Engineered zinc-binding sites confirm proximity and orientation of transmembrane helices I and III in the human serotonin transporter

Abstract

The human serotonin transporter (hSERT) regulates neurotransmission by removing released serotonin (5-HT) from the synapse. Previous studies identified residues in SERT transmembrane helices (TMHs) I and III as interaction sites for substrates and antagonists. Despite an abundance of data supporting a 12-TMH topology, the arrangement of the TMHs in SERT and other biogenic amine transporters remains undetermined. A high-resolution structure of a bacterial leucine transporter that demonstrates homology with SERT has been reported, thus providing the basis for the development of a SERT model. Zn2+-binding sites have been utilized in transporters and receptors to define experimentally TMH proximity. Focusing on residues near the extracellular ends of hSERT TMHs I and III, we engineered potential Zn2+-binding sites between V102 or W103 (TMH I) and I179-L184 (TMH III). Residues were mutated to either histidine or cysteine. TMH I/III double mutants were constructed from functional TMH I mutants, and Zn2+ sensitivity was assessed. Dose-response assays suggest an approximately twofold increase in sensitivity to Zn2+ inhibition at the hSERT V102C/M180C and approximately fourfold at the V102C/I179C mutant compared to the hSERT V102C single mutant. We propose that the increased sensitivity to Zn2+ confirms the proximity and the orientation of TMHs I and III in the membrane. Homology modeling of the proposed Zn2+-binding sites using the coordinates of the Aquifex aeolicus leucine transporter structure provided a structural basis for interpreting the results and developing conclusions.

Figure 1.

TMH I exhibits greater sensitivity to mutation than TMH III. [³H]5-HT uptake activity of hSERT C109A, TMH I, TMH III, and TMH I/III mutants was assessed in transiently transfected HeLa cells as previously described (Roman et al. 2003); nonspecific uptake was determined using 10 μM fluoxetine. Specific [³H]5-HT uptake of TMH I or TMH III single mutants (A) and TMH I/III double mutants (B) was normalized to total specific uptake of C109A. Bars represent mean ± SEM of at least three independent experiments performed in triplicate. (N.D.) Not detectable.

Figure 2.

Changes in SERT expression do not account for mutation-induced reductions in [³H]5-HT uptake. Surface binding assays were performed in HEK-293 cells transiently transfected with mutants possessing <50% parental uptake activity, as described in Materials and Methods. (A) V102C, V102C/M180C, W103H/I179C, W103H/M180C, and W103H/W182C exhibited a decrease in total binding compared to hSERT, but (B) the percentage of surface binding compared to the total protein was approximately the same for C109A and all of the mutants. Nonspecific binding was determined using 10 μM fluoxetine. Internal binding was determined as the binding of [³H]citalopram in the presence of 200 μM MPP⁺. Specific binding on the surface was calculated as (total binding − nonspecific binding) − (binding in the presence of MPP⁺ − nonspecific binding). Bars represent the mean ± SEM for assays performed in triplicate in at least three independent experiments, except $ = n of 2.

Figure 3.

hSERT TMH I/III double mutants exhibit significant Zn²⁺ inhibition. Inhibition of [³H]5-HT uptake by 1 mM Zn²⁺ in transiently transfected HeLa cells was performed as described in Materials and Methods; nonspecific uptake was determined using 10 μM fluoxetine. (A) Only the V102C mutant exhibited a significant inhibition by Zn²⁺ compared to C109A. (B) Mutation of residues V102C/I179C, V102C/M180C, W103H/M180C, W103H/W182C, W103H/A183C, and W103C/W182C resulted in statistically significant Zn²⁺ inhibition. The dashed lines represent the maximum Zn²⁺ inhibition by V102C alone (V102C/TMH III mutants) or C109A (W103H/TMH III and W103C/TMH III mutants). Bars represent the mean ± SEM for at least three independent experiments performed in triplicate. One-way ANOVA with Bonferroni's post-hoc test was performed for statistical analysis. (†) p < 0.001 vs. V102C; (**) p < 0.01 vs. C109A; (***) p < 0.001 vs. C109A; (N.D.) not determined.

Figure 4.

DTT increases inhibition by 1 mM Zn²⁺ at the hSERT V102C/A183C mutant. Assays were performed in transiently transfected HeLa cells in the presence or absence of 12 mM DTT, as described in Materials and Methods. (A) In KRH/glucose, V102C/A183C was insensitive to inhibition by 1 mM Zn²⁺ compared to V102C. Zn²⁺ inhibition at V102C/A183C was increased in the presence of DTT. (B) Total specific uptake of [³H]5-HT was reduced in the presence of DTT. (***) p < 0.001 vs. V102C without DTT; (†) p < 0.001 vs. total specific [³H]5-HT uptake in KRH/glucose. One-way ANOVA with post-hoc Bonferroni's multiple-comparison test was performed for statistical analysis. Data represent mean ± SEM from three separate experiments.

