Dual Action of Zn2+ on the Transport Cycle of the Dopamine Transporter*

Background: Zn2+ was originally proposed to inhibit dopamine uptake by the dopamine transporter (DAT). Results: Zn2+ depressed peak currents, and modulated steady state-currents through DAT in a manner dependent on internal Na+. Conclusion: At physiological intracellular Na+, Zn2+ activates both the forward and exchange transport mode of DAT. Significance: The action of Zn2+ is of interest for rescuing conformationally trapped DAT mutants.

The dopamine transporter shapes dopaminergic neurotransmission by clearing extracellular dopamine and by replenishing vesicular stores. The dopamine transporter carries an endogenous binding site for Zn 2؉ , but the nature of the Zn 2؉ -dependent modulation has remained elusive: both, inhibition and stimulation of DAT have been reported. Here, we exploited the high time resolution of patch-clamp recordings to examine the effects of Zn 2؉ on the transport cycle of DAT: we recorded peak currents associated with substrate translocation and steady-state currents reflecting the forward transport mode of DAT. Zn 2؉ depressed the peak current but enhanced the steady-state current through DAT. The parsimonious explanation is preferential binding of Zn 2؉ to the outward facing conformation of DAT, which allows for an allosteric activation of DAT, in both, the forward transport mode and substrate exchange mode. We directly confirmed that Zn 2؉ dissociated more rapidly from the inward-than from the outward-facing state of DAT. Finally, we formulated a kinetic model for the action of Zn 2؉ on DAT that emulated all current experimental observations and accounted for all previous (in part contradictory) findings. Importantly, the model predicts that the intracellular Na ؉ concentration determines whether substrate uptake by DAT is stimulated or inhibited by Zn 2؉ . This prediction was directly verified. The mechanistic framework provided by the current model is of relevance for the rational design of allosteric activators of DAT. These are of interest for treating de novo loss-of-function mutations of DAT associated with neuropsychiatric disorders such as attention deficit hyperactivity disorder (ADHD).
The dopamine transporter (DAT 2 , SLC6A3) terminates synaptic transmission by reuptake of extracellular dopamine into the presynaptic neuron. The transport cycle of DAT is consistent with the alternating access model (1): the transporter presents a substrate-binding site to the extracellular side of the membrane; upon binding of substrate and co-substrates (Na ϩ and Cl Ϫ ), conformational changes occlude the extracellular access pathway and expose the solutes to the intracellular space. This results in their transport into the cell. Subsequently, the transporter returns to the outward facing conformation. This reaction is assumed to be the rate-limiting step of dopamine transport (2). Crystallographic investigations of bacterial transporters, in conjunction with computational modeling, provide compelling evidence in support of this general mechanism (3)(4)(5). The crystal structure of the Drosophila melanogaster dopamine transporter (dDAT) confirmed that insights from these bacterial SLC6 transporter homologs can be extrapolated to SLC6 neurotransmitter transporters (6).
DAT and its close relatives, i.e. the transporters for serotonin and norepinephrine, are targets of several psychoactive substances. These chemically diverse ligands are used for both, therapeutic and recreational purposes, and can be classified with respect to their mode of action on DAT (1). They act as substrates (e.g. amphetamines), competitive inhibitors (e.g. cocaine), or non-competitive inhibitors (e.g. ibogaine), and have in common that they increase extracellular dopamine concentrations because they reduce the net-uptake of dopamine into the cell (1,(7)(8)(9).
However there is one notable exception: Zn 2ϩ , which is an endogenous allosteric modulator of DAT. The mode of Zn 2ϩ binding has been elucidated; it binds DAT at an extracellular site, which is coordinated by four residues (10 -12). The effect of Zn 2ϩ on the function of DAT, however, has remained enigmatic. Both, inhibition and activation of DAT were observed, depending on the experimental assay employed. Originally, Zn 2ϩ was found to inhibit substrate uptake (8,9) but to stimulate amphetamine-induced dopamine release (18). In addition, Zn 2ϩ potentiated dopamine-induced currents through DAT. This is inconsistent with reduced substrate influx (11). In fact, Zn 2ϩ was subsequently shown to either inhibit or stimulate [ 3 H]dopamine uptake in heterologous expression systems, depending on the selected cell line (13). It is evident that these findings are difficult to reconcile.
