Molecular mechanism of substrate recognition by folate transporter SLC19A1

Abstract

Folate (vitamin B₉) is the coenzyme involved in one-carbon transfer biochemical reactions essential for cell survival and proliferation, with its inadequacy causing developmental defects or severe diseases. Notably, mammalian cells lack the ability to de novo synthesize folate but instead rely on its intake from extracellular sources via specific transporters or receptors, among which SLC19A1 is the ubiquitously expressed one in tissues. However, the mechanism of substrate recognition by SLC19A1 remains unclear. Here we report the cryo-EM structures of human SLC19A1 and its complex with 5-methyltetrahydrofolate at 3.5-3.6 Å resolution and elucidate the critical residues for substrate recognition. In particular, we reveal that two variant residues among SLC19 subfamily members designate the specificity for folate. Moreover, we identify intracellular thiamine pyrophosphate as the favorite coupled substrate for folate transport by SLC19A1. Together, this work establishes the molecular basis of substrate recognition by this central folate transporter.

Fig. 1. Cryo-EM structure of the BRIL-SLC19A1/Fab/Nb complex.

a Schematic diagram of BRIL-SLC19A1, anti-BRIL Fab, and anti-Fab Nb. b [³H]-MTX uptake assay to verify the function of BRIL-SLC19A1. The results are normalized to the activity of wild-type SLC19A1. All experiments were done in triplicates. (n = 3, mean ± SD). ns, non-significant; **** P < 0.0001 (Student’s t-test). c Profile of size exclusion chromatography (SEC) for the complex purification and the SDS-PAGE results to show the protein purity. d Cryo-EM map of the BRIL-SLC19A1/Fab/Nb complex. e Overall structure of the BRIL-SLC19A1/Fab/Nb complex. α-helices are shown in cylinders and β-strands are in ribbon. The residues of BRIL and SLC19A1 that are involved in electrostatic interactions are shown with side chains. f Cartoon diagram for the TM domain of SLC19A1. The TM numbers are labeled, and the plasma membrane is indicated with dotted lines. g Ribbon presentation of the SLC19A1 structure in two views. Two half TM bundles are colored in blue (TM1–6) and yellow (TM7–12), respectively. A close-up view of the unwound region of TM1 is shown in an inset.

Fig. 2. Cryo-EM structure of the SLC19A1/5-MTHF complex.

a Chemical structures of several folate analogs: 5-MTHF, folate, and MTX. The constituent moieties of 5-MTHF are indicated. b Ribbon diagram of SLC19A1 in complex with 5-MTHF. The four TMs (TM1, 4, 7, and 10) interacting with 5-MTHF are colored in cyan and the other ones are colored in blue. 5-MTHF is shown in sticks and its cryo-EM densities are shown in black meshes. c Superposition of the apo and 5-MTHF-bound SLC19A1 structures. d The electrostatic potential (in units of k_BT/e, where k_B is the Boltzmann constant, T is the absolute temperature and e is the elementary charge) of the substrate-binding pocket in SLC19A1, as calculated at pH 7.0 and 0.15 M concentrations of monovalent cations and anions. 5-MTHF is shown in sticks. e Ribbon presentation of the substrate-binding site in SLC19A1. The residues participating in 5-MTHF binding are indicated with side chains (< 4.0 Å). Hydrogen bonds and salt bridges are depicted as dashed lines. f A schematic summary of the interactions between SLC19A1 and 5-MTHF. The distances of the hydrogen bonds are indicated in angstroms. g The [³H]-MTX uptake activities of SLC19A1 mutants. The results are normalized to the activity of wild-type SLC19A1. All experiments were done in triplicates (n = 3, mean ± SD). ns, non-significant; * P < 0.01; ** P < 0.005; *** P < 0.001 (Student’s t-test).

Fig. 3. Analyses of the substrate discrimination mechanism of SLC19 subfamily members.

a Sequence alignment of the three SLC19 family members. The partially conserved residues are indicated with blue boxes and the strictly conserved ones are further filled with orange color. The residues of SLC19A1 that are involved in 5-MTHF binding are indicated. The conserved and non-conserved ones are denoted by black and red arrowheads, respectively. b, c Ribbon presentation of the 5-MTHF binding site in SLC19A2 and SLC19A3. The structures of SLC19A2 and SLC19A3 are predicted by AlphaFold. 5-MTHF is modeled into these structures based on the superposition with our SLC19A1/5-MTHF structure. The cognates of SLC19A2 and SLC19A3 corresponding to the 5-MTHF interaction residues of SLC19A1 are indicated. d Functional verification of the non-conserved residues for SLC19A1 using the [³H]-MTX uptake assay. The results are normalized to the activity of wild-type SLC19A1. All experiments were done in triplicates (n = 3, mean ± SD). ns, non-significant; * P < 0.01; ** P < 0.005; *** P < 0.001 (Student’s t-test).

Fig. 4. Verification of TPP as the favorite coupled substrate of SLC19A1.

a The chemical structure of TPP. The constituent moieties are indicated. b Ribbon diagram of SLC19A1 in complex with TPP. TPP is shown in sticks and its cryo-EM densities are shown in black meshes. c The electrostatic potential of the TPP binding site. d Structural comparison of the TPP- and 5-MTHF-bound SLC19A1. e Ribbon presentation of the TPP-binding site in SLC19A1. The residues involved in the interaction with TPP are shown with side chains. f A schematic summary of the interactions between SLC19A1 and TPP. The distances of the hydrogen bonds are indicated in angstroms. g Conformational change of the side chain of Arg133 between apo and TPP-bound SLC19A1 structures. h Inhibitory effect of different compounds on the [³H]-MTX uptake activity of SLC19A1. All the molecules were tested at a concentration of 200 μM. The results are normalized to the activity of the control experiment in which no inhibitors are added. All experiments were done in triplicates (n = 3, mean ± SD). ns, non-significant; * P < 0.01; ** P < 0.005 (Student’s t-test). i Quantitative measurement of the potency of TPP in inhibiting the [³H]-MTX delivery by SLC19A1. The results are normalized to the MTX transport activity of SLC19A1 in the absence of TPP. All experiments were done in triplicates (n = 3, mean ± SD). IC₅₀ was calculated by fitting to a nonlinear regression model. j Quantification of the fluorescence-detection size-exclusion chromatography (FSEC)-based thermostability assay. The concentrations of molecules are indicated. The thermostability is calculated relative to the control experiment in which SLC19A1 was not incubated with any compounds. All experiments were done in triplicates (n = 3, mean ± SD). ns, non-significant; * P < 0.01 (Student’s t-test). k MST analysis to examine the affinity of SLC19A1 for TPP. All experiments were done in multiple replicates (n = 6–10, mean ± SD).

Fig. 5. Model of the substrate transport cycle of SLC19A1.

SLC19A1 utilizes the alternating access mechanism to reverse transport two substrates. Under physiological conditions, 5-MTHF and TPP are likely the favorite extracellular and intracellular substrates of SLC19A1, respectively. They compete for the same binding site within the central cavity of SLC19A1. The four key residues (Glu123, Arg133, Tyr281, and Arg373) for substrate recognition are indicated.