PESTICIDE BIOCHEMISTRY AND PHYSIOLOGY 58, 13–22 (1997) ARTICLE NO. PB972280
Glyphosate Uptake in Catharanthus roseus Cells: Role of a Phosphate Transporter F. Morin, V. Vera, F. Nurit,* M. Tissut,* and G. Marigo1 Laboratoire d’Ecophysiologie Ve´ge´tale, Centre de Biologie Alpine, and *Laboratoire de Physiologie Cellulaire Ve´ge´tale, Universite´ Joseph Fourier, BP 53, F-38041 Grenoble Cedex 9, France Received April 14, 1997; accepted July 14, 1997 In Catharanthus roseus L. cells, the study of glyphosate uptake indicates that, at low concentrations, glyphosate is transported against a concentration gradient, reaching an accumulation ratio of about 30, 18 h after the cells are transferred into a fresh medium. Evidence that part of the glyphosate uptake may be carrier mediated was provided by concentration dependence experiments which showed that glyphosate uptake exhibited a saturation phase at low concentrations (up to 50 mM). The role played by a phosphate transporter in this process is demonstrated by the existence of a lag period and the inhibition of glyphosate uptake in the presence of high concentrations of sodium phosphate. It is also shown by the effect of PFA, a powerful inhibitor of phosphate transport in animal cells, and also by the action of protein chemical reagents (PCMBS and DCCD). Compartmental analysis with isolated protoplasts and vacuoles indicated that [14C]glyphosate is distributed between the cytosolic and the vacuolar compartments, but the greater part is localized in the cytosol. Detailed studies carried out to investigate the requirements of the glyphosate transporter showed that, among the different constituents of the Gamborg’s nutrient medium, the major elements increasing the cellular glyphosate uptake were Ca2+, Mg2+, and the presence of iron. q1997 Academic Press
INTRODUCTION
(11) and promotes metabolism of indole-3-acetic acid (12, 13). It is probable that the additive effects of the observed phenomena are responsible for the ultimate death of plants. The effectiveness of glyphosate as a herbicide is also dependent on its translocation within the plant. Phloem transport of glyphosate has been demonstrated in several species (6, 14, 15, 16). However, in addition to the long-distance transport, it should be noted that glyphosate penetration of cells must occur in each cell containing the target enzyme. The cell penetration mechanisms are not yet completely understood. For example, in some species [Convolvulus arvensis, (16); Beta vulgaris, (17); Vicia faba, (18)] glyphosate is transported within the cells by passive nonfacilated uptake, but more detailed kinetic data have shown, in V. faba, that glyphosate may be taken up via a plasmalemma phosphate transporter (19). Contradictory information is also available concerning the cellular distribution of glyphosate, particularly at the vacuolar level. The presence of glyphosate has been reported in vacuoles of spinach plants (3).
Glyphosate [N-(phosphonomethyl)glycine] is an active ingredient produced by Monsanto Chemical Co. (St. Louis, MO). It is effective for the control of many annual and perennial weeds of both grass and broad-leaved species, as well as woody plants. Since the introduction of glyphosate as a nonselective herbicide (1), extensive studies have been conducted to determine its mode of action. Glyphosate is a potent inhibitor of the shikimate pathway enzyme 5enolpyruvylshikimic acid 3-phosphate synthase (EC 2.5.1.19) (2), which therefore induces a powerful accumulation of shikimate (3) and interferes with aromatic amino acid and phenylpropanoid biosynthesis (4). It has also been reported that glyphosate induces chromosomal aberrations (5), affects respiration (6) and ion absorption (7), inhibits chlorophyll formation (8, 9), and accelerates chlorophyll degradation (8, 10). It also modifies chloroplast ultrastructure 1
To whom correspondence should be addressed. Fax: 33 04 76 51 44 63. E-mail:
[email protected]. 13
0048-3575/97 $25.00 Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.
14
MORIN ET AL.