Here, we examined the hypothesis that Zn 2ϩ exerts an allosteric effect on the transport cycle of DAT. Based on our observations, we provide a mechanistic framework, which accounts for all reported actions of Zn 2ϩ . Most importantly, we show that the kinetics of the transport cycle dictate an intricate coupling between the allosteric effect of Zn 2ϩ and the intracellular Na ϩ concentration: at low [Na ϩ ] i binding of Zn 2ϩ affords accelerated cycling, but at high [Na ϩ ] i it acts as a brake.

Experimental Procedures
Whole-cell Patch-clamp-Patch-clamp recordings were performed with HEK-293 cells stably expressing human DAT (hDAT). The cells were seeded at low density 24 h before recordings. Substrate-induced hDAT currents were recorded under voltage clamp using the whole-cell configuration. Briefly, glass pipettes were filled with a solution consisting of 133 mM K-gluconate, 5.9 mM NaCl, 1 mM CaCl 2 , 0.7 mM MgCl 2 , 10 mM EGTA, and 10 mM HEPES adjusted to pH 7.2 with KOH. Substrate-induced peak currents were isolated using an internal solution comprising 152 mM NaCl, 1 mM CaCl 2 , 0.7 mM MgCl 2 , 10 mM EGTA, and 10 mM HEPES (pH 7.2 with NaOH), which eliminated the steady current component. For some measurements we used a solution containing 50 mM NaCl, 100 mM KCl, 1 mM CaCl 2 , 0.7 mM MgCl 2 , 10 mM EGTA, and 10 mM HEPES (pH 7.2 with KOH).
The cells were continuously superfused with external solution (140 mM NaCl, 3 mM KCl, 2.5 mM CaCl 2 , 2 mM MgCl 2, 20 mM glucose, and 10 mM HEPES adjusted to pH 7.4 with NaOH). Currents were recorded at room temperature (20 -24°C) using an Axopatch 200B amplifier and pClamp 10.2 software (MDS Analytical Technologies). Cells were voltage-clamped to a holding potential of Ϫ60 mV or 0 mV and the washout period following dopamine application was 30 s in all cases. Current traces were filtered at 1 kHz and digitized at 10 kHz using a Digidata 1320A (MDS Analytical Technologies). Drugs were applied using a DAD-12 (Adams & List, Westbury, NY), which permits complete solution exchange around the cells within 100 ms (14). Current amplitudes in response to application of dopamine or the combination of dopamine and Zn 2ϩ were quantified using Clampfit 10.2 software. Passive holding currents were subtracted and the traces were filtered using a 100 Hz digital Gaussian low-pass filter.
Radioligand Binding Experiments-Membranes were prepared from a defined number of stably transfected HEK293 cells expressing human DAT. The binding reaction was carried out as described previously (15): briefly, the assay volume was 0.1 ml, the membranes (corresponding to about 1-1.5 ϫ 10 4 * cells/reaction) were incubated in buffer containing 20 mM Tris-HCl, 2 mM MgCl 2 , 120 mM NaCl, and 3 mM KCl, 10 M ZnCl 2 , and [ 3 H]CFT (specific activity: 80 Ci/mmol; 6 to 7 concentrations covering the range of 3 to 120 nM) for 30 min at 20°C.
Nonspecific binding was determined in the presence of 10 M mazindol.
Modeling-The recorded currents were emulated with a previously published kinetic model of the transport cycle of DAT (2). This model was extended to account for binding of Zn 2ϩ to DAT. The time-dependent changes in state occupancies were evaluated by numerical integration of the resulting system of differential equations using Systems Biology Toolbox (16) and Matlab 2012a (Mathworks).
