Evidence for the Sequential Formation of Two Complexes Between an Uptake Inhibitor, GBR 12783 [1-[2-(Diphenylmethoxy)ethyl]-4-(3-Phenyl-2-Propenyl)piperazine], and the Neuronal Transporter of Dopamine


  • Jean-Claude Do-Régo,

  • Hervé Hue,

    1. Laboratoire de Mathématiques, U.F.R. de Médecine et Pharmacie, Saint Etienne du Rouvray, France
    Search for more papers by this author
  • Jean Costentin,

  • Jean-Jacques Bonnet

  • Abbreviations used : DA, dopamine ; DAT, dopamine transporter ; GBR 12783, 1-[2-(diphenylmethoxy)ethyl]-4-(3-phenyl-2-propenyl)-piperazine ; KRM, Krebs-Ringer medium ; TI, transporter-inhibitor complex.

Address correspondence and reprint requests to Dr. J.-J. Bonnet at UPRESA C.N.R.S. 6036, U.F.R. de Médecine et Pharmacie, BP 97, 76803, Saint Etienne du Rouvray, France.


Abstract : Incubation of a crude synaptosomal fraction from rat striatum with GBR 12783 at 37°C produced an inhibition of the specific uptake of [3H]dopamine that increased with time. The inhibition increased when GBR 12783 was present during preincubation and incubation (IC50 = 1.85 ± 0.1 nM) instead of incubation alone (IC50 = 25 ± 3.5 nM). Time-course studies of uptake inhibition demonstrated that a first collision transporter-inhibitor complex (TI) was formed immediately after addition of GBR 12783 so that the initial uptake velocity (Vo) decreased for increasing concentrations of inhibitor (Ki≥ 20 nM). TI slowly isomerized to a more stable complex TI* (K*i≤ 5 nM) with a value of t1/2 = 20-270 s. Fits of data to model 2 in which the steady-state uptake (VS) is set to zero were generally preferred, suggesting that formation of TI* could tend to irreversibility, as a consequence of a very low reverse isomerization. As expected, k, Vo, and VS tended to steady-state values in an asymptotic manner for high concentrations of GBR 12783. GBR 12783 at 2.5 nM produced a mixed inhibition of the uptake, with an increase in KM and a decrease in Vmax ; these effects were improved for 10 nM GBR 12783 and at 20°C. These results are discussed in relation to previous data concerning [3H]GBR 12783 binding. The present work gives the first experimental demonstration that dopamine uptake blockers can act according to a two-step mechanism of inhibition ; this is of great interest, because these inhibitors can oppose the effects of cocaine or amphetamine on the transporter according to a reaction that is partly nondependent on the concentration of the abused agent.

Cocaine abuse has become a major problem of public health in various countries. A recent study strongly suggests that central serotonergic systems could be involved in cocaine self-administration in mice (Rocha et al., 1998). However, conclusions of this study and a wide variety of reports indicate that the neuronal dopamine (DA) transporter (DAT) plays a pivotal role in cocaine properties (Ritz et al., 1987 ; Madras et al., 1989 ; Kuhar et al., 1991). Many studies are devoted to the identification of agents capable of blocking the effects of cocaine, or other abused drugs, on the DAT (Carroll et al., 1997). These blockers could be more efficient, as they behave as slow binding inhibitors, and especially as they could form two sequential complexes with the transporter.

Neuronal carriers and enzymes display some functional similarities so that there is a general agreement with base quantitative and qualitative descriptions of uptake of a neurotransmitter and its inhibition on enzymatic models. With reference to enzyme inhibition (similar work concerning uptake blockers is lacking), a slow binding of an inhibitor to the transporter could result from two different mechanisms. As illustrated in the following scheme, in mechanism A, the slow binding occurs because of the low value of the rate constant for the association of the transporter (T)-inhibitor (I) complex (k3), whereas mechanism B involves the rapid formation of an initial collision complex (TI) that undergoes a slow isomerization leading to the formation of a second complex TI* (Morrison and Walsh, 1988 ; Szedlacsek and Duggleby, 1995).

1 inline image

An uptake inhibition that occurs according to mechanism B is expected to elicit a decrease in the initial rate of uptake as the result of the fast formation of a collision complex, whereas the slow formation of the TI complex that characterizes mechanism A does not modify this initial rate of uptake (Morrison and Walsh, 1988). Reactions occurring in mechanism A and during the first step of mechanism B (i.e., the formation-dissociation of a first collision complex) are characterized by a dissociation constant Ki. An overall dissociation constant K*i can be defined in mechanism B. The extent to which K*i is lower than Ki depends on the relative magnitude of the values for k5 and k6. When k6k5, then K*iKi and formation of TI* greatly contributes to the inhibition (Morrison and Walsh, 1988 ; Szedlacsek and Duggleby, 1995).

