Skip to main content
PMC home page
eLife logoLink to eLife
. 2017 Jan 24;6:e19314. doi: 10.7554/eLife.19314

Direct assessment of substrate binding to the Neurotransmitter:Sodium Symporter LeuT by solid state NMR

Simon Erlendsson 1,2,3,4,5,†, Kamil Gotfryd 3,4,5,†,‡, Flemming Hofmann Larsen 6, Jonas Sigurd Mortensen 3,4,5, Michel-Andreas Geiger 7, Barth-Jan van Rossum 7, Hartmut Oschkinat 7, Ulrik Gether 3,4,5, Kaare Teilum 1,2,*, Claus J Loland 3,4,5,*
Editor: Gary L Westbrook8
PMCID: PMC5262378  PMID: 28117663

Abstract

The Neurotransmitter:Sodium Symporters (NSSs) represent an important class of proteins mediating sodium-dependent uptake of neurotransmitters from the extracellular space. The substrate binding stoichiometry of the bacterial NSS protein, LeuT, and thus the principal transport mechanism, has been heavily debated. Here we used solid state NMR to specifically characterize the bound leucine ligand and probe the number of binding sites in LeuT. We were able to produce high-quality NMR spectra of substrate bound to microcrystalline LeuT samples and identify one set of sodium-dependent substrate-specific chemical shifts. Furthermore, our data show that the binding site mutants F253A and L400S, which probe the major S1 binding site and the proposed S2 binding site, respectively, retain sodium-dependent substrate binding in the S1 site similar to the wild-type protein. We conclude that under our experimental conditions there is only one detectable leucine molecule bound to LeuT.

DOI: http://dx.doi.org/10.7554/eLife.19314.001

Research Organism: Other

eLife digest

All living cells need amino acids – the building blocks of proteins – in order to survive, yet few cells can make all the amino acids that they need. Instead, transporter proteins in cell membranes must take these molecules from the outside of the cell and release them to the inside. Some cells, including those in the brain, also release amino acids and molecules derived from them into the spaces outside of the cell to send signals to other nearby cells. Again, transporter proteins must move these signaling molecules back inside cells, to stop the signaling and to allow the molecules to be recycled. Importantly, problems with these uptake mechanisms have been linked to disorders such as depression, epilepsy and Parkinson’s disease.

One family of transporters involved in the uptake of amino acids are the “Neurotransmitter:Sodium Symporters”. Though these proteins are involved in processes that are fundamental to life, it remains unclear exactly how they work. Specifically, it has been heavily debated whether this family of transporters require one or two amino acid molecules to bind at the same time in order to help transport them across the membrane.

Now Erlendsson, Gotfryd et al. have analyzed a bacterial protein in the Neurotransmitter:Sodium Symporter family. This transporter takes up an amino acid called leucine into cells, and is commonly used as a model to understand this family of transporter proteins more generally. Using a technique called solid state nuclear magnetic resonance, Erlendsson, Gotfryd et al. could detect a single molecule of leucine bound to each transporter, but not a second one. This technique could also pinpoint that the leucine was located at the transporter’s central binding site. Leucine was never found at the proposed secondary binding site. Together these findings suggest that only one molecule of leucine binds to the transporter at any one time, and that it binds to the transporter’s central binding site.

Erlendsson, Gotfryd et al. have shown now how solid state nuclear magnetic resonance can be used to explore in detail how Neurotransmitter:Sodium Symporters move molecules across cell membranes. The next challenge is to use the same experimental setup to characterize other Neurotransmitter:Sodium Symporters. Doing so could potentially lay the groundwork for designing more specific and improved drugs to treat disorders like depression and Parkinson’s disease.

DOI: http://dx.doi.org/10.7554/eLife.19314.002

Introduction

The Neurotransmitter:Sodium Symporters (NSSs) are responsible for clearing neurotransmitters, such as dopamine, serotonin, norepinephrine, glycine and GABA from the synaptic cleft. The transporters are thereby crucial for the regulation of synaptic transmission in the CNS and alterations in their function have been linked to several psychiatric and neurological disorders such as depression, bipolar disorders, attention deficit hyperactive disorder (ADHD), epilepsy, and Parkinson’s disease (Broer, 2013; Kristensen et al., 2011). The understanding of the molecular mechanisms and structural (re)arrangements underlying NSS function has advanced significantly in recent years. The most detailed insight into structure-function relationships of NSSs comes from studies of the amino acid transporter, LeuT, from Aquifex aeolicus (Kantcheva et al., 2013; Kazmier et al., 2014; Malinauskaite et al., 2014, 2016; Piscitelli et al., 2010; Quick et al., 2012; Shi et al., 2008; Singh et al., 2007; Wang et al., 2012a, 2012b; Yamashita et al., 2005). Recent structures of the drosophila dopamine transporter (dDAT) (Penmatsa et al., 2013) and the human serotonin transporter (Coleman et al., 2016), which are eukaryotic members of the NSS family, confirm that LeuT is a reliable model protein and proves its value in understanding the molecular function of this class of transporters.

