Environ. Sci. Technol. 2004, 38, 2443-2448
Enhanced Accumulation of Phosphate by Lolium multiflorum Cultivars Grown in Phosphate-Enriched Medium NILESH C. SHARMA,† S H I V E N D R A V . S A H I , * ,† JINESH C. JAIN,‡ AND KASCHANDRA G. RAGHOTHAMA§ Biotechnology Center, Department of Biology, Western Kentucky University, Bowling Green, Kentucky 42101, Department of Civil Engineering and Geological Sciences, University of Notre Dame, Notre Dame, Indiana 46556, and Department of Horticulture, Purdue University, West Lafayette, Indiana 47907
Agricultural and animal husbandry practices combined with soil composition have caused phosphate overloading of farmlands in different parts of the U.S. and Europe. Movement of soluble phosphates (Pi) from phosphorus enriched soils results in degradation of natural aquatic systems, triggering serious environmental problems. Remediation of such sites using plants that tolerate and accumulate high concentrations of Pi in their aerial parts may be an attractive remediation technology. In the present study, Pi transport and accumulation potential of Marshall and Gulf ryegrass (Lolium multiflorum cultivars) were determined using a solution culture of seedlings. Ryegrass seedlings accumulated phosphorus (P) in excess of 2% of dry weight in their aerial parts when supplied with 5 g/L KH2PO4 in medium. Phosphorus accumulation was positively correlated with the concentration of phosphate (0-5 g/L KH2PO4) in medium. Plants grew well on medium containing 5 g/L KH2PO4, but concentrations above 5 g/L caused symptoms of toxicity. Scanning electron microscopy and energy-dispersive X-ray spectroscopy confirmed high P accumulation in different cell types of grass roots and shoots. Phosphate starvation and replenishment experiments point to the unique ability of these grasses to concentrate phosphate in the above-ground parts. It is hypothesized that the unique ability of these ryegrass cultivars may be due to the presence of efficient phosphate transport and sequestration mechanisms.
Introduction Enormous amounts of phosphate-rich manure (chicken, swine litter, and other animal wastes) are spread over the soil in regions of the world where intensive animal-based agriculture is in vogue. As a consequence, the soluble phosphate levels often exceed the crop requirement causing its movement to rivers, lakes, and other water bodies by * Corresponding author phone: (270)745-6012; fax: (270)745-6856; e-mail:
[email protected]. † Western Kentucky University. ‡ University of Notre Dame. § Purdue University. 10.1021/es030466s CCC: $27.50 Published on Web 03/17/2004
2004 American Chemical Society
methods such as run off and leaching (1, 2). Increased Piload has been implicated as one of the major factors responsible for accelerated eutrophication of surface waters at many locations in the U.S. and elsewhere (3, 4). While problems with eutrophication are long-standing, reports of Pfiesteria piscicida emergence in waterways in Mid-Atlantic U.S. pose a potential human health concern (5). This environmental issue may very well dictate the future expansion of animal-based agricultural practices and necessitates finding ways to reduce nonpoint-source Pi pollution. Methods to reduce the off-site movement of P from fields receiving manure have been investigated (6, 7). Amendment of soils with water treatment residuals (8), alum, or aluminum chloride (9) resulted in significant reductions in nonpointsource Pi runoff. However, these methods aim to reduce levels of Pi by fixing them in insoluble forms, which can be only a temporary solution to the problem. Alternatively, plantassisted extraction (phytoextraction) of phosphate could be an attractive strategy. If a crop, while growing, accumulates high amounts of Pi in their harvestable part, successive cropping can reduce excess Pi levels. Phytoremediation is an inexpensive, nonintrusive, socially desirable and often highly effective technique (10). Plant-based cleanup strategies offer a number of advantages over traditional cleanup methods, as well as over other bioremediation technologies. For a crop to be feasible in phytoextraction of phosphates, it should be capable of hyperaccumulating in its aerial parts (>1% DW) besides having high biomass and economic value (11). Plants, generally referred to as metal hyperaccumulators, have the inherent potential to survive and accumulate excessive amounts of metal ions (>0.1 % DW) in their biomass without incurring damage to basic metabolic functions (12). There are several reports of metal hyperaccumulators that are immensely useful in phytoremediation (13-16). However, the ability of vegetation to assist in the remediation of P remains largely unknown. Annual ryegrass (Lolium multiflorum) is a closely related and interfertile species with perennial ryegrass (Lolium perenne), and both are grown all over the world as key forage grasses (17). These are among the most palatable and highly digestible grasses for livestock (17). In a preliminary study, several grasses and vegetable crops were screened for tolerance to high levels of P. Annual ryegrasses, among them, showed promise; thus, they were chosen as ideal candidates in the present experiment. A hydroponic method was preferred in this study to eliminate the effects of variable soil pH and microflora on P solubility and mobility. Thus, the objectives of this investigation were to (1) determine Pi transport and accumulation potentials in Marshall and Gulf ryegrass, two cultivars of L. multiflorum, grown in nutrient medium enriched with high levels of Pi, (2) study the effect of pH on P accumulation pattern, and (3) assess the impact of P deprivation and replenishment phenomena on P accumulation in these grasses.
