Transport of lactate in Plasmodium falciparum-infected human erythrocytes

JOURNAL OF CELLULAR PHYSIOLOGY 149:469-476 (1991)

Transport of lactate in PIasmodium fakiparum-Infected Human Erythrocytes

JAMIL KANAANI AND HAGAI GINSBURG* Department of Biological Chemistry, Institute of Life Sciences, The Hebrew University of

lerusalem, lerusalem 9 1904, Israel

The intraerythrocytic human malarial parasite Plasmodium falciparum produces lactate at a rate that exceeds the maximal capacity of the normal red cell membrane to transport lactate. In order to establish how the infected cell removes this excess lactate, the transport of lactate across the host cell and the parasite membranes has been investigated. Transport of radiolabeled L-lactate across the host cell membrane was shown to increase ca. 600-fold compared to uninfected erythrocytes. It showed no saturation with [L-lactate] and was inhibited by inhibitors of the monocarboxylate carrier, cinnamic acid derivatives (CADs), but not by the SH-reagent p-chloromercuriphenyl sulfonic acid (PCMBS). These results suggest that L-lactate is translocated through CAD-inhibitable new pathways induced in the host cell membrane by parasite activity, probably by diffusion of the acid form and through a modified native monocarboxylate:H+ symporter. Continuous monitoring of extracellular pH changes occurring upon suspension of infected cells in isoosmotic Na-lactate solutions indicates that part of the lactate egress is mediated by anionic exchange through the constitutive, but modified, anion exchanger. The transport of L-lactate across the parasite membrane is rapid, nonsaturating, and insensitive to either CADs or PCMBS, or to the presence of pyruvate. L-lactate uptake increased transiently when external pH was lowered and decreased when ApH was dissipated by the protonophore carbonylcyanide m-chlorophenyl hydrazone (CCCP). These results are compatible with L-lactate crossing the parasite membrane either as the undissociated acid or by means of a novel type of lactate-/H+ symport.

Asexual malaria parasites propagating inside the erythrocytes of their vertebrate host are regarded as homolactate fermentors (Scheibel et al., 1979; Sher- man, 1979). Malaria-infected erythrocytes consume large amounts of glucose (Roth et al., 1982) and produce vast amounts of lactic acid (Pfaller et al., 1982), though less than the predicted stoichiometric quantities (Scheibel et al., 1979). The excessive oxidation of glucose has been alluded to activation of host cell glycolysis (Roth, 1990). However, a simple consider- ation of the maximal activity of rate-limiting glycolytic enzymes in normal erythrocytes (Jacobash et al., 1974) indicates that glycolysis can be accelerated a t most tenfold, compared to a 40-100-fold larger lactate output in malaria-infected cells. Hence, excessive lactate production in infected cells must originate from the parasite's glycolytic activity. The latter is undoubtedly supported by the tremendous increase in the specific activity of glycolytic enzymes in parasites, which is four to 520 times higher than that of their counterparts in the host erythrocyte (Vander Jagt e t al., 1990). The vast amounts of lactic acid produced by the parasite must egress from the infected cell in order to prevent acidification of the parasite's cytoplasm and to permit the regeneration of NAD. On its way out of the cell, lactic acid must cross the parasite cell membrane, the parasitophorous membrane, and the host cell membrane. Transport of lactate through the normal eryth-

0 1991 WILEY-LISS. INC.

rocyte membrane occurs predominantly by lactate/ proton cotransport mediated by the monocarboxylate carrier, some by simple diffusion as lactic acid and some through the Cl-/HCO; exchanger (Halestrap, 1976; Deuticke, 1982). The maximal lactate transport capacity of normal RBC is ca. 1.5 pmoles/lOs cells/hr (Fish- bein et al., 1988), while trophozoite-infected cells can produce up to 2.1 pmoles/lO' cells/hr (Pfaller et al., 1982; Vander Jagt et al., 1990). Hence, the parasite must in some way modify the permeability of its host cell to lactate so as not to be submerged in its own waste product. Parasites are demonstrably capable of inducing drastic alterations in the permeability of their host cell membrane to carbohydrates, amino acids, anions, and cations (reviewed by Ginsburg, 1990, and Ca- bantchik, 1990). They do so by inducing new permeability pathways or by altering the constitutive transport agencies of the host cell.

In the present work the transport of lactate across the host cell and the parasite membrane has been investigated. The capacity of both membranes to transport lactate was found to be very large, as needed. Some characteristics of lactate transport across those membranes have been investigated.

Received March 25, 1991; accepted June 28, 1991. *To whom reprint requestsicorrespondence should be addressed.

470 KANAANI AND GINSBURG

MATERIALS AND METHODS Parasite cultures

The FCR, strain of Plasmodium falciparum was cultivated in either 0'- or A+-washed human erythrocytes a t 37°C in RPMI-1640 (GIBCO) medium supple- mented with 10% (viv) 0' or A+ heat-inactivated human plasma, 25 mM Hepes (N-2-hydroxyethyl pip-

,d erazine-N-2-ethane sulfonic acid), 32 mM NaHCO and 10 mM glucose. The growth medium in 150 cm culture flasks was replaced daily followed by gassing with a mixture of 90% N,, 5% CO,, and 5% 0,.

