. The laser wavelength was set to 700 nm. Direct detectors Bio-Rad bialkali photomultiplier tubes (PMTs) were used for all imaging having higher quantum efficiency at the blue end of the spectrum, where

phenanthrene was visualized. Compounds containing fluorophores will fluoresce if excited by a range of wavelengths. Individual compounds show unique fluorescence emission spectra, allowing their unambiguous identification. Twophoton excitation at 700 nm was used, equating to absorption at the focal point of approximately 350 nm. The emission detection filters and dichroic mirrors used were as follows: emission filter A (UG 11/IR), dichroic mirror (UV 400 DCLP), and emission filter B (HQ 495/20). The UG 11/IR filter transmitted between 300 and 390 nm, which was detected by PMT 1, corresponding to the emission of phenanthrene between 345 and 390 nm using blue as the pseudo color channel. The HQ 495/20 filter transmitted between 485 and 505 nm detecting all plant components including the cuticle, hairs, cell walls, and chloroplasts using green as the pseudo color channel. Images were collected and processed using Bio-Rad Lasersharp 2000 imaging software, confocal assistant 4.02, and Amira 3.1.1. Images were made using XY and XZ scans, and 3D reconstructions of the samples were produced using Amira 3.1.1 software. Initially, the fluorescence characteristics were determined for pure phenanthrene applied to glass slides, when subject to two-photon excitation at 700 nm. Slides coated in a thin film of either pure octacosane wax or paraffin wax ∼25-µm thick were also added to the chamber for 1-6 days, to observe the uptake of phenanthrene into the wax, to ensure its emission profile as a diffuse layer could be determined. Control plants and pure wax samples showed no fluorescence within the blue channel (phenanthrene) for all samples and both species (see Figure 1 in the Supporting Information (SI)).

Results and Discussion Phenanthrene Uptake to Wax Slides within the Contaminated Chamber. Uptake of phenanthrene to wax slides was initially observed at the wax surface. Within 24 h the wax showed an even distribution of phenanthrene through approximately one-third of the wax volume, extending to two-thirds within 48 h, and throughout the full wax volume within 72 h. The pure octacosane wax showed a more even distribution of phenanthrene than the paraffin wax. Microstructural differences could be observed within the paraffin wax and associated with these were areas of more concentrated phenanthrene. Areas of greater phenanthrene in the paraffin wax were not at the surface of the wax, but within the wax matrix, suggesting the chemical had diffused into the wax, before accumulating within certain areas. It is believed this may be due to differences in the hydrocarbon chain lengths forming the waxes, producing differences in structure, to which the phenanthrene has a greater or lesser affinity. Phenanthrene Concentrations in the Chamber Air and in Leaves. At saturation, it is estimated the concentration of phenanthrene within the chamber would be ∼1.4 µg dm-3 at 25 °C (30). The measured vapor concentration was stable and about a factor of 4 below that (0.31-0.33 µg dm-3) (i.e., 0.31-0.33 mg m-3). Phenanthrene was therefore continually being replenished within the chamber air to replace any compound which had been removed by sampling, uptake by the plants, and degradation. Phenanthrene concentrations in the leaves after 5 days in the chamber ranged from 1100 to 4000 µg/g dry weight for spinach and from 3000 to 9000 µg/g for maize. Both plant species showed similar levels of phenanthrene uptake and variability between separate leaves on the same plant. Ambient phenanthrene concentrations are highly variable in the environment. Values at urban monitoring stations in London and Manchester are typically 20-160 ng m-3, for example, while values near point sources or roadsides could VOL. 40, NO. 3, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

