Accumulation and volatilization of different chemical species of selenium by plants

Abstract.Selenium (Se)removal from polluted waters and soils is especially complicated and highly expensive.Phytoremediation has been suggested as a low-cost,e cient technology for Se removal.Plants remove Se by uptake and accumulation in their tissues,and by volatilization into the atmosphere as a harmless gas.Unraveling the mechanisms of Se uptake and volatiliza-tion in plants may lead to ways of increasing the e ciency of the phytoremediation process.The objec-tives of this study were:(i)to determine the e ect of di erent Se forms in the root substrate on the capacity of some plant species to take up and volatilize Se;(ii)to determine the chemical species of Se in di erent plant parts after the plants were supplied with various forms of Se;and (iii)to determine the in¯uence of increasing sulfate levels on plant uptake,translocation,and vola-tilization of di erent Se species.Plants of broccoli (Brassica oleracea var.botrytis L.),Indian mustard (Brassica juncea L.),sugarbeet (Beta vulgaris L.)and rice (Oryza sativa L.)were grown hydroponically in growth chambers and treated for 1week with 20l M Se as Na 2SeO 4,Na 2SeO 3or L -selenomethionine (SeMeth)and increasing sulfate levels.The data show that shoots of SeO 4-supplied plants accumulated the greatest amount of Se,followed by those supplied with SeMeth then SeO 3.In roots,the highest Se concentrations were attained when SeMeth was supplied,followed by SeO 3,then SeO 4.The rate of Se volatilization by plants followed the same pattern as that of Se accumulation in roots,but the di erences were greater.Speciation analysis (X-ray absorption spectroscopy)showed that most of the Se taken up by SeO 4-supplied plants remained unchanged,whereas plants supplied with SeO 3or SeMeth contained only SeMeth-like species.Increasing the sulfate level from 0.25mM to 10mMinhibited SeO 3and SeMeth uptake by 33%and 15±25%,respectively,as compared to an inhibition of 90%of SeO 4uptake.Similar results were observed with regard to sulfate e ects on volatilization.We conclude that reduction from SeO 4to SeO 3appears to be a rate-limiting step in the production of volatile Se compounds by plants.Inhibitory e ects of sulfate on the uptake and volatilization of Se may be reduced substantially if Se is supplied as,or converted to,SeO 3and/or SeMeth rather than SeO 4.Key words:Ion uptake ±Selenium uptake ±Selenomethionine ±Speciation (selenium)±Sulfate ±Volatilization (selenium)IntroductionSelenium (Se),a required trace element in the diet of animals and humans,is poisonous to animals,humans and plants if ingested in relatively high concentrations.Recently,after the discovery of Se poisoning to wildlife in California,much attention has been focused on developing ways for the cleanup of Se-contaminated soils and waters and the prevention of future build-up of Se in the environment.Phytoremediation has been suggested as a low-cost,environmentally friendly,and highly e cient approach for the cleanup of Se pollution (Terry and Zayed 1998).Plants can scavenge Se from a large volume of soil and bioconcentrate it within their relatively small biomass,which can then be harvested and removed totally from the contaminated site.In addition,plants have the capacity to remove Se from the contaminated rooting medium by metabolizing it into a non-toxic volatile gas,e.g.dimethyl selenide (DMSe)(Lewis 1971;Terry and Zayed 1994).Among other factors,the rates of Se uptake and volatilization are governed largely by the chemical form of Se present in the root substrate.Plants take up Se as selenate (SeO 4),selenite (SeO 3),and/or organic Se.MostAbbreviations:SeMeth L -selenomethionine;XAS X-ray absorption spectroscopyCorrespondence to :N.Terry;E-mail:nterry@;Tel./Fax:1(510)6423510Planta (1998)206:284±292Accumulation and volatilization of di erent chemical species of selenium by plantsAdel Zayed,C.Mel Lytle,Norman TerryDepartment of Plant and Microbial Biology,111Koshland Hall,University of California,Berkeley,CA 94720±3102,USA Received:27February 1998/Accepted:30March 1998research regarding the availability of di erent forms of Se for plant uptake and accumulation has