Functional studies of LeuT have suggested the existence of a secondary substrate binding site (S2) located in the extracellular vestibule of LeuT approximately 10 Å from the primary substrate binding site (S1) (Khelashvili et al., 2013; Quick et al., 2009; Shi et al., 2008). The S2 site is suggested to be an allosteric trigger, essential for coupling the energy from the electrochemical gradient to the transport of the solute. The binding of leucine to the S2 site has been measured to have the same affinity (in nM range) as binding to the S1 but does not, as the S1 bound substrate, directly coordinate sodium (Quick et al., 2012). However, attempts to crystallize LeuT with substrate bound to the S2 site have so far been unsuccessful, and therefore the existence of the S2 site is supported primarily by radioligand binding assays and guided MD simulations (Quick et al., 2012; Zhao et al., 2011). Due to the lack of structural evidence, the existence of a high-affinity S2 site has been questioned (Piscitelli et al., 2010), supporting the need for employing new techniques for investigating ligand binding in NSS proteins.

Here we investigate the leucine binding properties of LeuT by magic angle spinning (MAS) NMR, aiming at a characterization of the proposed S2 binding site. Our approach offers several advantages: (i) We use microcrystalline samples of LeuT prepared under experimental conditions allowing for conformations capable of ligand binding to both S1 and S2 (Quick et al., 2012). (ii) NMR offers information on the full structural ensemble which is unlikely not to include conformers (even lowly populated) prone to bind leucine in S2. (iii) Leucine binding to S1 and S2 may be distinguished by characteristic chemical shifts that are expected to be different due to different chemical environments, i.e. interacting residues (Reyes et al., 2011).

Results and discussion

Prior to crystallization and NMR experiments we verified the functionality of the produced LeuT wild-type (WT) samples. We initially performed [3H]leucine saturation binding experiments and subsequently assessed Na+-dependency of [3H]leucine binding. All experiments were done at a DDM concentration commonly used in in vitro assays (i.e., 0.05% corresponding to 5.7x CMC). At this detergent concentration, LeuT was reported to retain binding to both S1 and the putative S2 site (Quick et al., 2012). In scintillation proximity binding assays LeuT WT bound [3H]leucine with a dissociation constant (Kd) of 12 ± 1 nM in the presence of 200 mM sodium. The EC50 value calculated for the Na+-dependent binding was 47 ± 4 mM (Figure 1—figure supplement 1A–B). These values are in agreement with those previously reported for LeuT (Shi et al., 2008; Singh et al., 2008).

As we were primarily interested in a simple readout reporting solely on substrate binding, we purified and kept LeuT in the presence of 1 mM 15N enriched L-leucine to ensure substrate binding and detection in both sites (Figure 1A). With a 15N natural abundance around 0.3%, the background from the protein amides and amines is sufficiently low to distinguish even weakly populated states originating from the enriched substrate only.

Figure 1. Assessment of L-leucine binding to LeuT WT by solid state NMR.

(A) Cartoon illustration of experimental approach. 15N enriched L-leucine substrate is added to detergent reconstituted LeuT, which is subsequently crystallized using large scale sitting drop vapour diffusion. Rod-shaped microcrystals form within 24 hr and can be readily harvested. (PDB ID: 3F3E) (B) LeuT WT purified in NaCl (red) and LeuT purified in KCl (black). 15N L-Leucine specific peak is indicated by an asterix with a chemical shift of 38.2 ppm. Spectra are tentatively intensity normalized to the 15N natural abundance signal from the LeuT backbone amides. Signal-to-noise is calculated to be 21. (C) 23Na-NMR of LeuT WT (red) and LeuT WT in KCl (black) in presence of leucine. Minor peak at −8.9 ppm represents the shape of one or two structural sodium molecules. Despite inequivalent location of the two sodium sites in the LeuT, the coordination mechanism is almost identical which might account for the observation of a single peak in the 23Na-NMR spectrum instead of two distinct peaks.

DOI: http://dx.doi.org/10.7554/eLife.19314.003

Figure 1.

Figure 1—figure supplement 1. Functional characterization of LeuT WT.

Figure 1—figure supplement 1.

LeuT WT was purified in KCl and eluted in 0.05% DDM (pH 8.0). Scintillation proximity assay-based measurements of (A) [3H]leucine saturation binding to 100 ng LeuT in the presence of 200 mM NaCl and (B) Na+-dependent [3H]leucine binding (100 nM) by 100 ng LeuT. Ionic strength was compensated with KCl. Data are displayed as means ± s.e.m., performed in triplicates, n = 3.
DOI: http://dx.doi.org/10.7554/eLife.19314.004

Figure 1—figure supplement 2. Microscopy image of LeuT microcrystals.

Figure 1—figure supplement 2.

LeuT microcrystals are needle-shaped and have a length of 1–10 μm. The microscopic image was solely used to assess the quality of the microcrystalline material.
DOI: http://dx.doi.org/10.7554/eLife.19314.005

Figure 1—figure supplement 3. 1D 15N CP/MAS spectrum of frozen and lyophilized LeuT WT samples.

Figure 1—figure supplement 3.

(A) Protein concentration: 2 mg/ml, leucine concentration: 1 mM. The spectrum is dominated by the signal arising from the free unbound leucine, at ~42 ppm. Spectrum was recorded on a 400 MHz Bruker shielded wide bore magnet equipped with a 3.2 mm MAS HCN operating at ~100 K. Spinning rate: 8000 kHz. CP contact time: 1000 us, recycling delay: 3 s. 24 k scans were required to record presented spectrum. (B) 15N CP/MAS spectra from dry L-leucine powder (top panel) and from lyophilized leuT samples (bottom panel). Free 15N L-Leucine has a distinct chemical shift at 118 ppm. For the LeuT WT NaCl preparation (red) several additional peaks appear around the dominating free state peak. Though these peaks vary slightly in intensity when compared to LeuT WT KCl (dark). These peaks do not originate from structural leucine. Spectra were recorded on a 700 MHz Bruker shielded wide bore magnet equipped with a 4 mm MAS HCN probe operating at 298 K. Spinning rate: 12500 KHz, CP contact time: 1500 us, recycling delay: 2.5 s. 65 K scans were required to record presented spectra.
DOI: http://dx.doi.org/10.7554/eLife.19314.006

Figure 1—figure supplement 4. 15N L-leucine spectrum substrate peak for LeuT WT.