909

FIGURE 2. A 3D reconstruction of the surface of a spinach leaf taken at the center of the leaf on the abaxial leaf surface showing the uneven surface topography. The protruding stomata can clearly be seen. reach ∼1 µg m-3 (31). Plant phenanthrene concentrations are correspondingly very variable in the environment; values of ∼500 ng/g have been reported for sludge-amended pasture grassland but this may easily be exceeded in urban/ contaminated locations, for example (32). The air and leaf concentrations in this study were about 3 orders of magnitude higher than might typically be experienced in urban environments. These elevated concentrations were employed to facilitate compound detection on the leaves by TPEM. Knowledge of the amounts of chemical within single leaves (above) enabled estimates to be made of the masses being observed within a single TPEM image. Estimated compound masses in the fg-pg range could be directly observed within the leaves. Air-to-Leaf Transfer of Phenanthrene and Uptake to the Leaf Cuticle. As noted in the Introduction and Figure 1, air-leaf transfer of phenanthrene can occur via particulate and gaseous deposition. Evidence for both transfer mechanisms was obtained in the study (see below). The two mechanisms resulted in different distributions of phenanthrene across the leaf surface. Individual particles could be clearly seen and resulted in distinctive “localized clusters” of phenanthrene on/in the cuticle (SI Figure 2), while gasphase transfer was responsible for a more diffusive coverage of phenanthrene across the cuticle (SI Figure 3). It is suggested from the observed results that gas-phase transfer resulted in more phenanthrene reaching the leaves than particle deposition. Transfer via Particulate Deposition. Individual particles about 1-11 µm in size were observed on the cuticle surface of both species (see SI Figure 2 as an illustration). These are believed to be fine dusts circulating in the air of the chamber. More particulates reached the adaxial surfaces (upper) than the abaxial surfaces (lower). This is likely to be through greater particulate deposition and retention to the upper leaf surface, with uptake being lower on the underside of the leaf due to reduced deposition. After dust-borne phenanthrene reached the leaf surface, different mechanisms were responsible for transferring phenanthrene into the cuticle. Particulate material contaminated with phenanthrene was observed on the surface of the cuticle and encapsulated up to ∼2 µm within the cuticular waxes (SI Figure 2). In addition, phenanthrene could be observed to diffuse from the particles into the cuticular matrix adjacent to them (SI Figure 2). The size of these diffuse areas increased over time. Transfer via the Vapor Phase. In this section we consider first the transfer of phenanthrene via the vapor phase to the plant surface and then the movement of this phenanthrene 910

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 3, 2006

into the cuticle. Both “steps” resulted in some variability in phenanthrene distributionsacross the leaf surface and subsequently through the cuticular layer. This variability provides important information about the air-leaf transfer process and the within-leaf storage/transfer processes. Distribution of Phenanthrene Across the Leaf Surface. The vapor-phase transfer of phenanthrene to the leaves of both maize and spinach resulted in a more general “diffusive” distribution across the leaf surface than particulate deposition. Nonetheless, important and interesting differences were seen between the species and across the leaves. In maize, phenanthrene distribution across the leaf cuticle was relatively uniform (SI Figure 3a). The maize leaves had a relatively uniform surface topography. However, the distribution across the spinach cuticle was variable and showed some distinct features. First, the cuticle enveloping the leaf edges contained consistently higher levels of phenanthrene than that at the center of the leaf (SI Figure 4). This was observed within 24 h, before the compound had diffused widely through the cuticular layers. Second, micro-surface irregularities away from the leaf margins showed greater levels of phenanthrene than the depressed areas of the leaf (Figure 2). These were associated with raised epidermal cells, or areas in close proximity to protruding stomata and veins. We interpret the variability in phenanthrene distribution across the leaf surface as evidence that microscale differences in boundary layer thickness influenced the rate of supply of phenanthrene to the leaf surface. Once turbulence has delivered phenanthrene from the bulk air to the laminar boundary layer above the leaf surface, its transfer through the boundary layer occurs via diffusion. This is the cause of “air-side resistance” to air-leaf transfer (1, 19, 22). As noted in the Introduction, differences in the boundary layer thickness cause differences in the rate of supply. It is thinner at the leading edge leaf margins, particularly for large, flat leaves such as spinach (22). The higher concentrations of phenanthrene at the margins therefore indicate air-side resistance is influencing the rate of compound supply to spinach in the chamber study. Similarly, micro-surface irregularities across the surface of the spinach leaf probably induce differences in the boundary layer thickness (19, 22), resulting in more rapid transfer to the “raised” regions than the depressed regions (Figure 2). Clearly these differences are operating on the microscale (on the orders of a few tens of µm) across the leaf surface. Phenanthrene Associated with Stomata and their Wax Plugs. Phenanthrene was visualized within the wax stomatal plugs or on the external walls of the stomatal pores (Figure