In fact, various “pure” inhibitors of DA uptake, devoid of marked releasing effects at inhibitory concentrations (Heikkila et al., 1975 ; Raiteri et al., 1979 ; Bonnet et al., 1984), could act as slow binding inhibitors. Consistent with this are their kinetics of binding to the DAT, on a time scale of seconds to minutes or hours (Bonnet et al., 1986 ; Reith et al., 1986 ; Madras et al., 1989 ; Boja et al., 1992 ; Xu et al., 1995), and the delay they need to offer a maximal protection against N-ethylmaleimide-induced inactivation of the binding site of [3H]GBR 12783 [1-[2-(diphenylmethoxy)ethyl]-4-(3-phenyl-2-propenyl)-piperazine] on the DAT (Héron et al., 1994). Concerning this point, it is noteworthy that transporter substrates differed from pure uptake inhibitors, because they offered an immediate protection of the binding site (Héron et al., 1994).

In some cases, the dissociation rate from the DAT seems to be low enough to allow a noncompetitive component of inhibition to occur. Thus, administration of GBR 12909 to rats resulted, ex vivo, in a mixed, competitive and noncompetitive, blockade of [3H]GBR 12935 binding (Rothman et al., 1991). On the contrary, GBR 12783 and GBR 12909 have been reported to inhibit DA uptake in vitro in a purely competitive manner (Bonnet and Costentin, 1986 ; Andersen, 1989), and this puzzling situation calls for further studies.

Thus, the present work was undertaken to investigate the mechanism of the inhibition of the neuronal uptake of DA provoked by a pure uptake blocker, GBR 12783. This compound rather specifically inhibits the neuronal DAT with a nanomolar affinity, and nanomolar concentrations of GBR 12783 are reported to block DA uptake in a competitive manner at 37°C (Van der Zee et al., 1980 ; Bonnet and Costentin, 1986). Particular attention has been paid to determine whether GBR 12783 interacts with the transporter according to mechanism A or mechanism B. For this purpose, calculations of initial and final steady-state velocities were performed starting from results of experiments in which uptake was begun by the simultaneous addition of [3H]DA and GBR 12783 to preincubated synaptosomes.


[3H]DA uptake

Male Sprague-Dawley rats weighing 150-300 g were purchased from Charles River (Saint Aubin lès Elbeuf, France). All procedures necessary to prepare synaptosomal suspensions were done at 0-2°C.

Animals were killed by decapitation, and their striata were dissected out and homogenized with 10 up-and-down strokes of a Teflon-glass homogenizer (800 rpm) in 10 volumes (wt/vol) of ice-cold 0.32 M sucrose solution containing 0.1 mM pargyline. The nuclear material was removed by centrifugation at 1,000 g for 10 min, and the supernatant S1 was used in uptake experiments.

Aliquots (50-75 μg of protein) of the crude synaptosomal fraction were preincubated at 37°C for at least 5 min, with 1,850 μl of modified Krebs-Ringer medium (KRM) containing 109 mM NaCl, 1 mM KH2PO4, 1 mM CaCl2, 27 mM NaHCO3, and 5.4 mM glucose (pH 7.4 ± 0.1). The duration of the preincubation was chosen to allow controls and assays to stay at 37°C during the same time. Incubation was performed in the presence of 20 nM [3H]DA (20-50 Ci/mmol). The reaction was stopped by the addition of 3 ml of ice-cold KRM containing 10-4M cocaine and immediate centrifugation (7,000 g, 10 min, 4°C). The pellet was washed with 1 ml of KRM containing 10-4M cocaine and centrifuged. The final pellet was sonicated in 250 μl of distilled water, and aliquots of the homogenate were used for the determination of radioactivity and protein concentration according to the method of Lowry et al. (1951) using bovine serum albumin as standard.

The radioactivity was determined by liquid scintillation spectrometry (BetamaticV, Kontron, Trappes, France) in 4 ml of Optiphase Hisafe II with 33-36% counting efficiency.

The specific uptake of DA was generally defined as the difference between the total uptake at 37°C and nonspecific accumulation at 0°C in the presence of 10-4M cocaine. In time-course studies, the nonspecific uptake values corresponded to “zero time” incubation assays to rule out changes in uptake that could occur during stopping procedures. Zero time incubation assays were determined by preincubating control or GBR 12783 assays at 37°C and stopping them immediately after the addition of [3H]DA.

Determination of unlabeled released DA by HPLC with electrochemical detection

The concentration of unlabeled DA released in the incubation medium by synaptosomes measured at the end of a 5-min preincubation reached a concentration of 8.1 nM (mean of three experiments). Assays were centrifuged (7,000 g, 10 min, 4°C), and aliquots of supernatants were filtered (0.45 μm, Millipore, type HAWP) and injected onto an HPLC column (Ultrasphere 5 μm, Beckman) coupled to an electrochemical detector set at +0.8 V. The mobile phase consisted of 0.1 M KH2PO4, 6 mM heptanesulfonic acid, 7.5% methanol, and 0.05% EDTA. In a previous work, the external concentration of DA observed after preincubation of striatal synaptosomes at 37°C has been shown to be stable during a further 1.5-min incubation (Bonnet et al., 1984).