However, more recently, Gout et al. (20) have shown, by 31P-NMR analysis, that glyphosate is found only in the cytoplasmic compartment of Acer pseudoplatanus cells. In order to continue the investigation of the glyphosate uptake and cellular distribution processes, it was decided to use a cell suspension culture of Catharanthus roseus L., in which the absorption of glyphosate was greatly dependent on the growth conditions and the composition of the culture medium. In this work, the definition of an “uptake medium,” coupled with kinetic data and the use of chemical agents, enabled us to demonstrate that part of the glyphosate uptake may be mediated by an energy-dependent phosphate carrier. MATERIALS AND METHODS
Cell Suspension Cultures Cell suspension cultures of C. roseus (L.) G. Don (cell line C20) were grown in the medium described by Gamborg et al. (21), supplemented with 4.5 mM 2,4-D, 0.28 mM kinetin, and the addition of 2% sucrose as the carbon source (standard medium). In some experiments, a phosphate-free medium was used to enhance the uptake of glyphosate. Cells were subcultured weekly by transferring a 15-ml cell suspension (4 g fresh wt) into 300-ml Erlenmeyer flasks containing 100 ml fresh medium. The cultures were maintained in constant logarithmic growth by subculturing prior to the onset of the stationary phase. Flasks were incubated at 258C on a rotary shaker (100–120 rpm) under low light. Protoplast and Vacuole Isolation Protoplasts from 1- to 4-day-old cultures were prepared by enzymatic digestion, following the procedure described by Marigo et al. (22). For cell wall digestion, 15 g of cells was first rinsed three times with 20 ml of 550 mM sorbitol, adjusted to pH 5.5, and then incubated for 35 min at 368C in the same solution, containing 2% Caylase 345 (Cayla SA, Toulouse, France) and 1% pectolyase Y23 (Seishim Pharmaceutical). The protoplasts were filtered through a 60-mm
nylon cloth and washed three times with the sorbitol solution after sedimenting by centrifugation at 200g, for 1.5 min. They were resuspended in 2 ml of 550 mM sorbitol and centrifuged through a sucrose gradient. After 5 min centrifugation at 200g, protoplasts that were free from cellular debris were recovered from the interphase of the 15 and 5% sucrose layers. Vacuole isolation was performed by osmotic lysis of the plasma membrane and purification by flotation as described by Renaudin et al. (23). Lysis was induced by incubating protoplasts for 30 min at 48C, in a medium containing 200 mM sorbitol, 5 mM EDTA, and 10 mM Hepes–KOH buffer (pH 7.0). The remaining intact vacuoles were suspended in a medium of 350 mM sorbitol, 8.5% Nycodenz, 5 mM EDTA, and 10 mM Hepes–KOH buffer (pH 7.0). After gentle agitation, 1 ml of 550 mM sorbitol was deposited on top of the vacuole suspension. After centrifugation at 160g for 3 min at 48C, the vacuoles were recovered in the upper phase. Following this standard procedure, contamination by intact protoplasts never exceeded 1%. In some experiments, in order to improve the purity of the isolated vacuoles, the vacuolar fraction was purified again by flotation through a silicon layer, according to a procedure described by Martinoia et al. (24). Briefly, this involves pipetting aliquots of 100 ml into a 400-ml microfuge tube. The samples are overlayered with 220 ml silicon oil 550 (d 5 1.068, Rhodorsil silicon, RhoˆnePoulenc) and 50 ml of 550 mM sorbitol are added to the top. After centrifugation (10,000g, 30 s) the upper phase is collected. Protoplasts and vacuole numbers were counted, using a Fuschs Rosenthal cell. Uptake Experiments Measurement of glyphosate uptake was carried out using cells of C. roseus, harvested at the beginning of the stationary phase of growth and transferred, under standard conditions, into a fresh culture medium (0-day-old culture). Measurements were also made of glyphosate uptake
GLYPHOSATE UPTAKE IN Catharanthus roseus CELLS
from a medium without the addition of phosphate salts. Some uptake experiments were performed on 1- to 7-day-old cultures growing in their original medium. For the uptake experiments, 100 ml of cell suspension was incubated with 0.03 mM [14C]glyphosate (1.89 GBq mmol21, Amersham, France) in the presence of various concentrations of glyphosate or of other compounds, as detailed in the figure legends and under Results. After various periods of time, 20-ml aliquots of the suspension were rapidly filtered, in order to measure the intracellular (Ci, in dpm g21) and the extracellular (Ce, in dpm ml21) concentrations of glyphosate, as described previously (25). To take into account any contamination by extracellular labels, reference samples were centrifuged immediately after the addition of [14C]glyphosate. The values thus obtained (zero time) were subtracted from the experimental measurements. Ci/Ce represented the accumulation ratio of glyphosate in the cells. The initial velocity of uptake was determined by measuring the glyphosate amounts taken up by the cells. Except where stated otherwise, data points are the means of three similar samples 6SD. For the cellular distribution studies, 4-day-old cells were incubated for 12 h in presence of 0.3 mM [14C]glyphosate (1.89 Gbq mmol21). The protoplasts and vacuoles were then rapidly isolated as described previously.
15
The decrease of the Pi concentrations in the standard medium was measured by the method of Bencini et al. (26). RESULTS
Uptake of Glyphosate by C. roseus Cells Growing in a Standard Medium Figure 1 represents a typical absorption curve for 0.03 mM [14C]glyphosate. When 7-day-old cells of C. roseus were transferred to a fresh medium, the uptake of glyphosate by the cells increased over the first hours of the experiment, reaching a maximum value after 14 to 18 h, depending on the experiment. A detailed analysis of the uptake shows the existence of a lag period before rapid absorption of glyphosate, which started at 6 h (also see Fig. 2). The maximum accumulation ratio was reached at 18 h (Ci/Ce 5 35, mean value for three different experiments) and then decreased gradually and
Analytical Techniques Glyphosate metabolism was studied by the TLC of cells and medium extracts on cellulose plates with the following eluting system, ethanol:water:NH3:trichloroacetic acid:acetic acid (55:35:2.5:3.5:2, v/v). After autoradiography using Kodirex X-ray film (Kodak), radioactive spots were scraped off and counted for radioactivity by scintillation counting in a 1900 TR Packard counter, after addition of 4 ml of Aqualuma (Amersham). The pH changes of the medium (pHe) were measured, at specific times, using a glass electrode (pH-meter Radiometer Copenhague Model PHM 82).