FIGURE 3. Phenanthrene within the waxy plugs of the spinach stomatal pores. Phenanthrene is shown in blue at either end of the stomatal pore. It can also be observed within the cuticular plugs surrounding the stomata. XZ image taken at the adaxial leaf surface. 3). Spinach has larger stomatal pores than maize and the effect was more pronounced than for maize. However, phenanthrene was not visualized on the internal walls of the stomatal pore or within the sub-stomatal cavity, beyond the stomatal plug. Elevated levels of phenanthrene were associated with the cuticles and underlying epidermal cells in close proximity to stomata, or other protruding regions of the leaf (Figure 2). The detection of phenanthrene within the stomatal plugs and on the external edge of stomatal pores suggests it is reaching, interacting, and entering the stomata via vaporphase deposition. This is further supported by the presence of phenanthrene within the guard cells of maize. However, it was not visualized on the internal stomatal surface or within the sub-stomatal cavity of the leaf, suggesting that detectable levels of phenanthrene were not passing beyond the stomatal plugs, but were being incorporated into this wax matrix. Again, this may be directly related to the boundary layer dynamics around the protruding stomata mentioned above. These observations indicate that transfer to the leaves via stomatal uptake was not as significant as cuticular transfer in this chamber study. Distribution Within the Cuticle. The cuticle was generally ∼1-2 µm thick in both species in this study. However, it extended down to 5-15 µm deep in the cuticular plugs (areas of thicker cuticle found between the inter-connecting cell walls of the epidermis; see Figure 1). After 24 h (the first sample interval) phenanthrene delivered via the gas phase was seen within the upper 1-5 µm of the cuticle of both species (SI Figure 3). However, it did not form a homogeneous layer throughout the cuticle (SI Figure 3), but occurred in distinct clusters, strips, or diffuse layers (SI Figure 3). The distribution of phenanthrene in maize cuticle was more uniform than that in spinach. It is important to note that the nonhomogeneous distribution through the cuticle was a separate phenomenon from the nonhomogeneous supply/ distribution of phenanthrene across the leaf surface discussed above. It is hypothesized that the clusters, strips, and diffuse layers of phenanthrene in the cuticle are caused by the variable nature of the cuticle itself, and the compound having varying affinity for the constituents which make up the cuticle (similar to that noted earlier in the paraffin wax). It is known

that cuticle composition varies in polarity and the specific composition of waxes between species, plants, and even across leaves (19, 33, 34). It has been suggested that lipophilic molecules diffuse along wax and cutin domains, for example (34). These observations could all be made in samples taken after just 24 h. Thereafter, increasing amounts of phenanthrene transferred to the leaves of maize and spinach. The cuticle “filled up” within 24-48 h, with levels remaining steady after this time. Initial accumulation occurred in the epicuticular wax of the cuticle. Areas of higher accumulation within the cuticular matrix were associated with visible structures within the cuticle, and conformed to the distribution of phenanthrene observed within the paraffin wax. Following uptake to the cuticle, phenanthrene moved into the epidermis. Accumulation Within the Cuticle and Transfer to the Epidermis. After 24 h and in addition to the concentrated clusters and strips noted above, phenanthrene was clearly visualized within the cuticular plugs of both species (Figure 4). The level of phenanthrene within the cuticular plugs was pronounced showing the highest accumulation of phenanthrene in the cuticle (Figure 4). Movement from the cuticle to the epidermis was rapid, occurring predominantly at the cuticular plugs, which appeared to play an important role in determining the overall movement of SVOCs into the epidermis. Importantly, the observations made here suggested that the leaf cuticle may not be the ultimate or dominant reservoir for such chemicals within plants; rather, movement through the cuticle appeared to be just the initial step in the movement of phenanthrene into the inner portions of the leaf. Although maize and spinach showed similar phenanthrene uptake into the leaf cuticle, they displayed significant differences in their accumulation, processing, and storage of phenanthrene once within the epidermis. As explained below, spinach accumulated and stored phenanthrene within the cytoplasm and vacuoles of the epidermal cells, with very little chemical visualized within the cell walls, whereas maize accumulated and stored phenanthrene within the cell walls, with very little visualized within the cellular cytoplasm. The epidermis lies below the cuticle and surrounds the photosynthetically active mesophyll on both the adaxial and abaxial leaf surfaces. It can be subdivided into the cell wall, cytoplasm cellular organelles, and cellular vacuole. The uptake of chemicals to each of these compartments will presumably vary depending on their composition, structure, and function, and the mobility of the chemical. For example, lipophilic organic chemical uptake to the lipid fraction of the cell wall may differ from that of the solute-rich cytoplasm or vacuole (Figure 1). Uptake into the Maize Epidermis. The visualization of phenanthrene within the maize cuticular plugs was relatively uniform over the leaf surface; with the cuticular plugs being extensively contaminated (Figure 4a). The epidermal cell walls below the cuticular plugs showed significant accumulation of phenanthrene after 72 h (Figure 5). Movement into the cell cytoplasm was low, being seldom visualized, after 96288 h (Figure 5). In the cellular cytoplasm phenanthrene occurred as well-defined masses of compound closely associated with the cell walls, or as small clusters of chemical enclosed within the cytoplasm. Phenanthrene accumulated and travelled within the maize epidermal cell walls over time, reaching the inner surface of the cell walls between 96 and 144 h, appearing to surround the epidermal cells. Movement into the Spinach Epidermis. In spinach, phenanthrene was visualized within the cuticular plugs after 24 h (Figure 4). From there, it moved into the epidermal cell walls and cytoplasm below the contaminated cuticular plugs within 48 h. It was not stored within the epidermal cell walls, VOL. 40, NO. 3, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

