Original Articles
Different uptake pathways between hydrophilic and hydrophobic compounds in lateral roots of Cucurbita pepo
2015 Volume 40 Issue 3 Pages 99-105
Details
2015 Volume 40 Issue 3 Pages 99-105
From previous reports, uptake and accumulation of organic compounds by plant roots are strongly related to water solubility. However, the relation between uptake pathways and water solubility remains unclear. Here, we used confocal laser scanning microscopy to observe the uptake of fluorescent hydrophilic and hydrophobic compounds in the living roots of Cucurbita pepo. We found strong regulation of the uptake of berberine, a hydrophobic compound at the endodermal Casparian strip, as previously reported, and similar regulation was also observed in the pericycle. Berberine was loaded into and transported upward through the protoxylem. Perylene, a highly hydrophobic compound, in contrast, passed through the Casparian strip and accumulated preferentially in the endodermis and pericycle. The results of our solvent extraction suggested that perylene diffused into the non-aqueous phase. Therefore, the uptake pathways for the hydrophilic and hydrophobic compoundss are different. These results offer a new way to understand the uptake of agrochemicals and pollutants and to select host plants for phytoremediation.
Plant roots take up water and nutrients. Water and solutes pass through the root epidermis, exodermis, cortex, endodermis, pericycle, and parenchyma before being loaded into vascular bundles (xylem and phloem). The Casparian strip, located in the exodermis and endodermis, works as a diffusion barrier.1) The plasma membrane domains constituting the Casparian strip adhere tightly to the intercellular walls2); a family of transmembrane proteins produced in the endodermis has recently been found to function in this adhesion.3) Although maturation of the exodermal Casparian strip is a later process of root development,1) maturation of the endodermal Casparian strip seems to begin before cell elongation begins.4) Therefore, it is reasonable to think that researching in the root regions after the formation of the endodermal Casparian strip is important for understanding the plant’s uptake of organic compounds. Almost all vascular plants develop an endodermis,5) and the endodermal Casparian strip actually regulates the uptake of various ions, heavy metals, and other compounds.6–8) The uptake of water in roots is considered to follow apoplastic, symplastic, and transcellular pathways.9) Many substances are excluded from the apoplastic passage by the endodermal Casparian strip and take the symplastic route to the stele.1)
On the other hand, previous studies have shown that the accumulation of organic pollutants in roots increases with their increased hydrophobicity, and that moderate hydrophobicity is important for efficient translocation into the aerial parts.10,11) These facts indicate that the water solubility of those organic compounds impacts their root uptake; therefore, it is important to study both pathways of hydrophilic and hydrophobic compounds to understand root functions. However, there have been few reports regarding the dynamics of hydrophobic compounds such as organic pollutants in plant roots, and it has been unclear how plants take up hydrophobic compounds through the roots. Recent studies have shown that among many plant species cucurbits can efficiently accumulate hydrophobic organic pollutants—insecticides, for example—in their shoots,12,13) and that their roots may have an important role in this,14) although the reasons are still unknown. For this reason, cucurbits might be suitable for studying the uptake of hydrophobic compounds.
Here, in order to reveal different uptake pathways for organic compounds with different solubilities, we used confocal laser scanning microscopy (CLSM) to directly observe the uptake of hydrophilic and hydrophobic compounds in living lateral roots of soil-grown Cucurbita pepo plants. Lateral roots have a Casparian strip similar to the main roots and are so thin that we can more easily observe their tissues than those of the main roots. Among the important components in the root structures, we focused on the tissues of the epidermis, cortex, endodermis, pericycle, and xylem, as are often the focus in studies of water uptake.9)
Seeds of C. pepo L. ssp. ovifera cv. ‘Patty Green’ and ‘Zephyr’ and of C. pepo ssp. pepo cv. ‘Gold Rush’ and ‘Magda’ were purchased from Johnny’s Selected Seeds (Albion, ME, USA). Seeds of C. pepo ssp. pepo cv. ‘Black Beauty’ were purchased from Tanenomori (Saitama, Japan). Among subspecies of C. pepo, ssp. pepo show higher accumulations of hydrophobic compounds than do ssp. ovifera.15) The seed coats were removed and soaked in 70% ethanol for 1 min and then in 5% hypochlorous acid for 20 to 30 min. They were then rinsed three times with sterile water. The seed skins were peeled off during the rinses. The sterile seeds were placed on MS agar medium16) that contained 0.8% agar and 3% sucrose and then cultured at 25°C under a 16-hr light/8-hr dark cycle. Seedlings 1 to 2 weeks old were transferred to autoclaved commercial soil (Green Plan Co., Ltd., Miyagi, Japan).