Erythrocytes harboring mature parasite stages (trophozoites and schizonts) were concentrated from culture using the gelatin flotation method (Jensen, 1978). After separation, cells were washed twice in wash medium (i.e., growth medium without plasma) and allowed to recover for 2 h r in culture conditions. Cells were counted in a cell counter (Analys Instrument, Stockholm, Sweden) and the percentage of infected cells (parasitemia) was determined by microscopic in- spection of Giemsa-stained thin blood smears.

Permeabilization of infected cells was achieved by incubating the cell suspension (2.5% hematocrit in wash medium) with Sendai virions (20-50 pg protein/ ml, equivalent to 400-600 hemagglutination units) for 6 min at 37°C and centrifugation for 30 sec in an Eppendorf microfuge (Kanaani and Ginsburg, 1988, 1989).

Influx of radiolabeled substrates Normal erythrocytes or cell preparations containing

more than 80% trophozoite-infected cells were used. Prior to influx assay, the cells were suspended a t 10% hematocrit in wash medium or in PBS. The cells were preincubated €or 5-10 min with 0.5 mM 4,4'-dinitrostilbene-2-2'-disuIfonic acid (DNDS) in the presence or absence of the indicated inhibitor. Influx was initiated by the rapid mixing of an equivalent volume of the same medium containing 0.5 mM DNDS, cold and radioactive substrate, and the indicated inhibitors (loading solution). At different time intervals, tripli- cates of 50 pl of the cell suspension were overlayed on top of 100 pl of ice-cold dibutylphthalate in a 0.4-ml polyethylene microcentrifuge tube and centrifuged for 15 sec in a Beckman microfuge. Zero time samples were taken as follows. The dibutylphthalate in the tube was overlaid with 25 p1 of ice-cold loading solution and the tube was placed horizontally in the microfuge. Cell suspension (25 pl) was carefully pipetted onto the tube wall, avoiding mixing with the loading solution. The centrifuge was then activated, driving the cells through the loading solution. The dwelling time in presence of radiolabel was about 15 msec (H. Ginsburg, unpub- lished results). After centrifugation the tip of the tube containing the cell pellet was cut off and placed in a tube containing 0.5 ml distilled H,O to lyse the cells. Residual dibutylphthalate was removed by centrifugation of the lysate for 2 min. An aliquot of 475 ~1 of the lysate was mixed with 25 p1 of 100% (w/v) trichloro- acetic acid (TCA), centrifuged, and 400 pl of the clear supernatant was taken for radioactivity counting. The remaining 25 pl lysate was diluted 1:4 with distilled water in 96-well plate and hemoglobin was determined

from its absorbance at 405 nm using Bio-Tek ELISA reader. Results in DPMiabsorbance were converted to mmoles/liter cell water, using specific activity of the substrate, the related number of cells obtained from the hemoglobin absorbance, and the fractional water volume. The fractional water volumes used for noninfected, infected, and virus-treated cells were taken as 0.67, 0.48, and 0.32, respectively.

Measurements of extracellular pH changes A sample of 50 p1 of normal or trophozoite-infected

erythrocytes was washed in 1 ml of ice-cold isotonic Na-citrate solution (pH 7.51, resuspended in 100 pl of the same medium, and rapidly injected into a closed chamber containing 2 ml C1-free isotonic unbuffered solution of either Na-lactate or Na-hexanoate (pH 7.4, 37°C). pH changes were monitored by means of a combined glass electrode connected to a pH meter (Corning, Model 240) and recorder (Yokagawa, Model 3077).

Chemicals L( + )lactate, a-Fluorocinnamate (a-FC), n-hexanoic

acid, pyruvic acid Type I1 (Na salt), carbonylcyanide m-chlorophenylhydrazone (CCCP), p-chloromercuriphenyl sulfonic acid (PCMBS), were purchased from Sigma Chemical Co. (St. Louis, MO). a-Cyano-P-(l- phenylindol-3-yl) acrylate (UK-5099) was obtained from Pfizer Central Research (Sandwich, U.K.). 4,4'- dinitrostilbene-2-2'-disulfonic acid (DNDS) was from ICN Pharmaceuticals (NY). L-['4C(U)llactic acid (177.3 mciimmole) was obtained from the Radiochemical Cen- tre (Amersham, U.K.).

RESULTS Transport of lactate across the membrane of

normal and infected erythrocytes We have previously shown that the transport of

lactate across the membrane of malaria-infected erythrocytes is substantially larger than that across the membrane of normal red cells (Kanaani and Ginsburg, 1991). However, this phenomenon has been demon- strated using the isoosmotic lysis technique (Ginsburg et al., 1983) where ammonium lactate has been used as the lysing solute. In order to assess whether this phenomenon also occurs at physiological lactate concentrations, the uptake of radiolabeled lactate from a medium containing 1-50 mM L-lactate has been tested as a function of time and concentration. All experiments have been performed in the presence of 0.5 mM DNDS to block transport of lactate through the C1-i HC0,- system (Halestrap, 1976). The time-dependence of L-lactate uptake into normal erythrocytes is shown in Figure 1 and that into cells infected with the trophozoite stage of P. falciparum is displayed in Figure 2. Since the time-course of L-lactate uptake into both normal and infected erythrocytes obeyed first order kinetics (e.g., [cell lactate1 = Limit . (1 - exp(-k . t)), where Limit is the equilibrium level, k is the rate constant, and t is the time of sampling), it is possible to calculate the initial rate of uptake from the product Limit . k. At 1.5 mM extracellular lactate, the initial rate of L-lactate uptake into normal cells a t