FIG. 1. Time course change for glyphosate uptake and extracellular pH (pHe) shown by Catharanthus roseus cells growing in a standard medium. 7-day-old cells were transferred to a fresh medium containing 0.03 mM [14C]glyphosate. Intracellular concentration (Ci, nmol g21 fresh wt, n), extracellular concentration (Ce, nmol ml21, V). M, pHe. Values of the accumulation ratio (Ci/Ce) are indicated in parentheses. The results are the mean of three independent experiments (six measurements) 6SD.
16
MORIN ET AL.
fell to values as low as 6.5, after 66 h. It continued to decrease in the last days of culture, reaching a mean value of 1 from the 3-day stage (data not shown). Metabolic studies indicated that, during the first 2 days of uptake, no significant metabolism of [14C]glyphosate occurred, since it represented about 85% of the radioactivity in the cells and 88% in the medium (reference sample 93%). A visible transformation of the herbicide occurred after 3 days of culture (data not shown), aminomethylphosphonic acid being the major metabolite produced. Consequently, the accumulation ratio of radioactivity was used as an indicator for the relative accumulation of [14C]glyphosate, for incubation times not exceeding 60 h. The changes observed in the cellular uptake of glyphosate (Ci) occurred in parallel with opposing changes in extracellular concentrations (Ce). After a decrease in Ce corresponding to the initial absorption of the herbicide, the extracellular concentration of the herbicide increased again after 18 h, coming close to the initial value in the final stages of the experiment. These data indicate, therefore, that the observed decrease in intracellular concentration that occurs between 18 and 60 h corresponds to a net cellular efflux of glyphosate and not to a degradation of the herbicide into 14CO2. They demonstrate also that this decrease can be associated with a glyphosate release from the cells and not just due to some cell death. Moreover, since at the concentration used in this study (0.03 mM) glyphosate has no phytotoxic effects on cell growth (data not shown), it can be assumed that the observed efflux is not related to the action of the herbicide inside the cytoplasm. The uptake of glyphosate is accompanied by a rapid increase of the extracellular pH (Fig. 1). Effect of Growth Conditions on Glyphosate Uptake and Accumulation Figure 2 shows the [14C]glyphosate uptake and its accumulation by 0- to 7-day-old cell cultures growing in their original medium. When the cells were transferred to fresh medium (0day-old cells), there was a lag period of about
FIG. 2. [14C]Glyphosate uptake by 0- to 7-day-old cell culture growing in its original medium, evaluated according to the Ci/Ce ratio. The experimental conditions were similar to those described in the legend to Fig. 1 (0.03 mM [14C]glyphosate). 0- (m), 1- (M), 3- (v), 5- (n), and 7- (V) dayold cell cultures. Values given are the mean of two independent experiments.
6 h in the absorption of glyphosate. After that, the glyphosate entered into the cells rapidly, reaching an accumulation ratio of about 13 at 10 h. The presence of the lag period was not observed for 1-day-old cells, but, after 3 h of incubation, the glyphosate that was taken up was also in turn effluxed into the medium. As shown by the very rapid Ci/Ce decrease, this effect does not originate from cell growth with arrested glyphosate penetration. For 3- to 7-day-old cell cultures, [14C]glyphosate rapidly reached an equilibrium between the cells and the culture medium showing an accumulation ratio that did not exceed a mean value of 1.5. The considerable accumulation of glyphosate found for cells transferred to fresh medium was not dependent on the age of the cells, since an identical distribution pattern, including the existence of a lag period, was also obtained for glyphosate with 3- and 5-day-old cells (data not
GLYPHOSATE UPTAKE IN Catharanthus roseus CELLS
shown). The composition and characteristics of the medium appear, therefore, to play a major role in the specific behavior of glyphosate. Under our experimental growth conditions, our results showed that Pi was almost completely taken up by C. roseus cells, from the standard medium, after 6 h of incubation (Fig. 3). Therefore, in order to try to explain the lag period, and the difference in glyphosate uptake observed between 0- and 1-day-old cells, we investigated the effect of various concentrations of phosphate on glyphosate behavior (Fig. 3). When the medium was deprived of phosphate, the rate of glyphosate uptake was approximatively linear for at least the first 4 h of the experiment. In contrast, in the presence of 1 mM NaH2PO4, the uptake process started after a lag period of 6 h minimum. At this time, the phosphate had
17
FIG. 4. Concentration dependence of [14C]glyphosate uptake as a function of glyphosate concentration in cell suspension of Catharanthus roseus. Since glyphosate uptake occurs linearly during the first hours of the experiment, the uptake assays were carried out for 1 h in the Pi-free medium. Uptake is expressed in nmol g21 fresh wt h21. The results are the mean of two independent experiments. (Inset) Lineweaver–Burk plot. 1/V, inverse of the initial uptake rate; 1/S, inverse of the glyphosate concentrations. The apparent kinetic parameters were as follows: Km 5 25 mM and Vmax 5 12.5 nmol g21 fresh wt h21.