2. Detection of berberine and perylene uptakeSoil attached to roots was carefully removed by hand in tap water. An intact thin lateral root was placed onto a cover glass, and excess water was blotted with paper. Either 80 µL of 0.05% (w/v) berberine hemisulfate (CAS: 633-66-9) or 80 µL of 0.005% (w/v) perylene (CAS: 198-55-0) (0.1% stock solution in dimethyl sulfoxide [DMSO], diluted 1 : 20 with water) solution was dropped on the cover glass without touching the root, and then another cover glass was placed over the first, creating contact (Fig. 1a). The molecular weight and log Kow of berberine hemisulfate are 384.39 and −0.99 (predicted value in the ChemSpider database),17) respectively, and this chemical showed high water solubility. Berberine is a hydrophilic fluorescent compound that stains lignin and suberin and is used as a histological stain and apoplastic tracer.18,19) The molecular weight and log Kow of perylene are 252.3 and 6.25,20) respectively, and this shows low water solubility. Fluorescence in the area around 1 cm from the root tip was observed by CLSM (FV300, Olympus Co., Tokyo, Japan) with argon laser excitation at 488 nm and a detection range of 510–630 nm. Parameters for detection sensitivity were set not to detect the autofluorescence of roots visually.
In plasmolysis tests, the roots were soaked in a 3% (w/v) NaCl solution for 5 min. When the solution was changed, the roots were rinsed in water for 5 min. All solutions used for testing perylene contained 5% DMSO. Only in an examination to evaluate the effect of DMSO on chemical uptake, was a berberine solution containing 5% DMSO used (Supplemental Fig. S3b).
Translocation was observed with gap experiments as shown in Fig. 1b. An area about approximately 1.5 cm from the root tip was placed between two cover glasses filled with water or 5% DMSO, and 0.05% (w/v) berberine hemisulfate or 0.005% (w/v) perylene was applied to the root tip on a glass slide separated by a gap of 1 to 2 mm to prevent movement of the solution to the observation area. Fluorescence was observed by CLSM.
Lateral roots of C. pepo plants grown in soil are better suited to the detection of fluorescent compounds by CLSM because they are thinner and more transparent than those grown in hydroponic solution or agar medium. Roots grown in such media probably developed second cell wall deposition more than in soil and showed high background levels of autofluorescence. Bright-field microscopy clearly revealed internal structures at a depth of 50 µm in soil-grown roots. The lateral roots consisted of root hairs, epidermis, cortex, endodermis, and stele, including pericycle and xylem (Fig. 2a). A lateral root was immersed in a 0.05% berberine hemisulfate solution and then continuously observed. Berberine uptake from the surface to the central stele via the cell wall—the apoplastic pathway—was observed in roots of all C. pepo cultivars (Fig. 2a and Supplemental Fig. S1). There were no clear differences in fluorescence patterns among the cultivars. Strong fluorescence around the epidermis was observed within 1 min after berberine application. After approximately 5 min, fluorescence was detected at the periphery of the endodermal apoplast and reached the protoxylem. Fluorescence in the protoxylem was preferentially localized in the cell wall, making the same pattern as the spiral thickening. The fluorescence in the protoxylem continued to increase until 10 min, but little was observed in the cell walls of the endodermis and pericycle. Blockage at the middle of the cell wall in the endodermis was sometimes observed (Supplemental Fig. S1, arrowheads). Although xylem vessels, that are the apoplast, were well stained, the inner endodermis and the pericyclic apoplast were barely stained by 10 min (Fig. 2a and Supplemental Fig. S1). We performed gap experiments to observe the translocation (Fig. 1b). Twenty min after application of the solution, fluorescence was observed in the protoxylem vessels away from the root tip in the absence of direct contact with berberine solution (Fig. 3a). No leakage of berberine from the protoxylem into any peripheral parts was observed, even though some berberine clearly stained the protoxylem.