471 LACTATE TRANSPORT IN MALARIA-INFECTED ERYTHROCYTES

I /

0.1 0 5 10 15 20 25 30

TIME (min)

Fig. 1. Effect of UK-5099 on the uptake of L-lactate into normal erythrocytes. After 24 hr in culture conditions, cells were preincubated for 5 min at room temperature (RT) in wash medium containing 0.5 mM DNDS in the presence (solid circles) or absence (open circles) of 0.1 mM UK-5099. Influx was initiated by the rapid mixing of equal volumes of cell suspension and of wash medium containing 0.5 mM DNDS, 3 mM L-lactate (0.8 pCiiml) in the presence or absence of 0.1 mM UK-5099. Samples were removed at various times of incubation at RT and the uptake of lactate was terminated by rapid centrifugation through a dibutylphthalate cushion as described in Materials and Methods. Experimental data were analysed by least square fitting to first order kinetics and the lines were drawn accordingly, using the following parameters: Control: k = 0.29 f 0.03 min-', Limit = 0.62 i 0.02 mmoliliter cell water; UK-5099: k = 0.057 ? 0.006, Lim- it = 0.46 r 0.02.

room temperature (RT) is 0.18 mmole/liter cell water/ min, and that into infected cells at O°C (ice) is 10.2 mmole/liter cell waterimin. Assuming that the temperature-dependence of L-lactate transport in infected cells is similar to that in normal cells (e.g., a 36-fold increase from 0°C to RT, Deuticke et al., 1982), the lactate permeability of infected cells is ca. 600-fold larger than that of normal cells.

Cinnamic acid derivatives are well characterized inhibitors of the monocarboxylate transport system of the normal erythrocyte membrane (Halestrap, 1976). Comparison of the initial rates of lactate uptake * inhibitor allows the evaluation of the inhibitory effect. Thus, 0.1 mM of UK-5099 caused 86% inhibition of L-lactate uptake into normal erythrocytes and 90% into infected cells (Kanaani and Ginsburg, 1991), while a-FC inhibited uptake into infected cells by 93%. These values should be compared to only 25% inhibition by UK-5099 using the isoosmotic lysis technique (Kanaani and Ginsburg, 1991).

Concentration-dependence of L-lactate uptake into infected cells

While monocarboxylate carrier-mediated uptake of L-lactate into normal erythrocytes is saturable (Deu- ticke, 1982; Halestrap, 1976), the uptake of lactate into infected cells is not. Results displayed in Figure 3 clearly show that the initial rate is linear with medium L-lactate concentration. The same linear relationship was observed for up to 50 mM (data not shown). Three millimolar of a-FC were equally inhibitory to uptake

6.0 I I

0 2 4 6 8 10 12 14 16

TIME (min)

Fig. 2. Effect of a-FC on the uptake of L-lactate into trophozoite- infected erythrocytes. Trophozoite-infected erythrocytes from gelatin enriched preparation were preincubated on ice for 10 rnin in wash medium containing 0.5 mM DNDS in the presence (solid circles) or absence (open circles) of 3 mM a-FC. Lactate uptake was measured and analyzed as described in Figure 1 except that incubation was done on ice. Control: k = 2.13 t 0.04, Limit = 4.79 f 0.03; a-FC: k = 0.24 2 0.07, Limit = 3.33 i- 0.34.

0 2 4 6 8 10

[LACTATE] (mM)

Fig. 3. Concentration-dependence of L-lactate transport into trophozoite-infected erythrocytes. Trophozoite-infected erythrocytes from gelatin enriched preparation were preincubated for 5 min on ice in wash medium containing 0.5 mM DNDS in the presence (solid circles) or absence (open circles) of 3 mM a-FC. Lactate uptake during 30 sec was measured as described in Figure 2. Results were best fitted by linear regression, yielding the following parameters: Control, slope = 1.01 i 0.06; a-FC, slope = 0.18 i 0.02.

irrespective of [L-lactate] up to 10 mM. This observation is compatible with a Ki = 0.1 mM for this inhibitor and a very high K, for L-lactate. At [L-lactate1 higher than 10 mM, the efficacy of inhibition decreased, as would be expected from the competitive inhibitory nature of a-FC (data not shown).

Effect of PCMBS on L-lactate transport PCMBS is a known covalent inhibitor of L-lactate

transport through the monocarboxylate carrier in hu-


2.5

2.0 1.5 I 1.0

0.5

0.0 0 5 10 15 20 25 30

TIME (min)

Fig. 4. Effect of PCMBS on the uptake of L-lactate into normal erythrocytes, intact, and virus-treated trophozoite-infected erythrocytes. Normal erythrocytes were in culture conditions for 2 hr. Trophozoite-infected erythrocytes were separated from culture by gelatin flotation method. Normal erythrocytes (circles), intact (trian- gles), and virus-treated (squares) trophozoite-infected erythrocytes were suspended to 20% hematocrit in PBS containing 50 mM sucrose and 0.5 mM DNDS in the presence (solid symbols) or absence (open symbols) of 0.5 mM PCMBS. Influx was initiated by the rapid mixing of equal volumes of PBS containing 50 mM sucrose, 0.5 mM DNDS, and 2 mM L-lactate (1.6 pCiim1) in the presence or absence of 0.5 mM PCMBS. Infected and virus-treated cells were incubated on ice and normal erythrocytes a t RT. Lactate uptake was measured and analysed as described in Figure 1. Parameters describing the drawn lines: Uninfected cells--Control: k = 0.18 i 0.03, Limit = 0.77 -t 0.05; PC- MBS: k = 0.1 i 0.03, Limit = 0.3 i 0.03. Infected cells-Control: k = 2.36 ? 0.03, Limit = 0.9 i 0.03; PCMBS: k = 2.17 -t 0.2, Lim- it = 0.74 ? 0.02. Virus-treated-Control: k = 3.45 ? 1.08, Lim- it = 2.45 ? 0.08; PCMBS: k = 3.51 i 1.08, Limit = 2.41 -t 0.08.