focused on the comparison among inorganic chemical species of Se with special emphasis on the two oxidized forms:SeO4and anic forms of Se may be more readily available to plant uptake and volatilization than inorganic forms (Williams and Mayland1992;Verkleij and MacNaeidhe 1992;Zijian et al.1991).Research on Se volatilization by microorganisms showed that microbial volatilization could be100-fold higher from L-selenomethionine (SeMeth)than from inorganic Se(Reamer and Zoller 1980).Unfortunately,there are very few studies available showing how plant uptake and volatilization of Se is in¯uenced by its chemical form.In the present work,we compared uptake and volatilization of three Se forms, namely SeO4,SeO3and SeMeth,in four di erent plant species,namely broccoli,Indian mustard,sugarbeet and rice.The objective of the research was to investigate the e ect of the presence of di erent Se forms in the root substrate on the capacity of a variety of plant species to take up and volatilize Se.This information is vital in order to elucidate the physiological basis of Se uptake and volatilization in plants.Another objective of our study was to determine the chemical species of Se in di erent plant parts after being supplied with various forms of Se.Once Se is taken up by plant roots in a particular chemical form,whether organic or inorganic,it undergoes certain metabolic changes that determine the®nal product as well as its translocation and accumulation in di erent plant tis-sues.If these changes are identi®ed,it will be possible to determine the key-limiting steps in the Se metabolic pathway,which then can be manipulated by modern molecular biological techniques.Several researchers have tried to devise tools for the fractionation and speciation of Se in plant tissues(Asher et al.1977; Gissel-Nielsen1979,1987).Most of these attempts involved either indirect measurements of the di erent forms of Se or treatments that may cause chemical changes in Se species.Synchrotron-based X-ray absorp-tion spectroscopy(XAS)techniques on the other hand, allow for direct measurement of the chemical species in vivo.This technique has recently been used in the environmental speciation of contaminants in various ecological samples,including waters,soils,and plants (Pickering et al.1995;Salt et al.1995,1997;Kra mer et al. 1996;Lytle et al.1996;Manceau and Gallup1997). Chemical speciation of Se in plant tissues using XAS techniques is extremely speci®c so that once adjusted it measures only Se with no interference from any other ing this technique,we were able to accu-rately determine the Se species that accumulated inside plant tissues in the forms in which it is actually present in the intact and functioning plant.The third objective of our study was to determine the in¯uence of increasing sulfate supply levels on the uptake,translocation,and volatilization of Se by plants supplied with Se as SeO4,SeO3,or SeMeth.The rate of Se uptake and volatilization depends partly on the concentrations of competing ions in the root medium.Earlier research showed that SeO4uptake and volatil-ization by broccoli plants were inhibited as the sulfate level increased in the growth medium(Zayed and Terry 1992,1994).Since high levels of Se and S naturally coexist in most seleniferous soils,it is necessary to understand the e ects of high sulfate levels on the uptake and volatilization of Se from various chemical forms of Se that may be found in most soils,e.g.SeO4, SeO3,and SeMeth.This will help develop more-e cient phytoremediation techniques suitable for areas where sulfate levels are high.For instance,if it is found that SeO3or SeMeth uptake and volatilization are consider-ably less a ected by the higher sulfate levels than SeO4, then chemical,physical,or biological treatments may be used to change the Se form so that higher rates of uptake and/or volatilization can be achieved.Unfortunately, there are no data available to show how the changes in sulfate levels will a ect SeMeth or SeO3uptake and volatilization by plants.Materials and methodsPlant culture.Plant species used in this study include broccoli (Brassica oleracea var.botrytis L.),Indian mustard(Brassica juncea L.),sugarbeet(Beta vulgaris L.),and