Different experimental conditions have been designed to validate the use of the following equations : (a) GBR 12783 increases the KM of the DA uptake process at 37°C (Bonnet and Costentin, 1986) ; (b) progress curves for GBR 12783-induced inhibition of uptake have been constructed by starting the reaction by the addition of GBR 12783 and/or [3H]DA to preincubated synaptosomes (Morrison and Walsh, 1988 ; Szedlacsek and Duggleby, 1995) ; and (c) occurrence of a tight inhibition of uptake has been avoided by testing GBR 12783 concentrations that were at least 10 times higher than that of the transporter [≤0.2 nM for 50-75 μg of protein/2 ml and a Bmax of 8-10 pmol/mg of protein (Billaud et al., 1993)].

Means of specific uptake values obtained in the presence of GBR 12783 were fitted to the general integrated equation corresponding to the sequential formation of two TI complexes, i.e., mechanism B (Morrison and Walsh, 1988 ; Szedlacsek and Duggleby, 1995) :

2 (model 1) (1) U=VSt+ (V0−VS)(1−e−kt)/k

and to a second equation corresponding to the same mechanism in which the reverse isomerization rate constant (k6) is very low and the formation of TI* becomes essentially irreversible (model 2) ; in these conditions, VS equals zero and Eq. 1 becomes

3 2 U=V0(1−e−kt)/k

In these equations, U, Vo, VS, and k represent, respectively, the specific uptake, the initial velocity, the final steady-state velocity, and the apparent first-order rate constant for the establishment of the equilibrium between TI and TI*.

Dissociation constants were calculated from Vo and VS values, respectively, using a simple inhibition equation :

4 3 Vi=Vu/(1+I/Kiapp)

in which Vi is the inhibited rate of uptake (Vo or VS), Vu is an estimate of the uninhibited rate of uptake, I is the concentration of GBR 12783, and Kappi is an apparent inhibition constant. Ki and K*i were calculated using an equation for classical competitive inhibition in which Vo (or VS) is related to Ki (or K*i) :

5 inline image

where Vmax and KM are the apparent kinetic constants for the neuronal uptake determined in similar experimental conditions and A denotes the concentration of DA (Morrison and Walsh, 1988). For calculations, the total concentration of DA (i.e., 20 nM [3H]DA + 8.1 nM unlabeled released DA) was considered. Using these inhibition constant values, k6 and k5 were calculated from the following equations (Eqs. 9 and 10 in Morrison and Walsh, 1988) :

(5) k=k61+I/(Ki*+Ki*A/KM)1+I/(Ki+KiA/KM)

which corresponds to

(6) k=k6V0/VS


(7) inline image

K*i and k5 were also calculated when the formation of TI* tended to irreversibility (model 2). In these conditions, the linear plot of 1/k versus 1/I gives vertical and horizontal intercepts corresponding to 1/k5 and -1/Ki(1 + A/KM), respectively (Eq. 22 in Morrison and Walsh, 1988).

Data analyses were performed using the curve-fitting programs Sigma Plot or Origin, which give standard deviation for each parameter. Results obtained with these programs were consistent, in contrast with Systat, which gave different results in some cases. The better fit of data to theoretical equations of uptake was that which gave the highest Schwarz's criterium (SC) (Schwarz, 1978). The SC value is given by SC = -n ln (SS) - k ln (n), in which n is the number of experimental values, k is the number of equation parameters, and SS is the sum of the squared differences. It is noteworthy that this global criterium constitutes a statistical comparison without level of significance. Geometric means and 95% confidence limits were calculated for KM and Vmax values, and the significance of changes was tested with a Dunnett's t test. IC50 values (concentrations of GBR 12783 inhibiting 50% of the control uptake) were calculated by nonlinear regression analysis of the specific uptake (Ligand, Biosoft, Cambridge, U.K.). They were compared using a Student's t test.


[3H]DA hydrochloride was purchased from Amersham and NEN (Les Ulis, France). Cocaine hydrochloride was obtained from La Coopérative Pharmaceutique Française (Melun, France). Millimolar solutions of GBR 12783 dihydrochloride (synthesized by Prof. Robba, Caen, France) were prepared in distilled water. Subsequent dilutions and solutions of other agents were performed in the incubation medium.