man erythrocyte (Deuticke et al., 1978). Since lactate transport in infected cells was found to be extensively accelerated in malaria-infected cells, it was deemed interesting to evaluate the susceptibility of L-lactate transport to the SH-modifying agent. Results shown in Figure 4 substantiate PCMBS inhibition in normal cells by 80%. However, in infected cells the presence of PCMBS caused a mere 24% inhibition. This result, in conjunction with the overall acceleration of lactate flux, may imply that the native monocarboxylate carrier has been modified by the intracellular parasite, and, in addition, that new pathways for lactate translocation not sensitive to PCMBS are induced by the parasite in the host cell membrane. These pathways are not inhibited by PCMBS (Ginsburg et al., 1983).

Extracellular pH changes due to lactate uptake When normal erythrocytes are suspended in isoos-

motic Cl--free lactate solutions, time-dependent pH changes are observed in the extracellular medium. These changes have been interpreted (Deuticke, 1971; Aubert and Motais, 1975) to result from the following processes. First there is a rapid exchange of intracellular C1- for extracellular OH- that results in medium acidification and disturbance of the Donnan ratio. Thereafter, lactate anion exchanges with intracellular C1-, thus restoring the Donnan equilibrium, OH- is

PHe 4 1 C 8 o h

7.4-1- ....

D

I

JI

1 rnin ““1 ” Fig. 5. Extracellular pH changes after addition of normal and trophozoite-infected erythrocytes to isotonic solutions of Na acetate or Na hexanoate. Normal erythrocytes (A,C) and trophozoite-infected erythrocytes (B,D), separated from culture by gelatin flotation method to a 90% parasitemia, were washed and suspended to a 5% hematocrit in wash medium in the absence (-1 or presence of either 0.6 mM (- - -1 or 5 mM ( . . acetazolamide. The cells were preincubated with acetazolamide for 10 min at 37°C. Aliquots of 1 ml of these cell suspensions were centrifuged, washed with ice-cold isotonic Na citrate (pH 7.51, and the pellet resuspended with 100 pl of the same solution. This cell suspension was added to 2 ml of C1-free isotonic Na-lactate (A,B) or Na-hexanoate (C,D) (pH 7.4, 37°C) in the presence or absence of acetazolamide, and external pH changes were monitored by a glass electrode connected with a pH meter and recorder.

taken up, and pH returns to its equilibrium level. The translocation of OH- is essentially mediated by the transport of HCO,, and the concentration of the latter depends on the activity of carbonic anhydrase. There- fore, inhibition of enzymatic activity by acetazolamide should reduce considerably both the extent of initial acidification and the rate of realkalinization. Results depicted in Figure 5A show that noninfected erythrocytes behave indeed as expected, and the effect of acetazolamide is dose-dependent. When infected cells are subjected to the same conditions, the extent of acidification is substantially reduced and the rate of realkalinization is increased. In presence of acetazolamide, acidification is diminished and no realkalinization is observed (Fig. 5B).

Anions that are sufficiently hydrophobic and for which there is no specific transport system penetrate by nonionic diffusion of their acid form. As a result of such ingress, the medium’s pH first increases due to removal of protons from the medium, and then i t falls, but does not return to the initial level due to the relatively slow Cl-iOH- exchange, again mediated by HCO, . Results obtained with Cl--free isoosmotic solutions of Na- caproate (hexanoate) are depicted in Figure 5C,D. While with normal cells an appreciable pH rise is observed, in the presence of the same number of infected cells, the extent of alkalinization is reduced, but the extracellular pH equilibrates a t a level identi- cal to that found with uninfected cells.

473 LACTATE TRANSPORT IN MALARIA-INFECTED ERYTHROCYTES

Transport of lactate across the parasite membrane

Since most of the lactic acid in infected cells is produced by the parasite, we have attempted to char- acterize the mode of transport across the parasite membrane. The experimental access to the parasite membrane is made possible by treating infected cells with Sendai viruses that insert permanent channels permeable to solutes up to 1,600 MW, specifically in the host cell membrane (Kanaani and Ginsburg, 1988, 1989). The integrity of the parasite membrane has been verified by observing the membrane potential-dependent accumulation of rhodamine 123 in the parasite (not shown). Testing the uptake of labeled lactate into virus-treated cells revealed that the initial rate of uptake is similar or larger than that into intact infected cells (Fig. 4 and Table 1). Analysis of the results presented in Figure 4 indicates that the half-time for equilibration in intact cells is 0.3 min, while that of the parasite is 0.2 min. This comparison indicates that translocation across the parasite membrane does not constitute a limiting step in the discharge of lactate from parasite to the extracellular medium. Since the volume of the parasite a t the trophozoite stage is ca. 40% of the intact cell, and t,,, is proportional to the volumelsurface area ratio (i.e., to the radiusl31, it is trivial to show that the specific permeability of the parasite membrane to lactate is very similar to that of the intact cell, since the ratio of the radii is 0.74 while that of the t,,,s is 0.67.