911

FIGURE 4. Phenanthrene within the cuticular plugs of (a) maize and (b) spinach leaves exposed to vapor-phase phenanthrene for 48 h. Phenanthrene is shown as blue, and the cuticle, cell walls, and stomata are shown in green. These are XY images made at the abaxial leaf surface. but diffused through them rapidly, entering the cytoplasm of the epidermal cells, where subsequent accumulation was visualized over time. The cytoplasm showed significant accumulation of phenanthrene between 48 and 96 h from the beginning of the experiment, before concentrating from the cytoplasm into the cellular vacuoles of the spinach epidermal cells after 144 h (Figure 5). The distribution of phenanthrene within the spinach epidermis was not uniform, but related to the distribution of phenanthrene within the overlying cuticle. Epidermal cells at the leaf edge showed consistently elevated concentrations of phenanthrene compared to those within the center of the leaf, reflecting the distribution observed in the cuticle. Adjacent cells did not accumulate the same levels, with raised cells or those in close proximity to protruding stomata often having higher levels than adjacent cells within depressions at the leaf surface. Importantly, the rapid movement of phenanthrene through the epidermal cell walls and subsequent accumulation within 912

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 3, 2006

FIGURE 5. (a) Phenanthrene within the epidermal cell walls of a maize leaf after 144 h exposure to phenanthrene. Image made at 25 µm depth from the adaxial leaf surface. A small amount of phenanthrene can be observed within the cellular cytoplasm. Phenanthrene is shown in blue, and the cell walls, stomata, and chloroplasts are shown in green. Phenanthrene can be seen as small diffuse regions extending from the cell walls into the cellular cytoplasm. (b) Phenanthrene within the cellular cytoplasm and vacuoles of spinach epidermal cells after 144 h exposure to phenanthrene. Phenanthrene is shown as blue, and the cell walls, chloroplasts, and stomata are shown in green. XY image taken at 26 µm depth. XZ image extending 104 µm into the leaf. The intense blue regions shown phenanthrene within the cellular vacuole, and the lighter blue represents the cytoplasm. the cytoplasm and vacuole of the epidermis indicates that this SVOC readily reaches a number of different leaf subcompartments, contrary to the common assumptions based