To examine the influence of the osmotic pressure on berberine diffusion, we transiently plasmolyzed the lateral root cells with a 3% NaCl as a hypertonic solution. We then restored the turgor with water and applied berberine solution. In this procedure (5-min plasmolysis, 5-min rinse with water, and then 10-min uptake of berberine), berberine fluorescence was observed in the cell walls of the endodermis and pericycle as well as in the epidermis, cortex, and protoxylem vessels within 5 min after application (Fig. 4a). The results also showed that plasmolyzed roots took up berberine faster than did non-plasmolyzed roots. To further explore the involvement of osmotic pressure in uptake, we soaked the roots in hypertonic solutions after the uptake of berberine in the reverse procedure (10-min uptake of berberine, 5-min rinse with water, and then 5-min plasmolysis). This procedure caused rapid, strong, localized berberine fluorescence in the protoxylem vessels (Fig. 4b), as compared to the uptake behavior without plasmolysis.
Perylene is a polycyclic aromatic hydrocarbon with high hydrophobicity. Weak perylene fluorescence was observed around the cell surface of the epidermis in the lateral roots of C. pepo within a few minutes after application, and the fluorescence in roots increased with time (Fig. 5a and Supplemental Fig. S2). After 46 min, strong fluorescence was localized in the endodermis and pericycle despite the still weaker fluorescence in the epidermis and cortex. In the protoxylem little fluorescence of perylene was observe within 46 min, indicating that most perylene molecules did not move to protoxylem in 46 min. This was the opposite pattern of that of berberine. The control 5% DMSO solution caused no fluorescence (Supplemental Fig. S3a). To evaluate the accumulation and localization of perylene in the tissues, we measured the fluorescence intensity in each tissue over time, correcting for the background levels by subtracting the intensities at 1 min after the root’s contact with the solution. The results clearly showed that the intensities in all tissues of all cultivars increased over time with a similar accumulation pattern (Supplemental Fig. S4). To evaluate the accumulation capacity of each tissue, we then calculated the relative intensities in each tissue to the average intensity in all tissues as follows: the corrected values in each tissue were divided by the average fluorescence in four tissues from the epidermis to the pericycle (Fig. 6). All cultivars showed a similar pattern after 31 min, in which levels in the endodermis and pericycle tended to be higher than those in the epidermis and cortex. The levels in the endodermis or pericycle of C. pepo cv. ‘Black Beauty,’ ‘Gold Rush,’ and ‘Magda’ were already higher at 6 min than in the immediately centrifugal cortex. The intensities in all tissues continued to increase until at least 46 min, and the pattern of accumulation was stabilized by 31 min. In gap experiments to observe the translocation of perylene, little fluorescence was observed in root tissues until 100 min after application (Fig. 3b).
To confirm whether the 5% DMSO, which was used to increase the water solubility of perylene, affects the tissues and accumulation patterns of fluorescence, a berberine solution containing 5% DMSO was applied, and the same results were obtained as those without DMSO (Supplemental Fig. S3b). Furthermore, we attempted to extract perylene from the roots by flushing with 5% DMSO and saw little change in fluorescence by 30 min (Fig. 7a). To examine the accumulation in the apoplast, we extracted the roots by plasmolysis with a 3% NaCl solution containing 5% DMSO after perylene uptake. The extraction did not affect perylene accumulation within 30 min (Fig. 7b). These results were not caused by toxicity or stress during incubation, as there was little to no root tissue autofluorescence in a long incubation with 5% DMSO and 3% NaCl (Supplemental Fig. S3a).