The influx of L-lactate into virus-treated infected cells was neither inhibited by cinnamic acid derivatives (Table 1) nor by PCMBS (Fig. 4). Pyruvate that competitively inhibits lactate transport by means of the monocarboxylate carrier in normal red cells (Hal- estrap, 1976) had no effect on the uptake of labeled L-lactate across the parasite membrane (results not shown).

Effect of pH gradient on lactate transport across the parasite membrane

The accumulation of lactate inside the parasite to concentrations larger than those of the extracellular medium suggests that the pH gradient across the parasite membrane may be mechanistically involved in lactate transport. To test this, the kinetics of lactate uptake into virus-treated cells were studied as a function of the pH of the bathing medium. Results depicted in Figure 6 show that when the medium's pH is reduced to 6, i.e., the pH gradient across the membrane is elevated, there is a transient accumulation of lactate inside the cells beyond the level observed when the medium pH is 7.6. In a second experiment the pH gradient has been dissipated by means of the protonophore CCCP. In this experiment cells were fed with 10 mM glucose. Results displayed in Figure 7 clearly show that the transient accumulation seen in control cells is significantly reduced when CCCP is present, i.e., when the pH gradient is dissipated.

DISCUSSION We have shown elsewhere that the permeability of

erythrocytes infected with the human malaria parasite

TABLE 1. Effect of a-FC and UK-5099 on the initial rate of zero- trans influx of L-lactate into intact and virus-treated trophozoite- infected erythrocytes'

Intact cells Virus-treated System Rate W Inhibition Rate % Inhibition

Control 3.67 3 49 -. _- ~~

a-FC 2.04 44.5 2.45 1.6 UK-5099 1.94 47.2 2.16 13.2

' Trophozoite-infected erythrocytes were separated from culture using the gelatin flotation method to a parasitemia of 95%. Cells were depleted of lactate and glucose by incubation in PBS + 50 mM sucrose at 37°C for 1 hr at 5% hematocrit. After 30 min incubation, the medium was renewed. Cells were washed again and resuspended at 10% hematocrit in PBS t 50 mM sucrose. Half of this cell suspension was treated with Sendai virus as described in Materials and Methods. Intact and virus-treated cells were then preincubated for 10 min on ice in PBS containing 50 mM sucrose, 0.5 mM DNDS in the presenceor absence of either 3 mM u-FC or 0.2 mM UK-5099. Influx was initiated by mixing an equal volume of ice-cold PBS containing 50 mM, 0.5 mM DNDS, 5 mM L-lactate(0.8pCi/mlOin thepresenceorabsenceofeither3mMa-FCor 0.2 mM UK-5099. Uptake of radiolabelled lactate was assayed as described in Materials and Methods. Results are given as mmoles IactateAiter cell water/30 sec.

Plasmodium fakiparum to lactate is dramatically increased (Kanaani and Ginsburg, 1991). Lactate uptake in these experiments has been monitored by the lysis of infected cells dispersed in isoosmotic solutions of ammonium lactate, a method that selectively evaluates the permeability of the host cell membrane (Ginsburg, 1990). Under these conditions, lactate influx was only moderately inhibited by cinnamic acid derivatives (CADs), which are known to inhibit competitively the native monocarboxylate carrier. Further probing of lactate transport using radiotracer techniques and physiological concentrations of lactate indicated that although lactate uptake into infected cells was ca. 600-fold faster than uptake into normal cells, i t was equally susceptible to CADs (Figs. 1,2). It should be emphasized that lactate concentrations in infected cells have been calculated using a value of 0.48 for the fractional water volume. This value was obtained assuming that the parasite occupies 40% of the volume of the infected cells and its fractional water volume is taken as 0.8, while the volume occupied by the host cell cytosol is 60% and its fractional water volume is 0.27 (Zanner et al., 1990). Thus, (0.4 x 0.8) + (0.6 x 0.27) = 0.48. If the latter value is an underestimate, the lactate concentrations would be somewhat lower (by a factor of ca. 1.5) and lactate permeability would be accordingly higher. This dramatic increase could be due either to the activation of native transport pathways or to the induction of new ones.

Further characterization of lactate transport across the host cell membrane indicates that, unlike the saturable nature of the major lactate transporter in normal erythrocytes (the monocarboxylate carrier, Km = 9.4 mM; Fishbein et al., 19881, lactate uptake into infected cells is linear with extraceIIular lactate concentrations up to 50 mM, i.e., it is not saturable. Yet, lactate uptake into infected cells is inhibited by CADs, to the same extent that i t is in normal erythrocytes, implying that the constitutive monocarboxylate carrier may have been modified. In contrast, PCMBS, which covalently binds to the monocarboxylate carrier and inhibits it, affects lactate transport into normal cells but only partially that into infected erythrocytes. This result would suggest that most of lactate uptake is


3.0 h

L 2.5 al

E - 0 . 0 A ' ' " " " ' I . ' 0 10 20 30 40 50 60

TIME (min)

Fig. 6. Effect of the external medium pH on the uptake of L-lactate into virus-treated trophozoite-infected erythrocytes. Trophozoite-infected erythrocytes obtained by the gelatin flotation method were permeabilized using Sendai virus as described in Materials and Methods. The cells were then suspended at 20% hematocrit in PBS containing 50 mM sucrose and 0.5 mM DNDS, a t pH 6 (open circles) or 7.6 (solid circles). Influx was initiated by the rapid mixing of an equivalent volume of PBS containing 50 mM sucrose, 0.5 mM DNDS, 2 mM L-lactate (0.8 pCiiml), at pH 6 or 7.6. Samples were removed at various times of incubation at 37°C and the uptake of lactate was terminated by rapid centrifugation through a cushion of 60% dibutylphthalate and 40% dioctylphthalate, as described in Materials and Methods.

mediated by a system different from the monocarboxylate carrier.