rice(Oryza sativa L.).All plants were cultured hydroponically in growth chambers at25°C with an irradiance of500l mol m A2s A1photosynthetic photon¯ux supplied over a16-h photoperiod.Seeds were sown in sand for two weeks during which time they were irrigated with half-strength Hoagland's solution.Germinating seedlings of similar size were transplanted and allowed to grow for another two weeks in half-strength Hoagland's solution.Selenium(20l M)was then added as Na2SeO4,Na2SeO3(Sigma Chemical Co.),or SeMeth(CalBio-chem-Novabiochem Intl.)to the nutrient solution.To test the e ect of increasing sulfate concentrations on Se uptake and volatiliza-tion,a group of broccoli plants was transferred to a modi®ed half-Hoagland's solution supplied with MgSO4in three di erent concentrations,i.e.0.25,1.0,10.0mM.For each sulfate treatment, Se(20l M)was added to the nutrient solution as Na2SeO4, Na2SeO3,or SeMeth.All the plants were allowed to grow for one more week before Se volatilization was measured.Determination of the rate of Se volatilization and analysis of Se in plant materials.One week following the start of Se treatment, plants were transferred to Plexiglas chambers and the rate of Se volatilization was determined for roots and shoots separately as described by Zayed and Terry(1994).For each treatment of each plant species,we obtained four individual volatilization measure-ments for roots and two measurements for shoots,with each shoot volatilization measurement representing two individual plants. After determining volatilization rates,plant shoots and roots were removed from the collection chambers,washed with running tap water,and oven-dried at70°C for48h.Dry plant tissues were digested and prepared for Se analyses using the HNO3/HClO4 procedure according to Mikkelsen(1987).Total Se in the plant extract was then measured using the atomic absorption/hydride generation method following Varian's procedure for the VGA-76 vapor generator.Speciation of Se in plant tissues by XAS analysis.To perform XAS analysis on Se in plant tissues,Se-treated broccoli plants were harvested,separated into leaf and root tissues,rinsed in deionized water,frozen in liquid nitrogen,ground to a®ne texture,and stored at A80°C.Selenium K-edge analyses of plant tissues were completed on Beam Line4-1of the Stanford Synchrotron Radiation Laboratory using an Si(111)double-crystal mono-chromator with a 1-mm entrance slit.Frozen leaf and root tissues were placed in a sample chamber at a 45°angle to the X-ray beam.X-ray ¯uorescence of Se in plant tissues was obtained by a series of replicate scans (3±15)dependent on Se concentration.XAS spectra were also collected for Se reference compounds.Background subtraction and normalization were carried out according to established procedures (Koningsberger and Prins 1988).ResultsSelenium concentrations in shoots and roots.Selenium accumulation was determined in shoots and roots of Indian mustard,broccoli,sugarbeet,and rice plants grown in nutrient solutions containing 20l M Se as SeO 4,SeO 3,or SeMeth (Fig.1A,B).Maximum Se levels (366±550mg kg A 1DW)in shoots of the di erent plant species were attained when Se was supplied as SeO 4,except for sugarbeet,where maximum Se levels (405mg kg A 1)were reached when Se was supplied as SeMeth (Fig.1A).In all species,shoots of SeO 3-supplied plants attained the least Se concentrations (<73mg kg A 1)compared to SeO 4or SeMeth-supplied plants.In roots of all plants,Se was bioaccumulated to the highest levels when they were supplied with SeMeth (655±2197mg kg A 1),followed by those supplied with SeO 3(230±690mg kg A 1),then those supplied with SeO 4(78±169mg kg A 1)(Fig.1B).Generally,Se transport from root to shoot was highly dependent on the chemical form of the Se supplied(Table 1).The Shoot/root ratio of total Se content in all plant species studied ranged from 0.6to 1for plants supplied with SeMeth and was less than 0.5for those supplied with SeO 3,while this ratio ranged from 1.4for rice to 17.2for Indian mustard when SeO 4was the form of supplied Se.Maximum Se accumulation per plant (shoot Se content +root Se content)was reached in plants supplied with SeO 4in broccoli and Indian mustard,while in sugarbeet and