Figure 1—figure supplement 4.

(A) Line broadening for window function 1 Hz, Signal-to-noise in calculated to be 21, Full width half height (FWHH) of the substrate peak is 31 Hz. (B) Line broadening for window function 10 Hz.
DOI: http://dx.doi.org/10.7554/eLife.19314.007

Figure 1—figure supplement 5. In-solution 1D 15N spectra of free 98% 15N L-leucine at different pH.

Figure 1—figure supplement 5.

(A) pH titration of the L-leucine amine. The 1D 15N spectra are recorded for 1024 scans at 25°C. (B) 15N chemical shift of the L-leucine amine as a function of pH. From the sigmoidal curve fit the pI is estimated to be 9.72 ± 0.06.
DOI: http://dx.doi.org/10.7554/eLife.19314.008

Figure 1—figure supplement 6. 1D 13C CP/MAS spectrum of microcrystalline LeuT WT samples.

Figure 1—figure supplement 6.

Spectra for samples prepared in NaCl or KCl are colored in red and black, respectively. Resonances originating from labelled leucine are indicated by their chemical shifts and their respective assignments. These results are in good agreement with the 15N 1D spectra, and only one set of chemical shifts from the ligand can be observed.
DOI: http://dx.doi.org/10.7554/eLife.19314.009
Open in a new tab

To achieve sufficiently narrow line widths of the NMR signals and to avoid any signal from unbound leucine, we produced microcrystalline samples of LeuT (Figure 1—figure supplement 2), and performed cross polarization (CP)-based NMR experiments at temperatures above the freezing point. In all other preparations tested (frozen, lyophilized and proteoliposomes) the signal from the unspecific or unbound leucine completely dominated the spectra (Figure 1—figure supplement 3A–B). Using the microcrystalline preparations, we were able to produce the high quality CP-based 15N detected spectra showing one significant (above 2σ – Figure 1—figure supplement 4) peak at 38.2 ppm that could be assigned to the amine of protein bound leucine (Figure 1B). In addition to the sharp signal from leucine, much broader signals between 110 and 130 ppm were also observed, which originate from the 15N natural abundance of the LeuT amides (Figure 1B). To further assess whether the intense signal at 38.2 ppm reflects sodium specific leucine binding to LeuT, we performed a parallel experiment substituting Na+ with K+. Sodium is required for leucine binding (Zhao et al., 2011). By the use of 23Na-NMR we confirmed the presence of only a negligible amount of residual NaCl (at 7.1 ppm), and that no detectable Na+ was coordinated in the protein (Figure 1C). In the absence of Na+, the signal at 38.2 ppm in the 15N 1D spectrum disappeared as expected for a signal originating from 15N-leucine bound to LeuT (Figure 1B). Worth of note, the amine NH3+ group of free leucine has a chemical shift of approximately 41 ppm at pH 8 (Figure 1—figure supplement 5A–B), demonstrating that the bound substrate resides in a not fully solvent accessible environment. Importantly, we were unable to detect any signal from any additionally bound leucine. Similarly, the 13C CP/MAS spectra from the same samples clearly displayed only one single set of sodium dependent leucine signals (Figure 1—figure supplement 6).

To investigate whether the origin of the substrate peak at 38.2 ppm was due to leucine binding either to the S1 or the S2 site, we recorded solid state NMR spectra of two variants with compromised leucine binding, F253A (S1) and L400S (S2) (Figure 2A,B). For these experiments we lowered the final concentration of the added enriched substrate to 5 μM to ensure proper detimental effect by the mutations. This concentration has previously been reported not to provide any detectable [3H]leucine occupancy in the S2 site of LeuT L400S, but saturated S1 binding (Quick et al., 2012). For LeuT WT the specific leucine peak was unaffected by lowering the free leucine concentration (Figure 2—figure supplement 1). The F253A mutant has previously been shown to impair binding to the S1 site (Billesbølle et al., 2015; Wang et al., 2012b). Thus, F253A serves as a S1 disturbing mutant at low substrate concentrations. In the F253A 1D 15N spectrum, we observed sodium dependent substrate binding with a chemical shift of 38.4 ppm and a slightly lower signal intensity, when compared to the WT spectra (Figure 2C). The shift in F253A was consistent for both high (1 mM) and low (5 µM) leucine concentrations (Figure 2—figure supplement 3). Most importantly, the chemical shift difference of ~0.2 ppm for the observed bound leucine peak, demonstrates that the ligand is affected by the local environment of the S1 binding site. The L400S mutation was previously suggested to abolish S2 leucine binding (Quick et al., 2012). The 15N spectrum obtained for leucine bound to L400S (Figure 2C, blue) completely resembled the spectrum obtained with WT protein (Figure 2C, red). We were not able to detect any change in intensity or chemical shift. This argues against the possibility that the substrate signal we observe in LeuT WT samples is reporting on a combination of S1 and S2 binding. Also, we reason that it would be highly unlikely for an S1 bound substrate and a putative S2 site-bound substrate to have the same chemical shift as the environment of the putative S2 binding site markedly differs from the S1 binding site. As a major difference, the proposed S2 binding site does not involve direct sodium binding (Shi et al., 2008)

Figure 2. Effects of S1 and S2 site mutations on the L-leucine chemical shift .