disappeared from the medium. For higher concentrations of sodium phosphate (2.5 and 5 mM), glyphosate uptake was notably reduced. The Dependence of Glyphosate Uptake on Glyphosate Concentration
FIG. 3. Effect of phosphate concentrations on glyphosate uptake. Uptake experiments were carried out in the phosphate-free medium (V), in the standard medium with 1 mM NaH2PO4 (n), and in a medium with 2.5 mM NaH2PO4 (M) or 5 mM NaH2PO4 (v). The reaction was started by the addition of [14C]glyphosate diluted with unlabeled glyphosate to give a final concentration of 3 mM (final sp act in the medium 18 Mbq mmol21). Uptake is expressed in nmol g21 fresh wt. For the standard medium, the decrease of the Pi concentrations (m) was also followed during the glyphosate uptake. The results are the mean of two independent experiments.
The initial rate of glyphosate uptake was measured in a phosphate-free medium, after a 1-h incubation period, for glyphosate concentrations ranging between 0.5 and 500 mM (Fig. 4). The concentration dependence data clearly show the existence of two processes involved in glyphosate uptake; a first, saturable, phase which occurs between 0.5 and 50 mM and a linear phase which occurs up to 500 mM glyphosate. The presence of the saturable phase suggests that part of the glyphosate uptake is carrier mediated. In contrast, the linear phase, for higher concentrations, indicates the existence of glyphosate diffusion across the plasma membrane.
18
MORIN ET AL. TABLE 2 Contribution of the Components of the Gamborg’s Medium on Glyphosate Uptake
Effect of Chemical Agents on Glyphosate Uptake To confirm the involvement of a carrier for lower concentrations, further investigation was made of the action of chemical agents on glyphosate uptake. The direct involvement of a membrane protein in the uptake process was demonstrated using PCMBS,2 a slowly permeant reagent which binds to the thiol group of the proteins. For 0-day-old cells incubated in a phosphate-free medium, addition of 0.5 mM PCMBS suppressed 60% of the cellular uptake (Table 1). The lipophilic carboxylic group reagent, DCCD, was also a potent inhibitor of glyphosate uptake. At a concentration of 0.2 mM, it inhibited the uptake process by 40%. This compound is used as a probe for proton-translocating enzymes, since it modifies a carboxyl group of the glutamyl or aspartyl residues of several proteins involved in H+ conductance. The protonophore CCCP, which acted via changes in membrane permeabilities to H+, also decreases the glyphosate uptake by 60%. Interestingly, PFA, which is well known as a competitive inhibitor of phosphate uptake in animal (27,28) and plant cells (19), also strongly inhibited the absorption of TABLE 1 Inhibitor Sensitivity of [14C]Glyphosate Uptake Glyphosate uptake (% of control)a Substances 0.50 0.20 0.05 5.00
mM mM mM mM
PCMBS DCCD CCCP PFA
0-day-old cellsb
1-day-old cellsb
3-day-old cellsb
39 60 39 31
42 48 37 29
89 92 40 93
a Cells were preincubated for 20 min with the chemical agents before the addition of 3 mM [14C]glyphosate (sp act 18 Mbq mmol21). The uptake was determined after 1 h of incubation. Values are the average of two experiments. b The effect of the various inhibitors were investigated on 0-day-old cells in a Pi-free medium, and 1- and 3-dayold cells in the original standard medium.
2
Abbreviations used: CCCP, m-chloromethoxycarbonyl cyanide phenylhydrazone; DCCD, N,N8-dicyclohexylcarbodiimide; PCMBS, p-choromercuribenzenesulfonic acid; PFA, phosphonoformic acid.