The voltage-dependence of individual rates were modeled according to Laeuger (17) assuming a symmetric barrier as k ij ϭ k0 ij exp(ϪzQ ij FV/2RT), with F ϭ 96485 Cmol Ϫ1 , r ϭ 8.314 JK Ϫ1 mol Ϫ1 and V the membrane voltage in volts and T ϭ 293 K. The extra-and intracellular ion concentrations were set to the values used in the experiments. Substrate uptake was modeled as (TiClS ϫ kSioff-TiCl ϫ Si ϫ kSion ϩ TiClSZn ϫ kSioff-TiClZn ϫ Si ϫ kSion) ϫ NC/NA. Where NC is the number of transporters and NA is the Avogadro constant. Currents were simulated assuming a transporter density of 25 ϫ 10 6 /cell.

Zn 2ϩ Causes a Modest Shift in the Concentration-dependence of Dopamine-induced Steady-state Currents through DAT, but
Does Not Affect Their Voltage-dependence-We analyzed the action of Zn 2ϩ on distinct partial reactions of the transport cycle by capitalizing on the high temporal resolution of the whole-cell patch-clamp technique. We selected stably transfected HEK293 cells, which expressed the human DAT to high levels; the number of DAT molecules was estimated in saturation binding experiments of [ 3 H]CFT and amounted to 27 Ϯ 3.9 ϫ 10 6 transporters/cell. Upon rapid substrate application, HEK293 cells expressing hDAT exhibit an inwardly directed current that comprises two components, a transient peak current and a steady-state current (Fig. 1A). These currents have been assigned to distinct events in the transport cycle: the peak current has been attributed to the conformational rearrangement that carries substrate through the membrane. The steadystate current is assumed to originate predominantly from the conversion of empty inward facing transporters to the outward facing conformation (2). In other words, the peak current can be used to study substrate translocation, whereas the steadystate current reads out the forward transport mode. Fig. 1A shows a typical current elicited by fast application of 10 M dopamine to a HEK-293 cell stably expressing DAT. The current was measured in the presence of physiological ion gradients, and the membrane potential was clamped to Ϫ60 mV. Application of 10 M dopamine elicited the initial peak current (see inset to Fig. 1A), which was followed by the sustained steady-state current mentioned above. The steady-state current deactivated upon removal of dopamine from the bath solution.
The magnitude of the steady-state current depended on the concentration of dopamine: i.e. it was close to saturation at 30 M (Fig. 1B). The concentration-response curve was adequately described by a saturation hyperbola. The EC 50  We verified that Zn 2ϩ did not change the current-voltagerelation of the steady-state current through DAT by recording currents between Ϫ60 and ϩ30 mV. At potentials more positive than ϩ30 mV, excessive noise precluded further analysis. The currents were induced by application of 30 M dopamine; the current amplitudes were normalized to the amplitude measured at Ϫ60 mV and plotted as a function of voltage: It is evident from Fig. 1C that (i) the current-voltage-relation of the dopamine-induced steady-state current was linear over the explored voltage range and (ii) that it was not affected by Zn 2ϩ .