A first set of experiments had been carried out at 20°C to slow down association and dissociation rates, and then to determine more easily the different steps that could occur during the formation of the TI complex. Comparison of Schwarz's criteria indicated that the best fit for control uptake data was that corresponding to a model of transport in which the initial rate of uptake was greater than the steady-state rate (model 1 in Table 1) and not a linear function of uptake versus incubation duration (Fig. 1). The same results were obtained at 12°C and in a series of experiments performed at 20°C in which incubations were stopped by filtration instead of centrifugation (Fig. 1 and Table 1). On the contrary, a linear model of uptake was statistically preferred for incubations at 37°C. As indicated in Fig. 1, deviation from linearity at 12-20°C was early ; this was supported by the increase of the origin of the linear regressions as an inverse function of the incubation temperature. In the same way, VS/Vo ratios tended to unity when incubation temperature was enhanced (Table 1). Except when indicated, further experiments were carried out at 37°C.

Table 1. Effect of temperature on the characteristics of the specific uptake of [3H]DA in control assaysMeans of specific uptake data measured on a 120-s (at 20°C) or a 210-s (at 12 and 37°C) incubation period and stopped by centrifugation or filtration were fitted using Sigma Plot to linear regressions and to Eq. 1, which corresponds to mechanism B (model 1). Estimates of uptake at the origin (O) are expressed as fmol/mg of protein. Velocity of uptake in linear regression (V), initial velocity (Vo), and steady-state velocity (VS) are expressed as fmol/mg of protein/s. The apparent first-order rate (k) is expressed as 10-3/s. When indicated, incubations were stopped by dilution and immediate filtration through Millipore cellulose ester filters (0.65 μM). Filters were washed once with 4 ml of assay buffer. Uptake values were corrected for the specific binding of [3H]DA on filters that was generated in the presence of 10-4M cocaine (nonspecific assays). Means (standard deviation) of four (centrifugation : 12 and 37°C), nine (filtration : 20°C), and 51 (centrifugation : 20°C) experiments carried out in duplicate are shown.A better fit of data was obtained with linear regressions
 Parameters for linear regression
Experimental conditionsV OSC
  1. aor Eq.1

  2. b, as indicated by the Schwarz's criterium (SC).

Centrifugation :    
12°C 88.1 (2.7) 1,656 (338)-170.8
20°C 388 (2.4) 1,048 (181)-107.6
37°C 899 (8.85) 84.6 (311)-193.2 a
Filtration :    
20°C 229.1 (8.4) 1,022 (428)-213.6
 Parameters for model 1
Centrifugation :    
12°C 237 (35.7)80.4 (2.3)54 (17)-153.9 b
20°C 536 (24.6)350 (181)42 (11)-97.9 b
37°C 919 (15.9)691 (2,794)1.1 (21)-196.8
Filtration :    
20°C 346 (21.2)204 (7.55)45 (15)-209.4 b
Figure 1.

Influence of the incubation temperature on the time course of the specific uptake of [3H]DA in controls. Crude synaptosomal suspensions were preincubated at 12-37°C and then incubated at the same temperature in the presence of 20 nM [3H]DA for varying times up to 120 s (20°C) or up to 210 s (12-37°C). Assays were stopped and rinsed by dilution in KRM containing 10-4M cocaine and centrifugation. Zero time incubation assays were used for the estimation of nonspecific uptake. Specific uptake values presented here are expressed as percentages of that observed for a 120-s incubation period at 37, 20, or 12°C. Presented are means ± SEM of four to 51 experiments carried out in duplicate. Characteristics of fits of the specific uptake values are shown in Table 1.


FIG. 1.

The inhibitory potency of GBR 12783 on the specific uptake of [3H]DA was increased significantly when a 5-min preincubation of the synaptosomal fraction was performed in the presence of the uptake blocker. Thus, its IC50 value decreased from 25 ± 3.5 nM (nH = 0.87 ± 0.15) when the inhibitor was omitted from the preincubation medium to 1.85 ± 0.1 nM (nH = 0.88 ± 0.4) when it was present (means ± SEM from four experiments performed in duplicate ; p < 0.001).

Saturation experiments

Saturation experiments carried out in the presence of 30-500 nM [3H]DA showed that the presence of 2.5 nM GBR 12783 during preincubation and incubation periods (5 + 1 min) resulted in an increase in the KM and a slight but significant decrease in the Vmax for the specific uptake (Table 2). Thus, the KM value in controls (224 nM) significantly increased to 355 nM in the presence of 2.5 nM GBR 12783, whereas Vmax decreased from 505 in controls to 478 pmol/mg of protein/min in the presence of GBR 12783 (Table 2). The addition of 2.5 nM GBR 12783 during only the incubation period still significantly increased the KM value. GBR 12783 at 10 nM provoked marked modification of saturation constants of the [3H]DA uptake, whether the inhibitor was present or not during the preincubation period (Fig. 2). Similar results were obtained in the presence of 5 nM GBR 12783 at 20°C (Table 2).