Several constitutive transport systems of the red cell membrane have been shown to be modified in infected cells (Gero et al., 1988; Tripatara and Yuthavong, 1986; Ancelin et al., 1985; Ginsburg and Krugliak, 1983). In most cases this is a result of some increase in the K, and a larger increment in the V,,,. However, the final result is a t most an increase of one order of magnitude in the transport capacity, compared to more than two orders of magnitude in the case of lactate. Hence, it seems plausible that the increment of lactate translocation in infected cells is due to the induction of new permeability pathways in the host cell membrane. Such pathways that mediate carbohydrate and amino acid transport are demonstrably inhibited by CADS (Kanaani and Ginsburg, 1991), thus explaining the inhibition of lactate uptake into infected cells by CADS. Most importantly, the lactate transport capacity of infected cells is much larger than the maximal rate of lactate production by infected cells, thus allowing the rapid disposal of this waste product. Tracer fluxes and their inhibition by CADS and PCMBS (when the constitutive anion transporter is inhibited by DNDS), indicate that the increased translocation of L-lactate across the membrane of the infected cell is mediated by an altered monocarboxylate carrier and new permeability pathways induced by the parasite.

In order to evaluate the contribution of host cell membrane anion transporter to lactate transport, the kinetics of extracellular pH changes that occur upon the dispersion of cells in unbuffered isoosmotic solutions of Na-lactate were analyzed (Deuticke, 1971; Aubert and Motais, 1975). In normal cells, such dis-

4.2 ra 1

0 10 20 30 40 50 60

TIME (min)

Fig. 7. Effect of CCCP on the uptake of L-lactate into virus-treated trophozoite-infected erythrocytes. Trophozoite-infected erythrocytes were separated from culture by gelatin flotation method, permeabilized with Sendai virus, washed, and suspended to 20% hematocrit in PBS containing 50 mM sucrose and 10 mM glucose, pH 7.0 and 37°C. Cells were preincubated in the same medium containing 0.5 mM DNDS, in the absence (open circles) or presence (solid circles) of 10 pM CCCP, for 15 min at 37°C. Influx was measured as described in Figure 6 .

persal results in a rapid acidification of the medium. This is due to the disturbance of the Donnan ratio that results in the exchange of intracellular C1- for extracellular OH- (driven by the C1- gradient). Subse- quently, realkalinization occurs due to reestablishment of the original Donnan ratio by Cl-Aactate- exchange and the ensuing efflux of OH-. Since the movement of OH- is mediated by HC03-, both the extent of the initial acidification and the consecutive rise in pH increase are, as expected, inhibited by the impairment of carbonic anhydrase by acetazolamide. In infected cells, the extent of the initial acidification is reduced and the rate of alkalinization is increased (tl,2 of pH increase in infected cells is 0.33 rnin compared to 0.75 min in uninfected cells; see Fig. 5). Both could be explained by a substantial increase in the permeability of the host cell membrane to lactate through the anion transporter. They are similar to differences observed when normal bovine cells are dispersed in the quickly permeating glycolate (tl,z = 2 min) as compared to their dispersion in pyruvate (t,,, = 5 min) (Deuticke, 1971). A similar result would be obtained when the transport of lactate is increased either by simple diffusion of the acid or through the accelerated symport. But this interpretation would not be compatible with the complete obliteration of pH rise in infected cells in the presence of acetazolamide (Fig. 5D). This may be explained by either a more efficient penetration of this inhibitor into the infected cell, or by an alteration in carbonic anhydrase susceptibility to the inhibitor in these cells.

The obligatory exchange mode of anion translocation by means of the anion exchanger suggests that lactate, as an anion, actually exchanges with C1- on the same carrier. However, the rate of pH changes will always depend on the rate-limiting translocation of OH-. According to this mechanism, one must also assume an

LACTATE TRANSPORT IN MALARIA-INFECTED ERYTHROCYTES 475

acceleration of OH- translocation, in order to comply with the rapid realkalinization. That this indeed occurs is indicated by the kinetics of pH changes in the presence of Na-caproate. Here the extent of pH rise is substantially smaller with infected than with noninfected cells. Such reduction is expected if Cl-IOH- exchange is increased in infected cells and counteracts the pH changes caused by the influx of caproic acid in its acid form. These interpretations would only hold if despite the acceleration of anion translocation the coupling of anion exchange is maintained in the altered membrane of infected cells. The alterationb) in the constitutive anion exchanger apparently do not modify its susceptibility to the specific anion exchange inhibitor DNDS (Ginsburg et al., 1981). The alternative interpretation, i.e., that increased translocation of both lactate and OH- (HCO,) occurs through the new permeability pathways and that their translocation is not driven by obligatory exchange but by the need for electroneutrality, can be excluded. If i t were true, one would neither observe the initial pH decrease that can be explained only by Cl-IOH- exchange nor the effect of acetazolamide.