All results with two compounds were obtained with a plant condition as illustrated in Fig. 1, and similar results were observed even when the root tip was removed from the solution and the hypocotyl was cut off.
As a hydrophilic compound, the clear uptake of berberine via the apoplastic pathway was obtained. Spiral thickening patterns in the xylem indicated secondary cell wall deposition in the protoxylem.21,22) There is preferential accumulation of berberine fluorescence in patterns because the secondary cell wall of the spiral thickening zone consisted of additional amounts of lignin22) and probably absorbed the berberine. Although a diffusion block in the endodermis was observed, the berberine moved to the protoxylem (Fig. 2a and Supplemental Fig. S1). It is unlikely that the berberine could pass through the Casparian strip to the protoxylem without accumulation, since berberine accumulated there under a plasmolyzed condition (Fig. 4a). Therefore, berberine could little diffuse in the apoplast of the pericycle and the inner endodermis under normal conditions. Thus, in roots, only berberine molecules that had entered the cytoplasm could move symplastically through all tissues to the pericycle and its inner tissues. After the berberine diffused into apoplastic vessels, fluorescence was not observed in the apoplast of the endodermis or pericycle. This observation indicates that berberine did not flow from the endodermal and pericycle symplast nor back from the protoxylem vessel to those apoplast areas (Fig. 2b). This suggests that the important functions of not only the Casparian strip but also the pericycle and the inner endodermal apoplast for regulating the uptake of solutes by the roots. Observation of the berberine fluorescence in only the protoxylem in the gap experiments (Fig. 2) showed that hydrophilic berberine diffused longitudinally through the xylem but not through the epidermal or cortical apoplast.
2. Effect of plasmolysis on uptake regulationPlasmolyzed roots showed an efficient and different uptake behavior of berberine in the endodermal and pericycle apoplast (Fig. 4a); in this condition, the Casparian strip did not normally work as a diffusion barrier. This result seems to be inconsistent with the following: Adhesion of the cell wall to the plasma membrane in the endodermal Casparian strip is not broken by plasmolysis.4) In addition, symplastic diffusion via plasmodesmata could decrease because of disconnection by plasmolysis.23) Thus, our plasmolyzed root results suggest that plasmolysis disrupts the function of the endodermal Casparian strip to regulate diffusion (but does not break the structure).
It is well known that increased osmotic pressure decreases cell size and turgor pressure,24) and in our experiments the reverse plasmolyzing procedure impacted berberine accumulation in the cell wall and caused its quick movement into the protoxylem (Fig. 4b). Under plasmolysis there would be little turgor pressure, and we think that reduction of turgor pressure is an important factor for reducing resistance to the diffusion of berberine in the cell wall as observed in Fig. 4b. Taken together, plasmolysis decreased the function of the Casparian strip and resistance to the diffusion of berberine. As a result, berberine moved easily or quickly to the secondary cell wall in the protoxylem under the two different plasmolyzing conditions. The results of transient plasmolysis may show that turgor pressure is an important factor in regulating the diffusion capacity of hydrophilic solutes in cell walls, even in the Casparian strip.