Medium compositiona Pi-free medium Without macronutrients Minus KNO3 Minus (NH4)2SO4 Minus MgSO4 Minus CaCl2 Without FeSO4 Without 2,4-D
Concentrations inb the standard Glyphosate medium uptake (%)c 1 mM 25 mM 1 mM 1 mM 1 mM 100 mM 4.5 mM
100 25 105 102 64 48 22 132
a The experiments were carried out on a Pi-free medium (control) or on the same medium deprived of some of the major components. b Standard medium concentrations of omitted elements. c The transport solution consisted of 3 mM glyphosate (final sp act 18 Mbq mmol21) and the uptake was measured after 1 h of incubation. Mean of two experiments.
glyphosate (70%), which is in agreement with the hypothesized involvement of a phosphate transporter. The same inhibitory effects of the chemical agents used also were found for 1-day-old cells cultured in their standard medium (Table 1). In contrast, for 3-day-old cells, all the protein reagents used were without notable effects on the glyphosate uptake, and only the inhibition exerted by CCCP was still significant. Composition of the Medium and Optimum Conditions of the Glyphosate Uptake Since glyphosate uptake was greatly dependent on the composition and the characteristics of the medium, we investigated the effects of the different elements of the Gamborg’s nutrient solution on this process. In Table 2 are reported the major components which affected the transport activity. Suppression of macronutrients (25 mM KNO3, 1 mM (NH4)2SO4, 1 mM MgSO4, 1 mM CaCl2) in the final medium decreases the uptake by about 75% with a specific effect of MgSO4 and CaCl2, respectively, 35 and 50% of inhibition when these salts were omitted. At the opposite, when the medium was deprived of 2,4D, the glyphosate uptake was stimulated by 32%
GLYPHOSATE UPTAKE IN Catharanthus roseus CELLS
indicating an inhibitory effect of the synthetic auxin. Finally, among the constituents which control the transport activity with the greater efficiency (MgSO4 and CaCl2), one of the most important was FeSO4. The absence of FeSO4 in the medium decreases the transport activity by 78%. Intracellular Location of Glyphosate The vacuole yield from the C. roseus cell suspension was found to be dependent on the age of the culture. The best results were obtained with cells 3–6 days after inoculation, during the logarithmic growth phase. Average yields were around 39% for 4-day-old cells, but were only 10% for vacuoles isolated from 1-day-old cells. Contamination was also greater in vacuoles isolated from young cells. For this reason, glyphosate distribution between cytoplasm and vacuole was investigated with 3- to 4-day-old cells. Since it would appear that one protoplast contains and releases one vacuole (23), glyphosate distribution was calculated on the basis of protoplast and vacuole numbers. Possible interference associated with the purity of the vacuolar fraction was also taken into consideration by comparing two protocols for vacuole purification. The data obtained showed that, whatever the mode of purification, 40 to 45% of the herbicide was found to be located in the vacuoles (Table 3). Assuming that the vacuoles occupy 90% of TABLE 3 Compartmentation of Glyphosate in Protoplasts and Vacuoles Isolated from Catharanthus roseus Cellsa Glyphosate 6
Protoplast Vacuoleb Vacuolec a
b
pmol in 10 of
% in vacuole
4.82 6 1.05 2.17 6 0.6d 1.92 6 0.5d
— 45 40
d
4-day-old cells were incubated for 12 h in presence of 0.3 mM [14C]glyphosate. b The protoplasts and the vacuoles were rapidly isolated according the standard procedure. c The vacuoles were purified in a subsequent step by flotation through a silicon layer. d Data represent the average value of four independent experiments.
19
the intracellular space, the cytoplasmic and the vacuolar glyphosate concentrations were calculated to be 1.8 and 0.13 mM, respectively, for the external concentration of 0.3 mM used in these experiments. DISCUSSION
In C. roseus cells, the study of glyphosate uptake clearly shows that after a lag period of about 6 h, the herbicide is accumulated against its electrochemical potential, this effect being particularly pronounced after the cells are subcultured in a fresh medium. Due to the transmembrane potential of plant cells, when equilibrium is reached the intracellular concentration of an anion, such as glyphosate, should be several times lower than the external concentration. By using the Nernst equation, and taking into account a mean value of 258 mV for the electrical potential in C. roseus cells (29), a theoretical distribution ratio of 0.1 can be calculated for the monoanionic species, which is the most permeant and preponderant form of glyphosate found at physiological pH. Therefore, if we consider the experimental values found for the glyphosate accumulation ratio, which are between 1.5 and 30, it may be concluded that energy is required to accumulate the herbicide in these cells. A possible explanation for the cellular accumulation of glyphosate is vacuolar storage of the herbicide after it enters into the cells. Compartmental studies to test this hypothesis were carried out, but it was found that, although glyphosate was detected in the vacuole, it cannot accumulate in this compartment. These data must be considered with some caution since it is not excluded that a glyphosate redistribution occurs during the preparative procedures of protoplasts and vacuoles. However, by using protoplasts isolated at different times of the enzymatic digestion of the cells (up to 2 h, data not shown) we demonstrated that the [14C]glyphosate accumulated initially in the cells was not removed significantly outside the protoplasts. The same determinations were not performed with vacuoles but, by considering the similar amounts of [14C]glyphosate found in vacuoles isolated according the standard procedure or purified in
20
MORIN ET AL.