FIGURE 6. Phenanthrene within the maize mesophyll. Phenanthrene is shown in blue as distinct clusters in the bottom half of the image, and the mesophyll, cell walls, chloroplasts, and stomata are shown in green. XY image taken at a depth of 98 µm from the leaf adaxial surface. on compound physicochemical property modeling (see ref 19). Summary Remarks on Transport Through the Epidermis. Following the uptake and movement of phenanthrene through the leaf cuticles, its behavior within the epidermis suggests two possible modes of transport within the leavess apoplastic and symplastic transport. Apoplastic water transport involves diffusion between and through cell walls, while symplastic transport involves movement into and through the cell cytoplasm of interconnecting cells via the plasmodesmata (24). In spinach, phenanthrene moved rapidly from the cuticle through the epidermal cell wall into the cytoplasm, passing from the apoplast to the symplast, where it accumulated. Once in the cytoplasm, phenanthrene was transported across the tonoplast and concentrated within the cellular vacuoles. The vacuoles generate turgor pressure within the leaf and have a storage function, accumulating the end products of metabolism, starches, proteins, and substances toxic to the plant (24). They are often referred to as the “toxic waste dump” of the leaf. As such, the epidermal vacuoles may represent an important sink for such chemicals within plants. Further work is required to understand the processes involved in the uptake of phenanthrene and other SVOCs into the cellular cytoplasm and vacuole. In spinach the cytoplasm and vacuole acted as significant reservoirs for phenanthrene, appearing to be more important than the lipid-rich cuticle or cell walls (Figure 5). In stark contrast, in maize phenanthrene moved/resided predominantly within the apoplast, remaining within and moving through the cell walls and intercellular spaces of the epidermis. These species differences were important for the next phase of their journey, as they migrated into the mesophyll. Movement into the Mesophyll. The mesophyll lies between the adaxial and abaxial epidermis, and is composed of thin-walled parenchyma cells, containing the photosynthetically active chloroplasts. The mesophyll is split into the regularly shaped (palisade) upper region, containing cells ∼80 µm deep and ∼30 µm round, and an irregularly shaped spongy lower region containing rounded cells of approxi-

FIGURE 7. Phenanthrene within the maize xylem. Phenanthrene is shown in blue binding to the thickened xylem walls shown in green (a) and forming concentrated bands within the xylem (b). XY images taken at a depth of 117 µm from the adaxial leaf surface. mately 30 µm diameter (24). Within the mesophyll there are substomatal cavities and intercellular air spaces, allowing the free exchange of gases essential to metabolic activity (Figure 1). Using this technique it was only possible to visualize into the second layer of mesophyll cells from either the abaxial or adaxial leaf surface. Movement into the Mesophyll in Maize. In maize phenanthrene was observed to reach the interface between the epidermal and mesophyll cells after 96 h, clearly moving from the cell walls of the epidermis to those of the mesophyll. After 144 h phenanthrene was seen well within the cell walls of the mesophyll to a depth of 100 µm on both the abaxial and adaxial surfaces. After 168 h phenanthrene had reached the second layer of the mesophyll at a depth of 110 µm, close to the vascular system of the leaf (Figure 6). Movement into the Spinach Mesophyll. Phenanthrene was visualized to reach the interface between the epidermal and mesophyll cells after 144 h, when it was located within the mesophyll cell walls. Uptake to the mesophyll occurred beneath contaminated regions of the epidermis, suggesting VOL. 40, NO. 3, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

913

FIGURE 8. Schematic diagrams showing the locations where phenanthrene was visualized within leaves of spinach (left) and maize (right) after 12 days exposure to an atmosphere contaminated with phenanthrene. Greater detail of the apoplastic and symplastic flow is shown in (b). that transfer to the mesophyll primarily followed transport through the epidermal cells, not through stomata. After 12 days, phenanthrene was located within the cytoplasm and at the surface of the chloroplasts of the mesophyll. However, it occurred at significantly lower concentrations in the mesophyll than in the adjacent epidermis, only a few µm above. This suggests that the mesophyll may not be as readily accessible to these chemicals as the epidermis, possibly due to a lower partition coefficient/affinity for phenanthrene than the epidermis, or that the plants had not been exposed for long enough for phenanthrene to reach high enough concentrations here, and/or that degradation was occurring. The mesophyll of spinach contained less phenanthrene than that of maize. Transfers to the Vascular System. The leaf vascular system is found within the mesophyll and has two main compartmentssthe xylem and phloem (24). The xylem transports solutes from the roots to shoots, while the phloem transports solutes between sites of assimilation to sites of use. The xylem is characterized by wall thickening, in the form of widely spaced stretchable rings, or loosely coiled helical structures, termed secondary walls. Phenanthrene was not detected within the vascular system of spinach. However, it was visualized within the xylem of maize after 288 h (at a depth of 115-135 µm), where it formed localized focused bands extending up to 95 µm in length (Figure 7) and increasing in number toward the leaf tip. It appeared to be predominantly bound to the walls of the xylem structure. These focused bands looked similar to those observed previously in the root cortex cells of maize grown within a contaminated environment (28). Phenanthrene was 914