(A–B) Cartoon representation displaying the location of F253 in the S1 site and L400 in the proposed S2 site based on PDB: 3USG (Wang et al., 2012a). (C) 15N 1D NMR spectrum of LeuT WT (red), F253A (green) and L400S (blue). Inset: Close-up of L-leucine specific peak. Spectra are tentatively intensity normalized to the 15N natural abundance signal from the LeuT amides.

DOI: http://dx.doi.org/10.7554/eLife.19314.010

Figure 2.

Figure 2—figure supplement 1. Comparison of spectra derived from LeuT WT purified and crystallized in either 1 mM (red) or 5 uM (purple) free leucine.

Figure 2—figure supplement 1.

Spectra are tentatively normalized to the LeuT natural abundance amide signal. As expected the substrate peak relates to LeuT concentration and not to the added free leucine concentration.
DOI: http://dx.doi.org/10.7554/eLife.19314.011

Figure 2—figure supplement 2. Power spectra of LeuT WT (red), F253A (green) and L400S (blue).

Figure 2—figure supplement 2.

To rule out that wrong phasing would cause the chemical shift difference observed for the substrate peak of the spectra presented in Figure 2C, we present the them also as power spectra (which is ultimately a squared magnitude spectra) where phase signs are not preserved. We have intensified the F253A signal to only demonstrate the chemical shift difference.
DOI: http://dx.doi.org/10.7554/eLife.19314.012

Figure 2—figure supplement 3. Comparison of LeuT WT and LeuT F253A in the presence of 1 mM free substrate.

Figure 2—figure supplement 3.

(A) Full spectrum displayed with a line broadening of the windows function of 100 Hz. Tentatively intensity normalized using natural abundance signals. (B and C) Close up of the substrate peak region showing preserved chemical shift difference between the substrate bound in LeuT WT (black dashed line) and LeuT F253A (blue dashed line) in two different free substrate concentrations.
DOI: http://dx.doi.org/10.7554/eLife.19314.013

Figure 2—figure supplement 4. Cartoon representation of S1 bound substrate (green).

Figure 2—figure supplement 4.

Sodium ions are depicted in blue. Measured chemical shift values in red (in ppm) and measured distances in black (in Å). Image is made from PDB file: 3F3E.
DOI: http://dx.doi.org/10.7554/eLife.19314.014
Open in a new tab

In conclusion, although all LeuT samples used in the present study were prepared in DDM at low concentration to exclude previously reported detrimental effects on the S2 binding site, we were only able to identify one single substrate signal at 38.2 ppm in the 15N spectra and one set of signals in the 13C spectrum (Figure 2—figure supplement 2). We note, however, that at our current signal-to-noise ratio (~20), we would not be able to detect species populated less than 5% of the structural ensemble. Based on the minor change in chemical shift in the F253A (S1) mutant, and the completely unaltered signal for the L400S (S2) mutant we reason that the observable bound leucine is located at the S1 binding site, thus supporting the idea that LeuT exhibit one single central binding site. We cannot exclude that the detection of S2 binding may only be possible upon the complete transition of the transporter towards a specific (yet unknown) conformation or that unfavourable crystal contacts might complicate S2 binding. However, several crystal structures have shown the binding of antidepressants, which overlaps with the putative S2 site, using these exact conditions (Singh et al., 2007; Zhou et al., 2009). As previously proposed for LeuT (Piscitelli et al., 2010) we speculate that the S2 substrate binding site, if present, is rather a transient site, responsible for optimal functionality of the transporter.

Materials and methods

Protein expression and purification

Expression of LeuT WT from Aquifex aeolicus was performed according to the protocol described previously (Billesbølle et al., 2015). LeuT WT was expressed in E. coli C41(DE3) transformed with pET16b encoding C-terminally 8xHis-tagged transporter (expression plasmid was kindly provided by Dr E. Gouaux, Vollum Institute, Portland, Oregon, USA). Briefly, isolated bacterial membranes were solubilized in 1% DDM (Anatrace, USA) in the presence of 1 mM 98% 15N-L-Leucine (Cambridge isotopes, Tewksbury, MA) and the protein was bound to nickel-charged affinity resin (Life Technologies, Carlsbad, CA). Subsequently, protein was eluted in 20 mM Tris-HCl (pH 8.0), 200 mM KCl, 0.05% DDM, 1 mM 15N-L-Leucine and 300 mM imidazole (KCl sample) or in the same buffer containing NaCl instead of KCl (NaCl sample). LeuT F253A and L400S variants were generated from the leuT gene using a QuikChange kit (Agilent Technologies, Santa Clara, CA) and purified similarly to the LeuT WT protein with the difference that 5 µM of 13C-15N-L-Leucine (Cambridge Isotopes) was used for co-purification, and the salt content in all buffers consisted of 50 mM NaCl and 150 KCl. The LeuT F253A variant in the presence of 1 mM substrate was prepared similar to the NaCl sample. Subsequently, all LeuT samples were dialyzed for approx. 36 hr at 4°C in the respective elution buffer without imidazole.