rice it was observed for plants supplied with SeMeth (Table 1).Selenite was the least accumulated form of Se in plant tissues.Of these plant species studied,Indian mustard was thebestTable 1.Ratio of shoot Se content to root Se content and total plant uptake of Se in Indian mustard,broccoli,sugarbeet and rice as in¯uenced by the chemical form of Se supplied Chemical form of Se suppliedBroccoliIndian mustard SugarbeetRiceSelenium shoot/root ratio Selenate 9.5517.2 6.121.44Selenite 0.470.300.240.40SeMeth 0.910.98NA a 0.58Total Se uptake (mg plant A 1)Selenate2.588.65 1.400.04Selenite 0.60 2.580.490.08SeMeth 2.005.555.140.38aNot availableFig.1A,B.Selenium concentrations in shoots (A )and roots (B )of Indian mustard,broccoli,rice,and sugarbeet plants supplied with 20l M Se as selenate,selenite,or SeMeth.Error bars indicate SD.NA not availableat accumulating Se in tissues regardless of the chemical form of supplied Se (Table 1).Increasing sulfate supply in the growth medium from 0.25mM to 10mM resulted in a progressive inhibition in SeO 4uptake,while it caused little or no e ect on SeO 3and SeMeth uptake,as judged by shoot and root Se concentrations (Fig.2).In SeO 4-supplied plants,in-creasing sulfate supply from 0.25mM to 1mM and 10mM decreased Se concentrations by 47%and 93%in shoots,respectively,and by 72%and 90%in roots,respectively.Plants supplied with SeO 3or SeMeth exhibited no signi®cant changes in Se concentrations with the increase in sulfate levels,except for roots of plants supplied with SeO 3,which exhibited a lower uptake at higher sulfate supply levels (33%lower uptake at both 1mM and 10mM compared to 0.25mM;Fig.2).In general,the increase in sulfate supply had amuch stronger inhibitory e ect on SeO 4uptake and accumulation than on that of SeO 3or SeMeth,especially when the sulfate level increased from 1to 10mM (Fig.2;Table 2).Rate of Se volatilization.The rates of Se volatilization by shoots and roots were measured for four di erent plant species supplied with 20l M Se as SeO 4,SeO 3,or SeMeth.The data show that for all four plant species the rate of Se volatilization was greatest when plants were supplied with Se as SeMeth and least when they were supplied with SeO 4(Fig.3).Selenium volatilization by shoots of broccoli,Indian mustard,sugarbeet,and rice plants supplied with SeMeth was 38-,98-,48-,and 15-fold higher than those supplied with SeO 4,respectively,and 25-,188-,63-,and 16-fold higher than those supplied with SeO 3,respectively.In roots,the rates of Se volatilization of broccoli,Indian mustard,sugarbeet,and rice plants supplied with SeMeth was 126,83,89,and 36times higher than those rates of plants supplied with SeO 4,respectively,and 66,51,73,and 20times higher than those of plants supplied with SeO 3,respec-tively.Roots also volatilized Se slightly faster (21±90%faster)from SeO 3than from SeO 4in all plant species studied (Fig.3).Root volatilization of Se from SeO 4,SeO 3and SeMeth was 5±14,8±30,8±48times faster than that of shoots,respectively,in the di erent plant species studied.Increasing the sulfate level in the growth medium from 0.25mM to 1mM resulted in a substantial reduction (>5-fold decrease)in the rate of Se volatil-ization by SeO 4-supplied broccoli plants.Further reduc-tion in the Se volatilization rate occurred with the increase in sulfate supply levels from 1mM to 10mM,but it was much less pronounced (Fig.4).In SeO 3-supplied plants,an increasing level of sulfate supply resulted in a slight inhibition in the rate of Se volatil-ization by roots,but had no signi®cant e ect on shoot Se volatilization rate.Rates of Se volatilization by SeMeth-supplied plants showed an inconsistent response to the changes in sulfate level (Fig.4).The data show that as the sulfate level increased from 0.25mM to 10mM the ratio of Se volatilization rate of SeO 3-supplied broccoli plants to that of SeO 4-supplied plants (SeO 3/SeO 4volatilization ratio)increased from 0.42to 3.26in shoots and from 0.64to 2.65in roots (Table 2).Similarly,the SeMeth/SeO 4volatilizationFig.2.In¯uence of increasing sulfate supply on Se concentrations in roots and shoots of broccoli plants supplied with 20l M Se supplied as sodium selenate (Selenate ),sodium selenite (Selenite ),or L -selenomethionine (Se-Met ).Error bars indicate SDTable 2.In¯uence of increasing sulfate levels in the nutrient solution on the Se concentration and volatilization ratios of plants supplied with SeO 3or SeMeth as compared to those supplied with SeO 4Sulfate level (mM)Se volatilization ratio Se concentration ratio SeO 3/SeO 4SeMeth/SeO 4SeO 3/SeO 4SeMeth/SeO 4Shoot 0.250.4242.90.040.201.00 2.1836.70.060.2410.0 3.26106.20.50 5.10Root 0.250.6434.90.78 3.921.00 1.30239.4 1.8811.810.02.65332.75.3829.5ratio increased from 42.9to 106.2in shoots and from 34.9to 332.7in roots along with the increase in external sulfate level (Table 2).Analysis by XAS of Se in reference compounds and plant tissues.The Se K-edge spectra of reference compounds of SeO 4,SeO 3,Se 0,and SeMeth are shown in Fig.5A.It is apparent by the shift in energy towards more-positive values that the reference compounds'K-edges are sensitive indicators of valence.There was a positive shift in the K-edge along the X-ray energy axis with the formal oxidation state of Se.Similar results have been reported earlier for a wider range of Se reference compounds by Pickering et al.(1995).For this study,the selection of Se reference compounds corresponded to Se treatments.The theory of the K-absorption edge transitions has been clearly discussed and modeled (Kutzler et al.1980).The results for the XAS K-edge analysis for shoot and root tissues of broccoli plants treated with 20l M Se as SeO 4,SeO 3,or SeMeth are shown in parisons of the XAS spectra of the leaf samples (Fig.5B)with those obtained for reference materials (Fig.5A)indicate that Se in the leaves of plants suppliedwith SeO 3or SeMeth is consistent with SeMeth.However,the XAS spectrum of leaves of SeO 4-supplied plants shows that most of the Se remains as SeO 4with a smaller fraction being converted into an organoselenium species most like the SeMeth reference compound.Similar results were observed for root samples treated with SeO 4,SeO 3or SeMeth,except that the fraction of the SeO 4-Se which was converted into organoselenium in SeO 4-supplied plants was much higher in roots than that in leaves (Fig.5C).DiscussionIn¯uence of the chemical species of supplied Se.Varia-tions in the chemical form of the supplied Se greatly in¯uenced the ability of plants to accumulate,translo-cate,and volatilize Se from roots and shoots.In general,plant leaves accumulated Se to the greatest levels when Se was supplied as SeO 4,followed by SeMeth and then SeO 3,while the highest concentrations in roots were attained when SeMeth was supplied,followed by SeO 3,and then SeO 4(Fig.1).Selenium translocation from roots to shoots is best described as SeO 4)SeMethFig.3.Rates of Se volatilization by shoots and roots of Indian mustard,broccoli,rice,and sugarbeet plants supplied with 20l M Se as selenate (SeO 4),selenite (SeO 3),or L -selenomethionine (Se-Met ).Error bars in-dicate SD>SeO3in all plant species studied(Table1).Rates of plant Se volatilization closely corresponded to Se accumulation in roots,except that the di erences in the rates of Se volatilization were greater than the di erences in Se concentrations in roots.This is especially true for plants supplied with SeMeth as compared to those supplied with SeO4or SeO3.Seleni-um volatilization by plant roots was20±73times faster from SeMeth than from SeO3,and was1.2±1.9times faster from SeO3than from SeO4(Fig.3).Selenium concentrations in shoots did not correlate with the rate of plant Se volatilization.This suggests that the rate of Se volatilization depends mainly on the uptake and accumulation of Se by roots.It has already been established that roots are the main sites of Se volatil-ization in plants(Zayed and Terry1994).The question of whether rhizosphere microorganisms are involved in this process or not remains unanswered.The greater di erences we observed in the rates of plant Se volatilization as compared to the di erences in Se concentrations as in¯uenced by the chemical species of Se can be attributed to the ability of the plant to metabolize di erent forms of Se.Chemical speciation of selenium in plant tissues using the XAS technique showed that when plants were supplied with Se as SeO4,most of the