To conclude the deliberations on lactate flux across the host cell membrane, i t should be emphasized that although a substantial part of this process is mediated by modified constitutive anion exchanger and monocarboxylate carrier, the major part of lactate flux is apparently mediated by the new permeability pathways, as evidenced by the degree of its inhibition by CADs (note that all uptake experiments were done in the presence of the anion transport inhibitor DNDS).

The transport of lactate across the parasite membrane (one cannot distinguish here between the cell membrane and the parasitophorous membrane) is ba- sically different from that across the host cell membrane. Although both processes are not saturable and insensitive to PCMBS, the former is not inhibited by CADs. Most important, however, is the observation that the extent of accumulation of lactate (above extracellular concentration) in virus-treated cells depends on ApH, as evidenced both by the effect of reduced extracellular pH and the effect of the protonophore CCCP. Such behavior would be expected either if lactate translocates by means of a lactate/H+ symport or by simple diffusion as the free acid (Roth and Brooks, 1990). The lack of inhibition by compounds that usually affect symport mechanisms in other types of cells, e.g., CADs or PCMBS, favors the simple diffusion mode of translocation. The presence of a novel type of lactate-/H+ symport, which is not inhibited by classical inhibitors of this system, however, cannot be excluded. Whatever may be the actual mechanism, the capacity of the parasite membrane to transport lactate is far superior to the maximal rates of lactate production by infected cells, which is mostly due to the parasite's glycolytic activity.

A comparison of lactate levels at steady state indicates that they increase in going from uninfected, to infected, and to virus-treated infected cells (compare controls in Figs. 1 and 2; Fig. 4). Accumulation in the latter case depends on the metabolic status of the parasite, since i t reaches higher levels when the uptake experiment is done in the presence of glucose (Figs. 6,

7). These observations are compatible with the ApH- dependence of acid distribution across membranes when transport is mediated either by the acid itself or by an anion:H+ symport mechanism (Roos and Boron, 1981). This dependence is given by the relationship: [TA],/[TA], = ( lopH'- pK + l)/( 10pH"-pK + l), where TA is total acid concentration, and subscripts i and o stand for intracellular and extracellular compart- ments, respectively. Since pH, and pH, are much larger than pK of lactic acid, the e uation can be simplified to yield: [TA],/[ TA], = 10pH'-p '. The steady-state levels of lactate observed in the different cell prepartions are compatible with a cellular pH of 7.1-7.2 in normal red cells, a pH of 6.9-7.0 in the cytosol of infected cells, and a value of 7.3-7.5 for the parasite cytosol (Friedman et al., 1979; Kruckeberg et al., 1981; Yayon et al., 1984). When lactate transport is inhibited by CADs, cells continue to produce lactic acid glycolitically, but the egress of the acid is inhibited, leading to a decrease in cell pH. Hence, CADs reduce the rate of lactate transport and the extent of lactate accumulation.

One can now suggest an integrated view of lactate disposal: glycolytically produced acid diffuses across the parasite cytosol along a concentration gradient. Such process prevents the acidification of the parasite. From the host cytosol lactate is translocated across the host cell membrane by three parallel pathways: (1) either as an anion or as acid through the new permeability pathways. This is undoubtedly the predominant mode; 2) through the accelerated monocarboxylate:H+ symporter; 3) as an anion through the modified anion exchanger. Inhibition of parasite growth by CADs (Kanaani and Ginsburg, 1991) could be due to the effect of these inhibitors on lactate translocation through the host cell membrane, which may result in the dumping of the parasite in its own waste product.

ACKNOWLEDGMENTS Thanks are extended to Professor W.D. Stein for his

continual interest in this work and his numerous helpful suggestions. We are grateful to Dr. D.A. Faulkner of Pfizer Central Research for the supply of UK-5099. This work was supported by The Fund for Basic Research, administered by The Israel Academy of Sciences and Humanities.

a

LITERATURE CITED Ancelin, M.L., Vial, H.J., and Philippot, J.R. (1985) Inhibition of

choline transport into Plasmodium-infected erythrocytes effective antiplasmodial compounds zn uitro. Biochem. Pharmacol. 34:4068- 4071.

Aubert, L., and Motais, R. (1975) Molecular features of organic anion permeability in ox red blood cell. J. Physiol. 246:159-179.

Cabantchik, Z.I. (1990) Properties of permeation pathways induced in the human red cell membrane by malaria parasites. Blood Cells 16:421432.

Deuticke, B. (1971) The transmembrane exchange of chloride with hydroxyl and other anions in mammalian red blood cells. Alfred Benzon Symposium IV, Kopenhagen, pp. 307-316.

Deuticke, B. (1982) Monocarboxylate transport in erythrocytes. J. Membr. Biol. 70939-103.

Deuticke, B., Beyer, E., and Forst, B. (1982) Discrimination of three parallel pathways of lactate transport in the human erythrocyte membrane by inhibitors and kinetic properties. Biochim. Biophys. Acta. 684t96-110.

Deuticke, B., Rickert, I., and Beyer, E. (1978) Stereoselective, SH- dependent transfer of lactate in mammalian erythrocytes. Biochim. Biophys. Acta. 507t137-155.


Fishbein, W.N., Davis, J.I., Foellmer, J.W., and Casey, M.R. (1988) Clinical assay of the human erythrocyte lactate transporter. Bio- chem. Med. Methods Biol. 39:351-359.

Friedman, M.J., Roth, E.F., Nagel, R.L., and Trager, W. (1979) Plasmodium falczparum: Physiological interactions with the human sickle cell. Exp. Parasitol. 47:73-80.