Table 2. Effect of GBR 12783 on saturation constants for the specific uptake of [3H]DAAliquots of a crude synaptosomal fraction were preincubated for 5 min or 15 min at 37 and 20°C, respectively. Incubation was started by the addition of [3H]DA (30-500 nM final concentrations ; 2 ml final volume for set 1 and 4 ml for set 2). GBR 12783 at final concentrations of 2.5, 5, or 10 nM was present when necessary during the preincubation and the incubation periods (I + PI) or during incubation alone (I). One-minute incubation assays were stopped by centrifugation. The specific uptake was defined as the difference between the total uptake at 37 or 20°C and the nonspecific accumulation at 0°C in the presence of 10-4M cocaine. Geometric means and 95% confidence limits of four to eight experiments (n) carried out in duplicate are shown. Values were compared with those obtained in the respective controls, using a Dunnett's t test.
Experimental conditionsKM (nM) Vmax (pmol/mg of protein/min)n
  1. ap < 0.05

  2. bp < 0.01, significantly different from the respective controls.

Set 1 : 37°C    
Control224 (208-240)505 (486-524)8
2.5 nM GBR 12783, I252 (241-263) a483 (469-497) 
4 2.5 nM GBR 12783, I + PI  
355 (316-399) b478 (460-495) a4 
10 nM GBR 12783, I327 (304-352) b450 (434-467) b4
10 nM GBR 12783, I + PI536 (498-577) b298 (283-314) b4
Set 2 : 20°C    
Control65.1 (46.2-91.6)134 (109-163)6
5 nM GBR 12783, I144 (80.3-257) a58 (39-0-86.3) b6
5 nM GBR 12783, I + PI161 (125-206) a41.9 (33.7-52.0) b6
Figure 2.

Effect of 10 nM GBR 12783 on Lineweaver-Burk plot of the specific uptake of [3H]DA. Aliquots of a crude synaptosomal fraction were preincubated for 5 min at 37°C in the presence (l + Pl) or in the absence of 10 nM GBR 12783. Incubation was started by the addition of 100 μl of [3H]DA (30-500 nM final concentrations) and 100 μl of GBR 12783 (10 nM final concentration) (l) when necessary (2 ml final volume). The 1-min incubation period was stopped by centrifugation. The nonspecific uptake was determined at 0°C in the presence of 10-4M cocaine. Means ± SEM of four to eight experiments carried out in duplicate are shown. The slope of the linear regression calculated from uptake values obtained with synaptosomal fractions preincubated and incubated in the presence of 10 nM GBR 12783 (1.81 ± 0.015 μM/nmol/mg of protein/min) was significantly higher than that obtained when GBR 12783 was omitted from the preincubation medium (0.73 ± 0.01 μM/nmol/mg of protein/min ; estimates ± SEM ; t = 59.96, p < 0.001).


FIG. 2.

Time course of GBR 12783-induced uptake inhibitions

Specific uptake values obtained in control experiments fitted well with a linear relationship between uptake and time (Fig. 3 and Table 3). Incubation of synaptosomal fractions for 10-210 s with increasing concentrations of GBR 12783 (2.5-160 nM) generated a family of curves depicting uptake inhibitions (Fig. 3) that are characteristic of plots expected when the reaction is started by the simultaneous addition of the substrate and a slow binding inhibitor (Morrison and Walsh, 1988).

Figure 3.

Progress curves for inhibition of specific uptake of DA by GBR 12783. Aliquots of a crude synaptosomal fraction were preincubated for different periods to allow each control and assay to stay 8 min at 37°C. Incubation at 37°C was started by the addition of 20 nM [3H]DA and GBR 12783 when necessary (2 ml final volume). The 10-210-s incubation period was stopped by centrifugation of the assays diluted in a KRM containing 10-4M cocaine. Zero time incubation assays were used for the estimation of nonspecific uptake. Presented are means ± SEM of four experiments carried out in duplicate. Parameters of fits of these uptake values are given in Table 3.

Table 3. Parameters of the fits of the progress curves of inhibition of the specific uptake of [3H]DA by GBR 12783Means of specific uptake values obtained in the presence of GBR 12783 and reported on Fig. 3 were fitted (Sigma Plot program) to Eq. 1 (model 1) and to Eq. 2 corresponding to the same mechanism in which the formation of TI* tends to irreversibility (model 2), giving estimates (and standard deviation) of uptake inhibition parameters. Estimate of uptake at the origin (O) is expressed as fmol/mg of protein. Velocity of uptake is expressed as fmol/mg of protein/s. The apparent first-order rate constant for the establishment of equilibrium between TI and TI* (k) is expressed as 10-3/s.A better fit of data was obtained withas indicated by the Schwarz's criterium (SC). All correlation coefficients were ≥0.996 for Eq. 1 and ≥0.998 for Eq. 2.
ControlV O 
 899 (9) 85 (311) 
Assays : model 1VoVskSC
  1. aEq. 1 or