Zn 2ϩ Alters the Amplitude and the Kinetics of Currents Associated with Dopamine Transport- Fig. 2A shows representative traces of currents induced by 30 M dopamine in the presence and absence of 10 M Zn 2ϩ . Binding of Zn 2ϩ to DAT changed the following properties of the dopamine-induced current: (i) it increased the steady-state current (inset to Fig. 2, A and B) consistent with previous reports, which documented potentiation of dopamine-induced currents by extracellular Zn 2ϩ (18,19) and (ii) decreased the peak current ( Fig. 2C). (iii) Moreover, the current deactivated more rapidly in the presence of Zn 2ϩ . The latter finding suggests that the DAT/Zn 2ϩ complex relaxes faster to its original state (T o NaClZn: see model described below). We tested this conjecture directly by monitoring the time course of peak current recovery after dopamine application and subsequent washout. For these measurements, we exploited the fact that the peak current amplitude is proportional to the number of transporters available for dopamine binding. Thus, the peak current amplitude provides a read-out for occupancy of the outward facing conformation (T o NaCl and T o NaClZn). Similarly, the time course of peak current recovery allows for estimating the time required for the transporter to return to the outward facing conformation. We applied 30 M dopamine for 0.5 s and subsequently removed dopamine from the bath solution for 0.1, 0.2, 0.5, 0.8, 1, 1.2, 1.5, and 2 s, respectively. We then re-applied 30 M dopamine to determine the fraction of transporters that had returned to T o NaCl/T o NaClZn and were therefore available to support another peak current. It is evident from Fig. 2, D and E that peak current amplitudes increased in magnitude as a function of the dopamine-free interval. In Fig. 2F, this time course of recovery is plotted in the presence and absence of 10 M Zn 2ϩ : the data were well described by a mono-exponential function. The fit yielded time constants of 0.29 s (0.26 -0.33 s) and 0.72 s (0.67-0.78 s) in the presence and absence of Zn 2ϩ . The time course of recovery is a measure of the turnover rate of DAT, i.e. the time needed to complete one conformational cycle. Thus, the analysis yields a turnover rate of DAT in the absence of Zn 2ϩ of 1.45 s Ϫ1 (versus 3.45 s Ϫ1 in the presence of Zn 2ϩ ). This rate agrees well with earlier estimates (2,20). The observations therefore suggest that the turnover rate of DAT is larger in the Zn 2ϩbound state. This interpretation, however, disagrees with studies, which documented inhibition of substrate uptake by Zn 2ϩ (11,12).
The Action of Zn 2ϩ on DAT Depends on the Intracellular Na ϩ Concentration-The analysis of the Zn 2ϩ action on the peak current by DAT is hampered by the confounding presence of the steady-state current. Therefore, we measured the peak current in the absence of the steady-state current. This can be achieved by using high intracellular Na ϩ concentrations ([Na ϩ ] i : 150 mM) (2,21). An intracellular concentration of 150 mM Na ϩ favors rebinding of Na ϩ to the inward facing transporter, thus precluding progression through the transport cycle. This eliminates the steady-state current, but keeps the peak current intact. Fig. 3A shows typical current traces. Shown are representative peak currents elicited by 30 M dopamine measured at 0 mV before and after application of 0.1, 1, and 10 M Zn 2ϩ . Zn 2ϩ reduced the peak current amplitude. The magnitude of the effect is quantified for 6 independent recordings on different cells in Fig. 3B. Importantly, the inhibition by Zn 2ϩ exceeded that seen in the measurements summarized in Fig.  2C: when the currents were measured in the presence of 6 mM Na ϩ i , 10 M Zn 2ϩ reduced the peak current by only 30%. In contrast at 150 mM Na ϩ i , 10 M Zn 2ϩ reduced the peak by ϳ70%. This suggests that Na ϩ i rendered the process that gives  rise to the peak current more sensitive to Zn 2ϩ . We then investigated whether Na ϩ i also alters the Zn 2ϩ action on the steadystate current. For this purpose, we set [Na ϩ ] i to 50 mM. As predicted, the steady-state current was lower at 50 mM Na ϩ i than at 6 mM Na ϩ i , because the progress through the cycle was impeded. More importantly, however, under these conditions, 10 M Zn 2ϩ led to a reduction of the steady-state current (Fig. 3,  C and D). This contrasts with the observed current potentiation at 6 mM Na ϩ i (see Fig. 2, A and B). Zn 2ϩ Dissociation from DAT Is State-dependent-It has been proposed that Zn 2ϩ binds to the outward facing state of DAT with higher affinity than to the inward facing state (10,22). To verify this conjecture by an experimental approach, we  designed a protocol to measure the dissociation rate of Zn 2ϩ from the outward facing conformation (Fig. 4A) and a complementary protocol (Fig. 4C) for the inward facing conformation. For the former, we relied on peak current suppression by Zn 2ϩ (see Fig. 2), which is expected to vanish upon Zn 2ϩ dissociation. In the protocol depicted in Fig. 4A, we exposed cells to Zn 2ϩ for 5 s and subsequently measured the recovery of the peak current as a function of the time interval of Zn 2ϩ washout. The time course of peak current recovery was adequately described by mono-exponential rise (Fig. 4B) and thus provided a measure of the dissociation rate of Zn 2ϩ ( ϭ 1.57 s). However, the modest peak current reduction by Zn 2ϩ (ϳ20 -30%) limited the dynamic range of this measurement. This is reflected by the large 95% confidence interval for the time constant estimated by the fit 1.57s (1.06 -3.01 s). These measurements were conducted with physiological ion gradients in place, a condition at which DAT was shown to adopt the outward facing conforma-tion. Hence, the measured rate most likely represents the dissociation rate of Zn 2ϩ from the outward facing conformation.