Functional characterization of LeuT WT

Functional characterization of the LeuT WT purified in KCl was performed using a scintillation proximity assay (SPA) (Quick and Javitch, 2007). Saturation binding of [3H]leucine (50.2 Ci/mmol; PerkinElmer, Waltham, MA) to purified LeuT WT was performed with 100 ng/well (1.66 pmol) of protein in buffer composed of 20 mM Tris-HCl (pH 8.0), 200 mM NaCl, 0.05% DDM, 20% glycerol in the presence of 1.25 mg/ml copper chelate (His-Tag) YSi beads (PerkinElmer). Sodium-dependency was measured at fixed [3H]leucine concentration of 100 nM with increasing concentrations of NaCl (NaCl was substituted with KCl for equal ionic strength) again using 100 ng/well LeuT WT. [3H]Leucine binding was monitored using MicroBeta liquid scintillation counter (PerkinElmer) and data were fitted to a one-site saturation or dose-response function, respectively, using Prism 7 software (GraphPad, San Diego, CA).

Preparation of microcrystals

Microcrystalline samples were produced by large scale sitting drop vapour diffusion method. 1 mL of the protein sample solutions were mixed 1:1 with the crystallization buffer composed of 100 mM NaCl (or KCl), 120 mM MgCl2, 28% PEG400, 100 mM MES or HEPES pH 6.5. The crystallization was carried out at 18°C. After approximately 20 hr a white precipitate could be harvested by centrifugation and transferred directly to the rotor. All samples were freshly prepared immediately before use. Microcrystals where visualized using a Leica M125 microscope with a 1.0x PlanApo objective.

Lyophilized protein preparation

Protein for lyophilisation was depleted of glycerol during dialysis, snap-freezed in liquid nitrogen and added to a freeze drier. The remaining powder could be transferred directly to the rotor.

Solution NMR

15N-L-Leucine was dissolved to a final concentration of 1 mM in the following buffer: 20 mM Tris-HCl (pH 8.0), 200 mM KCl, 0.05% DDM and added to an Economy WG5 NMR tube. The Experiment was run on a Bruker Avance III 500 MHz operating at a Larmor frequency of 50.667 MHz for 15N. The directly detected 15N spectra were recorded using a recovery delay of 2 s and an acquisition time of 500 ms. The total number of 1024 scans were used.

Solid sate NMR

Microcrystalline LeuT nitrogen spectra were recorded on Bruker Avance III 800 MHz wide bore (89 mm) spectrometer equipped with a 4 mm MAS HCN efree probe. Spectra were obtained at 275 K (measured temperature), at 12500 Hz magic-angle-spinning. 15N CP/MAS experiments were run for 60 K scans in blocks of 10 K scans and the magnet was fine-tuned between each block. Cross-polarization contact time was set to 1750 us. Initial recovery delay was set to 3 s. Protons were decoupled at 86 kHz during acquisition. 13C CP/MAS experiments were run for 2 K scans. Cross-polarization contact time was set to 2000 us. The initial recovery delay was set to 3 s. Spectra were displayed using a 1, 10 or 100 Hz line broadening for EM window function in topspin.

Sodium MAS NMR spectra were recorded on a Bruker Avance NMR spectrometer operating at a Larmor frequency of 105.8 MHz for 23Na using a double resonance probe equipped for 4 mm (o.d.) rotors. All spectra were recorded at room temperature employing a central transition selective 90 degree pulse (1.8 μs), a recycle delay of 2 s, an acquisition time of 40.9 ms, a spectral width of 75.19 kHz and a spin rate of either 9 or 10 kHz. The spectra are referenced to crystalline NaCl at 7.1 ppm.

Acknowledgement

We thank Trent Franks, Matthias Hiller, Daniel Stöppler for technical assistance with the acquisition of solid state NMR spectra. The work was supported in part by the Danish Independent Research Council – Sapere Aude (0602-02100B) (CJL), The Lundbeck foundation (R108-A10755, R151-2013-14302, R221-2016-847) (CJL, KTE), BioNMR (BIO-NMR-00232) (SE, KTE) and iNEXT (PID 1597) (SE, KTE).

Funding Statement

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Funding Information

This paper was supported by the following grants:

  • BioNMR BIO-NMR-00232 to Simon Erlendsson, Kaare Teilum.

  • iNEXT PID 1597 to Simon Erlendsson, Kaare Teilum.

  • Lundbeckfonden R151-2013-14302 to Kaare Teilum.

  • Lundbeckfonden R221-2016-847 to Kaare Teilum.

  • Lundbeckfonden R108-A10755 to Claus J Loland.

  • Det Frie Forskningsråd Sapere Aude, 0602-02100B to Claus J Loland.

  • Det Frie Forskningsråd DFF-4183-00581 to Claus J Loland.

  • bioSYNergy, University of Copenhagen's Excellence Program for Interdisciplinary Research to Claus J Loland.

Additional information

Competing interests

The authors declare that no competing interests exist.

Author contributions

SE, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article.

KG, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article.

FHL, Acquisition of data, Drafting or revising the article.

JSM, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article.

M-AG, Acquisition of data, Drafting or revising the article.

B-JvR, Acquisition of data, Drafting or revising the article.

HO, Analysis and interpretation of data, Drafting or revising the article.

UG, Drafting or revising the article.

KT, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article.

CJL, Conception and design, Analysis and interpretation of data, Drafting or revising the article.