  2. bEq. 2

2.5 nM GBR 12783961 (38)500 (93)10 (0.4)-208.6
5 nM GBR 12783947 (22)329 (31)13 (2)-195.1
10 nM GBR 12783844 (11)251 (11)15 (1)-178.3 a
20 nM GBR 12783633 (16)212 (9)20 (2)-182.3
40 nM GBR 12783398 (19)104 (6)27 (3)-157.8 a
80 nM GBR 12783191 (12)59 (3)30 (7)-165.1
160 nM GBR 12783124 (7)37 (2)30 (5)-158.8
Assays : model 2VoVs = 0kSC
2.5 nM GBR 12783912 (10)2.6 (0.2)-201.8 b
5 nM GBR 12783870 (6)5.1 (0.3)-185.5 b
10 nM GBR 12783759 (5)6.2 (0.1)-181.2
20 nM GBR 12783536 (4)6.0 (0.3)-173.2 b
40 nM GBR 12783308 (13)7.8 (0.2)-176.2
80 nM GBR 12783153 (5)7.8 (0.5)-155.9 b
160 nM GBR 1278399 (2)7.8 (0.4)-141.4 b

FIG. 3.


Fits of the specific uptake values to the general integrated Eq. 1 (model 1) or to Eq. 2, corresponding to model 2 in which the formation of TI* is essentially irreversible, gave estimates of uptake velocities (Vo, VS) that decreased in an asymptotic manner at higher GBR 12783 concentrations (Table 3 and Fig. 4A). It is noteworthy that Vo calculated for 5 nM GBR 12783 in model 1 was significantly higher than the rate of uptake in controls (p < 0.05). Kappi, Ki, and K*i calculated by fitting VS and Vo values to Eqs. 3 and 4 (legend of Fig. 4A and Table 4) were completely consistent with the aforementioned IC50 values. First-order rate constants for the establishment of equilibrium between TI and TI* (k) were calculated for both models (Table 3). Curves describing increases in k values as a function of GBR 12783 concentrations exhibit asymptotes for the highest tested concentrations (Fig. 4B) ; t1/2 values corresponding to these asymptotes were 23 and 89 s for models 1 and 2, respectively.

Figure 4.

Secondary plots of Vo, Vs, and k versus GBR 12783 concentrations. Presented are initial velocities Vo, steady-state velocities Vs, and apparent first-order rate constant (k) values reported in Table 3 and corresponding to specific uptake data fitted to equations for models 1 and 2. A : First of Vo and Vs to Eq. 3. For Vo (model 1) : Vu = 1,062 (40.9) fmol/mg of protein/s, Kiapp = 26.98 (4.14) nM. For Vs (model 1) : Vu = 988 (45.7) fmol/mg of protein/s, Kiapp = 2.96 (0.413) nM. For Vo (model 2) : Vu = 1,011 (37.9) fmol/mg of protein/s, Kiapp = 21.66 (3.07) nM. B : Plot of k values obtained using curve-fitting equations for models 1 and 2 against GBR 12783 concentrations. C : Linear plot of 1/k values obtained for models 1 and 2 versus reciprocals of GBR 12783 concentration (1/C). Linear regressions gave the following results : 1/k = 172.9 s nM/C + 37.35 s, (r2 = 0.925) for model 1 ; and 1/k = 625.2 s nM/C + 114.07 s, (r2 = 0.942) for model 2.

Table 4. Calculated constants for the inhibition of the DA uptake by GBR 12783 according to model 1Dissociation constants (Ki and Ki*), k5, and k6 were calculated using Eqs. 4-7, with A corresponding to 28.1 nM (20 nM [3H]DA + 8.1 nM unlabeled released DA), and KM and Vmax values corresponding to 224 nM and 8,416 fmol/mg of protein/s, respectively, as determined in saturation experiments performed under similar experimental conditions. Ki and k5 values for 2.5 and 5 nM GBR 12783 assays could not be calculated, because their initial uptake velocity was higher than the uptake velocity in controls.
 Ki (nM) Ki* (nM) k5 (10-3/S) k6 (10-3/S)
2.5 nM GBR 127832.555.15
5 nM GBR 127832.414.52
10 nM GBR 1278381.533.25107.304.45
20 nM GBR 1278337.045.2140.996.71
40 nM GBR 1278326.314.4534.437.01
80 nM GBR 1278318.234.7926.029.29
160 nM GBR 1278321.695.7824.378.86

FIG. 4.


Calculation of the true rate constants for the formation (k5) and dissociation (k6) of TI* in model 1 gave k5/k6 ratios that were roughly similar to Ki/K*i ratios (Table 4). The vertical intercept of the linear plot of 1/k versus 1/I corresponds to 1/k5, allowing a k5 value for model 2 (0.009 s-1) to be estimated (Fig. 4C). The same calculation gave a K5 value for model 1 (0.027 s-1), which agrees well with values reported in Table 4. The horizontal intercept of the linear plots gave Ki values of 4.63 and 5.48 nM for models 1 and 2, respectively (Fig. 4C).