Figure 5.

Zn²⁺ significantly potentiates MTSET inhibition of [³H]5-HT uptake at the hSERT V102C and W103C mutants. HeLa cells were transiently transfected with the indicated cDNAs. At the time of assay, cells were treated with 1mM MTSET for 10 min at 22°C in the presence or absence of 1 mM Zn²⁺ in Zn²⁺-binding buffer or with 1 mM Zn²⁺ alone, washed twice with Zn²⁺-binding buffer, and [³H]5-HT uptake was then performed as described in Materials and Methods. V102C was insensitive to 1 mM MTSET in the absence of Zn²⁺. In the presence of 1 mM Zn²⁺, V102C and W103C MTSET inhibition of [³H]5-HT uptake was potentiated, but C109A remained insensitive to MTSET. Comparison of inhibition at V102C, W103C, and W103H by Zn²⁺ alone or Zn²⁺ plus MTSET suggested that the potentiation of MTSET inactivation was due to the mutated cysteine and not to the unmasking of an endogenous cysteine; W103H inhibition by Zn²⁺ plus MTSET was not different from inhibition by Zn²⁺ alone. The increased inhibition by Zn²⁺ compared to that previously shown (Fig. 3) was most likely due to the use of Zn²⁺-binding buffer as opposed to KRH/glucose. Data represent the mean ± SEM for at least three independent experiments performed in triplicate.

Figure 6.

Zn²⁺ increases the rate of inactivation by MTSET at the hSERT V102C and W103C mutants. HeLa cells transiently transfected with V102C or W103C cDNAs were treated with increasing concentrations of MTSET in the presence or absence of 1 mM Zn²⁺ in Zn²⁺-binding buffer followed by assessing [³H]5-HT uptake as described in Materials and Methods. For MTSET inhibition assays performed in the absence of Zn²⁺, activity remaining was normalized to total uptake. In the presence of Zn²⁺, activity remaining was determined by normalizing MTSET uptake inhibition in the presence of Zn²⁺ to the uptake inhibition by 5 mM MTSET alone (i.e., in the absence of Zn²⁺). Nonspecific uptake was determined using 10 μM fluoxetine. (A) W103C exhibited an increase in the reaction rate of MTSET inactivation in the presence vs. the absence of Zn²⁺ with rates of 570 ± 90 M⁻¹ min⁻¹ and 290 ± 50 M⁻¹ min⁻¹, respectively. (B) MTSET inactivated V102C at a rate of 6300 ± 600 M⁻¹ min⁻¹ vs. 210 ± 40 M⁻¹ min⁻¹ in the presence or absence of Zn²⁺, respectively. Data represent the mean ± SEM for at least three independent experiments performed in triplicate.

Figure 7.

MTS-3-MTS cross-links V102C to M180C. HeLa cells transiently expressing hSERT C109A, M180C, V102C, or V102C/M180C cDNA were incubated with 1 mM MTS-3-MTS or MTSET in PBS/CM for 10 min at 22°C and washed twice before being assayed for [³H]5-HT uptake, as described in Materials and Methods. (A) MTS-3-MTS significantly inhibited [³H]5-HT uptake at V102C/M180C compared to V102C. Data were normalized to total uptake of the respective mutant. However, (B) MTSET inhibited uptake only at V102C/M180C. (#) p < 0.05 vs. V102C; (***) p < 0.001 vs. C109A. One-way ANOVA with post-hoc Bonferroni's multiple comparison test was performed for statistical analysis. Data represent mean ± SEM for three experiments performed in triplicate.

Figure 8.

Proximity of TMHs I and III in a SERT homology model. (A) A homology model of SERT was generated from the structure of LeuT_Aa using the hSERT sequence and energy minimization (Yamashita et al. 2005). The 12 TMHs of SERT are shown as ribbons, with each TMH assigned a unique color. TMH I is light blue, and TMH III is bright yellow. The area expanded in B is defined by the yellow box. (B) A close-up visualization of TMH I residues V102 and W103 and residues in TMH III. The C_α backbone of TMH I is a light blue ribbon, and the backbone of TMH III is shown as a bright yellow ribbon. Only residues examined in this study are shown. Approximate S-S distances between V102 and TMH III residues were estimated as if the residues had been mutated to cysteines in the homology model. Measured distances were V102–I179, 8.5 Å; V102–M180, 8.6 Å; V102–A181, inaccessible on opposite side of helix; V102–W182, 9.9 Å; V102–A183, 7.5 Å; V102–L184, inaccessible on opposite side of helix. These distances are in general agreement with the experimental data.