Integrated AlphaFold2 and DEER investigation of the conformational dynamics of a pH-dependent APC antiporter

Abstract

The Amino Acid-Polyamine-Organocation (APC) transporter GadC contributes to the survival of pathogenic bacteria under extreme acid stress by exchanging extracellular glutamate for intracellular γ-aminobutyric acid (GABA). Its structure, determined in an inward-facing conformation at alkaline pH, consists of the canonical LeuT-fold with a conserved five-helix inverted repeat, thereby resembling functionally divergent transporters such as the serotonin transporter SERT and the glucose-sodium symporter SGLT1. However, despite this structural similarity, it is unclear if the conformational dynamics of antiporters such as GadC follow the blueprint of these or other LeuT-fold transporters. Here, we used double electron-electron resonance (DEER) spectroscopy to monitor the conformational dynamics of GadC in lipid bilayers in response to acidification and substrate binding. To guide experimental design and facilitate the interpretation of the DEER data, we generated an ensemble of structural models in multiple conformations using a recently introduced modification of AlphaFold2 . Our experimental results reveal acid-induced conformational changes that dislodge the Cterminus from the permeation pathway coupled with rearrangement of helices that enables isomerization between inward- and outward-facing states. The substrate glutamate, but not GABA, modulates the dynamics of an extracellular thin gate without shifting the equilibrium between inward- and outward-facing conformations. In addition to introducing an integrated methodology for probing transporter conformational dynamics, the congruence of the DEER data with patterns of structural rearrangements deduced from ensembles of AlphaFold2 models illuminates the conformational cycle of GadC underpinning transport and exposes yet another example of the divergence between the dynamics of different families in the LeuT-fold.

Keywords: acid resistance; amino acid transport; membrane protein biophysics; structure prediction.

Fig. 1.

Architecture and pH-dependent transport activity of GadC. (A) The crystal structure of GadC consists of a four-helix bundle domain, a four-helix hash domain, two gating helices, and two ancillary helices not involved in transport (not emphasized). Additionally, the C terminus is wedged into the intracellular cavity, locking the transporter into a putatively inactive conformation. (B) Proposed antiport mechanism consisting of inward-facing, outward-facing, and occluded conformations. This mechanism implicates glutamate and GABA in driving the outward-to-inward and inward-to-outward conformational transition, respectively, and it forbids substrate-free isomerization. (C) Glutamate transport by wild-type GadC reconstituted into proteoliposomes is pH dependent. Error bars correspond to the SEM (n = 3).

Fig. 2.

Detachment of the C terminus of GadC is triggered by low pH. (A) Position of the C terminus, shown in pink, relative to the main transmembrane domain of the transporter. Inset: rotated view of the C terminus embedded in the intracellular vestibule. (B) Distance distributions monitoring the docking of the C terminus. At low pH, a broad long-distance component is observed in equilibrium with a distance component consistent with predictions made from the crystal structure (shown in the dashed line). (C) Titration measurement of the dissociation of the C terminus. Error bars correspond to 95% CIs calculated using the program GLADDvu (see Materials and Methods). (D) pH-dependent increases in conformational heterogeneity resolved by continuous-wave EPR. (E) Titration measurement of the mobile component of the CW spectra reveals a pH dependence similar to the distance distributions shown in B.

Fig. 3.

Alternative conformations of GadC modeled by AF2. (A) Representative models generated using AF2 cluster into inward-open, fully occluded, and outward-open conformations. To highlight structural changes between the AF2 clusters and the crystal structure, key helices in each AF2 model are shown as light-colored ribbons, with the crystal structure shown in darker-colored ribbons for reference. (B) Dimensionality reduction using principal component analysis shows clustering of predicted conformations. Model confidence was measured using pTM, with greater values indicating greater model confidence. A fourth cluster with doubly-open models (Bottom center) was not analyzed. (C and D) Close-up views of the representative outward-open model from the top and bottom, respectively.

Fig. 4.

Conformational changes in the bundle domain on the intracellular side. Left: cartoon depictions of GadC shown from the intracellular side. Labeled positions are indicated as black spheres in cartoons, with the bundle domain shown in red. Right: distance distributions between the bundle domain and reference sites (in blue and green) are largely unimodal at neutral pH and become more heterogeneous at low pH. Short-distance components consistent with an inward-closed model were also observed in some distributions. Substrates were not observed to affect the distance distributions. Dashed traces are predicted distributions from the structure and the outward-facing AF2 model.

Fig. 5.

Conformational changes in the bundle domain on the extracellular side. Left: cartoon depictions of GadC shown from the extracellular side, with labeled positions indicated as black spheres. Right: distance distributions overlapped with predictions made from the crystal structure at neutral pH. Low pH coincided with equilibrium shifts toward populations consistent with an outward-open model in a substrate-independent manner.

Fig. 6.

The bundle domain of GadC does not behave as a rigid body. Measurements between spin labels (indicated by black spheres) across the membrane indicate a degree of intra- and interhelical flexibility. Low-pH measurements with GABA and glutamate overlapped with apo low-pH measurements and are omitted for clarity.

Fig. 7.

No large-scale pH-dependent changes in the positions of IL1 and EL4. Top: Position of EL4 as measured from residues 269 and 280. Bottom: IL1. Cartoon depictions of GadC shown on the left along with positions of labeling sites indicated by black spheres.

Fig. 8.

TM10 undergoes substrate-dependent dynamics independent from the hash domain. (A) Cartoon depiction of GadC highlighting TMs 6a and 10 and the spin-labeling sites (black spheres). (B) Distance distributions between TM6a and TM10 (residues 204/365) demonstrate that glutamate, but not GABA (purple and yellow traces are superimposable), induces a shift in the population of distance components (the long-distance peak at 48 Å is likely an aggregation artifact also present in the apo pH 7.5 trace). (C) Proposed activation and glutamate-GABA antiport mechanism (Left). The C-terminus detaches at weakly acidic pH, allowing isomerization to proceed. Glutamate and GABA lower an energy barrier separating the inward- and outward-facing conformations. No large-scale substrate-dependent conformational changes were observed in GadC that were equivalent to binding of sodium and substrates to sodium-coupled symporters such as Mhp1 and vSGLT (shown on the Right). The greyed-out portion of the transport cycle of Mhp1 and vSGLT showing sodium/substrate symport is not directly translatable to our proposed mechanism of antiport.