As indicated by Schwarz's criterium values, model 2 was often preferred to model 1, suggesting that the formation of the second complex TI* could tend to irreversibility (Table 3).


Preliminary data provided evidence that control DA uptake at 12-20°C deviated from linearity and were best fitted to Eq. 1, in spite of experimental conditions designed to rule out nonspecific processes in specific uptake values (Fig. 1.) Calculated values of the origin of linear regressions increased as the incubation temperature decreased, also suggesting that deviation from linearity was most pronounced as the temperature decreased (Table 1). Even if this result was not expected, it is consistent with the curvilinear aspect of the time course of the DA uptake reported in a previous study performed at 25°C and using filtration to stop the assays (Xu et al., 1995). A possible explanation can be found in relation to the effects of temperature and Na+ on the transport(er) (Bonnet et al., 1990 ; Trendelenburg, 1991). Vmax for DA uptake decreased with temperature. This is probably the consequence of a lower rate of transporter functioning, because populations of DAT (8-10 pmol/mg of protein) were not affected by temperature (Bonnet et al., 1990 ; Billaud et al., 1994), and KM values consistently decreased with temperature (Table 2 : see also Bonnet et al., 1990). Thus, the rate of turnover for the DAT (Vmax/Bmax) calculated from values presented in Table 2 decreased from 56 to 15 DA molecules/DAT/min when the temperature was reduced from 37 to 20°C. This value should be in the range of 1-2 DA molecules/DAT/min at 12°C (Bonnet et al., 1990). The 5-min preincubation of synaptosomes in a high Na+-containing medium set up a population of DAT in an outwards facing position, allowing the uptake to reach a maximal rate immediately after addition of DA (Trendelenburg, 1991). Thus, rates of transporter turnover at 20 and 12°C were so low that DAT did not operate fast enough to maintain the initial population of DAT in a position favorable to uptake. Consequently, lower rates of uptake were observed.

Data obtained in the present work strongly indicate that the inhibition of [3H]DA uptake induced by GBR 12783 displays several characteristics of an enzymatic slow binding inhibition involving the sequential formation of two protein-inhibitor complexes (Morrison and Walsh, 1988 ; Szedlacsek and Duggleby, 1995). First of all, the inhibition of uptake by GBR 12783 slowly reached a maximal level with t1/2 values in the range of 20-270 s, depending on the concentration of GBR 12783 and model of inhibition that were considered (Table 3). For GBR 12783 concentrations of ≥40 nM, k and uptake velocities (Vo and Vs) tended to steady-state values in an asymptotic manner, indicating that inhibition did not rise with time because of an artefactual process, such as transporter inactivation or decrease in DA concentration (Fig. 4A and B). In these latter cases, k values should have continued to increase and uptake velocities should have decreased to zero, or tended to this value in a nonasymptotic manner. Finally, the decrease in Vo as a function of the inhibitor concentration demonstrates that transporter and GBR 12783 rapidly formed a first collision complex, which then underwent an isomerization to a more stable complex (Table 3 and Fig. 4A). Supporting this point, Ki* values were consistently lower than Ki (Table 3). Thus, GBR 12783 was able to reach the transporter rapidly, excluding a major involvement of diffusion barriers in the slow binding inhibition observed here. Indirect evidence for this has been reported in a previous study in which similar drug concentrations offered either an immediate protection of the [3H]GBR 12783 binding site against inactivation by N-ethylmaleimide, in the case of substrates, or a delayed one, in the case of pure uptake blockers (Héron et al., 1994).

Present evidence for an isomerization of a first complex is consistent with thermodynamic data obtained in binding studies with [3H]GBR 12783 and [3H]mazindol, because they already suggested that formation of a stable TI complex needs a conformational change of the DAT (Bonnet et al., 1990 ; Billaud et al., 1994). These studies showed that such a conformational change is likely to occur for the binding of every pure uptake blocker, including abused drugs, and consequently, the sequential formation of two TI complexes could be a general property of this class of agents. Also consistent with previously reported data are Ki* values presented in Table 3, which are the same as Ki and KD values obtained in experiments dealing with the inhibition of DA uptake at 37°C (Bonnet and Costentin, 1986) and binding of [H]GBR 12783 at 20-37°C (Bonnet et al., 1990 ; Billaud et al., 1993, 1994).