Experimental Procedures Seed Germination. Seeds of Marshall grass and Gulf ryegrass (L. multiflorum cultivars), provided by USDA Lab, Starkville, MS, were sterilized with sodium hypochlorite (1% v/v) and rinsed several times with sterile deionized water. They were then transferred to water-agar (0.8%) medium in Magenta boxes and maintained at 25 ( 2 °C under 12/12 light/dark regime in a growth chamber. Ten day-old seedlings were used for various treatments. Phosphate Treatment and Growth of Seedlings. Modified Hoagland’s salts mixture (115 mg/L ammonium nitrate, 2.86 VOL. 38, NO. 8, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2443
mg/L boric acid, 656.4 mg/L calcium nitrate, 3.0 mg/L ferric chloride, 240.7 mg/L magnesium sulfate, 1.81 mg/L manganese chloride, 0.016 mg/L molybdenum trioxide, 400.6 mg/L potassium nitrate, and 0.22 mg/L zinc sulfate) was used as a basal nutrient medium for the growth of seedlings. The basal medium was supplemented with KH2PO4 at 0.57-5 g/L for Pi treatments. Phosphate deficient control plants were grown in nutrient medium supplemented with appropriate amounts of K2SO4 to balance the supply of potassium. Five seedlings were aseptically transferred to culture vessels (250 mL) containing 25 mL of KH2PO4-enriched nutrient solution (described previously). Media were changed every 3 days, and an aerator provided aeration. Plants were grown in growth chamber at 25 ( 2 °C in 16/8 light/dark (180-200 µmol m-2 s-1 of cool fluorescent light) for different periods (2-4 weeks). Each treatment was replicated three times, and experiments were repeated (n ) 6). Analysis of P in Plant Tissue. Following 2-4 weeks of growth, plants from different treatments were harvested, washed thoroughly with deionized water, divided into root and shoot biomass, and air-dried until a steady weight was achieved (4-6 weeks). The ground samples were then weighed and placed in 15 mL Teflon beakers. Three mL of concentrated HNO3 was added to the sample, and the beaker was placed on a hotplate set at 100 °C overnight, until evaporated to dryness. The samples were allowed to cool and made up gravimetrically to a volume of 20 mL with 2% HNO3 (18). A VG Elemental Plasma Quad (model PQZ) ICP-MS was used for all data acquisition. Analyses were performed using an external calibration procedure, and internal standards were included to correct for matrix effects and instrumental drift corrections (18). Statistical Analysis. Statistical analysis of the data was carried out using SYSTAT (Version 9 for Windows, 1999, Systat Software Inc., Richmond, CA). Two-way ANOVA analysis of data was performed to verify significant differences in mean shoot P values at different treatments between two grasses and among different treatments for each grass. Data were normalized using log10 transformation. Effect of Solution pH on Pi Uptake. To study the effect of pH on Pi transport and accumulation by Lolium roots and shoots, pH of the nutrient solution (containing 5 g/L KH2PO4) was adjusted to 3.7-7.7 with either 0.1 N HCl or 0.1 N NaOH. For this study, nutrient solution was modified with chelated micronutrients [1.81 mg/L Mn-N-hydroxyethylethylene diaminetriacetic acid (HEDTA), 3.0 mg/ L FeHEDTA, and 0.22 mg/L ZnHEDTA] and changed everyday. Plants were grown under the conditions described previously for 2 weeks. Scanning Electron Microscopy. Plants grown on 5 g/L KH2PO4 for 2 weeks (selected from each replicate) were used as samples for scanning electron microscopic examinations. Samples were washed thoroughly with deionized water, quick-frozen by plunging into liquid N2, and fractured into small pieces. The fractured pieces were placed in Styrofoam cups, each containing liquid N2 and a 500 g steel block. Root and leaf pieces were freeze-dried by warming to room temperature in a bell jar evacuated with a rotary pump to 2.6 Pa. Dried samples (root and leaf) were mounted on stubs. Then samples were viewed uncoated in a JEOL (Tokyo) 5400 LV SEM at 15 kV low vacuum modes using a backscattered electron detector. Elemental analysis was carried out using an attached KEVEX Sigma Energy Dispersive X-ray Spectrometer (EDS). For comparison, plants grown on 0 and 0.57 g/L KH2PO4 were also examined under the same conditions. Effect of Phosphate Starvation and Replenishment on P Accumulation. To study the effect of transient phosphate starvation on Pi uptake and accumulation, control plants were grown in nutrient medium (in absence of KH2PO4) for 2 weeks. In the second treatment, plants were grown on 0 2444
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 8, 2004
FIGURE 1. Accumulation of P in Marshall and Gulf ryegrass grown in nutrient solution containing 0.57-5 g/L KH2PO4 for 2 weeks. Corresponding controls were set up with equal amounts of K2SO4. Each point represents mean of six replicates ((SE). [MS ) Marshall grass shoot, MSC ) Marshall grass shoot control, GS ) Gulf grass shoot, GSC ) Gulf grass shoot control, MR ) Marshall grass root, MRC ) Marshall grass root control, GR ) Gulf grass root, and GRC ) Gulf grass root control].