References

  1. Billesbølle CB, Krüger MB, Shi L, Quick M, Li Z, Stolzenberg S, Kniazeff J, Gotfryd K, Mortensen JS, Javitch JA, Weinstein H, Loland CJ, Gether U. Substrate-induced unlocking of the inner gate determines the catalytic efficiency of a neurotransmitter:sodium symporter. Journal of Biological Chemistry. 2015;290:26725–26738. doi: 10.1074/jbc.M115.677658. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Broer S. Diseases associated with general amino acid transporters of the solute carrier 6 family (SLC6) Current Molecular Pharmacology. 2013;6:74–87. doi: 10.2174/18744672113069990034. [DOI] [PubMed] [Google Scholar]
  3. Coleman JA, Green EM, Gouaux E. X-ray structures and mechanism of the human serotonin transporter. Nature. 2016;532:334–339. doi: 10.1038/nature17629. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Kantcheva AK, Quick M, Shi L, Winther AM, Stolzenberg S, Weinstein H, Javitch JA, Nissen P. Chloride binding site of neurotransmitter sodium symporters. PNAS. 2013;110:8489–8494. doi: 10.1073/pnas.1221279110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Kazmier K, Sharma S, Quick M, Islam SM, Roux B, Weinstein H, Javitch JA, McHaourab HS. Conformational dynamics of ligand-dependent alternating access in LeuT. Nature Structural & Molecular Biology. 2014;21:472–479. doi: 10.1038/nsmb.2816. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Khelashvili G, LeVine MV, Shi L, Quick M, Javitch JA, Weinstein H. The membrane protein LeuT in micellar systems: aggregation dynamics and detergent binding to the S2 site. Journal of the American Chemical Society. 2013;135:14266–14275. doi: 10.1021/ja405984v. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Kristensen AS, Andersen J, Jørgensen TN, Sørensen L, Eriksen J, Loland CJ, Strømgaard K, Gether U. SLC6 neurotransmitter transporters: structure, function, and regulation. Pharmacological Reviews. 2011;63:585–640. doi: 10.1124/pr.108.000869. [DOI] [PubMed] [Google Scholar]
  8. Malinauskaite L, Quick M, Reinhard L, Lyons JA, Yano H, Javitch JA, Nissen P. A mechanism for intracellular release of Na+ by neurotransmitter/sodium symporters. Nature Structural & Molecular Biology. 2014;21:1006–1012. doi: 10.1038/nsmb.2894. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Malinauskaite L, Said S, Sahin C, Grouleff J, Shahsavar A, Bjerregaard H, Noer P, Severinsen K, Boesen T, Schiøtt B, Sinning S, Nissen P. A conserved leucine occupies the empty substrate site of LeuT in the Na(+)-free return state. Nature Communications. 2016;7:11673. doi: 10.1038/ncomms11673. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Penmatsa A, Wang KH, Gouaux E. X-ray structure of dopamine transporter elucidates antidepressant mechanism. Nature. 2013;503:85–90. doi: 10.1038/nature12533. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Piscitelli CL, Krishnamurthy H, Gouaux E. Neurotransmitter/sodium symporter orthologue LeuT has a single high-affinity substrate site. Nature. 2010;468:1129–1132. doi: 10.1038/nature09581. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Quick M, Javitch JA. Monitoring the function of membrane transport proteins in detergent-solubilized form. PNAS. 2007;104:3603–3608. doi: 10.1073/pnas.0609573104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Quick M, Winther AM, Shi L, Nissen P, Weinstein H, Javitch JA. Binding of an octylglucoside detergent molecule in the second substrate (S2) site of LeuT establishes an inhibitor-bound conformation. PNAS. 2009;106:5563–5568. doi: 10.1073/pnas.0811322106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Quick M, Shi L, Zehnpfennig B, Weinstein H, Javitch JA. Experimental conditions can obscure the second high-affinity site in LeuT. Nature Structural & Molecular Biology. 2012;19:207–211. doi: 10.1038/nsmb.2197. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Reyes N, Tavoulari S. To be, or not to be two sites: that is the question about LeuT substrate binding. The Journal of General Physiology. 2011;138:467–471. doi: 10.1085/jgp.201110652. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Shi L, Quick M, Zhao Y, Weinstein H, Javitch JA. The mechanism of a neurotransmitter:sodium symporter--inward release of Na+ and substrate is triggered by substrate in a second binding site. Molecular Cell. 2008;30:667–677. doi: 10.1016/j.molcel.2008.05.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Singh SK, Yamashita A, Gouaux E. Antidepressant binding site in a bacterial homologue of neurotransmitter transporters. Nature. 2007;448:952–956. doi: 10.1038/nature06038. [DOI] [PubMed] [Google Scholar]
  18. Singh SK, Piscitelli CL, Yamashita A, Gouaux E. A competitive inhibitor traps LeuT in an open-to-out conformation. Science. 2008;322:1655–1661. doi: 10.1126/science.1166777. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Wang H, Elferich J, Gouaux E. Structures of LeuT in bicelles define conformation and substrate binding in a membrane-like context. Nature Structural & Molecular Biology. 2012a;19:212–219. doi: 10.1038/nsmb.2215. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Wang H, Gouaux E. Substrate binds in the S1 site of the F253A mutant of LeuT, a neurotransmitter sodium symporter homologue. EMBO Reports. 2012b;13:861–866. doi: 10.1038/embor.2012.110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Yamashita A, Singh SK, Kawate T, Jin Y, Gouaux E. Crystal structure of a bacterial homologue of Na+/Cl--dependent neurotransmitter transporters. Nature. 2005;437:215–223. doi: 10.1038/nature03978. [DOI] [PubMed] [Google Scholar]
  22. Zhao Y, Terry DS, Shi L, Quick M, Weinstein H, Blanchard SC, Javitch JA. Substrate-modulated gating dynamics in a Na+-coupled neurotransmitter transporter homologue. Nature. 2011;474:109–113. doi: 10.1038/nature09971. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Zhou Z, Zhen J, Karpowich NK, Law CJ, Reith ME, Wang DN. Antidepressant specificity of serotonin transporter suggested by three LeuT-SSRI structures. Nature Structural & Molecular Biology. 2009;16:652–657. doi: 10.1038/nsmb.1602. [DOI] [PMC free article] [PubMed] [Google Scholar]
eLife. 2017 Jan 24;6:e19314. doi: 10.7554/eLife.19314.015

Decision letter

Editor: Gary L Westbrook1

In the interests of transparency, eLife includes the editorial decision letter and accompanying author responses. A lightly edited version of the letter sent to the authors after peer review is shown, indicating the most substantive concerns; minor comments are not usually included.