a subsequent way by flotation through a silicon layer (Table 3), it may be assumed that, if a vacuolar remobilization takes place during the isolation step, this process was relatively slow. The fact that glyphosate is not accumulated in the vacuoles confirms the observations of Schultz et al. (3) which demonstrated that, in tomato and spinach plants, the greater part of the [14C]glyphosate imported into the leaf via the phloem was present in the cytosolic fraction of the mesophyll cells. Several observations support the idea that glyphosate transport in C. roseus cells is, in part, mediated by a carrier. Concentration dependence studies showed that glyphosate absorption exhibited a saturable phase at low glyphosate concentrations, between 0.5 and 50 mM, superimposed by a linear diffusion phase at higher concentrations (up to 500 mM). Moreover, the possible involvement of a carrier for the lower concentrations was confirmed by the existence of a lag period in glyphosate absorption when the cells were incubated with sodium phosphate, which suggests that Pi acts as a competitive inhibitor of glyphosate uptake. Competitive inhibition by phosphate has been described earlier by other authors with other plant material (19). The occurrence of the carrier component in C. roseus was also confirmed by the inhibitory effect of PFA, a powerful inhibitor of phosphate transport in animal cells (27,28) which inhibits competitively the glyphosate uptake in broad bean (19), but also by the similar action of two chemical reagents for proteins, PCMBS and DCCD. These data are in agreement with recent observations of Denis and Delrot (19) who, using isolated protoplasts of broad bean leaves, have demonstrated that glyphosate can be taken up via a phosphate transporter at the plasma membrane level. The data were also consistent with other indirect observations that showed some mutual interaction between phosphate and glyphosate uptake (7,30) Concerning the energization processes, the study of the [14C]glyphosate uptake, during the cell growth, led us to postulate the involvement of two energy-dependent uptakes in C. roseus: a basal process, present at the different stages
of the cell growth, superimposed by an active phase during the first hours after cell subculture. As shown by their characteristics, these mechanisms appear to be distinct. The initial process proceeds during the first 18 h, in parallel with an alkalinization of the medium (Fig. 1), and reaches an accumulation ratio of 30, thus necessitating a high motive force. If the cells were subcultured in a phosphate-free medium buffered to pH 7.8, conditions which abolish the pH component of the proton motive force, the glyphosate uptake was largely decreased (data not shown). Glyphosate uptake was also inhibited by the protonophore CCCP and by DCCD, an inhibitor of the proton-translocating enzymes. In conjunction with our data showing the involvement of a carrier mechanism for the uptake of low glyphosate concentrations, these observations, altogether, indicate that glyphosate is cotransported with H+ along a proton motive force gradient generated by the plasmalemma H+-ATPase. A similar mechanism has been also reported for Pi, in C. roseus cells, by Sakano et al. (31). In the basal process, in contrast, glyphosate uptake is little affected by the presence of the protein chemical reagents PCMBS and DCCD, nor by the phosphate competitive inhibitor PFA. These data indicate the existence of a passive nonfacilitated mechanism for the cellular absorption of glyphosate. However, glyphosate is still accumulated by the cells with a mean accumulation ratio value of 1 which suggests also the involvement of an energydependent process. A passive mechanism explaining cellular accumulation for ionic species is the weak acid mechanism (32). This model predicts a higher permeability toward the less charged species, the monoanionic form for glyphosate. Net charge of the molecule would increase (21 to 22 for glyphosate) as it moved from cell wall space at a slightly acidic pH (5.5) into the cytosolic compartment of the cells at pH about 7.5 that increases the accumulation ratio. In accordance with this theory, glyphosate uptake and accumulation would be expected to decrease after the action of the protonophore CCCP which abolishes the pH gradient across the plamalemma. Since an inhibitory effect of
GLYPHOSATE UPTAKE IN Catharanthus roseus CELLS