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 3, 2006

visualized within both the primary and secondary xylem structures. Phenanthrene was not visualized within the phloem. Plant Compartmentalization. During the 12-day uptake period, phenanthrene was observed within 9 distinct compartments as represented in Figure 8a. These were (1) sorbed to particulate matter at the leaf surface, (2) within the cuticle, (3) within the stomatal plug, (4) within the cuticular plug, (5) within the maize cell walls, (6) within the spinach cytoplasm and (7) vacuole, (8) within the mesophyll, and (9) within the maize xylem. This has important implications for understanding the fate of these chemicals within plants, and for determining the role that vegetation plays in the global cycling of these chemicals. It is often assumed that persistent and lipophilic compounds just partition into plant lipids, notably the cuticle. However, the uptake storage and accumulation of phenanthrene from the atmosphere to leaves of maize and spinach was not restricted to the cuticle. In fact, the results even suggested that the epidermis and mesophyll of these species displayed a greater capacity for chemical storage and accumulation than the cuticle. Two-compartment models generally consider a surface compartment which interacts directly with the atmosphere, and a larger internal storage reservoir (13, 25, 34-36). The data obtained here generally support this notion, with the cuticle (surface compartment) filling to reach a steady state within 24-48 h in both species. Subsequent, longer-term uptake supplied the epidermis and mesophyll of maize and spinach. Presumably, the “internal compartment” may incorporate the apoplast of the leaf, comprising the cell walls

and intracellular spaces, or the symplast of the leaf, comprising the internal cellular cytoplasm and vacuoles of the leaf. However, this can only be shown through future uptake and depuration experiments. This study may suggest that the internal compartments of the leaf combine the epidermis, mesophyll, and vascular system of the leaf and can be further divided into the apoplast and symplast. The apoplast includes the cell walls and intercellular spaces of the leaf and does not represent a significant volume of the leaf, whereas the symplast incorporates the cellular cytoplasm and vacuoles of the leafsa large leaf volume. As Figure 8b shows, the two species studied showed very different phenanthrene distribution between these compartments. In maize, it was observed to move through and accumulate within the apoplast, while in spinach it predominantly entered the symplast. These observations may indicate different “internal reservoirs”, depending on species, which may have important implications for routes and rates of movement, storage location and capacity, and overall fate within the leaf. Phenanthrene travelled more rapidly and further into the maize leaf through the apoplast than in the spinach leaf through the symplast. If the same amount of compound is taken into leaves with an apoplastic or symplastic internal reservoir, it will presumably move a greater distance into the leaf with the apoplastic reservoir, because the apoplastic reservoir requires only a small amount of chemical to “fill” a large volume of the leaf, while the symplastic reservoir will require a relatively large amount of chemical to “fill” a small volume of the leaf. This may help explain why phenanthrene was observed to move more rapidly into the maize mesophyll than spinach, and reached the vascular tissues in maize and not spinach. These observations highlight the importance of species differences in the uptake and storage of organic chemicals. Future experiments in which an input of chemical is introduced to plants which are subsequently moved to clean air conditions will enable the reversibility and dissipation of compound to be monitored, helping to elucidate the role different compound compartments or locations have in resupplying the air, and help track movement/ translocation through the leaf. This technique is currently limited by the depth of penetration into the leaf (∼200µm).

(7)

(8) (9) (10)

(11) (12) (13) (14) (15) (16)

(17)

(18) (19)

(20) (21)

Acknowledgments We are grateful to the UK Natural Environment Research Council (NERC) for provision of a Ph.D studentship (NER/ S/A2002/10394) to E.W.