Se remained unchanged in the leaf and root tissues with a small portion being converted to a SeMeth-like species;this portion being greater in roots than in leaves.However,plants supplied with SeO3or SeMeth contained only SeMeth-like Se species in both roots and leaves,with no SeO4,SeO3,or Se0observed. These data provide evidence for the rapid reduction of SeO3by plant roots and suggest that the reduction of SeO4to SeO3is a rate-limiting step in the production of volatile Se compounds by plants.This is because SeO3-supplied plants volatilized Se slightly faster than those supplied with SeO4,despite the fact that total Se uptake and accumulation was much greater in SeO4-supplied plants(4-to5-fold higher)than in SeO3-supplied plants.The much greater rates of Se volatiliza-tion from SeMeth-supplied plants,compared to SeO4 and SeO3-supplied plants,can be attributed to:(i)the overall rate of SeMeth uptake being similar to(broccoli and Indian mustard)or slightly higher than(sugarbeet and rice)that of SeO4,and much faster than that of SeO3(Table1);and(ii)all the Se in roots and shoots of SeMeth-supplied plants remaining as SeMeth-like spe-cies(volatilization from SeMeth is more favorable energetically than from inorganic Se).The di erences in plant Se uptake and movement due to the variation in the form of Se supplied may be attributed to the fact that plants take up and translocate SeO4,SeO3,and SeMeth by dissimilar mechanisms. Previous research has shown that while the uptake of SeO4and SeMeth was driven metabolically,uptake of SeO3had a major passive component(Ulrich and Shrift 1968;Asher et al.1977;Abrams et al.1990).Arvy(1993) demonstrated that within3h,50%of the SeO4taken up by bean plant roots moved to shoots,while in the case of SeO3most of the Se remained in the root and only a small fraction was found in the shoot.The addition of hydroxylamine,a respiratory inhibitor,to the nutrient solution inhibited SeO4uptake by80%while SeO3 uptake was inhibited only by20%.Similarly,Abrams et al.(1990)showed that SeMeth uptake by wheat seedlings was inhibited by metabolic inhibitors and by anaerobic conditions,suggesting an active uptake of SeMeth bywheat.Fig.4.In¯uence of increasing sulfate sup-ply on Se volatilization in shoots(top)and roots(bottom)of broccoli plants supplied with20l M Se supplied as sodium selenate (SeO4),sodium selenite(SeO3),orL-selenomethionine(Se-Met).Error bars indicate SDEarlier research also showed that Se volatilization by plants might be dependent on the chemical form of Se supplied.Lewis et al.(1974)revealed that 10±16times more Se was released from excised cabbage leaves taken from plants supplied with SeO 3than those taken from plants supplied with SeO 4.In another study,Asher et al.(1967)found that,during oven-drying,roots of SeO 3-grown plants released 11times more volatile Se than did roots of SeO 4-grown plants.Most recently,Zhang and Moore (1997),while investigating Se volatilization in wetland microcosms,found that the concentration ofdissolved organic Se is more important than dissolved inorganic Se in a ecting Se volatilization from wetland plants and sediments.Selenium volatilization by soil microorganisms has also been shown to be an order of magnitude higher from organic Se than from SeO 4,SeO 3,or elemental Se (Doran and Alexander 1977;Reamer and Zoller 1980;Frankenberger and Karlson 1994).Thus,it appears that the biological volatilization of Se by plants and microbes proceeds more rapidly if Se is supplied in more-reduced forms (e.g.SeO 3,or SeMeth as compared to SeO 4).In plants,when Se is supplied as SeO 4,most of the Se is transported to the shoot unchanged with very little Se remaining in roots.The detection of more SeMeth-like species in roots of SeO 4-supplied plants than in shoots partly explains why roots volatilize faster than shoots,despite the fact that roots contain much less total Se.The overall rate of plant Se volatilization is substantially lower from SeO 4than from SeO 3or SeMeth because of the diminished accumulation in roots and lower rate of metabolic Se conversion.On the other hand,when Se is supplied as SeMeth or SeO 4,a large volume of the absorbed Se remains in roots mainly as Se-Meth,leading to