FIGURE 2. Accumulation of P in Marshall and Gulf ryegrass grown in nutrient solution containing 5 g/L KH2PO4 over a period of 21 days. Corresponding controls were set up with equal amounts of K2SO4. Each point represents mean of six replicates ((SE). [MS ) Marshall grass shoot, MSC ) Marshall grass shoot control, GS ) Gulf grass shoot, GSC ) Gulf grass shoot control, MR ) Marshall grass root, MRC ) Marshall grass root control, GR ) Gulf grass root, and GRC ) Gulf grass root control]. mg/L KH2PO4 for 2 weeks, then transferred to a fresh medium containing 5 g/L KH2PO4 for 12 days. In the third treatment, plants were grown in a medium containing 5 g/L KH2PO4 for 2 weeks, and half of these were harvested, and the rest were transferred to a fresh medium containing 0 g/L KH2PO4 to be harvested after 12 days. In the last treatment, plants grown for 2 weeks in the medium containing 5 g/L KH2PO4 were replenished with the fresh solution containing same concentration of KH2PO4.
Results and Discussion Accumulation of P in Plant Tissue. Both grasses accumulated high concentrations of P (>2% of tissue dry weight) in their roots and shoots in a medium containing 5 g/L KH2PO4 (Figure 1). Root P content declined while shoot P raised rapidly with further increase of Pi in the medium. The P accumulation pattern in Marshall grass root and shoot was opposite of each other in the second week of growth in the treatment supplied with 5 g/L KH2PO4. However, after two weeks, both roots and shoots accumulated more P (Figure 2). Decline in root P concentration in the second week may be a consequence of translocation of Pi from root to shoot. However, root and shoot of Gulf ryegrass demonstrated
FIGURE 3. Effect of pH on P accumulation in Marshall ryegrass grown in nutrient solution containing 5 g/L KH2PO4 for 2 weeks. Each point represents mean of six replicates ((SE). (Small differences in SE do not appear as bars in some cases). increased P content up to week 2, reaching a plateau thereafter (Figure 2). Shoots of Gulf ryegrass accumulated more P than roots at all the concentrations tested (Figure 1).