Thank you for submitting your article "Direct assessment of substrate binding to the Neurotransmitter:Sodium Symporter LeuT by solid state NMR" for consideration by eLife. Your article has been reviewed by two peer reviewers, including Baruch Kanner (Reviewer #2), and the evaluation has been overseen by Gary Westbrook as the Senior Editor. The reviewers have discussed the reviews with one another and the Senior Editor has drafted this decision to help you prepare a revised submission.

Summary:

This paper analyzes NMR data on (micro)crystalline preparations of LeuT and two mutants in reference to previous structural and functional studies that led to conflicting views on the number of leucine binding sites: a central site (S1) on which there is general agreement and a potential additional site (S2) located in the external vestibule of the transporter. This S2 site has been proposed to play an important role in the transport mechanism. Clearly, a detailed study of ligand binding to membrane proteins is a central aspect for understanding their workings in cell membranes. The LeuT system has served as a powerful bacterial model system to understand the details of ligand and ion concentrations as well as the role of lipids and detergents. The crystal structures of the bacterial amino acid transporter LeuT have been determined in various conformations. It has been established that LeuT is a very useful model for Neurotransmitter:Sodium:Symporters. For these reasons, this work addresses an important, so far unresolved issue.

In this manuscript, the issue is addressed using a novel approach with solid state NMR to study leucine binding in the wild type and two mutant transporters. The latter are expected to perturb either of the proposed sites. The authors describe a single sodium dependent signal in the wild type with similar chemical shifts and magnitudes at high (1mM) or low (5μM) substrate concentrations. At the low concentration, an identical signal was observed with the S2 site mutant, whereas the amplitude of the signal of the S1 mutant was lowered and the shift was different. The reviewers were interested in the topic and approach, but raised several issues that will need to be addressed in a revised manuscript.

Essential revisions:

1) Reliability of the NMR data interpretation.

The main conclusion of this work, i.e., the presence of only one bound Leucine site, is largely based on one-dimensional NMR spectra of rather limited signal to noise. This is especially true for Figure 2 where, considering the base line, the signal to noise S/N is roughly 4:1 (and not as claimed in Figure 1, 21:1). Although the data shown in Figure 2 seem to have been recorded with similar acquisition times, the signal to noise seems to vary significantly with the best S/N for the L400S mutant.

As a result, the "tentative normalization" of the 38 ppm peak to the NH backbone signals relative to the 38.2 ppm peak is questionable and the signal modulation at the 38 ppm resonance for the 3 samples is comparable to the noise level. Why did the authors not conduct longer experiments to obtain a better signal to noise ratio? In addition, the 15N signal of the F253A mutant refers to the spectrum with the lowest signal to noise and the claimed 0.2 ppm peak shift could easily disappear with a slight change in phase correction. Even if this shift is really present, it is probably smaller than the intrinsic NMR line width.

2) General relevance and implications in reference to previous work.

MAS-NMR has been used to study proteins embedded in proteoliposomes for decades and it remains unclear why such experiments were not conducted here. Because the substrate remains labeled, they should be readily possible. To the reviewer, such data would greatly improve the general relevance of this study because they would allow to ultimately compare detergent and lipid bilayer data on the atomic level (see Quick et al., 2012).

3) Because in the literature the S1 mutant has been described as capable of binding the substrate, but with lower affinity, the authors attribute the lowered amplitude to the lowered substrate affinity but do not substantiate this claim by measuring the signal at the high substrate concentration. It is essential to do this measurement and it will be important to see if the chemical shift changes or not.

4) In the last sentence the authors unnecessarily soften up their conclusion, probably to try to be "politically correct". Assuming that the suggested experiment strengthens their conclusion, the sentence could read something like "Our data, using a novel approach to determine substrate binding, support the idea that LeuT exhibits a single central binding site".

eLife. 2017 Jan 24;6:e19314. doi: 10.7554/eLife.19314.016

Author response

Essential revisions:

1) Reliability of the NMR data interpretation.

The main conclusion of this work, i.e., the presence of only one bound Leucine site, is largely based on one-dimensional NMR spectra of rather limited signal to noise. This is especially true for Figure 2 where, considering the base line, the signal to noise S/N is roughly 4:1 (and not as claimed in Figure 1, 21:1). Although the data shown in Figure 2 seem to have been recorded with similar acquisition times, the signal to noise seems to vary significantly with the best S/N for the L400S mutant.

As a result, the "tentative normalization" of the 38 ppm peak to the NH backbone signals relative to the 38.2 ppm peak is questionable and the signal modulation at the 38 ppm resonance for the 3 samples is comparable to the noise level. Why did the authors not conduct longer experiments to obtain a better signal to noise ratio? In addition, the 15N signal of the F253A mutant refers to the spectrum with the lowest signal to noise and the claimed 0.2 ppm peak shift could easily disappear with a slight change in phase correction. Even if this shift is really present, it is probably smaller than the intrinsic NMR line width.