greatly enhanced rates of Se volatilizationIn¯uence of sulfate supply.The results of this study showed that increasing sulfate supply had a stronger inhibitory e ect on the uptake and volatilization of SeO 4compared to those of SeO 3or SeMeth (Figs.2,4;Table 2).Previous research has shown that SeO 4uptake and volatilization by plants are strongly inhibited by the presence of high concentrations of sulfate in the growth medium (Pratley and McFarlane 1974;Mikkelsen et al.1988;Mikkelsen and Wan 1990;Zayed and Terry 1992,1994;Barak and Goldman 1997).Antagonistic sulfate/selenate interactions have generally been reported in single cells,excised roots,and whole plants (Lauchli 1993).Several researchers reported reductions in SeO 4concentration in plant tissues ranging from 80%to 100%when sulfate level was increased in the growth medium (Leggett and Epstein 1956;Ulrich and Shrift 1968;Ferrari and Renosto 1972;Barak and Goldman 1997).Previous studies also showed that sulfate has a much greater e ect on plant uptake of SeO 4than SeO 3(Pratley and McFarlane 1974;Gissel-Nielsen 1973).However,other than the results of this study,there are no data available describing the e ects of increasing sulfate levels on SeMeth uptake and volatilizationbyFig.5A±parison of Se K-edge XANES of:A Se 0,SeO 3,SeO 4,and Se-Meth reference solutions (note the shift in energy towards more positive energy as the valence state of Se increases);B broccoli leaf tissues taken from plants previously supplied with 20l M Se as selenate,selenite,or selenomethionine;and C broccoli root tissues taken from plants previously supplied with 20l M Se as selenate,selenite,or selenomethionine (see Materials and methods for details)Table 3.In¯uence of sulfate supply level on shoot/root con-centration ratio of broccoli plants treated with 20l M Se supplied SeO 4,SeO 3or SeMeth at di erent sulfate levels Sulfate level (mM)Shoot/root concentration ratio SeO 4SeO 3SeMeth 0.25 2.640.140.131.00 5.920.190.1210.01.920.180.33plants.The ability of plants to acquire Se in sulfate-rich soils was related to the presence of organic Se forms in these soils(Abrams et al.1990).It is established that plants can actively take up SeMeth from solution (Abrams et al.1990).The question is:Why does sulfate inhibit SeO4uptake more than SeO3and SeMeth? Inhibitory e ects of sulfate on plant Se uptake,and therefore volatilization,are expected to be greater with respect to SeO4as compared to SeO3or SeMeth due to the competition between SeO4and sulfate for the uptake sites in the root cell membranes.Plants take up SeO4by the same carrier in root cell membranes as sulfate (Leggett and Epstein1956;Ferrari and Renosto1972). Selenite and SeMeth,on the other hand,are taken up by other mechanisms.Plant uptake of SeO3is mostly passive(Arvy1993),whereas,the uptake of SeMeth is thought to be mediated by uptake-permeases speci®c for S-containing amino acids.This conclusion comes from the results obtained by Sandholm et al.(1973),who showed that uptake of L-methionine and L-selenome-thionine by the alga Scenedesmus dimorphys was com-petitive,indicating that the two amino acids may be taken up by the same mechanism.It seems likely therefore that sulfur compounds will competitively inhibit the uptake of their Se analogues(i.e.sulfate inhibits SeO4and methionine inhibits SeMeth).How-ever,experiments conducted by Asher et al.(1977) indicate that neither sulfate nor sul®te caused signi®cant reductions in SeO3uptake by tomato root systems.They also reported that the presence of excess sul®te,and not sulfate,enhanced the transport of SeO3in tomato plants.In our study,increasing sulfate levels did not show any consistent e ect on long-distance transport of Se when supplied as SeO4,SeO3or SeMeth,as judged by the shoot/root concentration ratio(Table3).Rates of Se volatilization by SeO4-supplied plants were shown here to be greater than those of SeO3-supplied plants only at the lowest sulfate level (0.25mM).At all other sulfate levels(1±10mM),plant Se volatilization from SeO3exceeded that of SeO4.The abnormally higher rates of Se volatilization from SeO4 at0.25mM sulfate supply,as compared to SeO3,can be explained by the substantially increased Se accumulation