Each grass showed significant differences in shoot P contents among the various treatment groups (F5,49 ) 531.731; P < 0.0001). Annual ryegrasses grew well in medium supplemented with 5 g/L of KH2PO4 for 4 weeks, but concentrations of 10 and 15 g/L caused symptoms of toxicity in the form of leaf yellowing and necrosis. Thus, Pi concentration of 5 g/L was considered optimal for other parameters of this study. Increase in biomass (fresh weight) after 2 weeks of treatment (5 g/L KH2PO4) was more than 100% over control growth and about 48% in excess of the growth at a low dose (0.57 g/L) used in this study. Marshall ryegrass demonstrated better adaptation than Gulf ryegrass in Pi-enriched medium and showed less yellowing of leaves (data not presented). An earlier study by Delorme et al. (19) pointed to the phosphate phytoremediation potentials of a few crops [Indian mustard (Brassica juncea), canola (Brassica napus Westar), corn (Zea mays), collard (Brassica oleracea L. Acephala Group), alfalfa (Medicago sativa), soybean (Glycin max L.) etc.] grown in a greenhouse. They found the lowest P accumulation in canola shoot (0.2% of tissue dry weight) and the highest in collard and corn shoot (0.6 and 0.5% of tissue dry weight, respectively). Likewise, root P was recorded in the range of 0.2-0.5% dry weights for collard and corn. Only a few of these crops showed a shoot P/root P ratio of more than one. In the present study, both ryegrasses demonstrated far greater P accumulation potentials in roots as well as shoots (Figures 1 and 2). Shoot-to-root ratio of P
FIGURE 4. Scanning electron micrographs of the root of Gulf ryegrass seedlings grown in modified Hoagland’s medium in the presence (A) and absence (B) of 5 g/ L KH2PO4: (A) root section (scale marker ) 10 µm) shows bright fibrillar spots (arrowheads) in epidermal and cortical cells. (B) Control root section (scale marker ) 10 µm) shows no comparable spots (arrowhead indicates a dense, superimposed sheet of aluminum as a contaminant). Scanning micrographs of the leaf of Marshall ryegrass seedlings grown in modified Hoagland’s medium in the presence (C) and absence (D) of 5 g/L KH2PO4: (C) leaf section (scale marker ) 10 µm) shows abundance of bright fluffy structures (arrowheads) in epidermal and cortical cells. (D) Control leaf section (scale marker ) 10 µm) shows no such structures. VOL. 38, NO. 8, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2445
FIGURE 5. Electron dispersal X-ray spectroscopic analysis. (A) Spectrum of one of the bright spots (arrowheads) shown in Figure 4A. (B) Spectrum of control root shown in Figure 4B. (C) Spectrum of one of the bright fluffy structures (arrowheads) shown in Figure 4C. (D) Spectrum of control leaf shown in Figure 4D. accumulation was always more than 1 in Gulf ryegrass, while Marshall ryegrass equaled 1 after 21 days of growth in KH2PO4. The two grasses showed no significant differences in shoot P accumulation at all the tested concentrations of Pi (F1,49 ) 0.101; P > 0.05). High rate of P accumulation in annual ryegrass may be a reflection of enhanced transport and sequestration of P by these plants. In another study, phosphate uptake from a Pi-enriched solution was shown in perennial ryegrass as a function of time (20). Depletion of Pi from the nutrient solution by this plant was more than 20% after 7 days, but the level of fortified Pi was very low (0.014 g/L) and thus not comparable to this study on annual ryegrass. However, when wheat seedlings were grown in high concentrations of solution phosphate, P concentrations in plants increased to 1.3% (shoot DW) (21), but the accumulation was not in concentration-dependent manner as in annual ryegrass. Three cool-season turf grasses: Kentucky bluegrass (Poa pratensis), tall fescue (Festuca arundinaceae), and perennial ryegrass (Lolium perenne) were investigated for phosphate removal capacity from enriched soils (22). Clipping P differed significantly among these three grasses ranging from 0.3 to 0.45% of dry mass. These results also showed that genetic differences in P absorption might exist among turf grasses at both the interspecific and the intraspecific levels. Aggressive grasses such as Brachiaria species and a tropical forage legume, Arachis pintoi, known for P-uptake efficiency, when grown in greenhouse with different sources of soil P at the rate of 20-100 kg/ha, demonstrated manyfold less foliar P concentration than these annual ryegrasses (23). All these observations suggest that annual ryegrass has a far greater potential for removing excess P from the medium. The sterile hydroponic condition without any physical or chemical impedance for Pi uptake may have contributed to high rates of acquisition and accumulation. As suggested earlier, these studies only point to the potential of these grasses to accumulate the nutrient under unlimited supply and the minimal presence of chemical or biological agents that can bind dissolved P. Although these grasses can remove ef2446
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 8, 2004