We fully acknowledge that the intensities of the peaks in our spectra are low. Although it would be very nice with better signal-to-noise, we have already pushed the amount of NMR time used for recording a single 1D spectrum to the limit. The spectra presented in the paper were signal averaged in 6 blocks of 60K scans. The magnet was re-tuned before each block of scans, which resulted in just about 64 hours for each 1D experiment. Consequently, increasing the signal to noise by a factor of two would require at least 256 hours of experiment time, which in our opinion would be too much. It could risk a significant destabilization of the sample.

The data presented in Figure 1 clearly demonstrate the absence of additional Leucine binding sites that are populated more than 5% relative to the S1 site and that have a chemical shift difference of more than 30Hz (0.04ppm) compared to the S1 site.

In Figure 2, spectra of the variants F253A and L400S are shown. We did not obtain as much crystallized protein for these variants as for LeuT WT and the signal-to-noise for the S1 peaks in these spectra are indeed lower than for the WT spectrum (4:1 and 11:1, respectively). However, the peaks are sharp and their positions are well defined (intrinsic line widths of ~30 Hz). We agree with the reviewers that our presentation of the data can be improved. In the original figure we had processed the data with a line broadening of 100 Hz, hiding the resolution. To accommodate this issue, we have presented the data in the inset of Figure 2 with only 1 Hz line broadening. In addition, in Figure 2—figure supplement 2 we have produced power spectra, and aligned the intensities of the three constructs to demonstrate that the 0.2 ppm difference cannot be accounted for by minor phasing errors.

2) General relevance and implications in reference to previous work.

MAS-NMR has been used to study proteins embedded in proteoliposomes for decades and it remains unclear why such experiments were not conducted here. Because the substrate remains labeled, they should be readily possible. To the reviewer, such data would greatly improve the general relevance of this study because they would allow to ultimately compare detergent and lipid bilayer data on the atomic level (see Quick et al., 2012).

Proteoliposome preparations are very suitable for most MAS-NMR membrane protein studies, and we did indeed try this approach for the substrate binding but without any luck. As demonstrated in Figure 1—figure supplement 3, having either frozen or completely solid samples will cause the free leucine to dominate the spectrum completely. When recording proteoliposome data at room temperature we do not see any signal in the CP based experiments from substrate, which might be an effect of large flexibility or too low sensitivity (lipids will constitute most of the material in the ssNMR rotor in order to preserve the functionality of the transporter).

In the microcrystalline samples we have no unbound Leucine present as no signals are observed in the potassium purified protein, which serves as our negative control. Finally, microcrystalline samples have greatly improved resolution (more order) compared to fluid detergent/lipid preparations, which is crucial for our conclusions. To clarify these points, we have rephrased the text which now reads:

“To achieve sufficiently narrow line widths of the NMR signals and to avoid any signal from unbound leucine, we produced microcrystalline samples of LeuT (Figure 1—figure supplement 2), and performed cross polarization (CP)-based NMR experiments at temperatures above the freezing point. In all other preparations tested (frozen, lyophilized and frozen proteoliposomes) the signal from the unbound leucine completely dominated the spectra (Figure 1—figure supplement 3A-B).”

3) Because in the literature the S1 mutant has been described as capable of binding the substrate, but with lower affinity, the authors attribute the lowered amplitude to the lowered substrate affinity but do not substantiate this claim by measuring the signal at the high substrate concentration. It is essential to do this measurement and it will be important to see if the chemical shift changes or not.

We thank the reviewers for this insightful suggestion. As proposed we have performed the experiment for F253A at 1 mM free substrate concentration (Figure 2—figure supplement 3). To this end, we realize that the fourth paragraph of the Results and Discussion puts unnecessary emphasis on the slight change in intensity. Our conclusion that the bound leucine senses changes in the S1 local environment (caused by the F253A mutation) and therefore are bound in S1, rests primarily on the differences in chemical shift of the leucine signals between LeuT WT and F253A. The comparison of the signal intensities relies on tentative normalization based on the natural abundance resonances, which are not suitable for making conclusions on subtle differences in the affinity, as also noted above.

Taken together, we have changed the paragraph to read:

“In the F253A 1D 15N spectrum, we observed sodium dependent substrate binding with a chemical shift of 38.4 ppm and a slightly lower signal intensity, when compared to the WT spectra (Figure 2C). The shift in F253A was consistent for both high (1 mM) and low (5 µM) leucine concentrations (Figure 2—figure supplement 3). Most importantly, the chemical shift difference of ~0.2 ppm for the observed bound leucine peak, demonstrates that the ligand is affected by the local environment of the S1 binding site.”

4) In the last sentence the authors unnecessarily soften up their conclusion, probably to try to be "politically correct". Assuming that the suggested experiment strengthens their conclusion, the sentence could read something like "Our data, using a novel approach to determine substrate binding, support the idea that LeuT exhibits a single central binding site".

The last paragraph of the Results and Discussion now reads,

“Based on the minor change in chemical shift in the F253A (S1) mutant, and the completely unaltered signal for the L400S (S2) mutant we reason that the observable bound leucine is located at the S1 binding site, thus supporting the idea that LeuT exhibit one single central binding site.”

Articles from eLife are provided here courtesy of eLife Sciences Publications, Ltd

RESOURCES