Another point raised by the present study was the appearance of a noncompetitive component in the GBR 12783-induced inhibition of [3H]DA uptake (Table 2). This component was expected, because the low values of k6 (reverse isomerization rate constant) determined in the present study indicate that a 1-min period of uptake did not allow a subtotal fraction of TI* complexes to dissociate (Tables 3 and 4). It is noteworthy that this seems at variance with previous data suggesting that GBR 12783 and GBR 12909 exerted a pure competitive inhibition of the DA uptake (Bonnet and Costentin, 1986 ; Andersen, 1989). However, these previous data resulted from studies performed in the presence of low concentrations of GBR derivatives (1-3 nM), which induced rather minor modifications of the saturation constants ; in the present work, use of higher concentrations and/or lower incubation temperature greatly enhanced the effects of GBR 12783 on these parameters and clearly induced decreases in Vmax (Table 2). Furthermore, longer incubation periods used in previous studies (3-5 min) were more favorable to the dissociation of TI* complexes (Bonnet and Costentin, 1986 ; Andersen, 1989). These results stress the relationships between incubation duration and rates of interactions between the inhibitor and the transporter on the one hand, and involvement of a Vmax reduction in the inhibition of uptake on the other hand. This noncompetitive component is more likely to appear at physiological temperature with pure uptake blockers displaying the higher affinities, because the dissociation rate of tritiated pure uptake blockers seems to be an inverse function of their affinity (Vignon et al., 1988 ; Zimanyi et al., 1989 ; Héron et al., 1996).

Vo values calculated for low GBR 12783 concentrations tended to be higher than the rate of control uptake, a tendency that was significant for 5 nM GBR 12783 and model 1 (Table 3). In analogy with the modulated receptor hypothesis, one could suggest that a component of the action of the inhibitor on the transporter consists in raising the rate at which the latter recovers its efficient configuration for uptake. Alternatively, the apparent discrepancy in uptake rates could be due to the mode of calculation and the short period of incubation chosen to be within the linear phase of uptake for controls.

The preference for Eq. 2 versus Eq. 1 in modeling uptake inhibition data could also originate from the necessity of performing uptake assays during the linear phase of DA transport. Present experimental data have been obtained for a limited part of the progress curve describing the inhibition occurring according to model 2, because t1/2 values calculated for the formation of TI* ranged between 88 and 266 s (Table 3), indicating that the concentration of TI* began to reach a maximal level after 6-18 min of incubation (4 ×t1/2). Moreover, some results from previous studies indicate that the binding of GBR 12783 to the DAT, and consequently the resulting uptake inhibition, are reversible. So various uptake blockers and substrates behave as pure competitive inhibitors of the [3H]GBR 12783 binding to striatal membranes studied at equilibrium at 0, 20, or 37°C (Bonnet et al., 1990 ; Billaud et al., 1994 ; Saadouni et al., 1994). Furthermore, dissociation experiments of [3H]GBR 12783 binding from striatal membranes are completely consistent with reversibility ; in a 10 mM Na+ medium, t1/2 values decreased from 138 to 38 min when dissociation was performed at 0 and 20°C, respectively. A further rise in temperature even more markedly affected GBR affinity (Bonnet et al., 1990 ; Billaud et al., 1994) and brought the t1/2 value for dissociation to 2 min (data not shown ; but see Héron et al., 1994). This value agrees with the mean value for k6 in model 1 (Table 3), consistent with a two-step model of inhibition in which k4k6. Thus, considering that properties of the [3H]GBR 12783 specific binding to synaptosomal preparations and isolated membranes were rather similar (Billaud et al., 1993), it is probable that the DA uptake inhibition is reversible quite rapidly, in accordance with an uptake inhibition conforming to model 1.

Enzyme inhibition resulting from a two-step mechanism (mechanism B) is of greater interest in therapeutics than classical competitive inhibition and slow inhibition due to a slow rate of formation of a unique enzyme-inhibitor complex (mechanism A). As a matter of fact, in mechanism B, the second, isomerized complex is not sensitive to the concentration of other competitive species, in contrast to the initial formation of the first collision complex (Morrison and Walsh, 1988). Thus, the more stable the second complex is, the more independent to competition the resulting inhibition is and the more useful the inhibitor. In enzymology, several examples are found for which half-lives for dissociation of the second complex range from hours to days (Morrison and Walsh, 1988). Thus, a two-step inhibitor of DA uptake displaying a very high affinity and making a stable TI* complex with the DAT could oppose the binding of an abused agent according to a reaction that should be partly nondependent on the concentration of this agent (see also Rothman et al., 1989). Results from saturation experiments reported here are consistent with this, because they clearly demonstrated that a noncompetitive component of inhibition was observed when GBR 12783 was added simultaneously to a saturating concentration of DA (Table 2). However, as discussed before, it is likely that the GBR 12783-DAT complex can dissociate quite rapidly and, consequently, GBR 12783 just appears as a proto-type for highly potent and useful DA uptake blockers. Finally, considering structural and functional homologies between DAT and the more closely related neuronal transporters, one can suggest that a two-step inhibition of uptake could be also observed with blockers of norepinephrine and serotonin uptakes that display antidepressant and/or anorexiant properties.

To our knowledge, the present study establishes for the first time that the inhibition of the [3H]DA uptake can result from the sequential formation of two TI complexes. Slow binding inhibitors displaying similar or higher affinity than GBR 12783 could be particularly useful in preventing the binding of abused drugs to the DAT.