The other protocol to measure Zn 2ϩ dissociation is shown in Fig. 4C. Here, we first applied 30 M dopamine to induce a current. After 5 s, 10 M Zn 2ϩ was co-applied; this led to the expected potentiation of the steady-state current. Subsequently, Zn 2ϩ was removed from the solution and the steadystate current relaxed to its initial amplitude. The time course of this relaxation is another measure of Zn 2ϩ dissociation (Fig. 4C, inset) and it was well described by a mono-exponential function. The estimated time constant for Zn 2ϩ dissociation under this condition was 0.21 s Ϯ 0.06 s (Fig. 4C). Thus, this time constant (for relaxation of steady-state current) is ϳ10 times smaller than the one obtained by measuring recovery of peak current. This second protocol differs considerably from the first, because upon dopamine binding, DAT enters the transport cycle. Accordingly, under this condition, the transporter visits all conformations within the transport cycle, including those that are inward facing. The faster rate of Zn 2ϩ dissociation observed with the latter protocol is therefore consistent with the idea that Zn 2ϩ dissociates faster from the inward facing conformation of DAT.
It is evident that the dopamine test pulses employed in the first protocol (peak current recovery) elicit currents that display a mixed response: while the peak current suppression by Zn 2ϩ was still observed in test pulses following a Zn 2ϩ free interval of 2 s, the two other effects by Zn 2ϩ (current potentiation and accelerated current deactivation) were not observed in any of the test pulses including those following the shortest Zn 2ϩ free interval (0.2 s, see Fig. 4A). This differential effect is consistent with the fast Zn 2ϩ dissociation observed in the second protocol, i.e. under the condition, where DAT is allowed to undergo the transport cycle to generate the steady state current.
Effect of Zn 2ϩ on Dopamine-induced Membrane Depolarization-Meinild and coworkers showed that application of dopamine together with Zn 2ϩ gave rise to a larger depolarization than dopamine alone when the membrane voltage was measured in unclamped Xenopus laevis oocytes (18). We reproduced this finding in unclamped HEK293 cells stably expressing DAT. However, the average difference in voltage (V DA&Zn 2ϩ Ϫ V DA ) was only ϳ6 mV (Fig. 5, A and B). This may contribute to the reported inhibitory action of Zn 2ϩ on cellular substrate uptake, but it is unlikely to account for its reported magnitude (i.e. by 50%). Given our measured current-voltagerelation (see Fig. 1C), a change by 6 mV is predicted to only modestly affect substrate uptake by DAT.