Gero, A.M., Bugledich, E.M.A., Paterson, A.R.P., and Jamieson, G.P. (1988). Stage-specific alteration of nucleoside membrane permeability and nitrobenzylthioinosine insensitivity in Plasmodium falciparum-infected erythrocytes. Mol. Biochem. Parasitol. 27:159-170.

Ginsburg, H. (1990) Alterations caused by the intraerythrocytic malaria parasite in the permeability of its host cell membrane. Comp. Biochem. Physiol. 95A:31-39.

Ginsburg, H.. and Krugliak, M. (1983) Uptake of L-tryptophan by erythrocytes infected with malaria parasites (Plasmodium falciparum). Biochem. Biophys. Acta. 729:97-103.

Ginsburg, H., Krugliak, M., Eidelman, O., and Cabantchik, Z.I. (1983) New permeability pathways induced in membranes of Plasmodium falciparum infected erythrocytes. Mol. Biochem. Parasitol. 8:177- 190.

Ginsburg, H., Kutner, S., Krugliak, M., and Cabantchik, Z.I. (1981) Inhibition of P. falciparum growth in uitro by specific inhibitors of red blood cell anion transport. In: The Biochemistry of Parasites. G.M. Slutzky, ed. Pergamon Press, Oxford, pp. 86-96.

Halestrap, A.P. (1976) Transport of pyruvate and lactate into human erythrocytes: Evidence for the involvement of the chloride carrier and a chloride-independent carrier. Biochem. J . 156:193-207.

Jacobash, G., Minakami, S., and Rapoport, S.M. (1974) Glycolysis of the erythrocyte. In: Cellular and Molecular Biology of Erythrocytes. H. Yoshikawa and S.M. Rapoport, eds. Urban & Schwarzenberger, Munchen, pp. 55-92.

Jensen, J.B. (1978) Concentration from continuous culture of erythrocytes infected with trophozoites and schizonts of Plasmodium falczparum. Am. J . Trop. Med. Hyg. 27:1274-1276.

Kanaani, J., and Ginsburg, H. (1988) Compartment anqlysis ofATP in malaria-infected erythrocytes. Biochem. Int. 17:451459.

Kanaani, J., and Ginsburg, H. (1989) Metabolic interconnection between the human malarial parasite Plasmodium falciparum and its host erythrocyte: Regulation of ATP levels by means of an adenylate translocator and adenylate kinase. J. Biol. Chem. 264:3194-3199.

Kanaani, J . , and Ginsburg, H. (1991) Effect of cinnamic acid derivatives on in vitro growth of Plasmodium falciparum and on the permeability of the membrane of malaria-infected cells. Antimicrob. Agents and Chemother. (Submitted for publication).

Kruckeberg, W.C., Sander, B.J., and Sullivan, D.C. (1981) Plasmo- dium berghei: Glycolytic enzymes of the infected mouse erythrocyte. Exp. Parasitol. 51:438443.

Pfaller, M.A., Krogstad, D.J., Parquette, A.R., and Nguyen-Dinh, P. (1982) Plasmodium falciparum: Stage-specific lactate production in synchronized cultures. Exp. Parasitol. 54:391-396.

Roos, A,, and Boron, W. (1981) Intracellular pH. Physiol. Rev. 61:296434.

Roth, D.A., and Brooks, G.A. (1990) Lactate and pyruvate transport is dominated by a pH gradient-sensitive carrier in rat skeletal muscle sarcolemmal vesicles. Arch. Biochem. Biophys. 279:386394.

Roth, E.F. (1990) Plasmodium falciparum carbohydrate metabolism: A connection between host cell and parasite. Blood Cells 16:453- 460.

Roth, E.F., Raventoz-Suarez, C., Perkins, M., and Nagel, R.L. (1982) Glutathione stability and oxidative stress in P. falciparum infection in uitro: Responses of normal and G6PD deficient cells. Biochem. Biophys. Res. Commun. 109:355-362.

Scheibel, L.W., Adler, A,, and Trager, W. (1979) Tetraethylthiuram disulfide (Antabuse) inhibits the human malaria parasite Plasmo- dium falczparum. Proc. Natl. Acad. Sci. USA 765303-5307.

Sherman, I.W. (1979) Biochemistry of Plasmodium (malarial parasites). Microbiol. Rev. 43:453495.

Tripatara, A., and Yuthavong, Y. (1986) Effect ofinhibitors on glucose transport in malaria (Plasmodium berghei) infected erythrocytes. Inter. J. Parasitol. 16:441446.

Vander Jagt, D.L., Hunsaker, L.A., Campos, N.M., and Baack, B.R. (1990) D-Lactate production in erythrocytes infected with Plasmo- dium falciparum. Mol. Biochem. Parasitol. 425277-284.

Yayon, A., Cabantchik, Z.I., and Ginsburg, H. (1984) Identification of the acidic compartment of Plasmodium falciparum infected human erythrocytes as the target of the antimalarial drug chloroquine. EMBO J. 3:2695-2700.

Zanner, M.A., Galey, W.R., Scaletti, J.V., Brahm, J., and Vander Jagt, D.L. (1990) Water and urea transport in human erythrocytes infected with the malaria parasite Plasmodium falczparum. Mol. Biochem. Parasitol. 40:269-278.

Documents

Transport of lactate in Plasmodium falciparum-infected human erythrocytes