Supporting Information Available

(22)

(23)

Supplemental Figures 1-4. This material is available free of charge via the Internet at http://pubs.acs.org. (24)

Literature Cited (1) Bohme, F.; Welsch-Pausch, K.; McLachlan, M. S. Uptake of airborne semivolatile organic compounds in agricultural plants: Field measurements of interspecies variability. Environ. Sci. Technol. 1999, 33, 11, 1805-1813. (2) Kirkwood, R. C. Recent development in our understanding of the plant cuticle as a barrier to the foliar uptake of pesticides. Pestic. Sci. 1999, 55, 69-77. (3) Simonich, S. L.; Hites, R. A. Importance of vegetation in removing polycyclic aromatic hydrocarbons from the atmosphere. Nature 1994, 370, 49-51. (4) Horstmann, M.; McLachlan, M. S. Atmospheric deposition of semivolatile organic compounds to two forest canopies. Atmos. Environ. 1998, 32, 10, 1799-1809. (5) Simonich, S. L.; Hites, R. Organic pollutant accumulation in vegetation. Environ. Sci. Technol. 1995, 29, 12, 2905-2914. (6) Niu, J.; Chen, J.; Martens, D.; Quan, X.; Yang, F.; Kettrup, A.; Schramm, K.-W. Photolysis of polycyclic aromatic hydrocarbons

(25) (26) (27)

(28) (29) (30)

adsorbed on spruce [Picea abies (L.) Karst.] needles under sunlight. Environ. Pollut. 2003, 123, 39-45. Niu, J.; Chen, J.; Martens, D.; Henkelmann, B.; Quan, X.; Yang, F.; Seidlitz, H. K.; Schramm, K.-W. The role of UV-B on the degradation of PCDD/Fs and PAHs sorbed on surfaces of spruce (Picea abies (L.) Karst.) needles. Sci. Total Environ. 2004, 322, 231-241. Welsch-Pausch, K.; McLachlan, M. S. Photodegradation of PCDD/Fs on pasture grass. Organohalogen Compd. 1995, 24, 509-512. Wilken, A.; Bock, C.; Bokern, M.; Harms, H. Metabolism of different PCB congeners in plant cell cultures. Environ. Toxicol. Chem. 1995, 14, 2017-2022. Mackova, M.; Macek, T.; Ocenaskova, J.; Burkhard, J.; Dermnerova, K.; Pazlarova, J. Biodegradation of polychlorinated biphenyls by plant cells. Int. Biodeterioration Biodeg. 1997, 39, 317-325. Thomas, G. O.; Sweetman, A. J.; Lohmann, R.; Jones, K. C. Derivation and field testing of air-milk and feed-milk transfer factors for PCBs. Environ. Sci. Technol. 1998, 32, 3522-3528. Thomas, G. O.; Sweetman, A. J.; Jones, K. C. Input-output balance of PCBs in a long-term study of lactating dairy cattle. Environ. Sci. Technol 1999, 33, 104-112. Komp, P.; McLachlan, M. S. The kinetics and reversibility of the partitioning of polychlorinated biphenyls between air and ryegrass. Sci. Total. Environ. 2000, 250, 63-71. Simonich, S. L.; Hites, R. A. Vegetation-atmosphere partitioning of polycyclic aromatic hydrocarbons. Environ. Sci. Technol. 1994, 28, 939-943. Bakker, M. I.; Vorenhout, M.; Sum, D.; Kollo¨ffel, C. Dry deposition of atmospheric polycyclic aromatic hydrocarbons in three Plantago species. Environ. Toxicol. Chem. 1999, 18, 2289-2294. Jeffree, C. E. Structure and ontogeny of plant cuticles. In Plant Cuticle: An Integrated Functional Approach; Kersteins, G., Ed.; BIOS Scientific Publishers Limited: Oxford, UK, 1996; pp 3382. Price, C. E. A review of the factors influencing the penetration of pesticides through plant leaves. In The Plant Cuticle; Cutler, D. F., Alvin, K. L., Price, C. E., Eds.; Academic Press: New York, 1982; pp 237-2523. Bakker, M. I.; Koerselman, J. W.; Tolls, J.; Kolloffel, C: Localisation of deposited polycyclic aromatic hydrocarbons in leaves of Plantago. Environ. Toxicol. Chem. 2001, 20, 1112-1116. Barber J. L.; Thomas, G. O.; Kerstiens, G.; Jones, K. C. Current issues and uncertainties in the measurement and modelling of air-vegetation exchange and within-plant processing of POPs. Environ. Pollut. 2004, 128, 99-138. Schreiber, L. Polar paths of diffusion across plant cuticles: New evidence for an old hypothesis. Ann. Bot. 2005, 95, 1069-1073. Smith K. E. C.; Jones, K. C. Particulates and vegetation: Implications for the transfer of particulate-bound organic contaminants to vegetation. Sci. Total Environ. 2000, 246, 207236. Grace, J. Effects of wind on plants. In Plants and Their Atmospheric Environment; Grace, J., Ford, E. D., Jarvis., P. G., Eds.; The 21st Symposium of the British Ecological Society, Edinburgh; Blackwell Scientific Publishers: Oxford, 1979; pp 31-56. Schreiber, L.; Kirsch, T.; Riederer, M. Transport properties of cuticular waxes of Fagus sylvatica L. and Picea abies (L.) Karst: Estimation of size selectivity and tortuosity from diffusion coefficients of aliphatic molecules. Planta 1996, 198, 104-109. Dickinson, W. C. Integrative Plant Anatomy; Academic Press: London. 2000. Tolls, J.; McLachlan, M. S. Partitioning of semivolatile organic compounds between air and Lolium multiflorum (Welsh Ray grass). Environ. Sci. Technol. 1994, 28, 159-166. Wild, E.; Dent, J.; Barber, J. L.; Thomas, G. O.; Jones, K. C. A novel analytical approach for visualizing and tracking organic chemicals in plants. Environ. Sci. Technol. 2004, 38, 4195-4199. Wild, E.; Dent, J.; Thomas, G. O.; Jones, K. C. Real time visualization and quantification of PAH photodegradation on and within plant leaves. Environ. Sci. Technol. 2005, 39, 268273. Wild, E.; Dent, J.; Thomas, G. O.; Jones, K. C. Direct observation of organic contaminant uptake, storage, and metabolism within plant roots. Environ. Sci. Technol. 2005, 39, 3695-3702. Nadal, M.; Schumacher, M.; Domingo, J. L. Levels of PAHs in soil and vegetation samples from Tarragona County Spain. Environ. Pollut. 2004, 132, 1-11. Ambient Air Pollution by Polycyclic Aromatic Hydrocarbons (PAH), Position Paper; Prepared by the Working Group on