CCCP was also observed on the glyphosate uptake for the third day of culture (Table 3), it may be assumed that, in addition to the existence of a glyphosate diffusion across the plasma membrane, the weak acid mechanism could play also a part in the cellular transport and accumulation of this compound. The glyphosate accumulation, which occurs after the cell suspension was transferred to a fresh medium, decreases in the last days of culture. As shown by the high proton pumping activity observed in C. roseus cells (33), it may be suggested that the limitative factor, in the low glyphosate accumulation, was not the generation of a proton motive force but rather the low activity of the transporter itself. This hypothesis was confirmed by the low inhibitory effects of the chemical agents PCMB, DCCD, and PFA on the basal glyphosate uptake (Table 2). It was also supported by the apparent decrease of the transporter activity from 1 day (Fig. 2). The fact that the transporter activity decreases, after 1 day of culture, would be one of the reasons explaining the remobilization of the herbicide from the cells. Indeed, if the large accumulation of glyphosate which occurs during the first hours of the culture is not maintained by a high transport activity, the anionic species taken up by the cells will be, in turn, effluxed into the medium. These observations raised the problem of the regulation of the transport activity. So, to obtain more information about the requirements of the transporter, we investigated, in the last step of this work, the effects of the major components of the Gamborg’s nutrient solution. Beyond phosphate which acts as a competitive inhibitor of the glyphosate uptake, the major elements involved in the regulation of the transporter were the synthetic hormone 2,4-D, two salts (CaCl2 and MgSO4), and FeSO4. Concerning the synthetic auxin, the same inhibitory effect was reported by Okihara et al. (34) on the Pi transport. Auxins are structurally very similar to some inhibitors of the Pi transporter such as furosemide, in having a hydrophobic molecular structure and a carboxylic group. The inhibitory effect of 2,4-D, therefore, was also consistent with the involvement of a Pi transporter in the glyphosate uptake. Ca2+
21
and Mg2+, on the other hand, have been shown to stimulate proton pumping in C. roseus cells (33). In this way, these two species could be involved in the glyphosate uptake through an activation of the energization process. Another noteworthy point is the requirement of iron for the optimum activity of glyphosate transport. Iron is an important component of enzymes involved in redox reactions but there is no information on its effect on Pi transport and more specifically on glyphosate uptake. Finally, the formation of chelates which has been described for glyphosate and other cationic species (35) may be another possible alternative to explain the stimulating effect of Ca2+, Mg2+, and Fe2+. In conclusion, this study performed with cells of C. roseus indicated that, for concentrations lower than 50 mM, glyphosate absorption is, in part, mediated by a phosphate transporter. The role played by this transporter in controlling the phloem mobility of glyphosate and the specific implication of the ionic effectors are now under investigations. ACKNOWLEDGMENTS The authors thank Mrs. Anne Marie Viaud, Mr. Fabien Anthelme, and Mr. Jean Pierre Guichard for skilful technical assistance and Mrs. Willison for English corrections. This work was supported by a financial assistance from the Conseil Ge´ne´ral de Savoie and from the Conseil Re´gional RhoˆneAlpes, France (Programme fe´de´radeur Environnement, XIe`me Plan Etat Re´gion).
REFERENCES 1. D. D. Baird, R. P. Upchurch, R. P. Homesley, and J. E. Franz, Introduction of a new broad spectrum postmergence herbicide class with utility for herbaceous perennial weed control, Proc. North. Cent. Weed Control. Conf. 26, 64 (1971). 2. H. C. Steinru¨cken and N. Amrhein, The herbicide glyphosate is a potent inhibitor of 5-enolpyruvylshikimic acid-3-phosphate synthase, Biochem. Biophys. Res. Commun. 94, 1207 (1980). 3. A. Schultz, T. Munder, H. Holla¨nder-Czytko, and N. Amrhein, Glyphosate transport and early effects on shikimate metabolism and its compartmentation in sink leaves of tomato and spinach plants, Z. Naturforsch. 45, 529 (1990). 4. H. Holla¨nder-Czytko and N. Amrhein, The site of the inhibition of the shikimate pathway by glyphosate. I. Inhibition by glyphosate of phenylpropanoid synthesis
22
5.
6.
7.
8.
9. 10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