fectively excess Pi from the wastewater, further studies under field conditions will reveal their potential application in phytoremediation of phosphate rich soils. Effect of pH on P Accumulation in Annual Ryegrass. Phosphate availability to plants is strongly influenced by pH of the growth medium. At low pH, Pi is known to complex with Fe and Al, and similarly, at high pH it can complex with Ca2+ (24-26). In the present study, the nutrient solution contained negligible concentrations of Al, Fe, and Ca (Pibinding factors) in proportion to the supplied Pi (5 g/L KH2PO4). In the absence of these cationic factors, Pi accumulation would more truly reflect the ability of plants to acquire the nutrient at varying pH levels. These grasses exhibited greater uptake and accumulation at lower pH. Shoot accumulation at pH 3.7 was approximately more than 30 and 14% when compared to those at pH 7.7 and 4.7, respectively (Figure 3). Although the difference in total accumulations (root and shoot) at pH 3.7 and 7.7 was significant (P < 0.05), the accumulations at pH 4.7-6.7 did not vary significantly. These results thus indicate that the solution pH has influence on Pi uptake. Scanning Electron Microscopy. The channel of P transport and accumulation in leaves and roots of ryegrass was mapped out by scanning electron microscopy (SEM) and electron dispersive X-ray spectroscopy (EDS) techniques. The SEM of roots excised from experimental plants (grown on 5 g/L KH2PO4) showed higher accumulation of P in epidermal and cortical cells. Bright fibrillar spots of Pi can be seen in these cells (Figure 4A). These observations support the notion that epidermal and cortical cells are the primary entry points for P uptake. Although cells of the stellar region show intense formation of bright patches, P is distributed unevenly throughout the root section. In the micrograph of control roots (supplied with no Pi), such bright formations are conspicuously absent (Figure 4B). Leaf samples from the experimental plants showed concentrations of brighter P spots from pericycle to epidermis. Some of the epidermal and cortical cells revealed the presence of bright fluffy structures rich in P and K (Figure 4C). SEM of control leaf
FIGURE 6. Accumulation pattern of P in roots and shoots of Marshall ryegrass during a starvation and replenishment experiment. Each point represents mean of six replicates ((SE). [0] represents plants grown in nutrient solution with no KH2PO4. [1] represents plants grown in nutrient solution with no KH2PO4 for 14 days and subsequently transferred to a fresh medium containing 5 g/L KH2PO4 for 12 days. [2] represents plants grown in nutrient medium containing 5 g/L KH2PO4 for 14 days. [3] represents plants grown in nutrient solution containing 5 g/L KH2PO4 for 14 days and subsequently transferred to a nutrient solution with no KH2PO4 for 12 days. [4] represents plants grown in nutrient solution containing 5 g/L KH2PO4 for 14 days and subsequently transferred to the same fresh nutrient solution for 12 days. samples is clearly distinguishable from those of P supplemented plant specimen with respect to fluffy structures (Figure 4D). The EDS analysis (P peaks) of KH2PO4-exposed plant parts is also distinguishable from those of controls (Figure 5 A-D). SEM and EDS observations presented in this study lead to the belief that high levels of Pi flow into the cells via symplastic (via cell-to-cell connections) pathways. However, apoplastic (outward flow via cell walls) migration of Pi cannot be ruled out. Both apoplastic migration and symplastic transportation of heavy metals have been reported in heavy metal hyperaccumulator species (27). Starvation and Replenishment Experiment. It is generally observed that a brief period of Pi starvation leads to transcriptional activation of high affinity Pi transporter genes resulting in an enhanced uptake of this nutrient (28). In this experiment, Pi deprivation and replenishment conditions were created to reflect states of P deficiency and sufficiency and to observe how these states affect plants’ capability of P aquisition. Figure 6 depicts that P accumulation was greater at an initial dose of 5 g/L KH2PO4 than both conditions when (1) plants were supplied Pi after 14 days of starvation period and (2) plants were first supplied with Pi and then deprived for another 12 days. However, accumulations were higher when Pi augmentation followed the starved phase than the one preceding the starvation phase. A repeat of the same dose (5 g/L Pi) resulted in highest accumulations, particularly in shoots (Figure 6). These data suggest that when the Pi concentration is very high in the medium, the low affinity Pi uptake system constitutively operating in plants may overwhelm the high affinity Pi uptake system. Excretion of excess phosphate into the medium is one of the common mechanisms used by plants to maintain P homeostasis. It has been shown that under Pi sufficiency conditions, plants excrete nearly 80% of the absorbed Pi by the action of anion channels (29). It will be interesting to study how the P excretion mechanism is operating in a P accumulator such
as annual ryegrass. It is also interesting to note that these grasses do not demonstrate P toxicity symptoms when growing on excessive Pi (up to 5 g/L), whereas other plants show noticeable symptoms (30, 31). It appears that ryegrass may have a more efficient P sequestration mechanism to avoid P toxicity. Another attractive feature of this plant is the ability to sequester P in above-ground parts of the plant. A similar phenomenon was observed in a well-characterized mutant of Arabidopsis (Pho2) exhibiting an enhanced capacity to accumulate more P in above-ground parts (32). The gene responsible for the phenotype of this mutant is not yet characterized. Furthermore, in annual ryegrass, the negative correlation between shoot Zn and shoot P was not found. Shoot Zn level was recorded as 121 mg/kg DW when plants accumulated P (2.2 g/kg DW) in their shoots exposed to 5 g/L KH2PO4. The control plants supplied with no Pi contained only 27 mg Zn/kg DW in their shoots. Zinc status of plant tissue has been shown to regulate Pi uptake in both Pi-sufficient and -deficient barley roots (33). Shoot Zndeficiency (