3. Uptake of hydrophobic perylenePerylene is a polycyclic aromatic hydrocarbon, a fluorescent chemical. This class of compounds is known to have higher accumulations, especially in C. pepo ssp. pepo, than in other cucurbitaceous plants25) with similar hydrophobic compounds. We do not know the uptake of such hydrophobic compounds in living cucurbitaceous roots at the microscopic level. Although the 5% DMSO used for the solvent may be toxic to root cells somehow, it did not show a clear effect on the uptake of berberine or visible toxicity that increased autofluorescence (Supplemental Fig. S3). The perylene accumulation in all tissues continued to increase until at least 46 min (Supplemental Fig. S4), and the pattern of accumulation was almost stabilized by 31 min (Fig. 6). Therefore, the localized distribution of perylene was formed by the continuous free diffusion among root tissues rather than by preferential, irreversible binding to the endodermis and the pericycle. If such binding had occurred, the bias in the accumulation pattern observed in Fig. 5 would have continued to increase until the intensities in the endodermis or the pericycle were saturated. Consequently, the accumulation pattern showed the relative affinities of each tissue for perylene. The opposite manner to that of hydrophilic berberine suggests that the mechanisms preventing the diffusion of berberine to the endodermis and the pericycle are related, directly or indirectly, to the high affinities of perylene. A previous study that involved the live imaging of maize and wheat roots for uptake of anthracene and phenanthrene, kinds of polycyclic aromatic hydrocarbons, demonstrated that such hydrophobic compounds localized in the cortex of the exposed living roots.26) This study is partially consistent with our results for perylene uptake in the respect of its passage through the epidermis, whereas the previous researchers could not detect the radial movement of their compounds to inner tissues beyond the cortex. In addition to differences in the experimental conditions, the thickness of their root samples might make it difficult to detect fluorescence in such deep areas, thus causing the different findings between our study and theirs.
In addition, perylene accumulation behavior that was opposite to the gradient of the perylene concentration supplied needed the hydrophobic perylene to pass efficiently through the barrier of the Casparian strip. This, and little extraction with the solvent, suggests that perylene accumulates and diffuses in a non-aqueous phase, like a lipid bilayer. Because the membrane permeability of non-ionic compounds increases with their hydrophobicity,27) hydrophobic compounds, rather than hydrophilic ones, can diffuse preferentially in the membrane. Intercellular diffusion could be mediated through the membrane via the plasmodesmata, and the presence of many plasmodesmata from the epidermis until the pericycle28) might limit the diffusion of perylene until the pericycle, as observed in our results. Thus, we hypothesized that perylene would diffuse mainly in the membrane and not in the apoplastic pathway (Fig. 5b). It was reasonable to observe the poor translocation of perylene in the gap experiment within 100 min (Fig. 3b) on account of the poor release to the aqueous phase (Fig. 5b, A blunthead line between pericycle and protoxylem). This does not conflict with the results of a previous report showing much less accumulation of highly hydrophobic dioxins in the shoots of hydroponically grown C. pepo as compared with those in the roots.29)
All results shown were obtained with conditions as illustrated in Fig. 1, in low light and at room temperature, and the transpiration stream and uptake from root tips with immature tissue could somehow affect the results of the uptake behaviors of the two compounds in the observed areas. However, the influences on uptake in our conditions would be much smaller because we found no clear differences in both fluorescent patterns when the root tip was removed from the solution and the hypocotyl was cut off. We did not clarify whether the berberine and perylene accumulation patterns in the roots of Cucurbita plants were common to the uptake of other hydrophilic and hydrophobic compounds in plant roots. Nevertheless, the opposite manners of the uptake of hydrophilic berberine and hydrophobic perylene suggest that, to some extent, our finding reflects uptake mechanisms based on solubility. Thus, it may be possible to predict the uptake behaviors of compounds from their solubilities. Prediction of uptake will allow for improvement of agricultural technologies and phytoremediation by controlling the uptake of nutrients, agrochemicals, and pollutants.
KY and HI conceived and designed the experiments; KY performed the experiments; KY analyzed the data; and KY, HT, and HI wrote the paper.
This work was supported by the Ministry of Agriculture, Forestry, and Fisheries of Japan (Genomics for Agricultural Innovation) [GMB-0006 to H. I.]; and a Grant-in-Aid for Scientific Research A from the Ministry of Education, Culture, Sports, Science, and Technology of Japan [No. 23241028].