MORIN ET AL. in buckwheat (Fagopyrum esculentum Moench), Plant Physiol. 66, 823 (1980). W. S. Boyle and J. O. Evans, Effect of glyphosate and etephon on meiotic chromosomes, J. Hered. 65, 250 (1974). P. Sprankle, W. F. Meggitt, and D. Penner, Absorption, action and translocation of glyphosate, Weed Sci. 23, 235 (1975). B. J. Brecke and W. B. Duke, Effect of glyphosate on intact bean plants (Phaseolus vulgaris L.) and isolated cells, Plant Physiol. 66, 656 (1980). T. T. Lee, Effect of glyphosate on synthesis and degradation of chlorophyll in soybean and tobacco cells, Weed Res. 21, 161 (1981). L. M. Kitchen and W. W. Witt, Inhibition of chlorophyll accumulation by glyphosate, Weed Sci. 29, 513 (1981). B. E. Abu-Irmaileh and L. S. Jordan, Some aspects of glyphosate action in purple nutsedge (Cyperus rotundus), Weed Sci. 26, 700 (1978). W. E. Campbell, J. O. Evans, and S. C. Reed, Effects of glyphosate on chloroplast ultrastructure of quackgrass mesophyll cells, Weed Sci. 24, 22 (1976). T. T. Lee, Mode of action of glyphosate in relation to metabolism of indole-3-acetic acid, Physiol. Plant. 54, 289 (1982). T. T. Lee, T. Dumas, and J. J. Jevnikar, Comparison of the effects of glyphosate and related compounds on indole-3-acetic acid metabolism and ethylene production in Tobacco callus, Pestic. Biochem. Physiol. 20, 354 (1983). M. E. Schultz and O. C. Burnside, Absorption, translocation, and metabolism of 2,4-D and glyphosate in hemp dogbane (Apocynum cannabinum), Weed Sci. 28, 13 (1980). M. E. Schultz and N. Amrhein, Glyphosate induced accumulation of shikimic acid in tomato plants in relation to the translocation of glyphosate, Plant Physiol. 75, S-49 (1984). [abstract 273]. J. L. Honegger, J. M. Brooks, E. J. Anderson, and C. A. Porter, Glyphosate transport in plants, in “Phloem Transport” (J. Cronshaw, W. J. Lucas, and R. T. Giaquinta, Eds.), p. 601, A. R. Liss, New York, 1986. J. A. Gougler and D. R. Geiger, Uptake and distribution of N-(phosphonomethyl) glycine in sugar beet plants, Plant Physiol. 68, 668 (1981). H. El Ibaoui, S. Delrot, J. Besson, and J. L. Bonnemain, Uptake and release of a phloem-mobile (glyphosate) and of a non phloem-mobile (iprodione) xenobiotic by broadbean leaf tissues, Physiol. Ve´g. 24, 431 (1986). M. H. Denis and S. Delrot, Carrier-mediated uptake of glyphosate in broad bean (Vicia faba) via a phosphate transporter, Physiol. Plant. 87, 569 (1993). E. Gout, R. Bligny, P. Genix, M. Tissut, and R. Douce, Effect of glyphosate on plant cell metabolism: 31P and 13 C NMR studies, Biochimie 74, 875 (1992).
21. D. L. Gamborg, R. A. Miller, and M. Ojima, Nutrient requirements of suspension cultures of soybean root cells, Exp. Cell Res. 50, 151 (1968). 22. G. Marigo, H. Bouyssou, and L. Laborie, Evidence for a malate transport into vacuoles isolated from Catharanthus roseus cells, Bot. Acta 101, 187 (1988). 23. J. P. Renaudin, S. C. Brown, H. Barbier-Brygoo, and J. Guern, Quantitative characterization of protoplasts and vacuoles from suspension cultured cells of Catharanthus roseus, Physiol. Plant. 68, 695 (1987). 24. E. Martinoia, U. I. Flu¨gge, G. Kaiser, U. Hebert, and H. W. Held, Energy-dependent uptake of malate into vacuoles isolated from barley mesophyll-protoplast, Biochim. Biophis. Acta 806, 311 (1985). 25. G. Marigo, Y. M. Delorme, U. Lu¨ttge, and A. M. Boudet, Roˆle de l’acide malique dans la re´gulation du pH vacuolaire dans des cellules de Catharanthus roseus cultive´es in vitro, Physiol. Ve´g. 21, 1135 (1983). 26. D. A. Bencini, J. R. Wild, and O’Donovan, Linear on step assay gor the determination of orthophosphate, Anal. Biochem. 132, 254 (1983). 27. R. Beliveau, M. Jette´, M. Demeule, M. Potier, J. Lee, and H. S. Tenenhouse, Different molecular sizes for Na+-dependent phophonoformic acid binding and phosphate transport in renal brush border membrane vesicles, Biochim. Biophys. Acta 1028, 110 (1990). 28. H. Al-Mahrouq and S. A. Kempson, Photoaffinity labelling of brush-border membrane proteins which bind phosphonoformic acid, J. Biol. Chem. 266, 1422 (1991). 29. J. P. Rona, D. Cornel, A. M. Pennarum, M. Monestiez, M. Convert, and G. Marigo, Energetics of OH2 or H+ dependent nitrate uptake by Catharanthus roseus cells: Electrophysiological effects. Bioelectrochem. Bioenerg. 25, 213 (1991). 30. T. L. Mervosh and N. E. Balke, Effect of calcium, magnesium, and phosphate on glyphosate absorption by cultured plant cells, Weed Sci. 39, 347 (1991). 31. K. Sakano, H. Yasaki, and T. Mimura, Cytoplasmic acidification induced by inorganic phosphate uptake in suspension cultured Catharanthus roseus cells, Plant Physiol. 99, 672 (1992) 32. C. E. Crips, Insecticides, in “Proceedings of the 2nd International IUPAC Congress on Pestic Chemistry,” Vol. 1, p. 211, Tel Aviv, Israel, 1972. 33. M. Belkoura, R. Ranjeva, and G. Marigo, Cations stimulate proton pumping in Catharanthus roseus cells: Implication of a redox system? Plant Cell Environ. 9, 653 (1986). 34. K. Okihara, T. Mimura, S. Kiyota, and K. Sakano, Furosemide: A specific inhibitor of Pi transport across the plasma membrane of plant cells, Plant Cell Physiol. 36, 53 (1995). 35. H. E. L. Madsen, H. H. Christensen, and C. GottliebPetersen, Acta Chem. Scand. 32, 79 (1978).