Discussion
DAT is endowed with an endogenous Zn 2ϩ -binding site, which is absent in the closely related transporters for the other monoamines (10 -12). In the original description of the Zn 2ϩ action on substrate transport by DAT expressed in HEK293 cells, 10 M Zn 2ϩ decreased uptake of dopamine by approx. 50% (12). Based on this finding, Zn 2ϩ has been suggested to act as inhibitor of DAT by restricting a conformational change required for the transport process. This inhibitory action has been reproduced by others (10, 23). Nevertheless, there are other reports, which showed stimulation of substrate uptake by FIGURE 4. Measurements of the dissociation rate of Zn 2؉ from the outward (protocol 1) and the inward facing conformation (protocol 2). A, protocol 1: a HEK293 cell stably expressing hDAT was clamped to Ϫ60 mV and exposed to 10 M Zn 2ϩ for 5 s. Following a Zn 2ϩ -free interval of 0.2, 0.5, 1, 2, 5, and 10 s, respectively, the cell was challenged with 30 M dopamine for 5 s. Application of dopamine elicited a current comprised of a peak and a steady current component. The peak current was suppressed at the first time point (0.2 s after removal of Zn 2ϩ ). This suppression gradually vanished as the Zn 2ϩfree interval was prolonged. B, normalized peak currents as a function of the duration of the Zn 2ϩ -free interval. The straight line indicates a fit of a monoexponential function to the data-points (n ϭ 6). The time-constant for the peak current recovery as estimated by the fit was 1.57s (95% confidence interval: 1.06 -3.01 s). C, protocol 2: a HEK293 cell expressing DAT was clamped to Ϫ60 mV and challenged with 30 M dopamine. Dopamine induced a peak and a steady-state current. The latter increased upon Zn 2ϩ co-application (10 M). Upon removal of Zn 2ϩ the current relaxed to its prior level. The inset shows the time-course of current relaxation upon Zn 2ϩ removal. The red straight line indicates a fit of mono-exponential function to the current trace. The derived time constant from six independent measurements was 0.21 s (Ϯ0.06 s).
Zn 2ϩ or failed to detect an inhibition (13,18,24): A notable example is a study conducted by Pifl et al. where the nature of the Zn 2ϩ action depended on the expression system. While Zn 2ϩ suppressed substrate uptake in HEK-293 cells expressing DAT, it stimulated transport when DAT was expressed in SK-N-MC cells (13). Moreover, in voltage-clamped X. laevis oocytes, Zn 2ϩ did not cause any change in substrate uptake by DAT (18). It is unsatisfactory to ascribe these discrepancies to variation in the experimental assay and/or cell type employed without a rational explanation. The current experiments were designed to explore the mechanistic basis for the action of Zn 2ϩ by exploiting the high temporal resolution provided by electrophysiological recordings. The precision of the recordings is adequate to justify their incorporation into a kinetic model (Fig.  6A), which is based on the model proposed by Erreger et al. (2): the original model related the two types of currents, which were recorded in the present study, to the transport cycle of DAT. Here, we show that Zn 2ϩ affects both current components, i.e. the peak current reflecting the substrate translocation step and the steady state-current, which was suggested to originate predominantly from the return of the empty transporter. In agreement with previous reports, we found potentiation of the steady-state current (18,19). In addition, we observed that Zn 2ϩ reduces the amplitude of the dopamine-induced peak current. Finally, Zn 2ϩ has been suggested to stabilize DAT in the outward facing conformation (22). Stabilization of the outward facing conformation by Zn 2ϩ implies that the latter binds Zn 2ϩ with higher affinity than the inward facing conformation. Our experiments examined this conjecture and found two dissociation rates of Zn 2ϩ , which differed approximately by a factor of ten. The current kinetic model accounts for Zn 2ϩ binding by adding a Zn 2ϩ -bound state to each state in the original scheme. (see resulting scheme, Fig. 6A). Based on our observations, the affinity of Zn 2ϩ for the outward facing conformation was assumed to be 10-fold higher than for the inward facing conformation (see Fig. 4). The differing affinities required an adjustment in the rates of two reaction steps of the transport cycle to maintain microscopic reversibility. The reactions that were inevitably affected by these manipulations were the following: (i) conver-sion of the empty inward facing transporter to the outward facing conformation (i.e. the rate-limiting step) and (ii) isomerization of substrate-loaded transporters from the inward to outward facing conformation (i.e. the substrate exchange reaction).