in roots and shoots of broccoli at that level of sulfate due to the lesser competition between sulfate and SeO4.No such increase in uptake and accumulation was observed for SeO3at low sulfate levels,leading to larger di er-ences in tissue Se concentrations between the two Se anions.Rates of Se volatilization from SeMeth were faster than those of SeO4at all sulfate levels,but the di erences were greater at high sulfate supply than at low supply.This may be a re¯ection of the di erences in: (i)Se accumulation in roots;and/or(ii)energy require-ments for the formation of volatile Se.This study was supported by grants from the University of California Salinity/Drainage Task Force and the Electric Power Research Institute.The authors thank Stanford Synchrotron Radiation Laboratory for granted beam time(proposal2413MP) and on-line support.ReferencesAbrams MM,Shennan C,Zasoski J,Burau RG(1990)Selenome-thionine uptake by wheat seedlings.Agron J82:1127±1130 Arvy MP(1993)Selenate and selenite uptake and translocation in bean plants(Phaseolus vulgaris).J Exp Bot44:1083±1087 Asher CJ,Evans CS,Johnson CM(1967)Collection and partial characterization of volatile selenium compounds from Medic-ago sativa L.Aust J Biol Sci20:737±748Asher CJ,Butler GW,Peterson PJ(1977)Selenium transport in root systems of tomato.J Exp Bot23:279±291Barak P,Goldman I(1997)Antagonistic relationship between selenate and sulfate uptake in onion(Allium cepa):implications for the production of organosulfur and organoselenium com-pounds in plants.J Agric Food Chem45:1290±1294Doran JW,Alexander M(1977)Microbial formation of volatile Se compounds in soil.Soil Sci Soc Am J40:687±690Ferrari G,Renosto F(1972)Regulation of sulfate uptake by excised barley roots in the presence of selenate.Plant Physiol 49:114±116Frankenberger Jr WT,Karlson U(1994)Microbial volatilization of selenium from soils and sediments.In:Frankenberger Jr WT, Benson S(eds)Selenium in the environment.Dekker,New York,pp369±387Gissel-Nielsen G(1973)Uptake and distribution of added selenite and selenate by barley and red-clover as in¯uenced by sulphur.J Sci Food Agric24:649±655Gissel-Nielsen G(1979)Uptake and translocation of selenium-75 in Zea mays.``Isotopes and radiation in research on soil-plant relationships''.IAEA,Vienna,pp427±436Gissel-Nielsen G(1987)Fractionation of selenium in barley and rye-grass.J Plant Nutr10(9±16):2147±2152 Koningsberger DC,Prins R(1988)X-ray absorption:principles, applications,techniques of EXAFS,SEXAFS and XANES.Wiley,New York,p673Kra mer U,Cotter-Howells JD,Charnock JM,Baker AJM,Smith JAC(1996)Free histidine as a metal chelator in plants that accumulate nickel.Nature379:635±638Kutzler FW,Natoli CR,Misemer DK,Doniach S,Hodgson KO (1980)Use of one electron theory for the interpretation of near edge structure in K-shell X-ray absorption spectra of transition metal complexes.J Chem Phys73:3274±3288Lauchli A(1993)Selenium in plants:uptake,functions,and environmental toxicity.Bot Acta106:455±468Leggett JE,Epstein E(1956)Kinetics of sulphate absorption by barley roots.Plant Physiol31:222±226Lewis BG(1971)Volatile selenium in higher plants:the production of dimethyl selenide in cabbage leaves by the enzymatic cleavage of methylselenomethionine selenonium salt.PhD thesis,University of California,BerkeleyLewis BG,Johnson CM,Broyer TC(1974)Volatile selenium in higher plants.The production of dimethyl selenide in cabbage leaves by enzymatic cleavage of Se-methyl selenomethionine selenonium salt.Plant Soil40:107±118Lytle CM,Lytle FW,Smith BN(1996)Use of XAS to determine the chemical speciation of bioaccumulated manganese in Potamogeton pectinatus.J Environ Qual25:311±316 Manceau A,Gallup DL(1997)Removal of selenocyanate in water by precipitation:characterization of copper-selenium precipi-tate by X-ray di raction,infrared,and X-ray absorption spectroscopy.Environ Sci Technol31:968±976Mikkelsen R(1987)Materials and methods for determination of selenium in plants and soils.In:Workshop on analytical methods for selenium,other trace elements,and on quality control/quality assurance,Sacramento,Calif,24±25March 1987pp63±66Mikkelsen RL,Wan HF(1990)The e ect of selenium on sulfur uptake by barley and rice.Plant Soil121:151±153。