VOL. 40, NO. 3, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

915

(31)

(32) (33) (34)

916

Polycyclic Aromatic Hydrocarbons; ISBN 92-894-2057-X; European Commission: Luxembourg, 2001. Prevedouros, K.; Bro¨rstrom-Lunden, E.; Halsall, C. J.; Jones, K. C.; Lee, R. G. M.; Sweetman, A. J. Seasonal and long-term trends in atmospheric PAH concentrations: evidence and implications. Environ. Pollut. 2004, 128, 17-27. Smith, K. E. C.; Green, M.; Thomas, G. O.; Jones, K. C. The behaviour of sewage sludge derived PAHs on pasture. Environ. Sci. Technol. 2001, 35, 2141-2150. Hallam, N. D. Fine structure of the leaf cuticle and origin of leaf waxes. In The Plant Cuticle; Cutler, D. F., Alvin, K. L., Price, C. E., Eds.; Academic Press: London. Schreiber, L.; Kirsch, T.; Riederer, M. Diffusion through cuticles: principles and models. In Plant Cuticles: An Integrated

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 3, 2006

Functional Approach; Kersteins, G., Ed.; BIOS Scientific Publishers Limited: Oxford, 1996; p 109. (35) Hauk, H.; Umlauf, G.; McLachlan, M. S. Uptake of gaseous DDE in spruce needles. Environ. Sci. Technol. 1994, 28, 2372-2379. (36) Barber, J. L.; Thomas, G. O.; Kerstiens, G.; Jones, K. C. Study of plant-air transfer of PCBs from an evergreen shrub: implications for mechanisms and modelling. Environ. Sci. Technol. 2003, 37, 3838-3844.

Received for review July 31, 2005. Revised manuscript received November 1, 2005. Accepted November 10, 2005. ES0515046