The model evidently accounts for the observed changes in the dopamine-induced currents that occur upon Zn 2ϩ binding (Fig. 6C). In the model, Zn 2ϩ potentiates the steady-state current because the rate-limiting step in the transport cycle of DAT, i.e. the return step, is accelerated. Consequently, one prediction of the model is that Zn 2ϩ increases the efficacy of substrate transport by DAT. In light of this finding, Zn 2ϩ can be considered as an activator of transport. This conclusion is at odds with the reported inhibitory effect of Zn 2ϩ on substrate uptake, but fits with a previous study that showed stimulation of uptake by Zn 2ϩ (12,13). The model suggests that the peak current amplitude is reduced by Zn 2ϩ , because a sizeable fraction of transporters, which converted from the outward to inward facing conformation with dopamine bound, returns immediately. The returning fraction of transporters carries the charge that normally gives rise to the inwardly directed peak current into the opposite direction. As a consequence, the netcharge and hence the peak current amplitude is reduced. Importantly, the model predicts that Zn 2ϩ either increases or decreases substrate uptake by DAT, depending on the [Na ϩ ] i . This can be explained as follows: Zn 2ϩ enhances the conversion from the inward to the outward facing conformation regardless of whether the transporter is empty or substrate loaded. High [Na ϩ ] i prolong the dwell-time in the substrate-bound form; thus, under this condition, Zn 2ϩ will favor the substrate-exchange mode of the transporter. This leads to inhibition of uptake. Conversely, low [Na ϩ ] i (e.g. 6 mM) is predicted to permit completion of the transport cycle. Given that Zn 2ϩ accelerates the rate-limiting step, this must result in enhanced uptake.
Several lines of evidence support this notion: (i) if 150 mM Na ϩ i is applied internally (a condition that fully eliminates the steady-state current) the peak current is rendered more sensitive to the inhibition by Zn 2ϩ (ii) at 50 mM Na ϩ i , a condition at which the steady-state current is reduced but not fully eliminated, Zn 2ϩ suppresses the dopamine-induced steady-state current. The latter finding contrasts with the observed potentiation of the steady-state current by Zn 2ϩ at low [Na ϩ ] i (6 mM). (iii) Larger currents with Zn 2ϩ at low [Na ϩ ] i , on the other hand, are consistent with the hypothesis that Zn 2ϩ can enhance substrate uptake. This is further supported (iv) by our measurement of the turnover rate of DAT, which was greater in the presence of Zn 2ϩ . It is worth pointing out that our model can faithfully reproduce these observations (see Fig. 6, B-E). In addition, the model does emulate the currents obtained by the protocol employed to measure Zn 2ϩ dissociation from the inward facing conformation with amazing fidelity (Fig. 6F). Moreover, we show in Fig. 6G the simulated dependence of substrate uptake on [Na ϩ ] i . At ϳ10 mM [Na ϩ ] i , the effect of Zn 2ϩ on uptake is predicted to switch from stimulatory to inhibitory.
Our observations and our model provide a plausible explanation for the contradicting reports on the Zn 2ϩ action on DAT. We surmise that in those studies, where Zn 2ϩ exhibited an inhibitory activity, [Na ϩ ] i was high: [Na ϩ ] i accumulation is per se a consequence of heterologous overexpression of DAT in non-excitable cells such as HEK-293 cells, because (in contrast to neurons) they lack the capacity to rapidly clear the influx of Na ϩ . This view is supported by measurements of [Na ϩ ] i : Khoshbouei et al. demonstrated that, upon application of dopamine to DAT-expressing HEK-293 cells, [Na ϩ ] i ramped up to ϳ50 mM within a time period of 60 s (25). This is the concentration, where we observed inhibition of the transport cycle by Zn 2ϩ . In addition, it is the minimum time interval over which cellular uptake assays are conducted.
The study by Meinild et al. invoked an uncoupled Cl Ϫ conductance as a cause for the larger depolarization in the presence of Zn 2ϩ (18). However, our measurements of currents carried by wild-type DAT do not support this interpretation: if Zn 2ϩ had also triggered a Cl Ϫ conductance, it should have altered the voltage dependence of the steady-state current. Specifically, Zn 2ϩ should have shifted the reversal potential to more negative voltages. This, however, was not the case. Therefore we surmise that the Cl Ϫ conductance described by Meinild et al. is most likely a property of the mutant DAT Y335A, which the authors had used to conduct their analysis. Zn 2ϩ has also been shown to inhibit uptake of [ 36 Cl Ϫ ] by DAT-expressing HEK-293 cells (13). Our kinetic model accounts for this finding because Cl Ϫ uptake is limited when the transporter is trapped in the exchange mode. Hence, we consider this effect collateral rather then causative. In this context it is also worth pointing out that our model can explain why Zn 2ϩ enhances substrate release by DAT (23). The exchange mode is a prerequisite for substrate release (9,26,27). In accordance, if Zn 2ϩ binding favors substrate exchange, Zn 2ϩ is predicted to increase amphetamine-induced substrate release.
The stimulatory action of Zn 2ϩ on DAT may be of clinical relevance. The monoamine transporters are prominent targets of several therapeutic drugs. To date, all of these drugs are either inhibitors or substrates. Here, we show that an allosteric modulator like Zn 2ϩ can act as an activator of substrate uptake when DAT works in the forward transport mode. This novel type of pharmacological action is therapeutically desirable to treat loss-of function mutations of DAT (28). For example, recently, a de novo mutation of DAT associated with attention deficit hyperactivity disorder (ADHD) was shown to exhibit reduced activity in the forward and reverse transport mode  (2). To account for Zn 2ϩ binding, we added a Zn 2ϩ -bound state to each state in the original scheme. To account for the stabilization of the outward facing conformation of DAT by Zn 2ϩ , we assumed a 10-fold higher affinity for all outward-facing states (k on out ϭ 1 ϫ 10 7 , k off out ϭ 0.5, k on in ϭ 1 ϫ 10 7 , k off in ϭ 5). The rates of the model were adopted according to the data gathered in the present study. The rates were as follows (s Ϫ1 /Mol Ϫ1 ϫ s Ϫ1 ): k1 ϭ 5 ϫ 10 5 , k2 ϭ 2000, k3 ϭ 10 4 , k4 ϭ 100, k5 ϭ 10 7 , k6 ϭ 10, k7 ϭ 5 ϫ 10 6 , k8 ϭ 200, k9 ϭ 4 ϫ 10 7 , k10 ϭ 40, k11 ϭ 10 7 , k12 ϭ 10 5 , k13 ϭ 1, k14 ϭ 0.1, k15 ϭ 500, k16 ϭ 50, k17 ϭ 0.5, k18 ϭ 5, k19 ϭ 5, k20 ϭ 50. B, simulated peak current recovery after dopamine application assuming 2.5 ϫ 10 7 DAT molecules/cell. This measure of the turnover rate of DAT agrees well with the data in Fig. 2F. C-E, simulated currents at 6, 50, and 150 mM [Na ϩ ] i with/without Zn 2ϩ . The simulated traces match the experimental data shown in Figs. 2A, 3A, and 3C, respectively. F, simulated currents elicited by the protocol to measure Zn 2ϩ dissociation from the inward facing conformation of DAT. The simulation matches the data in Fig. 4C. G, simulated substrate uptake as a function of [Na ϩ ] i . The model predicts enhanced substrate uptake by Zn 2ϩ below 10 mM [Na ϩ ] i and inhibition of substrate uptake above 10 mM [Na ϩ ] i . (29). Most importantly, Zn 2ϩ restored the function of this mutation. This is consistent with the present model: Zn 2ϩ fosters the forward and reverse operating mode of DAT; thus, it is predicted to restore its function. It is therefore conceivable that a small diffusible compound could emulate the action of Zn 2ϩ and could be used as treatment for such disorders.